• Geophysics and Geophysical Exploration
• /
• v.27 no.3
• /
• pp.171-180
• /
• 2024

The processing of seismic data involves analyzing earthquake wave data to understand the internal structure and characteristics of the Earth, which requires high computational power. Recently, machine learning (ML) techniques have been introduced to address these challenges and have been utilized in various tasks such as noise reduction and velocity model construction. However, most studies have focused on specific seismic data processing tasks, limiting the full utilization of similar features and structures inherent in the datasets. In this study, we compared the efficacy of using receiver-wise time-series data ("receiver array") and synchronized receiver signals ("time array") from shotgathers for pretraining a Bidirectional Encoder Representations from Transformers (BERT) model. To this end, shotgather data generated from a synthetic model containing faults was used to perform noise reduction, velocity prediction, and fault detection tasks. In the task of random noise reduction, both the receiver and time arrays showed good performance. However, for tasks requiring the identification of spatial distributions, such as velocity estimation and fault detection, the results from the time array were superior.

• Geophysics and Geophysical Exploration
• /
• v.19 no.4
• /
• pp.220-227
• /
• 2016

Marine seismic data have not only primary signals from subsurface but also ghost signals reflected from the sea surface. The ghost decreases temporal resolution of seismic data because it attenuates specific frequency components. For eliminating the ghost signals effectively, the exact ghost delaytimes and reflection coefficients are required. Because of undulation of the sea surface and vertical movements of airguns and streamers, the ghost delaytime varies spatially and randomly while acquiring seismic data. The reflection coefficient is a function of frequency, incidence angle of plane-wave and the sea state. In order to estimate the proper ghost delaytimes considering these characteristics, we compared the ghost delaytimes estimated with L-1 norm, L-2 norm and kurtosis of the deghosted trace and its autocorrelation on synthetic data. L-1 norm of autocorrelation showed a minimal error and the reflection coefficient was calculated using Kirchhoff approximation equation which can handle the effect of wave height. We applied the estimated ghost delaytimes and the calculated reflection coefficients to remove the source and receiver ghost effects. By removing ghost signals, we reconstructed the frequency components attenuated near the notch frequency and produced the migrated stack section with enhanced temporal resolution.

• Economic and Environmental Geology
• /
• v.29 no.5
• /
• pp.629-637
• /
• 1996

In this paper a numerical experiment is conducted to determine the low acoustic impedance of a thin oil or gas reservoir from a seismogram by using the generalized linear inversion method. The seismograms used are normal incident synthetic seismograms containing p-wave primary reflections, multiples, and peg-leg multiples on the layers consisting of oil-, gas-, water-filled sandstone incased in shales. In this experiment the acoustic impedance, the location of reservoir boundary, thickness, and source wavelet are assumed initially and revised iteratively by the least-squares-error technique until the difference between the seismogram and calculated one is very small. This experiment shows that the acoustic impedance and thickness, about 10 m thick, can be determined by the inversion.

• Geophysics and Geophysical Exploration
• /
• v.1 no.1
• /
• pp.43-48
• /
• 1998

Recently seismic tomography has been widely used to visualize subsurface structure for resource explorations and construction site evaluation. We studied a way to include fresnel zone concept in the conventional ray-based traveltime tomography. The algorithm developed uses the same order of computing time as the conventional traveltime to mography but incorporates the rigorous wavepath concept of wave-equation tomography. Some experiments to synthetic and real data show reasonable results compared to conventional ray-based traveltime tomography.

• The Journal of Engineering Geology
• /
• v.23 no.4
• /
• pp.457-465
• /
• 2013

We integrated and correlated datasets from surface and subsurface geophysics, drilling cores, and engineering geology to identify geological interfaces and characterize the joints and fracture zones within the rock mass. The regional geometry of a geologically weak zone was investigated via a fence projection of electrical resistivity data and a borehole image-processing system. Subsurface discontinuities and intensive fracture zones within the rock mass are delineated by cross-hole seismic tomography and analyses of dip directions in rose diagrams. The dynamic elastic modulus is studied in terms of the P-wave velocity and Poisson's ratio. Subsurface discontinuities, which are conventionally identified using the N value and from core samples, can now be identified from anomalous reflection coefficients (i.e., acoustic impedance contrast) calculated using a pair of well logs, comprising seismic velocity from suspension-PS logging and density from logging. Intensive fracture zones identified in the synthetic seismogram are matched to core loss zones in the drilling core data and to a high concentration of joints in the borehole imaging system. The upper boundaries of fracture zones are correlated to strongly negative amplitude in the synthetic trace, which is constructed by convolution of the optimal Ricker wavelet with a reflection coefficient. The standard deviations of dynamic elastic moduli are higher for fracture zones than for acompact rock mass, due to the wide range of velocities resulting from the large numbers of joints and fractures within the zone.

• Geophysics and Geophysical Exploration
• /
• v.17 no.4
• /
• pp.187-201
• /
• 2014

Compared with the separated inversion of electromagnetic (EM) and seismic data, a joint inversion using both EM and seismic data reduces the uncertainty and gives the opportunity to use the advantage of each data. Seismic fullwaveform inversion allows velocity information with high resolution in complicated subsurface. However, it is an indirect survey which finds the structure containing oil and gas. On the other hand, marine controlled-source EM (mCSEM) inversion can directly indicate the oil and gas using different EM properties of hydrocarbon with marine sediments and cap rocks whereas it has poor resolution than seismic method. In this paper, we have developed a joint EM inversion algorithm using a cross-gradient technique. P-wave velocity structure obtained by full-waveform inversion using plane wave encoding is used as structure constraints to calculate the cross-gradient term in the joint inversion. When the jointinversion algorithm is applied to the synthetic data which are simulated for subsea reservoir exploration, images have been significantly improved over those obtained from separate EM inversion. The results indicate that the developed joint inversion scheme can be applied for detecting reservoir and calculating the accurate oil and gas reserves.

