NATION WIDE 222Rn AND 220Rn ATLAS FOR INDIA T.V. Ramachandan1, L.A. Sathish2* and S. Sundareshan3 1 Ex-Environmental Assessment Division, Bhabha Atomic Research Center, Mumbai- 450 085, India 2 Department of Physics, Government Science College, Bangalore – 560 001, India 3 Department of Physics, Vijaya College, Bangalore – 560 004, India *Corresponding Author (Sathish) Email: lasgayit@yahoo.com Abstract Considering the epidemiological effect of radon on human beings, an attempt is made to make a nation-wide atlas of 222 Rn and 220 Rn for India. More than 5000 measurements have been carried out in 1500 dwellings across the country, India. The solid state nuclear track detectors were deployed for the measurement of indoor 222Rn, 220Rn and their progeny levels. The mean annual inhalation dose rate due to 222Rn, 220 Rn and their progeny in the dwellings is found to be 0.97 mSv y1 (GSD 2.49). It is observed that the major contribution to the indoor inhalation dose is due to 222Rn and its progeny. However, the contribution due to 220Rn and its progeny is not trivial as it is about 20% of the total indoor inhalation dose rates. The dependence of indoor 222 Rn levels in dwellings has shown a significant difference between the nature of walls and floorings. The results are discussed in detail. Key words: 222Rn, 220Rn, inhalation, radiation doses. Corresponding Author: Dr.Sathish.L.A Assistant Professor Department of Physics Government Science College Nrupathunga Road, Bangalore – 560 001 India +91-80-9886639324 1 Introduction Ever since studies on uranium miners established the presence of a positive risk coefficient 222 for the occurrence of lung cancer in miners exposed to elevated levels of Rn and its progeny, there has been a great upsurge of interest in programmes concerned with the measurement of radon in the environment. This interest was accentuated by the observations of elevated radon levels in the indoor environment in many countries that led to the realization of residential radon as being a possible public health issue in the western world. It was also hoped that in conjunction with epidemiological studies, large-scale indoor surveys might lead to quantitative understanding of the low dose effects of 222 222 Rn Rn exposures. As a result of these, considerable amount of information is available on the levels of 222 Rn gas and its progeny in the indoor environment across the globe (UNSCEAR, 2000). In contrast, there exist a few studies relating to the measurements of 220 Rn in the environment (Doi and Kobayashi 1994; Doi et al, 1994) since it is assumed that the inhalation dose to the general population from 220Rn and its progeny is only about 10% of the inhalation dose due to 222 Rn (UNSCEAR, 2000). But, recent studies in many countries have revealed that this assumption if far from the truth (Steinhausler et al, 1994). In general, such studies are important in two ways. Firstly, any radiological impact assessment of nuclear facilities, either existing or those to be set up in the future, requires information on the exposure due to natural radiation prevalent in their vicinity. Secondly, the radiation risk coefficients are fairly well established at high doses and high dose rates, whereas little is known about the effects of radiation at low dose rates. Several epidemiological study programmes in different countries are in progress to estimate the population exposures due to natural radiation with a view to obtain the radiation risk coefficients at low dose rate levels. In this regard, radiation surveys in high background areas will provide an excellent setting for epidemiological studies relating to the effects of low doses of radiation. In view of these, a comprehensive estimate of the natural inhalation dose requires both 222 Rn and 220 Rn levels in the indoor and outdoor atmosphere. Sources of 222Rn and 220Rn Radionuclides such as 222Rn and 220Rn, from the uranium and thorium decay chains are noble gases produced by the decay of their immediate respective parent nuclides, 226 Ra and 224 Ra, present in natural rocks, uranium ores and soils (Fleischer, 1997). The decay products of 222 Rn and 220 Rn are the radioactive isotopes of polonium, bismuth, lead and thallium. decay products are divided into two groups; the short-lived 222 Rn daughters such as 222 Rn 218 Po 2 (RaA), 222 214 Pb (RaB), 214 214 Po (RaC1) with half-lives below 30 min, and long-lived Bi (RaC), 210 Rn decay products such as Pb (RaD), Bi (RaE), 210 Po (RaF). However, 220 Rn progeny has no long-lived group. Most important radionuclide in this chain is the lead isotope 212 Pb with a half-life of 10.6 h. These daughter products, being the isotopes of heavy metals, get attached to the existing aerosols, suspended particulate matters, in the atmosphere. Their elimination from the atmosphere occurs either by radioactive decay or by other removal processes such as plate-out or surface deposition and washout by rain. Vast differences in the half-lives of 222 Rn (3.8 d) and 220 Rn (55 s) is a crucial parameter in governing their release from the ground and subsequent distribution in the free atmosphere. When radium decays in soil grains, the resulting atoms of filled pores. The fraction of 222 222 Rn isotopes first escape from the mineral grains to air- Rn escapes into the pores is known as the emanation power fraction. Even though the detailed processes responsible for 222 Rn emanation from grains are not fully understood, it is believed that the main contribution to the emanation comes from the recoil processes (Nazaroff, 1988). The recoil range is about 0.04 - 0.06 µm in grain materials and about 60 µm in air (Tanner, 1980). Also, recoil-stopping distance of 220 222 Rn and Rn is lower in water than in air. Hence, the moisture content influences the emanation power fraction (Megumi and Mamuro, 1974; Strong and Levins, 1982; Ingersall, 1983; Stranden et al, 1984). Emanation power fraction of building materials for 220Rn is about 2-10 times smaller than that for 222 Rn, despite the greater recoil energy of 220 Rn atoms (Porstendorfer, 1994). Experimental studies on building show that it ranges from 0.2 to 30% for 222 222 Rn and 0.2 to 6% for 220 Rn (Porstendorfer, 1994; Barretto et al, 1972). Transport of Rn through the soil takes place by diffusion and/or with gases like CO2 and CH4 or water moving in the soil horizons. The diffusion coefficient for 222 Rn in different soil types varies from 10-9 to 10-5 m2 s-1 from water to air media (UNSCEAR, 1992). 222Rn and 220Rn enter the atmosphere mainly by crossing the soil-air or building material-air interface. Typical values of exhalation rate (amount of activity released per unit area of the surface per unit time) for 222 Rn in soil and building material are 0.02 and 5.0 × 10-4 Bqm2s-1, respectively. The same for 220 Rn are as high as 1 and 0.05 Bqm-2sl, respectively (Porstendorfer, 1994). 222 Rn and 220 Rn progeny aerosols in the atmosphere are generated in two steps. After the formation from the 222 Rn isotope by decay, the freshly generated radionuclides react very fast with trace gases and air vapors, and become small particles, called clusters or unattached radionuclides with diameters varying from 0.5 to 5 nm. In addition, these radionuclides attach to the existing aerosol particles in the atmosphere within 1 - 100s, forming the radioactive aerosols. Most of the newly formed decay product clusters are positively charged and have a high mobility 3 (Porstendorfer and Mercer, 1979). Mobility is characterized by the diffusion coefficient that mainly controls the formation of the radioactive aerosol by attachment and their deposition on surfaces and in the human lung. 222 Rn and 220 Rn in indoor environments mainly originate from emanation of the gases from the walls, floor and ceilings. Most terrestrial building materials have 3-4 orders of magnitude higher gas concentrations in pore spaces than in the atmosphere, permanently maintained by the continuous decay of its parent nuclides. High concentration leads to a large gradient between the materials and open air. Levels of 222 Rn and 220 222 Rn /220Rn Rn in the open atmosphere are governed by the balance between the exhalation rate and the atmospheric dilution processes. The external gamma dose rates have been more or less well mapped in India by several studies. A countrywide survey of outdoor natural gamma radiation levels using Thermo Luminescent Dosimeters (TLD) covering quite large number of locations scattered all over the country revealed that the average external gamma radiation dose for the country is about 775 µGy yr-1 (Nambi et al, 1986). Mishra and Sadasivan (1971) have projected a national average value of 707 µGy yr-1 based on natural radioactivity analysis of undisturbed soil samples from more than 30 different locations, all over the country, assuming a uniform cosmic ray component of 287 µGy yr-1. Of the terrestrial component, 48.7% of the contribution is from 40 K and the remainder is by the thorium (33.6%) and uranium series (17.7%) (Sadasivan et al, 2003). Tables 1 and 2 give the estimated natural radioactivity content in the building materials used for construction in India and the distribution of and 232 Th in Indian soil (Sadasivan et al, 2003). It can be seen from these tables that 238 40 U K is also a major source of radiation in the environment. A good database on the countrywide concentration levels of 238 U, 232 Th and 40 K in geological materials as shown in Table 3 (Sankaran et al, 1986). Table 4 gives the estimated ranges of 222 Rn entry rate from different sources in typical houses (ICRP, 1986). It is evident that soil has the highest entry rates followed by brick or concrete. Environmental measurements of earlier. Since 1970, indoor 222 222 Rn were mostly confined to outdoor atmospheric air Rn levels were measured with keen interest, and several large- scale surveys have been carried out by several agencies all over the world (Campos-Venuti et al, 1994; UNSCEAR, 2000). Typical worldwide indoor and outdoor levels of 222Rn are about 45 and 7 Bq m-3, respectively and that of outdoor 220 Rn level is estimated as 0.2 Bq m-3 4 (Mettler and Upton, 1995). An initial survey in Indian houses indicates that the indoor 222 Rn concentration varied between 2.2 to 56 Bq m-3 with a geometric mean of 15.1 Bq m-3 (Subba Ramu et al, 1993). The reported indoor 222Rn and 220Rn levels are tabulated in Tables 5 and 6 respectively, shows that the population weighted worldwide average 222 Rn concentration is 39 Bq m-3; while the geometric mean calculated for the data is 30 Bq m-3 with a geometric standard deviation of 2.3 (UNSCEAR, 2000). Average equilibrium equivalent concentration of 220 Rn (Table 6) range between 0.2 and 12 Bq m-3, while the ratio of 222 Rn/220Rn EEC varied from 0 . 0 1 to 0.5 worldwide. All this information facilitated the understanding of many environmental processes, which affect the distribution of 222 Rn and 220 Rn levels in indoors and outdoors and the related radiation exposure to man. However, there exist still many problems associated with the accurate assessment of exposures and radiation doses to general population due to 222Rn, 220Rn and their progeny. Measurement Methodology The present national survey covered 25 locations. About 1500 houses of different types of construction were surveyed on a time integrated quarterly cycle of 90 days covering all the four seasons of a calendar year. Solid State Nuclear Track Detector (SSNTD) based dosimeters (Nikolaeve and Ilic, 1999; Subba Ramu et al, 1994) were used for the survey. These are simple to use and less expensive as compared to some continuous measurement systems like the AlphaGuard. The latter is useful for occasional comparisons with the SSNTD based dosimeters. In view of this, SSNTD based dosimeters, described in the following section, were developed and calibrated for the national survey. Since the sampling is passive and integrated for long duration, the diurnal and seasonal variations in radon concentrations are being taken into account (Ilic and Suteg, 1997). SSNTD based dosimeter System developed is a cylindrical plastic chamber divided into two equal compartments (Nambi et al, 1994), each having an inner volume of 135 cm3 and height 4.5 cm. Dimensions of the dosimeter are chosen based on the ratio of the effective volume of the cup to its total volume to achieve maximum track registration for the cylindrical cup (Jha et al, 1982). The design of the dosimeter is well suited to discriminate 222 Rn and 220 Rn in mixed field situations, where both the gases are present as in the monazite deposited areas. Cellulose nitrate films of LR-115 type II manufactured by the Kodak Pathe are used as detectors. The 12 µm thick film cut into 2.5 cm × 2.5 cm size is affixed at the bottom of each cup as well as on the outer surface of the dosimeter. The exposure of the detector inside the 5 cup is termed as cup mode and the one exposed open is termed as the bare mode. One of the cups has its entry covered with a glass fiber filter paper that permeates both 222Rn and 220Rn gases into the cup and is called the filter cup. The other cup is covered with a semi-permeable membrane (Ward et al, 1977) sandwiched between two-glass fiber filter papers and is called the membrane cup. This membrane has permeability constant in the range of 10-8 -10-7 cm2s-1 (Wafaa, 2002) and allow more than 95 % of the entry of 220 Rn and Rn gas to diffuse while it suppress the Rn gas almost completely. Thus, the SSNTD film inside the membrane cup registers tracks contributed by 222 222 220 222 Rn only, while that in the filter cup records tracks due to Rn. The third SSNTD film exposed in the bare mode registers alpha tracks contributed by the concentrations of both the gases and their alpha emitting progeny. The dosimeter is kept at a height of 1.5 m from the ground and care is taken to keep the bare card at least 10 cm away from any surface. This ensures that errors due to tracks from deposited activity from nearby surfaces are avoided, since the ranges of alpha particles from 222 Rn /220Rn progeny fall within 10 cm distance. After the exposure period of 90 days, the SSNTD films are retrieved and chemically etched in 2.5 N NaOH solutions at 60 °C for 60 minutes with mild agitation throughout (Miles, 1997). The tracks recorded in all the three SSNTD films are counted using a spark counter. A methodology has been developed to derive the equilibrium factors separately for 222 Rn and 220 Rn using the track densities based on the ventilation rates in the dwellings (Mayya et al, 1998).One may expect deposition of activity on the SSNTD film in the bare mode exposure, which may pose as an unknown parameter in the calibration factor. But it has been proved that the LR-115 (12 µm) film does not register tracks from deposited activity (Eappen et al, 2004). This is because the Emax for LR-115 film is 4 MeV and all the progeny isotopes of 222Rn /220Rn emit alphas with energies greater than 5 MeV. Calibration Facility and Standardization of Dosimeter Experiments were carried out at the Bhabha Atomic Research Centre, Mumbai, India to estimate the calibration factors (Ramachandran et al, 1995) separately for 222Rn and 220Rn, in a calibration chamber of stainless steel of 0.5 m3 volume. 222 Rn (or 220 Rn) gas is introduced into the chamber from standard sources obtained from Pylon, Canada. The calibration chamber has provisions for imputing aerosols from an aerosol generator, which is a Sinclair LaMer type condensation aerosol generator. It gives a laminar flow of mono-dispersed aerosols of di-2-ethylhexyl sebacate condensed on NaCl nuclei. The temperature settings of the boiler and re-heater are adjusted to obtain mono-dispersed aerosols of 0.25 µm diameter, 6 which is close to the activity median aerodynamic diameter of 0.2 |µm reported for indoor aerosols (Yihe et al, 1996). Aerosol concentrations of the order of 104 to I05 particles per cm3 of air were generated to simulate the indoor environment conditions. Depletion of the aerosols inside the chamber was studied and accordingly input of the aerosols was regulated to maintain a near constant particle concentrations. The chamber has provisions for coupling an on-line Lucas cell system in conjunction with an AlphaGuard for continuous measurement of 222 Rn gas concentration. The AlphaGuard, kept inside the chamber recorded hourly averaged 222 Rn concentrations. The on-line Lucas cell system was coupled to an alpha counting setup and counts were taken synchronizing with the timing of the AlphaGuard. The comparison of 222 Rn measured by the two systems for a wide range of concentrations showed very good correlation of regression coefficient 0.97 and has a slope equal to unity (Eappen et al, 2001). Calibration factors (concentration conversion factors) for 222 Rn and 220 Rn are required to convert the recorded tracks in the exposed SSNTD films into 222Rn and 220 Rn concentrations. Calibration factors were estimated experimentally as well as theoretically for all the three modes of exposures. These are discussed in the following sections. Calibration factors (CFs) for 222Rn and 220Rn gases in the cup mode were determined through a series of experiments. CFs for 222Rn (kR) and for 220Rn (kT) in terms of tr cm-2 per Bq d m-3 can be obtained as: kR = 24T CR H and kT = 24T CT H where, T is the tracks per unit area (tr cm-2), CR is concentration of the CT is the level of 220 222 Rn gas (Bq m-3), Rn gas (Bq m-3) and H is the exposure time (hours). Experimentally obtained calibration factors for 222Rn and 220Rn are given in Table 7 for cup mode exposure. CF for 222Rn in the membrane compartment is found to be equal (0.019 tr cm-2 / Bq d m-3) to that in filter paper compartment (0.02 tr cm-2/Bq d m-3). CF for 220Rn in the membrane cup is essentially zero and that in the filter paper cup is 0.017 tr cm-2/Bq d m-3. The definition of the CF for the bare mode has certain ambiguities. In the earlier approach, the CF for the bare detector was defined as the track density rate obtained per unit WL (Barillion and Chambraudet, 2000; Durrani and Ilic, 1997). In reality, track formation rate in the bare mode is not a unique function of WL, but would depend on the equilibrium factor (F). If one defines the bare detector calibration factor as kB (tr cm-2/Bq d m-3) of each species, it may be easy to show that this quantity is independent of the equilibrium factor as well as the 7 incident energy of the alpha particle. For a given track density rate 7(tr cm-2 d-1) and working level (WR for 222 Rn and WT for 220 Rn in mWL units) and the corresponding equilibrium factors, FR and FT, the calibration factors as defined above can be obtained for and 220 -2 222 Rn (kBR) -3 Rn (kBT) respectively in terms of tr cm / Bq d m using the following equations. ! T "! FR " k BR = # $# $ % 3.7WR &% 1+2FR & T ! "! FT " k BT = # $# $ % 0.275Wr &% 2 + Fr & Based on this concept CFs was derived for the species matrix for 222 Rn, 220 Rn and their progeny concentrations. They were found to be nearly constant for a wide range of equilibrium factors (0.1 - 0.72) supporting the basic assumption of the new approach. Table 7 shows the results of the CFs for the bare mode exposure for 222 222 Rn and 220 Rn. The CF for Rn and 220Rn are estimated as 0.02 tr cm-2/Bq d m-3and 0.019 tr cm-2/Bq d m-3, respectively and are nearly identical. This confirms the assumption that the bare card calibration factors are the same for the alpha emitters since they are functions of only the difference in the ranges and the lower and upper cut off energies of the detector. Hence for practical use, an average value of 0.02 tr cm-2/Bq d m-3may be used as the CF for 222 Rn and 220Rn in the bare mode exposure. A Theoretical model has been developed to derive the calibration factors for 222 Rn and 220Rn for all the exposure modes (Eappen and Mayya, 2004). The theoretical model is based on certain parametric constants chosen after experimental verifications. These include the bulk-etching rate and the break down thickness for the spark counting technique. The present calculation uses bulk etching rate as 4.0 µm/h and break down thickness as 3.0 µm. In the model, the upper and lower cut off energies for normal incident alphas are translated as residual ranges using the range energy relationship. The sphere of influence for the upper and lower cut off energies from normal incident angle to critical angle can be obtained from integrating for the total area covered under solid angle for residual length of alpha particles lying within those incident angles. With these considerations, the observable tracks per unit area on the film per unit exposure time can be computed using the following equation. 2$ # R "R c F U &c Tr = d % d # sin # cos#dr 4$ !0 # != 0 r = RE "!RL (# ) 8 where η is the efficiency of track registration, C is the activity concentration of the species,φ is the solid angle suspending the area of influence, θ is the angle of incidence ranging from normal incidence (0°) to critical angle (θC), r is the radial distance from the point of emission, R E is the range of the alpha particle corresponding to its max energy and RL, R U are the lower and upper cut off ranges for track registration for an incident angle θ The integration extends over a region of influence, which is constructed by using detailed track development model. Eappen et al (2004) have discussed the typical regions of influence for 220 Rn and Rn and their progenies in bare mode exposure configuration. Authors have showed that the region of influence is located farther from the detector for 222 222 220 Rn progeny as compared to Rn and its progeny concentrations. For the cup mode exposure, integrations over the regions of influence would also include surface deposited activity contributions from the inner walls of the dosimeter. A code has been written in FORTRAN for calculating the calibration factors in different configurations using the theoretical model (Eappen et al, 2001). Several experimental studies were carried out in the calibration facility to determine the calibration factors under various equilibrium factor and gas concentration conditions. Theoretical and the experimental CFs obtained for the cup mode and bare mode exposures show close agreement with each other. Dosimetric Methodology Inter-laboratory standardization experiments for the etching characteristics conducted by all the participants using standard alpha source also showed good agreements. A theoretical methodology has been developed for evaluating the progeny concentrations using the twin cup 222 Rn - 220 Rn dosimeter system (Mayya et al, 1998). The mathematical basis used is similar to that developed by Planinic and Faj (1990, 1991) for radon dosimetry in which an auxiliary parameter, ventilation rate, was extracted from the equations relating the bare detector track densities to the gas and progeny levels. This approach is considered as most logical one for 222 Rn - 220 Rn dosimetry with bare and cup detector system. But this methodology is complicated in the mixed field situation by the fact that 220 Rn contribution has to be given as its ventilation dependant spatial profile for which only limited information is available in literature. So the data currently available in the literature are used for the parameters like wall loss rates, unattached fractions and indoor turbulence levels (Porstendorfer, 1994). In this method, it is assumed that SSNTD kept in the bare mode responds only to the airborne alpha emitters and not to the alpha activity deposited on it. It is also assumed that the bare card calibration factors are same for alpha emitters since it is a 9 function of only the difference in the ranges, lower and upper cut off energy of the detector. Let T1, T2 and T3 be the track densities recorded in the membrane mode, filter mode and bare mode, respectively. Let and kR be the calibration factors for 222 Rn gas in membrane compartment and filter compartment, respectively and kT be the calibration factor for 220Rn in the filter compartment. If d is the duration of exposure (days), the gas concentrations of 222Rn (Bq m-3) and 220 Rn (Bq m-3) the vicinity of the dosimeter can be determined from the observed track densities T1 and T2 using the following equations: CR = Since the 222 T1 dk R and CT = T2 ! dCR k R dkT Rn decay constant is far smaller than the usually encountered air change rates (ventilation rates), 220Rn may be assumed to be spatially uniform. The activity fractions of the progeny are governed by their wall loss rates for the fine and the coarse fractions and the ventilation rates. The bare track densities are also dependent on the ventilation rates, which represent the progeny fractions for both gases. However unlike 222Rn, 220 Rn is not uniformly distributed in the room due to its short half-life, but is expected to set up profiles (Doi and Kobayashi, 1994). The concentration CT would be considerably lower than that present near the ground and the walls, which are the 220Rn emitting surfaces. On the other hand, the thoron decay products, 212 Pb and 212Bi, being longer lived would mix more or less uniformly in the room and their activities will be fractions of a representative average 220 Rn concentration. A turbulent-diffusive transport model developed by Mayya et al, (1998) was used to obtain the bare track densities in terms of this concentration and the indoor ventilation rates. This method, which is known as the root finding method (RFM), is theoretically the most satisfactory approach for determining 222Rn, 220 Rn concentrations and their progeny working levels using the tracks recorded on the three SSNTD films. The progeny working levels were evaluated using the following relations: WLR = CR FR CR (0.104 FRA + 0.518 FRB + 0.37 FRC ) = 3700 3700 WLT = CT FT 275 = CT (0.908 FTB + 0.092 FTC ) 275 where FR and FT are the equilibrium factors for 222Rn and 220Rn progeny, respectively, which are related to the ventilation rate. However, in practice, it was found that small uncertainties in the recorded tracks propagate non-linearly leading occasionally to unacceptable solutions for the equilibrium factors. Very rich experience in measurements is required to eliminate 10 these uncertainties, which is expected to be realized in the coming few years. Until then, it was decided to estimate the progeny concentrations using the cup based gas concentrations and the universally accepted equilibrium factors published elsewhere (UNSCEAR, 2000). Information obtained from the third SSNTD is being used in conjunction with the RFM for building a database on the equilibrium factors. At present, the effective dose rate due to inhalation was estimated from the 222 Rn, 220 Rn and progeny concentrations using the UNSCEAR (2000) equilibrium factors as given in Table 8. Inhalation Dose Absorbed dose rates to the critical cells of the respiratory tract due to 222 Rn, 220 Rn and their progeny can be estimated on the basis of aerosol characteristics, its size distribution, unattached fraction, breathing fraction, and fractional deposition in the airways, mucous clearance rate and location of the target cells in the airways. Several models have been developed to assess the inhalation dose rates to the population due to 222 Rn, 220 Rn and their progeny (Jacobi, 1993; Subba Ramu, 1988). Lung dose distribution assessment carried out by different agencies from the year 1956 to 2000 show a large variation in dose conversion factors (UNSCEAR, 1993, 2000). The estimated dose conversion factors varied drastically based on the breathing rate as well as the target tissue mass. In the present study, the dose conversion factors reported by UNSCEAR (2000) have been used to estimate the indoor inhalation dose rates D (µSvh-1) due to 222Rn, 220Rn and their progeny as shown below: D = 10!3[(0.17 + 9 FR )CR + (0.11 + 40 FT )CT ] Numerical values given in the above relations are the dose conversion factors for gas and progeny concentrations. Results and Discussion Present survey covers 25 locations in different parts of the country. This database alone was not sufficient for obtaining a comprehensive mean value of the indoor 222Rn and 220Rn levels on a national scale. Hence, similar data generated and published by this centre as well as published by several groups elsewhere have also been used for the purpose. This data includes mainly the indoor 222 Rn levels and the equilibrium factors estimated earlier survey using single cup dosimeter covering more than 90 locations (UNSCEAR, 1993) and the data generated from the survey carried out around 12 nuclear installations in India using the twin chamber 222 Rn /220Rn dosimeters (Ramachandran et al, 1995). In the case of 220 Rn, the data 11 generated from 25 locations under this study and the data generated from the survey carried out around nuclear installations in India were used. Indoor 222Rn and 220Rn Level Estimated levels of indoor 222Rn and the equilibrium factors between 222Rn and its progeny in 105 houses of different types of construction at 84 locations in different parts of India by the single cup method are given in Table 9. The estimated 222Rn level at different locations varies from 6.4 Bqm-3 to 95.4 Bq m-3 with a geometric mean of 25.5 Bq m-3 (GSD 2.1). Equilibrium factors were estimated using the bare detector exposure mode along with the cup with membrane mode in these locations for 222 Rn progeny. From the calibration factors for the bare detector, the progeny concentrations are evaluated and the equilibrium factors were estimated using the standard equation. Equilibrium factors for 222 Rn progeny range between 0.21 and 0.95 (UNSCEAR, 1993) with a geometric mean of 0.54 (GSD 1.4). Estimated mean equilibrium factors range between 0.1 and 0.9, but most of the values are found to be within 30% of the typical value of 0.4 used by the UNSCEAR (1993) for inhalation dose calculations. Values thus computed using the standard relation is not strictly correct, since the bare detector exposure is not a function of WL, but depends on the F factor. A theoretical methodology has been developed incorporating this fact to extract the modified F values. Using this concept, the revised F values were evaluated and these values are found to vary from 0.12 to 1.2 with a median of 0.46 ± 0.2. Although, the median value of F is found to decrease in the revised estimates, the spread is found to be higher. Besides, the distribution is found to be skewed to the left, unlike the near symmetrical form shown by the pre-revised data in Fig.1. Mathematical analysis of the F distribution shows that the F values correspond to a mean ventilation rate of 2 per hour with a GSD of 3 (UNSCEAR, 1988). Results on the indoor 222Rn, 220Rn levels and the estimated inhalation dose rates are presented in Table 10. The geometric mean 222Rn levels at different locations range between 4.6 Bq m-3 and 147.3 Bq m3. The estimated geometric means of indoor 220 Rn levels at these locations range between 3.5 and 42.8 Bq m-3. Fig. 2 shows the lognormal distribution of indoor 222 Rn -3 levels at different locations in India, which gives a geometric mean of 23.0 Bq m (GSD 2.61). The lognormal distribution pattern of indoor 220 Rn levels is shown in Fig. 3 with the geometric mean of 12.2 Bq m-3 (GSD 3.22). In view of the large number of measurements carried out, the distributions pattern estimated can confidently be projected as national representations of indoor 222 Rn and 220 Rn levels in India. The relationship between indoor 12 222 Rn and 220Rn levels is indicated in Fig.4, which shows a good correlation between the two quantities. The relationship indicates that, in general, the indoor thoron concentration is about 50% of that due to indoor 222 Rn concentration, which is not trivial as considered earlier. All the data from the present study as well as other relevant data mentioned in this report have been used for preparing the maps of indoor 222 Rn and level. Fig. 5 and 6 illustrate these maps of indoor 222 220 Rn concentrations on a national Rn and 220 Rn levels respectively to represent the different concentration levels. 222 Rn level dependencies on different types of dwellings The variation of indoor 222Rn levels in various types of dwellings is examined using the data and the results are presented in Table 11. A scrutiny of this table reveals that the 222Rn levels are higher in houses constructed with plastered whitewashed walls and mosaic floors. Houses having wooden walls show lowest 222Rn levels. It can be noticed that irrespective of the type of walls, houses constructed with tile flooring show lower 222Rn levels. An analysis has been carried out to evaluate the statistical significance (95% confidence limits) of the difference in means of the indoor 222 Rn levels among different dwelling types and the results are given in Table 12. This is based on the assumption that the 222 Rn levels follow a normal distribution and that there is not much variation in the ventilation rates in dwellings. Although this analysis does not include all the geographical factors that govern the pattern of radon levels, it provides a general representation for the variation of indoor dwellings. Table 12 shows that the differences in the 222 222 Rn levels in Indian Rn levels among different types of floors are small or insignificant when the walls are plastered and painted. This shows that most of the 222Rn emanates from the walls and painted walls will reduce the 222Rn emanation. When the walls are of plastered and whitewash type, there are significant differences between mosaic and any other floor types. Also, whenever mosaic floors are used, the differences are significant between different types of walls. Hence this analysis shows that the combination of whitewash walls and mosaic floors may lead to higher levels of indoor the reason for this high 222 222 Rn. However, Rn levels in dwellings having this combination is not obvious from the present set of data. More detailed investigation and categorization are needed in this respect. Estimation of Ventilation Rates Several methods such as tracer gas techniques and SSNTD based techniques are being used to estimate the ventilation rates in dwellings. Usual method of determining the ventilation rate in a room involves the measurement of the rate of loss of a tracer gas from the room. 13 Various tracer gases like CO2, nitrous oxides and 85Kr are being used for these measurements. The diurnal variations of indoor radon levels also can be used to estimate the ventilation rate in rooms (Ramachandran, 2001; Shaikh et al, 1992). 