Effect of rock physics modeling in impedance inversion from seismic data using convolutional neural network

2019 ◽  
Author(s):  
Vishal Das ◽  
Ahinoam Pollack ◽  
Uri Wollner ◽  
Tapan Mukerji
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Seismic Data ◽  
Rock Physics ◽  
Impedance Inversion ◽  
Rock Physics Modeling ◽  
Physics Modeling

Integrated VTI model building with seismic data, geologic information, and rock-physics modeling — Part 1: Theory and synthetic test

Geophysics ◽  
2016 ◽  
Vol 81 (5) ◽  
pp. C177-C191 ◽  
Author(s):  
Yunyue Li ◽  
Biondo Biondi ◽  
Robert Clapp ◽  
Dave Nichols
Keyword(s):  
Seismic Data ◽  
Seismic Anisotropy ◽  
Model Building ◽  
Rock Physics ◽  
Seismic Inversion ◽  
Additional Information ◽  
Synthetic Test ◽  
Rock Physics Modeling ◽  
Physics Modeling ◽  
Geologic Information

Seismic anisotropy plays an important role in structural imaging and lithologic interpretation. However, anisotropic model building is a challenging underdetermined inverse problem. It is well-understood that single component pressure wave seismic data recorded on the upper surface are insufficient to resolve a unique solution for velocity and anisotropy parameters. To overcome the limitations of seismic data, we have developed an integrated model building scheme based on Bayesian inference to consider seismic data, geologic information, and rock-physics knowledge simultaneously. We have performed the prestack seismic inversion using wave-equation migration velocity analysis (WEMVA) for vertical transverse isotropic (VTI) models. This image-space method enabled automatic geologic interpretation. We have integrated the geologic information as spatial model correlations, applied on each parameter individually. We integrate the rock-physics information as lithologic model correlations, bringing additional information, so that the parameters weakly constrained by seismic are updated as well as the strongly constrained parameters. The constraints provided by the additional information help the inversion converge faster, mitigate the ambiguities among the parameters, and yield VTI models that were consistent with the underlying geologic and lithologic assumptions. We have developed the theoretical framework for the proposed integrated WEMVA for VTI models and determined the added information contained in the regularization terms, especially the rock-physics constraints.

Seismic attribute enhancement of weak and discontinuous gas hydrate bottom-simulating reflectors in the Pegasus Basin, New Zealand

2019 ◽  
Vol 7 (3) ◽  
pp. SG11-SG22 ◽  
Author(s):  
Heather Bedle
Keyword(s):  
New Zealand ◽  
Gas Hydrates ◽  
Gas Hydrate ◽  
Seismic Data ◽  
Rock Physics ◽  
Stability Zone ◽  
Seismic Attribute ◽  
Rock Physics Modeling ◽  
Physics Modeling ◽  
Hydrate Stability

Gas hydrates in the oceanic subsurface are often difficult to image with reflection seismic data, particularly when the strata run parallel to the seafloor and in regions that lack the presence of a bottom-simulating reflector (BSR). To address and understand these imaging complications, rock-physics modeling and seismic attribute analysis are performed on modern 2D lines in the Pegasus Basin in New Zealand, where the BSR is not continuously imaged. Based on rock-physics and seismic analyses, several seismic attribute methods identify weak BSR reflections, with the far-angle stack data being particularly effective. Rock modeling results demonstrate that far-offset seismic data are critical in improving the imaging and interpretation of the base of the gas hydrate stability zone. The rock-physics modeling results are applied to the Pegasus 2009 2D data set that reveals a very weak seismic reflection at the base of the hydrates in the far-angle stack. This often-discontinuous reflection is significantly weaker in amplitude than typical BSRs associated with hydrates. These weak far-angle stack BSRs often do not appear clearly in full stack data, the most commonly interpreted seismic data type. Additional amplitude variation with angle (AVA) attribute analyses provide insight into identifying the presence of gas hydrates in regions lacking a strong BSR. Although dozens of seismic attributes were investigated for their ability to reveal weak reflections at the base of the gas hydrate stability zone, those that enhance class 2 AVA anomalies were most effective, particularly the seismic fluid factor attribute.

