- Home
- Conferences
- Conference Proceedings
- Conferences
76th EAGE Conference and Exhibition 2014
- Conference date: June 16-19, 2014
- Location: Amsterdam, Netherlands
- Published: 16 June 2014
21 - 40 of 1028 results
-
-
Reconnaissance Processing to Understand Potential Value of Converted Waves - Experiences from Culzean
Authors A.S. Calvert, A. Merry, F.G. Bourgeois, E.S. Sturup-Toft, P. Salek, A. Rollet and S. PerrierSummaryThe focus of many ocean bottom seismic surveys is improved P-wave imaging using the hydrophone and vertical components. The horizontal components also record potentially valuable converted wave (C-wave) arrivals. The sensitivity of C-waves can provide insights into the rock properties and stress environment of not only the reservoir but also the overburden. C-wave processing projects often take significant time and sometimes produce disappointing results. We share our experiences conducting rapid reconnaissance processing to assess the potential value and challenges of conducting C-wave processing to constrain the geomechanical model of the HPHT Culzean Discovery. The geologic environment is conducive to obtaining high quality C-wave results although is more challenging beneath the top chalk. The full azimuth sampling available from the high density ocean bottom cable survey facilitates velocity analysis, assessment of shear wave splitting and illumination near a salt dome adjacent to the field. The shear wave splitting fast azimuth variations found in the shallowest layers are similar to the principle horizontal stress directions predicted from geomechanical modelling. Such evaluations can be conducted quickly and in parallel to fast track P-wave processing to allow early and more informed decisions about including the horizontal component data in the full production processing flow.
-
-
-
Converted-wave Processing of an Ocean Bottom Cable Dataset from the Arabian Gulf
Authors M.J. Grimshaw, J.P. Holden, E. Murray, K. Shaukat, B. Beck and F. Al-EnaziSummaryThe processing of PS waves from a 4-component converted-wave dataset can provide complementary information to that obtained from traditional P-wave imaging. However, processing the PS component requires additional care to ensure the full potential of the data is realised. This paper discusses the specific processing required for a shallow-water converted-wave dataset acquired with an Ocean Bottom Cable (OBC) system in the Arabian Gulf.
Vector fidelity, receiver-side shear statics, and compensation for shear-wave splitting in the overburden are key topics that had to be addressed in the pre-processing. After further corrections for deeper shear-wave splitting, the PS data were regularised with differing schemes: for analysis of gamma-effective and pre-stack time migration velocity analysis; and for the migration of the data itself.
To maintain data integrity it was important to preserve the signal on both the radial and transverse components through all the processing steps up to and including migration. This ensured a final shear-wave splitting analysis could be made post migration at the target level to minimise any residual shear-wave splitting effects prior to final interpretation and in preparation for a joint PP-PS inversion.
-
-
-
Shear-wave Splitting and Azimuthal Anisotropy in an Offshore Field in the Arabian Gulf
Authors J.P. Holden, B. Beck, M.J. Grimshaw, C. Giersz and F. Al-EnaziSummaryThe effect of shear-wave splitting, an expression of an azimuthal anisotropic medium, has been observed in shear-wave seismic reflection recordings since at least the 1980’s. Since those early experiments it has been recognised that the existence of this phenomenon will degrade shear-wave images unless the issue is addressed adequately within data processing.
A 4-component converted-wave dataset acquired in the Arabian Gulf exhibits such shear-wave splitting. Significant azimuthal anisotropy within the overburden, and locally in excess of 10%, results in a severe loss of coherency on the radial component. Correcting for the shear-wave splitting in this and subsequent intervals dramatically improves both the vertical and spatial resolution, and therefore the interpretability, of the data.
The resulting attributes from the analyses also yield information on the evolution of the subsurface. Of particular note is the overprinting of the azimuthal anisotropy in the post-Eocene overburden by the pre-Turonian (Cretaceous) faulted anticline. This is consistent with an interpretation of late uplift of this deep structure and the consequent extension of the overburden in a direction orthogonal to the structural axis.
