Performance of foundations in a rising groundwater environment
In recent years an increase in the groundwater level in the basal aquifer beneath London has been observed. The result of this water level rise, if it were to reach equilibrium levels of two centuries ago, would be to cause a reduction in effective stress levels in the founding strata beneath London. The effect that such an increase in pore pressure would have on foundations in overconsolidated clay was investigated. The performance of foundations in stiff clay during a rising ground water event was investigated by means of centrifuge model tests. The model tests included comparisons of the behaviour of bored piles with different factors of safety on load, piles with different length to base diameter ratios, comparison of shallow and deep foundation behaviour and the effect of different initial pore pressure distributions. In two tests piezocone tests were carried out in low and high pore water pressure regimes. Triaxial testing and numerical analyses were used to provide information for use in analysis of the centrifuge test results. The main findings of the project were: The geometry of a pile foundation (slender or under-reamed) and the manner by which load is transferred from pile to soil were seen to effect pile settlement relative to the ground surface during a rising groundwater event. Piles which require mobilisation of end bearing resistance at working loads will typically settle more than predominantly friction piles of the same length. For similar geometry piles the initial factor of safety will effect settlement during a rising groundwater event. Piles with lower initial factors of safety settle more than those with higher initial factors of safety during a rising groundwater event. Differential settlements between shallow and deep foundations were almost entirely due to the deep foundation settlement relative to the heaving ground surface where there was a surface perched water table. Where there was no perched surface water both shallow and deep foundations settled relative to the surface. Soil heave, in this latter case, was largely due to the high percentage loss in vertical effective stress near the surface compared to the case where a perched water table existed. Piled foundation load capacity was seen to reduce as a result of a rising groundwater event. Base capacity, measured under largely drained conditions, was seen to be linearly related to the mean normal effective stress in the ground as was cone end resistance of piezocone tests carried out a slow penetration rates. The piezocone tests also showed that the percentage loss in drained end bearing esistance was larger that the loss in undrained resistance. Finite element analyses investigated pile installation effects for model and prototype piles and the effect that they have on pile behaviour during a rising groundwater event. The results have shown that, on a smooth pile surface, the prototype piles will suffer a larger percentage reduction in shaft capacity than the model piles. The analyses were valuable for assessing the applicability of the centrifuge model data to prototype situations.