Performance Assessment of Near-Surface Facilities for Disposal of Low-Level Radioactive Waste

91823NCRP_Cover

7/20/06

9:46 PM

Page 1

152

NCRP REPORT No. 152

PERFORMANCE ASSESSMENT OF NEAR-SURFACE FACILITIES FOR DISPOSAL OF LOW-LEVEL RADIOACTIVE WASTE

PERFORMANCE ASSESSMENT OF NEAR-SURFACE FACILITIES FOR DISPOSAL OF LOW-LEVEL RADIOACTIVE WASTE

N C R P

National Council on Radiation Protection and Measurements

NCRP REPORT No. 152

Performance Assessment of Near-Surface Facilities for Disposal of Low-Level Radioactive Waste

Recommendations of the NATIONAL COUNCIL ON RADIATION PROTECTION AND MEASUREMENTS

December 31, 2005

National Council on Radiation Protection and Measurements 7910 Woodmont Avenue, Suite 400 / Bethesda, MD 20814-3095

LEGAL NOTICE This Report was prepared by the National Council on Radiation Protection and Measurements (NCRP). The Council strives to provide accurate, complete and useful information in its documents. However, neither NCRP, the members of NCRP, other persons contributing to or assisting in the preparation of this Report, nor any person acting on the behalf of any of these parties: (a) makes any warranty or representation, express or implied, with respect to the accuracy, completeness or usefulness of the information contained in this Report, or that the use of any information, method or process disclosed in this Report may not infringe on privately owned rights; or (b) assumes any liability with respect to the use of, or for damages resulting from the use of any information, method or process disclosed in this Report, under the Civil Rights Act of 1964, Section 701 et seq. as amended 42 U.S.C. Section 2000e et seq. (Title VII) or any other statutory or common law theory governing liability.

Disclaimer Any mention of commercial products within NCRP publications is for information only; it does not imply recommendation or endorsement by NCRP.

Library of Congress Cataloging-in-Publication Data National Council on Radiation Protection and Measurements. Performance assessment of near-surface facilities for disposal of low-level radioactive waste. p. cm. — (NCRP report ; no. 152) Includes bibliographical references and index. ISBN-13: 978-0-929600-89-5 ISBN-10: 0-929600-89-4 1. Low level radioactive waste disposal facilities—United States—Evaluation. 2. Radioactive waste disposal in the ground—United States—Evaluation. I. National Council on Radiation Protection and Measurements. TD898.15.P47 2006 363.72'89--dc22 2006018391

Copyright © National Council on Radiation Protection and Measurements 2006 All rights reserved. This publication is protected by copyright. No part of this publication may be reproduced in any form or by any means, including photocopying, or utilized by any information storage and retrieval system without written permission from the copyright owner, except for brief quotation in critical articles or reviews.

[For detailed information on the availability of NCRP publications see page 448.]

Preface The search for solutions to the challenges posed by the need for long-term disposal and isolation of low-level radioactive waste has been long and complex. The Low-Level Radioactive Waste Policy Act, passed in 1980 and amended in 1985, specified that the disposal of most low-level waste not generated at U.S. Department of Energy sites is the responsibility of states or State Compacts. A critical factor in the process of determining acceptable disposal practices for low-level waste at any site is a demonstration of compliance with regulatory performance objectives. NCRP was asked to evaluate current approaches to performance assessment for near-surface disposal facilities for low-level radioactive waste, and Scientific Committee 87-3 was established to prepare a report on this subject. This Report provides a review of concepts underlying performance assessments of near-surface disposal facilities for low-level radioactive waste and approaches to conducting such assessments. This review includes discussions on the nature and scope of performance assessment, accepted approaches to conducting all aspects of a performance assessment, and unresolved issues in conducting performance assessments and applying the results. The Report also discusses a number of policy issues that affect conduct of performance assessment. Examples of these issues include the time period for complying with performance objectives, application of drinking water standards, and interpretation of performance objectives for compliance purposes. It is not the objective of this Report to present recommendations for resolution of policy issues, although the importance of such issues and other social, political and economic factors is recognized. Serving on the Committee were:

Chairmen David C. Kocher (1999–2006) SENES Oak Ridge, Inc. Oak Ridge, Tennessee

Matthew W. Kozak (1992–1999) Monitor Scientific LLC Richland, Washington

iii

iv / PREFACE Members William E. Kennedy, Jr. Dade Moeller & Associates, Inc. Richland, Washington

Roger R. Seitz Bechtel BWXT Idaho Scoville, Idaho

Vern Rogers* Rogers & Associates Engineering Corporation Salt Lake City, Utah

Terrence Sullivan Brookhaven National Laboratory Upton, New York

NCRP Secretariat E. Ivan White, Staff Consultant Cindy L. O’Brien, Managing Editor David A. Schauer, Executive Director

The Council is grateful for the financial support provided by the U.S. Department of Energy and the U.S. Nuclear Regulatory Commission at various times during the preparation of this Report. The Council also wishes to express its appreciation to the Committee members for the time and effort devoted to the preparation of this Report. Thomas S. Tenforde President

*deceased

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.1 Purpose of Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.2 Scope of Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.3 Related NCRP Recommendations . . . . . . . . . . . . . . . . 14 2. Definition and Principles of Performance Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.1 Nature of Performance Assessment . . . . . . . . . . . . . . . 16 2.2 Definition of Performance Assessment . . . . . . . . . . . . 18 2.3 General Principles of Performance Assessment . . . . . 20 2.3.1 Performance Assessment as an Iterative Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.3.2 Performance Assessment as a Decision Tool . 22 2.3.3 Uncertainty in Results of Performance Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.3.4 Integration and Interpretation of Results . . . . 23 2.3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2.4 Balance Between Conservatism and Realism in Performance Assessment . . . . . . . . . . . . . . . . . . . . . . . 25 3. Context for Performance Assessment . . . . . . . . . . . . . . 28 3.1 Definition of Low-Level Radioactive Waste . . . . . . . . . 28 3.1.1 Earliest Descriptions of Low-Level Waste . . . 28 3.1.2 Current Definition of Low-Level Waste . . . . . 29 3.2 Sources and Properties of Low-Level Waste . . . . . . . . 34 3.3 ICRP Recommendations on Disposal of Radioactive Waste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.3.1 General Recommendations on Radiation Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.3.2 General Policy on Application of Protection Principles to Radioactive Waste Disposal . . . . 36 3.3.3 Application of Protection Principles to Disposal of Solid Radioactive Wastes . . . . . . . 37 3.3.4 Discussion of ICRP Recommendations . . . . . . 45 v

vi / CONTENTS 3.4

3.5

Requirements for Near-Surface Disposal of Low-Level Waste . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 3.4.1 Authorized Disposal Systems . . . . . . . . . . . . . . 47 3.4.1.1 Legal and Regulatory Specifications . 47 3.4.1.2 Historical Development of Disposal Technologies . . . . . . . . . . . . . . . . . . . . 48 3.4.2 Requirements for Protection of the Public . . . . 49 3.4.2.1 Licensing Criteria Established by NRC . . . . . . . . . . . . . . . . . . . . . . . . 49 3.4.2.2 Requirements Established by DOE . . 52 3.4.2.3 Implementation of the ALARA Requirement . . . . . . . . . . . . . . . . . . . . 56 3.4.2.4 EPA Views on Requirements for Disposal of Low-Level Waste . . . . . . . 57 3.4.2.5 Implications of Performance Objectives . . . . . . . . . . . . . . . . . . . . . . 59 3.4.2.6 Requirements of States and State Compacts. . . . . . . . . . . . . . . . . . . . . . . 61 3.4.3 Unresolved Issues in Performance Objectives for Low-Level Waste Disposal . . . . . . . . . . . . . 61 3.4.3.1 Time Period for Compliance. . . . . . . . 62 3.4.3.2 Inclusion of Doses Due to Radon . . . . 63 3.4.3.3 Performance Objective for Protection of Groundwater. . . . . . . . . . . . . . . . . . 64 3.4.3.4 Interpretation of Performance Objectives for Compliance Purposes . 66 3.4.4 Other Approaches to Regulating Waste Disposal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 3.4.4.1 Approaches to Regulating Radioactive Waste Disposal . . . . . . . . 67 3.4.4.2 Approach to Regulating Disposal of Hazardous Chemical Waste . . . . . . . . 69 3.4.5 Requirements for Protection of the Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 Other Concepts in Performance Assessment . . . . . . . . 74 3.5.1 Institutional Controls . . . . . . . . . . . . . . . . . . . . 74 3.5.1.1 Active Institutional Controls . . . . . . . 74 3.5.1.2 Passive Institutional Controls . . . . . . 75 3.5.2 Model Validation and Confidence in Model Outcomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

CONTENTS

3.5.3

/ vii

3.5.2.1 Quality Assurance . . . . . . . . . . . . . . . 78 3.5.2.2 Model Calibration . . . . . . . . . . . . . . . 79 3.5.2.3 Evaluation of Conservative Bias . . . . 80 Concept of Reasonable Assurance . . . . . . . . . . 81 3.5.3.1 Description and Interpretation of Reasonable Assurance . . . . . . . . . . . . 83 3.5.3.2 An Approach to Achieving Reasonable Assurance of Compliance. . . . . . . . . . 84

