Environmental Geochemistry and Health, 1992, ]4(4), page 113-120
Seasonal and spatial variations in Rn-222 and Rn-220 in soil gas, and implications for indoor radon levels Glenn Sharman Geology Department, University of Leicester, LE1 7RH, UK Abstract Rn-222 enters dwellings as a component of soil gas drawn from the soil by mass flow driven by the pressure difference between the house and soil beneath. In a site on Northampton Sand Ironstone (Aalenian), a preferred path of emanation (hotspot) was found. A difference of 63 Bq L-1 Rn-222 was recorded in July between this point and another 3 m away. Rn-222 in this hotspot shows 12% less variation annually than the surrounding rock. During winter, Rn-222 values within 1.6 m of the house were 44% lower than those at more than 4 m away. Rn-222 showed a 99.5% negative correlation with wind run, showing that on this soil wind pressure can significantly reduce radon in the soil at 500 mm depth. Rn-220 in soil gas correlated positively at the 99.5~ level with grass and air temperatures. Rn-220 was not associated with the hotspot.
Introduction The Northampton Sand Formation is a Jurassic sedimentary rock of Aalenian age, comprising ironstone in its lower part, and ferruginous and calcareous sandstones in its upper beds. It has been found to be the principal source of elevated radon in dwellings in Northamptonshim, with the main source considered to be a layer of uraniferous nodules near its base. Thorium is concentrated in ironstones relative to average sedimentary rocks (Sutherland, 1991). Twenty-seven percent of dwellings on Northampton Sand are above the Government's Action Level of 200 Bq m-3 (NRPB, 1990). Thirty-one percent of soil-gas readings on the same rock have more than 20 Bq L-1 Rn-222 (20,000 Bq m-3) (Sharman, 1991). Study of radon in soil gas in Northamptonshire by the author has included widespread and intensive work on this rock type, and one aspect of the study is reported here since it may have wider implications for other rocks of similar physical and chemical characteristics. The garden on three sides of a house was sampled on a closely-spaced grid for a period of 1 year. The observed high values of Rn in its soil gas means that counting errors are reduced and enable a study of the spatial relationship of radon in soil gas to radon in the building to be carried out over an extended period. These data also allow observation of spatial variation in Rn emanation from a relatively uniform rock. Seasonal and meteorological variations were also investigated. Previous Work As the potential hazard of high radon levels in dwellings has become known, so there have been attempts to find correlations between indoor Rn and Rn, Ra and U in subjacent soils. Akerblom et al. (1983) noted that Rn in soil
gas under houses is generally higher than that in ground adjacent to buildings, and suggested that buildings reduce both diffusion of Rn from the soil surface, and air exchange in the soil layer under houses. Eaton and Scott (1983) reported that Rn concentration in buildings depended on four factors; pressure differential between soil and house, resistance of the soil, Rn release rate per volume of soil around the foundations, and the average ventilation period of the house. Observation has shown that soils of the Northampton Sand Formation are free-draining and therefore permeable, due to fracturing and fissuring and high porosity. This is especially so in its lower part, the main ironstone group, which underlie the study area. Wilkening (1985) supported the work of Smith, Barretto and Pournis (1976) when he showed that the average distance an atom can move through dry soil by diffusion is 1.6 m (Rn-222) and 20 mm (Rn-220). As buildings tend to dry out the soils beneath, and to some extent around the foundations, radon in buildings is likely to be drawn from the ground within this distance of the foundations. Therefore, by sampling both within 1.6 metres of the building and at several metres away, a variation close to the building might be observed. Fracturing of the rock could lead to preferential emanation along paths of least resistance, which would lead to small-scale spatial variations. On this subject, Taipale and Winqvist (1985) concluded that Rn concentrations vary greatly in a small area, to the extent that the choice of house site to within a few metres can make a substantial difference. Their argument is supported by this study. Schery and Siegel (1986) indicated that channels through soil are important for Rn transport at times of high pressure gradients, but that channels contribute only moderately to overall exhalation. Meteorological variations have significant effects
114
Variations in Rn-222 and Rn-220 in soil gas
4
3
2
1
8
7
B
.5
I0
9
procedures are described in Ball el al. 1991. It is possible to differentiate between Rn-222 and Rn-220 due to the different half-lives of these isotopes. (-Rn-2200 tl/2 = 55 s, will decay much faster than Rn-222, tlj2 = 3.8 days). Radon activity is expressed as Becquerels per litre (Bq L-I). The instrument was calibrated using a calibration standard cell, which has a geometry similar to the Lucas cell used for counting. Cells were calibrated by the National Radiological Protection Board, and quality conliot was achieved by periodic cross-calibration with the British Geological Survey. Methodology
12
11
14
13
15
16
SCALE
17
3 METRES
18
N #,
Figure 1 Plan of the study site showing sample positions.
