Parameter Estimation for Infiltration in Unsaturated Soils Using a Bayesian Approach

Abstract eng:
Infiltration plays a significant role in the stability of slope under rainfall conditions. The calculated pore water pressure profiles using soil parameters from limited samples are significant different from the field measurements. In this study, a probabilistic method is proposed for estimating soil parameters of unsaturated soils using the field observation of porewater pressures. The proposed method is formulated based on the Bayesian framework and solved using the Markov Chain Monte Carlo simulation method. A case study of rainfall infiltration in a natural terrain site in Hong Kong is presented to illustrate the proposed method. Posterior distributions of the unsaturated soil parameters, initial condition and boundary conditions are obtained. It is found that almost all the parameters are well identified within a small region. The correlations among the three unsaturated soil parameters are also presented.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket