Search

Summary In this paper, we present an anisotropic AVO inversion method for the VTI media based on an improved form of Ruger’s equation. The AVO intercept, gradient and anisotropic P-wave velocity are inverted simultaneously. Synthetic data test demonstrates that this method provide accurate result of anisotropic P-wave velocity and more accurate AVO gradient than the isotropic inversion. We apply this method on a real seismic data; the anisotropy parameter e can be estimated according to the inverted anisotropic P-wave velocity.

Traditional shear-wave splitting analysis mainly focuses on velocity anisotropy, which is poorly understood in detecting oil/water. The main objective of this project is to evaluate the ability to discriminate oil and water from shear-wave splitting (SWS) amplitude. Anisotropic-viscoelastic model was established from Luojia area in Shengli oil field, fast and slow shear-wave were recorded using HTI viscoelastic equation. Both ratio and subtraction differences of SWS show the relation between SWS amplitude and fluid type. A detailed cross analyses were given to evaluate oil-waver boundary using SWS amplitude differences. The connecting-well seismic profiles suggest the possible use of SWS amplitude to discriminate between oil and water in fractured reservoirs.

Oil-water discrimination is of great significance to design and adjust development project for oilfields. For fractured reservoir, based on anisotropic S-wave splitting information, it be-comes possible to effectively solve such problem that the traditional longitudinal wave explo-ration is difficult to deal well with due to the similar bulk modulus and density of these two fluids. In this abstract, by analyzing the anisotropic character of HTI medium, reflection coef-ficient differences between fast and slow S-wave caused by fluid substitution were verified. Then through wave field response analysis of the three-layer fractured model, we found that water saturation causes lower amplitude difference between fast and slow S-wave, while the oil case is corresponding to higher amplitude difference. Therefore, a new class attribute was proposed, that is, the amplitude energy of the fast and slow shear wave, used for oil-water distinction. This new attribute was applied to the 3D3C seismic data of carbonate fractured reservoirs in the Luojia area of Shengli Oilfield in China. It shows that the prediction result of energy attribute has a good consistency with well information, and hence demonstrates the great advantages and potential of this new attribute in oil-water recognition.

The energy of micro-seismic data is weak and the signal-to-noise ratio is also very low. However, the quality of the monitoring data has a significant impact on effective signal recognition in Micro-seismic. It is one of the key factors in improving the signal-to-noise ratio when processing micro-seismic data. Due to high frequency, short duration, and low energy of the micro-seismic events, the effective signal can be overwhelmed by the random noise. So it is essential to preprocess the data before determining the location of the source, the origin time, and the strength of the source. But conventional filtering methods may not be fully qualified to eliminate noise and improve SNR. In this paper, we propose a method to remove the noise of the micro-seismic data, which is mainly based on using differences in geometrical features between the effective signal and the random noise in the reconstructed phase space to separate one from the other, so that we can acquire a better and more accurate effective signal.

During the exploration and development of shale gas, brittleness is now recognized as a crucial important parameter to evaluate the effect of hydraulic fracturing. We deduce a series of angle related functions to obtain information with all angles of shale layer. These functions are applied to six synthetic shale samples with different cementation. The synthetic samples are created by a new construction method, through which the synthetic shale samples in question are now with realistic mineral composition, lamination structure, low porosity and permeability, cementation and pressure sensitivity. To simulate natural shale with different cementation degrees under diagenetic conditions, we made 6 samples each with a different cement weight percentage: 20%, 17.5%, 15%, 12.5%, 10%, and 7.5%, named A1 to A6. The results indicate that the unconsolidated shale with less cementation shows stronger anisotropy and lower Young’s modulus. In the direction perpendicular to the lamination, the Poisson’s ratios are not sensitive to cementation. Yet in the parallel direction, with lamination structure developed to a certain extent, radial deformation is affected and Poisson’s ratio decreases. Cementation and fracture effects have a positive correlation. The anellipticity of the brittleness curve relates to the anisotropy parameter δ.

