Search

Recent advances in full-waveform inversion (FWI) algorithms have allowed it to work stably and effectively in different geological settings, especially in areas with geobodies of large impedance contrasts such as salt, despite its presently taken acoustic assumption. This has resulted in a leap in salt velocity model building over the conventional interpretation-driven approach, providing significantly improved salt models and a step-change in subsalt imaging. Furthermore, it produces a directly interpretable FWI Image, the normal derivative of the FWI velocities, that is now widely accepted as another step- change for subsalt imaging. However, elastic effects at the large impedance contrasts cause smearing of the salt boundary and a considerable salt halo in acoustic FWI results. To mitigate this issue, we have incorporated an elastic modeling engine into our acoustic Time-lag FWI algorithm (A-TLFWI) that has proven effective in mitigating the negative impact of large differences between synthetic and real data at sharp contrasts. Elastic TLFWI (E-TLFWI) can model the energy at large impedance contrasts with better amplitudes and phases. This intrinsically reduces data mismatch between synthetic and real data and thus further improves the convergence. Using an ocean bottom node (OBN) data set in the Gulf of Mexico (GoM), we demonstrate that E-TLFWI can significantly reduce the salt halo that typically occurs in A- TLFWI. Although it does not change the model kinematics much compared to A-TLFWI, as evident in RTM images, it provides FWI Images of more balanced amplitudes, improved focusing, and higher signal-to-noise ratio (S/N).

Full-waveform inversion (FWI) has become the norm in velocity model building for different surveys from land to marine, and from streamers to ocean bottom nodes. FWI using wide-azimuth (WAZ) streamer data in areas with complex geologic settings can fall short of resolving the complexity when starting from a poor initial velocity model. Previous studies have shown that iterating salt scenarios and Time-lag FWI (TLFWI), an effective and stable FWI algorithm for salt, can lead to a better initial model for the next round of FWI and eventually result in a step-change in the salt model and the subsalt image, even when only using WAZ streamer data. However, this approach relies on RTM images for the evaluation of model updates, which are highly susceptible to illumination issues. The recently developed FWI Imaging algorithm can bolster the approach by reducing the impact of poor illumination for subsalt imaging, easing model evaluation. In our case study in the Perdido area, we employ TLFWI and FWI Imaging to maximize the value of WAZ data for velocity model building and subsalt imaging.

Most seismic surveys carried out in offshore Brazil consist of narrow-azimuth towed-streamer (NATS) acquisitions. In areas such as the South Atlantic offshore basins, the geological complexity results in illumination issues for NATS acquisitions, especially in the deep pre-salt section. Both Full-Waveform Inversion Imaging (FWI Imaging) and Least-Squares RTM (LSRTM) can be used to tackle this issue. Furthermore, internal multiples are a common problem. When coupled with inhomogeneous illumination, these can cause strong distortions to the geological structures and render seismic images uninterpretable. Using a data set covering a complex salt mini-basin in Campos basin, we demonstrate the capacity of FWI Imaging to naturally alleviate distortions associated with internal multiples and illumination issues. We also compare it to LSRTM and highlight some of the differences.

Recent technology advancements have enabled us to revisit the Foz do Amazonas (2014) and Barreirinhas (2016) narrow-azimuth towed-streamer (NATS) seismic surveys and tackle specific imaging challenges in a way that the legacy processing could not. In an exploratory frontier such as the Brazilian Equatorial Margin (BEM), the petroleum system knowledge is limited, with few wells and no significant commercial discovery in the area yet. Thus, high- quality seismic data is crucial for reducing risk in the initial exploration phases.

