- Home
- A-Z Publications
- Geophysical Prospecting
- Previous Issues
- Volume 70, Issue 2, 2022
Geophysical Prospecting - Volume 70, Issue 2, 2022
Volume 70, Issue 2, 2022
-
-
Elastic anisotropy of the Marcellus Shale
More LessABSTRACTThe Marcellus Shale is one of the largest shale gas resources in the world and is anisotropic due to fine layering and the partial alignment of anisotropic clay minerals and organic matter with the bedding. This anisotropy can be approximated as Vertically Transversely Isotropic (transversely isotropic with a vertical axis of rotational symmetry) with five independent density‐normalized elastic stiffnesses A11, A13, A33, A55 and A66 in the two‐index notation and with axis of rotational symmetry along x3. Compressional and dipole shear‐wave data acquired in a horizontal well allows estimation of A11, A55 and A66, while the same data in a vertical pilot well allows estimation of A33 and A55. The ratio of vertically propagating P‐ to S‐velocity is and has a quadratic dependence on in the Upper Marcellus with minimum occurring for the largest volume fraction of kerogen. This relation allows estimation of A33 along a lateral well using measured values of A55 in the well. Comparison of Thomsen's anisotropy parameter ε is found to be mostly greater than anisotropy parameter γ. Estimating A13 as the average of A33 – 2A55 and A11 – 2A66 as proposed recently by Yan and Vernik allows Thomsen's anisotropy parameter δ and a parameter K0 that relates the horizontal and vertical effective stresses to be estimated. The results are expected to help in estimating horizontal stress needed for design of hydraulic fractures, and in interpreting seismic amplitude variation with offset required for reservoir characterization.
-
-
-
Deblending and merging of 3D multi‐sweep seismic blended data
Authors Woodon Jeong, Constantinos Tsingas and Mohammed S. AlmubarakABSTRACTSeismic blended source acquisition, also referred to as simultaneous source acquisition, is a cost‐effective technology that achieves a significant reduction in acquisition cycle time and increases seismic crew field productivity. The dispersed source array is a blended acquisition field technique that simultaneously employs sources emitting different types of sweeps (i.e. multi‐sweep), in terms of frequency bandwidth and length, which ultimately will result in a full broadband seismic data. In this paper, deblending of 3D multi‐sweep seismic blended data and the subsequent merging of the data volumes having different frequency bandwidths will be discussed. In specific data domains where the signal component is coherent, interference shots (i.e. blending noise) are randomly distributed in the data space according to its own shot firing time. Therefore, the deblending process, which separates interference shots from a signal component, becomes a noise attenuation problem. A sparse inversion methodology is applied in the frequency–wavenumber–wavenumber (f–kx–ky) domain to attenuate blending noise. By applying this deblending methodology to both dispersed source array's low‐ and mid‐high‐frequency bandwidths, we obtained high‐quality deblending results. For both frequency bandwidths of the deblended dispersed source array data, additional effort was made to combine the two datasets to a single broadband data volume. Consequently, deblending and merging of the dispersed source array blended data generated a broadband, deblended and well‐balanced seismic volume suitable for further processing and reservoir characterization applications.
-
-
-
Unsupervised dual learning for seismic data denoising in the absence of labelled data
Authors Yu Xing Zhao, Yue Li and Ning WuABSTRACTThe training and updating of the supervised deep learning model are both dependent on labelled data. Unfortunately, the data‐labelling process is usually expensive and time consuming, which makes obtaining labelled data a huge challenge, especially in the field of geophysics. Because of the above‐mentioned factors, some seismic data denoising methods based on supervised deep learning use synthetic seismic data for network training. Since synthetic seismic data are more or less different from real seismic data, the denoising results may have some amount of residual noise and some false signals when dealing with real seismic data. In response to the above problems, this paper introduces an unpaired domain‐to‐domain translation method based on the framework of two‐way generative adversarial networks, which does not need to use labelled data for training, and effectively solves the problem of lacking labelled data in the seismic data denoising tasks. We use an unpaired mixed training set containing synthetic seismic data and real seismic data to train the network, which effectively improves the denoising ability of the network for the real seismic data. The experimental results show that compared with the state‐of‐the‐art denoising methods such as denoising convolutional neural network, the proposed method can suppress the random noise and ground roll more thoroughly, and the denoising results basically would have no false signals.
