NUCLEAR CLEANUP THE STANDARDS CONFLICT BY DANIEL HIRSCH PRESIDENT AND EMILY CHURG AND TONY ZEPEDA RESEARCH ASSISTANTS COMMITTEE TO BRIDGE THE GAP NOVEMBER 2004 SUPPORTED BY A GRANT FROM THE CITIZENS’ MONITORING AND TECHNICAL ASSESSMENT FUND ABSTRACT The U.S. Department of Energy has recently violated a longstanding Joint DOE-EPA Policy which commits DOE to clean up all its nuclear facilities nationwide to the Environmental Protection Agency (EPA) Superfund (CERCLA) standards. The focal point of this conflict between DOE and EPA cleanup standards is the Santa Susana Field Laboratory (SSFL), a 2800- acre facility on the Los Angeles-Ventura County line in Southern California. Ten reactors, a plutonium fuel facility, and a “hot laboratory” for cutting up irradiated nuclear fuel were operated at the facility, which opened in the 1940s when it was remote from populated areas. Now large numbers of people live nearby. One of the reactors suffered a partial meltdown in 1959; two others experienced damage in 1964 and 1969 to 80% and 35% of their fuel, respectively. In 1989, DOE found widespread chemical and radioactive contamination at the site, and a cleanup program commenced. In 1995, DOE and EPA entered into a Joint Policy to assure that all DOE sites, whether or not they were on the National Priority List (Superfund), would be cleaned up consistent with EPA’s CERCLA standards. In March of 2003, DOE reversed course and, while claiming to still follow the 1995 Policy, announced it would not clean the site up to the EPA standards. It would remove only 1% of the contaminated soil and then release the site for unrestricted residential use. In December 2003, EPA issued findings that the site was not being cleaned up consistent with the 1995 Joint Policy and that under the circumstances, so much radioactivity could be left in place that residential use would be unsafe and the only safe use would be restricted dayhikes with limitations on picnicking. To date no detailed study has been done comparing the DOE and EPA cleanup standards. This report, supported by a grant from the Citizens’ Monitoring and Technical Assistance Fund, performs that analysis. The evaluation demonstrates that DOE’s decision not to comply with EPA’s cleanup standards will result in radionuclide concentrations being permitted that are hundreds, thousands, tens of thousands, and in some cases, hundreds of thousands of times higher than EPA’s primary cleanup goals. For most radionuclides, the associated risk exceeds even the uppermost permissible risk level of EPA under CERCLA. In some cases, those cancer risks rise to levels on the order of one cancer per ten people exposed, using the federal government’s official radiation risk figures. Should DOE proceed with acting in contravention of the 1995 DOE-EPA Joint Policy on cleaning up DOE sites consistent with CERCLA at SSFL, there would be significant ramifications for the DOE nuclear complex nationwide, and for public health. i NUCLEAR CLEANUP: THE STANDARDS CONFLICT Introduction The Department of Energy (DOE) nuclear complex consists of approximately one hundred sites across the country, many of which face daunting cleanup challenges because of extensive radioactive and chemical contamination of soil, surface water and groundwater. Decades of accidents, spills, and other releases have left behind a legacy of pollution which poses significant risks if not adequately remediated. All DOE nuclear sites are supposed to be cleaned up in accordance with the standards and procedures of the Environmental Protection Agency (EPA) under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA), commonly known as Superfund. DOE facilities are brought under CERCLA in two primary ways. Some DOE sites are on the National Priority List (NPL), i.e., they are Superfund sites, and automatically must comply with CERCLA requirements. All other DOE sites must also be cleaned up consistent with EPA’s CERCLA guidance, as a result of a 1995 Joint DOE-EPA “Policy on 1 Decommissioning Department of Energy Facilities Under CERCLA.” 1 The Joint Policy is included herein as an attachment. That Joint Policy requires that all DOE sites, irrespective of whether they are on the NPL, shall be cleaned up consistent with EPA’s CERCLA standards, guidance, procedures, and methods. This includes the use of EPA’s Preliminary Remediation Goals (PRGs) as the starting point for determining acceptable levels of contaminants that can remain after decommissioning the facilities, methods of calculating acceptable contaminant levels in a site-specific way in terms of appropriate inputs to models, and public participation procedures. Recently, DOE, while still claiming to follow the 1995 Policy, has begun to resist complying with its requirement to be consistent with EPA’s CERCLA standards. The facility where this conflict between DOE and EPA standards is being fought out most clearly is the Santa Susana Field Laboratory (SSFL), located on the boundary between Los Angeles and Ventura Counties in Southern California. This report examines the conflict between the DOE and EPA cleanup standards, focusing on SSFL as a case study, with significant implications for the DOE nuclear complex nationwide. SSFL Founded in the 1940s as a government facility for development and testing of nuclear reactors, in addition to testing rockets, SSFL was designed as a remote field laboratory to conduct work too dangerous to be performed in more populated areas. However, in the more than half a century since it was founded, Southern California’s population has mushroomed, and large numbers of people now live within a few miles of the site. 2800 acres, SSFL is situated on top of the Simi hills, overlooking Simi Valley to the north, Chatsworth, Canoga Park, and West 2 Hills to the east, Agoura, Calabasas and Woodland Hills to the south, and Thousand Oaks to the west. Over the years, approximately ten nuclear reactors operated at SSFL, in addition to several “critical facilities,” a plutonium fuel fabrication facility, a uranium carbide fuel fabrication facility, and a “hot lab” in which irradiated fuel shipped in from around the country was decladded and examined. The reactors had no containment structures, i.e., were built without the large concrete domes surrounding modern power reactors. Numerous accidents occurred at the site. In July 1959, the Sodium Reactor Experiment (SRE) suffered a power excursion, in which power rose out of control. With significant effort, the reactor was shut down; but, inexplicably, a few hours later it was started up again, without having been able to determine the cause of the incident. The reactor continued to operate for several more weeks, with