Robert J. Halstead
Technical studies sponsored by DOE and NRC (Refs. 1,2, 3, 4, 5) have identified three aspects of spent nuclear fuel(SNF) and high-level radioactive waste(HLW) transportation which could result in increased radiation exposures to transportation workers, members of the general public, and emergency response personnel: 1) during routine transportation operations, gamma and neutron radiation are continuously emitted through the cask walls; 2) a severe transportation accident could damage the cask radiation shielding, resulting in elevated gamma and neutron radiation levels around the damaged cask, and possibly release some portion of cask contents, resulting in contamination of a relatively large down-wind area with alpha-, beta,- and gamma- emitting isotopes; and 3) a terrorist attack using high energy explosives could breach a cask and disperse a portion of its contents, resulting in elevated gamma and neutron radiation from the damaged cask, and contamination of the nearby area with alpha-, beta-, and gamma- emitting isotopes.
Nevada Agency for Nuclear Projects staff and contractors have reviewed these DOE- and NRC-sponsored research reports as part of the Agency's overall assessment of nuclear waste transportation risks and impacts.(Refs. 6, 7, 8, 9, 10, 11) These studies are not sufficient for the assessment of potential health effects that must be addressed in the Yucca Mountain repository site environmental impact statement(EIS). As part of the EIS process, DOE must address potential radiological health effects for transportation along specific rail and highway routes likely to be used for shipments to a repository at Yucca Mountain and/or an interim storage facility at NTS. Nonetheless, the reports cited in the following discussion do at least provide a starting point for identification of the kinds of radiation exposures which could occur during SNF and HLW transportation, and suggest the kinds of baseline community health information which will need to be collected before, during, and after repository operations in the event that the Yucca Mountain project proceeds.
RADIATION EXPOSURES AND HEALTH EFFECTS
In the United States, members of the general public annually receive an average background radiation dose of about 360 millirem(mrem) from a combination of natural and man-made sources. The primary natural source, radon gas, contributes about 200 mrem, to the average annual dose equivalent, and medical x-rays, the primary man-made source, contribute about 40 mrem. (A typical chest x-ray results in a 10 mrem dose.) For purposes of this analysis, residents of Nevada are assumed to receive the national average annual dose.
The U.S. Nuclear Regulatory Commission(NRC) has established dose limits for workers exposed to radiation as part of their jobs, and these limits have been adopted (where not otherwise required) by the U.S. Department of Energy(DOE). The annual dose limit for radiological workers overall is 5.0 rem (5,000 mrem) whole body, and for declared pregnant workers, 0.5 rem(500 mrem) to the unborn child (embryo/fetus) over the nine-month gestation period. The average annual dose to the general public from nuclear industry activities is limited to 0.1 rem (100 mrem).
The NRC limits reflect the prevailing assumption among government (and industry and many academic) technical authorities that an individual must receive a whole-body dose of about 25,000 mrem (15,000 mrem for a pregnant woman) before there is a significant increase in the risk of serious human health effects, and a dose of about 500,000 mrem (500 rem) before probable death as a result of radiological health effects. On the other hand, government regulations also require that NRC licensees and DOE contractors follow the radiation control concept known as ALARA (As Low As Reasonably Achievable). The ALARA objective is to attain worker and public doses as far below the applicable limits as reasonably achievable given social, technical, economic and policy considerations. The ALARA concept recognizes the uncertainties associated with the risk of low level exposure to ionizing radiation. It should also be remembered that there is considerable technical controversy about the individual health effects of any additional exposures beyond background levels, and mythological debate about the use of collective population-doses to calculate latent cancer fatalities among the general population in probabilistic risk assessments.
EXPOSURES AND DOSES RESULTING FROM ROUTINE TRANSPORTATION OPERATIONS
This analysis of exposures resulting from routine(non-accident) transportation focuses on radiation doses received by: (1) workers conducting safety inspections of casks and vehicles; (2) individuals residing, working, or institutionally confined at locations near shipping routes; and (3) drivers and passengers of vehicles in traffic gridlock incidents who may be stranded for an extended period of time very near an undamaged shipping cask. Appendix A provides a partial listing of other circumstances in which workers and/or members of the public may receive significant radiation exposures during routine transportation operations
The Sandquist report( Ref.1) evaluated exposures to the public and workers from routine shipments of truck and rail casks containing five-year-old, medium-to-high burn up SNF. Specific fuel characteristics, cask designs, and cask capacities are less important for estimating routine exposures than the emission rate allowed under NRC regulations, 10 mrem/hour at 2 meters from the cask surface. Cask designs being developed for shipments to a repository assume the 10 mrem/hour emission rate. ( DOE considered and rejected the idea of limiting new cask design emissions to 2 mrem/hour at 2 meters, which would have cut the payload for the new truck casks in-half.) Using the PATHRAE model to estimate exposure rates (in microrem/minute) at various distances (in meters) from the cask center, Sandquist specified exposure times(in minutes) and distances(in meters) for various events (such as slow transit through residential areas), and calculated maximum individual exposures(in millirems) per event.