-
-
-
Attenuation compensation for wavefield‐separation‐based least‐squares reverse time migration in viscoelastic media
Authors Wei Zhang and Jinghuai GaoABSTRACTElastic least‐squares reverse time migration can image the multicomponent seismic recordings. However, on the one hand, without considering the intrinsic attenuation of the subsurface, it may produce blurred reflectivity images with incorrect positions of reflectors for seismic recordings with strong attenuation. On the other hand, the crosstalk artefacts created by the different wave modes may severely degrade the imaging quality. To alleviate these crosstalk artefacts and compensate for the attenuation effects, we have developed a time‐domain wavefield‐separation‐based least‐squares reverse time migration approach in viscoelastic media, which utilizes the viscoelastic wave equation on the basis of the standard linear solid model to simulate intrinsic subsurface attenuation. The key point of the proposed approach is that we utilize the separated gradient contribution of PP‐ and PS‐wave modes based on the wavefield separation technique in viscoelastic media to construct the P‐wave and S‐wave velocity images, respectively. Unlike the conventional wavefield‐separation‐based least‐squares reverse time migration approach, which generally uses the new stress–velocity equations to formulate, our proposed wavefield separation scheme fully depends on the conventional viscoelastic wave equation and its adjoint wave equation. By introducing the pure P‐wave stress in the forward and adjoint wavefields, the coupled wavefields can be well decomposed, which requires much less computational costs than the wavenumber‐domain wavefield separation scheme. Numerical examples using the layered and Marmousi‐2 models have shown that the proposed approach can improve the migration image quality under geological conditions with strong attenuation where elastic least‐squares reverse time migration may produce blurred and unfocused events. Meanwhile, the proposed least‐squares reverse time migration approach with the wavefield separation scheme has a better convergence rate and produces fewer crosstalk artefacts than that without the wavefield separation scheme.
-
-
-
First break picking with deep learning – evaluation of network architectures
Authors Paul Zwartjes and Jewoo YooABSTRACTIn recent years, various convolutional neural network architectures have been proposed for first break picking. In this paper, we compare the standard auto‐encoder and U‐net architectures as well as versions enhanced with ResNet style skip connections. The U‐net appears to have become the standard network for segmentation, judging from the number of published articles. Still, there is some variety in neural network architectural choices. In this paper, we assess the impact of neural network depth, width and input data size, as well as some small modifications for deep networks offered by the ResNet. In general, results improve as the networks get deeper, but with diminishing returns. The more complex the data, the more benefit the deeper networks bring. We use complete shot gathers, albeit rescaled for efficiency, to train the neural networks. For shot gathers with a simple piecewise linear moveout, this approach yields results with good accuracy when gathers are resampled to 128 × 128 samples. For shot gathers with more complex first break moveout, using our approach it is advised to stay close to the original dimension of each gather for best accuracy, at the expense of increased training times. A good trade‐off between network depth, image size and training times is to use a nine‐stage U‐net with 256 sample images. Despite the advantages in other applications, the basic U‐net outperforms a U‐net with ResNet features. We show that changing the input data dimensions for trained networks does not work, despite the fact the fully convolutional networks are independent of image size. The U‐net based first break picking is not sensitive to picking errors, as in many cases the neural network predictions are better than the training data where the training data have random mispicks. This suggests a practical application; namely, to train or re‐train a pre‐trained network on a single data set after conventional first break picking with the objective of improving conventionally picked first breaks.
-
-
-
Iterative Gaussian mixture model and multi‐channel attributes for arrival picking in extremely noisy environments
Authors Hang Wang and Yangkang ChenABSTRACTArrival picking is a traditional and important topic in the field of seismic exploration. Depending on the quality of seismic data, the picking results are prone to large errors. As commonly used arrival‐picking strategies, conventional clustering methods, such as K‐means, suffer from insufficient flexibility, for example the clustering boundaries are fixed given a set of attribute vectors. In order to overcome this drawback, we propose a new strategy which can provide flexible clustering results (i.e. the variable clustering boundaries) based on an iterative Gaussian mixture model and utilize the local correlation among adjacent traces to enhance the anti‐noise ability. First, we use local principal component analysis to roughly distinguish the areas of signal waveforms. Then, two multi‐channel attributes are calculated and input to the iterative Gaussian mixture model. These two attributes can correctly identify the arrivals and avoid redundant computation caused by the high‐dimensional attribute vectors. The final step is to establish an optimized Gaussian mixture model and to iteratively select the proper boundaries for each trace. Although the iterative Gaussian mixture model takes a longer calculation time, the synthetic and real data tests have shown superior results over conventional methods even in situations of extremely low signal‐to‐noise ratios.