/1\
|
upon radon concentrations in soil gas; indeed Ball et at. (1991) stated that Rn concentrations in soil gas is controlled as much by meteorological factors as by chemical and mineralogical factors. In 1964, Kraner el aL showed that as pressure decreases, air rich in Rn-222 is drawn from the ground and as pressure rises, air low in Rn is forced into the ground. Miller (1966) noted high radon in soil gas during and immediately after heavy rain. Miller and Ball (1969) showed that where topsoil is thin, absent or pervious, barometric pressure becomes the controlling factor in radon escape from soils, but plays only a marginal role where there is a well developed humic topsoil, in which case soil moisture content affected by relative humidity is a major factor. Small intra-granular pore spaces are easily closed by moisture, so decreasing the emanation of radon from the ground. Miller and Ostle (1973) went on to shown that where there is a thin or permeable topsoil, low pressure and strong wind reduce radon in the near-surface environment, drawing up radon from depth, whilst radon builds up in the ground during calm, anticyclonic conditions. Clements and Wilkening (1974) went further and suggested that changes in pressure of 1-2% can give rise to changes in Rn-222 flux of 20-60%. Ball et al. (1983) found Rn concentrations correlated strongly with soil temperature, and less strongly with air temperature. Instrumentation
An emanometer was used to determine the radon concentration in soil gas samples. Principles and
The garden site measured 18 m • 14 m, and a square grid was sampled at 3-metre intervals, to provide 18 sample locations (Figure 1). These were sampled at approximately monthly intervals. The bedrock is an oolitic, sideritic, limonitic ironstone (Hollingworth and Taylor, 1951), and produces a ferritic brown earth soil (Reeve, 1978). Soil on the site is brown and crumbly, welt developed, with ironstone fragments, and is flee-draining. The results were compared with meteorological records from the weather station at Nene College, Northampton, situated 18 km to the south-west of the site, and with monthly averages for the agro-ctimatic area 22 East; (Smith, !984). Neither of these sources would necessarily be able to identify localised showers in the study area, although each has its use; the former provides a record of changes associated with cyclonic conditions, whilst the averages show long-term variations. Since gas sampling was carried-out at the beginning of each month, weather values were averaged from mid-month to mid-month. Mean summer (May to October) and mean winter (November to April) values were based upon these monthly averages, the criteria used to separate summer from winter being temperatures above or below 10~ Analysis has involved dividing the data set to examine the two isotopes individually, and by distance from the house (since dwellings are known to alter emanation patterns). Spatial and seasonal variations were then examined. This can be summarised as follows: Rn-222 spatial variation. (a) Spatial results at >4 m from the house. (b) Spatial results at 1.6 m from the house. Rn-222 seasonal variation. (a) Seasonal results at >4 m from the house. (b) Seasonal results at 1.6 m from the house. Rn-220 spatial variation. Rn-220 seasonal variation. Results
Rn-222 spatial variation Figure 2(a) shows the annual average pattern of soil gas values. Maximum radon activity is found in the north and west of the site. Variation between adjacent positions can be great, the largest being 63 Bq L-1 difference between two adjacent readings in July. The coefficient of variation (standard deviation as a percentage of the mean) indicates
G. Sharman
115
Figure 2 Spatial variation of radon. (a) Rn-222; (b) Rn-220. the variation, ranging from 39% to 106%. The main problem with this plan is that it cannot reflect true spatial variations due to the presence of the house which is seen to alter the emanation pattern through the year. For this reason the data set was divided to illustrate spatial variation at (a) >4 m from the house (~ssumed to be outside of the influence of the house) and (b) 1.6 m (within the influence of the house).
