United States
Environmental Protection
Agency
Air And Radiation
(6602J)
EPA 402-R-99-008
August 1999
£EPA
Environmental Radiation
Protection Standards For
Yucca Mountain, Nevada
Draft Background Information
Document For Proposed
40CRF 197
-------
40 CFR Part 197
BACKGROUND INFORMATION DOCUMENT
FOR 40 CFR 197
ENVIRONMENTAL
RADIATION PROTECTION STANDARDS
FOR YUCCA MOUNTAIN, NEVADA
August 1999
U.S. Environmental Protection Agency
Office of Radiation and Indoor Air
Washington, DC 20460
-------
TABLE OF CONTENTS
Section Page
List of Acronyms xiii
Executive Summary ES -1
CHAPTER 1 INTRODUCTION 1-1
1.1 Purpose and Scope of the Background Information Document 1-1
1.2 EPA's Regulatory Authority for the Rulemaking 1-3
1.3 The National Academy of Sciences Recommendations 1-4
1.4 History of EPA's Rulemaking 1-7
1.4.1 Legislative History 1-8
1.4.2 The Development of EPA's Role in the Federal Program .... 1-11
1.4.3 Early Federal Action 1-12
1.4.4 40 CFRPart 191 '. 1-14
References 1-18
CHAPTER 2 HISTORY OF RADIATION PROTECTION IN THE UNITED STATES
AND CURRENT REGULATIONS 2-1
2.1 Introduction 2-1
2.2 The International Commission on Radiological Protection, the
National Council on Radiation Protection and Measurements, and
the International Atomic Energy Agency 2-2
2.3 Federal Radiation Council Guidance 2-7
2.4 Environmental Protection Agency 2-9
2.4.1 Environmental Radiation Exposure 2-10
2.4.2 Environmental Impact Assessments 2-10
2.4.3 Ground-Water Protection 2-11
2.4.4 Radionuclide Air Emissions 2-13
2.4.5 Disposal of High-Level Radioactive Waste and Spent
Nuclear Fuel 2-15
2.4.6 Evaluation of Radiation Dose 2-17
2.5 Nuclear Regulatory Commission 2-18
2.5.1 Fuel Cycle Licensees 2-19
2.5.2 Radioactive Waste Disposal Licenses 2-19
2.5.3 Repository Licensing Support Activities 2-21
2.6 Department of Energy 2-23
References 2-25
-------
TABLE OF CONTENTS (Continued)
Sectiqn page
CHAPTER 3 SPENT NUCLEAR FUEL AND HIGH-LEVEL WASTE DISPOSAL
PROGRAMS IN OTHER COUNTRIES 3-1
3.1 Belgium 3-3
3.1.1 Nuclear Power Utilization 3-3
3.1.2 Disposal Programs and Management Organizations 3-5
3.1.3 Regulatory Organizations and Their Regulations 3-6
3.2 Canada 3.7
3.2.1 Nuclear Power Utilization 3-7
3.2.2 Disposal Programs and Management Organizations 3-8
3.2.3 Regulatory Organizations and Their Regulations 3-10
3.3 Finland 3-11
3.3.1 Nuclear Power Utilization 3-11
3.3.2 Disposal Programs and Management Organizations 3-11
3.3.3 Regulatory Organizations and Their Regulations 3-12
3.4 France 3-13
3.4.1 Nuclear Power Utilization 3-13
3.4.2 Disposal Programs and Management Organizations 3-14
3.4.3 Regulatory Organizations and Then-Regulations 3-16
3.5 Germany 3-16
3.5.1 Nuclear Power Utilization 3-16
3.5.2 Disposal Programs and Management Organizations 3-17
3.5.3 Regulatory Organizations and Their Regulations 3-18
3.6 Japan 3-19
3.6.1 Nuclear Power Utilization 3-19
3.6.2 Disposal Programs and Management Organizations 3-20
3.6.3 Regulatory Organizations and Their Regulations 3-22
3.7 Spain 3_22
3.7.1 Nuclear Power Utilization 3-22
3.7.2 Disposal Programs and Management Organizations 3-23
3.7.3 Regulatory Organizations and Their Regulations 3-25
3.8 Sweden 3-25
3.8.1 Nuclear Power Utilization : 3-25
3.8.2 Disposal Programs and Management Organizations 3-25
3.8.3 Regulatory Organizations and Their Regulations 3-27
3.9 Switzerland 3_28
3.9.1 Nuclear Power Utilization 3-28
3.9.2 Disposal Programs and Management Organizations 3-28
3.9.3 Regulatory Organizations and Their Regulations 3-30
u
-------
TABLE OF CONTENTS (Continued)
Section
Page
3.10 United Kingdom 3-30
3.10.1 Nuclear Power Utilization 3-30
3.10.2 Disposal Programs and Management Organizations 3-31
3.10.3 Regulatory Organizations and Their Regulations 3-32
References 3-34
CHAPTER 4 U.S. PROGRAMS FOR THE MANAGEMENT -AND DISPOSAL OF
SPENT NUCLEAR FUEL AND HIGH-LEVEL RADIOACTIVE
WASTE AND THE EVALUATION OF YUCCA MOUNTAIN
4.1 Introduction 4-1
4.2 The Department of Energy 4-1
4.2.1 DOE's Office of Civilian Radioactive Waste Management
(OCRWM) 4-2
4.2.2 DOE Management and Disposal of Defense Wastes 4-4
4.3 The Nuclear Regulatory Commission 4-5
4.3.1 Legislative Requirements and Regulatory Framework 4-6
4.3.2 Status of NRC's Program 4-7
4.4 Nuclear Waste Technical Review Board 4-8
4.5 State and Local Agencies , 4-8
4.6 Native American Tribes 4-9
References 4-12
CHAPTER 5 QUANTITIES, SOURCES, AND CHARACTERISTICS OF SPENT
NUCLEAR FUEL AND HIGH-LEVEL WASTE IN THE UNITED
STATES ; 5-1
5.1 Introduction 5-1
5.2 Spent Nuclear Fuel 5-1
5.2.1 Commercial Spent Nuclear Fuel Inventory and Projection .... 5-2
5.2.2 DOE Spent Nuclear Fuel 5-4
5.3 Defense High-Level Radioactive Waste 5-7
5.3.1 High-Level Inventories at the Hanford Site 5-11
5.3.2 High-Level Waste Inventories at INEEL 5-12
5.3.3 High-Level Inventories at the Savannah River Site 5-13
5.3.4 High-Level Inventories at the West Valley Demonstration
Project 5-13
5.4 Significant Radionuclides Contained in Spent Nuclear Fuel and
High-Level Waste 5-14
References 5-17
m
-------
TABLE OF CONTENTS (Continued)
Section page
CHAPTER6 DOSE AND RISK ESTIMATION 6-1
6.1 Introduction 6-1
6.2 Dose Estimation 6-1
6.3 Cancer Risk Estimation 6-3
6.4 Genetic Effects 6-4
6.5 Developmental Effects 6-6
6.5.1 In-Utero Carcinogenesis , 6-6
6.5.2 Brain Teratology 6-7
6.5.3 Other Effects of Prenatal Irradiation 6-8
6.5.4 Summary of Developmental Effects 6-9
References 6-10
CHAPTER 7 CURRENT INFORMATION CONCERNING A POTENTIAL WASTE
REPOSITORY AT YUCCA MOUNTAIN 7-1
7.1 Principal Features of the Natural Environment 7-1
7.1.1. Geologic Features 7-1
7.1.2 Hydrologic Features 7-57
7.1.3 Climate Considerations 7-117
7.2 Repository Concepts under Consideration for Yucca Mountain 7-125
7.2.1 Conceptual Repository Systems 7-125
7.2.2 Design Concepts for Engineered Features 7-127
7.3 Repository System Performance Assessments 7-143
7.3.1 DOE Historic Performance Assessments 7-144
7.3.2 DOE's TSPA for the Viability Assessment 7-147
7.3.3 TSPA-VA Results 7-161
7.3.4 Reviews of the TSPA-VA 7-177
7.3.5 NRC Total System Performance Assessments 7-188
7.3.6 EPRI Total System Performance Assessments 7-197
7.3.7 Comparison of DOE, NRC, and EPRI TSPA Results for VA
Repository 7-205
References .- ' 7-207
CHAPTER 8 RADIOLOGICAL PATHWAYS THROUGH THE BIOSPHERE 8-1
8.1 Introduction 8-1
IV
-------
TABLE OF CONTENTS (Continued)
