Physical modeling of soil uncertainty

Abstract

This paper presents a new technique to prepare reduced scale soils models with controlled spatial variability. Soil spatial variability is obtained analytically by using a discrete random field generation method. Then, heterogeneous models are constructed by using nine homogeneous soils, which are created by mixing different percentages of kaolin and bentonite at water content of 1.5 times the liquid limit. Next, the heterogeneous models are subjected to a consolidation process that is developed in twostages: oedometric path and geotechnical centrifuge. Finally, the mechanical soil properties of the heterogeneous models were estimated by measurement of the unit weight and the water content of the random field elements. Results clearly show that the mechanical properties do not only depend on its intrinsic behavior, but also on its space locations. The method proposed in this paper offers new possibilities for the study of the effect of soil variability on the behavior of geotechnical structures. © 2014 Taylor & Francis Group.