Exposure levels of radiofrequency electromagnetic fields from mobile base stations in Mpumalanga province, South Africa

This study assessed the radiofrequency electromagnetic fields (RF-EMF) exposure levels proximal to primary school structures in the Mpumalanga province of South Africa. A calibrated Acoustimeter was used to measure the RF EMF levels at defined distance points of 50 and 100m away from the three mobile base stations, that provide a 4G network coverage. All the measurements were obtained from three mobile stations, each from one district municipality, using 30 minutes intervals for repeated measurements. The measured peak exposure level at 50m was 887 μW/m2 and 905 μW/m2 at 100m. No statistical difference was found when comparing the measurement data obtained at 50 and 100m from the three district municipalities. However, comparison of measurements obtained at 50 and 100m for all base stations were statistically significant, p-value< .002 and p< .003 respectively. The measured RF EMF exposure levels at both distance points were below the reference levels found under International Commission for Non-ionizing Radiation Protection (ICNIRP) RF guidelines. It is therefore essential for mobile base station companies to consider proximity of residential structures when installing new base stations in the future, and regularly maintain the components of mobile base towers, especially in the current state of power load shedding in South Africa.


Introduction
A constant use of new technology devices has led to an increase in the radiofrequency electromagnetic fields (RF-EMF) exposure pattern, both in the general public and occupational environments.In previous years, the need to determine the degree of RF-EMF exposure ascribed to various sources increased significantly [1][2][3].Exposure assessment techniques and methodologies have been developed to assess exposure to RF-EMFs [3,4].Various assessment methods such as spot monitoring, personal measurements, fixed-site monitoring, and micro-environmental measurements [5][6][7], have been used to establish compliance with the International Commission of Non-Ionizing Radiation Protection (ICNIRP) [8].The established reference level for incident power density between 2 and 300 GHz is 50 W/m² for occupational exposure and 10 W/m² for the general population measuring whole-body exposure averaged over 30-minute intervals.An assessment of cumulative RF-EMF exposure in five European countries suggested mobile phone base stations in outdoor urban environments as the most significant environmental RF-EMF exposure sources [9].This assessment prompted the monitoring of exposure levels from base stations near residential houses [10].Radiofrequency EMFs are primarily emitted by telecommunications devices (e.g.such as base stations).These devices generate low but constant RF-EMF exposures for the surrounding population [11].It is hypothesized that the output power of the mobile base stations is greater than devices operating near the body [12].However, field intensity decreases with distance, resulting in low whole-body exposure from environmental RF-EMF exposure sources, as opposed to highly localized exposure from sources operating close to the body (e.g., mobile phones, tablets, laptops, and Wi-Fi routers) [12,13].
The World Health Organization (WHO) advocated for RF-EMF exposure and the identification of exposure determinants surrounding the population [3].The quantification of the RF-EMF exposure has been performed mainly in the European countries [10,[14][15], with relatively no exposure data in the Republic of South Africa (RSA) [16].Researchers have noted base stations to be more prevalent in densely populated areas [17].According to Freudenstein et al. [14] and Lagorio et al. [18], the exposure contribution of the RF-EMFs depends on the distance from the source, frequency of exposure, duration of exposure, time of day, and the size of the exposure source.Furthermore, exposure patterns from base stations are controlled by antenna output power, transmission direction, and attenuation [1,13].
Monitoring and evaluating exposure levels is critical for the quantification of exposure levels near residential household [16].Mpumalanga, one of the provinces in RSA has a widespread mobile base station proximal to areas where people spend most of their time.The significant increase of base stations in the residential areas have resulted in a fundamental shift in population-based exposure patterns to RF-EMFs in the general environment [19].This study aimed to determine the RF EMFs exposure levels from mobile base stations in Mpumalanga province, South Africa.

