Influence of Meteorological Parameters on Indoor Radon Concentration Levels in the Aksu School

Abstract

The radon concentration activity in buildings is influenced by various factors, including meteorological elements like temperature, pressure, and precipitation, which are recognized as significant influencers. The fluctuations of indoor radon in premises are related to seasonal change. This study aimed to understand better the effects of environmental parameters on indoor radon concentration levels in the Aksu school. Indoor and outdoor temperature differentials heavily influence diurnal indoor radon patterns. The analysis indicates that the correlation between indoor radon and outdoor temperature, dew point, and air humidity is weak and negligible for atmospheric pressure, wind speed, and precipitation, as determined by the obtained values of R² and the Chaddock scale. The multiple regression model is characterized by the correlation coefficient r_xy = 0.605, which corresponds to a close relationship on the Chaddock scale.

1. Introduction

Radon release from the earth has been extensively researched for many years from various perspectives. Radon is a naturally occurring radioactive noble gas with varying occurrences in geological settings due to the decay of uranium present in different types of rocks and soils. As the properties of soil are affected by fluctuations in weather conditions, changes in soil-gas radon concentration are frequently observed over time [1]. The radon concentrations in soil and indoor air vary both in the short term (daily to weekly) and in the long term (seasonal) [2]. Different buildings and soil types may lead to varying radon levels and fluctuations [3].

It is generally accepted that elevated radon levels in buildings can substantially increase the risk of lung cancer. Given that indoor radon is the second leading cause of lung cancer after smoking, significant resources have been dedicated to overseeing, mapping, simulating, and forecasting these levels, as well as rectifying them where feasible and justified. Reports highlight the importance of conducting thorough and methodical assessments of indoor radon concentrations to determine the levels of radon exposure [4,5].

The radon concentration activity in buildings is influenced by various factors, including meteorological elements like temperature, pressure, and precipitation, which are recognized as significant influencers [6]. Based on several investigations, the radon concentration and its decay products in premises have shown significant variations in temperature, pressure, humidity, building material, ventilation conditions, wind speed, and other factors that can lead to large temporal and local fluctuations [7,8,9]. The levels of indoor radon can vary due to seasonal changes, as climate variations can result in different impacts on indoor and outdoor air [10,11]. During the winter, buildings tend to keep their windows and doors closed for longer periods due to rain, snow, or ice. This results in lower ventilation rates in rooms, causing indoor radon levels to rise and potentially reach harmful levels. In contrast, during the summer, people tend to open doors and windows, increasing the ventilation in their homes [11,12,13,14,15]. The relationship between ventilation rate and radon concentration is inverse. Increasing ventilation is crucial for decreasing indoor radon levels, as enhancements to ventilation systems typically result in a reduction of radon concentration by less than 50 percent [16,17,18]. During wintertime, indoor radon levels usually increase compared to other seasons, although this is not always true [19,20,21].

Preliminary past study of indoor radon measurements in the Aksu school has shown wide variability, from 14 Bq/m³ to 4703 Bq/m³ (Short-Term Radon Measurements) and up to 9000 Bq/m³ (radon monitoring measurement—monthly) [22,23]. The high levels of radon contamination found in the ground-floor classrooms can be attributed to the fact that a gold-mining site existed in the area where the school now stands back in the 1930s. The remaining ore materials under the building are emitting high levels of radon. The test results indicate that the construction of the school building in Aksu does not meet the necessary radiation safety standards. Investigations of underground areas, such as basements, indicate that rural houses in Aksu are not adequately protected from radon. Radon from the soil is entering through vents and cracks in the floor. Radon concentrations in underground areas in Aksu have been measured to range from 130 Bq/m³ to 5870 Bq/m³, which is only three to seven times higher than the measurements recorded on the ground floor for 60% of the buildings. Only 30% of buildings have well-shielded living spaces, resulting in a 20–50-fold reduction in radon levels compared to those in the underground areas [24].

The previous research indicates that the indoor levels of radon at the Aksu school changed according to weather conditions, and it was clear that proper ventilation played a crucial role in decreasing the concentrations. One study revealed a direct connection between radon levels, outdoor temperature, and relative humidity, demonstrating that meteorological factors affect radon concentrations [25].

This study aimed to better understand the effect of environmental parameters on indoor radon concentration levels in the Aksu school.