Section
Page
8.2 Past, Current, and Potential Use of the Yucca Mountain
Region 8-3
8.2.1 Past Use of the Yucca Mountain Region 8-3
8.2.2 Current Demographics and Land Use 8-13
8.2.3 Factors Affecting Future Use of the Region 8-14
8.3 Radiation Protection of Individuals 8-45
8.3.1 The Critical Group Concept 8-46
8.3.2 Probabilistic Scenario Modeling 8-47
8.3.3 Exposed Individuals and Exposure Scenarios for Yucca
Mountain 8-50
8.3.4 Details and Analyses for the Subsistence Farmer Scenario ... 8-52
8.3.5 Alternative Exposure Scenarios for Consideration at Yucca
Mountain 8-68
8.4 The Repository Intrusion Scenario: A Special Case 8-71
8.4.1 Site Resources as Potential Cause for Intrusion 8-73
8.4.2 Types of Human Intrusion 8-79
8.4.3 Parameters and Assumptions Associated With Ground
Water Withdrawal 8-85
8.4.4 Parameters and Assumptions Associated With Human
Intrusion 8-86
References 8-96
CHAPTER 9 YUCCA MOUNTAIN EXPOSURE SCENARIOS AND COMPLIANCE
ASSESSMENT ISSUES 9-1
9.1 Introduction 9-1
9.2 Gaseous Releases: A Secondary Pathway for Human
Exposure 9-3
9.2.1 Production and Early Containment of Carbon-14 9-4
9.2.2 Impacts of Thermal Loading on Gaseous Releases and
Transport : 9-4
9.2.3 Estimates of Travel Time 9-5
9.2.4 Distribution of C-14 in the Biosphere 9-7
9.2.5 Dose Modeling and Exposure Estimates 9-8
9.2.6 Potential Non-Radiological Impacts of C-14 9-10
9.3 Development Performance of Summary Scenarios and Compliance
Issues 9-11
9.3.1 Identification of Improbable Phenomena 9-11
9.3.2 Screening of Events and Processes 9-12
-------
TABLE OF CONTENTS (Continued)
Section
Page
9.3.3 Compliance With A Standard 9-14
9.3.4 Development of Site Performance Issues 9-17
References 9-26
CHAPTER 10 RADIOLOGICAL RISKS FOR DEEP GEOLOGICAL DISPOSAL AND
SURFACE STORAGE OF SPENT NUCLEAR FUEL 10-1
10,1 Background Information 10-1
10.2 Regulatory Limits 10-2
10.2.1 Power Reactors 10-3
10.2.2 Research Reactors 10-4
10.2.3 Independent Spent Fuel Storage Installations 10-4
10.2.4 DOE Facilities 10-5
10.2.5 Summary of Regulatory Limits 10-5
10.3 Report by the Monitored Retrievable Storage Review Commission . 10-6
10.3.1 At-Reactor Storage Options 10-6
10.3.2 Radiation Exposure Modeling Assumptions for At-
Reactor Storage of SNF 10-8
10.3.3 Model Assumptions for MRS Storage of SNF 10-10
10.3.4 Transportation Models for SNF With and Without MRS .. 10-10
10.3.5 Public Exposure from SNF Storage 10-11
10.4 Other Information Sources 10-14
10.4.1 An Assessment of LWRS Spent Fuel Disposal Options .. 10-15
10.4.2 Generic Environmental Impact Statement, Management
of Commercially Generated Radioactive Waste 10-16
10.4.3 Review of Dry Storage Concepts Using Probabilistic
Risk Assessment 10-17
10.4.4 Requirement for the Independent Storage of Spent Fuel
and High-Level Radioactive Waste 10-17
10.4.5 Environmental Assessment Related to the Construction
and Operation of the Surry Dry Cask Independent Spent
Fuel Storage Installation 10-19
10.4.6 Environmental Assessment Deaf Smith County Site,
Texas 10-19
10.4.7 Preliminary Assessment of Radiological Doses hi
Alternative Waste Management Systems Without an
MRS Facility 10-20
10.4.8 Monitored Retrievable Storage Submission to Congress .. 10-21
VI
-------
TABLE OF CONTENTS (Continued)
Section Page
10.4.9 The Safety Evaluation of Tunnel Rack and Dry Well
Monitored Retrievable Storage Concepts 10-22
10.4.10 Summary Assessment of Available Data 10-22
References 10-26
GLOSSARY G-l
APPENDICES
I. Demography and Ecosystems 1-1
II. Radionuclide Exposures to Persons in the Vicinity of the Nevada Test Site/Yucca
Mountain Site II-l
III. Soil Types Found in the Yucca Mountain Area III-l
IV. Well Drilling and Pumping Costs IV-1
V. New and Unusual Farming Practices V-l
VI. Current Information Regarding Ground-Water Flow and RadionuclideTransport
hi the Unsaturated and Saturated Zones '. VI-1
vn
-------
TABLES
No. Page
1-1 Significant Events in the History of High-Level Radioactive Waste and Spent
Nuclear Fuel Disposal 1-7
3-1 National and International Criteria and Objectives for the Disposal of Long-Lived
Radioactive Wastes 3-4
5-1 Historical and Projected Mass and Radioactivity of Commercial Spent Nuclear
Fuel 5-3
5-2 Historical and Projected Installed Nuclear Electric Power Capacity 5-4
5-3 DOE Spent Nuclear Fuel Inventory 5-8
5-4 Historical and Projected Cumulative Volume and Radioactivity of High-Level
Waste Stored in Tanks, Bins, and Capsules by Site 5-10
5-5 Radionuclide Inventories in Spent Nuclear Fuel and High-Level Wastes Expected
to be Disposed in a Yucca Mountain Repository 5-15
6-1 Estimated Frequency of Genetic Disorders in a Birth Cohort Due to Exposure of
Each of the Parents to 0.01 Gy (1 Rad) per Reproductive Generation (30 yr) 6-6
6-2 Possible Effects of Inutero Radiation exposure 6-9
7-1 Stratigraphy of the Southern Great Basin 7-11
7-2 Principal Stratigraphic Units 7-15
7-3 Known or Suspected Quaternary Faults within 20 Km of the Proposed Repository
Site 7-29
7-4 Significant Earthquakes within 320 Km of Yucca Mountain Site Since 1850 7-35
7-5 Hydraulic Conductivities Calculated from Pumping Test Data 7-86
7-6 Borehole Location and Depth Data for Wells Drilled to the Lower Carbonate
Aquifer in the Vicinity of and Downgradient of the Yucca Mountain Area 7-92
7-7 Design Parameters for Enhanced Design Alternatives (DOE99) 7-140
7-8 Principal Results of Enhanced Design Alternative Analysis 7-141
7-9 Comparison of EDA II and Viability Assessment Design Features (DOE99) 7-142
7-10 Principal Performance Factors for TSPA-VA Modeling (DOE98) 7-154
8-1 Range in Concentration of Dissolved Constituents hi Ground Water in the
Amargosa Desert 8-16
8-2 Hydrographic Basins in the Vicinity of Yucca Mountain 8-19
8-3 Water Appropriations by Hydrographic Basin in the Study Area 8-21
8-4 1993 Ground Water Pumpage Inventory for Basin No. 230 8-22
8-5 Wells and Boreholes hi the Amargosa Valley 8-26
8-6 Ground Water Budget for Hydrographic Basins in Study Area 8-35
8-7 Estimates of Acreage Under Cultivation for Feedstock 8-40
8-8 Ground Water Storage Values for Relevant Hydrographic Basins 8-41
viii
-------
IAELES (Continued)
Page
8-9 Concentration Ratios and Transfer Coefficients By Element 8-59
8-10 Dose Conversion Factors for a Resident Farmer in Current Biosphere By
Exposure Pathway and Radionuclide for Ground Water Source 8-64