222 Rn and its short-lived progenies, which are naturally present in air, are also being used as a tracer. In the SSNTD based techniques, 222 Rn gas and progeny concentrations are estimated in rooms using SSNTD dosimeters in membrane and bare modes of exposure on a time integrated scale. For steadystate 222 222 Rn and its progeny levels, the ratio of the working level (progeny concentration) to Rn gas concentration (Bq m-3) is evaluated. This ratio is related to the pseudo-ventilation rate and plate-out rate. Actual ventilation rate is obtained by subtracting the plate-out rates of attached and unattached fractions of 222 Rn daughters from the pseudo-ventilation rates. The ventilation rates estimated by earlier investigations (Shaikh et al, 1992) in Indian dwellings using this method are given in Table 13. The measured ventilation rates varied between 0.42 and 4.46 h1 with a mean of 2.08 h-1 (standard deviation of 49%). With respect to the type of dwellings, the ventilation rates varied from 0.42 to 2.82 h-1 in Chawls and from 0.52 to 4.46 h-1 in flats. This wide variation is acceptable due to differences in construction and atmospheric conditions. Ventilation rates in some other countries like UK and USA range from 0.93 to 2.89 h-1 and 0.03 to 1.16 h-1 respectively (Nero et al, 1983; Israeli, 1985). Being in the temperate region, Indian dwellings are expected to have higher ventilation rates compared to dwellings in cold regions. Inhalation Dose Rates The 222Rn, 220 Rn and their progeny concentrations are converted into inhalation dose rates to residents using the above equation and the results are presented in Table 10. This table includes contributions from 222 Rn and progeny as well as 220 Rn and progeny. The total estimated inhalation dose rates vary from 0.27 m Sv y-1 at Kalpakkam to 5.14 m Sv y-1 at Digboi with a geometric mean value of 0.97 m Sv y-1 (GSD 2.49). Inhalation dose rates due to 222 Rn and its progeny show a geometric mean value of 0.63 m Sv y-1 (GSD 1.52), while that due to 220Rn and its progeny show a geometric mean value of 034 m Sv y-1 (GSD 1.44). It can be seen from this table that the dose due to 220 Rn and progeny is about half of that due to 222Rn and progeny. This fact is illustrated in Fig.7, which shows good correlation between the total indoor inhalation dose and that due to inhalation dose rate due to 220 220 Rn and its progeny. Contribution of Rn and its progeny is seen to be nearly 17% of the total inhalation dose rate. Nambi et al, (1986) estimated the average external gamma radiation 14 dose rate in India as 0.80 m Sv y-1 based on TLD measurements. These data suggest that in normal background areas, the inhalation dose rates predominate over the external gamma dose rates. The distribution pattern of indoor inhalation dose rates is depicted in Fig.8, which is a lognormal distribution. The majority of measurements indicate that indoor inhalation dose rates range between 0.1 and 2.5 mSv y-1. The geographical variation of indoor inhalation dose rates is also of considerable interest. This information can be used to delineate the normal and high background radiation areas. Though the present survey data is not sufficient for such an exercise, an effort has been made to study the geographical variation of the indoor inhalation dose rates in India is depicted in Fig.9. This figure shows that about 11 locations record dose rates above 1 mSv y-1 and most of these locations lie in the northeastern part of the country. Remedial Action Levels Elevated levels of indoor 222 Rn may be encountered in work places other than uranium or non-uranium mines as well. An issue of concern today is to prescribe action levels (in terms of average indoor levels) above which intervention would be desirable to reduce the levels of human exposure. Action level is defined as the level of dose rate or activity concentration above which remedial actions or protective actions should be carried out in chronic exposure or emergency exposure situation. Choice of the action level is complex depending not only on the level of exposure but also on the likely scale of action, which has economic implications for the community and for the individuals (IAEA, 1994; ICRP, 1991). ICRP (1993) made a distinction between the existing exposure situations, where any action would have to be remedial, and future situations, which can be subjected to limitation and control at the stage of decision and planning. In this connection, it is pointed out that the distribution pattern of indoor 222 Rn follows a lognormal distribution, which means that there would be a very small fraction of the total that would have large values. The geometric mean and geometric standard deviation are appropriate for characterizing this type of distribution. Knowing the geometric mean and geometric standard deviation, it is possible to predict what fraction of the total population would exceed a given value of the parameter. ICRP (1993) has recommended that there is considerable merit in the definition of radon-prone areas so as to focus attention where it is most exigent and on action where it is most effective. A 222 prone area may be defined as the one in which about 1% of the buildings has 222 Rn Rn concentrations above 200 Bq m-3. The recommended action level is 200 Bq m-3 for such a building, which would correspond to an annual effective dose of 5 mSv. On the other hand, 15 UNSCEAR (1993) recommends an action level of 400 Bq m-3.The international recommendations for 222 Rn action levels are given in Table 14 (Sohrabi, 1997). The results presented here, show that the indoor 222 Rn levels in India are far below the action levels. Hence, it is clearly demonstrated that most of the dwellings in India do not warrant any action level with respect to indoor concentrations levels for 222 Rn levels. As per the new WHO recommendations the 222 Rn and 220 Rn are 200 and 100 Bq m-3, respectively. But, the study raises some concern about the high inhalation dose rates observed at the northeastern parts of the country. Conclusions A countrywide survey on 222 Rn and 220 Rn levels for India has been carried out in dwellings using Solid State Nuclear Track Detector based passive detector technique. A good database on the total external radiation across the country is supplemented with the inhalation component, which is mainly contributed by 222 Rn and 220 Rn and their progeny. Calibration factors for the measurements have been derived experimentally as well as theoretically. The results show that the 222 Rn gas concentrations at different locations vary between 4.6 and 147.3 Bq m-3with an overall geometric mean of 23.0 Bq m-3 (GSD 2.61). 220 Rn gas concentrations are found to be less than the 222Rn gas concentrations at these locations (3.5 to 42.8 Bq m-3) with an overall geometric mean concentration of 12.2 Bq m-3 (GSD 3.22). The inhalation dose rates due to 222 Rn, 220 Rn and their progeny ranged from 0.27 m Sv yr-1 at Kalpakkam to 5.14 m Sv yr-1 at Digboi with a geometric mean value of 0.97 m Sv yr-1 (GSD 2.49). In general, the indoor 220 Rn and progeny concentrations and corresponding inhalation dose rates are found to be about half of that due to 222 Rn and its progeny. The geographical distribution pattern shows comparatively high inhalation dose rates (> 2.0 m Sv yr-1) in the northeastern part of India, which is supported by observations of high concentration of uranium, and thorium in soil and rocks in this region. The study also reveals that most of the dwellings in India do not demand any action levels with respect to indoor 222Rn and 220Rn due to good ventilation prevailing in Indian dwellings. However, it raises some concern about the high inhalation dose rates observed in the northeastern part of the country. 16 References Barillion R and Chambraudet (2000) Alpha particle dosimetry using solid state nuclear track detectors: Application to radon and its daughters, J Radio-anal and Nucl. Chem, 243, 607 620. Barretto PMC, Clark, Ronald B, and Adams JAS (1972) Physical characteristics of radon-222 emanation from rocks, soils and minerals: its relation to temperature and alpha dose, Natural Radiation Environment II, Conf-720805-P2, USERDA, New York, 731 -740. Campos-Venuti G, Janssens A and Olast M (1994) Indoor Radon Remedial Action, The Scientific Basis and the Practical Implications, Proceedings of the First International Workshop, Rimini, Italy, Rad. Prot Dos., 56(1-4): 133-135. discriminating dosimeter, Radiation Measurements 38, 5 – 17 Doi M and Kobayashi, S (1994) The passive radon-thoron discriminative dosimeter for practical use, Hoken Butsuri, 29, 155 - 166. Doi M, Fugimoto K, Koyabshi S and Yenhara, H (1994) Spatial distribution of thoron and radon concentrations in the indoor air of a traditional Japanese wooden house, Health. Phys, 66, 43 - 49. Durrani, S A, and Radomir Ilic (1997) Radon Measurements by Etched Track Detectors: Applications in Radiation Protection, Earth Sciences and the Environment, World Scientific, Singapore, 77 - 102. Eappen KP and Mayya YS (2004) Calibration factors for LR-115 (type-II) based radon thoron Eappen KP, Ramachandran TV, Shaikh AN and Mayya YS (2001) Calibration factors for SSNTD based radon / thoron dosimeters, Rad. Prot. and Environment. 410 - 414. Eappen, KP, Ramachandran, TV, Shaikh, AN and Mayya, YS (2001) Calibration factors for SSNTD based radon / thoron dosimeters, Rad. Prot. and Environment, 410 - 414 Fleischer RL, (1997) Radon: Overview of properties, origin and transport, In: Radon Measurements by Etched Track Detectors, Applications in Radiation Protection, earth Sciences and the Environment, (Eds.Durrani, S. A., and Ilic, R), World scientific, Singapore, 1 -20. Ilic R, and Suteg T (1997) Radon Monitoring devices based on Etched Track Detectors, In: Radon Measurements by Etched Track Detectors, Applications in Radiation Protection, earth Sciences and the Environment, (Eds,, Durrani, S. A., and Ilic, R), World Scientific, Singapore, 103 - 128. Ingersall JG (1983) A survey of radio nuclide constants and radon emanation rates in building materials used in the US., Health Phys, 45, 363-368. 17 International Atomic Energy Agency (1994) International Basic Safety Standards for Protection against Ionizing Radiation and for the Safety of Radiation Sources, Safety Series No. 115 - 1, IAEA, Vienna International Commission for Radiation Protection (1986) Publication No. 50, Lung Cancer Risk from Indoor Exposures to Radon Daughters, Annals of ICRP 16, Pergamon Press, Oxford. International Commission for Radiation Protection(1993) Publication No. 65, Annals of ICRP 23, Protection Against Rn222 at Home and at Work., Pergamon Press, Oxford. International Commission for Radiation Protection. Recommendations, Pergamon Press, Oxford. (1991) Publication No. 60, Israeli M (1985) Deposition rates of radon progeny in houses, Health Phys, 49, 1969-1983. Jacobi W (1993) The history of the radon problems in mines and homes, Annals of the ICRP, 23(2), 39. Jha G, Raghavayya M, and Padmanabhan N (1982) Radon permeability of some membranes, Rad, Meas., 19, 307 - 308. Mayya YS, Eappen KP, and Nambi KSV (1998) Methodology for mixed field inhalation dosimetry in monazite areas using a twin-cup dosimeter with three track detectors, Rad. Prot. Dosimetry, 77, 177 - 184. Megumi K, and Mamuro T, (1974) Emanation and exhalation of radon and thoron gas from soil particles, J. of Geophys. Res., 79(23), 3357-3360. Menon, MR, Lalit, BY and Shukla, VK (1987) Natural radioactivity content in building construction materials in India, Bull. Rad. Prot, 14, 45 - 48. Mettler FA and Upton AC (1995) Medical Effects of Ionizing Radiation, 2nd edition, (Eds.) Saunders, W.B. Philadelphia. Miles JCH (1997) Calibration and standardization of etched track detectors, In Radon measurements by etched track detectors, applications in radiation protection, earth Sciences and the Environment (Eds, Durrani, S. A., and Ilic, R), World Scientific, Singapore, 143176. Mishra, UC and Sadasivan, S (1971) Natural radioactivity levels in Indian soil, J. of Sci.and Industrial Research, 30, 59 - 62. Nambi KSV, Bapat VN, David M, Sundaram VK, Sunta CM, and Soman SD (1986) Natural Background Radiation and Population Dose Distribution in India, BARC Report, Health Physics Division, Bhabha Atomic Research Center, Government of India, Mumbai. Nambi KSV, Subba Ramu MC, Eappen KP, Ramachandran TV, Muraleedharan TS, and Shaikh AN, (1994) A new SSNTD method for the measurement of radon-thoron mixed working levels in dwellings, Bull. of Rad. Prot., 17, 34 - 35. 18 Nazaroff WW (1988) Radon and its decay products in indoor air; (Eds. W. W. Nazaroff and A. V. Nero), Chapter 2, John Wiley & Sons, New York, pp. 57 - 112. Nero AV, Berk JV, Boegel CD, Hollowell JR, Ingersoll, and Nazaroff WW (1983) Radon concentrations and infiltration rates measured in conventional and energy efficient houses, Health Phys, 45,401 - 405. Nikolaeve VA and Ilic R (1999) Etched track radiometers in radon measurements: A review, Rad Meas., 30, 1-13. Planinic J and Faj Z (1990) Equilibrium factor and dosimetry of radon with a passive nuclear track etch detector, Rad. Prot. Dosimetry, 1, 97 - 109. Planinic J and Faj Z (1991) Dosimetry of radon and its daughters by two SSNTD detectors, Rad. Prot. Dosimetry, 35, 265 - 268. Porstendorfer J (1994) Properties and behavior of radon and thoron and their decay products in the air, J. Aerosol Sci, 25, 219 - 263. Porstendorfer J and Mercer TT (1979) Influence of electric charge and humidity up on the diffusion coefficient of radon decay products, Health Phys, 37, 191. Ramachandran TV, Eappen KP, Shaikh AN, and Mayya YS (2001) Indoor radon levels and equilibrium factors in Indian dwellings, Rad. Prot. and Envir, 420-422. Ramachandran TV, Subba Ramu MC, and Nambi KSV (1995) Simultaneous measurements of radon and its progeny using SSNTDs and evaluation of internal doses due to inhalation, Bull. Rad. Prot., 18, 109 - 114. Sadasivan S, Shukla VK, Chinnasaki S and Sartandel S J (2003) Natural and fallout radioactivity measurements in Indian soil, J. Radio, Anal. And Nucl. Chem., 256, 603 - 607. Sankaran AV, Jayswal V, Nambi K S V, and Sunta C M (1986) U,Th and K distributions inferred from region of geology and the terrestrial radiation profiles in India, Bhabha Atomic Research Center (Government of India) Report Shaikh AN, Muraleedharan TS, Ramachandran TV and Subba Ramu MC (1992) Estimation of ventilation rates in dwellings, The Sci. of the Tot. Envir, 121, 67 - 76. Sohrabi M (1997) High radon levels in nature and in Dwellings: Remedial actions, In Radon Measurements by Etched Track Detectors, Applications in Radiation Protection, Earth Sciences and the Environment, (Eds. Durrani, S.A., and Ilic, R), World Scientific, Singapore, 225 - 242. Steinhausler F, Hoffman W, and Lettner H (1994) Thoron exposure to man: A negligible issue? Rad. Prot. Dosi., 56, 1 - 44. Stranden E, Kolstad AK, and Lind B, (1984) The influence of moisture and Temperature and radon exhalation, Rad. Prot. Dosi., 7, 55-58. 19 Strong KP, and Levins DM, (1982) Effect of moisture content on radon emanation from uranium ore and tailing, Health Phys, 42, 27. Subba Ramu MC, Muraleedharan TS, and Ramachandran TV (1988) Assessment of lung dose from radon daughters in dwellings, Rad. Prot. Dosim., 22,187 - 191. Subba Ramu MC, Raghavayya M, and Paul AC, (1994) Methods for the measurement of radon, thoron and their progeny in dwellings, AERB Technical Manual, TM/RM - 1. Subba Ramu MC, Ramachandran TV, Muraleedharan TS, Shaikh AN, and Nambi KSV, (1993) Proc. of 7th.Nat. Conf. on Particle Tracks in Solids, Defense Laboratory, Jodhpur, 11 -25. Tanner AB (1980) Radon migration in the ground: A supplementary review, In proceeding of Natural Radiation Environment III (Eds., Gassel, T. F., and Lowder, W.M), Tech. Infor. Centre, US Department of Energy, Washington, D.C, CONF - 780422. United Nations Scientific Committee on the Effects of Atomic Radiation (1988) Sources, Effects and Risks of Ionizing Radiation, Report to the General Assembly, United Nations, New York. United Nations Scientific Committee on the Effect of Atomic Radiation (1992) Sources and Biological Effects, Report to the General Assembly., United Nations, New York. United Nations Scientific Committee on the Effects of Atomic Radiation (1993) Sources, Effects and Risks of Ionizing Radiation, Report to the GeneralAssembly, United Nations, New York. United Nations Scientific Committee on the Effect of Atomic Radiation (2000) Sources, Effects and Risks of Ionizing Radiation, Report to the General Assembly, United Nations, New York. Wafaa A (2002) Permeability of radon-222 through some materials, Rad. Meas., 35, 207 - 211 Ward WJ, Fleischer RL, and Mogro-Campero A (1977) Barrier technique for separate measurement of radon isotopes, Rev. Sci. Instr., 48, 1440 - 1441. Yihe J, Shimo M, Weihai Z, Liya Xu, and Guoqiu fang (1996) The size distribution of atmospheric radioactive aerosols and its measurement, Proc, of: 4th Int. Conf, on High Levels of Natural Radiation, Beijing, China, October, 21 - 25 (Eds., Luxin, Wei., Tsdtomu, Sugahara., and Zufan, Tao), Elseveir, Amsterdam, 1997, 97 - 109. 20 Table Caption Table 1: Natural Radioactivity Content in Indian Building Materials Table 2: Natural Radioactivity Content in Indian soil Table 3: Uranium, Thorium and Potassium Content in Indian Rocks Table 4: Volume Specific Entry Rate and indoor Radon Levels from Various Sources Table 5: Reported Indoor 222Rn Levels Around the World Table 6: Outdoor and Indoor 220Rn Levels around the World Table 7: Calibration Factors (CFs) for the Cup Mode and Bare Mode Exposures Table 8: Average concentration of 222Rn, 220Rn and their progeny in air and corresponding annual effective doses Table 9: Indoor 220Rn levels and equilibrium factors in Indian dwellings using Cup dosimeter Table 10: Indoor 222Rn, 220Rn levels and Inhalation Doses Table 11: 222Rn levels in different types of dwellings Table 12: Statistical significance between radon level and type of dwelling Table 13: Ventilation Rates in Indian Dwellings Table 14: Action Levels Reported in Literature 21 Table 1: Natural Radioactivity Content in Indian Building Materials (Menon et al 1987) 40 Material Cement Brick Stone Sand Granite Clay Fly ash Lime stone Gypsum 5 130 48 5 76 6 6 6 70 226 K - 385 - 1390 - 1479 - 1074 - 1380 - 477 - 522 - 518 - 807 232 Ra Th Radium equivalent Bq kg-1 16 - 377 8 - 78 40 - 440 21 - 48 26 - 126 88 - 311 6 - 155 5 - 412 24 - 311 1 - 5047 4 - 2971 22 -7759 4 - 98 103 - 240 25 - 525 7 - 1621 4 - 311 11 -1865 7 - 670 30 - 159 56 - 773 1 - 26 1 - 33 5 - 148 7 - 807 1 - 152 59 - 881 Table 2: Natural Radioactivity Content in Indian soil (Mishra et al 1971; Sadasivan et al 2003) 232 Location Ahmedabad Aligarh Bangalore Bhopal Bikanir Mumbai Kolkatta Cherrapunji Chingalput Coimbatore Cuttack Darjeeling Dehradun Delhi Dhanbad Gangtok Gulmarg Hyderabad Jaduguda Jaipur Jasilmar Jamnagar Jodhpur Th 226 40 Ra K -1 Bq kg 53.0 82.0 16.9 15.7 11.7 13.5 24.1 17.4 120.5 10.1 61.7 24.8 22.6 19.9 37.1 23.9 20.1 45.9 41.0 14.9 37.1 3.8 13.1 24.8 54.4 15.2 11.8 8.8 9.4 20.4 21.5 22.9 10.2 15.3 2417 25.9 19.2 53.0 26.1 15.9 15.2 179.1 11.6 49.0 2.9 10.7 526.6 530.1 486.7 376.8 439.6 169.6 662.9 37.7 408.2 266.9 722.2 678.2 803.8 536.9 345.4 854.1 555.8 1073.4 455.3 505.5 565.2 56.5 458.9 232 Location Kanpur Kharagpur Kakrapar Chennai Mangalore Meerut Nagpur Nainatal Nasik Ooty Poona Ranchi Shillong Srinagar Tehri Thiruvalla Trivandrum Tiruchirapali Visakapatnam Udagamandalam Jhansi Kaiga Thumba Th 226 40 Ra K -1 Bq kg 23.8 18.4 12.4 23.1 13.5 22.0 16.5 24.8 34.4 3.4 4.2 22.4 23.9 18.6 136.2 74.3 53.2 47.8 24.8 43.2 12.6 12.4 27.5 24.0 15.2 12.2 6.7 9.3 22.7 11.8 24.7 18.6 2.5 3.0 24.4 15.5 14.8 81.6 19.8 20.3 2070 163.2 114.6 20.4 12.2 8.9 850.9 72.2 94.4 766.2 151.2 112.3 307.7 979.7 290.6 87.8 87.9 1055.0 323.2 615.4 328.2 25.1 37.7 509.4 376.8 272.6 518.2 94.2 ----- 22 Table 3: Uranium, Thorium and Potassium Content in Indian Rocks (Sankaran et al 1986) 228 State 232 U Th Potassium (%) Bq kg Andaman & Nicobar Andhra Pradesh Arunachal Pradesh Assam Bihar Daman & Diu Delhi Goa Gujarat Haryana Himachal Pradesh Jammu & Kashmir Karnataka* Kerala Madhya Pradesh Maharashtra Manipur Meghalaya Mizoram Nagaland Orissa Pondicherry Punjab Rajasthan Tamil Nadu Tripura Uttar Pradesh West Bengal 31.5 33.2 34.9 63.0 40.9 55.7 32.6 33.0 55.7 32.6 32.6 43.4 33.0 45.1 44.0 31.7 95.2 66.7 35.5 89.1 35.4 27.4 32.6 36.7 27.4 33.1 32.9 47.9 40 K -1 27.4 40.9 98.2 129.3 36.9 24.5 30.4 30.5 24.5 30.4 30.4 29.0 30.5 47.6 31.6 33.4 36.2 32.0 28.8 39.5 110.3 33.1 30.4 32.1 33.1 28.5 33.8 45.1 1.22 1.65 2.00 2.41 1.62 1.63 1.87 1.33 1.63 1.87 1.87 1.76 1.33 1.80 1.48 1.64 1.63 1.67 1.87 2.00 1.61 1.52 1.87 1.64 1.52 1.55 2.03 1.86 378.2 511.5 620.0 747.1 502.2 412.3 579.7 412.3 505.3 579.7 579.7 545.6 412.3 558.0 458.8 508.4 505.3 517.7 579.7 620.0 499.1 471.2 579.7 508.4 471.2 480.5 629.3 576.6 Table 4: Volume Specific Entry Rate and indoor 222Rn Levels from Various Sources (ICRP 1986) Specific entry rate (Bq m–3 h–1) Source Brick or concrete Wooden Houses Soil Outdoor air Others (walls, natural gas) All sources Estimated mean Range 2-20 <1 1-40 2-5 < 0.1 6-60 1-50 0.05-1 0.5-200 0.3-15 0.01-10 2-200 Indoor 222Rn concentration *(Bq m–3) Estimated Range mean 3-30 0.7-100 <1 0.03 - 2 2-60 0.5-500 3-7 1-10 <0.1 0.01-10 10-1000 2-500 * Mean ventilation rate used is 0.7 h–1 (normal range 0.3 – 1.5 h–1) 23 Table 5: Reported Indoor 222Rn Levels Around the World (UNSCEAR 2000) Concentration (Bq m–3) AM GM MAX GSD Africa Algeria 30 140 Egypt 9 24 Ghana 340 North America Canada 34 14 1720 3.6 United States 46 25 3.1 South America Argentina 37 26 211 2.2 Chile 25 86 Paraguay 28 51 East Asia China 24 20 380 2.2 Hong Kong 41 140 India 57 42 210 2.2 Indonesia 12 120 Japan 16 13 310 1.8 Kazakhstan 10 6000 Malaysia 14 20 Pakistan 30 83 Thailand 23 16 480 1.2 West Asia Armenia 104 216 1.3 Iran 82 3070 Kuwait 14 6 120 Syria 44 520 North Europe Denmark 53 29 600 2.2 Estonia 120 92 1390 Finland 120 84 20000 2.1 Lithuania 55 22 1860 Norway 73 40 50000 Sweden 108 56 3900 West Europe Austria 15 190 Belgium 48 38 12000 2.0 France 62 41 4690 2.7 Germany 50 40 >10000 1.9 Ireland 37 1700 Luxemburg 110 70 2500 2.0 Netherlands 23 18 380 1.6 Switzerland 70 50 10000 U.K 20 10000 East Europe Bulgaria 22 250 Czech Republic 140 20000 Hungary 107 82 1990 2.7 Poland 41 32 432 2.0 Romania 45 1025 Slovakia 87 3750 South Europe Albania 120 105 270 2.0 Croatia 35 32 92 Cyprus 7 7 78 2.6 Greece 73 52 490 Italy 75 57 1040 2.0 Portugal 62 45 2700 2.2 Slovenia 87 60 1330 2.2 Spain 86 42 15400 3.7 Oceania Australia 11 8 420 2.1 New Zealand 20 18 90 Median 46 37 480 2.2 Population weighted average 39 30 1200 2.3 Region Country 24 Table 6: Outdoor and Indoor 220Rn Levels around the World (UNSCEAR 2000) Country North America United States of America China Hong Kong Japan Malaysia France Equilibrium equivalent concentration 220Rn/222Rn EEC ratio (Bq m–3) Outdoor Indoor Outdoor Indoor 0.5 0.04 0.09 (0.03-4.7) (0.03-0.3) 0.4 0.3 (0.1-0.5) 009 (0.03- 0.12) 0.5 (0.3-1.8) - United Kingdom - Germany - Republic of Moldova Romania Russian Federation Italy Slovenia Range 0.2 0.3 (0.1-0.6) 0.12 (0.05 - 0.37) (0.09 - 0.5) 0.2 (0.1-0.3) 0.8 0.8 (0.4-1.2) 0.7 (0.04-2.1) 1.1 (0.4-2.1) 0.8 (0.6-13.3) 0.3 (0.07-1.1) 0.5 (0.1 -1.0) 1.0 (0.1 -6.4) 1.1 (0.1-6.4) (1.1-7 .1) 12 (0.5-76) - 0.013 (0.2 -12) 0.01- 0.08 0.05 0.04 0.07 0.06 - 0.2 0.08 0.08 - 0.03 - 0.02 - - 0.04 0.05 0.05 0.04 - 0.09 (0.02 - 0.24) 0.11 (0.01 - 0.38) - 0.01- 0.5 25 Table 7: Calibration Factors (CFs) for the Cup Mode and Bare Mode Exposures (Mayya et al 1998; Eappen et al 2004) Mode of Exposure 222 Filter Calibration Factors (Tr cm–2/Bq d m-3) for Rn Membrane 220 Rn Filter Membrane Cup Mode Exposure Experimental Theoretical Bare Mode Exposure 0.02 ± 0.004 0.019 ± 0.003 0.017 ± 0.003 0.016 0.021 Experimental Theoretical - 0.0 - 0.020 ± 0.002 0.019 ± 0.003 0.019 0.019 Table 8: Average concentration of 222Rn, 220Rn and their progeny in air and corresponding annual effective doses (UNSCEAR 2000) Radionuclide Radon Total Thoron Location Concentration (Bq m–3) Effective dose equivalent (mSv/ Bq h m–3) Annual effective dose (µ Sv) Gas EEC+ Gas EEC Gas EEC Outdoor Indoor 10 40 6 16 0.17 0.17 9 9 3 48 Outdoor Indoor 10 10 0.1 0.3 0.11 0.11 40 40 2.0 8.0 95 1009 1155 7.0 84 101 1256 Total Total Annual Effective Dose Equivalent Due to 222Rn and 220Rn (µ Sv) + This is the equilibrium equivalent concentration (EEC) of radon/thoron and is the product of the concentration of radon/thoron and the equilibrium factor between radon/ thoron and its decay products. The equilibrium factor has been taken as 0.6 for outdoor and 0.4 for indoor in the case of radon. In the case of thoron F is taken as 0.01 for outdoor and 0.03 for indoor. These values are weighted for an occupancy factor of 0.2 for outdoor and 0.8 for indoor. 26 Table 9: Indoor 222Rn levels and equilibrium factors in Indian dwellings using Cup dosimeter (Ramachandran et al 1995) State Andhra Pradesh A & Nicobar Arunachal Pradesh Assam Bihar Chaidigarh Delhi Ooa Gujarath Haryana Himachal Pradesh Karnataka Kerala Maharashtra Madhya Pradesh Meghalaya Orissa Punjab Pondicheery Rajashtan Sikkim Tripura Tamil Nadu Uttar Pradesh West Bengal Sites 5 1 1 2 9 1 1 1 3 2 3 6 9 5 2 2 3 4 1 2 1 I 11 6 3 Radon levels (Bq m–3) Equilibrium factor No. of houses MAX MIN GM GSD MAX MIN GM GSD 5 1 1 5 15 1 1 1 3 4 3 6 9 5 2 2 12 4 1 2 1 1 11 6 3 41.8 15.6 27.6 88.7 92.6 29.9 39.8 23.4 26.4 96.8 43.6 56.9 51.3 35.2 45.8 33.7 64.8 93.0 14.2 66.9 55.9 59.3 51.9 95.4 95.4 6.4 10.0 16.9 43.7 7.4 19.3 12.7 8.8 9.4 7.0 10.4 16.7 7.1 7.6 12.2 11.6 14.6 9.0 6.9 7.9 25.1 25.1 5.9 10.5 6.4 17.5 13.4 20.6 67.6 40.9 25.6 18.4 15.5 15.0 32.1 18.3 14.9 17.0 21.2 21.3 17.3 30.1 44.7 9.9 21.4 38.3 40.0 15.3 27.0 25.5 1.7 1.2 1.2 1.1 1.9 1.2 1.4 1.4 1.4 2.6 1.4 1.6 1.6 1.4 1.5 1.4 1.6 2.2 1.3 2.3 1.3 1.5 1.7 1.8 2.1 0.93 0.58 0.35 0.87 0.92 0.42 0.90 0.59 0.94 0.85 0.80 0.86 0.95 0.88 0.67 0.92 0.67 0.87 0.63 0.77 0.34 0.59 0.75 0.81 0.95 0.27 0.46 0.24 0.48 0.32 0.33 0.36 0.42 0.39 0.36 0.36 0.29 0.21 0.30 0.32 0.30 0.21 0.30 0.42 0.26 0.31 0.28 0.20 0.31 0.21 0.46 0.51 0.29 0.67 0.61 0.36 0.59 0.53 0.65 0.54 0.57 0.46 0.51 0.48 0.44 0.48 0.42 0.55 0.48 0.44 0.32 0.41 0.44 0.54 0.54 1.4 1.1 1.2 1.1 1.2 1.1 1.4 1.2 1.3 1.4 1.3 1.4 1.5 1.3 1.3 1.5 1.4 1.4 1.2 1.5 1.0 1.3 1.3 1.3 1.4 27 Table 10: Indoor 222Rn, 220Rn levels and Inhalation Doses 222 No Location No. of Houses Rn (Bq m–3) GM GSD 01 Patiala 91 11.2 02 Chandigarh 40 15.9 03 Palanpur 30 29.2 04 Amritsar 70 14.0 05 Hamirpur 29 48.8 06 Tehri 121 41.6 07 Kumaun Hill 68 18.9 08 Hyderabad 72 4.6 09 Secunderabad 80 48.5 10 Chennai 100 14.3 11 Chennai suburbs 113 15.1 12 Kalpakkam 42 6.3 13 Mysore 70 21.5 14 Mysore surburbs 106 9.7 15 Kamptee 12 8.7 16 Nagpur 84 54.3 17 Guwahati 48 48.1 18 Shillong 29 59.7 19 Karimganj 7 37.6 20 Kailash sahar 5 31.3 21 Itanagar 65 41.1 22 Mizoram 17 27.6 23 Namrup 10 147.3 24 Digboi 20 60.5 25 Agarthala 57 34.7 222 –3 Mean Rn concentration (Bq m ) 220 Mean Rn concentration (Bq m–3) Mean total inhalation dose rate (mSv y–1) 2.2 1.7 1.7 2.0 1.8 1.7 1.5 2.1 2.1 2.3 1.7 1.8 2.7 2.7 2.3 3.3 1.7 2.0 1.5 1.6 1.7 1.7 1.4 1.7 1.7 220 Rn (Bq m–3) GM GSD 6.3 8.4 14.6 7.8 32.3 13.1 21.1 3.5 34.0 6.4 13.5 5.7 19.6 11.4 6.1 15.1 25.4 29.5 10.2 15.5 28.6 12.1 23.6 42.8 18.3 2.7 2.4 2.4 2.7 2.4 2.2 2.1 3.3 3.3 3.3 2.1 1.9 3.1 3.1 2.9 4.2 1.7 2.1 1.7 1.9 1.8 2.0 2.1 2.3 2.1 Inhalation dose (mSv y–1) 222 220 Rn + Rn + Progeny Progeny 0.37 0.07 0.53 0.10 0.96 0.17 0.46 0.09 1.61 0.37 1.37 0.15 0.62 0.24 0.15 0.04 1.60 0.39 0.48 0.08 0.50 0.16 0.21 0.07 0.71 0.23 0.32 0.13 0.29 0.07 1.79 0.17 1.59 0.29 1.97 0.34 1.24 0.12 1.03 0.18 1.36 0.33 0.91 0.14 4.87 0.27 1.15 0.21 0.21 0.07 Total inhalation Dose (mSv y–1) 0.44 0.63 1.13 0.55 1.98 1.52 0.86 0.19 1.99 0.55 0.66 0.28 0.94 0.45 0.36 1.96 1.88 2.31 1.36 1.21 1.69 1.05 5.14 1.36 0.28 23.0 12.2 0.97 28 Table 11: 222Rn levels in different types of dwellings Wall Type Flooring Bare Plaster and painted Cement Cement Mosaic Tile Stone Cement Mosaic Wood Tile Cement Mosaic White washed Wooden panel No. of houses GM (AM) (Bq m–3) GSD(SD) 4 121 95 12 1 11 7 4 4 3 8 20.8 (21.6) 20.2 (23.4) 18.1(21.4) 12.9(13.5) 28.5 (28.5) 15.1 (18.5) 34.8 (38.9) 15.4(17.8) 12.1 (13.0) 10.8 (10.9) 13.7(13.9) 1.3 (7.0) 1.7 (13.2) 1.8 (13.2) 1.4 (4.0) 1.0 (0.0) 1.8 (14.9) 1.7 (18.0) 1.7 (11.9) 1.5 (4.8) 1.1 (1.4) 1.2 (2.9) Table 12: Statistical significance between radon level and type of dwelling Wall Plaster and paint Plaster and whitewash Wood Floor Cement Mosaic Cement Tile Mosaic Tile Mosaic Cement Cement Tile Mosaic Wood Mosaic Tile Wood Tile Mosaic Cement Floor Cement Mosaic floor Wall Plaster/paint Plaster/whitewash Plaster/whitewash Plaster/paint Plaster/whitewash Wood Difference of means Statistical estimate 95 % Confidence limits Remarks 2.00 2.30 -2.5 to 6.5 No difference 9.90 1.66 6.7 to 13.8 Small difference 7.90 2.28 3.4 to 12.3 Small difference 20.40 8.15 4.1 to 36.4 Significant difference 5.50 5.09 -4.5 to 15.5 No difference 21.10 9.04 3.4 to 38.3 Significant difference 25.90 7.21 11.8 to 40.0 Significant difference 4.80 6.42 -7.8 to 17.4 No difference 3.00 1.31 0.4 to 5.6 Small difference Difference of means Statistical estimate 95 % Confidence limits Remarks 4.90 4.65 -4.2 to 14.0 No difference 17.50 7.08 3.6 to 37.4 Significant difference 25.00 6.89 11.5 to 38.5 Significant difference 29 Table 13: Ventilation Rates in Indian Dwellings (Shaikh et al 1992) Actual * Mean ventilation ventilation rate rate (h–1) (h–1) Type of dwelling Pseudo ventilation Rate (h–1) Plate out rate (h–1) Chawl 3.9 1.43 2.43 2.4 1.07 1.33 1.5 3.7 3.7 2.1 1.10 0.90 1.43 1.02 0.42 2.82 2.27 1.08 5.0 1.38 3.62 4.5 2.61 1.89 4.7 2.73 1.97 4.9 2.59 2.31 3.0 2.5 3.6 6.7 3.3 2.8 5.5 1.02 1.98 1.45 2.24 1.69 1.76 2.08 1.98 0.52 2.15 4.46 1.61 1.04 3.42 3.7 1.65 2.05 Bungalow A/C room Flat 1.73 2.76 2.14 2.15 30 Table 14: Action Levels Reported in Literature (Soharabi, 1997) Country Australia Austria Canada Denmark Action Level ( Bq m-3) Old New Building Building 200 200 400 400 800 200 200 Remarks and/or recommended time for remedial action ….. ……. ……. ……. A time frame band and on the basis of a life time (60 y) cumulative exposure of 15,000 Bq/m3 y; 10 times higher than the UK (NRPB) level Germany 250 250 Ireland 200 200 Sweden 200 70 Between 70 to 200 should be reduced by simple measurements if possible 200 200 A time frame band on the basis of a life time (60 y) cumulative exposure of 1500 Bq/m3 , a few years; for 750 to 7500 Bq/m3, within a few months; above 7500 Bq/m3, immediate action or evacuation. United States ICRP 65 IAEA-BSS CEC 150 200-600 200-600 400 150 200-600 200-600 200 WHO 200-300 200-300 United Kingdom 31 Figure Caption Fig. 1: Frequency distribution of equilibrium factors Fig. 2: Distribution pattern of indoor 222Rn levels Fig. 3: Distribution pattern of indoor 220Rn levels Fig. 4: Relation between indoor 222Rn and 220Rn concentrations Fig. 5: Indoor 222Rn levels in India Fig. 6: Indoor 220Rn levels in India Fig. 7: Relation between total indoor inhalation dose rates and that due to progeny 220 Rn and its Fig. 8: Distribution pattern of total indoor inhalation dose rates Fig. 9: Total indoor inhalation dose rates due to locations in India 222 Rn, 220 Rn and their progeny at different 32 Fig. 1: Frequency distribution of equilibrium factors 33 Fig. 2: Distribution pattern of indoor 222Rn levels 34 Fig. 3: Distribution pattern of indoor 220Rn levels 35 Fig. 4: Relation between indoor 222Rn and 220Rn concentrations 36 Fig. 5: Indoor 222Rn levels in India 37 Fig. 6: Indoor 220Rn levels in India 38 Fig. 7: Relation between total indoor inhalation dose rates and that due to 220Rn and its progeny 39 Fig. 8: Distribution pattern of total indoor inhalation dose rates 40 Fig. 9: Total indoor inhalation dose rates due to 222Rn, 220Rn and their progeny at different locations in India 41 COMPARISON OF TWO DAY CONSECUTIVE RADON MEASUREMENT RESULTS TO A 90 DAY AVERAGE MEASUREMENT RESULTS USING THE DATA FROM CONTINUOUS RADON MONITORS IN A TYPICAL SINGLE FAMILY HOME IN FREDERICK, MD, USA Payasada (Paul) Kotrappa and Frederick Stieff Rad Elec Inc. 5716-A Industry Lane Frederick MD 21704 ABSTRACT Two calibrated Radon Scout continuous radon monitors were installed in the basement of a typical single family home in Frederick, MD, USA, in accordance with the USEPA deployment protocols. These units were set to collect and average radon concentrations over three hour intervals. In addition to logging the average radon concentrations, the basement temperature, humidity, and barometric pressures were also collected. The units were run for three months in closed house conditions. The data was analyzed with an aim to find any correlation of three hourly peaks to outdoor environmental conditions such as precipitation, temperature, and wind. Further aim was to determine how the two day consecutive average radon concentrations compare with the three month average radon concentrations. This work describes the results for the first two quarters of 2010 and will continue for the entire year. This may answer some frequently faced queries by the radon measurement professionals, in the course of their work. INTRODUCTION In accordance with the USEPA radon testing protocols, short term indoor radon measurements are usually carried out for 2 to 7 days; and long term measurements are carried out for a period of more than 90 days. Radon concentrations at a given time depend upon the time of the day, and atmospheric conditions such as barometric pressure, precipitation, and outside temperature differentials. It is widely accepted that the emanation of radon from the ground and diffusion of radon into home is dependent upon these parameters. Such dependence is very complex. When short term measurements are conducted for a period less than 91 days, such as for two days, seven days or even for 30 days, such measured values can be different from the long term average. The purpose of this study is to compare 2 day averages, seven day averages, and monthly averages to the long term average. Such a comparison provides the type of uncertainties encountered when a short term measurements are done. Further whether there are some noticeable episodes of weather conditions. In fact USEPA has recommended: “Quote Radon Measurement Protocol: EPA-402-R—92-004(1992): 42 Short-term tests lasting for two to three days should not be conducted if severe storms with high winds (e.g.., >30 mph) or rapidly changing barometric pressure are predicted during the measurement period. Weather predictions available on local news stations can provide sufficient information to determine if these conditions are likely”. It is of interest to examine the data and find any correlation of high transient radon concentration with other major episodes such as light/heavy rains, heavy snow, slow /fast thaws, high winds, and unusually low or high temperatures. METHODOLGY A pair of calibrated Radon Scout Continuous Radon Monitors (CRMs), were installed in the basement of a typical single family home in Frederick, MD, USA. The Radon Scout CRMs are NRSB listed devices, based on USEPA evaluation. These instruments have the capability of recording hourly or 3 hourly readings, temperature, humidity and atmospheric pressures in the area where the monitors are located. The units were set up to measure 3 hourly average parameters. The readings were down loaded and displayed in an Excel® spread sheet. The data was analyzed to calculate consecutive two day averages, consecutive 7 day averages, and consecutive monthly (30 day) averages for each quarter. The home had a sub slab depressurization mitigation system installed in the basement, but this mitigation system was turned off during these studies. The home has typical forced air heating unit and a centralized air conditioning system. The thermostat was maintained at about 70 degrees F in winter and about 80 degrees F in summer. Closed house conditions that complied with USEPA short term radon testing protocols were maintained for the entire period of the studies. This is the normal operating conditions for the home. RESULTS Two sets of quarterly data were analyzed - one set for the period from Jan 20 to April 20, 2010 and another set for the period from April 20 to July 20, 2010. The first set of data covers part of winter and a part of spring. Second quarter covers part of spring and a part of summer. FIRST QUARTER (January 20 to April 20, 2010) Results are displayed in Figures 1 to 5. Figure 1 is a record of the radon concentrations over the stated periods. Note that the radon concentration averaged over 3 hours is recorded to provide better statistics. The concentrations varied from insignificant radon concentrations, to a high of 12 pCi/L with several peaks and dips. Such variations are attributed to the varying environmental conditions. Due to such variation, a two day measurement can lead to significant errors relative to a 3 month average. Figure 2 presents consecutive two day average concentrations. These show a spread from 1.9 pCi/L to 5.5 pCi/L. Any random 2 day measurement can be anywhere between these two numbers, depending upon the environmental conditions. The Standard Deviation of the 2 day measurements is about 40%. Figure 3 presents consecutive seven day average concentrations. These show a spread from 1.9 pCi/L to 5.0 pCi/L. Any random seven day measurement can be anywhere between these two numbers, depending upon the 43 environmental conditions. The Standard Deviation of the 7 day measurements is about 30%. Figure 4 presents consecutive monthly averages. The data indicates that even a monthly measurement can vary from 2.5pCi/L to 4 pCi/L, and can be quite different from the 3 month average. Figure 5 gives a correlation between the atmospheric pressure (as measured inside the monitoring area) and the radon concentration .There is no very clear correlation. Generally, lower pressures lead to higher radon concentration, but not all