Editor's Note: The U.S. Department of Energy's many production and research sites contain radioactive and hazardous wastes. These waste sites pose potential risks to the health and safety of remediation and waste management workers and the public. The risks, however, vary from site to site. Some sites undoubtedly present larger risks than others and should be cleaned up first. However, before the cleanup begins, DOE is required by law to prepare an environmental impact statement on any actions that may significantly affect the environment--even actions that would clean it up.

The impact statement must also consider the health and safety risks to workers and the public from the cleanup and management of the resulting wastes. In other words, attempts to reduce potential or existing health risks by cleaning up a waste site could produce other risks, altering the benefit-to-risk ratio. For example, cleanup might result in the release of contaminants into the groundwater or air, potentially exposing the public to hazardous substances. Workers removing contaminated soil may be injured while operating heavy equipment. Other cleanup personnel might suffer from heat stress as a result of wearing protective gear. Still other workers risk becoming ill from exposure to radiation or hazardous chemicals.

Clearly, it is a difficult task to weigh the relative risks of leaving wastes at their sites or removing, treating, and disposing of them to restore DOE sites to their original environmental condition. ORNL has taken the lead role in assessing the current risks of DOE waste sites to the public, workers, and ecological systems (i.e., wildlife and vegetation) and in predicting the risks of environmental restoration and management of the wastes produced in the cleanup process. In the following article, Bonnie Blaylock describes ORNL's work in computer modeling and human health risk assessment. In an accompanying article, Larry Barnthouse describes ORNL's ecological risk assessment project in support of DOE's impact statement.

Since the spring of 1992, Oak Ridge National Laboratory's Center for Risk Management (CRM) has been a key participant in the Department of Energy's Programmatic Environmental Impact Statement (PEIS). The preparation of this statement was mandated by the Secretary of Energy in 1990 to develop and implement an integrated environmental restoration (ER) and waste management (WM) program for DOE sites. The PEIS is conducted under the guidelines of the National Environmental Policy Act, which requires that federal agencies prepare environmental impact statements on major federal actions that may significantly affect the environment. The laboratories involved in the PEIS are evaluating several options for environmental cleanup and waste consolidation to trace potential risks from the ER cradle (hazardous waste sites) to the WM grave (storage and disposal facilities).

ORNL was chosen to perform human health and ecological risk assessments for this multi-million-dollar project because of the center's risk assessment expertise. The staff of the CRM (directed by Curtis Travis of the Health Sciences Research Division) performed the human health risk assessment portion of the PEIS, and the staff of Larry Barnthouse in the Environmental Sciences Division performed the ecological risk assessment portion of the PEIS (see the following article). In addition to health and ecological risks, other impacts evaluated in the PEIS include socioeconomic impacts, air and water quality impacts, costs, and transportation risks. ORNL is responsible only for evaluating the human health and ecological impacts.

Environmental Restoration, or "The Cradle"

For effective project management, the human health risk assessment was divided into two parts--ER and WM. The ER risk assessment team, led by Jill Morris and Irene Datskou, assessed the risks for the baseline conditions of various categories of waste sites across the DOE complex: facilities, buried waste, contaminated soils, liquid containment structures, surface water, and groundwater.

To manage the huge number of sites and contaminants that exist across the DOE complex, the ER team used a fate and transport model to simulate and predict the transport of contaminants through the environment to different areas where people might be living under both present and future site conditions. The linearity of the computer model allowed the team to use a unit risk approach to estimate risks for numerous exposure scenarios for various times (e.g., living on a waste site 500 years from now) for each of the waste sites being evaluated.

A New Approach to Complex-Wide Risk Assessment

The unit risk approach is a simple concept that was applied on a large scale to yield complex results. ORNL worked with DOE's Pacific Northwest Laboratory, which provided contaminant concentration data for each waste site and many of the environmental parameters needed to run the computer models used for the risk assessment. The ER team assumed that one unit (1 gram or 1 curie) of each contaminant was present at the site. This one unit was used in the computer simulation of environmental transport for different exposure pathways (such as drinking groundwater and swallowing soil), exposure scenarios (such as on-site resident or off-site population), and time periods (such as present and future). Once the simulation was complete, the analysts could tell how much contaminant remained over time at different locations in the environment. In addition, the ER team could estimate how much of that contaminant might be ingested, inhaled, or absorbed by potential populations or individuals located within a given distance of each site. Once the exposure was estimated, then the potential health risk was estimated using standard U.S. Environmental Protection Agency (EPA) toxicity values and radiation risk assessment methods. The result was a "unit risk," or risk per unit of contaminant for each exposure scenario and pathway. From there, the actual risks posed by a site under given conditions were estimated by scaling up to the actual amount of contaminants at the site. If 50 grams of a contaminant was present, the ER team multiplied the final concentration by 50 and calculated the subsequent exposures and risks. The final risks were presented in terms of cancer fatalities, cancer incidences (the probability of developing fatal or nonfatal cancer), adverse genetic effects, dose, and the potential for health effects from noncarcinogens, known as the hazard index.