8-11 Summary of Mean TEDE Results From CNWRA Unit Concentration Evaluations
for Water 8-65
8-12 Comparison of Inhalation, Drinking Water and Food Consumption Rate
Parameter Values From Various Sources 8-66
8-13 Likely Human Intrusion Scenarios for Different Types of Resources 8-80
8-14 Typical Borehole Characteristics 8-88
9-1 Annual Average Doses Resulting from the Release of lOOCi MCO2 for Distances
Outto50Miles 9-10
9-2 Potentially Disruptive Events and Processes 9-13
9-3 Techniques for Quantifying or Reducing Uncertainty in the Performance
Assessment 9-16
10-1 Spent Fuel Accumulation at Shutdown Commercial Light Water Power Reactors .. 10-9
10-2 Reduction in Dry Storage Needs At Reactor Facilities with Linked MRS 10-10
10-3 Life-Cycle Transportation Risk Measures 10-11
10-4 Total Life-Cycle Doses in Person-Rem from Spent Nuclear Fuel Management
With and Without MRS 10-12
10-5 Location of Spent Fuel With MRS in 2010 and Repository in 2013 10-13
10-6 Comparison of Public Exposures Resulting from Three SNF Storage Alternatives . 10-14
10-7 Public Doses for Normal Repository Operation and From Shaft-Drop Accident ... 10-16
10-8 At-Reactor Storage Accidents: Summary of Results 10-18
10-9 Preclosure Exposure Associated with a Reference Salt Repository 10-19
10-10 Public Dose Estimates for Reference Reactor and Repository Surface Facility ... 10-20
10-1 la Public Doses from Routine Operations at MRS and Repository 10-21
10-1 Ib Public Doses from Accident Releases at MRS and Repository 10-22
10-12 Normalized Population Doses 10-23
10-13 Summary Data of Public Doses Associated with SNF Storage At-Reactor,
MRS, and Repository 10-25
IX
-------
FIGURES
Page
7-1 Location of Yucca Mountain (DOE94a) 7-2
7-2 Boundaries and Larger Subdivisions of the Basin and Range Physiographic
Province (HUN74) 7-3
7-3 Physiographic Features in the Yucca Mountain Site Area (DOES 8) 7-5
7-4 Generalized Regional Stratigraphic Column Showing Geologic Formations
and Hydrological Units in the Nevada Test Site Area 7-7
7-5 Latest Precambrian through Mid-Paleozoic Paleography of the Great Basin
(Modified from DOE95a) 7-8
7-6 Late Devonian and Mississippian Paleogeography of the Great Basin
(Modified from DOE95a) 7-9
7-7 Simplified Geologic Map Showing the Distribution of Major
Lithostratigraphic Units hi the Yucca Mountain Area (Modified from
DOE95a) 7-12
7-8 East-West Geologic Cross Section for the Yucca Mountain Site 7-14
7-9 The Walker Lane Belt and Major Associated Faults (DOE88) 7-23
7-10 Major North-Trending Faults hi the Vicinity of Yucca Mountain (DOE95k) 7-25
7-11 Index Map of Faults at and near Yucca Mountain 7-26
7-12 Index Map of Known or Suspected Quaternary Faults in the Yucca Mountain
Region (Modified from DOE 95a) 7-28
7-13 Sketch map of the Western United States Showing Some Major Structural
Features. 7-32
7-14 Magnitude 3 or Greater Earthquakes within 320 Km (200 Miles) of Yucca
Mountain from 1850 to 1992 (Modified from DOE 95a) 7-38
7-15 Index Map Showing Outlines of Calderas in the Southwestern Nevada
Volcanic Field, and Extent of the Tiva Canyon and Topopah Spring Tuffs of
the Paintbrush Group (Modified from DOE95a) 7-43
7-16 Distribution of Basalts in the Yucca Mountain Region with Ages of Less Than
12MA(NRC96a) 7-46
7-17 Unsaturated Zone Hydrogeologic Units (USG84a) 7-60
7-18 Locations of Deep Boreholes hi the Vicinity of Yucca Mountain (USG96a) 7-63
7-19 Early Conceptual Model of Ground-water Flow in the Unsaturated Zone at
Yucca Mountain (USG84a) 7-71
7-20 Current Conceptual Model of Ground-water Flow in the Unsaturated Zone at
Yucca Mountain (LBL96) 7-72
7-21 Saturated Zone Hydrostratigraphy of Volcanic Rocks (USG96a) 7-81
7-22 Schematic North/south Cross-sectional Illustration of Thinnhig of Volcanic
Units Beneath the Amargosa Desert (USG85a) 7-84
7-23 Schematic Illustration of Ground-water Flow System in the Great Basin
(USG76a) 7-101
7-24 Death Valley Ground-Water Flow System (USG96a) 7-102
7-25 Alkali Flat-Furnace Creek Ranch Ground-Water Subbasin (USG96a) 7-104
-------
FIGURES (Continued;*
Page
7-26 Potentiometric Surface in the Death Valley Ground-Water Flow System
(USG96a) 7-107
7-27 Potentiometric Surface in the Amargosa Desert (USG90b) 7-110
7-28 Future Climates, Expressed in Terms of Overall Global Temperature Change
(GOO92) 7-121
7-29 Model Simulations of Past and Future Climate Conditions (BER91) 7-123
7-30 Repository Layout for the VA Reference Design.(DOE98) 7-128
7-31 Repository Location Within Yucca Mountain (DOE98) 7-130
7-32 North Portal Facilities Layout for the VA Reference Design Plan (DOE98) 7-131
7-33 21 PWR Waste Package Design for the VA Reference Design Plan (DOE98) ... 7-132
7-34 Defense HLW Package Design for the VA Reference Design Plan (DOE98) 7-133
7-35 Drift Cross - Section for the VA Reference Design Plan (DOE98) 7-135
7-36 Computer Code Configuration for the TSPA-VA (DOE98) 7-156
7-37 TSPA-VA Base Case Dose Rates for Periods up to 10,000 Years (DOE98) 7-163
7-38 TSPA-VA Base Case Dose Rates for Periods up to 100,000 Years (DOE98) .... 7-163
7-3 9 TSPA-VA Base Case Dose Rates for Periods up to One Million Years
(DOE98) 7-165
7-40 Uncertainties in the TSPA-VA Base Case Results (DOE98)
7-41 Structure of Performance Factors for NRC Performance Assessments
(NRC98) 7-188
7-42 Structure of NRC Computer Codes for Performance Assessments (NRC98) 7-189
7-43 NRC TSPA Results for Alternative Conceptual Models (NRC99a) 7-193
7-44 NRC TSPA Results for Mean-Values Data Set (NRC99b) 7-194
7-45 EPRI's IMARC Logic Tree (EPR98) 7-198
7-46 Results of EPRI's IMARC-4 Dose Evaluations (EPR98) 7-204
7-47 Comparison of DOE, NRC, and EPRI Performance Assessment Results
(derivedfromNRC99b) 7-206
8-1 Schematic Illustration of the Major Pathways From a Repository at Yucca
Mountain to Humans 8-2
8-2 Yucca Mountain and Surrounding Land Use 8-4
8-3 Winter Sites near Beatty and Belted Range 8-7
8-4 Major Winter Sites hi Ash Meadows and Pahrump Valley 8-8
8-5 Major Winter Sites in Northern and Central Death Valley 8-9
8-6 Map Showing Boundaries of Ground Water Subbasins hi the Study Area 8-18
8-7 Ground Water Usage in the Amargosa Desert 8-20
8-8 Location of Water Wells in the Amargosa Farms Area 8-25
8-9 Wells and Boreholes in the Amargosa Valley 8-33
8-10 Ground Water Pathway Model for Subsistence Fanner 8-54
XI
-------
FIGURES
No.
