Failure Prediction of Clay Embankments Subject to Weather-Driven Deterioration

Abstract

Embankments have been widely used in the construction of transportation and flood defense infrastructure. Embankments constructed from clays experience a suite of weather-driven deterioration processes that lead to a progressive loss of hydromechanical performance and eventually to failures whose consequences can be severe and costly. This study aimed to predict the time to failure of aging, deteriorating clay embankments supporting transportation infrastructure. A multiphase numerical modeling approach developed to simulate the long-term weather-driven hydromechanical behavior of clay embankments was used to model and simulate the behavior of a number of well-documented embankment failure case histories with known service life and available information to derive the necessary soil properties and climate records. Numerical models were developed for a total of 34 case studies, and numerical simulations were performed to predict the time to failure of the embankments due to progressive, long-term, weather-driven deterioration. Predictions compared well with actual times to failure reported for the simulated case studies. Further, geometries of modeled slides compared well with those measured for actual slides. The models offered insights into the conditions associated with failure, such as displacement patterns and pore pressure conditions. It was observed that as near-surface soil experiences cycles of wetting and drying, and accompanying cycles of swelling and shrinkage, irrecoverable swelling accumulates with time, causing swelling-induced slides. Overall, the numerical modeling approach was proven effective in producing data necessary to develop deterioration models that could improve infrastructure asset management. This study underscores the importance of understanding and predicting the long-term performance of clay embankments in civil engineering infrastructure. Through the use of long-term numerical modeling, this study could predict the time of weather-driven shallow slides in clay embankments. The ability to predict failures of embankments can offer significant improvements in infrastructure asset management. Predictive models can offer early warning to potential imminent failures, allowing for timely intervention to minimize the risk of sudden, unexpected failures, especially where consequences can be severe and costly.