Ground penetrating radar (GPR) exploration provides volumetric water content which can be calculated from relative dielectric permittivity of a ground medium. The volumetric water content is calculated using Topp's equation (Topp et al., 1980) in the TDR(Time Domain Reflectmetry) method. This equation has wide applications to soil water content estimation as a simple and easy way. Topp's equation was determined by the relative dielectric permittivity of soils and glass beads which have taken the volumetric water content under 50%. Care must be taken when applying Topp's equation for high water content materials and fine-grained cohesive soils.
 It is known that the frequency dependence (dielectric dispersion) is present for soils containing water, but dielectric dispersion has hardly been discussed in Topp et al.(1980).
 This report showed the relation between relative dielectric permittivity and volumetric water content less than 90% considering the dielectric dispersion of the soil samples. This result supports the ‘semi-disperse model’ by Wobschall (1977) and will provide the basis for a new indicator of the high accuracy volumetric water content estimation using GPR.

View full abstract

 It is important to simulate a strong ground motion induced by the occurrence of earthquakes for disaster mitigation and prediction. Though finite difference method (FDM) has been used to simulate seismic wave propagation, it is difficult to introduce traction-free boundary conditions in a model with an arbitrary ground surface. Since finite element method (FEM) has an advantage of introduction of traction-free boundary conditions, FEM was applied to simulate seismic wave propagation. However, the connectivity between elements and nodes are needed in pre-processing for FEM analyses. This will lead the complex data structure and time consuming process. On the other hand, many particle methods have been developed and applied to simulate solid analyses. A Hamiltonian particle method (HPM) is one of the particle methods and developed for accurate energy conserving method. In HPM, the traction-free boundary condition is automatically introduced like FEM. Furthermore, the data structure becomes very simple because the positions of particles are only needed in HPM. However, artificial forces are needed for suppressing local particle oscillations in HPM.
 In the present study, we apply HPM to simulate seismic wave propagation and evaluate the effect of artificial force with comparing to the results from FDM and FEM. The results show that HPM can reproduce seismic wave propagationwith sufficient accuracy.

View full abstract