Impact of urban geology on shallow groundwater

Increasing urbanization and climate-change-related measures have resulted in a growing demand for knowledge of the subsurface beneath cities and urban water management. Yet, knowledge of urban subsurfaces is not well documented and the urban anthropogenic geology's impact on groundwater is poorly understood. This study examines the impact of urban geology on the water balance and the dynamics of shallow groundwater at city-scale.  

An integrated surface-subsurface hydrological model was developed based on the MIKE SHE code for an urban domain in Odense, Denmark, covering an area of 10 km². In addition to basic hydrological processes, the model included urban processes in the form of overland drainage based on the degree of paved area, perimeter drains around major buildings, subsurface drainage, leakage from the sewer system, and groundwater abstraction. Three geological models were tested as input to the hydrological model. The hydrological models were run with two different horizontal resolutions, respectively a grid size of 10x10 and 50x50 m. The three geological models varied in complexity and representation of the near-surface urban geology: (1) V0, the base model, represented the layered regional geology beneath the urban area. (2) V1, a revised version of V0, included a representation of subsurface infrastructure; road and railroad base and embankment material, basements, and utility trenches. (3) V2, a revised version of V1, in addition included data from shallow geotechnical boreholes, yielding a representation of local areas with fill material. The urban near-surface geology in V1 and V2 were represented in a voxel model with sand/clay fraction classes. All versions of the hydrological model were calibrated based on the same setup, objective functions, and a calibration dataset consisting of 53 hydraulic head time-series and stream discharge observations within the model domain.

The results showed that the heterogeneity was smoothened when the hydrological model included a complex near-surface urban geology in a 50x50 m grid size and thus an effect of the urban geology was not reflected in the simulated head or the water balance. Meanwhile, the near-surface complexity in the V1 and V2 models led to a better model performance in terms of mean error and annual amplitude error, when the hydrological model had a 10x10 m grid size, which is closer to the scale of the heterogeneities.

The study illustrates that the manmade urban geology, in terms of subsurface obstacles and utility trenches, impacts shallow groundwater dynamics and flow paths. Moreover, it documents that it is possible to represent heterogeneous urban geology in a city-scale model, given the data is available. The results suggest that to simulate the effect of urban geology on shallow groundwater the computational grid needs to be of a size that can resolve the main subsurface infrastructures. In conclusion, a representation of the urban near-terrain geology improves the simulation of shallow groundwater and thus provides a better basis for urban planning, water management, and transport modeling.