2. Materials and Methods

2.1. Study Area

The research area is situated in the Akmola region of Northern Kazakhstan, which includes the town of Aksu. Aksu is situated in an area known for uranium deposits and past gold mining activities (Figure 1). Some buildings and facilities in Aksu were built on land that was previously used for gold mining. Mines and the industrial area are located very close to residential areas. The surrounding environment within 5–10 km of the uranium mining complex is not regularly monitored for external gamma radiation, or outdoor and indoor radon.

2.2. Indoor Radon Measurements

The Radon Scout Professional radiometer, manufactured in Germany, was used to conduct radon monitoring measurements. Standard calibration was performed at a radon concentration of 3000 Bq/m³ after determining the background effect. The radiometer operates in diffusion mode, ensuring that thoron does not affect the measurement results. The measuring chamber, equipped with a semiconductor detector and high-voltage electrode (SARAD GmbH, Dresden, Germany), is not affected by fluctuations in air humidity. The sampling time for the measurements was 240 min. For radon concentration results at or below the reference level of 300 Bq/m³, at least 60 min intervals should be used, particularly when measuring outdoors [26].

The measurements were conducted during the autumn and winter of 2021 in classrooms where high radon levels were identified. In 2022, measurements were conducted in the autumn and winter in specific classrooms: room 2 on the first floor, room 24 on the second floor, and room 36 on the third floor. To determine the average annual radon concentration, the Radon Scout Professional radiometer was installed on the ground floor in room 2 from May 2022 to April 2023, covering 12 months. [27].

Together with employees of the National Center for Expertise in Stepnogorsk and specialists of the Republican State Institution “Stepnogorsk City Department of Sanitary and Epidemiological Control of the Department of Sanitary and Epidemiological Control of the Akmola region of the Committee for Sanitary and Epidemiological Control of the Ministry of Health of the Republic of Kazakhstan”, a ventilation system was installed in the school and control measurements of radon erosion were carried out in the school premises. Ventilation pipes are installed along all classrooms on the first floor of the Aksu school and in the classrooms of the kindergarten. A mechanical ventilation system is used only in cold periods; a natural ventilation system is used in warm periods.

The heated ventilation system is installed during the winter in the ground floor classrooms of the school, minicenter, and basements. The VK-100 fan model (1, Mikhaila Kotzubinskogo St., Kiev, 01030, Ukraine) is installed in the school and designed for installation in a circular duct with a 100 mm diameter. It is used in supply and exhaust ventilation systems in various residential and industrial buildings, schools, sports and recreation complexes, and other premises. The fan has a galvanized steel housing and is equipped with a reliable electric motor on a ball bearing, significantly increasing the service life and allowing for mounting in a horizontal or vertical position. The maximum air consumption is 290 m³/h, and the rotation speed is 2180 revolutions per minute.

The effect of ventilation rate on indoor air quality may be seen from its impact on airborne pollutant concentration by examining the equation of ventilation in the premises [28].

$$R_p={\displaystyle \frac{Q}{V}}\times (D\times h)$$

(1)

where $R_p$—ventilation rate per person, m³/h person or L/s person;

• Q—fresh air flow through the room, m³/h;
• V—volume of the room, m³;
• D—occupancy density, m² person;
• h—height, m.

Air change per hour (ACH) can be calculated by Equation (2) [29]

$$ACH={\displaystyle \frac{Q}{V}}$$

(2)

where ACH—air change per hour, h⁻¹;

• Q—fresh air flow through the room, m³/h;
• V—volume of the room, m³.

2.3. Measurement of Meteorological Quantities

As part of the study, the results of atmospheric air temperature, humidity, precipitation, soil surface temperature, atmospheric pressure, and dew points were obtained from the official website, www.kazhydromet.kz (accessed on 10 February 2024), of the National Hydrometeorological Service of the Republic of Kazakhstan, the Republican State Enterprise—(RSE) “Kazhydromet”. The database was established using the guidelines for sharing data by the National Hydrometeorological Service, allowing for the open utilization of observational findings from RSE “Kazhydromet’s” meteorological stations [30].

