Fredrik Mossmark, Geo Innova AB (Sweden)
Lars O Ericsson, Chalmers University of Technology (Sweden)
Lars-Olof Dahlström, NCC AB (Sweden)
Hans Hultberg, IVL Swedish Environmental Research Institute (Sweden)
Malin Norin, NCC AB (Sweden)
When considering the degradation process and lifetime of the support system and equipment in underground facilities, the selection of materials is (normally) based on established criteria for the chemical composition of the groundwater. This is important for decisions regarding the steel quality and protection of reinforcement bolts, as well as the material used for the waterproofing system and lining. The criteria are imposed through groundwater sampling and analysis of groundwater prior to the construction of an underground facility. However, studies of the impact on groundwater chemistry from the construction of underground structures and experiments with groundwater extraction indicate that the groundwater chemistry is likely to change over time.
Underground facilities are known to cause hydrological changes, especially during the construction phase. However, extensive monitoring programmes of groundwater chemistry are unusual. To further investigate possible changes of water chemistry due to hydrological changes, an experiment with groundwater extraction has been carried out. The experiment was conducted through the constant extraction of groundwater for a period of five years (between the years 2000 and 2005) from within a small watershed (28000 m2) at Lake Grdsjön, located 50 km north of Gothenburg in Sweden. The area was also monitored during a few years before the extraction started and during the recovery phase. The area of the experiment is characterized by Precambrian crystalline bedrock covered by a thin overburden of glacial till and organic soils. The extraction caused the runoff from the watershed to decrease by nearly 50 % and the groundwater level to fluctuate more than at a nearby reference area.
The hydrological impact of the experiment, with increased groundwater recharge, lead to changes and increased seasonal variations in the chemical composition of the groundwater in the bedrock. The hydrochemical variations were caused by seasonal variations in both the amount of water available for groundwater recharge and the chemical composition of the recharging water. Compared to the reference area, the seasonal variations were observed to increase for all the parameters included in the criteria used by the Swedish authorities for selection of construction materials (pH, hardness (Ca), alkalinity, salinity (EC)). An established method to assess the impact of the water composition on the corrosion of steel materials is the use of Langeliers index. The experiment with groundwater extraction caused a larger fluctuation of Langeliers index in the test area compared to the reference area. The results from the experiment confirm the observations from previous tunnelling projects, and show that the methods commonly used to assess the expected future aggressivity of the groundwater in the planning for underground facilities should be reviewed.