What: This event is designed to create opportunities for data discussion and idea sharing among graduate students and other researchers working on research related to "water".
Presenter: Mike Blackhurst
Description: Lead service lines are the primary source of lead in drinking water. However, information on the locations of lead service lines is often missing or inaccurate, making it difficult to understand and reduce exposure risks. Customers often elect to sample for lead in their drinking water, but little is known about the relationship between sample results and service line materials. How should customers interpret these sample results? Can municipalities use tap water samples to guide service line replacements? This study merges approximately 1,800 tap water samples from 1,700 customers in Pittsburgh, PA to explore the relationship between service line materials and tap water lead concentrations. Linear models with and without repeated measures and with and without censoring at detection thresholds were developed. Results indicate moderate correlations between tap water concentrations and service line materials, controlling for seasonality. Lead levels are higher in June, July, and August, indicating a potential seasonal water quality effect. Counts of multiple detections or non-detects in tap water samples alone – with no other covariates – were highly predictive. Customers at which lead was detected at least twice are highly likely (>98%) to have lead in the service line, and customers at which lead was not detected at least twice are unlikely (>92%) to have lead in their service line. These results suggest that repeated customer tap water samples may be a helpful indicator of lead service lines and that sampling between June and August may be most informative.
Join Zoom Meeting
Meeting ID: 916 5706 3673
One tap mobile
8778535247,,91657063673# US Toll-free