- Home
- Conferences
- Conference Proceedings
- Conferences
72nd EAGE Conference and Exhibition incorporating SPE EUROPEC 2010
- Conference date: 14 Jun 2010 - 17 Jun 2010
- Location: Barcelona, Spain
- ISBN: 978-90-73781-86-3
- Published: 14 June 2010
101 - 200 of 797 results
-
-
Building Borehole-calibrated and Geologically Plausible Anisotropic Models Using Wells and Horizon-guided Interpolation
Authors A. V. Bakulin, Y. Liu and O. ZdravevaUniversal adoption of anisotropic depth imaging places stronger focus on delivering quality anisotropic models that increase confidence in the depth positioning of seismic volume while also optimizing image quality. Calibration with well data such as checkshots or markers from one or several wells is becoming a must. Anisotropy determined around boreholes requires careful geologically-driven extrapolation between wells. We present a simple workflow that addresses both aspects. Borehole calibration step includes traveltime-preserved smoothing of the checkshots and deriving anisotropy profiles at wells by manual inversion or localized tomography. Then horizon-guided interpolation creates volumes of Thomsen parameters propagated consistently with the subsurface geology. Once new anisotropy volumes are derived, the entire velocity cube is revised to preserve normal moveout velocities. At the final step, the updated model is stretched into new seismic image depth. We present a case study that applies this workflow to wide-azimuth seismic data from the Gulf of Mexico where a VTI depth model is built for an area of 100 outer continental shelf (OCS) lease blocks using 18 wells with checkshots.
-
-
-
Migration Velocity Analysis Aided by RTM
Authors N. Bienati, C. Andreoletti, R. Brajucha and F. DoniselliRTM is the most advanced imaging technology and it is starting to be routinely applied to production projects. Nonetheless, in order to fully exploit its potentiality RTM should be also integrated in the velociy model building loop. The first step along this direction is the application to salt interpretation, especially in case of very complex 3D geometries. In this paper we show a succesfull case history where RTM aided salt interpretation yielding a more accurate salt model and accordingly an improved imaging. We discuss also why RTM gave a unique added value that couldn't be achieved by conventional (and approximated) PSDM algorithms.
-
-
-
Complex-salt Model Building Using a Combination of Interactive Beam Migration and Layer-stripping RTM
More LessA new salt velocity model building methodology is proposed which allows effective testing of different salt interpretation scenarios. In this methodology, we combine the strength of efficiency from interactive beam migration with the accuracy of layer-stripping RTM to derive a more accurate salt geometry. Using interactive salt geometry editing and efficient beam migration, a large number of salt interpretation scenarios are quickly tested and narrowed down to a small number of likely salt interpretation cases. This is followed by a reduced number of layer-stripping RTM runs to single out the final salt velocity model. Redatuming the wavefield from surface to a user defined subsurface datum plays a pivotal role in this methodology; it enables improving the quality of beam migration and the efficiency of RTM.
-
-
-
Improved Ray-based Seismograms by Combining Modeling by Demigration with a Prestack Depth Migration Simulator
Authors T. Kaschwich and I. LecomteRay-based seismic modeling methods can be applied at various stages of the exploration and production process, e.g., for comparison, simulation or representation of seismic data. Ray tracing in its standard forms is not always able to produce complete synthetic seismograms, as may be required for processing tests. Standard reflection-based modeling gives good results in many cases, but could fail for some complex structures. For instance, in the presence of sharp edges on the modeled reflector, or in the vicinity of caustics, classical dynamic ray tracing does not give reliable amplitude values. Therefore, we propose an alternative approach where we do not use reflection modeling. Instead we use an optimized ray-based approach to calculate Green's functions, e.g., amplitudes and traveltimes, which will be applied within tailored modeling schemes. Here, we adapt the original modeling by demigration approach using an alternative PSDM simulator. Even a simple two-layer model reveals the weaknesses of the standard ray tracing, and demonstrates on the contrary that the synthetic seismograms calculated with the presented method are more complete and significantly improved. Finally, we document the applicability of this approach using a reservoir model of the Norne Field (Norwegian continental shelf).
-
-
-
Traveltime and Reflection Coefficients Accuracy of Staggered-grid Finite-difference Simulation of Seismic Waves
Authors O. Podgornova, V. Lisitsa and D. VishnevskyFinite-difference methods are widely used for simulation of wave propagation in heterogeneous media, in particular for a case of discontinuous parameters. Usually, the same numerical scheme is used in the whole computational domain whether there are discontinuities or not. However if medium parameters are not modified in the vicinity of the discontinuities the accuracy of the scheme would be lost. In current work we derive the modified elastic parameters to provide the second-order accuracy of the Virieux scheme on elastic-elastic interface in the sense of reflection-transmission behaviour. We support the theoretical results with numerical experiments.
-
-
-
Wave Equation Illumination
Authors C. M. Lapilli, A. Gonzalez, D. Nichols and J. P. PerdomoRecent advances in the imaging of prestack seismic data, in particular more accurate migration algorithms such as reverse-time migration (RTM), have made higher-quality images possible. However, tools to understand the factors affecting the quality of the results such as illumination are not readily available. Conventionally, illumination and resolution studies involve ray-based methods, and although they capture the effects of lateral variations in velocities, intensity and directional information of wavefield propagation, and irregular acquisition geometries, they suffer from severe limitations in accuracy in complex regions where the high-frequency asymptotic approximation of rays might fail. Wave-equation methods generally do not generate the directional information of the wave propagation, which prevents us from using a wave-equation method directly for illumination studies. In this work, we investigate and develop target-oriented illumination studies from wave-equation wavefield propagation. Possible applications of the proposed method include: computation of illumination maps for optimization of aperture for reverse-time migration, amplitude versus angle (AVA) or amplitude versus offset (AVO) targeted illumination compensation, survey design analyses for selection of the most suitable acquisition geometry, computation of point-spread functions for the analysis of resolution and uncertainty, among others.
-
-
-
Solving the Wave Equation Using Curvelets
Authors B. Sun, J. Ma, H. Chauris and H. YangSeismic imaging is a key step in seismic exploration to retrieve the Earth properties from seismic measurements at the surface. One needs to properly model the response of the Earth by solving the wave equation. We present how curvelets can be used in that respect. Curvelets can be seen from the geophysical point of view as the representation of local plane waves. The unknown pressure, solution of the wave equation, is decomposed in the curvelet domain. We derive the new associated equation for the curvelet coefficients and show how to solve it. In this paper, we focus on a simple homogeneous model to illustrate the feasibility of the curvelet-based method. This is a first step towards the modeling in more complex models. In particular, we express the derivative of the wave field in the curvelet domain. The simulation results show that our algorithm can give a multi-scale and multi-directional view of the wave propagation. A potential application is to model the wave motion in some specific directions.
-
-
-
Convolutional and Non-convolutional PMLs for High-order Schemes Optimized at Grazing Incidence for the Seismic Wave Equation
Authors R. Martin, D. Komatitsch, S. D. Gedney and E. BruthiauxThe perfectly matched layer (PML) absorbing boundary condition has proven to be very efficient from a numerical point of view for the elastic wave equation to absorb both body waves with nongrazing incidence and surface waves. However, at grazing incidence the classical discrete PML method suffers from large spurious reflections that make it less efficient for instance in the case of very thin mesh slices, in the case of sources located close to the edge of the mesh, and/or in the case of receivers located at very large offset. The PML can be improved at grazing incidence for the seismic wave equation based on unsplit convolution (CPML) or non convolution techniques (ADE-PML/Auxiliary Differential Equations PML). Unsplit CPMLs for the velocity and stress formulation of the seismic wave equation are classically computed based on a second-order finite-difference time scheme.To increase the accuracy of the algorithm, particularly for long time periods, we implemented an unsplit non-convolutional high-order time-stepping version of ADE-PML. ADE-PML and CPML are equivalent at the second-order in time. Applications to purely elastic, anisotropic, poroelastic and viscoelastic thin slices in 2D/3D configurations are shown.
-
-
-
2D Frequency-domain Seismic Wave Modeling in VTI Media Based on a Hp-adaptive Discontinuous Galerkin Method
Authors R. Brossier, Y. Gholami, J. Virieux and S. OpertoSeismic imaging in complex media by full waveform inversion requires efficient and accurate modelling tools to solve the full wave equation. In this study, we present a low order finite element discontinuous Galerkin (DG) method in the space-frequency domain, for 2D P-SV waves propagation in VTI media. DG method provides a high level of flexibility for simulation in complex media, thanks to the cell size and interpolation order adaptation that depends on local physical properties. The method is validated with a time-domain finite-difference method in an homogeneous strong anisotropy model and in a realistic heterogeneous model. Results show that VTI DG method allows to compute accurate solutions in complex media and is therefore suitable as a modelling engine for 2D elastic VTI full waveform inversion
-
-
-
3D Spectral Element Method Combined With H–refinement
Authors A. C. Lesage, R. Aubry, G. Houzeaux, M. Araya Polo and J. M. CelaIn this work, we propose a Spectral Finite Element method (SEM) for 3D acoustic wave equation modelling combined to an unstructured mesh adaptation method, the h-refinement approach. This method is well adapted to complex geometries and unstructured meshes, which is very important dealing with complex velocity models. Impulse response for modeling has been shown to validate the proposed SEM implementation. Moreover, it proved good parallelization speed up, which is critical for high performance computing purposes as required by Reverse Time Migration.
-
-
-
On the Accuracy of Ruger’s Approximations for PP Reflection Coefficients in HTI Media
Authors A. Ali and M. JakobsenWe here investigate the accuracy of Ruger’s approximation for PP reflection coefficients within the context of seismic AVAZ analysis in HTI media; that is simple reservoir model with a single set of vertical fractures. An individual comparison of forward and inverse modelling results have been done for the interface between an isotropic and HTI medium. The comparison is performed for different contrast (small and large) and anisotropy by changing the fracture density. The elastic stiffness tensor of the fractured reservoir of HTI medium was calculated using a combination of T-matrix with Brown-Korringa relations. We have calculated the percentage error for the Ruger’s approximation(s) with respect to the exact formula for reflection coefficients. The inversion of this non-linear forward model with respect to fracture density and azimuthal fracture orientation was done in a Bayesian setting, which provide information about uncertainties as well as the most likely values. The maximum percentage error for large contrast and large anisotropy (fracture density = 0.2) may reach up to 70%. An application to the synthetic case of characterizing a fractured damage zone, where fracture density and orientation depend on distance from the fault core was also provided to elaborate the workflow.
-
-
-
Up-down Deconvolution in the Presence of Subsurface Structure
More LessWavefield separation into up- and down-going waves is a fundamental step of ocean-bottom data processing. It provides receiver-side multiple attenuation; source side multiples in general require further processing. In this paper we consider the up-down deconvolution method as a means to simultaneously eliminate both source-side and receiver-side multiples. In theory, this method is strictly valid for a horizontally layered earth only but we demonstrate analytically that if the seabottom is flat it is robust in the presence of structure. We present a real data application example of up-down deconvolution and we study the performance of the method in the presence of complex geology using a 2D synthetic model. Finally, we find that up-down deconvolution provides excellent results on this complex synthetic, in agreement with our theoretical derivation.
