scholarly journals Removal of internal multiples with the common-focus-point (CFP) approach: Part 2 — Application strategies and data examples

Geophysics ◽  
2005 ◽  
Vol 70 (3) ◽  
pp. V61-V72 ◽  
Author(s):  
D. J. Verschuur ◽  
A. J. Berkhout
Keyword(s):  
Physical Model ◽  
Field Data ◽  
Velocity Model ◽  
Data Sets ◽  
Removal Method ◽  
The Past ◽  
Feedback Model ◽  
The Common ◽  
Internal Multiples ◽  
Focus Point

In the past, the surface-multiple-removal method based on the feedback model has been successfully applied to many different field data sets. The extension of surface to internal multiples can be made by replacing shot records with common-focus-point (CFP) gathers, a CFP gather representing focused data with one source in the subsurface and all receivers at the surface (or vice versa for a receiver gather). The internal-multiple-removal algorithm can be formulated in terms of boundary-related and layer-related versions. In the boundary-related version, the internal multiples are removed for one downward-scattering reflector at a time. In the layer-related version, the internal multiples are removed for a sequence of downward-scattering reflectors at a time. An exact velocity model is not required, but proper muting is critical; muting becomes straightforward in the CFP domain. The strategy for applying the two versions of the multiple-removal algorithm is demonstrated on physical-model and field data. One can conclude that the layer-related version is the most appropriate in most situations because it requires less user action and does not need exact knowledge of the multiple-generating boundary.

scholarly journals Improving the resolution of migrated images by approximating the inverse Hessian using deep learning

Geophysics ◽  
2020 ◽  
Vol 85 (4) ◽  
pp. WA173-WA183 ◽  
Author(s):  
Harpreet Kaur ◽  
Nam Pham ◽  
Sergey Fomel
Keyword(s):  
Field Data ◽  
Velocity Model ◽  
Image Resolution ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
Data Sets ◽  
Computationally Efficient ◽  
Validation Data ◽  
Network Output

We have estimated migrated images with meaningful amplitudes matching least-squares migrated images by approximating the inverse Hessian using generative adversarial networks (GANs) in a conditional setting. We use the CycleGAN framework and extend it to the conditional CycleGAN such that the mapping from the migrated image to the true reflectivity is subjected to a velocity attribute condition. This algorithm is applied after migration and is computationally efficient. It produces results comparable to iterative inversion but at a significantly reduced cost. In numerical experiments with synthetic and field data sets, the adopted method improves image resolution, attenuates noise, reduces migration artifacts, and enhances reflection amplitudes. We train the network with three different data sets and test on three other data sets, which are not a part of training. Tests on validation data sets verify the effectiveness of the approach. In addition, the field-data example also highlights the effect of the bandwidth of the training data and the quality of the velocity model on the quality of the deep neural network output.

Including and using internal multiples in closed-loop imaging — Field data examples

Geophysics ◽  
2018 ◽  
Vol 83 (4) ◽  
pp. R297-R305 ◽  
Author(s):  
Mikhail Davydenko ◽  
D. J. Verschuur
Keyword(s):  
North Sea ◽  
Field Data ◽  
Closed Loop ◽  
Velocity Model ◽  
Measured Data ◽  
Data Interpolation ◽  
Migration Process ◽  
Multiple Reflections ◽  
Additional Illumination ◽  
Internal Multiples

Nowadays, it is more widely accepted that multiple reflections should not be considered as noise, but as signal that can provide additional illumination of the subsurface. However, one of the challenges in seismic imaging is including all multiples in the migration process for field data in a reliable manner. Although including surface multiples in imaging has been demonstrated already on field data in recent years, the proper imaging of internal multiples is less established. We have determined successful field data applications on imaging that takes all internal multiples into account. This is done via so-called full-wavefield migration (FWM), an inversion-based method in which, given the migration velocity model, the angle-dependent reflectivity is iteratively estimated by minimizing the misfit between the modeled and the measured data. Its forward model is based on a multidimensional version of the so-called Bremmer series, which allows modeling of transmission effects and any type of multiple scattering in the subsurface and, thereby, is able to minimize the data misfit correctly. An application of FWM on deepwater field data from the Norwegian North Sea validates its capabilities to explain and image internal multiples. Furthermore, it is demonstrated on the same field data that the FWM framework can also be used for data interpolation and primary/multiple separation.

scholarly journals Removal of internal multiples with the common-focus-point (CFP) approach: Part 1 — Explanation of the theory