Joint estimation of porosity and saturation using stochastic rock-physics modeling

Geophysics ◽  
2006 ◽  
Vol 71 (5) ◽  
pp. O53-O63 ◽  
Author(s):  
Ran Bachrach
Keyword(s):  
Bayesian Estimation ◽  
Seismic Data ◽  
Rock Physics ◽  
Joint Estimation ◽  
Accuracy Comparison ◽  
Posterior Probability Density ◽  
Rock Physics Modeling ◽  
Posterior Probability Density Function ◽  
Reservoir Porosity ◽  
Physics Modeling

Sediment porosity and saturation affect bulk modulus, shear modulus, and density. Consequently, estimating hydrocarbon saturation and reservoir porosity from seismic data is a joint estimation problem: Uncertainty in porosity will lead to errors in saturation prediction, and vice versa. Porosity and saturation can be jointly estimated using stochastic rock-physics modeling and formal Bayesian estimation methodology. Knowledge of shear impedance reduces the uncertainty in porosity and thus also reduces uncertainty in saturation estimation. This study investigates joint estimation of porosity and saturation by using rock-physics, stochastic modeling, and Bayesian estimation theory to derive saturation and porosity maps of expected pay sands. In the field example, the uncertainty in porosity, quantified by the standard deviation (STD) associated with the posterior probability density function (pdf), derived from inversion of seismic data is much less than the uncertainty in the derived saturation. For a typical case, the STD associated with saturation is [Formula: see text] while porosity STD is about 1.34 porosity units given seismic-derived inversion attributes with reasonable accuracy. Comparison of these numbers with prior estimates showed that inversion of seismic data decreased the uncertainty in porosity to 15% of the prior uncertainty while saturation uncertainty was only reduced to 92% of the prior uncertainty. Although these results may vary from one location to another, the methodology is general and can be applied to other locations.

Integrated VTI model building with seismic data, geologic information, and rock-physics modeling — Part 2: Field data test

Geophysics ◽  
2016 ◽  
Vol 81 (5) ◽  
pp. C205-C218 ◽  
Author(s):  
Yunyue Li ◽  
Biondo Biondi ◽  
Robert Clapp ◽  
Dave Nichols
Keyword(s):  
Seismic Data ◽  
Model Building ◽  
Structural Information ◽  
Rock Physics ◽  
Velocity Model ◽  
Seismic Inversion ◽  
Well Logs ◽  
Velocity Model Building ◽  
Rock Physics Modeling ◽  
Physics Modeling

Velocity model building is the first step of seismic inversion and the foundation of the subsequent processing and interpretation workflow. Velocity model building from surface seismic data only becomes severely underdetermined and nonunique when more than one parameter is needed to characterize the velocity anisotropy. The traditional seismic processing workflow sequentially performs seismic velocity model building, structural imaging/interpretation, and lithologic inversion, modifying the subsurface model in each step without verifications against the previously used data. We have developed an integrated model building scheme that uses all available information: seismic data, geologic structural information, well logs, and rock-physics knowledge. We have evaluated the accuracy of the anisotropic model in the image space, in which structural information is estimated. The lithologic inversion results from well logs and the dynamic seismic information (amplitude versus angle) are also fed back to the kinematic seismic inversion via a cross-parameter covariance matrix, which is a multivariate Gaussian approximation to the numerical distribution modeled from stochastic rock-physics modeling. The procedure of building the rock-physics prior information and the improvements using these extra constraints were tested on a Gulf of Mexico data set. The inverted vertical transverse isotropic model not only better focused the seismic image, but it also satisfied the geologic and rock-physics principles.

Sign in / Sign up

Export Citation Format

Share Document