-
-
-
Efficient Elastic Wave-mode Separation in TTI Media
More LessWave mode separation is an indispensable step in elastic wave equation imaging. For isotropic media, the separation is typically done using Helmholtz decomposition. However, Helmholtz decomposition does not completely separate wave modes for anisotropic media. Wavefield separation operators for TI (transverse isotropic) models are constructed based on the polarization vectors evaluated at each point of the medium by solving the Christoffel equation using local medium parameters. These polarization vectors can be represented in the space domain as localized filters, which resemble conventional derivative operators. The wave mode separation for TI media is usually implemented as non-stationary filtering with local filters. However, the accurate separation in the space domain is computationally expensive especially in 3D. In this paper, we show an efficient method for wave-mode separation, which exploits the same general idea of projection on polarization vectors. The method consists of two steps: first separate wave modes in the wavenumber domain for a number of reference models, and second interpolate the wavefields in the space domain using the spatially-variable model parameters. An example shows that the separation by interpolation works well for models with complex geology.
-
-
-
WAZ Mirror Imaging with Nodes for Reservoir Monitoring – Dalia Pilot Test
Authors D. Lecerf, J. L. Boelle, A. Lafram and J. CantilloRepeated marine seismic recorded with towed streamer have been proved successful for imaging reservoir productions. Unfortunately major infrastructures (FPSO) constitute a “blind zone” for the reservoir illumination from the sea surface. Because undershooting surveys come with repeatability and HSE issues, nodes imaging appears to be a valuable solution. In 2009 a deep water nodes surveys was acquired by Total Angola. Since “Base” surveys are acquired usually with marine streamers, the first objective of this paper is to find out how to reconcile nodes and streamer data in order to provide comparable images. The second objective is to propose an azimuth compliant processing approach valid for an optimum node WAZ mirror imaging. Two original processing approaches are described: firstly, data cross-matching is done in angle domain in order to provide similar ray path and equivalent sea surface offset. Secondly, we show that the concept of offset vector binning using hexagonal tiles is applicable to the nodes acquisition geometry. Mirrored data migration in common offset vector domain provides CIGs with preserved offset and azimuth information. Post-migration processing like full azimuthal residual move-out and azimuthal illumination selection can then be applied for an optimal reconciliation between nodes and streamer data.
-
-
-
High-precision Measurement of Split S-waves in OBS Data and its Application to Marine Slope Stability Assessment
Authors R. Haacke, R. Exley and G. K. WestbrookSplit shear waves are highly sensitive to pore-fluid pressure. Measurement shear-wave splitting could provide a powerful way to monitor subtle changes of pore-fluid pressure if accuracy is high enough. This is particularly difficult when done with marine data, however; when the target is thin, or deep, the shear-wave splitting signal easily becomes overwhelmed by the effects of heterogeneity or P-wave anisotropy. We illustrate the potential for making highly-precise measurement of polarisation directions and time delays of split PS waves using an array of OBS deployed near the Storegga slide offshore Norway. We find a 50-m-thick zone about 150 mbsf, under 950 m of water, in which S-wave polarisation orientations are consistently flipped by 90-degrees. Data from a nearby borehole and results of traveltime inversion of PP and PS phases indicate that this zone is over-pressured. A reflection image across the OBS array shows clear evidence for slope failure detaching on planes within the overpressured zone. We suggest that high-precision measurement of S-wave splitting could be generally applied to the assessment and monitoring of submarine slope stability, with clear application to geotechnical engineering and development of seabed infrastructure.
-
-
-
PP/PS PSDM, AVO and Joint Inversion Applied to a Virgin Exploration Area in the North Sea
Authors G. Ronholt, Ø. Korsmo, H. Hoeber, A. Smith, P. Hughes and M. ØygarenWintershall Norge ASA have identified a Palaeocene seismic amplitude anomaly in blocks 33/5 & 6 in the North Sea. The anomaly is located in a virgin area for Palaeocene exploration and there is no analogous anomaly that has been described or drilled within this or adjacent parts of the North Sea (Hughes et al, 2010). Multi component seismic (OBC) was acquired in the summer of 2009 in order to de-risk the prospect. The combination of converted wave (PS) and compressional wave (PP) data can be used to de-risk a prospect before drilling in order to distinguish between lithology and fluid fill. The OBC data were processed by CGGVeritas in the autumn of 2009. A complex 4 component (4C) processing sequence was implemented and imaged through anisotropic PP- and PS-wave pre-stack depth migrations (PSDM). The resulting PS images are of high quality and we demonstrate successful applications of PS AVO analyses and joint PP/PS inversion.
-
-
-
Extract Subsurface Density Information Using Converted Waves – Workflow and Case Study in the Gulf of Mexico
Authors K. Zou and A. KoesoemadinataThe S-Zero Stack algorithm extracts robust density information of the subsurface from converted waves. The method is based on the separation of velocity effects from density effect analytically in a PS reflection equation (Aki & Richards, 1980); thus, a special stacking method can be designed to capture density variations at subsurface interfaces without going to inversion. This paper deals with practical aspects of the method. A reasonable workflow extracting subsurface density information is discussed. As a case study, the data of Eugene Island, Line 421 in the Gulf of Mexico is used and the resulting subsurface density anomalies are calibrated with the density log in Well EI 273 C9.
-
-
-
PP–PS Joint Inversion Applied to 2D–3C High Resolution Seismic Data – A Case of Junggar Basin, Northwest China
Authors L. Shen, Y. F. Dang, B. Lou, X. G. Miao, P. Wang and S. H. ZhangFor exploration in the Chepaizi area, Junggar Basin of Northwest China, lateral lithologic variations due to deltaic and fluvial sediments present major challenges to the characterization of oil sand reservoirs. The major oil reservoirs exist in so-called “sweet spots”. However, conventional P wave is not able to effectively distinguish true “bright spots” from false ones, which leads to high drilling risk. For example, a strong reflection amplitude is observed in the target layer for Che89, a highly productive oil well drilled in 2006. But only a few wells drilled on a similar reflection response turned out to be successful commercial discoveries. It is widely believed that not only the fluid content, but also lithologic variations, have led to such misleading outcomes. Effectively locating the hydrocarbon distribution by eliminating the impact of lithologic variations could be the key for successful oil exploration. In order to better characterize the reservoirs, PetroChina Xinjiang acquired a high resolution 2D-3C seismic line in 2007, intersecting two existing wells, Che89 (good producer) and Che97 (dry well), by employing DSU3 digital sensor units with advanced MEMs accelerometer technology. Both high-resolution amplitude-preserving processing and PP-PS joint inversion were employed for reservoir characterization.
-
-
-
Simultaneous Inversion for Elastic Parameters Using PP- and PS-waves Jointly in the Ray Parameter Domain
More LessIn seismic inversion for P-wave velocity, S-wave velocity and density simultaneously, it is difficult to solve the density parameter. We develop a simultaneous inversion by using PP- and PS-wave reflection coefficients jointly in the ray parameter domain. Before joint inversion,we need to appropriately calibrate two input data sets both in space and time with their true amplitudes. We show that the simultaneous inversion is superior in terms of accuracy and stabilization when compared to inversion using PP-wave alone. We apply the simultaneous inversion on a tight-sand gas reservoir. With the well-calibrated PP- and PS- waves and appropriately designed weighting factors, we are able not only to invert high-resolution P- and S-wave velocities, but also to estimate density accurately from which reservoir distribution can be identified.
-
-
-
A Comparison of Three Automatic Interpretation Techniques for Magnetic and Gravity Data
Authors D. Gerovska, M. J. Arauzo-Bravo and K. WhalerWe compare three automatic methods for interpretation of magnetic and gravity data grids: MaGSoundDST (Magnetic and Gravity Sounding based on the Differential Similarity Transform), MaGSoundFDST(Magnetic and Gravity Sounding based on the Finite Difference Similarity Transform) and DST Euler (Euler deconvolution based on the DST). The methods are window-based but produce a single solution per source. They account for the presence of linear background in the data and do not require reduction-to-the-pole in the magnetic data case. MaGSoundFDST requires only measured and upward-continued data, while MaGSoundDST and DST Euler use measured data and first-order derivatives. We applied the methods to a vertical component data grid from the Bolyarovo-Voden magnetic anomalous zone, southeast Bulgaria. MaGSoundFDST performed best in detecting the deeper central parts of sources that almost crop out on the surface.
-
-
-
3D Gravity Inversion of a Faulted Basement Relief Using the Total Variation Regularization
Authors V. C. F. Barbosa, C. M. Martins, W. A. Lima and J. B. C. SilvaWe present a gravity inversion approach that estimates the basement relief of a fault-bounded sedimentary basin. The sedimentary pack is approximated by a grid of 3D vertical prisms juxtaposed in both horizontal directions. The prisms’ thicknesses represent the depths to the basement and are the parameters to be estimated from the gravity data. To obtain 3D depth-to-basement estimates we use a nonsmooth function, named total variation regularization, which allows the reconstruction of a nonsmooth basement relief. We deduce a compact matrix form of the gradient vector and of the Hessian matrix of the approximation to the total variation function that allows using the regularized Gauss-Newton algorithm. We apply our method to synthetic data from a simulated fault-bounded sedimentary basin. By comparing our result with that obtained with the smoothing regularization, we note the improved resolution of the proposed method in defining the normal faults in the basement relief. Inversion of the gravity data from the onshore and part of the offshore Almada Basin, on Brazil’s northeastern coast, show good correlation with known structural features and confirm the potential of our method in imaging in-depth normal faults in the basement of a sedimentary basin.
-
-
-
A Modified Tilt–depth Method for Kimberlite Exploration
More LessThere are many enhancement techniques available for potential field data, such as horizontal and vertical derivatives, vertical continuation, pole reduction, etc. The Tilt angle (Miller and Singh, 1994) produces a balanced vertical derivative which can be particularly useful as an enhancement tool. Recently Salem et al (2007) introduced the Tilt-depth method, which produces depth estimates of vertical contacts which have vertical magnetisation (or have been pole-reduced). The method determines the depth as being half the distance between the ±45 contours of the Tilt angle. This paper extends the Tilt-depth method to the model of the gravity anomaly from a vertical cylinder, which can be useful in Kimberlite pipe exploration projects, particularly now that airborne gravity data is becoming more readily available.
-
-
-
Interpretation of 2-D Gravity Data using 2-D Continuous Wavelet Transform
Authors F. Salajegheh, A. Roshandel Kahoo and H. R. SiahKoohiRecently the continuous wavelet transform has been proposed for interpretation of potential field anomalies. Almost, all of the 2-D wavelet transform based methods for interpretation of potential field data use the Poisson wavelet. In this paper we introduced a 2-D wavelet based method that uses a new mother wavelet for interpretation of gravity data. The new wavelet is the first horizontal derivative of gravity anomaly of a buried cube with unit dimensions. Our wavelet was applied on both synthetic and real data and gave good results on both data.
-
-
-
Exact Finitely Expanded Gravity Anomaly of Uniform Thin Polygonal Sheets
Authors H. Holstein and C. AnastasiadesStandard closed form anomaly formulae for polyhedral targets exhibit numerical instability. Under finite floating point precision, the accuracy of the computed anomaly increasingly degrades with increasing target distance. The instability may with advantage be studied from the anomaly of a thin polygonal sheet, as the polyhedral formulae represent a superposition of sheet formulae. We derive an exact finite expansion of the thin sheet formula, corresponding to a point source term plus perturbation. This form exhibits the required absolute numerical stability. The result suggests generalization to absolutely stable polyhedral anomaly formulae.