The database holds data gathered through the state observation network and serves general purposes. It contains operational meteorological information derived from the processing and analysis of primary meteorological data. This data are extracted from the files (RES) following the initial processing of current meteorological information from stations conducted on the PERSONA MIS program (an automated system for the primary processing of current meteorological information from stations). The data span from 2000 to the present and are regularly updated every month as observation materials become available [31].

Kazhydromet RSE has 15 branches in the Republic and 228 meteorological data measurement stations. We received data from the substation of settlement Akkol, which is located near Aksu. The distance between these settlements is 100 km. At the stations, the following measurement devices are used:

• Measurement of the maximum temperature—TM-1-1 and TM-1-2 thermometer; TM-1 meteorological glass thermometers are designed to measure the maximum temperature over a certain period of time. The measurement range of the TM-1-1 is from −35 to 50 °C, and for TM-1-2 from −20 to +70 °C. The limit of permissible error of the thermometers after the introduction of corrections is not more than ±0.2 °C. The manufacturer is JSC “Thermopribor”, Klin, Moscow region, Volokolamsk highway, 44.
• Measurement of the minimum temperature—thermometer TM-2-3; meteorological glass TM-2 thermometers (hereinafter referred to as the thermometers) are designed to measure the minimum air temperature over a certain period of time. The measurement range of the TM-2-3 is from −50 to 40 °C, and the limit of permissible error of the thermometers after the introduction of corrections is not more than 0.5 °C. The manufacturer is JSC “Thermopribor”, Klin, Moscow region, Volokolamsk highway, 44.
• Measuring soil temperature—thermometer TM-3-1; meteorological glass TM-3 thermometers (hereinafter referred to as the thermometer) are designed to measure the temperature of the soil’s surface. The measurement range of the TM-2-3 is from −35 to 60 °C, and the limit of permissible error of the thermometers after the introduction of corrections is no more than 0.2 °C. The manufacturer is JSC “Thermopribor”, Klin, Moscow region, Volokolamsk highway, 44.
• TM10-1 meteorological glass thermometers (hereinafter referred to as the thermometers) are designed to measure the temperature of deep soil layers and measure the temperature of the surface layer of water in reservoirs. The measurement range of the TM10-1 is from −20 to 30 °C, and the limit of permissible error of the thermometers after the introduction of corrections is no more than 0.1 °C. The manufacturer is JSC “Thermopribor”, Klin, Moscow region, Volokolamsk highway, 44.
• TM4 meteorological glass thermometers (hereinafter referred to as the thermometers) are designed to determine the temperature and humidity of the air. They are used in pairs in station psychrometers. The measuring range of the TM4-2 is from −25 to +50 °C, and the limit of permissible error of the thermometers after the introduction of corrections is not more than 0.1 °C. The manufacturer is JSC “Thermopribor”, Klin, Moscow region, Volokolamsk highway, 44 [32].
• Humidity and temperature sensor HMP155—humidity and temperature measurement, humidity measurement range from 0 to 100%, temperature from 80 to +60 °C. The error is ±0.6% (0–40%), ±1.0% (40–10%) [33].
• The M63M-1 anemorumbometer is designed for remote measurement of instantaneous, maximum, and average wind speeds and directions. Measurement ranges: instantaneous wind speed, m/s from 1.5 to 60; maximum wind speed, m/s from 3 to 60; average wind speed, m/s from 1.2 to 40; and wind direction, degrees from 0 to 360. The averaging periods of the average wind speed are 2 and 10 min. The main measurement error is no more than, when measuring wind speeds, m/s ± (0.5 + 0.05 V), where V is the measured wind speed; and when measuring wind direction, degrees ± 10. The manufacturer is JSC Safonovsky Plant Gidrometpribor, Russia, 215500, Safonovo, Smolensk region [34].
• For the BRS-1M-1 barometer, the barometer is designed to operate under conditions established for the performance of UHL category 4.2 according to GOST 15150-69, but at ambient temperatures from 5 to 50 °C and a maximum relative humidity of 95% (at a temperature of 30 °C). The range is 450–825 mmHg. The error is 0.25. The manufacturer is LLC Enterprise “Barometer”, Moscow, Tkatskaya str., 19 [35].
• The regional meteorological service of the Akkol station also has an AMS 111 system, available for stationary or mobile meteorological stations, applicable in areas where there are difficulties connecting the industrial power supply. The temperature measurement range is from −65 °C to +75 °C, atmospheric pressure from 500 to 1100 gPa, soil temperature from −65 °C to +75 °C, operating humidity range from 0 to 100%, and wind sensor from 0 to 60 m/s [36].

