Derivation of a screening methodology for evaluating radiation dose to aquatic and terrestrial biota

Journal of Environmental Radioactivity 66 (2003) 41–59 www.elsevier.com/locate/jenvrad

Derivation of a screening methodology for evaluating radiation dose to aquatic and terrestrial biota Kathryn A. Higley a,∗, Stephen L. Domotor b, Ernest J. Antonio c, David C. Kocher d a

b

Oregon State University, Department of Nuclear Engineering, 100 Radiation Center, Corvallis, OR, 97331, USA US Department of Energy, Office Environmental Policy and Guidance, 1000 Independence Ave., SW, Washington, DC, 20585, USA c Pacific Northwest National Laboratory, PO Box 999, Richland, WA, 99352, USA d SENES Oak Ridge, Inc., Oak Ridge, TN 37830, USA Received 25 September 2000; accepted 20 November 2001

Abstract The United States Department of Energy (DOE) currently has in place a radiation dose standard for the protection of aquatic animals, and is considering additional dose standards for terrestrial biota. These standards are: 10 mGy/d for aquatic animals, 10 mGy/d for terrestrial plants, and, 1 mGy/d for terrestrial animals. Guidance on suitable approaches to the implementation of these standards is needed. A screening methodology, developed through DOE’s Biota Dose Assessment Committee (BDAC), serves as the principal element of DOE’s graded approach for evaluating radiation doses to aquatic and terrestrial biota. Limiting concentrations of radionuclides in water, soil, and sediment were derived for 23 radionuclides. Four organism types (aquatic animals; riparian animals; terrestrial animals; and terrestrial plants) were selected as the basis for development of the screening method. Internal doses for each organism type were calculated as the product of contaminant concentration, bioaccumulation factor(s) and dose conversion factors. External doses were calculated based on the assumption of immersion of the organism in soil, sediment, or water. The assumptions and default parameters used provide for conservative screening values. The screening methodology within DOE’s graded approach should prove useful in demonstrating compliance with biota dose

∗

Corresponding author. Tel.: +1-541-737-0675; fax: +1-541-737-048. E-mail address: [email protected] (K.A. Higley).

0265-931X/03/$ - see front matter Published by Elsevier Science Ltd. doi:10.1016/S0265-931X(02)00115-7

42

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

limits and for conducting screening assessments of radioecological impact. It provides a needed evaluation tool that can be employed within a framework for protection of the environment. Published by Elsevier Science Ltd. Keywords: Radiation dose; Plants; Animals; Screening methods

1. Introduction In the United States, the Department of Energy (DOE) has in place a radiation dose standard for the protection of aquatic animals, and is considering dose standards for terrestrial biota. These standards are: 10 mGy/d for aquatic animals, 10 mGy/d for terrestrial plants, and 1 mGy/d for terrestrial animals. The DOE’s Office of Environmental Policy and Guidance has also established a Biota Dose Assessment Committee (BDAC) to assist in the development of technical standards, methods, and guidance in assessing radiation doses to biota (USDOE, 1998). Although the biota-protection standards are dose-based, it is problematic to directly measure dose to any specific organism as a means of demonstrating compliance. The inherent complexity of environmental systems, and the vast array of biota which can potentially be exposed to radionuclides in the environment, make this a Herculean task. Instead, a more practical means of demonstrating compliance with the dose standards is needed. One of the simpler methods is to derive radionuclide concentration limits for environmental media (e.g., water, sediment, or soil) that would be acknowledged to be generally protective of all biota. However, this simplistic approach is of necessity very conservative for most exposure situations. Consequently, it was decided that a graded approach to evaluating compliance would be the most appropriate. DOE’s graded approach consists of a three-step process (Fig. 1), which is designed to guide a user from an initial, conservative general screening process to, if needed, a more rigorous analysis using site-specific information. The three-step process includes: (1) assembling environmental radionuclide concentration data and general knowledge of sources, receptors, and routes of exposure for the area to be evaluated; (2) applying a general screening methodology; and (3) conducting analysis through site-specific screening, site-specific analysis, or an actual site-specific biota dose assessment within an ecological risk framework, if needed (DOE, 2002). The focus of this paper is on the general and site-specific screening elements of the graded approach. The first step in evaluating compliance is to compare measured environmental concentrations with the very conservative (i.e., very restrictive or protective) concentration limits for environmental media (e.g., Biota Concentration Guides, or BCGs, for water, sediment, or soil) in a general screening process. To be useful in general screening, the concentration limits (BCGs) must be set so that doses received by real biota exposed to such concentrations are not expected ever to exceed the biota dose limits. Since the screening limits would be chosen to protect ‘all biota, everywhere’ they would, by their nature, be restrictive, and in many circumstances overly

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

Fig. 1.

43

DOE’s graded approach.

conservative with regards to specific environments. Users also need to be able to examine and revise, if appropriate, the screening limits in a site-specific screening process to more realistically reflect the conditions at their site. This approach parallels methods currently used to protect human health from residual levels of radionuclides in the environment (e.g., site-specific conditions can be considered in deriving allowable residual radionuclide concentration levels).

2. Method 2.1. Basis for general screening Several factors had to be considered in developing the general screening methodology. The method had to be simple, defensible, and user-friendly; it had to have broad applicability—encompassing both aquatic and terrestrial ecosystems; it had to address radiation dose in small organisms (e.g., mice) and large carnivores (e.g., cougars); and it had to provide a logical and consistent departure point if an in-depth evaluation of dose was required. Should such additional analysis be required, the method had to utilize existing data—either from the technical literature or from sitespecific monitoring—whenever possible. Lastly, the method had to be useful in evaluating the combined impacts of exposure to multiple contaminated media (water, sediment, and soil). The choice of organisms for this methodology evolved from consideration of the existing and proposed radiation protection standards for biota (DOE, 2002). In the