-
-
-
Analysing Full Tensor Gravity Data with Intuitive Imaging Techniques
Authors J. L. Dickinson, C. A. Murphy and J. W. RobinsonFull Tensor Gravity Gradiometry (FTG) has proven to be an effective tool for both the Oil and Gas and Mineral exploration industries since it was introduced in the late 1990s. The multi component gravity surveying technology provides a rich source of information that lends itself ideally suited to imaging and delineating key geological structure. New techniques involving invariant analysis have been developed that facilitate quick and efficient imaging to extract detailed geological information. This paper presents these imaging techniques applied to data collected over the Vinton Dome in Louisiana.
-
-
-
Potential Field Searchlights
More LessPotential field searchlights are a new schema (analogous to the schemata of smoke-rings in electromagnetic prospecting and rays in seismic) that assist the geophysicist to understand the measurement of any potential field. This provides a cognitive structure which, combined with experience, allows rapid and reliable decision-making.
-
-
-
Gravi-magnetic Anomalies of Uniform Thin Polygonal Sheets
Authors H. Holstein, D. Fitzgerald and C. AnastasiadesThin planar sheets are useful gravitational and magnetic models of dykes and veins treated as two-dimensional geophysical structures on the scale of the survey. Thus, the anomaly of a polygonal thin sheet of uniform surface density or magnetization in arbitrary orientation has practical interest. The limiting thin-sheet anomaly can be approached from the corresponding polyhedral parallelepiped under decreasing thickness, though the numerical limit cannot be reached this way on account of the floating point finite precision. We derive the analytical zero thickness limit for the gravity potential while maintaining finite total mass. We use the concept of gravi-magnetic similarity to extend the thin-sheet potential formula to include the potential, field and field gradient in both gravity and magnetic cases, thereby generalising other studies that have obtained isolated polygonal thin-sheet anomaly solutions. We compare the anomalies computed by the new formulae to those of corresponding finite thickness targets, and to the finite difference estimates of the field and field gradient obtained from numerically differentiated thin-sheet potentials. In both cases a second order rate of approach to the limit is observed, verifying the correctness of the new formulae. Thin-sheet solutions are attractive for their reduced computational burden compared to full parallelepiped solutions, while the stacking of thin sheets may be used to simulate variable density or magnetization targets. It is anticipated that thin-sheet solutions presented here will find wide application in gravi-magnetic modelling.
-
-
-
Automatic Fault Detection Based on Seismic Data Correlation Analysis
Authors N. L. Ivanova JSC, A. G. Averbukh JSC and E. R. Ahmetova JSCWe describe the problem and suggest multi-step technique for fault surfaces and zones automatic detection and tracking using 3D seismic data. We examine as fault indicator the multivariate seismic attribute, which components are reconstructed on the base of seismic cube correlation structure analysis. The algorithm of correlation matrix reduction to principal axes is applied. Further we consider the set of parameters, analyzed a behavior of correlation matrix eigenvectors and eigenvalues; such set identifies fault zones. In order to associate fault zones as separated connected fields (facies), we use the multivariate classification algorithms, in particular, cluster analysis methods. It allows us to introduce, to formalize and to assess in classifying procedures a similarity level for the multivariate 3D seismic fault indicators. The special step of fault detection and tracking procedure is fault boundaries and surfaces tracking. In image processing such operations are called contour detection and skeletonization. We use available approaches with respect to our task
-
-
-
Attributes Designed for Detection of Salt Diapirs Without Well Defined Boundaries
Authors A. Berthelot, A. S. Solberg and L. J. GeliusPrestack depth imaging can be particularly difficult in complex geologies such as salt diapirs. Iterative velocity model updating and manual interpretation of these structures is time-consuming. Automatic detection of salt structures has been investigated but only for cases when the salt interface is well defined. However, in many cases, no such sharp boundaries exist. We therefore propose a selection of attributes designed to reveal texture differences between the salt diapirs and the surrounding geology. We also demonstrate how such attributes can be used for automatic segmentation of the salt structure.
-
-
-
Re–binning Pre–stack Seismic Data for Fluid Contact Determination – AVO Screening and Visualisation
Authors A. K. T. Wever, M. Chandler, A. Huck and F. BlomCommon Contour Binning (CCB) is an approach, based on the principles of the CMP method, to re-organize and filter post- and pre-stack seismic data. It is a very powerful technique to highlight subtle hydrocarbon-related amplitude effects on both pre- and post-stack data. Pre-requisite is that such effects are structure conformable with available interpretation, and that they are flat in the domain of investigation. Under these conditions, CCB can identify fluid contacts which are not or hardly visible on conventional seismic displays. As CCB can also highlight subtle AVO effects on raw pre-stack data, it is a very effective technology for interpretive use of unconditioned pre-stack data early in the exploration cycle.
-
-
-
Nonstationary Phase Estimation for Analysis of Wavelet Character
Authors M. A. van der Baan, M. Perz and S. FomelPhase mismatches sometimes occur between final processed seismic sections and zero-phase synthetics based on well logs – despite best efforts for controlled-phase acquisition and processing. Statistical estimation of the phase of a seismic wavelet is feasible using kurtosis maximization by constant phase rotation, even if the phase is nonstationary. After estimation, we achieve space-and-time-varying zero-phasing by phase rotation. We demonstrate how a statistical analysis provides pertinent information about the data that can be used for zero-phasing, as a quality control to check deterministic phase corrections, or even as an interpretive tool for highlighting areas of potential interest.
-
-
-
Reservoir Properties and OGIP through Elastic Inversion and Multi-attribute Calibration – From Development to Exploration
Authors M. Galbiati and M. FervariThis paper discusses the value and limits of a seismic quantitative approach based on elastic inversion and multi-attribute calibration to predict reservoir properties and assess hydrocarbon volumetrics, through the different phases of development of gas fields in the offshore Nile Delta Abu Madi Fm., and their extension to assess nearby exploration opportunities in the same exploration play. The applied workflow allowed to provide not only reservoir quality mapping and conventional property volumes computation (Pseudo-porosity, Lithology, Fluid Saturation), but also the generation of 3D calibrated datasets of relatively unconventional reservoir properties (Porosity Thickness, Net Hydrocarbon in Place). De-risking for fluid saturation did not work in the exploration area, either because of lithological effects on angle dependent elastic and/or low gas saturation.
-
-
-
Cross-validation of Fault Enhancement and Elastic Dislocation Modelling
By J. J. LongFault enhancement are extracting discontinuities from the seismic volume. However, those discontinuities are not all genuine faults. Coherent noise, acquisition footprint or processing artefacts can be mistaken for sub-seismic faults. In order to discriminate between genuine faults and artefacts, we are using an independent method: elastic dislocation modelling. Elastic dislocation (ED) modelling assumes that the strain distributions around large-scale faults are primarily controlled by the coseismic slip on them. The stress distribution can be derived from the strain distribution. Finally the sub-seismic fault orientation and density can be estimated from the stress distribution. The fault enhancement and elastic dislocation don't rely on the same underlying assumption and have different limitations. Therefore, if a feature is detected by both methods the confidence that it is a genuine fault is increased. We have applied this approach to a Tunisian case study and successfully identify a remote feature, interpreted as an interference pattern between two major faults.
-
-
-
Interpretational Applications of Spectral Decomposition in Identifying Minor Faults
By X. D. WeiAt present, spectral decomposition is used in reservoir characterization and hydrocarbon detection. But there is very little literature on fault identification using spectral decomposition. In this paper, on the basis of the high-precision 3D seismic data from the Qiuling oil field, spectral decomposition is used for identifying minor faults. The phase tuning cube is used for interpreting fault planar variation and the common frequency cubes are used for describing the spatial features of faults. Integrating both the phase tuning cube and the common frequency improves the reliability of fault identification.
-
-
-
Using Azimuthal Curvature as a Method for Reducing Noise in Post Stack Curvature Volumes
Authors R. E. Daber and T. H. BoeFor post stack seismic volumes, a major challenge continues to be bringing out desired features while reducing unwanted ones. For seismic volumes that have a distinct noise component, noise often masks post stack attribute results and can hide features. Using post stack curvature attribute generation and specifying an azimuthal direction for calculating the curvature, unwanted directional noise can be significantly reduced and major features in resultant curvature volumes imaged more clearly.
-
-
-
The Role of Geophysics in Well Planning and Execution – A Case Study in the SNS
Authors O. Estaba, K. Veeken, C. Faasen, J. Goudswaard, M. Ligtendag and R. van BoomThe role of Geophysics in the planning and drillings phases of a well is diverse, varying with the challenges of any given well. The objective of this paper is a case study highlighting how Geophysics contributed in both phases, including input stemming from recognition of hazards whilst drilling, to the successful outcome of a difficult well in the Southern North Sea. The well, initially stalled due to economics, was re-activated as a result of new depth conversion insights using PSDM velocity modelling; its trajectory was designed and modified to address hazards interpreted on seismic. In Shell Upstream International - Europe, Geophysics plays a pivotal role in the decision making during the well trajectory design phase and drilling phase. Whilst drilling, output from a parallel full-field review became available that included reprocessed seismic data. The superior quality of these data resulted in improved fault definition at target level and made a significant contribution to de-risk the well.
-
-
-
A Multi Scale Approach on Large Seismic Volumes – Tunisia Case Study
Authors R. Lemaire, F. Pauget, S. Lacaze, T. Cheret, M. Mangue and C. Horno KortRecently global approaches have been developed to compute geological models directly from seismic data. The present paper proposes to adapt on a large data set the method based on the minimization of a cost function developed by Pauget et al, 2009. This seismic data-set presented is a merging of several seismic blocks, located in the Gulf of Gabes off-shore Tunisia. The proposed workflow aims to compute separately geological models on fifteen individual sub- blocks. The interpretation of each sub-blocks produced independent geological models. The sub models were then synchronized using the same global approach but at the scale of the whole area. By this way it provides a large scale geological model with the same level of accuracy obtained on the sub blocks. Such result and approach can be used for the regional interpretation and push back the frontiers of the seismic interpretation for large volumes. Stratigraphic as well as a structural interpretations were achieved on the whole area. It enhanced new features in the existing reservoir in production and helped in the regional understanding of the geology.
-
-
-
Phase Residues for Stratigraphic Discontinuities Detection
Authors K. Zhang, M. Matos, O. Davogustto and K. MarfurtSpectral decomposition has proven to be very successful in direct hydrocarbon detection, thin bed thickness determination, and lithology identification. To date, most publications have focused on the application of the spectral magnitude, and little has been introduced about interpreting the spectral phase although seismic acquisition and processing does a much better job in preserving phase than it does in preserving amplitude. This paper shows how to improve seismic spectral decomposition interpretation by detecting the phase discontinuities in the joint time-frequency spectral phase.