9-1 Schematic Illustration of the Major Pathways from a Repository at Yucca
Mountain to Humans 9-2
9-2 Retarded travel times of C-14 from the repository to the atmosphere for
particles released at 1,000 Years 9-6
9-3 Retarded travel times of C-14 from the repository to the atmosphere for
particles released at 10,000 Years 9-6
9-4 Annual Average Concentration for Uniform Continuous Source and Specific
Activity doe 100 Ci/year Source 9-9
9-5 An Illustration of Hypothetical Individual Dose Rates Associated with a
Disruptive Event Happening at Two Different Times after Disposal of
Radioactive Waste 9-15
Xll
-------
LIST OF ACRONYMS
ACHP Advisory Council On Historic Preservation
AEA Atomic Energy Act
AEC Atomic Energy Commission
AECB Canadian Atomic Energy Control Board
AECL Atomic Energy of Canada Limited
AFCN Belgian Nuclear Inspection Agency
AGNEB Swiss Interagency Working Group on Licensing of Nuclear Waste Facilities
AGR Advanced Gas-Cooled Reactor
AIRFA American Indian Religious Freedom Act
ALARA As Low As is Reasonably Achievable
ALI Annual Limit on Intake
ANDRA French Radioactive Waste Management Agency
ANL-W Argonne National Laboratory - West
Bq Becquerel
BEW Swiss Energy Office
BfS German Institute for Radiation Protection
BID Background Information Document
BMFT German Ministry for Research and Technology
BMU German Ministry for Environment, Protection of Nature and Reactor
Safety
BNFL British Nuclear Fuels Limited
BRGM French Bureau of Geological and Mineral Research
xni
-------
BSS Basic Safety Standards
BWR Boiling Water Reactor
CAA Clean Air Act
CCDF Complementary Cumulative Distribution Function
CEA French Atomic Energy Commission
CEC Council of the European Communities
CEDE Committed Effective Dose Equivalent
CEN Belgian Nuclear Research Center
CFR Code of Federal Regulations
Ci Curie
CLAB Swedish Centralized Spent Fuel Storage Facility
CNWRA Center for Nuclear Waste and Regulatory Analysis
CRPPH Committee on Radiation Protection and Public Health
CRWM Committee on Radioactive Waste Management
DACs Derived Air Concentrations
DCF Dose Conversion Factors
DDREF Dose, Dose Rate Effectiveness Factor
DOD U.S. Department of Defense
DOE U.S. Department of Energy
DSIN French Directorate for the Safety of Nuclear Installations
EBS Engineered Barrier System
EDE Effective Dose Equivalent
EDI Swiss Department of Interior
xiv
-------
EIA Environmental Impact Assessment
EIR Swiss Institute for Reactor Research
EIS Environmental Impact Statement
EMSL-LV Environmental Monitoring Systems Laboratory - Las Vegas
EnPA Energy Policy Act
EPA U.S. Environmental Protection Agency
EPRI Electric Power Research Institute
ERA Energy Reorganization Act
EURATOM European Atomic Energy Community
EVED Swiss Department of Transport, Communications, and Energy
FEIS Final Environmental Impact Statement
FERC Federal Energy Regulatory Commission
FFTF Fast Flux Test Facility
FRC Federal Radiation Council
GTCC Greater-Than-Class-C
Gy Gray
GW (e) Gigawatt - Electric
HEU Highly Enriched Uranium
HI Human Intrusion
HLW High-Level Waste
HSK Swiss Nuclear Safety Division
HTGR High-Temperature Gas-Cooled Reactor
IAEA International Atomic Energy Agency
xv
-------
ICPP Idaho Chemical Processing Plant
ICRP International Commission on Radiological Protection
IMARC Integrated Multiple Assumptions and Release Calculations
INEEL Idaho National Engineering and Environmental Laboratory
IPA Iterative Performance Assessment
IPSN French Institute for Nuclear Protecti on and Safety
IRQ Interagency Review Group
IRSR Issue Resolution Status Report
JAERI Japan Atomic Energy Research Institute
JNFL Japan Nuclear Fuel Services Limited
KASAM Swedish Consultative Committee for Nuclear Waste Management
KSA Swiss Commission for the Safety of Nuclear Installations
KTI Key Technical Issue
LET Linear Energy Transfer
LMFBR Liquid-Metal Fast-Breeder Reactor
LWR Light Water Reactor
MCLs Maximum Contaminant Levels
MCLG Maximum Contaminant Level Goal
MFRP Midwest Fuel Recovery Plant
MITI Japanese Ministry of International Trade and Industry
MPC Multi-Purpose Canister
mrem Millirem
MRS Monitored Retrievable Storage
xvi
-------
mSv
Millisieverts
MTHM Metric Tons of Heavy Metal
MTIHM Metric Tons of Initial Heavy Metal
MWd Megawatt Days
NAGRA Swiss Cooperative for the Storage of Radioactive Waste
NAS National Academy of Sciences
NCI National Cancer Institute
NCRP National Council on Radiation Protection and Measurements
NEPA National Environmental Policy Act
NHPA National Historic Preservation Act
NIREX British Nuclear Industry Radioactive Waste Executive
NPRM Notice of Proposed Rulemaking
NRC U. S. Nuclear Regulatory Commission
NRPB National Radiological Protection Board of the United Kingdom
NSC Japanese Nuclear Safety Commission
NTS Nevada Test Site
NUCEF Japanese Nuclear Fuel Cycle Engineering Facility
NWPA Nuclear Waste Policy Act
NWPAA Nuclear Waste Policy Amendments Act
NWPO Nuclear Waste Project Office
NWTRB Nuclear Waste Technical Review Board
OECD/NEA Organization for Economic Cooperation and Development/Nuclear Energy Agency
OCRWM Office of Civilian Radioactive Waste Management
xvii
-------
OMB Office of Management and Budget
ONDRAF Belgian Agency for Radioactive Waste and Fissle Materials
ORERP Off-Site Radiation Exposure Review Project
ORNL Oak Ridge National Laboratory
ORSP Off-Site Radiological Safety Program
PA Programmatic Agreement
PARCLAY Belgian Preliminary Demonstration Test for Clay Disposal
PBF Power Burst Facility
PICs Pressurized Ion Chambers
PNC Japanese Power Reactor and Nuclear Fuel Development Corporation
PPA Project Programmatic Agreement (Yucca Mountain)
PUREX Plutonium-Uranium Extraction
P WR Pressurized Water Reactor
QAP Quality Assurance Plan
R Roentgen
R&D Research and Development
RADWASS Radioactive Waste Safety Standards Radioactive Waste Management
RBE Relative Biological Effectiveness
rem Roentgen Equivalent Man
RBOF Receiving Basin for Off-site Fuels
RETF Japanese Recycling Equipment Testing Facility
RMEI Resaonably, Maximally Exposed Individual
ROD Record of Decision
xviii
-------
RSK German Reactor Safety Commission
SAB Science Advisory Board
SAR Safety Analysis Report
SCP Site Characterization Plan
SDWA Safe Drinking Water Act
SON French Agency Providing Architectural and Engineering Services
SHP Japanese Steering Committee on High-Level Radioactive Waste
8KB Swedish Nuclear Fuel and Waste Management Company ,
SKI Swedish Nuclear Power Inspectorate
SKN Swedish Board for Spent Nuclear Fuel
SNF Spent Nuclear Fuel
SNTZ Southern Nevada Transverse Zone
SRS Savannah River Site
SSI Swedish Institute for Radiation Protection
SSK German Committee on Radiological Protection
STA Japanese Science and Technology Agency
Sv Sievert
SZ Saturated Zone
TBM Tunnel Boring Machine
TEDE Total Effective Dose Equivalent
THORP Thermal Oxide Reprocessing Plant
TLD Thermoluminescent Dosimeter
TRU Transuranic
xix
-------
TSPA Total System Performance Assessment
TSPAI Total System Performance Assessment and Integration
UK United Kingdom
UNSCEAR United Nations Scientific Committee on the Effects of Atomic Radiation
URL Underground Research Laboratory
USDW Underground Sources Drinking Water
UZ Unsaturated Zone
VA Viability Assessment
WL Working Level
WLM Working Level Month
WIPP LWA Waste Isolation Pilot Plant Land Withdrawal Act
WVDP West Valley Demonstration Project
YMS Yucca Mountain Site
ZWILAG Swiss Cooperative of Nuclear Utility Operators
xx
-------
EXECUTIVE SUMMARY
ES. 1 INTRODUCTION AND BACKGROUND
The U.S. Environmental Protection Agency (EPA) is responsible for developing and issuing
environmental standards and criteria to ensure that public health and the environment are
adequately protected from potential radiation impacts from waste stored or disposed in Yucca
Mountain, Nevada. The Yucca Mountain site is located in Nye County, approximately 150
kilometers (90 miles) northwest of Las Vegas, Nevada, and on the southwestern boundary of the
Nevada Test Site. Yucca Mountain is an irregularly shaped elevated land mass six to 10 km
wide (four to six miles) and about 40 km (25 miles) long. A waste repository would be about
300 meters (one thousand feet) below the crest of Yucca Mountain and about the same distance
above the water table under the mountain.
The EPA is proposing, in 40 CFR Part 197, site-specific environmental standards to protect the
public from releases of radioactive materials disposed of or stored hi the potential repository to
be constructed at Yucca Mountain.1 These standards provide the basic framework to control the
long-term storage and disposal of three types of radioactive waste:
1. Spent nuclear fuel (SNF), if disposed of without reprocessing
2. High-level radioactive waste (HLW) from the reprocessing of spent nuclear fuel
3. Other radioactive materials that may be placed hi the potential repository
The other radioactive materials that could be disposed of hi the Yucca Mountain repository
include highly radioactive low-level waste, known as greater-than-Class-C waste, and excess
plutonium resulting from the dismantlement of nuclear weapons. However, the plans for
placement of these materials are uncertain and therefore, for the purpose of the present
rulemaking, the information presented hi this Background Information Document (BID) is
limited to spent nuclear fuel and high-level radioactive waste.
JIt is important to note that no decision has been made regarding the acceptability of Yucca Mountain for
storage or disposal. However, for the purposes of this document, the description of Yucca Mountain as "potential"
will generally not be used but is intended.
ES-1
-------
ES.1.1 Purpose And Scope of The Background Information Document
The BID presents the technical information used by EPA to understand the characteristics of the
Yucca Mountain site and to develop its proposed rule, 40 CFR Part 197. Most of the technical
information discussed in the BID is derived from investigations sponsored by the Department of
Energy (DOE). However, where appropriate, information from other sources, such as the
Electric Power Research Institute (EPRI) and U.S. Nuclear Regulatory Commission (NRC), and
Nevada state and local agencies is presented to supplement the DOE data base, to fill data gaps,
and to illustrate alternative conceptualizations of geologic processes and engineered barrier
performance.