-
-
-
Low‐rank seismic data reconstruction and denoising by CUR matrix decompositions
Authors Quézia Cavalcante and Milton J. PorsaniABSTRACTLow‐rank reconstruction methods assume that noiseless and complete seismic data can be represented as low‐rank matrices or tensors. Therefore, denoising and recovery of missing traces require a reduced‐rank approximation of the data matrix/tensor. To calculate such approximation, we explore the CUR matrix decompositions, which use actual columns and rows of the data matrix, instead of the costly singular vectors derived from singular value decomposition. By allowing oversampling columns and rows, CUR decompositions obviate the need for the exact rank. We evaluate three different procedures for randomly selecting columns and rows to obtain the CUR. Once the low‐rank approximation is estimated, data reconstruction is achieved by an iterative optimization scheme. To demonstrate the effectiveness of CUR matrix decompositions for multidimensional seismic data recovery, we present examples of 3D and 4D synthetic and field data. Results derived by CUR compare well to conventional eigenimage‐family methods.
-
-
-
Frequency‐dependent signal‐to‐noise ratio effect of distributed acoustic sensing vertical seismic profile acquisition
Authors Nour Alzamil, Weichang Li, Hua‐Wei Zhou and Harold MerryABSTRACTDistributed acoustic sensing provides seismic sensors distributed every few metres over each fibre optic cable of several kilometres in length, serving as an ultra‐dense array for both surface and downhole seismic acquisitions. It is well‐recognized now that distributed acoustic sensing acquisition configuration can have profound effect on data quality, notably the choice of the gauge length and lead‐in cable length on the resulting signal to noise ratio. In this paper, we present a detailed study on how these effects behave differently over various signalling frequency bands. We derive the signal to noise ratio as a function of the peak frequency, gauge length and lead‐in cable length, and demonstrate significantly different 80 SNR samples behaviours over low, middle and high frequency bandwidths, respectively. These results, not only reveal the underlying mechanism for signal and noise variation as function of acquisition parameters, but also help identifying the desired gauge and cable lengths that better serve particular applications, such as inversion and imaging which take information from different frequency bands. A field data has been used to demonstrate these frequency‐dependent behaviours.
-
-
-
Estimation of porosity and gas hydrate saturation by inverting 2D seismic data using very fast simulated Annealing in the Krishna Godavari offshore basin, India
Authors Anju K Joshi and Maheswar OjhaABSTRACTGas hydrate saturation and porosity are the two essential parameters for characterizing a gas hydrate reservoir. Generally, porosities determined at the well locations are interpolated and extrapolated over the seismic volume, which is not so appropriate to get an estimate of gas hydrate saturation. Here, we propose a method to predict both porosity and gas hydrate saturation directly by inverting seismic data using a global optimization technique known as Very Fast Simulated Annealing. Acoustic impedance, as a function of water saturation and porosity, is defined by a second‐degree polynomial equation used as a forward problem. These are used as the primary model parameters. Acoustic impedance derived from these values is then used to compute synthetic seismograms that are compared against observed traces. We demonstrate our approach on a 23.5 km two‐dimensional seismic line crossing four wells in the Krishna Godavari Basin, eastern Indian offshore. First, the proposed method is tested on well log data and then applied on post‐stack seismic data. The posterior probability density function and the correlation matrix between the model parameters (gas hydrate saturation and porosity) are calculated to measure the associated uncertainty in prediction. Inversion results illustrate that the approach can be efficiently used for the prediction of porosity and gas hydrate saturation directly from the seismic data.
-
-
-
Conductivity in partially saturated porous media described by porosity, electrolyte saturation and saturation‐dependent tortuosity and constriction factor
Authors Carl Fredrik Berg, W. David Kennedy and David C. HerrickABSTRACTModelling the relationships among bulk conductivity, electrolyte conductivity, porosity and fractional electrolyte‐saturation continues to be based on empirical relationships such as Archie's law or the Bruggeman correlation. Several authors have attempted to connect such empirical models to first principles. This article presents a complete first principles‐based description of conductivity in partially saturated porous media. A geometrical factor is introduced which, together with porosity and saturation, describes the conductivity of partially saturated porous media. This geometrical factor is separated into two intrinsic geometrical descriptors of the electrolyte distribution accounting for tortuosity and constriction. This tortuosity and constriction factor are obtained from integration over the electrical field lines traversing the electrolyte. Bulk and saturation‐dependent conductance descriptions are illustrated using three‐dimensional pore space models and simulated fluid distributions. Describing conductance through porosity, saturation and a geometrical factor, decomposed into separate terms accounting for tortuosity and constrictivity, permits more insightful understanding of conductance in partially saturated porous media. Use and efficacy of this full conductance description is illustrated throughout this article.
