Seasonal variations of decay rate measurement data and their interpretation

Measurement data of long-lived radionuclides, for example, (85)Kr, (90)Sr, (108m)Ag, (133)Ba, (152)Eu, (154)Eu and (226)Ra, and particularly the relative residuals of fitted raw data from current measurements of ionization chambers for half-life determination show small periodic seasonal variations with amplitudes of about 0.15%. The interpretation of these fluctuations is a matter of controversy whether the observed effect is produced by some interaction with the radionuclides themselves or is an artifact of the measuring chain. At the origin of such a discussion there is the exponential decay law of radioactive substances used for data fitting, one of the fundamentals of nuclear physics. Some groups of physicists use statistical methods and analyze correlations with various parameters of the measurement data and, for example, the Earth-Sun distance, as a basis of interpretation. In this article, data measured at the Physikalisch-Technische Bundesanstalt and published earlier are the subject of a correlation analysis using the corresponding time series of data with varying measurement conditions. An overview of these measurement conditions producing instrument instabilities is given and causality relations are discussed. The resulting correlation coefficients for various series of the same radionuclide using similar measurement conditions are in the order of 0.7, which indicates a high correlation, and for series of the same radionuclide using different measurement conditions and changes of the measuring chain of the order of -0.2 or even lower, which indicates an anti-correlation. These results provide strong arguments that the observed seasonal variations are caused by the measuring chain and, in particular, by the type of measuring electronics used.