It is generally accepted that anthropogenic activities have brought about the unequivocal warming of the earth. This reality has forced researchers and geotechnical engineers to enhance their fundamental understanding of soil-atmosphere processes. Anticipated changes in climate conditions such as extreme temperatures and rainfall will likely lead to changes in soil behaviour which will influence geotechnical works such as unsaturated embankments. Unsaturated soil strength is fundamentally governed by moisture content where wetter soils are generally weaker. Temperature has been shown to significantly influence the shear strength and volumetric behaviour of soil, however, its impact is seldom studied or included in designs. As such, altered temperature and rainfall loads will impact geot...[ Read more ]

It is generally accepted that anthropogenic activities have brought about the unequivocal warming of the earth. This reality has forced researchers and geotechnical engineers to enhance their fundamental understanding of soil-atmosphere processes. Anticipated changes in climate conditions such as extreme temperatures and rainfall will likely lead to changes in soil behaviour which will influence geotechnical works such as unsaturated embankments. Unsaturated soil strength is fundamentally governed by moisture content where wetter soils are generally weaker. Temperature has been shown to significantly influence the shear strength and volumetric behaviour of soil, however, its impact is seldom studied or included in designs. As such, altered temperature and rainfall loads will impact geotechnical works and lead to long-term seasonal deformations that may cause severe safety and maintenance problems. These long-term impacts considering temperature are neither well studied nor well-understood and should be investigated.

Centrifuge modelling has the advantage of capturing the stress-dependent behaviour of soil and seasonal soil behaviour can be modelled in a single test. A key contribution of this study is a newly developed centrifuge environmental chamber (CEC), employed to simulate different climate conditions in the centrifuge. Improving on the current state-of-the-art the new CEC can separately control temperature, relative humidity (RH), wind, radiation and rainfall in-flight. A humidity nozzle was introduced which permits control of RH by manipulating the air humidity ratio. The ability of the new CEC is validated, and it is possible to maintain a constant temperature while varying RH and vice versa.

Employing the newly developed CEC, four series of centrifuge tests were carried out to assess the influence of climate conditions on embankment performance, with a focus on temperature and RH effects. The impact of wetting-drying cycles at constant elevated temperature on a clay and silt embankment was investigated. Then, the impact of separate temperature and RH cycles on a clay embankment was investigated. Furthermore, tests on the impact of thermal cycles on the performance of clay and silt embankments compacted at different densities were carried out. In all the tests, embankments were instrumented to measure pore-water pressure, temperature and settlement. The results of the thermal cycle tests were back-analysed using Vadose/W 2012 through a transient thermo-hydraulic analysis. Thereafter, numerical parametric studies were carried out to assess the impact of saturated permeability (k_s) and the initial degree of saturation (S_r) on embankment performance.

Wetting-drying cycles at elevated constant temperature show heave for the clay embankments during initial heating, likely due to the expansion of pore-water and soil particles. Once the temperature stabilised thermal softening became the dominant mechanism as the increase in temperature reduced the preconsolidation pressure. In contrast, continuous settlement is observed for the silt embankment. This is because of heating-induced contraction, due to a reduction in preconsolidation pressure that leads to settlement.

The influence of separate temperature and RH cycles on a clay embankment show more significant shrink-swell behaviour during temperature cycles. This is because volumetric strains during temperature cycles are due to thermal expansion of both soil particles and water, while RH cycles only affect water (for the range of temperature and RH imposed). Thermal softening resulting in rearrangement of soil particles cause overall settlement of the embankment. In addition, radiation results in excess heave because of increased thermal expansion of the soil due to a higher temperature gradient and deeper influence depth.

The impact of thermal cycles on the performance of clay and silt embankments compacted at densities show that for the number of thermal cycles imposed clay embankments settle more than silt embankments. This is because thermal volumetric strains are less for silt and would suggest that silt embankments are more resilient against thermal cycles. The service life of clay embankments may be reduced to a quarter of the design life due to thermal cycles, while the effect on silt embankments is not as significant. Calculated secondary compression results show that secondary compression coefficients (C_α) for clay embankment should allow for temperature affects, while typical values for silt may be acceptable. A distinctly different differential settlement trend was observed for clay embankments, where the crest settle more, compared to clay embankments subject to wetting-drying cycles. Silt embankment show less significant differential settlement, because of its higher permeability, allowing the embankment to dry more uniformly leading more uniform stiffness changes.