Assessment of radioactivity in soil samples from Wolaita Sodo town, Ethiopia: implications for environmental and public health

Abstract

This study assesses the activity concentrations of the radionuclides ²³⁸U, ²³²Th, and ⁴⁰K in soil samples collected from Wolaita Sodo town, located in the Southern Nations, Nationalities, and Peoples' (SNNP) Region, Ethiopia. A gamma-ray spectrometer equipped with a NaI(Tl) detector was used for the measurements. The concentrations of ²³⁸U, ²³²Th, and ⁴⁰K varied from 3.25 ± 1.5 to 13.84 ± 2.2 Bq.kg⁻¹, 0.4 ± 0.9 to 85.12 ± 3.4 Bq.kg⁻¹, and 34.43 ± 2.7 to 748.07 ± 5.9 Bq.kg⁻¹, respectively. The average activity concentrations were 7.83 ± 1.9 Bq.kg⁻¹, 40.74 ± 2.7 Bq.kg⁻¹, and 161.63 ± 3.9 Bq.kg⁻¹ for ²³⁸U, ²³²Th, and ⁴⁰K, respectively. The average radium equivalent activity was 192.25 Bq.kg⁻¹, well below the recommended safety limit of 370 Bq.kg⁻¹. The average gamma dose rate, and annual effective dose rate were 35.68 nGy.h⁻¹, and 0.18 mSv.y⁻¹, respectively. The internal and external indexes are below the recommended limit set by UNSCEAR and ICRP. However, the estimated excess lifetime cancer risk and indoor radon concentrations are slightly higher. Despite this, the overall radiological impact on the environment and public health in the study area remains negligible. This study provides valuable baseline data for radiation protection and informs urban and environmental policy in the region.

Introduction

Naturally occurring radionuclides are recognized for their potential to cause significant radiation exposure to humans. Numerous epidemiological studies have documented the carcinogenic risks associated with the prolonged inhalation of Naturally Occurring Radioactive Materials (NORMs) [1]. These radionuclides exist in diverse forms, arising from radioactive decay, nuclear reactions, cosmic radiation interacting with the Earth's atmosphere, and geological materials. These radionuclides emit radiation in the form of alpha, beta, and gamma rays [2–5]. The primary contributors to naturally occurring radiation emanate predominantly include uranium (²³⁸U), thorium (²³²Th), potassium (⁴⁰K), and their associated radionuclides [6–8]. These can be found in soil, rocks, water, and minerals [9, 10].

While low-level natural radiation exposure is a normal part of everyday life and is generally considered to be harmless [6–8], extended contact with radiation intensities may lead to notable health effects [2–5, 11–14]. To mitigate risks, regulatory bodies have established exposure limits. For instance, the International Commission on Radiological Protection (ICRP), United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) and the World Health Organization (WHO), recommended that; (a) radium equivalent concentration should not exceed 370 Bq.kg⁻¹, (b) indoor radon levels should remain below 200 Bq.kg⁻¹, (c) internal and external hazard indexes should stay under 1, and (d) the annual gamma dose rate should be lower than 1.5 mSv.year⁻¹ [13, 15]. Assessing environmental radioactivity is crucial for public health safety, particularly due to concerns about natural and human-made sources [12, 15–22]. Soil plays a critical role in accumulating radionuclides, which is particularly relevant in rapidly urbanizing regions. Activities such as mining, agriculture, and industrial processes can elevate natural radiation levels, making it necessary to systematically monitor soil radioactivity as part of environmental health and safety management [23].

This study offers a unique contribution by being the first to assess soil radioactivity in Wolaita Sodo Town, Ethiopia, from an environmental and public health perspective. As Ethiopia undergoes rapid urbanization, understanding the levels of both naturally occurring and anthropogenic radionuclides in urban soils is essential for protecting public health and guiding policy development. Although several studies have assessed natural radioactivity in other parts of the world [12, 15–22], there is a critical lack of data from Ethiopian urban environments, which are influenced by unique geological conditions. Specifically, Wolaita Sodo lies along the western margin of the East African Rift Valley, underlain by Pliocene peralkaline ignimbrites and Tertiary basaltic sequences [24]. The area also lies downhill from Mount Damota, a significant volcanic complex contributing to elevated radon gas emissions and background radiation from primordial radionuclides like uranium and thorium [25, 26]. These geological and environmental characteristics make the region particularly relevant for studying soil radioactivity.

This research aims to address the gaps in environmental radioactivity data for Wolaita Sodo by evaluating key radionuclides and their potential risks. The study’s findings can inform urban planning, environmental policy, and public health strategies amid Ethiopia’s rapid development. Additionally, the research contributes to international efforts in radiation protection, providing valuable insights into future studies and risk mitigation strategies. The novelty of this work lies not only in addressing a critical local data gap but also in its potential to influence global frameworks for radiation safety.

Materials and methods

Description of the study area

The radionuclide concentration in the area is fundamentally dependent on the area's geological and pedagogical settings. Figure 1 depicts the study area, Wolaita Sodo town, in the Wolaita Zone of the South Nations, Nationalities, and Peoples' (SNNP) national regional state of Ethiopia. The town is approximately 330 kilometers from the capital, Addis Ababa. As a key agricultural hub and a rapidly developing urban center, Wolaita Sodo is experiencing significant population growth and development that may impact the local environment and radiological risks. Geographically, the town is part of the Main Ethiopian Rift, specifically at the margin of the southern Main Ethiopian Rift, which represents the least evolved rift segment compared to the Northern and Central Main Ethiopian Rift; the rift evolution shows a decrease in rift maturity from the north to the south [27]. In the Wolaita Sodo area, deformation is dominated by marginal faulting rather than axial tectonic-magmatic activity with fault systems are concentrated along the rift margins. Also, volcanic centers, i.e., Damota Volcano and Hobitcha Caldera, are significant features in the area, with volcanic activity extending into the quaternary period. Stratigraphically, the area includes tertiary basaltic sequences, Pliocene ignimbrites, and Pleistocene alluvial and lacustrine sediments interbedded with volcanic materials [28].

