Christian Moeck1, John Molson2, Mario Schirmer1,3
1Eawag, Swiss Federal Institute of Aquatic Science and Technology, Dübendorf, Switzerland (Christian.email@example.com); 2Département de géologie et de génie géologique, Université Laval, Québec City, Canada; 3Centre of Hydrogeology and Geothermics (CHYN), University of Neuchâtel, Neuchâtel, Switzerland
A Null-Space Monte-Carlo (NSMC) approach was applied to account for uncertainty in the calibration of a three-dimensional groundwater model of a major water supply system in Switzerland. The NSMC approach generates different parameter realizations of the hydraulic conductivity field using the pilot point methodology. Subsequently, particle tracking (PT) was applied to each calibrated model, and the resulting particle distribution is interpreted as the spatial streamline density distribution of multiple sources. The adopted approach offers advantages over classical PT which does not evaluate uncertainty originating from the description of subsurface heterogeneity. It is demonstrated that uncertainty in the description of subsurface heterogeneity strongly influences streamline spreading distributions. This spreading in streamlines is evident in locations where the information content of the head observations does not sufficiently constrain the estimated parameters. Even though uncertainty in the predictions is apparent, the pumped drinking water is most likely artificially-infiltrated groundwater originating from the local artificial infiltration system. Only the central and western part of the well field are predicted to potentially extract regional groundwater. Although the probability is small, mixing between young artificially-infiltrated water and regional groundwater was observed in the central locations in the field. The fact that only a few realizations resulted in streamline paths that support the field observations supports the need for a rigorous type of uncertainty assessment.