4. Framework for Performance Assessment . . . . . . . . . . . 88 4.1 Data Collection, Conceptual Models, and Mathematical Models . . . . . . . . . . . . . . . . . . . . . . . . . . 89 4.1.1 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . 90 4.1.2 Development of Conceptual Models . . . . . . . . . 91 4.1.3 Selection and Implementation of Mathematical Models . . . . . . . . . . . . . . . . . . . . 93 4.2 Process for Conducting Performance Assessments . . . 95 4.2.1 Historical Perspective . . . . . . . . . . . . . . . . . . . 96 4.2.2 General Process for Conduct of Performance Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 4.2.2.1 Description of Context for Performance Assessment. . . . . . . . . 101 4.2.2.2 Description of Disposal System. . . . 103 4.2.2.3 Development and Justification of Scenarios . . . . . . . . . . . . . . . . . . . . . 103 4.2.2.4 Formulation and Implementation of Models. . . . . . . . . . . . . . . . . . . . . . . . 104 4.2.2.5 Conduct of Calculations (Consequence Analysis) . . . . . . . . . . . . . . . . . . . . . . 105 4.2.2.6 Interpretation of Results . . . . . . . . . 106 4.2.2.7 Modifications of Assessment . . . . . . 107 4.2.2.8 Iterations of Performance Assessment . . . . . . . . . . . . . . . . . . . . 108 4.2.2.9 Summary . . . . . . . . . . . . . . . . . . . . . 109 5. Performance Assessment Models . . . . . . . . . . . . . . . . . 110 5.1 General Approach to Modeling of Disposal Systems . 110 5.1.1 Decoupling and Simplifying an Analysis . . . 113 5.1.2 Analysis by Modules . . . . . . . . . . . . . . . . . . . . 114 5.1.3 Analysis of Time Dependence . . . . . . . . . . . . 117 5.1.4 Organization of Section . . . . . . . . . . . . . . . . . 117

viii / CONTENTS 5.2

5.3

5.4

Cover Performance and Infiltration . . . . . . . . . . . . . . 117 5.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 5.2.2 Types of Covers . . . . . . . . . . . . . . . . . . . . . . . . 121 5.2.3 Degradation of Covers . . . . . . . . . . . . . . . . . . 123 5.2.4 Approaches to Estimating Infiltration . . . . . . 125 5.2.5 Summary and Conclusions . . . . . . . . . . . . . . . 129 Performance of Concrete Barriers . . . . . . . . . . . . . . . . 130 5.3.1 General Approach to Modeling of Concrete Barriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 5.3.2 Water Flow Through Concrete . . . . . . . . . . . . 132 5.3.3 Degradation of Concrete . . . . . . . . . . . . . . . . . 134 5.3.3.1 Sulfate Attack . . . . . . . . . . . . . . . . . . 135 5.3.3.2 Freeze/Thaw Cycling . . . . . . . . . . . . 136 5.3.3.3 Calcium Leaching . . . . . . . . . . . . . . . 136 5.3.3.4 Alkali-Aggregate Reaction . . . . . . . . 137 5.3.3.5 Corrosion of Reinforcing Steel . . . . . 137 5.3.3.6 Combination of Reactions . . . . . . . . 139 5.3.4 Application of Models . . . . . . . . . . . . . . . . . . . 140 5.3.5 Example Analyses of Long-Term Performance of Concrete Barriers . . . . . . . . . . . . . . . . . . . . 140 5.3.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 Source Term . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 5.4.1 Inventories of Radionuclides . . . . . . . . . . . . . 144 5.4.2 Radionuclide Release Rates (Source Term) . . 147 5.4.3 Disposal Facility Concepts . . . . . . . . . . . . . . . 153 5.4.4 Waste Containers . . . . . . . . . . . . . . . . . . . . . . 155 5.4.5 Waste Forms . . . . . . . . . . . . . . . . . . . . . . . . . . 158 5.4.5.1 Waste-Form Performance: Aqueous Phase . . . . . . . . . . . . . . . . . . . . . . . . . 160 5.4.5.1.1 Surface Rinse with Partitioning. . . . . . . . . . . 160 5.4.5.1.2 Diffusion-Controlled Release . . . . . . . . . . . . . . 162 5.4.5.1.3 Dissolution (Constant) Release . . . . . . . . . . . . . . 165 5.4.5.1.4 Solubility-Limited Release . . . . . . . . . . . . . . 166 5.4.5.2 Waste-Form Performance: Gas Phase . . . . . . . . . . . . . . . . . . . . . . . . . 167 5.4.5.3 Ingrowth of Radionuclides . . . . . . . . 167

CONTENTS

5.4.6

5.5

/ ix

Transport in Disposal Facility . . . . . . . . . . . . 168 5.4.6.1 Aqueous-Phase Transport . . . . . . . . 168 5.4.6.2 Gas-Phase Transport . . . . . . . . . . . . 170 5.4.7 Interfaces with Other Performance Assessment Models . . . . . . . . . . . . . . . . . . . . . 171 5.4.8 Source-Term Issues . . . . . . . . . . . . . . . . . . . . 171 5.4.8.1 Radionuclide Inventory Issues . . . . 172 5.4.8.1.1 Unit Source Term. . . . . . 172 5.4.8.1.2 Inaccurate Estimation of Inventories . . . . . . . . . . . 173 5.4.8.2 Waste-Container Issues . . . . . . . . . . 174 5.4.8.2.1 Insufficient Characterization of Containers . . . . . . . . . . . 175 5.4.8.2.2 Distributed Failure of Containers . . . . . . . . . . . 175 5.4.8.3 Waste-Form Issues . . . . . . . . . . . . . . 175 5.4.8.3.1 Changes in Waste Types and Characteristics . . . . 176 5.4.8.3.2 Insufficient Waste-Form Characterization . . . . . . 176 5.4.8.3.3 Insufficient Data on Release Rates . . . . . . . . . 177 5.4.8.3.4 Homogeneity of Wastes . 177 5.4.8.3.5 Issues of Geochemistry and Solubility . . . . . . . . . 178 5.4.8.4 Issues of Radionuclide Transport . . 179 5.4.8.4.1 Steady-State Flow . . . . . 179 5.4.8.4.2 Uniform Flow Fields . . . 179 5.4.8.4.3 Role of Geochemistry in Transport . . . . . . . . . . . . 180 5.4.8.4.4 Role of Microbial Processes . . . . . . . . . . . . 181 5.4.8.4.5 Role of Colloids. . . . . . . . 181 5.4.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 Unsaturated Zone Flow and Transport . . . . . . . . . . . 183 5.5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 183 5.5.2 Interfaces with Other Performance Assessment Models . . . . . . . . . . . . . . . . . . . . . 185

x / CONTENTS 5.5.3

5.6

5.7

General Discussion of Unsaturated Zone Flow and Transport . . . . . . . . . . . . . . . . . . . . . . . . . 186 5.5.4 Data Requirements . . . . . . . . . . . . . . . . . . . . . 191 5.5.5 Modeling of Unsaturated Flow . . . . . . . . . . . . 194 5.5.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 Aquifer Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 5.6.1 Modeling of Aquifer Flow . . . . . . . . . . . . . . . . 197 5.6.2 Issues in Solving Flow Equation . . . . . . . . . . 200 5.6.2.1 Development of Steady-State Conditions . . . . . . . . . . . . . . . . . . . . . 200 5.6.2.2 Scale and Heterogeneity . . . . . . . . . 200 5.6.2.3 Boundary Conditions . . . . . . . . . . . . 202 5.6.2.4 Flow in Fractured Media . . . . . . . . . 203 5.6.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 Radionuclide Transport in Groundwater and Surface Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 5.7.1 Phenomena That Influence Transport in Groundwater . . . . . . . . . . . . . . . . . . . . . . . . . . 206 5.7.1.1 Sorption. . . . . . . . . . . . . . . . . . . . . . . 206 5.7.1.2 Advection . . . . . . . . . . . . . . . . . . . . . 208 5.7.1.3 Diffusion . . . . . . . . . . . . . . . . . . . . . . 209 5.7.1.4 Dispersion . . . . . . . . . . . . . . . . . . . . . 210 5.7.2 Combination of Phenomena That Influence Transport in Groundwater . . . . . . . . . . . . . . . 213 5.7.3 Methods of Solution of Groundwater Transport Equation . . . . . . . . . . . . . . . . . . . . . 215 5.7.3.1 Analytical Solutions . . . . . . . . . . . . . 216 5.7.3.2 Green's Function (Semi-Analytical) Solutions . . . . . . . . . . . . . . . . . . . . . . 216 5.7.3.3 Finite-Element and Finite-Difference Solutions . . . . . . . . . . . . . . . . . . . . . . 217 5.7.3.4 Stream-Tube Solutions. . . . . . . . . . . 218 5.7.4 Boundary Conditions . . . . . . . . . . . . . . . . . . . 219 5.7.5 Modeling of Transport in Surface Water . . . . 221 5.7.5.1 Modeling of Discharges to Surface Water . . . . . . . . . . . . . . . . . . . . . . . . . 222 5.7.5.2 Modeling of Transport in Rivers and Streams . . . . . . . . . . . . . . . . . . . . . . . 223 5.7.5.3 Modeling of Transport in Lakes. . . . 224 5.7.5.4 Modeling of Transport in Sediment. 225

CONTENTS

/ xi

5.7.6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 Atmospheric Transport Analysis . . . . . . . . . . . . . . . . 226 5.8.1 Models for Estimating Suspension of Particulates . . . . . . . . . . . . . . . . . . . . . . . . . . . 227 5.8.1.1 Modified Mass Loading Model. . . . . 227 5.8.1.2 Resuspension Factor Model . . . . . . . 229 5.8.1.3 Resuspension Rate Model . . . . . . . . 233 5.8.2 Release of Gases by Diffusion . . . . . . . . . . . . 235 5.8.3 Advective Transport . . . . . . . . . . . . . . . . . . . . 236 5.8.4 Atmospheric Transport Models . . . . . . . . . . . 237 5.8.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 5.9 Biotic Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 5.9.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 5.9.2 Biotic Transport Processes . . . . . . . . . . . . . . . 242 5.9.2.1 Transport Enhancement . . . . . . . . . 243 5.9.2.2 Intrusion and Active Transport . . . 243 5.9.2.3 Secondary Transport . . . . . . . . . . . . 244 5.9.3 Pathways of Human Exposure . . . . . . . . . . . 244 5.9.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 5.10 Exposure Pathways and Radiological Impacts . . . . . 245 5.10.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 245 5.10.2 General Recommendations . . . . . . . . . . . . . . 249 5.10.3 Exposure Scenarios for Off-Site Members of the Public . . . . . . . . . . . . . . . . . . . . . . . . . . 250 5.10.3.1 Definition of Environmental Conditions and Living Habits . . . . . 250 5.10.3.2 Exposure Scenarios for Different Release and Transport Pathways . . 251 5.10.4 Exposure Pathway Models . . . . . . . . . . . . . . . 252 5.10.4.1 General Considerations . . . . . . . . . . 252 5.10.4.1.1 Components of Exposure Pathway Models . . . . . . . 253 5.10.4.1.2 Multiplicative-Chain Models. . . . . . . . . . . . . . . 255 5.10.4.1.3 Specific-Activity Models 256 5.10.4.2 Models of Foodchain Pathways . . . . 258 5.10.4.2.1 Terrestrial Foodchain Pathways . . . . . . . . . . . . 258 5.10.4.2.2 Aquatic Foodchain Pathways . . . . . . . . . . . . 258 5.8