Sandquist correctly cautions that: "These exposures should not be multiplied by the expected number of shipments to a repository in an attempt to calculate total exposures to an individual; the same person would probably not be exposed for every shipment, nor would these maximum exposure circumstances necessarily arise during every shipment." With appropriate modifications and qualifications, however, Sandquist's approach and some of Sandquist's assumptions can be used to calculate cumulative exposures for certain individuals and groups of exposed individuals.
Worker Exposures Due to Safety Inspections of Casks and Vehicles.
Workers responsible for safety inspections of SNF and HLW shipments could receive yearly occupational doses significantly in excess of annual background doses. Sandquist assumed each truck cask inspection would take about 12 minutes , at a distance of 3 meters from the cask center (near the personnel barrier), and result in a dose of 2 mrem per event. Inspections of truck casks entering Nevada will likely require 45 - 75 minutes, based on actual experience in other western states with the more rigorous inspection protocols developed by the Commercial Vehicle Safety Alliance(CVSA), and may also involve swipe sampling inside the personnel barrier to determine cask surface contamination levels. Rigorous mechanical and radiological safety inspections at Nevada ports of entry could very well result in an average dose of 10 mrem per person per truck cask arrival. An inspector who conducted two truck inspections per week could receive a cumulative annual dose ranging from 200 to 1,000 mrem. At one inspection per day, 5 days a week, an inspector could receive an annual dose of up to 2,500 mrem.
Exposures to Members of the Public Residing, Working, or Institutionally Confined at Locations Near Shipping Routes.
Individuals who reside, work, or are institutionally confined at certain locations within 6 to 40 meters (20 to 130 feet) of a nuclear waste highway route, or within 6 to 50 meters (20 to 160 feet) of a nuclear waste rail route, could potentially receive yearly radiation doses equal to a significant percentage of, or even in excess of, average annual background doses. Such exposures could occur under circumstances where: (1) residences, workplaces, or certain institutions (especially schools, prisons, or long-term health care or retirement facilities) are located near route features or segments which would require nuclear waste trucks or trains to stop and start again, or travel at very slow speed; (2) the number of shipments is high enough, one to several casks per day, that opportunities for exposures occur frequently at the same locations, and (3) the individuals residing, working, or confined at near-route locations are regularly present to be exposed to a significant portion (if not all) of the shipments which occur annually.
Based on route-specific impact studies conducted by Agency contractors and personnel, there is a high probability that all three circumstances exist along some of the routes likely to be used for shipments to a repository at Yucca Mountain or to an interim storage facility at the Nevada Test Site. Legal-weight truck (LWT) routes of special concern would include US 95 from the I-15 interchange in downtown Las Vegas to Lathrop Wells, and the so-called NDOT B Route, US 93A, US 93, US 6, and US 95 from West Wendover to Lathrop Wells (especially where vehicle stops and/ or left turns are required in West Wendover, McGill, Ely, Tonopah, Goldfield, and Beatty).These routes could carry between 600 and 2,700 truck casks per year. Rail route locations of particular concern would potentially include areas in Jean, Arden, Las Vegas, North Las Vegas, Moapa, and Caliente along the Union Pacific mainline from Salt Lake City to Los Angeles. These routes could carry between 300 and 500 rail casks per year. Heavy haul truck (HHT) route segments of special concern would include US 93 west from Caliente to Oak Springs Summit, State Route 375 through Rachel, US 6 intersection with US 95 in Tonopah, and US 95 through Tonopah, Goldfield, and Beatty. This route could carry an average of 500 - 600 slow-moving HHT shipments per year. (It is also possible that HHT shipments could be routed through North Las Vegas or Las Vegas). (Ref. 12.)
Using exposure rates (in microrem/minute) generated by the the PATHRAE model, Sandquist specified exposure times(in minutes) and distances(in meters) for routine transportation events (such as slow transit through residential areas and areas with pedestrians, truck stops for driver's rest and refueling), and calculated maximum individual exposures(in millirems) per event. Although Sandquist cautioned against using these exposures to calculate 30 year cumulative doses, these exposures, when appropriately qualified, can be used to estimate maximum potential annual doses to individuals near truck cask shipping routes, as follows:
Distance from cask center 6 m/20 ft 10 m/33 ft 15 m/49 ft 40 m/131 ft
Dose Rate(microrem/min.) 70 40 20 6
Maximum Dose, 6 min. exposure(mrem) 0.4 0.2 0.1 0.04
Maximum Dose, 2 min. exposure(mrem) 0.14 0.08 0.04 0.01
Max. Annual Ind. Dose, 600 trucks(mrem) 84-240 48-120 24-60 6-24
Max. Annual Ind. Dose, 2,400 trucks(mrem) 336-960 192-480 96-240 24-96
It is possible that there are locations along highway routes in Nevada where exposure times could average 6 minutes per truck shipment. It is likely that there are locations where exposure times could average 2 minutes per truck shipment ( for example, major intersections along the NDOT B Route in West Wendover, Ely, and Tonopah). Depending upon the number of truck shipments and distance from the route, maximally exposed individuals near highway routes could potentially receive annual doses ranging from 6 mrem to 960 mrem, equivalent to 2% to 266% of the average annual background radiation dose. Further study of route-specific details is necessary for more precise dose estimates.
