ABSTRACT
Groundwater lowering in the inner city of Copenhagen, Denmark, is generally not permitted due to the sensitive nature of the foundations of existing buildings, and to avoid spreading existing groundwater contamination. Outside the inner-city, groundwater lowering is often undesirable due to risks towards settlement of infrastructure, nature and contamination. To ensure dry conditions during construction of shafts and other large excavations, and to protect nearby buildings as well as the environment, groundwater control systems have been established in each of the 28 underground stations/shafts of the Cityringen metro, including the Nordhavn and Sydhavn extensions. The temporary groundwater control systems have comprised of cut-off walls and a network of pumping and recharge wells. There is often a notable discrepancy between the discharge rates predicted by modelling in the design phase, and those measured during construction. Possible explanations for this discrepancy include differences between the modelled design and the actual constructed dewatering system, quality of the cut-off wall, inadequate ground investigations or misinterpreted results, pressure-dependent ground permeability and uncertain assumptions regarding anisotropy. This paper describes the basis for the dewatering activities and the outcome of the three-dimensional (3D) groundwater models developed during the detailed design and compares this with the observed groundwater monitoring data. Inconsistencies between the design and observed behaviour of the groundwater control systems are reviewed and the potential reasons are investigated.