xii / CONTENTS 5.10.5 Selection of Model Parameter Values . . . . . . 259 5.10.6 Sources of Generic Data on Model Parameter Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 5.10.6.1 Dose Coefficients . . . . . . . . . . . . . . . 260 5.10.6.1.1 Internal Exposure . . . . . 261 5.10.6.1.2 External Exposure . . . . . 268 5.10.6.1.3 Summary of Dose Coefficients . . . . . . . . . . . 272 5.10.6.2 Usage Factors . . . . . . . . . . . . . . . . . . 272 5.10.6.3 Transfer Factors for Foodchain Pathways . . . . . . . . . . . . . . . . . . . . . . 274 5.10.6.3.1 Terrestrial Foodchain Pathways. . . . . . . . . . . . . 274 5.10.6.3.2 Aquatic Foodchain Pathways. . . . . . . . . . . . . 278 5.10.6.3.3 Summary of Transfer Factors. . . . . . . . . . . . . . . 279 5.10.7 Uncertainties in Dose Assessment Models . . 279 5.10.7.1 Uncertainties in Transfer and Usage Factors. . . . . . . . . . . . . . . . . . . . . . . . 279 5.10.7.2 Significance of Uncertainties in Transfer and Usage Factors. . . . . . . 280 5.10.7.3 Sources of Uncertainty in Dose Coefficients . . . . . . . . . . . . . . . . . . . . 282 5.10.7.4 Significance of Uncertainties in Dose Coefficients . . . . . . . . . . . . . . . . . . . . 282 5.10.8 Approaches to Estimating Risk . . . . . . . . . . . 283 5.10.9 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 6. Inadvertent Human Intrusion . . . . . . . . . . . . . . . . . . . . 288 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 6.2 Role of Inadvertent Human Intrusion in Radioactive Waste Disposal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 6.2.1 Historical Perspective . . . . . . . . . . . . . . . . . . . 291 6.2.2 Regulatory Requirements for Near-Surface Disposal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 6.2.2.1 NRC Requirements. . . . . . . . . . . . . . 292 6.2.2.2 DOE Requirements . . . . . . . . . . . . . 294 6.3 Widely Used Scenarios for Inadvertent Intrusion . . . 295 6.3.1 Scenarios for Acute Exposure . . . . . . . . . . . . . 295

CONTENTS

6.4

6.5

6.6

6.7 6.8

/ xiii

6.3.1.1 Construction Scenario . . . . . . . . . . . 295 6.3.1.2 Discovery Scenario . . . . . . . . . . . . . . 296 6.3.1.3 Drilling Scenario . . . . . . . . . . . . . . . 297 6.3.2 Scenarios for Chronic Exposure . . . . . . . . . . . 297 6.3.2.1 Agriculture Scenario . . . . . . . . . . . . 297 6.3.2.2 Resident, Nonagriculture Scenario . 299 6.3.2.3 Postdrilling Scenario . . . . . . . . . . . . 299 6.3.2.4 Groundwater Pathway for Chronic Intrusion Scenarios . . . . . . . . . . . . . 300 6.3.3 Comparison of Standard Scenarios for Inadvertent Intrusion . . . . . . . . . . . . . . . . . . . 301 6.3.4 Other Scenarios for Inadvertent Intrusion . . 302 Selection of Site-Specific Scenarios for Inadvertent Intrusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304 6.4.1 Application of Widely Used Scenarios to Site-Specific Assessments . . . . . . . . . . . . . . . 304 6.4.2 Judgmental Factors in Selecting Exposure Scenarios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 6.4.3 Summary of Principles of Scenario Selection 308 Inputs to Dose Analyses for Inadvertent Intruders . 308 6.5.1 Time of Occurrence of Intrusion . . . . . . . . . . 309 6.5.2 Radioactive Decay . . . . . . . . . . . . . . . . . . . . . . 309 6.5.3 Waste Dilution Following Disposal . . . . . . . . 310 6.5.4 Consideration of Radionuclide Transport . . . 312 Outputs of Dose Analyses for Inadvertent Intruders 313 6.6.1 Scenario Dose Conversion Factors . . . . . . . . 313 6.6.2 Waste Acceptance Criteria Based on Intruder Dose Assessment . . . . . . . . . . . . . . . . . . . . . . 314 Effects of Inadvertent Intrusion on Off-Site Releases of Radionuclides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317

7. Uncertainty, Sensitivity and Importance Analysis . . 320 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320 7.2 Description of Importance Analysis . . . . . . . . . . . . . . 321 7.3 Purpose of Importance Analysis . . . . . . . . . . . . . . . . . 322 7.4 Nature of Uncertainties in Performance Assessment 323 7.4.1 Characteristics of Uncertainties . . . . . . . . . . 323 7.4.1.1 Type-A and Type-B Uncertainties . 324 7.4.1.2 Classification of Model Uncertainties . . . . . . . . . . . . . . . . . . 325 7.4.2 Uncertainty in Models . . . . . . . . . . . . . . . . . . 326

xiv / CONTENTS 7.4.3 7.4.4

7.5

7.6

7.7

7.8

Uncertainty in Future Site Conditions . . . . . 327 Uncertainty in Model Parameters . . . . . . . . . 328 7.4.4.1 Measurement Errors . . . . . . . . . . . . 328 7.4.4.2 Insufficient Data. . . . . . . . . . . . . . . . 329 7.4.4.3 Dependence of Measurements on Scale . . . . . . . . . . . . . . . . . . . . . . . 330 Mathematical Methods of Treating Uncertainty . . . . 330 7.5.1 Introduction to Mathematical Methods . . . . . 331 7.5.2 Propagation of Model Uncertainty . . . . . . . . . 332 7.5.3 Propagation of Future Uncertainty . . . . . . . . 334 7.5.4 Propagation of Parameter Uncertainty . . . . . 338 7.5.4.1 Deterministic Methods. . . . . . . . . . . 338 7.5.4.2 Probabilistic Methods. . . . . . . . . . . . 340 7.5.4.2.1 Monte-Carlo Analysis. . . 340 7.5.4.2.2 Perturbation Analysis . . 343 7.5.4.2.3 Possibilistic Analysis . . . 344 Role of Sensitivity Analysis in Importance Analysis . 344 7.6.1 Need for Sensitivity Analysis . . . . . . . . . . . . . 344 7.6.2 Methods of Parameter Sensitivity Analysis . 345 Application of Uncertainty Analysis to Importance Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 346 7.7.1 General Structure of Uncertainty Analysis . . 346 7.7.2 Evaluation of Different Methods of Uncertainty Analysis . . . . . . . . . . . . . . . . . . . 347 7.7.3 Evaluation of Approaches to Importance Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351

8. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353 8.1 Purpose and Scope of Performance Assessment . . . . 353 8.2 Basic Elements of Performance Assessment . . . . . . . 354 8.2.1 Development of Conceptual Models . . . . . . . . 355 8.2.2 Development and Application of Mathematical and Physical Models . . . . . . . . 356 8.2.3 Integration and Interpretation of Results . . . 357 8.3 Components of Performance Assessment Modeling . . 358 8.3.1 Cover Performance and Infiltration . . . . . . . . 358 8.3.2 Performance of Concrete Barriers . . . . . . . . . 359 8.3.3 Source Term . . . . . . . . . . . . . . . . . . . . . . . . . . 360 8.3.4 Unsaturated Zone Flow and Transport . . . . . 361 8.3.5 Aquifer Flow . . . . . . . . . . . . . . . . . . . . . . . . . . 363

CONTENTS

/ xv

8.3.6

8.4 8.5 8.6 8.7

Radionuclide Transport in Groundwater and Surface Water . . . . . . . . . . . . . . . . . . . . . . . . . 364 8.3.7 Atmospheric Transport . . . . . . . . . . . . . . . . . 366 8.3.8 Biotic Transport . . . . . . . . . . . . . . . . . . . . . . . 366 8.3.9 Exposure Pathways and Radiological Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367 8.3.10 Overview of Components of Performance Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . 368 Inadvertent Human Intrusion . . . . . . . . . . . . . . . . . . 369 Uncertainty, Sensitivity and Importance Analysis . . 370 Iterative Nature of Performance Assessment . . . . . . 371 Reasonable Assurance of Compliance with Performance Objectives . . . . . . . . . . . . . . . . . . . . . . . . 372

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396 The NCRP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439 NCRP Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458

Executive Summary In the United States, low-level radioactive waste is defined as any radioactive waste arising from operations of the nuclear fuel cycle that is not classified as high-level waste (including spent fuel when it is declared to be waste), transuranic waste, or uranium or thorium mill tailings. Low-level waste is generated in many commercial, defense-related, medical, and research activities. Owing to its definition only by exclusion and its many sources, low-level waste occurs in a wide variety of physical and chemical forms, and it contains a wide range of concentrations of many different radionuclides. Most low-level waste, except relatively small volumes that contain high concentrations of radionuclides with half-lives on the order of 30 y or longer, is intended for disposal in facilities located on or near the ground surface. Decisions about acceptable nearsurface disposals of low-level waste are based in large part on the need to comply with regulatory requirements that have been established by the U.S. Nuclear Regulatory Commission (NRC) and the U.S. Department of Energy (DOE). These regulatory requirements include performance objectives that define allowable radiation exposures of the public at future times due to releases of radionuclides to the environment. In order to determine whether a particular facility complies with regulatory performance objectives for long-term protection of the public, a performance assessment of the facility must be conducted. As used in this Report: Performance assessment is an iterative process involving site-specific, prospective modeling evaluations of the postclosure time phase of near-surface disposal systems for low-level waste with two primary objectives: • to determine whether reasonable assurance of compliance with quantitative performance objectives can be demonstrated; and • to identify critical data, facility design, and model development needs for defensible and cost-effective licensing decisions and to develop and maintain operating limits (i.e., waste acceptance criteria). 1

