Radon Update




Members of the Committee who participated actively in writing and editing this report include: A.B. Brill, Comm. Chairman, with significant contributions by Committee members, D. V. Becker, D.R. Brill, K. Donohue, B. Greenspan, S. Goldsmith, H. Royal, E. Silverstein, and E. Webster. K. Kase, Stanford Linear Accelerator Center provided valuable advice and assistance in the preparationIt of the document.


  • Permission was granted by Dr. A.B. Brill for use of this article at this WWW site. It has been reproduced from its publication in The Journal of Nuclear Medicine, Vol. 35, No. 2; Feb 1994 and material submitted by the author. This article may not be reprinted without the author's permission. The article has been form modified for WWW publication. For correspondence or reprints contact: A.B Brill, University of Massachusetts Medical Center, 55 Lake Ave. North, Worcester, MA 01655.
  • Numbers in parenthesis point to the list of referenced sources at the end of the paper. This document is best viewed with a web browser that can read tables and superscripts. Numbers shown as " 3x10-2 " are read as 3 times ten to power of minus 2 (i.e., 0.03). Some editor notes have been added to the text and are shown in brackets [ ].
  • Many of the units used in science are broken down into smaller units or expressed as multiples, using standard metric prefixes. As examples, a kilobecquerel (kBq) in 1000 becquerels, a millirad (mrad) is 10-3 rad, a microrem (�rem) is 10-6 rem, a nanogram is 10-9 grams, and a picocurie is a 10-12 curies. These units are examples used frequently throughout this short paper. To find definitions of terms you're not familiar with, look on our glossary page.
  • Common abbreviations used on this page are: m - meter, m3 - cubic meter, g - gram, kg - kilogram, Bq - becquerel, Sv - sievert, Gy - gray, Ci - curie, ppm - parts per million, yr- year, hr - hour, L - liter


The potential hazard of radiation exposures to radon gas and its daughter products from natural background has been highlighted in the press and has become a matter of concern and a source of confusion to the public. Home owners are besieged with devices to measure radon levels, and may not know what to do about the results they get. The Environmental Protection Agency (EPA) has issued guidelines (1-7), as has the National Council on Radiation Protection and Measurement (NCRP), the International Commission on Radiation Protection (ICRP) and other groups concerned with radiation protection matters. The guidelines and recommended actions are in general agreement, although they differ in detail. A matter of concern is that the media have chosen the lowest level of the guideline, which the public translates into the upper limit of "safe dose". It is not surprising that there is widespread confusion regarding the nature and severity of the problem, the risk magnitude, the steps that should be taken to cope with different circumstances, and the costs associated with different actions. The material distributed in the Radon Update is intended to provide information needed to help understand these issues and to provide a compilation of the relevant facts for those individuals interested in the potential health effects of environmental radon.

Many articles have been published in the scientific literature dealing with the issue of human risk from radon exposures, and many of these appear in publications by the National Academy of Sciences (NAS), NCRP, EPA , Department of Energy (DOE) (8-15), as well as in the radiation related journals, primarily the Health Physics Journal, and its Newsletter, and Radiation Research. Several general references are given at the end of the reference section to complement the information presented herein. There are different perspectives on the significance of environmental radon, and the referenced documents illustrate the range of opinion regarding this issue.

Several key points are underscored in this report:

