4
Environmental Pathways

HUMAN EXPOSURE TO radionuclides occurs as a result of transport along various environmental pathways. It is the overlap in space and time of the "region of influence" of decaying radionuclides with the presence of a person that results in radiation exposure of tissues and organs. The objective of an environmental pathway analysis is to estimate human exposure rates or to determine radionuclide concentrations in air, water, and foodstuffs. This information is used in the assessment of radiation doses to representative or specific individuals.

The environmental pathway analysis, which in most cases uses the source term assessment as an input, may involve an iterative procedure for making estimates of exposure rates and of environmental concentrations of radionuclides. In the scoping study, which is carried out to determine if there is a need for a full-scale dose reconstruction study, only the most important environmental pathways and radionuclides are considered, and readily available information on the characteristics of the site and on the population distribution is used to estimate the exposure rates and environmental concentrations at a limited number of locations of interest. In a full-scale dose reconstruction study, the number of estimates of exposure rates and environmental concentrations is gradually expanded, and their quality is improved as: (1) the full source term is taken into consideration, (2) the results of detailed studies of the characteristics of the site with respect to the environmental behavior of the released radionuclides are used, (3) all environmental pathways that



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 27
4 Environmental Pathways HUMAN EXPOSURE TO radionuclides occurs as a result of transport along various environmental pathways. It is the overlap in space and time of the "region of influence" of decaying radionuclides with the presence of a person that results in radiation exposure of tissues and organs. The objective of an environmental pathway analysis is to estimate human exposure rates or to determine radionuclide concentrations in air, water, and foodstuffs. This information is used in the assessment of radiation doses to representative or specific individuals. The environmental pathway analysis, which in most cases uses the source term assessment as an input, may involve an iterative procedure for making estimates of exposure rates and of environmental concentrations of radionuclides. In the scoping study, which is carried out to determine if there is a need for a full-scale dose reconstruction study, only the most important environmental pathways and radionuclides are considered, and readily available information on the characteristics of the site and on the population distribution is used to estimate the exposure rates and environmental concentrations at a limited number of locations of interest. In a full-scale dose reconstruction study, the number of estimates of exposure rates and environmental concentrations is gradually expanded, and their quality is improved as: (1) the full source term is taken into consideration, (2) the results of detailed studies of the characteristics of the site with respect to the environmental behavior of the released radionuclides are used, (3) all environmental pathways that

OCR for page 27
could lead to radiation exposures are considered, (4) all available historical measurements of environmental radiation are used to bypass environmental transfer models or to validate those models, and (5) the uncertainties attached to the results of the environmental pathway analysis are estimated. It is important to have epidemiologists involved in the full-scale dose reconstruction study in order to ensure that the information developed is appropriate for epidemiologic decisions and planning. The purposes of this chapter are to describe the environmental pathways that need to be considered in dose reconstruction studies and to list criteria for their assessment. TRANSPORT OF RADIONUCLIDES AND OTHER CONTAMINANTS Humans come into contact with radioactive and other materials by means of a variety of pathways, and the movement of the materials can be affected by many physical, chemical, and biologic processes. An important feature of any dose reconstruction must be a critical analysis of pathways so that those of most relevance can be identified. The environmental releases most frequently encountered in dose reconstruction studies are direct releases to the atmosphere or to the hydrosphere; they are considered in turn. Direct Releases to the Atmosphere Analysis of the environmental pathways from atmospheric releases to humans requires study in four areas. First, the meteorologic processes that govern atmospheric dispersion and the precipitation processes that deposit gaseous and particulate emissions must be considered. Second, the pollutant concentrations in ground-level air must be quantified. Radiation exposure by inhalation and exposure to external irradiation from the radioactive materials present in the cloud are derived at this stage. Next, the amounts and the physical and chemical forms of the materials deposited on the ground need to be determined. This information can be used to estimate exposures to external irradiation and, occasionally, through ingestion of soil. Finally, the behavior of radionuclides after they are deposited on the ground must be ascertained. They can be resuspended in the air; dissolved in groundwater; or enter the food chain through agricultural processes, through irrigation systems into crops or into the drinking-water supply. Radioactive materials from sources such as nuclear power plants are typically released between the ground surface and an elevation of 100 m into a region of the atmosphere called the planetary boundary layer. The

