Skip Nav Destination
Filter
Filter
Filter
Filter
Filter

Search Results for
Variation of reflectivity with angle (AVA)

Update search

Filter

- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number

- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number

- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number

- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number

- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number

- Title
- Author
- Author Affiliations
- Full Text
- Abstract
- Keyword
- DOI
- ISBN
- EISBN
- ISSN
- EISSN
- Issue
- Volume
- References
- Paper Number

### NARROW

Peer Reviewed

Format

Subjects

Journal

Publisher

Conference Series

Date

Availability

1-20 of 523 Search Results for

#### Variation of reflectivity with angle (AVA)

**Follow your search**

Access your saved searches in your account

Would you like to receive an alert when new items match your search?

*Close Modal*

Sort by

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2005 SEG Annual Meeting, November 6–11, 2005

Paper Number: SEG-2005-0281

... Summary AVO (amplitude

**variation**with offset) and seismic attribute analyses often assume a flat model for the calculation of**reflection****angles**from the offset distance. If the dip is not considered, the calculated**AVA**(amplitude**variation**with**angle**) will be incorrect in steeply dipping areas...
Abstract

Summary AVO (amplitude variation with offset) and seismic attribute analyses often assume a flat model for the calculation of reflection angles from the offset distance. If the dip is not considered, the calculated AVA (amplitude variation with angle) will be incorrect in steeply dipping areas. Raytrace modeling and AVO analysis in this study indicates substantial muting of AVO attribute values for steeply dipping beds compared to more flat lying beds with similar rock properties. If dip above about 20° is present on a prospect, modeling should be done to understand its influence on AVO and other seismic attributes and the AVO analysis should include the structure. With increased dip, certain rock property contrasts produce reduced HCI and AVO signatures that could be mis-interpreted because they do not appear anomalous. This study used a velocity profile typical of the Gulf of Mexico outer shelf area. Other basins with different velocities should be modeled to determine the variations in dip effects. Important factors influencing how dip impacts AVO are the maximum offset of the seismic data, the maximum reflection angle at the reflector of interest, and the shooting direction of the survey relative to the dip direction. Further modeling is often needed to determine the AVO effect for given acquisition parameters, reflector depth, rock property contrast, and the velocity field in a prospective area. Introduction In amplitude variation with offset (AVO) analysis, the angle of reflection must be calculated to determine the actual angle of reflection at the reflecting interface for various offset distances. Then the expected amplitude variations with angle (AVA) can be calculated using Zoeppritz’s equations or by using one of the approximations (Bortfeld, Shuey, etc). The calculation of the angle of reflection in an AVA analysis is usually accomplished using a simple model with a flat reflecting interface. An example of such a model (Figure 1) includes a water bottom and increasing velocity with depth. Curved rays are overlain on the model to a common midpoint (CMP) gather, which is also a common reflection point (CRP) gather in this flat case. The reflection angle relationship to offset can be calculated from the raypaths. Recently several prospects in the Gulf of Mexico that were seismically investigated included bright spots with significant dip (up to 30°). This paper investigates the AVO effects of the assumption of a flat model when there is actually dip on the reflecting surface. Models with dips from 0 to 60° were made and raytraced. CRP gathers were analyzed to determine the changes in angle of reflection with dip for the same shooting geometry. Amplitudes were then calculated for a reservoir rock property contrast at the angles derived from the models. These amplitudes and attributes derived from them were compared to see the effect of reflector dip Theory and Methods Initial models for this study were made with dip of the study were made with dip of the reflector varying from 0 to 45° by 15° increments. The maximum offset was 19,600 feet (6 km). Figures 1 through 3 display these initial models.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2015 SEG Annual Meeting, October 18–23, 2015

Paper Number: SEG-2015-5814947

... Summary

**AVA**inversion method utilizing**angle**information to obtain elastic parameters of subsurface rock is vital to reservoir characterization. Under the assumption of blocky layered media, a novel**AVA**inversion algorithm combining spectral**reflectivity**inversion with sparse Bayesian...
Abstract

Summary AVA inversion method utilizing angle information to obtain elastic parameters of subsurface rock is vital to reservoir characterization. Under the assumption of blocky layered media, a novel AVA inversion algorithm combining spectral reflectivity inversion with sparse Bayesian learning is presented in this paper. The method retrieves sparse P-wave, S-wave and density reflectivity series by sequentially adding, deleting or re-estimating hyper-parameters, without pre-setting the number of nonzero reflectivity spikes. Moreover, the method can highlight the stratigraphic boundaries and improve vertical resolution of inverted parameters, which is helpful for the subsequent interpretation. Model data and real data examples illustrate the effectiveness of the method. Introduction Pre-stack seismic data contains more subsurface elastic property information than post-stack seismic data because of the amplitude variation with angle (AVA) feature which is related to the parameters contrasts at interface. The migrated pre-stack seismic data can be transformed to the angle domain as AVA data. The parameters such as Pwave velocity, S-wave velocity and density are then inverted, leading to an AVA inversion problem. AVA inversion in the time domain, which is commonly used, makes use of all frequency information of seismic data. However, information beyond ~100 Hz or below ~5 Hz in seismic data is redundant or relatively inaccurate. Moreover, pre-stack seismic data has a lower SNR (signal to noise ratio), thereby making AVA inversion vulnerable. Frequency-domain AVA inversion we proposed has an optional frequency range, which can flexibly utilize the dominant information, get rid of some noise and improve the computation efficiency. Sparse Bayesian Learning (SBL), which has been proposed and proven to be an effective and accurate method for regression and classification problems (Tipping 2001), is used to solve the mentioned AVA inversion problem in this paper. The application of SBL in seismic data, which was introduced by Yuan and Wang (2013), is to identify thin beds below tuning thickness, highlight stratigraphic boundaries and improve vertical resolution for post-stack seismic data. In this paper, we extend it to pre-stack AVA inversion.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2010 SEG Annual Meeting, October 17–22, 2010

Paper Number: SEG-2010-0398

... SUMMARY Certain seismic data anomalies, associated with low-Q targets, have recently been identified as instances of “absorptive

**reflectivity**”. The anelastic**reflection**coefficients underlying this type of anomaly, and their**variations**with frequency (AVF) and/or**angle**(**AVA**), represent...
Abstract

