Numerical study of the imaging capabilities of the pre-Jurassic complex of the Tomsk and Novosibirsk regions

Background and aim. The complexity of the structures of the Paleozoic deposits of Western Siberia requires the use of specialized methods for seismic data processing. However, the standard time processing procedures are still used in Western Siberia. Therefore, in this work, the goal is to study of seismic processing procedures for the construction of high-quality images of the pre-Jurassic complex in Western Siberia. Materials and methods. A comparative analysis of time and depth processing was carried out in the paper on realistic synthetic data and models from Western Siberia containing the pre-Jurassic complex. Numerical examples are calculated for synthetic data obtained from two realistic seismic models. To create the first model, various geological and geophysical data from the Tomsk region are used. The most difficult areas of the Paleozoic in this model are steeply dipping carbonate structures and intrusive formations with steep slopes and outcropping to the erosion surface. Another model was built based on the seismic data processing results in the area of the Maloichskoye and Verkh-Tarskoye fields in the Novosibirsk region. Based on these data, the main horizons and a system of sub-vertical faults, characteristic of the pre-Jurassic deposits of the Novosibirsk region, were identified. Seismic data processing was carried out with an emphasis on the possibility of object-oriented migration. Results. It is shown that the time processing of seismic data is insufficient and the need for deep processing to construct kinematically correct images of pre-Jurassic deposits. We also compared migration algorithms based on Gaussian beams and found that object-oriented migration gives the best quality results.

Prospects for using the reflected electromagnetic wave-CDP method to search for small geological oil-and-gas prospective objects

The publication presents the key results of testing the reflected electromagnetic waves method at one of the fields of the Yamal-Nenets Autonomous District, including a description of the field research methodology, the principal aspects of the use of sounding equipment for subsurface studying based on the GROT-12 series deep ground-penetrating radars, as well as comparative examples of successful experience in the electromagnetic field processing using software designed for seismic data processing. For the first time in the world practice, the depth of the study averaging 500 - 550 meters, was achieved and confirmed by the speed characteristics of the medium.

Geophysical Aspects of Reservoir Characterization of Tight Gas Play in the Dnieper-Donets Basin

The objective of this work is to describe a comprehensive approach integrating seismic data processing and sets of wireline logs for reservoir characterization of one of the tight gas plays of the Dnieper-Donets basin. This paper intends to discuss a case study from seismic data processing, integrating seismic attributes with formation properties from logs in a geocellular model for sweet spot selection and risk analysis. The workflow during the project included the following steps.Seismic data 3D processing, including 5D interpolation and PSTM migration.Interpretation of limited log data from 4 exploration and appraisal wells.Seismic interpretation and inversion.Building a static model of the field.Recommendations for drilling locations.Evaluation of the drilled well to verify input parameters of the initial model. The static model integrated all available subsurface data and used inverted seismic attributes calibrated to the available logs to constrain the property modelling. Then various deterministic and stochastic approaches were used for facies modeling and estimation of gas-in-place volume. Integrating all the available data provides insights for better understating the reservoir distribution and provided recommendations for drilling locations. Based on the combination of the geocellular model, seismic attributes and seismic inversion results, the operator drilled an exploration well. The modern set of petrophysical logs acquired in the recently drilled well enforced prior knowledge and delivered a robust picture of the tight gas reservoir. The results from the drilled well matched predicted formation properties very closely, which added confidence in the technical approach applied in this study and similar studies that followed later. It is the fork in the road moment for the Dnieper-Donetsk basin with huge tight gas potential in the region that inspires for exploration of other prospects and plays. A synergy of analytical methods with a combination of seismic processing, geomodeling, and reservoir characterization approaches allowed accurate selection of the drilling targets with minimum risk of "dry hole" that has been vindicated by successful drilling outcome in a new exploration well.

Effective velocities artifacts caused by floating datum and compensation methods

Seismic data processing from a floating datum is accompanied by difficulties in estimating effective velocities. These difficulties are associated with the roughness of the datum surface, which, if ignored, leads to artifacts in the estimated effective velocities. The study presents the results of a quantitative analysis of the distortion of effective velocities with model data, as well as the technique to minimize the distorting effect of the elevation on effective velocities. The essence of the method is bringing the sources and receivers within one CDP to a local constant level. This approach has been tested on modeled and real data. It showed a significant reduction in the effect of floating level roughness on kinematic parameters. At the same time, there is no need to modify the processing flow whatsoever.

Passive seismic data processing methods to identify contrast intrasalt interlayers in the geological section of the Astrakhan Arch

The article examines the problem of processing microseismic noise (MN) to identify and evaluate occurrence depth of contrasting geological objects – intersalt interlayers with a potentially high formation pressure. If it is impossible to use artificial wave sources, statistic processing of passive seismic data becomes critical. Due to the accumulation of power spectral density (PSD) during a long-term recording of MN the deterministic medium effect on a random signal spectrum is identified. PSD modulation when the surface or the bottom of the layered medium is exposed to white noise is expressed in terms of the Green’s function (GF) of a wave equation. Relevant GF variations corresponding to the layers form the basis for accumulated PSD approximation, and indicate the depth and contrast of the target features.

Nonstatinary deconvolutive Radon transform

Different versions of the Radon transform (RT) are widely used in seismic data processing tofocus the recorded seismic events. Multiple separation, data interpolation, and noise attenuationare some of RT applications in seismic processing work-flows. Unfortunately, the conventional RTmethods cannot focus the events perfectly in the RT domain. This problem arises due to theblurring effects of the source wavelet and the nonstationary nature of the seismic data. Sometimes,the distortion results in a big difference between the original data and its inverse transform. Wepropose a nonstationary deconvolutive RT to handle these two issues. Our proposed algorithm takesadvantage of a nonstationary convolution technique. that builds on the concept of block convolutionand the overlap method, where the convolution operation is defined separately for overlapping blocks.Therefore, it allows the Radon basis function to take arbitrary shapes in time and space directions. Inaddition, we introduce a nonstationary wavelet estimation method to determine time-space-varyingwavelets. The wavelets and the Radon panel are estimated simultaneously and in an alternative way.Numerical examples demonstrate that our nonstationary deconvolutive RT method can significantlyimprove the sparsity of Radon panels. Hence, the inverse RT does not suffer from the distortioncaused by the unfocused seismic events.

Integrated Geo Cloud Solution for Seismic Data Processing

Oil industries generate an enormous volume of digitized data (e.g., seismic data) as a part of their seismic study and move it to the cloud for downstream applications. Moving massive data into the cloud can pose many challenges, especially to Commercial-off-the-shelf geoscience applications as they require very high compute and disk throughput. This paper proposes a digital transformation framework for efficient seismic data processing and storage comprising of: (a) Novel Data storage options, (b) Cloud-based HPC framework for efficient seismic data processing, and (c) MD5 hash calculation using the MapReduce pattern with Hadoop clusters. Azure cloud platform is used to validate the proposed framework and compare it with the existing process. Experimental results show a significant improvement in execution time, throughput, efficiency, and cost. The proposed framework can be used in any domain which deals with extensive data requiring high compute and throughput.