The scope of the BID encompasses the conceptual framework employed by the Agency for
assessing radiation exposures and associated health risks. In general terms, this assessment
discusses the radioactive source term characterization, movement of radionuclides from the
repository at Yucca Mountain through the appropriate environmental exposure pathways, and the
estimates of potential doses to members of a representative group of people living in the region
around the repository site. It is not intended to be a technical critique of the investigations
conducted by DOE and other parties. Nor is it a regulatory compliance or criteria document.
The BID is simply a summary of the technical information considered by EPA hi developing the
rationale for and specifics in 40 CFR Part 197.
This executive summary highlights key chapters of the BID, particularly information concerning
efforts hi other nations to develop deep geological repositories (Chapter 3); current efforts to
develop a repository at Yucca Mountain (Chapter 4); the types and inventories of waste likely to
be disposed in Yucca Mountain (Chapter 5); geologic and hydrogeologic characteristics of the
repository site and anticipated repository performance (Chapter 7); and pathways for human
exposure to radionuclides potentially released from the site (Chapter 8). The reader is referred to
the full text of the BID for information regarding ways in which radiological dose and risk are
estimated (Chapter 6); potential exposure scenarios and compliance assessment issues for the
Yucca Mountain repository (Chapter 9); and the comparative radiological risks associated with
deep geological disposal and surface storage of spent nuclear fuel (Chapter 10),
ES-2
-------
ES.1.2 EPA's Regulatory Authority For The Rulemaking
The proposed standards governing environmental releases from the potential Yucca Mountain
repository have been developed pursuant to the Agency's responsibilities under the Energy
Policy Act (EnPA) of 1992 (Public Law 102-486). Section 801 of this Act directed EPA to
promulgate standards to ensure protection of public health from releases of radioactive material
in a deep geologic repository to be built in Yucca Mountain by setting standards to protect
individual members of the public. The EnPA also required EPA to contract with the National
Academy of Sciences (NAS) to advise the Agency on the technical bases for the Yucca Mountain
standards. The EnPA directed that these standards will apply only to the Yucca Mountain site
and are to be based upon and consistent with the findings and recommendations of the NAS:
• ...the Administrator shall, based upon and consistent with the findings and
recommendations of the National Academy of Sciences, promulgate, by
rule, public health and safety standards for protection of the public from
releases from radioactive materials stored or disposed of in the repository
at the Yucca Mountain site. Such standards shall prescribe the maximum
annual effective dose equivalent to individual members of the public from
releases to the accessible environment from radioactive materials stored
or disposed of in the repository.
ES.1.3 The National Academy of Sciences Recommendations
In the EnPA, the Congress asked the Academy to address three issues in particular:
• Whether a health-based standard based upon doses to individual members
of the public from releases to the accessible environment will provide a
reasonable standard for protection of the health and safety of the general
public;
• Whether it is reasonable to assume that a system for post-closure
oversight of the repository can be developed, based upon active
institutional controls, that will prevent an unreasonable risk of breaching
the repository's engineered or geologic barriers or increasing exposure of
individual members of the public to radiation beyond allowable limits;
and
• Whether it will be possible to make scientifically supportable predictions
of the probability that the repository's engineered or geologic barriers will
be breached as a result of human intrusion over a period of10,000 years.
ES-3
-------
To address these questions, the Academy assembled a committee of 15 persons representing a
range of scientific expertise and perspectives. The committee conducted a series of five technical
meetings at which more than 50 nationally and internationally known scientists and engineers
were invited to participate. In addition, the committee received information from the NRC,
DOE, EPA, Nevada State and county agencies, and private organizations, such as the Electric
Power Research Institute.
The committee's conclusions and recommendations are contained in its final report, entitled
Technical Bases for Yucca Mountain Standards, which was issued on August 1,1995. In this
report, the committee addressed the key issues posed by Congress and reached the following
conclusions:
• ...an individual-risk standard would protect public health, given the
particular characteristics of the site, provided that policy makers and the
public are prepared to accept that very low radiation doses pose a
negligibly small risk.
• ...it is not reasonable to assume that a system for post-closure oversight of
the repository can be developed, based on active institutional controls,
that will prevent an unreasonable risk of breaching the repository's
engineered barriers or increasing the exposure of individual members of
the public to radiation beyond allowable limits.
• ...it is not possible to make scientifically supportable predictions of the
probability that a repository's engineered or geologic barriers will be
breached as a result of human intrusion over a period of 10,000 years.
In addition, the report offered the Agency several general recommendations as to the approach
EPA should take hi developing 40 CFR Part 197. Specifically, the NAS recommended:
...the use of a standard that sets a limit on the risk to individuals of
adverse health effects from releases from the repository.
... that compliance with the standard be measured at the time of peak risk,
whenever it occurs. (Within the limits imposed by the long-term stability
of the geologic environment, which is on the order of one million years.)
• ... that the consequences of an intrusion be calculated to assess the
resilience of the repository to intrusion.
ES-4
-------
The EPA does not believe it is bound to adopt all of the positions advanced by the NAS in the
Yucca Mountain rulemaking. The Agency has used the NAS report as the foundational starting
point for the rulemaking. The Agency has carefully considered the recommendations of the
NAS, but the role of the NAS recommendations is not to replace the rulemaking authority of the
Agency. The Agency will tend to accord greatest deference to the judgements of NAS about
issues having a strong scientific component, the area where NAS has its greatest expertise. The
EPA will reach final determinations that are congruent with the NAS analysis whenever it can do
so without departing from the Congressional delegation of authority to promulgate, by rule,
health and safety standards for protection of the public. The Agency believes that such
determinations require the consideration of public comments and the Agency's own expertise
and discretion.
ES.1.4 Prior Agency Action
In December 1976, EPA announced its intent to develop environmental radiation protection
criteria for radioactive waste to ensure the protection of public health and the general
environment. These efforts resulted in a series of radioactive waste disposal workshops, held in
1977 and 1978. Based on issues raised during workshop deliberations, EPA published a Federal
Register notice on November 15,1978 of intent to propose criteria for radioactive wastes and to
solicit public comments on possible recommendations for Federal Radiation Guidance. In March
1981, EPA withdrew the proposed "Criteria for Radioactive Wastes" because it considered the
implementation of generic disposal guidance too complex given the many different types of
radioactive waste.
In 1982, Congress enacted the Nuclear Waste Policy Act (NWPA), which established the current
national program for the disposal of SNF and HLW. The Act assigned to DOE the responsibility
of siting, building, and operating an underground geologic repository for the disposal of these
wastes and directed the EPA to "promulgate generally applicable standards for the protection of
the general environment from off-site releases from radioactive material hi repositories." In that
same year, under the authority of the Atomic Energy Act (AEA), the EPA proposed a set of
standards under 40 CFR Part 191, "Environmental Standards for the Management and Disposal
of Spent Nuclear Fuel, High-Level and Transuranic Radioactive Wastes." After a number of
public hearings and comment periods, the EPA issued the final rule under 40 CFR Part 191 on
August 15,1985. Sections of this rulemaking were remanded by a Federal Court in 1987 and
repromulgated by EPA in 1993.
ES-5
-------
In December 1987, Congress enacted the Nuclear Waste Policy Amendments Act (NWPAA).
The 1987 Amendments Act redirected the nation's nuclear waste program to evaluate the
suitability of the Yucca Mountain site as the location for the first SNF and HLW repository.
Activities at all other potential sites were to be phased out.
In October 1992, the Waste Isolation Pilot Plant Land Withdrawal Act (WIPP LWA) was
enacted. While reinstating certain sections of the Agency's 1985 disposal standards, the Act had
the effect of exempting the Yucca Mountain site from these generic disposal standards.
However, also in October 1992, the EnPA directed the EPA to set site-specific radiation
protection standards for the Yucca Mountain disposal system.
ES-6
-------
ES.2 CURRENT U.S. PROGRAMS FOR YUCCA MOUNTAIN
The DOE, NRC, and EPA each have legislatively defined roles in the management and disposal
of SNF and HLW at the proposed Yucca Mountain disposal site. As stated in the NWPA, DOE
is responsible for developing, constructing, and operating repositories for disposal of these
wastes. The NRC has responsibility to license the repository and related facilities, and EPA is to
promulgate radiation protection standards which the NRC is to adopt as the basis for its licensing
actions. Affected state and local governments and Native American tribes have an oversight role
in the program. The NWPAA designated the Yucca Mountain site in Nevada as the only site to
be evaluated by DOE as a potential location for disposal of SNF and HLW, and established the
Nuclear Waste Technical Review Board (NWTRB) to provide oversight of the DOE program.