44

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

US, biota standards had been set for aquatic animals, and were being considered for terrestrial plants and animals. Accordingly, the screening methodology had to accommodate these three general categories. A fourth, riparian animal, was added after recognizing that the riparian pathways of exposure combined aspects of both the terrestrial and aquatic systems. The pathways of exposure, for each of the four organism types, were developed based on consideration of the likelihood of dose occurring through a specific route, or ‘pathway’. Relevant pathways can involve both internal exposure (derived from incorporation of radionuclides into the tissues of organisms) and external exposure (derived from exposure to penetrating radiations emitted by radionuclides in the environmental media). Based on the potential pathways of exposure, BCGs were derived for surface water, sediment, and soil. Calculated using conservative assumptions, the BCGs are intended to preclude the relevant biota from being exposed to radiation levels in excess of the relevant existing or recommended biota dose limits. 2.2. Basis of the BCGs The methodology used to demonstrate compliance with a dose limit is based on the fact that biota dose is a function of the contaminant concentration in the environment, and is the sum of internal and external contributions. It is possible, given a unit concentration (i.e., 1 Bq kg⫺1) of a contaminant in a single medium (e.g., soil) to estimate the potential dose rate to a receptor from both internal and external exposures (several assumptions must be made to do so, and these are described in the following sections). Once the dose rate from this unit concentration of contaminant has been calculated, it can be used to back calculate a concentration of the contaminant in the medium that will generate a dose rate at any specified limit: BCGi(Bq·kg⫺1) ⫽

DL (Gy·y⫺1) . DCFi(Gy·y⫺1 / Bq·kg⫺1)

(1)

Where BCGi (Bq kg⫺1) is the concentration of nuclide i in some defined medium which, based on the screening level assumptions, corresponds to a dose rate of DL (Gy y⫺1) to a specific organism; and, DCFi (Gy y ⫺ 1 / Bq kg ⫺1) is the dose conversion factor for radionuclide i in the medium from which the organism is exposed (for consistency with the DOE Technical Standard (DOE, 2002) and for dimensional accuracy, the dose rates used in the equations presented in this report are expressed in Gy y⫺1. However, it is recognized that the relevant time scales are on the order of Gy d⫺1, or shorter). Eq. (1) is simplistic in that it does not specify the source of the exposure—internal or external. For the DOE BCGs, where both modes of exposure need to be considered, the equation was revised as follows: BCGi(Bq·kg⫺1) ⫽ (DCFi, internal(Gy·y

(2) ⫺1

DL(Gy·y⫺1) . / Bq·kg )·LPi) ⫹ (DCFi, external(Gy·y⫺1 / Bq·kg⫺1)) ⫺1

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

45

Where DCFi,internal (Gy y⫺1/Bq kg⫺1) is the dose conversion factor for radionuclide i incorporated in the tissue of the organism; LPi is the ratio of the radionuclide concentration in the medium external to the organism to the concentration in the tissues of the organism; DCFi,external (Gy y⫺1/Bq kg⫺1) is the dose conversion factor for radionuclide i in the medium external to the organism; and, all other terms have been defined. If multiple contaminated media are present, then the dose assessment can be performed for each, with the results individually expressed as a ratio to the standard. Compliance is demonstrated when this ‘sum of fractions’ (an approach commonly used in evaluating compliance for humans exposed to radionuclides discharged to air, soil and water) is less than one. 2.3. External dosimetry External dose conversion factors (DCFs) provide dose rates from external exposure per unit concentration of a radionuclide in a specific environmental medium (e.g., water, soil, or sediment) surrounding an organism. In the derivation of the external DCFs described for this work, only penetrating radiations (photons and electrons) were included, non-penetrating radiations (e.g., alpha particles) were not. Animals with a thick epidermis or other radioresistant outer covering will not be affected by the dose from external alpha radiation (Till, 1978). For very small organisms (approximately less than 0.05 cm), the statistics of radioactive decay and the lifespan (or lifestage) of the organism are important factors in determining dose from alpha particle emitting contaminants. For purposes of simplifying the determination of external dose factors, the contribution from alpha emitting radiations was not included. The principal environmental (source) media of concern are contaminated water and sediments for exposure of aquatic and riparian animals and contaminated soil and water for exposure of terrestrial biota. While implicitly included in the derivation of limiting media concentrations, separate DCFs for air were not provided. The external DCFs for the active air pathway are very similar for all organisms, including humans. Air release limits for humans are sufficiently restrictive that at DOE sites with active air releases, biota are unlikely to receive significant doses solely from this exposure pathway. The active air pathway would contribute only a small fraction to the total biota dose. External DCFs were derived based on the following assumptions: 앫 The source medium (water, sediments, or soil) is assumed to be infinite in extent and to contain uniform concentrations of radionuclides. These assumptions result in reasonably realistic estimates of dose rates for radionuclides which are dispersed in the source medium, because the range of electrons emitted in radioactive decay is no more than a few centimeters and the mean-free-path of emitted photons is no more than a few tens of centimeters (Shleien et al., 1998). This approach is also conservative in cases where the radionuclide is not homogeneously distributed throughout the volume, as it overestimates the dose rate.

46

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

앫 The exposed organism is assumed to be very small (less than the mean free path of the electrons emitted in decay). This assumption results in overestimates of external dose rates for any finite-sized organism, because the attenuation of photons and electrons as they penetrate an organism is ignored. Because all the energy emitted by radionuclides in a uniformly contaminated and infinite source medium is absorbed uniformly throughout the medium, the dose rate in the organism is essentially the same as the dose rate in the medium itself. The absorbed dose rate can thus be calculated directly from the energy of photons and electrons emitted per disintegration of the radionuclides in the medium. 앫 Because the organism is assumed to be very small, the energies of all photons and electrons emitted by radionuclides are taken into account in calculating the screening-level external dose conversion factors. This approach is particularly conservative for low-energy electrons when the irradiated tissues of concern lie below the body surface of an organism. In reality, lower-energy electrons could not penetrate to the location of these tissues. Taking into account the energies of all photons and electrons in radioactive decay is tantamount to assuming that the radiosensitive tissues of concern lie on the surface of a very small organism. The approach to calculating screening-level external dose conversion factors is simple, because the dose coefficients are calculated based only on the known energies and intensities of photons and electrons emitted in the decay of radionuclides. Screening-level external dose conversion factors for exposure of aquatic animals to radionuclides in sediments and water are calculated based on the assumptions that the organism is located 100% of the time at the water-sediment interface. Thus, it is assumed that the organism is exposed at the boundary of two semi-infinite and uniformly contaminated media. The assumption of exposure at the boundary of a semiinfinite medium results in an absorbed dose rate in the organism that is one-half of the dose-rate in an infinite source volume. The calculation of the screening-level external dose conversion factors for aquatic animals then proceeds as follows. The total energies of all photons and electrons emitted in the decay of radionuclides are given in units of MeV per disintegration (i.e., per Bq-s). For exposure to contaminated sediments, the desired units for the external dose conversion factors are Gy y⫺1 per Bq kg⫺1. The emitted energy in MeV per disintegration is expressed in terms of the desired units for the external dose conversion factors by multiplication of the known factors relating energy in MeV to joules (J), absorbed energy in J kg⫺1 to Gy, and time in seconds to years:

冉

1

冊冉

冊冉 冊冉

MeV J kg Gy s · 1.6 × 10⫺13 · · 3.156 × 107 Bq ⫺ s MeV J y

冊

(3)

Gy·y⫺1 . ⫽ 5.05 × 10⫺6 Bq·kg⫺1 As noted above, the external dose coefficient at the sediment-water interface is one-half of the value for exposure in an infinite medium. Therefore, given the total

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

47

energies (E) of photons and electrons in MeV per disintegration of a radionuclide, the external dose coefficient (dext) for exposure to contaminated sediments is given by: (dext)sediments

冉

Gy·y-1 Bq·kg-1

冊

冉 冊

⫽ (2.525 × 10⫺6)Ephotons+electrons

MeV . dis

(4)

For exposure to contaminated water, the desired units for the external dose conversion factors are Gy y⫺1 per Bq m⫺3. If the density of water is assumed to be 1 g cm⫺3, the external dose coefficient for exposure to contaminated water at the sediment-water interface is obtained from a calculation similar to that for contaminated sediments given above as:

冉

(dext)water

冊

Gy·y⫺1 Bq·m⫺3

冉 冊

⫽ (2.525 × 10⫺9)Ephotons+electrons

MeV . dis

(5)

The screening-level external dose conversion factors for exposure of aquatic animals to selected radionuclides in contaminated sediments and water calculated as described above are given in Table 1. The energies of all photons and electrons per disintegration of the radionuclides are obtained from the compilation by Kocher (1981), which summarizes the data contained in a handbook of decay data tables (Kocher, 1980). For most radionuclides, the decay data compiled by Kocher are in good agreement with the data compiled by the ICRP (1983). Screening-level external dose conversion factors for exposure of terrestrial biota to radionuclides in soil are calculated based on the assumption that the organism is immersed 100% of the time in an infinite and uniformly contaminated source region. This assumption takes into account that some terrestrial animals reside well below ground for a substantial fraction of the time, and it is appropriately conservative for purposes of screening. For exposure to contaminated soil, the desired units for the external dose conversion factors are Gy y⫺1 per Bq kg⫺1. Therefore, based on the calculations for contaminated sediments discussed previously, the external dose coefficient for exposure to contaminated soil is given by: (dext)soil

冉

Gy·y⫺1 Bq·kg-1

冊

冉 冊

⫽ (5.05 × 10⫺6)Ephotons + electrons

MeV . dis

(6)

The screening-level external dose conversion factors for exposure of terrestrial biota to selected radionuclides in contaminated soil calculated as described above are given in Table 1. The values for contaminated soil are twice the values for contaminated sediments. Several radionuclides—including 90Sr, 95Zr, 125Sb, 137Cs, 144Ce, 226Ra, 228Ra, 235U, and 238U—have radioactive decay products that are sufficiently short-lived that they can be assumed to be in activity equilibrium with the parent radionuclide in each environmental medium. For these radionuclides, the external dose conversion factors are the sum of the values for the parent and its indicated short-lived decay products, taking into account the branching fractions in the decay of the parent. The decay products are listed in Table 2.

48

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

Table 1 Dose conversion factors derived for the general screening methodology Parent Nuclide

241

Am Ce 135 Cs 137 Cs 60 Co 154 Eu 155 Eu 3 H 129 I 131 I 239 Pu 226 Ra 228 Ra 125 Sb 90 Sr 99 Tc 232 Th 233 U 234 U 235 U 238 U 65 Zn 95 Zr 144

Half-Life

432.2 y 284 d 2.3×106 y 30 y 5.26 y 16 y 1.811 y 12.3 y 1.57×107 y 8.05 d 2.41×104 y 1600 y 5.75 y 2.77 y 29.12 y 2.13×105 y 1.4×1010 y 1.58×105 y 2.44×105 y 7.04×108 y 4.47×109 y 245 d 65.5 d

Internal, Gy year⫺1 Bq⫺1 kg wet

5.6×10⫺4 6.8×10⫺6 3.4×10⫺7 4.3×10⫺6 1.3×10⫺5 7.6×10⫺6 6.2×10⫺7 2.9×10⫺8 4.5×10⫺7 2.9×10⫺6 5.3×10⫺4 3.0×10⫺3 3.6×10⫺3 2.7×10⫺6 5.7×10⫺6 5.1×10⫺7 4.1×10⫺3 4.9×10⫺4 4.9×10⫺4 4.5×10⫺4 4.4×10⫺4 3.0×10⫺6 8.4×10⫺6

External

Water Gy year⫺1 Sediment Gy year⫺1 Bq⫺1 kg Bq⫺1 m3 dry

Soil Gy year⫺1 Bq⫺1 kg dry

1.4×10⫺10 3.4×10⫺9 1.4×10⫺10 2.0×10⫺9 6.6×10⫺9 3.8×10⫺9 3.1×10⫺10 1.4×10⫺11 2.0×10⫺10 1.4×10⫺9 1.4×10⫺11 6.8×10⫺9 3.4×10⫺9 1.4×10⫺9 2.8×10⫺9 2.1×10⫺10 3.0×10⫺11 9.3×10⫺12 3.2×10⫺11 9.4×10⫺10 2.3×10⫺9 1.5×10⫺9 4.2×10⫺9