For each site, different assumptions about land use were used to estimate the future risks to various receptors (people who may be exposed to contaminants). Land use included restricted use, where the public was denied access to the site; unrestricted use, where a hypothetical homesteader could live on top of the site; and mixed land use, where a combination of both occurred; but the groundwater use remained restricted. The risks to the off-site public within a 50-mile radius of the installation and to a hypothetical homesteader living on the site boundary were examined.

In addition to estimating baseline conditions of each site for different land use scenarios, the ER team estimated risks during remediation for various remediation alternatives. Louis, Berger, and Associates, an engineering firm based in New Jersey, was asked to determine several remediation options for each site based on available technologies and costs. Once the firm selected the technologies, the CRM's ER team used standard emission rates recommended by the EPA for those technologies and performed more extensive computer modeling to estimate potential risks to the public during remediation. For example, radioactive substances could be carried into the air by excavation and removal of contaminated soil.

Using a similar unit risk approach to automate the process, a separate group within the ER team estimated risks to remediation workers performing the remediation activities. The group divided each technology into its composite worker activities and then estimated the number of person-hours the activities required for completion. They then estimated risks to workers from exposure to contaminants using various computer models for inhalation, ingestion, and direct radiation exposure. In addition, the group estimated safety risks such as fatalities and injuries from general construction activities, such as constructing a building or operating heavy equipment. The worker assessment accounted for protective clothing and equipment in the assessment of potential risks, which added another potential safety risk: heat stress fatalities and injuries, often caused when workers become overheated from wearing protective equipment.

(See "Worker Risk Assessment: Breaking Ground").

Waste Management, or "The Grave"

Like the runner in a relay race handing off the baton, the results from the ER assessment feed into the WM assessment. The ER assessment shows the scope of the problems within the environmental management (EM) program. The baseline risks at each site trigger various remediation alternatives, depending on the magnitude of the risk. Cleanup of the DOE waste sites generates volumes of ER waste that must be treated, stored, and/or disposed of within the WM program.

The WM risk assessment team, headed by Pat Pehlman, estimates the potential human health risks posed by various waste consolidation alternatives. In addition to the ER waste that is generated during cleanup, the WM program must address the volumes of various types of waste that are already being treated and stored at various installations throughout the complex. To determine the best ways to treat, store, and dispose of these wastes, the WM team evaluated risks from treatment, storage, and/or disposal of each type of waste at various locations across the complex. The team evaluated risks from high-level waste, low-level waste, low-level mixed waste, transuranic waste, and hazardous waste. Risks were presented in the same terms as for ER (i.e., cancer fatalities, cancer incidences, adverse genetic effects, and a hazard index).

The WM program focuses on three primary areas of waste handling: treatment, storage, and disposal (TSD). Treatment comprises processes such as incineration, solidification, or vitrification; and treatment processes differ depending on the waste type and treatability classification of the waste. Storage refers to either current or interim storage. Current storage is storage at the installation without waste transfer to another location. However, sometimes waste must be retrieved from current storage and treated or shipped to another location for storage or disposal. Interim storage is a temporary stage when waste is transferred from one location to another to await treatment or disposal. Disposal is a stage of permanent accumulation at DOE sites or at federally prepared facilities such as Yucca Mountain in Nevada or the Waste Isolation Pilot Plant in New Mexico.

The WM team works with DOE's Argonne National Laboratory, which provides contaminant emission rates from treatment processes and potential accident scenarios that were evaluated during the treatment or storage stages. The WM team has developed an automated unit risk approach that works much like the ER approach. For on-site and off-site populations, a unit inventory of contaminant (1 gram or 1 curie) is first assumed to be released to the air from a treatment or storage facility. For disposal, a unit inventory of contaminant is assumed to be disposed of, and the groundwater pathway is evaluated during the disposal stage. Fate and transport computer models are used to estimate the resulting exposures of individuals living near the facilities and of on-site plant employees. The models simulate environmental transport and account for different exposure pathways and routes. Actual risk calculations for various waste consolidation alternatives are then performed by scaling unit risks according to actual contaminant amounts and then adjusting the results by scenario-specific parameters (e.g., effective stack height, release periods). These calculations are performed in a data base that was constructed specifically for this application.

For the Programmatic Environmental Impact Statement, potential receptors for atmospheric releases are defined as (1) the public within a 50-mile radius of the installation and (2) on-site employees who are not directly involved with waste handling activities. Because the maximum exposure to contaminants released to groundwater can occur several years or lifetimes following the initial release, the receptors analyzed for the groundwater pathway (i.e., disposal) are the most-exposed generation, which is a hypothetical farm family of four.

In addition to the public and plant employees, a separate group within the WM team evaluated the risks to waste management workers who are directly involved with waste handling activities. In many cases, TSD facilities do not exist at an installation and, for the purposes of the PEIS, are assumed to be constructed. Health and safety risks to workers from construction-related hazards are also assessed. The WM worker risks are also estimated using a unit risk approach. Unit doses are estimated per contaminant and exposure route for each WM module or specific TSD task.

