- Home
- Conferences
- Conference Proceedings
- Conferences
77th EAGE Conference and Exhibition - Workshops
- Conference date: June 1-4, 2015
- Location: Madrid, Spain
- Published: 01 June 2015
41 - 60 of 128 results
-
-
Optimizing the Input Model for Waveform Inversion Using Image-domain Wavefield Tomography with Illumination Compensation
Authors E.F. Diaz Pantin and P.C. SavaSummaryImage domain wavefield tomography exploits focusing characteristics of extended images for updating the velocity field. In order to make good use of this information, we must understand how such images behave if the migration velocity is accurate. This is not trivial since focusing depends on the acquisition setup, as well as on illumination variation caused by the geology separating the acquisition array from the imaged structure, the data bandwidth, etc. We address this problem using a combination of migration/demigration to construct penalty functions that characterize focusing by incorporating acquisition parameters and data bandwidth. Moreover, instead of sampling the extended images at preset distance along the surface, we sample the image by constructing common image-point gathers, which are also much more economical from a computation point of view. Coupled with image residuals exploiting illumination-based penalty functions, we obtain robust wavefield tomography in areas of poor or uneven illumination. Models obtained with this type of methodology are good starting points to more sensitive, but less robust waveform inversion methods.
-
-
-
Migration Velocity Analysis Combining Reflected and Direct Waves in a Crosswell Configuration
Authors C.A. Lameloise and H. ChaurisSummaryMigration Velocity Analysis is classically restricted to reflected wave fields. In Crosswell Seismic Profiling (CSP) configuration, transmitted waves are usually considered. Here, in a CSP configuration, we combine direct and reflected waves in order to perform a Migration Velocity Analysis. The idea is to take advantage of the different illuminations of the subsurface provided by these different wave fields. The velocity update shows that the part related to reflected waves mainly emphasizes zones around reflectors, whereas transmitted waves illuminate the subsurface in a more homogeneous way. We discuss in a heterogeneous faulted model an automatic velocity method, based on different waves, separately or combined.
-
-
-
Is Image-warping a Robust Tool for Image Domain Tomography?
Authors F. Perrone and P. SavaSummaryMigration Velocity Analysis in the subsurface-domain measures velocity errors via (extended) imagedomain residuals with respect to an ideal reference image and then updates the velocity model in order to minimise those residuals. Because of the similarity between images with similar extension parameter (shot number, offset, incidence angle, etc.), image-warping presents a robust approach to compute image residuals in different subsurface domains. However, since similarity measures cannot in general distinguish between signal and coherent noise, kinematic artefacts that contaminate the migrated images in certain extended domains can lead to strong and spurious events in the image perturbations and hinder the robustness of the tomographic inversion.
-
-
-
Utilization of Multiples in Simultaneous Image Building and Velocity Estimation
Authors A.J. Berkhout, D.J. Verschuur, X.R. Staal and M. DavydenkoSummaryIn the past the seismic industry has made large investments in the removal of multiples from the seismic response ('data linearization'), with the purpose to make the data suitable for our linear velocity estimation and linear migration algorithms. Now we are beginning to understand that multiples contain most valuable information that should not be removed.
Instead they should be utilized, making velocity estimation and migration fully consistent with the underlying physics. In this presentation it is shown that velocity estimation can function as an integral part of the nonlinear migration process: Joint Migration Inversion (JMI).
It is also shown that JMI is insensitive to the initial velocity distribution.
-
-
-
Affordable Full Subsurface Image Volume - An Application to WEMVA
Authors T. van Leeuwen, R. Kumar and F. J. HerrmannSummaryCommon image gathers are used in building velocity models, inverting for anisotropy parameters, and analyzing reservoir attributes. In this paper, we offer a new perspective on image gathers, where we glean information from the image volume via efficient matrix-vector products. The proposed formulation make the computation of full subsurface image volume feasible.
We illustrate how this matrix-vector product can be used to construct objective functions for automatic MVA.