-
-
-
Migration without Velocity Model in Curvelet Domain
Authors A. Bona and M. AlMaghlouthWe present a formulation of velocity-less time migration in curvelet domain. In particular, we first decompose pre-stack gathers to curvelets, which unlike wavelets also contain directional information. Then, we use this directional information to perform migration that does not require any velocity model. This migration is performed directly with the curvelet coefficients. The resulting image is reconstructed by applying the inverse curvelet transform. The process is illustrated on synthetic data and compared to Kirchhoff migration.
-
-
-
Dip-consistent Interpolation of Seismic Prestack and Poststack Data Into Mute Zones Using CRS Techniques
Authors G. Gierse, J. Pruessmann, G. Eisenberg-Klein, H. Trappe and M. ZehnderNear-surface mute zones hamper the derivation of near-surface velocities, and degrade the near-surface image. Operator noise at mute edges may obscure the near-surface image after migration. In order to avoid such problems an interpolation method for prestack and poststack data is proposed using the Common-Reflection-Surface (CRS) method. The dip-consistent interpolation takes advantage of the CRS attributes which provide a highly detailed description of the seismic wavefield including the lateral continuation within a CMP aperture. CMP/offset regularisation and prestack data interpolation are then performed simultaneously during partial CRS stacking into so-called CRS gathers. In addition to a strong signal enhancement and better offset coverage, this partial stacking defines a new mute function in the CRS gathers. The live data is thus extended into mute zones within the CMP aperture of the CRS processing. Stacking or prestack migration transfers this interpolation to poststack domain. Applications in prestack and poststack domain show a good continuation of flat and dipping events. The increased zone of live data, and the improved offset coverage of the CRS gathers, may be be exploited for a fine-tuning of near-surface velocities, and thus for a better focusing and resolution in near-surface imaging.
-
-
-
Common Scatter Point Data Mapping
Authors S. Dell, D. Gajewski and C. VanelleThe Common Scatter Point (CSP) data mapping is a tool in seismic-reflection imaging that generates CSP gathers from common midpoint (CMP) data. A CSP gather is similar to a CMP gather, however, the move out is based on the distance from the sources and receivers to the scatter point location. A key concept of the CSP data mapping is a new parameterization of the double square root (DSR) equation. The parameterized mapping operator allows to map the summed amplitude directly into the apex of the traveltime surface. The algorithm is formulated in the space-time domain, which allows prestack migration velocity analysis and multiparameter stacking like the Common Reflection Surface (CRS) stack. The regular geometry of the CSP gather and the absence of the diffraction enables an improved velocity analysis. The multiparameter stacking of CSP gathers using, e.g, CMP or CRS stacking provides enhanced images as compared to prestack time migration. Application to field data demonstrates that the method is stable and produces highly-focused time-migrated images.
-
-
-
Improving Prestack Migration with CRS Techniques–A Case Study
Authors M. Spinner and N. A. MuellerBoth the Common-Offset Common-Reflection-Surface (CO CRS) stack and Prestack Seismic Data Enhancement (PSDE) are techniques that allow to regularize and enhance the signal-to-noise ratio of prestack data. Such data provides an improved input to time and depth migration and should therefore also result in improved prestack migration results. Here, we present the application of eni’s 2D/3D-narrow-azimuth CO CRS stack implementation to a 2D land dataset followed by a Kirchhoff Prestack Time Migration. The results are compared to a Prestack Time Migration of the original prestack data, a PSDE-based time migration, and a Kirchhoff Poststack Time Migration of the Zero-Offset CRS stack. As will be shown, the CO-CRS stack provides best input to migrate the dataset.
-
-
-
Solution of the 3D EM Problems with Induced Polarization in the Time Domain Using Optimized Krylov Subspace Reduction
Authors M. Y. Zaslavsky and V. L. DruskinWe developed a new algorithm for the solution of the large scale 3D time domain CSEM problems with induced polarization. After spacial discretization we obtain a large system of time-convolution equations. Then this system is projected onto a small subspace consisting of the Laplace domain solutions for a preselected set of Laplace parameters. In order to obtain the solution in the time domain we perform the inverse Fourier transform of the projected system. Our numerical experiments with Cole-Cole model of induced polarization effects show that the introduction of the induced polarization does not have significant effect on convergence of our algorithm.
-
-
-
Numerical Investigation of the Range of Validity of a Low-frequency Approximation for CSEM
Authors S. Bakr and T. MannsethRecently, a novel approximate hybrid method for modeling marine CSEM, simplified integral equation (SIE) modeling, has shown excellent accuracy in modeling the low-frequency response from a thin resistive target in a 2D setting (Bakr and Mannseth, Geophysics 74 (5), 2009). SIE solves the Poisson equation in a region containing the target, and subsequently uses rigorous integral equation (IE) modeling to obtain the EM fields in the receivers. SIE thus replaces the computationally intensive dense-matrix part of the rigorous IE method by sparse-matrix calculations based on a low-frequency approximation of Maxwell's equations. The computational performance of SIE was found to be orders of magnitude better than that of IE (Bakr and Mannseth, 79th SEG Annual Meeting, SEG Expanded Abstracts, 669-673). In the present paper, we investigate numerically the range of validity of SIE in 3D with respect to variation in problem parameters (frequency, electric conductivity, etc.). It is found that the accuracy of SIE is very good for vertically thin resistive targets for typical marine CSEM frequencies. The accuracy is also quite stable wrt. target shape and conductivity contrast, especially for resistive targets.
-
-
-
Error Modelling in Bayesian CSEM Inversion
Authors J. Gunning and M. E. GlinskyMis-characterisation of the noise has significant potential to disrupt reservoir parameter estimates and uncertainties in geophysical inversion. In all form of geophysical inversion, the "effective noise" used in the data misfit absorbs effects from approximate forward modelling in additional to environmental processing and measurement noise. For risk assessment, inversions require parameter uncertainties, and these are best approached from a Bayesian angle. But parameter uncertainty estimates are dependent on the noise model at leading order. Modelling noise in particular can be strongly correlated, and will corrupt parameter uncertainty estimates if the correlations are not taken into account. In CSEM inversion, structural and resistivity parameters can be particularly difficult to disentangle, and their separation is rather vulnerable to systematic components of the noise. We present several ideas to manage the effect, two of which are easily incorporated into standard optimization and sampling schemes.
-
-
-
1D Layered Resolution Analysis of Two Marine EM Exploration Methods
Authors N. H. Cuevas, D. Alumbaugh and J. ChenIn this work the sensitivity/resolution to the properties of a resistive layer in the horizontal source-receiver configuration is compared to that of a vertical source-receiver time domain method. The question of sensitivity of the methods in a 1D layered model is addressed from an analytical stand point, i.e. by analyzing the close form solutions of the guided mode dominating the far offset and that of the image terms describing the solution at close offsets. It is found that the spatial distribution of the guided mode and its high frequency behavior appears to have more resolution to the target than that provided by the spectral behavior of the image terms.
-
-
-
The Use of CSEM Within an Integrated Exploration Project
Authors A. Lovatini, E. Medina, T. Campbell and K. MyersThe CSEM method can provide additional information on rock resistivity. This makes its utilization within exploration workflow quite significant even if not largely adopted. The paper presents an example of its integration with other geophysical methods that can reduce its limitations exploiting complementary stregths.
-
-
-
A Broadband Marine CSEM Demonstration Survey to Map the Uranus Salt Structure
Authors M. Vöge, A. A. Pfaffhuber, K. Hokstad and B. FotlandTo apply a broad spectrum of signal frequencies for a marine electromagnetic survey (0.01 Hz to 500 Hz) is a unique way for detailed mapping of geology in conjunction to hydrocarbon exploration. We present results from a demonstration research survey over the Uranus salt structure (Nordkapp Basin, Barents Sea) involving purpose built broadband receiver systems containing electric and magnetic field sensors as well as four component seismometers. EM data interpretation in tight combination with seismic models indicates a deep salt body rather than the shallow diapir interpreted from seismic alone. The deep salt body was confirmed by an exploration well. The positive results of this proof of concept survey triggered numerous commercial surveys with similar configurations.
-
-
-
CSEM Using Ocean Bottom Cables and Streamer
Authors L. Engell-Sørensen, B. A. Farrelly, R. Henman, S. E. Johnstad, K. Kristiansen and I. VikenThe use of OBC for CSEM is proven to work, and in fact to work better than nodes. Work is now in progress on using a towed streamer or equivalent platform.
-
-
-
Feasibility of EM Monitoring – Acquisition and Inversion
Authors B. Rondeleux and S. SpitzIn the last years the possibility of using the electric resistivity as a DHI has strongly raised the interest of the oil and gas exploration community in the EM methodology. But the ability to directly address the nature of the fluids points also to EM in problems related to the reservoir management (OWC detection, CSS application). This paper discusses some aspects of the feasibility of EM to monitor, from the earth surface, the motion of fluids in a reservoir. We have chosen this particular subject owing to the large contrast in the electrical properties between the fluids in the reservoir. Indeed if the spatial variation of a contrast of the order of 1:100 at a depth interesting the oil industry cannot be handled, it is doubtful that EM would be useful when more subtle changes are involved.
-
-
-
Salt Imaging with Gravity Gradiometry and Magnetic Data – Nordkapp Basin, Barents Sea
Authors C. Stadtler, C. Fichler, K. Hokstad, B. Fotland, C. Gram, P. Hanssen, E. A. Myrlund and S. WieneckeReflection seismic investigation of salt diapirs may have – as in the Nordkapp Basin in the Norwegian Barents Sea – considerable imagining problems to define the shape of salt structures. To improve salt imaging, joint interpretation of full-tensor gravity gradiometry data (FTG), high-resolution magnetic data, controlled-source electromagnetic data and magnetotelluric data, together with 2D and 3D seismic data was applied. This paper focuses on major improvements of the gravity and magnetic modelling process, and presents as such a more sophisticated workflow, which utilizes gravity gradiometry and high resolution magnetic data as well as advanced isostatic investigations. The modelling results contributed to more confidence in the salt imaging, especially with respect to the base of the salt diapirs, which supported volume estimation of potential hydrocarbon traps and gained a better tectonic understanding of the Southern Nordkapp Basin and its salt structures, as a part of the integration workflow with EM and seismics. Advantageous of the novel, iterative modelling process are the utilization of potential field properties, the iterative workflow and the combination of gravity and magnetic data with advanced isostatic investigations.
-
-
-
Using Geophysical Methods to Characterize a Fault Zone – A Case Study from the Møre-Trøndelag Fault Complex, Mid-Norway
Authors A. Nasuti, J. Ebbing, C. Pascal, J. F. Tonnesen and E. DalseggThe Møre-Trøndelag Fault Complex (MTFC) is one of the most prominent fault complexes in Scandinavia. Existing regional gravity and aeromagnetic data used to map the fault in large scale and delineate the main geophysical features related to MTFC. For studying the fault zone in more detail, we acquired multiple geophysical data sets across two segments of the MTFC. To detect the fault zones and their structural attributes, a series of magnetic, resistivity, shallow refraction and deep reflection seismics profiles were measured across these fault zones. Interpretation of the magnetic data shows the distinctive signature of near-vertical faults (~60°-65° towards the south), trending NNE–SSW. Quantitative interpretation of the data points to a width of 90 to 150 m for the Tjellefjorden Fault and 700 m for the Fannefjorden Fault. Inversion of 2D resistivity data reveals a three-layered subsurface until 130 m depth. The layers represent the thin low resistive topsoil underlain by weathered bedrock, and the resistive bedrock. Within the resistive bedrock distinct low resistivity zones can be observed, which can be associated with highly fractured bedrock. These low resistive zones correlate to low velocity zones in the shallow refraction profile.