OCR for page 27
elements of the airborne plume are affected by turbulent eddies in the layer that diffuses the effluent material as the plume is transported downwind. Generally, the combined influences of diffusion and transport are called dispersion (Brenk and others 1983). Extensive discussions of atmospheric transport processes can be found in Meteorology and Atomic Energy (Slade 1968) and the Handbook on Atmospheric Diffusion (Hanna and others 1982). If the emission rate is known and approximately constant or randomly distributed in time (or if it can be inferred) and if the average or specific meteorologic conditions are known, then the movement of the materials can be described by a variety of models of atmospheric transport (Brenk and others 1983). The object of the atmospheric transport models is a description of the ground-level concentration of materials as a function of time. Unfortunately, atmospheric modeling is subject to a great deal of uncertainty, and it is preferable, therefore, to have actual measurements of the airborne concentrations wherever possible. The presence of radionuclides in air may lead to two modes of radiation exposure—inhalation and external irradiation. If the individual considered is outdoors, the measured or calculated radionuclide concentration in outdoor air at that location is used without modification in the estimation of the radiation dose, using an appropriate model for the inhalation dose. However, people spend most of their time indoors, and indoor concentrations of most air contaminants are as a rule smaller than are the outdoor concentrations because of the filtration provided by buildings. It is frequently assumed, on the basis of limited experience, that indoor air concentrations of radionuclides in particulate form are about one-third those found outdoors if building leakage is the only migration path. If the individual is outdoors, the measured or calculated radionuclide concentration in outdoor air at that location can be used without modification in the derivation of the external radiation dose. If the individual is indoors, the structure of the building attenuates the radiation emitted by radionuclides in the outdoor environment. The magnitude of the shielding factor is highly variable (typically between 0.001 and 0.5) because it depends, among other factors, on the type of building and on the floor level considered. The exposure geometry varies according to whether the irradiation is due to a contaminated overhead plume or whether the individual is immersed in a near-ground contaminated atmosphere. Also, people do not spend all of their time at the same location. Information on the whereabouts of people is particularly important in the aftermath of an accident. If countermeasures, such as relocation, were taken, knowledge of the radiation field distribution along the routes of

OCR for page 27
relocation and at the relocation site is needed. These are important factors in the determination of the total exposure of the relocated populations. Particles and volatile matter released as airborne effluents can be deposited on vegetation and the ground surface. Wet deposition processes include rainout and washout. Rainout occurs when the pollutant becomes involved in precipitation formation processes within a cloud and is subsequently removed from the atmosphere with the precipitation. Washout is the removal of the effluent from the plume by entrainment in falling precipitation. In the absence of precipitation, effluent material also can be removed from the atmosphere through gravitational settling onto the ground, vegetation, or other ground cover, such as buildings. The transfer of airborne contaminants from ground-level air to the ground surface, including vegetation, is usually modeled through the use of the deposition velocity concept, which is the quotient of the deposition of radioactivity (in becquerel, Bq) per unit area (in square meters, m2) to the ground surface and of the time-integrated concentration in ground-level air (Bq/m3). Choosing an appropriate value for the deposition velocity is difficult and can be a major source of uncertainty, possibly of four or five orders of magnitude. However, use of consistent methods of soil sampling and analysis must be documented for such data to be considered reliable. Deposition velocity varies as a function of particle size, wind speed, surface conditions, and most significantly as a function of the physical-chemical form of the contaminant (Sehmel 1980). The presence or absence of rainfall also can be crucial, and separate methods of predicting the washout or rainout of contaminants as a function of physical-chemical form must be used (Engelmann 1970). Wet deposition also can occur as a result of "washout" of these components by rain and other forms of precipitation. Dry and wet deposition are "integrating pathways"; the areal amount of deposited material is proportional to the time integral of the airborne concentration. Such processes can lead to localized areas of higher ground deposition or "hot spots" resulting in higher dose rates and doses. It is difficult to determine where and when hot spots occur, particularly in retrospect, because it requires knowledge of releases and concomitant atmospheric conditions (stability, wind speed, and the geographic patterns of precipitation such as rain, sleet, hail, fog, or snow). Both wet and dry deposition depend on the physical-chemical form of the radionuclide. The aerodynamic diameter of particles significantly affects deposition rates; higher deposition rates occur for large particles «0.2 µm (gravitational settlement) and for very small particles »0.2 µm (Brownian motion). Particle size can change with time (and distance) because of sorption on atmospheric particles, preferential depletion from deposition or washout, or agglomeration (Megaw 1965). Reactive gases

