Impact of urban geology on model simulations of shallow groundwater levels and flow paths

Abstract

This study examines the impact of urban geology and spatial discretization on the simulation of shallow groundwater levels and flow paths at the city scale. The study uses an integrated hydrological model based on the MIKE SHE code that couples surface water and 3D groundwater simulations with a leaky sewer system. The effect of the geological configuration was analyzed by applying three geological models to an otherwise identical hydrological model. The effect of spatial discretization was examined by using two different horizontal discretizations for the hydrological models of 50 and 10 m, respectively. The impact of the geological configuration and spatial discretization was analyzed based on model calibration, simulations of high water levels, and particle tracking. The results show that a representation of the subsurface infrastructure, and near-terrain soil types, in the geological model impacts the simulation of the high water levels when the hydrological model is simulated in a 10 m discretization. This was detectable even though the difference between the geological models only occurs in 7 % of the volume of the geological models. When the hydrological model was run in a 50 m horizontal discretization, the impact of the urban geology on the high water levels was smoothed out. Results from particle tracking show that representing the subsurface infrastructure in the hydrological model changed the particles' flow paths and travel time to sinks in both the 50 and 10 m horizontal discretization of the hydrological model. It caused less recharge to deeper aquifers and increased the percentage of particles flowing to saturated-zone drains and leaky sewer pipes. In conclusion, the results indicate that even though the subsurface infrastructure and fill material only occupy a small fraction of the shallow geology, it affects the simulation of local water levels and substantially alters the flow paths. The comparison of the spatial discretization demonstrates that, to simulate this effect, the spatial discretization needs to be of a scale that represents the local variability in the shallow urban geology.