Results on radon concentrations obtained in last decade have been reviewed. Radon dosimetry was discussed and its dependence on the unattached RnDP (smaller than 10 nm) commented on. A need for simultaneous monitoring of RnDP and measurements of the concentration and size distribution of nano aerosols has been accentuated.
COBISS.SI-ID: 23025447
Time series of radon concentrations have been analysed applying artificial neural networks. The entire database is divided into two subsets: SA subset with the data for seismically active periods and N-SA subset with the data for seismically non-active periods. First, the N-SA subset was used to train the program to predict radon concentration from the environmental data. When then the entire database was analysed, disagreement generally appeared between the predicted and measured radon concentrations. We were able to correctly predict 10 seismic events out of 13 within the 2 years period.
COBISS.SI-ID: 23182375