Geophysics ◽  
2005 ◽  
Vol 70 (3) ◽  
pp. V45-V60 ◽  
Author(s):  
A. J. Berkhout ◽  
D. J. Verschuur
Keyword(s):  
Wave Theory ◽  
Data Driven ◽  
Numerical Examples ◽  
Model Driven ◽  
The Common ◽  
Internal Multiples ◽  
Unified Description ◽  
Focus Point

Removal of surface and internal multiples can be formulated by removing the influence of downward-scattering boundaries and downward-scattering layers. The involved algorithms can be applied in a model-driven or a data-driven way. A unified description is proposed that relates both types of algorithms based on wave theory. The algorithm for the removal of surface multiples shows that muted shot records play the role of multichannel prediction filters. The algorithm for the removal of internal multiples shows that muted CFP gathers play the role of multichannel prediction filters. The internal multiple removal algorithm is illustrated with numerical examples. The conclusion is that the layer-related version of the algorithm has significant practical advantages.

scholarly journals Mitigating elastic effects in marine 3-D full-waveform inversion

2019 ◽  
Vol 220 (3) ◽  
pp. 2089-2104
Author(s):  
Òscar Calderón Agudo ◽  
Nuno Vieira da Silva ◽  
George Stronge ◽  
Michael Warner
Keyword(s):  
Wave Equation ◽  
Field Data ◽  
Waveform Inversion ◽  
Velocity Model ◽  
P Wave ◽  
Data Sets ◽  
Acoustic Wave Equation ◽  
Data Set ◽  
Velocity Models ◽  
Full Waveform

SUMMARY The potential of full-waveform inversion (FWI) to recover high-resolution velocity models of the subsurface has been demonstrated in the last decades with its application to field data. But in certain geological scenarios, conventional FWI using the acoustic wave equation fails in recovering accurate models due to the presence of strong elastic effects, as the acoustic wave equation only accounts for compressional waves. This becomes more critical when dealing with land data sets, in which elastic effects are generated at the source and recorded directly by the receivers. In marine settings, in which sources and receivers are typically within the water layer, elastic effects are weaker but can be observed most easily as double mode conversions and through their effect on P-wave amplitudes. Ignoring these elastic effects can have a detrimental impact on the accuracy of the recovered velocity models, even in marine data sets. Ideally, the elastic wave equation should be used to model wave propagation, and FWI should aim to recover anisotropic models of velocity for P waves (vp) and S waves (vs). However, routine three-dimensional elastic FWI is still commercially impractical due to the elevated computational cost of modelling elastic wave propagation in regions with low S-wave velocity near the seabed. Moreover, elastic FWI using local optimization methods suffers from cross-talk between different inverted parameters. This generally leads to incorrect estimation of subsurface models, requiring an estimate of vp/vs that is rarely known beforehand. Here we illustrate how neglecting elasticity during FWI for a marine field data set that contains especially strong elastic heterogeneities can lead to an incorrect estimation of the P-wave velocity model. We then demonstrate a practical approach to mitigate elastic effects in 3-D yielding improved estimates, consisting of using a global inversion algorithm to estimate a model of vp/vs, employing matching filters to remove elastic effects from the field data, and performing acoustic FWI of the resulting data set. The quality of the recovered models is assessed by exploring the continuity of the events in the migrated sections and the fit of the latter with the recovered velocity model.

Estimation of multiple scattering by iterative inversion, Part I: Theoretical considerations

Geophysics ◽  
1997 ◽  
Vol 62 (5) ◽  
pp. 1586-1595 ◽  
Author(s):  
A. J. Berkhout ◽  
D. J. Verschuur
Keyword(s):  
Numerical Experiments ◽  
Velocity Model ◽  
Neumann Series ◽  
Related Method ◽  
Feedback Model ◽  
Iterative Inversion ◽  
Internal Multiples ◽  
Surface Operator ◽  
Theoretical Considerations ◽  
Upper Boundary

A review has been given of the surface‐related multiple problem by making use of the so‐called feedback model. From the resulting equations it has been concluded that the proposed solution does not require any properties of the subsurface. However, source‐detector and reflectivity properties of the surface need be specified. Those properties have been quantified in a surface operator and this operator is estimated as part of the multiple removal problem. The surface‐related multiple removal algorithm has been formulated in terms of a Neumann series and in terms of an iterative equation. The Neumann formulation requires a nonlinear optimization process for the surface operator; while the iterative formulation needs a number of linear optimizations. The iterative formulation also has the advantage that it can be integrated easily with another multiple removal method. An algorithm for the removal of internal multiples has been proposed as well. This algorithm is an extension of the surface‐related method. Removal of internal multiples requires knowledge of the macro velocity model between the surface and the upper boundary of the multiple generating layer. In part II (also published in this issue) the success of the proposed algorithms has been demonstrated on numerical experiments and field data examples.