SUMMARY Certain seismic data anomalies, associated with low-Q targets, have recently been identified as instances of “absorptive reflectivity”. The anelastic reflection coefficients underlying this type of anomaly, and their variations with frequency (AVF) and/or angle (AVA), represent a potentially important source of information for determining subsurface properties of relevance to reservoir characterization. Series expansions of absorptive reflection coefficients about small parameter contrasts and incidence angles can be used to separately estimate variations in anelastic target parameters through a range of direct formulas, both linear and with nonlinear corrections. Algorithmically, it is a differencing of the reflection coefficient across frequencies that separates Q variations from variations in other parameters. This holds for both two-parameter (P-wave velocity and Q) problems and five-parameter anelastic problems, and would appear to be a general feature of direct absorptive inversion. INTRODUCTION The geophysics literature contains numerous reports of frequency dependent seismic data anomalies associated with attenuating targets. Some researchers have attributed these to the presence of a strong absorptive reflection coefficient, which, indeed, according to wave theory places a characteristic imprint on the data. This represents a potentially important source of information of direct relevance in reservoir characterization, because, as we will demonstrate, variations in the reflection coefficient associated with a plane contrast in anelastic or an-acoustic medium parameters (discussed in theory by, e.g., White, 1965; Borcherdt, 1977; Kjartansson, 1979; Krebes, 1984; Lam et al., 2004; de Hoop et al., 2005; Borcherdt, 2009) contain in principle sufficient information to separately determine relative changes in these parameters at the point of contrast. That this is true is in fact a prediction of absorptive inverse scattering theory, which, in the limit as a set of absorptive volume scatterers combine to produce a specular reflector, has been shown to make use of these absorptive reflection amplitudes (Innanen andWeglein, 2007; Innanen and Lira, 2010). In this paper we derive a range of formulas designed to exploit this information, providing direct estimates of target absorptive medium properties, both in linearized forms and with non-linear corrections, given the reflection coefficient (which must be first derived from seismic data) as input. An absorptive reflection coefficient can be analyzed mathematically by considering either its frequency variations (i.e., AVF), or its angle variations (i.e., AVA). Both have been studied: numerically, absorptionspecific reflection coefficient variability has been reported at large angles when synthetic viscoelastic data were examined at fixed frequencies (Samec and Blangy, 1992). This latter observation appears to be supported by other recent investigations and discussions (Chapman et al., 2006; Lines et al., 2008; Ren et al., 2009; Quintal et al., 2009). In this paper we will focus on AVF variations. The estimation approach we take is straightforward: we express the reflection coefficient in terms of plane-wave variables incidence angle and frequency, expand it about small parameter contrasts and incidence angles, and directly invert these series to determine the properties of the target. We treat two cases: the reflection coefficient associated with, first, a two-parameter acoustic/absorptive contrast, and, second, a five-parameter anelastic contrast.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2015 SEG Annual Meeting, October 18–23, 2015

Paper Number: SEG-2015-5899855

... offsets were recalculated to

**angles**for**AVA**-analysis. Rpp was calculated for each CMP gather as amplitude of the**reflected**wave (zero-phase signal after deconvolution) normalized to the amplitude of the direct wave.**Reflectivity**values were averaged within the corresponding Fresnel zone which...
Abstract

Summary Combination of P-wave velocity analysis and AVA-study of seismoacoustic data can provide information about Vp , Vs , bulk density ? and attenuation of the media. However, efficient AVA-analysis of ultra-high resolution dataset requires minor distortions in recorded signals. In this study, a specific field experiment with a deep-towed ultra-high resolution seismoacoustic system was conducted in order to minimize sea surface influence. Moreover, data were acquired using omnidirectional sparker source and hydrophone string to achieve accuracy in amplitude recovery. This allowed successful follow-up application of deterministic layer-stripping AVA-inversion for shallow subsurface. The approach is demonstrated on the White Sea field-data example. Introduction Seismoacoustic investigations can provide Vp/Vs ratio and bulk density ? of the media, which are related to geoengineering soil properties such as porosity, permeability, shear modulus etc. (Rechtien, 1996). Combination of conventional P-wave velocity analysis with AVA-study has been first tested for estimation of elastic properties of subbotom deposits by Riedel, Thilen, 2001. AVO/A-analysis (amplitude-variation-with-offset/angle) is based on Zoeppritz equations (Zoeppritz, 1919) that link transmitted and reflected plane wave amplitudes on the interface of two elastic half spaces (with different impedances and Vs ) with an incident angle ?. In the conventional seismic surveys (frequencies 10-100 Hz), linear approximations (Bortfeld, 1961, Aki, K. and Richards, 1980, Shuey, 1985) of the equations are widely used for reservoir characterization. However, for subbotom sediments with relatively low Vp , ? , close to zero Vs and high investigation frequencies (usually from 300 to 3000 Hz), AVA-functions are relatively flat up to the incident angle of 30 0 , so that consideration of wider range is required for sufficient study. As most of the Zoeppritz approximations break down for oblique angles larger than 30 0 , an exact solution is preferable. (Riedel, Thilen, 2001). Nevertheless, the extraction of P-wave reflectivity function Rpp(?) (seismoacoustic case) for ultra-high resolution dataset is not straightforward as the registered amplitudes are strongly affected by source and receiver directivity at high frequencies. Another problem is that conventional acquisition technique implies towing source and streamer at the optimal depth of &#955/4 (&#955 – reference wavelength) to achieve constructive inference of desired signals and a so-called ghost waves. That leads to notches in amplitude spectrum which appear at different frequencies in case of bad weather conditions (rough sea surface). These notches, in turn, make follow-up signal processing procedures (such as signature deconvolution) unstable and require rigorous adaptive deghosting techniques.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2009 SEG Annual Meeting, October 25–30, 2009

Paper Number: SEG-2009-0256

...) , with predefined thresholds.

**reflection**coefficient migration reservoir characterization international exposition seg houston 2009 accuracy alex malkin canning malkin fracture density gradient upstream oil & gas**angle**gather amplitude azimuthal**ava**analysis workflow inversion...
Abstract