(a) Spatial variation at >4 m from the house: This sub-set shows spatial variation within a small area which is typical of the variation found on Northampton Sand Ironstone (NSI) on undisturbed ground. The annual mean value is 20
Bq L-1 which agrees with NSi results regionally. There is a consistent spatial relationship, with relatively high spots being found in the same place in successive months. This is indicative of preferred path~ of emanation. The mean radon in the hotspots is 29 Bq L -1, with a mean coefficient of variation of only 32% annually. The mean value for the remainder of the results at >4 m is 14Bq L -1, with a mean coefficient of variation of 44%. This shows that seasonal variations appear to have a lesser effect upon preferred pathways o f emanation than in areas of probably unfractured rock. Other reasons for variation in Rn concentration could include the concentration of mineralogical hosts for U and
116
Variations in Rn-222 and Rno220 in soil gas movement of radon into the building as a component of mass flow of soil gas.
D
Figure 3 Seasonal variation of radon ranked by month. (a) Rn-222; (b) Rn-220. Th in the ground, radium-226 (h/2 = 1,602 yr) and radium-228 (tl/2 = 5.75 yr), and groundwaters which may move decay products of U in the soil/rock, as well as Rn-222 which could migrate 60-70 m (Smith et al., 1976). Radium mobility may be reduced in oxidising environments, where Fe and Mn oxides, clay minerals and organic matter scavenge Ra during precipitation, and in calcareous soils (Levinson et at., 1982). (b) Spatial variation at 1.6 m from the house: Values at 1.6 m from the house are significantly lower than those at >4 m away. The mean annual value is 13 Bq L -1. The hotspot is apparently unaffected by the presence of the building and the remainder of the results at 1.6 m have an annual mean of 9 Bq L-1, with a coefficient of variation of 58%. This is a considerably larger variation than the 44% at >4 m away, and must be partly attributable to the
Rn-222 Seasonal variation Figure 3(a) shows the results for the t8 sample positions ranked by month. Values rose sharply from April (mean of 10 Bq L -1) to peak in May/June ~.25/26 Bq L7 l) then maintained relatively constant v~ues (20-22 Bq L -1) until September. It also illustrates the wide range of Rn-222 values during the summer months (from 4 metres from the house. Radon emanation is therefore not as liable to alteration by se,~onal variations at these points as it is in the surrounding rock, During the period November to April Rn-222 values within 1.6 m of the building were 44% lower than those at 4 m from the house as opposed to the 25% drop in the summer. Between summer (May to October) and winter, values near the house fell by 50%, compared with a 33% drop in values at 4 m away. This leads to the conclusion that the house caused a measurable reduction in radon in the subjacent soil. The results at more than 4 metres from the house are typical of Northampton Sand Ironstone results which leads to the conclusion that the house does not affect soil gas radon at this distance. Variation in Rn-222 emanation was found to correlate negatively at the 99.5% significance level with wind run and positively at the same level with potential transpiration. Wind movement over the soil surface is therefore sufficient
120
Variations in Rn-222 and Rn-220 in soil gas
to reduce radon in the soil to at least 500 mm depth in the Northampton Sand Ironstone. There was a small range in the Rn-220 results which were not associated with preferred emanation paths. Rn-220 correlated positively at the 99.5% level with air and grass temperatures and hours of insolation. These are interpreted as factors controlling radon emanation. Seasonal adjustment to soil gas values can be made which would allow annually corrected values to be obtained.
Acknowledgements The author is grateful to Dr Diana Suthcrland for her encouragement and enthusiasm, and to whom the project owes so much. Also Dr Keith Ball of the BGS for willingly giving of his time, vast experience and resources, and Professor John Hudson for his advice and constructive criticism. Also appreciation must be given to Iris Sharman who allowed her garden to be used in this study, and to Christina Rout of Nene College for time spent consulting meteorological records. This work is dependent upon the financial support of the combined County, District and Borough Councils of Northamptonshire, and a grant from the Research Board of the University of Leicester to Professor J.D. Hudson. Thanks to all concerned.