WM modules can be interchangeably arranged to form a "treatment train." A module is a self-contained waste handling or treatment process where workers are located to treat or handle waste as it is processed through a facility. A treatment train of several modules might include receiving and inspection, sorting, compaction, incineration, solidification, packaging, and shipping. Workers are located inside the building where the module's process occurs, and fugitive emissions from treatment processes may be released into the atmosphere of the building. A facility can contain several treatment modules simultaneously, and a module may contain several subactivities with different unit exposures. The unit exposures are summed to yield the total exposure for a given module, and the exposures for modules are summed to yield the total exposure for a given treatment train.

The WM computer models require many different types of information to estimate risks from TSD facilities. Such information includes facility dimensions, treatment and storage capacities, stored waste inventory, engineered safety controls, storage and disposal criteria, and TSD technologies. For disposal, in most cases, two engineered disposal options are considered: (1) installations in the eastern United States are assumed to use the tumulus option, and (2) installations in the western United States are assumed to use the shallow land burial option.

A tumulus is an aboveground, vault-type disposal facility made entirely of reinforced concrete. The waste is placed in a metal and concrete cask and then placed inside the vault. Once the vault is filled, it is closed and covered with a clay soil cap to reduce infiltration of contaminants into the groundwater.

Shallow land burial uses a long, narrow, unlined trench for waste disposal. Waste is stacked on the earthen floor, the voids between the waste containers are filled with earthen material, and the top of the disposal unit is covered with dirt. A third disposal option, a below-ground vault, is also evaluated only for the Savannah River Site in South Carolina.

The WM team estimated risks for normal operating and accident conditions assumed to occur during either the treatment or storage stage. They estimated risks for a variety of scenarios or consolidation alternatives. First, they estimated risks for a no-action scenario, which represents the TSD activities if they were to cease today. Then, they evaluated a current program alternative that represents the WM program in its current state, including planned transportation of waste and planned TSD activities. The remaining alternatives were evaluated to determine the impacts of different waste consolidation options. These alternatives are regionalization, centralization, and decentralization. For each of these alternatives, the WM team evaluated many different "cases" or permutations of waste consolidation. The difference between these alternatives is the number of sites that treat waste, store waste, or dispose of waste.

To determine the potential impacts of regionalizing the handling of hazardous waste, for example, the team might evaluate the impacts of six sites treating hazardous waste, four sites storing the waste, and three sites disposing of the waste. If the alternative mandates that fewer installations across the complex treat, store, or dispose of waste, those installations must take on more of the waste from other installations, increasing the risks at the chosen installations and increasing the transportation risks by requiring more waste shipments. However, if an alternative specifies that many different installations must treat, store, or dispose of waste (i.e., decentralization), then more installations must build facilities to accommodate these processes, and costs and risks will increase at these installations.

Evaluation of the different waste consolidation alternatives shows the trade-offs between cost and risk that must be weighed by decision makers. The integration of the ER and WM program within the PEIS will present decision makers with a large body of information on different options for cleanup and waste treatment, storage, and disposal. This information is intended to be useful for demonstrating the options available within ER and WM.

Further Results of the PEIS

For this project, the CRM developed risk assessment methodologies for estimating human health risks (for the public and for workers) associated with ER and WM activities. Both methodologies are broad in scope and are useful for DOE risk assessment applications beyond the PEIS. The worker risk methodology, developed in conjunction with Pat Scofield of the DOE Office of Environmental Compliance and Documentation, is a ground-breaking development because DOE has previously had no consolidated guidance for assessing occupational health risks at its waste sites. The risk assessment methodologies have been validated by pilot studies and have been successfully applied in installation-wide risk assessments of ER activities at two DOE installations.

In addition to these methodologies, our staff has automated the unit risk approach for both public and worker risk assessment for ER and WM activities. The automated methodology and data bases developed for the PEIS have applications beyond the PEIS because they provide an efficient, user-friendly method of estimating site-specific or installation-wide human health risks. In the ER worker risk portion of the package, for example, the user can choose the technologies used to clean up a particular site, the number of personnel involved, the contaminants present, and the level of protective gear the workers are wearing.

The PEIS is an ambitious DOE endeavor. When it is completed (the draft is scheduled for release by May 1995), it will yield complex-wide human health, transportation, and ecological risk estimates that can be used in conjunction with the other evaluated impacts to integrate the ER and WM programs in the most cost-effective manner. The project has provided an opportunity to develop important risk assessment tools for DOE and other agencies, and ORNL's Center for Risk Management expects to continue having a national impact in the risk assessment arena.

Biographical Sketch

Bonnie Blaylock joined the research staff of ORNL's Health Sciences Research Division in 1991. She holds an M.A. degree in English from the University of Tennessee at Knoxville. Blaylock was a key member of the team that produced the risk assessment methodologies used for the programmatic environmental impact statement described in the article.

Where to?

[ Next article | Search | Mail | Contents | Review Home Page | ORNL Home Page ]