We use a new method to construction synthetic sandstones containing a controlled fracture distribution and geometry. The new construction method provides synthetic rock which has similar mineral composition, cementation, porous morphology as natural rocks. We use discs of decomposable material to create fractures in rock samples. The synthetic rock samples are more realistic to natural rock, compared to any synthetic experimental samples in previous experiments about fractures. Sets of synthetic samples with controlled fracture geometry and orientation are measured with ultrasonic investigation system when saturated by air and water, and both P wave velocity and S wave velocity are measured for anisotropy analysis.

This paper studies the influence of shear-wave splitting on the azimuthal behaviour of PS converted waves in HTI media. Theoretical analysis and synthetic study show that it is more accurate to separate the fast P-SV1 component from the slow P-SV2 component before compensating for azimuthal anisotropy, especially in water-saturated fractures. NMO corrections to the P-SV1 component in dry and water-saturated models can be improved by the application of the velocity ellipse.

The microseismic method is a crucial technology for locating the fracture location in the hydraulic fracturing process. Conventionally, the velocity model is usually constructed by well logs, seismic data or calibration shots which ignore the influence of fracturing domain. In this paper, we study the real-time influence of hydraulic fracturing on microseismic propagation. Based on the fluid seepage equation, fracture mechanics and the critical pressure criterion, we simulate the 3D hydraulic fracturing process and obtain the microseismic events and pore pressure distribution. Then the velocity model is constructed with the Coates-Schoenberg method and fracture compliances. The 3D ray tracing method is applied to model the microseismic travel time and direction. Simulation results show that, the deviation caused by the fracturing domain varies considerably from different receiver locations, the overall deviation increases with fracturing progress, and the deviation distribution of travel time and direction are quite similar.

Exploration and exploitation of shale gas need better understanding of rock physics relations between reservoir parameters and elastic properties. Kerogen content plays a significant role in elastic properties due to its low density, and hydrocarbon-filled pores in organic matter generated during different maturity levels will enhance the effect of kerogen in shales. In this study, we develop a shale rock physics model by incorporating Kuster and Toksoz theory and the self-consistent approximation method to quantify the effects of such factors on elastic properties in shales. Modeling results show that both kerogen content and kerogen-related porosity decrease velocities and density of shales, but kerogen content tends to affect elastic properties more profoundly, and the effect of kerogen-related porosity on the elastic properties only becomes more obvious as kerogen content goes up beyond about 0.1. We also find that the increase in kerogen content increases Poisson's ratio, while the variation in kerogen-related porosity has little effect on Poisson's ratio. Finally, for the reflector model designed in this study, the variations in kerogen content and kerogen-related porosity result in significant and predictable variations in AVO intercept and gradient.

For multicomponent seismic data joint processing and interpretation, event registration of C-waves in the P-wave time domain is a critical step, which is one of the “bottlenecks” that limits further application of the multicomponent data. In order to solve this problem, we use the Needleman-Wunsch (NW) algorithm, a nonlinear global optimization method developed for amino acid sequence alignment in proteins, to align C-wave events. The objective function is to maximize the waveform similarity to compensate for the different reflectivities of P-wave and C-wave. Both numerical modeling and field data examples are used to illustrate the methodology and its efficiency and robustness for event registration.

The presence of vertical aligned fractures in the Bakken Formation results in azimuthal seismic AVO responses. In this paper, we design a realistic geologic model of the fracture zone in the Bakken Formation, and investigate corresponding azimuthal seismic AVO responses. In the model, based on results from formation images, we assume that fractures and cracks cut through all seven major geologic units in the Bakken Formation with a specific ratio of fracture intensity about 3:3:2:1:2:1:1. Corresponding anisotropy parameters are calculated based on Hudson's theory. We then employ the reflectivity method to generate elastic seismograms. AVO curves picked for four interested geologic units indicate azimuthal variations in amplitudes especially at far offset. The decrease in amplitude at far offset corresponds to the increase in azimuth, and the AVO gradient also presents a linear variation with azimuth. However, because picking of the exact amplitudes is difficult due to interference and the presence of overlying fractures, it may be appropriate to treat the formation as an integrated part and investigate azimuth variations in RMS amplitudes. Comparison results indicate that the stacked RMS amplitudes act as a better indicator of the variations in crack density and fluid saturation than the corresponding AVO gradient.

