Investigation of the activity of Rn-222 along a small stream in the Representative Basin of Juatuba-MG

For thousands of years, water has been the focus of experimentation toward solving the challenges associated with human water supply, navigation, irrigation, and sanitation. The use of tracers to study water resources is an efficient approach that can facilitate the modeling of many hydrological scenarios. The goal of this paper is to show results of research that tracked the presence of Rn-222, a natural tracer, in the surface waters of a small watercourse in southeastern part of Brazil. RAD 7, which is an electronic and portable radon detector, was the main instrument used in this survey. We analyzed 117 water samples and converted the radon activity results to effective radiation doses with respect to the hypothetical human consumption of these waters. We also analyzed the sediments of the watercourse. The obtained data showed that the radon activity in the studied waters varies between 0.52-76.96 Bq/m3. We determined the effective dose of all samples to be less than 1 mSv y−1, and its consumption to present no risk to human health. The existence of connections between surface and subsurface waters in the stream is possible, and radon peaks may indicate the existence of discharge zones into the surface water body.


INTRODUCTION
A tracer can be defined as any substance or particle (chemical or biological) that can be used to follow, either punctually or continuously, the behavior of a particular system or component, such as water flow in an underground or open environment (DAVIS et al, 1980). Tracers are very useful tools for investigating many processes, and can be very helpful in clarifying many natural phenomena. According to the scenario under study, different kinds of tracers could be used, including radioactive, chemical, fluorescent, and biological ones. Tracers have been used in oilfield applications (SERRES-PIOLE et al., 2012) as well as to study the deposition of particulate matter on urban vegetation (VOLTAGGIO et al., 2016). Researchers in Australia studied lead pollution in marine sediments with the help of tracers (ALYAZICHI; JONES; MCLEAN, 2016), and medical scientists have recently used tracers for a variety of purposes (OOMS et al., 2014), (COLE et al., 2014).
The science of hydrology, along with many other research areas, routinely uses tracers to solve problems. Examples include the use of viruses as tracers to study residence times in aquifers (HUNT; BORCHARDT;BRADBURY, 2014), and the use of natural tracers to track groundwater flow in a mining area (COZMA et al., 2016). Tracers have also been used in the management of a nuclear site, for the protection of water resources, in simulations of a hypothetical near-surface repository for low-level radioactive waste (TESTONI; LEVIZZARI; DE SALVE, 2015), and in the evaluation of environmental impacts caused by the bottom discharges of a small hydro power plant (FERREIRA et al., 2013).
Rn-222, a natural tracer that has been used in many hydrological studies, is produced by the alpha decay of Ra-226 in the decay series of U-238, and has a half-life of 3.8 days. Rn-222 is more suitable as a water tracer than Rn-220, which is produced by the radioactive decay of Th-232 in the Ra-224 series, and has a half-life of just 56 seconds. Since it is chemically inert, Rn-222 can be used as a water tracer in streams, rivers, reservoirs, oceans, and aquifers, which makes possible to disregard biogeochemical reactions and facilitate its measurement. It is well known that radon in groundwater is enriched compared to that in surface waters. This feature results from the fact that groundwater is in contact with mineral grains that contain Ra-226, and surface water is subject to turbulence generated by environmental conditions, which allows radon to escape (BURNETT et al., 2008).
However, the presence of radon in water can be a problem, since there are safety limits for its ingestion by humans. The World Health Organization (WHO) established guidelines regarding the quality of drinking water, and included radiological aspects as one parameter to be observed. It has also emphasized that the ingestion and inhalation of radon can be responsible for a number of health problems (WHO, 2011).
Since the harmful effects of radon were first recognized and discussed, researches have been undertaken around the world to investigate the radon concentrations in surface and subsurface waters, oceans, wells, and springs, and to address the effects of its consumption. For example, in a survey of natural waters conducted in Romania (1511 samples), researchers found a correlation between the radon concentration and the geological structure (COSMA et al., 2008). In Serbia, a similar study of 44 samples showed that waters from a volcanic region had a radon activity much higher than the recommended level (TODOROVIC et al., 2012). In Greece and Cyprus, the radon activity in waters has also been studied (NIKOLOPOULOS;LOUIZI, 2008) In Brazil, researchers have also studied the presence of radon in water, including the mechanics of its transfer from rocks and soils to water, as well as its presence in the alkaline massif of Poços de Caldas and in the Paraná sedimentary basin (BONOTTO; LIMA, 1997). The activity of radon in water has also been studied in the Bauru aquifer (SANTOS; BONOTTO, 2011), in three different regions in the state of Bahia (COSTA; AZEVEDO, 2012), in the municipality of Águas da Prata (BONOTTO; LIMA, 1997), and in the metropolitan region of Curitiba (CORRÊA et al., 2015), among others.
The current study aims to track the presence of Rn-222 along a small stream in a rural area in the southeastern region of Brazil. We converted the radon activity in the water samples into effective doses to determine whether or not its consumption by humans would present a health risk. We then compared these values with data from other countries. In addition, we analyzed samples of sediments collected along the stream to determine their radium activity levels (since radium becomes radon by radioactive decay).