-
-
-
3D Gravity and Magnetic Model of a Triassic Large Igneous Province Vent, Central Alaska Range, Alaska
Authors G. G. Connard, J. M. Glen and J. M. SchmidtThe Amphitheater Mountains synform on the southern flank of the Alaska Range in south-central Alaska provides one of the most complete sections of a Large Igneous Province worldwide, revealing, in near continuous section: basal mafic and ultramafic sill complexes, associated mafic and ultramafic rocks, lower submarine lavas, and overlying subaerial lavas. Three-dimensional (3D) modeling of gravity and magnetic data shows details of the Amphitheater Mountains structure, flood basalt thickness, and geometry of ultramafic sills. Modeling is based on a compilation of existing regional and newly-acquired high-resolution profile gravity data, a compilation of regional and high-resolution aeromagnetic surveys, and numerous rock-property data (including density, susceptibility, and magnetic remanence measurements of outcrops, hand samples, and drill cores). The final 3D model includes 12 layers with up to 2 km of Nikolai basalts in the center of the asymmetric, west-plunging, 50 km long Amphitheater synform. The western end of the synform appears to be terminated by a steep N-S structure. The outcropping Fish Lake and Tangle ultramafic sills, apparently connected beneath the surface, are traceable to 5 km depth, and modeling indicates variations in thickness (1 to 3 km), shape, and orientation which may identify paths of magma flow within the system.
-
-
-
South Atlantic Kinematics and The Evolution of Tristan da Cuhna Hotspot Tracks
Authors D. E. Bird and S. A. HallDocumenting the relationship between the formation of the Tristan da Cuhna hotspot tracks and the opening of the South Atlantic Ocean basin, particularly as continental extension ended and oceanic crust began to form, can help us understand the role of magma sources and crustal evolution. We examine seafloor spreading magnetic anomaly profiles and calculate new reconstruction poles for the South American and African plates, we then use the results of a basin-scale 3D density inversion to compare the evolution of the Rio Grande Rise and Walvis Ridge hotspot tracks from ~130 Ma to 10 Ma.
-
-
-
Generation of a New High Resolution Magnetic Database for Europe for Resource Evaluation – Progress Report
Authors K. M. U. Fletcher, J. D. Fairhead, K. Lei, A. S. K. Salem, C. Ayala and P. L. M. CabanillasThis contribution has two aims – to present a new high resolution 1 km grid magnetic field solution for the European area that includes the Atlantic continental margins and the Mediterranean and to map the regional variation in the depth to magnetic basement using the Tilt-Depth and Local wavenumber interpretation methods. Special focus is given to the magnetic field associated with the Iberian Peninsula.
-
-
-
Extended Tilt Depth – A Fast Depth Imaging Method Using Local Wavenumbers of Magnetic Data
Authors A. S. K. Salem, J. D. Fairhead and E. AboudThe Extended Tilt-depth method builds on the existing Tilt-depth method to remove one of its constraints - that of relying on the source model being a vertical contact. To achieve this, there is a need to go to the tilt of the local wavenumber, a second order derivative which is independent of structural index. The Extended Tilt-depth method is thus a new interpretation tool for the initial evaluation of high resolution aeromagnetic survey data. The method is tested using a theoretical model and a field example from the North Western Desert of Egypt. In the former, the method estimated the correct depth for the model. For the field data, The Extended Tilt-depth method successfully identifies a number of important tectonic elements known from geologic mapping and known areas of sedimentary basins. Further testing and application of this new method, using known models and field examples, is needed to fully appreciate its strengths and weaknesses.
-
-
-
Some Remarks on Gravity Reduction by Vertical Field Continuation
Authors P. L. Smilde and M. H. KriegerFully Bouguer reduced gravity data cannot be assumed to be located at the reduction level. After the effect of the topography has been removed by Bouguer and terrain reduction, performing the normal field reductions may be interpreted as a virtual movement inside the normal field down to the reduction level. But this “movement” does not take into account the effect of the mass anomaly. There are two ways to handle this issue: Either by modelling at original varying observation heights or by applying a vertical field continuation to a constant level. Advantages and disadvantages of different approaches are discussed.
-
-
-
Interactive 3D Gravity and Magnetic Modelling in IGMAS+ and the Integration in the Depth Imaging Workflow
Authors B. Lahmeyer, H. J. Götze, S. Schmidt, M. R. Alvers, C. Plonka and C. FichlerIn order to be successful in challenging depth imaging projects the integration of all available data is important. We present the interactive 3D gravity and magnetic modeling tool IGMAS+, its integration in the depth imaging workflow in Statoil and case histories from different geological settings.
-
-
-
Quantitative Integration of Geology and Geophysics for Reservoir Modelling – A Case Study
Authors M. S. Sams, I. Millar, W. Satriawan, D. Saussus and S. BhattacharyyaGeology, rock physics and geophysics are combined quantitatively in a geostatistical inversion to produce reservoir models of a tidally influenced shoreline to deltaic environment. The geological input is in the nature of structural elements, stratigraphic elements, depositional environment and prior facies probability volumes. The rock physics is introduced through depth trends, stratigraphic trends and relationships between the petrophysical and elastic rock properties. The geostatistical inversion combines this quantitative information with angle dependent seismic and well log data to produce models that are consistent with all a priori knowledge. The impact of the geological facies probability volumes versus the seismic influence on the results can be estimated by comparing models with those produced using probabilities alone and with those using seismic inversion alone. The geological facies probability volumes ensure that the resultant models are consistent with the depositional environment and the seismic data constrain the lateral variations in facies and property distribution. This modelling is applied to a case study from Vietnam where the results are used to mitigate risk in field development.
-
-
-
Characterization of Thin Turbiditic Sands in the Cantabrian Sea through Bandwidth Extension
Authors V. Zucconi, J. G. de Castilho, A. Perez Garcia, H. Poore, J. Pierce and J. H. Yu3D seismic data acquired in the Ballena permits has been reprocessed by REPSOL during 2009 using the technique known as Bandwidth Extension (Smith et al., 2008). The target of the reprocessing was to broaden the seismic bandwidth and increase the temporal resolution of the 3D data over the Xana accumulation, in order to better image the oil-bearing thin turbiditic sands which are the main reservoir of the area. As a result of the reprocessing, peak frequency of the data increased from 22 Hz to 58 Hz and the vertical resolution improved from 25 m down to 12 m. Successive seismic prestack inversion of the bandwidth extended data led to the interpretation of two separate Santonian turbidite lobes of oil-bearing sands which were drilled by the wells MCC-3 and MCC-4, and a third different Cenomanian turbiditic sand body drilled by the well MCC-2. New structural and thickness maps have been elaborated and offered a better understanding of the geology and reservoir sand distribution.
-
-
-
Geomodel Update Using 4-D Petrophysical Seismic Inversion on the Troll West Field
Authors K. Gjerding, N. Skjei, A. Norenes Haaland, I. Machecler and T. ColéouThis paper presents results from the R&D collaboration between CGGVeritas and Statoil on 4-D Petrophysical Seismic Inversion (PetroSI-4D) on the Troll field. The proposed 4-D petrophysical inversion workflow inverts simultaneously multiple angle stacks from several time-lapse seismic surveys in order to estimate the time evolution of the hydrocarbon saturation and update the geometrical and petrophysical properties of an input geomodel. Inversion results can be used directly for well planning. The 4-D seismic data on the Troll field are used to monitor production effects such as the thinning of the oil column, while the main production challenge is the optimisation of in-fill drilling locations. The main objective of our study is to improve the existing fine-scale geo-model and coarser scale flow model that host more than 10 years of production information. As permeability is not directly accessible from seismic measurements, the targets are porosity estimation in the oil leg for sand quality prediction and saturation changes.
-
-
-
Integration of Seismic Data and Uncertainties in the Facies Model – Application to the Snorre Field
Authors P. Nivlet, S. Ng, M. A. Hetle, K. M. Børset, A. B. Rustad, P. Dahle, R. Hauge and O. KolbjørnsenIn this paper, we prove the benefit of a consistent and quantitative use of 3D seismic data to constrain the facies model built on a complex reservoir, Snorre, located in the Norwegian Sea. The workflow starts with the application of Bayesian seismic inversion to 3D angle stacks, allowing the estimation of elastic parameter probability distributions over the reservoir. Uncertainty decrease with respect to the prior model is quantified for each elastic parameter. The estimated uncertainties are then used input to design a filter which is applied to well logs to propagate seismic uncertainties in the next step: Seismic facies modeling. We use therefore a non-parametric probabilistic classification algorithm, supervised by the filtered well logs and blocked facies. The resulting sand and shale probabilities are then controlled at blind wells, with an overall match with facies proportions from wells around 70%. Finally, the probabilities are used as a quantitative 3D constraint to stochastic object-based channel modeling techniques. By drawing different realizations from this model, we prove that the use of seismic data allows better focusing the spatial distribution of the channels.
-
-
-
Ocean Bottom Seismic in the Oseberg South Area
Authors J. Daniels, J. Dangerfield, P. Riste, M. Skaug and V. HaugenThe Oseberg South area of the North Sea holds substantial reserves but the development has been very difficult in some fields with very poor seismic data. The image problems were due to the presence of dolomitized sand bodies varying rapidly in thickness and position in the overburden. The solution was to use ocean bottom seismic and the results were impressive. The technique evolved rapidly with sparser acquisition reducing cost but maintaining the excellent illumination. The seismic data changed from only just interpretable to good enough to provide excellent interpretation and inversion. The next survey planned is ten times the size of the first.
-
-
-
Looking for Gas in All the Tight Places
By T. L. DavisMulticomponent seismic data are used to illustrate that sweet spots of high permeability can be identified in a tight gas sand play in western Colorado. These sweet spots coincide with low clay content, high secondary porosity and natural fractures. In addition, the same technology can be used to investigate seal integrity in the petroleum system.
-
-
-
Volcanic Rock Estimation and Uncertainty Evaluation from Surface and Crosswell Seismic Data
Authors D. Grana and C. D‘AgostoIn this paper we propose a methodology for the evaluation of the uncertainty associated to petrophysical properties estimation from seismic, analyzing different sources of uncertainty and different types of seismic data. The proposed methodology is based on a full bayesian approach to the inversion from seismic to petrophysics and it is divided into three steps: elastic inversion, probabilistic upscaling and petrophysical properties estimation by means of statistical rock physics models integration. We verify the validity of the methodology and the reliable propagation of estimated properties uncertainty using real well logs and synthetic seismic. Finally we apply the workflow to evaluate the uncertainty associated to lithological properties in an oil clastic reservoir in North Africa, using well logs and seismic data coming from different sources: surface and crosswell seismic.
