| 6 | Savannah River Nuclear Weapons Facility: managing the legacy of the military's nuclear factory |
Introduction
Before the Second World War, the idea of building nuclear weapons was a scientific possibility shared among a relatively small group of physicists (McMillan 2005; Walker 2005). The first nuclear weapons were built and used by the United States to end the resistance of Japan. Subsequently, tens of thousands of nuclear weapons have been built, almost all by the United States (see below) and the former Soviet Union. With proliferation, the United Kingdom, France, China, Pakistan, India, Israel, and North Korea have, or are assumed to have, produced nuclear weapons.
The US government began nuclear weapons production during the 1940s under the “Manhattan” project, and while it maintains thousand of nuclear weapons, the Department of Energy (DOE) has been dismantling unused weapons and managing their by-products (Office of Environmental Management 1995). This chapter is about the use of the EIS process in the DOE's efforts to manage this nuclear defense waste legacy.
I had four reasons for writing a chapter on the nuclear weapons legacy. First, the complexity of managing this legacy is remarkable; I think unprecedented. The United States literally created a distributed nuclear factory at over 130 locations in more than thirty states to design, build, and test nuclear weapons and manage the by-products of these weapons (ibid.). The nuclear bomb effort has required unprecedented cooperation among scientists, engineers, and many other professionals at multiple sites. I wanted an EIS that would provide a snapshot of the complex process of managing important pieces of the nuclear waste cycle.
Second, because some of the nuclear and chemical materials are extremely hazardous and will remain so for tens of thousands of years or longer, the science and engineering has been extraordinarily challenging, in many cases requiring first-of-a-kind design, engineering, and evaluation. This type of EIS is the most technically demanding one described in the book. It often contains numerous flow charts, tables and terse technical summaries and jargon, and it represents many individual projects. I tried to focus on the major points and on a few components of the EIS in order to provide some depth. It is important for readers to see for themselves that some EISs are difficult to read, and doubtless difficult to write.
Third, the cost of this national effort at over 100 locations has already been $90–100 billion, and could reach over $300–375 billion (Office of Environmental Management 1995, 1997; interview with Henry Mayer, Executive Director, National Center for Neighborhood and Brownfields Redevelopment, Rutgers University, May 18, 2010). The cost of this waste management program is larger than that of any other government-run waste management program in the United States, and I believe larger than any other in the world, albeit it is not the most serious nuclear-related problem. The construction and operation of projects described in this EIS by far exceed the cost of any other project described in this book. What is notable about these costs is the juxtaposition of high cost and relatively minor near-term risks. Like the destruction of the chemical weapons on Johnston Island in Chapter 5, the risks are relatively small and are driven by legal mandates on the US government, stakeholder agreements with the DOE, and by the US government's ethical commitment to close the circle on these weapons of mass destruction (Carter 1987; Office of Environmental Management 1995). No US resident is being supplied with a road or train for travel, an energy product to heat their home or business, a historical place to visit, water to drink, or another obvious useful service. These EISs represent a set of documents written to address legal mandates and treaties, which implies a different role for project cost and politics than is normally found in business.
Fourth, the larger DOE sites are in relatively remote locations, where there are no immediate neighbors. But these sites are not ignored. Four regions, including Savannah River and Hanford in the United States, have depended on the DOE's weapon production, decommissioning, and waste management programs for their economic health. Accordingly, there is great local concern about DOE's expenditures, and trepidation about them stopping. The public reaction to new waste management projects is not like those I have seen elsewhere. Many strongly support new facilities. For example, a recent study showed that a near majority favored locating new nuclear power plants in the Savannah River region, and more than a majority favored locating new nuclear waste management facilities in this region. Both figures were notably higher than a national sample (Greenberg 2009). In addition, each of the major sites has an officially appointed citizen advisory board, and their reactions to the EISs are interesting (see below for further discussion).
Given these four reasons for wanting a case study of nuclear waste management, I could have chosen from over a dozen national programmatic and site-specific EISs on a variety of subjects. I have worked on nuclear waste management on and off since the 1970s, and so it was important that I not pick any single project that I had studied in detail, and therefore might, consciously or otherwise, have a bias toward or against because of my own research. Instead, I wanted to focus on a major planning document that involved numerous projects.
I chose to focus on one seemingly innocuous final EIS written in the mid-1990s for the Savannah River Site (SRS) in South Carolina. This EIS (US Department of Energy 1995a) illustrates the four reasons I wanted to present an analysis of this legacy management program: complex science and engineering, extremely hazardous materials, high costs, and a notable public involvement process. This site-wide environmental management EIS has embedded within it dozens of individual actions at the site and elsewhere across the DOE nuclear complex. For example, one set of EISs was written for the Defense Waste Processing Facility (DWPF) (US Department of Energy 1995a,b), which takes high-level nuclear waste and mixes it to produce large glass canisters (which look like large metal logs) that immobilize the highly radioactive waste (see below). The DWPF was my first choice for a nuclear waste EIS because it was such an interesting project and EIS. However, it is a very technical document and, more importantly, would not provide the opportunity to examine the full set of issues that this site-wide document presents; furthermore, the DWPF is part of the EIS to be reviewed here.
Before reviewing the specific EIS, I provide context about nuclear weapons and their wastes, and regulation of military nuclear waste. The EIS will be easier to understand in the appropriate context.
US nuclear weapons and wastes
US nuclear weapons are complicated devices designed to explode only when authorized to do so. This means that they have complex safeguards to prevent accidental or deliberate and malicious detonation. There are a number of basic types of nuclear weapon. So-called “atomic bombs” use high-grade conventional explosives and uranium that has been enriched with fissionable materials. The high-grade explosives trigger fission of the enriched uranium, that is, the enriched uranium atoms split apart causing a chain reaction. US nuclear weapons use the artificial element plutonium as the fissionable material. The explosives momentarily compress a ball of fissionable material in a metal sphere sufficiently to reduce neutron leakage so that the neutrons released during fission generate a “chain reaction” that produces more neutrons in each successive generation (Walker 2005, Greenberg et al. 2009).
Modern US nuclear weapons are so-called “thermonuclear” devices. They use conventional high-grade explosives to trigger fission, which then triggers fusion of heavy hydrogen atoms in tritium gas. In other words, atoms are fused together, rather than split apart as they are in fission. Fusion is what powers our Sun. Thermonuclear weapons release energy from both fusion and fission.
To put the nuclear factory and the Savannah River waste management role in a broad context, the Brookings Institution (1998) compiled facts about US nuclear weapons mostly from published US government sources. The United States built more than 70,000 nuclear warheads and bombs, 67,500 missiles, 4680 bombers, and eighteen ballistic missile submarines to carry these weapons. Fissile materials produced were 104 metric tons of plutonium and 994 metric tons of enriched uranium. In 1966, which was the peak year for nuclear weapons in the United States, the nation had 32,193 nuclear weapons and bombs. Stephen Schwartz (1998) estimated that the US government spent about $5.5 trillion on its nuclear weapons and delivery systems. When updated for the changing value of the dollar, this is over $100 billion a year ($100 billion a year from 1943 to 1998 in 1998 dollars; $131 billion a year on average in 2009 dollars). The DOE's environmental management program at the Savannah River has averaged over $1 billion a year for a decade (Greenberg et al. 2003; Mayer, interview, 2010, op. cit.).