Estimation of exposures from rail transportation is more difficult, primarily because of uncertainties about service options (dedicated trains versus general freight service), number of casks per shipment, and continuous rail shipment or intermodal transfer to HHT. At various times, DOE has considered locations in Jean, Arden, Las Vegas, North Las Vegas, and Caliente for rail spur origination and/or rail cask transfers. Maximally exposed individuals located within 20 meters (66 feet) of rail interchange/transfer points could potentially receive annual doses in the range of 150 mrem, assuming 500 rail cask/shipments per year and an average exposure time of 10 minutes per rail cask received. Further study is necessary for more precise dose estimates.
Exposures to Occupants of Vehicles Trapped in Traffic Gridlock Incidents Near an Undamaged Shipping Cask.
Drivers and passengers of vehicles in traffic gridlock incidents could receive potentially significant radiation doses as a result of being trapped next to or near an undamaged truck cask for an extended period of time. Sandquist evaluated such events, and concluded that occupants of stopped vehicles in lanes adjacent to the cask vehicle could receive a maximum dose of 3 mrem, assuming a distance of 5 meters from the cask center and an exposure time of 30 minutes. In response to inquiries from the U.S. Nuclear Waste Technical Review Board(NWTRB), DOE personnel in 1990 prepared an analysis which concluded that the maximum dose from a gridlock incident could be as high as 40 mrem. DOE provided the following analysis to the NWTRB:
Maximum Exposure of Critical Group in Gridlock
- Group located 1m from vertical plane of trailer
- 4-8 people in vehicles closest to trailer
- Gridlock lasts 2-4 hours
- No remedial action to move group members
- Exposure rate to group, 5-10 mrem/hr
- Exposure to group member, 10-40 mrem
- Exposure would be 2-8% of IAEA annual public dose equivalent limits
Source: Ref. 13
The risks associated with gridlock incidents involving SNF shipments, and gridlock risk reduction strategies, have received little serious study and many questions relative to health effects remain unanswered. For example:
Implications for Community Health Studies.
Routine transportation operations could result in a variety of whole body exposures to gamma and neutron radiation:
(1) Vehicle inspectors could receive 100 to 250 or more exposures equivalent to medical x-rays (10 mrem or more) per year, resulting in cumulative annual doses of 1,000 to 2,500 mrem, equal to as much as 50% of the occupational dose allowable for nuclear industry workers. The same inspectors could conceivably continue to receive the same annual dose for a period of 10, 20, or 30 years or even longer.
(2) Persons who reside (or work or attend school or are institutionalized) at certain locations very near nuclear waste highway routes could under certain circumstances receive hundreds or even thousands of very low-level exposures (ranging from 0.01 mrem to 0.4 mrem per shipment) per year, resulting in cumulative annual doses of 6 mrem to 960 mrem, equal to 10% to 250% of average annual background radiation dose. These maximally exposed individuals could theoretically continue to receive the same annual dose for a period of 10, 20, or 30 years or even longer.
(3) Persons caught in gridlock incidents involving SNF shipments could receive up to 40 mrem, a dose equivalent to four medical x-rays, over a period of 4 hours. It is unlikely (but certainly not impossible) that an individual would be caught in a gridlock incident involving SNF more than once.
These potential exposures and the resulting doses are lower than the thresholds usually considered to cause a high probability of adverse health effects. However, the potential doses are high enough relative to normal background doses to justify consideration in planning for data collection as a part of community health studies. The following should be considered:
(a) Vehicle inspectors should have complete medical examinations before beginning work on SNF shipments, and should be reexamined annually. Monitoring of white blood cell and platelet counts would be particularly important, although one would not necessarily expect to see impacts at exposures less than 10,000 mrem.
(b) Vehicle inspectors should be equipped with personal dosimeters, and actual doses should be monitored collectively and individually, probably at least monthly, depending upon the number of casks inspected.
(c) Cask inspection records should be carefully monitored, and actual emission levels should be tracked in aggregate, by cask type, and by individual cask.
(d) An ALARA assessment should develop exposure reduction strategies for vehicle and cask inspections (e.g., construction of specialized inspection bays at ports of entry, redesign hand-held instrumentation, and use of remote instrumentation and/or robots)
(e) All potential rail and highway routes should be surveyed to identify locations where the dose per shipment is likely to exceed some predetermined level (for example, 0.04 mrem/shipment, the calculated dose for a 2 minute exposure at 15 meters distance from the cask center). The traffic flow rates, demographics, building types, etc. at these locations should then be evaluated to determine the potential for actual exposures. The potential for human exposures could be much less than suggested, or it could be greater. This is one potential impact area where intuition should not be relied upon.
(f) Depending upon the conclusions of the above route survey (item e above), it may be useful to collect baseline data on cancers and genetic disorders in any corridor communities which appear to have potential for high exposures (for example, if an elementary school classroom is found to be located within 15 to 30 meters of a traffic light or stop or yield sign along a primary shipment route).
(g) A system of fixed radiation monitors should be installed at various locations along shipment routes, at various distances, to record actual exposure rates. Preferably this should be done before shipments begin to establish baseline data.
(h) It should be assumed that gridlock incidents will occur during SNF transport. The State should press DOE to develop a gridlock risk reduction strategy and to formulate a clear policy on response to gridlock incidents, including assessments of exposures.
