Search

A new method for calculation of isotope-specific activities and activity concentrations in measurement systems for atmospheric radioxenon is presented. The method results in simple matrix-vector equations, and requires the definition of fewer spectral regions-of-interest than previous algorithms. The most important difference compared to the current method is however the calculation of decision limits, which results in false detection rates closer to the selected confidence level of 95% compared to the methods used today. This is achieved by introducing a Bayesian correction of the background estimate. The results have implications for the understanding of the atmospheric radioxenon background, for example for the observed low levels of 133mXe, an important isotope in the area of nuclear explosion detection.

Motivated by the need for consistent use of concepts central to the reporting of results from measurements of atmospheric radioactivity, we discuss some properties of the methods commonly used. Different expressions for decay correction of the activity concentration for parent-daughter decay pairs are presented, and it is suggested that this correction should be performed assuming parent-daughter ingrowth in the sample during the entire measurement process. We note that, as has already been suggested by others, activities rather than activity concentrations should be used when nuclide ratios are calculated. In addition, expressions that can be used to transform activity concentrations to activity ratios are presented. Finally we note that statistical uncertainties for nuclide ratios can be properly calculated using the exact solution to the problem of confidence intervals for a ratio of two jointly normally distributed variables, the so-called Fieller׳s theorem.

Detailed deterministic calculations of electron–photon coincidence spectra for 131m Xe and 133m Xe, including Auger electrons and X-rays from atomic relaxation, have been performed in order to increase the understanding of observed features in spectra produced by radioxenon measurement systems used in nuclear explosion monitoring. Energies and intensities of Auger electrons and X-rays agree well with earlier calculations, and the predicted intensities and energies for 131m Xe show good agreement with observed features in electron spectra. The results can be used in the development of radioxenon calibration and concentration analysis techniques.