Best Practices for Risk-Informed Decision Making Regarding Contaminated Sites: Summary of a Workshop Series (2014)

3


Holistic Approaches to Remediation

3.1 OVERCOMING BARRIERS TO RISK-INFORMED REMEDIATION

Michael Truex, a senior program manager in the Environmental Systems Group at Pacific Northwest National Laboratory (PNNL), presented on approaches to overcoming remediation barriers at the Hanford, Washington, site. Many processes were used for weapons production at Hanford, including manufacturing of fuel elements, chemical separation, and plutonium finishing. Each of these processes generated different waste, and the entirety of the Hanford site covers 586 square miles. Thus, within the whole of the Hanford site are smaller sites where remediation efforts are more focused.

Recent work at Hanford resulted in a framework (Figure 3-1) that considers the unique aspects of multiple Hanford sites, which will require a longer timeframe to reach remediation endpoints. The framework

• uses conceptual models, which are a foundation for technical efforts and communication;
• uses the subsurface system and site context to inform remedy approach and timeframe;
• maintains protection while addressing future risk and cleanup;
• adapts as a plume evolves and responds to actions over time, enabling adaptation and transitions along a longer timeframe.

The conceptual models are dynamic; as information is collected during site characterization, it is incorporated back into the model. For example,

FIGURE 3-1 A framework for holistic remediation approaches.
SOURCE: Truex 2013.

data on subsurface systems relate to the fate and transport of contaminants, site context, exposure pathways, and size of the plume. All of these aspects are important in determining the remedy and timeframe. There often are not enough data for plume behavior at complex sites, so decisions may need to be adapted as more data are gathered.

Mr. Truex highlighted four examples at Hanford where application of these framework principles has been successful: soil vapor extraction, the 100-N Area, the 100-F Area, and the Central Plateau. The 100-N Area, which is located along the Columbia River north of the Central Plateau, includes a strontium groundwater plume adjacent to the river. Similarly, the 100-F Area is a reactor site that is also adjacent to the Columbia River and has multiple contaminants. The Central Plateau contained storage tanks where processing that created a large volume of waste took place (Figure 3-2).

3.2 SOIL VAPOR EXTRACTION

Significant amounts of carbon tetrachloride were disposed of at the Hanford site surface, which resulted in contamination of the vadose zone and groundwater. Because carbon tetrachloride is a volatile contaminant, soil vapor extraction was applied. This remedy is very effective at removing contaminant mass for most portions of the vadose zone, but in lower permeability zones, the residual contaminant diminishes more slowly. Over time, it became apparent to PNNL scientists and contractors that, although the contaminant mass was decreasing in the more permeable subsurface, the

FIGURE 3-2 Schematic of the Hanford Site.
SOURCE: Truex 2013.

returns were diminishing with the soil vapor extraction approach. Questions were raised about the length of time the extraction should continue and about the levels to which the mass should be diminished to be considered protective of the groundwater. The conceptual model, which defines the characteristics of the system and the source mass discharge, was used in conjunction with transport calculations to determine the mass discharge that is protective of the groundwater.

3.3 100-N AND 100-F AREAS

The 100-N Area is a reactor site along the Columbia River where trenches were used to dispose of water laden with strontium-90, explained Mr. Truex. These disposal activities resulted in a plume of strontium-90 migrating away from the trenches and through the subsurface. During operation, water was continually added, and the hydraulic driving force pushed the plume farther away from the trenches. The site is currently inop-

erative, and questions remain about how much risk the plume poses to the Columbia River. A conceptual model helps to shed light on the transport of the strontium-90 plume and factors that may affect that transport. The half-life for radioactive decay of strontium-90 is approximately 30 years, and there is sorption to the aquifer solids that greatly reduces transport of the plume. It was determined that the strontium-90 nearest to the river posed the only risk, because the strontium-90 in the plume farther back would decay before reaching the river. An apatite barrier was applied to sorb and sequester the strontium closest to the river and to protect the river while the remaining strontium decays in situ. This is a protective remedy, because it mitigates risk to the receptor (Columbia River), and is an example of application of a passive remedy for the majority of the plume.