OCR for page 27
such as elemental iodine (I2), can be preferentially deposited on vegetation and ingested by grazing animals. Particulate iodine deposits and organic iodides, such as methyl iodide (CH3I), have even lower deposition rates. Soluble gases and larger airborne particles are more subject to washout by precipitation than are inert gases or small particles. If a single contaminating event has taken place and if measurements have been made (such as external gamma exposure rate or deposition of one or more radionuclides or stable materials), it is often possible to begin the dose reconstruction with such data and to bypass the need to model the transport of radionuclides up through this stage. Even if the contamination has been chronic, it is often useful to sample the soil to measure the longer lived components of the contaminant materials and to infer the deposition of the shorter lived components. The alternative is to depend on atmospheric transport and deposition models, which are much less reliable than are direct measurements. The use of soil-monitoring data is particularly valuable when the longer lived contaminant has the same chemical and physical properties as do the shorter lived components. The presence of radionuclides on the ground can lead to human exposure by external irradiation. Here again, if the individual considered is indoors, the structure of the building acts as a shield against the radiation emitted by radionuclides present in the outdoor environment though there can be cumulative adsorption or deposition on the roof. Radionuclides deposited on the ground can become resuspended in the atmosphere through aeolian processes, that is, through the action of the wind. They also can be transported on the ground surface by runoff, or they can migrate vertically into deeper layers of soil. Resuspension of deposited radionuclides by wind or mechanical disturbance (for example, by transport on moving vehicles or people) can lead to inhalation intake or increased deposition at downwind locations. The amount of resuspended material depends on the characteristics both of the radionuclide particles and of the ground surface (surface roughness and vegetative cover) as well as on atmospheric factors such as wind speed. Resuspension is generally a minor exposure pathway if releases from the primary source are continuous, but it can become relatively significant after the primary releases cease. Much of the research on resuspension mechanisms has been done in arid environments, but after the Chernobyl reactor accident in 1986 the extent of such processes also has been studied carefully in more humid ecosystems. Resuspension could be a major factor in the transport of contaminants from piles of mill tailings and from desiccated storage lagoons. Most of the plutonium transported off-site from the Rocky Flats Plant in Colorado was in windborne soil that had been contaminated by leaking oil drums. This pathway was unsuspected by the operators of the plant

OCR for page 27
(Hammond 1971), and demonstrates the need for vigilance in considering possible modes of transport. Although there are models that might help researchers to predict the effects of windborne transport (Langer 1989), the process is so complicated that it is preferable to use measurements to infer the source term or to describe the starting point for use in the dose reconstruction (Hardy and Krey 1971). On-site contamination of soil and other surfaces has been found at many locations. During periods of normal rainfall, some fraction of the contaminants can be carried off-site by surface runoff. For example, this is a factor that may have influenced the estimates of uranium depositions between two studies at the Feed Materials Production Center in Fernald, Ohio (Voillequé and others 1991, Stevenson and Hardy 1993). In other cases, such as at Mound Laboratory, an episodic rain event was responsible for significant transport of radionuclides across the site boundary (Rogers 1975). Also, at Goiania, Brazil, runoff led to significant contamination of water bodies and sediments (Godoy and others 1991). Especially for non-radioactive contaminants, such as trichloroethylene, and for tritiated water, transport by groundwater is often a major consideration. In many cases, the movement of nitrates and phosphates is significant. Materials that have been deposited or spilled on the surface soil are transported through the vadose zone (the unsaturated soil layer—the region above the permanent groundwater aquifer) and eventually through groundwater. This is a pervasive problem at Department of Energy sites. The possible interception of contaminated plumes by wells that supply drinking water is one of the more challenging pathways to be modeled. Because many of the factors that control transport are poorly known, it is preferable to depend on actual concentration measurements of contaminants instead of on modeling. Radionuclides can enter the terrestrial food chain by direct deposition onto forage and food crops or by plant uptake from soil or from irrigation water. Contaminated plants can then be consumed directly by humans or ingested by animals with transfer of the radioactivity to human food products (meat, milk, or eggs). Animals can also inhale airborne contamination. Fortunately, plutonium and other transuranic actinides are relatively insoluble in water and their uptake by plants and animals via water pathways is very low. Estimating deposition onto plants is similar to estimating total deposition, except that a retention factor is included to estimate the fraction of the total deposition that adheres to vegetation. Investigation of plant uptake of radionuclides from soil requires considering the food crop, the edible portion (underground roots and tubers tend to show less radionuclide uptake than fruits, for example), soil char-

