| Radon Monitoring in Tourist Caves in South-West Western Australia
Bill Chandler Western Radiation
Introduction As a result of reports in ‘New Scientist’ and ‘The West Auatralian’ newspaper, the operators of the tourist caves in the Augusta/Margaret River and Yallingup/Busselton regions of Western Australia decided to commission a study of Radon gas and its decay products in their most popular tourist caves. The caves were first ‘screened’ for a 2-week period and follow-up monitoring was carried out in those caves that might be considered to present a possible hazard to the health of tourist guides or other cave workers. The draft guidelines for Radon exposure produced by the International Commission for Radiological Protection(ICRP)(Ref 1) were used as a basis for further monitoring. Being aware of Occupational Safety & Health legislation which requires that employers must exercise due care and diligence for the safety and health of their employees, the operators reported the results of the screening tests to the Western Australian Department of Health (Radiological Council) and requested advice on the health implications for their employees. The Radiological Council(Ref 2) recommended monitoring and action levels as recommended by the ICRP be adopted by the Operators of the tourist caves as an interim measure until the National Health & Medical Research Council’s Australian Ionising Radiation Committee (AIRAC) considers the ICRP recommendations. Western Radiation Services were commissioned by the cave operators to carry out the studies required to provide sufficient information for assessment of employees radiation exposure and to recommend a method for future monitoring that would be both accurate and cost effective. Monitoring Programme An initial screening programme using E-PERMTM radon monitors was carried out at the Jewel and Mammoth Caves operated by the Augusta/Margaret River Tourist Bureau (A/MRTB) and at the Yallingup Cave operated by the Busselton Tourist Bureau (BTB). The Moondyne Cave (A/MRTB) was re-opened during the testing period and included in the monitoring programme. Those caverns with average Radon concentrations greater than 400 Becquerels per cubic metre (Bq/m3) were monitored for monthly averages using E-PERMTM monitors and for Radon progeny using both grab sampling techniques and semi-continuous monitors. The grab sampling methods used were those of Kusnetz(Ref 3) and Tsivoglou (Ref 4) with the Rolle(Ref 5) method being used for semi-continuous monitoring. At the request of the Radiological Council(Ref 2) of Western Australia the monitoring was extended to include Thoron (Radon-220) and its progeny. Thoron was monitored using E-PERMTM ion chambers with a permeation time of approximately 30 minutes to measure Radon-222 only and a rapid response ion chamber to measure the combined Radon and Thoron. Thoron progeny was measured using the method of Rock. Fluctuations in gas and progeny concentrations were observed to occur diurnally and these were measured by operating the semi-continuous progeny monitor (Thomson & Nielsen Model TN-IR-21) for 24 hour periods in each of the caves of interest. The initial programme was run from October, 1992 to April, 1993 and will be modified to operate on a seasonal basis for on-going dose assessment purposes. Results The results of the screening tests are shown in
TABLE 1
Monthly monitoring and Radon progeny checks were carried out and the results are shown below in Table 2 & 3. Table 2
Note 1: K = Kusnetz, T = Tsivoglou (modified). Note 2:1 mWL = 1 milli Working Level = 20.8 x 10-9J/ m3 Note 3: F = calculated equilibrium factor. Table 3
Note: K = Kusnetz, T = Tsivoglou (modified). Further monitoring to determine the contribution of Radon-220 (Thoron) to the measured Radon concentrations was carried out at the request of the Radiological Council of W.A. The results are shown below in Table 4. Table 4
Note 1: K = Kusnetz; T = Tsivoglou (modified). Note 2: R = Radon-222; Th = Radon-220 (Thoron) Note 3: Equilibrium factor F is based on total radon concentrations. Table 5
Testing using Thomson & Nielsen Monitor and
Note: The 1500 measurement for Thoron was based on the filter used for the Rolle count. The relatively high Kusnetz and Tsivoglou readings for the same time and location may have been caused by the higher volume manual sampling pump operating in close proximity to the T&N instrument. Radon & Radon Progeny Concentrations
Testing using Thomson & Nielsen Monitor and Manual Sampling.
The data from Tables 5 and 6 are shown graphically in Figure 1. It is
noticeable that the data from the Moondyne Cave on the previous day closely
follows the variations measured at the Jewel Cave Organ Pipe between 1100
and 1530.
