Environmental Geochemistry and Health 1993 15(1) page 27

Geological and geochemical factors affecting the radon concentration in homes in Cornwall and Devon, UK T.K.Ball 1 a n d J . C . H . Miles 2

1 British Geological Survey, Keyworth, Nottingham NG12 5JH, UK 2 National Radiological Protection Board, Chilton, Didcot, Oxon OXl I ORQ, UK Abstract Recently collected data for radon levels in houses in Devon and Cornwall are compared with geological and geochemical information. The region is underlain by granites intruded into folded sedimentary rocks. The highest incidence of affected houses is on granites. The granites are characterised by moderate uranium concentrations, a deep weathering profile and uranium in mineral phase which is easily weathered. However, while the uranium may be removed, radium, the immediate precursor of radon, can remain in situ. Radon is emanated easily from the host rock, and high values of radon in ground and surface waters and soil gases have been detected. The granite areas are also characterised by high values of uranium in stream sediments and waters. In contrast, other zones of high uranium in stream sediment samples do not necessarily exhibit high house radon concentrations, especially when underlain by relatively impermeable rocks. Permeable ground can give rise to high incidences, of affected houses despite having uranium levels close to the crustal abundance. It is concluded that the most efficient method of identifying zones of high radon potential is the soil gas radon survey.

Introduction

Radon (Rn) is a naturally occurring radioactive gas. It is produced in the decay series of both uranium (U) and thorium (Th) (Table 1). Following its discovery, it remained a chemical curiosity for decades, being promoted at times as a '~health giving' gas at various spas. During the uranium prospecting boom of the 1950s and 1960s, the value of Rn-222 (commonly known as just 'radon') as a path-finder for the parent uranium was recognised. M u c h of the i n f o r m a t i o n c o n c e r n i n g the environmental behaviour of the gas stems from this period. Early work on the geochemistry of radon has been reviewed by Miller and Ostle (1973), Dyck (1972), Smith et al. (1976), Durrance (1986) and Ball et al. (1991). Geochemistry The three naturally occurring radon isotopes, with atomic masses of 219, 220 and 222, are found in the decay series of U-235, Th-232 and U-238 respectively. Rn-219 (actinon) has a very short half-life, of about 4 seconds. It occurs in the decay chain of U-235, present as 0.7% of natural uranium. The abundance of Rn-219 in gases from most geological sources is therefore limited. Rn-222 (radon, half life 3.82 days) is a member of the U-238 decay series, the stable end

product of which is Pb-206 (Table 1)o The U-238 decay chain may be divided into two portions separated by the radium isotope, Ra-226, w.hich has a half-life of about 1,600 years. Most of the radionuclides before Ra-226 have long half-lives, whilst those below it have relatively short half-lives. Following complete separation of uranium from its daughter products, 1,000,000 years is required for the decay series to achieve 91% of secular equilibrium, in which state the daughter products are produced at the same rate that the U-238 decays. Radium's geochemical behaviour is markedly different from uranium, being generally much less mobile in the oxidising (surface weathering) environment. Depending on its mineralogy, uranium may be removed from the weathering zone and soils, but the radium frequently remains behind. The radon and the high energy gamma activity from Bi-214, a further decay product, are relatively unaffected by the loss of the uranium. Rn-220 (thoron) has a half life of 54.7 s and is a member of the Th-232 decay series. The time taken for secular equilibrium to be attained in the thorium decay series is about 70 years. Radon is the most water soluble of the noble gases and may be transported from its site of generation. Being an inert gas, radon is relatively unaffected by the chemical buffering reactions that often control the generation of other gases in rocks and their weathering products.

28

Radon concentrations in homes

Table 1 The U-238 and Th-232 decay series (from Lederer and Shirley, 1978). Isotope

Decay

Half-life

Uranium-238 decay series

U-238 Th-234 Pa-234(m) Pa-234 U-234 Th-230 Ra-226

c~ [3 [~ ~ c~ c~ ~

Rn-222

c~

Po-218 Pb-214 At-218 Bi-214 Po-214 T1-210 Pb-210 Bi-210 Po-210 T1-206 Pb-206

~ ~ ~ ~ ~ [~ 13 ~ a 13 stable

4.468 × 109 yr 24.1 day 1.18 min 6.7 hr 2.48 × 105 yr 7.52 × 104 yr 1,602 yr 3.825 day 3.05 min 26.8 rnin 2s 19.7 rnin 1.64 x 10--4 s 1.32 min 22.3 yr 5.02 day 138.3 day 4.19 min

Thorium-232 decay series

Th-232 Ra-228 Ac-228 Th-228 Ra-224

~ [~ [3 a ¢x

1.39 × 10•0 yr 5.75 yr 6.13 h 1.913 yr 3.64 day

Rn-220

o~

Po-216 Pb-212 Bi-212 T1-208 Bi-~12 Po-212 T1-208 Pb-207

o~ 0.145s [3 10.64 h ~37% 60.5 rnin [~ 3.1 min 1363% 60.5 rain ~ 3.04 x 10-7 s [3 3.1 min stable