  • The risk of environmental radon levels is not higher now than in the past, when residential exposures were not considered to be a significant health hazard. It has now been raised from a nominal natural background contributor to human exposures to the prime contributor based on a new way of calculating dose, and not on increased exposures in the home due to tighter more energy efficient houses, or new information on radiation dose and/or biologic risk.
  • The great majority of the radiation dose is not from radon itself, but from the short-lived alpha particle-emitting radon daughters, most notably Po-218 (radioactive T1/2=3 minutes), and Po-214 (radioactive T1/2=0.164 milliseconds), along with beta particles from Bi-214 (T1/2=19.7 minutes). [T1/2 is physical half-life].
  • Measurement of radon in homes is simple and relatively inexpensive and may be accomplished by various means. Cannisters remain the method of choice for use by the average homeowner.
  • There is no way to avoid radiation exposures completely, much of which comes from the natural environment as well as from radioactivity in our own bodies.
  • Radon gas can penetrate houses from many sources in many fashions. It is not possible to radon-proof a home, but it is possible to reduce its level. The most important contributor to indoor radon is the soil from which radon can be drawn through large and small subsurface gaps in the house foundation. Houses that are in direct contact with the ground will have higher radon levels than houses with an air space under the dwelling. Radon levels in the upper floors of a multi-story building are lower than on the ground floor. If one lives in a high radon area, it is prudent to determine the level in the home and to take appropriate action if the level is found to be high, i.e. greater than 4-8 pCi/liter (NCRP recommended level is 8 pCi/liter, the EPA recommended level is 4 pCi/liter). [pCi is the activity of the substance and is equal to 10-12 Ci or 2.2 radioacitve emission per minute].
  • The shape of the dose response curves for miners exposed to alpha-emitting particles in the work-place is consistent with current biologic knowledge. It is linear in the low dose range and saturates in the high dose range. No detectable increase in lung cancer frequency is seen in the lowest exposed US miners, i.e. those with exposures <120 wlm, the relevant dose interval for most homes. in most cases, however, it is not possible to exclude a positive correlation due to wide uncertainty ranges on both the dose (x) and effect (y) axes. [wlm is a unit of expsoure to radon, working level month and is defined below]
  • Evidence for a health effect from radon exposure is based upon data gathered from epidemiologic studies of miners, and animal studies. Extensive radiobiologic data predict a linear dose response curve in the low dose region due to poor biological repair mechanisms for the high density of ionizing events that alpha particles create. If this is the mechanism that underlies lung cancer induction, there is reason to believe that there will be some degree of increased lung cancer risk even from low level environmental radon exposures. However, no compelling evidence for increased lung cancer risk has yet been demonstrated from "acceptable" levels (<4-8 pci/liter). increased lung cancer risk is primarily in cigarette smokers, and abatement of that risk can best be accomplished by changes in smoking status.
  • Mitigation of radon in the home can be accomplished by a variety of means. The level to which remediation should be directed is under dispute, to the extent that some groups even suggest the level should be maintained at less than or equal to 0.2 pCi/liter (the average outdoor level). The societal cost of testing and mitigation at the EPA recommended level (4 pCi/liter) is 44.5 billion dollars (1991 dollars), and rises to 101.2 billion dollars if the action level is set at 2 pCi/liter. At the NCRP action level, 8 pCi/liter, the cost is estimated at approx. 15 billion dollars.

Radon, Rn-222 (T 1/2= 3.82 days), is a daughter product of radium, Ra-226, which in turn is derived from the longer-lived antecedent, U-238. Thoron, Rn-220 (T 1/2 = 56 seconds) is a daughter of thorium, Th-232, which is present in larger amount in the earth's crust than radon. Because of thoron's short half-life, it is essentially all gone before it leaves the ground, and is of no significant radiobiologic consequence. These radionuclide series are present in slowly decreasing amounts in the environment (geologic time scale), due to radioactive decay of their parents, which has been known and understood since the end of the last century.

Widely varying radon levels exist in different regions related to geological circumstances. New concern regarding radon exposures is traceable to the discovery that there are more houses with high radon levels than previously realized and to the use of a new method of expressing and summing doses from partial body exposures, such as the lung dose from radon daughters (7-16). This method of expressing dose was promulgated by the ICRP and the NCRP based on defined weighting factors which make it possible to sum partial body doses and thereby estimate a total body dose which would have a quantifiable risk. This quantity is defined as the Effective Dose (ED) (16). Thus, the previously estimated partial body environmental radon dose to the tracheobronchial epithelium (TBE) (2500 mrem/year.) was not included in whole body dose calculations because that exposure was limited to a small fraction of the body.

The new method of calculation multiplies the 2500 mrem/year. dose to the TBE by a weighting factor (WF) which allows the dose to the TBE to be included in the ED from environmental radiation exposure Different WFs have been proposed, including 0.12 (EPA), 0.08 ( NCRP) and (NAS-NRC BEIR V), and 0.06 (ICRP). These WFs raise the radon contribution to the whole body from 0 mrem to 300, 200, and 150 mrem respectively. NCRP quotes an uncertainty of +/- 50% in these numbers. Based on these estimates, radon in equilibrium with its daughters delivers 2 times more dose than previously accepted as the total dose received from all sources of natural background exposure (approx. 100 mrem/year on the average in the United States). Thus, it is not surprising that adoption of the effective dose notion by many radiation protection groups (including the NCRP and the EPA in the United States), has led to increased concern regarding the potential health effects of radon. It should be noted that lung cancer risk coefficients from radon are not increased. There are no new cases of lung cancer that led to the increased dose estimate. In fact, the new estimates of radiation dose, imply a lower risk coefficient. That is, when the same number of lung cancer cases that occur are attributed to the higher doses (ED), the risk per unit exposure is decreased. The effective dose concept is discussed at greater length in NCRP Reports #93 (17) and #100 (18) , and ICRP #60 (7).