ES.2.1 The Department of Energy
The DOE's Office of Civilian Radioactive Waste Management (OCRWM) was established by
Congress in the NWPA specifically to provide management for the disposal of SNF from
commercial nuclear power reactors. Under a 1985 Presidential Executive Order, the repository
established by DOE is also to be used for disposal of HLW from DOE operations. The OCRWM
charter includes responsibility for receipt of SNF from reactors at the reactor sites; interim
storage of received SNF, as necessary, prior to disposal; transport of SNF to the site(s) for
interim storage and disposal; and siting, design, licensing, and operation of a central interim
storage facility and disposal facilities. In addition to its work at Yucca Mountain, DOE has
developed alternative designs for a central interim storage facility (known historically as a
Monitored Retrievable Storage (MRS) facility), but, as of mid-1999, the Department has not
established a site for such a facility.
In accord with the NWPAA, DOE has been evaluating Yucca Mountain as the disposal site for
SNF and HLW. Characterization of the site is proceeding with surface-based and sub-surface
activities. Design concepts for the engineered features of the repository are being developed.
Recent DOE activities produced the Viability Assessment (VA) report, which is a
Congressionally mandated appraisal of the viability of the Yucca Mountain project for geologic
disposal of nuclear wastes. The VA report contains:
• A site description and a design for engineered features of the repository and waste
package
ES-7
-------
• A Total System Performance Assessment, based on available data, describing the
probable safety performance of the VA reference design (TSPA-VA)
A plan and cost estimate for completing the license application (LA) to NRC for
repository construction
• Cost estimates for constructing and operating the repository
The VA report was published in December 1998. It will be followed by a draft environmental
impact statement (EIS) planned for 1999 and a final EIS in 2000. The site-suitability
recommendation, required by the NWPA, is planned to be submitted to the President in 2001 and
a License Application (LA) would be submitted to NRC in 2002 if the site is found suitable for
disposal. Significant recent accomplishments of the DOE program include:
• Completion of the Exploratory Studies Facility (ESF) and Cross Drift tunnels for
gathering experimental data at the proposed repository horizon
• Initiation of various types of experiments in the tunnel alcoves and niches
• Initiation of development of a market-driven approach to storage/transportation
technology
Completion of the Viability Assessment in December 1998
• Assessment of improved repository design options based on TSPA-VA results
ES.2.2 The Nuclear Regulatory Commission
The NRC is responsible for licensing and regulating the receipt and possession of SNF and
HLW, at privately owned facilities and at certain facilities managed by DOE. This licensing
responsibility includes the waste management and disposal facilities at Yucca Mountain. The
NRC currently licenses temporary storage facilities at reactor sites, as well as commercial storage
facilities at West Valley, New York, and Morris, Illinois.
NRC licensing of a repository at Yucca Mountain will be accomplished through review of a
License Application submitted by DOE after completion of site approval procedures set forth in
the NWPA. If the LA is found acceptable for review, NRC would review it to determine if there
is reasonable assurance of compliance with regulatory standards. If expectation of compliance is
established, DOE will be authorized to construct the repository. Subsequently, the LA will be
amended to seek approval to receive and emplace wastes for disposal.
ES-8
-------
Confirmatory testing is expected to continue throughout construction and disposal operations.
After disposal is completed (a process expected to span about 50 years), the LA would be
amended to request closure of the repository. After closure is authorized, post-closure
monitoring would be expected to be required.
The NWPA requires both EPA and NRC to publish radiation-protection standards and
regulations, for the storage and disposal of HLW. As previously noted, the EnPA directed the
EPA to develop radiation protection standards for the Yucca Mountain site and for the NRC to
develop implementing regulations that conform to the EPA's Yucca Mountain standards. The
NRC's proposed (February 1999) 10 CFR Part 63 regulations require use of multiple barriers
(natural and engineered) to achive compliance with regulatory standards, and implement the
Commission's principles of defense-in-depth and risk-informed regulation. The proposed rule
addresses licensing procedures, radiation exposure standards, criteria for public participation,
records and reporting, monitoring and testing programs, performance confirmation, quality
assurance, personnel training and certification, and emergency planning.
The NRC's proposed 10 CFR Part 63 regulations would be modified as necessary to conform to
EPA's 40 CFR Part 197 standards after they are established.
ES.2.3 Nuclear Waste Technical Review Board
The NWPAA established the Nuclear Waste Technical Review Board comprised of eleven
members recommended by the NAS and appointed by the President. These individuals are
experts in the fields of science, engineering, or environmental sciences and represent a broad
range of scientific and engineering disciplines, including hydrology, underground construction,
hydrogeology, and physical metallurgy. No member of the Board may be employed by DOE, its
contractors, or the National Laboratories. The current Board is composed of individuals with
academic and public and private sector experience. The Board's mandate is to evaluate the
technical and scientific validity of activities undertaken by DOE, regarding various aspects of the
U.S. SNF and HLW management. For example, the NWTRB provided comments in April 1999
on DOE's Viability Assessment in a report entitled Moving Beyond the Yucca Mountain Viability
Assessment - A Report to the U.S. Congress and the Secretary of Energy.
ES-9
-------
The NWTRB meets periodically in open public meetings. The Board reports to Congress and to
the Secretary of Energy at least twice a year on technical issues associated with the Nation's SNF
and HLW disposal program.
ES.2.4 State Governments and Native American Tribes
In both the NWPA and the NWPAA, the Congress provided for active State and Native
American tribe participation in the Yucca Mountain site evaluation process. The legislation
provides for financial assistance to the State of Nevada, and for affected tribes and units of local
government, for participation in program activities. The State of Nevada and affected tribes or
units of local government may also request assistance to mitigate any economic, social, public
health and safety, and environmental impacts that are likely to result from site characterization
activities at Yucca Mountain.
The Nevada legislature created the State's Nuclear Projects/Nuclear Waste Project Office
(NWPO) in 1985 to oversee Federal high-level nuclear waste activities in the State. Since then,
the NWPO has dealt primarily with the technical and institutional issues associated with DOE's
efforts to characterize the Yucca Mountain site. In addition, the counties contiguous to Nye
County, the host county for Yucca Mountain, have been determined to be affected parties and are
participating in program oversight.
ES-10
-------
ES.3 SPENT NUCLEAR FUEL AND HIGH-LEVEL WASTE DISPOSAL
PROGRAMS IN OTHER COUNTRIES
As in the United States, other countries that use nuclear power are establishing long-term
programs for the safe management and disposal of SNF and HLW. These countries include
Belgium, Canada, France, Germany, Japan, Spain, Sweden, Switzerland, and the United
Kingdom. Management strategies of these countries may include SNF storage at and away from
reactor sites, SNF reprocessing, HLW vitrification and storage, partitioning and transmutation of
the waste into short-lived or stable forms, and disposal in deep geologic media.
Deep geologic disposal is considered by the international scientific community to be the most
promising method for disposing of long-lived nuclear waste. As a consequence, all of the
countries discussed in this document envision emplacing solid radioactive waste in a deep
geologic formation located within their national borders.
Only the United States and Germany have identified candidate locations for disposal of HLW,
i.e., the Yucca Mountain site in Nevada and the Gorleben site in Germany. Other countries are to
varying degrees engaged in technical evaluations of the potential suitability of indigenous
geologic formations for disposal. Some nations, such as France, have several geologic
formations such as clay and granite, that might be used for disposal, and each alternative is being
evaluated. Others, such as Canada, have focused on one type of geologic formation. (Canada is
evaluating a crystalline rock formation in a setting with low seismic activity.) In addition,
several countries, such as Canada and Sweden, have established underground research
laboratories (URL's) and extended their research programs to include participation by other
nations with similar candidate geologies.
The disposal strategies of all nations assume that waste isolation will be maintained by reliance
on a combination of engineered and natural barriers between the emplaced waste and the
environment. Currently the United States is, as a result of site characterization data
interpretations, placing increasing emphasis on the role of engineered barriers in a potential
repository site at Yucca Mountain. This is, in part, due to the unique repository environment
and associated disposal strategy. Other countries, because of the characteristics of their available
geologic formations, are also placing emphasis on engineered barrier systems and are designing
these systems to ensure their long-term performance as a barrier to radionuclide release.
ES-11
-------
Various nations and international agencies, in addition to the United States, have begun to give
consideration to regulations and regulatory standards for SNF and HLW disposal. Some nations
have developed broad risk or dose criteria, and some have supplemented such criteria with
additional qualitative technical criteria concerning features of the disposal system. International
organizations, such as the Nuclear Energy Agency, provide opportunities for discussion of
regulatory criteria and also provide programs on issues of common interest.
