The most popular seismic attributes fall into broad categories - those that are sensitive to lateral changes in waveform and structure, such as coherence and curvature, and those that are sensitive to lithology and fluid properties - such as AVO and impedance inversion. Unfortunately neither of these two attribute families works well in differentiating depositional packages characterized by subtle changes in stratigraphic column or the lateral changes in texture. Automatic seismic facies analysis aims to classify similar seismic traces based on amplitude, phase, frequency and other seismic attributes. This research shows how this technique helps to visualize the variation in rock properties and the possible relationship between these rock types and their seismic expression. There are many different supervised and unsupervised clustering algorithms. In supervised training, the clusters are predefined and patterns are assigned to it in subsequent trainings. Unsupervised clustering is data driven without any a priori information. The self-organizing map (SOM) (Kohonen, 2001) is one of the most effective unsupervised pattern recognition techniques, and is generally used for the automatic identification of seismic facies. We describe a SOM implementation that begins by over-defining the number of initial clusters from the input dataset. These "prototype vectors" are organized using Kohonen SOM neighborhood training rule. These clusters are projected onto a 2D or 3D latent space and are displayed against the continuous 2D or 3D colorbar. Then these clusters are compared back to the input seismic traces and colors are assigned to these seismic traces. The colored output map of seismic facies is then defined by the interpreter's eye and brain, in conjunction with well and other geophysical attributes.

/pdf/automatic-seismic-facies-classification-with-kohonen-self-4ix2tnjyte.pdf

Automatic Seismic Facies Classification with Kohonen Self Organizing Maps - a Tutorial

Current schemes for removing near-surface effects in seismic data processing use either static corrections or wave-equation datuming (WED). In the presence of rough topography and strong lateral velocity variations in the near surface, the WED scheme is the only option available. However, the traditional procedure of WED downward continues the sources and receivers in different domains. A new wave-equation global-datuming method is based on the double-square-root operator, implementing the wavefield continuation in a single domain following the survey sinking concept. This method has fewer approximations and therefore is more robust and convenient than the traditional WED methods. This method is compared with the traditional methods using a synthetic data example.

Wave-equation global datuming based on the double square root operator

Adaptive acquisition footprint suppression based on a 3D stationary wavelet transform: A case study from China

The term acquisition footprint is commonly used to define patterns in seismic time and horizon slices that are closely correlated to the acquisition geometry. Seismic attributes often exacerbate footprint artifacts and may pose pitfalls to the less experienced interpreter. Although removal of the acquisition footprint is the focus of considerable research, the sources of such artifact acquisition footprint are less commonly discussed or illustrated. Based on real data examples, we have hypothesized possible causes of footprint occurrence and created them through synthetic prestack modeling. Then, we processed these models using the same workflows used for the real data. Computation of geometric attributes from the migrated synthetics found the same footprint artifacts as the real data. These models showed that acquisition footprint could be caused by residual ground roll, inaccurate velocities, and far-offset migration stretch. With this understanding, we have examined the real seismic data volume...

Pitfalls in seismic processing: An application of seismic modeling to investigate acquisition footprint

Summary Joint inversion of different geophysical datasets is an effective way to eliminate non-uniqueness in geophysical inversion problems. In this paper, we focus on a case study of joint inversion of seismic traveltime and gravity observations. The results are encouraging and we can have confidence that, comparing traveltime tomography alone, joint inversion of seismic traveltime and gravity data improved resolution, reduced velocity biases, formed the basis for successful statics solutions, and imaged geologic structure. The final 3D velocity model provided sufficient resolution and accuracy as the input for successful prestack depth-migration velocity analyses.

3D joint inversion of seismic traveltime and gravity data: a case study

We present comparisons of some of the available algorithms for acquisition footprint suppression. Fourier transform based Kx-Ky filtering is very efficient for noise removal, but can also remove geologic features. New wavelet transforms based techniques show promising results that can overcome these disadvantages. They allow for decomposition of the seismic signal on different scales and filtering of only those that have noise. The combination of 1D decomposition (wavelet transform) and 2D filtering (FFT or wavelet transform) gives the best results but it is computationally intensive. We use attribute analysis to evaluate properties of filtering techniques and notice improvement in both lineament resolution and noise removal.

Comparison of Some Algorithms For Acquisition Footprint Suppression And Their Effect On Attribute Analysis

Due to permit restrictions, crooked 2D and irregular 3D seismic reflection data were acquired within a 330 km area in central coastal California to image geologic structure and faults from the near-surface to depth of > 10 km. Key steps to achieve successful depth imaging included, nonstationary coherent noise modeling and removal, 3D processing of all 2D and 3D data, 3D tomography, residual statics using first-arrivals, and adaptive fast beam migration. No sonic log velocity data were available so multiple stages of 3D velocity inversion including joint velocity-density inversion with gravity data were used to develop velocity models and statics for prestack-depthmigration velocity analyses. Improved signal-to-noise in the 3D volumes provided constraints to design wavenumber spectral-balancing and wavelet transform denoising parameters to improve imaging in the 2D data.