Since the end of the Cold War, the number of weapons has dropped substantially. For example, in July 2007, President George W. Bush set a goal of cutting the number of deployed strategic nuclear weapons in half by 2012. This means that the United States would have between 1700 and 2200 operationally deployed weapons by 2012, the lowest number since 1950, which is the range required under the bilateral agreement between the United States and the former Soviet Union. Both countries have agreed to reduce their weapons-usable plutonium reserves by 34 metric tons (Greenberg et al. 2009). On April 8, 2010, President Obama and Russian President Dmitry Medvedev agreed to further reductions in warheads and launchers (Baker and Bilefskey 2010).
The tens of thousands of US nuclear weapons that are no longer part of the assured mutual destruction policy have been taken apart so that they no longer can be used as weapons. Each weapon has thousands of components. The vast majority of these are complex electronics, which are not discussed in this chapter. The actual bomb material, the so-called “physics package,” has only several hundred components. The fissionable material, consisting of plutonium that was part of the nuclear weapon, is the focus of environmental management.
The vast majority of the radioactivity in US nuclear waste is in the spent fuel rods from nuclear power generation. However, although the majority of radioactivity is in the spent commercial fuel rods, most of the volume of nuclear waste is military. The United States is no longer producing nuclear weapons, and yet there is a long legacy from decades of nuclear weapons production and associated research. The military waste is not as “hot” as the commercial civilian waste (it was not in reactors very long), but a good deal of it is hazardous. The DOE is processing a substantial amount of defense-related waste at SRS, and the EIS reviewed in this chapter is central to that national effort. In fact, the DWPF at Savannah River is the largest such processing facility of defense waste in the world, at least until the facility at Hanford in the State of Washington is completed.
Federal laws and regulations
Nearly all of what the EIS in this chapter proposes is required or conditioned by federal laws and regulations. Nuclear waste is categorized by federal laws, regulations, and rules. It is important that the reader understand that the classification does not necessarily correspond directly to hazard levels. Some low-level waste can contain highly radioactive constituents. With that caveat, about 90% of radioactive waste, by volume, is classified as low-level, and only about 0.3% is high-level. However, high-level waste contains about 95% of the total radioactivity of all nuclear waste (Greenberg et al. 2009).
The four main categories as they pertain to defense waste at SRS are as follows:
- High-level waste. These wastes are produced by reactions inside nuclear reactors, including those produced by military programs at DOE sites.
- Transuranic waste. This waste contains elements with atomic numbers (number of protons) greater than 92, the atomic number of uranium (hence the term transuranic, or “above uranium”.) Most of this waste was produced from defense-related activities. With regard to waste management, transuranic waste includes only waste material that contains transuranic elements with half-lives greater than 20 years and concentrations greater than 100 nanocuries per gram. If the concentrations of the half-lives are below the limits, it is possible for waste to have transuranic elements but not be classified as transuranic waste. At present the United States permanently disposes of transuranic waste generated from military facilities at the Waste Isolation Pilot Plant (WIPP) in New Mexico, near Carlsbad. At the time this EIS was written, WIPP was not open, and there is considerable discussion of transuranic wastes in the EIS.
- Low-level waste. Whatever is not high-level, used fuel from a power plant, transuranic, or by-product material (e.g. uranium mill tailings) is classified as low-level. It includes everything from radioactive garbage (e.g. mops, syringes, and protective gloves) to highly radioactive metals from inside nuclear reactors. Low-level waste is typically stored on-site by licensees, either until it has decayed away and can be disposed of as ordinary trash, or until amounts are large enough for shipment to a low-level waste disposal site in containers approved by the US Department of Transportation. Low-level waste has four subcategories according to activity level and lifespan: classes A, B, C, and “greater than class C” (GTCC). On average, class A is the least hazardous, while GTCC is the most hazardous.
- Mixed wastes. Some radioactive wastes have been mixed with non-radioactive hazardous substances, such as organic solvents or other toxic chemicals. Much of this waste (especially the transuranic waste) contains substantial quantities of long-lived radionuclides, such as plutonium-239 and technetium-99. These hazardous components are regulated by the Environmental Protection Agency (EPA).
The EIS described in this chapter was written when the imminent opening of a national geologic repository was assumed. In 1982, Congress passed the Nuclear Waste Policy Act (NWPA, P.L. 97-425), which stipulated disposal of high-level waste in a permanent deep geologic repository (Kubiszewski 2006). First, the defense high-level waste was to be encapsulated in a glass matrix (vitrified) and then placed in canisters (see above). The NWPA required the DOE to look for at least two suitable sites: one in the west, and one east of the Mississippi River. In 1987, the law was amended (National Safety Council 2001; MacFarlane and Ewing 2006) to say that Yucca Mountain in Nevada would be the only nuclear waste repository for high-level waste. The containers would be placed in underground tunnels. The tunnels, also known as drifts, would be about 1000 feet below the surface and, on average, 1000 feet above the water table to minimize natural exposure to moisture.
The DOE submitted a license application to the Nuclear Regulatory Commission in June 2008, requesting authorization to construct a national repository for the disposal of spent nuclear fuel and high-level radioactive waste at Yucca Mountain. The application consisted of a letter describing its purpose, expecting that permanent disposal would begin in 2017. However, this schedule has been delayed by funding limitations, problems obtaining permits, and litigation (for studies and articles see the DOE website on the Mountain repository at www.energy.gov/environment/ocrwm.htm). The Yucca-only decision has become a major political issue. Nevada has fought the Yucca-only policy (Reid 2007). When the SRS impact statement was written, a single repository at Yucca Mountain was assumed. The chances of it ever being used at as a permanent repository are slim, certainly not any time in the near future. While the Yucca site has not received any waste, WIPP, near Carlsbad, New Mexico, began receiving transuranic wastes in 1999 (National Safety Council 2001).
Also, at the time the SRS EIS was prepared, reprocessing of high-level nuclear waste was not assumed for the United States. The Global Nuclear Energy Partnership (GNEP) is a new program to study the impacts of reprocessing spent fuel rods and reuse the uranium, plutonium, and other transuranics for fuel (Greenberg et al. 2009). The process would produce nuclear fuel as well as radioactive wastes. Some of the uranium fuel would be exported to other countries as part of a global strategy to prevent proliferation of nuclear weapons. Countries could then have nuclear power plants without establishing their own uranium enrichment or reprocessing facilities, which would reduce the chances of extracting enriched nuclear materials for nuclear weapons. (It is also extremely expensive to obtain fissionable weapons-grade materials in this fashion.) The nuclear fuel could be burned as fuel in existing reactors, or in a new type of reactor to be developed and built in the United States. If it succeeded, the reprocessing and recycling would reduce the total volume and toxicity of highly radioactive wastes (ibid.). But GNEP is controversial and has been challenged on economic, environmental, moral, political, public health, technological, and other grounds. GNEP was not considered when the EIS in this chapter was written, and at the time this book was being written, the future of the GNEP concept was undetermined.
Meanwhile, US military high-level waste is being stored primarily at the Hanford, Savannah River, and Idaho sites. The main issue with regard to waste of defense origin is the high-level waste that remains in massive, largely underground storage tanks. These nuclear wastes in tanks were mixed with other substances, substantially complicating every aspect of environmental management. Some of the tanks have leaked, but all of them must be carefully monitored. Waste sometimes is transferred from one tank to another.
Savannah River Site environment and long-term stewardship
With regard to the local environment, the SRS near Aiken, South Carolina is about 20 miles southwest of Augusta, Georgia (see Figure 6.1). Established in 1950, the 310-square-mile SRS is among the DOE's largest facilities, with 15,000–20,000 employees and a budget well over $1 billion a year (Office of Environmental Management 1995) when this EIS was prepared.
In 1950, E. I. DuPont de Nemours was asked by the federal government to build and operate the facility. Its historic mission in the weapons complex was to produce plutonium (weapons-grade material) and tritium (US Department of Energy 2010). In 1952, the site produced so-called “heavy water,” and in 1953, the first production reactor began to operate. In 1956, the facilities that locals often call the “bomb plant” were completed. The defense mission has not disappeared. For example, in 1986, construction of a new tritium facility began.
While the bomb mission was prominent, other activities were occurring. In 1963, spent nuclear fuels from other sites were received. In 1972, the site was designated as the first National Environmental Research Park. In 1981, the environmental cleanup program began. This included shutting down of reactors, constructing treatment and burial facilities, and starting construction of the DWPF.
Savannah River Site
When the Cold War ended, SRS's role in the DOE Complex changed primarily to waste management and ecological research. SRS's role has continued to change depending on national needs; it may take on new military roles, and is currently being considered as a location for an energy park that would pursue science in solar energy, wind, biomass, coal, and nuclear energy (Gilbertson 2009).
The DOE's major sites in Idaho, Nevada, South Carolina, New Mexico, Tennessee, and Washington are surrounded by prairies, wetlands, deserts, and forests. Roughly 10% of the land is contaminated with radioactive materials released from the facilities, and there is also some chemical contamination from site operations (Burger 2000; Burger et al. 2003). Roughly 90% of the land is not contaminated, and usually represents the few remnants of old-growth forests and other vegetation in the region that have not been disturbed for more than half a century. SRS is illustrative, containing large tracts of forested areas. Almost 80% of the land was deliberately left as buffer areas. In 1997, the author wrote an article that characterized the site as “bombs and butterflies” (Greenberg et al. 1997), although another commentator felt that “mosquitoes” was a more accurate descriptor than “butterflies”.
The active or formerly active parts of these sites are so far from population concentrations that the hazards they contain pose relatively little risk to people living in the region, although worker risk cannot be discounted. Radiation and chemical exposures to the animals and plants can be a problem. An exposure could occur, for example, when an animal ingests, absorbs, or breathes a radionuclide, or when a plant draws a contaminant out of the soil. The effect varies in accordance with the amount of the dose, length of exposure, and vulnerability of individual organisms to damage. It is important to know that the DOE's mission includes environmental restoration, and DOE's stewardship program includes ecosystem management and integration of economic, ecological, social, and cultural factors in land-use decisions (DOE Order 430.1). When the DOE prepares an EIS for SRS, or for another of the major sites, it includes that ecological responsibility.
Major concerns related to the impact of nuclear materials are exposures to top-level predators (e.g. eagles and hawks) through ingestion, and the health and safety of wildlife populations that may be eaten, hunted, photographed, or viewed, or that are just part of the environment. Threatened or endangered species are always of special concern. Stripping the top layer of soil to reduce radiological contamination can harm existing animal and plant populations more than the radiation levels that existed before remediation (Burger 2000; Burger et al. 2003).
Remediation described in an EIS can destroy intact ecosystems that will never recover because of the degree of soil or other disruption. Remediation could be so extensive, for example, that it would destroy the seed bank and allow invasive species to move in, eliminating native species. Also, intrusion will mean movement of nuisance and other animals out of the area to escape remediation, perhaps to nearby suburban areas.
While we may not see what goes on in water bodies, aquatic animals are particularly at risk because contaminants move quickly through water. Also, some animals that do not move far from the contaminant source, or spend more time in the actual contaminated medium (fish in water), will have higher exposure. For many species, leaving waste sites alone would often be the best remedy. Cleaning up contaminated sites can re-expose animals by disrupting groundwater, water, sediment, and soil, potentially releasing contaminants back into the environment.
Ecological risks from radioactive contaminants will decrease if there is no new contaminant release because radioactive substances gradually decay. Yet the risk of some radionuclides will not decrease significantly for many decades or centuries. If action must be taken, the best approach is to avoid new roads being built through the system, and to disrupt the soil, trees, water, and the rest of the ecosystem as little as possible. DOE site-wide EISs must grapple with the dilemma of balancing the removal of hazards and disrupting ecosystems. The document does not provide much context for the non-expert about these trade-offs and their implications for on-site ecosystems.
Some radioactive contaminants with short half-lives will rapidly naturally decay into benign and other not-so-benign substances. However, some of the long-lived contaminants will be radioactive for tens of thousands of years or more. In addition, polychlorinated biphenyls (PCBs), solvents, and heavy metals do not decay much and can remain hazardous in perpetuity. To ensure that human health is protected for many generations, long-term stewardship must be carried out at sites like SRS even after active remediation (Greenberg et al. 2009). A remediated site could be cleaned up and even closed, but contain hazards in the form of stored or buried wastes, entombed facilities, such as reactors, or residual contaminants that are left in place.
Engineered controls for this purpose include physical containments around landfills, vaults, tank farms, or other waste units. They include operating, maintaining, inspecting, and monitoring caps, erosion control systems, environmental sampling, wells, and pump-and-treat groundwater remediation systems.
Institutional controls are legal tools to reduce exposure by ensuring that land- and water-use restrictions are maintained. They include restrictions on land or water use, well-drilling prohibitions, deed notices, easements or other legal advisories or measures, and long-term information management. Access obstacles such as fencing, markers, and signs can be considered passive controls. Federal ownership in perpetuity is itself an institutional control. The reader would not necessarily know that these controls are in place, assumed, or required by reading this document.
The final Savannah River waste management EIS, 1995
This EIS concerns the management of legacy and newly created high-level radioactive waste, low-level radioactive waste, hazardous and nonradioactive waste, radioactive and hazardous waste (mixed), and transuranic wastes at SRS. It does not consider domestic (sanitary waste) or domestic or foreign spent nuclear fuel. The wastes it focuses on must be managed to protect human health and the environment, and be in compliance with regulations, yet the choices are expected to be as cost-effective as possible.
The complexity of this EIS cannot be understated, and it is a reflection of the site itself. When the US government acquired the land in 1951, 60% of the 310 square miles of land area was forest and 40% was cropland and pasture (US Department of Energy 1995a). Almost 60 years later, about 69% was upland forest (predominantly pine) managed by the US Forest Service, 22% was wetlands and water, and 9% was waste management and other activities.
Almost all of the land that is used by the DOE for production, waste management, and administration is concentrated in a core area (Figures 6.2 and 6.3). Outside this developed area, as noted earlier, the site is an ecological treasure, a rare place where humans have had little impact for at least 60 years. Satellite images show a massive forested area with a few open spots. Several roads run through the site. Important for the EIS, the site's waste management operations are influenced by national-level decisions about weapons production and decommissioning, as well as waste management. Readers looking down at the site from above would see three kinds of activity that involve waste management focused in the inner area. One set of activities is ongoing operations related to chemical separation and processing of waste located in the high-level waste tanks. A second set of activities is decontamination and decommissioning of hundreds of buildings. The third major environmental management-related activity is environmental restoration. Each of these is central to decision-making about the site and is reflected in this EIS. In addition, activities at other sites across the DOE Complex impact SRS and are part of this EIS.
This EIS lists sixteen NEPA-related reviews issued between 1987 and 1995 that influence what is presented in this EIS. Five of the sixteen were DOE-wide assessments about tritium supply for nuclear weapons and recycling; nuclear weapons nonproliferation about spent nuclear fuel from foreign reactors; nuclear stockpile stewardship; and two others. The remainder were SRS-specific EISs and initial environmental assessments, including ground-water protection, incineration, mixed waste treatment, the DWPF, and plutonium management.