OCR for page 27
acteristics, and farming practices (liberal use of fertilizers can reduce radionuclide uptake by crops). Human exposure occurs through the consumption of contaminated foodstuffs from the natural environment (such as berries, mushrooms, or wild game) or in agricultural foodstuffs (such as green vegetables, milk, or beef). Radionuclide concentrations in the consumed foodstuffs can differ substantially from those in the raw agricultural products at the point of production. For example, leafy vegetables contaminated by direct deposition of radionuclides exhibit greater concentrations on the outer leaves, and removing those leaves as well as washing the vegetables will eliminate a substantial fraction of the contamination. Boiling or frying food also can reduce radionuclide concentrations. People do not necessarily consume local agricultural products. Even common foodstuffs like milk or vegetables often are transported large distances from the site of production to the point of consumption. Although the movement of foodstuffs must be accounted for, it is important primarily if it results in a reduction in the consumption of contaminated food by the local population, particularly in the case of milk. There are also seasonal variations in uptake via the food chain. Grazing practices are site-specific and can include the seasonal movement of grazing animals from one pasture to another. Finally, countermeasures after accidents can lead to drastic decreases when highly contaminated foodstuffs are removed from commerce and replaced with clean products. Even if contaminated foodstuffs are not removed, people often voluntarily avoid these foodstuffs and change their dietary habits to reduce their intake. Direct Releases to the Hydrosphere Waterborne process effluents are often released to water in rivers, lakes, or seas. Another source of emission to waterborne pathways is through sanitary sewers. Although this is not likely to be a source of exposure to the public, the potential should not be overlooked, in particular because sewage sludge is sometimes converted to agricultural fertilizer. Sanitary waste is usually released to streams after biologic digestion and chlorination, leaving the dissolved radionuclides in solution. The steps involved in the analysis of the environmental pathways from releases into the hydrosphere to humans are conceptually similar to those related to atmospheric releases. They entail the determination of the pollutant concentrations in locations where water is drawn for drinking or irrigation, the estimation of the amounts deposited on sediments, which can lead to exposure by external irradiation, and analysis of the subse-

OCR for page 27
quent behavior of radionuclides transported by water or deposited on sediments. In general, discharges to fresh water pose a potentially greater threat to public health than discharges to the marine environment because there is less potential for dilution in freshwater; freshwater is used for drinking and irrigation; and for most radionuclides, bioconcentration factors are higher in fresh water than they are in marine waters. Aquatic dispersion depends a great deal on the nature of the body of water (pond, lake, river, or saltwater body) that receives the effluent. In particular, radionuclides will accumulate in closed water bodies, such as small ponds and lakes. The nature of the receiving water body is also significant in determining the amount of contact aquatic organisms have with the radionuclides. The concentration of radionuclides in freshwater systems can be modeled as a function of time (Jirka and others 1983) in a fashion similar to that for atmospheric releases. The major pitfall in models of freshwater systems is uncertainty in the amount of dilution between the release point and the downstream location of water withdrawal. The possible occurrence of stratification, or the lack of mixing of sections of water, can cause uncertainty as well. These uncertainties are particularly pronounced for episodic releases, because it is not always possible to determine whether the plume bypassed or was captured at the water withdrawal location. Again, the use of actual measurements of concentration is preferable to the prediction of radionuclide concentrations by aquatic dispersion models. An important factor to consider in the analysis of the environmental behavior of radionuclides is the partitioning between water and sediments in a stream bed. The behavior of radionuclides in freshwater, estuarine, and marine ecosystems depends on their physical-chemical form. Larger particles of insoluble materials can settle out near the discharge point. Radionuclides that were dissolved initially can precipitate as a result of chemical reactions in the receiving water body or they can be adsorbed on sediments or suspended particles (e.g., cesium on clay particles). The chemical form of the radionuclide in an aquatic ecosystem can change with time and distance from the release point. Radionuclides can become less soluble because of chemical changes and precipitate onto bottom sediments. Physical changes in the receiving waterway may also affect radionuclide sedimentation. Widening of a river can result in lower flow velocities and increased sedimentation, as will the presence of impoundments such as dams. Changes in the chemical composition of the waterway, such as occur in an estuary, can affect radionuclide sedimentation. In particular, increased radionuclide deposition has been observed