* to soft tissues Modified from reference 9.

** Dose equivalent to bronchi from radon daughter products. Assumed weighting factor for effective dose=0.08.


Almost all measurements of radon levels in the home or outdoors are expressed as the concentration of radon in units of picocuries per liter of air (pCi/liter), or in SI units as Becquerels per cubic meter (Bq/m3), or radon daughters are expressed in working levels (WL) . A working level month (WLM) is defined as 170 hrs (2l.25 working days/month x 8 hrs/day) in a work place at one WL. Thus, a 12 hour a day exposure in the home at one WL, corresponds to approx. 26 working level months per year i.e. 2.1 X the occupational exposure, assuming equal radon levels at home and in the work place. Exposure rate is typically given in working level months per year (WLM/year).

The WL unit was developed for use in radon occupational exposure assessment since often there was incomplete information on the degree of equilibrium with daughter products. It is the dose delivered in one liter of air that results in the emission of 1.3 x 105 MeV of potential alpha energy. (19). The amount of time spent in the mine or in the home determines the number of WLM associated with a particular exposure level, but because most people spend more time at home than at work, the WLM could be higher than from a comparable mine radon daughter concentration. Typical outdoor levels in the U.S.A. are given by NCRP # 78 as 0.2 pCi/liter (11).

The correspondence between WLs and radon concentration in air in pCi/liter depends on the extent to which radon daughters (which impart dose to the tracheobronchial epithelium dose, "TBE") are in equilibrium with the parent radon. At complete equilibrium, one pCi/liter results in an exposure equal to 0.01 working levels. The assumption is generally made that inside buildings the radon decay product/radon equilibrium is 50%. Thus, inside buildings, 1 pCi/liter = 0.005 WL, or 1 WL= 200 pCi/liter. (Note: Consideration must also be given to radionuclide attachment and distribution. See dosimetry section)

Modified from DOE, 1990. *1 WL (Occup. exposure) X 12 Month/year (i.e. 8h/d, 5d/wk) = 12 WLM/year and 1 WL in the home conveys a higher dose due to occupancy time: 1 WL (Home) X 51.6/2 "Month/year" (i.e. 12hr/d, 7 d/wk occupancy) = 25.8 WLM/year, (0.2 pCi/liter = Typical Outside Level = 120 mrad/year (TBE)), Assuming, 8 hrs. outdoors (33%) @ 0.2 pCi/liter, and 16 hrs indoors (67%) @ 1 pCi/liter, NCRP Estimates: Average Radon level = 0.75 pCi/liter = 0.004 WL , which corresponds to 51.6 x 0.004 = 0.2 WLM/year, TBE dose from environmental radon depends mostly on indoor levels. If home levels are at the NCRP action guideline (8 pCi/liter), then Total Radon Exposure = 0.029 WL (i.e. 7.25 X above the average, i.e.0.004 WL).

120 mrad/year (TBE) x 20 (RBE) = 2400 mrem/year (TBE)

2400 mrem/year x 0.12 (EPA WF) = 300 mrem/year (ED)

2400 mrem/year x 0.06 ICRP WF) = 150 mrem/year (ED)

2400 mrem/year x 0.08 (NCRP WF) = 200 mrem/year (ED)

NOTE: There is a twofold difference in the Effective Dose (ED) based upon the weighting factor (WF) and the relative biological effectiveness (RBE) used. Further, the tracheobronchial epithelium (TBE) dose, varies by as much as a factor of three between most widely assumed models (8).


Radon 222 is a decay product from the U-238 decay chain, illustrated in Fig. 1. The external dose from Rn-222 and its airborne progeny is a very small fraction of the natural external radiation dose received by individuals. However, inhalation of radon and its daughters may be followed by deposition of potentially large amounts of energy, i.e. absorbed dose in the tracheobronchial epithelium from the short-lived a- and b- particle emitting decay products, primarily from Po-218, Pb-214, Bi-214, and Po-214.