Introduction Summary Azimuthal AVA workflow based on full-azimuth 3D angle gathers in depth is presented. The benefits of this approach are described and demonstrated, as well as the AVAZ inversion algorithm. An example from fractured carbonates reservoir from the North Sea is presented. Unique preprocessing and inversion workflow is discussed, illustrating the significance of appropriated data preconditioning to the reservoir detection. Final results are also analyzed using attribute crossplotting. Wide-azimuth seismic surveying is aimed to uncover and describe fractured reservoirs. The anisotropy (mainly HTI) which is a fundamental property of these fractured zones reveals itself in the form of azimuthal variations in amplitude as a function of offset and reflection angle, in addition to the moveout variations with azimuth. The conventional procedure which is currently used to perform AVAZ analysis uses the available pre-stack migrated data, mostly time migrated, organized in azimuth sectors. This data is composed of limited number of 2D gathers from different (surface) azimuths which are in turn fed into an AVAZ inversion procedure to obtain the standard attributes. The new approach for migrating multi-azimuth data developed by Koren et, al. (2008) opened new horizons for azimuthal AVA analysis. With this approach densely sampled full azimuth 3D angle gathers, in depth, are available for AVAZ analysis, providing high resolution, azimuth-angle domain information in true subsurface coordinates. The accuracy of subsurface azimuth and reflection angle estimation, as well as the reliability of amplitude preservation and recovery provided by this method makes these gathers the ideal data for AVAZ inversion. In addition, the densely sampled 3D gathers supports greatly improved accuracy and stability of the inversion process, which is very unstable when applied to multi-2D gathers of varying azimuths. Figure 1 shows an example of a migrated full azimuth 3D angle gather. This is a 3D gather unfolded into a 2D display. There are about one thousand traces in each gather, densely sampling the azimuth axis and the reflection angle axis. Theory and Method According to Ruger (1998), P-wave reflection coefficient R( ?,? ) for HTI media can be described as: where f is the azimuth angle, ? is the reflection angle, ß is the fracture orientation angle, I the intercept, G the gradient and G aniso is the anisotropic gradient. Substituting G 1 =G and G 2 = G aniso + G, we obtain We perform AVO inversion for each depth sample by fitting surfaces based on Equation (2) to the full azimuth 3D gather. We use either L1 or L2 norm, and produces traces of I, G 1 ,G 2 and ß. To linearize equation (2) we follow Grechka & Tsvankin (1998) and Jenner (2002). According to (2) G 1 is the gradient estimation across and G2 – along fracture orientation. Unfortunately, because of p periodicity of squared sine and cosine, the two solutions: ( I , G 1 ,G2, ß) and ( I , G 2 ,G1 ,ß- p/2) are completely equivalent. We select the solution which is closer to ß bkgr , a background value which we define a-priory . Orientation becomes meaningless when anisotropy is very small. We estimate this condition by comparing envelopes E( I ), E(G) and E(G aniso ) /E(G) , with predefined thresholds.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2006 SEG Annual Meeting, October 1–6, 2006

Paper Number: SEG-2006-0466

... s lower peninsula. The survey was optimized for both

**reflection**imaging purposes and the gathering of a wide range of incidence**angles**. With this we can do a highly detailed interpretation using the post stack migrated data, and utilizing the prestack data for**AVA**(amplitude**variation**with**angle**...
Abstract

SUMMARY A high-resolution crosswell seismic data set was collected over a Silurian Niagaran reef in Michigan’s lower peninsula. The survey was optimized for both reflection imaging purposes and the gathering of a wide range of incidence angles. With this we can do a highly detailed interpretation using the post stack migrated data, and utilizing the prestack data for AVA (amplitude variation with angle) analyses. With the wide range of incidence angles and by comparing AVA observations with model predictions we can better determine and map the various litho-facies and fluid distributions within a carbonate reef oil reservoir. In addition to the extremely high resolution of the crosswell images, reflections are obtained from boundaries that have near-zero reflectivity at small angles of incidence because of the large reflectivity that occurs near the critical angle. This has implications in allowing operators of carbonate reef oil reservoirs to better image the interior structure of the reservoirs and to identify those areas that may still contain amounts of oil after initial production. INTRODUCTION During the summer of 2005 a cross well seismic data set was collected over a carbonate reef, at a dedicated test site in Manistee County, Michigan. The test site offers a unique setting for study as this area has a huge library of collected data that spans over thirty years of research and study. This library contains past 2D, 3D, vertical seismic profiles (VSP), reverse vertical seismic profiles (RVSP), and an entire suite of well logging data, data that are beneficial in the interpretation and analysis of the crosswell data set. The survey was designed with two main objectives. The first was to obtain a high-resolution, high frequency reflection image of the target reef. The second objective was to obtain as wide a range of incidence angles as possible, yielding incidence angles that were only limited by borehole geometry. With these objectives we can produce highly detailed interpretations of the reef from the stacked reflection images, and also conduct unique AVA analyses due to the very wide incidence angles collected. Completion of both of these objectives will provide valuable data on the internal structure of the reef and the possible fluid distributions. GEOLOGIC SETTING The test site consists of two monitor boreholes that lie on opposite sides of, and on the flanks of, a Silurian Niagaran carbonate reef. This reef is just one of the many reefs located in the northern Michigan reef trend that stretches in a general southwest to northeast trend for approximately 170 miles and is 10-27 miles wide. The western flank well, the Burch 1-20, is an open hole below the casing that ends at 2,944 ft with a total depth of 6,780 ft. The eastern flank well, the Stech 1- 21, is fully cased to total depth of 6,500 ft, and sits much farther up the flank of the reef than the Burch 1- 20. There is also a producing oil well, the Merit 1-20, centered between the two monitor wells on the reef crest.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2010 SEG Annual Meeting, October 17–22, 2010

Paper Number: SEG-2010-3263

... is then used as a

**reflectivity**that is fed into a demigration process which creates modeled data. Remigration of such a data set then results in a gather on which any amplitude**variation**is a measure of illumination effects alone. The resulting illumination**AVA**signature can then be used to assess the effect...
Abstract

Summary The imaging of seismic data in areas of complex structure, such as subsalt, is problematic because the acoustic wavefield distorts significantly as it passes through such complexity, causing uneven illumination of subsurface reflectors. Conventional imaging methods do not correctly compensate for this uneven illumination, which distorts the amplitudes of seismic images in complex areas, and renders Amplitude versus Offset (AVO) / Amplitude versus Angle (AVA) analysis unreliable. Here we develop a workflow based on one-way wave-equation demigration-remigration for AVA de-risking in sub-salt or other geological settings where irregular illumination causes amplitude irregularity on migrated gathers. Introduction The last decade has been a period of highly successful subsalt exploration in the Gulf of Mexico, West Africa and other similar areas. As a result, numerous discoveries in these areas are making their way into development and production. One of the key aspects in the continuing development of these areas is well planning, which often must be done in complex geologic settings where seismic data is challenging. Since AVA is often used to assess the potential for well location, any irregularities in AVA response due to uneven illumination resulting from complex overburden introduces substantial risk in AVA analysis and well placement. The main goal of this paper is to develop a method that extends well known zero-offset or stacked wave-equation illumination analysis into the angle-gather domain, where it becomes an appropriate tool for assessing the effects of complex overburden on AVA response. The central idea behind such an analysis is to first create an angle gather that has a perfect AVA response (i.e. a constant amplitude as a function of angle). This gather is then used as a reflectivity that is fed into a demigration process which creates modeled data. Remigration of such a data set then results in a gather on which any amplitude variation is a measure of illumination effects alone. The resulting illumination AVA signature can then be used to assess the effect of illumination on AVA response, resulting in a useful AVA risk analysis. In this paper we develop a workflow for AVA risk assessment based on the transfer of illumination AVA signatures to known modeled AVA signatures from welllog welllog information in subsalt and near-salt areas. In the next section we describe the theory behind our method, and then proceed to an application of the method to an area of significant salt complexity in the Gulf of Mexico. Theory Illumination effects in migration generally arise because the propagation processes in a migration algorithm are an imperfect inverse to the full wave-propagation processes in the earth, even when those processes are well described by the acoustic wave equation. In such situations, modern treatments of illumination analysis for wave-equation migration are based upon a least squares approximation instead. Within such a treatment, the adjoint to migration is viewed as an approximate modeling operator, and the objective function is taken as the square sum over source, receiver, and time, of the difference between modeled and actual data.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2012 SEG Annual Meeting, November 4–9, 2012

