articles

 

V.N. Troyan

This article is devoted to the 100–year anniversary of Georgy Pertashen’, a founder of Laboratory of Dynamics of Elastic Media at Leningrad State University and presents an adapted version of the first chapter of [Петрашень, Нахамкин, 1973]. This topic has been selected due to the wide interest of geophysicists in the migration problem and continuation of wave fields that has kept growing in the last 30 years. In [Петра-шень, Нахамкин, 1973] the authors for the first time gave the clear mathematical problem statement of wave field continuation and laid the grounds of many migration methods that are widely applied today. The publi-cation has been one of the engines that boosted the advance of Soviet (Russian) Theoretical Geophysics, if compared with the well-known and highly cited monograph by Сlaerbout [1976]. The one by G.I. Petrashen’, S.A. Nakhamkin had been published in 1973 and has now been a rarity, so it’s essential to attract attention of the young generation of geophysics to it. For the book is not only written in very good language, but also contains detailed explanations of the physics and possible applications for all the calculations given.

Forward and inverse wavefield continuations, introduced by G.I. Petrashen and S.A. Nahamkin, have significantly broadened the possibilities of traditional methods of seismic data processing. Multiple prediction, correction for near-surface heterogeneities, velocity-depth model construction, seismic migration, etc. in complex seismic and geological conditions require application of the wavefield continuation methods. Numerous examples presented in the paper prove the efficiency of the approach.

This paper presents and discuss an approach to seismic imaging of fracture zone using multicomponent surface data. It is based on a specific imaging procedure, which consists of a weighted summation of multicomponent multishot/multioffset data. Additionally, for more essential suppression of regular reflections footprint, spectral removal is applied. Numerical experiments with synthetic data set computed for the typical seismogeological model of Yurubcheno-Tаkhomskoye area are presented and discussed.

Multiple suppression by the parabolic Radon transform is widely used in seismic processing. If the primary and multiple reflections have different moveouts, they will appear in different regions of the transform domain. The known approach represents a derivation of a linear non-adoptive 2D filter that performs flattening and regularization of NMO corrected gathers. In cases of low signal-to-noise ratio such a method works poorly, converting noise to horizontal events.
We suggest a robust non-linear adoptive moveout based multiple suppression. In contrast to the Radon transform methods, the subtracted wavefield consists of coherent multiple events found by an artificial intelligence. The proposed method has low computational cost, the irregular offsets are not impeding. Synthetic and real data examples have verified the advantages of the developed AI technique.

This paper is an overview of the results derived from shallow multicomponent seismic (down to 200 m) within the near-surface permafrost interval. The results were obtained using 3C receivers, vertical (Z) and horizontal (Y) impact excitation, and PP-, PS-, and SH-waves. The most reliable results for the study area were generated by processing the Yy data for SH-reflections. The close similarity between SH- and PP-wave spectra, coupled with much lower (by a factor of 4 to 5) velocities, explains the high resolution of non-converted shear wave images in the shallow subsurface intervals.

An updated technology is developed to compute coherence, a broadly used seismic attribute. The new method is based on a more realistic seismic data model which allows both signa amplitudes and noise variances to vary across the traces in a sliding window. A universal approach is suggested to improve the horizontal resolving power of any method for estimating coherence. Some methodology aspects that promote the enhancement of coherence efficiency are analyzed. The suggested methods are tested and compared with conventional analogues on synthetic data sets. The new measure of coherence is used, along with seismic curvature, to study fractured zones in the Donets Basin.

Data collected along a regional seismic profile that traverses the East Barents basin are used to study the properties of the wave field Ia. This field is formed on Permian boundary where carbonate sediments give way to terrigenous. Seismic models have been created for three structural units of the geological section along the profile. Numerical modeling of wave fields for the three blocks provides evidence of the reflection amplitude variations. Wave attenuations are affected by refraction at high-velocity intrusions. The high-velocity layers screen reflections coming from below, and the respective wave effects have been studied. Wavelet analysis is applied to study the wave field spectra.

The article presents the results of monitoring, processing and interpretation of recorded data from the Verhneilimpeyskaya well N 277. The data features influenced the choice of data processing methods. The selected methods made it possible to build a geological model in presence of a powerful artificial source of noise.