H53E-1707
Modelling water table drawdown and recovery during tunnel excavation in fractured rock: estimating environmental impacts and characterizing uncertainties in a heterogeneous domain
Friday, 18 December 2015
Poster Hall (Moscone South)
Jon Sege1, Yandong Li2, Ching-Fu Chang1, Jianqin Chen3, Ziyang Chen2, Yoram Rubin1, Xiaojun Li3, Zhu Hehua3, Changhong Wang1 and Carlos A Osorio-Murillo4, (1)University of California Berkeley, Berkeley, CA, United States, (2)Tongji University, Department of Geotechnical Engineering, Shanghai, China, (3)Tongji University, Shanghai, China, (4)Brigham Young University, Civil and Environmental Engineering, Provo, UT, United States
Abstract:
This study will develop a numerical model to characterize the perturbation of local groundwater systems by underground tunnel construction. Tunnels and other underground spaces act as conduits that remove water from the surrounding aquifer, and may lead to drawdown of the water table. Significant declines in water table elevation can cause environmental impacts by altering root zone soil moisture and changing inflows to surface waters. Currently, it is common to use analytical solutions to estimate groundwater fluxes through tunnel walls. However, these solutions often neglect spatial and temporal heterogeneity in aquifer parameters and system stresses. Some heterogeneous parameters, such as fracture densities, can significantly affect tunnel inflows. This study will focus on numerical approaches that incorporate heterogeneity across a range of scales. Time-dependent simulations will be undertaken to compute drawdown at various stages of excavation, and to model water table recovery after low-conductivity liners are applied to the tunnel walls. This approach will assist planners in anticipating environmental impacts to local surface waters and vegetation, and in computing the amount of tunnel inflow reduction required to meet environmental targets. The authors will also focus on managing uncertainty in model parameters. For greater planning applicability, extremes of a priori parameter ranges will be explored in order to anticipate best- and worst-case scenarios. For calibration and verification purposes, the model will be applied to a completed tunnel project in Mount Mingtang, China, where tunnel inflows were recorded throughout the construction process.