Figure 1.

Location of the study area a) soil map of Wolaita zone including wolaita Sodo town (bounded with the red line) b) elevation map and sampling points posted as triangles.

The formation of soil is primarily influenced by five key factors: parent materials, topography, climate, organisms, and time [29]. These elements play a crucial role in shaping soil characteristics. In the Wolaita Sodo area, the volcanic and tectonic history, along with interactions between the bedrock and climate conditions, have significantly impacted soil development. Additionally, topography stands out as a dominant influence in the study area, particularly due to the prominence of Mount Damota in the northern part of Wolaita Sodo town. As shown in Fig. 1, the area is characterized by various soil groups, including Andosols(T), Nitosols(N), Xelosols (x), Plandosols (w), and Cambisols(B). Wolaita Sodo town is covered in Nitosols soil group, which is derived from volcanic rocks that are possibly from tertiary basaltic sequences [27, 28].

Figure 1(a) shows the soil map of the Wolaita Zone, Ethiopia. Furthermore, the elevation of the area, including the sampling locations, ranges between 1817 and 2214 meters above sea level (Fig. 1(b)). Over the past three years, the specific humidity at 2 meters above ground level has consistently ranged between 10 and 12 g/kg, with the soil wetness index remaining close to 1 throughout this period. The area’s precipitation has varied between 0.6 and 1.9 mm/day, based on observations from the same three-year period.

Radionuclides activity measurement

Natural radioactive elements in soil typically originate from the mechanical and chemical weathering of rocks, with rain and water flows transporting the disintegrated material into the soil. Therefore, the concentration of natural radioactive elements in soil tends to reflect the composition of the parent rocks from which the soil was derived. A total of 36 soil samples were systematically collected from Wolaita Sodo town to ensure representative coverage of the area. The samples were taken from an average depth of 30 cm to minimize contamination from dust and to ensure the collection of pure soil. Each sample was dried at 200 °C for 10 hours to remove moisture, then crushed and sieved through a 200 μm mesh. A 100-gram portion of each sample was placed in a cylindrical plastic container with a diameter of 8.9 cm (3.5 in.) and a length of 7.6 cm (3 in.). To allow equilibrium between short-lived radionuclide progenies and their parent isotopes, the sealed samples were stored in calibrated plastic containers for one month before analysis.

The concentration of radionuclides (⁴⁰K, ²³⁸U, and ²³²Th) in the soil samples were measured using a low background multichannel gamma-ray spectrometer system [9, 22, 30]. The system employed a Sodium Iodide (NaI (Tl)) detector coupled with a multichannel analyzer and controlled by Genie software. The detector, with a crystal size of 3 in. × 3 in., was manufactured by GF Instruments, Geophysical Equipment and Service, and stainless-steel housing shield to reduce background interference. Firstly, the NaI(Tl) detector probe was activated and fitted into an empty plastic container prepared to hold the sample. Secondly, before measurements took place, the detector automatically calibrated and stabilized itself using its internal cesium (¹³⁷Cs) reference at an energy of 661.2 keV, with an efficiency of 8.7%. The background radiation was then measured and subtracted to correct for radiation interference. Then, a 100-gram soil sample was placed in contact with the detector probe and exposed for three minutes. The detector measured the complete spectrum and automatically displayed the calculated results: ⁴⁰K (in %), ²³⁸U (in ppm), ²³²Th (in ppm), and gamma dose rate (in nGy/h). The concentration of ⁴⁰K is determined directly, while the concentrations of ²³⁸U and ²³²Th are based on the detection of the radioisotopes ²¹⁴Bi and ²⁰⁸Tl, which are part of their respective decay series. The natural dose rate value (in nGy/h) is calculated from these concentrations according to IAEA recommendations. The measurement uncertainties were calculated to accurately determine the specific activities of each radionuclide in the environmental samples.

Each soil sample was analyzed using the gamma-ray spectrometer. Before measuring the radioactivity of the soil samples, background radiation was assessed by sealing an empty plastic container within the detector. The background value was then subtracted from the measurements of the soil samples to ensure accurate results.

Radium and radon concentration measurement

Commercially available LR-115 Type 2 strippable films, manufactured by DOSIRA France, contain an active layer of cellulose nitrate on a 100 μm transparent polyester base. These films were used to measure radon concentrations. In this study, 36 pieces of detector film, each measuring 2 × 2 cm, were employed. Soil samples were placed at the bottom of leak-proof plastic containers (using the “can technique”) with a height of 10 cm and a radius of 3.5 cm. The detector was secured inside the mouth of the container, facing the sample, to capture alpha radiation. The containers were sealed for one month to allow equilibrium between radium and radon progeny. During the exposure period, the detector films were exposed to alpha radiation emitted by the soil samples. After one month, the detectors were etched for 120 minutes in 2.5 N NaOH at 60 °C, and the alpha tracks were counted using an optical microscope. The observed alpha particle track density was determined using the equation [25].

(1)

where ρ is the track density, K is the sensitivity factor, and its value is 0.0312 tracks m⁻²/day/Bq/m³, T_e is effective time, and C_Ra is the effective radium content. The effective exposure time is calculated using the following equation.

(2)

The effective radium content C_Ra can be calculated using equation (4) [26].

(3)

where h is the distance between detector and surface of the sample, A is the area cross-section of the cylindrical can, and M is the mass of the sample. The concentration of radon in the soil samples can be calculated using the following equation [31].

(4)

where C_Rn is the concentration of radon, C_Ra is the effective radium content of the given sample, is the decay constant for radon (hour ^{− 1}), T is the actual exposure time (hour).

The mass and surface exhalation rate of radon gas has been estimated based on equation (5) and (6) [31];

(5)

(6)

Where λ_Ra is the decay constant for radium, λ_Rn is the decay constant of Radon, T_e is effective exposure time.

