5th Asia Pacific Meeting on Near Surface Geoscience & Engineering

A geophysical investigation was undertaken on a 150-year-old puddle core earthfill dam to determine seepage pathways and deformation. The dam is currently in operation and is undergoing a risk position in portfolio risk assessment. A Dam Safety Review (DSR) was to carry out to evaluate risks that associated with the dam. The geophysical study was deployed to define and characterise the thickness of existing fill and reservoir dam’s internal structure using four non-invasive, near-surface geophysical methods: Electrical Resistivity Imaging (ERI), Ground Penetrating Radar (GPR), Frequency-domain Electromagnetics (FEM) and seismic refraction.

The geophysical survey was verified with on-going geotechnical investigation. The results determine seepage pathways with using FEM and ERI, as well as potential voids within the dam channel walls that have been defined by GPR. These results have integrated into a three-dimensional geological model, dam foundation, existing dam structure and seepage pathways were included to assist on interpretation of dam condition. Collectively, the model of this study has been delivered to client to inform its future safety review and design.

Surficial and subsurface soil-rock characterization is challenging in tropical crystalline basement terrain, especially Penang Island, Malaysia. This is because of the varying complex soil profiles and geohydrodynamics of the litho-structural units. The spatial variability of the study area’s soil-rock conditions, interfaces, and architecture from three sites were investigated using velocity-resistivity, N-values (from in-situ soil penetration tests), and geostatistical relationships. The borehole-derived logs correlated with resistivity and seismic P-wave velocity (Vp) models clearly show that the eastern to northern sections of the area are characterized by thick, saturated, and loose silty sand to sandy materials (with clay) and deep-weathered/fractured zones. However, the southern section comprises a large amount of sand, boulders, and resistive bedrock at shallow depths with pronounced multiple fractures.

As a result, infrastructure must be adequately reinforced/piled to rest on the stable bedrock. Intended wells require proper casing and completion to prevent silt intrusion and occlusion, especially in the central and northern parts. Furthermore, landslides may be induced at steep sections due to the area’s relief topography and oversaturation of the clay/silty soils under heavy rains. The empirical relationships developed are adequate for N-values prediction in tropical granitic environments, especially Vp with N-values with 94.3% accuracy.

The work demonstrates a method for employing the FBG MLMS and sequential multi- depth injections for an HT survey at a low to moderate heterogeneity sedimentary aquifer. While the desired degree of imaging accuracy (resolution and accuracy of the estimated parameters) is usually problem-specific, the results suggest that increasing the number of injection points at various depths could improve imaging accuracy. Moreover, the FBG MLMS efficiently collects groundwater measurements at different depths inside a well. Thus, using MLMS wells, the increase in the injection/pumping points can lead to an effective increase in the number of cross-hole tests; in turn, the collected head observations can effectively reach ergodicity.

We carried out the 3D microtremor array measurement to demonstrate the contribution of the geophysical method in creating the 3D geological model employed to evaluate the liquefaction potential. The investigation site is Kamisu city, where extensive liquefaction occurred during the Tohoku earthquake in 2011. Because of limited access to private properties, we decided to carry out the 3D microtremor array measurement by irregular arrays on accessible roads. Line, L-shape, T-shape, and cross-shape arrays with a small receiver spacing of 5 m and grid arrays with a large receiver spacing of 40 m are combined, and we performed the CMP-SPAC method to obtain dispersion curves. Then, the 3D S-wave velocity model is estimated from all dispersion curves. Relatively low S-wave velocity areas in the 3D velocity model are highly correlated with damage to houses and roads. Additionally, the iso-surface map of the S-wave velocity of 200 m/s extracted from the 3D S-wave velocity model delineates the boundary of the filled sediments in the past mined areas. Because the loose sediment is one of the causes of the liquefaction damage, the result indicates the usefulness of the 3D microtremor array measurement for updating the geological model for a liquefaction evaluation.

The 6 initial models were tested to obtain the best 1D velocity model, called NTh1D model, to improve earthquake location in the northern part of Thailand. After testing the layer thickness of IASP91 model, the 2 km grid size and 32 km crustal thickness of the northern Thailand show the also show the minimum error. The NTh1D model shows more accurate hypocenter solution compare with original IASP91 and other models tested in this study. The station corrections were also reported to produced better earthquake locations.

Marine geotechnical engineering is the basis of marine engineering. Shear wave velocity is one of the most important parameters that can be used for geotechnical considerations. In contrast to borehole measurement, underwater surface wave method can provide shear wave velocity profile even 2D image with seabed surface measurement without boring. Considering the most efficient way to conduct the UMASW, air gun as seismic source plus on-boat multi-channel seismometer with streamers would be the optimum measurement equipment. It requires the streamer close enough to the seabed to acquire a good signal but no guideline to indicate how close would be close enough. A preliminary study based on the numerical study was conducted to find out the allowable distance between suspended streamer and seabed and the effect of water depth. According to the numerical results, water depth would be an important information if conducting UMASW in water depth less than 5m and suspended streamers can be used but 6m (or larger than1/4 the largest Scholte wave wavelength) away from the seabed is not recommend.

Time-harmonic loading over layered elastic half-spaces has applications in various science and engineering fields. While various approaches have been proposed in solving the related boundary-value problems, in this paper, we propose a new approach, which is based on the novel Fourier-Bessel series system of vector functions and the dual variable and position method (DVP). While the DVP method was proposed recently and verified to be computationally stable and efficient, the Fourier-Bessel series system of vector functions is newly introduced. Similar to the cylindrical system of vector functions, the normal (dilatational) and shear (torsional) deformations (waves) can be separated and solved in terms of the LM- and N-types of the new vector function system. The new formulation is coded, and the corresponding algorithm/program is applied to a couple of cases. It is shown that, by comparing previous approaches, this new series system of vector functions is equally accurate, but much more computationally powerful. Since it is substantially time saving in calculation, it is hopeful that this new approach would have broad applications related to transient response and inverse problems in elastodynamics of layered systems.

The estimation of shear wave velocity profile is an important issue in near surface geophysics.

Love wave analysis is a useful tool for shear wave velocity estimation for its simplicity over the Rayleigh wave, and the 1D inversion in the transformed domain reduces the non-linearity and completely avoid the need for source estimation. In this study, a 2D forward model of dynamic response simulating for isotropic, elastic-layered system under transversal strip loading is presented. Focusing on examining the full wavefield in frequency-phase velocity domain (i.e., full velocity spectrum (FVS)), we compare the FVS of the proposed method to FVS from true three dimensional (3D) solutions. A post-processing approach is proposed to modify the 2D solution to better fit the 3D solution. This modified 2D solution could replace the true but computationally intensive 3D solution for forward modelling of FVS. Inversion of VS profile in the transformed domain can be significantly improved based on the presented forward model.

A new generation time-domain electromagnetic (TEM) system has been developed for measuring conductivity in the near surface. It is unique in that it is backpack-mounted and incorporates the latest developments in electronics and signal processing to enable continuous operation whilst the operators are walking. This enables data collection in hard to access areas. Latest development sin electronics and signal processing allow collection of data in urban environments.

The system has been used in a wide range of applications including mineral exploration; investigation of tailings dams and other impoundments, clay mapping, groundwater studies, waste disposal sites, saline incursion, in-mine resource definition, acid mine drainage, archaeology, diamond exploration and void detection.

Results will be presented from studies of tailings storage facilities, mineral exploration, geological mapping and clay mapping.