Mapping of Mean Radon Concentrations, Using Survey Data and Covariates

Although elevated indoor radon concentration can occur almost anywhere, some areas of the U.S. have a much higher fraction of homes with elevated levels than do other areas. Knowledge of what areas tend to have elevated indoor concentrations would allow education and regulatory measures (such as required housing construction techniques) to be focused on populations most at risk from radon. In this paper, we discuss a quantitative approach to predicting annual-average living area radon concentration distributions by county, using survey data (including both short-term basement measurements and long-term living-area measurements), geologic province boundaries from the U.S. Geological Survey, aerometric radium measurements, and climate information. The resulting predictions are unbiased, quantitative, and have known uncertainties that vary from county to county depending on the available information within each county. Computer programs to implement the statistical methods, as well as databases and documentation, are being made available through Lawrence Berkeley National Laboratory.