Radiological effect calculations of ²³⁸U, ²³²Th and ⁴⁰K

Radium equivalent concentration

Radium equivalent concentration (²²⁶Ra_eq) measures the effective activity of all radionuclides (⁴⁰K, ²³⁸U, and ²³²Th), converted into an equivalent concentration. This concept uses ²²⁶Ra as a reference due to varying radioactivity and radiation dose potential among radionuclides, applying equilibrium factors or dose conversion factors based on their radiation emissions and dose conversion coefficients [32–35]. The calculation of Radium equivalent (Ra_eq) assumes that a dose rate equivalent to 370 Bq/kg of ²²⁶Ra is caused by 259 Bq/kg of ²³²Th and 4810 Bq/kg of ⁴⁰K. This relationship can be expressed using equation (7) [9, 30].

(7)

Where C_Ra, C_Th and C_K are the average activity concentration of ²³⁸U, ²³²Th and ⁴⁰K in the soil samples respectively.

Absorbed gamma dose rates

The absorbed gamma dose rate indicates the rate of gamma radiation emitted per unit time from the radionuclide source [16, 36]. The average absorbed gamma dose rate in air above the surface of the earth due to the fundamental radionuclide’s can be estimated using the formula given by equation (8) [15, 21, 37].

(8)

Where, is the absorbed dose rate at 1 m from the ground, C_U, C_Th, and C_k are the activity concentrations of ²³⁸U, ²³²Th and ⁴⁰K respectively.

Similarly, the annual effective radiation dose rate received from the soil was calculated by the following formula in Equation (9) [13].

(9)

To estimate the, the outdoor usage factor (0.2) and the conversion coefficient from the absorbed dose rate in the air to the effective dose (0.7 Sv.Gy⁻¹) were used. This number assumes that people spend 20% of their time outside their homes.

Internal and external radiation index

Internal and external indexes help quantify and evaluate radiation exposure levels, assess risks, and give a clue to establish radiation protection guidelines and safety standards [38]. The external (H_ex) and internal (H_in) hazard index reflects the risk of external and internal exposure to ionizing radiation. The H_ex and H_in of the radiation emitted from the soil samples were calculated according to equations 10 and 11, respectively.

(10)

(11)

Excess lifetime cancer risk

Excess Lifetime Cancer Risk (ELCR) quantifies the additional cancer risk resulting from ionizing radiation exposure, indicating an increased probability compared to the baseline risk in an unexposed population. It often represents the estimated fraction of the population expected to develop cancer due to radiation exposure from soil, which can migrate and accumulate in dwellings, causing carcinogenic effects. The possibility of developing carcinogenic effects in an individual’s life can be estimated by calculating the ELCR using the equation (12) [17, 20, 39].

(12)

Where , is the annual effective dose, DL is the average duration of lifetime in the study area (62 years) and RF is the fatal risk factor assumed to be 0.05 Sv⁻¹ for public as per ICRP-106 [40].

Results and discussion

Activity concentrations of radionuclides

In this study, we investigated the activity concentrations of ²³⁸U, ²³²Th, and ⁴⁰K in the soil samples of the study area. The measured activity concentrations of ²³⁸U, ²³²Th, and ⁴⁰K are shown in Table 1 and Fig. 2. It was found that the activity concentration of ²³⁸U ranged from 3.25 ± 1.5 to 13.84 ± 2.2 Bq.kg⁻¹, with an average value of 7.83 ± 1.9 Bq.kg⁻¹, and a standard deviation of 2.59. The average value is much lower than the worldwide average value of 35 Bq.kg⁻¹ [13]. The activity concentration of ²²⁶Ra emanating from ²³⁸U varied from 89.3 to 522.3 Bq.kg⁻¹ with an average value of 265.7 Bq.kg⁻¹. The average value is less than the recommended value of 370 Bq.kg⁻¹. However, the overall equivalent activity concentration could be higher. Additionally, the activity concentration of ²³²Th varied from 0.4 ± 0.9 to 85.12 ± 3.4 Bq.kg⁻¹ With an average value of 40.74 ± 2.7 Bq.kg⁻¹ and a standard deviation of 21.32. The higher variation between the maximum and minimum values could be due to the low abundance of ²³²Th minerals bearing in the local geology, weathering and erosion processes in the study area. The average value is higher than the worldwide average value of 30 Bq.kg⁻¹, which is recommended by UNSCEAR 2000 [13]. The difference is attributed to soil and rock types as well as geochemical composition and origin of soil type in a particular area. The results showed an increase in ²³²Th concentration of more than 36.33% compared to the international recommended value. It suggests that prolonged exposure to the soil in the study area could present considerable health risks.

Table 1.

Radioactivity concentrations in the study area.

