Copenhagen Metro: groundwater control in sensitive urban areas

Abstract

The construction of the Copenhagen Metro commenced in 1997 and the tunnel part in inner Copenhagen was finished in 2002. The metro includes an 8 km long bored twin tunnel, 7 deep stations and 10 emergency shafts. These structures are placed deep into the underlying limestone and well below the groundwater table. A temporary groundwater lowering was unavoidable. The tunnels were constructed by use of earth pressure balanced full face Tunnel Boring Machines (TBMs). Copenhagen is an over 800 years old city founded by the sea. Numerous historical buildings, palaces and other monuments are key attractions for tourists and for the city population. In many parts of the city these structures are founded in fill layers or alternatively on timber piles or other wooden structures that are sensitive to groundwater lowering. The city authorities do not allow any groundwater lowering in the upper aquifers in inner Copenhagen, due to the past record of groundwater lowering induced deterioration of timber piles and settlement damage to inadequately founded buildings. The city authorities are also concerned with the groundwater quality and with the water quality of the harbour and of city lakes. The Frederiksberg municipality, located within the city of Copenhagen, has a local drinking water supply with a catchment area covering a large part of Copenhagen. The water quality within this catchment area is of special concern to both authorities and the public and is therefore closely monitored. In order to provide efficient and economic groundwater control and meet the requirements of the authorities, it was essential to predict the groundwater lowerings that would inevitably occur. Predictions were based on three-dimensional groundwater modelling carried out for both detailed site-specific models and on a regional scale. As many areas were affected by more than one construction site, the predicted values for each site modelled were read back into the regional model. The output from the regional model then provided draw down levels over the entire Copenhagen area. Areas where the draw down was unacceptable could then be identified. The detailed models for the problem areas could then be re-run incorporating discharge reducing measures, such as increased depth of grouting or recharge of groundwater. Modelling was also carried out to show the effect of the construction processes on the groundwater quality. This was a key issue in many of the applications that were needed to obtain permits for tunnelling and construction works from the authorities. Groundwater control was carried out using a combination of cut-off walls and in sensitive areas of recharge of either extracted water or harbour water. Cut-off walls were constructed using secant pile walls and grouting. The optimal depth of the cut-off walls was determined on the basis of the groundwater modelling. Recharge system configuration was similarly designed on the basis of the modelling results. Recharge was carried out by infiltrating harbour water in areas with salty groundwater. In areas with fresh groundwater, extracted groundwater was reinfiltrated after extensive water treatment. Treatment was done to remove iron oxide and fine limestone particles that would otherwise clog filters. The example shows how groundwater control partly based on groundwater modelling helped to meet the challenges posed by a sensitive urban environment and demanding and competent authorities. (A). Reprinted with permission from Elsevier. For the covering abstract see ITRD E124500.