2.9×10⫺7 6.8×10⫺6 2.8×10⫺7 4.0×10⫺6 1.3×10⫺5 7.7×10⫺6 6.2×10⫺7 2.9×10⫺8 4.0×10⫺7 2.9×10⫺6 2.8×10⫺8 1.4×10⫺5 6.9×10-6 2.9×10⫺6 5.7×10⫺6 4.3×10⫺7 6.1×10⫺8 1.9×10⫺8 6.5×10⫺8 1.8×10⫺6 4.6×10⫺6 3.0×10⫺6 8.4×10⫺6

1.4×10⫺7 3.4×10⫺6 1.4×10⫺7 2.0×10⫺6 6.6×10⫺6 3.8×10⫺6 3.1×10⫺7 1.4×10⫺8 2.0×10⫺7 1.4×10⫺6 1.4×10⫺8 6.8×10⫺6 3.4×10⫺6 1.4×10⫺6 2.8×10⫺6 2.1×10⫺7 3.0×10⫺8 9.3×10⫺9 3.2×10⫺8 9.4×10⫺7 2.3×10⫺6 1.5×10⫺6 4.2×10⫺6

For several radionuclides, however, the external dose conversion factors do not include possible contributions from decay products that are sufficiently long-lived that they may not be in activity equilibrium with the parent radionuclide, even though the contributions from the decay products may be significant. The radionuclides of concern (with the decay products in parentheses) include 226Ra (210Pb) and 228Ra (228Th). If separate data on the concentrations of the shorter-lived decay products in sediments, water, or soil are not available, the decay products could be assumed to be in activity equilibrium with the parent, and the dose coefficients for the parent and the decay products should be added. This approach may or may not be conservative, depending on differences in the environmental behavior of the parent and its decay products. For most radionuclides, the concentration in aquatic animals relative to the concentration in water would be considerably greater than unity (Kennedy and Strenge, 1992). Therefore, the dose rate from internal exposure, calculated for purposes of

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

49

Table 2 Progeny included in internal and external dose conversion factors Parent Nuclide

Internal DCF

External DCF

137

Cs

137m

137m

226

Ra

222 210

228

Ra

228 212

90

Sr

232

Th

90

Ba

Rn, 218Po, 218Pb, Pb, 210Bi, 210Po

216

Ra, Po,

235

U

231

Th

238

U

234

Th,

95

Zr

95

Bi,

Ac, 228Th, 224Ra, 220Ra, Pb, 212Bi, 212Po, 208Tl

Nb

214

Po,

216

222 Rn, 214Pb, 214Bi, and 214Po. Possible contributions to dose coefficient from 210Pb decay products are not included, but dose coefficient for decay product is listed separately.

Po, Possible contributions to dose coefficient from 228 Th decay products are not included, but dose coefficient for decay product is listed separately. 90

Y

228

214

Ba

Y

228 212

Ac, 228Th, 224Ra, 220Ra, Possible contributions to dose coefficient from Pb, 212Bi, 212Po, 208Tl 228Ra and 228Th decay products are not included, but dose coefficients for decay products are listed separately. 231

234

Pa

Th

Short-lived decay products include

234

Th,

234

Pa

95

Nb

screening by assuming that all radiations emitted in the decay of radionuclides in an organism are absorbed in the organism, would usually be considerably higher than the screening-level dose rate from external exposure. In addition, for most radionuclides, the solid/solution distribution coefficient (Kd) in sediments would be considerably greater than unity (Onishi et al., 1981). Therefore, for the assumption of exposure at the sediment-water interface, the screening-level dose rate from external exposure to contaminated sediments should be higher in most cases than the corresponding dose rate from external exposure to contaminated water. Based on these arguments, the screening-level external dose conversion factors for exposure of aquatic animals to contaminated water in Table 1 are unlikely to be important for most radionuclides in determining screening-level concentrations in water. Rather, the screening-level concentrations of most radionuclides in aquatic environments should be based on considerations of external exposure to contaminated sediments and internal exposure. A comparison of the screening-level external dose conversion factors obtained in this report with those values given by Amiro (1997) also merits attention. The calculations of Amiro assumed that the organism is located 0.1 m below the surface of a semi-infinite, uniformly contaminated body of sediment, water, or soil. Compared with the assumptions of exposure in an infinite medium (soil) or at the boundary of

50

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

a semi-infinite medium (sediments and water) used in this paper, Amiro’s assumption is less conservative for exposure to contaminated soil but more conservative for exposure to contaminated sediments and water. In addition, the external dose conversion factors of Amiro were calculated for a human phantom, rather than a point receptor, and the calculated values for photons apply at the body surface and the calculated values for electrons apply at a depth of 70 µm in tissue. For high-energy photon emitters, the photon dose rate at the body surface of a human phantom is slightly higher than the dose rate in the source medium itself, but the difference is not significant. However, the tissue depth used by Amiro in calculating the electron dose rate is considerably less conservative than the assumption used in this paper. In Amiro’s work, the minimum electron energy that results in a non-zero dose at a depth of 70 µm is about 70 keV (Kocher and Eckerman, 1981) but in the methodology presented here, all such lower-energy electrons are taken into account. Finally, in the approach to screening developed by Amiro, the external dose conversion factors cannot be calculated simply on the basis of the energies of photons and electrons in radioactive decay, and results for radionuclides not considered by Amiro are not readily obtainable. 2.4. Internal dosimetry Internal dose conversion factors (Gy y⫺1 per Bq kg⫺1) were derived for unit concentrations of each radionuclide. 앫 Reference decay energies and abundances were taken from ICRP 38 (1983) for each radionuclide and its progeny (the external DCFs were derived based on the compilation by Kocher (1981), whereas the internal DCFs use ICRP 38 (1983 values). The two separate references simply reflect the choice of source data by the co-authors. However, the decay data compiled by Kocher are in good agreement with the data compiled by the ICRP (1983).) 앫 The default dose factor includes buildup of progeny with half-lives less than 100 y (progeny included in the calculations are listed in Table 2). 앫 The calculations assume all of the energies of radioactive decay were absorbed in the tissue of the organism (i.e., the organism was presumed to be extremely large). 앫 The radionuclides were presumed to be homogeneously distributed in the tissue. 앫 The default internal dose factors include a dose-modifying factor of 20 for alpha particles. The use of radiation weighting factors is being debated (Kocher and Trabalka, 2000), but the selection of a value of 20 for alpha particles is consistent with the current practice in radiation protection. The methodology was constructed such that the dose-modifying factor can be changed. The dose factors were calculated as the sum of all decay energies and multiplied by appropriate unit conversion factors. The resultant dose factors are presented in Table 1. For exposure to internal contaminants, the dose coefficients were calculated in units of Gy y⫺1 per Bq kg⫺1 of wet tissue using the following equation: DCFinternal, i