Sample ID	226Ra (Bq.kg−1)	222Rn (Bq.m−3)	238U (Bq.kg−1)	232Th (Bq.kg−1)	40K (Bq.kg−1)	232Th/238U
S-1	129.96	108.78	5.28 ± 1.7	51.90 ± 3.0	262.92 ± 4.5	9.83
S-2	168.42	140.97	5.68 ± 1.7	43.20 ± 2.9	68.86 ± 3.2	7.60
S-3	115.82	96.94	7.71 ± 1.9	16.10 ± 2.2	109.55 ± 3.6	2.08
S-4	89.30	74.74	6.09 ± 1.8	18.50 ± 2.3	97.03 ± 3.5	3.04
S-5	255.51	213.86	7.88 ± 1.9	0.40 ± 0.9	34.43 ± 2.7	0.05
S-6	192.74	161.32	5.68 ± 1.7	39.50 ± 2.8	90.77 ± 3.5	6.95
S-7	232.96	194.99	5.28 ± 1.7	51.90 ± 3.0	97.03 ± 3.5	9.83
S-8	157.37	131.72	8.12 ± 1.9	45.70 ± 2.9	106.42 ± 3.6	5.63
S-9	182.57	152.81	6.50 ± 1.8	46.90 ± 2.9	93.9 ± 3.5	7.22
S-10	198.70	166.32	6.50 ± 1.8	51.90 ± 3.0	150.24 ± 3.9	7.98
S-11	195.83	163.91	7.31 ± 1.8	53.10 ± 3.0	165.89 ± 4.0	7.27
S-12	128.20	107.30	10.96 ± 2.0	43.20 ± 2.9	153.37 ± 3.9	3.94
S-13	257.72	215.71	9.74 ± 2.0	54.30 ± 3.0	466.37 ± 5.2	5.58
S-14	209.98	175.75	10.15 ± 2.0	54.30 ± 3.0	109.55 ± 3.6	5.35
S-15	236.50	197.95	11.37 ± 2.1	18.50 ± 2.3	162.76 ± 4.0	1.63
S-16	357.18	298.96	11.77 ± 2.1	17.30 ± 2.3	97.03 ± 3.5	1.47
S-17	336.50	281.94	9.74 ± 2.0	50.60 ± 3.0	122.07 ± 3.7	5.20
S-17	274.52	229.77	8.93 ± 1.9	54.30 ± 3.0	150.24 ± 3.9	6.08
S-19	341.05	285.46	6.50 ± 1.8	34.60 ± 2.7	150.24 ± 3.9	5.32
S-20	195.83	163.91	6.50 ± 1.8	43.20 ± 2.9	178.41 ± 4.1	6.65
S-21	449.13	375.92	13.84 ± 2.2	42.40 ± 2.9	184.67 ± 4.1	3.06
S-22	352.32	294.89	3.65 ± 1.5	39.50 ± 2.8	118.94 ± 3.7	10.82
S-23	369.12	308.95	8.12 ± 1.9	30.90 ± 2.6	178.41 ± 4.1	3.80
S-24	439.12	367.78	6.09 ± 1.8	63.00 ± 3.2	194.06 ± 4.2	10.34
S-25	447.80	374.81	9.74 ± 2.0	85.12 ± 3.4	244.14 ± 4.4	13.18
S-26	405.81	339.66	6.09 ± 1.8	35.80 ± 2.7	147.11 ± 3.9	5.88
S-27	177.71	148.74	5.28 ± 1.7	24.70 ± 2.5	209.71 ± 4.3	4.68
S-28	240.04	200.91	4.87 ± 1.7	51.90 ± 3.0	212.84 ± 4.3	10.65
S-29	522.07	436.97	3.25 ± 1.5	58.00 ± 3.1	134.59 ± 3.8	17.87
S-30	370.00	309.69	8.53 ± 1.9	9.90 ± 2.0	748.07 ± 5.9	1.16
S-31	296.18	247.90	10.56 ± 2.0	19.80 ± 2.4	128.33 ± 3.8	1.87
S-32	218.38	182.78	8.93 ± 1.9	44.50 ± 2.9	56.34 ± 3.1	4.98
S-33	183.90	153.92	10.96 ± 2.0	34.60 ± 2.7	109.55 ± 3.6	3.15
S-34	330.66	276.76	12.59 ± 2.1	34.60 ± 2.7	75.12 ± 3.3	2.75
S-35	199.37	166.87	4.87 ± 1.7	39.50 ± 2.8	62.60 ± 3.1	8.11
S-36	307.00	150	6.90 ± 1.8	63.00 ± 3.2	147.11 ± 3.9	9.13

Figure 2.

Concentration of radionuclides in the soil samples of the study area.

Moreover, the activity concentrations of ⁴⁰K varied from 34.43 ± 2.7 to 748.07 ± 5.9 Bq.kg−1 with an average value of 161.63 ± 3.9 Bq.kg⁻¹, and a standard deviation of 125.76. The highest value of ⁴⁰K was measured in the Dipoturf area (S-30), this may be due to felsic and felspathic sediments of chemical weathering and complexes of organic constituents in the aqueous phase near aquifer-bearing formation [24]. In contrast the minimum value was recorded near St. Kuskuam church (S-5). The average value of 40 K is lower than the worldwide average activity concentration of 400 Bq.kg⁻¹ reported by the UNSCEAR, 2000 [13]. As shown in Fig. 2 the ⁴⁰K concentration is much higher than ²³²Th and ²³⁸U.

In this study, we also evaluated the ratio of ²³²Th/²³⁸U which refers to the relative abundance of ²³²Th compared to ²³⁸U in each sample or environment. On average, the ratio of ²³²Th/²³⁸U in the Earth's crust is approximately 4, indicating that ²³²Th is typically four times more abundant than ²³⁸U. However, it is important to note that this ratio can vary significantly depending on specific geological conditions, mineral deposits, and other factors. The result presented in Fig. 2 showed that the concentration of ²³²Th/²³⁸U varied from 0.05 to 17.87, with an average value of 6.1, which is higher than the world’s average value.

Finally, we assessed the radium equivalent concentration to predict the potential radiation hazard associated with the presence of radionuclides in the soil. Radium, a naturally occurring radioactive element that is part of the ²³⁸U decay chain, is found in trace amounts in most soils around the world. The measured results showed that the radium equivalent activity concentration varied from 34.96 to 598.70 Bq.kg⁻¹, with an average value of 192.25 Bq.kg⁻¹. The average radium equivalent activity concentration is above the maximum recommended limit of 370 Bq.kg⁻¹. This indicated that long-term exposure to the soil and indoor environment could pose health hazards. However, further investigation is needed to confirm this significant effect.

Descriptive statistics are used to summarize and describe the main features of the dataset of the radionuclides shown in Table 2. From the minimum and maximum values, we can see that the radionuclide concentration exhibits a wide range of variations across the soil samples. In all the radionuclides, the mean value is greater than the standard deviation, indicating lower variability. ²²⁶Ra, ²²²Rn, ²³⁸U, and ⁴⁰K show positive skewness, indicating a tendency towards higher values with varying degrees of skew. While ²³²Th shows minimal negative skewness, suggesting a nearly symmetric distribution with a slight tendency towards lower values. Additionally, it exhibits a strong positive skewness, indicating a pronounced right skew with a significant tail on the higher end of the data range. ²²⁶Ra, ²²²Rn, and ²³⁸U all have negative kurtosis, indicating flatter distributions with lighter tails and fewer outliers compared to a normal distribution. In contrast, ²³²Th shows positive kurtosis, meaning it has a peakier distribution with heavier tails and more outliers. Moreover, ⁴⁰K exhibits extremely high positive kurtosis, indicating a very peaked distribution with exceptionally heavy tails and a high number of extreme values or outliers.