-
-
-
3D RTM-based Wave Path Tomography Tested at a Realistic Scale
Authors P.M. Bakker, S. Gerritsen and Q. CaoSummaryA new method of 3D RTM-based Wave Path Tomography is presented. Input of the method are RMO picks from wide-azimuth RTM angle gathers. The depth sensitivity of picked events to a velocity perturbation is derived by Born approximation of the associated wave fields and the images. For reasons of efficiency, this is done only for the stationary shots, pertaining to selected picks at certain angle slots in the gather. In the sensitivity kernel calculations, the corresponding recorded data are restricted to those in the vicinity of the estimated stationary receivers. Each branch of a sensitivity kernel, linking stationary shot or receiver to the picked event, is computed by RTM imaging. The method is successfully tested at a realistic scale for a wide-azimuth synthetic example, and for a case with real data from an OBS survey.
-
-
-
Application of Plane-wave Extended Waveform Inversion to an Onshore Seismic Dataset
Authors Y. Liu, W.W. Symes, C.H. Zhu, Y.Q. Chen and M.Q. LuoSummaryExtended waveform inversion (EWI) fills the gap between migration velocity analysis (MVA) and waveform inversion (WI). It’s less prone to local minimum when the starting velocity is inadequate for conventional WI. To improve the efficiency of EWI, we extend the velocity model in the plane-wave source domain. To weaken the absolute amplitude information and make plane-wave encoding applicable in the real data processing, we use normalized wavefields in the data-misfit term of objective function. Numerical tests show that EWI can successfully and efficiently build a high-resolution velocity model for the onshore seismic dataset.
-
-
-
Strength and Limitation of FWI - What Can Tomographic Approaches Bring?
Authors G. Lambare and T. AllemandSummaryFull waveform inversion is now well established for velocity model building in areas investigated by recorded diving waves. At larger depths, significant limitations appear associated with a lack of resolution in the long vertical wavelengths of the velocity. To remedy this, new wave equation based approaches have been proposed. They reveal strong connections with migration velocity analysis and tomographic techniques. In the new context of an overlap between the resolution of velocity analysis and of imaging (created by the progress of migration velocity analysis and of low frequency acquisitions) the challenge is to combine these approaches with full waveform inversion in order to get high-resolution structurally conformable velocity models. We review and analyze the proposed methods.
-
-
-
Joint Full Waveform Inversion of Diving Waves and Reflected Waves for Velocity Model Building
Authors W. Zhou, R. Brossier, S. Operto and J.M. VirieuxSummaryAt depths where there is no sampling by diving waves, FWI behaves as a least-squares migration of the short-spread reflections, hence providing a reconstruction of the short-scale reflectivity at the expense of the long wavelengths of the velocity. Recently, it has been proposed to modify the FWI formalism such that the long wavelengths of the velocity can be updated from reflected waves using some prior knowledge on the reflectivity and an explicit scale separation between the velocity macro-model and the reflectivity. This scale separation allows one to emphasize the forward-scattering regime in the sensitivity kernel of the FWI, referred to as reflection FWI (RFWI). The drawback of the RFWI is to discard the valuable information on the shallow subsurface carried out by diving waves. A new FWI formalism, referred to as joint FWI (JFWI) is proposed and takes advantage of the long-wavelength information carried out by both diving waves and reflected waves to build a smooth velocity model. This formalism leads to a workflow which iteratively cycles the update of the smooth velocity model by JFWI and the update of the short-scale impedance model by classical FWI of short-spread reflections. Application to a synthetic Valhall model illustrates the performance of JFWI.