high radiation readings and other signs of problems, until it was shut down at the end of the month. At that point, the reactor operators discovered that a significant fraction of the fuel had suffered melting. Tetralin, a coolant used for the pump seals, had leaked into the sodium coolant of the reactor; carbonaceous material formed, blocking the coolant channels, causing the fuel to overheat and melt. Approximately one-third of the fuel experienced melting. A photograph of some of the melted fuel is found on the next page. 3 4 Radioactive gases were released from the reactor into holding tanks and then bled into the atmosphere over a period of weeks. The extent of the radioactive releases remains uncertain to this date. Some monitors went off scale; few measurements of the sodium coolant were taken, and these were contradictory; and the ratios of volatile radionuclides found in the coolant suggest significant releases from the coolant to the environment may have occurred. In 1964 and again in 1969, two other reactors suffered significant core damage. The System for Nuclear Auxiliary Power (SNAP) 8ER reactor operated for a year without its operators realizing the fuel was cracking. After shutdown, it was determined that 80% of the fuel had cracked. A few years later, the same type of accident occurred with the SNAP 8DR, with about a third of its fuel suffering damage. The hot lab suffered fires resulting in spread of contamination. The sodium burn pit, an open-air pit for cleaning sodium-contaminated components, got contaminated when radioactively- and chemically-contaminated items were burned there, in contravention of safety requirements. Other spills and releases occurred as well over the decades of operations. In 1989, a DOE investigation found widespread chemical and radioactive contamination on the property. Widely publicized in the local press, the revelations led to substantial concern among community members and elected officials, resulting in a challenge to and subsequent shutdown of continued nuclear activity at the site. Cleanup commenced, and EPA was brought in at the request of local legislators to provide oversight. 5 In 1995, DOE and EPA entered into the Joint Policy referenced above, committing that SSFL, and all other DOE sites, would be cleaned up consistent with EPA’s CERCLA standards. In March 2003, however, DOE issued a final Environmental Assessment (EA) and Finding of No Significant Impact (FONSI), which set cleanup standards for SSFL that are at great variance from EPA’s cleanup requirements. The EA conceded that there are approximately 405,000 cubic meters of soil contaminated above EPA’s primary cleanup standard, but then chose as its preferred alternative to clean up only approximately 5000 cubic meters, intending to leave behind 99% of the soil it concedes is contaminated, and then release the land for unrestricted residential use. In August 2003, the Senate Appropriations Committee issued a report on the Energy and Water Appropriations, urging DOE to live up to its commitments in the 1995 Joint Policy and clean up SSFL to EPA’s CERCLA standards. Shortly thereafter, DOE responded to the Senate, claiming it was in fact consistent with both the Joint Policy and EPA’s CERCLA standards. In December 2003, however, EPA issued formal findings that the cleanup was not consistent with CERCLA, that sufficient contamination would remain such that unrestricted residential use would not be appropriate, and that the only safe use under the circumstances would be restricted day hikes with limitations on picnicking. A significant conflict thus exists between DOE and EPA standards for cleaning up DOE sites such as SSFL. We will now examine the differences. 6 The EPA CERCLA Standards EPA cleanups are based on risk, not dose. EPA’s primary remediation goals are set so as -6 to not leave behind residual contamination sufficient to cause a more than one-in-a-million (10 ) -6 risk to an individual. If there are acceptable reasons why one cannot meet the 10 goal, one can request to be able to fall back from that cleanup goal somewhat, if one can show one meets the nine balancing criteria in CERCLA. One is then only permitted to increase the risk level the minimum amount necessary, and in no case to a risk in excess of approximately one in ten- -4 thousand (10 ). The amount of a contaminant that will produce a certain risk is determined by use of various EPA models and input assumptions about exposure scenarios. For example, EPA sets acceptable inputs for such things as breathing rates, soil resuspension, backyard garden vegetable usage, and so on, which can then be adjusted for site-specific factors. EPA bases these calculations on the exposure scenario that would produce the highest risk and which is reasonably foreseeable. For example, if a site is zoned rural residential, then the rural residential pathway is employed if it produces the highest risk for the same amount of contaminant in soil. All contaminants – chemical and radioactive – must be summed and the collective risk must meet the overall risk goal. For example, if a site has both chemical and radioactive 7 contamination, it would be impermissible under EPA guidance to ignore the chemical contamination and set radioactive cleanup goals at the upper limit of the risk range. EPA has published Preliminary Remediation Goals (PRGs) which set default concentrations of scores of radionuclides in environmental media such as soil for various 2 exposure scenarios. If there are strong site-specific reasons, one can alter the inputs into the model with the use of EPA’s PRG calculator. The DOE SSFL Cleanup Standards DOE, despite its commitment in the 1995 Joint Policy to follow EPA’s CERCLA guidance, has instead adopted cleanup standards that are in general orders of magnitude more lax than EPA’s PRGs. 3 DOE has set cleanup standards for approximately twenty-five radionuclides. It has done so based on its model, RESRAD, and its inputs thereto, rather than EPA’s PRGs and PRG calculator. Furthermore, DOE has restricted its model to the suburban residential exposure pathway, rather than the rural residential pathway for which the site is currently zoned and which produces higher doses for the same concentration of radioactivity. EPA’s CERCLA guidance 2 The most recent EPA PRGs for radionuclides, dated August 2004, are included here as an attachment. The concentrations are in pico-curies of radioactivity per gram of soil (pCi/g). The column marked “Residential Soil” is for the suburban residential scenario; “Agricultural Soil” is for the rural residential scenario. 3 See attached table, identifying the radionuclides, their symbols, and half-lives. 8