Study area
The Juatuba Basin ( Figure 1) is located in the upper part of the São Francisco River in the state of Minas Gerais in southeastern region of Brazil. This watershed covers 442 km 2 , is located 60 km from Belo Horizonte (the state capital), complying part of the cities of Mateus Leme, Igarapé, and Itaúna. The main affluents that feed into and form the Juatuba River are the Serra Azul and Mateus Leme streams, which have drainage areas of 265 km 2 and 155 km 2 , respectively. For this study, we chose the Matinha stream, which is 2.2 km long and located in a rural area.
Along the watercourse, 117 water samples were collected, being 18 meters the average distance between the sampling points. The study area is adjacent to a region known as the "Iron Quadrangle," which, according to a geological survey, consists of rocks from the Archeozoic, Lower Proterozoic, and Cenozoic periods. The region has a predominance of gneiss, gray granitic rocks, quartz, orthoclase, muscovite, biotite, and epitope (DRUMOND 2004).
It is valid to mention that within the scope of this project, we had previously performed some Rn-222 measurements in the Juatuba Basin (FERREIRA et al., 2015).

RAD 7
We used this detector ( Figure 2) to read the activity of Rn-222 in the water samples collected from the studied stream. The RAD7 has two modes for measuring radon in water: a GRAB mode for a single value and a SCAN mode for continuous measurements. In this study, we used the GRAB mode. The RAD7 has three main operational steps: 1. Drying -the device removes any water present in the system to prevent damage to the equipment. Following this step, the internal moisture content of the device should be ≤6%, which makes feasible the reading of samples; 2. Analysis -over a time period of 30 minutes, the device performs four readings and, for each one, displays its radon activity and standard deviation. Then, it calculates the average of these four sample values; 3. Cleaning -after reading a sample, the RAD7 must be cleaned in an open circuit to eliminate all radon from the system prior to performing the next readings.
Rn-222 decays by alpha emission to Po-218, and the RAD7 calculates the radon concentration on the basis of the Po-218 alpha peak. Figure   The RAD 7 detector can read activities in a range from 10 pCi.L −1 to 400.000 pCi.L −1 . For its operation in this project, we used the Wat250 protocol, which means that the water samples were stored and read in 250-ml flasks. Then, we connected a functional kit known as RAD H 2 O to the system for its operation. We collected all water samples from close to the bottom of the stream with the help of a peristaltic pump powered by a 12 V battery.
We obtained the initial concentration of radon using Equation 1, where A(t) is the concentration of Rn-222 measured at time t, A 0 is the initial concentration of radon in the sample, λ is the decay constant of Rn-222 (0.18), and t is the elapsed time between collecting and reading the sample.
We note that the RAD 7 detector has been used in many other projects that required measurement of Rn-222 in water (AKAWWI 2014), (RAVIKUMAR et al. 2014), (LE et al. 2015).

Sediment and background analyses
To determine if there is any correlation between the presence of Rn-222 in the water and the existence of Ra-226 in the sediments of the watercourse, we conducted a sampling campaign in May 2016. Using a small shovel, we collected sediments from the source to the end of the stream, at 100-m intervals, from the bottom and both sides, to yield 22 composite samples. We analyzed the samples by gamma spectrometry, a non-destructive analytical technique that enables the identification and quantification of gamma-emitting isotopes in a wide variety of matrices. The model we used for sample counting was a high-resolution HPGe detector with a nominal efficiency of 50%, manufactured by CANBERRA TM . This detector is a coaxial model 5019 (DSA-2000) coupled to a microcomputer with a multichannel-spectrum acquisition board and a Genie2K program.

