On the relation between outdoor 222Rn and atmospheric stability determined by a modified Turner method

In practice, information about atmospheric stability is often obtained from discrete stability classes determined from routine meteorological observations. However, changing concentrations of the radioactive gas 222Rn present in the atmosphere are also considered a good indicator of vertical dispersion and atmospheric stability. A complex, in-depth analysis between these different approaches of atmospheric stability assessment has not been performed so far, and was the main motivation behind this study. The study presents atmospheric radon data measured in Bratislava (Slovakia) and stability indexes (SI) calculated according to a modified Turner method during a period of one year. Basic features of the diurnal and seasonal variations of these variables are discussed. It was found that the time series of radon activity concentration (RAC) lags approximately 5 h behind that of the Turner stability classes adjusted for temperate climate regions. Various time lags were also identified between RAC and meteorological variables used to determine the stability classes. Evaluation of seasonal trends revealed a low variability of mean monthly values of stability classes compared to the variability of mean monthly values of RAC. Another notable difference between RAC and stability indexes was found - while the stability index can both increase and decrease with wind speed, concentration of outdoor radon was never observed to increase with increasing wind speed. In spite of the mentioned discrepancies, the time series of RAC and SI are generally in a good agreement. This is especially true if one compares the deviations of RAC and SI from their mean daily values, when the differences in their seasonal variability are eliminated. Deviations of RAC can be used to calculate diurnal variations of stability indexes. Analysis of a complete year of data also revealed a roughly linear relationship between average values of RAC and calculated stability indexes, because in large datasets random processes tend to cancel each other out.