Screening houses for vapor intrusion risks: a multiple regression analysis approach

Area of Science:

• Environmental Science
• Toxicology
• Risk Assessment

Background:

• Vapor intrusion, the migration of volatile organic compounds from groundwater to indoor air, is a key exposure pathway at hazardous waste sites.
• Current screening methods, like the U.S. Environmental Protection Agency's (EPA) generic attenuation factor, are often logistically challenging and may not accurately identify all at-risk homes.
• A need exists for more refined screening tools to assess indoor air quality risks associated with subsurface contamination.

Purpose of the Study:

• To develop and validate a screening-level model for predicting the vapor attenuation factor.
• To improve upon the EPA's generic approach by incorporating environmental, contaminant, and household characteristics.
• To reduce the rate of false negatives in identifying homes potentially affected by vapor intrusion.

Main Methods:

• Analysis of the EPA's vapor intrusion database, comprising nearly 2,400 indoor air and subsurface concentration samples from 15 states.
• Development of a multilevel regression model to predict the vapor attenuation factor.
• Evaluation of the model's performance against the EPA's current screening approach.

Main Results:

• The vapor attenuation factor is significantly influenced by soil type, depth to groundwater, season, foundation type, and contaminant molecular weight.
• The developed model demonstrates a reduced rate of false negatives compared to the EPA's generic screening method.
• Site-specific data analysis provides a more nuanced understanding of vapor intrusion dynamics.

Conclusions:

• The refined screening model offers a more accurate assessment of vapor intrusion risk than generic approaches.
• Incorporating key variables improves the identification of homes requiring further investigation for indoor air contamination.
• This research contributes to better public health protection at hazardous waste sites by enhancing exposure pathway assessment.