Guillaume Cambois

Surface-consistent deconvolution in the log/Fourier domain

In the surface‐consistent hypothesis, a seismic trace is the convolution of a source operator, a receiver operator, a reflectivity operator (representing the subsurface structure) and an offset‐related operator. In the log/Fourier domain, convolutions become sums and the log of signal amplitude at a given frequency is the sum of source, receiver, structural, and offset‐related terms. Recovering the amplitude of the reflectivity for a given frequency is then a linear problem (very similar to a surface‐consistent static correction problem). However, this linear system is underconstrained. Thus, among the infinite number of possible solutions, a particular one must be selected. Studies with real data support the choice of a spatially band‐limited solution. The surface‐consistent operators can then be calculated very efficiently using an inverse Hessian method. Applications to real seismic data show improvement compared with previous techniques. Surface‐consistent deconvolution is robust and fast in the log/F...

Flam - A Simultaneous Source Wide Azimuth Test

Summary A wide azimuth acquisition test was performed in the Gulf of Mexico where two sources out of four in total were fired simultaneously. All firing times were randomized within a small window to minimize the interference from the near simultaneous source as well as minimizing the effect of remnant coherent noise from the preceding shot. Results show direct comparisons between a conventional wide azimuth acquisition and the simultaneous source method as well as decimated results which represent apple to apple comparison in terms of fold and shot positions. The simultaneous source method has the potential to significantly improve wide azimuth acquisition as well as other acquisition geometries.

Multiple elimination using a pattern‐recognition technique

Standard prestack multiple elimination techniques, such as predictive deconvolution or Radon transforms, fail in the presence of complex structures.

Borehole transmission tomography for velocity plus statics

The accuracy of velocity tomograms reconstructed from borehole transmission traveltime data is highly sensitive to traveltime statics. We present a least‐squares tomography algorithm that includes a traveltime static term. The algorithm solves for both the velocity field and the traveltime statics simultaneously. This enables us to separate traveltime signal from traveltime noise, reducing the tomographic velocity artifacts caused by the statics. The incorporation of a priori constraints on the poorly determined spectral components of the velocity field further improves accuracy by reducing velocity artifacts as a result of uneven ray coverage. Application of the algorithm to numerical crosswell data results in velocity and statics’ estimates that are accurate to within 1 percent. Application of the algorithm to Exxon’s Friendswood tomography data results in velocity and statics’ estimates that correlate with independent data.

Surface‐consistent phase decomposition in the log/Fourier domain

In the log/Fourier domain, decomposing the amplitude spectra of seismic data into surface‐consistent terms is a linear problem that can be solved, very efficiently, one frequency at a time. However, the nonunique definition of the complex logarithm makes it much more difficult to decompose the phase spectra. The instability of phase unwrapping has previously prevented any attempt to decompose phase spectra in the log/Fourier domain. We develop a fast and robust partial unwrapping algorithm, which makes it possible to efficiently decompose the phase spectra of normal moveout‐corrected (NMO‐) data into surface‐consistent terms, in the log/Fourier domain. The dual recovery of amplitude and phase spectra yields a surface‐consistent deconvolution technique where only the average reflectivity is assumed to be white, and only the average wavelet is required to be minimum‐phase. Each individual deconvolution operator may be mixed‐phase, depending on its estimated phase spectra. For example, surface‐consistent tim...

A LINEAR FILTER APPROACH TO DESIGNING OFF-DIAGONAL DAMPING MATRICES FOR LEAST-SQUARES INVERSE PROBLEMS

Many geophysical data processing applications are formulated as either linear or linearized‐nonlinear inverse problems. Usually, the linear systems that describe these problems are ill‐posed. For example, seismic deconvolution is an ill‐posed inverse problem because the seismic wavelet is band limited. Borehole transmission tomography is ill‐posed because of insufficient ray coverage around the tomographic model (Worthington, 1984). Surface‐consistent statics’ estimation is intrinsically ill‐posed because of the geometrical relationships between the parameters defined by the surface-consistent model (Taner et al., 1974).

AVO attributes and noise: Pitfalls of crossplotting

Summary AVO crossplotting has been widely used in the past few years as a way of deriving improved hydrocarbon indicators from seismic data. By crossplotting the standard AVO attributes of intercept and gradient, it is possible to obtain an optimum combination of the two (the fluid factor), which is insensitive to the AVO effect of brine-saturated shales and sands. Any remaining AVO anomaly can then be attributed to hydrocarbons or other lithologic factors. In addition, the physical location of an AVO anomaly on the crossplot gives an indication as to the geological setting of the potential reservoir. However, like the stack, intercept and gradient are sensitive to noise. While the intercept standard-deviation is slightly higher than the stack, the gradient standard-deviation is much larger and dramatically increases with travel-time. This partly explains the scale difference between the two attributes that is always observed with real data. Furthermore, in the presence of noise, intercept and gradient become statistically correlated. This correlation biases fluid factor calculations so that this attribute is reduced to a mere far-offset stack. Since stack and gradient do not correlate statistically, their crossplot can be used to validate or dismiss a trend observed in an intercept versus gradient crossplot. If the trend still exists, albeit slightly rotated, in the stack versus gradient crossplot, it is a lithologic effect; if not, it is a statistical artifact. Applications to real data have shown that, in general, what could be interpreted as a background shale trend is in fact noise-related. Consequently, the calculated fluid factor corresponds to a far-offset stack, which is actually a legitimate hydrocarbon indicator, but not as good as the theoretical fluid factor. Improved hydrocarbon indicators can nonetheless be obtained using the statistical independence of stack and gradient.

Can we surgically remove swell noise

The weather is one of the main causes of down-time in marine seismic acquisition. When the swell noise generated by a rough sea reaches a given level, acquisition has to stop. When the weather gets even rougher the cables have to be pulled out of the water to avoid losing them. There is however a range of weather conditions during which shooting is safe but not performed because the amount of noise exceeds client specifications. The surgical removal of swell noise, that is to say the total elimination of noise and the full restoration of signal, would increase the level of noise that can be accepted, possibly to the point where shooting stops only when it is unsafe. This would significantly reduce acquisition downtime and therefore cost. Known noise attenuation techniques are applied on a test data set to check whether they fulfill the > condition. Although most of them fail to pass, a recently developed method called gives results that are sufficiently close to deserve the > label. The next step is of course to install an on-board procedure that will surgically remove swell noise using this technique, possibly in real time.

An introduction to this special section Instrumented Oil Fields

The instrumented oil field consists of deploying permanent instrumentation to monitor an oil field and modify production continuously or on demand. This concept has evolved from recent advances in both downhole permanent instrumentation and time-lapse monitoring. Both of these technologies are new, and combining them, with permanent installations in mind, requires further steps in research and development. As a result, this special section on the instrumented oil field can be viewed as a “road ahead” section to help the reader to construct not only a vision of the instrumented oil field but also to become familiar with the many issues surrounding this new technology.

Wide-azimuth acquisition: True 3D at last!

When it comes to 3D, processing has clearly been lagging behind acquisition, perhaps even to the point of holding back the deployment of this technology. One reason is that computer capacity always had to catch up with increasing survey size and true 3D processing; but the real obstacle was (and still is) the additional degree of complexity associated with azimuth.