(7)

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

⫽

冉 冊

冉

51

冊冉 冊

1 s J Gy (⌺⌺YjEjQjMev)(3.1536 × 107 ) 1.6022 × 10⫺13 . Bq·s ˙ j y Mev J·kg-1

Where DCFinternal,i=Gy y⫺1 per Bq kg⫺1 of wet tissue for radionuclide i; Yj=yield (abundance) of radiation j per disintegration of nuclide i; Ej=energy (MeV) of radiation j for nuclide i; and Qj is the dose modifying factor (quality factor) for radiation j of nuclide i. 3. Results 3.1. Calculating BCGs in the general screening phase For most radionuclides, the single most important predictor of biota dose is the method used to estimate internal tissue concentrations. For the general screening phase of the graded approach, lumped parameters (LPs) were used to provide estimates of concentrations of contaminants in tissues of organisms, and ultimately derive the BCG corresponding to each radionuclide, medium, and organism type. The technical literature contains reference to empirically based parameters that describe concentrations of contaminants in an organism relative to the surrounding medium. These ratios are called ‘concentration ratios’, ‘concentration factors’, or ‘wet-weight concentration ratios’. These lumped parameter (e.g., Biv) values are available for many radionuclides for plant:soil and for aquatic species:water. In a few instances they are also available for animal:soil or sediment. The advantage of using one of these factors is that it allows the prediction of tissue concentration based on simple measurements of contamination in environmental media such as water, sediment and soil. The selection of a value for this lumped parameter becomes problematic, however, when considering the range of organism types meant to be covered by the graded approach. For example, there are very limited data available for riparian and terrestrial animals (i.e., very few animal:water, animal:soil, and animal:sediment concentration ratios). Tabulated values of the lumped parameters used in this methodology are provided in Table 3. As the graded approach methodology evolved it became apparent that these data gaps (e.g., for selecting appropriate lumped parameters) needed to be addressed. Two alternative approaches for deriving and selecting lumped parameters were evaluated. These are discussed in the subsequent two papers (Higley et al., 2002a,2002b). BCGs were calculated for each of the biota (aquatic animal, riparian animal, terrestrial animal, and plant) for exposure to radionuclides in water, soil and sediment. The equations used to calculate each follow. 3.2. Sediment BCGs for aquatic animals The conceptual model for aquatic animals places the organism at the sedimentwater interface. In this screening model, sediment presents an external dose hazard

52

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

Table 3 Lumped parameters used in the screening methodology Nuclide

241

Am Ce 135 Cs 137 Cs 60 Co 154 Eu 155 Eu 3 H 129 I 131 I 239 Pu 226 Ra 228 Ra 125 Sb 90 Sr 99 Tc 232 Th 233 U 234 U 235 U 238 U 65 Zn 95 Zr 144

Aquatic animal (wet weight) to water

4.0×102 9.0×103 2.2×104 2.2×104 2.0×103 6.0×102 6.0×102 2.0×10⫺1 2.2×102 2.2×102 1.0×103 3.2×103 3.2×103 1.0×102 3.2×102 7.8×101 8.0×101 1.0×103 1.0×103 1.0×103 1.0×103 1.7×104 1.6×103

Riparian animal

Terrestrial Terrestrial plant (wet animal weight) to soil

Animal (wet weight) to sediment

Animal (wet weight) to water

3×10⫺3 5×10⫺4 3×10⫺1 3×10⫺1 1×10⫺2 4×10⫺3 3×10⫺3 4×10⫺1 3×10⫺1 1×10⫺1 3×10⫺3 3×10⫺2 3×10⫺2 4×10⫺4 2×100 5×10⫺2 2×10⫺3 4×10⫺3 4×10⫺3 4×10⫺3 4×10⫺3 2×100 3×10⫺3

1×101 3×101 5×104 5×104 2×102 2×101 2×101 8×10⫺1 6×102 2×102 3×101 8×102 8×102 3×10⫺1 6×103 3×101 1×100 3×101 3×101 3×101 3×101 2×105 4×101

Animal to soil Animal (wet (wet weight) weight) to to soil water 8×10⫺3 4×10⫺2 1×101 1×101 2×10⫺1 4×10⫺2 4×10⫺2 1×100 4×10⫺1 4×10⫺1 1×10⫺2 1×10⫺1 1×10⫺1 1×10⫺2 4×100 8×100 1×10⫺3 4×10⫺3 4×10⫺3 4×10⫺3 4×10⫺3 3×10⫺1 3×10⫺2

4×10⫺3 6×10⫺3 1×102 1×102 8×10⫺2 5×10⫺3 4×10⫺3 1×100 3×100 3×100 3×10⫺3 6×10⫺2 6×10⫺2 4×10⫺4 8×101 3×100 2×10⫺3 4×10⫺3 4×10⫺3 4×10⫺3 4×10⫺3 7×100 4×10⫺3