-
-
-
An Auxiliary Bump Functional to Overcome Cycle Skipping in Waveform Inversion
Authors P. Bharadwaj, W.A. Mulder and G.G. DrijkoningenSummaryTo overcome the local minima problem in FWI, we propose to use an auxiliary data-domain objective function during inversion. It reduces the data to a simpler form by squaring, followed by blurring to ensure that events that are too far apart can still interact during the inversion. As it effectively replaces seismic arrivals by bumps, we call it the bump functional. This objective function is less sensitive to cycle skipping. Its rôle is to guide the inversion towards the global minimum by pulling the trapped solution out of the local minima associated with the least-squares functional whenever necessary.
Waveform inversion cannot be performed with only the auxiliary objective function because it is insensitive to the polarity of the arrivals and the source signature. Therefore, we alternate between minimization with this and the classic least-squares functional. We confirm the validity of the approach using a simple numerical example with reflection data.
-
-
-
Diffraction Imaging - A Tool for Detection Small Scale Subsurface Heterogeneities
By E. LandaSummarySeismic diffraction is a carrier of information coming from subsurface objects of sub-wavelength scale. Isolating diffraction from the full wavefield and imaging them separately is a first step in establishing super-resolution of structural details. Although the importance of diffracted waved has been long time recognized in seismic exploration, in practice they are ignored in standard data processing and imaging. Naturally fractured reservoirs are an important target for the oil and gas industry. Usually information about the fractures comes from coherency cube interpretation or azimuthal analysis of the effective media. It is only seismic diffraction can directly indicate sharp structural and lithological changes in the subsurface.
Separation diffraction from specular reflection in the data domain is the first step of diffraction imaging. It can be efficiently done using differences in kinematic and dynamic properties of reflected and diffracted events. At second step, focusing diffractive component allows us reliably visualize small and medium scale elements of the subsurface such as faults, pinchouts, karsts, fractures etc.
I will demonstrate diffraction imaging on synthetic and real data examples.
-
-
-
Super-resolution by Moveout Correction and Migration of Surface-related Resonant Multiples
Authors B. Guo, Y. Huang and G.T. SchusterSummarySurface-related resonant multiples can be migrated to achieve better resolution than primary reflection data. However, such multiples are only recorded by near-offset traces. As a result, the migration image using resonant multiples suffers from poor signal-to-noise ratio. In this report, I mitigate this problem by aligning the first-order surface-related multiples across different offsets with the zero-offset resonant multiples. Such an alignment is based on the moveout characteristics of the first-order surface-related multiples. Stacking the aligned multiples improves the signal-to-noise ratio of the zero-offset resonant multiples as well as the resulting migration image. Results with both synthetic and field data results validate this method and show migration images with super-resolution characteristics.
-
-
-
Handling the Conflicting Dip Problem in the CRS/i-CRS Methods
Authors J. Walda and D. GajewskiSummaryMost current implementations of the CRS operator suffer from the occurrence of conflicting dip situations in the acquired data. To address this properly we apply the idea of the CDS. We use the i-CRS operator that can be related to the CRS operator, and show, that conflicting dips can be resolved well in multiparameter processing. The results are promising and reveal a lot of potential for further applications. This is shown by a diffraction separation technique applied to field data obtained in the Levantine Basin.
-
-
-
3D CRS-based Prestack Diffraction Separation and Imaging
Authors P. Bakhtiari Rad, C. Vanelle and D. GajewskiSummaryImaging of seismic diffractions is a challenge since it is inherently a 3D problem. Diffractions carry useful information about the subsurface and allow to identify the presence of small-scale heterogeneities and structures e.g. fractures, pinch-outs, thin lenses etc. Thus, diffraction separation and imaging can lead to higher resolution, which is of particular interest for reservoir characterisation and exploration.
In this work, we suggest a 3D workflow based on common-reflection-surface (CRS) method for prestack diffraction separation and imaging in time domain. The workflow combines the ideas developed for diffraction separation with the partial CRS stack technique. It comprises not only the diffraction separation facility but also includes a prestack data enhancement, i.e., an improved SNR in diffraction-only data. Application to a 3D synthetic model confirms its effectiveness in prestack diffraction separation. It also demonstrates potential for time migration velocity analysis using diffraction-only data.