Seismic imaging beyond depth migration

Geophysics ◽  
2001 ◽  
Vol 66 (6) ◽  
pp. 1895-1912 ◽  
Author(s):  
A. J. Berkhout ◽  
D. J. Verschuur
Keyword(s):  
Seismic Imaging ◽  
Time Lapse ◽  
Velocity Model ◽  
Surface Layers ◽  
Near Surface ◽  
Depth Migration ◽  
New Type ◽  
Complex Propagation ◽  
The Common ◽  
Focus Point

If seismic imaging is formulated in terms of two focusing steps—focusing in emission and focusing in detection (or vice versa)—the output of the first focusing step yields a new type of seismic gather, the common‐focus‐point (CFP) gather, which is available for data analysis and information extraction. One important consequence of this novel option is that the involved focusing operators can be updated without updating the underlying velocity model. Introducing the concept of “dynamic focusing,” it is proposed to verify the validity of focusing operators by comparing the “gather of focus‐point responses” with the “gather of focusing operators.” 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. In addition, in CFP migration, the second focusing step can be extended to produce both angle‐averaged reflection information and angle‐dependent reflection information. The CFP approach to seismic migration allows new solutions in the situation of complex near‐surface layers, subsalt targets, multicomponent processing, and time lapse analysis.

scholarly journals Identification of image artifacts from internal multiples

Geophysics ◽  
2007 ◽  
Vol 72 (2) ◽  
pp. S123-S132 ◽  
Author(s):  
Alison E. Malcolm ◽  
Maarten V. de Hoop ◽  
Henri Calandra
Keyword(s):  
Synthetic Data ◽  
Velocity Model ◽  
Single Scattering ◽  
Image Artifacts ◽  
Coherent Noise ◽  
First Order ◽  
Inverse Scattering Series ◽  
The Common ◽  
Internal Multiples ◽  
Seismic Images

First-order internal multiples are a source of coherent noise in seismic images because they do not satisfy the single-scattering assumption fundamental to most seismic processing. There are a number of techniques to estimate internal multiples in data; in many cases, these algorithms leave some residual multiple energy in the data. This energy produces artifacts in the image, and the location of these artifacts is unknown because the multiples were estimated in the data before the image was formed. To avoid this problem, we propose a method by which the artifacts caused by internal multiples are estimated directly in the image. We use ideas from the generalized Bremmer series and the Lippmann-Schwinger scattering series to create a forward-scattering series to model multiples and an inverse-scattering series to describethe impact these multiples have on the common-image gather and the image. We present an algorithm that implements the third term of this series, responsible for the formation of first-order in-ternal multiples. The algorithm works as part of a wave-equation migration; the multiple estimation is made at each depth using a technique related to one used to estimate surface-related multi-ples. This method requires knowledge of the velocity model to the depth of the shallowest reflector involved in the generation of the multiple of interest. This information allows us to estimate internal multiples without assumptions inherent to other methods. In particular, we account for the formation of caustics. Results of the techniques on synthetic data illustrate the kinematic accuracy of predicted multiples, and results on field data illustrate the potential of estimating artifacts caused by internal multiples in the image rather than in the data.

Localizing microseismic events on field data using a U-Net based convolutional neural network trained on synthetic data

Geophysics ◽  
2021 ◽  
pp. 1-47
Author(s):  
N. A. Vinard ◽  
G. G. Drijkoningen ◽  
D. J. Verschuur
Keyword(s):  
Hydraulic Fracturing ◽  
Field Data ◽  
Synthetic Data ◽  
Region Of Interest ◽  
Velocity Model ◽  
Mitigation Measures ◽  
Large Field ◽  
Data Sets ◽  
Data Set ◽  
Full Waveforms