Similarly, the 100-F Area is also a reactor site along the Columbia River. This site was characterized as having trichloroethylene, strontium, nitrate, and chromium plumes. At this site, the decommissioning of facilities resulted in source reduction, and with an understanding of the dynamics of the underlying plumes and the relationship between the groundwater and interaction with the river, natural attenuation of contaminants appears feasible and is predicted to meet goals based on numerical modeling. Natural attenuation at this site will be long term, and monitoring will be necessary to verify that the plume behaves as predicted and that the river remains protected from contaminants.

3.4 CENTRAL PLATEAU

Mr. Truex then explained remediation efforts in the Central Plateau region of the Hanford site, which held 149 single-shell tanks and 28 double-shell tanks. Approximately 67 single-shell tanks were known or suspected to have leaked waste into the surrounding soil. Interim actions have been taken to reduce migration of subsurface contamination, and final remedial actions will be coordinated with remediation activities elsewhere on the Central Plateau. As liquids were generated during production, they were disposed to the environment through planned and unintended releases via ponds and trenches. Within the Central Plateau region, there are multiple subsurface contaminants, such as uranium, technetium, and iodine in the vadose zone and groundwater. The vadose zone underlying the Central Plateau has properties that slow contaminants as they seep down. Because the source of the contamination is no longer present and the vadose zone retains much of the contamination, the groundwater has not yet been contaminated under this area. Many questions remain regarding how far contaminants will seep and to what extent the groundwater will be impacted. The long timeframe and minimal near-term risk because of a long path to potential exposure supports a more holistic approach to better assess the

system. This approach includes predictive modeling estimates, target actions that will be protective and reduce future risk, monitoring to verify behavior and responses of contaminants, and a progressive and adaptive remedial strategy.

These brief examples highlight key decision factors that are appropriate for complex sites and present a more holistic view that informs development of a description of the system and an appropriate response to meet remediation objectives, said Mr. Truex. Source reduction has been key in transitioning many of the plumes to natural attenuation or other terminal phases. Also, combining much of the contamination into a single operable unit with common issues has been a useful approach to addressing complex systems. Ongoing efforts to make data from the Hanford site more accessible through a web-based geographic information system (GIS) database will enhance data usage by researchers and the public and will facilitate communication. It is important to recognize that communicating and identifying the scientific basis for predictive assessments are important and that addressing complex sites in terms of maintaining protectiveness has been successful.

Jeffrey Griffin, associate laboratory director of the Environmental Stewardship Directorate at Savannah River National Laboratory, presented examples of overcoming barriers to risk-based remediation at the Savannah River Site (SRS). Dr. Griffin stated that the strategy at SRS is to match a remediation solution to the problem and to align resource investments to reduce risk. Soil and groundwater remediation strategies take into account a source, such as a facility or tank, and a zone around the source that is the high-impact zone (Figure 3-3). Addressing the high-impact zone requires a more aggressive response, such as source removal and excavation of contaminated sediments.

The next area farther out from the source is the intermediate-impact zone, which is typically characterized as having moderate levels of contamination that is non-uniform in distribution and fairly mobile. Contaminants in this zone represent a long-term risk to humans and the environment. The remediation approach to this zone would still be active, such as pumping groundwater and treating it for disposal elsewhere or re-injecting it back into the aquifer. This approach is often determined by a cost-benefit analysis of the technologies and regulatory end states. The third zone farthest from the source is a dilute or baseline zone, which is characterized by low levels of contamination. Although this zone may be less complex, it may also be larger in volume and comprise a more extended area. Risk to humans and the environment in this zone is smaller, but not negligible. Remediation approaches may be passive, such as natural attenuation. This conceptual framework of the risk associated with each contamination zone and the different remediation approaches that may be used in each zone

FIGURE 3-3 Soil and groundwater remediation strategy: matching solutions to the extent of contamination.
SOURCE: Griffin 2013.

illustrates how investment is aligned to the risk reduction that must be achieved. Ideally, a site would move from active remediation, which is a high-cost and high-energy effort, to more passive remediation, which would be a low-cost and low-energy effort.