Study setting
A quantitative, measurement study was conducted in Mpumalanga, a province in the RSA's northeastern region (Figure 1) with an estimated population of 4 720 497 people.The study was carried out in the three district municipalities (Ehlanzeni [EDM], Gert Sibande [GSDM], and Nkangala [NDM]) of the province.The province was chosen based on its terrain, geographical dispersion of base stations, and population density.One mobile base station was selected from each residential area of the district, and measurements were conducted from three residential densities; urban, peri-urban, and rural.The purpose of sampling from these residential densities was to represent a wide range of RF EMFs exposure situations in settings were different network-based activities are taking place.

Measurements
An Acoustimeter was used to take measurements at each defined distance point in the general environment, proximal to the base station.The Acoustimeter measures peak and average signals using a combination of LEDs and number readings on a bright text screen.It accurately measures all incoming radiation from 200 MHz to 8000 MHz (8 GHz) with peak exposure levels in V/m and Average exposure levels in μW/m².This study only considered the average exposure levels in μW/m².The AM11 has a measurement range of 200 -8 000 MHz ±6 dB ±0.02 V/m with a sensitivity of 1 μW/m² -100 000 μW/m².At each location, a proximal area was identified and a distance of 50 and 100 m from the base station was chosen.The distance points were marked in the direction of residential infrastructures.i.e. schools, residential houses, business centers, etc.The purpose was to represent exposure in various settings where people spend most of their time, either during the day or in the evenings.The Acoustimeter was positioned on a tripod at the height of one meter above the ground, and repeated measurements from each distance point were recorded at 30 minutes intervals.For each distance point in one district, a total of 21 repeated measurements were taken, amounting to 42 measurements per mobile base station.An overall of 126 measurements were gathered, and measured values were further compared with the RF-EMF ICNIRP of 2020 [20] reference levels for general public exposures.Figure 2 illustrate measurement points per base station.

Data analysis
Data were captured in the Excel spreadsheet and transferred to IBM SPSS version 27 for analysis.Descriptive data was analysed for frequencies, percentages, means (SDs), and ranges.Measurement units were captured in µW/m² for comparability with other studies.Shapiro-Wilk test was used to assess the normality of the data for all RF-EMF measurements.The exposure levels from the three base stations were further compared using a pairwise comparison test.A statistical significance level of (α) 0.05 was applied.
Comparing exposure levels at both distances, a statistical significant difference was found at 50m (pvalue < 0.002), and at 100m (p< 0.003).The chi-Square test was further used to find differences between the measurements.A statistical non-significant difference between the measurements (p< 0.389) was found.The pairwise comparison suggested a statistical significant difference between measurements at 50 and 100m (mean 497.60 µW/m², p<0.001) (Table 1).An overall of 126 measurements were done from the base stations.The results for statistical tests performed to compare the measurements for the three district municipalities at 50 and 100m are provided in Table 2.
Table 2. Independent t-test for base stations measured at 50m and 100m from the three district municipalities.
The average power density decreased drastically between two distance points.The minimum average power density from individual base stations surveyed in the three districts of Mpumalanga province at 50m was 161 µW/m² while the maximum was 887 µW/m².At 100m the minimum average was 50 µW/m² and maximum was 90µW/m².Therefore, the RF-EMF exposure hazard index in the study areas at 50m were higher but lower than the permitted RF-EMF exposure limits for the general public.At 100m, though exposure has decreased with distance, it was also below the permitted exposure limits.

Discussions
According to Röösli et al. [21], personal exposure is firstly determined by environmental exposure levels, individual behavior such as the use of wireless communication devices, and time spent in various microenvironments.The current study used the incident power density as a measurable variable and Acoustimeter (AM11) model to characterize RF-EMF exposure to base stations.Furthermore, the average RF-EMF exposure levels were measured during daytime over 30min, to determine general public exposure.
Therefore, the results indicated that within the general environment, the base stations RF-EFM measured at an increased distance of 100m, the mean incident power density (69.98 µW/m²) were below the ICNIRP reference levels (10 W/m²) guidelines (Table 3).Studies conducted in Europe show a constant increase of public exposure, especially in urban environment.Koppel et al. [22] attributed the increase of exposure levels to the installation of base stations in the general environment.An assessment of RF-EMF exposure by Gallastegi et al. [23] by spot and personal measurements in the locations suggested similar pattern.Table 3. Reference levels for RF EMFs exposure averaged over 30 min for frequency band 100 kHz to 300 GHz.

