Effects of Cover Depth on Ground Movements Induced by Shallow Tunnelling

"INTRODUCTION Shield tunnelling is often used in constructing underground infrastructure in cities due to the ability to limit settlements and damage to existing buildings. However, in an urban environment with soft overburden and buildings on pile foundations such as the North-South Line project in Amsterdam, there is a tendency to design the tunnel well below the surface and below the pile tip level in order to reduce interaction between tunnelling process and piles. This results in deep tunnels and deep station boxes. When the tunnels are located close to the surface and above the pile tip level, this would reduce the required depth of the station boxes and the construction cost. Moreover, other benefits of shallow tunnels are the low operational cost in the long-term and shorter traveling time from the surface to the platforms. Still, the tunnels should be constructed in such a manner that existing buildings are not structurally damaged, which results in a minimum required distance between tunnelling process and existing buildings. In this paper, the extent of the area that is influenced by tunnelling will be investigated in order to determine the limit distance from tunnelling to existing foundations without inducing too large building deformation. From analysing empirical data of many shield tunnels, Peck (1969) firstly presented the settlement trough on the surface induced by tunnelling in soft soil as a Gaussian distribution. This is also confirmed by other authors (Cording and Hansmire, 1975; Mair et al., 1993; Ahmed and Iskander, 2010). In this study, the Gaussian curve is used to investigate the ground movement when tunnelling in order to find the effects on existing structures. Based on the results from centrifuge test and empirical data, Mair et al. (1993) showed that the subsurface settlement profile distributes as the Gaussian curve also. The width of settlement trough at the depth z depends on the depth of the tunnel z0 and a coefficient K depending on depth. Other studies by Moh et al. (1996), Grant and Taylor (2000) and Jacobsz (2003) based on Mair et al. (1993) proposed a limited change of K in various kinds of soil."