Although the performance standards and the criteria for the various national regulations are
similar, each nation has established specific requirements to meet its needs. Current information
concerning the provisions of national and international criteria and objectives for the safety of
long-lived radioactive waste disposal is presented in Chapter 3 along with a summary of the
waste management programs of Belgium, Canada, Finland, France, Germany, Japan, Spain,
Sweden, Switzerland, and the United Kingdom.
ES-12
-------
Table 3-2. International Programs for HLW and SNF Disposal
COUNTRY
Belgium
Canada
Finland
France
Germany
Japan
Spain
Sweden
Switzerland
United
Cingdom
ESTIMATED
WASTE
AMOUNTS
2,500 MTHM
by 2000
34,000 MTHM
by 2000
2,440 MTHM
Not available
9,000 MTHM
by 2000
20,000 MTHM
by 2000
5,200 MTHM
(vitrified waste)
8,000 MTHM
by 2010
1,800 MTHM
by 2000
4,000 mj
by 2000
MANAGEMENT
ORGANIZATIONS
National Agency for Radioactive
Waste and Fossil Materials
(ONDRAF)
Atomic Energy of Canada Ltd.
(AECL)
Posiva Oy
National Radioactive Waste
Management Agency (ANDRA)
Institute for Radiation Protection
Steering Committee on High-
Level Radioactive Waste
Spanish National Radioactive
Waste Company (ENRESA)
Swedish Nuclear Fuel and Waste
Management Company (8KB)
National Cooperative for the
Storage of Radioactive Waste
(NAGRA)
British Nuclear Fuels, pic
(BNFL)
PROGRAM
STATUS
Conducting studies at the
Mol-Dessel Underground
Research Lab (URL)
Conducting studies at
Whireshell URL
Investigating three potential
sites
Working to identify suitable
site locations
Selected potential repository
site at Gorleben
Experiments being conducted
at Tono uranium mine and
Kamaishi iron ore mine
Developing conceptual
design
Conducting feasibility
studies; operates URL at
Apso
Considering appropriate
repository medium
Concentrating on deep
disposal of low to
intermediate level waste
PRIMARY REGULATORY
AGENCY
Federal Nuclear Inspection Agency
(AFCN)
Atomic Energy Control Board (AECB)
Finnish Centre of Radiation and
Nuclear Safety
Directorate for Safety of Nuclear
Installations (DSIN)
Federal Ministry for Environment,
Protection of Nature and Reactor
Safety
Atomic Energy Commission
Spanish Nuclear Safety Council
Ministry of the Environment and
Natural Resources
Nuclear Safety Division with the
Federal Department of Transport,
Communications, and Energy
Nuclear Installations Inspectorate;
Radiochemical Inspectorate; Ministry
of Agriculture, Fisheries, and Food;
UK Atomic Energy Authority;
Secretaries of State for Scotland and
Wales
TIME FRAME
Operation around 2030
Repository established by
about 2025
Repository operation
expected around 2020
Repository operations not
expected before 2020
Repository construction
to begin around 2000
Repository operation
expected by 2035 to 2045
Not available
Repository operation by
2008
Repository viability to be
determined by 2000;
commissioning of
repository will not occur
before 2020
Need for repository not
expected before 2040
ES-13
-------
ES.4 WASTE CHARACTERISTICS
Current national plans call for existing and yet-to be-produced inventories of SNF and HLW to
be disposed of in a Yucca Mountain geologic repository if it is approved for disposal. Each of
these waste forms is described below.
ESA1 Spent Nuclear Fuel
Spent nuclear fuel is defined as fuel that has been withdrawn from a nuclear reactor following
irradiation and whose constituent elements have not been separated by reprocessing. Generators
of SNF include: commercial nuclear power reactors, which consist of pressurized water reactors
(PWR) and boiling water reactors (BWR); reactors which are used in government-sponsored
research and demonstration programs in universities and industry; experimental reactors, e.g.,
liquid-metal, fast-breeder reactors (LMFBR) and high-temperature gas-cooled reactors (HTGR);
DOE Naval and nuclear-weapons production reactors; and Department of Defense (DOD)
reactors.
Commercial power reactors are by far the largest source of SNF. Approximately 98 percent of
the SNF from these reactors is stored at the reactor sites where it was generated. Spent nuclear
fuel from government research and production reactors is currently stored at various DOE
facilities. The fuels at these DOE facilities are Government-owned and, like commercial fuels,
there are no plans for reprocessing.
The fuel for commercial nuclear reactors consists of uranium dioxide pellets encased in
zirconium alloy (Zircaloy) or stainless steel tubes. During reactor operation, fission of some of
the uranium produces energy, neutrons, and radioactive isotopes known as fission products. The
neutrons cause further fission reactions and thus sustain the nuclear chain reaction. In time, the
uranium, is depleted to such a level that power production becomes inefficient. Once this occurs,
the fuel bundles are deemed "spent" and are removed from the reactor. Reprocessing of
commercial SNF to recover the unfissioned uranium and the by-product plutonium for reuse as a
fuel resource is currently not taking place in the United States.
The radioactive materials associated with SNF fall into three categories: 1) fission products; 2)
actinide elements (atomic numbers of 89 and greater); and 3) activation products. Typically,
fresh SNF contains more than 100 radionuclides as fission products. Fission products are of
ES-14
-------
particular importance because of the quantities produced, their high radiological decay rates,
their decay-heat production, and their potential biological hazard. Such fission products include:
strontium-90; technetium-99; iodine-129 and -131; cesium isotopes; tin-126; and krypton-85 and
other noble gases.
Activation products include tritium (hydrogen-3), carbon-14, cobalt-60, and other radioactive
isotopes created by neutron activation of reactor components, fuel assembly materials, and
impurities in cooling water or in the fuel pellets. The actinides include uranium isotopes and
transuranic elements, such as plutonium, curium, americium, and neptunium. The exact
radionuclide composition of a particular SNF sample depends on the reactor type, the initial fuel
composition, the length of time the fuel was irradiated (also known as "burnup"), and the elapsed
time since its removal from the reactor core.
As of January 1999, SNF from commercial reactor operations in inventory at various locations
amounted to 37,700 metric tons of initial heavy metal (MTIHM)2. Based on the DOE/EIA Low
Case assumptions of nuclear power capacity through the year 2030, the SNF inventory is
expected to increase to 87,900 MTIHM. This is the amount that would be produced by existing
commercial reactors under current licenses.
ES.4.2 Defense High-Level Radioactive Waste
High-level radioactive wastes are the intensely radioactive materials resulting from the
reprocessing of SNF, including liquid waste produced directly in reprocessing, and any solid
material derived from such liquid waste. High-level waste is generated by the chemical
reprocessing of spent research and production reactor fuel, irradiated targets, and fuel from U.S.
Naval propulsion reactors. The fission products, actinides, and neutron-activated products of
particular importance are the same for HLW as for SNF assemblies.
Historically, weapons program reactors were operated mainly to produce plutonium.
Reprocessing to recover the plutonium was an integral part of the weapons program. Naval
propulsion reactor fuel elements were also reprocessed to recover the highly enriched uranium
Commercial SNF reported in certain DOE documents is in units of metric tons of initial heavy metal
(MTIHM) to avoid difficulties arising from the need to estimate ranges of varied heavy-metal content that result
from different levels of enrichment and reactor fuel bum up. A metric ton (tonne) is 1,000 kilograms,
corresponding to about 2,200 pounds.
ES-15
-------
that remained after use. DOE decided in 1992 to phase out the domestic reprocessing of
irradiated nuclear fuel of defense program origin, so only minimal amounts of HLW are expected
to be added to the current inventory.
High-level radioactive waste that is generated by the reprocessing of SNF and targets contains
more than 99 percent of the nonvolatile fission products produced in the fuel or targets during
reactor operation. It generally contains about 0.5 percent of the uranium and plutonium
originally present in the fuel. Most of the current HLW inventory, which is the result of DOE
national defense activities, is stored at the Savannah River Site (126,300 m3), the Idaho National
Engineering and Environmental Laboratory (INEEL) (11,000 m3), and the Hanford Site (239,000
m3). A limited quantity of HLW is stored at the West Valley Demonstration Project (2,180 m3).
The HLW has, to date, been through one or more treatment steps, e.g., neutralization,
precipitation, decantation, or evaporation. It is currently planned that this HLW will be
solidified, using a vitrification process, for disposal. Vitrification of HLW is in progress at West
Valley and the Savannah River Site. A vitrification facility for HLW at Hanford is being
designed.
