A CASE FOR MORE DATA TO ASSESS VARIABILITY AND LONG-TERM TRENDS FOR TCE IN AMBIENT AND INDOOR AIR

In 1999, residents living near a former electroplating facility in Asheville, North Carolina (USA) noticed an oily substance in a wetland-spring complex they had been using as a drinking-water source. Subsequent sampling revealed trichloroethylene (TCE) concentrations exceeding 20,000 ppb in the springs and 30,000 ppb in groundwater upgradient from the springs. Further studies revealed over a dozen individual seeps, which have been sampled irregularly over the past 17 years. Spring flow increases or decreases—some seeps dry up completely—following long periods of higher or lower precipitation; aqueous TCE concentrations have fluctuated over several orders of magnitude.

Much of the TCE volatilizes as it discharges at the springs, raising concerns about both ambient outdoor air quality and vapor intrusion near the site. One residence, located about 20 meters from the edge of the contaminated spring complex, is particularly vulnerable. A 24-hour time-integrated sample collected from the crawlspace in December 2007 contained 20.5 µg/m³ TCE. Concentrations in the crawl space were lower in August 2008 (7.4 µg/m³), but similar to indoor living-space air (6.8 µg/m³) and outdoor ambient air (8.6 µg/m³). TCE vapors were not detected in soil gas samples collected below the crawl space, indicating volatilization from the contaminated spring complex was the primary source for vapor intrusion. Residents were evacuated in June 2014 after a reading of 11 µg/m³ and a decision by EPA Region 4 to enforce a lower limit for indoor residential air. Contractors installed a vapor control system later that year to promote volatilization and capture and treat TCE vapors. While largely successful at reducing ambient TCE levels (and therefore indoor air as well), increased concentrations during monthly sampling in late 2015 indicated a need to make adjustments to the system.

This case study demonstrates the need for early recognition of potential vapor pathways, a sufficient spatial distribution of samples, and increased frequency of sampling. Sampling protocols need to remain flexible so they can be adapted for site-specific conditions and adjusted in response to new information.