OCR for page 27
in the vicinity of the freshwater-saltwater interface as a result of a change of ionic potential, and adsorption on benthic organisms can be significant in brackish or estuarine waters. Better characterization of the environment and of the physical-chemical form of the effluents will allow more accurate estimates of the radionuclide concentrations in aquatic foodstuffs and reduce uncertainty. Radionuclides in water and sediments can be absorbed by aquatic organisms that are part of the human food chain, although the extent of absorption varies with the form of the radionuclide because changes in form affect the transport and uptake of the radionuclide. For example, radionuclides in particulate form are especially subject to concentration by shellfish or other filter-feeding organisms. One might also have to consider cross-contamination of terrestrial farm animals through the use of contaminated fish bone as feed. The presence of chemically similar stable elements can greatly alter uptake of radionuclides by aquatic organisms and, consequently, their concentrations in the food chain. For example, the degree of uptake of 137Cs by freshwater fish is inversely proportional to the potassium concentration in water, and the uptake of 90Sr is inversely proportional to the concentration of stable calcium in water. Consequently, the bioaccumulation of cesium and strontium isotopes is much lower in marine systems than it is in freshwater because of the larger amounts of potassium and calcium in seawater. Estimation of radionuclide intake from aquatic food chains requires an estimate of the radionuclide losses during food preparation. It is important that the characteristics of specific foods be considered so that all appropriate pathways are evaluated. (Small fish, such as smelts, are often consumed with their bones; larger fish are usually boned. The intake of radionuclides such as 90Sr, which tends to concentrate in bones, is apt to be higher for consumers of small fish.) The potential pitfalls in reconstructing doses associated with the discharge of radionuclides to the aquatic environment depend on whether the discharges were to the freshwater or marine environment and whether they were chronic or episodic. The factors that introduce large uncertainties into an aquatic dose reconstruction include retention or delay of groundwater migration in unsaturated subsurface flow, the removal of radionuclides by sedimentation processes and their resuspension during floods, and the bioavailability of radionuclides in solution and in the bottom sediment. In all cases, the chemical form of the radionuclide and the chemistry of the receiving water and sediment greatly influence bioavailability. Also, the influence of high water levels after storms in remobilizing sediments and the possibility of contamination of agricultural lands by flooding might need consideration.

OCR for page 27
FURTHER CONSIDERATIONS Even if there is an abundant base of monitoring data for environmental media, the estimates of exposure through some or possibly all of the environmental pathways usually require the use of mathematical models, i.e., quantitative approximations of the processes that affect the transfer of radioactive substances in the environment to the point at which there is human contact. The models are combinations of equations and parameters usually formulated in computer codes. The parameters require fairly precise numerical estimates if the model is to be more than heuristic. Appropriate Use of Mathematical Models When actual measurement data are either absent or incomplete, mathematical models are needed to estimate concentrations of radionuclides in air, water, soil, food, fodder, and human organs. Models are used to extrapolate information from situations where measurements have been made to situations where measurements are lacking. The model's complexity varies depending on the nature of the assessment question and the degree of resolution required to answer the question. Because mathematical models are merely representations of reality, any simplifications and assumptions inherent in their construction, implementation, and execution must be examined carefully through uncertainty and sensitivity analysis of the results. It is essential that the computer codes be properly verified to ensure the absence of coding errors. Every attempt should be made to validate the predictions of these codes against relevant but independent data sets. Caution must be exercised in the uncritical use of ''off-the-shelf" assessment codes that have been developed for the purpose of regulatory analysis. Their equations and data bases support generic assessments for reference situations, but because they usually are designed to determine compliance with regulations, they are seldom applicable to realistic estimates of exposure. The user is forced to use the code as a black box and typically can change only the assumptions about parameter values. Because the user is denied access to the source code, structural modifications cannot be made to adapt the model structure to the unique situation presented at a given site (Hoffman 1993). Finally, although computer codes can be verified, peer reviewed, and sanctioned by specific government agencies, their results still rely on the professional judgment of the user, and different users might get different results.