-
-
-
Surface–subsurface Integration Reveals Faults in Gulf of Suez Oilfields
Authors A. Laake, M. Sheneshen, C. Strobbia, L. Velasco and A. CuttsReservoir mapping in the Gulf of Suez petroleum system is challenging because rifting broke up the reservoirs by transform and cross-faults. These faults are still active and outcrop at the surface thereby posing a risk for drilling. We have developed a methodology that integrates fault outcrop mapping, seismic near surface characterization as well as ant tracking of faults and geobody extraction on a seismic PSTM cube. The interpretation of satellite imagery provides the outcrop structural geology. Shallow seismic techniques comprise Rayleigh wave velocity mapping and rayparameter interferometry. The methodology utilizes a combination of GIS and reservoir modelling software for the surface-subsurface integration. The joint analysis of Rayleigh wave data with satellite imagery provides a near surface structural geologic model, which can be interpreted for shallow drilling risks related to fault outcrops. The suite of near surface geological products – Rayleigh wave velocity mapping, short offset rayparameter interferometry and shallow fault mapping – is enabled by the acquisition, processing and interpretation of point-receiver seismic data. For the first time detailed structural geology comprising faults and lithology changes was imaged in the near surface, a data regime that is conventionally contaminated by the seismic acquisition footprint.
-
-
-
Joint Interpretation of Multiple Passive Seismic Data Volumes
Authors A. Goertz, B. W. Artman, B. Schechinger, P. Krajewski and M. KoerbeGeologic interpretation is best realized by joint interpretation of multiple geophysical techniques. After recording a 25 station, 7.5km linear passive array for over 48h, we divide the data into four time subsets characterized by total energy content. We investigate the 4 volumes with methods where the subsets satisfy the assumptions underlying the techniques. The first volume contains high total energy associated with anthropogenic noise. We analyze it with Rayleigh wave ellipticity velocity inversion. The second volume is 45s around a M5.9 teleseismic event. This data is analyzed for site amplification indications. The third subset has low energy content from 01:00 Sunday. This data is analyzed for departures from the background Rayleigh wave solution provided by volume 1. Fourth, 10-180s intervals are selected from the Sunday morning data and depth imaged with time-reversal algorithms. Finally, we interpret the results in an integrated framework. The first two volumes characterize the regional ambient wavefield, and indicates that the shallow subsurface is largely constant across the profile. The second two volumes indicate that there is observable body wave content providing local subsurface information from depth. 1D excess energy anomalies and 2D energy focusing are consistent with each other and an oil reservoir.
-
-
-
A Low Frequency Passive Seismic Array Experiment Over an Oilfield in Abu Dhabi – Implications for Hydrocarbon Detection
Authors K. A. Berteussen and M. Y. AliA low frequency passive seismic experiment utilising an array of forty-nine 3-component broadband seismometers were conducted over an oilfield in the emirate of Abu Dhabi in the United Arab Emirates (UAE). The experiment was conducted to better understand the characteristics and origins of microtremor signals (1–10 Hz) which had been reported as occurring exclusively above several hydrocarbon reservoirs in the region. High resolution frequency-wavenumber (f-k) spectral analyses were performed on the recorded array data to determine apparent velocities and azimuths of the wavefronts for the microseism and microtremor events. The f-k analyses confirmed that the microtremor events have an azimuth pointing towards a major oil pipe line west of the study area and a motorway east of the study area, indicating that they are probably being excited by continuous movement of the oil through the pipeline and by traffic noise. No body waves with an infinite apparent velocity were observed.
-
-
-
Signal–to–noise Study on a Low–frequency Passive Seismic Survey
More LessSpectral attributes of the low-frequency (LF) ambient seismic wave field have been found to correlate with hydrocarbon (HC) reservoirs throughout the world. A major challenge today for using the LF attributes as a HC detection tool is surface wave noise. Knowing the signal-to-noise ratio (SNR) can provide insight for a better understanding of the LF phenomenon as well as a guide for noise tolerance of HC detection based on LF attributes. We analyse a noisy dataset acquired in an urban area in Germany over a known oil reservoir with constrained synthetic noise models to address the SNR question in the area. Seven SNR scenarios were modelled and these synthetic datasets were used to train neural networks for HC detection. The performance of these predictors on the real dataset was used as an objective measure to estimate the SNR present in the actual data attributes. We estimate the SNR for the field data to be greater than 0.06 but less than 1.01. Based on the synthetic data alone, we estimate the minimum SNR allowable for reliable HC detection to be 0.06 for this survey.
-
-
-
Automatic Survey Design for Passive Seismic Monitoring
Authors B. Khadhraoui and D. G. RaymerThis paper presents a new automatic method for designing optimised surveys for passive seismic monitoring. Given a number of surface and/or downhole sensor tools to be deployed, their hardware specifications, a set of possible locations that can host them, a volume to be monitored, and velocity and noise models, this algorithm returns the best receiver network according to a chosen cost function. The search method utilized is Particle Swarm Optimisation (PSO) with an adequately chosen population of initial guesses. The discrimination between network candidates is done through the joint analysis of the smallest detectable event magnitudes and estimated uncertainty of the hypocenter parameters. The method is designed to return a network that detects as many events as possible subject to the smallest estimated hypocentre parameter uncertainties. Depending on the application, the method runs from a few seconds to several hours.
-
-
-
An Approach to Analyse Microseismic Event Similarity
Authors M. Scholze, K. Stürmer, J. Kummerow and S. A. ShapiroWe introduce two parameters to analyse the similarity of close seismic events. The first is an extension of the normalised cross correlation coefficient (composite correlation measure, CM), and the second parameter depends on differential arrival times (composite differential arrival time measure, TM). TM quantifies the separation of pairs of seismic events. The relationship between TM and interevent distance Delta-r is derived, and theoretical TM-Delta-r-distributions are calculated for different source-receiver geometries. We find that a combined analysis of the three parameters CM, TM and Delta-r is practical for identification of time measurement errors and mislocations. The envelope of the CM-TM plot delineates event pairs with the same source mechanism (multiplets). We define an area around this envelope in order to account for statistical uncertainties. Therefore we develope two different approaches to obtain a proper measure of this area.
-
-
-
Anisotropic Velocity Model Inversion for Imaging the Microseismic Cloud
Authors T. Mizuno, S. Leaney and G. MichaudWe present an anisotropic velocity model inversion method that uses multiple sources at known locations to calibrate the velocity model for microseismic hydraulic fracturing monitoring (HFM). We parameterize the perturbation of the initial velocity structure using 2 parameters for the vertical P- and S-wave velocities and 3 parameters for anisotropy, reducing the number of unknown parameters and making calibration fast. This workflow is applied to actual HFM jobs where the initial velocity model is constructed from sonic logs and the calibrated model is used for event detection and location. The inversion for both vertical velocity and anisotropy for this dataset leads to an improvement in location accuracy.
-
-
-
Microseismic Sources in Anisotropic Media
Authors W. S. Leaney and C. H. ChapmanMicroseismic source mechanisms are of interest in hydraulic stimulation because of what they may reveal about the induced fracture network. Since rocks are in general anisotropic, particularly shales, it is of interest to consider the impact that anisotropy may have on microseismic sources. While the theory presented is valid for general anisotropy we consider only VTI in our examples. A second aspect of microseismic sources generated by hydraulic stimulation that needs to be considered is that of fracture opening or closing. This results in a composite source mechanism including a dipole component, so a composite mechanism is included in our analysis. We show the influence of anisotropy at the source by looking at radiation patterns. We first consider a double-couple mechanism with a vertical fault and strike-slip displacement and then an inclined fault with a dip-slip component to displacement. We then consider a composite source mechanism including a component of displacement normal to the fault plane and describe the process of defactoring the source medium from the moment tensor, allowing the correct source parameters to be recovered. We conclude that knowledge of medium properties local to the source is important to recover correctly the source parameters.
-
-
-
Modelling Frequency-dependent Anisotropy Observations Made on Microseismic Data
Authors O. Al-Harrasi and M. KendallHere we observe the dependence of anisotropy on frequency in a microseismic data set acquired from a carbonate gas field. The shear-wave splitting observations are modelled using a poroelastic model which considers fluid communication between grain size pore spaces and larger scale fractures. A grid search is performed over fracture parameters (radius, density and strike) to find the model that best fits the real data. The results show that fracture size varies from the micro-scale within the shale cap rocks, to the meter-scale within the gas reservoir, to the centimetre-scale within the non-producing part of the carbonate formation. Cumulatively, the results depict the potential for characterising fracture systems using frequency-dependent anisotropy.
-
-
-
Microseismic Moment Tensors – A Path To Understanding Growth of Injection Induced Fractures
Authors A. M. Baig, T. I. Urbancic and M. PrinceMicroseismic moment tensors allow the characterization of hydraulic fracture treatments when monitored with a multiple observation well triaxial sensor configuration. Each moment tensor is a mixture of an isotropic component corresponding to volumetric change, an double-couple part corresponding to pure shear, and a compensated linear vector dipole roughly corresponding to the opening and closure of cracks. We map these components of each mechanism and examine their behaviour in time and space. The events for one study of a hydraulic fracture feature mechanisms corresponding to the opening of cracks occurring progressively farther out from the treatment well will events behind show more crack-closure type mechanisms. Correlating this progression with treatment parameters shows how the injection of proppant can be related to the different behaviours of the mechanisms with time. We contrast this study with the moment tensors for events during a cyclic steam treatment showing events that are more characterized by events that show no indication of occurring on cracks but are instead potentially showing changes in pore volume.
-
-
-
Beyond the Dots in the Box – Microseismicity-constrained Fracture Models for Reservoir Simulation
Authors P. Duncan, S. Williams-Stroud, L. Eisner, A. Hill and M. Thorntonlocations for these events and a geometrical interpretation of these ‘dots in the box’. In this study we show how additional information obtained from observed microseismic events, namely the source mechanisms, were used to generate a discrete fracture network. Both volumetric and shear-only source mechanism inversion was carried out on microseismic events from the treatment of a shale gas reservoir in the continental US. The source mechanisms revealed fracture orientations more accurately than could be inferred from microseismic event locations alone. The activity associated with different mechanisms is interpreted as indicating reactivation of existing fractures in the rock, as well as suggesting generation of new fractures. Failure analysis using source mechanisms on individual events allows an integrated understanding of the complex fracture interactions taking place in the reservoir, and also provides a more complete understanding of the stress conditions in the reservoir during the treatment. Fracture orientations, locations, and failure mechanisms are translated into discrete fracture network (DFN) models that can be used to verify the extent and character of the fractures created or reactivated during the fracture treatment, and may ultimately be used to generate fracture flow properties for reservoir simulation.
-
-
-
Source Mechanicms in the Intense Seismicity Induced by Fluids Injection at the Soultz HDR Site, Alsace, in 2003
Authors Z. Jechumtálová, J. Horálek, L. Dorbath and J. ŠílenýWe have inverted the peak amplitudes of direct P waves of 45 micro earthquakes with magnitudes of between M = 1.4 and 2.9 which occurred during the 2003 massive fluid injection in the borehole of the Soultz site. The full moment tensor expression of the mechanism was applied. Surprisingly, the mechanisms of all are dominantly pure shears. The T-axes are fairly stable, being concentrated subhorizontally roughly in the EW direction. On the contrary, the P-axes are ill-constrained varying in the NS direction from nearly vertical to nearly horizontal, which points to heterogeneous stress in the Soultz injected volume. This is in agreement with the stress pattern from in-situ measurements. The time-space distribution of the events analysed suggests that the injection activated two segments of the natural faults existing in the area showing different source mechanism patterns. The dip-slip regime is characteristic for fault segment I where the seismicity occurred during and also after injection, while the strike-slip regime prevails in segment II where the seismicity was triggered only after the injection shut in. This indicates that the tensile fractures may have occurred on a smaller scale than the pure shear microearthqukes investigated.
