Notes from Underground: Sampling a Subway System
By Ed Rutkowski
NEW ORLEANS, Louisiana (June 2, 2026)—In 2023, the Washington Metropolitan Area Transit Authority (WMATA) hired a contractor to clean emergency equipment located in underground subway stations. Prior to cleaning, the dust was tested for lead, and one sample, taken from a cabinet that had been relatively undisturbed for roughly 15 years, was found to have a lead concentration over 2,100 micrograms per square foot. As explained by WMATA industrial hygienist Trenell Boggans in an educational session at AIHA Connect, this finding prompted additional testing focused on areas in tunnels where employees were likely to contact dust, such as ventilation shafts and rooms containing mechanical equipment. Concentrations in these samples ranged from below the limit of detection to greater than 2,000 µg/ft2.
The high concentrations, while troubling, were understandable given the nature of subways. The WMATA system opened for riders in 1976. According to current ridership statistics, the system’s 1,500 buses and 1,200 rail cars move roughly 850,000 riders around Washington, D.C., Maryland, and Virginia each day. Still, media reports on subway pollution in transportation systems around the world have increased in recent years, and the finding of lead in the Metro system alarmed some employees and members of the public.
In June 2024, Boggans and his colleague Douglas Fallon set out to determine the components of dust in WMATA tunnels and how it could affect workers and riders. Analysis of bulk samples identified iron, carbon, and silica, among other substances. While lead was present, it was “a drop in the bucket” compared to other substances, Boggans said.
To further investigate, Boggans and Fallon conducted an exposure assessment that involved sampling 14 tasks across 12 similar exposure groups. They found that copper, iron oxide, and manganese were present in most samples, with iron oxide the most prevalent. To ensure their sampling accurately reflected actual exposures, they applied the Bayesian Decision Analysis statistical approach.
“BDA is a tool that helps us account for the uncertainty around our datasets,” Fallon explained. The goal, he said, is to achieve 95 percent confidence that the 95th percentile of exposure is below the relevant occupational exposure limit. OEHS professionals typically have small datasets, and “BDA is purpose-built for that,” Fallon said.
The sampling methodology addressed tasks that included pressure washing of subway tunnels, cleaning of HVAC units, maintenance on components of the ventilation system, replacement of ductwork, demolition of concrete, and torch cutting of rails. Lead was detected in approximately 20 percent of samples, but BDA showed that the concentrations were well controlled compared to the OSHA permissible exposure limit.
Reassured by these findings, Fallon and Boggans also wanted to find out whether Metro riders were being exposed to unhealthy air. They used direct-reading instruments to analyze ambient air at three transfer stations during periods of high ridership. As expected given the number of people who use the WMATA system, readings were slightly elevated for carbon monoxide. Fallon and Boggans were pleased that the time-weighted average concentrations of the contaminants they did find were below the EPA’s National Ambient Air Quality Standards.
There is more work to do: Fallon and Boggans would like to determine the constituents of the roughly 60 percent of sampled tunnel dust that was uncharacterized. They are exploring the possibility of installing particulate monitors, and they would like to apply health-based OELs adjusted for shift length as well as additive or synergistic effects. Ultimately, their goal is to collaborate with other transit organizations and share best practices for reducing dust.
Ed Rutkowski is editor in chief of The Synergist.