References Akerblom, G., Andersson, P. and Clavensjo, B. 1983. Soil gas radon a source for indoor radon daughters. Rad. Prot. Dosim., 7, 49. Ball, T.K., Cameron, D.G., Colman, T.B. and Roberts, P.D. 1991. Behaviour of radon in the geological environment: a review. Quart. J. Engineering Geology, 24, 169-182. Ball, T.K., Nicholson, R.A. and Peachey, D. 1983. Effects of meteorological variables on certain soil gases used to detect buried ore deposits. Trans. last. Min. Metall., 92, B183. Clements, W.E. and Wilkening, M.H. 1974. Atmospheric pressure effects on 222Rn transport across the earth-air interface. J. Geophys. Res., 79, 5025. Eaton, R.S. and Scou, A.G. 1983. Understanding radon transport into houses. Rad. Prot. Dosim., 7, 251. Hollingwonh, S.E. and Taylor, J.H. 1951. The Northampton Sand Ironstone. Stratigraphy, structure and reserves. Mere. Geological Survey Great Britain. HMSO, London. Kraner, H.W., Schroeder, G.L. and Evans, R.D. 1964. Measurements
of the effects of atmospheric variables on radon-222 flux and soil gas concentrations. In: The Natural Radiation Environment, pp.191-215. Chicago. Levinson, A.A., Bland, C.J. and Lively, R.S. 1982. Exploration for U o r e deposits. In: Ivanovich, M. and Harmon, R.S. (eds), Uranium Series Disequilibrium: Applications to Environmental Problems, pp.351-381. Clarendon Press, Oxford. Miller, J.M. 1966. Interim Report on the Measurement af Radon in Soil Ab" as a Prospecting Technique. Report No.272. Institute of Geological Sciences, Metalliferous Minerals and Applied Geochemistry Unit. Miller, LM. and Ball, T,I~ 1969. Second Progress Report on the Measurement of Radon in Soil Air as a Prospecting Technique. Report No.286. Institute of Geological Sciences, Metalliferous Minerals and Applied Geochemistry Unit. Miller, J.M. and Ostle, D. 1973. Radon measurements in uranium prospecting. In: Uranium Exploration Methods (Conference Volume), pp.237-247. International Atomic Energy Agency, Vienna. Morse, R.H. 1976. Radon counters in uranium exploration. In: Uranium Exploration Methods (Conference Volume), pp.229-239, International Atomic Energy Agency, Vienna. Nason, R. and Cohen, B.L. 1987, Correlation between z ~ a l in soil, 222Rn m 9 sod. gas and 222Rn reside , . adjacent . houses. Health Phys., 52, 73-77. NRPB (National Radiation Protection Board). 1990. Human exposure to radon in homes. In: Documents of the NRPB~ VoI.1, No.1. HMSO, London. Reeve, M.J. 1978. Soils in Northamptonshire 1. Sheet SP66 (Long Buckby). Soil Survey Record No.54. Rothamsted. Schery, S.D. and Siegel, D. 1986. The role of channels in the transport of radon from the soil. J. Geophys. Res., 91(t2), 366. Sharman, G. 1991, Radlometric investigation of radon in soil gas over Jurassic rocks of Northamptonshire, England. Environ. Geochem. Health, 13(3), 146-147. Smith, A.Y., Barretto, P.M.C. and Poumis, S. 1976. In: Exploration for Uranium Ore Deposits, p.185. I.A.E.A. Publication STI/PUB/434. Smith, L.P. 1984. The Agricultural Climate of England and Wales (1941-70). Reference book No.435. Ministry of Agriculture, Fisheries and Food, HMSO, London. Sutherland, D.S. 1991. Radon in Northamptonshire, England: geochemical investigation of some Jurassic sedimentary rocks. Envh'on. Geochem. Health, 13(3), 143-I45. Taipale, T.T. and Winqvist, IZL 1985. Seasonal variations in soil gas radon concentration. Sci. Total Environ., 45, 121. Wilkening, Mo 1985. Radon transport in soil and its relation to indoor activity. Sci. Total Envh'on., 45, 219.
(Manuscript No.267: received February 24, 1992 and accepted after revision April 14, 1992.)