EXPOSURES AND DOSES RESULTING FROM VERY SEVERE ACCIDENTS INVOLVING RELEASE OF CASK CONTENTS
An SNF transportation accident severe enough to release any portion of the cask contents virtually defines the concept of a low probability, high consequence event. Even though extremely unlikely, such an accident could occur during truck or rail shipments, and could result in very serious human health, environmental, and economic consequences. A release of contents implies a physical pathway through the cask shell. There is both a loss of shielding and a loss of containment, potentially exposing humans to radioactive contamination and irradiation. Radiation exposures must be evaluated for five groups of people:
(1) persons at the immediate scene of the accident at the time of occurrence;
(2) persons responding to the accident, especially emergency personnel, but also probably members of the general public, including good Samaritans, friends and relatives of potential accident victims, and thrill seekers;
(3) persons residing or working within the area contaminated by the accident, from the time of the accident until completion of clean-up activities [it is usually assumed that this number will be greatly reduced by mandatory evacuation];
(4) personnel involved in recovery and cleanup activities, which may require more than a year; and
(5) future residents of the contaminated area, which, depending upon the level of cleanup, may receive residual exposures from radionuclides deposited in soil or water [this group could also include future consumers of food products or water originating in the cleaned up area].
The timing of events is an important consideration in assessing the potential for radiation exposures from a very severe accident involving fire. Even in a fire exceeding the conditions assumed in the NRC regulations (1475 degrees Fahrenheit for 30 minutes), one would not expect immediate failure of neutron or gamma shielding, cask closure seals, or SNF pellet cladding. For a truck cask, an engulfing fire of at least one to six hours, and for a rail cask, at least two hours to twenty hours, depending upon temperature, would probably be required for a release of contents. [Personal communications with researchers at UNR College of Engineering suggests that a regulatory fire would have to burn 5-6 hours for a truck cask and 18-20 hours for a rail cask in order to fully oxidize the fuel pellets.] This means that the potential for radiation exposure could be greatly reduced by early establishment of a controlled-entry perimeter around the accident site ( at the 1 mrem/hour level, or about 500 to 1,000 feet radius from the cask) and by timely evacuation of downwind areas, particularly any population concentration within a mile of the cask. (Ref. 14)
Severe Accident Scenarios and Consequences.
The Sandquist report evaluated three types of very severe accidents involving release of radioactive materials from SNF casks shipped by rail. Sandquist assumes SNF characteristics and shipping cask design features similar but not identical to what we would expect for shipments to the proposed Yucca Mountain repository or storage facility at NTS. The rail casks used for shipments to Nevada are expected to have capacities at least 50% larger (21 PWR) and perhaps as much as 100 % larger (28 PWR). Sandquist assumed rail casks using wet neutron shields (water jackets) and dry cavities (filled with an inert gas); rail casks used for shipments to Nevada would most likely have solid neutron shields (such as borated polypropolylene). SNF shipped to an interim storage facility would be at least 5 years cooled. For shipments to a repository, SNF is assumed to be cooled ten years on average, although some will be 20-40 years old, and 5 year old fuel could be shipped. Current inventory projections assume higher SNF burn up levels than Sandquist. All of these factors affect the source term assumptions: larger cask capacities and higher burn up result in more curies and more gamma and neutron emitting fission products; greater cooling time reduces total curies and reduces gamma and neutron emissions from the fission products with shorter (less than ten years) half-lives. For purposes of this analysis, Sandquist's cask design and net source term assumptions are reasonably close to what is currently expected for shipments to a repository or interim storage facility.
The most severe accident scenario assessed by Sandquist involves "a combined impact and burst rupture[ of the SNF] accompanied by enhanced release due to oxidation." A rail cask containing 14 five-year-old PWR fuel assemblies is involved in a high-speed impact followed by a long duration, high temperature fire fed by some external source of petroleum fuel. All SNF inside the cask is assumed to be oxidized, and a pathway created either by a valve failure, failure of the cask closure seal, or a small breach caused by a fine stress crack in the cask shell. The release consists of fine particles of the most chemically volatile radioisotopes and radioactive gases. The accident scenario is one of several very low probability events considered credible by a workshop of nuclear industry and government experts convened by Sandia National Laboratories in 1980.(Ref. 15) Following the Sandia workshop report, Sandquist estimates the probability of such an accident at no more than two accidents per million shipments. However, even more severe, credible accidents can be postulated, with more severe consequences, particularly when human errors in cask fabrication, loading, or emergency response, or failure to evacuate nearby populations, are included in the risk assessment. (Ref. 16)
Sandquist calculated a total release of about 6,159 curies. Ignoring the Kr-85, this represents a release of about one-tenth of one percent of the total inventory, as follows:
ENVIRONMENTAL RELEASE OF NUCLIDES IN WORST CASE
SPENT FUEL RAIL ACCIDENT(IMPACT, BURST, AND OXIDATION)
Nuclide Cask Inventory(Ci)* Environmental Release(Ci)
Co-60(as crud) 645 8.06
Kr-85 42,700 4,780
Sr-90 417,000 0.379
Ru-106 114,000 4.67
I-129 0.213 0.001
Cs-134 192,000 326
Cs-137 613,000 1,040
Pu-239 2,870 0.0026
Totals(Ci) 1,380,000 6,159
*Based on a cask inventory of 14 PWR spent fuel assemblies, each 5 years out of the reactor
Source: Ref. 1.