The radiation dose to the bronchial mucosa from these high LET radiations depends on radionuclide deposition and residence time. Particle deposition depends on three mechanisms: impaction, sedimentation and diffusion. Deposition and residence time depend on whether the radioactivity is attached to air borne dust particles, or is unattached (following inhalation unattached daughters are able to deposit deeper in the lung than dust particle-attached radon daughters). Respiratory factors (breathing rate and depth, mucociliary clearance, and site of impaction in the bronchial tree) influence depth of penetration into the lung with deeper particles having a longer residence time. Dose to the TBE from radon per se is negligible since its intrapulmonary residence time is short with respect to its half-life. The high absorbed dose is from the decay of radon daughters attached to the TBE.

Although the location of the critical target for lung cancer induction is not known it is assumed to be the basal cell at the 4th generation of the tracheobronchial tree and beyond , and dose delivered to the mucous covered cell is calculated to the basal cell nucleus at this location. The depth of mucous covering the critical target strongly influences the dose received from the short-range energetic alpha emissions as does the integrity and activity of the muco-ciliary escalator that carries particles in a retrograde fashion out of the lung. Alpha particles contribute more than 85% of the TBE dose, and this dose will be deposited within 30 um of the decay site.

Dose calculations depend on the airborne radiation levels and concentration of radon and its progeny and on the modeling assumptions noted above (20). The radiation levels can now be measured with reasonable accuracy and precision. Present calculations for an average indoor and outdoor exposure (0.75 pCi/liter) to a cell 22 um deep, in a 4th generation airway, range from 140-340 mrad/year, with the highest doses to 10 year old children. (Note: a continuous exposure to radon at a concentration of 1 pCi/liter would result in an annual exposure to radon progeny of 0.25 WLM/year, which corresponds to 188 mrad/year, or 3750 mrem/year for an adult, assuming a quality factor of 20 for alpha particles.) (1).


There are 3 classes of measurement techniques that are used: 1) grab sampling, 2) continuous active sampling, and 3) integrative sampling.(21) Grab sampling provides instantaneous measures of radon or radon progeny in air. Since values fluctuate widely depending on various factors, grab sampling techniques are used in industrial monitoring. Continuous active sampling involves multiple measurements at closely spaced time intervals over a long period. These are costly and only recommended when other measures indicate a problem and the source of radon entry needs to be pinpointed precisely. Integrative sampling devices are passive, and collect data on radon levels over a fixed period of time.

Typical integrative devices are charcoal cannisters, or alpha track film dosimeters. The charcoal devices (Fig. 2) come in a cannister, which is opened and placed in selected locations. Radon in air diffuses into the cannister and is adsorbed onto charcoal. Following exposure for 2-7 days, depending on the particular device and the instructions for its use, it is sent back to the supplier who assays it by counting the gamma rays from the daughter nuclide (e.g. Pb-214). Since Rn-222 has a half-life of 3.8 days, if the cannisters are exposed for several weeks or longer, the results will be indicative primarily of the activity sampled toward the end of the exposure interval. Some of the cannisters have an additional filter that affects the integration period, and make the cannister insensitive to thoron (Rn-220).

Figure 2. . Charcoal cannisters can be manufactured simply using a small can covered by a screen. The charcoal is contained in the space below the screen, which is held in place by a ring. A top is fitted over this arrangement until exposure, at which time it is removed. The top is replaced and sealed at the end of exposure, and the entire can placed on a NaI gamma counting system for analysis. ( From Cothern and Smith, 1987, p. 67).

A second type of integrative sampling detector is the alpha particle track etch detector (Fig. 3). This device can be used to average data over longer periods of time, as the track etch evidence of exposure does not decay. However, dust and electrostatics make them less reliable and they are only sensitive to radon gas activity.

Ordinarily, charcoal cannisters are used to measure activity in the area where occupants spend the most time. Indoor radon levels are normally highest in winter. If levels are not elevated at that time, additional measurements should not be necessary. If high activity levels are found, then additional measurements should be made in other parts of the house, including the basement which usually is the highest area, and at other times of year. Should more intensive programs be warranted to pin point and remedy high levels, resources recommended by State, or local health or environmental protection agencies are available. These also advise on testing resources and can provide lists of radon testing laboratories that performed successfully in the EPA proficiency testing program.

