Effects of Depth-Dependent Uncertainty of Material Properties on Deformational Responses Induced by Tunnels

Urban tunneling is increasingly adopted to meet the growing demand for underground transportation infrastructure in densely populated cities. However, the unavoidable ground settlements induced by tunneling can threaten tunnel integrity and stability of adjacent structures. It is commonly acknowledged that Young’s modulus of soil is one of the most sensitive parameters for tunneling-induced deformational responses. The conventional methods always use either deterministic values or depth-independent spatial variability characteristics of Young’s modulus and rarely consider its depth-dependent nature, which is consistent with field observations. This study investigates the degree of influence of depth-dependent uncertainty of Young’s modulus on deformational responses in tunnel systems. A depth-dependent random field model that reflects specific nonstationary characteristics of Young’s modulus is adopted to more rationally describe the actual soil characteristics and enhance the accuracy of settlement predictions while improving the safety estimations of nearby built structures. Results indicate that depth-dependent material uncertainty significantly influences the magnitude and aggravates the uncertainty in deformational responses. Because probabilistic analysis with spatially variable properties is computationally intensive, this study also investigates the possibility of simplifying random field theory-based probabilistic analysis to a deterministic problem with the development of design charts that can give equivalent probabilistic responses for different scenarios of material uncertainty, reducing the need for computationally intensive and time-consuming simulations. Furthermore, the proposed design charts are validated using a real case study of the Saket Station tunnel, New Delhi.