A Digital Twin Prototype for Protecting Surface Source Waters

In water safety planning, drinking water quality is protected by multiple barriers, beginning with source protection before treatment and distribution. Protection requires timely information about conditions beyond directly monitored locations. This study presents a digital twin prototype integrating networked environmental observations, a three-dimensional hydrodynamic water quality model, and source-protection applications at the Brusdalsvatnet reservoir in Norway. The framework connects state estimation for operational monitoring with contamination-scenario analysis and model-informed planning of uncrewed surface vessel sampling. Compared against observations from a single moored station throughout the 2024 season, the modelled water temperature has a root-mean-square error of 1.73 ∘C near the surface and 2.15 ∘C averaged across depth, with the largest discrepancies at intermediate depths in summer. The contaminant scenario and sampling route demonstrate analytical capabilities without field validation. The results establish a proof of concept for source-water decision support and identify remaining challenges to making its predictions reliable.