Groundwater models form the basis for investigating subsurface processes that relate to groundwater flow. Urban cover, however, usually inhibits the collection of new subsurface or geological data. Therefore, models usually depend on existing, poor-quality, or scarce datasets. The geological domain is an integral part of any groundwater model, and as such, understanding the model’s sensitivity to the geological interpretation is key to constraining uncertainty. This research uses a recent advancement in mitigating uncertainty in geological modeling to investigate how different geological interpretations affect groundwater model uncertainty. Using the Ouseburn catchment, Newcastle upon Tyne, UK, as a case study, it estimates baseflows and uses them to develop an ensemble of coupled distributed groundwater recharge and groundwater flow models using SWAc and MODFLOW, and performs a Monte Carlo analysis on the different model formulations. Results indicate that even though river baseflows are not highly affected, there is a connection between simulated groundwater level sensitivity and areas of high geological uncertainty. As the interest in the urban subsurface grows, constraining uncertainty in groundwater models is especially important for urban planning, policy making, water resources, and groundwater flooding protection. Therefore, constraining uncertainty from geological datasets is key to robust groundwater modeling.
Ntigkakis et al. (Mon,) studied this question.