Depth imaging with crooked 2D and irregular 3D seismic data in rugged terrain

Tomostatics Using Deformable Layer Tomography

Summary Simultaneous sources and continuous recording have been widely used recently in seismic exploration to improve acquisition efficiency and coverage density. Many works have been done on the simultaneous sourcing side, using deblending to recover data as if it were recorded in a traditional way from the blended data, with deblending performed in a domain where the signal and blended noise are relatively easy to identify and separate. However, we found that little has been done on the continuous recording side to take full advantage of it. Obvious as it is that continuous recording will give us unlimited data in the temporal direction, we would show that sufficient data extension is necessary for accurate deblending. High-order multiples (up to 4th order) and deep events must be modelled to have a reliable water bottom, primary signals and multiples. With extended data, it is more challenging to do deblending. We found it is more stable and accurate to use sub-L1 norm for regularization in 3D FK-domain inversion-based deblending using the Iterative Shrinkage-Thresholding Algorithm. 

Sufficient Data Extension and Sub-L1 Norm Regularization in 3D Inversion-Based Deblending

PreviousNext No AccessSecond International Meeting for Applied Geoscience & EnergyMonument Butte 3D — Reprocessing case study in Uinta BasinAuthors: Mitch PrestonXinxiang LiSeth BermanPhillip KeithChristopher EggerJosef HeimNebojsa PralicaDavid BrookesJames BeckMitch PrestonTGSSearch for more papers by this author, Xinxiang LiTGSSearch for more papers by this author, Seth BermanOvintivSearch for more papers by this author, Phillip KeithOvintivSearch for more papers by this author, Christopher EggerTGSSearch for more papers by this author, Josef HeimTGSSearch for more papers by this author, Nebojsa PralicaTGSSearch for more papers by this author, David BrookesTGSSearch for more papers by this author, and James BeckTGSSearch for more papers by this authorhttps://doi.org/10.1190/image2022-3751001.1 SectionsAboutPDF/ePub ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InRedditEmail AbstractA 3D land seismic survey was processed using amplitude and azimuth-friendly workflow. The sequence includes multiple iterations of surface consistent amplitude balancing and surface consistent deconvolution, along with the attenuation of various types of noise in several domains at different processing stages. The resulting final PSTM gathers and stack volumes provide uplift compared to previous processing results. These improvements include broader bandwidths in the low and high end of the spectrum, enhanced structural feature imaging, and improved S/N. Furthermore, using 5D interpolation, data was regularized in offset and azimuth domains allowing surface-related multiple reflections to be identified and successfully removed. The azimuth-separated PSTM gathers preserve the azimuthal variations in the data. The azimuthal velocity variation analysis (VVAz) on the PSTM gathers successfully captured stress anisotropy caused by local and regional features. The anisotropy results corroborate drilling-induced tensile fractures and borehole breakouts on image logs.Keywords: processing, anisotropy, multiples, land, bandwidthPermalink: https://doi.org/10.1190/image2022-3751001.1FiguresReferencesRelatedDetails Second International Meeting for Applied Geoscience & EnergyISSN (print):1052-3812 ISSN (online):1949-4645Copyright: 2022 Pages: 3694 publication data© 2022 Published in electronic format with permission by the Society of Exploration Geophysicists and the American Association of Petroleum GeologistsPublisher:Society of Exploration Geophysicists HistoryPublished Online: 15 Aug 2022 CITATION INFORMATION Mitch Preston, Xinxiang Li, Seth Berman, Phillip Keith, Christopher Egger, Josef Heim, Nebojsa Pralica, David Brookes, and James Beck, (2022), "Monument Butte 3D — Reprocessing case study in Uinta Basin," SEG Technical Program Expanded Abstracts : 2665-2669. https://doi.org/10.1190/image2022-3751001.1 Plain-Language Summary KeywordsprocessinganisotropymultipleslandbandwidthPDF DownloadLoading ... 

Nebojsa Pralica

Papers

Comparison of Some Algorithms For Acquisition Footprint Suppression And Their Effect On Attribute Analysis

Depth imaging with crooked 2D and irregular 3D seismic data in rugged terrain

Tomostatics Using Deformable Layer Tomography

Sufficient Data Extension and Sub-L1 Norm Regularization in 3D Inversion-Based Deblending

Monument Butte 3D — Reprocessing case study in Uinta Basin