The 17th International Symposium on Recent Advances in Exploration Geophysics (RAEG 2013)

Data processing of magnetotelluric (MT) survey has been based on the Fast Fourier Transform (FFT). The FFT processing gives us the response functions (RFs) of the earth fast and easily. However, applying FFT processing to MT data may not be optimum always, since the source of MT is the transient fluctuation of electric current in the ionosphere. The FFT assumes time series to be a stationary so that we focused on an IIR filter called ‘pole on pedestal’ that extracts the signal at a specific frequency. Combining this IIR filter and the Hilbert transform, the RFs are calculated in time domain. For removing the segment contaminated by noise before calculating the RFs, we applied the maximum entropy method (MEM) to the selection of the segments contaminated by large noise. The MEM searches and removes these contaminated segments easily. As a result, we developed the time domain processing of MT data using MEM and IIR filter, and applied this processing to the real data acquired at the Nankai trough. Comparing the conventional and novel data processing, our novel data processing gives us more optimum RFs than the conventional processing. In addition, we discussed the time-domain separation of up-going/down-going electromagnetic waves based on the plane wave decomposition. All our attempts support an idea that the time-domain analysis of magnetotelluric data would give us more accurate and detailed subsurface information.

Better understanding of failure mechanism of rocks benefits in many fields from rock engineering to earth sciences. Especially, it is essential to understand how fractures initiate and propagate under various loading conditions in order to clarify real rock fracture processes. For the interpretation of rock failure, many attempts have been made experimentally or using fracture mechanics theory. Although much of the knowledge available today is based on experimental observations and the theory successfully represented the propagation of predefined cracks, the failure mechanics are not fully understood by experimental results and it is difficult to describe the initiation and coalescence of cracks using the theory. Thus, in the recent years, numerical modeling, which may be less restrictive, has been applied to study crack behaviors, and we also approach to the rock failure based on numerical simulations. To represent rock failure, we use a Hamiltonian Particle method (HPM), one of particle methods. In the HPM, we do not need to use grids or meshes to discretize the rock model, and thus we could deal with the failure relatively easily. In spite of this advantage of the HPM, the applicability of the method to the failure phenomena have yet to be revealed fully. In the present study, we apply the HPM to rock failure under some different specimens and different loading conditions. As a result of our simulations, the HPM successfully reproduce failure pattern of brittle fracture observed rock fracture experiments and can observe micro cracks initiation and propagation. This suggests that the HPM is the useful tool to analyze rock failure.

Stress field deep under the ground, like seismogenic depth, is important to know seismic activity, state of oil reservoir, etc. However we can only know stress field near the earth surface using the current equipment and technique. In this study we focus on seismic scattering waves having information of the earth crust where the seismic waves travel through. From the scattering waves, we try to estimate the stress field deep under the ground. At first we employ a 2-D finite different method to reveal a relationship between the seismic scattering and the crustal stress field, i.e., magnitude and direction of the stress. As the result, it is revealed that coda-Q, which is derived from attenuation ratio of a seismic wave, has a proportional relationship with magnitude of the stress and changes periodically against direction of the magnitude. Next it is examined if the relationship can be seen in real field data using seismograms obtained by Hi-net. And it is revealed that coda-Q has the relationship with change in strain obtained by GPS observation and theoretical dislocation calculated from fault movement; both are proxy of the crustal stress field. We conclude that change in crustal stress field can be estimated from monitoring of coda-Q.

Seismic full-waveform inversion (FWI) method has been used to estimate subsurface velocity structure. FWI directly utilizes observed waveforms that could include information on the properties of subsurface materials. In seismic time-lapse surveys, we observe the difference between waveforms as a function of time for the change such as fluid alteration. Residual waveforms between the observed before and after a certain time interval are used to estimate the changes in the fluid distribution in terms of seismic velocities in FWI method. In contrast to the previous FWI applications, our research focuses directly on the properties in the hydrocarbon reservoir in order to estimate the fluid distribution and alteration. We simulate the wave propagation based on the Biot theory that includes the effects of fluid in porous media. The simulation model is composed of a block of sandstone saturated with water and gas. We assume a transition zone around the fluid contact, whose vertical profile of the saturation rate varies gradually in time in this zone. The result inspires that the combination of elastic parameters is necessary for estimating the seismic velocity contact in fluid transition zone relating to the fluid-contact movement by FWI.

