• 한국지구물리탐사학회:학술대회논문집
• /
• 2007.06a
• /
• pp.209-214
• /
• 2007

In Korean geology that crystalline rock is dominant, the properties of subsurface including the anisotropy are distributed complexly and changed abruptly. Because of such geological environments, cross-hole seismic traveltime tomography is widely used to obtain the high resolution image of the subsurface for the engineering purposes in the geotechnical sites. However, because the cross-hole tomography has a wide propagation angle coverage relatively, its data tend to include the seismic velocity anisotropy comparing with the surface seismic methods. It can cause the misinterpretation that the cross-hole seismic data including the anisotropic effects are analyzed and treated with the general processing techniques assuming the isotropy. Therefore, we need to consider the seismic anisotropy in cross-hole seismic traveltime tomography. The seismic anisotropic tomography algorithm, which is developed for evaluation of the velocity anisotropy, includes several inversion schemes in order to make the inversion process stable and robust. First of all, the set of the inversion parameters is limited to one slowness, two ratios of slowness and one direction of the anisotropy symmetric axis. The ranges of the inversion parameters are localized by the pseudo-beta transform to obtain the reasonable inversion results and the inversion constraints are controlled efficiently by ACB(Active Constraint Balancing) method. Especially, the inversion using the Fresnel volume is applied to the anisotropic tomography and it can make the anisotropic tomography more stable than ray tomography as it widens the propagation angle coverage.

• Economic and Environmental Geology
• /
• v.57 no.1
• /
• pp.1-15
• /
• 2024

The Melanesian region in the western Pacific is dominated by complex plate tectonics, with the largest oceanic plateau, the OntongJava plateau, and a hotspot, the Caroline Islands. To better understand the complex geodynamics of the region, we estimate P- and S-velocity models and 𝛿 (V_P/V_S) model by using relative teleseismic travel times measured at seismometers on land and the seafloor. Our results show high-velocity anomalies in the subduction zones of the Melanesian region to a depth of about 400 km, which is thought to be subducting Solomon Sea, Bismarck, and Australian plates along plate boundaries. Along subduction zones, positive 𝛿 (V_P/V_S) anomalies are found, which may be caused by partial melting due to dehydration. A broad high-velocity anomaly is observed at 600 km depth below the Ontong-Java plateau, with a negative 𝛿 (V_P/V_S) anomaly. This is thought to be a viscous and dry remnant of the Pacific plate that subducted at 45-25 Ma, with a low volume of fluids due to dehydration for a long period in the mantle transition zone. Beneath the Caroline Islands, a strong low-velocity anomaly is obseved to a depth of 800 km and appears to be connected to the underside of the remnant Pacific plate in the mantle transition zone. This suggests that the mantle plume originating in the lower mantle has been redirected due to the interaction with the remnant Pacific plate and has reached its current location. The mantle plume also has a positive 𝛿 (V_P/V_S) anomaly, which is thought to be due to the influence of embedded fluids or partial melting. A high-velocity anomaly, interpreted as an effect of the thick lithosphere beneath the Ontong-Java plateau, is observed down to 300 km depth with a negative 𝛿 (V_P/V_S) anomaly, which likely indicate that little fluid remains in the melt residue accumulated in the lithosphere.

• Geophysics and Geophysical Exploration
• /
• v.5 no.3
• /
• pp.178-184
• /
• 2002

For economy and convenience, seismic refraction survey is widely used in surveying for large civil engineering work. The purpose of this study is to obtain the numerical responses of various models using Kirchhoff migration, and to analyze its application to the real data processing. Synthetic traveltime curve was calculated by vidale's algorithm, and various models such as 2 or 3 layer model and irregular topography model are tested to simulate the response of real structure. In order to compare the effect of initial velocity model, true velocity models, inversion results by tomography, smooth velocity models are used as an initial guess. The responses of model data show that the algorithm of this study is more sensitive to initial velocity model than the reflection survey, so choosing a suitable initial velocity model will be the most important thing in real data processing.

• Geophysics and Geophysical Exploration
• /
• v.11 no.1
• /
• pp.1-14
• /
• 2008

We present a method for interpreting seismic records with arrivals and waveforms having characteristics which could be generated by extremely inhomogeneous velocity structures, such as non-typical oceanic crust, decollement at subduction zones, and seamounts in oceanic regions, by comparing them with synthetic waveforms. Recent extensive refraction and wide-angle reflection surveys in oceanic regions have provided us with a huge number of high-resolution and high-quality seismic records containing characteristic arrivals and waveforms, besides first arrivals and major reflected phases such as PmP. Some characteristic waveforms, with significant later reflected phases or anomalous amplitude decay with offset distance, are difficult to interpret using only a conventional interpretation method such as the traveltime tomographic inversion method. We find the best process for investigating such characteristic phases is to use an interactive interpretation method to compare observed data with synthetic waveforms, and calculate raypaths and traveltimes. This approach enables us to construct a reasonable structural model that includes all of the major characteristics of the observed waveforms. We present results here with some actual observed examples that might be of great help in the interpretation of such problematic phases. Our approach to the analysis of waveform characteristics is endorsed as an innovative method for constructing high-resolution and high-quality crustal structure models, not only in oceanic regions, but also in the continental regions.

• Geophysics and Geophysical Exploration
• /
• v.24 no.2
• /
• pp.53-66
• /
• 2021

Velocity model building is an essential procedure in seismic data processing. Conventional techniques, such as traveltime tomography or velocity analysis take longer computational time to predict a single velocity model and the quality of the inversion results is highly dependent on human expertise. Full-waveform inversions also depend on an accurate initial model. Recently, deep neural network techniques are gaining widespread acceptance due to an increase in their integration to solving complex and nonlinear problems. This study investigated cases of seismic velocity model building using deep neural network techniques by classifying items according to the neural networks used in each study. We also included cases of generating training synthetic velocity models. Deep neural networks automatically optimize model parameters by training neural networks from large amounts of data. Thus, less human interaction is involved in the quality of the inversion results compared to that of conventional techniques and the computational cost of predicting a single velocity model after training is negligible. Additionally, unlike full-waveform inversions, the initial velocity model is not required. Several studies have demonstrated that deep neural network techniques achieve outstanding performance not only in computational cost but also in inversion results. Based on the research results, we analyzed and discussed the characteristics of deep neural network techniques for building velocity models.