Browsing by Author "Rahikkala, Sonja"

The construction of the Savio railway tunnel caused an unexpectedly large decrease in the groundwater heads in part of the observation points in the vicinity of the railway tunnel after the completion of the tunnel construction work. The study aimed to investigate the ability of the groundwater model to forecast such groundwater drawdown caused by a railway tunnel in the urban areas of East Hakkila, Kaskela and Hakunila. The uncertainty of the groundwater model and its sensitivity to certain parameters were studied and analyzed. The groundwater model was built using the GMS (Groundwater Modeling System) numerical groundwater modeling program. For the groundwater flow model, the bedrock mesh was built in Leapfrog Geo - program based on gravimetric measurements and rotary drilling. The groundwater model was calibrated manually and automatically to correspond to the groundwater heads measured in the area. The circumstances after the construction of the railway tunnel were then simulated. The model-calculated groundwater heads were compared to the observed groundwater heads and the observed tunnel leakage. Sensitivity analysis was performed for the horizontal conductivity parameters and recharge parameters of the model. The groundwater model before the construction of the tunnel mimicked the measured groundwater heads and flow directions in the study area. The simulation of the time after the construction of the tunnel caused groundwater drawdown of 0-16 meters at 0-730 meters from the railway tunnel. In the simulation of the time after tunnel construction, the groundwater heads calculated by the model differed by 0,01-24 meters from measured groundwater heads. According to the results of the sensitivity analysis, the model was most sensitive to changes in the horizontal conductivity of the upper bedrock layer and granite. The recharge parameters and soil layer had a marginal impact on the groundwater model. The uncertainty of the model was affected by the few measurement points in relation to the size of the area and the absence of field measurements of the hydraulic conductivity of the parameters. To reduce the uncertainty, the model should be calibrated based on the measured conductivity values of the bedrock. Despite the uncertainty, computational models give indications of the direction of change and are important tools in questions related to nature and the built environment.