%0 Journal Article
%K Environmental justice
%K Machine learning
%K Lead in school drinking water
%K Lead leaching
%K Public data mining
%A G.P Lobo
%A Johanna Laraway
%A Ashok J Gadgil
%B Science of the Total Environment
%D 2021
%G eng
%R https://doi.org/10.1016
%T Identifying schools at high-risk for elevated lead in drinking water using only publicly available data
%U https://www.osti.gov/pages/biblio/1821159
%V 803
%8 08/2021
%X <p>Estimating the risk of lead contamination of schools' drinking water at the State level is a complex, important, and<br />unexplored challenge. Variable water quality among water systems and changes in water chemistry during dis-<br />tribution affect lead dissolution rates from pipes and fittings. In addition, the locations of lead-bearing plumbing<br />materials are uncertain. We tested the capability of six machine learning models to predict the likelihood of lead<br />contamination of drinking water at the schools' taps using only publicly available datasets. The predictive fea-<br />tures used in the models correspond to those with a proven correlation to the dominant, but commonly unavail-<br />able, factors that govern lead leaching: the presence of lead-bearing plumbing materials and water quality<br />conducive to lead corrosion. By combining water chemistry data from public reports, socioeconomic information<br />from the US census, and spatial features using Geographic Information Systems, we trained and tested models to<br />estimate the likelihood of lead contaminated tap water in over 8,000 schools across California and Massachusetts.<br />Our best-performing model was a Random Forest, with a 10-fold cross validation score of 0.88 for Massachusetts<br />and 0.78 for California using the average Area Under the Receiver Operating Characteristic Curve (ROC AUC) met-<br />ric. The model was then used to assign a lead leaching risk category to half of the schools across California (the<br />other half was used for training). There was good agreement between the modeled risk categories and the actual<br />lead leaching outcomes for every school; however, the model overestimated the lead leaching risk in up to 17% of<br />the schools. This model is the first of its kind to offer a tool to predict the risk of lead leaching in schools at the<br />State level. Further use of this model can help deploy limited resources more effectively to prevent childhood<br />lead exposure from school drinking water.</p>