2 / EXECUTIVE SUMMARY This definition emphasizes that performance assessment focuses primarily on a decision about compliance with performance objectives, rather than the much more difficult problem of predicting actual radiological impacts on the public at far future times. The purpose of this Report is provide a review of concepts underlying performance assessments of near-surface disposal facilities for low-level waste and approaches to conducting such assessments. This review includes discussions on the nature and scope of performance assessment, accepted approaches to conducting all aspects of a performance assessment, and unresolved issues in conducting performance assessments and applying the results. Challenges in conducting and defending performance assessments at specific sites also are emphasized. An understanding of general principles of performance assessment is important, because such understanding can lead to an efficient process and defensible product and can reduce the potential for misinterpretation of results. General principles of performance assessment are summarized as follows: • Performance assessment should be an iterative, flexible process of integrating modeling, data collection, and design activities in a manner that identifies those aspects of engineered and natural barriers in a disposal system of importance to a decision about compliance with regulatory performance objectives. The performance assessment process is important during all time phases of a facility from site selection and facility design through operations and postclosure monitoring and surveillance. • Performance assessment is a process that is intended to provide reasonable assurance of compliance with performance objectives; absolute assurance of compliance generally is not attainable by any means unless disposal of only very small quantities of radionuclides is allowed. • Since there is substantial uncertainty in models and important parameters used in performance assessment and some physical, chemical, and biological processes that affect the long-term performance of a disposal system may not be well understood, use of subjective scientific judgment is an essential aspect of performance assessment. Therefore, a variety of results that investigate the consequences of different plausible assumptions should be presented, rather than a single projected outcome. • An integration and interpretation of assumptions and results, in which conceptual models of the performance of a disposal

EXECUTIVE SUMMARY

/ 3

system and their bases and the results of calculations are presented in a manner that reflects the many judgments involved and the importance of different aspects of an assessment to a demonstration of compliance with regulatory performance objectives, is a critical aspect of performance assessment. Quantitative performance objectives for near-surface disposal of low-level waste in NRC and DOE regulations are expressed in terms of limits on annual dose to members of the public, either equivalent dose to specific organs or tissues or effective dose equivalent. Therefore, projections of maximum concentrations of radionuclides in the environment at assumed locations of exposure are required. Although regulations are well established, a number of issues regarding performance objectives and their implementation are not yet fully resolved or are controversial. These include (1) the time period for compliance and the weight, if any, that should be given to projections of performance beyond the compliance period in determining acceptable disposals, (2) whether projected doses due to radon are included in performance objectives, which is potentially important in determining acceptable disposals of waste that contains radium, thorium, or uranium, (3) whether performance objectives should include a separate requirement for protection of groundwater resources in accordance with drinking water standards, which is important when drinking water standards would be more restrictive in determining allowable releases of many radionuclides than the existing performance objectives that apply to all exposure pathways combined, and (4) interpretation of performance objectives for compliance purposes—i.e., how highly uncertain results of performance assessments should be compared with fixed performance objectives in judging compliance. At most sites, movement of water is considered to be the most important means by which radionuclides may be released from a disposal facility and transported to locations where exposures of the public could occur. Even for the simplest types of near-surface facilities (e.g., an unlined trench with backfill and cap consisting of earthen materials), performance assessment requires an integration of results of a number of different types of models to provide an overall description of the performance of a disposal system. The usual approach to performance assessment is to model various components of the system separately and then link the components in a sequential fashion, with appropriate boundary and continuity conditions, to describe overall system performance. The different components that generally must be considered in a performance assessment are the following:

4 / EXECUTIVE SUMMARY • an analysis of cover performance and infiltration, the primary purpose of which is to estimate the flux of water (i.e., incident precipitation) that infiltrates through a natural or engineered cover system to locations of disposed waste or an engineered barrier (e.g., a concrete structure) above the waste; the performance of a cover system in inhibiting atmospheric releases of radionuclides in gaseous form also can be important at some sites and for some wastes; • an analysis of the performance of concrete barriers (e.g., vaults, modular canisters, or bunkers) that are used in many disposal facilities to enhance containment of low-level waste by (1) providing structural support for an earthen or engineered cover system, (2) delaying and inhibiting inflow of water to locations of disposed waste, (3) supplying additional adsorbing materials to retard movement of radionuclides into the surrounding environment, and (4) delaying and inhibiting release of radionuclides in leachate from a facility; • an analysis of the source term, the purpose of which is to estimate the rate of release of radionuclides from a disposal facility into the surrounding environment, either the vadose zone when releases are in liquid form or the atmosphere when releases are in gaseous form, by considering rates of release from waste forms and waste containers, transport within a disposal facility, and transport through any engineered barriers used in constructing the facility; • an analysis of flow and transport in the unsaturated (vadose) zone, which uses results of source-term modeling of liquid releases as input and provides estimates of releases into groundwater (zone of saturation); • an analysis of flow and transport in groundwater (zone of saturation), which uses results of modeling of flow and transport in the vadose zone, as well as any additional direct recharge to an aquifer that may occur due to runoff from a cover, as input and provides estimates of concentrations of radionuclides in groundwater at assumed locations of exposure; an analysis of flow and transport in surface water also may be required when groundwater into which releases occur discharges to the surface at locations close to a facility; • an analysis of atmospheric transport, which uses results of source-term modeling of gaseous releases or other means of transport of buried waste to the ground surface and release to the atmosphere as input and provides estimates

EXECUTIVE SUMMARY

/ 5

of airborne concentrations of radionuclides at assumed locations of exposure; • an analysis of biotic transport, which considers actions of plants and animals that could affect transport of buried waste to assumed locations of exposure; biotic transport is not often considered explicitly in performance assessment, but it can serve to enhance release and transport of radionuclides and can be important at some sites; and • an analysis of exposure pathways and radiological impacts, which uses results of environmental transport models (i.e., groundwater or surface water flow and transport, atmospheric transport, or biotic transport models) that estimate concentrations of radionuclides in environmental media at assumed locations of exposure as input and provides estimates of transport through various exposure pathways to human receptors and the resulting radiation doses. Atmospheric and biotic transport often are considered to be unimportant compared with transport in water. However, these processes can be important in some environments and for some facility designs, and their importance generally should be evaluated in site-specific analyses. NRC and DOE regulations also require that near-surface disposal facilities provide protection of hypothetical inadvertent intruders who are assumed to come onto a disposal site after loss of institutional control and access disposed waste by such means as excavating to construct a foundation for a home or drilling. Protection of inadvertent intruders at sites licensed under NRC regulations is provided by the NRC’s waste classification system, which specifies limits on concentrations of radionuclides in Class-A, -B, and -C waste and technical requirements on disposal of waste in each class. At DOE sites, a site-specific assessment of potential impacts on inadvertent intruders is required for the purpose of establishing limits on concentrations of radionuclides; these limits can vary greatly depending on site conditions and the design of a facility. Under either regulations, compliance with a requirement to protect inadvertent intruders is based on analyses of potential radiological impacts in assumed intrusion scenarios. Standard scenarios that are often used are reviewed in this Report. Given that there are separate requirements to protect members of the public and inadvertent intruders, determinations of acceptable near-surface disposals of low-level waste essentially involve achieving a balance between acceptable releases of radionuclides beyond the site boundary and acceptable residual concentrations in a disposal

6 / EXECUTIVE SUMMARY facility after loss of institutional control. If excavation into waste and residence on a disposal site in a homesteader scenario are considered to be credible occurrences, criteria that define adequate protection of inadvertent intruders usually are more restrictive in determining acceptable disposals than performance objectives for protection of the public or the environment. Therefore, selection of credible intrusion scenarios can be very important in determining acceptable disposals in near-surface facilities. Performance assessments generally must consider uncertainties in models and parameter values, which result in uncertainty in results of modeling (e.g., projected doses to the public), and the sensitivity of model outputs to changes in assumptions and variations in parameters. Since the primary purpose of performance assessment is to provide a demonstration of compliance with regulatory performance objectives, a particular kind of uncertainty and sensitivity analysis, which is termed importance analysis, is emphasized in this Report. Importance analysis is an integration and interpretation of results obtained from the performance assessment process for the purpose of identifying assumptions and parameters which, when changed within credible bounds, can affect a decision about regulatory compliance. This type of analysis, which focuses on uncertainties and sensitivities that are important to a decision about compliance with performance objectives, is different from more traditional uncertainty and sensitivity analyses, which are concerned with representing uncertainty in the actual behavior of a disposal system and outcomes of waste disposal. An understanding of the distinction between importance analysis and traditional sensitivity and uncertainty analysis is important in conducting performance assessments efficiently and defending the results. Models of varying degrees of sophistication and complexity have been developed for all components of a performance assessment. However, detailed modeling of all aspects of the performance of a disposal system is beyond current capabilities and, indeed, is not required to achieve defensible results and robust decisions about the acceptability of waste disposals. Since there are a large number of radionuclides in low-level waste and a large number of potential pathways for transport and exposure, simple screening analyses to select for further analysis only those radionuclides and pathways that contribute significantly to projected doses to the public are an important initial step in making a performance assessment tractable. Once radionuclides and pathways are selected for further analysis, many stylized and simplifying assumptions normally are used in performance assessment in the interest of expediency. Examples of such assumptions include the following:

EXECUTIVE SUMMARY

/ 7

• The potential importance of future climate change on infiltration and release and transport in water either is not considered or is modeled by assuming an abrupt change to an expected climate at some future time. • Infiltration through natural or engineered cover systems usually is modeled by assuming steady-state conditions, and transients that might occur as a result of episodic precipitation events are ignored. Failure of a cover system usually is modeled by assuming an instantaneous change or a series of instantaneous changes to natural conditions at some future time. • Degradation or failure of engineered barriers, either physical or chemical, usually is modeled by assuming an instantaneous change from an initial condition to a failed state at some future time or a constant rate of failure. • When there are highly heterogeneous distributions of radionuclides and a multiplicity of waste forms in a disposal facility, source terms usually are modeled by averaging radionuclide distributions over individual disposal units and assuming no more than a few idealized representations of waste forms. • The physical structure of unsaturated and saturated geologic media and their geochemical properties usually are assumed to be homogeneous and isotropic. • A graded approach to modeling flow and transport in the unsaturated (vadose) zone may be taken in which all or parts of the unsaturated zone are ignored (e.g., releases from a disposal facility are assumed to directly enter an underlying aquifer) or, less conservatively, a unit-gradient model that assumes steady-state flow and a flow rate equal to the infiltration rate is used. • The interface between models of flow in the vadose and saturated zones often is represented by simple boundary conditions (e.g., zero pressure head). • A linear sorption isotherm, described by the equilibrium solid/solution distribution coefficient (Kd), usually is assumed to represent all geochemical effects on transport in the disposal facility following release from a waste form and waste package and transport in the vadose and saturated zones. • All physical and chemical processes that affect release and transport of radionuclides often are assumed to be independent of radionuclide concentrations in waste, and the performance of the disposal system (e.g., projected dose to the public) is assumed to depend linearly on those concentrations.

