(Mis)Classification of Water Systems in the United States

The definitive federal census and repository for regulated water system data in the United States, universally used by researchers and policymakers, is the Safe Drinking Water Information System (SDWIS). We consider the veracity of the data used to identify and differentiate water systems. This research proposes a more exact classification scheme and typology, with a structural-function logic, and applies it to the community water systems (CWSs) in the Great Lakes region. Applying our proposed typology of water systems, the analysis highlights the striking asymmetry between the presence of governmental primary systems and nongovernmental ancillary systems, as well as the hybrid and interconnected nature of many water systems. The ideal action is not for researchers to try to fix the data but for USEPA to fix the system by which the data are generated. We provide policy recommendations for renovating SDWIS to improve system accountability, research credibility, and policy development