Paper Number: SEG-2012-0161

..., and interference of

**reflections**.**Angle**-dependent RCs extracted at subsurface**reflection**points (RPs), by prestack migration, are better at handling these problems (Resnick et al., 1987; de Bruin et al., 1990). A specific example is the com- parison of amplitude**variations**with offset (AVO) and with an- gle (**AVA**...
Abstract

SUMMARY The residual moveout in an angle-domain common-image gather (ADCIG) can be used for migration velocity analysis. ADCIGs also have potential for amplitude variation with angle (AVA) analysis. However, previous AVA analyses are not very successful, especially for deeper layers, because transmission losses are not properly compensated, and phase shifts occur at near- and post-critical reflections. Phase is less affected than amplitude by transmission losses in prestack images from reverse-time migration. We consider amplitude and phase analysis of a target reflector in an ADCIG. Tests on a modified Marmousi2 elastic model show that the phase variation with angle (PVA) is better preserved than the AVA. This is significant for target-oriented inversion because distortions in the overburden are minimized in PVA. ADCIGs potentially extend AVA/PVA analysis to 2D/3D models.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2007 SEG Annual Meeting, September 23–28, 2007

Paper Number: SEG-2007-2160

... is the high computational cost. INTRODUCTION Amplitude-preserving migration aims at making the AVO/

**AVA**stud- ies for hydrocarbon detection quantitative and rigorous by reliably esti- mating**angle**-dependent**reflection**coefficients. Many factors affect the amplitude**variations**; here, we consider only...
Abstract

SUMMARY Analysis of amplitude variations with offset (AVO) or with angle (AVA) for reservoir description and for hydrocarbon detection requires reliable amplitude information be obtained from prestack migration. A 3-D true amplitude prestack reverse-time depth migration is developed by compensating the geometrical spreading, intrinsic Q losses and transmission losses. Tests on synthetic models illustrate the feasibility. One practical limitation is the high computational cost. INTRODUCTION Amplitude-preserving migration aims at making the AVO/AVA studies for hydrocarbon detection quantitative and rigorous by reliably estimating angle-dependent reflection coefficients. Many factors affect the amplitude variations; here, we consider only the propagation effects. Geometrical spreading strongly reduces the wavefield amplitudes during both down and up going propagations. Anelasticity of the earth makes wave propagation in scalar or acoustic media only an approximation to propagation in real earth, and wave energy is partitioned at interfaces where velocity, density, or Q changes. All of these factors significantly distort the wavefield amplitudes during propagation. Most current true-amplitude migrations involve only the geometrical spreading (Bleistein, 1987; Zhang et al., 2003, 2007). Some previous algorithms can also handle absorption and dispersion (Mittet et al., 1995). None has demonstrated proper treatment of the transmission loss; and, there is no complete formulation which compensates them all. 3-D prestack reverse-time migration (Chang and McMechan, 1986) is used as framework to integrate the compensation of propagation effects into one algorithm. Since reverse-time migration uses full twoway extrapolations, the geometrical spreading and its compensation are implicitly included in both up and down going waves. Intrinsic attenuation can be handled by including a Q-dependent term in the wave equation. A two pass, and recursive procedure is developed to estimate the angle-dependent reflection and transmission coefficients from the surface to the target reflector. First a brief description of the theory and the methodology is presented in the following sections. Following that, numerical tests on a 3-D dipping layer model and a 3-D salt model are presented to illustrate the feasibility. BRIEF OF TRANSMISSION COMPENSATION ALGORITHM A layer stripping scheme is adopted; and a recursive, two-pass algorithm for transmission compensation based on reverse-time migration is developed. Each pass is a modified reverse-time migration with compensation embedded. The first pass extracts the information needed for compensation; the second pass applies the compensation (Figure 1). The compensations are repeated for each major reflector, with the velocity ratio estimates updated at each reflector. All the extrapolations are done through a smooth background model; all compensations are defined and applied, based on deviations from this model. First pass During the source wavefield extrapolation in the first pass, at each computational grid point, we save the maximum amplitude over the propagation time and the corresponding time and we estimate the wavefront orientation using the constant time trajectory defined by the image time. During the receiver wavefield extrapolation, we calculate the angle-dependent reflection coefficients at the points that satisfy the imaging condition. After the first pass through all the common-source gathers, a stacked migrated image is obtained and is used to estimate the reflector orientation.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2005 SEG Annual Meeting, November 6–11, 2005

Paper Number: SEG-2005-0817

... Introduction Summary An integrated seismic interpretation project was performed using a pattern recognition toolkit to characterize the seismic signature and amplitude

**variation**with**angle**(**AVA**) response of Tertiary oil sands encountered in a control well and identify other areas...
Abstract

Introduction Summary An integrated seismic interpretation project was performed using a pattern recognition toolkit to characterize the seismic signature and amplitude variation with angle (AVA) response of Tertiary oil sands encountered in a control well and identify other areas of similar signature in a 1,265 km2 3-D survey located in the Bay of Campeche, offshore Tabasco State, Mexico. To do this, two different multi-attribute templates were designed in order to characterize the seismic response in the zone of interest. One pre-stack time migrated seismic volume and three angle stack volumes were utilized in the project. The 3-D survey interpreted in this project is located offshore Tabasco State, in the Bay of Campeche, Mexico (Figure 1). It was acquired in 2003 and covers an area approximately 23 kilometers by 60 kilometers. Pattern recognition terminology and methodology Geophysical and petrophysical properties Logs from a total of 4 wells were used in this project for calibration of log data to seismic; one well encountered pay in the objective interval (Figure 2). Even with sufficient velocity control at the main calibration well, a good character tie with the seismic data was difficult to obtain. This was primarily due to the fact that there is very little difference in impedance between sands and shales in this zone (Figure 3). Pay sands, wet sands and shales all have approximately the same impedance ranges at the main calibration well; sands have slightly higher impedance (<5%) than shales at the other locations. This indicates that the zero-offset reflectivity at the top of a sand would be very low, resulting in a low-amplitude peak or trough. We conclude that the sands in the objective interval in the 3-D volume exhibit Class 2 AVA behavior (Rutherford and Williams, 1989). On the stacked section, Class 2 sands can exhibit a range of amplitudes, from low to high. A total of four 3D volumes were provided for this project: a Pre-Stack Time Migration volume (PSTM) as well as three angle stack volumes: Low Angle (0-12 degrees); Mid Angle (12-24 degrees); and High Angle (24-36 degrees). These volumes are referred to as the “AVA” volumes (Amplitude Versus Angle). The purpose of the volumes was twofold: first, to determine the amplitude versus angle relation of the oil sands’ response at the control well and use this relation in an extraction template; and second, to high-grade the set of leads which were extracted using pattern-recognition technology according to their AVA response. Initial AVA modeling and fluid substitution indicated that the reservoir sands encountered in the control well should exhibit strong AVA response. The pattern recognition toolkit applied in this project is a process of calibration of seismic trace “fragments” (the samples between successive zero-crossings) to petrophysical logs, isolation of the seismic character (vertical and horizontal) associated with the target geology, then searching the entire data volume for similar instances (Wentland et al., 1999). In this workflow, a relational database is created of fragment-based features and patterns that occur naturally in the 3-D seismic data.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2007 SEG Annual Meeting, September 23–28, 2007

