Multiple seasonality in soil radon time series

Abstract

The dynamics governing the movement of the radon are complex and dependent on many factors. In the present study, we characterise the nature of temporal variations of 2-hourly and daily radon measurements in several monitoring sites of the Italian Radon mOnitoring Network (IRON) in Italy. By means of continuous wavelet transformation, a spectral analysis in time-frequency domain is performed. The results reveal that there are sub-daily, daily and yearly persistent periodicities that are common for all the stations. We observe structural seasonal breaks, that occur at the same frequency but at distinct time. Variations in radon concentration and local temperature are studied in terms of frequency contents and synchronicity. When analysing several long time series together, it is evident that the phase difference at low frequency movements (365-day period) between the radon and local temperature time series is depending on the sites’ location and therefore strongly controlled by local factors. This could at least partially explain the apparently contrasting results available in the literature obtained investigating smaller dataset about the relationships between temperature and radon variations. On the other hand, results show that all radon time series are characterised by marked cycles at 1 and 365-days and less evident cycles at 0.5-day and 180-days. They would be all ascribable to environmental-climatic factors: the short-period cycles to temperature and pressure variations, the long-period cycles also to seasonal rainfall variations.

Monitoring soil radon during the 2016–2017 central Italy sequence in light of seismicity

“…This is encouraging, even if we obtain a higher number of false alarms for the corrected time series. Nevertheless, we have to remind that there are strong indications from multi-stations, multi-variate analyses 26 that local site effects play a major role in modulating radon signal interactions with environmental parameters variation. Such evidence prevents us from the possibility to univocally assess the meteorological correction parameters and, consequently, the results of our approach to deal with atmospheric conditions should be considered as a complementary attempt to give further insights on our dataset and not necessarily more reliable or robust.…”

Stochastic Models for Radon Daily Time Series: Seasonality, Stationarity, and Long-Range Dependence Detection

“…The PTRL site is characterized by a contained seasonal variability also if compared with other sites of the same network, even equipped with a borehole probe which should be more immune from such effects (c.f. Figure 2 in Siino et al (2019b)). The radon concentration is measured in Bq/m 3 , becquerel per cubic meter.…”

“…The radon signals usually present a complex dynamic structure that is directly and indirectly influenced by several factors, such as environmental and climatic conditions of the site and characteristics of the ground soil, tide, and solar effect. (Pinault and Baubron, 1996;Piersanti et al, 2015;Siino et al, 2019b). All these factors have a different effect on the signal, as they can result either in a trend, seasonal, or stochastic component.…”

“…All these factors have a different effect on the signal, as they can result either in a trend, seasonal, or stochastic component. For instance, climate or tidal forces reflect in a multiple seasonality of the radon time series: hourly, diurnal, multi-day, annual, and even multi-annual cycles have been detected in different studies worldwide (Crockett et al, 2006;Udovičić et al, 2014;Yan et al, 2017;Crockett et al, 2018;Siino et al, 2019b;D'Alessandro et al, 2020). Of particular interest is the role of Rn as a potential earthquake precursor (Barbosa et al, 2015;Woith, 2015;Baskaran, 2016;Morales-Simfors et al, 2019) because the fracturing processes in the crust could enhance the mobility of Rn toward the surface (Toutain and Baubron, 1999;Woith, 2015).…”

“…Similarly, anomalies can be the result of weather episodes which cannot be explained by meteorological variables. Whatever the cause, these anomalies can be masked within the signal, and a way to bring them to light would be to de-noise the signal from the trend and/or periodic components (Baykut et al, 2010;Siino et al, 2019b;D'Alessandro et al, 2020). As a matter of fact, it is a challenging task to untangle and properly quantify all of these effects on the radon fluctuations because Rn time series present generally a nonstationary behavior, not constant variability over time and a long-term memory (Donner et al, 2015).…”