Seasonal and spatial variations in Rn-222 and Rn-220 in soil gas, and implications for indoor radon levels Glenn Sharman Geology Department, University of Leicester, LE1 7RH, UK Abstract Rn-222 enters dwellings as a component of soil gas drawn from the soil by mass flow driven by the pressure difference between the house and soil beneath. In a site on Northampton Sand Ironstone (Aalenian), a preferred path of emanation (hotspot) was found. A difference of 63 Bq L-1 Rn-222 was recorded in July between this point and another 3 m away. Rn-222 in this hotspot shows 12% less variation annually than the surrounding rock. During winter, Rn-222 values within 1.6 m of the house were 44% lower than those at more than 4 m away. Rn-222 showed a 99.5% negative correlation with wind run, showing that on this soil wind pressure can significantly reduce radon in the soil at 500 mm depth. Rn-220 in soil gas correlated positively at the 99.5~ level with grass and air temperatures. Rn-220 was not associated with the hotspot.
Introduction The Northampton Sand Formation is a Jurassic sedimentary rock of Aalenian age, comprising ironstone in its lower part, and ferruginous and calcareous sandstones in its upper beds. It has been found to be the principal source of elevated radon in dwellings in Northamptonshim, with the main source considered to be a layer of uraniferous nodules near its base. Thorium is concentrated in ironstones relative to average sedimentary rocks (Sutherland, 1991). Twenty-seven percent of dwellings on Northampton Sand are above the Government's Action Level of 200 Bq m-3 (NRPB, 1990). Thirty-one percent of soil-gas readings on the same rock have more than 20 Bq L-1 Rn-222 (20,000 Bq m-3) (Sharman, 1991). Study of radon in soil gas in Northamptonshire by the author has included widespread and intensive work on this rock type, and one aspect of the study is reported here since it may have wider implications for other rocks of similar physical and chemical characteristics. The garden on three sides of a house was sampled on a closely-spaced grid for a period of 1 year. The observed high values of Rn in its soil gas means that counting errors are reduced and enable a study of the spatial relationship of radon in soil gas to radon in the building to be carried out over an extended period. These data also allow observation of spatial variation in Rn emanation from a relatively uniform rock. Seasonal and meteorological variations were also investigated. Previous Work As the potential hazard of high radon levels in dwellings has become known, so there have been attempts to find correlations between indoor Rn and Rn, Ra and U in subjacent soils. Akerblom et al. (1983) noted that Rn in soil
gas under houses is generally higher than that in ground adjacent to buildings, and suggested that buildings reduce both diffusion of Rn from the soil surface, and air exchange in the soil layer under houses. Eaton and Scott (1983) reported that Rn concentration in buildings depended on four factors; pressure differential between soil and house, resistance of the soil, Rn release rate per volume of soil around the foundations, and the average ventilation period of the house. Observation has shown that soils of the Northampton Sand Formation are free-draining and therefore permeable, due to fracturing and fissuring and high porosity. This is especially so in its lower part, the main ironstone group, which underlie the study area. Wilkening (1985) supported the work of Smith, Barretto and Pournis (1976) when he showed that the average distance an atom can move through dry soil by diffusion is 1.6 m (Rn-222) and 20 mm (Rn-220). As buildings tend to dry out the soils beneath, and to some extent around the foundations, radon in buildings is likely to be drawn from the ground within this distance of the foundations. Therefore, by sampling both within 1.6 metres of the building and at several metres away, a variation close to the building might be observed. Fracturing of the rock could lead to preferential emanation along paths of least resistance, which would lead to small-scale spatial variations. On this subject, Taipale and Winqvist (1985) concluded that Rn concentrations vary greatly in a small area, to the extent that the choice of house site to within a few metres can make a substantial difference. Their argument is supported by this study. Schery and Siegel (1986) indicated that channels through soil are important for Rn transport at times of high pressure gradients, but that channels contribute only moderately to overall exhalation. Meteorological variations have significant effects
114
Variations in Rn-222 and Rn-220 in soil gas
4
3
2
1
8
7
B
.5
I0
9
procedures are described in Ball el al. 1991. It is possible to differentiate between Rn-222 and Rn-220 due to the different half-lives of these isotopes. (-Rn-2200 tl/2 = 55 s, will decay much faster than Rn-222, tlj2 = 3.8 days). Radon activity is expressed as Becquerels per litre (Bq L-I). The instrument was calibrated using a calibration standard cell, which has a geometry similar to the Lucas cell used for counting. Cells were calibrated by the National Radiological Protection Board, and quality conliot was achieved by periodic cross-calibration with the British Geological Survey. Methodology
12
11
14
13
15
16
SCALE
17
3 METRES
18
N #,
Figure 1 Plan of the study site showing sample positions.