-
-
-
Evidential data integration to produce porphyry Cu prospectivity map, using a combination of knowledge and data‐driven methods
Authors Shokouh Riahi, Abbas Bahroudi, Maysam Abedi, Soheila Aslani and David R. LentzABSTRACTProducing an accurate and valid mineral prospectivity map is one of the most significant parts of mineral exploration studies. For this purpose, it is needed to obtain valid evidential layers and integrate them with an accurate methodology. Knowledge and data‐driven methods are two primary techniques applied to combine various evidential layers for mineral prospectivity mapping, of which each of them includes a variety of analytical techniques. In this study, in the first step, satellite data, aeromagnetic and airborne radiometric data, stream sediment geochemical data and geological data were applied to create valid remote sensing, geophysical, geochemical, lineaments and lithological evidential layers of the study area that are an essential factor in recognition porphyry copper mineralization, then in the second step, based on the known mineralization occurrences data, the evidential layers were weighted. Finally, these layers were integrated using fuzzy logic and index overlay methods in a combination of knowledge and data‐driven way. Validation of each layer was done using available data in the second step. The final mineral prospectivity map was evaluated, and the confirmation of this layer detected that the final mineral prospectivity map obtained from data‐driven multi‐index overlay method has a higher ore prediction rate of 76%, which identifies 24% of the area as potential zones for further exploration.
-
Volumes & issues
-
Volume 72 (2023 - 2024)
-
Volume 71 (2022 - 2023)
-
Volume 70 (2021 - 2022)
-
Volume 69 (2021)
-
Volume 68 (2020)
-
Volume 67 (2019)
-
Volume 66 (2018)
-
Volume 65 (2017)
-
Volume 64 (2015 - 2016)
-
Volume 63 (2015)
-
Volume 62 (2014)
-
Volume 61 (2013)
-
Volume 60 (2012)
-
Volume 59 (2011)
-
Volume 58 (2010)
-
Volume 57 (2009)
-
Volume 56 (2008)
-
Volume 55 (2007)
-
Volume 54 (2006)
-
Volume 53 (2005)
-
Volume 52 (2004)
-
Volume 51 (2003)
-
Volume 50 (2002)
-
Volume 49 (2001)
-
Volume 48 (2000)
-
Volume 47 (1999)
-
Volume 46 (1998)
-
Volume 45 (1997)
-
Volume 44 (1996)
-
Volume 43 (1995)
-
Volume 42 (1994)
-
Volume 41 (1993)
-
Volume 40 (1992)
-
Volume 39 (1991)
-
Volume 38 (1990)
-
Volume 37 (1989)
-
Volume 36 (1988)
-
Volume 35 (1987)
-
Volume 34 (1986)
-
Volume 33 (1985)
-
Volume 32 (1984)
-
Volume 31 (1983)
-
Volume 30 (1982)
-
Volume 29 (1981)
-
Volume 28 (1980)
-
Volume 27 (1979)
-
Volume 26 (1978)
-
Volume 25 (1977)
-
Volume 24 (1976)
-
Volume 23 (1975)
-
Volume 22 (1974)
-
Volume 21 (1973)
-
Volume 20 (1972)
-
Volume 19 (1971)
-
Volume 18 (1970)
-
Volume 17 (1969)
-
Volume 16 (1968)
-
Volume 15 (1967)
-
Volume 14 (1966)
-
Volume 13 (1965)
-
Volume 12 (1964)
-
Volume 11 (1963)
-
Volume 10 (1962)
-
Volume 9 (1961)
-
Volume 8 (1960)
-
Volume 7 (1959)
-
Volume 6 (1958)
-
Volume 5 (1957)
-
Volume 4 (1956)
-
Volume 3 (1955)
-
Volume 2 (1954)
-
Volume 1 (1953)