-
-
-
Fault Activation During Hydraulic Fracturing
Authors S. C. Maxwell, M. Jones, R. Parker, S. Leaney, M. Mack, D. Dorvall, D. D‘Amico, J. Logel, E. Anderson and K. HammermasterMicroseismic imaging was used to image hydraulic fracturing during a gas well stimulation. Some time after the end of the injection, there was an increase in the seismic deformation rate. Investigation of the frequency-magnitude characteristics during the pumping were consistent with other hydraulic fracture results, although the activity recorded after the end of pumping was more consistent with observations of natural seismic deformation along faults. The ratio of p- to s-wave amplitudes also varied for events recorded during the pumping compared to those occurring after the end of pumping, suggesting a different failure mechanism. In this example, it appears that the hydraulic fracture induced movement on a nearby fault. Geomechanical modeling was also performed to examine induced stresses associated with the stimulation, and investigate possible fault deformation.
-
-
-
Detetection of Mechanical Failure During Hyraulic Fracturing Through Passive Seismic Microseismic Monitoring
Authors A. De La Pena, L. Eisner, M. P. Thornton and S. Williams-Stroud Ph. DMicroseismic monitoring is extensively used for detection of microseismic events induced by hydraulic fracture stimulation and detected microseismic events are assumed to be caused by fracture growth in a formation. However, not all microseismic events are created equal as some events might be caused by mechanical changes in the completion. Source mechanism analysis of the micro-earthquakes proved to be an excellent tool to discriminate between the two types of events. We show a non-shear event of stage 4 that seems to be related to a mechanical failure on the casing, as the energy radiation pattern is mostly horizontal and sub parallel to the horizontal section of the well and consistent with tensile opening. Fracture related events are well explained by double couple dip-slip mechanisms occurred at time of fracture propagation while pumping.
-
-
-
Inter Event Times of Fluid Induced Seismicity
Authors C. Langenbruch and S. A. ShapiroWe show that the temporal occurrence of seismic events induced by borehole fluid injections is a Poissonian process. We demonstrate this by analyzing the distribution of inter event times between successive events induced in six different case studies. We show that during stationary phases of seismicity the occurrence of events follows a homogeneous Poisson process, while the temporal organization of a complete seismic sequence can be described according to a non homogeneous Poisson process. These results prove the independency of fluid induced events. The underlying process leading to seismic events is hence not influenced by the occurrence of an event. Our results build a basis for upcoming seismic risk studies in geothermal and hydrocarbon industry.
-
-
-
Scaling of Fluid-induced Seismicity by Coulomb Stress Transfer Model
Authors A. Y. Rozhko and J. TronvollThe borehole fluid injection is commonly used for CO2 storage and during stimulation of hydrocarbon and geothermal reservoirs. Usually this process induces micro-earthquakes with magnitude up to 4. The spatio-temporal distribution of the induced seismicity can be used for monitoring of fluid movement in the reservoir (passive seismicity) and for characterization of physical properties of the reservoir. Here we propose the method to explain and predict the shape and the spatio-temporal distribution of the induced seismicity cloud. This method is based on calculation of propagation of critical Coulomb Failure stress perturbation, which, as demonstrated, triggers the micro-seismicity. It is shown that non-linear field observations are well explained by coupling of linear fluid diffusion into the deformation of linear poroelastic rocks. It is demonstrated that depending on borehole fluid injection conditions, the seismicity triggering fronts can be scaled with different scaling laws; from the cubic root of time to almost a time-independent function. The proposed analytical solution can also be used potentially to predict the shape of back-triggering (suppression) front due to ceasing of fluid injection and to predict the conditions for induced seismicity during drawdown of pressure at borehole.
-
-
-
Analysis of Seismicity Resulting from Time-dependent Fluid Injection Source Pressures
Authors C. Dinske and S. A. ShapiroFluid induced seismicity is often governed by linear pressure diffusion. We relate the perturbed pore pressure to the induced seismicity to study the fluid - rock interaction, to examine physical processes and to characterize the reservoir. Solutions of the diffusion equation exist for the condition of constant source strength. But in some injection experiments, such as in Basel (Switzerland), the source strength is not constant over time. Here we present the solution of the diffusion equation which considers the special problem of linear rising source strength. Using the solution, we accordingly modify already established methods for a seismicity based reservoir characterization (SBRC). These methods are based on a statistical approach and consider features of induced seismicity such as the spatial event density and the seismicity rate. We apply those methods to the Basel microseismic data. Our analysis result in consistent estimates of hydraulic properties of the stimulated reservoir. We obtain a scalar permeability of around 75 microDarcy by assuming an effective isotropic medium. Furthermore, we study the criticality field which describes the strength of preexisting fractures in the reservoir. We observe that it is bounded by a minimum criticality of approximately 2000 Pa and a maximum criticality of about 0.75 MPa.
-
-
-
Dual-sensor Streamer Acquisition and its Impact on Reservoir Characterisation Studies
Authors C. Reiser and C. RibeiroThe main objectives of this paper are to present the benefits of the Dual-Sensor Streamer acquisition from a seismic reservoir characterisation point of view. The dual sensor streamer technology has been described at length in various publications emphasizing its operational benefits as well as the improved seismic imaging. This time, we look at its impact at the reservoir level through some case studies. The main questions are: what impact this technology brings at the reservoir level and what will be the uplift for the reservoir geoscientist to work with such a dataset. The dual-sensor streamer technology provides a significant increase of the seismic bandwidth at both the low and the high end of the seismic amplitude spectrum and hence clearly adds value at the reservoir. We illustrate the resulting benefits through the use of examples from two case studies.
-
-
-
Improving Seismic Inversion through Detailed Low Frequency Model Building
Authors A. Huck, G. Quiquerez and P. de GrootThis paper describes a comparison between two absolute acoustic impedance inversion runs. The only difference between the two runs is the low frequency model that is supplied as input to the inversion algorithm to constrain the solution. Both models are constructed using three out of four available wells. In run No. 1 the model is constructed in the conventional manner: only two mapped seismic horizons are used to guide the interpolation of acoustic impedance well logs. In run No. 2 nineteen additional horizons are used to guide the interpolation algorithm in the construction of the low frequency model. The additional horizons are auto-tracked from a “steering-cube”, i.e. local seismic dip- and azimuth information. The inverted acoustic impedance results obtained in run No. 2 (21 horizons) follow the seismic events much more closely than the inverted results obtained in run No. 1 (2 horizons). Also quantitatively run No. 2 correctly retrieves the trend whereas run No. 1 under-predicts by 4%. In general it is hypothesized that in many geologic settings the following statement is true: "the higher the number of horizons used to construct the low frequency model, the higher the accuracy of the inverted results.
-
-
-
Updating Low Frequency Model
Authors D. Marquez, P. R. Mesdag, L. de Groot and V. AubinImprecise information about high contrast layers directly above the reservoir causes residual sidelobes and other artifacts within the reservoir, leading to incorrect imaging close to high contrast layers. We propose a method to update the P-impedance low frequency model in order to account for high contrast P-impedance layers. The updating methodology is integrated in an iterative inversion scheme in which, after a first pass inversion, information of a laterally variable high contrast P-impedance layers is interpreted and used to update the P-impedance low frequency model for a new run of the inversion. The result is a better imaging in zones close to high contrast layers. A case study is presented from a gas field located within the Dutch sector in the Southern North Sea. The Rotliegend reservoir in the gas field, which comprises high porosity Aeolian sandstones, lies just below the Top Rotliegend horizon, a marker that locally is the base of a hard Zechstein salt layer. The high impedance Zechstein Salt layer varies in thickness and is in places interleaved with shale. This causes its impedance response to vary rapidly within the seismic band width.
-
-
-
2D High-resolution Reservoir Characterization by Model-free Seismic Inversion with Sparsity Promotion
Authors D. Tetyukhina, L. van Vliet, K. Wapenaar and S. M. LuthiThe low vertical resolution of seismic data is the main motivation for implementing stochastic inversion because it allows the information from all available sources to be integrated to a consistent image of the reservoir. We present an inversion approach for poststack, time-migrated, seismic data for high-resolution reservoir characterization. The method is a 2D trace-based technique and uses well data as a priori constraints. The novelty of this method is that we adopt a `super-resolution' technique that favors sparse solutions. The Bayesian objective function employed in the method is modified by adding two terms – a vertical and a horizontal operator that favor a sparse inversion solution and continuity of the 2D inversion result. Specifically, the study focuses on fluvio-deltaic clinoform systems which are known to have complex internal lithofacies distributions and the potential reservoir units are often close to or below seismic resolution. The method obtains a 2D acoustic impedance image at subseismic resolution when applied to a seismic data set in the North Sea. In addition, the method has the capability to increase the vertical resolution of the resulting 2D geological model to a level desired by the user.
-
-
-
Semi-automatic Determination of the Number of Seismic Facies in Waveform Classification
Authors W. Xiong, Z. H. Wan, M. S. Chen and H. Y. ZhangWaveform classification is a popular method in seismic facies analysis. It has been successfully applied to oil and gas reservoir prediction. However, the decision of the number of seismically-determined facies in field applications is not straightforward. Generally the task is performed by trial and error of a subjective user, which consequently increases the workload of the interpreter and the uncertainty of reservoir facies prediction. This problem is addressed here by proposing a self-organizing neural network in which the structure of the output layer is one-dimensional. Using this technique, the number of seismic facies can be semi-automatically determined. This method is applied to a synthetic example and a field example. It is shown that the proposed method is capable of estimating the number of classes effectively and enhancing the discrimination of seismic facies in an efficient way.
-
-
-
Understanding Seismic Anisotropy from Fractures Observed in Wells
Authors L. Cherel, J. Bruneau, N. Dubos-Sallée, K. Labat, J. F. Barthelemy and J. M. DanielThe description of fracture networks is a critical input for the building of reservoir models and reservoir simulation. This paper presents the integration of the physical characteristics of fractures defined in wells to interpret the azimuthal anisotropy of the amplitude of seismic signals. The proposed method converts the well data into elastic tensors to compute anisotropic reflection coefficients. Synthetic traces are then created by convolving the resulting reflection logs with source wavelet. The comparison of the synthetic seismograms with real azimuthal seismic data is used to evaluate the effect of fractures on these data. It turns out that taking into account the fractures in the forward modelling has only a minor effect on the correlation between synthetic and real traces: the variations observed between azimuthal sectors are strong and inconsistent with an anisotropic behaviour. These variations are probably due to poor seismic quality or to other causes such as layering or stress effects.