Frequency range Incident Power Density (W/m²)
Frei et al. [24] used personal exposimeters to evaluate the levels of exposure and the impact of various RF-EMF sources in Switzerland.Mobile phone base stations were significant exposure contributors, accounting for 52.6% of total outdoor exposure.There is scarcity of literature around outdoor RF EMF exposures resulting from mobile base stations.Many studies have looked at mobile phone use and Wi-Fi RF EMF exposures in the microenvironments.A study by Ramirez-Vazquez et al. [8] assessing personal exposure to Wi-Fi in Mexican microenvironments measured over 24h showed the temporal characterization of personal exposure to RF-EMF from Wi-Fi.The registered mean for RF-EMF from the Wi-Fi, minimum values registered at night were 129.4µW/m² and 93.9µW/m², and maximum values registered at daytime, were 303.1µW/m² and 168.3µW/m² for the Wi-Fi 2G band and Wi-Fi 5G band, respectively.It is evident that a great variability on the data could be seen from this study and measurements obtained from microenvironments.Ramirez-Vazquez et al. [8] further suggest that this observation could be that throughout the measuring procedure, the measured data moves multiple times between the minimum and the maximum limits.Further explaining that the Wi-Fi operates on demand, which means that if it is not in use, there is no signal.
Furthermore, an exposure assessment by Peyman et al. [25] measured the electric field strengths of 15 laptops and 12 different access points (during Wi-Fi transmission) in UK schools.The highest electric field strength values measured at 0.5 m surrounding laptops and access points were 2.9 mV/m and 5.7 mV/m, respectively.The measurements showed that access point field strengths were more significant than laptop field strengths.The results also showed that the electric field strength of all Wi-Fi devices decreased fast with distance.At 0.5 m, the highest power density values for laptops and access points measured 22 µW/m² and 87 µW/m², decreasing to 4 µW/m² and 18 µW/m² at 1 m.The similarity of results obtained in the current study, and those of other studies is that measured values decrease drastically with an increased distance between measurement points.This is the first study in RSA to provide preliminary data on RF EMFs levels from mobile base stations, particularly in the northeastern region of RSA.A manufacturer-calibrated Acoustimeter was used to measure the RF levels and distance points were used to account for exposure variation per given distance from mobile base stations.This study did not classify the repeated measurements into different time categories, e.g.07:30 to 09:00am, 12:00 to 16:00pm and 18:00 to 21:00pm, and this presented a major limitation.However, the repeated measurements were taken on 30 min intervals, which accounted for time categories when people were most likely to perform internet-based activities.Mpumalanga is one of the fastest growing provinces in South Africa, with an expected increased use of EMF-related devices [26].There is a need for future studies to look at exposure patterns of residents from both outdoor and indoor RF sources in the broader context of South Africa.

Conclusions
There is scarcity of data in RSA regarding the exposure patterns of South African to RF EMF sources.The results of this study suggest that the population of Mpumalanga province in South Africa is exposed to outdoor RF EMFs below the ICNIRP guidelines.However, due to the presence of RF exposure levels in environments where people spent most of their time, it is essential, in the future, for mobile base station companies to consider the proximity of residential structures prior installing base stations, and regularly maintain the components of mobile base towers, especially in the current state of power load shedding in South Africa [27].

Figure 2 .
Figure 2. Illustration of measurement and distance points.

Table 1 .
Descriptive statistics of combined mean power density of all base stations.