Table 2.

The descriptive statistics of ²³⁸U, ²³²Th, ⁴⁰K, ²²⁶Ra and ²²²Ra concentrations in the soil samples.

Sample	226Ra (Bq.kg−1)	222Rn (Bq.m−3)	238U (Bq.kg−1)	232Th (Bq.kg−1)	40K (Bq.kg−1)	232Th/238U	Ra.eq (Bq.kg−1)
Minimum	89.3	74.74	3.25 ± 1.5	0.4 ± 0.9	34.43 ± 2.7	0.05	34.96
Maximum	522.07	436.97	13.84 ± 2.2	85.12 ± 3.4	748.07 ± 5.9	17.87	598.7
Skewness	0.54	0.61	0.40	−0.17	3.41	0.91	2.36
Kurtosis	−0.47	−0.45	−0.54	0.49	14.06	1.41	7.19
Average	265.70	219.44	7.83 ± 1.9	40.74 ± 2.7	161.63 ± 3.9	6.11	192.25
SD	107.28	89.14	2.59	17.17	125.76	3.74	100.82

Furthermore, the spatial map of the elemental concentrations of ²²⁶Ra, ²³⁸U, ²³²Th, and ⁴⁰K was produced using the Phyton programming language, shown in Fig. 3. The aim of illustrating the distribution is to understand the geological impact on the distribution of radiation. The maps of ²³⁸U, ²³²Th, and ⁴⁰K exhibit color variations in Fig. 3, depicting possible changes in the natural radiation within the study area. These radiation changes could be due to the presence of different minerals containing radiation, distinct lithology, volcanic rocks, dark shale, and phosphate. The area under investigation is near the Ethiopian Rift Valley, which is why the concentration is significantly higher. In the Rift Valley, there is an active volcano, which contributes to the activity of natural radiation. With the aid of color contrast, the study area was marked by high and low radiation concentrations. High radiation is indicated by yellow color, while lower radiation is represented by green and blue colors. From fig. 3(a), the higher concentrations of uranium were observed in the highland areas relative to the lowlands, especially in increments near the Damot Hill. Finally, the maximum thorium concentration was observed in the two basins that encompass the center of the town, as shown in fig. 3(b). Additionally, as shown in Fig. 3(c), extreme concentrations of potassium were observed in the lowland areas relative to highland areas, which may be due to the weathering effects.

Figure 3.

Spatial map of radionuclides derived from machine learning based gridding a) ²²⁶Ra b) ²³⁸U, c) ²³²Th, and d) ⁴⁰K.

The histogram shows the frequency distribution of all corresponding radionuclides in the examined soil samples (see Fig. 4). This analysis underscores the diversity in soil samples, and the distribution of variables. The plot shows a suggestion that there is widespread variability in the data. As shown in figs. 4 (a) and (b), the ²²⁶Ra and ³³⁸U exhibit a normal (bell-shaped) distribution with a slight left skew distribution, i.e., the tail of the graph extends to the right and most of the data are on the left. More frequent ²²⁶Ra values are observed at around 200 Bq/kg. The distribution indicates that the highest peak value is on the recommended limits, but most of the samples fall above the permissible limit. The distribution of ²³²Th is approximately bell-shaped with spread distribution and shows variability in the thorium content of the soil samples, shown in Fig. 4(c). The highest histogram peak is observed around 50 Bq/kg, which is above the worldwide average value of 35 Bq/kg [13]. On the other hand, ⁴⁰K showed a spread distribution with left skewed (i.e., the tail of the graph extends to the right and most of the data are on the left activity concentration representatives), and outliers are observed (see Fig. 4(d)). The peak histogram distribution was observed at around 100 Bq/kg, which is much lower than the worldwide average value of 400 Bq/kg [13]. It is also noted that some histogram bars are isolated from the rest of the histogram, which may indicate outliers among the data points.

Figure 4.

Frequency distribution of radionuclides in soil samples.

Furthermore, we compare the current results for ²³⁸U, ²³²Th, and ⁴⁰K with findings from other studies conducted in various countries. Table 3 compares the values of radionuclides found in this study with those from the literature. The study's findings demonstrate lower concentrations of radioactive elements, particularly ²³⁸U and ⁴⁰K, compared to both the global average and other countries, indicating reduced natural radioactivity and, consequently, lower radiation exposure risk in the study area. Although ²³²Th levels are slightly above the global average, they remain moderate [13]. From a radiological safety perspective, these results are promising, suggesting minimal concerns about natural radiation exposure and aiding the development of effective radiation safety guidelines in line with global standards.

Table 3.

Average activity concentrations of ²²⁶Ra, ²³²Th and ⁴⁰K for all soil samples under investigation are compared with those from other countries.

Country	238U	232Th	40K	Reference
Nigeria	141.34	22.50	13.71	[3]
Turkey	25.00	50.00	228.00	[6]
Turkey	28.00	40.00	667.00	[20]
India	99.30	112.90	308.90	[21]
Saudi Arabia	44.10	29.30	251.50	[30]
Egypt	193.00	63.00	1034.00	[32]
Oman	22.80	39.90	253.16	[37]
China	518.90	2567.70	3820.40	[41]
Nepal	59.06	69.59	592.46	[42]
Ethiopia	51.90	68.32	220.00	[43]
UNSCEAR	30.00	35.00	400.00	[13]
Ethiopia	7.83 ± 1.9	40.74 ± 2.7	161.63 ± 3.9	This study

Radiological effect assessment of radionuclides

Radiological effects assessment is the process of evaluating and estimating the potential health effects that may result from ionizing radiation exposure. The amount of radiation dose can be estimated using various dosimetry methods, environmental measurements, and computer simulations [44, 45]. Once the radiation doses are estimated, a health risk evaluation is conducted to assess the probable adverse health effects. Epidemiological studies, animal experiments, and other scientific data are used to quantify the risks associated with different radiation doses. The assessed radiation risks are compared with established radiation protection standards and guidelines. These standards provide reference levels or limits for radiation exposure that aim to protect individuals and the public from excessive radiation doses [44, 45].