54.7 s

Health Aspects Initially, radon was regarded as a fairly harmless or even beneficial component of gases from geological sources. More recently its importance as the major contributor to the radiation dose received by the human body in the UK has been recognised. The National Radiological Protection Board (NRPB) has shown that approximately 50% of the total dose for the average person in the UK results from combined radon and thoron daughter products (Clarke and Southwood, 1989). The major health hazard from radon and its daughters has been identified as an increased risk of lung cancer (O'Riordan et al.,

1987). This is based upon epidemiological studies on uranium miners (ICRP, 1987; NRC, 1988) and supported by animal experiments (USDOE, 1988). In January 1990, the NRPB issued revised advice on radon in homes, resulting in a new action level for radon of 200 Bq m -3. Parts of the country in which there were one percent or more of the homes above the action level were to be regarded as 'Affected Areas"'. The Board recommended that within Affected Areas, the government should designate localities where precautions against radon in new houses were required. The first Affected Area to be defined consists of the counties of Devon and Cornwall in the South-west of England (Miles et aL9, 1990).

Radon Mapping The potential for high radon concentrations in buildings in an area depends on factors related to the local geology such as the uranium and Ra-226 distribution, and permeability of the substrata. Whether this potential is realised as a high radon concentration in a building depends on factors related to the structure of the building and the way it is used. Soil gas is drawn into a building by a slight underpressure indoors which results from the warmer air rising. Hence radon problems in houses are due to bulk flow of soil gas carrying radon, with a relatively small contribution by the diffusion of radon through building materials. The introduction and retention of radon and radon daughters may d e p e n d u p o n v e n t i l a t i o n , i n t e g r i t y o f the foundations, wind speed and direction. Maps showing the potential for high radon concentrations in homes can be produced using two approaches: by studying the geological factors that influence radon potential, or by measuring the radon concentrations in a sufficiently large sample of homes. Each approach has its strengths and weaknesses. Measurement of radon in homes is more definitive, as the potential for high levels is demonstrated by the results of the measurements. M o r e o v e r this approach does not require an understanding of the mechanisms causing the high concentrations in order to map them. However, this strength is also its weakness, as it only allows m a p p i n g in areas w h e r e large n u m b e r s of measurements have been made in homes, and predictions of other areas with radon problems is not possible. Mapping of the radon potential, based upon an understanding of the geological factors affecting this potential, allows the mapping to be extended to other areas where the same factors may be at work. The two assessment techniques may be seen as complementary, with the geological work identifying areas where there are likely to be problems, and measurements in houses defining the magnitude .and extent of the problem. This paper compares the results of the two approaches (geology vs house measurements) in south-west England in order to clarify the understanding of the geological factors involved. It identifies the areas where the

T.K. Ball and J.C.H. Miles

29

Percent of homes > 200 Bq m -3

1-3 3-10 ~1~

',10-30 30

>

....iii[~....

Figure 1 Summary of house data. Percentage above action level in 5 km squares for Devon and Cornwall. SWE

Granite Bampton Posf-Triassic New Red Sandstone Carboniferous

~

.

Devonian Pre-Devonian

N

+ • -:::::::. ,,::.v:." + :...v.v.. v.v: _... . . . . . . : ))Z.:.)Z-:.:... ÷

+

+

-I-

I

4.-

iiiiiii! !iii ..................................... +

Littlehom

" ''"""""::::::::::::::::: : +

St. lv#s ~ii:::i:i:::!:!:i:i:!!!iiiiii!ii:!:i!i]!!!iiii!:~ 'Sf Ausfell

~r~mgst3nage Start

0

Point

80

km

Lizard

Figure 2 Simplified geological map of south-west England (after Edmunds, Mckeown and Williams, 1975). present understanding or data are insufficient to allow radon potential mapping on the basis of the present geological data-base.

Radon Concentration in Homes The NRPB has carried out surveys for radon in homes in Cornwall and Devon, using passive etched-track detectors, to determine the long-term (three to six months) average radon concentration.

Results for more than 8,000 homes have been mapped (Miles et aL, 1990). The distribution of radon concentrations in homes is approximately log-normal, both for large or small areas, a property taken into consideration when the data were treated in terms of areal distribution. The data are considered in relation to a 5 km grid in order to maintain confidentiality. Not all such squares have sufficient data to enable a statistically

30

Radon concentrations in homes

valid estimate to be made of the frequency of high values. However, the distribution relationships for nearby grid squares with adequate data are used to derive estimates of the mean in squares with few data. For detailed discussion of the procedures for estimating and smoothing the data see Miles et al. (1990). Figure 1 shows the percentage of homes exceeding the action level in the 5 km grid squares in Cornwall and Devon. In Figure 3B the data is recast, using the procedures described above, but with final smoothing and drawing of the contours showing >30% and >10% by eye. It should be noted that the contours cannot be drawn with a better accuracy than 5 km because of the nature of the data, especially in sparsely populated areas, and because even in areas of apparent geological homogeneity there is a wide spread in domestic radon concentrations.