Sandquist calculates the following consequences:
(1) The maximally exposed individual receives a dose of about 10.2 rem, primarily from inhalation of radionuclides. Sandquist describes the maximally exposed individual as an emergency responder located 70 meters directly downwind from the point of release for a period of "a few hours" ("...no protective equipment is worn and no attempt is made to avoid inhalation of radionuclides in the atmosphere.") According to Sandquist, the 10.2 rem dose "is considered to have no consequence other than a possible small increase in the probability of incurring cancer in later years."
(2) The radionuclides released by the accident are carried downwind ( in the plume of smoke from a petroleum fire) and contaminate an area of about 110 square kilometers (42.5 square miles) at levels of 0.2 microcuries per square meter or greater. Three radionuclides - Co-60, Cs-134, and Cs-137 - account for over 99 percent of the activity deposited on the ground. [Kr-85 and other radioactive noble gases are assumed to dissipate harmless.] Within the contaminated region, an area of about 2.2 square kilometers (0.9 square miles) is contaminated at levels of 10 microcuries per square meter; an area of 4.3 square kilometers (1.7 square miles) is contaminated above 5 microcuries per square meter.
(3) The release from the worst case rail accident in a typical urban area results in "22 latent health effects"[ about half cancers, and half genetic disorders]. In a rural area, "the same accident could result in about 0.035 latent health effects." The exposed urban and rural populations would be expected to "experience about 470,000 and 730 cancer fatalities, respectively, from all other causes in the same time period. Clearly, the severe but credible rail cask accident does not contribute significantly to the number of cancer fatalities in the region."
(4) Cleanup of the contaminated area, assuming that the accident occurs in a rural area, could cost as much as $620 million (1985$) and require 460 days. Sandquist assumes cleanup "to a level that reduces individual dose rates from deposited radionuclides down to a maximum value of 500 mrem/yr." In an urban area, the cleanup cost is estimated to exceed $2 billion.
There are many deficiencies in Sandquist's report:
(a) The environmental releases and respirable fractions of radionuclides released in various types of accidents were derived from Wilmot's 1981 report(Ref. 17), which is based primarily upon a May, 1980 workshop sponsored by Sandia National Laboratories. The report is a classic example of reliance upon expert judgement in the absence of empirical evidence. Total release is constrained by the assumption that any breach or pathway will be less than 6.4 square centimeters (1 square inch). Wilmot explicitly excluded consideration of larger pathways resulting from human error or sabotage. Health effects resulting from a release are very sensitive to estimates of the percentage of Cs-134 and Cs-137 that is respirable. Subsequent research efforts, such as the NRC's Modal Study, have done little to resolve uncertainties.(Ref. 8) Sandquist (like Wilmot and the Modal study) fails to consider human error as an exacerbating factor in accident consequences.(Ref. 9)
(b) The report underestimates the potential for severe rail accidents in highly-populated places. In recent years there have been a number of rail accidents involving high-speed derailments and collisions, and long duration, high-temperature fires, in small towns and suburbs near large cities. The slower speeds at which trains travel through major metropolitan areas, which would tend to reduce accident impact forces, are frequency offset by other risk factors, such as proximity to petroleum and chemical storage facilities, collocation of petroleum and natural gas pipelines within railroad rights-of way, and high population densities near tracks. The difficulty of large-scale evacuation, especially if an accident occurs at night or under adverse weather conditions, must be considered.
(c) The methodology used to calculate population-dose estimates and the resulting health effects is grossly superficial. The 1,380 curie release in an urban area with a population density of 3,860 people/square kilometer, assuming no cleanup, generates a 50-year ground gamma population dose of 112,000 person-rem within 80 kilometers of the release point. The population dose of 112,000 is multiplied by 0.0002 latent health effects/person-rem to derive 22 latent health effects. This approach ignores the likelihood that individual doses would be higher, and health effects more serious, in the portion of the contaminated area with the highest contamination levels. A more critical issue, not addressed, is the potential for massive numbers of exposures resulting from inhalation of the plume if nearby populations are not, or cannot be, evacuated.
(d) Gamma and neutron radiation emitted from the cask as a result of shielding loss is not considered. This is a serious omission, since 5-year old PWR SNF has a very high surface dose rate, as high as 25,000 to 50,000 rem/hour at one meter in air. Even 10-year old SNF has a dose rate of 10,000 to 20,000 rem/hour. (Ref. 16) A small loss of shielding could result in significantly increased gamma and neutron radiation within 10 meters of the cask, creating an additional hazard for emergency responders and cleanup workers.
(e) While rail casks carry much larger payloads (and thus larger radioactive source terms) than truck casks, the risk of radiation release and exposure to the general public from a severe accident may be greater for truck shipments. This could be the case, for example, where shipments on the Interstate highway system or U.S. highways must pass through highly populated areas because no bypasses are available (for example, the intersections of I-15 and U.S. 95 in Las Vegas, or U.S. 6 and U.S. 95 in Tonopah).