Paper Number: SEG-2007-0303

... Summary There are numerous publications about amplitude

**variation**with offset (AVO) or with incident**angle**(**AVA**) in surface seismic data and their applications in the petroleum industry. However, this extended abstract demonstrates an**AVA**phenomenon in borehole sonic images of a horizontal...
Abstract

Summary There are numerous publications about amplitude variation with offset (AVO) or with incident angle (AVA) in surface seismic data and their applications in the petroleum industry. However, this extended abstract demonstrates an AVA phenomenon in borehole sonic images of a horizontal well. With a three-transmitter and thirteen-receiver-array sonic tool, Borehole Acoustic Reflection Survey (BARS) has been acquired in two long horizontal wells in an Oman carbonate field. The data demonstrate an AVA story which may inspire future borehole seismic AVA applications. Introduction An oil field in Oman is consisting of low relief carbonate reservoir of approximate 80 m thickness with a thin oil column, 10 to 15 m thickness. The reservoir is trapped by water-reactive shale. The top reservoir is very difficult to interpret from seismic data because the reflector is weak and interfered with seismic multiples (Fig. 1). Long (1.0 to 2.5 km) horizontal producers are implemented to produce this economically marginal field. These horizontal wells are required to be placed 0.5 to 1.5 m below the top reservoir to maximize recovery and for borehole safety. Unfortunately this is very challenging even with advanced geosteering technologies because resistivity contrast is used to guide the drilling and the reservoir and seal resisitivity contrast is weak (Fig. 1). Borehole Acoustic Reflection Survey (BARS) is tested in two wells to provide extra depth control to reduce structure uncertainty, to aid nearby future well placement and to validate the result of geosteering. This abstract focuses on the AVA phenomenon at the second well. Data Acquisition The borehole seismic data were acquired post-drill using SonicScanner* tool which has three monopole transmitters (MU, ML and MF) operating at 8000 Hz, two orders of magnitude higher than a good surface seismic source. Thirteen receiver arrays are placed between MU and ML. The tool configuration results in transmitter-receiver spaces of 1-7 ft for MU and ML and 11-17 ft for MF. Each receiver array contains eight hydrophones distributed around the axis of the tool at 45 deg which significantly improves signal azimuthal discrimination from Dipole Shear Sonic Imager (DSI*). BARS was intended to be logged over 1800 m horizontal length in the well. However, only 600 m was logged due to a tool (telemetry) failure. Data for the last 200 m of the logged section are discussed below. The first 400 m of data are very noisy because the wellbore is within 1 m below the top reservoir (a good geosteering result). BARS Image Results Figure 2 shows the BARS final image result with top reservoir reflection interpreted. A vertical line at 802 m horizontal distance to the wellhead separates contributions from MF and ML. When MF and ML images are examinated separately (Fig. 3), they reveal an intriguing phenomenon: the reflection on the MF image terminates around 802 m where the reflection on the ML image starts. And the ML reflection is weaker than the MF reflection which is partially buried in headwaves. What controls the truncation and the characteristics of the reflection signal from the two different transmitters?

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2007 SEG Annual Meeting, September 23–28, 2007

Paper Number: SEG-2007-0264

... estimated through Walden relations and

**angles**computed with the kinematic de-migration.**Variation**of about 5° can be observed in a range of 40-60o. This represents significant variability for**AVA**studies. III. Offset to**Angle**transformation up to 60° incidence. IV. Common**angle**spectral harmonization...
Abstract

Summary Recent works have shown the benefit of seismic data processing in common angle domain especially for AVA studies and stack resolution enhancement. When wide angle datasets are exploited, difficulties and reservations come with the genuine signification and the validation of the reflection angle value. The workflow presented includes two original methods providing accurate angle CDP gather from time migrated offset CDP gather. The dependencies offset, angle and time are computed through a kinematic pre-stack time demigration process taking into account local dips, the effective velocity and anelipticity model. The kinematic demigration takes advantage of the existing relationships between the reflection angle and the differential stretch in time of the wavelet for different offsets. This approach gives a better angle estimation than conventional methods especially for angles larger than 40o and for dipping events. In addition to the mapping of offset, angle and time correspondences, a direct (and reverse) angle transform is defined in order to provide seismic traces regularly distributed in angle. The “angle regularization” is performed by an iterative Fourier reconstruction minimizing the spatial frequency leakage. Because the amount of stretch is nearly stationary in time (if there are no variations of dip in time) for a common angle time-migrated trace, a spectral harmonization along angles can be easily implemented by single spectral matching operators. This stretch compensation impacts on the stack quality and resolution. Introduction In reservoir characterization, wide angle seismic data give valuable information for improving lithology and fluids analysis. For example, AVO/AVA studies need incidence angle larger than 40 degrees for inverting density. Wide angle data may also provide significant improvements to the structural image by additional target illumination. Because imaging introduces about 50% wavelet stretch at 60o incidence angle, conventional time processing sequences are not appropriate for such dataset. Data recorded beyond traditional 40° limit are generally muted to avoid distortion of the signal and large anisotropy effect. Using wide angle data for density inversion, Roy et al. (2006) has shown the benefit of being in angle domain for analytically correcting the stretch of the signal. The combination of the Walden and the Dix equations are frequently used to estimate reflection angle up to 35-40°. Beyond this limit the estimation suffers from too many assumptions to be reliable. Therefore, Pèrez and Marfurt (2006) have implemented an “Angle-binned” Kirchhoff time migration that delivers angle binned migrated traces using internally computed angle values. According to the authors, this makes the migration computationally more expensive than one whose bins over offset. This paper proposes to preserve the common offset prestack time migration algorithm (used in the industry and highly optimised) and presents an approach to computing common angle volume through a combination of a specific kinematic pre-stack time de-migration and a regularization process in angle domain. Incidence angle estimation with kinematic de-migration Our approach is based on kinematic pre-stack time demigration. For any position in a gather and for a given associated dip, time de-migration makes it possible to recover the associated shot and receiver positions).