Figure 3. Alpha track devices consist of a material, such as film, which sustains damage along the track of an alpha particle. The material is then placed into etching fluid, which enlarges the track by extending the region of damage. Once the tracks have been sufficiently enlarged to become visible, their density at the surface of the material is determined and related to dose. (Photo courtesy of Terradex) 19

A new method of estimating the long-term integrated radon exposures was developed recently in Sweden measures the amount of Po-210 in vitreous glass found in the home (23-25). Short-lived radon daughters plate out on the glass and undergo alpha decay leading to the formation of Pb-214 which decays to Pb-210 (22 year T 1/2). The activity of Pb-210 or its daughter product Po-210 can be used to estimate cumulative exposures to residents from radon daughter concentration in the home. The activity of the glass is measured in the home using large surface area ionization chambers or surface barrier detectors which assay the amount of the 5.3 MeV alpha energy emitted. (Fig. 4) The phenomenon is based on the fact that when the alpha particle is emitted the daughter nucleus (Pb-210) recoils in the opposite direction and gets embedded in the glass close to the surface. One would presume that 50% of the recoils would result in deposited activity in the glass, but the ratio is closer to 30%. Factors such as heat circulation patterns in the room, and the frequency with which surface grime is washed from the window does not appear to seriously affect the estimated dose. (26)

Figure 4. Surface plate out followed by alpha recoils embeds long-lived daughter activity in a surface. Range of recoiling nuclei in glass is about 50 nm. (26)

Another new technique promises to be useful for estimating cumulative in vivo absorbed dose from radon. The technique measures the Pb-210 content in the skull. Pb-210 emits a 47 keV gamma ray (4% abundance), which can easily penetrate the soft tissue that intervenes between the skull and the 5 large area thin crystal NaI scintillator detectors placed about the subject's head. Assuming 14% of the bone mass is contained in the skull (27), and the effective half-life of Pb-210 in the body (12-18 years), one can estimate the cumulative dose from radon in measured subjects (28).

The alpha recoil method makes it possible to estimate the dose from radon daughters accumulated over the lifetime of window or picture frame glass in a particular residence, or in a miner's lamp, if one can be located. Residential measurements should make it possible to rank houses on an exposure index, and miner's lamp readings could be important in ranking mines. The measurement of Pb-210 levels in vivo is also likely to be useful for ranking individuals into dose groups. However, Pb-210 levels are complicated by other factors including radium in the diet (food, water sources) and one can not differentiate recent from old exposures which will make it difficult to estimate person-years at risk for individual subjects. Nonetheless, all these new methods should be useful in epidemiology studies, but it is likely that it will still be difficult to estimate TBE dose accurately.


The use of average values of dose from natural background radon suggests that dose is rather uniform, whereas in fact radon levels vary markedly in different regions of the country, based on geologic factors, relations to mines and mine tailings, as well as levels of radium and radon in water supplies (29). In general, high levels of radon are associated with granite igneous rocks, shale and dirty quartz sedimentary rocks, phosphate deposits and some beach sands, which may contain high levels of radon progenitors, i.e. uranium, or thorium. Basalt has relatively little uranium, i.e. half of the average value found in rocks of all kinds, whereas the granite strata contain upwards of twofold increases above average values (0.7 pCi/gram). Figure 5 shows a map of United States locations with potentially high radon levels based on geologic formations. Rock types identified as giving rise to potentially high radon levels are listed below (29):

Rock types that are high radon sources in the U.S.A. include :

  1. Uraniferous metamorphic rocks and granites: Sheared faults in these formations cause some of the highest indoor levels in the US, particularly in the Rocky and Appalachian ranges, and the Sierra Nevada.
  2. Marine black shales: Sources of high radon throughout the US and especially the central region from Ohio to Colorado.
  3. Glacial deposits derived from uranium-bearing rock and sediment. Major components of glacial deposits in the northern Midwest. They have high radon emanation due to large surface area, and high permeability due to cracking when dry.
  4. Soils derived from carbonate, especially karstic terrain which are high in uranium and radium.
  5. Uraniferous fluvial, deltaic, marine, and lacustrine deposits, which provide most of the US uranium, and are located in the Western US.