Hydraulic fracturing plays an important role when it comes to the extraction of resources in unconventional reservoirs. The microseismic activity arising during hydraulic fracturing operations needs to be monitored to both improve productivity and to make decisions about mitigation measures. Recently, deep learning methods have been investigated to localize earthquakes given field-data waveforms as input. For optimal results, these methods require large field data sets that cover the entire region of interest. In practice, such data sets are often scarce. To overcome this shortcoming, we propose initially to use a (large) synthetic data set with full waveforms to train a U-Net that reconstructs the source location as a 3D Gaussian distribution. As field data set for our study we use data recorded during hydraulic fracturing operations in Texas. Synthetic waveforms were modelled using a velocity model from the site that was also used for a conventional diffraction-stacking (DS) approach. To increase the U-Nets’ ability to localize seismic events, we augmented the synthetic data with different techniques, including the addition of field noise. We select the best performing U-Net using 22 events that have previously been identified to be confidently localized by DS and apply that U-Net to all 1245 events. We compare our predicted locations to DS and the DS locations refined by a relative location (DSRL) method. The U-Net based locations are better constrained in depth compared to DS and the mean hypocenter difference with respect to DSRL locations is 163 meters. This shows potential for the use of synthetic data to complement or replace field data for training. Furthermore, after training, the method returns the source locations in near real-time given the full waveforms, alleviating the need to pick arrival times.

scholarly journals Deep Semisupervised Zero-Shot Learning with Maximum Mean Discrepancy

2018 ◽  
Vol 30 (5) ◽  
pp. 1426-1447 ◽  
Author(s):  
Lingling Zhang ◽  
Jun Liu ◽  
Minnan Luo ◽  
Xiaojun Chang ◽  
Qinghua Zheng
Keyword(s):  
Semantic Space ◽  
Experimental Results ◽  
Data Sets ◽  
Maximum Mean Discrepancy ◽  
The Past ◽  
Benchmark Data ◽  
Training Stage ◽  
Category Labels ◽  
The Common

Due to the difficulty of collecting labeled images for hundreds of thousands of visual categories, zero-shot learning, where unseen categories do not have any labeled images in training stage, has attracted more attention. In the past, many studies focused on transferring knowledge from seen to unseen categories by projecting all category labels into a semantic space. However, the label embeddings could not adequately express the semantics of categories. Furthermore, the common semantics of seen and unseen instances cannot be captured accurately because the distribution of these instances may be quite different. For these issues, we propose a novel deep semisupervised method by jointly considering the heterogeneity gap between different modalities and the correlation among unimodal instances. This method replaces the original labels with the corresponding textual descriptions to better capture the category semantics. This method also overcomes the problem of distribution difference by minimizing the maximum mean discrepancy between seen and unseen instance distributions. Extensive experimental results on two benchmark data sets, CU200-Birds and Oxford Flowers-102, indicate that our method achieves significant improvements over previous methods.

scholarly journals Adaptive overburden elimination with the multidimensional Marchenko equation

Geophysics ◽  
2016 ◽  
Vol 81 (5) ◽  
pp. T265-T284 ◽  
Author(s):  
Joost van der Neut ◽  
Kees Wapenaar
Keyword(s):  
Field Data ◽  
Synthetic Data ◽  
Velocity Model ◽  
Thin Layers ◽  
Multiple Reflections ◽  
Data Set ◽  
Marchenko Equation ◽  
Recorded Data ◽  
Internal Multiples ◽  
Multiple Elimination

Iterative substitution of the multidimensional Marchenko equation has been introduced recently to integrate internal multiple reflections in the seismic imaging process. In so-called Marchenko imaging, a macro velocity model of the subsurface is required to meet this objective. The model is used to back-propagate the data during the first iteration and to truncate integrals in time during all successive iterations. In case of an erroneous model, the image will be blurred (akin to conventional imaging) and artifacts may arise from inaccurate integral truncations. However, the scheme is still successful in removing artifacts from internal multiple reflections. Inspired by these observations, we rewrote the Marchenko equation, such that it can be applied early in a processing flow, without the need of a macro velocity model. Instead, we have required an estimate of the two-way traveltime surface of a selected horizon in the subsurface. We have introduced an approximation, such that adaptive subtraction can be applied. As a solution, we obtained a new data set, in which all interactions (primaries and multiples) with the part of the medium above the picked horizon had been eliminated. Unlike various other internal multiple elimination algorithms, the method can be applied at any specified target horizon, without having to resolve for internal multiples from shallower horizons. We successfully applied the method on synthetic data, where limitations were reported due to thin layers, diffraction-like discontinuities, and a finite acquisition aperture. A field data test was also performed, in which the kinematics of the predicted updates were demonstrated to match with internal multiples in the recorded data, but it appeared difficult to subtract them.

Sign in / Sign up

Export Citation Format

Share Document