Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Berkhout, A.J.
Summary The CFP-approach to seismic imaging is very suitable for the processing of converted waves. Using multicomponent receiver data, it offers the possibility for estimation of Pwave and S-wave one-way propagation operators and corresponding target images. Throughout this procedure, the reflection and conversion points can be kept at the identical positions. Therefore, many problems in traditional converted wave processing are avoided.
- North America > United States > Utah (0.17)
- Europe > Netherlands (0.15)
ABSTRACT Focal beams are the ideal tools for analyzing the acquisition footprint because the source and detector geometry are treated separately. A perfect illumination with both the source and detector geometry ensures the absence of an acquisition footprint. Practically, it is not affordable to have a perfect illumination with both the source and detector geometry. Usually, the acquisition footprint is the strongest in the cross-line direction. In general the detector geometry will be mainly responsible for the acquisition footprint in that direction. Therefore optimization of the detector geometry and the cross-line roll-along distance is the obvious choice to reduce the acquisition footprint. An example illustrates this process.
Summary The surface multiple removal method has been successfully applied to several synthetic and field datasets. The extension to internal multiples can be made by including data-driven propagation operators towards the internal multiple generating boundary, such that the use of explicit velocity-depth model information is avoided. By CFP redatuming to a suitable level between reflectors, the upward reflectivity can be taken into account such that all internal multiples generated within a complete layer are addressed in one step, without using a velocity model. Introduction During the last decade, the interest in multiple elimination techniques has grown significantly. One of the reasons is that still today the problem of multiples in seismic (marine) data is not solved completely and can be one of the major hurdles to be taken in today’s 3D processing flows.
- Europe > Netherlands (0.30)
- North America > United States > Texas (0.17)
- Geophysics > Seismic Surveying > Seismic Processing (1.00)
- Geophysics > Seismic Surveying > Seismic Modeling > Velocity Modeling (0.35)
Summary Compared with velocity-driven time and depth migration, operator-driven CFP-migration can be considered as the most general approach to seismic imaging: it does not require a velocity model, and it automatically takes into account unknown complex propagation effects such as conversion, anisotropy and dispersion. Introduction For each subsurface gridpoint seismic migration can be written in terms of two focusing steps, i.e. focusing in detection followed by focusing in emission. In the first focusing step (focusing in detection) each shot record is transformed into a single trace by the focusing operator for the gridpoint under consideration (Fresnel-zone stacking). Together those traces define the so-called common Focus-point gather (CFP-gather), each trace being positioned at the source location of its corresponding shot record. One event in the CFP-gather is the focus-point response.
The AVP-function represents the distortion of angle dependent The acquisition geometry of a 3-D seismic survey should be reflection information by the acquisition geometry. For a designed in such a way that it fulfils the imaging requirements stationary spread the ray-parameter-frequency domain product while satisfying the economical constraints. Focal source and of the source and receiver beam gives the AVP-function. The focal detector beams are defined which describe the focusing focal beams and focal functions contain information on the properties of the source and detector geometry. These beams following aspects that are required for high quality seismic can be computed and evaluated separately. The computation of pre-stack processing: these beams can be done very efficiently using a Fourier - resolution, transform of the source and receiver sampling function. In the - amplitude accuracy, case of a stationary spread the potential resolution and - quality of AVP-information, accuracy of AVP information can be obtained from these - amount of noise suppression, beams.
- Geophysics > Seismic Surveying > Surface Seismic Acquisition (1.00)
- Geophysics > Seismic Surveying > Seismic Processing > Seismic Migration (0.69)
Summary Using the CFP technology, any near surface complexities can be described in terms of propagation operators which can be easily updated. With these fully data driven updated operators, a one-way time image can be constructed and a complete wavefield redatuming can be applied to the seismic data by simply defining one of the sub-weathering reflectors as the new acquisition datum. Introduction Conventional redatuming methods of seismic data involve applications of static time shifts to the input traces. These time shifts, commonly known as “statics”, can be applied pre-stack and/or post-stack and their aim is to compensate for the effects of the near surface.
