The experience gained after seven decades of research shows that large–scale sampling points in each area of interest and over an extended time are required to collect a statistically more robust data sample, meaning that numerous sensor networks for radon are required to give a new level of statistical significance of the measurements. The advanced machine learning algorithms for time series combined with computational power will enable a deeper analysis of the parameters affecting radon and other observations. The prospect is to learn which radon anomalies can be associated with earthquakes. Technically, there should be no obstacle to building extensive radon networks along seismic fault zones supporting increased public awareness.
Inspired by such suggestions, the Artemis project has been developed to advance large–scale monitoring of radon and other observables in groundwater in the Abruzzi region, Italy, the Ionian Islands, Greece, and selected sites in Switzerland. The project intends to produce a sensor network of about 100 units – the detection based on gamma-ray measurements in groundwater using a novel sensor of CsI developed in SGI, Germany and calibrated in JURO, Czech Republic. The radon concentration, temperature, barometric–hydraulic pressure, conductivity, and pH will be measured with a vision of prolonged monitoring duration. The data time series will be analysed using A.I. to obtain correlation results between the radon signal, the earthquake's precursor time and magnitude, and the building of an early public warning system.
https://doi.org/10.1007/s10967-024-09710-4
Please sign in or register for FREE
If you are a registered user on Research Communities by Springer Nature, please sign in