SRS is approximately 300 square miles and has a radioactive waste burial ground, operational disposal pits around the site, and seepage basins and associated process sewer lines, similar to the Hanford site. Closure at the site is an integrated effort of Area Completion Projects to address project-specific needs through applied technology, said Dr. Griffin. This effort provides opportunities for innovative development of holistic strategies and results in cost-effective, schedule-efficient, and improved cleanup. Site cleanup is being accomplished by grouping 515 individual waste units into completion areas based on geography and priority. There are currently 14 consolidated completion areas, and 400 of the 515 waste sites have been completed.

A key component of the approach at SRS has been the Core Team Process, which is a formalized, consensus-based process in which individuals with decision-making authority, including representatives from the Depart-

ment of Energy (DOE), Environmental Protection Agency (EPA), and the South Carolina Department of Health and Environmental Control reach agreement on key remediation decisions. Each organization sends empowered decision makers, and the Core Team reaches specific and actionable agreements. The Core Team engages with technical staff and stakeholders for input into the decision-making process. This process has proven successful at expediting decision making, improving project focus, streamlining documentation, and minimizing the comment and review process, said Dr. Griffin.

Dr. Griffin presented two key Core Team case studies focusing on the F-Area Seepage Basin and the T-Area. Billions of liters of radioactive waste were dumped from separations processing at the F-Area Seepage Basin, which resulted in contamination of the groundwater. The original remediation plan at this location was to pump out and treat groundwater for strontium-90, iodine-129, and uranium. The plan was implemented several years ago, was projected to run for 30 years at a cost of up to $10 million to $12 million per year. Using the Core Team approach, more innovative practices and alternative remedies were considered for this location. The innovative thinking was to work with natural flow patterns rather than fight against them, and sequester and treat groundwater contaminants in situ. This system required developing technologies to employ enhanced natural attenuation of contaminants in groundwater using hydrologic flow models, engineered structures, and injection of chemical amendments. This system placed barriers underground to direct groundwater to a treatment zone where alkaline and silver chloride injections resulted in contaminant precipitation out of the water column. The system was successful: the pump-and-treat system was turned off, and spending was reduced from $12 million to approximately $1 million per year.

Although the T-Area did not have radionuclides, it is close to the river and had substantial amounts of solvents and degreasers. The buildings were a source of contamination and were removed, and pump-and-treat and soil vapor extraction systems were installed in 1992 to extract trichloroethylene (TCE). These systems were expected to run for 30 years and cost approximately $1 million per year. TCE concentrations decreased from about 5,000 micrograms per liter to about 600 micrograms per liter over 15 years of operation. Because the concentration levels were lower, replacement of the pump-and-treat and soil vapor extraction systems with an enhanced attenuation approach was considered. The Core Team Process was used to gain regulatory acceptance, and transition to the attenuation approach began. Edible oil, such as vegetable oil, was injected into the ground in emulsified form to create zones for the treatment of TCE by bacterial degradation. The objective was to encourage the TCE plume to flow through a series of anaerobic and aerobic treatment zones to take advantage of natural deg-

radation processes. It is projected that, with the attenuation approach, the standards will be met in approximately 10 years. Dr. Griffin commented that they continually work to move away from active to more passive remediation strategies, particularly as budgets become more constrained, because these strategies allow for standards to be met more cost-effectively and often in less time.

3.5 ALTERNATE END STATES

John Applegate, Walter W. Foskett professor of law and executive vice president for academic affairs at Indiana University, described sustainability and alternate end states for site remediation. Dr. Applegate described the evolution of risk relating to cleaning up contaminated sites. Originally, the Comprehensive Environmental Response, Compensation, and Liability Act

(CERCLA), and environmental law more generally, held the basic premise that a site would simply be cleaned up. It was either clean or not. With the growth of risk assessment, however, the realization emerged that there was a spectrum rather than absolutes, and the question of how clean is clean was continually raised. Also emerging from the discussion of risk was the discussion of future use. Understanding exposure is a component of establishing risk; risk is affected by changing exposure to a given contaminant as well as through the cleanup levels achieved at a given site. It became more evident that when end uses of facilities were taken into consideration, calculations of risk and the levels needed to be achieved would change.