the time. SECOND QUARTER Figure 6 to 10 gives data for this period. Comments for the first quarter are generally applicable for this quarter. However the quarterly average is higher (4.7 pCi/L) compared to the first quarter average of 3.4 pCi/L. EPISODES AND CORRELATIONS WITH RADON PEAKS The radon concentration of a 7.0 pCi/L and above is considered as significant peak for the fist quarter. This is about twice the quarterly average. Table-1 gives the list of major episodes for the first quarter and their association, if any, with the radon peaks. NS stands for “not significant”. Table 2 gives the list of major episodes for the second quarter and their association with major radon peaks. DISCUSSIONS AND CONCLUSIONS: Steck and his associates (Steck 1990; Steck 1992; Steck (2009) have done similar studies extending to several years. Their conclusions regarding the in adequacy of making short term measurement is similar to the one that is made in this paper. The Spring/Summer season (second quarter) gave higher quarterly average radon concentrations, compared to the Winter/Spring season. During the first quarter (Table-1), major snow storms and high winds did not show any association with high radon peaks. However, in four cases high radon is associated with the rain. There are four cases where high radon is not associated with the rain. During the second quarter (Table-2), there are only in two cases where there is an association of high radon with rain. High wind is not associated with radon peaks. Other radon peaks are not associated with rain. The association of radon peaks with rain is not unexpected. Moist soil has higher radon emanation coefficient in moist to wet conditions. This leads to higher emanation of radon. However, there may be cases of very wet conditions that may also block the radon. In spring months, the ground is generally wetter which “releases” more radon. This may be the reason for higher average concentration in spring compared to winter, but fewer peaks are associated with the rain. 44 Table-1 Episodes and Radon concentration peaks in first quarter Dates 1/25-1/26 1/30 2/5- 2/6 2/9-2/10 2/11-2/12 2/14-2/15 2/25 3/6 3/12-3/13 3/14-3/15 3/22 3/28-29 4/3 4/7 4/17 Radon Peaks Rain Snow Wind pCi/L Inches Inches MPH 11.8 NS NS NS NS NS NS 6.8 10.1 8.8 8.6 9.4 6.8 7.1 7.6 Heavy 2 to 3 2.5 to 3 Rain continued 1.0 - High wind Heavy 4 to 6 Heavy 20-30 Heavy 15-20 Snow drift 2- 3 Snow melt - Very high 45 Table-2 Episodes and Radon concentration peaks in second quarter Dates Radon Peaks Rain pCi/L Inches Snow Inches 4/25-4/27 5/3-5/4 5/7-5/8 5/23-5/24 5/31-6/1 6/3-6/8 6/13- 6/16 10.5 11.4 NS 12.2 8.3 13.3 13.1 - Heavy 1.71 0.26 - Wind MPH Very high wind 45 REFERENCES Steck D.J, A Comparison of EPA-screening measurements and annual radon in statewide surveys. Health Physics 58:523-530 (1990) Steck D.J, Spatial and temporal indoor radon variations. Health Physics: 62:351-355 (1992) Steck, D.J, Annual average indoor radon concentration over two decades. Health Physics: 96:37-47 (2009) Jan 20 to April 20, 2010 3 hourly average radon concentration 3 month average : 3.4 pCi/L Radon pCi/L 14 12 10 8 6 4 2 0 1 4 5 89 133 177 2 21 265 309 35 3 397 44 1 485 529 57 3 617 66 1 7 05 Cons ecutive 3 hours (J an 20 to April 20, 2 01 0) Figure 1 Co nsec utive 3 h ourly r ad on Con centr atio n for p eriod Ja n 20 to Ap ril 20, 2010 Radon pCi/L 2 day average radon concentration 3 month average: 3.4 pCi/L 6.0 5.0 4.0 3.0 2.0 1.0 0.0 46 0.0 10.0 20.0 30.0 40.0 50.0 60.0 EXPERIMENTAL DETERMINATION OF THE EFFECTIVENESS OF RADON BARRIERS Michael Kitto1,2,* and Edward Perazzo3 1 Wadsworth Center, New York State Department of Health, P.O. Box 509, Albany, NY 12201 2 School of Public Health, State University of New York at Albany, Rensselaer, NY 12144 3 Siena College, Loudonville, NY 12211 Abstract Several types of membrane materials (radon retarders) are available for placement under concrete slabs as barriers to the upward movement of soil gas into buildings. Selection of barrier material is seldom based on its resistance to air permeation, as such information is not readily available. For the current study, the permeability of several membrane materials, which may be used as radon barriers, were tested in the laboratory using three methods. All membrane materials were found to significantly reduce radon permeation, but the efficacy of resistance varied considerably amongst the membranes. Introduction Convective flow and diffusion are the primary driving forces behind radon entry into buildings, with the former often being the dominant force due to reduced pressure in a building relative to the outdoor environment. A barrier, such as polyethylene sheeting, is frequently placed under the concrete slab during building construction as membrane material and vapor barrier to reduce radon transport into the building as a result of both convective flow and diffusion. The purpose of the membrane material is to retard gases and aerosols (e.g., radon, methane, water vapor) that emanate through the soil from being transported into the living area of the home. The membrane is typically laid over the layer of gravel placed under the foundation. The effectiveness of a membrane for reducing the movement of radon (or soil gas) into the building is dependent upon the material composition, material thickness, and sealing of the membrane seams. There have been several studies of the efficacy of barrier membranes for reducing radon penetration. Chen et al. (2009) examined 10 membranes commonly used in Canada, and concluded that membranes of higher density are better barriers of radon permeability. Similarly, Daoud and Renken (2001) determine that several membrane materials are sufficiently impermeable to be used under a concrete slab for radon reduction. The methodologies utilized were somewhat different than the approach used in the present study. Materials and Methods For this study, several potential membrane materials were examined. In addition to the 47 membranes available from distributors of radon mitigation supplies, various thicknesses of polyethylene sheeting were included in the study. Table 1 provides a listing and description of the membranes that were studied. The thicknesses of the membranes varied from about 1 mil (0.001 inch; 0.0254 mm) to 16 mil; colors included blue, black, clear, and white; and, as shown in Fig. 1, thickness was well correlated (r2 = 0.97) with area density (i.e., unit weight). 400 data regression 300 2 200 r 2 = 0.97 Area density (g/m ) 100 0 0 5 10 15 20 Membrane thickness (mil) Fig. 1. Membrane thickness is strongly correlated with area density for the materials studied. Table 1. Description of membranes tested. Membrane ID Thickness (mil) Supplier 1 2-mil poly 4-mil poly 1.2-mil polyolefin Low-density poly PVC film 6-mil poly Polypropylene Supplier 2 Supplier 3 Black B Supplier 4 Supplier 5 Supplier 6 Supplier 3A 3-mil poly Black A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 6 2 4 1.2 2 1.5 6 2 16.3 6 3 3 10 4.5 6 3 3 Area density (g/m2) 100 46 88 13 44 12 148 47 390 151 53 84 263 86 151 58 78 Description clear; white fibers clear clear clear clear clear clear clear white and blue blue black white white black blue clear black 48 Three methods were applied to determine the effectiveness of the membrane materials. Not all membranes were tested by each method. Initial measurements were conducted with the membrane material placed between a continuous radon monitor (CRM; Pylon AB-5 with passive radon detector) and a radium-lined crock (i.e., Revigator) filled with water (Fig. 2A). The second approach utilized a 226Ra source (0.9 nCi) that was place inside a hemisphere, and allowed the radon to pass through the membrane material and into the CRM (Fig. 2B). Since the membrane was not affixed to the source or detector, there was the possibility of radon leakage around the membrane for setup 2A and 2B. The third approach used to examine the membranes involved placement of the 226Ra source inside a 4-L glass container that was covered with the membrane material. This setup was sealed inside a 50-L airtight chamber with a CRM (Fig. 2C). Though the membranes were sealed onto the opening of the glass jar, some membranes were too rigid to produce a tight fit. A large hose clamp was affixed to minimize leakage, but radon loss was possible for the rigid membranes. For all three approaches, counts were typically accumulated for at least one day for each membrane. Fig. 2. Experimental setups that were used to examine the permeability of membrane materials: continuous radon monitor mounted over water-filled Revigator (A); 226Ra capsule and continuous radon monitor (B); and covered glass jar containing 226Ra capsule in airtight chamber (C). Results and Discussion Two similar, but separate, experiments were conducted with the membranes covering the top of the revigator. The count rates determined near the end of the measurements (i.e., equilibrium) using the initial experimental setup showed that a large fraction of the radon was blocked by even the thinnest membrane (Fig. 3). The 2-mil polyethylene membrane reduced the measured radon levels by nearly 83%. The reduction increased to 91% for the 3-mil polyethylene membrane, and even greater reduction for the other membranes. The membrane commercially sold by Supplier 2 reduced the radon concentration to near the background level. 49 100,000 Supplier 4 Supplier 6 det. bkgrd. 4-mil poly Supplier 1 6-mil poly 2-mil poly Supplier 2 no barrier no barrier 3-mil poly 10,000 1,000 100 counts / 15-minute interval 10 200 300 400 500 600 700 800 900 1000 minutes over revigator Fig. 3. Count rates measured with membranes separating continuous radon monitor and revigator (initial setup 2A). A summary of results (Fig. 4) of the follow-up experiment using the same configuration showed a similar pattern, but with somewhat greater reductions. All membranes provided at least a 90% reduction in radon compared to the count rate measured without any membrane. 10000 1000 100 Counts/15 min 10 1 Al-foil Supplier 3 EPDM rubber Supplier 2 Supplier 6 Supplier 4 Supplier 6A 2-mil polypro 0.75-mil PVC film 6-mil poly 5-mil poly Supplier 1 4-mil poly 3-mil ploy 2-mil poly 2-mil low-d poly Membrane over revigator Fig. 4. Count rates measured with membranes separating continuous radon monitor and revigator (repeated setup 2A). 50 7000 6000 5000 ~21,000 4000 3000 Counts / Hour 2000 1000 0 no barrier 2-mil low-d poly 1.2-mil polyolefin Supplier 1 Supplier 2 Black "B" 2-mil poly 2-mil polypro Supplier 1A Supplier 5 Supplier 4 0.75-mil PVC Supplier 3 4-mil poly 6-mil poly Membrane Fig. 5. Count rates measured with membranes separating continuous radon monitor and 226Ra capsule (setup 2B). Results from the measurement of count rates (Fig. 5) emitted from the 226Ra capsule (setup 2B) are somewhat different than those determined using the revigator (setup 2A). In general, the membrane distributed by Supplier 3 showed the significant radon reduction and thin polyethylene sheeting provided the least resistance to radon permeability. As with the Revigator, no effort was made to seal the setup, so there was the possibility of radon leakage around the edge of the membranes. This possibility is supported by the lower radon reductions (70-84%) that were achieved for the various membranes using this setup. Lastly, the 226Ra source inside the glass jar (setup 2C) produced results that are similar to the other experimental setups. Membranes from Suppliers 2 and 3 provided the greatest resistance to radon permeation (Fig. 6), with a ~95% reduction in the levels measured for the unobstructed (open) source. As with setup 2A, the commercial membrane from Supplier 1 provided the least radon reduction (41%). As explained in the Experimental section, radon leakage around the edge was possible for the thicker membranes that could not be tightly compressed between the threads of the glass jar and the lid. 51 10000 1000 100 Counts / hour 10 1 no barrier Supplier 1 Supplier 5 Polyolefin 2 mil poly Black "B" 4 mil poly Supplier 6 Black "A" 6-mil poly 3-mil poly Supplier #4 0.75-mil PVC 2-mil polypro Supplier 3 Supplier 2 Membrane Fig. 6. Count rates measured with membrane separating continuous radon monitor and 226Ra capsule inside airtight chamber (setup 2C). Most publications regarding membrane permeability by radon express results in terms of gas diffusion coefficients (e.g., 10-11 to 10-14 m2 s-1). Daoud and Renken (2001) did provide percent reductions for membranes that range from 71-98% for the various materials that were tested. These reductions are similar to those determined in this study, though the materials tested were somewhat different. There are no known studies of the long-term durability of these membrane materials in a subslab environment. Based on (unpublished) measurements of various stone aggregates that are often placed under the concrete (basement) slab of a home (2000 ft2 and 0.5 ft thick), at least 1 µCi of radon is present at equilibrium from the aggregate alone. Assuming a barrier that is 90% effective is used, the radon above the barrier and available for transport into the basement, from the aggregate, is roughly 100,000 pCi. These levels of radon emphasize the importance of selecting a high-performance radon barrier, and sealing of the seams and holes that may occur during placement of the barrier at a building site. Conclusions The permeability of thin-film membrane barriers to radon was determined for several different materials that may be placed under concrete slabs to reduce radon transport. Although none of the three setups were airtight for all of the membranes, the results are similar and can be considered qualitative. All of the membrane materials effectively reduced radon transport. Membranes of greater density typically provided an improved resistance to radon movement, and often allowed only 5-10% of the radon to pass through. 52 References Chen, J.; Ly, J.; Schroth, E.; Hnatiuk S.; Frenette E.; Blain, M-F. Radon diffusion coefficients of vapour barrier membranes used in Canadian building construction. Radiat. Environ. Biophys. 48: 153-158 (2009). Daoud, W.Z.; Renken, K.J. Laboratory assessment of flexible thin-film membranes as a passive barrier to radon gas diffusion. Sci. Total Environ. 272: 127-135 (2001). 53 
 THE DISTRIBUTION OF INDOOR RADON CONCENTRATIONS IN A LARGE POPULATION OF HOMES Willard E. Hobbs Radon Reduction & Research Abstract-- One of the least understood qualities of radon and its associated risk is the relationship between its concentration at the source and the statistical distribution of concentrations found in indoor environments. In this presentation I will clarify several of the important factors and provide a simple statistical model for discussing and communicating distribution qualities of our primary source of ionizing radiation. A few of these are as follows: (1) Radium, like most other minerals in the earth’s crust, has an overall log-normal distribution in concentration; (2) The factors which affect the concentration of indoor radon do so in a multiplicative fashion; (3) The measured data sets of indoor radon concentrations follow a log-normal distribution which is characterized by a skewed distribution with a long tail to hazardous levels; (4) A fundamental comparison property for such a distribution is the geometric ratio of levels and not the arithmetic difference; and (5) The response of the soil gases to the building stock and ventilation parameters thereof provide an additional factor of approximately 2. These facts provide critical information for developing health strategies for states, communities and individual homeowners. Introduction In this paper, information on the distribution of radon concentration in California homes will be discussed. Section II summarizes qualitative information on the lognormal distributions. (Note: An appendix at the end of the paper provides some technical details on this distribution.) Section III discusses the requirement for a scientific sample in studying radon distributions is discussed. Section IV explains how various factors result in the wide spread of observed radon concentrations and their relative magnitudes. In Section V data from the California counties and from Summerland, California, provide understanding of the variance in the observed radon levels. Finally, in Section VI the practical implications of these results and how they can aide in the reduction of radon exposure to the general population. The Lognormal Distribution It is well-known that the range of radon concentrations found in US homes is very broad. The result is that the specific concentration in a given home, chosen randomly, cannot be accurately predicted. The best way to determine the radon concentration in a home is to make a measurement. When analyzing data sets, the distribution of radon concentrations is not a common distribution, but a right-skewed one called a lognormal distribution. A lognormal distribution is simply one where the logarithms of the data values have a normal (Gaussian) distribution. (The distribution name is therefore descriptive.) Since the distribution is strongly skewed it has a long tail extending to high magnitudes. Thus, there is usually a reasonable probability that any house could contain a radon concentration that is unacceptable. 
 54 
 The normal distribution is characterized by two parameters: the mean (location of the center) and the standard deviation (the relative degree of the spread). For a lognormal distribution, these are called the geometric mean and the geometric standard deviation. The geometric mean for a sample of n data points is defined as the nth root of the product of values. This is equal to the exponential of the average of the logs of all the data points (any units and any base can be used). The geometric standard deviation (GSD) does not have a simple arithmetic algorithm like the geometric mean. It is defined as the exponential of standard deviation of all the log-data. This is a dimensionless factor which has the same value no matter what units or exponent/log base is used. Conventional units (pCi/liters) and natural logarithm base = 2.718281828 are used here. The role of the distribution parameters are reflected in the shape of the lognormal distribution. Let (µg,σg) be the mean and standard deviation of the log (data values) respectively; the geometric mean and geometric standard deviation are M=exp(µg) and GSD=exp(σg). For a normal distribution about 68% of the distribution is found within one standard deviation of the mean. Thus, if a house is chosen at random, there is about probability P = 68% that the log of its radon concentration (henceforth called logradon) will be in the interval µg - σg < ln r < µg + σg (1) or equivalently M / GSD < r < M × GSD (2) There is about 95% probability that the log-radon value will be in the interval µg ± 2σg, or equivalently the radon concentration itself would be within a factor of GSD2 of M. The geometric mean is equal to the median or 50th percentile, the population center of the distribution. To illustrate the lognormal distribution using a specific distribution, suppose the median is 2 pCi/liter and the geometric standard deviation is 2.5. The radon concentration in a randomly chosen house would have 68% probability of being in the interval 0.8 to 5.0 pCi/liter and 95% probability of being the interval 0.32 to 12.5 pCi/liter. This distribution has rather high radon levels; the probability of a house having radon greater than 4 pCi/liter (z-value = [ln(4)-ln(2)]/ln(2.5) = 0.756) is 22.5%. It is convenient to think of the log of the radon level as the basic data unit. The distribution of the logs will be near-normal and the various normal statistical applications are applicable. For example, it is appropriate to do such things as estimate the population parameters from sample data or compare two populations using sample data. In Fig. 1, graphs of the probability distributions for P(µg=ln(2),σg=ln(2.5)) are shown: 
 55 
 Figure1. Lognormal distribution of radon with median M=2 pCi/l and GSD=exp(σg)=2.5. Increasing the geometric standard deviation flattens the lognormal curve just like increasing the standard deviation expands, and flattens, the more familiar bell-shaped normal curve. It greatly increases the probability that a portion of the distribution will be found at high magnitudes. The GSD is therefore a critical parameter for the lognormal curve. As an example, we know about the standard normal curve with µ=0 and σ=1. If these data were log values, then we would have eµ=1 and eσ=e=2.718. The curves for this appear very similar to those shown in Fig. 1. The bell-shaped curve is shifted over about one unit to the left. For the skewed lognormal curve, values for the horizontal axis are multiplied by 0.5 and the vertical axis is multiplied by 2.0 (the area under the curve remains 1=100%). To emphasize the skewness of the lognormal distribution here are some other parameters of the distribution shown in Fig. 1. The most likely value (the mode) is equal to 0.86, the median (given) is 2.0, and the arithmetic mean is 3.0, all in pCi/liter. The first quartile (25%-tile) is 1.1 pCi/l and the third quartile (75%-tile) is 3.7 pCi/l (note: a factor of 1.85 on either side of 2.0). The standard relation for the normal curve is ln r = µg + z σg; this allows the z-value to be related to the probability value for a range of radon levels. For radon between ln(r1) and ln(r2), P( r1 < r < r2 ) = normalcdf (ln r1, ln r2, µg, σg ) = normalcdf ( z1, z2, 0, 1 ) (3) where normalcdf is the cumulative distribution function for a normal curve. z-values are used with the standard normal curve and the probabilities are easily evaluated using a table, a calculator, or the internet. Scientific sampling The discussion in Section II, while certainly valid, assumes knowledge of the population geometric mean and geometric standard deviation which are fundamentally unknowable quantities. This technically involves information about the whole population, and, if that were known, there would be little reason to calculate the summary values. So, it seems to have limited value. The remedy is to use scientific sampling to estimate the required parameters from a manageable number of data. The gold-standard for a scientific sample is called a “simple random sample.” This is a sample created where all samples of a given size n have an equal chance of being chosen. It is analogous to putting the names of all members of the population into a hat, shaking them up, and drawing 
 56 
 the sample. This is impractical for large populations. It requires that the whole population be delineated. A multistage sampling procedure is usually taken where demographics are considered and an honest attempt to achieve a representative sample is made. This is called an unbiased or scientific sample. Various collections of radon data have been made, but most of them involve what is called voluntary or self-selected data. Voluntary samples are notoriously bad. One of the most dramatic cases was illustrated by Ann Landers in her newspaper column of Nov 3, 1975. She had surveyed her readers to determine the fraction that regretted having their children. An overwhelming number (~70%) of those who replied said that they did regret having their children. Of course, most of those who replied were unhappy so naturally they were negative. And clearly, this is not representative of the whole population. A self-selection process is at work in radon measurements also. I use as an example in my own zip code 93105. As we have determined (Hobbs, 1996), by counting houses from aerial photographs, there are 10,167 houses in this zip code with 275 (2.7%) on the high-uranium Rincon formation. (This is a little higher ratio than that the total city of Santa Barbara: 27,225 homes with 325 on the Rincon.) Based on this, we calculated that as many as 5% of the homes in 93105 would be above 4 pCi/liter. Yet the California DHS data base (Blood, 2002) shows that 54% of the homes tested in 93105 are above 4. To their credit, the web site does caution readers that these results were voluntarily submitted and they do not constitute a scientific survey. They urge everyone to do their own test. The values for 93105 are biased by the over-representation of the Rincon values. I know this because I contributed many of the values in this data base for 93105. There is a large Rincon downwash area in the eastern part of the zip code that has most of the high radon homes. Most of the homes I test are for real estate transactions. The real-estate agents and geologists know that this is an area of high radon so they emphasize testing to their clients. Whenever I am asked about a home in this area, I reply, “This is a high potential area, you definitely should test for radon.” In
proving
foresight
may
be
vain:
 The
best
laid
schemes
of
mice
and
men
 Go
often
askew,
 And
leave
us
nothing
but
grief
and
pain,
 For
promised
joy!
 In the early 90’s, as part of the Radon Abatement Act, the California Department of Health Services did undertake to survey all the counties in California to assay the average radon concentration in homes. They have a special office in Sacramento which has statistics professionals and we were told the survey would provide accurate results. This was released after the discovery of the high-uranium Rincon Formation in Santa Barbara County. We were expecting big numbers. The Santa Barbara results were average: geometric mean 0.63 pCi/l, arithmetic mean 1.25 pCi/l, and GSD = 3.35, using n=120 measurements. They did publish all of their results and identified each of the measurements by zip code (Liu, 1990). We have a nearby zip code that is exclusively built on Rincon formation soil, the tourist 
 57 
 town of Summerland, zip 93067. It was where Carlisle (Carlisle, 1993) had found the first high levels of radon in California and the levels there are consistently high. So, I studied the output and realized, “There are no data from Summerland.” I called Steve Hayward of the State Health Service and he investigated the situation. He said that the California data base for homes is from the property tax rolls (Hayward, 1995). They had reduced the base by only considering homes that are occupied by their owners by checking to see if the tax address agreed with the property address. There are none in Summerland because they do not have home delivery of US mail; they all have free boxes at the post office. It turns out that this is not the end of the story. Some years later we found another community built exclusively on exposed Rincon, Los Alamos 93440. This place also has consistently high indoor radon concentrations. You guessed it: they also do not have home delivery so none of their homes were considered in the State survey. So the State had innocently systematically excluded two high-radon communities in Santa Barbara County from their analysis, a definite cause of bias. Defining your survey population is important in making a proper survey. I believe it is important to have geologists identify different areas of potential concern before making a survey. This was partially done in the State survey by having more measurements done in counties where high levels were anticipated. For example, Ventura County (once part of Santa Barbara County) was sampled n=159 times, the largest in the state. It had been a region of intense uranium prospecting in the 1950’s and many mining claims were filed. This State survey (Liu, 1990) was used in making the EPA radon map which is ubiquitous. A relatively simple proposal for a random sampling would be to chose geographic coordinates randomly and find the nearest houses. We mentioned drawing names from a hat earlier; this process would be analogous to throwing darts at a map on the wall (blind-folded, of course). As shown in the Appendix, a scientific sample of 40 to 50 homes would give a good estimate of the median radon level. Variation in the Distribution As Nero, et al., (Nero, 1986) emphasize in their seminal paper on radon distributions, radon data in various categories are very accurately represented by lognormal distributions. They perform various statistical tests to verify this. They also provide discussion as to why this is the case. A lognormal distribution results when the magnitude of the data results from a product of factors. If there are enough, it does not matter what the shape of the distribution of the factors themselves is. For lognormal factors, there doesn’t have to be that many factors. The process is similar to the generation of the more familiar Gaussian distribution by additive factors. Nero, et al., (Nero, 1986) note that the analysis of indoor radon shows that it results from numerous factors, many are multiplicative but some additive. Because of the apparent closeness of the data to a lognormal curve, the multiplicative factors dominate the process. Suppose we have the magnitude of the indoor radon in some randomly chosen house is r. Conceptually, we can collect the factors for this level into three groups and combine them r = fs fw fb 
 (4) 58 
 where fs is the combination of the factors related to the soil (radium concentration, permeability, etc.), fw is the combination of the factors related to the weather (wind speed, rain amount, etc.), and fb is the combination of the factors related to the building (leakage area, footprint area, etc.) We know that radium concentration in soil samples is itself distributed according to a lognormal distribution (Wedpohl, 1969). It is convenient to think of the magnitude of the radon as coming from the soil and the weather and building factors spreading the indoor concentrations (like filters). In fact, it is reasonable to assume that each of these factors itself is a lognormal distribution over the homes in an arbitrary area (containing at least a thousand homes). The product radon level r will most likely have a lognormal distribution. Remember the log of the radon concentration has a normal distribution, and mathematically ln r = ln fs + ln fw + ln fb (5) All these logarithms have simple Gaussian normal distributions and we know how to work with these. In particular, the variance of the logradon values can be written as a quadrature sum σg2 = σs2 + σw2 + σb2 (6) The variation in the product is the sum of the variation in the logarithms of the various factors. The geometric standard deviation of the radon concentrations is written GSD = exp ! s2 + ! w2 + ! b2 (7) Recall from Eq. (2) that this is the factor of the distribution median (×÷) that will make an interval containing the 68% of the distribution. A rough estimate of the magnitude of the GSD may be obtained by averaging the ratio of a few pairs of randomly chosen data points. This is because the GSD is related to the average ratio of the data points, similar to the normal standard deviation being related to the arithmetic difference of data points. This is quick, but very rough, method to estimate the GSD. Analytic procedures for estimating distribution parameters are described in the Appendix. Next, I show that values for GSD are often in the range of 2 to 3. Observations of Radon Variation in California It is provocative to ponder the life of a radon-222 atom. As it progresses from birth in the soil as radium daughter to its final death in decay as it transmutes into polonium-218, there are numerous physical factors which influence its path. While it may be possible to estimate the branching ratio for the many possibilities, I chose instead to simply look at the final indoor concentration as measured by a radon detection device. In the fall of 1990, The California 
 59 
 Department of Health Services performed numerous (>2000) short-term measurements of radon concentrations in homes in California (Liu, 1990). County number Median number Median GSD Orange 41 0.8 1.52 Tuolumne 30 1.1 1.86 Santa Clara 84 0.8 2.87 San Joaquin 30 1.1 2.94 Tulare 73 1.5 1.89 Placer 102 0.5 3.04 122 1.1 1.92 Marin 68 0.5 3.13 San Diego Contra Costa 73 0.6 2.07 101 0.3 3.15 73 0.7 2.07 Sonoma Santa Barbara 120 0.6 3.35 Los Angeles 89 0.7 2.11 Napa 33 0.6 3.49 123 0.9 2.28 Solano 59 0.5 3.50 Alameda 79 0.7 2.29 San Mateo 50 0.4 3.57 Siskiyou 37 0.7 2.32 Shasta 96 0.4 3.73 Butte 51 0.5 2.39 Nevada 32 0.6 4.07 159 0.7 2.48 Sacramento 68 0.3 4.49 45 0.8 2.83 Humboldt 50 0.1 5.42 Kern Fresno Ventura El Dorado GSD County TABLE I. Summary values for radon concentrations in selected California counties These data have been published and the central portions of the distributions appear to be well approximated by lognormal distributions. The logarithms of all the data points have been used to calculate their mean and standard deviations in each of the counties. The exponentials of these values have been evaluated and these are listed as the medians (geometric means) and GSD’s (geometric standard deviations) in Table I. These are statistics and I treat them as unbiased estimates best estimates of the corresponding population parameters. A more complete discussion would involve finding a confidence intervals. The counties have been ordered in ascending GSD. An attractive feature of these data was that they were all taken at approximately the same time of year (the fall). According to Nero (Nero, 1986) an important source of radon variation is the outdoor temperature. Thus, the radon measurements in the winter are about a factor of 2 larger than the measurements in the summer. The ability to average over variations due to weather (temperature, wind and rain) is part of the rationale in making a long-term (1-year) measurement. On the scale of a county, taking all the measurements at the same time would also reduce the variation due to weather; all the homes would have approximately the same ambient conditions. So the GSD’s in Table I are primarily due to variations in the buildings and the soils. Most of the GSD factors have values between 2.0 and 3.5. These statistics are calculated by simple direct manipulation of the county data. In a few cases it is clear that there is more than one lognormal population in a county. If the data are ordered and plotted on lognormal paper, most of the points will be in a straight line until the last few points. 
 60 
 The points diverge in a systematic fashion. This indicates the occurrence of a high-radon subpopulation, a geologic radon hot spot. This was discussed in some detail in a previous paper by Meada and myself (Hobbs, 1996). The logradon data clearly follows a dominant Gaussian distribution until well out in the tail, then the last few data transition to another Gaussian: a bump on the tail. The GSD can always be calculated, but in a multimodal case, its magnitude is abnormally increased by the bump. Several California counties show this evidence of a subpopulation hot spot. This is the case in Santa Barbara County. Here we have 95% of the homes on nominal California radium soil (about 1-2 ppm eU) and the remaining 5% on Rincon soil (about 25 ppm eU) (Rosen, 2002). The official California survey completely omitted Summerland which is built exclusively on this soil. I have been taking measurements in Summerland for the last 20 years. Here are my 36 measurements in Summerland. 9.5
 2.0
 30.1
 21.1
 4.3
 2.9
 1.9
 6.1
 11.1
 3.5
 3.5
 2.1
 3.8
 1.1
 1.9
 5.1
 2.4
 3.6
 4.7
 5.9
 6.0
 6.8
 1.0
 4.5
 2.7
 12.3
 5.5
 5.4
 1.3
 1.4
 3.5
 2.0
 4.7
 3.4
 10.6
 1.8
 TABLE II. Hobbs radon measurements (pCi/l) in Summerland, CA 93067; 1990-2010. These measurements were often made pursuant to real-estate transactions and were short-term (2-4 days). Sometimes the houses were empty, but sometimes there were occupants. They were made throughout the years, but I have more business in the springtime. Summerland is right on the coast and doesn’t have much weather (marine temperate); the temperature is almost always 45-85ºF. What all the houses have in common is the black Rincon soil. The logradon data is reasonably represented by a Gaussian distribution or equivalently the radon data is reasonably represented by a lognormal distribution. The data and curve are compared. 
 data curve geometric mean geometric SD 4.0 2.18 4.0 2.18 median arithmetic mean 3.7 5.2 4.0 5.5 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 8
 