2.4. Statistical Analysis

The statistical analysis was conducted using StatTech version 3.1.10, developed by StatTech LLC v. 4.5.0 in Russia (developer—Stattech LLC, Kazan, Russia). Descriptive statistics for quantitative variables are presented in

Table 1. Descriptive statistics for quantitative variables.

Variables	M ± SD/Me	95% CI/Q₁–Q₃	n
Air temperature, Me (°C)	2	−9–16	353
dew point, Me (°C)	−2	−13–6	353
soil surface temperature, Me (°C)	1.9	−9.9–19.4	353
atmospheric pressure at station level, M ± SD (mm Hg)	974 ± 10	973–975	353
atmospheric pressure at sea level, Me (mm Hg)	1020	1011–1029	353
the magnitude of the barric trend, Me (hPa)	1	0–1	353
wind speed, Me (м/s)	2	2–3	353
precipitation1, Me	0	0–0	353
air humidity, Me (%)	68	57–78	353
radon concentration, Me (Bq/m3)	364	99–2096	353

. The normality of the quantitative variables was evaluated using the Shapiro–Wilk test for sample sizes less than 50, and the Kolmogorov–Smirnov test for sample sizes greater than 50.

If the data do not follow a normal distribution, quantitative information is represented using the median (Me) and the first and third quartiles (Q1–Q3) (Table 1).

Descriptive statistics were used to summarize categorical data, while the Mann–Whitney U-test was applied to compare two groups based on a non-normally distributed quantitative variable. For comparisons involving three or more groups and non-normally distributed quantitative variables, the Kruskal–Wallis test and Dunn’s criterion with Holm correction were used as a post hoc method. The analysis of multifield contingency tables involved the use of Pearson’s chi-square test for expected values greater than 10. Spearman’s correlation coefficient was employed to assess the relationship between two non-normally distributed quantitative variables, and ordinary least squares linear regression was used to develop a prognostic model for the dependence of a quantitative variable on predictors. Additionally, logistic regression was utilized to create a prognostic model for the probability of a binary outcome, with Nagelkerke R² serving as a performance measure for the model.

3. Results

According to a previous study, the indoor radon concentrations decreased after cleaning the ventilation shaft of the school, but during the cold period, indoor radon concentrations exceeded the permitted level [25]. The results of measuring radon concentration in the school premises after installing the supply ventilation system in the classrooms are shown in

Table 2. The results of radon concentrations in the school premises after installing the supply ventilation system in the classrooms.

Place of Measurement	The Results of Radon Concentration, Bq/m3
	Before Turning on the Ventilation System	30 min after Turning on the Ventilation System	60 min after the Ventilation System Is Turned on
1st floor	552 ± 184	255 ± 85	44 ± 13
Bedroom No. 1 (children’s garden)	683 ± 223	433 ± 144	153 ± 51
Gym	491 ± 158	335 ± 107	85 ± 27
Canteen	489 ± 152	120 ± 38	106 ± 35
3rd floor, № 32	531 ± 177	146 ± 44	-
3rd floor, № 34	225 ± 71	64 ± 21	-
3rd floor, № 36	388 ± 127	42 ± 14	-
2nd floor, № 21	220 ± 69	174 ± 55	-
2nd floor, № 26	132 ± 41	-	-
2nd floor, № 22	457 ± 147	172 ± 49	-
2nd floor, № 24	158 ± 47	15 ± 4	-

.

As it can be seen from Table 2, before the ventilation system was turned on, in the classrooms of the first floor and the kindergarten the radon concentration ranged from 489 to 683 Bq/m³, on the second floor from 132 Bq/m³ to 457 Bq/m³, and on the third floor from 225 Bq/m³ to 531 Bq/m³. After turning on the ventilation system for 30 min in the classrooms of the 2nd and 3rd floors, the radon concentration decreased below the reference level. In the classrooms on the first floor, it ranged from 120 Bq/m³ to 433 Bq/m³. An hour after connecting the ventilation system in the classrooms on the first floor, the radon concentration ranged from 85 Bq/m³ to 153 Bq/m³, which is lower than the reference level. In connection with the above, recommendations were made to turn on the ventilation system for an hour before the start of classes at the school and the kindergarten.

