Applications of Probabilistic Risk Assessment Techniques for Ecological Risk Assessment: A Predictive Methodology. Corey McDaniel, Probabilistic Risk and Hazards Analysis Group Technology and Safety Assessment Division, Los Alamos National Laboratory, Los Alamos, New Mexico
This paper presents a methodology for combining ecological risk assessment (ERA) with probabilistic risk assessment (PRA) techniques to predict the potential effect of an accident at a facility on an ecosystem. Present ecological risk assessment methodologies evaluate the damages to an ecosystem already affected by contamination. The current ERA process measures ecological risk in terms of the effects on assessment endpoints and evaluates the ecological significance of those effects. PRAs traditionally have been conducted on complex facilities, primarily nuclear reactors, that could pose significant human health risks and use fault and event trees to determine the probable occurrence of a certain hazard. Incorporating PRA quantification techniques into present ERA methodologies allows a proposed facility to be analyzed so that the risk it poses to the ecosystem may be evaluated before the facility is built. This combined methodology was implemented on a proposed hazardous/mixed waste storage and treatment facility (HMWSTF) at the Los Alamos National Laboratory; the results quantify the probabilities of several worst-case accident scenarios and present the resulting ecological effects of such accidents. The results of this approach have been used to evaluate the overall ecological risk associated with this proposed facility. This approach to ERA has shown that a probabilistic ecological risk assessment (PERA) can identify and predict credible potential ecological hazards before a facility is built, allowing recommendations and design changes to be made before site construction begins. The HMWSTF PERA verified the feasibility of using a PERA approach to predict the frequency and ecological consequences of several worst-case scenarios associated with a potentially hazardous facility.
The PERA methodology developed for the HMWSTF followed two initial risk assessments that estimated the potential human health risk from the facility. The first assessment, a preliminary hazard analysis (Pl4A), was conducted to identify all of the potential hazards associated with the facility based on information given in the conceptual design plans. Next, a limited-scope PRA was performed to quantify the risk to workers and offsite individuals from the potential worst-case release scenarios identified by the PHA. The combined results of the PHA and PRA concluded that the facility would pose no significant health risk to the public. These reports were included in the site Environmental Assessment (EA). An ERA was chosen as the next logical step in the evaluation process for the HMWSTF because the first two assessments did not address the potential effects of the hazards on the ecosystem.
The first step in developing a methodology specific to this PERA was to select the scenarios that would result in the most serious ecological hazards. The initiating events selected were seismic events, internal fire, and external fire. The waste inventory at the Laboratory's existing storage facility was examined to determine the types and quantities of hazardous materials available for release from the proposed facility. The four contaminants selected, which we determined posed the most severe hazards to the ecosystem, were PCBs, lead, mercury, and 238U. When the potential contaminants were selected, the containers and relative storage locations were analyzed to develop scenarios for releasing the contaminants beyond the site boundary. Several scenarios ranging from waste drum ruptures to exploding waste containers were analyzed. Each scenario was developed to produce the worst-case release of the contaminant.
The PRA portion of the analysis quantified the probability for each release scenario and then predicted the amount of contaminant that would be released beyond the site boundary. Events leading to a release were developed using fault- and event-tree analysis. The fault trees and event trees then were entered into the CAFTA PRA code (developed by SAIC Inc.), which calculated the frequency of release for each scenario. Actual quantities of the contaminant that would be released were calculated. The results of the PRA provided offsite contaminant concentrations for each release scenario. After the offsite concentrations were determined, the ecological segment of the risk assessment began.
The next step in the PERA was to determine the contaminant's acute and chronic ecological effects on the ecosystem. ECOTRAN, a code developed by Los Alamos, was used to model the transport of the contaminants through the ecosystem and to the plants and animals. The model used site-specific data to determine the uptake and concentrations in predetermined endpoint species. The model was able to provide tissue, organ, and total body uptake for individual species and populations, as well as ecological system uptake. The depositions were used to evaluate contaminant effects on the species using empirical data.
The ERA methodology was developed using primarily EPA guidance documents and text published by experts in the ERA field. Regulatory drivers such as the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) and several private organizations that have performed traditional ERAs also were consulted. The EPA Risk Management Rule under the Clean Air Act (CAA) requires the assessment of the effect of a worst-case scenario on the environment. The effects of each worst-case scenario were measured using endpoints selected by an ecological endpoint analyst. The ecological analysis of the effects of the predicted contamination provided detailed acute and chronic results of population changes in the ecosystem.
The PERA methodology predicted the frequency and severity of ecological damage presented by the HMWSTF during a worst-case accident scenario. This risk assessment was conducted to show that ERAs can be conducted in a manner similar to PRAS, which have traditionally looked only at human health risks. Human health risk assessment methodologies have been developed and regulated for many years; however, the ERA methodology is continuously evolving and has no set regulatory requirements at this time. As ERA methodologies continue to improve it will be useful to incorporate human health risk assessment techniques into ERAS. This PERA has shown the usefulness of coupling ERA and PRA to predict ecological hazard effects and has shown the importance of assessing potentially hazardous sites before they actually contaminate an ecosystem. Los Alamos PERAs have already been proposed for future sites as well as existing ones that have not yet released any contamination. By performing PERAS, Los Alamos can assess the potential risks associated with any number of sites and then prioritize them according to the ecological risk for each site. Risk-based prioritization has used the results of human health risk assessments for many years. In the future, ERAs will likely play an equally important role in risk-based decision-making activities.