Wellhead-protection zones are commonly delineated on the basis of advective travel-time analysis without considering any aspects of model uncertainty. In the past decade, research efforts have produced quantifiable risk-based safety margins for protection zones. These margins are based on well-vulnerability criteria (e.g., travel times, exposure times, peak concentrations) and take model and parameter uncertainty into account. There are three main reasons why practitioners still refrain from applying these new techniques. (1) They fear the additional areal demand of probabilistic safety margins; (2) probabilistic approaches are allegedly complex, not readily available and require huge computing resources, and (3) uncertainty bounds are fuzzy, whereas final decisions are binary. The primary goal of this paper is to show that these reservations are unjustified. We present a straightforward, computationally affordable framework that offers risk-informed decision support for robust and transparent wellhead delineation under uncertainty. We show that reliability levels can be increased by exchanging delineated low-risk areas for previously nondelineated high-risk areas. We also show that further improvements may often be available with only little additional delineated area. As proof of our concept, we illustrate our key points with the example of a pumped karstic well catchment, located in Germany.
© 2014, National Ground Water Association.