The previous research revealed that indoor radon levels fluctuated depending on weather conditions, and it is clear that effective ventilation significantly reduces the concentrations [25]. This study found that there are positive correlations between the levels of radon, outside temperature, and relative humidity, indicating that radon concentrations are affected by meteorological factors.

Figure 2 shows a comparative analysis of radon concentration in the seasons spring (March, April, and May), summer (June, July, and August), autumn (September, October, and November), and winter (December, January, and February). According to the data obtained, when comparing radon concentrations, statistically significant differences are revealed depending on the season (p < 0.001) (applied method: the Kruskal–Wallis test).

As it can be seen from Figure 2, it is possible to see an increase in the concentration of radon in the spring and winter period of the year; this is due to the diminished natural ventilation of the rooms. To fully clarify the reason for the difference, we analyzed the relationship of radon concentration with meteorological parameters such as temperature, relative humidity, atmospheric pressure, and soil surface temperature.

The statistical correlation between radon concentration and environmental parameters such as temperature, humidity, and atmospheric pressure was analyzed to determine if there is a relationship between them. Figure 3 shows a correlation analysis of the association between the outside air temperature and radon concentration.

As it can be seen from Figure 3, with a 1 °C decrease of air temperature, a 65.56 ± 27.36 Bq/m³ change in radon concentration should be expected, for which there was a close correlation (by the Chaddock scale). According to the coefficient of determination R² of the resulting model, 26.1% of the observed variance of radon concentration was explained. According to prior studies, in the summer, the radon concentration is lower than in the cold period of the year [9]. This is due to the fact that during the warm period, especially on hot days, the premises are often ventilated and there is also no stack effect, which allows for reducing the concentration of radon in the premises. These data are confirmed by the anticorrelation between the outdoor air temperature and the radon concentration inside residential premises. In winter, with the strengthening of the stack effect caused by higher temperature differences between the internal and external environment, the concentration of radon increases.

For a detailed analysis of the relationship between the outside air temperature and the radon concentration, the outside air temperature was divided into groups. Based on the data obtained during the analysis of radon concentration, depending on the outside air temperature by groups, statistically significant differences were revealed depending on air temperature by groups (p < 0.001) (applied method: the Kruskal–Wallis test). Figure 4 shows a comparative analysis of radon concentration and outside air temperature by groups.

Figure 4 shows that the radon concentration starts higher in groups from 0 to −5 and −25 to −20 °C, and then the radon concentration decreases above −30; this may be due to the fact that the soil surface temperature under buildings can also freeze, which reduces air exchange in the soil. In studies on measuring radon flux density in different seasons, scientists revealed the fact that with a decrease in soil surface temperature, the activity of radon release decreases [37].

We conducted a correlation analysis of the relationship between dew point and radon concentration. There was a noticeable tightness of feedback between radon concentration and dew point. Figure 5 shows a correlation analysis of the association between dew point and radon concentration.

In the studies of a number of authors, the dew point temperature also negatively correlates with the radon concentration in the premises [13,18]. With a 1 °C decrease of dew point, a 70.08 ± 25.40 Bq/m³ change in radon concentration should be expected, for which there was close correlation (by the Chaddock scale). According to the coefficient of determination R² of the resulting model, 21.4% of the observed variance of radon concentration was explained.

Other meteorological parameters were demonstrated to have less influence on radon levels in the house. The atmospheric pressure generally exhibited variations of periodicity longer than the diurnal cycles, and radon concentrations were not noticeably influenced by pressure changes. Figure 6 shows a correlation analysis of the association between atmospheric pressure and radon concentration.

As it can be seen from Figure 6, a weak correlation positive association between radon concentration and atmospheric pressure at station level was estimated. With a 1 mm Hg increase of atmospheric pressure at station level, a 24.33 ± 9.52 Bq/m³ change in radon concentration should be expected, for which there was a weak correlation. According to the coefficient of determination R² of the resulting model, 1.8% of the observed variance of radon concentration was explained.

Figure 7 shows a correlation analysis of the association between air humidity and radon concentration.