Proceedings Papers

Paper presented at the SPE/AAPG/SEG Unconventional Resources Technology Conference, July 20–22, 2015

Paper Number: URTEC-2151459-MS

.... The

**variation**of seismic**reflection**amplitude with**angle**of incidence (**AVA**) of seismic waves**reflected**from an interface between two geological layers may be used to obtain predrill information on RQ and CQ. The information from**AVA**inversion may be analyzed conveniently using rock physics templates, in which...
Abstract

Shales make up a large proportion of the rocks in most sedimentary basins, and form the seal and source rocks for many hydrocarbon reservoirs. In unconventional shale plays, the shale acts as the source rock, reservoir, and seal. Organic-rich shales represent an enormous energy resource, but production from wells in such plays shows considerable lateral variation. Economic production from such formations requires good reservoir quality (RQ), representing the multiple properties defining reservoir potential, and good completion quality (CQ), representing the multiple properties defining the potential for creating and sustaining large surface area in contact with the reservoir. Reliable predrill methods for determining the spatial variation in RQ and CQ are required to optimally locate wells in these low-permeability reservoirs. The variation of seismic reflection amplitude with angle of incidence (AVA) of seismic waves reflected from an interface between two geological layers may be used to obtain predrill information on RQ and CQ. The information from AVA inversion may be analyzed conveniently using rock physics templates, in which the results of inversion are plotted together with the predictions of rock physics models that take into account variations in RQ and CQ. In this paper, the ability of rock physics templates to distinguish between organic-rich and organic-lean shales is investigated using log data from the Eagle Ford Shale.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2005 SEG Annual Meeting, November 6–11, 2005

Paper Number: SEG-2005-0242

... The physics to explain the

**variations**of**reflected**seismic amplitudes as a function of the incidence**angle**(**AVA**) is well known. Under the assumption of two homogeneous and isotropic layers,**AVA**behavior is specified by the changes in P-wave velocities (Vp), S-wave velocities (Vs) and densities across...
Abstract

Summary In this paper, we present a method for detecting automatically CDP gathers with clear discrepancies between the observations and the model used to define the AVA attributes. The technique is presented using a linear model of AVA; however, any regression model can be used. Bootstrap and least median square (LMS) regression are the main components of the method. An application to data from the Glitne field (North Sea) is presented to illustrate the applicability of the methodology. Introduction The physics to explain the variations of reflected seismic amplitudes as a function of the incidence angle (AVA) is well known. Under the assumption of two homogeneous and isotropic layers, AVA behavior is specified by the changes in P-wave velocities (Vp), S-wave velocities (Vs) and densities (r) across the interface. Hence, AVA attributes respond to the contrast of elastic properties, a fact that inevitably brings non-uniqueness when inverting AVA for Vp, Vs and r, even when the assumptions are fully satisfied. Another approach to exploit AVA effects is to define and extract attributes from the seismic data and use them as direct hydrocarbon indicators. Although sometimes it is forgotten, this type of procedure also has a non-unique solution. Drilling based on seismic bright spots alone has resulting in many dry wells. This has motivated the need to understand the rock physics behind the amplitude anomalies as a requirement for current AVA analyses. Uncertainties in interpretation of attributes as well as in the value of the attribute itself need to be taken into account. Statistical rock physics methods (Mukerji et al, 2000) are a way to assess uncertainties in the AVA attributes to reservoir properties transformation. However, it is still unusual in AVA studies to include a systematic uncertainty estimation of the attributes values and its impact on the final predictions. Most of the AVA attributes are computed by fitting some type of regression to seismic amplitudes (e.g. Shuey, 1985; Fatti et al, 1994). Because of the unavoidable presence of noise, some criterion of error minimization has to be used to adjust a regression model to the data. Commonly, the ordinary least squares (OLS) method is used. In OLS the residual square error (RSE) is minimized. As is well known, the least median square estimator (LMS) is a more robust method for data fitting. It is based on minimizing the median squared residual (MSR). Ferré et al (1999) present some advantages of using LMS. The seismic amplitudes used in the computation of AVA attributes can have different types of problems. For example, residual NMO correction and low signal to noise ratio can generate inconsistencies between the extracted amplitudes and the assumed AVA model. It is very difficult to attempt to identify and quantify error contributions of all sources of discrepancies individually. Even more complicated is to do it automatically. On the contrary, as we shown in this paper, what is feasible is to attempt an automatic detection of data inconsistencies. That is, to identify the CDP gathers where there are clear discrepancies between the observations and the model used to define the attributes.

Proceedings Papers

Publisher: Offshore Technology Conference

Paper presented at the Offshore Technology Conference, May 6–9, 2002

Paper Number: OTC-14147-MS

... to the

**angle**volumes the seismic-wavelet effect is minimized thus reducing the likelihood of wavelet**variations**within the gathers causing false**AVAs**. Additionally inversion unravels both the amplitude shape and strength into an interval attribute as opposed to**reflectivity**that is an interface attribute...
Abstract

Abstract Elastic inversion provides an estimate of elastic parameters by the use of Amplitude Variations with Angle data. This information gives the geoscientist additional parameters for the interpretation of lithology, porosity and fluid type. By applying a stratigraphic inversion method to the angle volumes the seismic-wavelet effect is minimized thus reducing the likelihood of wavelet variations within the gathers causing false AVAs. Additionally inversion unravels both the amplitude shape and strength into an interval attribute as opposed to reflectivity that is an interface attribute. This allows a volumetric interpretation of the data producing more accurate understanding of the reservoir volume. Introduction Seismic attributes play a critical role in the efforts to reduce risk associate with reservoir prediction and description. Post stack inversion has proven to be a significant tool that allows geologic calibration and a volumetric understanding of the impedance data. This method allows the interpreter to go beyond contact event interpretation and begin understands the amplitude strength and wavelet shape and its relationship to the vertical and lateral impedance profiles. This begins the understanding of the spatial geologic parameters. Acoustic impedance alone may not always be enough to describe the geologic parameters in complex environments. Acoustic impedance is the result of the Pwave velocity and density. The Pwave velocity is a function of the mineral constituents, the manner in which they are arranged and the pore volume fluid. In cases where two of these unknowns are constant the third parameter may be interpreted. However in cases where these are not constant ambiguous results will occur. When attempting to relate geologic parameters such aslithology, porosity, and fluid type in complex environments additional parameters related to the rock physics must be utilized. To provide these attributes elastic impedance inversion is used. When these additional elastic parameters are used with acoustic impedance they offer a strong tool for increasing the quality of the interpretation of lithology, porosity, and fluid type. Fig.1 shows the crossplot relationships of common lithologies and fluid types. Fig. 1 Crossplot of P-wave velocity and V p /V s for various lithologies and fluids.(Available in full paper) Method Overview A common practice of the AVO analysis is to obtain the intercept and gradient from prestack data. These correspond to A and B terms in Shueys equation. R(?)= A + Bsin 2 ? + C sin 2 ?tan 2 ? After these two parameters are calculated they may be transformed into parameters such as P reflectivity, S reflectivity, and Poisson's reflectivity. Although valuable attributes to assist in the interpretation of fluid type this method does not address the variations in wavelet and NMO stretching as a function of offset. In a previous paper by Cambois 1 it was shown that variations in the wavelets may produce a false avo phenomenon referred to as "intercept leakage". To address the wavelet variations Connolly 2 developed a method called "elastic impedance". This method involves inverting the angle volumes for the angle dependant elastic impedance. These angle dependant elastic impedances can then be converted to P and S impedance.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the SEG International Exposition and Annual Meeting, September 15–20, 2019

