This report examines the current policy, legal, and regulatory framework pertaining to used nuclear fuel and high level waste management in the United States. The goal is to identify potential changes that if made could add flexibility and possibly improve the chances of successfully implementing technical aspects of a nuclear waste policy. Experience suggests that the regulatory framework should be established prior to initiating future repository development. Concerning specifics of the regulatory framework, reasonable expectation as the standard of proof was successfully implemented and could be retained in the future; yet, the current classification system for radioactive waste, including hazardous constituents, warrants reexamination. Whether or not consideration of multiple sites are considered simultaneously in the future, inclusion of mechanisms such as deliberate use of performance assessment to manage site characterization would be wise. Because of experience gained here and abroad, diversity of geologic media is not particularly necessary as a criterion in site selection guidelines for multiple sites. Stepwise development of the repository program that includes flexibility also warrants serious consideration. Furthermore, integration of the waste management system from storage, transportation, and disposition, should be examined and would be facilitated by integration of the legal and regulatory framework. Finally, in order to enhance acceptability of future repository development, the national policy should be cognizant of those policy and technical attributes that enhance initial acceptance, and those policy and technical attributes that maintain and broaden credibility.
Transportation risk, like any risk, is defined by the risk triplet: what can happen (the scenario), how likely it is (the probability), and the resulting consequences. This paper evaluates the development of transportation scenarios, the associated probabilities, and the consequences. The most likely radioactive materials transportation scenario is routine, incident-free transportation, which has a probability indistinguishable from unity. Accident scenarios in radioactive materials transportation are of three different types: accidents in which there is no impact on the radioactive cargo, accidents in which some gamma shielding may be lost but there is no release of radioactive material, and accident in which radioactive material may potentially be released. Accident frequencies, obtainable from recorded data validated by the U.S. Department of Transportation, are considered equivalent to accident probabilities in this study. Probabilities of different types of accidents are conditional probabilities, conditional on an accident occurring, and are developed from event trees. Development of all of these probabilities and the associated highway and rail accident event trees are discussed in this paper.
The dose limits for emissions from the nuclear fuel cycle were established by the Environmental Protection Agency in 40 CFR Part 190 in 1977. These limits were based on assumptions regarding the growth of nuclear power and the technical capabilities of decontamination systems as well as the then-current knowledge of atmospheric dispersion and the biological effects of ionizing radiation. In the more than thirty years since the adoption of the limits, much has changed with respect to the scale of nuclear energy deployment in the United States and the scientific knowledge associated with modeling health effects from radioactivity release. Sandia National Laboratories conducted a study to examine and understand the methodologies and technical bases of 40 CFR 190 and also to determine if the conclusions of the earlier work would be different today given the current projected growth of nuclear power and the advances in scientific understanding. This report documents the results of that work.
Uncertainty may be introduced into RADTRAN analyses by distributing input parameters. The MELCOR Uncertainty Engine (Gauntt and Erickson, 2004) has been adapted for use in RADTRAN to determine the parameter shape and minimum and maximum of the distribution, to sample on the distribution, and to create an appropriate RADTRAN batch file. Coupling input parameters is not possible in this initial application. It is recommended that the analyst be very familiar with RADTRAN and able to edit or create a RADTRAN input file using a text editor before implementing the RADTRAN Uncertainty Analysis Module. Installation of the MELCOR Uncertainty Engine is required for incorporation of uncertainty into RADTRAN. Gauntt and Erickson (2004) provides installation instructions as well as a description and user guide for the uncertainty engine.
The radiological transportation risk & consequence program, RADTRAN, has recently added an updated loss of lead shielding (LOS) model to it most recent version, RADTRAN 6.0. The LOS model was used to determine dose estimates to first-responders during a spent nuclear fuel transportation accident. Results varied according to the following: type of accident scenario, percent of lead slump, distance to shipment, and time spent in the area. This document presents a method of creating dose estimates for first-responders using RADTRAN with potential accident scenarios. This may be of particular interest in the event of high speed accidents or fires involving cask punctures.
RADTRAN is an internationally accepted program and code for calculating the risks of transporting radioactive materials. The first versions of the program, RADTRAN I and II, were developed for NUREG-0170 (USNRC, 1977), the first environmental statement on transportation of radioactive materials. RADTRAN and its associated software have undergone a number of improvements and advances consistent with improvements in both available data and computer technology. The version of RADTRAN currently bundled with RadCat is RADTRAN 6.0. This document provides a detailed discussion and a guide for the use of the RadCat 3.0 Graphical User Interface input file generator for the RADTRAN code. RadCat 3.0 integrates the newest analysis capabilities of RADTRAN 6.0 which includes an economic model, updated loss-of-lead shielding model, and unit conversion. As of this writing, the RADTRAN version in use is RADTRAN 6.0.
Residents of Southern Nevada have expressed interest in the risks posed by transportation of radioactive materials to the Nevada Test Site for disposal. Residents of the urban sections of Clark County, in particular, are concerned because the trucks carrying this material pass through fairly densely populated areas on both primary and secondary highways. Excellent studies by the Desert Research Institute (Miller, et al, 2005; 2007) measured the external radiation doses from these trucks and provided a benchmark for estimating the doses sustained by the population along the transportation routes. The studies did not, however, assess the doses to the population in the event of an accident involving these trucks. The present study estimates doses to the population along these routes and to various individual receptors, both for routine, incident-free transportation and for transportation accidents.
RADTRAN is a program and code for calculating the risks of transporting radioactive materials. The first versions of the program, RADTRAN I and II, were developed for NUREG-0170 (USNRC, 1977), the first environmental impact statement on transportation of radioactive materials. RADTRAN and its associated software have undergone a number of improvements and advances consistent with improvements in computer technology.
The RADTRAN Loss of Shielding (LOS) Model was benchmarked using MicroShield 6.20®. This analysis considers an intact spent fuel truck cask as well as a set of damaged truck casks. Ratios of dose rates are calculated for casks with a loss of lead shielding to those of intact casks, and are then compared to ratios generated by the LOS model. LOS Model results were considered verified if two main constraints were satisfied. First, the dose rate profiles for both the LOS and MicroShield 6.20® calculations must have the same general shape and behavior. Additionally, the largest factor difference between any two points of the dose rate profiles may not exceed an order of magnitude. Reasonable agreement is shown for large-fraction LOS scenarios; however the differences in results are not satisfactory for cases with small fractions of slump.