8 / EXECUTIVE SUMMARY Challenges in conducting performance assessments and defending the results in a regulatory setting, which is tantamount to defending important assumptions, increase as the quantities of radionuclides that are intended for disposal in a facility increase and compliance with performance objectives cannot be demonstrated by using clearly conservative (pessimistic) models for highly uncertain components of performance. The lack of relevant site-specific data or data over time and spatial scales of importance is a frequent concern. Some of the challenges in modeling the performance of a disposal system when water is the medium in which releases are assumed to occur and a high level of performance of natural and engineered barriers is required to demonstrate compliance with performance objectives are summarized as follows: • Cover performance and infiltration: It can be difficult to justify that an engineered cover will perform as designed and constructed to control infiltration over time periods much beyond the period of institutional control. • Concrete barriers: There is little relevant data to predict the structural integrity and load-bearing capabilities of concrete over long time periods, so it is difficult to justify assumptions about structural integrity at times beyond several hundred years. Assumptions about infiltration through degraded concrete structures, including the relative importance of flow in fractures and pores, can be an important issue. • Source term: Inventories of radionuclides that often are expected to be the most important in releases to groundwater (e.g., 14C, 99Tc, and 129I) can be difficult to estimate. Modeling of releases can be challenging when radionuclide distributions are heterogenous and several waste forms are used. Although grout provides a homogeneous waste form for liquid wastes, modeling of long-term changes in hydrologic and geochemical properties of grout waste forms can be difficult when there are little relevant data and judgment must be relied on. • Unsaturated (vadose) zone flow and transport: Modeling at a detailed level is data-intensive and difficult to defend at specific sites owing to the complex, nonlinear relationships between moisture content, pressure (suction) head, and hydraulic conductivity and their dependence on soil type. A defensible conceptual model of flow and transport in unsaturated fractured rock is not yet available.

EXECUTIVE SUMMARY

/ 9

• Saturated zone flow and transport: A groundwater velocity field is not directly measurable but must be generated using a model that is based on data on hydraulic head from monitoring wells and data on pump or core hydraulic conductivity tests. A velocity field so generated is non-unique, and multiple interpretations of data, with different effects on performance, may be reasonable at any site. Other issues of potential concern in modeling saturated zone flow include assumptions about boundary conditions, use of transient data to model steady-state conditions, the spatial scale and heterogeneity of a velocity field, modeling of flow in fractured rock, and the applicability (scaling) of laboratory data to field conditions. Issues in modeling transport include the simplistic nature of the Kd concept, justification of assumed Kd values at specific sites, treatments of diffusion and dispersion, and a lack of site-specific data on dispersivities. In contrast, modeling of atmospheric transport and exposure pathways and radiological impacts rarely is difficult or controversial, in part because the relevant processes are well understood and there are extensive studies to validate models normally used in performance assessments. A central issue that must be confronted in all performance assessments is whether simple and clearly conservative models should be used in demonstrating compliance with performance objectives or whether more complex and rigorous modeling should be undertaken in an effort to provide more realistic projections of outcomes at times far into the future. Either approach may be desirable for many reasons, and both have their difficulties. The point of view taken in this Report is that an appropriate balance between conservatism and more realistic approaches to performance assessment is largely a matter of judgment that should be applied on a site-specific basis. All performance assessments should attempt to incorporate some degree of realism to demonstrate an appropriate level of understanding of the long-term performance of disposal systems. The goal should be to provide a cost-effective and defensible assessment that is commensurate with the hazards posed by wastes that are intended for disposal at a specific site. At some sites with highly desirable characteristics, use of simple and conservative models for some aspects of system performance may not affect a decision about regulatory compliance. At other sites, however, efforts at more realistic modeling may be required. At any site, it is important to recognize that performance assessment is conducted to inform decisions about the

10 / EXECUTIVE SUMMARY acceptability of waste disposals, and that it is not necessary to obtain realistic projections of outcomes to render such decisions in a defensible manner. Although performance assessment involves a significant amount of subjective scientific judgment and there are important limitations in regard to predicting actual outcomes, these factors do not compromise the essential role of performance assessment in regulatory decision making.

1. Introduction Low-level radioactive wastes are generated in a variety of commercial, defense-related, medical, and research activities. Most low-level waste1 generated in the United States, except relatively small volumes that contain the highest concentrations of radionuclides, is intended for disposal in facilities located on or near the ground surface. This intention is based largely on generic analyses, such as those performed by the U.S. Nuclear Regulatory Commission (NRC, 1981a; 1982a), which indicated that near-surface facilities should be capable of providing adequate protection of public health and the environment at times far into the future. Decisions about acceptable near-surface disposals of low-level waste at specific sites are made on the basis of applicable laws and regulations. Such decisions take into account many scientific, technical, economic and social factors. Regulatory requirements that apply to disposal of low-level waste in near-surface facilities include performance objectives that define allowable radiation exposures of the public at future times. In order to determine whether a particular disposal facility complies with performance objectives, a performance assessment of the facility must be conducted. A performance assessment essentially is a prospective evaluation of potential radiation exposures of the public at times after a disposal facility is closed. 1.1 Purpose of Report The purpose of this Report is to provide a review of concepts underlying performance assessments of near-surface disposal facilities for low-level waste and approaches to conducting such assessments. This review includes discussions on the nature and scope of performance assessment, accepted approaches to conducting all aspects of a performance assessment, and unresolved issues in conducting performance assessments and applying the results. 1 In this Report, the term “low-level waste” is used to refer to low-level radioactive waste, and similarly with “high-level waste.” Since the terms “low-level waste” and “high-level waste” are not used to describe wastes that contain hazardous chemicals, their use to describe different radioactive wastes should not cause confusion.

11

12 / 1. INTRODUCTION Many discussions in this Report also apply to deep geologic repositories, which are intended for disposal of high-level and transuranic radioactive wastes. In particular, the conceptual foundations and general principles of performance assessment described in this Report are applicable to geologic repositories, even though models used in performance assessments for geologic repositories and applicable regulatory criteria may differ from those for nearsurface facilities. Discussions in this Report generally do not apply to disposal of large volumes of uranium mill tailings, since this activity is conducted in accordance with a different set of legal and regulatory requirements that do not call for prospective evaluations of long-term radiological impacts on the public in determining acceptable disposal practices at specific sites.2 This Report focuses on performance assessments of proposed or currently operating facilities for near-surface disposal of low-level waste. Assessments to evaluate potential radiological impacts of past disposal practices at inactive or abandoned sites for the purpose of determining whether there is a need for site remediation to mitigate those impacts are outside the scope of this Report. Although technical aspects of such assessments may have much in common with performance assessments discussed in this Report, remediation of past disposal practices is addressed under a different legal and regulatory framework, and the need to assess potential impacts of past disposal practices at times far into the future is not yet established. 1.2 Scope of Report This Report is intended to provide a general overview of performance assessment, and to identify sources of current information about performance assessment and the different scientific disciplines involved in conducting performance assessments. An important emphasis is the level of detail that has been shown by past experience to be required in performance assessments of specific disposal facilities. It is not the intent of this Report to provide a comprehensive review of all aspects of performance assessment that may be important at a particular facility. 2

Performance assessments to evaluate potential radiation exposures of the public at mill-tailings disposal sites were conducted by the U.S. Environmental Protection Agency in developing regulatory requirements in the form of design objectives that apply at all sites (EPA, 1982; 1983). The acceptability of mill-tailings disposal at any site then is demonstrated by meeting the design objectives.

1.2 SCOPE OF REPORT

/ 13

Discussions in this Report also consider a number of policy issues that affect conduct of performance assessment. Examples include the time period for complying with performance objectives, application of drinking water standards to protection of groundwater resources, and whether impacts of inadvertent human intrusion on the normal performance of a near-surface disposal facility are evaluated in demonstrating compliance with performance objectives. It is not the purpose of this Report, however, to present recommendations for resolution of any such issues. Similarly, although the importance of social, political and economic factors is recognized and discussed to a limited extent, this Report is not concerned with addressing those factors and integrating them into the process of developing and licensing new facilities. This Report is organized as follows. Sections 2 and 3 provide background information of importance to understanding discussions on technical approaches to performance assessment, including a definition of performance assessment and a discussion of goals and principles of performance assessment (Section 2) and discussion of the broader context for performance assessment (Section 3). Sections 4 and 5 present current views on suitable technical approaches to conducting performance assessments. Section 4 presents a conceptual framework for conducting performance assessments, including general discussions of a recommended approach to performance assessment, principles for treatment of uncertainty, and development of confidence in results as part of the performance assessment process. These discussions emphasize advantages of an iterative approach to performance assessment involving interactions among site characterization and data collection, facility design, and modeling activities. Section 5 discusses models and databases that are considered suitable for use in performance assessments of near-surface disposal facilities for low-level waste, as well as important sources of uncertainty in those models. The approach taken in this Report is to divide performance assessment into several modules, each of which represents a particular aspect of a disposal system (e.g., longevity of engineered barriers, release of radionuclides from a disposal facility, transport of radionuclides in the environment, pathways of human exposure). Specific computer codes that may be used in performance assessment are rarely discussed in this Report. A code is no more than a particular numerical representation of models embodied in it, and it is far more important to focus on selection of credible models than on particular implementations of those models. Sections 6 and 7 present discussions on topics that are important to performance assessment but do not fit in discussions in