The impact of nonlocal decisions on waste management at SRS is illustrated by the Pantex and Idaho EISs summarized in this volume. Pantex, located near Amarillo, Texas, assembles and decommissions nuclear weapons. The EIS for the continued operation of Pantex influences the type and amount of waste that would come to SRS. The Idaho nuclear facility programs for spent nuclear fuel are also part of the consideration of this SRS waste management EIS, as was the WIPP facility for transuranic wastes.
Savannah River Waste Management Area
Major developed areas
My key point is that, in response to federal mandates, the DOE created proposed options to manage multiple sources of waste that were created in the past, are still to be created, and involve transport of products across the United States to and from the site. The author views this site as a giant chess board with pieces moved around at various locations by chess masters located at the site, Washington DC, and other DOE facilities.
Four options for the Savannah River 1995 EIS
This EIS offers a no-action alternative and three alternatives for managing nuclear waste for the next 30 years (1995–2024). The options are different from any of the other alternatives presented in this volume. In the other chapters, the alternative action was relatively distinct, although some alternatives had overlapping activities. In this case, all four options have overlapping actions, and this EIS has no real no-action alternative. The four alternatives are labeled “no-action,” “limited treatment,” “moderate treatment,” and “extensive treatment.”
All of these alternatives have implications that can be measured by radionuclide releases and storage, exposure to workers, the public, and ecosystems, jobs, and in other ways. In order to begin with a uniform and relatively easily comprehensible summary metric, I have provided a summary table of new structures that the document says will be required under the various alternatives (Table 6.1).
“No-action” option
The “no-action” option does not literally mean no action will be taken. It does mean a continuation of current practices, because a true no-action alternative would violate the DOE's requirements to manage the wastes. This no-action alternative continues past practices of constructing facilities to store new radioactive wastes. The no-action alternative would leave both transuranic and mixed waste untreated, in essence, in storage. This would mean that the DOE would not meet its legal agreements.
Beginning with structures that would exist with all four options, the DOE would construct vaults for the disposal of low-level waste and for hazards and mixed wastes. It would also include a treatment facility, a long-lived water storage structure, a high-level waste evaporator, and a new waste-transfer facility. The no action alternative, like the other alternatives, also relies on the DWPF to convert high-level wastes into vitrified containers.
Table 6.1 Summary of new waste management facilities proposed by four alternatives and three waste forecasts*
* Waste forecast: min = minimum; exp = expected; max = maximum.
Adapted from US Department of Energy 1995a, EIS-217, Table S-1, p. S-21.
Table 6.1 shows that the DOE would be required to build seventy-three new waste tanks in three areas of the site that would not be required with the other options (see the no action column of Table 6.1). It would also require building shallow land disposal trenches, low-activity and intermediate level waste vaults, and a Resource Conservation and Recovery Act (RCRA)-permitted disposal facility.
Before describing the other alternatives, two other issues must be noted. All of these options, according to the document, would protect public health and the environment. In other words, the DOE's legal mandates are satisfied by any of the three.
The DOE focused on five criteria when it evaluated the options:
- waste processing variables, such as achievable volume reduction, secondary waste generated, and efficacy of decontamination and decommissioning;
- engineering variables, such as effectiveness of technology, its maturity, and needed maintenance;
- safety, public health, and environmental impacts, including worker, public, environmental, and transportation risk;
- public acceptance and political consequences, including regulatory requirements, permits, and schedule;
- cost-effectiveness in the near-term and over the next 30 years.
Local site managers, as noted above, do not have complete control over what they will do on the site. Analysts recognized this complication, and thus estimated the minimum, expected, and maximum amount of waste the site could receive for the following five waste categories: liquid high-level radioactive waste; low-level radioactive waste; hazardous non-radioactive waste; mixed (radioactive and non-radioactive hazardous) waste; and transuranic waste.
Hence, rather than looking at four options, the reader looks at ten (one no-action; and three each for the limited, preferred, and extensive options). Because of the complicated options, I have chosen to present the preferred option last in the text rather than first.
Limited treatment option
This option meets required mandates, but not much more. For example, low-level waste would be treated by existing compactors already in place before storage. Hazardous waste would be recycled, then sent off-site for treatment and disposal, or incinerated in the Consolidated Incineration Facility (CIF).
The waste forms would be safe, but not represent state-of-the-art treatment. A negative implication is that there would be more waste generated. Yet an advantage is that potential worker and public exposure would decrease in the short run because there would be less handling of the waste. Table 6.1 shows that this approach requires more waste storage vault and RCRA-permitted disposal facilities than the other two options.
Extensive treatment option
In contrast to the limited treatment option, this one would require far more extensive treatment of wastes to reduce their toxicity and volume and create stable waste forms that would be difficult to move. The long-term toxicity impacts would be reduced, but the trade-off is that there inevitably would be more exposure to workers and possibly the surrounding population due to additional handling and processing of the waste. This option means more reliance on the incineration facility to burn wastes, repacking and then storage or shipment of waste. Table 6.1 shows more reliance on shallow land disposal trenches, but what is not shown in the table is the greater reliance on vitrification for low-level waste at SRS and even other hazardous wastes.
Moderate treatment – the preferred alternative
Most moderate treatment options can be seen as a way of reaching a politically and/or economically acceptable solution. Arguably, the same is true in this case. But there are health and safety reasons for this choice. The EIS takes a position that the DOE chooses to concentrate its resources on the wastes it considers most likely to impact human and environmental health, and will place less emphasis on treatment for less hazardous wastes. A good example is transuranic wastes. Under the extensive treatment scenario, all transuranic wastes are to be vitrified. Under the least treatment option, they would all be repackaged in accordance with federal law. The preferred option distinguishes between plutonium-238 and highly radioactive plutonium-239 components of transuranic wastes on the one hand, and all the other transuranic forms on the other hand. The first two would be vitrified and the remainder repackaged. In essence, this is a risk-based approach to allocating limited waste management resources.
Public reports of the EIS focus on the DOE's choice of the moderate treatment option. The most obvious differences in Table 6.1 are between the minimum and maximum waste inventories that would have to be managed. The number of long-lived low- and high-level waste storage buildings does not vary among the alternatives. But the number of mixed waste and transuranic storage pads is markedly greater if the site generates and receives additional waste products, as do the number of shallow land disposal tranches, low-activity and intermediate waste vaults, and RCRA-permitted disposal facilities. This distinction is clearly highlighted in Table 6.1.
Behind these notable differences in the impacts of the options are policy-related assumptions about what is going to happen both on- and off-site. The DOE's analysts took 1993 and 1994 data and made assumptions about what could conceivably happen. Space does not permit a full recitation of what the waste forecasts assume. For example, the maximum waste forecasts make different assumptions about aluminum-clad spent nuclear fuel coming to SRS for processing from Idaho; about plutonium and tritium from Pantex coming to SRS; and nine other questions. Depending upon the answers to these questions, SRS could have a great deal more waste, which was envisioned in the maximum waste forecast. For example, the expected waste forecast for the years 2000–24 assumes that 182 facilities inside the central area will be gutted and 423 outside the area will be taken down to their foundation. In strong contrast, the maximum waste results assume that all 182 in the central area will be taken to foundation and all 423 outside the area will be taken to “greenfield.” The difference between gutting, taking to foundation, and removing implies a lot more waste being moved around the facility.
