Statistical interpretation of a frequency-of-monitoring program designed for the protection of underground uranium miners from overexposure to radon daughters

"This paper presents the statistical analyses of several sets of data acquired in several programs of sampling of radon daughters in uranium mines. The measurements were taken over periods of days or weeks for given locations or location occupations within these mines. Because the sampling applies to specific intervals of time and specific locations, a method is suggested f or the eventual application of a monitoring program which would be suitable on an annual basis for these locations. The present data and analyses thereof are not justifiably applicable to a monitoring regimen on an annualbasis, without a number of additional assumptions.The aim of the present analyses of concentration measurements is the determination of the minimum number of measurements required to achieve an annual radon daughter exposure estimate, for individual workers, having an error no greater than 50% at the 95% confidence level. The criterion of accuracy is taken from a statement in the American National Standard, Radiation Protection in Uranium Mines (ANSI NI3.8-I973), and is currently used, in somewhat modified form, as a guide to uranium mining licensees in Canada. Uncertainties associated with occupancy time are ignored in the present analysis. IntroductionOne of the principal aims of any scheme of radiation protection is that of improving the accuracy of exposure estimation for workers. The example presented here is from the area of radon daughter exposure estimation for uranium miners. The conventional means of estimating workers' exposures is grab sampling of radon daughter concentrations combined with occupancy time weighting in the various work and travel locations. Uncertainties exist in each of these components, and in an environment of highly variable radon daughter concentrations (such as an active uranium mine) one is always uncertain as to the representativeness of the grab sample (compared to the average concentration) unless a sophisticated monitoring scheme has been implemented based on previous experience in such an environment.Improvements in the accuracy and precision of employee radon daughter exposure records depend, primarily, on three factors;"