Use of literature mining for early identification of emerging contaminants in freshwater resources

Abstract Chemical and microbial contaminants in the aquatic environment pose a potential threat to humans and to ecosystems. Humans may be exposed to contaminants in water resources when used for drinking water production, agriculture, aquaculture or recreation. Climatological, social and demographic changes, as well as the increasing sensitivity of analytical techniques, may result in the augmented detection of contaminants. Recent research has shown that it takes about 15 years from the time of the first scientific study mentioning the presence of a contaminant in the environment for the issue to peak in scientific attention and regulatory action. One possible factor influencing this lengthy period is that the first article becomes lost in the vast number of publications. In this study, we therefore developed a methodology using literature mining to identify the first scientific study which reports the presence of a contaminant in the aquatic environment. The developed semi-automated methodology enables health and environment agencies to inform policy makers about contaminants in the aquatic environment that could be significant for public and environmental health in national, international and river basin settings. The methodology thereby assists the proactive governance of emerging contaminants in the aquatic environment. This was illustrated by a retrospective analysis of the period of emergence in the Netherlands of: (1) perfluorooctanoic acid in surface water, and (2) biological industrial wastewater treatment systems as potential infection sources for Legionnaires´ disease.