Overview
This community-engaged field study deployed portable voltammetric sensors (a mercury-free “E-Tongue” device) alongside ICP-MS confirmation to monitor lead in tap water across 317 homes in four Massachusetts municipalities. First-liter and fifth-liter sequential samples were collected by residents after 6 h or more stagnation, following a protocol similar to the US EPA Lead and Copper Rule revised 2021 (Trigger Level 10 µg/L). GIS spatial analysis identified geographic clustering of elevated lead samples. The highest single first-liter sample measured 168.2 µg/L (confirmed at 25.1 µg/L in the corresponding fifth liter) in Andover. Fifth-liter exceedances above the action level in Andover indicated potential lead service lines in that municipality.
Key numbers
First-liter results across all four municipalities (317 homes, µg/L):
- Above 10 µg/L (US EPA LCRI action level, 2024): 7 samples (4 from Andover, 2 from Lowell, 1 from Lawrence)
- 5 to 10 µg/L: 3 samples
- 2 to 5 µg/L: 21 samples
- Below 2 µg/L: remainder
Highest single first-liter measurement: 168.2 µg/L (Andover; sensor result); confirmed by ICP-MS at 25.1 µg/L (fifth-liter sample from same home)
Fifth-liter results (317 homes):
- Above 10 µg/L: 3 samples (all Andover, suggesting lead service line presence)
- 5 to 10 µg/L: 2 samples
- 2 to 5 µg/L: 10 samples
Sensor performance: voltammetric E-Tongue MDL 1.6 µg/L; linear range 5 to 200 µg/L; lead recovery 84 to 105 percent; RSD below 10 percent.
Water chemistry by municipality: Andover most alkaline; Lawrence neutral; Lowell and Dracut intermediate.
Methods (brief)
Community-engaged sampling with resident-collected first-liter and fifth-liter samples (1 L per sample) after 6 h or more stagnation, confirmed by certified ICP-MS or graphite furnace atomic absorption if sensor result was above the action level. Portable anodic stripping voltammetry (E-Tongue; mercury-free bismuth electrode). Confirmatory analysis by EPA Methods 200.8 (ICP-MS) or 200.9 (GFAA). GIS spatial analysis using natural neighbor interpolation and kriging.
Evidence fitness
Supports: demonstrating feasibility of community-wide distributed lead monitoring using portable sensors validated against reference methods; establishing baseline lead concentrations across four Massachusetts municipalities; identifying geographic clustering of elevated lead and potential lead service line locations via fifth-liter sampling.
Does not support: establishing a city-level 90th percentile for LCR compliance purposes (self-selected volunteer sample, not a Tier I high-risk site survey); generalizing sensor performance across all tap water matrices without site-specific validation; substituting sensor results for regulatory compliance measurements without ICP-MS confirmation.
Limitations
Residential sampling was volunteer-based, potentially oversampling residents with elevated concern about water quality. Sample size (317 homes across four municipalities) is insufficient for statistically robust municipality-level distributions. The E-Tongue MDL of 1.6 µg/L means that samples below this threshold are non-detect, potentially underestimating low-level exposure. Sensor results require confirmatory ICP-MS when above the action level; the study confirmed a subset, but routine community monitoring with sensors alone would miss the precision needed for regulatory compliance. Stagnation time was self-reported as 6 h or more without confirmation.
Related evidence
Update history
No substantive edit history is available in this build. The full commit record is available in git.