As shown in Table 4 and Fig. 5, the calculated gamma absorbed dose rate varies from 5.32 to 91.99 nGy.h⁻¹, with an average value of 35.68 nGy.h⁻¹, which is below the recommended value [46]. According to UNSCEAR 2000, 65% of the world's average outdoor exposure is due to terrestrial gamma radiation of dose 59 nGy.h⁻¹ [46]. The calculated annual effective dose equivalent varies from 0.03 to 0.45 mSv.y⁻¹, with an average value of 0.18 mSv.y⁻¹, which is higher than the world-wide average value of 0.051 mSv.y⁻¹ [46]. The study area presented a higher radiation rate. These radioactive elements can mix with food and then pass up the food chain to plants, animals, and humans. On the other hand, the external hazard index of naturally occurring ionizing radiation was between 0.03 and 0.57, with an average value of 0.22. Internal exposure to radon and its progeny is controlled by the internal index, which is between 0.05 and 60 with an average value of 0.24. Both the external and internal hazard indexes are less than the recommended value.

Table 4.

Absorbed dose rate, external and internal hazard index, annual effective dose equivalent and excess lifetime cancer risk of the soil samples collected from Wolaita Sodo, Ethiopia.

S.ID	(nGy.h−1)	(mSv.y−1)	Hex	Hin	ELCR 10−3	Ex(M) × 10−6 (Bq.kg−2 day−1)	Ex(S) × 10−5 (Bq.m−2 day−1)
S-1	44.75	0.22	0.27	0.28	0.68	4.30	2.0
S-2	31.59	0.15	0.20	0.21	0.48	5.60	2.5
S-3	17.85	0.09	0.11	0.13	0.27	3.90	1.7
S-4	18.03	0.09	0.11	0.12	0.27	3.00	1.3
S-5	5.32	0.03	0.03	0.05	0.08	8.50	3.8
S-6	30.27	0.15	0.19	0.20	0.46	6.40	2.9
S-7	37.83	0.19	0.23	0.25	0.58	7.80	3.5
S-8	35.79	0.18	0.22	0.24	0.54	5.20	2.4
S-9	35.25	0.17	0.22	0.24	0.54	6.10	2.7
S-10	40.62	0.20	0.25	0.27	0.62	6.60	3.0
S-11	42.37	0.21	0.26	0.28	0.64	6.50	2.9
S-12	37.55	0.18	0.23	0.26	0.57	4.30	1.9
S-13	56.74	0.28	0.33	0.36	0.86	8.60	3.9
S-14	42.05	0.21	0.26	0.29	0.64	7.00	3.2
S-15	23.21	0.11	0.14	0.17	0.35	7.90	3.6
S-16	19.93	0.10	0.12	0.15	0.30	1.20	5.4
S-17	40.15	0.20	0.25	0.27	0.61	1.10	5.1
S-17	43.19	0.21	0.27	0.29	0.66	9.20	4.1
S-19	30.17	0.15	0.18	0.20	0.46	1.10	5.1
S-20	36.54	0.18	0.22	0.24	0.56	6.50	2.9
S-21	39.70	0.19	0.24	0.28	0.60	1.50	6.7
S-22	30.50	0.15	0.19	0.20	0.46	1.10	5.3
S-23	29.85	0.15	0.18	0.20	0.45	1.20	5.5
S-24	48.96	0.24	0.30	0.32	0.74	1.50	6.6
S-25	91.99	0.45	0.57	0.60	1.20	1.50	6.7
S-26	30.57	0.15	0.19	0.20	0.46	1.40	6.1
S-27	26.10	0.13	0.15	0.17	0.40	5.90	2.7
S-28	42.47	0.21	0.26	0.27	0.65	8.00	3.6
S-29	42.15	0.21	0.26	0.27	0.64	1.70	7.8
S-30	41.11	0.20	0.22	0.24	0.63	1.20	5.6
S-31	22.19	0.11	0.13	0.16	0.34	9.90	4.4
S-32	33.35	0.16	0.21	0.23	0.51	7.30	3.3
S-33	30.53	0.15	0.19	0.22	0.46	6.10	2.8
S-34	29.85	0.15	0.18	0.22	0.45	1.10	5.0
S-35	28.72	0.14	0.18	0.19	0.44	6.60	3.0
S-36	47.37	0.23	0.29	0.31	0.72	4.76	3.97
Min.	5.32	0.03	0.03	0.05	0.08	1.10	1.30
Max.	91.99	0.45	0.57	0.6	1.4	9.90	7.80
Skew	1.53	1.55	1.59	1.68	1.52	−0.03	0.55
Kurt	6.93	6.90	7.46	7.79	6.91	−1.40	−0.45
Ave.	154.10	0.76	0.22	0.24	0.54	2.90	1.62
STD	13.79	0.07	0.09	0.09	0.21	2.82	1.58

Figure 5.

The radiological indexes in the study area.

Finally, the approximation of excess lifetime cancer risk varies from 0.08 × 10⁻³ to 1.20 × 10⁻³, with an average value of 0.54 × 10⁻³. The average value is slightly higher than the recommended action level of 0.29 × 10⁻³ [14, 39, 46]. The value recommended by the ICRP for occupational exposures typically falls between 0.05% and 5% per Sievert (Sv) of effective dose [11]. Similarly, the UNSCEAR provided a reference value of 0.2% to 4% per Sv of effective dose [13]. According to the result, out of 1000 people, 29 may have a chance of developing cancer due to long-term radiation exposure. According to these measured values, it can be concluded that the radiological effect in the study area is not significant.

The descriptive statistics summarize the main features of the dataset of radiological parameters, as shown in Table 4. Most indices show a right-skewed distribution, meaning they have a tail extending towards higher values. Many of the indices have high positive kurtosis, which means they have leptokurtic distributions with heavy tails and sharp peaks.