In the early 2000s, DOE issued a risk-based end states directive, which made the consideration of future use more central to decision making. This shift meant that long-term stewardship would need to be considered concurrently. Sustainability is literally about time, and it is a useful concept under which risk and long-term stewardship fit well, Dr. Applegate said. Sustainable development in an international context is not about preservation but is about development while providing for the future—what is known as intergenerational equity. Sustainability is about accepting current usage while conserving the quality of the environment and avoiding taking too much from the future, he said. It is about conserving and investing in resources, which is also easily applicable to the remediation context. Sustainability is also often thought to be a process of inclusion, particularly of stakeholders. In that sense, it is similar to the risk-informed decision making that the National Research Council has recommended in the past, he said.

Dr. Applegate presented Rocky Flats as an example of the use of an alternate end state. Rocky Flats is located 15 miles northwest of Denver, where primarily plutonium triggers for nuclear weapons were fabricated. Each step in that process resulted in waste that had environmental consequences. Although a fairly standard industrial process, the process involved exotic materials, such as plutonium. There was extensive contamination of surface soils and underlying aquifers, as well as of numerous buildings that needed to be decontaminated and demolished. Early in the remediation process at Rocky Flats, future end uses were considered. The Rocky Flats National Wildlife Refuge Act of 2001 established and allocated permanent federal ownership of the entire site between DOE and the Fish and Wildlife Service of the Department of the Interior. This meant that examination of risk-based end states at Rocky Flats was enhanced by a given end state, for which the risk was able to be appropriately set. Currently, people are not allowed access to the wildlife refuge, and a retained area for cleanup remains.

Rateb (Boby) Abu-Eid of the Nuclear Regulatory Commission commented that Rocky Flats is a success story because of the focus on risk at the site’s end state, which allowed for agreement among stakeholders and regu-

lators. Similarly, the Nuclear Regulatory Commission has NUREG-1757, which provides guidance for decommissioning activities and future land use scenarios. Dr. Abu-Eid stated that conducting an assessment, analyzing costs, and exploring options in discussion with stakeholders and the public is the appropriate approach. He noted, however, that obstacles such as demand for a residential scenario instead of an alternate end state, such as a wildlife refuge, exist.

David Maloney, director of technology for CH2M HILL, discussed the application of risk-based management in site remediation. There are three enablers of the process: the contract, technologies used during cleanup and innovations implemented during the project, and regulatory flexibility. For Rocky Flats, CH2M HILL estimated in 2000 that it would take $37 billion and until 2060 for completion. The first 5 years of the project involved an ongoing dialogue with stakeholders and regulators to better understand the site and project completion. At the end of those 5 years, it was estimated that the project could be completed by 2010 for $7 billion. DOE requested that it be completed 4 years sooner and for $1 billion less, and, in return, CH2M HILL would share the savings from the project. The company completed the project at $350 million under target and in 2005 instead of 2006, which Dr. Maloney attributed largely to technology.

Needed are many technologies, as well as a complete flow sheet with detailed activity levels that includes the risk of the technologies being successful, to help chart the path forward. Dr. Maloney stated that the flow sheet for Rocky Flats, for example, built into the 40,000 activities the technology costs, schedule risks, and externality risks. CH2M HILL’s contract with DOE allowed for flexibility to use whatever methodology was needed to reach the end state, which was key to moving forward and completing the project. The end state never changed in the last 5 years of the contract once the wildlife refuge was agreed upon. Flexibility allowed for the incorporation of new technologies as they emerged from one year to the next. This type of contract was more of a partnership with DOE and state regulators, one that required those involved to have the right attitude to make such a partnership work.