Search

The shaking table tests were conducted on small scale embankment slopes to study their behaviors under dynamic conditions. The model embankment slopes were built in a rigid box with a length of 1.0 m, width of 0.90 m and height of 0.48 m. Medium grained local sand with a water content of 3% was used in the study. Three slope angles, 30°, 35° and 40° were considered for the embankment. Three slope heights of 15 cms, 18 cms and 21 cms were considered for each slope to observe the effect of slope height during a ground motion. All the nine slope models were tested under the peak ground acceleration of 0.1 g. The acceleration and deformation at the base and the crest of model slopes were monitored to find the responses of embankments under dynamic condition. The failure surfaces developed during the induced motion were fairly shallow and confined within the slope surfaces. The failure surfaces were circular in the upper part of the slopes. At the toe of the slopes, bulging and translational movements were observed. The amplitude of crest acceleration was found to be larger toward the outward slope direction which implied development of a sliding surface and separation of the soil body near the crest. Amplification in acceleration response was observed with the increase in height of model slopes and slope angle. The numerical simulation of the model tests was performed using a commercial program called FLAG. The numerical analyses modeled the crest settlements reasonably well but did not predict the heaving and outward movement of soil at the toe of the slopes. No distinct slip surface was predicted by the numerical analyses. The amplification of response at the crest computed by FLAG was lower than the corresponding test value. The increase in amplitude of acceleration toward the outward slope direction, observed in shaking table tests, was also not predicted by FLAG. [ABSTRACT FROM AUTHOR]