9×10⫺2 8×10⫺3 3×100 3×100 1×10⫺1 1×10⫺1 9×10⫺2 1×100 3×100 1×100 9×10⫺2 4×10⫺1 4×10⫺1 5×10⫺3 3×101 8×10⫺1 5×10⫺2 5×10⫺2 5×10⫺2 5×10⫺2 5×10⫺2 2×101 3×10⫺2

to the aquatic animal, with the BCG therefore based on a semi-infinite exposure model. Uptake of contaminants from the sediment to the organisms are implicitly addressed via the empirical organism to water lumped parameter discussed in following sections. The equation used to derive the aquatic animal BCGs for exposure to a single nuclide in contaminated sediment is: BCG(sediment)i,aquatic

animal

⫽

365.25DLaa . CFaa ·DCFext,sediment,i

(8)

Where BCG(sediment)i,aquatic animal (Bq kg⫺1) is the concentration of nuclide i in sediment which, based on the screening level assumptions, corresponds to a dose rate of DLaa, (0.01 Gy d⫺1) to the aquatic animal; 365.25 (d per y) is a conversion factor; DLaa (0.01 Gy d⫺1) is the recommended dose limit for aquatic animals; DCFext,sediment,i (Gy y⫺1 per Bq kg⫺1) is the external dose conversion factor used to

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

53

estimate the dose rate to the tissues of the aquatic animal from nuclide i in the sediment; and CFaa (dimensionless) is the correction factor for area or time. This correction factor is set at a default of 1. 3.3. Water BCGs for aquatic animals The conceptual model for aquatic animals places the organism at the sedimentwater interface. In this screening model, water presents both an internal and external dose hazard to the aquatic animal. Lumped parameters (e.g., bioaccumulation factors) are used to estimate the extent of internal contamination (and by extension, the dose), and external exposure is assessed with a semi-infinite source term. The method used to derive the screening-level aquatic animal BCGs for exposure to a single nuclide in contaminated water is: BCG(water)i,aquatic animal ⫽

(9)

365.25DLaa . CFaa[(0.001 Biv,aa·DCFinternal,i) ⫹ (DCFexternal, water, i)]

Where BCG(water)i,aquatic animal (Bq m⫺3) is the concentration of nuclide i in water which, based on the screening level assumptions, corresponds to a dose rate of DLaa (0.01 Gy d⫺1) to the aquatic animal; DLaa (0.01 Gy d⫺1) is the recommended dose limit for aquatic animals; 0.001 is the conversion factor for L to m3; Biv,aa,i (L kg⫺1) is the fresh mass aquatic animal to water concentration factor for nuclide i; DCFinter⫺1 per Bq kg⫺1) is the internal dose conversion factor used to estimate the nal,i (Gy y dose rate to the tissues from nuclide i in tissues; DCFexternal, water,i (Gy y⫺1 per Bq m⫺3) is the external dose conversion factor used to estimate the dose rate to the aquatic animal from submersion in contaminated water; and, all other terms have been defined. 3.4. Sediment BCGs for riparian animals The conceptual model for riparian animals also places the organism at the sediment-water interface (as does the aquatic animal model). However, in this screening model, sediment presents both an internal and external dose hazard to the riparian animal. Lumped parameters are used to estimate the extent of internal contamination (and by extension, the dose), and external exposure is assessed with a semi-infinite source term. The equation used to derive the riparian animal BCGs for exposure to a single nuclide in contaminated sediment is: BCG(sediment)i,riparian animal ⫽

(10)

365.25DLra . CFra[(LPra,sed,i·DCFinternal,i) ⫹ (DCFext,sediment,i)]

Where BCG(sediment)i,riparian animal (Bq kg⫺1) is the concentration of nuclide i in sediment, based on the screening level assumptions, corresponds to a dose rate of

54

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

DLra (0.001 Gy d⫺1) to the riparian animal; DLaa (0.001 Gy d⫺1) is the recommended dose limit for riparian animals. LPra,sed,i (dimensionless) is the fresh mass riparian animal to sediment concentration factor of nuclide i; CFra (dimensionless) is the correction factor for area or time for the riparian organism. This correction factor is set at a default of 1; and all other terms have been defined. 3.5. Water BCGs for riparian animals As noted previously, the conceptual model for riparian animals has the animal situated at the sediment-water interface. In assessing potential contributors to dose, water presents both an internal and external dose hazard. As before, lumped parameters are used to estimate the extent of internal contamination. External exposure is assessed with a semi-infinite source term. The equation used to derive the screening-level riparian animal BCG for exposure to a single nuclide in contaminated water is as follows: (11)

BCG(water)i,riparian animal ⫽

365.25DLra . CFra[(0.001LPra,water,i·DCFinternal,i) ⫹ (DCFext,water,i)]

Where BCG(water)i, riparian animal (Bq m⫺3) is the concentration of nuclide i in water, which based on the screening level assumptions, corresponds to a dose rate of DLra (0.001 Gy d⫺1) to the riparian animal; LPra,water, i (L kg⫺1) is the fresh mass riparian animal to water concentration factor of nuclide i; and all other terms have been defined. 3.6. Soil BCGs for terrestrial plants In this screening model, soil provides both an internal and external dose hazard to the plant. The conceptual model for terrestrial plants is based on the entire plant being surrounded by soil. While many plants may have a substantial portion of their mass above ground, the BCG thus derived, will be conservative. Lumped parameters (e.g., bioaccumulation factors) are used to estimate the extent of internal contamination (and by extension, the dose), and external exposure is assessed using an infinite source term. The lumped parameters used in the model account for aerial deposition onto plant surfaces with subsequent uptake. The equation used to derive the BCGs for terrestrial plant exposure to a single nuclide in contaminated soil is: BCG(soil)i,terrestrial plant ⫽

365.25DLtp . CFtp[(Biv,tp,i·DCinternal,i) ⫹ (DCFext,soil,i)]

(12)

Where BCG(soil)i,terrestrial plant (Bq kg⫺1) is the concentration of nuclide i in soil which, based on the screening level assumptions, numerically equates to a dose rate of DLtp (0.01 Gy d⫺1) to the terrestrial plant; DLtp (0.01 Gy d⫺1) is the recommended dose limit for terrestrial plants; Biv,tp,i (dimensionless) is the fresh mass terrestrial plant to soil concentration factor; CFtp (dimensionless) is the correction factor for