As it can be seen from Figure 7, a moderate correlation and positive association between radon concentration and air humidity was estimated. With a 1% increase in air humidity, a 54.66 ± 19.21 Bq/m³ change in radon concentration should be expected; the correlation on the Chaddock scale was average. According to the coefficient of determination R² of the resulting model, 15.6% of the observed variance of radon concentration was explained. According to the data obtained, the radon concentration is moderately positively correlated with the relative humidity of the outdoor air. In the studies of Spasić, D. and Gulan, L., no clear correlation was found between indoor and outdoor radon concentrations and relative humidity and precipitation [38]. In the work of Aquilina, N.J. and Fenech, S., radon conjugation in combination also weakly correlates with temperature load and relative humidity [39].

Throughout the day, the wind speeds followed a daily cycle, peaking in the early afternoon and reaching their lowest levels after midnight. Even though it was anticipated that the higher wind speeds would create negative pressure differences in the school, which could result in the transport of radon through advection, there was no apparent connection between the variations in radon concentration and the daily wind-speed pattern, except when the wind cycle aligned with the temperature fluctuation cycle. Figure 8 performed a correlation analysis of the association between wind speed and radon concentration.

As it can be seen from Figure 8, a weak correlation and positive association between radon concentration and wind speed was estimated. With a 1 m/s increase of wind speed, a 350.55 ± 115.42 Bq/m³ change in radon concentration should be expected, for which there was a weak correlation. According to the coefficient of determination R² of the resulting model, 8.5% of the observed variance of radon concentration was explained.

A meteorological parameter that did appear to affect the indoor radon concentrations was precipitation, as periods following significant rainfall often exhibited decreased indoor radon levels. Figure 9 shows a correlation analysis of the association between precipitation and radon concentration. As a result of comparing radon concentration depending on precipitation, statistically significant differences were established (p = 0.007) (method used: the Kruskal–Wallis criterion).

Figure 9shows a weak correlation positive association between radon concentration and precipitation was estimated. With a 1 mm/h increase of precipitation, a 29.50 ± 9.57 Bq/m³ change in radon concentration should be expected, for which there was close correlation (by the Chaddock scale). According to the coefficient of determination R² of the resulting model, 0.2% of the observed variance of radon concentration was explained. Indicators of radon concentration and precipitation have statistically significant differences: radon concentration increases with precipitation of 0–1.9 mm, and with an increase in precipitation, radon concentration decreases. This is because, with a low moisture content, the process of radon emanation increases with an increase in soil moisture [40,41], which leads to an increase in the radon flux density. With a high moisture content, radon output to the surface is prevented by water in the pores, i.e., water shields the radon output from the ground, therefore, the radon flux density decreases [38]. Precipitation saturates the soil and prevents the release of radon due to the deterrent effect [42,43]. The release of radon from the soil depends on the amount of precipitation, and the radon diffusion coefficient increases with temperature [44].

The regression analysis approach is largely insufficient to constrain the very complex dynamics of radon, and time series, multiple linear regression, and multivariate analysis should be performed [45]. The indoor radon concentrations are influenced by meteorological parameters, and multiple linear regression was performed, as shown in

Table 3. Analysis of radon concentration conditioning on air temperature, dew point, soil surface temperature, atmospheric pressure at sea level, wind speed, precipitation, and air humidity.

	B	Std Error	t	P	R2
Intercept	−2065	427	−5	<0.001 *	0.36
dew point	−61	7	−9	<0.001 *
wind speed	336	52	6	<0.001 *
air humidity	34	6	5	<0.001 *

* differences are statistically significant (p < 0.05).

, to estimate the dependence of radon concentration on quantitative variables. The number of observations was 352.

The observed association of radon concentration with dew point, wind speed, and air humidity is presented by a linear regression equation:

Y_{radon concentration} = −2065 − 61X_{Dew point} + 336X_{wind speed} + 34X_{air humidity}

where Y—radon concentration value, X_{Dew point}—dew point (°C), X_{wind speed}—wind speed (m/s), and X_{air humidity}—air humidity (%).

According to this equation, with a 1 °C decrease of dew point, a 60.71 ± 20.24 Bq/m³ change in radon concentration should be expected; with an 1 m/s increase of wind speed, a 335.92 ± 111.89 Bq/m³ change in radon concentration should be expected; and with a 1% increase of air humidity, a 33.96 ± 11.32 Bq/m³ change in radon concentration should be expected.