(f) In spite of these deficiencies, the 12-year-old Sandquist report is still the best available reference for estimating exposures from severe SNF transportation accidents.
Implications for Community Health Studies.
A severe SNF transportation accident could release a small amount of the Cobalt-60, Cesium-134, and Cesium-137 contained in the SNF, and result in human radiation exposure through a number of mechanisms: inhalation of radionuclides carried by the plume, external gamma radiation from radionuclides in the plume, external gamma radiation from radionuclides deposited on the ground, inhalation of radionuclides re-suspended in dust from the ground, and ingestion of water contaminated with radionuclides deposited on surface water and soil. Sandquist concluded that radiation exposures from consumption of food grown on contaminated land were "negligible" compared to the first five pathways. Sandquist did not address direct gamma and neutron radiation from the cask as a result of damage to the shielding.
(1) Persons within 70 meters of the accident scene for a period of a few hours could receive doses in excess of 10 rem, primarily through inhalation (lung dose), but also direct gamma radiation from the plume(whole-body dose). Sandquist does not explicitly address long-term retention of respirable particulates in lung tissue, but this would probably be a concern with Cs-137, which emits both beta and gamma radiation and has a half-life of about 30 years. It should also be noted that site-specific circumstances in Nevada could significantly affect both near-site atmospheric concentrations available for inhalation (increasing the dose), and the number of persons exposed by inhalation and direct gamma radiation from the plume. Under certain weather conditions, plume movement in a canyon or valley could be more constrained than the dispersion modeled by Sandquist. A severe accident in Las Vegas or Caliente could be even more severe.
(2) Persons residing or working in the region contaminated by the accident could receive doses in excess of 500 mrem per year, depending upon the schedule and level of cleanup. These would be primarily whole-body gamma doses from Co-60, Cs-134, and Cs-137 deposited on the ground.
(3) Although not explicitly addressed by Sandquist, direct gamma and neutron radiation from the damaged cask could pose a significant threat to emergency response personnel, recovery workers, and accident victims. Depending upon the degree of shielding loss, exposure rates in the range of 25 to 250 rem/hour at 2 meters, and 2 to 20 rem/hour at 8 meters are easily conceivable, and could result in whole-body doses in excess of 10 rem.
These potential exposures and the resulting doses are lower than the thresholds usually considered to cause a high probability of adverse health effects. However, the potential doses are very high relative to normal background doses, and could exceed the annual allowable dose limits for nuclear industry workers. In planning for data collection as a part of community health studies, the following should be considered:
(a) All potential rail and highway routes should be surveyed to identify highly-populated locations within one-mile of the corridor. Baseline data on cancers, genetic disorders, and respiratory system illnesses and deaths in communities near shipment corridors should be collected. This would obviously be complicated for rail and highway routes through Las Vegas.
(b) It should be assumed that an accident resulting in release of radioactive materials could occur during SNF transport. Prior to the beginning of shipments, an appropriate State of Nevada agency should develop protocols for identifying persons potentially exposed to radiation as a result of an accident, and for monitoring their physical and mental health in the aftermath.
(c) Emergency response personnel along SNF transportation corridors should have complete medical examinations before shipments begin, and should be reexamined annually. Monitoring of white blood cell and platelet counts would be particularly important for any responders actually involved with SNF accidents, although one would not necessarily expect to see impacts at exposures less than 10,000 mrem.(Ref. 14)
(d) Emergency response personnel should be equipped with personal dosimeters, and actual doses should be monitored collectively and individually, in the aftermath of any SNF accident.
(e) An ALARA assessment should develop exposure reduction strategies for emergency response personnel and accident cleanup workers.
(f) A system of fixed radiation monitors should be installed near highly populated areas along shipment routes to record actual exposure rates. Preferably this should be done before shipments begin to establish baseline data. These instruments would also provide valuable information on radiation levels in the event of an accident.
EXPOSURES AND DOSES RESULTING FROM SUCCESSFUL TERRORIST ATTACK
While there is no definitive method for predicting the probability of a terrorist event involving spent fuel shipments, research studies sponsored by DOE and NRC have demonstrated that terrorists armed with high-energy explosive devices could breach a truck shipping cask and cause a release of radioactive materials. These studies established that a release of one percent of the fuel mass contained by the cask is credible.
Unlike the release from a severe accident involving fire discussed previously, only a very small fraction of the terrorist release would be expected in the form of a respirable aerosol, and absent an accompanying fire, the terrorist release would not be expected to travel far and contaminate a large area. However, the terrorist release would likely involve much a larger fraction of cask contents, so that the total radioactivity released from a truck cask would be considerably larger than the accident release from a rail cask. Moreover, the terrorist attack would clearly create a much greater hazard of direct gamma and neutron radiation from the damaged cask. Depending upon the physical geography of the attack location and weather conditions at the time of incident and immediately thereafter, radiative contamination from a terrorist incident could also spread beyond the site of the attack.
