Identifying and predicting karst water inrush in a deep tunnel, South China

Identifying the source of tunnel inrush is essential for its prediction and mitigation. In this work, hydrological, hydrochemical, and isotopic methods were used to identify the source of water inrush into the deep Xiakou tunnel of south China. In addition, a rainfall-runoff model was calibrated to predict its anticipated runoff volume and temporal characteristics. These methods reveal that water inrush in the northern tunnel originates in the Permian-Triassic karst aquifer to the north, and that a groundwater divide exists between two major springs (Xianglong Spring and Xiakou Spring). The southern tunnel also discharges karst water during heavy rainfall conditions, but mainly drains Jurassic fissure water in the dry season. The peak flows, peak times, and recession processes of five inrush events were effectively simulated with our calibrated, rainfall-runoff model. A catchment area of 16.3 km2 was delineated based on the water inrush point identified by optimal model parameters. The catchment analysis shows that about 53% of rainfall recharges into the water inrush point. These methods presented in this study facilitate the estimation of recharge areas in mountainous karst areas and provide a basis for the design of hydraulic structures for the control and mitigation of tunnel inrush.