Radium and radon concentrations

Indoor radon is a naturally occurring radioactive gas that can seep into buildings from the ground. It is colorless, odorless, and tasteless, making it difficult to detect without specialized equipment. Radon is formed from the decay of uranium in soil, rock, and water, and it can accumulate in enclosed spaces, particularly in poorly ventilated areas. Prolonged exposure to high levels of indoor radon can increase the risk of lung cancer, making it a significant public health concern. Testing for radon levels and implementing mitigation measures are important steps to reduce exposure and minimize health risks associated with indoor radon [47, 48].

Besides radium equivalent calculation, radium concentration was measured using a plastic track detector. The results revealed that the radium concentration in the soil samples varied from 89.30 to 522.07 Bq.kg⁻¹ with an average value of 265.70 Bq.kg⁻¹. The result showed that the average radium equivalent activity concentration is below the maximum recommended limits of 370 Bq.kg⁻¹. However, long-term exposure may cause health hazards. On the other hand, the estimated indoor radon concentration in the soil samples varies from 74.74 to 436.97 Bq.m⁻³ with an average value of 219.44 Bq.m⁻³. This result is above the recommended value of 200 Bq.m⁻³ [49]. Radon gas can seep from soil into the indoor environment. The alpha particles emitted from the inhaled radon gas may pose health hazards.

Our analysis of soil samples reveals that radon inhalation exceeds the recommended threshold. In our previous study, we found that the inhalation dose demonstrates a higher magnitude in construction materials compared to that derived from soil samples [26]. These findings underscore the necessity for further investigation into the complex interplay of environmental and artificial factors contributing to elevated inhalation doses in both outdoor and indoor environments. The predominance of moisture-retentive soils like Nitosols and Andosols further contributes to radon emanation and elevated exhalation rates, warranting attention from health authorities for risk assessment and mitigation measures [26].

The estimated indoor radon concentration in the soil samples varied from 74.74 to 436.97 Bq m⁻³, with an average value of 219.44 Bq m⁻³. These values establish a significant baseline for understanding radon levels within the context of the area's geological composition and history. The investigation into radon emanation in the Wolaita Sodo area has yielded crucial insights regarding the local geological framework. The calculated radon mass exhalation rate in the region varied from 1.1 to 9.9 Bq kg⁻².day⁻¹, with an average value of 4.76 Bq kg⁻².day⁻¹. Similarly, the radon surface exhalation rate ranged from 1.3 to 7.8 Bq kg⁻².day⁻¹, averaging 3.97 Bq kg⁻².day⁻¹. The emanation of radon is fundamentally influenced by the geological and pedological settings of the region. The study area is situated within the Main Ethiopian Rift, particularly at the fringe of the southern segment, which is characterized by its relatively low degree of geological evolution compared to the more mature northern and central segments of the rift [27]. This results in varied structural stability and fault dynamics that affect radon emanation. Moreover, the area is marked by marginal faulting and features volcanic centers like the Damota Volcano, contributing to the presence of radon-rich materials.

The region's stratigraphy includes Tertiary basaltic sequences and volcanic deposits, which enhance radon release due to their permeability. Soil formation is shaped by parent materials, topography, climate, organisms, and time, with Nitosols derived from volcanic activity being particularly conducive to higher radon production. Mount Damota's topography likely influences radon surface exhalation and airflow patterns. Additionally, the study area is one of the agricultural hubs in Ethiopia. Thus, the radon exhalation rate increases to the Earth's surface, raising contamination levels through food streams and indoor radon concentrations.

Multivariant statistical analysis

We applied multivariate statistical analysis to examine multiple radiation parameters to understand the relationships, patterns, and effects among them. We quantified the linear relationships between the variables using the Pearson correlation coefficient, as shown in Fig. 6. The coefficients provide insights into the relationships between various radionuclides and related parameters, which could be useful for understanding radioactivity in environmental or geological studies. A coefficient of 1 signifies a perfect positive correlation, 0 means no correlation, and − 1 indicates a perfect negative correlation. Values from 0.7 to 1 reflect a strong positive correlation, those 0.3 to 0.7 denote a moderate correlation, and values 0 to 0.3 represent a weak positive correlation.

Figure 6.

The Pearson correlation between the radionuclides and radiological parameters.

²³²Th shows a highly positive correlation with gamma dose rate (0.86), H_ex (0.88), H_in (0.87), and ELCR (0.85). This indicates that as ²³²Th increases, these variables tend to increase as well, and they share a strong linear relationship. D_γ is highly correlated with H_ex (1.00), H_in (0.99), and ELCR (1.00). This suggests that D_γ, H_ex, H_in, and ELCR move together in a similar manner. However, ²²⁶Ra and ²²²Rn have a perfect positive correlation (1.00), suggesting they are directly proportional or possibly measuring the same thing in this context. ²²⁶Ra and ²²²Rn both show moderate positive correlations (0.33–0.34) with D_γ, H_ex, H_in, and ELCR. ⁴⁰K has moderate positive correlations (0.32–0.39) with D_γ, H_ex, H_in, and ELCR. It shows very weak or no correlation with ²³⁸U (−0.03) and ²³²Th (−0.03). On the other hand, ²³⁸U has weak positive correlations (0.05–0.12) with most other parameters, except for a weak negative correlation (−0.14) with ²³²Th. The unexpected lack of correlation between ²³⁸U and other radioactive parameters may be attributed to various factors including geological factors, its unique chemical behavior and oxidation states, environmental factors like pH and redox conditions, disturbances in the uranium decay chain, and minimal variation in uranium concentrations across samples. These factors contribute to its differing distribution and correlation with other elements.

²³²Th shows a very low negative correlation with ⁴⁰K (−0.03), indicating almost no relationship between these variables. ²³²Th appears to be a significant contributor to overall radiation measures, more so than ²³⁸U or ⁴⁰K in this dataset. Moreover, CLCR shows strong positive correlations with ²³²Th (0.85), D_γ (1.00), H_ex (1.00), and moderate correlations with others, suggesting that it aligns closely with these variables.