Typically the maximum Ra-226 concentration in phosphate ores is about 40 pCi/gram (about 50 times greater than average concentration in soil). Thus, ore that is close to the surface, or residues from mining that are left on the surface, can give rise to very high local concentrations. In the U.S.A. this problem is mostly localized to Polk County, Florida, and although not a great contributor to global levels, there is concern within those communities, and local abatement efforts are underway. In some mining communities in Colorado, local releases from uranium mining residues, and mine tailings, can be significant sources of atmospheric radon. Typical emanation rates may exceed 300 pCi/m2 s. (30). In 1983 EPA established regulations that average releases from tailing sites may not exceed 20 pCi/m2 s (which is 40 times greater than the average from soil). Releases from coal residues and the burning of natural gas and coal complete the list of major contributors to atmospheric radon.

Figure 5. Map of the United States showing areas with potentially high radon levels in soil gas, based primarily on geological reports and modification of national uranium resource evaluation data. 19 XXX(From Cothern and Smith, 1987, p. 108)

It should be noted that indoor levels of radon are not related simply to geologic factors. The relation depends on many factors, including degree of fracturing of the bedrock, and on the intervening pathway. Radon mobility through soil may vary by up to l06 fold depending on soil porosity (30). Rock permeability is now recognized as a key factor influencing radon availability at the surface, even in low uranium containing rock types, such as limestone. (31)

Figure 6. Map of the United States showing the distribution of radon in ground water supplies in the U.S.A. by County. (19) XXX(From Cothern and Smith, p.118)

Another potentially important source of radon exposures is from radon outgassing from high levels in water. Radon concentrations in surface waters are usually very low. Since municipal water supplies are typically aerated, this results in diminished radon levels. Rural household wells are a potentially bigger problem. Deep aquifers have highly variable radon evels. Levels depend on uranium content of the rock and distribution of the aquifer relative to the rock, and on groundwater flow patterns. Thus, areas with granite-based aquifers may have highly variable levels, as noted in Table 3.


The major source of radon-222 in the atmosphere (at least 80%) is from emanations from soil from rock formations close to the ground surface (11) , from the decay of U-238 through Ra-226 to Rn-222(Table 4).

Radon dissolved in ground water is the second most important potential source of atmospheric radon. Nonetheless, in most locations it is a minor source of human exposure in view of the small absorbed dose following oral ingestion. In some locations where water from highly radioactive deep wells is used, it can be a significant contributor. Thus, in Maine, New Hampshire, some regions of the Appalachian mountains, and Florida, concentrations found in some private wells exceed 10,000 pCi/liter. When water use is high in the home, air levels are found to be elevated due to outgassing from the water. (32).

Turbulent or heated water (flowing in wash basins, showers, washing machines, flush toilets, etc.) is a source of elevated radon levels in the home, as these activities liberate dissolved radon into the home atmosphere. The amount so released depends on the radon content of the water (which varies widely between regions) and the amount used (70 to 250 gallons in a typical household per day). On the average 70% of radon contained in household water is released into indoor air (35).

The effect of human inhabitance on home radon levels is illustrated graphically in Fig. 7 which is a record of radon levels in a Houston apartment during a two day period. Radon concentrations in air increased three to five fold during times the apartment was occupied.

Figure 7. FIG. 7 Continuous radon monitor record from Houston apartment indicating influence of occupancy on radon concentration. (From Pritchard and Gesell, 1981, p. 602)

NCRP Commentary No. 6 discusses the main sources of indoor radon and gives specific geographic areas in the US where high levels exist (5) . The EPA stratified survey was conducted in 125 counties in all 50 states. An average level of 1.25 pCi/liter (46 Bq/m3.) was found, with 6% of the housing units exceeding the EPA action limit of 150 Bq/m3. or 4 pCi/liter (13).


Since radon is constantly escaping from the ground, it is always present in the air, but under certain circumstances the concentration of radon in a building can be increased significantly over its normal outdoor level. Most buildings have a confined air space with limited air movement and only a slow exchange with outside air. Consequently the concentration of any particulates or gases released into the building atmosphere will tend to increase above the concentration normally found in outside air. Radon can enter a building in a number of ways and once inside, the concentration of its particulate progeny will increase as the radon decays. Thus, high concentrations of radon in soil gas in soils with high transport efficiency (i.e. loose, porous, dry soil) can lead to elevated radon concentrations in buildings.

