Investigating Radon Concentrations in the Cango Cave, South Africa

Abstract

Radon concentrations in the tourist part of the Cango cave were measured using 25 strategically placed electret ion chambers. Airflow rates were also measured and found to be less than 1 m/s throughout the cave. An IDW interpolated radon concentration overlay was constructed using QGIS and overlayed on maps of the cave. The maximum radon concentration of 2625 Bq/m³ was measured in the Grand Hall, located in the central part of the cave following a narrow passage. The initial part of the cave near the entrance exhibited normal cave breathing characteristics, with radon concentrations of less than 300 Bq/m³. The deepest section of the cave, however, demonstrated an unexpected decrease in radon levels, temperature, and humidity. The average radon concentration in the Cango cave, measured at 1265 Bq/m³, is relatively low compared to other caves worldwide that need mitigation measures according to the International Commission on Radiological Protection (ICRP).

1. Introduction

The Cango cave is located approximately 29 km from the town of Oudtshoorn in the Western Cape province of South Africa. This karstic cave comprises an extensive system of caverns and tunnels, filled with stalactites, stalagmites, and helictites. These features were carved out of Precambrian limestone over the last 450 Mya. The primary rock type in the Cango cave is limestone (CaCO₃) from the Cango cave group with minor dolomite (CaMg(CO₃)₂), aluminum oxide (Al₂O₃), and quartz (SiO₂) deposits [1]. Most of the unique properties of the rock formations in the cave are the result of calcium carbonate deposits left by slow alluvial redeposition. Figure 1 shows the typical geology found in the cave.

The Cango cave contains evidence suggesting that it has been known to humans for thousands of years. Early inhabitants include the Khoikhoi and San people, who likely used the cave for shelter. However, the cave gained widespread attention in the late 18th century when a local farmer, Jacobus Van Zyl, rediscovered it. Current conservation efforts are crucial to preserving the delicate environment of the Cango cave. Measures are in place to reduce the impact of tourism, such as limiting the number of visitors and controlling the temperature and humidity within the cave. The Cango cave, like several other caves, has a restricted entrance which hampers ventilation and thus impacts the temperature, humidity, and overall air quality inside the cave.

The cave is divided into three sections [2]. Cango I, the first 750 m from the cave entrance to the Devil’s Workshop, has been open to tourists since 1897. The adjacent section, Cango II, is the 300 m long intermediate part of the cave and has never been open to the public. Although Cango II was only discovered in 1972, it is estimated that the airflow between Cango I and II has always been unimpeded. Adjoining Cango II is Cango III, the 1500 m terminal section of the cave. It was explored in 1975 by lowering the water level in the sump joining these two sections, which effectively seals this section from the first two parts of the cave.

Radon, specifically the ²²²Rn nuclide, is a radioactive gas present in the progeny of naturally occurring uranium. Despite its short half-life of 3.8 days, Radon is one of the most significant contributors to lung cancer after smoking. [3]. Karstic geology typically has low levels of uranium, although concentrations may increase due to alluvial redeposition because of uranium’s water solubility [4]. Enclosed cavities in this geology, such as karstic caves, are therefore susceptible to radon accumulation due to inadequate ventilation. Consequently, it is important to assess radon levels in confined areas occupied by humans. Appropriate measures can be taken to decrease radon exposure, thereby reducing the associated risks.

Radon measurements have been conducted in the Cango cave, but a comprehensive survey, including a radon concentration map, is lacking. Additionally, these measurements were carried out during the warmer parts of 2004 and 2005, and the seasonal variation of radon concentrations could affect the measurements. Based on the results of these measurements, a radon protection program for tour guides was subsequently implemented at the cave.

This study’s objective is to measure and map the radon concentrations in Cango I, the popular tourist section of the Cango cave, which is continuously visited by tourists and their accompanying tour guides. The survey was conducted during the temperate autumn season. This study also attempts to examine the cave’s ventilation and provide information on seasonality, ventilation, and radon distribution within the cave.