Figure 7 shows the dendrogram, which illustrates a hierarchical cluster where each sample is grouped into clusters based on their similarities. The height of the branches represents the distance between clusters, with closely related branches indicating more similar clusters. The red dashed line marks the threshold for cutting the dendrogram, forming distinct clusters. Each vertical line that the red line intersects corresponds to a separate cluster. This dendrogram highlights the hierarchical structure of the radionuclide data. Such analysis provides valuable insights into specific groupings and assists in determining the optimal number of clusters. The type of clustering depends on the choice of distance metric and linkage method, and understanding these aspects allows researchers to interpret the dendrogram effectively and extract meaningful information from the data.

Figure 7.

Agglomerative hierarchical clustering analysis of radionuclides and associated radiological parameters.

Furthermore, the elbow method is used to confirm whether hierarchical agglomerative clustering produces clusters like those identified by K-means clustering. Figure 8 illustrates the optimal number of clusters for the dataset related to radionuclide concentrations. The x-axis represents the number of clusters, while the y-axis shows the inertia, defined as the sum of squared distances between samples and their closest cluster centroid. The optimal number of clusters corresponds to the point where the rate of decrease in inertia sharply slows, forming a visible “elbow” in the plot. The blue line traces how inertia decreases as the number of clusters increases, with a steep decline initially, followed by a gradual leveling-off. The red dashed line marks the “elbow point,” suggesting that the optimal number of clusters is approximately 3. Hence, selecting 3 clusters would strike a balance between model performance and simplicity. For the radionuclide dataset, this result indicates that the concentrations and associated parameters can be effectively grouped into three distinct clusters. However, while the elbow method provides a straightforward way to identify the optimal cluster count, it is not always definitive. Additional factors, such as domain knowledge and the specific characteristics of the data, should also be taken into consideration when finalizing the clustering approach.

Figure 8.

The optimal number of clusters of radionuclides and associated radiological parameters.

Figure 9 shows a violin plot that displays the distribution of the data across different variables. Plotted along the x-axis are the variables ²³⁸U, ²³²Th, ⁴⁰K, ²²⁶Ra, and ²²²Rn, while the y-axis represents the range of values for each variable. The shape of each violin reflects the distribution of data, with wider sections indicating higher density. Inside each violin, a black box plot shows the median, quartiles, and potential outliers, providing a detailed overview of data variability.

Figure 9.

Violine plot for the radionuclides.

²²⁶Ra and ²²²Rn exhibit broader distributions, suggesting greater variability, while variables like ²³⁸U have a much narrower range of values. Specifically, the distribution of ²³⁸U is very narrow and centered at lower values. ²³²Th also shows a narrow distribution, centered around low to moderate values. For ⁴⁰K, the distribution has a moderate spread, with a wider range than ²³⁸U and ²³²Th. The distributions of ²²⁶Ra and ²²²Rn are wide, extending to higher ranges, indicating significant variability compared to the other variables. It is noted that violin plots effectively display both the density of the data and summary statistics, offering a comprehensive view of data variability and distribution.

We used linear regression to analyze the relationship between radio nuclides indices, Fig. 10. The scatter plot displays data points with a positive linear trend line (red) and a confidence interval (shaded area). This indicates a positive linear correlation between Ra with D (Fig. 10(a)), H_in (Fig. 10(b)), H_ex (Fig. 10(c)), and ELCR (Fig. 10(d)). Overall, each plot illustrates a positive relationship between Ra and each dependent variable, with trend lines and confidence intervals highlighting potential predictive relationships. Additionally, from Fig. 11, we can observe that ²³²Th is positively correlated with D, H_in, H_ex, and ELCR. Moreover, from Fig. 12 we can observe that ⁴⁰K is positively correlated with D, H_in and ELCR. Furthermore, ⁴⁰K shows a positive trend with a wider confidence interval.

Figure 10.

Correlation of radium with a) D_γ, b) H_in, c) H_ex, and d) ELCR.

Figure 11.

Correlation of thorium with a) D_γ, b) H_in, c) H_ex, and d) ELCR.

Figure 12.

Correlation of potassium with a) D_γ, b) H_in, c) H_ex, and d) ELCR.

Conclusion

Activity concentrations of ²²⁶Ra, ²³⁸U, ²³²Th, and ⁴⁰K in the soil samples collected from Wolaita Zone, South Nations, Nationalities, and Peoples (SNNP) national regional state of Ethiopia) were measured by using a gamma ray spectrometer with a NaI (Tl) detector. The radiological effects were calculated accordingly by using mathematical expressions. The average activity values of ²³⁸U, ²³²Th, and ⁴⁰K were found to be 7.83 ± 1.9 Bq.kg⁻¹, 40.74 ± 2.7 Bq.kg⁻¹, 161.63 ± 3.9 Bq.kg⁻¹ respectively. The results showed that the average values of ²³⁸U and ⁴⁰K are lower, and ²³²Th is higher than the world's average value. The average value of radium equivalent activity is 192.25 Bq.kg⁻¹, which is below the recommended value of 370 Bq.kg⁻¹. Additionally, the average indoor radon concentration of 219.44 Bq.m⁻³ was above the recommended limit of 200 Bq.m⁻³. The average value of the absorbed gamma radiation dose rate is 35.68 nGy.h⁻¹, generating an annual effective dose of 0.76 mSv.y⁻¹. Which is below the ICRP recommended value for the general population. Moreover, the value of excess lifetime cancer risk is slightly higher. According to the multivariate statistical analysis, ²³²Th exhibited pronounced associations with multiple radiological parameters. Cluster analysis identified three distinct groups, and violin plots revealed variable distribution patterns, particularly for ²²⁶Ra and ²²²Rn. Linear regression confirmed positive trends between ²²⁶Ra, ²³²Th and ⁴⁰K and radiological parameters. The descriptive statistics showed positive skewness and varied kurtosis among radionuclides, supporting the development of effective radiation safety guidelines. Based on these results, it is concluded that the soil of the study area may not have a significant radiological hazard.

Conflict of interest

None declared.

Funding

None declared.

Declaration of competing interest

The authors declare no competing interests.

Data and code availability

Data will be made available on request. The source code will be available with a reasonable request to the second author.