-
-
-
Elastic Impedance Minus Shear Impedance - A Powerful Pre-stack Inversion Attribute from OBC Data, Campos Basin, Brazil
Authors L.M.R. Martins and T. DavisSummaryPre-stack seismic inversion is a useful technique to predict the petrophysical properties of the reservoir. Using a four component (4-C) ocean bottom-cable (OBC) seismic survey we set out to improve the reservoir characterization in a deep-water turbidite field in the Campos Basin.
Elastic Impedance (EI) gives us the opportunity to calibrate and invert nonzero-offset seismic data, whilst acoustic impedance is useful for zero-offset data. On the other hand, shear impedance (SI) was derived from converted PS data in a pre-stack inversion process. EI in combination with SI inverted data produce a better characterization tool for fluid saturated reservoirs.
In this paper we combine EI and SI results from pre-stack inversion to better characterize the deep water turbidite reservoir.
-
-
-
Dan Field Ocean Bottom Node (OBN) Survey - A Shallow Water Case History
Authors J. Zaske, P. Hickman, H. Roende, S. Mukund, S. Halliday and S. PerrierSummaryIn 2012 Maersk Oil acquired an Ocean Bottom Node (OBN) seismic survey over the Dan Field located in shallow water in the Danish North Sea. The survey took place shortly after a regional multi-field 4D monitor streamer survey which included the area around the Dan Field. The primary objective of the OBN survey was to acquire data in an area affected by surface infrastructure not accessible with streamer technology. Secondary objectives included evaluating the imaging results in comparison with streamer technology on a field wide scale and evaluating the matching quality between the vintage streamer and OBN survey data.
Significant challenges due to the shallow water environment were overcome during operations as well as processing. The results show a clear improvement with the OBN data in the area affected by the platforms when compared to the streamer data in-filled with conventional two boat undershooting. The OBN image is of a similar or better quality over most of the field when compared to the streamer result allowing for different strategic options for future seismic surveys in this area.
-
-
-
Moveout Approximation for Converted Waves in Layered Orthorhombic Medium
More LessSummaryIn this work we study the hyperbolic moveout approximations for shear and converted waves in a layered orthorhombic medium with flat interfaces in the horizontal plane of symmetry, distinguishing two types of conversion - PS1 and PS2, with different shear polarizations. Note that the azimuthally dependent NMO velocity function for shear or converted waves is similar to that of compression waves - only the coefficients are different. Given a package of orthorhombic layers with different parameters, one can establish an equivalent effective model consisting of a single layer with the same vertical time as the original package, described by the fast and slow effective NMO velocities, the effective azimuth of the slow velocity, and the vertical compression velocity. The latter is normally obtained from non-seismic information such as check-shot or well logs. Note that the effective azimuths are different for models describing PP, PS1 and PS2 waves.
-
-
-
General Definition of Reflection-point Coverage for P- and PS-wave COV Data
By J.E. GaiserSummaryCommon-offset vector (COV) data have served industry well for survey design, generating well-sampled datasets for noise suppression, offset and azimuth analyses and inversion, and image improvements. For P-waves acquired on a flat recording surface, the basic cross-spread theory assumes that reflection-points (RPs) occur below common mid-points (CMPs). This assumption is ray-based, for 3D propagation in isotropic laterally homogeneous media. However, converted P- to S-waves (PS-waves) and other geometries violate this theory because of their asymmetric ray paths. In this paper I apply a general ray-based theory to define COV illumination area for both P- and PS-waves of various geometries. This involves a first-order approximation in ray-parameter, p, of reflection-point (RP) position as a function of t0V2 terms: vertical traveltime and NMO velocity. This definition facilitates corrections to improve RP consistency within a COV gather or volume. I describe corrections applied to offset-vector tile (OVT) distortion for conventional land orthogonal cross-spread data, and for marine geometries of ocean-bottom cable and node (OBC and OBN) geometries where source and receiver have a different datum. Also, I show that this theory can be applied to illumination distortion from structure (plane dipping layers) and discuss corrections for COV data in the context of migration.