 61 
 min first quartile (Q1) median (Q2) third quartile (Q3) max 1.0 2.05 3.7 0 2.3 4.0 5.95 30.1 no bound (3 data w/ 3.5 pCi/l) mode 3.5 6.5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 distribution
 
 
 
 Figure 2.4 2. Summerland radon The geometric mean and standard deviation are calculated from the radon measurements. These statistics become the parameters to generate the continuous probability distribution. In the table the characteristics of the parametric curve are compared with the basic statistics. There are small discrepancies between the data and the theoretical distribution which are expected. The soil of Summerland is quite homogeneous. In addition to its high uranium content, another important property is that it is very expansive when wet. It is usually dry and forms a spider-web of cracks. This is actually its most identifying characteristic for me. The numerous cracks would lead to enhanced gas-emanation. This, in turn, enhances the probability that homes on the Rincon will contain unacceptable levels of radon. Even so, all the homes in Summerland have essentially identical soils. I estimate that the soils lead to a small variation in the indoor radon levels; less than 10%, so σs2 < [ln(1.1)]2 = 0.009. The temperatures of Summerland are very mild. Because of the fog and chill at night and mornings, it is common for heating year round in the homes. (We have just completed a record summer in Southern California. There were numerous records set for low temperatures. The high temperature in July was a cool 64oF.) There are occasional breezes, inversions, and showers but I estimate that the weather may contribute less than a 30% variation in radon level; this would mean σw2 < [ln(1.3)]2 = 0.069. The remaining source of uncertainty in the indoor radon comes from the characteristics of the homes and their operation. This is related to the magnitude of soil gas infiltration rate compared to the total ventilation rate of the house. If this were expressed as a percentage, we would expect a distribution similar to the Summerland radon distribution shown in Fig. 2 above. At this point we can use the data and our assumptions to make an estimate of the building-factor uncertainty, σb2. Rearranging Eq. 6, σb2 = σg2 - σs2 - σw2 
 (8) 62 
 From the measurements σg = ln(GSD) = ln(2.18) = 0.779, substituting __________________ σb = √0.607 – 0.069 – 0.009 = 0.727 (9) The exponential of this is a geometric factor representing the building effects of radon, a geometric standard deviation for the building GSDb = exp(0.727) = 2.07 (10) Building design features and flaws along with the variation in the operation of the building contribute, on the order of, a factor of 2.0 to the indoor radon level. In addition, this value is consistent with various data from the State of California (Table I). Conclusion and Summary The distribution of indoor radon is important to understanding its nature and to developing a strategy to protect citizens from radon exposure. A scientific survey is critical in generating radon data. Many data sets are self-selected (voluntary) and are essentially useless in making inferences. Self-selected data can be used to indicate that radon measurements have been made, but little else. The specific radon concentration in a home is the result of numerous factors. As such the central part of the distribution is well approximated by a lognormal distribution. The logarithms of the data form a Gaussian “bell-shaped” profile. The fundamental parameters of the Gaussian curve are the mean (log geometric mean) and standard deviation (log geometric standard deviation). When plotted against radon (no log), the distribution is strongly skewed to the right, containing points with relatively high magnitude, several times the mean. The broad radon distribution emphasizes some platitudes about testing. Since the radon concentration results from many factors, the best way to determine a radon level is to make a measurement. Also, radon levels may vary from house to house in seemingly unpredictable ways. And most important, almost any house could possibly have a radon level greater than the EPA action level of 4.0 pCi/liter. 
 
 63 
 Much of the tedious work with radon data can be obviated by simply taking the logarithm of the radon level first and using the logradon values as the basic data. The logradon values have a simple normal distribution. If you have a random sample you can find confidence intervals for the parameters, compare samples from different populations, etc. A reasonable value for the logradon standard deviation is simply 1.0 (or a geometric standard deviation of about 2.7). These are reasonable values for back-of-the-envelope calculations. It is easy to make a histogram of the logradon values and check if it appears consistent with a normal distribution. The many factors which contribution to the specific radon concentration in a home can be collected into three groups: the soil factors, the weather factors, and the building factors. Each of these is itself a complicated collection of factors and parameters. Thus, it is reasonable to deduce that each of these factors is well represented by a lognormal distribution. This leads to the quadrature formula (Eq. 6) for the log of the geometric standard deviation σg2 = σs2 + σw2 + σb2 where the terms on the right are the squares of the logs of the geometric standard deviations (GSD’s) of the factor groups. By looking at sets of radon data collected under different conditions, it appears that each of the geometric standard deviations for the groups can be as large as 2 or even larger. This means that each group of phenomena could cause variation in radon concentration on the order of a factor of 2. For a lognormal distribution a factor one GSD contains 68% of the data. A simple example is the comparison of the 1-year radon measurements compared with the 4-day measurements. Long-term measurements average over the ensuing radon variation due to meteorological phenomena and consequently have a smaller GSD. Nero, et al., (Nero, 1986) discuss this further. Both the mean and the standard deviation contribute the indoor radon level. Carlisle & Azzouz. (Carlisle, 1993) note that as the housing stock ages, there is more soil-gas infiltration resulting in a rise of the mean value. The houses do not all age the same way and the distribution will be broadened. I have personal experience with this phenomenon. About 15 years ago, I was asked to make measurements of a cross section of homes in the Winchester Canyon area of Goleta, California. This is near an area of Rincon soil. All of the measurements had magnitudes less than 1.1 pCi/l. Recently, I have had occasion to test some of these homes for real-estate transactions. The recent measurements have increased and fall in the range 1.2-4.5 pCi/l. Both of these sets had a few data points (about 5), but they seem to validate the earlier results (Carlisle, 1993). The building set of radon factors is the one over which we have most control. We severely limit the amount soil gas infiltrating into a building by changing the building characteristic with a mitigation system. An understanding of the distribution of indoor radon concentrations, as outlined in this paper certainly validates the requirement of building with radon-resistant construction techniques in high potential regions. These are regions where the other sets of factors give rise to a high probability that the home would have a high average radon 
 64 
 concentration, particularly areas of low ambient temperatures (weather) and high soil radium content (soil). These techniques include such things as a soil-gas ventilation system built into the foundation. There is a tremendous need to develop requirements for energy efficient homes. This should be done taking into consideration the total indoor environment including all the various indoor air pollutants including radon. With this in mind, it is hoped that the various radon-resistant features of home construction will become common place in the future. It is also important that they be built as robust as possible to continue preventing radon entry for the life of the house. References 1. 
 W. E. Hobbs and L. Y. Maeda, “Identification and Assessment of a Small, Geologically Localized Radon Hot Spot,” Environment International Vol. 22, pp. S809-S817, Elsevier Science Ltd, 1996. 65 
 2. R. Blood, Radon Database for California, October 2002. http://www.consrv.ca.gov/cgs/minerals/hazardous_minerals/radon/DHS_Radon_Database. 3. D. Carlisle and H. Azzouz, “Discovery of radon potential in the Rincon Shale, California – A case history of deliberate exploration,” Indoor Air 3, pp 131-142. Munksguard, 1993. 4. K. S. Liu, S. B. Hayward, J.R. Girman, B. A. Moed, and F. Y. Huang, “Survey of residential indoor radon concentrations in California,” Final Report CA/DOH/AIHL/SP-53, Berkeley, CA, 1990. 5. S. B. Hayward, private communication. 6. A. V. Nero, M. B. Schwehr, W. W. Nazaroff and K. L. Revzan, “Distribution of Airborne Radon-222 Concentrations in U.S. Homes,” Science, Vol 234, pp. 992-997, 21 Nov 1986. 7. K. H. Wedepohl (ed.), Handbook of Geochemistry, Berlin, Springer-Verlag, 1969. 8. Art Rosen, Physics Professor, Cal Poly San Luis Obispo. Several canisters of Rincon Shale soil were assayed using the gamma spectrometer at the radiological lab of Cal Poly SLO in 2004. There was also a canister from San Luis Obispo County which showed the same as Santa Barbara County; ppm eU mean parts per million equivalent uranium (by mass). Problem: What is the mean of the population of homes with a nominal lognormal radon distribution of GSD = 2.7 (σg=1) that has 20% of the homes with radon levels greater than 4.0 pCi/l. The z-value for this is inverseNorm(1-0.2) =0.84. Using Eq. A5, from ln r = µ g + ! g z we have ln 4 = µg + (1)(0.84), so µg = 0.545; the median is e0.545 = 1.7 pCi/l and the arithmetic mean is calculated r = exp (µ g + ! g2 / 2 )= 2.8 pCi / l 
 
 66 
 Appendix Evaluating parameters for a radon lognormal distribution The notation and methods of working with the lognormal distribution are reviewed. The lognormal distribution is a probability distribution P(r) of the form # %(ln r % µ g ) 2 $ exp & ' 2! g2 &( ') P(r )dr = dr 
 r 2"! g2 or # %(ln r % µ g ) 2 $ exp & ' 2! g2 &( ') P(ln r )d (ln r ) = d (ln r ) 
 (A1) 2"! g2 The expression on the left is in normal units (e.g. pCi/l) and has a distribution skewed to the right. The expression on the right is in logradon units and is perfectly symmetrical. They are equivalent. In these µ g is the natural log of the geometric mean and ! g is the natural log of the geometric standard deviation; both are in logradon units. The geometric mean is also the median (also called the 2nd quartile or 50th percentile). The arithmetic mean is " ln r = # r P(r ) dr = µ g + ! g2 / 2 
 
 
 
 
 
 (A2) 0 In the linear theory of cancer, the risk is proportional to r = exp (µ g + ! g2 / 2 ). There values in the tail are accentuated by the skewness through the ! g term. The mode rm is found at dP/dr = 0 rm = exp (µ g " ! g2 ) (A3) This means the densest part of the distribution will be less than the median. It will be common to make a measurement that results in a value at this low level concentration. The specific probability of a given interval of radon concentrations for population ( µ g , ! g ) can be found using standard normal curve values, the normal cumulative distribution function. In particular, it is interesting to estimate the proportion of houses in a population that would have indoor radon levels greater than 4.0 pCi/liter. P (r > 4) = normalcdf (ln 4, ", µ g , ! g ) = 1 # normalcdf (#", ln 4, µ g , ! g ) (A4) 
 Again, we point out that the magnitude of the term ! g is important in this evaluation. Evaluation of the geometric parameters—is straightforward working from the premise that the data is from a lognormal population. The first step is to take the logarithms of all the radon measurements. As discussed in the body of the paper, the factors contributing to the magnitude of the indoor radon level generally combine through multiplication and this results in a lognormal distribution. We assume that the radon data is from a lognormal population. If all the data are known (like in Summerland), then the mean and standard deviation of the radon logarithms are estimates of the parameters ( µ g , ! g ) . Usually, however, the low-magnitude measurements are only roughly known. This is sometimes indicated by being below a limit. For 
 67 
 example, it is common to see a radon measurement simply stated as less than 0.5 pCi/l. This makes the direct
calculation impractical. Other methods for estimating the mean and standard deviation of the distribution make use of the standard normal curve. The dimensionless z-value for a radon datum r is defined so that ln r = µ g + ! g z (A5) The z-value itself is associated with a probability or position in the distribution. Suppose, for example, there are 55% of the data with values less than r1 = 1.0 pCi/l and 20% of the data between 1.0 and r2 = 2.0. The z-values are inputs to the cumulative standard normal distribution, so they can be found with the inverse-cdf. The values are z1 = inverse-normal 55% = 0.1257 and z2 = inverse-normal 75% = 0.6745. Combining with the logs: ln 1 = 0 and ln 2 = 0.6931, the equations are 0 = µ g + (0.1257)! g (A6) 0.6931 = µ g + (0.6745)! g The solution is µ g = !0.1587 and ! g = 1.2630 . These are logarithms and their exponentials are the geometric mean M = 0.9 pCi/l and GSD = 3.54. For the California county data, there were scientific samples of reasonable size taken for most of the counties. The magnitudes of these data less than one were simply given as “< 1.0”. These were counted and the z-value associated with their fraction were the first point as above. Then the remaining data were ordered and the z-values associated with their cumulative fraction calculated. When the log of the data is plotted as a function of the z-value, invariably the main body of the distribution would be characterized by a surprisingly straight line. The log of the geometric mean µ g is its y-intercept and the log of the geometric standard deviation ! g is its slope. These can be evaluated by the least-squares method. This is a good approach since it gives equal weight to each data point and the errors in the position of individual points would be averaged out when many are used (an application of the law of large numbers) . The population of interest is assumed to have a lognormal distribution. This allows confidence intervals to be calculated for the various parameters. Consider Ventura County in California as an example. There were 159 randomly chosen homes with radon data; 90 had radon measurements less than 1.0 pCi/l. When the other 69 data points are plotted on lognormal probability graph paper they clearly form a straight line. The y-intercept and slope are found to be µ g = !0.21 and ! g = 0.91 which are the unbiased estimates. The 95% confidence intervals are found to be "0.4325 < µ g < "0.1475 0.819 < ! g < 1.023 (A7) The confidence intervals go inversely with
 n , and 40-50 data points results in a reasonable bound. The best estimates for the arithmetic mean of Ventura County homes is r =1.1 pCi/and the percentage of homes above 4 pCi/l is p4 = 3.2%. The 95% confidence intervals for these statistics are 0.9 < r < 1.46 pCi / liter (A8)
 2.2% < p4 < 4.7% 
 68 IMPROVING INDOOR AIR QUALITY BY REDUCING RADON AND VAPOR INTRUSION THROUGH THE USE OF ETHYLENE VINYL ALCOHOL (EVOH) R. B. Armstrong1, M. ASME, M. SPE, M. IGS. Department of Research and Technical Services, Kuraray America Inc., 11500 Bay Area Boulevard, Pasadena, TX 77507; PH (281) 474-1576; FAX (281) 474-1572; email: robert.armstrong@kurarayamerica.com Keywords: indoor air quality, radon diffusion, ethylene vinyl alcohol, EVOH, coextrusion, diffusion resistance, volatile organic compound transport. Abstract Ethylene Vinyl Alcohol (EVOH) is a random copolymer of ethylene and vinyl alcohol, commonly used as a barrier to hydrocarbons in automotive fuel systems and agricultural pesticide and herbicide solvents. EVOH provides extremely high resistance to the migration of gases and volatile organic compounds (VOC’s). The incorporation of EVOH in radon resistant new construction has the potential to dramatically reduce the diffusion of radon and other harmful vapors into building spaces and significantly improve indoor air quality. Recent tests of the radon diffusion coefficient of EVOH are compared to previously published results, which show EVOH has radon barrier properties several orders of magnitude better than polyethylene such as LDPE or HDPE. For radon resistant new construction (RRNC) the use of EVOH in a composite with commonly used materials such as HDPE, LLDPE or PP would dramatically reduce the diffusion of radon and VOC’s through plastic sheeting or vapor retarders. Potential applications for a high barrier vapor retarder (HBVR) include radon and vapor barrier membranes for RRNC and brown field remediation. Introduction Radon is a naturally occurring gas, which decays with a half life of approximately four days and emits radioactive alpha particles that are now known to be the primary cause of lung cancer in non-smokers, and the second most important cause of lung cancer after smoking. The International Agency for Research on Cancer (IARC), a World Health Organization (WHO) agency specializing in cancer, and the US National Toxicology Programme has classified radon as a human carcinogen. In 2009 the WHO Handbook on indoor radon reported that recent estimates of the proportion of lung cancers attributable to radon range from 3 to 14%. In the United States approximately 21,000 people die each year from radon-related lung cancer. The dose-response relation seems to be linear without evidence of a threshold, meaning that the lung cancer risk increases proportionally with increasing radon exposure(WHO, 2009) The WHO recently reduced the recommended maximum level of radon gas to 100 Becquerel’s per cubic meter 1 The author is an employee of Kuraray America, Inc. a producer of ethylene vinyl alcohol (EVOH) resin. 69 (Bq/m3) or 2.7 picoCurie per liter (pCi/L), which is ten times lower than the recommended maximum level corresponding to a 10-3 risk level in 1996 (WHO, 1996). In the US and Canada a recommended action level of 4 pCi/L has been established for a number of years. In the US and Canada national agencies set guidelines for radon levels in residences and workplaces, and guidelines for radon mitigation and radon resistant new construction. In the US each state sets local guidelines or standards which vary significantly. Radon resistant new construction and overall indoor air quality are also within the scope of several of the new “green” building standards which address the related issue of vapor intrusion into buildings sited on or adjacent to brownfield sites. Vapor intrusion or VI is defined as the migration of volatile chemicals in building spaces from groundwater or soil. The chemicals of primary concern were either hydrocarbons such as benzene, ethyl benzene, toluene and xylene (BTEX) or chlorinated solvents such at trichloroethylene (TCE) or perchloroethylene (PERC). The chronic effects of long term exposure to very low levels of these chemicals motivated the EPA to issue guidelines for limits on vapor intrusion in 2002 to 2005, which are now being acted upon by most states, with California, Colorado, Wisconsin, New York and New Jersey being the most active. The action level for remediation or mitigation based on VI is often very low, typically being the level of a volatile chemical with a one in a million chance of causing cancer over a 20 year period (translates to 1 µg/m3 for toluene in indoor air). To prevent radon and vapor intrusion into residential and commercial buildings a variety of techniques are employed, typically a passive barrier or active venting system or a combination of the two, which can be part of building construction or mitigation after radon or vapor intrusion is detected. A passive barrier system normally includes a membrane that forms a barrier between the ground and the foundation of the building. The installation of the barrier membrane must be carried out with care so gas cannot enter via overlaps and pipe penetrations that are not sealed. To prevent gas build-up beneath the membrane, it is necessary to provide a means for gas to disperse into the atmosphere. Radon testing and mitigation efforts are not keeping pace with construction of new houses that suffer from radon problems, so promoting radon resistant new construction is becoming much more important. Current EPA guidelines suggest using 3mil (75 micron) polyethylene sheeting as a radon barrier in conjunction with an active vent system, however houses built to the current standard can exceed the action limit and require additional mitigation efforts. Several groups including the Czech Technical University have tested a variety of materials offered as radon barrier membranes and suggested that determining the radon diffusion coefficient of barrier membranes is a key step in determining the optimum type and thickness of radon barrier membranes. Properties of EVOH Ethylene Vinyl Alcohol (EVOH) is a random copolymer of ethylene and vinyl alcohol widely used to protect materials from oxidation and for containment of volatile organic hydrocarbons because of its outstanding barrier to gases, solvents and hydrocarbons (Lagaron et al. 2001). EVOH offers extremely good resistance to the migration of volatile organic compounds (VOC’s), hydrocarbons and organic solvents, with the rate of solvent diffusion in EVOH being several orders of magnitude lower than in polyethylene. 70 Many polymers exhibit softening, swelling or environmental stress cracking when exposed to solvents, while EVOH retains key physical properties in the presence of organic solvents, acid and alkali solutions and non-ionic surfactants. The morphology of EVOH is a combination of a highly ordered crystalline structure interspersed with disordered amorphous regions. The permeability of EVOH is controlled by factors such as crystallinity, chain stiffness, free volume, cohesive energy density, and extrinsic factors such as temperature and moisture (Koros, 1990). Like other hydrophilic polymers, EVOH can exhibit large permeability increases as temperature or moisture content rises due to an increase in the free volume (Lopez-Rubio et al 2003). By varying the ethylene content (mol % of ethylene) and volume fraction of crystallinity (Φ) of EVOH this effect is minimized (Armstrong, 2003). Table 1 illustrates the magnitude of differences in material properties by comparing the permeability of EVOH and HDPE to common gases. Table 1 Gas barrier properties of 32 mol% EVOH vs. HDPE Gas EVOH* HDPE** Nitrogen 0.019 190 Oxygen 0.25 2300 Carbon Dioxide 0.6 17526 Sulfur Dioxide 0.3 21844 Methane 0.4 2845 Volumetric permeation rate in (cc.20µ/m2.day.atm) Conditions: 23°C – 0% RH (ASTM D1434T) * ASTM D1434 at Kuraray lab – 32mol% EVOH **Permeability Properties of Plastics and Elastomers, Massey, 2nd Edition Published data shows that the barrier property of EVOH to VOC’s is extremely good. Table 2 compares the diffusion coefficients of trichloroethyelene and toluene in EVOH and HDPE. Note that the EVOH testing was conducted at 100% solution concentration, while the HDPE tests were conducted with dilute solution concentrations ranging from only 2 to 5 mg/L. Table 2 Diffusion coefficient (Dg) of TCE and Toluene in EVOH vs. HDPE Solvent EVOH * HDPE** Trichloroethyelene 3.1x10-17 4.0x10-13 -17 Toluene 3.1x10 3.0x10-13 Diffusion coefficient Dg in m2/s *Kiwa NV report April 2008 for EVAL Europe N.V ** Sangam and Rowe, (2001) Geotextiles and Geomembranes 19 329-357 Experiments & Results To determine the radon diffusion coefficient of ethylene vinyl alcohol (EVOH) and compare this material property with other materials commonly used as radon barrier membranes, samples of 44mol% EVOH film of 0.6mil (15 microns) produced by the Kuraray Company Limited of Japan and designated ‘EF-E’ were submitted to the radon test laboratory in the Civil Engineering Department of the Czech Technical University in Prague. To determine the radon diffusion coefficient of these samples the laboratory followed the method K124/01/09 which is accredited by the Czech Accreditation Institute, and allows for the determination of the radon flux through the tested material placed 71 between two cylindrical containers. Radon diffuses from the lower container, which is connected to the radon source, through the sample to the upper container. From the known time dependent curves of the radon concentration in both containers the radon diffusion coefficient can be calculated. The details of the method are outlined in the test report and also in be found in published papers (Jiránek, 2008). The results of tests conducted by the Czech Technical University are in Table 3, with the radon diffusion coefficient D reported in units of m2/s. The radon diffusion coefficient is reported as a material property of the composite or monolayer sample in all cases. Table 3 Radon Diffusion Coefficient (D) in EVOH vs. other barrier materials Material Radon diffusion coefficient (D) * EF-E (EVOH) 1.3x10-14** Bitumen coated Al foil 3.9x10-14 Polyurethane (PU) coating 2.3x10-12 Chlorinated polyethylene 3.5x10-12 High density polyethylene (HDPE) 5.8x10-12 Polyethylene (PE) 1.0x10-11 Plasticized polyvinylchloride – (PVC-P) 1.9x10-11 Low density polyethylene (LDPE) 2.5x10-10 *Radon diffusion coefficient D in m2/s **Czech Technical University Test Report No 124008/2010 1-4-2010 All other values from Jiránek, Rovenská and Froňka (2008) The radon diffusion coefficient for the EF-E sample of 1.3 x 10-14 m2/s with uncertainty of ±0.1 m2/s is one of the lowest values ever reported by the laboratory, even lower than bitumen coated aluminum foil. The reported uncertainty in result for EF-E (EVOH) sample was also significantly lower than for the bitumen coated foils which is attributed to the excellent flex crack resistance of EVOH. Discussion The significance of these results can be illustrated using several of the design methods that determine radon barrier membrane thickness. The design methods currently utilized for such purposes and used for this comparison include: • Method that requires the radon diffusion coefficient D of the radon barrier membrane must be below a strict limit value, which has been suggested should be <1x10-11 m2/s. • Method that requires that the thickness d of the radon-proof membrane must be at least three times greater than the radon diffusion length. • Method that requires the thickness of the membrane be calculated for each house according to the radon diffusion coefficient in the membrane itself, radon concentration in the soil on the building site and house parameters (ventilation rate, area in contact with the soil etc). In the case of the first design method, 44mol% EF-E EVOH has a radon diffusion coefficient of 1.3 x10-14 m2/s which is three orders of magnitude lower than the suggested 72 limit of <1x10-11 m2/s. If the second design method is employed, EVOH can be compared to HDPE and LDPE using the design factor requiring that radon barrier membrane thickness should be three times greater than the diffusion length. The diffusion length itself is a function of material properties and the established equation: l = (D/λ)1/2 [m] Equation 1 Where l is the diffusion length in meters (m) D is the radon diffusion coefficient (m2/s) λ is the radon decay constant which is 0.00756 hr-1 or 2.1x10-6 s-1 The results of diffusion length design method calculations are in Table 4. The required thickness for EVOH is within the capabilities of existing producers of radon barrier membranes, however a 236 micron (9mil) thick barrier membrane of EVOH would be relatively expensive and would not possess other functional characteristics such as flexibility during installation and easy heat sealability that typical polyolefins such as LDPE, HDPE and PP possess. Table 4 Radon barrier membrane thickness by diffusion length method Material l (m) 3l (m) 3l(microns) 3l (mil) EVOH 7.86x10-5 2.36 x10-4 236 9 -5 -4 HDPE 166.2x10 49.86x10 4986 196 LDPE 1091.0x10-5 327.3 x10-4 32733 1289 The third design method detailed by others (Jiránek, Rovenská and Froňka 2008) involves a calculation of the minimum thickness of membrane required to control diffusion of radon into a defined control volume (house or building), assuming intrusion of radon by convection to be negligible. By way of a performance comparison the minimum thickness of membrane required for a typical two storey house with a full basement using EVOH, HDPE or LDPE was calculated using the design calculation proposed by Jiránek, Rovenská and Froňka. d ! l. arcsin h # 1 .l.".C s (A f + Aw ) C dif .n.V [m] Equation 2 Where d is the minimum thickness of radon barrier membrane in meters (m) l is the diffusion length (m) established by Equation 1 α1 is a safety factor that accounts for inaccuracies in soil gas radon concentration measurements. Values of α1 can be estimated according to the soil permeability (for highly permeable soils α1 = 7, for soils with medium permeability α1 = 3 and for low permeable soils α1 = 2.1) λ is the radon decay constant which is 0.00756 hr-1 or 2.1x10-6 s-1 73 Cs is the radon soil gas concentration (Bq/m3) Af is the floor area in direct contact with the soil (m2) Aw is the area of the basement walls in direct contact with soil (m2) n is the air exchange rate (h-1) V is the interior air volume (m3) Cdif (Bq/m3) is a fraction of the reference level of indoor radon concentration Cref caused by diffusion. In accordance with the method detailed by Jiránek, Rovenská and Froňka the value of Cdif was set at 10%, meaning that the radon diffusion into the house will be limited to 10% of indoor radon concentration, with the balance reserved for radon intrusion by convection. The design calculation input for a typical size two story house with a full basement was: α1 - safety factor was 3 for soils with medium permeability λ - radon decay constant was 0.00756 hr-1 or 2.1x10-6 s-1 Cs - the radon soil gas concentration at 74,000 (Bq/m3) or 2000 (pCi/L) Af - the floor area in direct contact with the soil was 220 m2 Aw - the area of the basement walls in direct contact with soil was 98 m2 n - the air exchange rate was 15 h-1 V - the interior air volume was 600m3 Cdif – was 0.1 The results of specific membrane thickness design method calculations are in Table 5. The thickness of EVOH required in this typical case was only 3.7 microns (0.14 mil), which is at least two orders of magnitude lower than the thickness of HDPE and four orders of magnitude lower than LDPE. This is consistent with all established data and experience with the material properties of EVOH, with barrier properties being typically several orders of magnitude better than polyolefins. Table 5 Radon barrier membrane thickness by specific membrane thickness method Material d (m) d (mm) d(microns) d(mil) EVOH 3.67 x10-6 0.0037 3.7 0.14 HDPE 1.45 x 10 -3 1.45 1447.7 57.0 -2 LDPE 2.80 x 10 28.00 28000.4 1102.4 Although material property data and design calculations indicate that a minimal thickness of EVOH would be sufficient to provide significantly improved protection from radon and vapor intrusion, very thin films of EVOH are not practical for installers of radon and vapor intrusion systems. A potential answer lies in the use of coextrusion. Coextrusion allows for the combination of materials in a composite with properties optimized for a target application. In industries as diverse as food packaging, agricultural chemical containers, automotive fuel tanks, pipe and tube, flexible or rigid coextrusions of five to seven layers and varying geometries exist. Utilizing materials such as EVOH in a radon barrier membrane would require at least three layers in a coextrusion, although five or six layer structures would allow for optimizing the amount of EVOH at perhaps no more than 2 to 4% of the total thickness of a radon barrier film – such as a 3 mil PE (75 micron) film with 4% EVOH. Making this technology transfer less of a challenge is 74 the fact that many of the equipment manufacturers and suppliers well known within the barrier membrane industry, including Gloucester Engineering, Brampton Engineering, Cloeren Inc., Davis Standard Inc and Extrusion Dies Industries LLC also supply equipment to industries that are currently utilizing multilayer coextrusion. A model high barrier radon or vapor membrane structure containing a layer of EVOH is presented below in Figure 1. It should be noted that there is a wide variety of possible structures and manufacturing methods that could be employed to produce radon and vapor barrier membranes, and this example by no means exhausts the design possibilities. Polyethylene or polypropylene Adhesive EVOH SUBSOIL – EVOH is not in contact with ground Figure 1 Model radon and vapor barrier membrane with EVOH Conclusion Tests of the radon diffusion coefficient of ethylene vinyl alcohol (EVOH) have shown that EVOH has radon barrier properties several orders of magnitude better than polyethylene such as LDPE or HDPE. Designers, builders and regulators could take advantage of these properties for radon resistant new construction (RRNC) where the use of EVOH in a composite with commonly used materials such as HDPE, LLDPE or PP would dramatically reduce the diffusion of radon through plastic sheeting or vapor retarders. An opportunity exists to create and utilize a superior passive radon barrier membrane that consistently and reliably controls radon levels in buildings and houses to < 1 picoCurie per liter. To realize this goal will require more than lab scale testing of the radon barrier properties of EVOH alone. A demonstration of a model high performance radon and vapor intrusion barrier membrane in a housing development or subdivision may be required to overcome conventional thinking that all vapor barriers (polyethylene) are the same, and must always be supplemented with an active venting system. Other potential applications for a high barrier membranes incorporating EVOH include VOC barriers and in brown field remediation, where EVOH could dramatically reduce the diffusion of harmful vapors into building spaces and significantly improve indoor air quality. 75 References Armstrong, R., Effects of Polymer Structure on Gas Barrier of EVOH and Considerations for Package Design. In TAPPI PLACE Conference 2002, Boston, MA, September 9-12, 2002 [CD-ROM]; The Association of Pulp and Paper Industry Conference on Polymers, Laminates, Adhesives, Coatings and Extrusions: Norcross GA, 2003; Session 4, Paper 3. Crank, J. The Mathematics of Diffusion, 2nd ed.; Oxford Science Publications: Oxford, UK, 1975. Czech Technical University Test Report No 124008/2010 Dated 1-4-2010. Test conducted at the Czech Technical University in Prague, Faculty of Civil Engineering Test Laboratory accreditation No. 1048 OL 124. Jiránek M., Froňka A.: New Technique for the Determination of Radon Diffusion Coefficient in Radon-proof Membranes. Radiation Protection Dosimetry, 2008. Jiránek M., Rovenská K., and Froňka A.: Radon Diffusion Coefficient – A Material Property Determining the Applicability of Waterproof Membranes as Radon Barriers: Proceedings of the American Association of Radon Scientists and Technologists 2008 International Symposium Las Vegas NV, September 14-17, 2008. Koros, W.J. In Barrier Polymers and Structures; Koros, W.J., Ed.; American Chemical Society: Washington, DC, 1990; pp 1-22. Lagaron, J. M.; Powell, A.K.; Bonner, J.G. 2001. Polymer Testing, 20/5, 569-577. Lopez-Rubio, A.; Lagaron, J.M.; Gimenez, E.; Cava, D.; Hernanandez-Munoz, P.; Yamamoto, T.; Gavara, R. Macromolecules 2003, 36, 9467-9746. WHO handbook on indoor radon: a public health perspective; Zeeb H., and Shannon F., Ed.; World Health Organization: Geneva, Switzerland, 2009. WHO report on indoor air quality: a risk based approach to health criteria for radon indoors; Suess H, and Cardis E, Ed.; World Health Organization: Regional Office for Europe, Copenhagen, Denmark, 1996. 76 EXAMINATION OF OHIO INDOOR RADON DATA Ashok Kumar1, Akhil Kadiyala1, Vijay Devabhaktuni2, Arjun Akkala2, Dilip Varma Manthena1 Department of Civil Engineering1, Department of Electrical Engineering and Computer Science2 The University of Toledo, 2801 W. Bancroft Street, Toledo, OH 43606 Abstract It is estimated that 700 to 1300 people die annually in Ohio from radon-induced lung cancer. The Ohio Radon Information System (ORIS) available online at http://www.radon.utoledo.edu has been developed to create the much-needed awareness among Ohio’s citizens on radon concentrations in their localities and the problems associated with radon exposure. Over the years, the database (made available by testing laboratories, Ohio Department of Health (ODH), Ohio Environmental Protection Agency (EPA), and universities) has been expanded to 159,340 radon observations from homes; 1,341 radon observations from schools; 1,283 radon observations from drinking water; 28,062 radon observations from licensed mitigation contractors; and 77,581 radon observations from licensed testers and specialists. This paper presents various statistics of the analyses performed on the radon databases developed in the last 22 years. For various reasons, radon concentration data are not available for each and every zip code, in Ohio. In places where data are unavailable, radon data may be estimated by applying effective interpolation techniques allowing for comprehensive radon mitigation planning. This paper presents two interpolation techniques for estimating radon concentration values in missing zip codes. Initial results concerning the relative performance of such techniques are shown, and the impact of the interpolated data on radon awareness in Ohio is discussed. Introduction Radon is a colorless, naturally-occurring, radioactive, inert gas formed by the natural breakdown of uranium in soil, water and rock. Radon gas drifts upward through the ground to the surface of the soil and seeps into the buildings through foundation cracks. Radon gas is formed naturally by radioactive decay of uranium present in geologic materials. The major sources of radon gas in Ohio are ‘Ohio shale’ and soil. Elevated radon levels have been discovered in every state. The United States Environmental Protection Agency (U.S. EPA) estimates that as many as eight million homes throughout the country have elevated levels of radon (EPA Report, 2004). If the indoor radon concentrations exceed the EPA recommended action level of 4 pCi/l, immediate measures should be taken to reduce the radon level to 2 pCi/l (Kumar, 2001). Radon exposure is responsible for about 21,000 lung cancer deaths per year in the United States (EPA Report, 2003). Ohio Department of Health (ODH) initiated an indoor radon gas program in the late 1980’s to reduce the number of deaths attributable to radon. In the 1990s, ODH started encouraging the reduction of radon concentrations in houses and schools to a safe level through a number of mitigation methods. In 2001, Ohio passed a law that required radon mitigation contractors to 
 77 report mitigation data on homes to the ODH (ODH Chapter, 2001). As a result of the new law, two new databases were developed to study radon mitigation systems and to track observations by testers. The University of Toledo (UT), under several research grants from the ODH and the Ohio Air Quality Development Authority (OAQDA), developed the Ohio Radon Information Systems (ORIS), and a website (Harrell, 1993; Harrell, 1991; Kumar, 2001; Kumar, 1990; and Ojha, 2001). As of June 2010, the radon database developed and maintained by UT has 159,340 radon observations from homes; 1,341 radon observations from schools; and 1,283 radon observations from drinking water. In addition, as of June 2009, the mitigation database has been expanded to 28,062 radon observations and the tester database has been expanded to 77,581 radon observations. The purpose of developing and maintaining the ORIS database is to analyze radon data across the state of Ohio and produce results that help create public awareness and understanding of the hazards of radon gas and, therefore, reduce any of its concentration levels in places that surpass the EPA’s action limits. This paper summarizes the key results obtained from analysis of ORIS data and compares the performance of two different interpolation techniques that help predict radon concentrations in unmeasured zip codes. The two interpolation techniques used in this study are kriging and cokriging. Methodology The ORIS consists of five different databases or modules: home database, school database, water database, mitigation database, and tester database.
 Home Database The home database provides information on the radon gas concentrations measured using radon detectors in Ohio homes. The home database was initially handled using the ORACLE/MS Access database (Joshi, 2002). However, due to the yearly increase in the number of radon records in the homes database it is currently handled by SQL Server 7.0/MS Access and Microsoft Excel 2007. Data for homes is provided by various organizations and radon testing laboratories as electronic files. The raw data is processed before inclusion in the database. As of June 2010, there are 159,340 radon data points in the homes database. Queries were built in MS Access to analyze the radon statistics based on counties and zip codes. The statistics computed include maximum (Max.), minimum (Min.), arithmetic mean (AM), geometric mean (GM), standard deviation (SD), coefficient of variance (CV), median (Md), quartile 1 (Q1), and quartile 3 (Q3). Using Geographical Information Systems (GIS) software, zip code and county maps are drawn to visually represent the radon concentrations in Ohio. These statistics and color-coded GM maps, for the state of Ohio based on counties and zip codes, are available online on the ORIS website. Queries were also built to identify zip codes and counties with radon GM concentrations > 4 pCi/l and 8 pCi/l. In this study, the unmeasured zip codes or zip codes where the data were not available were estimated using two different interpolation techniques that include kriging and cokriging. Some studies have provided discussion on estimating radon concentrations for missing zip codes (Kumar, 2007; Manthena, 2009; and Akkala, 2010). School Database 
 78 The School database is very small compared to the home database. MS Access is used to handle the school database. The radon measurements in schools in Ohio are provided by the ODH. Queries such as schools tested by county; number of schools tested each year; schools with radon GM concentrations > 4 pCi/l, > 8 pCi/l, > 20 pCi/l; percentage of rooms having radon concentrations > 4 pCi/l in each school; and schools with more than 15 rooms and radon concentrations > 4 pCi/l are used to analyze the data. The school statistics query provides the number of schools tested in each county, AM, GM, SD, and variance of radon concentrations for 1,341 schools in the 63 counties where data were measured as of June 2010. The visual representation of the county map with the percentage of school rooms with radon GM concentrations > 4 pCi/l is drawn using GIS and is available online on the ORIS website. Water Database The water database contains data collected from both public water wells and private water wells. The public water wells database was provided by the Ohio EPA while radon data for the private water wells were obtained from various Ohio universities’ MS theses and research programs. Queries were built in MS Access which show the zip code, county, radon concentrations, and date of testing for both the public and private water wells. As of June 2010, of the 1,283 water supply systems in the water database; 216 are public water wells, and the remaining 1,067 are private water wells. It is difficult to obtain new data in this area. Mitigation Database Licensed mitigation contractors perform tests and submit the results to the ODH. This data is manually entered in to an Excel spreadsheet. The data contains information of 1) license number of the contractor; 2) name of the contact; 3) phone number of the contact; 4) address of the contact; 5) county, zip code, city and state; 6) system type; 7) start and completion dates; 8) premitigation and post-mitigation levels; 9) quarter of the year; and 10) year in which measurements were made. The missing data values are assigned as “NA” for alpha numeric type and “-1” for numeric values. The mitigation data (Kumar, 2003) is imported to MS Access to analyze, store, and update the mitigation data. Currently (June 2009), the mitigation database consists of 28,062 radon observations. Queries were designed in MS Access for analyzing mitigation data. These queries determine the total number of tests performed by each licensed contractor, average removal efficiency by each type of system, counties with pre-mitigation radon level > 4 pCi/l, premitigation radon level between 4 pCi/l and 20 pCi/l, and pre-mitigation level greater than 20 pCi/l. Tester Database The radon tests conducted by licensed testers are submitted to the ODH and these records are then passed to the UT Civil Engineering Department. As of June 2009, the tester database has been expanded to 77,581 radon observations. The tester database includes information on 1) license number of the tester; 2) name of the contact; 3) street address of the contact; 4) zip code, city, county, and state; 5) device code; 6) test type; 7) start and finish dates of the test; 8) radon concentration level; 9) quarter of the year; and 10) and year in which the measurements were made. 
 