2. Materials and Methods

2.1. Measurements

The radon concentrations in the cave were measured using electret ion chambers (EICs), specifically E-PERM^® S-chambers with short-term (ST) electrets from Rad-Elec Inc. The electrets were deployed in an SST configuration and were systematically spaced from the entrance up to the Devil’s Workshop, which marks the farthest point in the tourist area of the cave. The EICs were placed in significant sections of the cave, with more than one EIC placed in the larger caverns. A total of twenty-five EICs were carefully placed at the locations indicated on the cave map in Figure 2. The EICs were positioned a minimum distance of 1 m from any wall or impediment and directly on top of the geological formations. The radon measurements were conducted over a two-day period by first leaving the EICs for 24 h in the deeper part of the cave. The larger caverns and tunnels close to the entrance area were then measured during a second 24 h period. An Alba-Windwatch^® airflow meter from SILVA (www.silva.se) was used to determine airflow patterns in various parts of the cave. The general cave ventilation was estimated using these airflow patterns, which provided valuable information. The airflow was measured at different times of the day and at locations where the cave narrows. No significant airflow was detected, with all the airflow measured at less than 1 m/s.

2.2. Cave Layout and Structure

One of the largest sections of the Cango cave is the Van Zyl Hall. This cavern is approximately 98 m long, 49 m wide, and 15 m high, and includes a stalagmite formation that reaches 9 m in height. The Van Zyl Hall, Throne Room, Bridal Chamber, Grand Hall, and Lot’s Chamber are all large caverns in the initial part of the tourist cave. However, some of these caverns are linked by smaller tunnels. Following these caverns is a long, narrow passage called the Lumbago Tunnel, which provides access to a cavern known as King Solomon’s Mines. The remaining caverns in the tourist section are reachable through a narrow passageway, dubbed the Tunnel of Love, that leads from King Solomon’s Mines. This passageway is accessed via a staircase, as shown in Figure 3. The cavities beyond the staircase are all comparatively small compared to the chambers leading up to this section.

The entrance to the Cango cave helps to mitigate radon buildup by allowing for ventilation from the surface, but it is relatively small. Even though the Van Zyl Hall is situated near this entrance, four EICs were placed in this region due to its size, the high number of tourists, and the potential health risks associated with this area. Additionally, four EICs were placed at various locations between the Van Zyl Hall and the entrance area, where tourists typically spend additional time wandering.

2.3. Radon Calculation

The initial potential (V_i) of each electret was measured using a surface potential electret voltage reader (SPER), also from Rad-Elec Inc., before placing the EICs. The SPER was calibrated before and after the survey by using three reference electrets. These include a blank and two stable high-voltage electrets. The EICs were left in this location for at least 24 h, whereafter the final potential (V_f) was measured. The electret potentials may be at risk of depletion if they are left for an extended period due to the likelihood of unusually high radon concentrations in specific areas of the cave. Consequently, 24 h measuring periods were chosen. The difference in potential was then used to calculate the airborne radon concentration (RnC) using Equation (1) [5].

$$RnC={\displaystyle \frac{(V_i-V_f)}{\left(T\right)\left(CF\right)}}-BG$$

(1)

where CF is a linearization coefficient given by

$$CF=0.314473+0.260619\times \ln \left({\displaystyle \frac{V_i+V_f}{2}}\right)$$

(2)

with T the measuring time in days and BG the radon equivalent of natural gamma ray background. The background radiation correction was estimated at 32.2 Bq/m³. This value was proposed by Shahbazi-Gahrouei et al. [6] and was used in previous radon measurements in the Cango cave [7]. Additionally, Bezuidenhout [8] used the geology of the Cango cave to predict the dose rate of the area to be between 50 and 100 nGy/h, which also relates to a radon equivalent background of 32 Bq/m³ [9].

Three systematic errors were quadratically combined to estimate the total error of the EICs [5]. The first error is associated with the electret thickness and the general chamber parameters, including the volume. The second error considers inaccuracies in reading the EICs initial and final potentials, whereas the third error estimates the uncertainty of the natural gamma radiation background. The average total systematic error of the measurements was found to be 6.2%.