The 24 hour measurements taken in the Jewel and Yallingup caves are shown graphically in Figures 2 & 3. The variations with time can be clearly seen and amply demonstrate that spot measurements are not sufficiently accurate for hazard assessment. It is also noticeable that each of the caves has an individual pattern of variation in the radon progeny concentrations. Since the variations occur during normal operating hours of the caves, it is suggested that an overall mean of the radon progeny concentrations be related to the long-term mean of the radon concentrations to provide an average equilibrium factor that can be used for personnel exposure calculations. Discussion The screening tests showed that the Jewel, Moondyne and Yallingup caves were within the range of radon gas concentrations that require further monitoring based on the draft recommendations of the ICRP. These recommendations were aimed at workplaces and dwellings and suggested that an equilibrium factor (F) of 0.4 would be acceptable for such locations. Information on cave atmospheres from the USA suggests that F may be greater than 0.4 during periods of very low air exchange rates. Data from Jewel and Yallingup caves showed that the F varied with time of measurement, so 24-hour monitoring for radon progeny was carried out in both caves to determine the variation pattern and to obtain an average F that could be used for personnel exposure calculations. Table 7
Table 8
Using average radon concentrations of 1100 Bq/m3 for the Jewel Cave and 1130 Bq/m3 for Yallingup we obtain an F facor of 0.32 and 0.37 respectively. This factor appears to be reasonable when compared to earlier spot measurements. The graph of the data for the Yallingup Cave (Fig.3) shows a clear day/night breathing pattern for the cave, with the radon progeny concentrations reaching a peak at 11:00 pm and gradually reducing to a low plateau at about 2.30 pm the following day. The pattern for the Jewel Cave (Fig.2) shows two distinct peaks at 5:00 am and 5:00 pm with the low level plateau being of much shorter duration. The major difference between the caves is that Jewel is apparently linked to at least two other caverns (Easter and Moondyne caves) whereas Yallingup appears to be a single cave system. Short term monitoring at the Jewel and Moondyne caves (Fig.1) seems to show a similar pattern for radon progeny concentrations in the Camel section of Jewel and the Moondyne cavern with a time offset, but this will need to be confirmed by further 24 hour measurements at both locations. If (as believed) the caves are connected then air movement would occur between the caverns as well as to and from the surface. This would give rise to multiple peaks and troughs as seen on Fig.2. It can be seen that although many caves are similar in appearance and appear to be similar in atmosphere, the air movement in the caves can be extremely complex, necessitating that monitoring for radon gas and progeny be individually carried out for each cavern. This is true even when a cavern is part of a multiple cave system. The health hazard associated with exposure to radon gas has been determined by the ICRP and the units used here (mWL) can be related to breathing rate of personnel and time of exposure to arrive at a Committed Effective Dose Equivalent (CEDE) per annum. The current limits for personnel not classified as radiation workers are set at 1 mSv/a for 50 years exposure. Where personnel are not full-time long-term employees, the lifetime exposure of 50 mSv can be apportioned at a maximum of 5mSv/a for up to 10 years. Assuming a working year of 2000 hours and continuous exposure during working hours a radon progeny concentration of 8 mWL would reach the limit of 1mSv/a and 40 mWL would reach 5 mSv/a. The actual figure for maximum exposure would be lower than 40 mWL since there is also a gamma radiation exposure associated with the decay of radon that will need to be added to the radon progeny CEDE to obtain total exposure. In order to comply with Occupational Health and Ionising Radiation exposure regulations it will be necessary for some cave operators to take action to minimise the exposure of personnel. Since it is not a viable option (in most cases) to ventilate efficiently the caves the most effective method of exposure control is to limit the time spent in the caves by employees. Record keeping of employee working hours becomes very important when it is necessary to monitor their exposure levels, with careful recording of actual times spent in the various locations. Once the parameters for a particular cave have been determined for daily, seasonal and annual concentrations of radon gas and the equilibrium factors for radon progeny, it should be acceptable for operators to monitor only radon gas on a seasonal average basis and apply the equilibrium factor to obtain an exposure assessment for their personnel. This would be the most economic method of ongoing monitoring in those caves where it is required. The ICRP recommend that monitoring should be carried out where radon levels exceed 400 Bq/m3 and action should be taken where the levels exceed 1000 Bq/m3. References 1. Protection Against Radon in Buildings and Workplaces:
Report by an ICRP Task Group: August 1992.
Acknowledgements The authors wish to acknowledge the assistance and co-operation of all the personnel at the Augusta/Margaret River and Busselton Tourist Bureaus and to thank Mr Keith Tritton and Mr Barry Brown for permission to publish this paper. |