Calculation of dose from ingestion of water
We used Equation 2 to determine the annual effective dose due to the presence of Rn-222 in drinking water: where Deff is the effective dose from ingestion, K is the ingestion dose conversion factor for Rn-222, G is the volume of water consumed, C the concentration of Rn-222, and t the duration of consumption. In this work, we used a dose conversion factor (K) of 3.5×10 −9 Sv Bq −1 (1) to calculate the effective dose (NRC, 1999). The daily average water consumption per capita (G) has been determined to be typically between one to two liters per day (PINTI et al., 2014), which was used to calculate the annual effective dose of Rn-222 in drinking water (YALCIN et al., 2011). In this work, we assumed a mean value of −1.5 l/day and the parameter t represents a year, or 365 days. Table 1 presents the results of Ra-226 in the sediments collected along the Matinha stream, based on the gamma spectrometry analysis.

RESULTS
From September 2014 to May 2016, we collected 117 water samples in the Matinha stream for Rn-222 analyses. Table 2 shows the obtained results.
Using all the measurements performed during the execution of the project (24 months), we plotted the Rn-222 activity results and the distances from the source (Figures 4-6).

DISCUSSION
From a total of 117 water samples, all of the effective dose values from the ingestion of radon are much smaller than 0.1 mSv y −1 . Considering that Brazil has established a national standard regulation of 1 mSv for the annual limit that an individual can receive (BRASIL, 2014), our study results show that the possible ingestion of this water represents no risk to public health. In fact, the literature indicates that spring water generally presents higher radon concentrations.
The high variability of the Rn-222 activity concentrations can be explained by their different origins and the lithology of the aquifer host rocks (LLERENA et al., 2013). Larger values are found in groundwater, based on the geology of the studied site (FONOLLOSA et al., 2016).  If we compare the results of this work with those from survey data of many other countries and different sources, such as wells, non-bottled mineral water, springs, and underground, surface, and drinking water, the Rn-222 activity concentrations found in the Matinha stream present no anomalous readings.
The United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR, 2000) reported that, on average, 90% of the dose attributable to radon in drinking water comes from inhalation rather than ingestion. Thus, it is much more important to control the inhalation of radon than its ingestion from drinking water. Epidemiologic studies regarding this issue have been conducted in the USA, Canada, Japan, Brazil, and Germany, among other countries (HYSTAD et al., 2014;ETANI et al., 2017;KREUZER et al., 2015;LARA et al., 2015;CORLIN et al., 2016).
In our sediment analysis, and since radon escapes from rocks surrounded by underground waters, it is likely that only the spectrometry analysis of surface sediments will fail to generate results by which we can establish a correlation with the Rn-222 activities of surface waters.
Lastly, we note that along the Matinha stream, there is a station that monitored rainfall from 2012 to 2015, under the scope of another research project. During recent years, the whole region has endured long dry seasons that have lowered the levels of almost all reservoirs. However, the water level of the stream, as monitored by a station located downstream of the study area, has not changed significantly (Figure 7). This also indicates the presence of discharge sections along the Matinha stream (MARTINEZ; RAIBER; COX, 2015). The use of Rn-220 as a natural tracer could be an efficient tool for identifying the location of these sections, since its half-life of 55 seconds is very short (CHANYOTHA et al., 2014).

CONCLUSIONS
The presence of Rn-222 in the surface waters of the Matinha stream indicates a possible connection between surface and subsurface waters. The peaks of Rn-222 along the watercourse likely indicate the presence of discharges sections.
The radon activity in the studied waters does not present a public health risk, since the effective dose values associated with its hypothetical ingestion are small. The largest Rn-222 activity value was below the limits established by national legislation. We found no anomalous readings in the obtained data.
In Brazil, further studies and analysis should be routinely conducted in underground, mineral, and drinking waters, since these have Rn-222 activity concentrations much higher than those in surface waters. Considering that the human health effects of the inhalation and ingestion of radon cannot be disregarded, this subject should also be more discussed among members of the research teams involved with hydro resources.