Some case studies of road embankments damaged and undamaged by the 2011 off the Pacific coast of Tohoku earthquake were conducted as the first step for the final target to rationalize the current seismic design method. The relationship between the conventional safety factor evaluated by pseudo-static analysis and the degree of actual damage was investigated based on the results from drained triaxial compression tests on retrieved fill materials. The value of the seismic coefficient, which should be employed in the conventional design method for preventing severe damage due to the recent large earthquakes, was also discussed. In addition, the applicability of Newmark’s sliding block method to the evaluation of earthquake-induced displacement of road embankments was presented.

The Korean government recently has been focusing to minimize natural disaster damages including earthquake damage under the slogan ‘ Safe Korea’. A part of this effort is to set up Korean earthquake defense GIS system including several earthquake hazards such as liquefaction and land sliding. The purpose of this study is to make micro and macro liquefaction hazard maps in Korea based on liquefaction potential index calculated by simplified liquefaction potential assessment. The difference between micro and macro maps is consideration on the seismic amplification in soils. In case of micro liquefaction hazard map, it has been used to perform site response analysis. In other case of macro liquefaction hazard map for the whole country, soil amplification factor can replace the site response analysis because we must perform so many site response analyses to calculate LPIs. Accordingly, we verified the application of the soil amplification factor. At that time, S city was considered in its feasibility by the target. In this study, we developed a specific program to calculate LPI automatically. Finally, macro liquefaction hazard maps under various accelerations for the whole country could be drawn.

Damage to the embankments of abutment backfills often affects emergency routes by introducing sharp drops in the road level between the bridge and backfill. Therefore, it is important to clarify the seismic performance of abutment backfills with reinforced double-wall facing to maintain the road level. The length of the necessary tie-bars is determined by the internal stability of the reinforced area. It follows that anchor plates may be installed in intersecting or binding configurations, which are not ordinary conditions, depending on the width of the road. In this study, a series of centrifugal shaking table tests was performed to investigate the seismic behavior of a multi-anchor wall with double-wall facing. The test results showed various deformations with different reinforcement positions; perpendicularity was best maintained when using a binding tie-bar configuration. Smaller deformation was observed for a wall using intersecting anchor plates compared to that using an ordinary configuration. This implies that crossing the reinforcement areas can improve, rather than adversely affect, the stability of the full structure.

Damages of the reinforced soil structure due to high ground water level by seismic wave were reported, even such soil structure has high performance against earthquake as common knowledge. This paper described the effectiveness of the drainage pipe into the anchored reinforced soil wall which has high level of ground water to improve stability of such structure during earthquake. A series of centrifugal shaking table tests was carried out to clarify the deformation mechanism of multi-anchor wall by earthquake during seepage flow. Test results showed the high seismic performance of such structure in spite of smaller degree of compaction of backfill compared with ordinary construction. Behaviors of reinforced region entirely against the earthquakes were observed. This implies that it is important for maintenance the deformation of reinforced soil wall to prevent the soil mass increasing with ground water rising in order to control an inertial force during earthquakes.

This paper reports on the influence of initial static shear on large deformation behavior of very loose (D_r = 24-30%) Toyoura sand subjected to undrained cyclic loading. A series of isotropically consolidated torsional simple shear tests were carried out on hollow cylindrical specimens up to single amplitude shear strain exceeding 50%. Two types of cyclic loading patterns, namely reversal stress and non-reversal stress, were employed by varying the magnitude of combined initial static shear and cyclic shear stresses. The observed types of failure were distinguished as liquefaction and residual deformation based on the difference in the effective stress paths and the modes of development of cyclic residual shear strain. Test results revealed that, similar to the case of medium-dense Toyoura sand (D_r = 44-50%) previously investigated by the Authors, under reversal stress loading, failure could be associated with liquefaction followed by extremely large deformation during cyclic mobility. Contrarily, under non-reversal stress loading, a progressive accumulation of residual (plastic) deformation brought specimens to failure although liquefaction did not occur. Moreover, the presence of initial static shear does not always lead to a decrease in the liquefaction resistance or strain accumulation of very loose sand. In fact, its resistance can increase or decrease with an increase in initial static shear stress, but it strongly depends on the combination of static and cyclic shear stresses and, thus, on the type of loading. However, under the same magnitude of combined shear stresses applied, very loose sand is much weaker against large cyclic shear strain accumulation than the medium-dense sand.

We present a numerical model to simulate blast induced vibration propagation. The model applies normal pressure to the walls of a cavity, which represents the fractured zone formed by the blasting. The material damping is modeled using the Rayleigh damping formulation. The model is verified through comparisons with two sets of analytical solutions. The comparisons demonstrate that the numerical model accurately predicts the attenuation both at near-field and far-field. The near-field attenuation is not greatly influenced by the damping, whereas the far-field vibration is dependent on the material damping. The far-field attenuation also is shown to be sensitive to the frequencies selected for the Rayleigh damping formulation. We propose a procedure to select optimum frequencies for the Rayleigh damping formulation. The proposed numerical model can be used to estimate the dynamic response nearby structures, as well as to develop site-specific source-to-site attenuation relationship.

Recently, a bentonite sheet is often employed in deteriorated earth-fill dams for preventing water leakage. However, it is concerned that a slipping type of failure may occur at the boundary between the bentonite sheet and the fill material in the event of seismic attack. In this paper, seismic behavior of a small-sized earth fill model inside which a bentonite sheet was mantled was examined in shaking table test, and it was simulated by numerical analysis. It is manifested in the shaking table test that a shallow slip surface developed in the fill above the bentonite sheet, and no harmful deformation was observed along the bentonite sheet when subjected to seismic motion, noting also that the shallow failure was successfully simulated by pseudo static limit equilibrium analysis. In addition, it was confirmed that the seismic response of the model earth-fill was unaffected by the existence of bentonite sheet as examined by FEM seismic numerical simulations.