Experimental Evaluation of Internal Erosion Using Groove Erosion Test

Abstract

Internal erosion by piping causes failure in earthen structures by widening the preexisted cracks that eventually lead to the breach formation being subjected to seepage. This complex behavior was studied by developing a new experimental device named groove erosion apparatus. Until now, the cohesive sediment detachments have been mostly modeled by using a linear excess shear stress approach. A recent study has shown limitations in predicting the erosional behavior of all soils with the available approach because of the varied soil response during erosion while modeling cracks. Therefore, the nonlinear excess shear stress and Wilson models came into the literature. Hence, an attempt was made to identify the reliability of both linear and nonlinear models and to measure the critical shear stress and erosion rate index of different soils at different water content values. It was observed that with the increase in fines the critical shear stress of the soil increases almost linearly for excess shear-stress models. However, the erosion rate index showed different trends when plotted against percentage of fines for linear and nonlinear models. Mutiregression analysis was employed to establish the relation between critical shear stress and engineering properties of soil. Results suggested that no single model can predict the erosional behavior for all soil types.