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

55

area or time. This correction factor is set at a default of 1; DCFext, soil,i (Gy y⫺1 per Bq kg⫺1) is the external dose conversion factor used to estimate the dose rate to the plant tissues from nuclide i in surrounding soils; and all other terms are as previously defined. 3.7. Water BCGs for terrestrial plants The conceptual model for terrestrial plants is based on the entire plant being surrounded by soil. However, the potential for exposure to contaminated water—from soil pore water or from irrigation—exists. As a compromise to the methodology, external exposure from water was added. In this screening model, the BCG for water is based on a semi-infinite exposure model. The equation used to derive the BCGs for terrestrial plant exposure to a single nuclide in contaminated water is: BCG(water)i,terrestrial plant ⫽

365.25DLtp . CFtp·DCFext,water,i

(13)

Where BCG(water)i,terrestrial plant (Bq m⫺3) is the concentration of nuclide i in water which, based on the screening level assumptions, corresponds to a dose rate of DLtp (0.01 Gy d⫺1) to the terrestrial plant; and all other terms are as previously defined. 3.8. Soil BCGs for terrestrial animals The screening conceptual model for terrestrial animals has the animal surrounded by soil. In assessing potential contributors to dose, soil presents both an internal and external dose pathway. As before, lumped parameters are used to estimate the extent of internal contamination (e.g., as might occur from ingestion or inhalation). External exposure is assessed with an infinite source term. The equation used to derive the terrestrial animal BCGs for exposure to a single nuclide in contaminated soil is: BCG(soil)i,terrestrial animal ⫽

365.25DLta . Cta[(LPta,soil,i·DCFinternal,i) ⫹ (DCFext,soil,i)]

(14)

Where BCG(soil)i, terrestrial animal (Bq kg⫺1) is the concentration of nuclide i in soil which, based on the screening level assumptions, corresponds to a dose rate of DLta (0.001 Gy d⫺1) to the terrestrial animal; DLta (0.001 Gy d⫺1) is the recommended dose limit for terrestrial animals. LPta,soil,i (dimensionless) is the fresh mass terrestrial animal to soil concentration factor of nuclide i; CFta (dimensionless) is the correction factor for area or time for the terrestrial organism. This correction factor is set at 1 for the general screening phase of the calculations; and all other terms have been defined. 3.9. Water BCGs for terrestrial animals The conceptual model for terrestrial animals is based on the entire animal being surrounded by soil. However, the potential for exposure to contaminated water from soil pore water or by drinking from contaminated ponds or rivers—exists. Water

56

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

presents both an internal and external dose hazard. As before, lumped parameters are used to estimate the extent of internal contamination (e.g., as might occur from ingestion). A semi-infinite exposure model is used for the external exposure. The equation used to derive the terrestrial animal BCGs for exposure to a single nuclide in contaminated water is: BCG(water)i,terrestrial animal ⫽

(15)

365.25DLta . Cta[(0.001LPta,water,i·DCFinternal,i) ⫹ (DCFext,water,i)]

Where BCG(water)i,terrestrial animal (Bq m⫺3) is the concentration of nuclide i in water which, based on the screening level assumptions, corresponds to a dose rate of DLta (0.001 Gy d⫺1) to the terrestrial animal; LPta,water,i (L kg⫺1) is the fresh mass terrestrial animal to water concentration factor of nuclide i; and all other factors have been defined. 3.10. Tabulated values of BCGs The results of applying the BCG equations are shown in Tables 4 and 5. The initial value of the ‘lumped parameter’ (i.e., the Biv) used in the general screen, is specifically chosen to produce conservative media limits. This quickly removes from further consideration contamination levels that would not cause biota to receive doses above acceptable limits. This initial screening is expected to be sufficient in the majority of cases involving radionuclide contamination. However, some sites may exceed this initial screen. This does not mean that they are causing biota to receive doses above the acceptable limit. It is recognized that actual Biv values for a single radionuclide may range over several orders of magnitude, depending upon biotic and abiotic features of the environment. The site-specific screening element of the graded approach allows a user to examine this lumped parameter and, using data either directly from the site, or from the technical literature, select a value that is more representative for the specific site conditions and receptors. In doing so, the screening calculation would be repeated and a new BCG provided. If the newly calculated BCG is still more restrictive than that measured in environmental medium at the site, a more rigorous analysis of potential exposure, considering other factors relating to site-specific conditions and receptors (e.g., spatial and temporal averaging of media radionuclide concentrations; organism occupancy time in the evaluation area), can be undertaken.

4. Electronic spreadsheets All of the equations presented in this text have been encoded into a series of electronic spreadsheets, termed the RAD-BCG Calculator. The spreadsheets were built using Microsoft’s Excel and incorporate Visual Basic commands which help automate the user’s progress through the assessment process. The RAD-BCG Calcu-

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

57

Table 4 Biota concentration guides for water and sediment for evaluation of an aquatic system

Nuclide 241

Am Ce 135 Cs 137 Cs 60 Co 154 Eu 155 Eu 3 H 129 I 131 I 239 Pu 226 Ra 228 Ra 125 Sb 90 Sr 99 Tc 232 Th 233 U 234 U 235 U 238 U 65 Zn 95 Zr 144

BCG for Water, Bq/m3

Organism responsible for limiting dose in water

BCG for sediment, Bq/kg

Organism responsible for limiting dose in sediment

2×104 6×104 2×104 2×103 1×105 8×105 1×107 1×1010 1×106 5×105 7×103 2×102 2×102 1×107 1×104 2×107 1×104 7×103 7×103 8×103 8×103 5×102 3×105

Aquatic Animal Aquatic Animal Riparian Animal Riparian Animal Aquatic Animal Aquatic Animal Aquatic Animal Riparian Animal Riparian Animal Riparian Animal Aquatic Animal Riparian Animal Riparian Animal Aquatic Animal Riparian Animal Riparian Animal Aquatic Animal Aquatic Animal Aquatic Animal Aquatic Animal Aquatic Animal Riparian Animal Aquatic Animal

2×105 1×105 2×106 1×105 5×104 1×105 1×106 1×107 1×106 2×105 2×105 4×103 3×103 3×105 2×104 2×106 5×104 2×105 2×105 1×105 9×104 5×104 9×104

Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian Riparian

Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal

lator contains approximately 14 worksheets. Users enter information on water, sediment, and soil concentrations by nuclide. The spreadsheet automatically determines compliance by calculating fractional ratios to the biota dose standards. The spreadsheets allow the user to work with alternative biota dose limits and effects data, and to modify a variety of input parameters, including the radiation-weighting factor. If the site exceeds the screen, one can move to the spreadsheet which provides additional information on which organism is driving the BCG. These spreadsheets are available from the authors. A DOE Technical Standard (DOE, 2002) also accompanies the spreadsheet to provide additional guidance on modifying parameters used in the calculation of media limits.