A more detailed example is how the three options plan to deal with the residuals of spills, so-called “spill units.” The minimum forecast assumes that forty of the 134 spill units (30%) would have their wastes removed. In contrast, the maximum waste forecast assumes all 134 would have their wastes removed. The expected waste forecast assumes that half (sixty-seven) would have their waste removed.
Before highlighting some of the impacts, it should be noted that the 30-year study period for this plan leaves the site with ongoing environmental management operations that will last well into the twenty-first century. Table 6.2 (opposite) lists the facilities that manage the high-level nuclear waste.
Environmental impacts considered: some illustrations
The 310-square-mile SRS site sits on the Atlantic Coastal Plain, and includes parts of Aiken, Allendale, and Barnwell Counties in South Carolina. The impact sections examine potential environmental effects during both the construction and operations of proposed new facilities. The report examines impacts on air, water, animals, and plants, and on worker and public health for those who live in the area. The report also includes presentations about economic and social impacts.
None of the environmental conditions is striking. Impacts vary much more by amount of waste received than by waste management options (no-action and three alternatives). The key to understanding the message is that the waste forecasts mostly depend on the extent of decontamination, decommissioning, and restoration of the site. Rather than go through every category, I focus here on some of the key impacts that were addressed (see Table 6.3, p. 158).
Table 6.2 Facilities that would operate beyond planning period, 1995–2024
Type |
Function |
Defense waste-processing facility |
Vitrifies high-level radioactive waste |
Z-area saltstone manufacturing and disposal facility |
Saltcrete processing and disposal |
F/H-area effluent treatment facility |
Treatment of routine process effluent and wastewater |
In-tank precipitation |
Removal of radionuclides from highly radioactive salt solution |
Savannah River Technology Center |
Research and development |
Replacement tritium facility |
Separates tritium from targets |
Type III liquid high-level waste tanks |
Storage of liquid high-level waste, sludge and saltcake |
Consolidated incineration facility (stops operating under maximum treatment alternative in 2006) |
Destroys selective radioactive and other hazardous wastes |
New special recovery facility of 221 FB-line |
Recovery of plutonium scrap |
Powerhouse, water treatment and support facilities, analytical laboratories |
Produce on-site energy, treatment of powerhouse effluent and laboratory services |
Adapted from US Department of Energy 1995a, EIS-217, Table 2-3, p. 2-21.
Land-use impacts
Table 6.1 shows the difference in number of facilities, which is reflected in land-use impacts. The no-action alternative would require an estimated 160 more acres (quarter of a square mile). This compares with 107, 157, and 1010 acres (the last is 1.6 square miles) for the preferred minimum, expected, and maximum waste forecast estimates, respectively. The document indicated that each forecast is consistent with the site land-use plan, which is to concentrate development in the central area (see Figure 6.2).
Ecological impacts
Under the no-action alternative, 160 acres of forest would be required. The requirement for the preferred option would be 117 more acres under the expected waste forecast. Under the minimum waste forecast, only 90 acres would be required. The notable difference is that 960 acres (1.5 square miles) would be required for the preferred option and maximum waste combination. This represents less than 1% of the forested area on the site. Yet threatened and endangered species and wetlands could be affected, although the DOE presumably will do surveys to avoid this. Avoidance of specific ecologically important areas and other management options would also reduce impact on endangered species that inhabit these historical forests.
Table 6.3 Environmental impacts considered for Savannah River Nuclear Weapons Facility
Geologic resources |
|
Groundwater resources |
|
Surface water resources |
|
Air resources |
Construction |
Ecological resources |
|
Land uses |
|
Socioeconomics |
Construction Operations |
Cultural resources |
|
Aesthetics and scenic resources |
|
Traffic and transportation |
Traffic |
Occupational and public health |
Occupational health and safety |
Facility accidents |
|
Unavoidable adverse impacts and irreversible or irretrievable commitments of resources |
|
Cumulative impacts |
Existing facilities, |
Adapted from US Department of Energy 1995a, p. 4-191-B.
Large and mobile animal species inhabiting the undeveloped portions of the site (e.g. fox, raccoon, deer) should be capable of avoiding the construction equipment. However, small and relatively immobile species (small mammals, reptiles, amphibians) would not. Dispersion of the survivors would, in turn, pressure other species in adjacent areas. The net result is fewer total species and possibility less diversity, despite the very small magnitude of these impacts. Timing of the land clearing would be important. The DOE would try to avoid the spring and summer, when the impacts on nests and breeding would be maximum. Impact on wetlands and streams could be minimized if the DOE required the installation of erosion control and used best practices.
Socioeconomic impacts
The Savannah River region is heavily dependent on the DOE's activities. In an earlier study, the author and colleagues (Greenberg et al. 2003) estimated that 16% of the gross regional economic product in counties near the site came directly from site activities. In the nearest areas, the proportion is more than half the product. In a large urban area such as Atlanta, 170 miles to the west of the site, SRS would account for less than 1% of the local economy. Accordingly, economic impact is important in site-specific EISs when the region is otherwise rural, and SRS has experienced serious regional economic recessions when the DOE shrunk the workforce while the United States as a whole was growing. On the other hand, the area has been less impacted than places such as Detroit and other private manufacturing centers when national recessions have occurred (Greenberg et al. 2003). Reflecting its economic importance, the local media have made this and similar rural sites a major priority (Greenberg et al. 2008). Any possibility of new jobs and regional income is important to many people.
The economic impacts are based on the estimated construction and operations personnel required to implement the remedial options described in the EIS. Impacts on socioeconomic resources can be evaluated by examining the potential effects from both the construction and operation of each waste management alternative on factors such as employment, income, population, and community resources in the SRS region.
Over 100 jobs are estimated to be required for constructing the no-action alternative facilities in the peak construction year. Operations employment is much higher, about 2450 jobs a year during the period 2003–24. This is about 12% of the mid-1990s SRS employment. While this is a sizeable number of jobs, the site was expected to lose over 4000 jobs (from 20,000 to 16,000). Hence DOE expects that these added jobs would be filled through the reassignment of existing workers.
With regard to job creation, the maximum waste forecast presents a notable difference. The maximum peak employment estimate is 330 jobs during the period 2003–05. Operations employment associated with this maximum forecast was estimated to be 10,010 from 2002–05. This is about half of DOE's forecasted regional employment in 2005. The report assumes that half of workers could come from the existing labor force and the other half would be added. If this had occurred, it would have meant more people, more income, and indirect benefits to the local economy. There would be more demand on schools and other services, but more money to pay for these services, and as a result a net addition to the regional economy of between 1.5% and 3%.
Groundwater impacts
One of my issues with the EIS is that parts of the document are not readable to anyone without a technical background, multiple documents in front of them, and/or a technical glossary (see below for a lengthier discussion). I illustrate this with direct quotes from the report about groundwater. I chose groundwater because water resources are always a major public concern. With regard to the no-action alternative, I quote from several pages that focus on the use of trenches and vaults to contain radioactive wastes:
The disposal of stabilized waste forms (ashcrete, glass) in slit trenches was not evaluated in the Radiological Performance Assessment and is subject to completion of performance assessments and demonstration of compliance with performance objectives required by DOE Order 5820.2A (“Radioactive Waste Management”). Therefore, DOE was unable to base an analysis of stabilized waste in slit trenches on the Radiological Performance Assessment. The analysis presented in the draft EIS did not account for the reduced mobility of stabilized waste forms in slit trenches. The final EIS assumes that releases from these wastes in slit trenches would not exceed the performance objectives specified by DOE Order 5820.2A. As a result of the modified assessment approach, exceedances for uranium and plutonium isotopes identified in the draft EIS under some alternatives and waste forecasts are no longer predicted to occur. DOE would re-evaluate the performance assessment and, if necessary, adjust either the waste acceptance criteria or the inventory limit for the storage or disposal units to ensure compliance with these criteria, or standards which may become applicable in the future. The results of applying this assessment methodology to the different storage and disposal facilities are presented below.