14 / 1. INTRODUCTION Section 5 on modeling of particular aspects of a disposal system. These topics include assessments of inadvertent human intrusion (Section 6) and treatment of uncertainty and sensitivity by means of importance analysis (Section 7). Finally, Section 8 provides summary comments on performance assessment, with particular emphasis on technical challenges in conducting performance assessments. 1.3 Related NCRP Recommendations The National Council on Radiation Protection and Measurements (NCRP) previously issued three reports that are relevant to performance assessments of near-surface disposal facilities for low-level waste. NCRP Report No. 76 (NCRP, 1984a) presents recommendations on models and databases for use in assessing radiation doses to the public following release of radionuclides to the atmosphere, surface water, or groundwater. Models of transport in the environment and exposure pathways discussed in Report No. 76 are relevant to performance assessment. However, some approaches to modeling environmental transport and exposure pathways discussed in that report may be more detailed than necessary or useful in performance assessment. For example, modeling of episodic releases and seasonal dependencies of exposures, which can be important in assessing doses resulting from actual or expected releases from operating facilities and in reconstructing doses from past releases, are not needed in prospective assessments of the performance of disposal facilities at far future times. NCRP Report No. 123 (NCRP, 1996a; 1996b) presents recommendations on screening models to assess impacts of releases of radionuclides to the environment, including screening models of atmospheric transport, transport in surface water, disposal of radionuclides in the ground, and transport in terrestrial and aquatic food chains. Those models are generic and are intended to be applied at any site. Screening involves use of simple models that employ clearly conservative (pessimistic) assumptions. Such models can be used to demonstrate that operating facilities comply with regulatory requirements or to eliminate unimportant radionuclides and exposure pathways from further consideration in a dose assessment. Screening models described in Report No. 123 may, in many cases, provide a suitable first iteration for related aspects of performance assessments of low-level waste disposal facilities, to be followed by use of more site-specific models of increasing sophistication. However, when generic screening models are used in performance assessment, including models recommended by NCRP,

1.3 RELATED NCRP RECOMMENDATIONS

/ 15

proper justification for their use at a disposal site of concern must be provided. Screening models for disposal of radionuclides in the ground, which include models for releases to groundwater and exposures to buried waste by several pathways assuming loss of institutional control at 10 y after facility closure, and transport of radionuclides in terrestrial and aquatic food chains could be particularly useful in performance assessment. Models of transport in surface water, which are discussed in this Report, also could be useful at some disposal sites. NCRP Report No. 129 (NCRP, 1999) presents recommendations on screening levels (concentrations) of radionuclides in surface soil. For each of several assumptions about future uses of contaminated land and assumptions about exposure pathways and pathway models in each land-use scenario, radionuclide-specific screening factors expressed as annual effective doses per unit activity concentration in surface soil [Sv (Bq kg)–1] are derived. The review and evaluation of exposure pathway models used to derive screening levels in surface soil essentially updates recommendations in NCRP Report No. 76 (NCRP, 1984a). In contrast to the approach to screening in NCRP Report No. 123 (NCRP, 1996a; 1996b), which uses point values of all parameters, the screening analysis in Report No. 129 incorporates assumptions about uncertainties in all parameters used in estimating annual effective doses. Median values and 95th percentiles of uncertain screening factors are derived, and the 95th percentiles are used to obtain screening levels in surface soil that correspond to an annual effective dose of 0.25 mSv. Thus, the intent is that at any site at which a particular land-use scenario is considered appropriate, application of the derived screening levels should ensure that, with a high level of confidence, annual doses to members of the public who might occupy contaminated land would be less than the assumed dose criterion. Screening levels developed in Report No. 129 could be particularly useful in performance assessments in regard to assessing doses following dispersal of radionuclides over the land surface (e.g., by irrigation with contaminated groundwater) and in assessing doses from inadvertent intrusion into a facility.

2. Definition and Principles of Performance Assessment This Section presents a general discussion of objectives and principles of performance assessment, and the definition of performance assessment used in this Report. The role of performance assessment in radiation protection of the public is discussed further in Section 3.3. 2.1 Nature of Performance Assessment Performance assessment is concerned with prospective evaluations of waste disposal systems. A disposal system is comprised of multiple engineered and natural barriers (ICRP, 1998), which are intended to inhibit movement of radionuclides from locations of waste emplacement into the general environment beyond the boundary of a disposal site and to inhibit intrusion into waste by water, plants, burrowing animals, and humans. Examples of engineered barriers include concrete disposal vaults, impermeable caps above buried waste that are constructed with man-made or natural materials, and grouting of waste to control ingress and egress of water and release of radionuclides. An important natural barrier at any near-surface disposal site is the ability of native soils to sorb radionuclides and, thus, retard their migration. Performance assessment is an essential activity in developing disposal facilities and gaining approval by regulatory authorities, because it provides the only available link between measurable properties of a disposal facility or waste and potential long-term radiological impacts of waste disposal on the public. This link is crucial because, unlike many more familiar engineered systems, there usually are not intuitively evident relationships between measurable properties of a disposal system and consequences of waste disposal. For instance, one cannot draw general conclusions that a geologic stratum with high permeability is either favorable or unfavorable as a disposal site. Performance assessment is used to integrate available information about the long-term behavior of a disposal facility for the purpose of obtaining a defensible 16

2.1 NATURE OF PERFORMANCE ASSESSMENT

/ 17

conclusion regarding the ability of the facility to protect the public in accordance with applicable regulatory criteria. Given the complexities associated with assessments of natural and engineered disposal systems, a critical part of this integration process involves iterative feedback to identify data, modeling and design needs for the process of gaining regulatory approval of a facility. Performance assessment also can be important after regulatory authorities have approved a disposal facility. Additional data may be gathered during disposal operations or after facility closure, and a performance assessment may need to be maintained and updated until institutional control is relinquished. Thus, performance assessment should be viewed as a tool for risk management, not just as a means of gaining approval of a facility by regulatory authorities. Unlike safety analyses of nuclear reactors, airplanes and bridges, for example, that address relatively short-term behavior of engineered systems on the basis of large amounts of empirical data, performance assessment addresses the long-term behavior of complex systems for which relatively little empirical data are available.3 It is, therefore, important to distinguish performance assessment from other, more standard engineering problems. The primary distinction is that performance assessment relies more heavily on judgments, which are necessitated by a lack of observations of the long-term performance of disposal facilities. Indeed, judgments are essential and form the basis of calculations of long-term performance. There needs to be a frank acknowledgment that required judgments do not always have a firm basis in relevant measurements on real systems. Some judgments will be based on established principles of science and engineering, but others will be little more than informed opinion. Performance assessments commonly involve a combination of simple screening models and more complex analyses. Screening models may be used, for example, to eliminate unimportant radionuclides and transport or exposure pathways from further consideration in an assessment or to address in a clearly conservative 3Experience

with low-level waste disposal facilities that have operated in the past can provide useful information for performance assessments of currently operating or planned facilities. For example, observations at historical facilities that did not perform adequately can provide information on designs and environmental conditions that are not likely to prove satisfactory. However, the number of such facilities is limited, and most of them did not include engineered barriers and waste forms of the type frequently used in currently operating or planned facilities.

18 / 2. DEFINITION AND PRINCIPLES OF PERFORMANCE ASSESSMENT manner aspects of the performance of a disposal system for which little information is available and modeling is difficult, such as transport of radionuclides in unsaturated soil. In conducting a performance assessment, considerable effort often is spent in developing an appropriate balance between use of realistic and conservative models; this important issue is discussed further in Section 2.4. The key to conducting performance assessments is to structure an analysis so that it is defensible on the basis of available information on the long-term performance of a disposal facility. 2.2 Definition of Performance Assessment The definition of performance assessment in this Report emphasizes several concepts that are important to the conduct, interpretation and use of performance assessments, and that distinguish performance assessment from typical engineering analyses. As used in this Report: Performance assessment is an iterative process involving site-specific, prospective modeling evaluations of the postclosure time phase of near-surface disposal systems for low-level waste with two primary objectives: • to determine whether reasonable assurance of compliance with quantitative performance objectives can be demonstrated; and • to identify critical data, facility design, and model development needs for defensible and cost-effective licensing decisions and to develop and maintain operating limits (i.e., waste acceptance criteria). This definition emphasizes that, for purposes of this Report, performance assessment is a process that focuses primarily on a decision about compliance with regulatory requirements, rather than the much more difficult problem of predicting actual outcomes of waste disposal (i.e., actual radiological impacts on the public and the environment). The following paragraphs summarize key aspects of this definition. Further discussions are contained in the following section. The term “iterative process” refers to an expectation that performance assessment probably will require two or more sequential sets of calculations as additional data are collected during site characterization activities and, perhaps, during facility operations and postclosure monitoring of a facility. In general, an iterative

2.2 DEFINITION OF PERFORMANCE ASSESSMENT

/ 19

approach helps to minimize a tendency for performance assessment to become focused on details of models without considering the relative importance of specific aspects of an assessment to the primary concern of demonstrating compliance with regulatory requirements; it also emphasizes that performance assessment is more than a modeling exercise. Modeling of disposal systems must be conducted in coordination with data collection, facility design activities, and long-term monitoring. When applied efficiently, the performance assessment process provides input to management decisions regarding needs for additional data collection, design activities, and monitoring. In general, iteration of performance assessments is necessary when new information becomes available that could affect a licensing decision, but care should be taken to avoid unnecessary iterations that would impede the licensing process.4 The term “site specific” refers to the need to base performance assessments on relevant data for a facility being considered. Generic performance assessments have been useful in developing regulations, investigating the cost-effectiveness of various designs, and building confidence that near-surface disposal facilities at well-chosen sites should be protective of public health and the environment, but models used in generic assessments should not be applied to specific sites and facility designs without proper justification. The optimum approach to collecting site-specific data depends primarily on conditions at a site, the design of a facility, and the need for additional data in demonstrating compliance with performance objectives. The term “postclosure” emphasizes that performance assessment is concerned only with the consequences of waste disposal following emplacement of all waste and closure of a facility.5 Routine and accidental releases of radionuclides prior to closure are better addressed using standard safety assessments of site operations and monitoring. However, performance assessment can be helpful in guiding monitoring activities during the preclosure time phase, and such monitoring can provide useful information on postclosure performance. 4In this Report, the term “licensing” refers generally to a process of obtaining approval of a disposal facility by any regulatory authorities; it does not necessarily refer only to approval by NRC or an Agreement State. 5This may not be the case at a geologic repository for disposal of spent nuclear fuel and high-level wastes, because such a facility may remain open for decades after waste emplacement is complete.