OCR for page 27
Uncertainty Analysis Pathway modeling of exposure should include quantitative estimates of uncertainty to measure the degree of confidence that can be placed in the exposure estimate. The dominant contributors to uncertainty should be identified, and this information should be used to acquire additional data to reduce uncertainty. Quantitative uncertainty analysis, however, requires a rigorous definition of the target end point of the assessment because different results will be obtained depending on whether the objective is to estimate exposures for representative individuals or for actual persons. If the latter, further information about age, gender, diet, lifestyle, and residential history is needed in the modeling process. Probability density functions (pdf's) are typically needed to express uncertainty in model parameters for which the true value is unknown. In the absence of site-specific data, the derivation of parameter pdf's requires quantification of the subjective degrees of belief of the investigators or outside experts. There may be divergence among experts as to the best parameter pdf. The resolution of this issue will require the acquisition of new experimental data or the use of formal approaches for eliciting subjective information from groups of experts. There are other issues of concern: It is necessary to distinguish between uncertainty that arises from unexplained variability in the observed data and uncertainty that results from a lack of knowledge about a true but unknown value. It is important to evaluate known dependencies among model parameters and their effect on the uncertainties in assessing dose and on the epidemiologic association between dose and health outcomes. It is also important to include the model's uncertainty itself along with uncertainty in model parameters. This might require resolution through sensitivity analysis. Once uncertainty has been propagated through the exposure pathway equations, different methods can be used to express confidence in the model result. Some investigators have used a geometric mean and geometric standard deviation; others have used an 80% to 95% probability interval based on expert judgment; still others use the entire joint pdf of the model result. The one most appropriate for dose reconstruction depends on the type of decision that is to be made with the exposure information. While there is no standard for all situations, it is recommended that uncertainties be quantified as a confidence interval (IAEA 1989). If the uncertainty is due to natural variability, a pdf may be appropriate but confidence intervals about the pdf should be provided. Confidence statements should include all sources of uncertainty.

OCR for page 27
Several areas of uncertainty have been identified in past and current pathway analyses: It is difficult to estimate the precise wind trajectory fields for sites of episodic releases or those for which meteorologic data are inadequate. Determining the physical-chemical forms of material released and transported in air, can be problematic but is important because these forms can affect the rates of removal from the plume and the retention of materials in the human lung. It is difficult to estimate the amount of wet versus dry deposition and whether the deposited material is intercepted and retained on natural or synthetic surfaces, such as trees or buildings. Site-specific values need to be determined for the food chain transfer of radionuclides. Element-specific transfer coefficients and rate constants must be quantified, and estimates of the transport of contaminated foodstuffs outside or into the region of concern must be made. Accounting for unusual pathways of exposure, such as the contamination of cisterns through wet deposition, and the uptake of possibly high levels of contamination by wild game and waterfowl can be difficult. Correctly quantifying the attributes of the human receptor is another area of concern. The locations of residence and work, recreational activities, dietary habits, characteristics of metabolism, state of health, and the effectiveness of shelter all must be considered. Many additional attributes must be properly quantified to translate the estimate of exposure into an estimate of dose and an estimate of health risk. These issues are more significant in the process of obtaining exposure estimates for actual persons than for representative (hypothetical) individuals. SUMMARY AND RECOMMENDATIONS One important feature of any dose reconstruction must be a critical analysis of all possible environmental pathways to identify those of most relevance to the population and to those special groups that might have been the most exposed. The environmental releases most frequently encountered are those directly to the atmosphere or to the hydrosphere. Conceptually, the steps involved in the analysis of the environmental pathways from releases into the atmosphere or the hydrosphere to humans are similar. They entail the determination of the pollutant concentrations in locations where the radionuclides could be inhaled or ingested; the estimation of the amounts deposited on the ground or in sediments, which can lead to exposures by means of external irradiation; and the analysis of the subsequent transport of the radionuclides through physi-

OCR for page 27
cal or biologic processes that will bring the contaminants into contact with humans. The committee makes four recommendations: Insofar as possible, measurements of environmental radiation or of radionuclides should be used in the environmental pathway analysis. For example, if a single contaminating event has taken place and if measurements have been made (such as external gamma exposure rate or deposition of one or more radionuclides or stable materials), it is often possible to begin the dose reconstruction without the need to model the transport of radionuclides up through this stage. Even if the contamination is chronic, it is often preferable to take suitable soil samples to measure the longer lived components of the contaminant materials (such as 137Cs or 129I) and to infer the deposition of the shorter lived components (such as 131I), rather than to depend on atmospheric transport and deposition models, which are much less reliable than are direct measurements. Even if there is an abundant base of monitoring data, mathematical models are usually needed to extrapolate information from situations where measurements have been made to situations where measurements are lacking. Every attempt should be made to validate the predictions of the models against relevant data sets. Caution should be exercised in the use of "off-the-shelf" computer codes that may have been developed for other purposes such as regulatory analyses. Environmental pathway analyses should include quantitative estimates of uncertainty to indicate the degree of confidence that can be placed in exposure estimates. For accidents, there should be careful scrutiny of any countermeasures, such as removal of contaminated foodstuffs from commerce. Even if contaminated foodstuffs are not removed, people often voluntarily avoid contaminated foodstuffs and change their dietary habits. For routine releases, attention should be paid in the assessment of ingested radionuclides to the movement of foodstuffs into the region of interest because people do not necessarily consume local agricultural products.