(US Department of Energy 1995a, pp. 4-10,4-11)
This 193-word example is typical of this EIS and many others prepared by the DOE; that is, it is full of technical language, reference to other sections of the document, and other reports. The author is not accusing the writers of providing inaccurate information or of deliberate obfuscation with technical complexity. However, relatively few people would have the expertise to understand these sections and, even if they did, the author doubts if they would have the patience to read hundreds of pages of text. The appendices referred to are even more technically oriented.
The essence of the message about the trenches and vaults is that institutional controls are essential to make sure that the waste vaults and trenches are maintained. If institutional controls were abandoned and the equipment and vaults were permitted to degrade, there eventually would be leaks into the surrounding land and then to groundwater. The extent of the impact is uncertain because of limited data and modeling.
Worker and public health impacts
This EIS considers various potential exposures and health effects among workers and the surrounding public. The DOE calculated chronic emissions and accident scenario exposures and consequences. These analyses are some of the more interesting science-based work in the EIS. This site as a whole and the new facilities are very remote from the nearest residential concentration some 25 miles away. Accordingly, the off-site human health impacts are very small, and in essence are not measurable. I will present the worker latent cancer estimates. I picked cancer as the major concern because there is a great deal of radiation on these sites and cancer is such a prevalent disease, making it difficult to distinguish site-related exposures from other contributing factors. I note that some DOE sites have a legacy of potentially toxic beryllium, asbestos, and other metal contamination. But these vary by site, and are not serious problems at every major DOE site in comparison with radiation, which is a consistent concern.
With regard to the no-action alternative, exposures result from handling effluent destined for treatment facilities, waste tank farms, and storage pads. The DOE used industry standard codes and assumptions about exposures at various distances from the facilities. DOE regulations (10 CFR 835) require that annual doses to individual workers not exceed 5 rem per year (roentgen equivalent in man, a measure of the effects of ionizing radiation on humans). The DOE assumed that exposure to the maximally exposed involved worker at SRS would not exceed 0.8 rem per year due to engineering and administrative controls. Using this set of assumptions, the DOE estimated latent cancer fatalities. The probability that the average involved worker would develop a fatal cancer sometime during his/her lifetime as the result of a single year's exposure to waste management-generated radiation would be approximately 1.0×10–5, or 1 in 100,000. For the worker exposed to the administrative limit (0.8 rem), the probability of developing a fatal cancer sometime in his lifetime as a result of a single year's exposure would be 3.2×10–4, or approximately 3 in 10,000. For the total involved site workforce, the collective radiation dose could produce up to 0.022 additional fatal cancers as the result of a single year's exposure; over the 30-year period, the report summarizes that the involved workers could have 0.65 additional fatal cancer as a result of exposure.
The preferred alternative presented in the EIS implies the operation of the CIF, vitrification facilities, a mixed and hazardous waste containment building, the mobile soil sort facility, compaction facilities, and the transuranic waste characterization/certification facility. Emissions from these facilities would slightly increase potential adverse human health impacts compared with the no-action alternative for the three waste forecasts. The report examined potential releases and controls to estimate quantities of radionuclides released by each process.
Adding across these activities, the DOE estimated a dose of 0.037 rem per year, which is below the SRS administrative guideline of 0.8 rem per year. The probabilities and projected numbers of fatal cancers from 30 years of waste management operations for the preferred alternative leads to an individual worker to have a 1 in 44,000 probability of developing a fatal cancer due to exposure to SRS waste management activities. Given a workforce of 2154 workers (see above), the estimate is one additional fatal cancer from the 30 years of waste management activities considered in this EIS.
The estimate for the minimum waste forecast is almost identical. The larger amount of handling associated with the maximum waste forecast leads to a higher probability of a worker contracting a fatal cancer as the result of a 30-year occupational exposure to radiation. But “higher” is not much higher. The EIS estimates that two people in the workforce of 2501 could develop a fatal cancer sometime during their lifetimes as the result of a 30-year exposure.
The report places these estimates in context, noting that in the US, 23.5% of the population died of cancer during this period. This means that, if this percentage of deaths from cancer remained constant, 491 workers in a work-force of 2088 involved workers would normally be expected to die of cancer. This is not a false statement; however, it is not an entirely accurate statement because it fails to take into account the “healthy worker” effect, which shows that a smaller percentage of employed people would be expected to die from cancer than the general population, because the disabled and ill are less likely to be employed. Also, comparisons of this type often offend people who reject the idea of accepting any additional cancer burden that could be avoided. In essence, however, the estimates imply that that worker cancer effects would be undetectable.
Recapitulating, the limited treatment alternative requires more disposal capacity and facilities, coupled with more sophisticated methods of containment (more vaults and less shallow land disposal), because this option does not reduce or immobilize wastes to the extent of the other options. Yet the other options also require many of the same facilities, but fewer, with different emphases. The alternative involving extensive treatment would produce higher operations-related impacts than those in the alternative involving limited treatment, because more handling and processing of wastes generally produces more emissions and greater worker exposure. The moderate treatment alternative is a hybrid that uses options from the limited and extensive options, and produces impacts that fall between the two. The no-action alternative would require more storage facilities than the other three alternatives. Mixed and transuranic wastes would not be treated or disposed of during the three-decade period. This policy would increase the likelihood of health and environmental impacts, including accidents and worker radiological exposure, compared with the other alternatives. In essence, assuming no new technologies, the impacts would be deferred under the no-action alternative. Overall, impacts are small for each of the alternatives because the site is remote, and wastes are already heavily protected by multi-billion dollar engineering structures and systems and a well-paid and dedicated labor force.
Public reactions
Public participation at this site is different than one would find in response to the overwhelming majority of EISs. Nine of the DOE's largest sites have site-specific advisory boards (SSABs), including Savannah River (Office of Environmental Management 2010a,b). The boards provide the Assistant Secretary for Environmental Management and the site managers with advice, information and recommendations about waste management, technology use, and site restoration options. The DOE views the SSABs as official public representatives and as a way of building public trust.
Members are chosen with regional demographics in mind. As of June 2010, the SRS SSAB had twenty-five members, including eleven women and seven African Americans. The SRS full board has monthly meetings, and its committees meet as needed. All of their minutes are published on the DOE's website, which allowed the author to read those for 1995, which is when this EIS was finalized.
The waste management EIS was mentioned in the January 1995 meeting by Tom Heenan, former assistant manager for environmental quality at SRS, and again by him in the September meeting, when he stated that SRS had chosen a middle path. These were the only notations about this EIS. However, this does not mean that the SSAB was not interested. In fact, throughout the year, the minutes present discussions of almost every major element in part of the waste management EIS. These discussions include high-level waste, the DWPF, groundwater contamination, on-site land-use plans, transuranic waste, receipt of foreign spent fuel, clean-up of contaminated basins on the site, options regarding high-level waste tanks, transportation of nuclear materials through South Carolina, and permissible recreation on the site. At almost every meeting, the Board discussed the pressure on the site's environmental management budget and expressed their concern that the site was not receiving its appropriate share of the DOE environmental management budget.