Paper Number: SEG-2019-3216604

... presents oil production data as evidence of aligned (or azimuthal or fracture) porosity connected to the borehole (fracture porosity) in an onshore naturally-fractured carbonate oil reservoir. The change with azimuth of the near-

**angle**amplitudes (6-15) and the**AVA**gradient (amplitude**variation**with**angle**...
Abstract

ABSTRACT P-P reflection amplitudes vary by offset and by azimuth in the presence of one set of vertical aligned fractures. At a boundary wherein the orthorhombic axes are misaligned by 70 at a shale-carbonate boundary, simple models of azimuthal (az’l) P-P amplitudes do not apply. This paper presents oil production data as evidence of aligned (or azimuthal or fracture) porosity connected to the borehole (fracture porosity) in an onshore naturally-fractured carbonate oil reservoir. The change with azimuth of the near-angle amplitudes (6-15) and the AVA gradient (amplitude variation with angle, AVAz) have been calculated and displayed and tied to oil production data. The industry standard az’l amplitude measurement, AVAz, ties the oil calibration data moderately. The az’l variation of the near-angle amplitude data ties the oil production data quite well. Aligned porosity (fracture porosity) causes az’l P impedance. The az’l variation of the near-angle amplitudes is the response to az’l P impedance. Presentation Date: Monday, September 16, 2019 Session Start Time: 1:50 PM Presentation Time: 4:20 PM Location: 217D Presentation Type: Oral

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2017 SEG International Exposition and Annual Meeting, September 24–29, 2017

Paper Number: SEG-2017-17467683

... sensitive to offset

**variation**than**reflection****angle**in CIGs. Mahmoudian and Margrave (2009) used**angle**-domain common image gathers (ADCIGs) for**AVA**analysis and found that they uniquely define ray couples for each point in the subsurface. The purpose of this paper is to delineate and compare the advantages...
Abstract

ABSTRACT When we delineate oil and gas reservoir in complex structural or lithologic controls. AVO analysis will be applied with either common-image gathers (CIGs) or common-midpoint (CMP) gathers. Different gathers may draw distinct conclusions, thus correct gathers are vital for AVO analysis. Biondi et al.(2004) depicted the differences between CMP gathers and CIGs in the quality and accuracy of seismic imaging. Reshef (2008) discussed the sensitivity of velocity variation both in angle and offset domains and found that interval velocity analysis for prestack depth migration (PSDM) is more sensitive to offset variation than reflection angle in CIGs. Mahmoudian and Margrave (2009) used angle-domain common image gathers (ADCIGs) for AVA analysis and found that they uniquely define ray couples for each point in the subsurface. The purpose of this paper is to delineate and compare the advantages of CIGs and CMP gathers in the offset domain. Comparing velocity model, computational cost and images both for generating CIGs and CMP gathers, we find that CIGs are better than CMP when in large dip angle and more sensitive to velocity variation. While in small dip angle and flat layers, CMP gathers have almost the same image as CIGs but cost-effective in offset domain. Presentation Date: Thursday, September 28, 2017 Start Time: 9:45 AM Location: 370D Presentation Type: ORAL

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2008 SEG Annual Meeting, November 9–14, 2008

Paper Number: SEG-2008-2937

... at constant

**angles**along each trace can be directly used to invert for reservoir lithology fluid properties using various amplitude-**variation**-withangle (**AVA**) inversion methodologies. Offset-to-**angle**transformation is straightforward for isotropic models. In the presence of anisotropy, such a transformation...
Abstract

Summary Reservoir seismic applications require offset domain seismic data to be converted into angles. While it is straightforward to convert offsets to angles under the isotropic assumptions, such a transformation is complex in the presence of anisotropy. Correct interpretation of angle dependent reflection amplitudes requires offset domain data to be corrected for the non-hyperbolic anisotropic moveout and distinction between the phase and the group angles. Some of these fundamental issues relating to the angle dependent P-wave seismic reflections for transversely isotropic elastic medium with a vertical symmetry axis (VTI medium) are reviewed here with illustrations. Introduction Reservoir characterization applications require prestack seismic data in offset domain to be converted into angles. Reflection amplitudes in angle domain data representing seismic reflection information at constant angles along each trace can be directly used to invert for reservoir lithology fluid properties using various amplitude-variation-withangle (AVA) inversion methodologies. Offset-to-angle transformation is straightforward for isotropic models. In the presence of anisotropy, such a transformation is however not so straightforward. First of all, it is necessary to account for the non-hyperbolic moveout in anisotropic medium. Second, offset-to-angle transformation using the velocity field obtained from an anisotropic velocity analysis generates group angle traces and to compare them with reflection coefficients, it is necessary to compute the equivalent phase angles for those group angles. Finally, most of the phase-to-group-angle and phase-to-group-velocity transformations that are required for such analysis use weak anisotropy assumptions which aren''t strictly valid in many parts of the world where the anisotropy is strong. This work reviews some of these fundamental concepts associated with angle-domain analysis of seismic data for anisotropic elastic medium. Background Theory Assume a right-handed Cartesian coordinate system xyz with z-axis pointing vertically downward. Fig. 1 compares the P-wave phase velocity computed using the exact formula of equation (3) with the approximate formula of equation (4). Fig. 1 demonstrates that the approximate formula computes the P-wave phase velocity reasonably accurately for weak to moderate anisotropy. Phase velocity computed for all possible phase angles and azimuth in three dimensions is a surface known as the phase velocity surface. The curves shown in Fig. 1 are in fact the cross-sections of such phase velocity surfaces in the x-z plane. Since a VTI medium is azimuthally isotropic, these curves for any vertical plane at an arbitrary azimuth will be similar to the one shown in Fig. 1. Fig.3 compares the group angle for various phase angles using the exact and weak-anisotropic approximation for Taylor Sandstone and Green River Shale whose phase velocities are shown in Fig.1. Notice that even though the approximate phase velocities match closely with the exact phase velocities, approximate group angle calculations do not match as good with the exact group angles even when the anisotropy is weak (Fig 3a). Also notice that group angles can differ by as much as 10° from the phase angle for weak anisotropy (Fig 3a). For moderate anisotropy, the difference can be as high as 30° (Fig 3b).