Geological Setting A very simplified geological sketch map of the area is shown in Figure 2 (after Edmonds et al., 1975). The rocks of Cornwall and Devon comprise dominantly argillaceous and arenaceous sediments of the Devonian and Carboniferous age. These were folded, faulted and intruded by a granite batholith. In the east, these rocks are overlain by Mesozoic strata. The oldest rocks occur in the Lizard (an igneous complex with compositions from ultrabasic to granitic) and near Start Point (schists and gneisses). The main part of the peninsula is underlain by rocks of the Armorican Geosyncline. Sedimentation was rapid during the Devonian. In the west, flysch type sedimentation prevailed, but further east the Devonian strata are deeper water mudstones, interbedded with volcanic rocks and thick limestone lenses. In the Exmoor area the Devonian contains a greater proportion of deltaic material derived from the Old Red Sandstone continent to the north. Shallow marine conditions further south led to the accumulation of thick intercalated sequences of shales, siltstones and thin sandstones. Carboniferous sedimentary rocks (The Culm Measures), occupy an area of some 3,000 km 2. The lower (Dinantian) part of the succession comprises dark grey s h a l e s , with s a n d s t o n e s , lavas, agglomerates, lenticular limestones and cherts, representing slow accumulation under quiescent conditions disturbed by periodic volcanicity. The Upper Carboniferous comprises the Namurian Crackington Formation, a thick sequence of turbiditic sandstones with interbedded mudstones. This is overlain by the Westphalian Bude Formation comprising thick and massive sandstones with interbedded thin siltstones and mudstones. Granites were intruded into the folded Devonian and Carboniferous. Present day granite outcrops reveal only the uppermost few hundred metres of the Cornubian Batholith. Exposures

exhibit three main magmatic types (Stone and Exley, 1986) of which the types B and C are biotitic and type E is lithium-mica bearing. Some 90% of the outcrop is type B, a medium- to coarse-grained adamellite with numerous textural variations; much of the remainder is occupied by type C, which is similar but finer grained. The New Red Sandstone, underlying about a quarter of Devon, is a major terrestrial red-bed succession resulting from erosion of upland areas and accumulation of the debris in surrounding basins. It i n c l u d e s strata of u p p e r m o s t Carboniferous, Permian and Triassic age, and outcrops mainly in East Devon. The strata dip eastwards beneath Jurassic beds. The irregular western margin reflects the rugged pre-New Red Sandstone topography; its mountain ridges and valleys, which generally trend E-W, formed a series of cuvettes extending from the south to the north coast of the SW peninsula. The Crediton valley is the most outstanding cuvette extending some 40 km west of the main outcrop. The rocks comprise b r e c c i a s , c o n g l o m e r a t e s , s a n d s t o n e s , and mudstones. Some of the breccias contain porphyry boulders thought to represent erosion from volcanic edifices which overlay the Dartmoor granite. The Jurassic strata comprise impure limestones with shales overlain by a mostly m u d s t o n e sequence. The Cretaceous is represented by basal sandstones, mudstones and clays which are succeeded by the chalk.

Uranium Mineralisation Uranium has been mined in Cornubia, either as a major component or as a by product, for about a century. The main producer was the South Terras Mine near St Austell although substantial quantities were raised near St Ires (Dines, 1956); uranium minerals occur in small amounts at numerous localities. Two types of occurrence are recognised : within high temperature tin/tungsten/copper lodes and in later 'crosscourses' (for summary see Ball et aL, 1982). The general paragenesis of uranium and associated minerals is similar in both environments. The uranium is later than the tin/tungsten and some of the copper rnineralisation, and is often associated with mesothermal or lower t e m p e r a t u r e a s s e m b l a g e s . The main ore is pitchblende and even when intensively phosphatised near the surface, the phosphates pseudomorph pitchblende. S p a t i a l l y m o s t of the u r a n i u m vein mineralisation occurs within about 2 km either side of the granite contact (Figure 3A). Uranium mineralisation ranges from small, high-grade, to more disseminated occurrences (Bowie etal., 1973). Small uraniferous mineral occurrences near Start Point, and also the uraniferous/vanadiniferous nodules in the New Red Sandstone are not related to a granite contact. These latter have been known for decades, (Carter, 1931; Perutz, 1939; Ponsford,

T.K. Ball and J.C.H. Miles

1.

31

................

Dartmoor

2. 3. 4. 5. 6. 7. 8. 9. 10.

i

Tregonning-Oodolphin Bosworgey St. Agnes Cligga Castle an Dinas Hemerdon Redmoor / Carnmenellis f Littleham 45 .

Geological and geochemical factors affecting the radon concentration in homes in Cornwall and Devon, UK.

Recently collected data for radon levels in houses in Devon and Cornwall are compared with geological and geochemical information. The region is under...
987KB Sizes 0 Downloads 0 Views