At some meetings, DOE officials and citizens from the surrounding area praised the SRS SSAB for representing citizen interests. This is not to say that every meeting was free of friction and controversy. After several heated exchanges among members, for example, the Board discussed what they called etiquette. They also discussed if Board members were representative of the surrounding communities, and how many meetings could be missed before someone was asked to resign from the Board. The people who serve on this Board, and others that the author has visited, take their responsibility to the community seriously.
With regard to the process for this EIS, the DOE completed the draft EIS and the EPA published a notice of availability for the document in the Federal Register, which started the period of public comment on the draft EIS. Following standard practice, the DOE accepted and responded to letters, telephone messages, faxes, and presentations at public hearings near the site in Barnwell, South Carolina; Columbia, South Carolina; North Augusta, South Carolina; Savannah, Georgia; Beaufort, South Carolina; and Hilton Head, South Carolina. This is quite a few hearings within 100 miles of the site. Yet the DOE received only ten letters and heard statements from five people.
Remarks focused on the CIF; impacts of the wastes on pubic health; and public participation. Government comments primarily sought clarification and/or indicated no opposition to the EIS. The EPA endorsed the proposed action in its response and asked for more information. As previously noted (Chapter 1), this is a common response. William Lawless of the Savannah River SSAB posed several questions at the public meetings.
By reading the letters and documents, I found that that the public's concerns were related to this EIS, but were more focused on issues raised in other EISs about this site and the DOE complex. The DOE responded to all the points and indicated that some points were outside the scope of this EIS, and that they would forward the questions to the appropriate groups. For example, one concern was decomposition of organic materials present in low-level wastes. The testifier suggested that the incinerator ash be vitrified, and that buried contaminated metals be retrieved and processed by smelting before sale or reburial. The DOE respondent replied that these ideas would be consistent with the extensive treatment alternative, but that the EIS does not establish the level of restoration across the site. This, he noted, awaited an agreement between the DOE, EPA, and State of South Carolina.
Concern was expressed about the underlying science and validation of accident scenarios in the EIS. The DOE responded that they used the experiences of similar facilities elsewhere to estimate impacts, and they noted that independent peer review was employed to review their analysis of, for example, flood damage (the large tanks are underground).
Another commenter wanted more discussion of plutonium storage. The DOE responded that the transuranic wastes likely have some plutonium, and that plutonium storage per se was not part of the EIS. They then pointed to more than a half dozen other EIS documents that did discuss it.
Several stakeholders wanted to know more about the CIF and the proposed vitrification facilities. The DOE provided some additional discussion, including the legal context and EPA's and South Carolina's roles. As part of this interchange, a suggestion was made to recover what otherwise would be waste energy from the incineration facility. The DOE responded that it was not economically effective to do so and also would require a permit from the EPA, which would be difficult to secure.
Other questions were posed about the adequacy of monitoring and plans for reducing or increasing the site labor force. The DOE provided straightforward answers to these questions, which are consistent with what is presented in the EIS. In one case, the DOE noted that they inadvertently omitted an issue from the EIS; they then provided the data.
Compared with some other EISs, few of the remarks were made with much drama or responded to in that way. I was not at these meetings, so I cannot judge the expressions or body language. However, the transcript suggests intellectual exchanges rather than emotional confrontations.
Changes in the plans proposed in the 1995
waste management EIS
Circumstances change and require that plans captured in EISs be alterable. Each federal department and agency has developed processes for altering its plans. In the case of the EIS and the DOE, it can issue a supplement or a new EIS. The DOE prepares what it calls a “supplement analysis” to determine if a new EIS is required or no further NEPA review is needed. Here I illustrate this process for the SRS waste management EIS.
In May 1997, the US Department of Energy (1997) decided to take several additional steps to manage mixed low-level radioactive and transuranic wastes that it said were consistent with the preferred alternative described in EIS-0217. One step was to send elemental mercury and other mercury-contaminated low-level radioactive waste off-site for treatment. Residuals from the process were to be returned to SRS. Second, the supplement calls for vitrifying uranium chromium solutions and contaminated soils. And the supplement calls for building and operating two additional buildings to manage mixed and low-level radioactive waste and transuranic wastes. The five-page supplement assesses the impacts as small and states that any that would be found could be mitigated (ibid.).
This supplement is one of many actions across the DOE Complex that influences this 1995 waste management EIS. Indeed, SRS publishes “What's New” on its home page (www.srs.gov) that lists, among other things, current NEPA actions that affect SRS. A recent interesting supplement serves as a second illustration. In the 1960s, the DOE produced plutonium containing oxide materials at the SRS site, so-called “low-assay plutonium” or LAP material. It was shipped to the Hanford site for scientific analyses but never opened. The supplement calls for shipping twelve drums containing approximately five kilograms of LAP material by truck (three drums per truck) back to SRS. The material there would be stored and then treated in the DWPF.
Interview
Dr Kevin Brown is a research scientist in the Department of Civil and Environmental Engineering at Vanderbilt University, where he applies life-cycle and risk-analysis methods to radioactive waste management. I interviewed him on June 29, 2010. From 1985–2002, he worked at the Savannah River site for DuPont and Westinghouse on some of the more challenging radiological waste management problems in the world.
My questions focused on several of the projects described in the EIS and the overall concept behind a site-wide EIS. For context, he noted that SRS has no single-shell tanks per se and the SRS tanks are either double-shell or built in a teacup design, which means that leakage is into the teacup (annulus) and not into the environment. He could only think of hearing about one 10-gallon leak into the environment at SRS. This is in strong contrast to the Hanford site (Washington), which has a majority of single-shell tanks that do not have the teacup design, and hence has had a good deal of leakage to the environment. Furthermore, SRS had only two primary types of separation process and segregated high iron containing high-level nuclear waste from high-aluminum-containing wastes. This greatly simplifies the treatment process compared with a site such as Hanford, where there were many more separation processes and little attempt to segregate wastes.
Kevin Brown noted that the DWPF has been in operation for almost 15 years. (He was in the control room when it went hot.) About 10% of the waste volume goes to the DWPF, and this fraction accounts for about 55% of the curies and an even larger proportion of the long-lived radioactive elements. He feels that the operation has been smoother than he had anticipated. One major reason is that the materials retrieved from the tank have been more consistent than he anticipated, enhanced by the blending scheme that helps achieve a degree of consistency. This waste is then mixed with glass-forming frit and vitrified. The canisters of vitrified waste (see above for a description) are stored on-site in a special building. He praised the site for continuously monitoring the process, and observed that the production of the canisters was not optimized, because site personnel were concerned about changing a process that was working well and was not causing human health or environmental impacts.
The remaining 90% of the volume and about 45% of curies will be grouted (saltstone) and store permanently on site in concrete vaults. So far, the grout vaults have been effective in containing the waste. However, Kevin Brown suggested that 15–20 years is a short time to judge the ultimate success of this kind of waste management project and more information is needed. In fact, some of his current research focuses on the stability of the engineered system comprising the grout and vault. Overall, he believes that the EIS did a good job of capturing the important details of the DWPF and grout facilities.
The CIF was built in 1995 at a cost of $102 million and tested in 1997. The CIF could not efficiently incinerate the combustible wastes, which included radioactive elements. At an operating cost of $20 million per year, the site judged the facility not cost-effective for the limited feed material available, and instead site management found alternative management methods. In other words, the CIF was not used as had been anticipated in the EIS.
Transuranic wastes were another issue in this EIS. Transuranic waste has been going to WIPP in New Mexico. Indeed, the SRS Citizens Advisory Board congratulated the DOE on its 1000th shipment of transuranic wastes to WIPP. It also praised the DOE for examining options to treat more radioactive transuranic wastes on-site or ship them to WIPP.