The applicability and feasibility of eddy current detection method for the measurement of wall thinning and surface crack of steel structure have been practically confirmed by field and laboratory experiments. Recently, we could roughly understand where and how large the defects are by this method. However, it is difficult to estimate the exact size and shape of them. For more accurate inspections, there has been a growing demand to quantitatively evaluate the defects. Therefore, we have developed a numerical simulator to consider whether we could develop the high accuracy eddy current method. Eddy current method uses the information of excitation and induced magnetic field. In order to calculate the induced magnetic field, we used a 2.5 dimensional finite-difference frequency domain technique (2.5D-FDFD) to solve Maxwell's equations numerically. In this technique, we assumed the two-dimensional structure and the three-dimensional electromagnetic field. We used two-layer structure consisting of seawater and steel plate containing defects. To estimate characteristic of the induced magnetic field, we simulated for a variety of defects and compared what effect appear. As a result, we could confirm the effect of surface defects of steel plate on receiving magnetic field intensity. The induced magnetic field intensity increases near the edge of the defects and decays above the defects. The larger depth and width of the defects are, the more attenuate the magnetic field intensity becomes. Our simulation results indicated that we could obtain the response of magnetic field intensity whose detectable scale of defects is no smaller than mm order.

Electromagnetic waves with single or limited frequencies from VLF transmitters generate secondary induced components of magnetic field, which are used to detect localized changes in electrical conductivity contrasts. This method, so-called VLF-EM, has been the powerful tool for mapping subsurface geological structures because of its low cost and short survey terms compared with electrical resistivity survey. However, it has not been tested to estimate a pseudo-resistivity section, both the apparent resistivity and the depth of conductive anomaly by using the measured magnetic components with a single frequency. In this study, the Normalized Full Gradient (NFG) method, generally used for the downward continuation of the potential filed data, was applied to the magnetic components at the surface. The VLF-EM data set was obtained numerically on a synthetic model. The cross section of NFG values derived from horizontal component of magnetic field clearly indicates high peaks at edges of a low resistivity anomaly buried below the surface. The peak of NFG values from the vertical component correspond with the centre of the anomaly. On the basis of the results, we estimated the pseudo-section of apparent resistivity from the VLF-EM data weighted with the NFG values at each depth. We confirmed that the weighted apparent resistivity values are lower in the vicinity of low resistivity anomaly than the surrounding area, although the estimated value is a little higher than the original value. We conclude that our simple technique give an approximate subsurface resistivity structures quickly, which is useful for geological interpretations.

We present a 3D joint inversion method to estimate two physical parameters, density and magnetization of subsurface materials using field intensity measurements. In the method, we introduce the fuzzy c-means (FCM) clustering technique to relate gravity with magnetic data. In the approach, the subsurface structure is discretized to a set of rectangular prisms. For estimating the density and magnetization of each prism, we minimize the quadratic norm of the residuals between the observed and the forward-modeled. Two regularization terms, i.e. the roughness and the similarity of the two physical parameters, are introduced in our joint inversion to control the degree of model roughness and similarity. We determine their regularization parameters using the L-curve criterion. We apply our method to a numerical model which represents submarine massive sulphides (SMS). The joint inversion results, which have the advantages of both gravity and magnetic inversion, show better accuracy and resolution than the individual ones.

The 2D shear wave velocity profile of strata is estimated using the active and passive surface wave seismic tests. The experimental dispersion curves were obtained after the recorded signals were transformed by the slant stack procedure. The phase velocity in the relatively high frequency range can be obtained using the dispersion curves deduced from the active tests. On the other side, dispersion curves obtained from the passive tests can be used to estimate the phase velocity in the relatively low frequency range. From the higher frequency portion of the dispersion curves that stand for the fundamental mode, we obtained the phase velocities about 190 m/s for the sandy surface fill. Theoretical dispersion curves can be constructed by the thin-layer-stiffness-matrix method. For theoretical dispersion curves, the soil layers of the test site were modeled as the sandy surface fill overlying a half space soil layer. A real-parameter genetic algorithm was programmed to minimize the difference between the theoretical and experimental dispersion curves. We prove that the real-parameter genetic algorithm is capable to reduce the error between experimental and theoretical dispersion curves. The estimated 2D geometry of the sandy surface fill using the active and passive surface wave seismic tests was verified with the borehole data.