2.4. Radon Map

The cave map was constructed using polygons in QGIS. The locations where the radon concentrations were measured were then added to the map, and their corresponding concentrations were interpolated using the Inverse Distance Weighting (IDW) function of QGIS. Adelikhah et al. [10] found the IDW method more suitable for predicting mean indoor radon concentrations compared to other interpolation techniques due to the lower mean absolute error (MAE) and root mean square error (RMSE). A graduated red color ramp based on continuous intervals was selected for the interpolated radon layer and overlayed on the map of the cave. A similar process was followed to produce a side view of the radon concentrations in the cave.

3. Results and Discussion

The calculated radon concentrations for the measurement locations in the cave are listed in

Table 1. A list comparing the average radon concentrations measured in the four zones.

Location	Radon Concentrations (Bq/m3)
Old Exit Passage	446 ± 36
Entrance	257 ± 31
Entrance Hall	214 ± 30
Fern Garden	494 ± 38
Van Zyl’s Hall (Entrance)	494 ± 38
Van Zyl’s Hall (Side tunnel)	549 ± 40
Van Zyl’s Hall (Organ)	631 ± 44
Van Zyl’s Hall (Exit)	470 ± 39
Throne Room	680 ± 46
The Vestry	756 ± 48
Bridal Chamber	1438 ± 79
Bridal Bypass	589 ± 43
Rainbow Room	759 ± 49
Fairy Palace	2464 ± 128
Grand Hall	2625 ± 135
Sand Chamber	2472 ± 128
Lot’s Chamber	2455 ± 127
Smyth’s Ladder	2201 ± 114
Lumbago Entrance	1748 ± 92
Lumbago Tunnel	1683 ± 90
Crystal Palace	1472 ± 80
King Solomon’s mines	1326 ± 72
Prince Albert Chamber	1132 ± 66
Post Office	951 ± 58
Devil’s Workshop	1638 ± 88

, with Figure 4 and Figure 5 showing maps of the cave with radon concentration overlays.

The radon concentrations listed in Table 1 were then plotted as a function of distance, from the entrance to the deepest part at the Devil’s Workshop. The resulting graph is shown in Figure 6. It is evident that the concentrations from the entrance to the Bridal Chamber stay relatively constant. This is followed by an abrupt increase from the Bridal Chamber through Fairy Palace and into the Grand Hall. Fairy Palace is a narrow part of the cave that links the Bridal Chamber to the lower-laying Grand Hall. The highest radon concentrations are found here, and in the caverns leading up to Lot’s Chamber, whereafter it systematically decreases towards the Devil’s Workshop. A similar pattern was observed by Nemangwele [7]. This trend is further illustrated in the interpolated radon overlay of Figure 4 and Figure 5.

The Cango cave only has one known opening, resulting in poor ventilation in the deeper parts of the cave. It was further found that some of the narrower halls, chambers, and tunnels enhanced the accumulation of radon, thus leading to elevated levels in these areas. Temperature and pressure changes cause typical cave breathing through the cave entrance [11], which reduces radon concentrations in the initial part of the cave. The narrow sections of the cave, however, would inhibit the flow of atmospheric air resulting in elevated radon concentrations. This is clearly illustrated in Figure 6 with the initial low radon concentrations substantially increasing after the narrow Fairy Palace passage. The narrowing of this passage is clearly visible in the top and side views of the cave (Figure 4 and Figure 5).

Contrary to the findings in the Cango cave, the Sudwala cave in Mpumalanga, South Africa, which formed in similar karstic geology, showed the expected radon buildup in its deeper caverns [12]. This was attributed to cave breathing diffusing radon gas into these caverns. In the Cango caves, however, Figure 4, Figure 5 and Figure 6 show a decrease in radon gas in the caverns after the Grand Hall and Lot’s Chamber. This may be due to a change in the geological formation or a change in cave surface area relating to radon exhalation. A study by Cigna [13] measured humidity and temperature in the Cango cave and found a decreasing trend in these two parameters. As shown in Figure 7, this trend correlates with the radon concentrations and may be attributed to the large number of tourists who follow the Heritage tour, ending at Fairy Palace, compared to the smaller number who take the deeper Adventure tour from Fairy Palace to the Devil’s Workshop.