Soil is the major source of radon. Studies underway by the US Geologic Survey show that soil-gas radon levels vary widely in small areas (within a housing lot) and are not well correlated with the radium content of the soil. Pressure-driven flow is the major means of transport from soil into buildings because the pressure inside buildings is usually lower than that in the soil, especially in the winter. Houses with no barrier between the soil and the interior, e.g. with a dirt floor in the basement or crawl space, are especially vulnerable. Houses with porous foundations, e.g. concrete block or fieldstone, present only a minimal barrier to flow. Even houses with poured concrete basement floors and foundations usually have routes of entry for soil gas through joints, penetrations, cracks, sumps, and drains. Radon can enter a house from soil gas through ground level drainage systems, through flaws in a concrete floor slab, and through concrete block walls (33).

The water supply can be a route of entry if there is a significant amount of radon in the ground water, and if the home water supply is derived directly from deep wells (34). Differences in water use patterns, ventilation and air flow can cause significant temporal variations in radon levels indoors. However, soil gas radon content may be the greatest determinant of home levels. Table 5 shows the variations in contributions to radon in the home (35).

a) From reference 35 b) May be a factor of 10 to 100 times higher in certain regions (34)

Radon concentrations indoors will generally be highest in the basement or on the ground level since the major source is influx from the soil under and around the house. First floor concentrations will be lower by about a factor of two. Indoor radon concentrations are typically a factor of 2 to 3 higher than outdoor levels. The radon concentration in the upper levels and in apartments above the first floor are usually of no concern.

In addition to soil and water sources of indoor radon, home construction materials can be a significant contributor. Table 6 indicates the emission rate measured from various building materials. Clearly, the concrete used in building depending on its origin can be a major contributor, and in all cases, concrete is a more significant radon source than other building materials.

From reference 33. (a) The impervious (glassy) nature of these products retards radon release.

The number of homes with elevated radon levels varies in different regions of the country. The shape of the measured distribution is log normal. The distribution is highly skewed with most homes in the low dose region. Based on measurements in 552 homes from 19 studies conducted in regions without unusually high radon concentrations, average levels measured in single-family homes by Nero (36) were found to be 0.96-1.66 pCi/liter based on geometric and arithmetic mean calculations. In Nero's data, 2.5% of the houses were above 8 pCi/liter, which is the action level recommended by NCRP.

An EPA survey of 11,000 homes from 125 counties nationwide found the average annual radon concentration in US housing units is 1.25 +/- 0.06 pCi/liter, with a median value of 0.67 pCi/liter. They estimate that 6.01 +/- 0.58% of housing units (6 million homes) exceed the EPA action level of 4 pCi/liter. (37)

Figure 8. Percent of single-family homes with different values of radon-222 concentration. (19)

The percent of population exposed to different radon levels in the Reading Prong region vs. the U.S.A. average is tabulated below .

The distribution of population dose to residents in the Reading Prong region is given in Table 7. The levels in Pennsylvania are higher than in New Jersey which is significantly higher than in the USA per se. The difference is most notable in the portion of the population exposed at levels above 8 and 20 pCi/liter. It is in the high dose regions of the country that greatest attention to measurement and remediation needs to be focused.

On rare occasions radon levels have been found in houses which exceed those measured in uranium mines. This situation has occurred where the house is above a deep fissure in a granite shelf containing higher than normal levels of radium. The radon outgasses from the fissure through the soil under and around the home and enters the home. The high pressure under ground, relative to the pressure in the home forces radon into the building. This is especially true in the winter, when hot, low density air is vented from chimneys and other openings and is replaced by cold radon-containing air from the environment. When the ground adjacent to the house is frozen, its soil permeability is diminished. A high access channel for radon entry into the home is found under the house, where the ground is warmer.

When a house is located on a shelf of granite with a deep fault, the surface area from which Rn-222 escapes includes the depth of the fault with granite on both sides, as well as the ground surface. A similar situation was noted where a granite shelf lay below an empty large salt dome, through which the Rn migrated rapidly, and reached the surface with less time for radioactive decay. Clearly, such rare, very local circumstances justify quick remedial action. Public concern about radon has been raised and remedial action has been recommended at levels below which increased risk has been documented. It is not surprising that little in the way of remedial action has been taken at levels in the 4-8 pCi/liter interval. EPA is concerned, however, that testing even in high radon areas has not been widely carried out.

Note: This article is continued on the next page

Go to the top of the next page