 79 Database maintenance procedures adopted for the tester database are similar to the maintenance procedures adopted for the mitigation database. Tester results are entered manually into an Excel spreadsheet, processed, and checked for accuracy to avoid transcription errors. The data is then imported to MS Access to run queries. Queries are run to determine the number of records tested by each licensed tester, counties with radon level > 4 pCi/l, radon statistics that include the number of radon measurements, Max., Min., AM, GM, SD, and variance based on county and zip codes. The queries are built in such a way that they do not consider missing data during the analysis.
 Interpolation Techniques Radon data is not available in each and every zip code due to inapproachability, cost effectiveness, and time constraints. Two interpolation techniques: kriging and cokriging are investigated in this study to estimate the radon concentrations for unmeasured zip codes. These interpolation techniques use the available radon data in known locations to estimate the radon data for unmeasured zip codes which will help to render an effective plan to mitigate the radon concentrations in Ohio. An overview of the two interpolation techniques is given: Kriging: Kriging is a geostatistical technique, generally used to interpolate the value of a random field (e.g., the elevation, z, of the landscape as a function of the geographic location) at an unobserved location from values at observed locations. This method not only produces prediction surface, but also provides an error and uncertainty surfaces. Kriging is mainly divided into two different functions: predicting and quantifying the spatial structure of the data. This interpolation technique is very flexible and allows the user to investigate graphs of spatial autocorrelation. This technique uses statistical models that allow a variety of map outputs including predictions, prediction standard error, standard error of indicators, and probability. Cokriging: Cokriging interpolation technique is similar to kriging that perfoms better estimates using a secondary variate, sampled more intensely than the primary variate. If the primary variate is difficult or expensive to measure, cokriging uses secondary variate to predict the data without having to more intensely sampling the primary data. This method uses radon as the primary variate and uranium as the secondary variate. Ordinary cokriging technique has been found to give the most reproducible estimations (Ahmed, 1987). More information about these two interpolation techniques can be found in the open literature. Our current research is on the use of Artificial Neural Networks to predict radon concentrations in unmeasured zip codes (Akkala, 2010 and Akkala, 2011). Results and Discussion The analysis of radon data from the five databases help in better understanding the radon problem in Ohio. The analysis of radon data and results obtained from running the queries provide information to the concerned authorities to take necessary steps in evaluating various steps to mitigate radon concentrations to acceptable levels (EPA action limit of 4 pCi/l). Home Database The analysis of home database showed that out of 88 counties in Ohio, 29 have radon GM levels more than 4 pCi/l, with Licking being the only county with radon GM level greater than 8 
 80 pCi/l. Of the 1544 zip codes homes data, 32.64% zip codes have radon GM levels greater than 4 pCi/l and 8.04% zip codes have radon GM levels greater than 8 pCi/l. Based on the 159,340 radon homes database, the GM of radon concentrations in the state of Ohio is 3.99 pCi/l. Maximum radon concentration of 927.6 pCi/l is accounted for the zip code 43952 in “Jefferson” county. Figure 1 provides the visual representation of radon GM concentrations in Ohio on a county and zip code basis respectively, thereby giving a better idea on the radon distribution in Ohio. Figure 1: Geometric Mean of Radon Concentration in Ohio Counties and Zip Codes School Database Sixty three (63) counties and 1,341 schools have been tested under the ODH School Testing Program as of June 2010. Analysis of the radon data in schools on a county basis revealed a school in Belmont County having a maximum radon concentration greater than 85.5 pCi/l. The school also had 11 out of 39 school rooms tested with radon concentration greater than 4 pCi/l. Figure 2 provides the visual representation of counties in Ohio with percentage of school rooms over 4 pCi/l. It can be observed from Figure 2 that Pike County schools have approximately 5060% of schools with radon concentrations over 4 pCi/l. Overall 28% of the schools in Ohio have a potential for at least one room in excess of the EPA action level of 4 pCi/l. Water Database Table 1 provides a summary of the results associated with radon concentrations in private wells. Of the 1,067 private water wells, 65 wells have radon concentration greater than or equal to 1000 pCi/l. Of these 65 wells, 28 are located in Logan County and 13 in Delaware County (Table 1). Seven private wells showed radon concentration over 3000 pCi/l. However, none of the public water supply systems have radon levels more than 1500 pCi/l. Only two public wells showed radon concentration greater than 1000 pCi/l. 
 81 Figure 2: Counties with % of School Rooms with Radon Concentration >= 4 pCi/l Mitigation Database Mitigation testing program results provide an insight into the effectiveness of the program in Ohio. As of June 2009, of the 28,062 radon mitigation observations; 21,607 (76.99%) records are complete with both post-mitigation and pre-mitigation levels. Over the years, there has been an increase in the percentage of complete records submitted. Currently, the percentage of complete records is 94%, vs. 20.38% in 2001. This shows that the efforts of the ODH are productive which increased data quality. Table 2 presents the average removal efficiency by each type of system for the year 2008. The analysis of the mitigation systems revealed that the Sub Slab Depressurization (SSD) system performs well in mitigating the radon concentrations to below 4 pCi/l (refer Table 2). Figure 3 shows the variation of removal efficiency with pre-mitigation and post-mitigation levels for the best mitigation system for the year 2008. It can be observed that efficiency of the system decreases with a decrease in the pre-mitigation level, because it is difficult to reduce the radon concentrations below a certain level. Figure 3 also shows that the removal efficiency decreases with an increase in the post-mitigation level. Tester Database The query results obtained from the tester database provide an insight into the radon concentration levels in different counties and zip codes across Ohio. Of the 77,581 records reported as of June 2009; 77,123 records (99.41%) are complete. Table 3 provides the statistics for counties with GM radon levels >= 4 pCi/l. There are 29 counties that have radon test levels above 4 pCi/l. Harrison (GM = 10.25 pCi/l) and Perry (GM = 8.65 pCi/l) are the two counties having GM radon levels > 8 pCi/l. Table 1: Radon Concentration in Private Water Wells in Ohio Counties 
 82 Counties Total No. of Wells Max. Radon Conc. (pCi/l) Min. Radon Conc. (pCi/l) Butler Champaign Clark Clermont Crawford Darke Delaware Erie Fulton Greene Hamilton Hancock Hardin Henry Huron Logan Marion Miami Montgomery Morrow Ottawa Paulding Preble Sandusky Seneca Union Warren Williams Wood Wyandot 7 80 8 1 78 1 60 181 3 4 2 7 49 1 149 212 74 4 6 93 2 1 3 3 9 5 4 8 6 6 571 1491 1386 163 1021 231 2314 3104 172 703 380 470 996 510 2010 7511 1574 413 637 3425 150 190 782 130 220 334 542 245 560 180 217 73 172 326 13 231 2 20 119 200 213 180 44 510 5 25 26 174 249 25 130 190 184 80 80 82 340 148 200 96 No. of Wells with Avg. Radon Radon Conc. ≥ Conc. (pCi/l) 1000 pCi/l 415.00 0 355.85 2 436.88 1 163.00 0 143.42 1 231.00 0 599.12 13 285.82 6 147.00 0 438.75 0 296.50 0 322.86 0 238.73 0 510.00 0 230.82 7 553.99 28 257.61 3 248.00 0 406.67 0 303.15 4 140.00 0 190.00 0 507.67 0 96.67 0 129.89 0 210.40 0 442.00 0 185.25 0 320.00 0 132.00 0 Table 2: Average Removal Efficiency by Each Type of System for Year 2008 Type of System Number of Records Average % Removal Standard Deviation SSD SUMP/DTD SSD/SMD DTD 2706 604 574 162 82.50 85.40 86.25 86.00 13.25 9.50 9.75 14.25 SUMP/DTD/SMD 41 85.25 10.50 SUMP/SSD 38 77.00 18.00 
 83 SUMP VENTILATON SSD/DTD 33 23 88.33 83.50 4.67 12.67 Figure 3: Variation of Removal Efficiency with Pre-Mitigation and Post-Mitigation Levels for the Best Performing System for Year 2008 Table 3: Statistics for Counties with Tester Radon Concentration >= 4 pCi/l County Harrison Perry Ross Logan Mason Van Wert Erie Carroll Madison Pickaway Darke Marion Auglaize Coshocton Pike Licking Knox Columbiana Champaign Highland Fairfield Preble Guernsey No. of Records 12 9 16 30 2 4 131 34 37 35 17 72 13 8 164 585 138 116 24 6 274 21 4 Max. Min. AM GM SD Variance 50.50 36.60 44.60 46.10 10.10 16.70 273.20 62.10 36.60 22.60 30.10 53.60 13.10 7.80 16.50 296.60 258.00 77.10 36.80 8.80 54.00 21.70 6.70 0.80 1.70 0.50 0.40 5.20 0.80 0.05 0.40 0.50 0.40 0.60 0.20 0.50 0.70 0.50 0.20 0.40 0.20 1.00 2.30 0.20 0.10 1.30 10.74 10.70 8.15 9.13 7.65 6.82 13.34 7.87 7.46 6.31 7.07 8.30 5.52 5.30 5.79 9.75 14.60 8.83 5.72 4.88 7.58 5.37 4.50 10.25 8.65 7.91 7.56 7.25 6.77 6.76 5.69 5.59 5.48 5.32 5.32 5.08 5.08 4.99 4.98 4.95 4.93 4.90 4.88 4.86 4.61 4.50 6.61 12.22 2.65 7.49 3.46 1.20 16.28 8.78 7.44 3.95 7.44 7.83 2.48 2.40 3.10 13.80 26.13 10.09 6.07 83.93 62.29 70.83 86.55 45.29 40.75 104.27 84.24 94.59 69.38 101.99 89.95 54.11 77.56 53.51 135.96 158.11 105.32 59.69 7.41 4.47 97.47 69.15 
 84 Morrow Stark Tuscarawas Delaware Franklin 29 1920 526 1797 7891 22.50 111.00 134.00 735.00 939.00 0.30 0.10 0.05 0.10 0.05 5.24 6.84 7.28 6.62 6.88 4.25 4.21 4.02 3.99 3.98 3.76 9.03 10.01 9.96 10.07 69.50 128.76 127.25 132.88 144.01 Comparative Performance of Interpolation Techniques The radon data set available at The University of Toledo, collected from various radon testing organizations, consists of radon data for 1075 zip codes until 2010, whereas there are 1862 zip codes mentioned in the GIS data file of Ohio. Using kriging and cokriging interpolation techniques, radon concentrations for unmeasured zip codes are predicted and different statistical performance measures are computed to determine the relative performance of the two techniques. Figure 4 presents the spatial distribution of radon concentration maps obtained on using kriging and cokriging interpolation techniques. Both these maps exhibit similar pattern of radon concentration distributions across the state of Ohio. Relatively high concentrations are observed in the central and western parts of Ohio as can be seen from Figure 4. It was also observed that these maps exhibit similar pattern to the uranium distribution in Ohio. The uranium map can be found online on the ORIS website. Figure 4: Spatial Distribution of Radon Concentrations in Ohio using Kriging and Cokriging Interpolation Techniques (Respectively) Table 4 presents the summary of different statistical performance measures used to compare kriging and cokriging interpolation techniques. The performance measures used to compare the two techniques are model bias (MB), normalized mean square error (NMSE), correlation coefficient (Corr.), factor of two (Fa2), fractional bias (FB), and fractional standard deviation (FS) (Hanna, 1991). An interpolation technique is considered to be ideal and perfect if the FB and NMSE are equal to zero. However, no technique is perfect in making accurate predictions. Kriging and cokriging interpolation techniques are deemed acceptable if the performance measures meet the criteria of having (i) NMSE ≤ 0.5, (ii) -0.5 ≤ FB ≤ 0.5, and (iii) Fa2 ≥ 0.80. It can be seen from Table 4 that both kriging and cokriging techniques meet all the three 
 85 requirements of NMSE, FB, and Fa2. One can observe both the techniques to over predict (negative bias) radon concentrations. Also, the predicted results using both these techniques have shown close correlation with the observed data as can be seen from correlation coefficient values in Table 4. One can also observe cokriging interpolation technique is slightly better than kriging interpolation technique, because the performance measures for cokriging technique are found to be relatively closer to the ideal values of FB, NMSE, and Fa2. Table 4: Performance Measures for Interpolation Techniques Performance Measures Observed Kriging Cokriging Mean 3.42 3.53 3.49 Sigma 2.74 2.47 2.34 MB 0.00 -1.25 -1.23 NMSE 0.00 0.47 0.45 Corr. 1.00 0.876 0.883 Fa2 1.00 0.99 0.99 FB 0.00 -0.358 -0.353 FS 0.00 0.101 0.114 Percentage Change in Zip Codes Exceeding 4 pCi/l Based on Kriging and Cokriging Interpolation Techniques Kriging and cokriging interpolation techniques were used to predict radon concentrations in unmeasured zip codes and these results showed that there are significant number of zip codes that exceeded 4 pCi/l. The analysis of available zip code radon data and the interpolated radon data for missing zip codes using kriging technique showed that 32.60% of zip codes have concentrations above 4 pCi/l as compared to 28.68% of zip codes based on measured radon data, while cokriging technique showed 31.90% of the zip codes to exceed EPA action limit of 4 pCi/l. These results indicate that more mitigation work is ahead for radon planners in Ohio. Conclusion An integrated Ohio radon information system has been successfully compiled from the data provided by government agencies, university researchers, and commercial testing companies. The information available from the database is useful in assessing the extent of the radon problems in Ohio’s homes, public water systems, and schools. It was also possible to determine the best mitigation system to control radon gas problem in Ohio homes and identify the counties and zip codes with radon test levels greater than EPA action limit of 4 pCi/l. The radon website developed during the project helps in creating awareness among Ohio’s citizens on radon issue and provides information on steps to reduce radon exposure. After predicting the radon concentrations for unmeasured zip codes using the two interpolation techniques, it was observed that cokriging technique showed relatively better performance than kriging technique. Interpolated data shows that more mitigation work is ahead for many zip codes that were not known before. Acknowledgements The authors are thankful for the research grants awarded by the Ohio Department of Health (ODH) and the United States Environmental Protection Agency (US EPA) to The University of Toledo, which made the development of such a radon gas management system possible. The contributions of earlier investigators of the grants (Dr. Jim Harrell and Dr. Andrew G. Heydinger, and many graduate students who worked on this project over the years) are all greatly 
 86 acknowledged. The authors also acknowledge the contribution of a number of staff members from the ODH. The views expressed in this paper are those of authors. 
 87 References 1. Ahmed S, de Marsily G. Comparison of Geostatistical Methods for Estimating Transmission Data on Transmitivity and Specific Capacity. Water Resources Research 1987; 23: 1717-1737. 2. Akkala A, Devabhaktuni VK, Kumar A. Interpolation Techniques and Associated Software for Environmental Data. Environmental Progress & Sustainable Energy 2010; 29(2): 134-141. 3. Akkala A, Devabhaktuni VK, Kumar A. Spatial Interpolation Techniques for Environmental Data: Theory to Practice, Chapter 2, Advances in Environmental Research, Volume 10, Editor: Justin A. Daniels, Nova Science Publishers, Inc., 2011 (In press). 4. Hanna SR, Strimaitis DG, Chang, JC. Hazard Response Modeling Uncertainty (A Quantative Method) Volume 1: User’s Guide for Software Evaluating Hazardous Gas Dispersion Models. Sigma, Research Corporation, Westford, MA, 1991. 5. Harrell JA, Belsito ME, Kumar A. Radon Hazards Associated with the Ohio Shale. Environmental Geology and Water Sciences 1991; 18: 17-26. 6. Harrell JA, McKenna JP, Kumar A. Geological Controls on Indoor Radon in Ohio. Investigation No. 144, Division of Geological Survey, Dept. of Natural Resources, State of Ohio 1993; 36 pp. 7. Joshi A, Manne GK, Kumar A. Management of Ohio's Radon Data with MS Access/SQL Server 7.0. Environmental Progress 2002; 21 (4): D8-D12. 8. Kumar A, Harrell JA, Heydinger A. Ohio Goes Online to Combat Indoor Radon. EM (by A&WMA) 2001; 2: 10-12. 9. Kumar A, Heydinger A, Harrell, JA. Development of an Indoor Radon Information System for Ohio. Final Report for Ohio Air Quality Development Authority, University of Toledo, 1990. 10. Kumar A, Maroju S, Bhat A. Application of ArcGIS Geostatistical Analyst for Interpolating Environmental Data from Observations. Environmental Progress 2007; 26(3): 220-225. 11. Kumar A, Tandale A, Kalapati RS, Ghose S. Management of Radon Mitigation Data in the State of Ohio. Environmental Progress 2003; 22 (3): O5-O10. 12. Manthena DV, Kadiyala A, Kumar A. Interpolation of Radon Concentrations Using GIS Based Kriging and Cokriging Techniques. Environmental Progress 2009; 28(4): 487-492. 13. Ohio Department of Health. Radon Licensing Program Rules and Regulations. Ohio Administrative Code, Chapter 3701-69, Ohio Department of Health 2001. 14. Ojha S, Thomas SJ, Kumar A. Experience in Integrating Geographical Information Systems (GIS) to an Indoor Radon Database. Environmental Progress 2001; 20(3): O7O10. 15. U.S. EPA. EPA Assessment of Risks from Radon in Homes. U.S. EPA 402-R-03-003, U.S. EPA, Air and Radiation (6608J) Washington, D.C., 20460, 2003. 16. U.S. EPA. A Citizen's Guide to Radon: The Guide to Protecting Yourself and Your Family from Radon. U.S. EPA 402-K-02-006, U.S. EPA, Indoor Environments Division (6609J) Washington, D.C., 20460, 2004.
 