20 / 2. DEFINITION AND PRINCIPLES OF PERFORMANCE ASSESSMENT The term “prospective modeling evaluations” is used rather than “modeling predictions” to emphasize that performance assessment is directed primarily at building sufficient understanding of the long-term behavior of a disposal system to make a defensible decision about compliance with performance objectives. Such an understanding is achieved by identifying those assumptions about the performance of engineered and natural barriers in a disposal system that are most important to obtaining the projected outcome and then showing that plausible changes in those assumptions would not affect a decision about compliance. The term “reasonable assurance” emphasizes the inexact nature of performance assessment and the crucial role of judgment in conducting performance assessments and in evaluating results. Further discussion of the important concept of reasonable assurance is provided in Section 3.5.3. The term “operating limits” is included in the definition to emphasize that results of performance assessment are used to identify acceptable operating conditions at a disposal facility, especially waste acceptance criteria in the form of limits on concentrations or inventories of radionuclides and requirements on physical and chemical properties of waste forms. Operating limits may change during the period of waste disposal in response to improved understanding of system behavior or changes in facility design, properties of waste, waste containers, waste emplacement, or the closure concept. Any changes in operating limits must be supported by revisions of a performance assessment to incorporate such new information. 2.3 General Principles of Performance Assessment This Section presents a discussion of general principles of performance assessment, based on the definition given in the previous section. It is important to recognize that performance assessment is essential to management of a disposal site and waste disposal operations. An understanding of principles of performance assessment can reduce the potential for misinterpretation of results, and can lead to an efficient process and defensible product. General principles of performance assessment discussed in this Section are identified as follows: • Performance assessment should be an iterative, flexible process of integrating modeling, data collection, and design activities in a manner that identifies those aspects of

2.3 GENERAL PRINCIPLES OF PERFORMANCE ASSESSMENT

/ 21

engineered and natural barriers in a disposal system of importance to a decision about compliance with performance objectives. • Performance assessment is a process intended to provide reasonable assurance of compliance with performance objectives. • Since models and parameters used in performance assessment are uncertain and some processes that affect system performance may not be well understood, a variety of results should be presented rather than a single projected outcome. • An integration and interpretation of assumptions and results, in which bases for conceptual models and results of calculations are presented in a manner that reflects the judgments involved and the importance of different aspects of an assessment to a licensing decision, is a critical aspect of performance assessment. In general, the performance assessment process provides a means of building confidence in judgments and models used to determine whether reasonable assurance of compliance with performance objectives can be obtained. 2.3.1

Performance Assessment as an Iterative Process

Characterization of performance assessment as a process, rather than a set of calculations, is important. An early view was that site characterization should be completed and a conceptual facility design developed, after which a performance assessment would be conducted to support a license application (Starmer et al., 1988). However, experience has shown that efficient application of the performance assessment process involves substantial interaction between modeling and data collection and facility design activities. These interactions occur during the period prior to licensing and during facility operations and are dynamically linked. That is, site data and information on facility design are needed to make modeling assumptions and to assign parameter values, evaluation of modeling results can lead to an identification of additional data needs or design changes, and these in turn can lead to use of altered assumptions and parameter values in modeling. Introduction of unanticipated wastes can lead to changes in modeling assumptions and facility design. In general, feedback between different activities identifies aspects of the site, design or models for which altered

22 / 2. DEFINITION AND PRINCIPLES OF PERFORMANCE ASSESSMENT assumptions are needed to improve projected performance in a defensible manner. Thus, performance assessment is a management tool to be used throughout the preclosure phase of facility operations. Performance assessment also is a useful tool for risk management during the postclosure time phase. It can be used to determine needs for institutional control at a site, and it can be used to guide monitoring activities that normally are undertaken after facility closure. Such monitoring activities can provide additional information on the validity of a performance assessment in supporting a licensing decision. Important aspects of data collection, facility design, and modeling are best identified using a flexible, iterative process that allows feedback from each activity to be incorporated in subsequent iterations of other activities (Case and Otis, 1988; DOE, 1985; IAEA, 1997a; 1999; Kozak, 1994a; Kozak et al., 1993; NRC, 2000; Seitz et al., 1992a). Efficiency is improved by starting with simple analyses and using findings of those analyses to identify areas that require more detailed consideration in subsequent analyses. Sensitivity analysis is an important contributor to this feedback. Sensitivity analysis provides an increased understanding of particular aspects of a disposal system that influence overall performance, and it helps to provide justification for those aspects for which additional effort would provide the most benefit to a performance assessment and an appropriate basis for a licensing decision. 2.3.2

Performance Assessment as a Decision Tool

Performance assessment is an essential tool for regulatory decision making. Performance assessment is used to identify conditions at a disposal facility that support a finding of reasonable assurance of compliance with performance objectives. An understanding of how this use of performance assessment differs from its application to the more difficult problem of predicting the actual long-term performance of disposal systems is critical to the conduct of performance assessment and associated data-collection activities and to proper interpretation of results. In the context of performance assessment, it is desirable to develop conceptual models of the long-term performance of disposal systems that bound the range of reasonably foreseeable conditions. Thus, performance assessment can be used, in an inverse sense, to identify conditions that may cause performance objectives to be exceeded. This places the emphasis on defending why such conditions are not likely to occur or on needed changes in facility design

2.3 GENERAL PRINCIPLES OF PERFORMANCE ASSESSMENT

/ 23

to minimize the potential for such conditions to occur. Because of inherent uncertainties in any performance assessment, judgment will be necessary in assessing the defensibility of conceptual models. 2.3.3

Uncertainty in Results of Performance Assessment

Models and parameters used in performance assessment are characterized by varying degrees of uncertainty. Thus, performance assessment generally should provide more than a single result in demonstrating compliance with performance objectives, with different results developed on the basis of a variety of plausible assumptions. For example, although a single set of results may provide the primary basis for a comparison with performance objectives, additional results should be provided in the context of a sensitivity and uncertainty analysis to lend credence to the demonstration of compliance. The type of uncertainty of primary concern to this Report is uncertainty of importance to a licensing decision, rather than uncertainty in the actual outcome (projected dose). To emphasize this distinction, this Report uses the term “importance analysis” (Section 7). Importance analysis is used to identify assumptions and parameter values in a performance assessment that have an impact on a decision regarding compliance with performance objectives. This information can then be used to identify areas where further data, design enhancements, or modeling are needed to reduce uncertainty (increase confidence) in a decision. Since results of performance assessment depend on assumptions, data and design of a facility, changes in any of these can result in changes in conclusions resulting from an analysis. 2.3.4

Integration and Interpretation of Results

Given the inexact nature of performance assessment and the need to consider a variety of conditions in an assessment and, thus, to produce a range of projected outcomes, it is logical to ask how those outcomes can be interpreted for comparison with fixed performance objectives. An essential but highly challenging aspect of performance assessment is an integration and presentation of results in a manner that facilitates this interpretation. The term “integration” refers to a need to summarize a variety of results and to identify assumptions and data that are important to a licensing decision, as determined from the analyses performed. The term “interpretation” refers to a need to present results and critical assumptions in a manner that provides the basis for

24 / 2. DEFINITION AND PRINCIPLES OF PERFORMANCE ASSESSMENT identifying conditions under which a disposal facility can be expected to comply with performance objectives. This part of a performance assessment will require substantial judgment, as do all other parts. Peer review can be a useful means of building confidence in assumptions and the defensibility of an analysis. 2.3.5

Summary

Performance assessment is most efficient when conducted in an iterative manner, as shown conceptually in Figure 2.1. These steps in an iterative approach are developed more fully into procedural steps in Section 4. At this stage of the discussion, the important point is that performance assessment produces an evolution of ideas about the behavior of a disposal system by an integration of observations, conceptualizations and interpretations of technical analyses. Analyses become more refined as additional data, information on facility design, detailed models, or better supported assumptions are used. Data collection, design, and modeling activities are interdependent parts of a process that require regular feedback among the multiple disciplines involved. Results of each iteration of a performance assessment should provide feedback to help identify critical data and design needs. Performance assessment should be viewed as a process, the primary purpose of which is to determine whether there is reasonable

Fig 2.1. Conceptual representation of iterative approach to conduct of performance assessment; process normally begins with observation and development of conceptual models.

2.4 BALANCE BETWEEN CONSERVATISM AND REALISM

/ 25

assurance that doses to members of the public and impacts on the environment that would result from disposal of waste in a particular facility will be below applicable regulatory requirements. A defensible performance assessment includes the use of results obtained throughout different phases of the process to demonstrate an improved understanding of the importance of assumptions regarding a site and engineered features with respect to conclusions obtained from an analysis of long-term performance. Finally, the most important task in performance assessment is to integrate and interpret results in a manner that facilitates their use in providing justification for a decision about compliance of a disposal facility with applicable regulatory requirements. Acceptance of the need for judgment in conducting performance assessment and in interpreting results is an essential aspect of the licensing process. 2.4 Balance Between Conservatism and Realism in Performance Assessment A central issue that must be confronted in all performance assessments is whether simple and clearly conservative models should be used to describe various aspects of the long-term performance of a disposal system or whether more complex and rigorous modeling should be undertaken in an effort to provide more realistic projections of outcomes at times far into the future. The following discussion describes the point of view on this issue taken in this Report. Incorporation of realism in performance assessment is desirable for many reasons. For example: more realistic assessments could be used to justify that disposal of greater amounts of waste in licensed near-surface facilities would be safe, which could alleviate needs for more costly disposal options; more realistic assessments could more accurately convey to the public and other stakeholders the expected magnitude of doses or risks at far future times, which could alleviate unwarranted concerns about radiological impacts of waste disposal; more realistic assessments are potentially important in comparing alternative disposal systems or facility designs; efforts at realistic assessments may lead to significant advances in modeling and development of improved databases to support modeling activities; and too great a reliance on conservative models in evaluating the performance of operating or planned near-surface disposal facilities may provide precedents that, when applied in evaluating needs for cleanup of radioactively contaminated sites, could lead to costly but unwarranted remediation.