-
-
-
Model Based Registration of PP and PS Waves in the Depth Domain
More LessSummaryWe introduce an approach to perform a model based PP and PS registration in depth domain which involves converting the RMS time model to the depth model and performing PSDM. We have developed a GUI tool for depth model building and a PSDM tool which is based on the ray-tracing method and Kirchhoff summation. This approach is applied to a real 2D dataset. Time processing is applied to the PP and PS waves to obtain time images and RMS velocity models separately. Then the depth models are obtained for the PP and PS waves. For PS waves, we build the depth model as close to the P wave depth model as possible. Finally, we apply the prestack depth migration to produce migrated depth images. The results show that the events in PP and PS depth images are well correlated. The model building in this approach is straightforward and this approach is practical.
-
-
-
S-ray Approximation Tomography for PP and PS Horizon Co-Depthing
Authors L.A.D. D’Afonseca, M. Dumett, O.J. Birkeland, P. Guillaume, T. Krishnasamy, T.M.G. Santiago and C. GuerraSummaryDepth imaging of multicomponent data involves additional challenges of event registration and co-depthing. Marine multicomponent data usually comes from ocean bottom cable (OBC) or ocean bottom node (OBN) acquisitions. Therefore, two data sets need to be processed: P-waves (PP) and converted waves (PS). A P-velocity model can be built with techniques similar to the conventional streamer data processing, but the S-velocity demands special treatments. We propose a method for estimating P- and S-velocities by simultaneously inverting PP and PS seismic reflection residual moveout data and matching PP and PS horizons to well markers in a TTI anisotropic medium. We present a work flow to update the Vs component of a TTI velocity model using a tomographic co-depthing of PP and PS horizons based on an S-ray approximation for PS-horizon map migration. This method is integrated to tomography and allows fitting PP and PS residual moveouts simultaneously to horizon matching. Field data shows that this strategy yields more realistic velocity models than those obtained by simple 1D velocity model updates, providing depth images of superior quality.
-
-
-
PS-wave Processing and Velocity-model Building from a Sparse OBN Array
Authors R.R. Haacke, L. Casasanta, P. Chiarandini, N.V.V. Golla, A. Kalil, S. Drummie, I. Meades and A. StrudleySummaryPS-wave processing is particularly challenging for sparse ocean-bottom acquisitions in deep water. Problems in signal processing revolve around the extreme asymmetry of raypaths (due both to the mode conversion and also to the acquisition geometry), as well as severe aliasing and low fold in conversion-point gathers that would be well focused with denser acquisitions. Aliasing and low fold also cause problems in PS-wave velocity-model building, since migrated gathers are typically noisy and multidimensional interpolation or regularisation methods produce inaccurate images in the shallow subsurface. These problems require methodological modifications both in the processes applied and in the workflows with which they are implemented. For example, PS-wave directional-designature and surface-related multiple suppression are particularly effective with receiver-domain radial-down deconvolution. Meanwhile, meaningful updates of principle shear velocity and anisotropy without use of image gathers for tomography or quality control can be achieved using common-receiver controlled-beam migration straight to stack. Examples from data acquired in deep water to the west of the Shetland Isles (UK) show the efficacy of such modifications, and indicate that PS data from such sparse acquisitions can provide useful images of the subsurface.
-
-
-
Application of Excitation Amplitude Imaging Condition to PS Wave Imaging in Elastic Reverse-time Migration
Authors Q.Z. Du, M.Q. Zhang, X.F. Gong and Y. LiuSummaryFor reverse-time migration, the imaging condition is one of the key factors that will not only determine the image quality but also influence the storage cost and computational efficiency. In this paper, we extend the highly-efficient and well amplitude-preserved excitation amplitude imaging condition from acoustic reverse-time migration to elastic reverse-time migration. In this paper, we firstly provide the detailed realization process of this imaging condition in elastic reverse-time migration. Considering the polarity reversals in PS image, we then propose an innovative way to combine the polarity reversal correction method with this imaging condition in PS wave imaging. Numerical examples demonstrate that this imaging condition is well amplitude-preserved and can obtain polarity-consistent PS image of comparable quality to that of source-normalized crosscorrelation imaging condition. Meanwhile, the disk occupation can be almost negligible and the computational efficiency is highly improved compared with those of crosscorrelation-based imaging conditions.