 88 SHOULD ONLY ACTIVE RADON RESISTANT NEW CONSTRUCTION BE REQUIRED IN ALL ZONE ONE COUNTIES IN ALABAMA? James L. McNees, CHP Office of Radiation Control* Alabama Department of Public Health Montgomery, Alabama Abstract Some individuals within standards development groups involved with radon are discussing a possible revision of the existing Radon Resistant New Construction (RRNC) standards into a consensus national standard that would advocate the requirement that all new RRNC installations in Zone One counties be active radon systems which include a radon fan. This paper investigates the potential for needless installations of radon fans and associated energy waste that such a requirement might cause. Many publications have documented that at radiation levels near natural background, the Linear No Threshold Theory of radiation risk is at best a weak theory. Some publications, including some directly related to radon exposure, have demonstrated no detectable risk to occupants of residences with radon concentrations in the range of the WHO reference level of 2.7 pCi/L. Certainly occupants of homes with radon concentrations at or below half of the WHO reference level are not subject to sufficient risk to justify an active radon removal system. The author uses existing publications on the operational cost of an active system to investigate the wasted energy that would result in over half of the new homes constructed in Alabama’s Zone One counties should the installation of active RRNC ever become required. This, plus the unnecessary expense of unneeded radon fans, leads to the conclusion that for the State of Alabama any RRNC standard should require both a passive system capable of being made active and an initial occupancy radon test to determine if a radon fan is needed. Requiring active-only RRNC would be an excessive waste of money and energy. Introduction The Alabama Radon Program is a partnership between the Alabama Department of Public Health (ADPH) and the Alabama Cooperative Extension System (ACES) with the latter being the primary supplier of information dessemination and outreach. As a public service ACES sells radon test kits to Alabama citizens at near wholesale prices for citizens to utilize to test their *Financial assistance for the distribution and sale of the radon test kits and the tabulation of the resulting data utilized in the paper was provided by the U.S. Environmental Protection Agency as part of the SIRG grant provided to the State of Alabama. 89 residences. The results of those tests are reported by the vendor to ACES which maintains a data base of results. Results are screened to remove repeat and post mitigation tests and to remove results where the citizen indicates that the residence has a radon mitigation system or was built with radon resistant construction features. These results are then tabulated by zip code and by county to provide information of the location and magnitude of Alabama’s radon problem. This paper utilizes the Alabama Radon Program’s 2009 test results as an indicator of the need for Alabama residences to be built with radon resistant new construction techniques that would mandate that an active soil depressurization system be installed at the time of construction. The assumption is made that the Alabama Radon Program’s 2009 test results for a given county are representative of the radon levels that would be expected in new construction built in that county without radon resistant construction features. It is further assumed that if a passive RRNC radon stack and appropriate sealing were to be included in these houses, the radon concentration would be no greater than if no RRNC system were installed and can reasonably be assumed to be lower. Radon Test Results In 1993 the U.S. Environmental Protection Agency (EPA) designated 13 Alabama counties as Zone One. (EPA, 1993) As allowed by that document the State of Alabama added two additional counties, Jefferson and Shelby to the Zone One designation. In 2009 there were 711 radon test results from Zone One counties reported to the Alabama Radon Program that met the criteria for inclusion into the data base. Of those, 141 or 19.8% were greater than or equal to the EPA Action Level of 4.0 pCi/L and 235 or 33.1% were greater than or equal to the WHO Reference Level of 2.7 pCi/L. Conversely, for Alabama’s Zone One counties 80.2% were less than the EPA Action Level and 66.9% were less than the WHO Reference Level. (Alabama Radon Program,2010) If two-thirds of the residences are already below the WHO Reference Level, the lower of the two action levels, is it justifiable to force the purchasers of new residences in all of Alabama’s Zone One counties to be required to purchase a radon fan and then to pay for its operation? Based on two decades of tabulation of radon test results, the Alabama Radon Program has established Colbert and Madison counties as the two Alabama counties having the greatest radon problem. In 2009 there were 38 radon test results from Colbert County that met the criteria for inclusion into the data base. Of those, 11 or 28.9% were greater than or equal to the EPA Action Level of 4.0 pCi/L and 19 or 50% were greater than or equal to the WHO Reference Level of 2.7 pCi/L. Conversely, for Colbert County, 71.1% were less than the EPA Action Level and 50% were less than the WHO Reference Level. In 2009 there were 304 radon test results from Madison County that met the criteria for inclusion into the data base. Of those, 78 or 25.7% were greater than or equal to the EPA Action Level of 4.0 pCi/L and 121 or 39.8% were greater than or equal to the WHO Reference Level of 2.7 pCi/L. Conversely, for Madison County, 74.3% were less than the EPA Action Level and 60.2% were less than the WHO Reference Level. (Alabama Radon Program,2010) Even in Alabama’s two highest radon counties it is reasonable to expect that approximately half of the new residences built without RRNC will be below the WHO Reference level and two-thirds will be below the EPA Action Level. 90 Effect of a Passive Radon Stack Currently the Alabama Radon Program recommends the installation of RRNC with appropriate sealing and a passive radon stack for new residential construction in our high radon areas, followed by a radon test at occupancy and then by another radon test in the opposite season if the initial test was made during the air conditioning season. If the initial test was done in the spring or fall the second test is recommended to be done in the winter heating season. Only if the radon test results indicate a radon problem should the resident be advised to purchase the installation of a fan for the system. 91 The reason for the opposite season retest is because Alabama has found that in Alabama’s high radon counties that approximately one third of the residences that test below 4.0 pCi/L in the summer air conditioning season will have radon concentrations greater than 4.0 pCi/L if retested in the winter heating season. (McNees and Roberts,2007) Alabama has only anecdotal data of the reduction in radon concentration resulting from the installation of passive radon systems. In several incidences over the last 30 years the author observed that passive ventilation reduced the radon concentration by about half. The author can find no published study on this matter except for the Finnish study published in 2010. In that study it was documented that passive radon stacks reduced the radon concentration by 55% when installed along with appropriate sealing in Finland’s high radon areas. (Arvela,2010) The basic principle of warm air rising out the passive stack during winter is the same for both Alabama and Finland. However, there are differences in construction styles that could affect the soundness of applying the Finnish study to Alabama data. In Finland they utilize an insulation layer between the foundation and the soil. In both Finland and Alabama the predominate style of new home construction is a slab on grade. The difference being in Alabama the slab is on top of an aggregate layer which is on top of the soil. At the 2008 AARST International Radon Symposium, Bernard Collignan presented evidence that a passive system can “run efficiently a significant part of the year if it is properly dimensioned, and mainly during cold conditions, where it is more necessary to have a good protection against radon.” (Collignan,2008) This is the same situation as in Alabama where radon levels are typically highest in the winter. (McNees and Roberts,2007) Problems in Alabama with passive systems being unsuccessful when made active are most often a result of incorrect passive system installation and not with the fact that the system did not include a radon fan at installation. For the radon distribution found in Alabama the author advocates that local building codes should require that the system be in installed passive, with appropriate sealing, followed by a radon test at occupancy and then by another radon test in the opposite season if the initial test was made during the air conditioning season. If the initial test was done in the spring or fall, the second test is recommended to be done in the winter heating season. Then, if the radon test results exceed the national Action Level, a radon fan should be added and made active. Alternative Action Levels The EPA has established our national Action Level for radon to be 4.0 pCi/L. (EPA,2005) Some have suggested that for RRNC we should utilize the WHO Reference Level of 2.7 pCi/L. Others have suggested that we utilize what they call the Technically Achievable Threshold with a value of 2.0 pCi/L. In accordance with current national guidance, for an existing home to be acceptable for occupancy following a real estate transaction, the radon concentration should be below the EPA Action Level of 4.0 pCi/L. If our nation were to adopt a policy that for a newly constructed residence to be acceptable for occupancy it should have radon concentrations of less than some new lower concentration such as 2.0 pCi/L, such a policy would be a major inconsistency. Having such an inconsistent policy concerning the acceptable level of radon 92 concentration for occupancy would result in a tremendous loss of credibility for national radon reduction efforts. The EPA national Action Level does not have universal agreement. In addition to the WHO Reference Level, it also differs from that of the National Council of Radiation Protection and Measurement (NCRP) and from that of the International Commission on Radiological Protection (ICRP). NCRP in their Report No. 77 recommends that for the purposes of remedial action the annual average exposure rate should be no more than 2 WLM per year. Where WLM is Working Level Months. The annual average exposure rate of 2 WLM per year can be translated into an average WL of 0.04, which using the 50% equilibrium assumption of Report No. 77, corresponds to 8 pCi/L. (NCRP,1984) Granted NCRP Report No. 77 was published in 1984, but it has not been rescinded by the NCRP. The ICRP in its 2007 recommendations established an upper reference level for residential occupancy of 600 Bq/m3 or 16 pCi/l. (ICRP,2007) ICRP has since issued a statement on radon giving revised upper reference levels to take account for the more recent epidemiological analyses. (ICRP,2009) Their upper value of the reference level for radon gas has been revised to 300 Bq/m3 or 8 pCi/L. The point being that there is not universal scientific acceptance that either the EPA national Action Level of 4.0 pCi/L or the WHO Reference Level of 2.7 pCi/L is the appropriate average annual radon concentration to justify an active radon mitigation system. For the purposes of this paper the author assumes that our nation will reduce the national Action Level from the current 4.0 pCi/L to the WHO Reference Level of 2.7 pCi/L and that this revised Action Level will be consistently applied to both the previously occupied and the newly constructed residences with respect to if the radon concentration is acceptable for occupancy. If requiring RRNC to be installed “active-only” is not justified at an assumed future national Action Level of 2.7 pCi/L, it certainly is not justified at the existing national Action Level of 4.0 pCi/L. Colbert and Madison are the two Alabama counties that have been shown to contain the greatest radon problems in the state, yet radon test results for houses in those counties are less than 2.7 pCi/L 50% and 60.2% of the time respectfully. These test results were for houses built with no radon control systems. Consider what the results would be if these houses had been built as the State of Alabama currently recommends, i.e. with proper sealing and a passive radon stack installed such that it is capable of being made an active ASD system. Because warm air rises everywhere in the world, the same stack effect that worked to remove radon in Finland would also reduce the radon in these houses. Thus, if they were constructed with RRNC having appropriate sealing and passive radon stacks it is reasonable to expect that more that half of the new homes built in Alabama’s highest radon counties would be less than even the WHO Reference Level of 2.7 pCi/L or 100 Bq/m3. Applying the results of the Finish study, half might well be less than 1.35 pCi/L or 50 Bq/m3. 93 Does the risk really exist all the way to zero? The argument is often made that there is still risk reduction to be achieved all the way down to zero radiation dose. Such is the premise of the Linear No Threshold Theory of radiation risk. We utilize the Linear No Threshold Theory extensively in the formulation and application of regulatory rules and practices. Scientific research however, produces many interesting results that cause the Linear No Threshold Theory to be questioned at doses near natural background. As long ago as 1898 Atkinson observed that irradiated algae grew faster than nonirradiated controls. (Atkinson,1898) In 1919 Davey noted the increased life span of irradiated insects. (Davey,1919) Following the atomic bomb blasts it was observed that Japanese who received 11 to 120 R in 1945 appeared to live longer than those who received none or more. (Encyclopedia Britannica,1974) Numerous studies during the second half of the twentieth century demonstrated the increased life span in mammals receiving low doses of radiation. So many such observations were observed and published during the latter half of the twentieth century, that the increased life span of experimental animals exposed to low levels of ionizing radiation was acknowledged by the U.S. National Council on Radiation Protection and Measurements in their Handbook 39, entitled Basic Radiation Protection Criteria. (NCRP,1971) That Handbook outlined the basis of what became many of radiation regulations that are enforced today by regulatory agencies such as the Alabama Office of Radiation Control. The nationwide study of radon concentrations by county by B. L. Cohen suggested a possible inverse relationship between residential low-dose radon levels and lung cancer mortality. (Cohen,1997) The more recent Worcester County, Massachusetts study of Thompson, et.al., found no increase in lung cancer when compared to radon concentrations in radon levels below 150 Bq/m3 (4.0 pCi/L). In fact the Thompson’s article concludes with the statement, “The possibility of a hormetic effect on lung cancer at low radiation doses cannot be excluded.” (Thompson,2008) While the modern publications of Darby, Krewski, and Lubin all demonstrate that breathing greater concentrations of radon are related to lung cancer, they all conclude that in the range of 1.35-2.7 pCi/L (50-100 Bq/m3) that the effects of radon on the incidence of lung cancer is uncertain. (Darby,2005) (Drewski,2005) (Lubin,2005) The scientific truth is that there is no conclusive evidence that Alabama citizens living in residences where the radon concentration is already below the 1.35-2.7 pCi/l (50-100 Bq/m3) range would receive any benefit from the expense of the installation and operation of a radon reduction fan. How much unnecessary expense would it be? The State of Alabama is currently recommending the installation of a passive RRNC as outlined in the EPA publication Building Radon Out for new residential construction in their Zone One counties. (EPA,2001) Alabama builders have told the author that once adopted on a wide scale basis the increased cost of the passive RRNC as recommended would be in the range of $350 to 94 $500. A fan would then be added and operated only in the event the radon levels were found to be above the national Action Level. If the building codes in Alabama were revised to require active-only RRNC for new residential construction in the 13 counties that the EPA designated as Zone One, then over half of the purchasers of those new homes would be paying for a radon fan, associated installation cost, and operational costs that they do not need. The EPA sponsored moisture study estimated the cost of operation of such systems to be between $83 and $191 per year. (Turk and Hughes,2008 ) In 2009, Leo Moorman presented and published a more rigorous analysis of estimated cost associated with the operation of an active radon removal system. For climates similar to the high radon areas of north Alabama, Moorman’s methodology estimates a total energy cost of $300 to $325 per year. (Moorman,2009) There are those who say that the occupants could turn the fan off and test for radon and then decide if the fan off radon level was acceptable for continued occupancy. But the author advocates that it would be wiser and more acceptable to the occupants to require installation of a passive system with appropriate sealing followed by a radon test at occupancy to determine if a fan needs to be purchased, installed, and operated continuously. Conclusion There may be some Zone One counties in the USA with a distribution of indoor radon concentrations such that the preponderance of new construction would be above the national Action Level if only passive systems were installed. In that case, requiring active-only RRNC might be justified. Even in the Alabama counties that have the highest occurrence of elevated indoor radon, such is not the situation. In all of Alabama’s Zone One counties, a majority of newly constructed homes if correctly built with passive RRNC would have no need for an active radon fan. Thus, adoption of a building code requiring active-only RRNC is not justified anywhere in Alabama. Each state or tribal government should examine the distribution of indoor radon concentrations within their own jurisdictions and determine if a building code requiring active-only RRNC is appropriate and justified. 95 References 1. EPA, “EPA’s Map of Radon Zones Alabama," EPA 402-R-93-021, September 1993. 2. Alabama Radon Program, data base of program test results, May 2010. 3. McNees and Roberts, “Summertime Short-term Negative Radon Tests Need To Be Retested in Winter,” Health Physics Journal, July 2007. 4. Arvela, “Radon Prevention in New Construction – Sample Survey 2009,” STUK-244, 2010. 5. Collignan, “Experimental Study On Passive Sub-Slab Depressurisation System,” Proceedings of the AARST International Radon Symposium, Las Vegas, NV, 2008. 6. EPA, "A Citizen's Guide To Radon" EPA 402-K02-006, revised September 2005. 7. NCRP, Exposures From the Uranium Series With Emphasis On Radon and Its Daughters – Report 77, National Council on Radiation Protection and Measurement, 1984. 8. International Commission on Radiation Protection, ICRP Publication 103, Annals of the ICRP, 2007. 9. International Commission on Radiation Protection, Web Site http://www.icrp.org, Statement On Radon, ICRP Ref 00/902/09, 2009 10. Atkinson, “Report upon some preliminary experiments with the Rontgen rays on plants.” Science, July 1898. 11. Davey, “Prolongation of life of Tribolium consusum apparently due to small doses of Xrays.” Journal of Experimental Zoology, Volume 28, 1919. 12. Encyclopedia Britannica, “Biological Effects of Radiation,” Volume 15, page 378, 1974. 13. NCRP, Basic Radiation Protection Criteria – Handbook 39, National Council on Radiation Protection and Measurement, 1971. 14. Cohen, “Lung cancer rate vs. mean radon level in U.S. counties of various characteristics.” Health Physics Journal, Volume 72, 1997. 15. Darby, “Radon in homes and risk of lung cancer: “Collaborative analysis of Individual data from 13 European case-control studies.” British Medical Journal, January 2005. 16. Krewski, “Residential radon and risk of lung cancer: A combined analysis of 7 North American case-control studies.” Epidemiology, March 2005. 96 17. Lubin, “Risk of lung cancer and residential radon in China: Pooled results of two studies.” International Journal of Cancer, 2004. 18. EPA, “Building Radon Out: A Step by Step Guide on How to Build Radon-Resistant Homes,” EPA 420-K-01-002, April 2001. 19. Turk and Hughes, “Exploratory Study of Basement Moisture During Operation of ADS Radon Control Systems, Revised 3/20/2008,” U.S. Environmental Protection Agency Indoor Environments Division, Washington, DC, 2008. 20. Moorman, “Energy Losses and Operational Costs of Radon Mitigation Systems”, Proceedings of the AARST International Radon Symposium, St. Louis, MO, 2009. 97 RADON-RELATED LUNG CANCER DEATHS AND MITIGATION COST EFFECTIVENESS IN A RADON-PRONE REGION Daniel J. Steck Physics Department, St. John’s University Collegeville, MN 56321 ABSTRACT Can a significant number of radon-related lung cancers be averted by mitigating existing single-family houses whose radon is above the action level? Would societal resources be used cost effectively by a comprehensive measurement and mitigation policy using the existing or some other radon action level? Simple questions, complex answers. The risk of lung cancer depends on the radon exposure in living spaces. Radon concentration distributions and population densities can vary on several spatial scales, so an analysis for an entire country may reach different conclusions than a regional analysis. New information has been discovered about indoor radon distributions, residential radon risk, effectiveness of installed mitigation systems, and costs of mitigation systems since the last analyses. Costeffectiveness analysis itself has also advanced. A protocol developed for the WHO International Radon Project and applied to the United Kingdom reached the conclusion that remediating existing homes in the UK was not cost effective. Those authors also thought it likely that this conclusion would apply to most developed countries. A cost-effectiveness analysis is underway for the Upper Midwest, a radon-prone region of the United States. This analysis uses actual regional radon distributions and measurements of effectiveness of radon mitigation, system costs, and risk models. The preliminary results suggest that a regulatory policy for radon measurement and mitigation with widespread compliance could save many lives at a cost much lower than direct medical treatment. INTRODUCTION Members of the general public and some public health officials have been slow to take action to decrease radon exposures in the United States (US). Among the reasons for this inaction is a failure to appreciate the magnitude of the risk that radon can pose in some regions and the potential to avoid lung cancers through an aggressive regulatory policy of radon mitigation. Earlier analyses of the cost effectiveness of mitigation have focused on a voluntary measurement and action policy using the national radon distribution and highly uncertain estimates of the actual radon reduction achieved by private contractors (USEPA 1992, Ford et al. 1999, Lin et al. 1999). These studies as well as those done in Europe often differ substantially in their conclusion of the utility of different radon mitigation policies (Stigum et al. 1996, Gray et. al 2009, Haucke 2010). For example, a recent analysis concluded that while basic preventive measures in all new houses in the United Kingdom (UK) would be cost effective, remedial work on existing houses could not prevent most radon-related deaths; a conclusion that those authors believed was likely to apply to most developed countries (Gray et al. 2009). Some of the differences in the studies arise from 98 different choices of the style of economic analysis, costs included, radon distributions, risk models, population characteristics, and the comparative measures of the effectiveness of the policies (Mason and Brown 2010). The goals of this work are to estimate the potential lung-cancer deaths that can be avoided and to do an analysis of cost effectiveness from a public-health perspective for the radonprone Upper Midwest using regional distributions for radon exposures in living spaces and actual achievable performance of radon reduction through active slab depressurization and ventilation. (Insufficient radon measurement data preclude an analysis of preventive measures in new houses which are now taking place in the region.) MATERIALS AND METHODS Data from Iowa (IA) and Minnesota (MN) were used as representative of the entire Upper Midwest (UM) which should properly include eastern sections of the Dakotas and Nebraska and western sections of Wisconsin and Illinois. Approximately 7 million people live in IA and MN and a similar number live in “adjacent regions” of neighboring states. Radon-related risk of lung cancer Population-weighted indoor radon concentrations are needed to calculate the radon-related risk of lung cancer using lifetime risk models such as the recent EPA model (USEPA 2003). Earlier nationwide analyses used the national radon distribution (Marcinowski et al. 1994) whose geometric mean (GM) is 0.67 pCi/L and geometric standard deviation (GSD) is 3.12. The distribution of indoor radon concentrations in living spaces in the Upper Midwest was drawn from unbiased randomized surveys across MN (Steck 2005, 2006) and the participants in the Iowa Radon Lung Cancer Study (Field et al. 2000). The MN results are the long-term radon concentration measurements averaged over the two lowest living spaces from more than 2500 single-family homes. The distribution was log normal with a GM of 2.73 pCi/L and a GSD of 2.16. The result for each Iowa home was an average of several annual measurements on the first floor. The Iowa distribution had a GM of 2.55 pCi/L and a GSD of 2.02. Since both radon concentration and single-family house density can vary substantially across a state, the analysis was carried out on a county basis. Bayesian estimated geometric mean radon concentrations were calculated for each county to improve the estimates for those counties where only a small number of houses was sampled (Price et al. 1996). Risk reduction through mitigation in single-family homes Long-term, post-mitigation radon concentrations were measured in the two lowest living levels of 150 MN homes. These houses were randomly selected from the client list of five regional mitigators (Steck 2008). Post-mitigation radon concentrations were 0.8 pCi/L, on average. Monte Carlo simulations were used to calculate the average risk reduction in a county by subtracting the post-mitigation radon concentration from the pre-mitigation radon concentration that was generated from the county’s radon distribution. To simplify the exposure calculation, the single-family home population from the 2000 census and the mitigation performance were assumed to be constant over a 74-year lifetime. The total single-family home population in IA and MN was approximately 6 million people with home occupancy of approximately 2.5 per house. 99 Cost-effectiveness analysis Most recent economic studies of radon policies have used the Cost-Effectiveness Analysis (CEA) methodology. The WHO’s International Radon Project recommends this approach and provides guidance (WHO 2009). An example for the UK has been published (Gray et. al 2009). While WHO emphasizes the cost per Quality Adjusted Years of Life Lost ($/QAYLL) as the preferred measure of effectiveness, the Cost per Life Saved (CLS) is often calculated and used for comparison to other protective actions. In this preliminary analysis, central value estimates of regional costs are used in lieu of a using cost distributions needed for a fuller analysis. These estimates, shown in Table 1, are based on averages from local sources. For example, the installation cost is the average of several hundred installations in Minnesota (Steck 2008). The heat penalty costs are an average across the region’s climate taken by a new study that realistically evaluates heating and cooling costs in a variety of climates (Moorman 2009) Table1: Central value estimates for parameters in four CEA cases Parameter Action Level pCi/L (Bq m-3) Rn measurement $ Base Case 4 (150) 50 Lower Costs 4 (150) 25 Higher Costs 4 (150) 100 Lower action level 2.7 (100) 50 Installation $ Fan Replacement $/y 1400 12 800 10 2500 14 1400 12 Fan power (W) Electric $/kWh 80 0.10 24 0.08 150 0.12 80 0.10 Heat Penalty $/y 190 70 400 190 RESULTS AND DISCUSSION The results described below are preliminary. Additional sensitivity and economic adjustments are underway to refine and extend the analysis. Risk of lung cancer The cumulative percentage radon-related risks of lung cancer as a function of radon concentration for the exposed populations are shown in Figure 1 for the US distribution (green) and the Upper Midwest (red). 100 Fig 1 Cumulative population radon-related risk of lung cancer The distinct differences between the US and Upper Midwest risk levels is evident near both the current action level (4 pCi/L) and the reference level recommended by WHO (100 Bq m-3 or 2.7 pCi/L). For the Upper Midwest, 20% of the risk occurs from exposure to less than 4 pCi/L and 10% to less than 2.7 pCi/L, while the US values are 70% and 60%, respectively. The success of radon mitigation systems installed in the Upper Midwest suggests that there is a significant potential to reduce radon exposures, because the average radon concentration in homes prior to mitigation (7 pCi/L) can be lowered to about 1 pCi/L. Potential for avoiding radon-related lung cancers During a 74-year period, approximately 50,000 Minnesotans and 20,000 Iowans could be saved from dying from radon-related lung cancer if all single-family homes with a radon concentration greater than 4 pCi/L in the living spaces were mitigated. These represent about a 50% life-saving rate. A map of the number of potential lives saved by county is shown in Figure 2. The magnitude of these potential savings exceeds many other causes of homerelated death, for some of which preventive and corrective actions have been mandated. For example, carbon-monoxide detectors are required in MN homes. Completely preventing all carbon-monoxide fatalities would save 30 deaths per year in MN compared to the ~700 per year who could be saved by sustained, universal radon mitigation. Cost-Effectiveness Analysis Table 2 shows the cost per life saved and per year of life saved for the Upper Midwest population using point cost estimates for central values (base case), two estimates that include cost variation, and a policy with a lower radon action level. 101 Table 2: Cost-effectiveness Analysis results for four cases Base Case Cost per Life Saved (1000s of 2010 $) 190 Cost per Year of Life Saved (1000s of 2010 $) 12 Lower Costs 80 4 Higher Costs 390 20 Lower action level 220 13 Potential lives saved by mitigating single family homes with Rn above 4 pCi/L 0 - 100 100 - 250 250 - 500 500 - 1000 1000 - 10000 Fig 2 Map of the potential number of lives that could be saved by county in Iowa and Minnesota under a universal policy that requires measurement and mitigation single-family houses with Rn concentrations > 4 pCi/L. These values of cost per life saved (CLS) in Table 2 are of the same order of magnitude as the results of US analyses which evaluated a full, or nearly full, scenario of mitigation compliance (Ford 1999, Lin 1999). These CLS values pale compared to some benchmarks like the value of $7 million for a statistical life saved used by the EPA for some cost-benefit 102 analyses (Dockins et al. 2004). They are lower than the CLS for direct medical treatment for lung cancer, ~$1 million (USEPA 1996). Most of the US CEA studies focused on radon policies that relied on voluntary action taken by homeowners following “outreach and education” by public health organizations. Limited studies of the public’s response to these approaches concluded that this approach had a low efficiency for mitigation of homes that measured above the action level (Doyle et al. 1991, Ford and Eheman 1997). This condition led to a substantially higher CLS; estimates above $1 million. However, by focusing on radon-prone areas where higher compliance may be achievable, the CLS dropped substantially (Ford et al. 1999, Lin et al. 1999). While it is beyond the scope of this work to develop or propose a policy that would lead to better compliance of reduction of radon exposure, the author’s personal experience with public action related to radon in the Upper Midwest region suggests that community action supported by competent technical assistance and cost-sharing could create a high level of compliance with radon-reduction programs in single-family homes. A more robust sensitivity and uncertainty analysis is underway to improve the estimates of the effects of changes in measurement and mitigation policy, estimated costs, and risk models on the cost effectiveness of radon mitigation in the Upper Midwest. ACKNOWLEDGMENTS The author wishes to thank Rachel Dols for assistance in collecting and analyzing data for the Upper Midwest. 103 REFERENCES Dockins C, Maguire M, Simon N, Sullivan M. Value of Statistical Life Analysis and Environmental Policy: A White Paper U.S. Environmental Protection Agency National Center for Environmental Economics April 21, 2004 http://yosemite.epa.gov/ee/epa/eerm.nsf/vwAN/EE-0483-01.pdf/$file/EE-0483-01.pdf accessed 5 August 2010. Doyle JK, McClelland GH, Schulze WD, Elliott SR, Russell GW. Protective responses to household risk: a case study of radon mitigation. Risk Anal. 11, 121-134,1991. Field RW, Steck DJ, Smith, BJ, Brus CP., Neuberger JS., Fisher, EF, Platz CE, Robinson, RA, Woolson RF., Lynch CF. Residential Radon Gas Exposure and Lung Cancer: The Iowa Radon Lung Cancer Study. American Journal of Epidemiology. 151(11), 1091-1102; 2000. Ford ES, Eheman, CR. Radon retesting and mitigation behavior among the U.S. population. Health Phys. 72, 611614, 1997. Ford ES, Kelly AE, Teutsch SM, Thacker SB, Garbe PL. Radon and lung cancer: a costeffectiveness analysis. Am J Public Health. 89(3):351-357, 1999. Gray A, Read S, McGale P, Darby S. Lung cancer deaths from indoor radon and the cost effectiveness and potential of policies to reduce them. Br. Med. J. 338, a3110, 2009. Haucke, F., The cost effectiveness of radon mitigation in existing German dwellings. A decision theoretic analysis, Journal of Environmental Management (2010), 10.1016/j.jenvman.2010.06.015. Lin C, Gelman A, Price P and Krantz DH. Analysis of Local Decisions Using Hierarchical Modeling, Applied to Home Radon Measurement and Remediation. Stat. Sci., 14: 305337, 1999. Marcinowski F, Lucas, R , Yeager W M. National and regional distributions of airborne radon concentrations in U.S. homes. Health Physics 66 699706; 1994. Mason J, Brown MJ. Estimates of Costs for Housing-Related Interventions to Prevent Specific Illnesses and Deaths. J Public Health Management Practice, 16(5) E-Supp, S79– S89; 2010. Moorman L. Energy losses and operational costs of radon mitigation systems. Nineteenth International Radon Symposium, St. Louis, MO, September 2009. Available at http://www.aarst.org/radon_research_papers.shtml. 104 Price P N, Nero A V, Gelman A. Bayesian prediction of mean indoor radon concentrations for Minnesota counties. Health Physics 71 922936;1996. Stigum H, Strand T, Magnus P. Should radon be reduced in homes? A cost-effect analysis. Health Phys. 2003;84:227-235. Steck DJ. Residential Radon Risk Assessment: How well is it working in a high radon region?. 15th International Radon Symposium, San Diego CA. September 2005 Available at http://www.aarst.org/radon_research_papers.shtml. Steck DJ. Building codes, house age and indoor radon in Minnesota. Sixteenth International Radon Symposium, Kansas City Mo. September 2006 Available at http://www.aarst.org/radon_research_papers.shtml. Steck DJ. Post mitigation radon concentrations in Minnesota. Eighteenth International Radon Symposium Las Vegas NV September 2008. Available at http://www.aarst.org/radon_research_papers.shtml. U.S. Environmental Protection Agency. Technical support document for the 1992 Citizen’s guide to radon. Washington D.C.; U.S. Government Printing Office; 400-K92-011; May 1992. USEPA Cost of Illness Handbook 1996 http://www.epa.gov/oppt/coi/index.html accessed 25 August 2010. WHO handbook on indoor radon: a public health perspective; Hajo Zeeb and Ferid Shannoun, eds. World Health Organization Press, Geneva 2009. 105 2010 International Radon Symposium, Columbus, OH DESIGNING EFFICIENT SUB SLAB VENTING AND VAPOR BARRIER SYSTEMS FOR SCHOOLS AND LARGE BUILDINGS Thomas E. Hatton President Clean Vapor, LLC 32 Lambert Road, P.O. Box 688, Blairstown, New Jersey 07825 Phone (908)362-5616 Fax (908)362-5433 thatton@cleanvapor.com www.cleanvapor.com ABSTRACT The author discusses the basic components of designing efficient sub slab radon venting and vapor barrier systems for schools and large buildings. There are two new construction case study schools: one located in Georgia and the other in New Jersey. Both have active radon venting plans as supplied by the architect. Both architect designs required considerable review and plan amendments. The school in New Jersey has a conventional under slab pipe collection system, multiple riser pipes and a standard polyethylene vapor barrier. The school in Georgia has an efficient under slab plenum box collection system, a single riser pipe and a Liquid Boot vapor barrier system. This paper compares sub slab vacuum field extension, the leakage associated with the different vapor barrier systems, electrical consumption, as well as heat and cooling efficiencies. An energy consumption cost analysis demonstrates support for efficient designs and long term sustainability. INTRODUCTION For professionals involved in the construction of new commercial and school buildings or an addition to an existing building and there is a potential for radon or vapor intrusion exposure, the information in this paper will provide design considerations for the purpose of integrating cost effective preconstruction engineered controls to prevent entrainment of soil gases. There are many factors that need to be considered when deciding to build a new building. The health and safety of the occupants should be in the forefront of the planning and design process. Understanding radon and vapor intrusion and how soil gas entrainment can be prevented is a key item in constructing a building that has safe indoor air. Protecting human health by preventing radon and soil gas impacts needs to be addressed during the planning and construction phase. Radon removal systems for commercial buildings that are integrated with construction are far more cost effective, function better and are less obtrusive than retrofit systems. 106 The integration and enforcement of indoor air quality standards for vapor intrusion are increasingly becoming Certificate of Occupancy requirements for new commercial buildings. Since many schools and commercial buildings are constructed over reclaimed properties of former industrial sites, vapor barrier and ventilation systems that were originally designed to control radon may need to serve a dual capacity of a radon and vapor intrusion mitigation system. The author has participated in upgrading two ventilation systems that were originally designed to vent only radon at school sites but needed to also serve as vapor intrusion mitigation systems. WHAT CAUSES RADON TO BE DRAWN INTO A BUILDING? Air pressure in the lowest level of buildings is usually lower than pressure in the soil beneath the building. Negative pressures that are induced by buildings draw both radon and other soil contaminates into occupied building space where inhalation and human health risk from exposure occurs. Radon can enter the building through expansion joints, sumps, slab cracks, open block joints, utility penetrations or any opening that can serve as a pathway. 