5. Conclusions A screening methodology was developed as a principal element of DOE’s threestep graded approach for evaluating radiation doses to aquatic and terrestrial biota. Limiting concentrations of radionuclides in water, soil, and sediment were derived for 23 radionuclides. Four organism types (aquatic animals, riparian animals, terrestrial

58

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

Table 5 Biota concentration guides for water and soil for evaluation of a terrestrial system

Nuclide 241

Am Ce 135 Cs 137 Cs 60 Co 154 Eu 155 Eu 3 H 129 I 131 I 239 Pu 226 Ra 228 Ra 125 Sb 90 Sr 99 Tc 232 Th 233 U 234 U 235 U 238 U 65 Zn 95 Zr 144

BCG for Water, Bq/m3

Organism responsible for limiting dose in water

BCG for sediment, Bq/kg

Organism responsible for limiting dose in sediment

7×106 1×108 3×108 2×107 4×107 8×107 1×109 9×109 2×108 7×107 7×106 3×105 3×105 3×108 2×106 6×108 2×106 1×107 1×107 2×107 2×107 6×106 8×107

Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial

1×105 5×104 1×104 8×102 3×104 5×104 6×105 6×106 2×105 3×104 2×105 2×103 2×103 1×105 8×102 2×105 6×104 2×105 2×105 1×105 6×104 2×104 4×104

Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial Terrestrial

Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal

Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal Animal

animals, and terrestrial plants) were selected as the basis for methods development. Internal doses for each organism type were calculated as the product of concentrations of radionuclides in environmental media, bioaccumulation factor(s) and dose conversion factors. External doses were calculated based on the assumption of immersion of the organism in soil, sediment, or water. The assumptions and default parameters used provide for conservative screening values. The methodology makes use of available data concerning environmental transfer factors for biota, is designed to allow use of environmental radionuclide concentration data typically collected as part of routine environmental surveillance programs at contaminated sites, and allows for the use of site and organism-specific input data. The screening methods are encoded into a set of user-friendly electronic spreadsheets, providing a workable, useroriented tool in the absence of more sophisticated models. The screening methodology within DOE’s graded approach should prove useful in demonstrating compliance with biota dose limits and for conducting screening assessments of radioecological impact. It provides a needed evaluation tool that can be employed within a framework for protection of the environment.

K.A. Higley et al. / J. Environ. Radioactivity 66 (2003) 41–59

59

References Amiro, B.D., 1997. Radiological dose conversion factors for generic non-human biota used for screening potential radiological impacts. Journal of Environmental Radioactivity 35 (1), 37–51. DOE (US Department of Energy), 2002. A Graded Approach to Evaluating Radiation Doses to Aquatic and Terrestrial Biota. US Department of Energy, Final Technical Standard No. DOE-STD-1153-2002. July 2002, Washington, D.C. Higley, K.A., Domotor, S.L., Antonio E.J. (2002a). A kinetic-allometric approach to predicting tissue concentrations for biota. Journal of Environmental Radioactivity (submitted for publication). Higley, K.A., Domotor, S.L., Antonio E.J. (2002b). A probabilistic approach to obtaining limiting estimates of radionuclide concentrations in biota. Journal of Environmental Radioactivity (submitted for publication). ICRP (International Commission on Radiological Protection), 1983. Radionuclide transformations— energy and intensity of emissions. ICRP Publication 38, Annals of the ICRP 11–13, Pergamon Press, Oxford, UK. Kennedy, W.E., Jr. D.L. Strenge, 1992. Residual radioactive contamination from decommissioning, Pacific Northwest Laboratory, NUREG/CR-5512, PNL-7994, Vol. 1, Richland, WA. Kocher, D.C. (1980). A radionuclide decay data base—index and summary table, Oak Ridge National Laboratory, NUREG/CR-1413, ORNL/NUREG-70, Oak Ridge, TN. Kocher, D.C. (1981). Radioactive decay data tables—a handbook of decay data for application to radiation dosimetry and radiological assessment. US Department of Energy, DOE/TIC-11026, Oak Ridge, TN. Kocher, D.C., Eckerman, K.F., 1981. Electron dose-rate conversion factors for external exposure of the skin. Health Physics 40, 467–475. Kocher, D.C., Trabalka, J.R., 2000. On the application of a radiation weighting factor for alpha particles in protection of non-human biota. Health Physics 79, 407–411. Onishi, Y., Serne, R. J., Arnold, E. M., Cowan, C. E., Thompson F. L., 1981. Critical review: radionuclide transport, sediment transport, and water quality mathematical modeling; and radionuclide adsorption/desorption mechanisms. Pacific Northwest Laboratory, NUREG/CR-1322, PNL-2901, Richland, WA. Shleien, B., Slayback Jr, L.A., Kirby, B.K. (Eds.), 1998. Handbook of health physics and radiological health, third ed. Williams & Wilkins, Baltimore, MD. Till, J.E., 1978. The effect of chronic exposure to 238Pu(IV) citrate on the embryonic development of carp and fathead minnow eggs. Health Physics 34, 333–343. USDOE (US Department of Energy), 1998. Establishment of biota dose assessment committee and invitation for membership. Washington, D.C.: DOE; memorandum from A. Wallo III; 1998 (August 3).