The resulting regression model is characterized by the correlation coefficient r_xy = 0.605, which corresponds to a close relationship on the Chaddock scale. The model was statistically significant (p < 0.001). The resulting model explains 36.6% of the observed variance of radon concentration.

Table 4. The results of ventilation rate per hour and air change per hour in the classrooms in the school and kindergarten.

Place of Measurement	${\mathit{R}}_{\mathit{p}}$, L/s Person	ACH, h−1
1st floor	7	4
Bedroom No. 1 (children’s garden)	6	3.3
Gym	4	0.6
Canteen	5	1.6
3rd floor, № 32	7	4
3rd floor, № 34	7	3.6
3rd floor, № 36	7	3.1
2nd floor, № 21	8	4
2nd floor, № 26	7	3.8
2nd floor, № 22	6	3
2nd floor, № 24	8	3.3

presents the estimated air changes and ventilation rates per person for classrooms in the school and kindergarten.

As it can be seen from Table 4, the ventilation rate for all classrooms and kindergarten bedroom does not exceed the recommended level (8 L/s person) [46,47]. The gym has the maximum capacity in the school and the highest site volume at 450 m³, with the lowest ventilation rate per person. The ACH values varies from 0.6 to 4. These results fall within the bare minimum recommended by the healthy buildings site [48].

4. Discussion

Correlation analyses between indoor radon levels and outdoor weather conditions were conducted. As a result of the constant measurement of the radon level in the premises, taking into account the environmental factors of the outdoor air, attention should also be paid to the change in the temperature of the outdoor air, since this affects the temperature difference between indoor and outdoor air, and therefore the pressure drop between the indoor and outdoor environment. In various studies, authors also indicate that the daily fluctuations of radon are mainly associated with fluctuations in outdoor air temperature, which lead to frequent airing of the room in the summer, or to a decrease in ventilation of the heated period in winter. Ventilation of the room affects the dynamics of the radon level (i.e., chimney and ventilation) [49,50].

The levels of radon showed a regular pattern throughout the day, reaching a low point during the daytime and peaking at nighttime due to variations in atmospheric convection. On a seasonal time scale, an increase in the concentration of radon in the spring and winter period of the year is seen, and this is due to the natural ventilation of the room. The inverse relationship between indoor radon concentration and wind speed could be clarified by the drop in outdoor pressure as outdoor wind speed rises, thus resulting in a lower indoor radon concentration as the outdoor wind speed increases [51]. Another of the climatic parameters, precipitation also affects radon levels in the premises, and according to our data, statistically significant differences (p = 0.007) between the radon concentration and the amount of precipitation have been established. These results are in good agreement with the results of other studies, which also indicate that the radon flux density at a low moisture content increases the process of radon emanation with increasing soil moisture, and at a high moisture content, radon output to the surface is hindered by water in the pores. According to our data, the influence of atmospheric humidity on indoor radon levels is moderately positively correlated, and these results are consistent with previous observations and studies conducted by various authors [11,52,53]. The obtained results of the analysis of the relationship between indoor radon levels and climatic parameters indicate the importance of taking into account outdoor air temperature and atmospheric precipitation when conducting monitoring and short-term measurements of radon levels.

5. Conclusions

In this study, indoor radon concentrations have been correlated with meteorological parameters. Indoor and outdoor temperature differentials heavily influence diurnal indoor radon patterns. The analysis indicates that the correlation between indoor radon and outdoor temperature, dew point, and air humidity was weak and negligible for atmospheric pressure, wind speed, and precipitation, as determined by the obtained values of R² and the Chaddock scale. The multiple regression model is characterized by the correlation coefficient r_xy = 0.605, which corresponds to a close relationship on the Chaddock scale.

In our study, there is a weak correlation between outdoor air relative humidity and indoor radon levels. These findings are consistent with the work of Spasić D. and Gulan L., who also did not find a clear correlation between indoor and outdoor radon concentrations, relative humidity, and precipitation. Additionally, Aquilina, N.J. and Fenech, S. discovered that the combination of radon concentration weakly correlates with temperature and relative humidity.

The ventilation rate for all classrooms and kindergarten bedroom does not exceed the recommended level. The results fell within the bare minimum recommended.