-
-
-
Near-surface Imaging Challenges for Marine PS-wave Data
Authors L. Casasanta and S. GraySummarySparse seafloor seismic acquisitions present new challenges when processing and imaging converted wave (PS-wave) data. The severe anti-symmetric nature of PS-wave reflections for deepwater acquisitions, and the strong aliasing and low fold derived from sparse shot or receiver arrays, necessitate the solution of significant practical problems to achieve high-quality seismic images near the seafloor. The problems of PS-wave imaging with deep-water, sparse ocean-bottom receiver arrays (OBN or OBC) has motivated new investigation and led to interesting results.
For sparse acquisition geometries, wavefield aliasing is more acute when processing in some domains than others. For example, shot-carpet acquisitions recorded with sparse ocean-bottom nodes are well sampled in the common-receiver (CR) domain but poorly sampled in the common shot (CS) domain. Migration of these data is more attractive in the CR domain, in which the process of anti-aliasing during migration may be more effectively carried out than in the CS domain. Nevertheless, the summation of all the migrated CR wavefields still suffers from aliasing caused by the large receiver spacing. Further, low S-wave velocities are responsible for narrow illumination cones for all the CR migrated images, and the coarse receiver spacing cannot “heal” these images to produce a coherent stacked image near the seafloor. Here we propose some practical solutions to mitigate the effects of receiver spacing. We use Gaussian beam migration (GBM) to track surface ray parameters into the subsurface. We use this information both to redatum CR recorded wavefields above the seafloor and to anti-alias CR migration. Also, we show that PS-wave common-offset-vector (COV) GBM can provide adequate anti-aliasing when the receivers are relatively close together (OBC). Synthetic examples illustrate the effectiveness and limitations of these solutions for both OBC and OBN data.
-
-
-
Wavefield Decomposition of Field Data, Using a Shallow Horizontal Downhole Sensor Array and a Free-surface Constraint
Authors N. Grobbe, J. van der Neut, C. Almagro Vidal, G. Drijkoningen and K. WapenaarSummarySeparation of recorded wavefields into downgoing and upgoing constituents is a technique that is used in many geophysical methods. The conventional, multi-component (MC) wavefield decomposition scheme makes use of different recorded wavefield components. In recent years, land acquisition designs have emerged that make use of shallow horizontal downhole sensor arrays. Inspired by marine acquisition designs that make use of recordings at multiple depth levels for wavefield decomposition, we have recently developed a multi-depth level (MDL) wavefield decomposition scheme for land acquisition. Exploiting the underlying theory of this scheme, we now consider conventional, multi-component (MC) decomposition as an inverse problem, which we try to constrain in a better way. We have overdetermined the inverse problem by adding an MDL equation that exploits the Dirichlet free-surface boundary condition. To investigate the successfulness of this approach, we have applied both MC and combined MC-MDL decomposition to a real land dataset acquired in Annerveen, the Netherlands. Comparison of the results of overdetermined MC-MDL decomposition with the results of MC wavefield decomposition, clearly shows improvements in the obtained one-way wavefields, especially for the downgoing fields.
-
-
-
Pore-scale Analysis of Electrical Properties in Laminated Rock Using Digital Rock Physics
More LessSummaryA significant percentage of the world’s estimated hydrocarbon reserves are contained in thinly layered reservoir formations. In order to analyze the electrical properties of laminated rocks, a laminated digital rock is reconstructed via process-based method. Then a finite element method special designed to analysis the electrical properties of porous media is imposed on the model. Differing from the electrical properties of homogenous rock, the RI-Sw curve of laminated rock does not show a linear relationship, due to the bedding effect and anisotropy. The RI-Sw curve can be divided into two linear segments with different saturation exponent. The first segment with lower slope is at high value of water saturation, and the second one with higher slope is at low water saturation. If we take the capillary pressure into consideration, the first segment of RI-Sw curve with low value of slope should correspond to macro-porous layers and the high slope segment should correspond to the micro-porous layers. Traditional Archie equations cannot well describe the electrical behavior of laminated rock partially saturated with formation water. Therefore, the electrical behavior of laminated rock should be considered in calculating water saturation.