-
-
-
Illuminating Linapacan Limestone Fracture Sweetspots Through a Combination of Seismic Inversion and Ant Tracking
Authors V. W. T. Kong, A. A. Morado, J. P. Micu and C. M. PangThe West Linapacan field is situated some 60 kilometers offshore Palawan Island, The Philippines. This field, of Upper Eocene to Lower Miocene age fractured limestone, was discovered in 1990, with some production during a 3 year period. The planned redevelopment of the West Linapacan field requires that a better understanding of the Linapacan limestone, in terms of reservoir fracture intensity distribution, and, along with it a need to understand the fracture orientation sets within the limestone for further well trajectory planning. 3D recently reprocessed seismic data covered the West Linapacan field. The angle sub-stack seismic were used in the simultaneous seismic inversion technique over the field, computing for multiple rock physics data cubes, such as acoustic impedance, shear impedance and density. The shear-rich data thus obtained was used as input in the Ant Tracking process to image fracture clusters and fracture azimuth sets within the limestone reservoir. The calibrated density data cube was also used to delineate the lower density areas, indicative of higher fracture zones. The two prong method showed a good convergence whereby the low density zones also showed up as more intensely fractured zones, together with the illuminated fracture azimuth sets.
-
-
-
Calibration of Pre-stack Simultaneous Impedance Inversion Using Rock Physics
Authors S. W. Singleton and R. KeirsteadThis paper is the third part in a reservoir characterization series. Its objective is to demonstrate the necessity of understanding the rock property responses of a reservoir so that the project results can correctly interpreted. The first step is to check and correct acoustic and density well log curves. For the current study a combination of Raymer for density and Greenburg-Castagna for Vs were applied in the shallow zone above the reservoir. Within the turbidite reservoir section a laminated sand fluid substitution was used to understand its behavior as fluid content varies, and a matrix substitution to understand its behavior as sand content varies. Synthetic gathers were calculated for all models using both ray traced and full waveform algorithms. These exercises showed that AVO analysis could be used to detect fluid changes in the seismic data but not for detecting sand content changes. Rock physics crossplots, however, could make this distinction. The seismic inversion was calibrated to acoustic impedance (AI), shear impedance (SI), and Poisson’s Ratio (PR) well log curves and clearly revealed that acoustic anomalies seen in this prospect were the result of sand content changes and not the result of fluid saturation changes.
-
-
-
Dual Signals Separation in Shallow Borehole Hydrophone Data
Authors L. Petronio and F. PolettoWe present synthetic and real data examples to demonstrate the applicability of the dual hydrophone separation method in shallow boreholes. As in dual streamer (over/under) offshore seismic survey, we apply the time derivative and space derivative to obtain downgoing and upgoing wavefields.
-
-
-
Optimization Approach to Automatic First Arrival Picking for 3C 3D VSP Data
By E. BliasA new method for accurate estimation of 3C 3D VSP first arrival times that analyzes all three downhole component traces is proposed. A new optimization technique includes four main steps. We first pick the first break times manually for the first gather of the 3D VSP dataset and calculate parameters that describe its wave-shape. Next we predict the first arrival time for an adjacent source point assuming that the average velocity is similar to the velocity at the first shotpoint. We then calculate an objective function for a moving trace window and minimize it with respect to time shift and slope properties. This function combines four main attributes that describe the P-wave first arrival wavelet on 3-C VSP data: linear polarization, signal/noise ratio, similarity in wave shapes for adjacent shotpoints, and their stability in a time interval after the first arrival wavelet. Finally, we average the computed first break times determined in the overlapping windows. Using an optimization approach for several traces and three components makes this method very robust even for data with low signal to noise ratios.
-
-
-
Imaging of 3D VSP Data with the Common-offset Common Reflection Surface Stack
Authors M. Spinner, F. Miranda, N. A. Mueller and P. Marchettindent imaging technique for the simulation of common-offset sections with improved signal-to-noise ratio. Applied in a continuous manner, the CO CRS can be utilized to create a “new” regularized prestack dataset which provides an improved input for a subsequent prestack depth migration. We discuss an adapted strategy for eni’s CO CRS implementation which handles 3D VSP geometries and show for the first time 3D results from a complex real data case.
-
-
-
Walkaway VSP Multi–wave Imaging over a Gas Cloud Area
More LessIt has been proven in many cases that gas cloud imaging can be improved by using a converted wave or shear wave. In the main structure of a CS gas field, the reservoir crest is difficult to image with conventional P-wave data because the gas filled overburden attenuates P-wave energy and distorts its velocity field. Previous multi-component surface exploration has indicated that the converted-wave or shear wave is more suitable for imaging sub-gas-cloud sediments, but there are some difficulties in the shear wave static correction. This paper demonstrates how to image the structure with 3-C walkaway VSP data. The results show that the P-wave, converted wave and S-wave images are all clearer than the surface section and reveal that VSP can play a more important role in gas development in this area.
-
-
-
Improved Synthetic Seismic Ties after Log Analysis – Case Study in North America
Authors A. B. Sanz Fernandez and T. BuerkertMain inputs for synthetic gather generation include: the wavelet, density, compressional-DT and shear-DT logs. The synthetic trace generated for the well location is then correlated with real seismic data. At this stage various problems arise that are usually attributed to defects in the seismic information: incorrect surface corrections, errors in time-depth conversion, sparse data, and various processing artifacts. Often, the problems with seismic ties are reverted back to the log information itself. This paper shows that log data can be a significant source of error when improperly managed. Most problems belong to: i) bad logs ii) insufficient data usually supplemented by linear interpolations. In both cases, density and Pwave irregularities, as well as the artificial contrast between the linear segment and the 'real' data, can give rise to fabricated seismic reflections. Conversely, sharp contrasts in real formation properties can be blurred in logs due to adverse hole conditions. In both cases, log substitution of data objectively identified as bad improves the well-seismic similarities. Both editions and substitutions should be based in sound log interpretations and rock physics, rather than statistical estimations per se. An example and practical solutions in a well from North America are shown.
-
-
-
Application of a Novel Through Casing Acoustic Imaging Technique to Identify Gas Migration Paths in a Salt Body
Authors T. J. Bradley, X. Tang, D. Patterson and J. SchaafA North Sea well was drilled through a thick evaporite sequence. During drilling, gas shows were observed in the halite section by the mudloggers. No openhole wireline logs were acquired over the interval of interest; however, a full-wave monopole and dipole acoustic log was acquired over the interval of interest after casing had been set. This paper describes the application of a novel near-wellbore acoustic imaging technique applied to through-casing dipole acoustic data. The acoustic imaging processing detected geologic reflector structures through the casing up to 70 feet from the borehole. Two types of reflectors with different dip angles were mapped out by the imaging. The low dip-angle reflectors are assigned to the formation beds/structures crossing the borehole. High-angle reflectors correspond to fractures and/or faults of large extent in the formation. These acoustic imaging results were then compared to the gas shows observed during drilling and showed a direct correlation to the high-angle reflectors. This technique made it possible to estimate the fracture presence away from the borehole and pathways for gas migration into the borehole.
-
-
-
Wave Field Simulation and Data Acquisition Scheme Analysis for LWD Acoustic Tool in Very Slow Formation
More LessWe have examined multipole acoustic LWD tool responses in very slow velocity formation at different frequencies employing the real axis integration method. The semblance spectra in the time domain and the dispersion characteristics of array waveform have been analyzed. These analyses have demonstrated that in the measurements of the monopole LWD system for very slow velocity formations, P-wave energy is present in the form of a more dispersive leaky P mode. The P-wave velocity can actually be evaluated from the leaky P-wave velocity at lower source frequencies.There is no interference with the formation flexural wave by the collar wave at lower frequencies, due to the very low S-wave velocity. The dipole LWD system may be used to measure the P- and S- wave velocities simultaneously for very slow formations if the source frequency is set sufficiently low to excite the leaky P wave and flexural waves as large as possible. In such cases, dispersion analysis is an effective method to distinguish the dipole leaky P and collar waves.The cut-off frequency for the collar wave moves to lower frequencies with a decrease in S-wave velocity.
-
-
-
Borehole Radar Modeling for Reservoir Monitoring Applications
Authors M. Miorali, E. Slob and R. ArtsThe use of down-hole sensors and remotely controlled valves in wells provide enormous benefits to reservoir management and oil production. We suggest borehole radar measurements as a promising technique capable of monitoring the arrival of undesired fluids in the proximity of production wells. The high radar sensitivity to contrasts in fluid saturation changes would make this technique suitable for steam chamber monitoring in Steam Assisted Gravity Drainage (SAGD) processes and water front monitoring in thin oil rim environments. We use 2D FDTD modeling to evaluate the most significant technical and reservoir constraints. Technical constraints are given by the wellbore. A high dielectric medium surrounding the EM source is necessary to avoid destructive interference with the reflections of the metal casing. The main reservoir constraint is given by the conductivity of the formation where the radar is located; high conductivity makes attenuation and phase distortion too high for wave propagation. Additional constraints are given by the complexity of the reservoir and the dynamic of the fluids. Time-lapse changes in the heterogeneity of the background formation strongly affect the retrieval of the target reflections and gradual fluid saturation changes reduce the amplitude of the reflections.
-
-
-
Improved Pore Pressure Prediction from Seismic Data
Authors H. M. Helset, M. Lüthje, I. Ojala, A. Lothe, M. Jordan, K. Berg and I. Ribland NilssenThe objective of this study is to develop improved methods for predicting pore pressure from seismic data. Basin modelling techniques have been combined with Rock Physics velocity modelling to obtain an accurate and robust prediction of pore pressures prior to drilling. The seismic wave velocities depend on porosity, pore pressure and the burial history of the sediments. Knowledge of velocity-depth trends is important when predicting pore pressure from seismic data. Information provided by basin modelling can help establish the appropriate velocity-depth relation. Shales make up over 80% of sediments and rocks in sedimentary basins. Hence, a proper description of velocities in the shaly intervals is therefore important for seismic data interpretation and for accurate estimation of the overburden properties. We have taken a model-based approach in order to make pore pressure predictions away from well-control. Rock Physics based models are used for calculating velocities at a given state of stress, mineralogy and diagenesis. The burial and diagenetic histories are used as input to the velocity model. The pressure predictions from seismic velocities have been coupled with pore pressure predictions from basin modelling. The model has been successfully applied to case studies from the North Sea.
-
-
-
Overpressure and Compaction in the Peciko Field, Kutai Basin, Indonesia
Authors A. Ramdhan and N. GoultyThe Peciko Field contains gas in multiple stacked reservoirs within a Miocene deltaic sequence. In the deeper reservoirs, gas is trapped hydrodynamically by high lateral overpressure gradients. We have analysed overpressure and compaction in this field by using wireline log, pressure, temperature, and vitrinite reflectance data. The top of overpressure is located below 3 km burial depth, below the depth range for transformation of smectite to mixed-layer illite/smectite. Density-sonic and density-resistivity crossplots for mudrocks show unloading responses within the transition zone into hard overpressure below 3.5 km depth. Vitrinite reflectance measurements indicate that the start of unloading coincides with the onset of gas generation. Moreover, mudrock density continues to increase with depth in the overpressured section. We conclude that gas generation and chemical compaction are responsible for overpressure generation, contradicting previous interpretations that disequilibrium compaction is the principal mechanism for generating overpressure in this area. The transition into hard overpressure may be identified from reversals on the sonic and resistivity log responses in the mudrocks.
-