Abstract: A laboratory physical model test was carried out to investigate the impact of underground structures on pumpinginduced land subsidence, in which the underground structure was represented by a concrete plate. The vertical displacement and pore water pressure were documented through sensors, from which temporal and spatial distribution patterns of vertical displacement and pore water pressure had been analyzed. It has been shown that the presence of the plate and the depth of the plate in soil have significant influence on the values and distribution patterns of vertical displacement and pore water pressure, and so does the distance of the plate to the pumping well. The presence of the plate blocks the water flow in soils and impedes water replenishment from lateral sources, prolonging the process for the pore water pressure and soil displacement to reach a stable state. It also increases the subsidence of the soil layer and the drop in pore water pressure. The subsidence increasingly changes with the relative depth of the plate in soil, however, the relationship between them is nonlinear and with a critical value (85% in this experiment). When the relative depth of the plate is smaller the critical value, the maximum subsidence of the sand layer changes gently with the depth of plate in soil, whereas it varies significantly when the relative depth of the plate is greater than the critical value. In the numerical simulations of land subsidence in the presence of underground structures, consideration should not only be given to the obstructive effects of the structures on water flow but also to their hindrance to soil displacement. The findings of this paper can provide valuable insights for the prediction and mitigation of pumping-induced land subsidence.

Key words: underground structures, laboratory model experiment, land subsidence, pumping, barrier effect