- North America > United States (0.48)
- Asia > Middle East > Saudi Arabia (0.30)
- Geophysics > Seismic Surveying > Seismic Processing > Seismic Migration (1.00)
- Geophysics > Seismic Surveying > Seismic Modeling > Velocity Modeling (0.97)
Summary The surface multiple removal method has been successfully applied to several synthetic and field datasets. The main characteristic of this wave theory based method is that the seismic data itself are used as the multi-dimensional multiple prediction operator. This avoids the necessity to bring in any subsurface information. It requires that sources and receivers be at the multiple generating interface, which is easily fulfilled for regular surface data. The extension to internal multiples can be made by including data-driven propagation operators towards the internal multiple generating boundary, such that the use of explicit velocity-depth model information is avoided.
Summary In the CFP-approach to seismic imaging the migration process is formulated in terms of two separate steps: focusing in detection and focusing in emission. The latter means amongst others that in CFP-migration the upgoing and downgoing wave fields are separately addressed. This property of CFP-migration is most favorable in the situation of mode conversion: the operator for focusing in detection may address another wave type than the operator for focusing in emission, without making the migration algorithm more complex. In addition, operator updating may replace velocity updating and angle-averaged reflection output may be extended to angle-dependent reflection output.
Summary The CFP-approach to seismic processing yields new solutions in the situation of complex geology. Introduction Prestack depth migration with the integral method (such as Kirchhoff-summation) involves one double sum for each subsurface gridpoint, the summation representing weighted superposition over the detector and source positions. In CFP-migraton the double sum is split into two single sums, one summation being carried out over the detector positions (focusing in detection) and one summation being carried out over the source positions (focusing emission).
Abstract The surface-related multiple removal method has been successfully applied to several synthetic and field datasets during the last decade. The main characteristic of this wave theory based method is that the seismic data itself are used as the multi-dimensional multiple prediction operator. This avoids the necessity to bring in any subsurface information. It requires that sources and receivers be at the multiple generating interface, which is easily fulfilled for regular streamer data. For OBC data some modifications are required. With this method all possible surface-related multiples can be attenuated. This makes it attractive when several strong subsurface reflectors are present (e.g. water bottom and top of salt) that can create surface multiples which obscure the deeper (e.g. the sub-salt) structures. Furthermore, the extension to internal multiples can be made by including data driven propagation operators towards the internal multiple generating boundary. Introduction During the last decade, the interest in multiple elimination techniques has grown significantly. One of the reasons is that still today the problem of multiples in seismic (marine) data is not solved completely and can be one of the major hurdles to be taken in today's 3D processing flows. Currently, there is not something like the process to solve all multiple problems. The seismic processor needs a toolbox of methods from which the best one (or the best combination) can be selected to solve a specific problem. In the past multiple attenuation was primarily achieved by making use of two different approaches: statistical predictionerror filtering and deterministic move-out filtering. Statistical prediction-error algorithms make use of the property that a multiple tail may be well predicted from its earlier arriving primary reflection via the auto-correlation functions (Robinson, 1957). For 1-D media prediction-error filtering is very successful, particularly in the linear Radon domain (Taner, 1980). However, the more complex the medium response the less effective prediction-error filtering. In addition, lengthy prediction-error filters may distort the primary reflection events. Later the deconvolution method has been extended to a multi-dimensional one by (Taner et al., 1993), which increased the ability to follow more complex structures. Algorithms in the second approach, deterministic move-out filtering, make use of the property that primary reflections may have a significantly smaller move-out than the interfering multiple reflections. For media with large positive velocity gradients move-out filtering may be very successful, particularly in the parabolic Radon domain (Hampson, 1986) and the hyperbolic Radon domain (Foster et al.., 1992). However, in the situation of small velocity gradients, or even velocity reversals, the difference in move out between primaries and multiples may become very small, particularly at small offsets, and move-out filtering may seriously attenuate the primary reflections as well. In addition, in the situation of complex medium responses application of the parabolic or hyperbolic Radon transform may not be meaningful any more. In the late seventies, it was felt that particular attention should be paid to the relative strong surface-related multiples. Kennett (1979) described a 1-D forward model for surfacerelated multiples and proposed a 1-D inversion scheme.