The caverns in the deeper sections of the cave, past the Lumbago Tunnel, are significantly smaller than those in the initial part of the cave. This is evident in the cave map in Figure 4 and illustrated by the photos in Figure 8. The reduction in cave volume in the latter part of the Cango cave should consequently increase radon concentrations, humidity, and temperature in these areas. The decrease in these elements, however, suggests additional cave openings to the surface in the deeper sections of the cave that assist with ventilation. The absence of airflow through the confined Lumbago Tunnel also suggests that there may be ventilation openings in the cave’s deeper region. Further investigation of this, however, is needed.

More than 70% of caves worldwide demonstrate average radon concentrations above 1000 Bq/m³ [14]. In 1994, the ICRP implemented a recommended action level of 1000 Bq/m³ to reduce radon exposure in caves [15]. The average radon concentrations in the Cango cave were found to be 1130 Bq/m³ and 1265 Bq/m³ for the Heritage and Adventure tours, respectively.

Table 2. A list of caves with average radon concentrations above 1000 Bq/m³ from different parts of the world.

Cave	Mean Radon Concentrations (Bq/m3)	Reference
Jenolan Cave, Australia	4578	[14]
Shawan Cave, China	47,419	[16]
Moravian Karst caves, Czech Republic	1235	[17]
Various caves, Great Britain	2907	[18]
Various caves, Great Britain	35,890	[19]
Various caves, Great Britain	9306	[20]
Petralona Cave, Greece	25,179	[21]
Various caves, Hungary	3300	[22]
Various caves, Hungary	2468	[23]
Various caves, Ireland	4127	[24]
Various caves, Poland	1166	[25]
Various caves, Russia	2390	[26]
Postojna Cave, Slovenia	1412	[27]
Postojna Cave, Slovenia	25,020	[28]
Altamira Cave, Spain	3564	[29]
Altamira Cave, Spain	3286	[30]
Rull Cave, Spain	1762	[31]
Various caves, Switzerland	25,000	[32]
Various caves, United States	1927	[33]
Mammoth Cave, United States	2589	[34]
Various caves, United States	1475	[35]
Big Bone Cave, United States	11,678	[36]
Carlsbad, Cavern USA	1821	[37]
Hollow Ridge, USA	4733	[38]
Global average (All caves)	6160	[14]
Cango Cave, South Africa	1265	This study

provides a list of other caves with average radon concentrations exceeding 1000 Bq/m³. It is evident that the radon concentrations measured in the Cango cave are substantially below the global average, and only marginally above the recommended action level of the ICRP. Interestingly, there is no substantial difference between the radon risk of the Heritage and Adventure tours in the Cango cave. The extended duration of the Adventure tour, however, increases the exposure to elevated radon levels, thereby increasing its risk. A potential method of mitigating radon accumulation in the cave is to investigate the prospect of an additional ventilation opening that could assist in improving and regulating the ventilation within the cave.

4. Conclusions

Radon concentrations in the Cango cave were measured using electret ion chambers (EICs). A total of 25 EICs were strategically placed throughout the tourist part of the cave. The measurements were conducted for 24 h periods, and no significant airflow was detected within the cave. Systematic errors were combined to estimate a final average error of 6.2% for the radon measurements. The results demonstrated that the Cango cave, which has only one known opening, has poor ventilation, especially in the central part of the cave. Compared to other caves with radon concentrations exceeding 1000 Bq/m³, the Cango cave ranks among the lower ranges. The radon concentrations measured in the Cango cave also compare reasonably well with other caves with similar lithology.

The variations and rapid increase in radon concentrations in the central part of the cave were influenced by insufficient ventilation due to narrowing cavities. Cave breathing in the initial part of the cave diffuses radon gas between the different sections of the cave. These results are generally comparable to findings in a similar karstic cave in South Africa, where the distance from the entrance resulted in increased radon gas concentrations [12].

The decrease in radon concentrations in the deepest part of Cango I may be attributed to possible additional openings to the surface. Further investigation into ventilation openings should therefore be considered. The similarity between measurements conducted by Nemangwele [7] during the summers of 2004 and 2005 and those conducted in autumn suggests insignificant seasonal differences in radon accumulation. Although Nemangwele employed a similar methodology, the seasonal and temporal variability of radon necessitates additional measurements across all four seasons to accurately assess the average annual radon concentrations in the Cango cave.