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2010 SEG Annual Meeting, October 17–22, 2010

Paper Number: SEG-2010-0409

... porous media model 1

**ava****variation**velocity dispersion phase**angle****reflection**coefficient incident**angle****reflection**magnitude increase reservoir characterization upstream oil & gas reservoir reservoir model**variation**dispersion Frequency-dependent**AVA****variations**in thinly layered...
Abstract

Summary Amplitude-Versus-Offset (AVO) technology has successfully helped to detect hydrocarbon reservoir for more than two decades. However, the Zoeppritz equation only considers the elastic properties of the media, the nonelastic behaviors are ignored. There are still some problems that the traditional AVO technology doesn’t handle adequately. Although the frequency-dependent AVO technology has been brought forward, a theory is lacking to guide it. Based on White’s patchy saturation model, we have investigated characteristics of the frequency dependent Amplitude Versus incident-Angle (AVA) at an interface between a non-dispersive medium and a patchysaturated dispersive medium. And then, numerical modeling based on Biot’s poroelastic wave theory was performed on three selected reservoir models. The numerical modeling results confirmed our analytical analysis. These variations could provide insight for frequency-dependent AVO analysis. Introduction For more than two decades, with the quick development in seismic exploration, AVO technology has achieved remarkable advancement and been extensively implemented in oil industry. However, the Zoeppritz equation only considers the elastic properties of the rocks. The non-elastic properties, such as velocity dispersion and attenuation, are ignored. There are still some problems that the traditional AVO technology doesn’t handle adequately. For years, geophysicists have noticed low-frequency seismic anomalies associated with hydrocarbon reservoirs (Taner et al., 1979), and this topic is gaining more and more attention (Goloshubin et al., 2000; Castagna et al., 2003; Korneev et al., 2004; Chapman et al.,2006). Therefore, we should consider the effects of dispersion and attenuation on traditional AVO anomalies. Although some researchers have done some significant attempts on the frequency-dependent AVO analysis (Yoo et al., 2005; Marmalyevskyy et al., 2006; Chapman et al.,2006; Odebeatu et al., 2006; Liu et al., 2006), a theory is still lacking to guide it. Based on patchy-saturated model, we try to investigate characteristics of the angle-dependent reflection coefficient as a function of frequency at an interface between a non-dispersive medium and a patchysaturated dispersive medium and expect to provide some insights for frequency-dependent AVO analysis. Frequency-dependent Amplitude versus Incident Angle To systematically investigate dispersion effects on the magnitude and phase angle of angle-dependent reflection coefficients, we select three reservoir models that represent three types of reservoirs commonly encountered in oil exploration. The overburden shale, rock-frame, and porefluid properties of models 1, 2 and 3 are same as Ren et al. 2009. For each model, the reservoir consists of 1-m thick layers with the same rock frame, but brine-saturated layers alternate with gas-saturated layers. Although the stratified model might not be realistic physically, it does represent the attenuation associated with White’s patchy-saturation model (Dutta and Seriff, 1979). Moreover, the stratified layering simplified the numerical modeling. For model 1, the reservoir is consolidated sand, and the porosity and permeability are small. In the amplitude versus incident-angle domain, Figure 2a shows that the reflection magnitude decreases with increasing incident angle, which agrees with traditional Class I AVO response. In the amplitude versus frequency domain, Figure 2a shows that when the incident-angle is less than 30º, the reflection magnitude increases toward higher frequencies.

Proceedings Papers

Publisher: Society of Exploration Geophysicists

Paper presented at the 2008 SEG Annual Meeting, November 9–14, 2008

Paper Number: SEG-2008-2402

...

**variations**of the velocity model. The attributes may be used at various steps of the processing sequence: stretch factor to design de- stretching operators,**reflection****angle**to design partial**angle**stacks for AVO/**AVA**studies and kinematic invariants to perform**reflection**tomography. Introduction Pre-stack...
Abstract

Summary In many areas time imaging still represents the majority of seismic imaging activity in the industry. In this context prestack time migration remains a central process. We propose a new approach for estimating various kinematic attributes associated with pre-stack time migrated results. Our approach is based on a kinematic demigration of locally coherent events characterized by their position and dip in common offset time migrated images. The reflection angle, the instantaneous velocity or the geological dip can be estimated with the assumption that time imaging exactly positions laterally the events. In addition, the stretch factor and the "kinematic invariants" (the demigrated facet in the unmigrated domain!) can be recovered accurately independently of the positioning accuracy of time imaging. As opposed to many conventional approaches, the estimation is directly based on the travel time curves used in pre-stack time migration with no assumption regarding the travel time curves, the dip or the lateral variations of the velocity model. The attributes may be used at various steps of the processing sequence: stretch factor to design destretching operators, reflection angle to design partial angle stacks for AVO/AVA studies and kinematic invariants to perform reflection tomography. Introduction Pre-stack time migration (PreSTM) still represents the majority of seismic imaging activity in the industry. The reason for this is the speed and robustness of time imaging and its ability to focus seismic reflectors for most geological settings. Various kinematic attributes have been introduced for the analysis of PreSTM gathers. For example, the reflection angle can be used to produce angle stacks, or the instantaneous velocity can be used for deriving an approximate interval velocity model. However in most studies the estimation of these attributes is based on simplified approximations, as for example in Walden (1991) who assumes a layered model and Dix relations for reflection angle determination. In the domain of depth migration many methods have been proposed for deriving relevant and accurate attributes for the analysis of the migrated images. Bleistein (1987) for example proposed a technique of "migration of attributes" to estimate accurately various kinematic attributes by a technique of double migration. As shown by Guillaume et al., (2001) or Chauris et al. (2002) an efficient kinematic demigration process can be also used to obtain these attributes (for example those necessary for tomography). In this paper we extend this later approach to PreSTM and propose in addition some kinematic attributes specific to time imaging, such as instantaneous velocity and geological dip. First we present kinematic time demigration based on gradients of travel times and then we show how to derive kinematic attributes useful for the analysis of PreSTM results. Computation of kinematic attributes The "kinematic invariants" are the first kinematic attributes that we have derived from a time migrated image (Lambaré et al., 2007b), but some additional kinematic attributes may also be valuable. The key information for deriving all these attributes is the "gradient" of travel times (Figure 2), which provides information on the illumination pattern at the imaging point.

Advertisement