/1\
|
upon radon concentrations in soil gas; indeed Ball et at. (1991) stated that Rn concentrations in soil gas is controlled as much by meteorological factors as by chemical and mineralogical factors. In 1964, Kraner el aL showed that as pressure decreases, air rich in Rn-222 is drawn from the ground and as pressure rises, air low in Rn is forced into the ground. Miller (1966) noted high radon in soil gas during and immediately after heavy rain. Miller and Ball (1969) showed that where topsoil is thin, absent or pervious, barometric pressure becomes the controlling factor in radon escape from soils, but plays only a marginal role where there is a well developed humic topsoil, in which case soil moisture content affected by relative humidity is a major factor. Small intra-granular pore spaces are easily closed by moisture, so decreasing the emanation of radon from the ground. Miller and Ostle (1973) went on to shown that where there is a thin or permeable topsoil, low pressure and strong wind reduce radon in the near-surface environment, drawing up radon from depth, whilst radon builds up in the ground during calm, anticyclonic conditions. Clements and Wilkening (1974) went further and suggested that changes in pressure of 1-2% can give rise to changes in Rn-222 flux of 20-60%. Ball et al. (1983) found Rn concentrations correlated strongly with soil temperature, and less strongly with air temperature. Instrumentation
An emanometer was used to determine the radon concentration in soil gas samples. Principles and
The garden site measured 18 m • 14 m, and a square grid was sampled at 3-metre intervals, to provide 18 sample locations (Figure 1). These were sampled at approximately monthly intervals. The bedrock is an oolitic, sideritic, limonitic ironstone (Hollingworth and Taylor, 1951), and produces a ferritic brown earth soil (Reeve, 1978). Soil on the site is brown and crumbly, welt developed, with ironstone fragments, and is flee-draining. The results were compared with meteorological records from the weather station at Nene College, Northampton, situated 18 km to the south-west of the site, and with monthly averages for the agro-ctimatic area 22 East; (Smith, !984). Neither of these sources would necessarily be able to identify localised showers in the study area, although each has its use; the former provides a record of changes associated with cyclonic conditions, whilst the averages show long-term variations. Since gas sampling was carried-out at the beginning of each month, weather values were averaged from mid-month to mid-month. Mean summer (May to October) and mean winter (November to April) values were based upon these monthly averages, the criteria used to separate summer from winter being temperatures above or below 10~ Analysis has involved dividing the data set to examine the two isotopes individually, and by distance from the house (since dwellings are known to alter emanation patterns). Spatial and seasonal variations were then examined. This can be summarised as follows: Rn-222 spatial variation. (a) Spatial results at >4 m from the house. (b) Spatial results at 1.6 m from the house. Rn-222 seasonal variation. (a) Seasonal results at >4 m from the house. (b) Seasonal results at 1.6 m from the house. Rn-220 spatial variation. Rn-220 seasonal variation. Results
Rn-222 spatial variation Figure 2(a) shows the annual average pattern of soil gas values. Maximum radon activity is found in the north and west of the site. Variation between adjacent positions can be great, the largest being 63 Bq L-1 difference between two adjacent readings in July. The coefficient of variation (standard deviation as a percentage of the mean) indicates
G. Sharman
115
Figure 2 Spatial variation of radon. (a) Rn-222; (b) Rn-220. the variation, ranging from 39% to 106%. The main problem with this plan is that it cannot reflect true spatial variations due to the presence of the house which is seen to alter the emanation pattern through the year. For this reason the data set was divided to illustrate spatial variation at (a) >4 m from the house (~ssumed to be outside of the influence of the house) and (b) 1.6 m (within the influence of the house).