The more general question is: What is the utility of this kind of site-wide EIS? Kevin Brown thinks this type of EIS is unusual; there are few that try to be so comprehensive. They are valuable to give a snapshot of management's thinking, but management have the option of changing their mind about individual projects, and circumstances change, which may mean that an EIS like this one does not even include all the plausible options for treating these wastes. In a sense, this kind of EIS is a site-wide scorecard that can be used to assess a comprehensive snapshot of site management thinking.
Evaluation of the five questions
Information
There is a massive amount of information in this document, so much that it is easy to lose track of its purpose. The focus, according to DOE, was the choice between no action and three levels of action regarding nuclear and some non-nuclear wastes. The choice the DOE made is risk-based, namely, to concentrate its resources on the most hazardous materials. It decided that this option would reduce worker exposures, compared with the more aggressive option, and manage the wastes more effectively compared with the least aggressive option.
Unfortunately, that message is buried in a sea of information. The quality and quantity of information brought to bear on the implications of this decision is impressive. Unfortunately, it is hard to access – often key numbers are in tables and/or graphs that are supposed to help explain the results. Sometimes the table and graphs are helpful, other times not. Some of the most important material is in even less scrutable appendices and graphs.
The most important shortcoming is the absence of a framework to support the choice of the chosen alternative. It would have been helpful to have had this decision grounded in basic risk-assessment and risk-management principles. The document addresses accidents, ecological impacts, and so on in encyclopedic detail. But it doesn't have an introductory section that describes the principles that it is silently evoking (Kaplan and Garrick 1981; Berlin and Stanton 1989):
Some sections contain the information to address these questions, for example, worker cancer risks (see above). But the discussion is so terse, and so unconnected to anything else about public or ecological health, that this reader had no idea how seriously this impact was compared with injuries from falls, accidents, and so on. Because the DOE's decision has consequences for risk prevention and cost, it merited an unequivocal effort to directly link risk events, risk likelihood, and consequences (Berlin and Stanton 1989).
More than any other project in this book, this EIS and this genre of EIS concern multiple projects. Because there is no framework, the reader is left to struggle with the relative significance of these different projects (DWPF, incinerator, trenches, vaults). It also seriously understates the importance of the maximum waste forecast, which needed to be unequivocally related to degree of site clean-up and future land-use options. The volume has too many dangling facts, and not enough integration and interpretation of the risk-related principles behind the choice of those facts. Overall, the tone and writing are acceptable; that is, the words are clear. But the message about the waste forecast and its relationship to key policy decisions is lacking.
On the technical side, I would have liked more discussion of the issue of uncertainty in modeling some of the impacts that were discussed, especially groundwater and air. However, this would be for the technically inclined.
Comprehensiveness
This document is so comprehensive that it borders on the incomprehensible. There are more than two dozen waste management projects on this site, each attached to one or more other EISs. Decisions about future land use, degree of clean-up, and off-site decisions about transuranic and other waste forms add another layer of complexity. The authors tried hard to discuss everything, and in the process, as noted above, missed an opportunity to integrate using a widely recognized risk-analysis framework. I have no objections to reading an almanac or encyclopedia, but I do when the document is supposed to be a comprehensive summary supporting a key decision. Even if they could not, or chose not to, use an overall framework, they never made it clear what these projects had to do with each other. Kevin Brown, who is intimately familiar with the site, knows many of these connections. Having them in the report would have helped.
In other chapters, I have critiqued the economic and social analyses for being superficial. In this case, I found the economic and social impacts to be the most carefully presented in the document. A reader can understand how many jobs might be created by each of the different management options and waste amount options, how the management options were associated with jobs and income, and where workers would likely come from. In contrast, the water resources discussion (see the example above) and the accident scenarios were truly a challenge. I had to read each multiple times, prepare a table for myself to keep track of the numbers and assumptions, go to appendices, and then thumb back and forth because the discussions are separated from each other in the text.
Coordination
The cooperating federal agencies were involved in the process, and their comments were not particularly challenging, leading this author to conclude that either the other agencies had had discussions with DOE before writing their comments and/or the EIS was mostly about issues that they had already addressed as part of their response to other EISs. That is, the key decision was for them perhaps a foregone conclusion.
Accessibility to other stakeholders
Even in that pre-web-communication era, there were more than enough opportunities for public participation. However, the results were not what would be expected from facilities that handle highly toxic materials. The SSAB did not spend much time on this EIS (assuming that the minutes are a reflection of their concerns). Fewer than a dozen citizens testified at the public hearings. The limited amount of testimony is misleading because this document is much more of a recitation of multiple projects than it is a focus on any single decision. The SSAB had discussed nearly every one of these projects (e.g. DWPF, incinerator, groundwater), often many times. The public had heard about each of the individual projects on multiple occasions, often for many years. There was not much new for them in this EIS.
The “new” information is the choice of the mid-level treatment option and three waste levels. The only striking finding is the difference between the maximum waste impact and the other options. The report makes no attempt to hide the information. Indeed, the assumptions are described and illustrated with interesting and useful graphics. The maximum waste forecast does have implications for the future use discussions that the SSAB had been discussing, and yet I find no evidence of a strong reaction in the public comments. Perhaps this lack of reaction is due to the fact that even the maximum waste forecast uses only a small amount of land and confines almost all of the waste management actions to the central developed area. I found it fascinating that Tom Heenan of DOE mentioned this EIS on several occasions, and, assuming the minutes are a reflection of reality, there were no probing questions from the SSAB.
An EIS like this one, not built around a hot-button issue, nevertheless needs to be completed and is quite costly. Hence I believe that web-based and visualization technologies should be employed to make these kinds of EIS come to life for nongovernmental organizations, citizens, and the media. In Chapter 8, I formally suggest this option. Here, I illustrate it with a discussion of the DWPF. In 1982, President Carter was concerned about nuclear weapons-grade materials, and he moved the United States away from a leadership role in converting and reusing weapons-grade nuclear materials for fuel. The SRS DWPF is the most obvious implication of that decision. The SRS vitrification is a 42,000-square-foot, $2 billion-plus complex that blends aqueous nuclear waste with a borosilicate glass-forming compound, which then is poured into a stainless steel container to form massive canisters approximately 10 feet high, with a 2-foot diameter and weighing about 4000 pounds. These “logs” are stored on-site (US Department of Energy 1994).
The conversion of the waste material from liquid to a vitrified form is an important risk-based decision because it all but eliminates the possibility of the migration of the radioactive material. The DWPF has its own EISs. But it is mentioned multiple times in this one. The author can envision several drawings of the DWPF, the canisters it produces, their storage on the site, and maps that show where they are located relative to the concrete vaults where the less curie-laden radioactive waste is managed. A second set of graphical overlays could be made of the incinerator, and again its location shown on a map. Ultimately, a cumulative map would show the concentration of activities in the developed part of the site, and be accompanied by graphs and discussion that show the risk avoided by these choices. The cumulative maps of the three options and no-action option could be displayed against the backdrop of the associated risks.
Fate without an EIS
I can't imagine that the DOE's decision was changed by this EIS. The DOE could not afford to cease work at the site because its legal agreements compel it to move forward and reach milestones. Furthermore, the region was economically distressed during this time, and the jobs and economic benefits are valued in this area. Yet there are likely more short-term health impacts on workers associated with the most aggressive waste management option. Furthermore, the cost (not reported) of the aggressive option likely was prohibitive.