Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Earth science
Reexamination Certificate
2000-09-08
2002-10-08
McElheny, Jr., Donald E. (Department: 2862)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Earth science
Reexamination Certificate
active
06463388
ABSTRACT:
BACKGROUND OF THE INVENTION
As it is well known, seismic amplitudes have been traditionally used to identify and quantify lithology and fluid types present in the subsurface. This reduces the overall risk in the hydrocarbon exploration and exploitation stages. Post stacked seismic data often show amplitude patterns that resemble the acquisition geometry. This is commonly referred to as “Acquisition Geometry Imprint (AGI).” One example where such patterns are often very noticeable is in time sliced data. These amplitude distortions can limit the value of the seismic amplitudes in applications such as hydrocarbon reservoir characterization and monitoring.
More specifically, the AGI distortion originates when seismic data is acquired with variable offsets, fold, and/or azimuth distributions. When the offset, fold, and/or azimuth distribution is not uniform, steeply dipping residual coherent noise or residual multiple reflections, or any other seismic interference, will intercept the primary reflections at different offsets and/or azimuths having an effect in the stack response. This effect causes the stack to produce spatial periodic variations in the stacked amplitude, thus resulting in organized spatial amplitude distortions that can mimic the field geometry (Gulunay et al. 1994), incorporated herein by reference. Thus, this acquisition geometry imprint creates a systematic distortion in the seismic data. This phenomenon is observed more clearly as high or low amplitude bands or strips in time slices taken from three dimensional stacked seismic data volumes. As mentioned earlier, this spurious energy masks the true relative amplitudes that are useful for interpretation purposes.
For all these reasons, the attenuation of the AGI effect is vital to provide high fidelity amplitudes after data processing. The impact of this correction is important because it affects the accuracy of the characterization and monitoring of hydrocarbon reservoirs. There is a long felt need for a method of attenuating the acquisition geometry imprint.
SUMMARY OF THE INVENTION
The present invention provides a method and system for attenuation of acquisition geometry imprint in seismic data. The method of one embodiment of the present invention includes receiving seismic data. In one example embodiment, this data is in the form of common midpoint (CMP) gathers with normal moveout correction (NMO) applied. In this example, the CMP gathers contain the data in the time-offset-azimuth domain. The amplitudes of the seismic data are raised to a predetermined power. The reciprocal of the raised amplitude data is computed to produce weights. A time slice across a CMP gather of the weights is selected, and the weights of the time slice are stacked. The original input amplitudes are scaled using the weights and stacked. The stacked scaled weighted amplitudes are divided by the stacked weights to produce the output amplitudes.
In another embodiment of the present invention, the amplitudes of seismic data with the normal moveout correction (NMO) applied are squared. The squared amplitude data is then inverted to produce weights. A time slice of the weights and seismic data is selected, and the amplitudes of the time slice of the inverted data are divided into sets. In each of the sets, the inverted data is multiplied by the raw amplitudes to obtain the scaled data. The scaled data is stacked, and the weights are also stacked. Next, a model amplitude corresponding to each set is developed and centered at the middle of the minimum and maximum offset in each set of amplitudes. The representative offset values for each set are transformed, and a parabola is fitted to the transformed representative offset values for each set. The originally received offsets of the received seismic data are then transformed. Model amplitudes are computed at all the originally received transformed offsets, and the model amplitudes are compared to the real amplitudes and a difference value is calculated. If the difference value exceeds a predetermined threshold, the real amplitudes are scaled to match the model amplitudes, and the scaled amplitudes are used to produce a new parabolic model of scaled amplitudes across the time slice for the second time. The parabolic model of the amplitudes is used to predict the output amplitude at zero offset for the current time slice. Then, the process continues to the next time slice until the maximum recorded time is reached.
REFERENCES:
patent: 3398396 (1968-08-01), Embree
patent: 6151555 (2000-11-01), Van Bemmel et al.
patent: 6343256 (2002-01-01), Winbow et al.
Gulunay, et al., “3D Data Acquisition Artifacts Removal—Spot Editing in the Spatial-Temporal Frequency Domain”, 56thMeeting and Technical Exhibition, Vienna, Austria, Jun. 6-10, 1994.
Hill, et al., “Acquisition Footprint and fold-of-stack plots”, The Leading Edge, Jun. 1999, pp. 686-695.
McElheny Jr. Donald E.
PGS Americas, Inc.
LandOfFree
Method and system for attenuation of acquisition geometry... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for attenuation of acquisition geometry..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for attenuation of acquisition geometry... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2959566