(a) Spatial variation at >4 m from the house: This sub-set shows spatial variation within a small area which is typical of the variation found on Northampton Sand Ironstone (NSI) on undisturbed ground. The annual mean value is 20
Bq L-1 which agrees with NSi results regionally. There is a consistent spatial relationship, with relatively high spots being found in the same place in successive months. This is indicative of preferred path~ of emanation. The mean radon in the hotspots is 29 Bq L -1, with a mean coefficient of variation of only 32% annually. The mean value for the remainder of the results at >4 m is 14Bq L -1, with a mean coefficient of variation of 44%. This shows that seasonal variations appear to have a lesser effect upon preferred pathways o f emanation than in areas of probably unfractured rock. Other reasons for variation in Rn concentration could include the concentration of mineralogical hosts for U and
116
Variations in Rn-222 and Rno220 in soil gas movement of radon into the building as a component of mass flow of soil gas.
D
Figure 3 Seasonal variation of radon ranked by month. (a) Rn-222; (b) Rn-220. Th in the ground, radium-226 (h/2 = 1,602 yr) and radium-228 (tl/2 = 5.75 yr), and groundwaters which may move decay products of U in the soil/rock, as well as Rn-222 which could migrate 60-70 m (Smith et al., 1976). Radium mobility may be reduced in oxidising environments, where Fe and Mn oxides, clay minerals and organic matter scavenge Ra during precipitation, and in calcareous soils (Levinson et at., 1982). (b) Spatial variation at 1.6 m from the house: Values at 1.6 m from the house are significantly lower than those at >4 m away. The mean annual value is 13 Bq L -1. The hotspot is apparently unaffected by the presence of the building and the remainder of the results at 1.6 m have an annual mean of 9 Bq L-1, with a coefficient of variation of 58%. This is a considerably larger variation than the 44% at >4 m away, and must be partly attributable to the
Rn-222 Seasonal variation Figure 3(a) shows the results for the t8 sample positions ranked by month. Values rose sharply from April (mean of 10 Bq L -1) to peak in May/June ~.25/26 Bq L7 l) then maintained relatively constant v~ues (20-22 Bq L -1) until September. It also illustrates the wide range of Rn-222 values during the summer months (from 4 metres from the house. Radon emanation is therefore not as liable to alteration by se,~onal variations at these points as it is in the surrounding rock, During the period November to April Rn-222 values within 1.6 m of the building were 44% lower than those at 4 m from the house as opposed to the 25% drop in the summer. Between summer (May to October) and winter, values near the house fell by 50%, compared with a 33% drop in values at 4 m away. This leads to the conclusion that the house caused a measurable reduction in radon in the subjacent soil. The results at more than 4 metres from the house are typical of Northampton Sand Ironstone results which leads to the conclusion that the house does not affect soil gas radon at this distance. Variation in Rn-222 emanation was found to correlate negatively at the 99.5% significance level with wind run and positively at the same level with potential transpiration. Wind movement over the soil surface is therefore sufficient
120
Variations in Rn-222 and Rn-220 in soil gas
to reduce radon in the soil to at least 500 mm depth in the Northampton Sand Ironstone. There was a small range in the Rn-220 results which were not associated with preferred emanation paths. Rn-220 correlated positively at the 99.5% level with air and grass temperatures and hours of insolation. These are interpreted as factors controlling radon emanation. Seasonal adjustment to soil gas values can be made which would allow annually corrected values to be obtained.
Acknowledgements The author is grateful to Dr Diana Suthcrland for her encouragement and enthusiasm, and to whom the project owes so much. Also Dr Keith Ball of the BGS for willingly giving of his time, vast experience and resources, and Professor John Hudson for his advice and constructive criticism. Also appreciation must be given to Iris Sharman who allowed her garden to be used in this study, and to Christina Rout of Nene College for time spent consulting meteorological records. This work is dependent upon the financial support of the combined County, District and Borough Councils of Northamptonshire, and a grant from the Research Board of the University of Leicester to Professor J.D. Hudson. Thanks to all concerned.
