Evaluation of Variability in Radionuclide Measurements for the Radionuclide Rule

摘要

In 2000, EPA promulgated a revised final regulation for radionuclides in drinking water. This Rule retained the maximum contaminant level (MCL) at 5 pCi/L for combined Ra226/Ra228 but requires separate monitoring requirements for Ra228. Previously, analysis for Ra228 was required only when Ra226 measurement was greater than 3 pCi/L. With the promulgation of the Radionuclide Rule and completion of the first round of required monitoring, many utilities throughout the country who were previously in compliance have found themselves now uncertain of their compliance status as a result of the added compliance monitoring parameters. In some regions radionuclide activities from groundwater supplies at the same site appear to vary by more than two fold over time. This may be due to natural variability in concentrations of some of the radionuclides, but it is also likely due to issues related to method and laboratory variability. To address this issue MWH evaluated compliance data from a number of utilities who have done repeat monitoring over time and then designed and conducted a blind interlaboratory evaluation of performance for routine samples among 5 multi-state certified labs. This interlab study was designed to address both intralab variation among blind replicate samples at levels near the MCL for radium 226, radium 228 and gross alpha activity, along with assessing laboratory precision and accuracy, and equally important, to examine the interlab variability among well qualified laboratories for the same samples. Participating laboratories were not aware that the study was a blind proficiency test, so samples were processed in the same way that normal compliance samples were processed. A total of 14 samples were submitted to each laboratory in several discrete batches. Each batch consisted of several replicates and also some Youden pairs to determine the single lab precision at applicable concentration ranges for compliance assessment and historical data. Results of the interlab study demonstrate that while Radium-226 measurements appear to be both accurate and precise, even at low levels, radium-228 analysis and gross alpha is much less rugged, resulting in potential false positives and false negatives for compliance determinations.