-
-
-
Numerical Estimation of Carbonate Properties Using a Digital Rock Physics Workflow
Authors M. Osorno, D Uribe, E.H. Saenger, C. Madonna, H. Steeb and O. RuizSummaryDigital rock physics combines modern imaging with advanced numerical simulations to analyze the physical properties of rocks. In this paper we suggest a special segmentation procedure which is applied to a carbonate rock from Switzerland. Starting point is a CT-scan of a specimen of Hauptmuschelkalk. The first step applied to the raw image data is a non-local mean filter. We then apply different thresholds to identify pores and solid phases. Because we are aware of a non-neglectable amount of unresolved micro-porosity we also define intermediate phases. Based on this segmentation determine porosity-dependent values for the p-wave velocity and for the permeability. The porosity measured in the laboratory is then used to compare our numerical data with experimental data. We observe a good agreement. Future work includes an analytic validation to the numerical results of the p-wave velocity upper bound, employing different filters for the image segmentation and using data with higher resolution.
-
-
-
Quantitative Investigation of Microstructures within Porous Rocks by Using Very High Resolution X-ray Micro-CT Imaging
Authors G. Zacher, M. Halisch, P. Westenberger and F. SiekerSummaryToday’s high-resolution X-ray CT with its powerful tubes and great detail detectability lends itself naturally to geological and petrological applications. Those include the non-destructive interior examination and textural analysis of rocks and their permeability and porosity, the study of oil occurrences in reservoir lithologies, and the analysis of morphology and density distribution in sediments – to name only a few. Especially spatial distribution of pores, mineral phases and fractures are important for the evaluation of reservoir properties. The possibility to visualize a whole plug volume in a non-destructive way is undoubtedly the most valuable feature of this type of rock analysis and is a new area for routine application of high resolution X-ray CT. All presented geological CT volume evaluations were performed with GE’s phoenix nanotom, a 180 kV/15 W nanofocus CT system tailored specifically for extremely high-resolution scans of samples weighing up to 3 kg and with voxel-resolutions down to < 300 nm.
-
-
-
Uncertainty Quantification in Rock Physics Modelling
By D. GranaSummaryRock Physics describes how reservoir rock properties, such as porosity and mineralogy, affect elastic properties, such as seismic wave velocities. If the rock properties are known, then we can predict the corresponding elastic properties and the seismic response using a rock physics model. Since measurements of rock properties in the reservoir are uncertain, we propose to apply rock physics models to random variables rather than deterministic values. If the input random variable can be described by a probability distribution, then we can formulate a probabilistic approach to rock physics models to estimate the exact probability distribution of the rock physics predictions. In this paper, we present the probability theory of this new approach, the application to two common models, such as Gassmann’s equation and Raymer’s relation, an illustrative example and a real case study.
-
-
-
Constraining Petrophysics with Rock Physics
By M. SamsSummaryIn the context of seismic reservoir characterisation the availability of high quality and consistent petrophysical analysis is essential. Achieving consistency is often compromised by poor data quality, lack of sufficient data and ambiguities in the data. Rock physics modelling can help to improve the consistency by ensuring that the petrophysical interpretations are also constrained by the elastic logs. There are three ways that rock physics can be used. First as a simple quality control, second to provide quantitative feedback to the petrophysics and third in a joint interpretation. The choice of method depending on the confidence achieved in the rock physics model.
-
-
-
Combined Modeling of Burial History and Rock Physics Properties - A Barents Sea Demonstration
Authors P. Avseth, I. Lehocki and T. VeggelandSummaryIn this study we show how we can combine basin modeling principles with rock physics models. We apply the Walderhaug diagenetic modeling approach to simulate porosity and rock texture evolution during geological time, and use these as input to contact theory rock physics models which allows us to predict seismic properties as a function of the burial history. We also demonstrate how this approach can be used to estimate tectonic uplifts, and show an example from the Barents Sea.
-
-
-
Unlocking the Secrets of Seismic Anomalies in the Load Transfer Domain
Authors A.S. Selnes, A.E. Edwards, N.W. Whitfield, K.W. Waters and C.M. MarzocchiSummaryWe will in this paper describe the importance of how integration of elastic properties and pressure is essential in order to better understand and constrain AVO signatures in the sub-surface, and particularly how amplitudes in the load transfer domain may differ from what is expected at a given prospect level, and thus may be overlooked as false negatives if not accounted for. This is particularly important in a regional context for exploration purposes (e.g. regional studies).
-