The 19th International Symposium on Recent Advances in Exploration Geophysics (RAEG 2015)

Seismic full-waveform inversion (FWI) is a method developed for estimating velocity distribution in a medium in which seismic waves travel through. Seismic velocities are expressed by the combination of two elastic constants and density in a homogeneous elastic medium, but density is usually estimated using an empirical formula or is fixed to a constant value in practice. The majority of elastic FWI studies have ignored the influence of density that could be an important parameter. Inverting for three elastic parameters is difficult problem even in FWI for a homogeneous elastic medium when solving for velocities that are composed of the three parameters coupled with each other, and the coupling effects prevent one from the appropriate estimation of the elastic parameters. Since, density is the useful parameter for hydrocarbon characterization, we investigate the best strategy to estimate density structure and reveal the difficulty of the estimation of density structure. We conduct several numerical experiments to estimate density structure. Our inversion results and grid analysis of misfit function show that density is affected by other parameter in the inversion and sequential inversion procedure is the most suitable strategy to estimate density.

The applicability and the feasibility of eddy-current testing method for the detection of defects such as wall thinning, surface crack, pinholes, etc. of steel structure have been practically confirmed by field and laboratory experiments. Qualitative analysis of these defects has been empirically understood by the current practice of eddy current methods based on analog analysis. There has, however, been a growing demand to quantitatively evaluate the defects. We tackle this problem by use of time series of induced magnetic field caused by the defects. In the present study, we proposed a new digital analysis process to use the induced magnetic field waveform, and validated the effectiveness of the method using numerical simulations. First, we developed a high-speed electromagnetic simulator using a fictitious wave domain method to reproduce the conventional eddy current testing method for realistic sheet-piles seawater model. Then, we confirmed the features of the induced magnetic field waveforms by a few types of defects. Paying attention to the residuals of the induced magnetic field waveform, we developed a novel migration procedure for detecting accurate position of cracks or defects. Using our approach, we could obtain crack images without phase lag. For the evaluation of the crack position and wall thinning of the sheet piles, we also applied attribute analysis traditionally used in the field of seismic survey. Through the application of the digital signal processing using induced magnetic field waveform, we could successfully develop a processing scheme with a high degree of accuracy of the eddy current testing method.

The objective of this study is the evaluation of reliability of estimated hydraulic conductivity and specific storage structure from self-potential (SP) data under two difficult conditions. As one of difficult condition, the lack of flux data is assumed. This situation is assumed because the sampling rate of flux data would become lower than the SP data when the monitoring data is focused on. The effect of metal casing pipe is evaluated as the other difficult condition case. Our numerical tests show that only the estimation result of hydraulic conductivity structure can be reliable when the sampling rate of the flux data is too low. It is also demonstrated that the metal casing changes the value of estimation results of hydraulic conductivity and specific storage around the casing pipe. We also demonstrate that the effect of metal casing pipe can be removed by the inversion scheme proposed in this study.

Hydraulic fracturing is a technique used in the development of shale oil reservoirs, enhanced geothermal system in hot dry rock, etc. The behavior of hydraulically induced fractures is of great interest in such applications. Although hydraulic fractures tend to propagate in the direction of maximum principle stress in theory, the direction of hydraulic fractures does not always correspond with that expected from regional stress due to pre-existing fractures. Furthermore, it has been indicated that the behavior of hydraulic fracture is influenced by the viscosity of injection fluid. Although the effects of pre-existing fractures and the viscosity of injection fluid on the effectiveness of the hydraulic fracturing are individually investigated, the prediction of the combination effect of them still remains as a big challenge. In order to examine the degree of the effectiveness of using high viscous fluid in hydraulic fracturing in naturally fractured rock, we developed a 2D numerical simulation code using discrete element method (DEM) that includes the coupling of solid-fluid interaction, and performed a series of numerical tests. The results show that hydraulic fracture is trapped by the pre-existing fracture in the low viscous fluid case, while it propagates beyond the pre-existing fracture in the high viscous case. The results also show that hydraulic fracture propagation could easily be retarded when the interaction angle of hydraulic fracture and pre-existing fracture is high. We conclude that high intersection angle between the hydraulic fracture and pre-existing fracture requires high viscous fluid for effective hydraulic fracturing.