References Akerblom, G., Andersson, P. and Clavensjo, B. 1983. Soil gas radon a source for indoor radon daughters. Rad. Prot. Dosim., 7, 49. Ball, T.K., Cameron, D.G., Colman, T.B. and Roberts, P.D. 1991. Behaviour of radon in the geological environment: a review. Quart. J. Engineering Geology, 24, 169-182. Ball, T.K., Nicholson, R.A. and Peachey, D. 1983. Effects of meteorological variables on certain soil gases used to detect buried ore deposits. Trans. last. Min. Metall., 92, B183. Clements, W.E. and Wilkening, M.H. 1974. Atmospheric pressure effects on 222Rn transport across the earth-air interface. J. Geophys. Res., 79, 5025. Eaton, R.S. and Scou, A.G. 1983. Understanding radon transport into houses. Rad. Prot. Dosim., 7, 251. Hollingwonh, S.E. and Taylor, J.H. 1951. The Northampton Sand Ironstone. Stratigraphy, structure and reserves. Mere. Geological Survey Great Britain. HMSO, London. Kraner, H.W., Schroeder, G.L. and Evans, R.D. 1964. Measurements
of the effects of atmospheric variables on radon-222 flux and soil gas concentrations. In: The Natural Radiation Environment, pp.191-215. Chicago. Levinson, A.A., Bland, C.J. and Lively, R.S. 1982. Exploration for U o r e deposits. In: Ivanovich, M. and Harmon, R.S. (eds), Uranium Series Disequilibrium: Applications to Environmental Problems, pp.351-381. Clarendon Press, Oxford. Miller, J.M. 1966. Interim Report on the Measurement af Radon in Soil Ab" as a Prospecting Technique. Report No.272. Institute of Geological Sciences, Metalliferous Minerals and Applied Geochemistry Unit. Miller, LM. and Ball, T,I~ 1969. Second Progress Report on the Measurement of Radon in Soil Air as a Prospecting Technique. Report No.286. Institute of Geological Sciences, Metalliferous Minerals and Applied Geochemistry Unit. Miller, J.M. and Ostle, D. 1973. Radon measurements in uranium prospecting. In: Uranium Exploration Methods (Conference Volume), pp.237-247. International Atomic Energy Agency, Vienna. Morse, R.H. 1976. Radon counters in uranium exploration. In: Uranium Exploration Methods (Conference Volume), pp.229-239, International Atomic Energy Agency, Vienna. Nason, R. and Cohen, B.L. 1987, Correlation between z ~ a l in soil, 222Rn m 9 sod. gas and 222Rn reside , . adjacent . houses. Health Phys., 52, 73-77. NRPB (National Radiation Protection Board). 1990. Human exposure to radon in homes. In: Documents of the NRPB~ VoI.1, No.1. HMSO, London. Reeve, M.J. 1978. Soils in Northamptonshire 1. Sheet SP66 (Long Buckby). Soil Survey Record No.54. Rothamsted. Schery, S.D. and Siegel, D. 1986. The role of channels in the transport of radon from the soil. J. Geophys. Res., 91(t2), 366. Sharman, G. 1991, Radlometric investigation of radon in soil gas over Jurassic rocks of Northamptonshire, England. Environ. Geochem. Health, 13(3), 146-147. Smith, A.Y., Barretto, P.M.C. and Poumis, S. 1976. In: Exploration for Uranium Ore Deposits, p.185. I.A.E.A. Publication STI/PUB/434. Smith, L.P. 1984. The Agricultural Climate of England and Wales (1941-70). Reference book No.435. Ministry of Agriculture, Fisheries and Food, HMSO, London. Sutherland, D.S. 1991. Radon in Northamptonshire, England: geochemical investigation of some Jurassic sedimentary rocks. Envh'on. Geochem. Health, 13(3), 143-I45. Taipale, T.T. and Winqvist, IZL 1985. Seasonal variations in soil gas radon concentration. Sci. Total Environ., 45, 121. Wilkening, Mo 1985. Radon transport in soil and its relation to indoor activity. Sci. Total Envh'on., 45, 219.
(Manuscript No.267: received February 24, 1992 and accepted after revision April 14, 1992.)