26 / 2. DEFINITION AND PRINCIPLES OF PERFORMANCE ASSESSMENT However, a desire for realism in performance assessments must be tempered by a realization of difficulties in taking this approach. A fundamental difficulty is that results of performance assessments cannot be tested by comparison with observed outcomes when projected impacts of waste disposal often occur many generations into the future.6 In addition: realistic assessments of all aspects of the long-term performance of a disposal facility would require extensive amounts of site-specific data that generally are not available when planning for a new facility begins; collection of required site-specific data would likely be time consuming and unreasonably expensive,7 or it may not be possible to obtain some data that would be required for rigorous modeling; and a demonstrated ability to realistically model some phenomena, such as flow and transport in unsaturated soil, under any conditions that might be encountered in the environment is not yet in hand. If attempts at realistic assessments are to be viewed as credible, they must be accompanied by full disclosure of uncertainties in all aspects of an analysis, including uncertainties in the present state of knowledge of chemical and physical processes that occur in complex natural and engineered systems and uncertainties in model structures used to represent those processes, as well as uncertainties in model parameters. This is a formidable challenge for many aspects of the performance of waste disposal systems. These difficulties illustrate that determination of “realistic” outcomes of waste disposal is not solely an objective exercise. Rather, any approach to modeling requires a significant amount of subjective scientific judgment, and it is practically impossible on the basis of current knowledge to evaluate how those judgments would affect differences between projected and actual outcomes. Although use of clearly conservative models for at least some aspects of the performance of waste disposal systems can circumvent many of the difficulties with attempts at more realistic 6Observations of outcomes of past disposal practices are of limited use when older facilities were not located or designed in accordance with current practices, when observations have been carried out for no more than a few decades, or when observed outcomes resulted from unsatisfactory designs and environmental conditions (see Footnote 3 on page 17). 7The time and costs required to obtain extensive amounts of sitespecific data are more difficult to justify at near-surface disposal facilities for low-level waste than at geologic repositories for spent nuclear fuel and high-level waste or transuranic waste, given that there will be very few geologic repositories and that the latter types of waste usually pose much higher intrinsic hazards.

2.4 BALANCE BETWEEN CONSERVATISM AND REALISM

/ 27

assessments described above, this approach also is not without its challenges. For example: conservative models often are implemented using generic data, and there is a need to demonstrate that assumed models and data are conservative at specific sites; assumptions that would lead to conservative projections of outcomes may not be intuitively obvious for some aspects of system performance; and it may be difficult to explain to the public and other stakeholders that an expected outcome is substantially less than a projected outcome obtained using conservative models. In addition, use of conservative models in inappropriate ways can result in misleading comparisons of alternative disposal systems or facility designs. Perhaps the most important concern with overzealous use of conservative models is that near-surface disposal of some wastes would be foreclosed when projected outcomes are compared with performance objectives even though, in reality, disposal of those wastes would not compromise compliance with performance objectives or protection of the public. The point of view taken in this Report is that an appropriate balance between conservatism and more realistic approaches to performance assessment is largely a matter of judgment that should be applied on a site-specific basis, and that there is no single prescription that would be appropriate at all sites. Performance assessments generally should attempt to incorporate some degree of realism to demonstrate an appropriate level of understanding of the long-term performance of disposal systems. The goal at any site should be to provide a cost-effective and defensible assessment that would be commensurate with hazards posed by low-level wastes that are intended for disposal at that site. At well chosen sites with highly desirable characteristics, use of simple and conservative models for some aspects of system performance may not affect a decision on the acceptability of wastes that are intended for disposal. At other sites with less favorable characteristics, efforts at more realistic modeling may be required to demonstrate that wastes that are intended for disposal would be acceptable. At any site, it is important to keep in mind that performance assessment is conducted to inform a decision about the acceptability of waste disposals, and that it is not necessary to obtain realistic projections of outcomes to render such decisions in a defensible manner.

3. Context for Performance Assessment This Section discusses several topics that provide important context for the conduct of performance assessments of low-level waste disposal facilities. These topics include: (1) the definition of low-level waste in the United States; (2) sources and properties of low-level waste; (3) recommendations of the International Commission on Radiological Protection (ICRP) on application of principles of radiation protection to disposal of solid radioactive waste; (4) legal and regulatory requirements for disposal of low-level waste in the United States, with particular emphasis on performance objectives for near-surface disposal systems that address long-term protection of the public and the environment; and (5) other concepts that are important in demonstrating compliance with applicable performance objectives, including the role of active and passive institutional controls, model validation and confidence in model outcomes, and “reasonable assurance” of compliance. 3.1 Definition of Low-Level Radioactive Waste The historical development of the definition of low-level radioactive waste in the United States and the current definition and its implications are discussed in detail elsewhere (Kocher, 1990; NCRP, 2002). Only a summary discussion is given here. 3.1.1

Earliest Descriptions of Low-Level Waste

The term “low-level waste” was first used formally in the radioactive waste management program of the U.S. Atomic Energy Commission (AEC) in the late 1950s (Lennemann, 1972). This term referred originally to liquid wastes produced in chemical reprocessing of spent nuclear fuel that generally contained the lowest concentrations of radionuclides of any liquid wastes produced in fuel reprocessing. Liquid low-level wastes usually were released to holding ponds or lagoons or directly to surface water. Beginning about 1960, the concept that low-level waste generally contained 28

3.1 DEFINITION OF LOW-LEVEL RADIOACTIVE WASTE

/ 29

the lowest concentrations of radionuclides also was applied to solid wastes when AEC initiated interim shallow-land burial services for solid radioactive wastes generated in the commercial sector (Lennemann, 1967). Although low-level radioactive wastes, either liquid or solid, usually were released to or disposed of in the near-surface environment, the earliest descriptions of these wastes were not based on considerations of protection of the public. Rather, they were based primarily on operational requirements for safe handling and storage of wastes at generating sites (Lennemann, 1967; 1972). Liquid and solid low-level wastes were wastes that contained concentrations of radionuclides sufficiently low that only minimal requirements on shielding and other protection systems for workers were required during waste operations. Numerical limits on concentrations of radionuclides in liquid or solid low-level wastes were developed independently at each AEC site that generated waste (Beard and Godfrey, 1967; Marter, 1967), but there was no attempt to develop limits that would be generally applicable to all sites or limits that would apply to both liquid and solid low-level wastes. Early limits were expressed in terms of total activity concentrations of all radionuclides combined. Thus, classification of radioactive wastes as “low level” was based essentially on concentrations of shorterlived radionuclides that did not pose a long-term hazard to the public following disposal or release to the environment. The earliest descriptions of low-level radioactive waste used at AEC sites were not retained when this term was defined in law. In particular: liquid and solid low-level wastes no longer were defined separately; low-level waste no longer was defined on the basis of requirements for protection of workers at waste generating sites, nor was the definition based on considerations of protection of the public at waste disposal sites; and, low-level waste no longer was defined as waste that contains relatively low concentrations of radionuclides only. 3.1.2

Current Definition of Low-Level Waste

The current legal definitions of low-level radioactive waste in the United States are contained in the Nuclear Waste Policy Act (NWPA, 1982), as amended, and the Low-Level Radioactive Waste Policy Amendments Act (LLRWPAA, 1985). In the Nuclear Waste Policy Act, low-level waste is defined as: • radioactive waste that is not high-level waste, spent nuclear fuel, transuranic waste, or uranium or thorium mill tailings; or

30 / 3. CONTEXT FOR PERFORMANCE ASSESSMENT • radioactive waste that NRC, consistent with existing law, classifies as low-level waste. LLRWPAA contains a similar definition, except transuranic waste is not excluded from low-level waste. Thus, low-level waste is defined only by exclusion. The two definitions in law exclude high-level waste, spent nuclear fuel, and uranium or thorium mill tailings, but they differ in regard to whether low-level waste excludes transuranic waste. This inconsistency has little practical significance, because LLRWPAA is concerned only with disposal of low-level waste at sites licensed by NRC or an Agreement State, and there currently is very little transuranic waste that could be sent to such sites (DOE, 1997a). Most transuranic waste has been generated in defense-related activities at AEC and U.S. Department of Energy (DOE) sites, and this waste is excluded from low-level waste on the basis of the definition in the Nuclear Waste Policy Act. The definition of low-level waste depends on the definitions of high-level waste, spent nuclear fuel, transuranic waste, and uranium or thorium mill tailings. Definitions of the different waste classes that arise from operations of the nuclear fuel cycle are summarized in Table 3.1. The definition of high-level waste is the key to the classification system for wastes that arise from operations of the nuclear fuel cycle, because the definitions of low-level waste and transuranic waste exclude high-level waste. The definition of high-level waste is based on its source, not its radiological properties. High-level waste contains high concentrations of fission products, resulting in high levels of decay heat and external radiation, and high concentrations of long-lived, alpha-emitting transuranium radionuclides. High-level waste thus requires extensive safety systems to protect workers during waste handling and storage and highly confining and isolating disposal systems to protect the public (e.g., a geologic repository). Since high-level waste is defined on the basis of its source, other wastes with similar radiological properties that are not produced directly in fuel reprocessing are not classified as high-level waste. Such wastes, as well as incidental wastes that arise in fuel reprocessing, operations at reprocessing plants, or further processing of reprocessing wastes that have been excluded from high-level waste on a case-by-case basis, are classified as low-level waste or transuranic waste, depending on the concentrations of long-lived, alphaemitting radionuclides.

3.1 DEFINITION OF LOW-LEVEL RADIOACTIVE WASTE

/ 31

TABLE 3.1—Summary of current definitions of different classes of radioactive wastes that arise from operations of nuclear fuel cycle in the United States.a Waste Class

Definition

High-level waste

Primary wastes, either liquid or solid, that arise from chemical reprocessing of spent nuclear fuelb

Spent nuclear fuel

Irradiated nuclear fuel that has not been chemically reprocessedc

Transuranic waste

Waste that contains more than 4 kBq g–1 of alpha-emitting transuranium radionuclides with half-lives >20 y, excluding high-level waste

Uranium or thorium mill tailings

Residues from chemical processing of ores for their source material (i.e., uranium or thorium) content

Low-level waste

Waste that is not high-level waste, spent nuclear fuel, transuranic waste, or uranium or thorium mill tailingsd

a Adapted from Kocher (1990) and NCRP (2002); definitions are simplified representations of current definitions in law. bCertain incidental wastes that arise in fuel reprocessing, operations at fuel reprocessing plants, or further processing of reprocessing wastes have been excluded from high-level waste on a case-by-case basis. Excluded incidental wastes generally have lower concentrations of fission products and long-lived, alpha-emitting radionuclides than primary reprocessing wastes that are classified as high-level waste. cSpent nuclear fuel is not waste until so declared. In some laws and regulations, spent nuclear fuel is not distinguished from high-level waste. d Low-level waste also does not include NARM not associated with operations of the nuclear fuel cycle.

The definition of transuranic waste is unique in two respects. First, the definition is quantitative, in that minimum concentrations of particular radionuclides are specified. Second, the definition is based on considerations of protection of the public following waste disposal, because waste with concentrations of long-lived, alpha-emitting transuranium radionuclides