Search

Imaging PS-wave data acquired in the shallow water at Alwyn North with ROV-deployed ocean-bottom nodes presented particular challenges due to the sparsity of the receivers. Having ensured vector fidelity of all recorded wavefields, the processing flow made simultaneous use of the PP and PS wavefields at several junctures, including construction of the imaging velocity-depth model, requiring all wavefields to be processed in parallel and in a consistent manner.

Changes in water velocity produce significant 4D noise in time-lapse images. To be addressed accurately, the water-velocity problem requires two major ingredients: 1) water velocity must be estimated accurately at all acquisition times and for all shot/receiver locations, 2) time-variable corrections to the data must be dynamic to treat the full wavefield accurately. We present a new approach to parameterization of water-velocity changes which minimizes sensitivity to water depth. Dynamic correction of the wavefield is then achieved by designing 3D time-variable phase-shift operators. This is applied in a tau-px-py least-squares modeling process.

Data acquired in shallow water environments often exhibit a strong acquisition pattern relating to variations in incidence angle, azimuth and source signature from inner to outer streamers. These variations must be accurately compensated for during processing to reveal a spatially consistent 3D image of the data. With modern broadband acquisition this problem becomes more challenging as the frequency spectrum is wider. This North Sea case study, acquired with a multi-level source and variable depth streamer, presents an enhanced processing sequence containing 3D processing steps using different parameters for each frequency range. Results show the advantage of such a sequence to achieve a high definition image from the very shallow to the deep reservoir target.

A recent 3D Broadseis survey was performed in the southern offshore area of Gabon, showing a wealth of detailed information in the seismic data. Understanding how these structures relate to the tectonic evolution of this basin requires the integration of the concurrently acquired gravity and magnetic data. 3D gravity and magnetic models test key geological questions relating to the evolution of this basin, and aid in outlining the nature of the extension and the composition of the crust.

In this paper a method is described to update the low frequencies of the elastic parameters coming from pre-stack time lapse inversion in a data driven manner. In the process no assumptions are made regarding the nature of the reservoir changes. Also no assumptions are made regarding the relationships between elastic parameters or reservoir parameters. The method described here can be seen as a 4D extension to a 3D updating method described by Mesdag et.al. 2010. In this way independent measures are obtained of time lapse absolute P-Impedance and time lapse absolute Vp/Vs (or  and ). These can then be compared with propagation parameters derived from 4D time shifts or velocities.

We propose a data-driven interferometry technique to remove low frequency aliased and non-conical surface waves in cross-spread domain. Despite insufficient sampling of the constructive regions in the cross-spread domain, the proposed approach has been designed for effectively handling any kind of 3D sparse geometry from Narrow to Wide-Azimuth land data using prior regularization and/or densification. This implementation provides a cost-effective workflow for large datasets and produces good removal of spatially aliased non-linear surface waves with minimal primary leakage as observed in a sparse acquisition land survey where the characteristics of the surface waves vary strongly.

We present a Reflection FWI (RFWI) workflow to update the velocity model using the low-wavenumber component of the FWI gradient of reflection data. This is achieved by alternately using high-wavenumber and low-wavenumber components to update density and velocity models, respectively. With synthetic examples, we discuss the limitations and requirements of this approach and propose possible ways to overcome some of the limitations. Finally, the method is applied to a deep-water survey in the Gulf of Mexico, where significant improvement is observed.

Data from towed streamer surveys in shallow water depths are usually contaminated by different orders of seafloor reflections called water-layer-related multiples (WLRMs). The model-based water-layer demultiple (MWD) method effectively removes WLRMs. The initial development of 2D MWD was followed by a 3D implementation to better handle out-of-plane WLRMs. This 3D implementation was further improved by using a selective-input strategy to handle the inconsistency of the high-frequency multiple patterns between adjacent sail lines. We also used time-variant apertures based on multiple contribution gathers (MCGs) to include sufficient large apertures, avoid aliasing problems, and compensate for apex shifting. Using narrow azimuth towed streamer (NATS) data from the Hibernia field, we demonstrate the benefit of selective-input MWD with time-variant apertures for attenuating WLRMs in shallow water data.

Full waveform inversion usually fails to update the sediment velocity in close vicinity of the top of salt (TOS) because of the strong velocity contrast between the sediment and salt. This phenomenon is a common challenge in practice, and it is not yet well understood. We investigate the relationship between FWI’s sediment velocity update and the accuracy of the TOS. The results indicate that the initial model with inserted salt helps update the sediment model, and a more accurate TOS provides a better sediment velocity update. However, an accurate TOS is not available unless the sediment velocity, particularly right above the TOS, has been correctly updated. We therefore propose a workflow to obtain a reasonable TOS for the FWI sediment velocity update using iterative FWI and salt interpretation. Using 2D synthetic data and real data, we demonstrate that our workflow yields a better FWI update above the salt than compared to using the sediment model as the initial model.