-
-
-
Zero-offset Based Prediction of Common-offset Diffraction Traveltimes
Authors A. Bauer, B. Schwarz and D. GajewskiSummaryThe imaging of diffracted waves is a crucial challenge in seismic processing, because they carry important information about small-scale subsurface structures. A key step of diffraction imaging is their separation and enhancement in the pre-stack data volume, which requires common-offset processing. However, due to the higher dimensionality of the problem, common-offset stacking is computationally more expensive than the stable and commonly used zero-offset processing. In this work, we motivate a straightforward decomposition principle for diffractions, which establishes a direct connection between zero-offset and common-offset diffraction wavefield attributes based on the decoupling of diffraction raypaths. We show, theoretically and on simple waveform data, that each common-offset diffraction operator can be decomposed exactly into two zero-offset operators. This allows the direct prediction of common-offset diffraction attributes solely based on their zero-offset counterparts. Application of the new method to complex data reveals its ability to reliably image diffractions in the common-offset domain using only results from zero-offset processing as input. The promising results in terms of both image and attributes reveal a high potential for improved pre-stack diffraction separation and diffraction-stereotomography.
-
-
-
Imaging Spatio-temporal Changes of the Earth with Coda Waves
Authors L. Margerin, T. Planès, J. Mayor, M. Calvet, E. Larose and V. RossettoSummaryCoda-wave interferometry is a powerful technique which exploits waveform perturbations observed in the coda to monitor changes of the propagation medium. In this work, we use a radiative transfer approach to model two relevant observables which may be employed for monitoring purposes: travel-time shifts and de-correlation of waveforms. These observables are sensitive to weak changes of the background velocity or to the addition of mechanical defects, respectively. We develop specific sensitivity functions for each type of observable and evaluate them numerically using analytical solutions of the radiative transfer equation. Our theory can model arbitrarily anisotropic wave fields and is not limited by a diffusion approximation. We show that the coherent wave plays a crucial role in the sensitivity of coda waves in the weak scattering regime. In particular, the coherent wave is responsible for the algebraic divergence of both the travel time and de-correlation sensitivity kernels at the source and receiver. The de-correlation kernel shows an additional zone of high sensitivity in the vicinity of the single-scattering ellipse. These sensitivity functions may be employed to develop a linearized tomographic approach to the monitoring of medium changes in the weak scattering regime.
-
-
-
Imaging in Random Media
By L. BorceaSummaryI will present an introduction to the mathematical theory of wave propagation in heterogeneous media with uncertain micro-scale, modelled as random media. I will summarize the scattering effects in such media and how to mitigate them in order to obtain high fidelity imaging results.
-
-
-
Focusing Prestack Depth Imaging Approaches
Authors F. Hlousek, O. Hellwig and S. BuskeSummaryThree focusing imaging approaches are presented which are based on Kirchhoff prestack depth migration. All approaches can be formulated as a weighted diffraction stack with different weighting functions. In any case the smearing along the two-wave-traveltime isochrone is limited to the physically relevant part at the specular reflection or diffraction point. The focusing characteristics of these approaches can be very valuable for increasing the image quality, both for hydrocarbon exploration data sets as well as for imaging in complex crystalline crust.
-
-
-
Can We Image Everything with Primaries?
Authors D.J. Verschuur, A.J. Berkhout, M. Davydenko and S.R. StaalSummaryFor true amplitude imaging with reduced artifacts a closed-loop approach for seismic migration needs to be employed (‘least-squares migration’). However, today’s least-squares migration algorithms do not take into account multiple reflections and transmission effects. The use of multiples will enlarge the illumination of the subsurface and can be of crucial importance when primaries are not properly measured due to acquisition constraints and background noise. In this paper it is shown that the full wavefield migration (FWM) process can image subsurface structures via internal multiples. This is successfully demonstrated for imaging near-surface structures for land data when their primaries are not properly measured.
-