• Geophysics and Geophysical Exploration
• /
• v.10 no.1
• /
• pp.19-28
• /
• 2007

In this study, we propose a joint inversion method, using genetic algorithms, to estimate an S-wave velocity structure for deep sedimentary layers from receiver functions and surface-wave phase velocity observed at several sites. The method takes layer continuity over a target area into consideration by assuming that each layer has uniform physical properties, especially an S-wave velocity, at all the sites in a target area in order to invert datasets acquired at different sites simultaneously. Numerical experiments with synthetic data indicate that the proposed method is effective in reducing uncertainty in deep structure parameters when modelling only surface-wave dispersion data over a limited period range. We then apply the method to receiver functions derived from earthquake records at one site and two datasets of Rayleigh-wave phase velocity obtained from microtremor array surveys performed in central Tokyo, Japan. The estimated subsurface structure is in good agreement with the results of previous seismic refraction surveys and deep borehole data. We also conclude that the proposed method can provide a more accurate and reliable model than individual inversions of either receiver function data only or surface-wave dispersion data only.

• Geophysics and Geophysical Exploration
• /
• v.8 no.1
• /
• pp.41-49
• /
• 2005

We have developed a random heterogeneous velocity model with bimodal distribution in methane hydrate-bearing Bones. The P-wave well-log data have a von Karman type autocorrelation function and non-Gaussian distribution. The velocity histogram has two peaks separated by several hundred metres per second. A random heterogeneous medium with bimodal distribution is generated by mapping of a medium with a Gaussian probability distribution, yielded by the normal spectral-based generation method. By using an ellipsoidal autocorrelation function, the random medium also incorporates anisotropy of autocorrelation lengths. A simulated P-wave velocity log reproduces well the features of the field data. This model is applied to two simulations of elastic wane propagation. Synthetic reflection sections with source signals in two different frequency bands imply that the velocity fluctuation of the random model with bimodal distribution causes the frequency dependence of the Bottom Simulating Reflector (BSR) by affecting wave field scattering. A synthetic cross-well section suggests that the strong attenuation observed in field data might be caused by the extrinsic attenuation in scattering. We conclude that random heterogeneity with bimodal distribution is a key issue in modelling hydrate-bearing Bones, and that it can explain the frequency dependence and scattering observed in seismic sections in such areas.

• Geophysics and Geophysical Exploration
• /
• v.21 no.2
• /
• pp.103-111
• /
• 2018

The recent research aim of seismic trace interpolation is to effectively interpolate the data with spatial aliasing. Among various interpolation methods, the Matching Pursuit interpolation, that finds the proper combination of basis functions which can best recover traces, has been developed. However, this method cannot interpolate aliased data. Thus, the multi-component Matching Pursuit interpolation and moveout correction method have been proposed for interpolation of spatially aliased data. It is difficult to apply the multi-component Matching Pursuit interpolation to interpolating the OBC (Ocean Bottom Cable) data which is the multi-component data obtained at the ocean bottom because the isolation of P wave component is required in advance. Thus, in this study, we dealt with an effective single-component matching Pursuit interpolation method in OBC data where P-wave and S-wave are mixed and spatial aliasing is present. To do this, we proposed the Ricker wavelet based single-component Matching Pursuit interpolation workflow with moveoutcorrection and systematically investigated its effectiveness. In this workflow, the spatial aliasing problem is solved by applying constant value moveout correction to the data before the interpolation is performed. After finishing the interpolation, the inverse moveout correction is applied to the interpolated data using the same constant velocity. Through the application of our workflow to the synthetic OBC seismic data, we verified the effectiveness of the proposed workflow. In addition, we showed that the interpolation of field OBC data with severe spatial aliasing was successfully performed using our workflow.

• Journal of the Korean Geophysical Society
• /
• v.3 no.1
• /
• pp.37-48
• /
• 2000

In order to investigate in-line ground motions caused by earthquakes, we examine the multicomponent complex trace analysis method (MCTAM) for the synthetic data and apply it to real earthquake data. An experimental result for synthetic data gives correct information on the arrival times, duration of individual phases, and approaching angles for body waves. Rayleigh waves are also easily identified with the MCTAM. A deep earthquake with magnitude of 7.3 was chosen to test various polarization attributes of ground motions. For P waves, instantaneous phase difference between the vertical and the in-line horizontal components ${\phi}(t)$, instantaneous reciprocal ellipticity ${\rho}(t)$, and approaching angle ${\tau}(t)$ are computed to be ${\pm}180^{\circ},\;0{\sim}0.25,\;and\;-30^{\circ}{\sim}-45^{\circ}$, respectively. For S waves, ${\phi}(t)$ tends to vary while ${\rho}(t)$ have values of $0{\sim}0.3\;and\;{\tau}(t)$ remains near vertical, respectively. A relatively low frequency signal registered just prior to the S wave event is interpreted as a P-wave phase based on its polarization characteristics. Velocities of P and S waves are computed to be 8.633 km/s and 4.762 km/s, and their raypath parameters 0.074 s/km and 0.197 s/km. Dynamic Poisson's ratio is obtained as 0.281 from the velocities of P and S waves.