Observations of the ground and structural behaviours induced by a deep excavation in loose sands

Abstract

Large-scale deep basement excavations are commonly used in metropolitan areas; however, previous studies focused primarily on excavations in impermeable clay instead of permeable sand with high groundwater levels. In this study, a unique opportunity provides a complete picture of the ground and structural behaviours induced by deep excavations in loose to medium sand, which have been rarely studied in the past. With a maximum excavation depth of 16.8 m, the pit was retained by a 0.9-m-thick and 32-m-deep reinforcement concrete diaphragm wall with 4-level struts. The maximum wall displacement (δvh) reached 56 mm at the end of excavation, and the ratio of δvh to the excavation depth was 0.32%. In contrast, the ratio of the maximum surface settlement to the excavation depth was lower, i.e. only 0.2%. Pumping inside the excavation site did not appear to induce a significant impact outside the excavation site due to the retaining wall embedded in a layer of impermeable clay at a depth of 28.5–30.5 m. The plain strain ratio from the observed settlement is interpreted; the ratio indicates a difference of 2–39%. This difference may be connected to the soil anisotropy, or the maximum surface settlement at each monitoring section may not be detected. The wide range of the variance might also be connected with the small measured surface settlements and the difficulty of detecting such small displacements on the ground level in field surveys.