The combination of near-site contamination and irradiation could result in radiation exposures to five groups of people:
(1) persons at the immediate scene of the attack at the time of occurrence;
(2) persons responding to the attack, especially law enforcement and emergency personnel, but also probably members of the general public[ evacuation may be delayed or hindered by confusion and panic];
(3) personnel involved in recovery and cleanup activities;
(4) persons residing or working within the area contaminated by the attack until completion of clean-up activities this number should be greatly reduced by mandatory evacuation and exclusion]; and
(5) future users and/or residents of the contaminated area, which, depending upon the level of cleanup, may receive residual exposures from radionuclides deposited on soil or buildings [if the attack occurred in a rural area, this group could also include future consumers of food products or water originating in the cleaned up area].
A full-scale test of shipping cask response to a terrorist attack conducted at Sandia National Laboratories determined the following consequences:
TERRORIST ATTACK SCENARIO AND CONSEQUENCES
Scenario: Terrorists capture and attack truck cask containing 1 PWR assembly with military high-energy explosive device (M3A1)
Hole Diameter: 152.5 mm (6.0 inches)
Fuel Rods Damaged: 111 of 223 (50%)
Fuel Mass Fractured: 20.82 kg (10%)
Fuel Mass Released: 2.55 kg [5.6 pounds] (1%)
Released as Aerosol: 2.94 g [1/10 ounce]
Source: Ref. 3.
A number of reviewers, including Agency staff and contractors, have criticized the Sandia full scale attack test, and related scale-model tests and subsequent analyses conducted at Sandia and Battelle Columbus Laboratories, on the grounds that the tests did not represent a worst case scenario.(Ref. 7) Reviewers have also criticized the NRC's use of the Sandia and Battelle studies to propose a reduction (later withdrawn) in the security requirements for spent fuel shipments. Agency staff and contractors are preparing a separate report on the vulnerability of shipments to terrorist attack, and the potential for consequences significantly more severe than those demonstrated by the Sandia full-scale test. For this analysis, a one percent release of cask contents is sufficient to illustrate the potential magnitude of exposures and doses.
Over the past decade, nuclear utility fuel management practices have moved towards higher initial fuel enrichments and higher burn up rates, resulting in spent fuel with higher radioactivity and thermal output than assumed in earlier studies [such as the Sandquist report]. The agency contractor preparing this analysis developed the following estimated inventory of a new generation shipping cask (such as the General Atomics GA 4/9) based on updated SNF characteristics published by Oak Ridge National Laboratories [a more detailed analysis is required to address the consequences of a release including the full population of radionuclides present in SNF, such as Cobalt-60, Technetium-99, and Iodine-129(Ref. 18)]:
ESTIMATED INVENTORY, BY MAJOR RADIONUCLIDE, OF A TRUCK CASK CONTAINING 4 REFERENCE PWR SPENT FUEL ASSEMBLIES*
Nuclide Radioactivity(Curies) Percent of Total
Kr-85 10,188 1.20
Sr-90 121,860 14.39
Y-90 121,860 14.40
Cs-134 13,356 1.58
Cs-137 176,760 20.89
Ba137m 167,400 19.77
Pm-147 16,416 1.94
Eu-154 10,260 1.21
Pu-241 172,260 20.36
Cm-244 5,184 0.61
Other 30,895 3.65
Total 846,439 100.00
*Cask contains 4 Westinghouse 17x17 assemblies, total weight 1.8 MTIHM, Initial
Enrichment 3.72%, Burn up of 40,000 Mwd/MTIHM, Decay time 10 years
Source: Calculated from Ref. 19, Table 2.4.13.
Agency contractors are currently preparing a detailed assessment of the radiological consequences of several scenarios involving terrorist attacks on shipping casks with high energy explosives. THE FOLLOWING SCENARIOS, RELEASE ESTIMATES AND DOSE RATE ESTIMATES ARE PRELIMINARY AND SUBJECT TO REVISION. For purposes of this analysis, the potential for radiation exposures results from the following:
Radioactive materials dispersed from the cask.
The attack on a GA 4 truck cask results in a one-percent release, about 8,464 curies, including all of the major radionuclides contained in SNF. The outer boundary of the blast effect/shrapnel zone of the assumed weapon is estimated at 100 meters radius from the cask, encompassing an area of about 31,416 square meters or about 7.8 acres. Assuming that the blast hole is located on the top of the cask, it is hypothesized that 50% of the radionuclides are deposited within 50 meters of the cask, resulting in an average contamination level of about 0.5 curies/square meter [over 7,854 square meters] in the area closest to the cask. The remaining 50% is hypothesized to be deposited within 100 meters of the cask [over an area of about 23,562 square meters] resulting in an average contamination level of about 0.18 curies/square meter. However, a relatively large portion of the Cs-134 and Cs-137, and other volatile nuclides, may be released in the form of a respirable aerosol, and be more widely dispersed. This dispersal creates enormous potential for direct irradiation and contamination.
Direct radiation from the damaged cask.
