SummaryThis study draws a relationship between filter mass loading, percent loss using the mass loading data collected, and previous studies of self-absorption. The mass loading consists of particulate dust, radioactive particulates, and filter material. A study by Higby [1984] calculated a minimum burial depth for an alpha particle to be lost due to absorption (100% loss) of about 3. . Mass loadings in this latter study included dust loading plus the mass of the front layer of filter. This study examined light dust loadings (averaging about 0.1 mg cm -2 ) on filter material and compared this with other literature data to estimate losses at typical mass loadings on filters from PNNL sampled exhaust sites.During an 18-month period, 116 samples were collected and analyzed from 8 different building stacks. Under normal operating conditions at the stacks monitored by Effluent Management, the mass loading of sample filters averages 0.09 + 0.12 (2σ) mg cm -2 (excluding negative values and outliers) and ranges from 0 mg cm -2 to 0.24 mg cm -2 . This study presents two different methods of relating percent loss due to selfabsorption to filter mass loading: exponential and linear relationships based on data from Luetzelschwab et al. [2000] and Higby [1984]. In general, samples have losses of less than 19% using the conservative exponential model and less than 7% using the linear model; therefore, a correction factor of 0.85 remains conservative.For higher accuracy, the Effluent Management group recommends that each filter be weighed before and after installation on the sampling system. Having tare weights and gross weights allow the mass loading of each filter and any applicable correction factors to be determined on a case-by-case basis.