Search

An automated and real-time field PSTM system, called TeraMig, was applied during the acquisition of a land 3D WAZ survey for PDO. The system was used to migrate one million vibrated points (around 10 billion traces) and was able to generate a real-time field PSTM cube as the seismic shots were recorded. The intermediate PSTM cubes were available in the field for instantaneous quality control. Daily reports including seismic data were sent to the end users (i.e., client, processing center). Immediately after the last recorded VP, the complete PSTM volume was ready to be delivered.

We present how to create realistic geology conformal anisotropic velocity models and reduce depth misties in standard tomographic and high-resolution FWI depth velocity modelling. We assume that localized variations in both velocity and anisotropy are caused by changes in the lithology and we use well information to establish anisotropy/velocity correlation for imaging velocity models.

A conventional CMP based COV binning will create footprint and migration noise for acquisitions where sources and receivers are at different depth levels and/or PS waves are imaged. The OVT muting & weighting technique is a way to overcome this problem. The OVT muting and weighting technique is a way to bin acquired seismic data in such a way that a uniform, or near uniform, illumination is obtained at all depth levels in the model beneath a certain given shallow depth level. The basic idea is to adjust the number of traces in a OVT gather and create a mute based on the depth- and offset dependent CMP-CRP difference computed by ray tracing. This will improve the CRP coverage considerably. This is done on pre-migration data and will create gathers that can be migrated by e.g. Kirchhoff migration. The method can be used for any acquisition technique but will be especially powerful for Ocean Bottom Cable (OBC) or Ocean Bottom Nodes (OBN) acquisitions. It can be used in both Kirchhoff time and depth migration.

The seismic monitoring solution presented here is a permanently buried, fully automatic, and continuous seismic acquisition and processing system. It ensures remarkably repeatable daily seismic. Our specific calendar oriented 4D processing flow is described and applied on a monitoring system installed for Shell on their Peace River project to provide daily monitoring of a heavy oil production pad. The main observation is that 4D attributes vary a lot even when looking at very short calendar periods. This continuous monitoring information gives significant insights into reservoir activities and offers new opportunities to better understand the short term dynamics of the reservoir.

In marine seismic surveys, seismic interference (SI) remains a considerable problem when marine seismic data sets are acquired in close vicinity of each other. We present a method for attenuating SI noise using a sparse Tau-P transform. Using a synthetic example, we demonstrate that this method effectively attenuates SI noise while preserving the seismic signals. Applying our method to real data confirmed our observations from the synthetic data example. Compared to the conventional Tau-P based method, our method leaves fewer residuals without observable damage to the primary signals.

The anelastic effects of the overburden cause seismic amplitude attenuation, wavelet phase distortion and seismic resolution reduction. It is desirable to correct the frequency dependent energy attenuation and phase distortion in a prestack depth migration. The situation becomes more challenging in complex geological regions with the presence of absorption, such as complex shallow gas clouds. The wavefields can be severely complicated and distorted by strong velocity contrast with the surroundings and serious attenuation, hence mask the image of the deep objectives. To deal with such challenges associated with absorption in complex geological regions which quite often needs to account for multi-pathing and anisotropy in wave propagation, we have developed a stable visco-acoustic TTI Reverse Time Migration based on our derived formulation for visco-acoustic wave propagation in TTI medium to compensate for the anelastic effects in the seismic data. We will demonstrate our visco-acoustic TTI Reverse Time Migration with both synthetic and field data examples.

We present a method for estimating the shape of salt bodies by using ray-based nonlinear slope tomography. The data to invert mainly consist of residual moveout (RMO) observed below salt shapes on migrated common image point gathers. We compute Fréchet derivatives made of traveltime derivatives with respect to depth Bsplines parameters describing salt bounding surface. Salt shape is progressively updated after each linearized tomographic step. We demonstrate the method on a 2D synthetic example and on a high-fold 3D land dataset from the Sultanate of Oman.

While most marine baseline surveys do not use broadband techniques, increasingly more monitor surveys are being adapted to broadband techniques that often use different receiver-depth profiles (e.g., deeper or variable-depth). A regular matching filter that has been commonly used in 4D time-lapse processing may be insufficient to normalize the wavelet difference between baseline data and monitor data because of the large difference in receiver-depth profiles. Deghosting can effectively remove the wavelet difference among vintages caused by different receiver depths. However, deghosting different vintages separately may cause inconsistent deghosting of common events and thus create false 4D signals. We propose a joint inversion scheme that uses both the baseline and monitor data (or more vintages) that deghosts common events consistently while preserving the difference among vintages. Using synthetic and field data, we demonstrate that joint deghosting baseline and monitor data provides a more accurate ghost removal and a more reliable 4D difference over separate deghosting of both data.

In the oil rich Bohai area, Ocean Bottom Cable (OBC) acquisition has become the new trend with the benefit of operational flexibility, better illumination, better multiple elimination and better S/N for the targets at middle to deep depth. However, the presence of azimuthal anisotropy can cause severe imaging challenges in the Wide Azimuth OBC data, particularly fault imaging which is sensitive to velocity accuracy. Fault imaging can be smeared and fault shadow can be observed within complex strike-slip fault system if the azimuthal dependency of wave propagation is not properly honored. In this paper, we will present a new orthorhombic model building flow and demonstrate with Luda OBC data that our approach can successfully reconcile the structural discrepancies between seismic images from different azimuths, thus provide a clear and sharp fault image with the WAZ stacking process.