The weapon is assumed to create a 6 inch-diameter hole in one side of the cask, perforating two PWR assemblies, and damaging the gamma shielding on the other [unpenetrated] side of the cask. The surface dose rate of the SNF is assumed to be at least 10,000 rem/hour at one meter in air. It is hypothesized that the blast hole creates a point source equal to about 1,000 rem/hour one meter from the surface of the nearest exposed cross-section of damaged SNF. Using the inverse square law, the dose rate is estimated at 250 rem/hour at 2 meters distance, 63 rem/hour at 4 meters, 16 rem/hour at 8 meters, and 4 rem/hour at 16 meters. Assuming that the blast hole is on the side of the cask [as opposed to the top or bottom], the gamma and neutron radiation from the damaged cask presents a serious radiological hazard to attack victims, law enforcement and emergency response personnel, and recovery workers within 8 to 16 meters of the side of the cask containing the blast hole. Beyond that distance, the dose rate is probably less significant than the exposures from radionuclides deposited on the ground. The radiation field on the other side of the cask would be higher than allowed under NRC regulations, due to shielding damage, but much lower than on the damaged side.
Implications for community health studies.
Until further analysis is completed, the implications of exposures and doses from a terrorist attack involving a release should be considered to be the same as for a severe accident involving a release, except that the releases, exposures, and resulting doses could be significantly greater, and concentrated in considerably smaller areas.
1. Sandquist, G.M., et al., Exposures and Health Effects from Spent Fuel Transportation, RAE-8339/12-1, Prepared for U.S. DOE, Office of Civilian Radioactive Waste Management, Salt Lake City: Rogers and Associates Engineering Corporation, November 29, 1985.
2. Finley, N.C., et al., Transportation of Radionuclides in Urban Environs: Draft Environmental Assessment, NUREG/CR-0743, Prepared for U.S. Nuclear Regulatory Commission, Albuquerque: Sandia National Laboratories, July 1980.
3. Sandoval, R.P., et al., An Assessment of the Safety of Spent Fuel Transportation in Urban Environs, SAND82-2365, Prepared for U.S. DOE, Albuquerque: Sandia National Laboratories, June 1983.
4. Schmidt, E.W., et al., Final Report on Shipping Cask Sabotage Source Term Investigation, NUREG/CR-2472, Prepared for U.S. Nuclear Regulatory Commission, Columbus, OH: Battelle Columbus Laboratories, October 1982.
5. U.S. NRC, "10 CR Part 73, Modification of Protection Requirements for Spent Fuel Shipments: Proposed Rule," Federal Register, Vol. 49, No.112 (June 8, 1984), 23867 - 23872.
6. Nevada Agency for Nuclear Projects/Nuclear Waste Project Office, A Report on High-Level Nuclear Waste Transportation: Prepared Pursuant to Assembly Concurrent Resolution No. 8 of the 1987 Nevada Legislature, NWPO-TN-001-88, December, 1988
7. Audin, L., "Analyses of Cask Sabotage Involving Portable Explosives: A Critique," Draft Report, Prepared for Nevada Nuclear Waste Project Office, October, 1989.
8. Audin, L., Nuclear Waste Shipping Container Response to Severe Accident Conditions: A Brief Critique of the Modal Study, NWPO-TN-005-90, Prepared for Nevada Nuclear Waste Project Office, December, 1990.
9. Resnikoff, M., Probabilistic Risk Assessment and Nuclear Waste Transportation: A Case Study of the Use of RADTRAN in the 1986 Environmental Assessment for Yucca Mountain, NWPO-TN-006-90, December, 1990.
10. Halstead, R.J., et al., State of Nevada Comments on the OCRWM From-Reactor Spent Fuel Shipping Cask Preliminary Design Reports, NWPO-TN-009-90, December, 1990.
11. Freudenburg, W.R., Organizational Management of Long-Term Risks: Implications for Risk and Safety in the Transportation of Nuclear Wastes, NWPO-TN-013-91, Prepared for Nevada Nuclear Waste Project Office, September, 1991.
12. Planning Information Corporation, The Transportation of Spent Nuclear Fuel and High-Level Waste: A Systematic Basis for Planning and Management at National, Regional, and Community Levels, Prepared for Nevada Nuclear Waste Project Office, September 10, 1996
13. Darrough, E., DOE/OCRWM, Presentation before U.S. Nuclear Waste Technical Review Board, Transportation and Systems Panel, Washington, DC, October 22, 1990
14. Klimenko, N.A. , and Redington, J.F. , Radiological Emergencies: A Handbook for Emergency Responders, Greenbelt, MD: Bradford Communications Corporation, 1982.
15. Wilmot, E.L., et al., Report on a Workshop on Transportation Accident Scenarios Involving Spent Fuel, May 6 - 8, 1980, SAND80-2012, Prepared for U.S. DOE, Albuquerque: Sandia National Laboratories, February 1981.
16. Resnikoff, M., The Next Nuclear Gamble: Transportation and Storage of Nuclear Waste, New York: Council on Economic Priorities, 1983.
17. Wilmot, E.L., Transportation Accident Scenarios for Commercial Spent Fuel, SAND80-2124, Prepared for U.S. DOE, Albuquerque: Sandia National Laboratories, February 1981.
18. Lipschultz, R.D., Radioactive Waste: Politics, Technology, and Risk, A Report of the Union of Concerned Scientists. Cambridge, MA: Ballinger, 1980.
19. Oak Ridge National Laboratory, Characteristics of Potential Repository Wastes, DOE/RW-0184-R, Prepared for U.S. DOE/OCRWM (July, 1992).