1, 2 Radon entering a building is the result of three primary variables: (1) source strength of radon in the soil; (2) entry routes; and (3) pressure differentials that draw radon gas from the soil into the building. Understanding these components and the effects they have on the transfer of soil gas to indoor air will help determine which preconstruction countermeasures should be integrated into the building. WHEN IS VAPOR INTRUSION ALSO A CONCERN? The potential for vapor intrusion should be assessed at all properties when the owner is performing due diligence early in the property evaluation process. The starting point is thorough Phase I and Phase II studies. A consultant needs to be chosen who is familiar with the geographic area and who has provided radon and vapor intrusion assessments for similar properties. The consultant should investigate the history of all properties within 100 feet of the site. If there is an existing building on the site that is targeted for redevelopment, indoor air, radon and soil gas samples should be conducted even if the building itself is scheduled for demolition. This will provide valuable information about a new building’s potential for radon and vapor intrusion. On site groundwater and soil gas samples should be collected. Vapor intrusion pathways can be a concern from 100 feet vertically and laterally. The risk associated with a chemical depends on it’s toxicity, concentration and potential to migrate into a building. The site’s complexity, geology and characteristics of the building will all influence the rate at which soil gases are drawn into a building. The Johnson Ettinger or J&E model is often used to project the attenuation rates of soil contaminants into a building. Data uncertainty is always an issue when interpreting the results of a model. The other unknown factor is the quality of building construction. Items such as small unsealed floor wall joints or minor openings around utility penetrations can create entry pathways and have a significant influence on the rate at which soil gases are transferred into a building. Both EPA and many state agencies recognize the importance of identifying the 107 potential for vapor intrusion and have established guidance for sampling and data evaluation. The Federal EPA has published the Brownfields Technology Primer: Vapor Intrusion Considerations for Redevelopment, March 2008,3 and a User's Guide for Evaluating Subsurface Vapor Intrusion into Buildings, February 2004.4 At this time, at least 21 states have Vapor Intrusion Guidance documents. Your state regulator should be contacted to determine the regulations that would apply to the subject property. Once the chemicals of concern, construction variables and state regulations have been understood, a decision can be made on how to move forward with integrating radon and vapor intrusion resistant construction If it is determined that there are also vapor intrusion concerns, the designer of the sub slab ventilation system needs to choose a vapor barrier that is chemically resistant to the contaminants of concern. Chemical vapor barriers represent a significant expense and the building owner should be encouraged to use a Professional Engineer for this aspect of the project. This will allow the designer of the vent system to optimize the venting installation and to coordinate with other trades to minimize damage to the liner. A coordination meeting should occur sometime during the design phase to determine the sequence of construction and maximize efficiencies of all the components. Pipe routes through the building need to be planned so vent locations are sufficiently away from fresh air intakes and passive relief vents. An early and proactive evaluation of vapor intrusion impacts will make panning easier and provide more time to design the best mitigation solution. RADON SHOULD ALWAYS BE CONSIDERED WHEN CONSTRUCTING A NEW BUILDING A wide variety of strata and soils have been determined to be radon emitters. Check the EPA map for radon potential. Many states will have databases with radon testing statistics that are grouped by zip code. Some states, such as New Jersey, have Tier classifications and mandatory radon code requirements for new school construction. If there is an existing building on the site that is targeted for redevelopment, conduct a radon test. This will provide valuable information about a new building’s potential for radon. HOW IS RADON ENTRY AND VAPOR INTRUSION PREVENTED? Once the soil contaminants potential for entering a building has been assessed, the next step is to select which preconstruction countermeasures should be implemented to minimize radon or vapor intrusion. There are five basic components to effective radon resistant construction. They are: 1) permeable sub slab support material, 2) venting all occupied ground contact slab areas, 3) properly sized under slab and riser piping, 4) a sealed vapor barrier and 5) if an active system is specified, a properly sized blower is needed to maintain sufficient negative pressure beneath the slab. Passive vent systems have components one through four but do not have a blower to mechanically draw soil gases from under the slab collection piping to above the roof. Active Soil Depressurization Systems (ASD) are powered by blowers that create vacuum beneath the slab 108 and actively vent sub slab gases through solid conveyance piping to above the roof. If there is a significant vapor intrusion concern or if geological or statistical test data supports the likely possibility of indoor elevated radon, it is prudent to take a conservative approach and include the activation of the system piping. The integrity of the vapor barrier and efficiency of passive vent systems are two main variables in determining the effectiveness of a passive system. Punctures or tears in the vapor barrier that may occur during the construction process will diminish the effectiveness of a passive system. Passive systems do not address the pressure differentials that are the main driving forces that draw radon and soil vapors into buildings. The benefit of a passive system is that it can be activated after an indoor air exceedance has been detected. After the sub grade has been proof rolled by removing undesirable items, drying, leveling and compacting the soil, a permeable layer of crushed stone should be installed. Eight inches or more of ASSHTO #57 stone is preferable. ASSHTO #57 stone is a highly permeable, course aggregate with 95% to 100% passing through a one inch (25mm) sieve, 25% to 60% passing through a half inch (12.5mm) sieve and less than ten percent passing through a Number 4 (4.75mm), 3/16” sieve. It has been the observation of Figure 1: Areas Isolated by Thickened Slabs the author that if there is not sufficient crushed stone above and below the pipe, then the slab directly above the pipe is prone to cracking. To minimize slab cracking that is associated with the pipe, there should be a minimum of four inches of crushed stone in addition to the diameter of any conveyance piping. If six inch pipe is used, the ground beneath the pipe may need to be trenched to assure sufficient stone for slab support. There should be a minimum of two inches of crushed stone above and below any sub slab conveyance pipe. All slab areas within the occupied area that have contact with the soil need to be included in the sub slab vapor collection system. Grade changes and thickened slabs beneath concrete masonry walls often isolate under slab areas (Figure 1). Because the radon collection system is usually the last item to be incorporated into the construction plan, a thorough review should occur to make sure all areas have been included. There should be no under slab areas that are outside of the influence of the radon collection system. The most efficient way to vent sub slab soil gas is through a centrally located soil gas collection plenum box. A plenum box is embedded in the crushed stone layer and constructed of hollow concrete blocks turned on their side with an empty space in the center from which a conveyance or vent pipe transports soil gases above the roof line. See Figure 2 below. There should be a minimum of eight inches of crushed stone beneath and beside the plenum box. Ground floor classrooms, offices and storage rooms are often partitioned by Concrete Masonry Unit (CMU) 109 walls that are supported by haunches or thickened slabs. If the architect does not specify a depression in the ground beneath the thickened slab for stone then the thickened slab will be in direct contact with the soil creating a “mini foundation” around the perimeter of the thickened slab that is isolated from the vented areas beneath the slab. Interior footings at grade changes will also create isolated sub slabs. GeoVent®, which is a one thick rectangular shaped roll out plastic and fabric covered conveyance plenum, or perforated collection pipe can provide a conduit to connect isolated slab areas to a central sub slab plenum box (See Figure 2). Depending on the leakage associated with the vapor barrier, the configuration of the under slab conveyance piping and the design of the plenum box, a single properly sized collection system can service up to 15,000 square feet of floor space. If the service area of the plenum box is greater than 4,000 square feet, the main vent pipe to the roof should be six inches in diameter. Figure 2: Connecting Isolated Slab Areas with a Central Plenum Box The design goal is to create a minimum sub slab negative pressure of -0.016” of water column (W.C.) (4 pascals) at the area that is most distant from the plenum box using a blower that consumes no more than 140 watts and can move 200 cubic feet per minute (CFM) at 1.0” W.C. static pressure. Even though it has been demonstrated that pressure differentials as low as - 0.004” W.C. (1 pascal) can successfully arrest the attenuation of soil gases, -0.016” W.C. (4 pascals) is used as a design goal because the design specialist has no control over construction conditions that can reduce the efficiency of the system. Factors that can obstruct the free flow of sub slab soil gas through the stone layer and truncate the extension of the vacuum distribution are sand particles mixed in with the crushed stone, elevated sub slab utility conduits that are in the stone, soil that is left over from burying utility lines after the ground has been leveled (trench overburdening), and conveyance piping that has been crushed or distorted by unscheduled vehicle traffic. SIZING THE CONVEYANCE PIPING 110 Sizing the conveyance pipe is based on the square feet of the area to be vented and the number of pipe fittings used between the under slab plenum box and the vent termination point. Drag coefficient tables exist for different pipe diameters and assorted fittings.5 Since coordinated drawings are usually not part of the design phase, the person designing the system should plan on twice the number of pipe fittings as anticipated when calculating the pressure drop associated with a riser pipe system. A pressure drop table for common pipe sizes and air flow is included in Figure 3 below. Equivalent footage of pipe for each fitting is included in Table 2 below. The most commonly used riser pipe material is PVC because of its availability, low cost and low air flow drag coefficients. No Hub Cast Iron pipe is used when there is concern of exceeding the flame spread or smoke index especially if the conveyance piping passes through a return air plenum. Protective pipe enclosures or steel pipe should be considered in areas of vehicle or fork lift traffic. Figure 3: Pipe and Airflow Pressure DropTable Source: Brodhead Conveyance piping can be joined together beneath the slab to minimize vertical risers. A three inch riser pipe can service up to 1500 square feet, a four inch riser can service up to 4000 square feet, and six inch riser pipe can service up to 15,000 square feet.6 Sub slab conveyance pipe should have 5/8” condensate drain holes that face down at four foot intervals to minimize water blockages inside the pipe. If factory perforated pipe is used, one set of holes shall face down. Note that significantly larger sub slab areas have been depressurized when the slab barrier and perimeter foundation is very air tight and the soil has low permeability. 111 Pipe Diameter Per 100’ of PVC Pipe at 100 CFM 6” 0.007 4” 0.04 3” 1.4 Table 1: Inches of Water Column Pressure Drop in 100 feet of Pipe at 100 CFM Source: Brodhead Pipe Diameter Straight Pipe Resistance Per 100’ of PVC Pipe at 100 CFM 6” – 90 elbow 15’ 0.00105 4” – 90 elbow 6’ 0.0024 3” – 90 elbow 4’ 0.056 Table 2: Inches of Water Column Pressure Drop from 90 degree Elbows Source: Brodhead When designing a depressurization system and specifying blowers, it is important to include the projected piping pressure losses. Speculating the final active system air flow is one of the most difficult parts of the design process. Air flow is a function of blower capacity, piping size, fittings and layout, sub slab aggregate resistance, soil permeability and slab and foundation leakage. 7 Clean ASSHTO #57 stone is highly permeable and can provide excellent slab support material through which to extend vacuum. As demonstrated by the comparison of the two case study schools, the performance required from the blower to achieve the specified vacuum field is largely determined by the slab leakage and quality of the vapor barrier seal. If there is clean crushed stone and four inch conveyance piping, a blower that can move 200 CFM at -1.0” W.C. can create a vacuum field of -0.02” W.C. or greater over a 4000 square foot area. Reducing the slab leakage can significantly increase the coverage area. The primary design goal should always be highly permeable sub slab material and minimal slab leakage. SELECTING A VAPOR BARRIER Selecting the right vapor barrier is a critical part of the project. The vapor barrier can also be the most expensive part of the system. The type of vapor barrier and the quality of the seal will determine the efficiency and effectiveness of the protective measure. The most commonly used material that is often referred to as a vapor barrier is 6 mil polyethylene. Under American Concrete Institute (ACI) standards, 6 mil (0.15mm) polyethylene is not classified as a vapor barrier but as a vapor retarder.8 This is because polyethylene and polyolefin vapor barriers generally have a permanence (water vapor transmission rate) (WVTR) of less than 0.3 perms, as 112 determined by ASTM E 96. A number of vapor retarder materials have been incorrectly referred to and used by designers as vapor barriers. True polyethylene and polyolefin vapor barriers are products that are generally not less than 10 mil (0.15 mm) have a permeance WVTR of 0.1 perms when tested in accordance with ASTM E 96. Polyethylene vapor barrier material is often made of “post consumer” recycled materials. The main limitations of these commonly used polyethylene and polyolefin vapor retarders is that they are not chemical resistant and the available sealing tape and lapping procedures make them difficult to seal at perimeter walls and around utility penetrations. These vapor retarders will often pull back from the attachment points during the concrete pour and cure process, leaving significant gaps for radon and vapor intrusion. These types of vapor retarders provide the lowest level of protection. The second most commonly used vapor barrier materials are 10, 12, and 15 mil polyethylene or polyolefin vapor barrier material such as Stego-Wrap which has a WVTR of less than 0.1 perms. These are true vapor barriers. The limitations to these materials are limited chemical protection and low puncture and tear ratings when evaluated against reinforced products of similar thickness. A third group of vapor barriers is reinforced and cross laminate polyethylene and polyolefin materials such as the Raven Vapor Block 15®. They are generally three ply materials with woven scrim between two polyethylene sheets. These vapor barrier materials have puncture and tear resistant ratings that can be as much as 50 times greater than a 6 mil polyethylene vapor retarder. If radon is the main soil gas contaminant of concern, thicker and cross laminated materials can provide greater efficiency than a standard polyethylene vapor retarder. This is largely because the materials are less likely to be punctured or torn and the seams and utility penetrations are usually sealed to a higher standard using manufacturer supplied tapes and cloth binders. The Liquid Boot ® is a spray-applied gas vapor barrier product that provides a near gas tight seal. It was designed to provide an impenetrable vapor membrane for water, volatile organic compounds and methane gas.9 The Liquid Boot ® application process involves first applying a base non-woven geotextile fabric over the stone and then spraying a hot emulsified asphalt/latex top covering. This hot material binds to the support fabric, column pads and side foundation walls to provide a minimum thickness of 60 dry mil. The total thickness including the support fabric is 73 mil. Once the application process is completed and top emulation has cured, a port in the vapor barrier is accessed and the underside is pressurized by using centrifugal blowers to supply colored indicator smoke. This will identify leaks for repair or enable the installer to certify that there are no leaks. Installers of this material must be licensed by the manufacturer. The most important part of the effectiveness of any vapor barrier system is a tight seal to foundation walls and around the utility penetrations of the membrane. A filter fabric layer is recommended to protect all vapor barriers from punctures associated with construction debris and the underlying stone. The concrete slab installer must not be allowed to puncture the liner to drain off extra water that may be associated with the concrete finishing process. QUANTIFYING THE EFFECTIVENESS OF THE RADON SYSTEM 113 The effectiveness of any soil depressurization system should be quantified after the slab is poured and allowed to cure for at least 28 days. The test is performed by temporarily installing the specified blower and measuring the sub slab vacuum field. During the construction phase, soil probes routed through the slab are embedded in the crushed stone prior to installing the vapor barrier. Probes are embedded because drilling through the concrete slab creates an unnecessary risk of damaging sub slab utilities and will void most vapor barrier warranties. Probes should be located distant from the plenum box near the projected end of the negative pressure field. Depending on the potential for radon or soil vapor entry, these probes could be as numerous as one per isolated foundation area. At least one probe should be installed per 5000 square feet of slab area and for each different slab elevation. Each blower system should have at least one soil probe. Soil test probes can be as simple as ¼ inch polyethylene tubing with a polyester particulate filter on the end that is in the stone. More durable probes are constructed of perforated two inch PVC pipe connected to ½ black iron gas pipes. These probes usually go from the gravel bed through the concrete form that surrounds a support column and terminates inside the column pocket a few feet above grade. The lines are then capped to prevent debris from blocking the tubing. Gas line can be purchased in a variety of lengths with a threaded end that is fitted with a cap. The more durable PVC and steel and probes have a greater probability of surviving the concrete pour and power trowel process because small plastic tubing is easily damaged. The efficiency of the system is measured by temporally activating the system by hooking up the blower that has been specified for the system activation. The pressure field extension is measured at the embedded probes. A micromanometer that can measure to a sensitivity of -0.001” W.C. should be used. The micomanometer tubing is temporarily sealed inside the soil gas probe. If vacuum field measurements at the probe most distant the sub slab collection piping exceeds 0.036” W.C. (9 pascals) the top of the acceptable vacuum range specified by ASTM10 the procedure can be repeated using a blower that uses less electricity. If the piping is incomplete at the time of testing, pressure drops as a result of the additional piping should be included in the calculation. Lower wattage blowers that will save energy should be installed to reduce operating expenses. Airflow exhaust measurements should be conducted using a pitot tube or vain anemometer. The selected blower model, final static riser pipe vacuum, vacuum field measurements at the embed probe locations and final exhaust airflow values should be recorded and included in the construction documents that are presented to the owner as part of the project record. Unless provisions are made during the design phase, soil probes shall remain capped and will become encased artifacts of the building and not accessible for future use. Indoor air sampling should occur once the building is weather tight and the air handling systems are operational. There is value in having one consultant working all the way through the design installation and application phase. This maintains continuity throughout the project and should provide a single standard of measurement methodology and quality control. GREEN ENERGY AND SUSTAINABILITY CONSIDERATIONS 114 EPA defines Green remediation as “considering all the environmental effects of a remedy implementation and incorporating options to maximize the net environmental benefit …” (EPA 2008). When designing a new construction radon system, long term energy considerations need to be factored into the design process. Greater design efficiency reduces operational costs and extends the time that an active venting can be sustained for a fixed capital expenditure. A streamlined under slab collection plenum system with minimal conveyance piping fittings will increase the efficiency of sub slab vacuum distribution and reduce the energy required by the blower. There are three main components that need to be considered when attempting to lower the operational energy costs of a radon or vapor intrusion mitigation system. They are: the cost of operating the blower(s) that will maintain the negative pressure field beneath the slab, the cost of the heat that is being drawn out of the building and the cost of the cooled conditioned air that is being drawn out of the building. An addition cost that must be considered is the costs of replacing the blowers themselves. More blowers will result in higher operations and maintenance costs. Selecting a sealed vapor barrier system that minimizes leakage is the largest variable in reducing ongoing energy costs. The cost to heat or cool the conditioned air that is drawn into the collection system can be a greater operational expense than the electrical cost to operate the radon fans. Installing a tightly sealed vapor barrier system and optimizing the blower size can save up to $1,000.00 annually in heating, cooling and electric costs per 10,000 square foot of floor space. After every reasonable attempt has been made to prevent radon entry by selecting the best vapor barrier and vent system design, there are still convective forces that are working to draw soil gases into a building. Even with the best vapor barrier system, soil contaminants can still be drawn in though the hollow cores of bock walls and unsealed electrical conduits that penetrate the slab. COMPARATIVE CASE STUDY The case study compares the efficiencies of two active radon systems that were installed as part of new construction at two schools. The first school is a three story school with 38,000 square feet of ground coverage and is located in New Jersey. The original design specified a standard 6 mil polyethylene vapor barrier and vent pipes compliant with New Jersey’s Radon Hazard Sub Code. The New Jersey Radon Hazard Sub Code was originally drafted to apply to residential homes. 11 Education facilities were included immediately prior to the 1995 ratification and the inclusion was intended to serve as an interim vehicle until the Radon Hazard Sub Code for Schools could be drafted and codified. This code, that was originally intended to be applied to homes, does not facilitate cost effective or energy efficient venting practices for schools or large buildings. The second school is a three story new school that was constructed in Georgia. Several “Green Energy” concepts were integrated into the construction of this school including indoor air quality goals. The size of the new construction footprint is 24,000 square feet. A decision was made by the architect early in the design process to select a vapor barrier that would be as gas tight as 115 possible to minimize the entrainment of radon. selected. The Liquid Boot ® vapor barrier system was At both schools, Clean Vapor was consulted to improve the original architects design and implement venting efficiencies as a part of construction process. In both cases, a condition of accepting the design improvements was that they could not alter the owner approved construction costs since the budgets had been finalized. NEW JERSEY SCHOOL The specifications in the New Jersey school originally called for 4 inches of crushed stone over proof rolled ground, Schedule 80 PVC sub slab piping, 39 three inch plumbing T’s set into the gravel, 6 mil polyethylene vapor barrier with limited sealing instructions, 39 vertical three inch risers routed from the T’s in the gravel to the roof and 39 -145 watt roof mounted centrifugal blowers. The proscribed method of measuring the pressure field extension testing was to drill holes through the slab and measure the vacuum fields with a micromanometer. The New Jersey School design was amended to replace the schedule 80 PVC pipe with schedule 40 PVC pipe. The 39 3” T fittings set in the gravel were replaced by a four inch under slab pipe collection system that included all previously isolated areas. At the end of each sub slab solid pipe run is a “T” fitting with chamfered extensions to increase the intake surface area of the fitting. The new design reduced the number of risers from 39 three inch risers to 14 four inch risers and blowers. The standard vapor barrier was installed per the original specification. Sub slab vacuum was still measured by drilling holes through the slab and collecting measurements with a micromanometer. Careful notes were made on the location of all electrical conduits and sub slab utilities prior to pouring the slab to determine where test holes could be installed. See 14 system riser layouts in Figure 4 below. Figure 4: New Jersey School Amended Sub Slab Layout Design 116 After the slab was poured, RadonAway RP-265 blowers were installed and the systems were tested for exhaust airflow and vacuum distribution. The exhaust airflow for all fourteen risers averaged 74 cubic feet per minute CFM. The vacuum field extension results averaged -0.0792” W.C. All test locations exceeded the -0.02” W.C. specified pressure differential under weather tight heated building conditions. The testing was conducted in the early spring at approximately 45 degrees Fahrenheit outside temperature. It was observed that the air exhausted from the blowers was relatively warm when compared with the outside air temperature. An analysis of the additional cost of the conditioned air being discharged was undertaken. Note that this type of school new construction radon mitigation design is commonly used in New Jersey. GEORGIA SCHOOL The original design for the Georgia school included a low permeable compacted road bed material beneath the slab. Road bed material is mechanically fractured, course gravel mixed with course and fine sands. The matrix of the material has almost no void spaces with 35% to 90% of the material passing through a Number 10 (2.0 mm) sieve. It can be tamped into a very hard low permeability fill material that is structurally very good for supporting concrete slabs but is not suited for extending a sub slab vacuum field. The risers and collection sumps consisted of three separate six inch riser pipes with one, 145 watt blower for each riser. Each riser was specified to be connected to a collection sump that consisted of a six inch section of twenty-four inch diameter pipe set vertically in the road bed material. 3/8” holes on 1¼” centers were to be drilled in a staggered pattern through the side walls of the twenty four inch pipe to transfer the soil gas from the sub slab, through the perforated pipe. See Figure 5 below. The perforated two foot diameter pipe was to serve the function of a perforated collection sump. The radon from each two foot diameter sump was to be vented through six inch conveyance pipe to a location above the roof. There were several areas of the slab that were isolated from the three collection sumps. The specified vapor barrier was Liquid Boot®. The plan called for drilling holes through the concrete and the Liquid Boot® vapor barrier to allow measurement of the sub slab negative pressure field once blowers were installed. Note that drilling these holes would have voided the Liquid Boot® warranty. Clean Vapor was consulted to redesign the system and install an effective energy conserving system that would minimize the entrainment of radon from the soil into the building. 117 Figure 5: Original 24 inch Collection Sump Design The compacted road bed material was replaced with ASSHTO #57 stone. The three perforated twenty-four inch diameter pipes were each replaced with concrete block soil gas collection plenum boxes. See photo of this collection plenum in Figure 2 above and the location of the collection plenums in Figure 6 below. Four inch perforated sub slab conveyance piping was installed in the gravel bed from all separate slab areas to the collection plenum box. Six inch soil gas conveyance pipes were routed from each of the three sub slab collection plenums to a common area where the three separate riser pipes came through the concrete. The plan for drilling test holes through the slab and vapor barrier was abandoned. Five vacuum test soil probes were installed in the stone bed during construction. See their location in Figure 6 below. SOIL PROBE # 1 RADON COLLECTION PLENUM # 1 SOIL PROBE #3 RISER # 1 6" UNDER-SLAB PVC PIPE SOIL PROBE # 5 RISER # 2 RISER # 3 6" UNDER-SLAB PVC PIPE RADON COLLECTION PLENUM # 3 RADON COLLECTION PLENUM # 2 SOIL PROBE #2 SOIL PROBE # 4 Figure 6: Georgia School Amended Design - Three Riser Systems Joined Together with One Blower 118 After the slab was poured, the three risers were individually tested for air flow and vacuum distribution. The vacuum field extension results were significantly greater than any measurements previously recorded by Clean Vapor for similar applications. A decision was made to join all three riser pipes together and repeat the measurements. A single Fantech HP 220 blower capable of creating 2.7” of static vacuum operating at 125 watts created a nearly uniform vacuum field beneath the 24,000 square foot slab of -1.8167” W.C. The total exhaust air flow was only 119 CFM. The vacuum distribution and minimal leakage design goals were substantially exceeded. Suction Device HP 220 Suction System Static Vac " WC S-1 -1.825 S-2 -1.826 S-3 -1.827 Air Flow CFM 119 Exhaust Diameter 6" Test Probe Suction Device On T-1 -1.8167 T-2 -1.7806 T-3 -1.8174 T-4 -1.8120 T-5 -1.7904 Table 3: Sub Slab Static Vacuum Distribution The Georgia school with the efficient conveyance piping design and Liquid Boot® vapor barrier had twenty times the static vacuum of the New Jersey school (-1.816” versus -0.079”). The New Jersey school’s fourteen blowers were moving a total of 1036 CFM versus the 119 CFM from one blower at the Georgia school. Using the Liquid Boot® appeared to reduce the CFM requirement from 0.027 CFM/ft2 for the New Jersey school to 0.005 CFM/ft2 for the Georgia school and it created a sub slab vacuum that was twenty times stronger. Clearly, the Green Energy initiatives would result in operational energy savings. The savings and net benefit were evaluated against the increased capital expenditure of installing the Liquid Boot. To create some consistency in the model, a cost of $0.14 kWh was used for electricity and $12.50 per 1,000 cubic feet of natural gas. Since the New Jersey school would have more heating days and the Georgia school more cooling days, the latitude of the schools was standardized to the degree days of a six month heating season. The operational cost of blower motors required to distribute the additional conditioned air from Roof Top Ventilators (RTU’s) to the classrooms and the costs of replacing the radon blowers after a projected six year life span were not included in the operational cost calculations. The costs were standardized to 10,000 square feet for the purpose of illustration and easy math for those who would be tasked with evaluating costs for future school construction. 119 New Jersey School Georgia School Approx. Square Feet 38,000 24,000 Riser per Sq Ft One 4” per 2,700 One 6” per 24,000 Average Vacuum Sub Slab -0.0792 -1.8167 Blowers Fourteen Blowers Operating at 118 watts One Blower Operating at 125 watts Annual Blower Electrical Cost * $516 $77 Est. Leakage per 10,000 sqft 272.6 CFM 49.5 CFM Table 4: Operational Cost Analysis New Jersey School Georgia School Heating: Factor H/E 1.4703 2.25 Cooling: * Factor C/E 0.2108 0.323 Overall Factor (E+H+C)/E = 2.681 3.57 Table 5: Energy Cost Analysis per 10,000 Square Feet of Floor Area New Jersey Georgia Savings Electric* $457 $63 $394 Heating: ** $ 672 $77 $595 Cooling: * $96 $11 $85 Total Savings $1,074 Table 6: Operational Cost Analysis per 10,000 ft2 of Floor Area Standardized to a 6 month HDD and CDD *Electric Cost $ 0.14 per kWh *Natural Gas Costs $ 12.50 per 1000 ft3 The next step in the evaluation was to compare the benefits derived from the energy savings with the increased construction costs and determine if this Green Energy Initiative is also fiscally sound. 120 The comptroller of a North Jersey Construction Company was contacted and the installation costs for both the standard vapor barrier and Liquid Boot acquired. Union labor rates were applied to the installation of both materials. The Liquid Boot cost for this model is $5.50/ft2 versus the traditional vapor cost of $0.25/ft2. A 5% finance cost over 20 years was applied to the project. The results are plotted in Figure 7 below which reveals in just less than ten years the Green Energy Initiative with the efficient vent system and the Liquid Boot® paid for itself in operational savings. Note in the chart the initial cost for 10,000 ft2 for the Liquid Boot® is $55,000.00 versus traditional vapor barrier costs of $2,500.00. The decision to issue energy credits or use public funds to finance energy saving projects is usually based on a twenty year return. The findings of this case study indicate that the additional costs expended to design and install highly efficient sub slab vent systems for the purpose of preventing radon entry and vapor intrusion are fiscally sound. Figure 7: Initial Capital Expenditure Versus Operational Energy and Finance Costs Depending on the square feet of ground coverage and labor rates, Liquid Boot® costs can be as low as $3.00/ft2. It is recommended that the installation cost variables be considered when modeling specific sites. 121 SUMMARY and CONCLUSIONS Both the Georgia and New Jersey school used ASSHTO #57 stone as a sub slab material. The New Jersey school used four inch perforated pipe in the stone to collect soil gas and included no central collection plenum boxes. The New Jersey school used a standard 6 mil polyethylene vapor barrier. The level of vacuum achieved at the Georgia school was twenty-two times greater than the New Jersey school. The difference was an efficient under slab soil gas collection plenum box and a near gas tight vapor barrier provided by the Liquid Boot® System. The annual savings associated with the efficient collection plenum and Liquid Boot per 10,000 square feet of floor area was estimated to be $ 1,074.00. The additional installation cost of a sealed hot spray emulsified vapor barrier (Liquid Boot®) should pay for itself in energy savings in approximately ten years. Energy efficient radon and vapor intrusion mitigation systems that are designed and integrated during the planning and construction phases have demonstrated to be cost effective, energy efficient and highly effective in preventing radon entry and vapor intrusion. 122 REFERENCES 1. United States Environmental Protection Agency Third Printing with Addendum, June 1994 Radon Prevention in the Design and Construction of Schools and Other Large Buildings EPAl625/R-921016. 2. Brodhead, Bill. 2002, 12th Annual International Radon Symposium, Reno, Nevada. Designing Commercial Sub-Slab Depressurizations Systems. WPB Enterprises, Inc. 2846 Slifer Valley Rd., Riegelsville, PA. 3. United States Environmental Protection Agency, March 2008 Brownfields Technology Primer: Vapor Intrusion Considerations for Redevelopment, EPA 542R-08-001. 4. United States Environmental Protection Agency, February 2004, User's Guide for Evaluating Subsurface Vapor Intrusion into Building. 5. Brodhead, Bill. 1996, 6th Annual International Radon Symposium Orlando, FL. Airflow Pressure Drop in Typical Radon Piping. WPB Enterprises, Inc. 2844 Slifer Valley Rd., Riegelsville, PA. 6. New Jersey Proposed Radon Hazard Subcode for Schools Buildings Draft revised November 2009. 7. Moorman, Leo Ph.D. 2008, 18th Annual International Radon Symposium, Las Vegas, NV, Solving Turbulent Flow Dynamics Of Complex, Multiple Branch Radon Mitigation Systems. Radon Home Measurement and Mitigation, Inc. Fort Collins, CO. 8. American Concrete Institute (ACI) Committee 302 March 23, 2004 Guide for Concrete Floor and Slab Construction ACI 302.1R-04. 9. CETCO Liquid Boot® Technical Data Spray-Applied Gas Vapor Barrier www.cetco.com CETCO 1001 S Linwood Ave, Santa Ana, CA. 123 10. ASTM E2121-08 2008 Standard Practice for Installing Radon Mitigation in Existing Low-Rise Residential Buildings. 11. New Jersey Register, Vol. 39. No. 22, November 19, 2007 Title 5. Department of Community Affairs Chapter 23. Uniform Construction Code Subchapter 10. Radon Hazard Subcode N.J.A.C. 5:23-10. KEY WORDS Vapor Barrier Radon Entry Vapor Intrusion Radon Hazard Sub Code for Schools ASSHTO # 57 Stone Liquid Boot Green Energy Cost Savings ACKNOWLEDGEMENTS Bill Brodhead, WPB Enterprises, Inc. Leo Moorman, PhD Radon Home Measurement and Mitigation, Inc 124 THE SPATIAL AND VOLUMETRIC VARIATIONS OF RADON IN BANGALORE CITY, INDIA LA Sathish1*, K Nagaraja2, HC Ramanna1, V Nagesh1, S Sundareshan3 and TV Ramachandran4 1 Department of Physics, Government Science College, Bangalore – 560 001, India 2 Department of Physics, Bangalore University, Bangalore – 560 056, India 3 Department of Physics, Vijaya College, Bangalore – 560 004, India 4 Ex-Environmental Assessment Division, Bhabha Atomic Research Center, Mumbai- 450 085, India *Corresponding Author (Sathish) Email: lasgayit@yahoo.com Abstract Radon levels have been measured in houses at ten different locations and rooms of various sizes ranging from 30 to 310 m3 for Bangalore city, India. The study was focused on the basis of quality of construction, age of building, similar nature of walls and types of floorings etc. Solid state nuclear track detectors were used for measuring the concentrations. The average spatial values of 222 Rn and 220 Rn concentrations were found to be 33.4 ± 6.1 and 21.6 ± 2.5 Bq m-3 respectively. However, the volumetric concentrations were ranged between 4.0-93.0 Bq m-3. The annual dose rate due to 222Rn, 220Rn and their progenies for the population in the studied location ranged from 0.1 to 0.5 mSv. It is alarming that the dwellers of lower volume receive relatively a higher dose rate and the result shows significant radiological risk. The magnitude and its effects of doses are discussed in detail. Key words: Radon, dwellings, volume, dose rates Corresponding Author: Dr.Sathish.L.A Assistant Professor Department of Physics Government Science College Nrupathunga Road, Bangalore – 560 001 India +91-80-9886639324 
 125 Introduction Measurement of indoor radon is significant due to the exposure of radon and its daughters, which contributes more than 50% of the total dose from natural sources on human being UNSCEAR (2000). The three radon isotopes, radon – 219 222 Rn, thoron – 220 Rn and actinon – Rn are gaseous and may be released from the ground, rocks of the Earth’s crust and also from building materials and accumulate with their short-lived progeny in closed spaces, particularly in dwellings. The dose deriving from the existence of 222 Rn in the air is directly linked to the inhalation of its short-lived daughters, which are deposited in the respiratory organs, if deeply inhaled; emit alpha-particles that are in contact with bronchial and pulmonary epithelium. On account of these, the dose deriving from the exposure of 222 Rn in closed spaces has been placed in direct relation to the risk of lung cancer UNSCEAR (2000). Some factors that influences the diffusion of radon from soil into the air are existence of uranium and radium in soil and rock, emanation capacity of the ground, porosity of the soil and/or rock, pressure gradient between the interfaces, soil moisture and water saturation grade of the medium (Schery and Gaeddert, 1984). The concentration of indoor radon also depends on ventilation rate of the dwellings. It is important to note that even though reduced ventilation rate aids to enhance the concentration of radon and its daughters in air. Solid state nuclear track based dosimeters are employed for the long - term integrated measurements (Stranden 1980; Abu-Jarad and Fremlin, 1983). Measurements on volumetric variations of 222 Rn and 220Rn in dwellings are limited and this work seems to be first of its kind. The paper reports the relationship between 222 Rn, 220 Rn and volume of rooms. The measurement for radon and thoron concentrations were also carried out by using plastic track detectors and the results obtained are discussed in detail. The data is continuously obtained for a period of three years since 2007, covering more than 150 dwellings. Study Area The location selected for the present study is Bangalore city, India and is shown in Fig. 1. The district lies between the latitudes 12˚39' to 13˚13' N and longitudes 77˚22' to77˚52' E. The climate is having four distinct seasons, viz., summer season (March to May), rainy season (June to August), autumn season (September to November) and winter season (December to February). April is usually the hottest month with the mean daily maximum temperature of 
 
