Abstract: The “groundwater pumping type” earth fissure disasters caused by excessive exploitation of groundwater widely occur worldwide. Their developments are contributed by both excessive exploitation of groundwater and complex geological and
hydrogeological conditions. The type of earth fissure in aquifer system with abrupt thickness change is a general type of earth fissures. In order to achieve a deeper understanding of the mechanism of this type of earth fissures, the paper designs a numerical experimental test in which earth fissure developes in an aquifer system with abrupt thickness change. Firstly, the finite element method is used to calculate the stress-strain distribution of the aquifer system, and the interface element method is used to calculate rupture development. Then the Sobol global sensitivity analysis method is applied to quantify the impact of four control factors, including aquifer thicknesses ratio, changes in pore water pressure, compression coefficient, and cohesion, on the depth of earth fissure development. The results of numerical test indicate that during the process of pumping groundwater, due to the decrease of pore water pressure, differential compaction and shear stress concentrate continuously increase at the location with abrupt changes in aquifer thickness, which leads to the formation of earth fissure on the ground and further downward expansion. The sensitivity analysis results indicate that the aquifer thicknesses ratio is the most sensitive control factor for the development of earth fissures, the sensitivity of pore water pressure changes is second, the sensitivity of compression coefficient is weak, and cohesion is an insensitive control factor. There is an obvious collaboration between the thicknesses ratio and the changes in pore water pressure. The greater the difference in thickness between aquifers, combined with the greater decrease in pore water pressure, the deeper the development of earth fissures.

Key words: Earth fissure, abrupt thickness change, sensitivity analysis, Sobol method, numerical simulation