Monument field is located in the northwest part of Walker Ridge in the Gulf of Mexico, with a Paleogene sandstone reservoir ?9 km below sea surface. The complicated overburden of low reflectivity shale bodies and complex salt geometries with steep salt flanks, multi-level weld systems, and large overhangs poses tremendous challenges to subsalt imaging, especially when the input data is suboptimal. Legacy images of streamer data from conventional velocity model building (VMB) driven primarily by manual salt interpretation, which could not systematically address the complicated overburden, show broken and distorted subsalt structures and thus render large uncertainties for further reservoir characterization. The well-proven Time-lag FWI (TLFWI) provides a leap in resolving the complex salt model and brings a step change to subsalt imaging. FWI Imaging leads to another step change in subsalt imaging by maximizing the illumination power of existing data through least-squares fitting of the full-wavefield data. Reprocessing using an iterative TLFWI and FWI-guided scenario VMB workflow and FWI Imaging technology makes the most out of the existing streamer data and obtains a step-change improvement in the subsalt image in this area. However, limited constraints from input streamer data can still bring challenges and uncertainties to the current results. This is especially true in deep sections beyond diving wave penetration depth, which require better input data with long offset, full azimuth, and good low frequency, such as Ocean Bottom Node (OBN), for the next level of subsalt imaging. Meanwhile, the strong elastic effects at the salt boundaries degrade the current velocity model obtained from TLFWI using only acoustic modeling. Therefore, we apply the recently developed elastic Time-lag FWI (E-TLFWI) to this streamer data, which gives a further improved velocity model with sharper salt boundaries and higher S/N, and better reverse time migration (RTM) image and FWI Image, especially around the complex salt boundary.

Free-surface multiple prediction of land data is challenging due to the unknown downward reflecting free-surface combined with poorly sampled recording of the shallow multiple generators. We use a wave-equation deconvolution approach to derive an image of the shallow multiple generators using the multiple periodicity. The image is then used to model multiples which are subsequently subtracted from the input data. The approach adjusts to variations in the shallow subsurface and the multiple model requires minimal adaptation.

Heavily faulted and karstified carbonate overburden poses potential geohazards for well drilling in West Papua, offshore Indonesia. Hence, high-resolution seismic images that capture geological details, such as karstified faults and small-scale caves within the carbonate, are highly desirable. In this study, two aspects play a pivotal role. First is the velocity model, which can capture velocity contrasts that are highly spatially varying within the carbonate. Second, considering the shallow nature of the targeted carbonate layer and the sampling limitation of the Ocean-Bottom Cable (OBC) acquisition, is a solution that handles the associated illumination issue. For the former, we demonstrate with Time-Lag Full-Waveform Inversion (TLFWI) that a high- resolution velocity model that conforms to geology can be obtained, which in turn generates significant uplift in the image using primary reflections. For the latter, we propose to use multiple imaging to further enhance the resolution of the structures within the carbonate. The resulting image provides new insight into the understanding of the fault and karst systems for the well drilling plan.

The Trion field in the western Gulf of Mexico (GoM) exhibits folded sediment beddings and strong attenuation bodies, which pose great challenges for seismic imaging. After a decade of processing effort employing the latest imaging technology, the available towed-streamer data for this area hit a technical limit. Improvements to velocity models and images were incremental, and it was determined that better data were needed to make a step change in image quality. To this end, an ocean bottom node (OBN) acquisition was carried out in late 2020 to record full- azimuth and long-offset seismic data with good low- frequency content. Time-lag FWI (TLFWI) using this OBN data was able to improve the tilted orthorhombic (T-ORT) velocity models, which led to better migrated images from least-squares (LS) Kirchhoff and reverse time migration (RTM). However, the reservoir image below the shallow absorption anomalies remained unsatisfactory. As an alternative imaging product, the FWI Image from TLFWI showed better structural continuity and more coherent amplitudes at the reservoir than LS-Kirchhoff or LSRTM.

Targeting of stratigraphic traps is rapidly becoming a successful exploration strategy, especially in relatively mature basins where a high proportion of structural traps have been tested. The development of robust geological and trap models, integrated with the results of seismic inversion, are key steps in mitigating the larger uncertainty and risk associated with exploration for stratigraphic traps. This study demonstrates the benefit of a fully integrated workflow, involving the interpretation and analysis of geophysical and geological data, to ensure the development of a robust prospect model. In this case, a new stratigraphic trapping concept has been developed within the Upper Jurassic Hanifa Formation a unit which is generally considered to consist of low porosities based on existing well control