 126 
 Fig. 1: Map of Bangalore Metropolitan, India ˚ 30-35 C and mean daily minimum at 20-24 ˚C. The geology of this part forms predominantly a granite terrain with numerous varieties of granites, granitic gneiss, pegmatite and charnockites and so on. The rocks around the study area are called Close pet granites (Ningappa et al, 2008). These rocks are younger than the peninsular gneiss, made up of several types of potassium granites with variable color, texture and multiple intrusion relationship. The common rocks are pink, grey and porphyrite gneisses with large feldspars, black dolerite. These rocks form geological band of a width 15–25 km. Most of the studied houses in Bangalore city were constructed with cement and bricks that were made up of local 
 
 127 soil and few were mud houses (Ningappa et al, 2008). The soil radioactivity reported in earlier studies is close to background levels from other regions of the country (Mishra, and Sadasivan 1974). The radioactivity reported for the building materials collected from this region is higher compared with soil radioactivity (Ramachandran et al, 2003). All the monitored houses were on the ground floor. About fifteen houses were chosen in all the monitored locations. Analysis is made on location wide, season wide and room volume. Methodology Twin cup dosimeters developed in Bhabha Atomic Research Centre (BARC), Mumbai, India were used in this study. The dosimeter has two cylindrical cups of equal volumes having radius 3.1 cm and height 4.1 cm. The cups are having provision to hold SSNTD films inside the cups and a third SSNTD film outside the cup for progeny measurements. A schematic diagram of the dosimeter is given in Fig.2. The dosimeter is designed to discriminate 222Rn 


3 M Radon Compartment Membrane Filter 









 1 F 2 Radon +Thoron Compartment Glass Fiber filter 
Filter
 1. 2. 3. Radon Cup mode SSNTD Film Radon + Thoron Cup mode SSNTD Film Bare mode SSNTD Film Fig. 2: Schematic diagram of twin cup radon-thoron dosimeter and 220 Rn in mixed field situations, where both the gases are present like in monazite rich deposit areas. Track detector used in the dosimeter is cellulose nitrate films, commercially called LR-115 films, made by Kodak Pathe. Films of size 3 cm × 3 cm were affixed at the bottom of each cup as well as on the outer surface of the dosimeter. The exposure of the detector inside the cup is termed as cup mode and other one exposed openly is termed as bare 
 128 mode. One of the cups has its entry covered with a glass fiber filter paper that permeates both 222 Rn and 220 Rn gases into the cup and is called filter cup. The other cup is covered with a semi permeable membrane sandwiched between two glass fiber filter papers called membrane cup (Ward et al, 1977). These types of semi permeable membranes have diffusion coefficient for radon gas in the range of 10–8–10–7 cm2 s–1 that permeates more than 95% of the 222 Rn gas while it suppress the entry of 220 Rn gas to (Wafaa, 2002) more than 99%. Thus, the SSNTD films inside the membrane cup register tracks that attributes to alone, while the filter film records tracks due to both 222 Rn and 220 222 Rn gas Rn gases. The third film exposed in the bare mode registers alpha tracks produced by both the gases and their alpha emitting progeny. Eappen and Mayya (2004) have reported that LR-115 (12 µm) film does not register tracks from deposited activity since Emax for LR-115 (12 µm) is 4 MeV and all the progeny isotopes of 222 Rn /220Rn emit alphas with energies more than 5 MeV. Thus, uncertainty due to deposited activity on film surface is removed for the bare detector estimate; a reason to choose LR-115 (12 µm) film for bare card estimate. The dosimeters were kept at a height of about 1.5 m from the ground, considering least disturbance to the occupants. Care is taken while placing the dosimeter such that the active surface of the SSNTD film used in bare mode exposure is kept at a minimum distance of 10 cm away from any surface to avoid tracks due to attenuated alphas reaching from these surfaces. Measurements were completed in each dwelling for a calendar year covering the four seasons prevailing in the area. After exposure, the dosimeters were retrieved and SSNTD films were removed from the dosimeter for etching. The films were then etched in 10% NaOH solution at 60 ºC for 90 minutes (Eappen and Mayya, 2004). The tracks recorded on LR-115 films were counted using a spark counter (Cross and Tommasino, 1970; Samyogi et al, 1978). Tracks are converted to gas concentrations using Eqs. (1) and (2). C R ( Bqm "3 ) = C T ( Bqm "3 ) = Tm d ! Sm T f " d ! C R ! S rf d ! S tf (1) (2) where Tm is the track density of the film in membrane compartment (Tr cm-2), d is the period of exposure in days (d), Sm refers to the sensitivity factor of membrane compartment (Tr cm-2)/(Bq d m-3), Tf is the track density of the film in filter compartment (Tr cm-2), Srf is the Sensitivity of 222 Rn in filter compartment (Tr cm-2)/(Bq d m-3) and, CR and CT are the concentrations (Bq m-3) of 222Rn and 220Rn, respectively. We followed the protocols given by 
 129 Eappen and Mayya (2004) for processing the exposed films; hence sensitivity factors Sm and Srf are taken from their work for computing the gas concentrations. The progeny concentrations in terms of Working Level (WL) can be written as: Rn ( mWL) = C R ! FR 3.7 (3) RT ( mWL) = CT ! FT 0.275 ( 4) Where FR and FT are equilibrium factors for radon and thoron respectively and can be equated with progeny fractions of respective gases as shown in Eqs. (5) and (6). FR = 0.104 FRA + 0.514 FRB + 0.37 FRC (5) FT = 0.91FTB + 0.09 FTC (6) Where FRA, FRB, FRC, FTB and FTC are activity fractions of 218Po, 214Pb, 214Bi, 212Pb and 212Bi, respectively. Mayya et al (1998) have obtained these activity fractions through ventilation parameters applying a root finding method using the deposition velocities for attached and unattached fractions of the progeny nuclides. Since the data in this study is not sufficient for (volumetric variations) deriving ventilation dependent Fxx factors, bare card results are not used in deriving F values. Such an exercise will be tried later after large number of measurements data are collected from future study. For the present study, inhalation dose is computed using UNSCEAR (2000) F values. Indoor occupancy factor for the population is taken as 0.8 and the annual inhalation dose (mSv y–1) is calculated using Eq. (7). ( ) D mSv y !1 = 7000 " [(0.17 + 9 FR )C R + (0.11 + 40 FT )CT ]"10 !6 (7 ) Results and Discussion Volumetric Variations of Indoor 222Rn and 220Rn: The natural radioactivity contents of soil samples of Bangalore region reported by earlier studies are 15.2, 16.90 and 486.7 Bq kg-1 for Sadasivan (1971) and the concentrations of 226 Ra, 238 U, 232 232 Th and Th and 40 40 K respectively Mishra and K in the building rocks of -1 Karnataka region are 33, 30.5 and 412.3 Bq kg respectively Ramachandran et al (2003). However, major quantity of bricks used for the construction of the buildings in Bangalore 
 130 city are brought from places in the city out skirts called Nelamangala, and Magadi and a small quantity from Hoskote, Ramanagara and Channapattana. The average activity concentrations of 226 Ra, 232 Th and 40 K in the soils of Nelamangala and Magadi are 31.3 ± 0.6, 52.6 ± 0.9 and 303.1 ± 6.1 Bq kg-1 and 16.9 ± 0.6, 57.5 ± 1.1 and 1073 ± 15.6 Bq kg-1 respectively Shiva Prasad et al, (2008). Rooms were broadly classified into ‘6-groups’ on the basis of volume ranged from 30 to 310 m3 such as 30–40, 45–60, 65–75, 80–100, 110–120 and 200–310 m3. About 7 rooms were selected in each dimension at ten different locations. Hence, the total number of rooms covered in each volume is 42 rooms. However, the total number of rooms monitored is 42 × 10 locations = 420 rooms. These 420 rooms have been analyzed for four seasons and lead to 1680 measurements. The total number of films (LR-115 detectors) exposed during this period of measurement is more than 5000. The frequency distribution of 222Rn and 220Rn levels in dwellings is presented in Figs. 3 and 4. Radon ( Bq m-3) 10 3 (GSD - 2.1) GM = 23.0 Bq/m 40 9 Number of houses 8 20 7 6 0 0 20 40 60 80 100 Cumulative frequency (%) 5 4 3 2 1 0 5 15 25 35 45 55 65 75 85 95 -3 Radon concentration (Bq m ) 
 Fig. 3: 222Rn levels in dwellings Geometric means of indoor 222 Rn and 220Rn levels in the study area are 23.0 and 20.0 Bqm-3 with GSDs 2.1 and 2.0 respectively. Cumulative frequencies against the 222 Rn/220Rn values showed linear regression with correlation coefficient equals 1 for both the cases. A linear correlation with correlation coefficient nearing one indicates a common factor predominant in the various categories of rooms governing the gas concentrations in these houses. A 
 
 131 correlation with dwellings volume and gas concentrations is attempted in this study which is explained below. 3 (GSD - 2.0 GM = 20 Bq/m ) Thoron( Bq m-3) 40 Number of houses 15 20 0 10 0 20 40 60 80 100 Cumulative frequency (%) 5 0 5 15 25 35 45 55 65 75 -3 Thoron concentration (Bq m ) Fig. 4: 220Rn levels in dwellings Inhalation dose is computed using UNSCEAR (2000) dose conversion factors. Inhalation dose calculated from the total results varied from 0.27 - 4.45 mSv y-1 with a geometric mean of 1.34 mSv y-1 (GSD 2.1). Table 1 show the range and average values of 222 Rn and 220 Rn levels in room volume ranging from 35 to 300 m3. The higher concentrations were observed in a room of lower volume than in larger volume. Table 1: 222Rn & 220Rn levels in category of rooms 222 Volume of room (m3) 30 – 40 45 – 60 65 – 75 80 – 100 110 –200 200 - 310 Rn (Bqm3) Range Aver. ± SD Min. Max. 67.3 93.0 81.1 ± 9.3 48.5 62.0 54.1 ± 4.4 39.8 47.4 43.4 ± 2.9 25.2 35.1 30.7 ± 3.8 12.9 20.5 16.7 ± 2.7 07.1 10.5 07.3 ± 2.2 220 Rn (Bqm3) Range Aver. ± SD Min. Max. 42.3 69.4 57.5 ± 9.7 27.5 36.8 31.0 ± 3.6 18.8 27.1 22.6 ± 3.4 13.2 17.4 15.7 ± 1.2 09.6 12.5 11.0 ± 1.1 06.6 09.0 06.9 ± 1.3 
 
 132 A plot of 222 Rn and 220 Rn concentrations is made against room volume in Fig.5. It is seen from the figure that the concentrations decrease with increase in volume of the rooms. However, in the case of 220Rn the effect is almost nullified beyond room volumes greater than 150 m3. If we consider that the exhalation rate for 222Rn and 220Rn from the room surfaces is almost same, assuming that the materials used for construction in these houses are similar, it is expected that the gas concentrations will decrease with increase in volume of the room since the surface to volume ratio decreases with increase in room volume. 100 90 Radon (B) Thoron (D) -3 Gas concentration (Bqm ) 80 70 Equation 60 R^2 y = A1*exp(-x/t1) + y0 0.99 0.91 Value 50 40 30 Std. Error B y0 6.77 1.95 B A1 125.79 16.43 B t1 56.18 7.91 D y0 8.58 1.26 D A1 123.37 49.24 D t1 32.03 6.87 20 10 0 0 50 100 150 200 250 300 3 Volume of room (m ) 
 Fig.5: Gas concentrations with volume of dwellings A plot of volume against ratio of area to volume (A/V) is shown in Fig. 6. It could be seen from the plot that the A/V ratio also showed an exponential fit in decreasing order with a correlation coefficient 0.99. It is interesting to note that the fitting parameter t in Fig. 6 is 61.4 which closely match with effective decay value for radon (56.2) in Fig. 5. This clearly indicates that the radon values inside dwellings covered under the study is predominantly depended on A/V ratio inside the houses. Effect of ventilation seems negligible when the measurement was carried out for long durations. However, the results of thoron were different compared to radon. The t value is almost half (32) to that of A/V ratio. One can speculate certain other phenomenon governing the thoron values. It is only logical to say that 
 
 133 predominance of thoron profile inside the room exists to some extend and in rooms having larger volumes concentration of thoron is profound from surfaces closer to dosimeter placement. Area to Volume ratio (A/V) 2.0 Equatio y = A1*exp(-x/t1) + y0 n R2 0.99 Value Std. Erro C y0 0.99443 0.02996 C A1 1.48873 0.08904 C t1 61.432496.36398 1.8 1.6 1.4 1.2 1.0 0 50 100 150 200 250 300 Volume of room (m3) Fig.6: Correlation between A/V ratio and volume of dwellings Spatial variation of indoor 222Rn and 220Rn levels: The construction materials used for building the houses are predominantly of cement, concrete and bricks made up of local soil. About 40% of the bricks used for the construction of the buildings in Bangalore city are from the city out skirts called Nelamangala, 46% from Hoskote and the remaining from Magadi, Ramanagara and Channapattana etc. The reported values of natural radioactivity contents of 226 Ra, 232 Th and 40 K for Bangalore soil are 15.2, 16.9 and 486.7 Bq kg-1 respectively (Mishra and Sadasivan, 1971), whereas the contents of 238 U, 232 Th and 40K in the building rocks of Karnataka region are reported as 33.0, 30.5 and 412.3 Bq kg-1 respectively (Ramachandran et al, 2003). About 150 dwellings in ten different locations of Bangalore city, India were selected on the basis of construction and age of the building to see the effective dose rates due to indoor 222 Rn, 220 Rn and their progeny levels in dwellings during different seasons of the year. The houses were categorized on the basis of 
 
 134 ventilation that depends on number of windows, doors and usage pattern (such as closed, open, partially open/close) to identify them as poor (no or 1-window), moderate (2-windows) and good (3 and above windows) ventilated houses. The annual average values of 222 Rn, 220 Rn and their dose rates in the different locations of Bangalore city are summarized in Table-2 including the number of houses monitored in each area during April 2007 to April 2010. Table 2: Annual average concentrations of 222Rn, 220Rn and their effective dose rates Number of Name of the Dwellings Location monitored Rajajinagar 15 Srinivasanagar 15 Sheshadripuram 15 Srirampuram 20 Padhmanabhanagar 15 Jayanagar 15 Banashankari 12 Malleshwaram 13 Vijayanagar 15 Gandhinagara 15 AM ± SD AM ± SD Dose rate 220 Rn Rn mSv y–1 Bq m–3 17.2 ± 1.2 16.1 ± 1.4 0.1 40.0 ± 1.9 29.2 ± 4.3 0.2 31.8 ± 3.1 19.8 ± 2.0 0.2 26.3 ± 3.2 18.8 ± 1.6 0.3 27.5 ± 1.7 25.9 ± 2.0 0.2 25.3 ± 1.6 19.7 ± 1.2 0.2 26.5 ± 2.0 21.4 ± 2.2 0.2 27.9 ± 2.9 17.8 ± 1.5 0.2 25.5 ± 3.8 8.3 ± 1.2 0.2 85.9 ± 2.3 38.3 ± 5.4 0.5 33.4 ± 6.1 21. 6 ± 2.5 0.2 ± 0.03 222 The arithmetic mean of 222Rn concentration varies from 17.2 ± 1.2 to 85.9 ± 2.3 Bq m–3 with a mean of 33.4 ± 6.1 Bq m–3, whereas for 220 Rn it vary from 8.3 ± 1.2 to 38.3 ± 5.4 Bq m–3 with a mean of 21.6 ± 2.5 Bq m–3. The lower values of 222Rn concentrations were observed in Rajajinagar and higher in Government Science College of Gandhinagara. The reason may be due to the fact that the activity concentrations (226Ra) in the surrounding area (Mallathalli 23.7±0.7) are lower compared to the Gandhinagara (Lalbagh -111.6±1.2). The lower and higher concentrations of 220 Rn were seen in Vijayanagar (Mallasandra: 29.5±0.9) and Government Science College of Gandhinagara (Lalbagh: 95.4±1.5), respectively. This is again due to the activity concentrations of 232 Th in the respective area (Ashok et al, 2008). Hunse et al, (2010) have reported that the radon in water in Rajajinagar (166.62±8.08 Bq L-1) is low and higher concentrations are in Cubbon Park (Government Science College: 764.05±35.4 Bq L-1). The results show that there is a direct correlation between radon in 
 135 water and indoor radon. The radons in water of other location are in between these two values with fair correlation between radon in water and indoor radon. The average indoor radon concentration reported for dwellings of different cities across the world varies between 8.7 Bq m–3 for Australia and 190 Bq m–3 for Saxony and Turingia of Germany, with a weighted arithmetic mean for all the cities considered of 40 Bq m–3 (UNSCEAR, 1993). The effective radiation dose due to 222 Rn and 220 Rn ranged between 0.1 – 0.5 mSv y-1 with an arithmetic mean of 0.2 mSv y-1. The observations made for Bangalore region were also of the same order reported elsewhere. The annual average concentrations of 222 Rn and 220 Rn for the different seasons and temperature of Bangalore city are tabulated in Table 3. The obtained concentration shows a clear seasonal variation. Higher concentrations of 222Rn and 220Rn in winter months and lower in summer months. This may be due to the enhanced radon exhalation and reduced ventilation as observed elsewhere (Virk and Sharma, 2000). Table 3: Typical Seasonal Variation of 222Rn and 220Rn concentrations Mean Temperature (˚C) December - February 20 Winter 35 Summer March - May June - August 30 Rainy 24 Autumn September - November Season Period 222 Rn 220Rn Bq m–3 42.6 26.3 16.2 14.0 23.4 18.8 24.9 19.2 Radon levels in closed environment are affected both by the degree of exchange with outdoor air as measured by the ventilation rate and by changes in the entry rate of radon rich air from the underlying soil and rocks. Since majority of the houses are well ventilated in summer season, indoor radon concentrations might be expected to be lower for summer than in winter season (Wilkening, 1986). To get a clear idea of the spatial variations, the observed values are compared with the surveys made in different areas. The range of 222 Rn, 220 Rn and their progenies for each location are given in Table 4. The elevated radon levels are seen in poor ventilation houses of all the locations where most of the houses were built by local soil and sedimentary gravel. 
 
 136 Some buildings with higher radon levels were found on gravel but all the lower values observed in Rajajinagar area. This may be due to the lower activity concentrations of 226 Ra (Ashok et al, 2008) and also low radon in water (Hunse et al, 2010) in the surrounding region. Table 4: Area wise range of 222Rn, 220Rn and their progeny levels Name of the Location Rajajinagar Srinivasanagar Sheshadripuram Srirampuram Padhmanabhanagar Jayanagar Banashankari Malleshwaram Vijayanagar Gandhinagara RANGE 222 220 Concentration Concentration Rn Rn 222 220 of Rn of Rn progeny progeny Bq m–3 m WL 4.0 – 36.8 5.5 – 35.4 0.02 – 0.9 0.02 – 0.5 29.8 – 50.3 13.7 – 56.9 0.12 – 1.9 0.04 – 0.7 5.8– 100.0 2.7 – 72.9 0.02 – 1.6 0.02 – 0.9 10.9 – 65.9 6.1 – 30.9 0.06 – 1.1 0.02 – 1.9 4.0 – 76.0 3.4 – 70.1 0.01 – 1.5 0.02 – 3.5 4.0 – 80.7 4.8 – 63.1 0.02 – 2.2 0.01 – 1.7 5.8 – 89.4 1.3 – 66.6 0.02 – 4.4 0.01 – 1.3 5.8 – 92.9 2.0 – 47.9 0.02 – 2.2 0.02 – 4.8 11.7 – 99.4 6.7 – 37.5 0.03 – 1.4 0.02 – 0.9 73.6–100.0 10.9 – 72.9 0.23 – 4.4 0.03 – 1.0 Figure 7 shows the frequency distribution of 222Rn concentrations in 150 houses. About 81% of indoor 222Rn levels are found to vary between 4 and 39 Bq m–3. The higher concentrations (40 - 80 Bq m-3) were observed in 15% of the studied houses, this may be due to the buildings without the basic concrete slab or the slab that was not properly built or already damaged (Vaupotic et al, 1999). Nearly 4% of buildings show radon concentrations above 80 Bq m–3 with a maximum of 100 Bq m-3 and they were 40 year old. The poor construction of houses leads to the several cracks in foundation, walls, basic slabs thorough which radon can easily enter the rooms (Vaupotic et al, 1999). The observed values of radon concentration are found comparable with variation observed in the country and ranges from 6.4 to 95.4 Bq m–3 with a geometrical mean 25.5 Bq m–3 (Ramachandran, 2003). In general the radon concentration was found higher in mud houses than in cement houses. The ground floor of such houses is directly constructed on the top of soil with a coating of mud. The ground floor allows more radon to diffuse inside the houses because of higher porosity of materials used (Ramola et al, 1995). 
 
 137 Fig 7: Frequency distribution of radon concentrations (Bq m-3) Figure 8 shows the frequency distribution of indoor 220 Rn concentration. About 83% of the dwellings have shown the concentrations below 30 Bq m-3, 11% ranged between 31- 49 Bq m-3 and 6% of the dwellings showed the concentrations above 50 Bq m-3 with a maximum of 72.9 Bq m-3. The reported values of mean indoor radon and thoron concentrations for India is 
 Fig 8: Frequency distribution of thoron concentrations (Bq m-3) 
 138 23.0 and 12.2 Bq m–3 respectively and the total inhalation dose rate is 0.9 mSv y–1 (Ramachandran et al, 2003). A comparison of indoor 222Rn and 220Rn concentration for different seasons of all the studied locations are shown in Table 5. Due to poor ventilation, the radon is accumulated inside the houses and thus results higher concentration. The 222Rn and 220Rn concentrations were found higher in winter and low in summer. The high values in winter are mainly because of ventilation factor (Vaupotic et al, 1999). The indoor 222Rn and 220Rn were influenced mainly by the ventilation condition of the house. In Government Science College high 220 222 Rn and Rn concentrations in summer is observed than in winter. This anomaly observed in the college is may be due the fact that the class rooms will be closed for longer duration in summer holidays. Table 5: Location wise seasonal variations of indoor 222Rn and 220Rn concentrations Name of the Location Rajajinagar Srinivasanagar Sheshadripuram Srirampuram Padhmanabhanagar Jayanagar Banashankari Malleshwaram Vijayanagar Gandhinagara winter Rn 220Rn Bq m-3 24.9 18.6 36.6 19.9 43.9 25.9 38.8 21.8 41.9 35.2 37.4 22.4 41.8 31.5 50.3 26.9 61.9 21.3 73.6 49.7 222 summer Rn 220Rn Bq m-3 10.9 14.2 18.2 16.2 18.3 15.5 17.3 12.2 15.6 14.3 12.9 14.6 13.7 14.4 14.3 12.4 26.2 16.6 24.5 20.1 222 Rainy Rn 220Rn Bq m-3 14.8 15.3 24.9 15.3 28.3 18.1 20.5 20.5 22.6 24.8 23.2 20.1 22.2 18.2 19.7 13.9 35.6 17.8 40.3 24.5 222 Autumn Rn 220Rn Bq m-3 18.2 16.2 29.6 21.3 36.6 19.9 28.5 20.7 29.6 29.3 27.3 21.5 27.9 21.3 27.4 18.2 43.8 20.4 51.4 25.8 222 The emanation of radon also contributes higher radon from rocks and local stones. In addition, the mud houses have small doors and a small window, which remain closed for most of the time to conserve the energy (Ramachandran et al,2003). The observed winter/summer ratio was found maximum while the winter/autumn ratio was found minimum. The winter /summer ratio in different locations are found to vary between 1.9 and 3.7 and this ratio is high compared to the ratios of winter/rainy and winter/autumn. Again this is correlated with the ventilation condition of houses. The concentrations of 222Rn 
 139 and its progeny also follow the same trend as it was recorded maximum in winter and minimum in summer (Virk and Sharma, 2000; Wilkening, 1986; Vaupotic et al, 1999). However the 220 Rn and its progeny concentration was found maximum during winter and minimum during rainy. This behavior is may be due to low emanation rate of 220 Rn during rainy season and also it may be due to the possibility of brief half life, it cannot escape easily from the soil capillaries that are mostly occupied by water during the rainy season (Sathish et al, 2001). Conclusions It is observed that the concentrations of indoor 222 Rn, 220 Rn and their progeny levels are higher in poor ventilated houses than in well ventilated houses. The higher 222 Rn and 220 Rn concentrations are may be due to the presence of radioactive contents in the building. Radon levels in houses were found to be inversely related to room sizes. Thoron levels did not show much effect with increase in room volumes. Inhalation dose measured in the houses were comparable with natural background areas. Acknowledgements The research work is supported by the University Grants Commission in the form of grants under the Research Funding Council for major research project, X Plan, UGC, New Delhi, India. (F.No.32-46/2006 (SR) dated 22-02-2007). The cooperation extended by all the residents is highly appreciated. 
 140 References Abu-Jarad F and Fremlin JH (1983) Effects of internal wall covers on radon emanation inside houses, Health Physics, 44, pp. 243-248. Cross WG and Tommasino L (1970) Rapid reading technique for nuclear particle damage tracks in thin foils. Radiation Effects. 5: 85-89. Eappen KP and Mayya YS (2004) Calibration factors for LR-115 (type-II) based radon thoron discriminating dosimeter. Radiat. Meas. 38: 5-17. Mayya YS, Eappen KP and Nambi KSV (1998) Methodology for mixed field inhalation dosimetry in monazite areas using a twin cup dosimeter with three track detectors. Radiat. Prot. Dosim. 77: 177-181. Mishra UC, and Sadasivan S (1971) Natural radioactivity levels in Indian soil. J. Sci. and Ind. Res., 30, 59-62. Ningappa C, Sannappa J and Karunakara N (2008) Study on Radionuclides in granite quarries of Bangalore rural district, Karnataka, India, Radiation Protection Dosimetry, 131(4), 495–502. Ramachandran TV, Eappen KP, Nair RN, Mayya YS and Sadasivan S; Department of Atomic Energy, Government of India, Bhabha Atomic Research Center, Mumbai, India. Report BARC/2003/E/023: p 43 (2003). Ramola RC, Rawat RBS and Kandari MS, (1995) Estimation of Risk from Environmental Exposure to Radon in Tehri Garhwal, Nuclear Geophysics, 9, 383-386. Samyogi G, Hunyadi I and Varga Z (1978) Spark counting of alpha radiograms recorded on strippable cellulose nitrate LR-115 film. Nuclear Track Detectors. 2, 191-197. Sathish LA, Sannappa J, Paramesh L, Chandrashekara MS and Venkataramaiah P (2001) Studies on indoor radon/thoron and their progeny levels at Mysore city, Karnataka State, Indian Journal of Pure and Applied Physics, 39, 738-745. Schery SD and Gaeddert DH (1984) Factors affecting exhalation of radon from a gravelly sandy loam, Journal of Geophysical Research, 89, 7299-7309 Shiva Prasad NG, Nagaiah N, Ashok GV and Karunakara N (2008) Concentrations of 232 Th, and 40K in the soil of Bangalore region, INDIA. Health Phys. 94(3), 264-271. 226 Ra, Stranden E, (1980) Thoron and radon daughters in different atmospheres, Health Physics, 38, 777-785. United Nations Scientific Committee on the Effects of Atomic Radiations (UNSCEAR). Sources and Effects of Ionizing Radiation. UNSCEAR 1993 report to the general assembly, with annexes. pp. 73-98. (New York: United Nations) 
 141 UNSCEAR. Report to the General Assembly with Scientific Annexes, United Nations. Annexure B, pp. 84 -156 (2000). Vaupotic J, Sikovec M and Kobal I, (1999) Systematic Indoor measurements in Slovenian schools, Health Physics, 78, 559-562. 222 Rn and Gamma ray Virk HS and Navjeet Sharma, (2000) Indoor 222Rn/220Rn survey report from Hamirpur and Uuna districts, Himachal Pradesh, India, Applied Radiation and Isotopes, 52, 137-141. Wafaa A (2002). Permeability of Radon -222 through some materials. Radiat. Meas. 35, 207211. Ward WJ, Fleischer RL and Morgo C (1977) Barrier technique for separate measurement of Radon isotopes. Revised Science Instruments, 48, 1440-1441. Wilkening M , (1986) Seasonal Variation of indoor United States, Health Physics, 51, 427-436. 222 Rn at a location in the Southwestern 
 142