Reflections from a single thin bed or a stack of thin layers are distinct from reflections associated with a single interface. In this study, we apply the propagator matrix approach to calculate frequency-dependent reflection coefficients resulting from layered structures, and propose a model-based scheme for the simultaneous inversion of thickness and material properties of the structured layers. We demonstrate the method using a single thin bed model for the inversion of porosity and thickness of the thin layer. We parameterize the thickness and porosity by empirical rock physics models and define error functions by the mean square roots of the difference between the predicted and observed AVO responses. The purpose is to find the values of thickness and porosity by minimizing the designed error function. The method produces inversion results with reasonable accuracy. We then apply the approach to the Bakken Formation for the characterization of the fracture zone and associated thickness. The self-consistent approximation rock physics method is employed for the parameterization of the crack density which is ready to be inverted, and the inversion results present reasonable agreements between inverted and real values of crack density and associated thickness of the fracture zone in the Bakken Formation.

Microstructure characterized by pore aspect ratio is a critical parameter which reveals the connectivity of fluid-filled cracks and pores and therefore has a strong influence on permeability in shales. Pore aspect ratio also has a great effect on elastic properties. We use the self-consistent approximation method to build a shale rock physics model for the inversion of pore aspect ratio. Inverted results show that the value of aspect ratio presents much less variety in the Barnett Shale compared to that in surrounding carbonates, which may imply the presence of more complex systems of fractures and cracks in the surroundings. We then investigate the correlations between geomechanical properties and microstructure parameters, and find that the increase in crack density decreases Young's modulus and shear modulus in surrounding carbonates but such correlation is not obvious in the Barnett Shale. In addition, the effect of crack density on Poisson's ratio is weak in all the relevant formations. Another work of this study is to predict S-wave velocity by using the inverted aspect ratio as a constraint. The prediction results indicate good agreements between predicted and real S-wave velocity for all three tight formations.

Conventional elastic/ray impedance approximations are derived as a scalar to heuristically present seismic reflectivity based on the assumptions of weak impedance, isotropic media, or weak anisotropic media. In this paper, we derive the exact elastic impedance tensors of qP- and qS-waves for orthorhombic media based on the stress-velocity law. They represent the unique mechanical properties of the medium, thus are called elastic mechanical impedance (EMI). The impedance tensor can be reduced for an isotropic medium with a vertical symmetry axis (VTI) or a horizontal axis (HTI). Because no assumptions are made during the derivation, the new impedance shows good accuracy at large angle and can be used to characterize unconventional reservoirs with strong anisotropy, such as shale gas and coal-bed methane reservoirs. An approximation of P-wave EMI is also discussed for a TI medium. It is expressed by vertical velocities and thus is applicable to seismic inversion and interpretation. Application tests on real log data and seismic data show the robust interpretation capability of EMI in lithology characterization compared with conventional impedances.

We develop a technique for inversion of fracture properties from azimuthal AVO. The method is based on the decomposition and reconstruction of the amplitude changes with azimuth in terms of principal components, which are calculated via singular value decomposition (SVD). We present the method and illustrate its application to surface seismic with a synthetic example of multi-azimuth common-offset gathers with different fracture densities.

Understanding pore-pressure and saturation changes is important in mature reservoirs. Here we analyze shear-wave splitting in a 3D3C onshore survey from Shengli Oilfield, China, where the thin sand-reservoir has been undergone production through water-flooding which altered the fluid composition and the pore-fluid pressure. Dividing the data into orthogonal azimuthal sectors and processing each sector separately reveals significant shear-wave splitting. The amount of shear-wave splitting can be correlated with the degree of water saturation. Furthermore, the slow shear-wave component shows amplitude dimming in water-flooded areas, whereas the zone of original oil in place shows only weak shear-wave splitting. Rock physics modeling based on the evolution of microcracked rocks and anisotropic fluid substitution incorporating both saturation and pressure changes confirm the observations. The saturation changes have little effect on the P and the fast shear-wave as confirmed by core analysis in the laboratory. However, the substitution of water for oil changes the fluid viscosity that has a strong effect on the slow (quasi) shear-wave. Moreover, the fluid substitution due to water flooding also changes the pore-fluid pressure that modifies the crack aspect ratios, further enhancing shear-wave splitting. These observations reveal the potential of using shear-wave splitting for oil-water discrimination.