Mineral scaling is a key process for various geophysical phenomena. Especially it is well known that silica scale is the most difficult one to control. The evaluation of the contribution of the chemical kinetic or the hydrodynamic process on the silica scaling is of importance towards the control of membrane scaling. To meet this goal, we compare the simulation result of the amount and the distribution of silica deposition predicted by the chemical kinetic and the hydrodynamic deposition process and the data from a laboratory or a field experiment. We solve the fluid, temperature and the dissolved silica concentration field by using the lattice Boltzmann method. From our simulation result, it is found that the kinetic process is not sufficient for representing the real silica deposition and that the hydrodynamic effect on growth of silica scale is important for both qualitatively and quantitative evaluations. It is, therefore, necessary to emphasize the physical adhesion should be taken into account for reproducing the process of silica scaling. To further predict the physical adhesion of silica particles, we need to analyze various interaction forces among silica particles in flow, membrane, wall, etc. and to establish the simple silica scale growth microscopically.

In The Nankai Trough seismogenic zone, cabled real-time observation system, DONET and IODP C0002G borehole observatory, have been operating to monitor seismic activity, crustal deformation and tsunami propagation since Aug. 2011 and Jan. 2013, respectively. For elucidating dynamic processes from preseismic state to the generation of mega-thrust earthquake, which occurs repeatedly in subduction zones, it is important to observe and monitor the stress state, i.e., a key parameter governing its fault dynamics in the vicinity of seismogenic fault. In this study, we performed active and passive seismic data processing to obtain seismic anisotropy, as a proxy of stress state, by using dataset acquired by three-component seismometers installed in the DONET and IODP C0002G observatories. In the passive data processing, we applied a seismic interferometry method to ambient noise records acquired by horizontal components of each seismometer. After the application of cross-dipole analysis to the acquired records, several coherent events have become visible. These events were perceived to be reflected S-wave from each layer below seafloor, and S-wave splitting caused by seismic anisotropy was observed. We then estimated anisotropy direction and amplitude beneath each seismometer in shallow sediment layer. For the active seismic dataset that was acquired in azimuthally aligned airgun shot locations for each seismometer performed in Nov. 2013, we steered horizontal records for each pair of shot and seismometer to form radial and transverse components to each shot location at a distance of ca. 3km from the seismometer. In the steered records, P-S converted waves from bottom of shallow sediments were clearly visible. The horizontal axis of symmetry to fast S-wave direction was estimated through the fitting of a simple sinusoidal curve to the aximuthal amplitude distribution in radial and transverse components for S-wave anisotropy in the shallow sediments in the Nankai Trough. We finally compared the obtained S-wave anisotropy from the passive data with the active one. The comparison in the order of anisotropy and the orientation of the horizontal axis of symmetry showed good agreement with each other, especially in landward area. Some differences in the complicated structure zone were probably caused by the signal-to-noise ratio deterioration due to the influence of the dimensionality of the sub-seafloor structure and the seafloor topography around the observatory in the active dataset, and to the contamination of seafloor microseisms in the passive data. We now plan to develop a new scheme including layer stripping method, 3-D rotation method, etc., to improve the quality of our analysis.

Full waveform inversion (FWI) of seismic reflection data has been widely used with different 2D/3D data acquisiton geometries. The method develops a reliable subsurface model by minimizing misfit between observed and simulated waveforms. Vertical Cable Seismic (VCS) is a recent seismic reflection method which uses vertical array of hytdrophones in order to record acoustic energy generated by seismic sources. VCS can acquire high resolution data by deploying hydrophone arrays very near to the seafloor which results in a higher signal to noise ratio. Because of the especial geometry of VCS data it is challenging to develop a velocity model based on the conventional processing techniques. We used a simulation experiment to evaluate the FWI results on seismic reflection data acquired using VCS geometry. Although dominant events in the data were reflections rather than diving waves, which are important for FWI, we could obtain a promising velocity model.

In the magnetotelluric (MT) method, underground impedance profile is estimated through the observation and the processing of both electric and magnetic components. However, there are mainly two problems in the observation of electric component. One is that the effect of the static shift biases the apparent resistivity at the surface, and the other is the difficulty in the observation of electric field at topographically planar surface suitable for the assumption of place wave incidence. Coupled with these known observation problems, the analysis of electromagnetic data is costly in the computation, especially in 3D studies. In this study, we propose a new method to use magnetic components at the surface in order to reduce the burden of computation that may help to cope with these problems. We first adopted a migration method which is usually used in seismic data processing and applied this method to the magnetic field observed at the surface. If magnetic field is used for the analysis, we could control the static shift that appears in electric field. Second, the mobility of magnetometers makes it possible to obtain field data much easier even at non-planar surface. Also, a migration method leads us to reduce the computational costs. Through the numerical calculation analyses, it was confirmed that the application of the proposed migration method to magnetic component is effective and efficient to process data. In this migration method, the larger value of reflection intensity is estimated around correct area if the proper resistivity structure is assigned to the migration. This means that the resistivity structures in the subsurface have an important role in the migration results. The combination of this electromagnetic migration method and inversion has the possibility to reveal more detailed subsurface resistivity structure.