The DOE defense HLW at INEEL results from reprocessing nuclear fuels from naval propulsion
reactors and special research and test reactors. The bulk of this waste has been converted to a
stable, granular solid (calcine). At the Savannah River and Hanford Sites, the acidic liquid waste
from reprocessing defense reactor fuel is or has been made alkaline by the addition of caustic
soda and stored in tanks. During storage, this alkaline waste separates into three phases: liquid,
sludge, and salt cake. The relative proportions of liquid and salt cake depend on how much
water is removed by waste treatment evaporators during waste management operations.
Both alkaline and acidic HLW was generated at West Valley. The alkaline waste was generated
by reprocessing commercial power reactor fuels and some Hanford N-Reactor fuels. Acidic
waste was generated by reprocessing a small amount of commercial fuel containing thorium.
Projecting DOE defense HLW inventories is based on specific assumptions and may be subject
to change. New treatment methods and waste forms are possible and may affect the future
projections. Since all DOE defense production sites are progressing toward closure, there should
be minimal amounts of waste added to the current inventory. Interim storage of DOE HLW will
be required and will most likely continue to be at the site where the waste is produced. Current
ES-16
-------
DOE policy states that DOE HLW will not be accepted at the geologic repository until six years
after initial receipt of commercial SNF.
ES.4.3 Significant Radionuclides Contained in Spent Nuclear Fuel and High-Level Waste
Of the 70,000-tonne capacity limit for Yucca Mountain set by the NWPA, about 40,785 MTHM
and 22,210 MTHM represent spent PWR and spent BWR fuel, respectively. About 7,000
MTHM of vitrified defense HLW and SNF represents the balance of the total specified
repository inventory. For the Yucca Mountain site, radionuclide-specific activity levels are
estimated by assuming that all spent nuclear fuel had been removed from the reactors 30 years
before emplacement withbum-ups of 39,651 MWd/MTHM for PWR fuel and 31,186
MWd/MTHM for BWR fuel. Although the burn-up of SNF from which HLW is derived is
generally uncertain, this is thought to affect the adjustment for decay only marginally. In
addition, the radionuclide inventories hi a repository at Yucca Mountain stemming from defense
HLW are expected to be much less than those from commercial SNF.
The radionuclide inventory of the repository will change with time due to radioactive decay and
ingrowth of radioactive decay products. For example, inventories of the initially prominent
fission products cesium-137 and strontium-90, which have approximately 30-year half-lives, will
decay to insignificant levels within 1,000 years, while some decay products, such as neptuniurn-
237 with a half life of 2.1 million years, will not contribute significantly to doses until about
50,000 years after repository closure. Activity levels for very long-lived radioisotopes will be
low but nearly constant for periods on the order of a million years. Overall, the radioisotope
inventory of the wastes placed in the repository will decrease by about five orders of magnitude
during the first 100,000 years after closure, and remain virtually constant thereafter.
ES-17
-------
ES.5 CURRENT INFORMATION ON A POTENTIAL WASTE REPOSITORY AT
YUCCA MOUNTAIN
ES.5.1 Geologic Features of the Yucca Mountain Site
In terms of geologic designations, the site is situated in the southern section of the Great Basin,
which is characterized by north-south mountain ranges separating narrow, flat valleys. As is
typical of mountains in the region, Yucca Mountain is essentially a tilted fault block, with the
west side steep and nearly vertical and the east side sloping to the adjacent valley floor. The
crest elevation of the mountain is on the order of 1,500 to 2,000 meters (5,000 to 6,000 feet)
above sea level and is about 650 meters (2,000 feet) above the adjacent valley floors.
The geologic features of the southern Great Basin are highly complex and varied, with rock
formations ranging in age from 500 million to less than 400,000 years. The geologic structure at
Yucca Mountain is dominated by a series of layers of rocks that were produced by explosive
volcanic eruptions and are known as tuffs. The tuff layers have widely varying physical
characteristics and are on the order of 10 to 15 million years old. The host rock of the potential
repository is known as Topapah Spring tuff, which is a hard, fractured rock about 13 million
years old.
Geologic features of the region that are important to the integrity of a radioactive waste
repository in Yucca Mountain include faulting, seismicity, volcanism, and stability of the
geologic regime.
ES.5.1.1 Major Fault Features of the Yucca Mountain Area
The geologic formations, of which Yucca Mountain is a part, contain numerous major faults as a
result of deformation caused by tectonic movement. The faults are indicative of past and
potential movement of the geologic structures and they are potential pathways for water to
transport radioactivity released from the repository to the biosphere. The location of faults and
the extent of recent movement along the faults are important to the location and design of surface
facilities at the disposal site and to the design of the underground facilities into which the wastes
would be placed for disposal.
There are more that 80 known or suspected Quaternary faults and fault rupture combinations
within 100 kilometers (km) of the Yucca Mountain site. The DOE has determined that 38 of
these faults are capable of generating a peak acceleration of one-tenth the acceleration of gravity
ES-18
-------
(0.1 g) or greater at the ground surface of the proposed repository site; these are classified as
relevant earthquake sources. An updated compilation of faults prepared by the U.S. Geological
Survey identifies 67 faults with demonstrable or questionable evidence of Quaternary movement
and capability for accelerations of at least 0.1 g at an 84% confidence limit. The NRC-supported
program of the Center for Nuclear Waste Regulatory Analyses (CNWRA) has identified 52 Type
I faults within a 100-km radius of the mountain. Eleven known or suspected Quaternary faults
exist within 20 km (12 miles) of Yucca Mountain.
The three major faults hi the immediate vicinity of the proposed disposal site are the Ghost
Dance fault, which passes through Yucca Mountain; the Bow Ridge fault, which is just to the
east of the mountain; and the Solitario Canyon fault, which is just to the west of the mountain.
According to DOE's interpretation of available data, no movement on any of these faults has
occurred during the past 10,000 years.
ES.5.1.2 Seismology of the Yucca Mountain Area
The fault systems and the seismic history of the Yucca Mountain area are the result of regional
tectonics, which are dominated by the interaction of the North American and Pacific plates. The
tectonic processes that are stretching the Great Basin and produced its major land forms are the
result of the Pacific Plate moving northwest relative to the North American plate; the typical
geologic structures of the region were developed on the order of 11 million years ago. The
relative plate movements produced the past and recent seismic activity characteristics of the
region, outlined below.
Seismicity in the region of Yucca Mountain is concentrated hi several zones. The Southern
Nevada Transverse Zone (SNTZ) is nearest to the Yucca Mountain site and is the most
significant to repository performance. Historic earthquakes hi the SNTZ have been of moderate
magnitude with no documented surface rupture. The most recent earthquake in the vicinity of
Yucca Mountain was the Little Skull Mountain event, of Richter magnitude 5.6, in June 1992.
This earthquake was centered 20 km southeast of Yucca Mountain and was associated with the
Landers, California earthquake earlier that year. It caused minor structural damage to the Yucca
Mountain project field office near the mountain but had no apparent effect on geologic features
near the mountain. It was the largest earthquake ever recorded hi the vicinity of the site, based
on nearly 100 years of records.
ES-19
-------
Assessments of available data indicate that Yucca Mountain has not been subject to ground
accelerations at the surface in excess of 0.2 g for over several tens of thousands of years. At the
proposed waste emplacement depth of about 300 meters, the effects of ground motion are
expected to be insignificant. Empirical evidence of damage to 71 rock tunnels in Alaska,
California, and Japan resulting from earthquake shaking indicates that tunnel damage does not
occur at peak surface accelerations of less than approximately 0.2 g and only minor tunnel
damage occurs when the peak surface acceleration is between 0.2 and 0.5 g. Since ground
acceleration is not expected to impose significant design demands on either the underground
repository or surface facilities at Yucca Mountain, DOE does not consider seismicity to be a
significant factor in repository performance.
The Department also believes that future tectonic events are unlikely to significantly alter the
hydrologic characteristics of the Yucca Mountain site. This position is based on the assumption
that the current state of faults and fractures at the site is the result of cumulative past tectonic
events. The CNWRA has proposed, however, that a single tectonic event can cause significant
changes in hydrologic characteristics. Currently, there are five alternative tectonic models which
may form the basis for future assessment of relationships between tectonic phenomena and the
hydrologic regime.
ES.5.1.3 Volcanism
Yucca Mountain is composed of layers of volcanic rocks which originated in silica-rich eruptions
at what is now the Timber Mountain volcanic basin complex starting about 10 km north of
Yucca Mountain. The principal eruptions took place approximately 11 to 15 million years ago,
and ceased about 7.5 million years ago. After the silicic volcanism ended, there were two
episodes of basaltic volcanic rock formation. The most recent of these, which produced minor
ash deposits in the Lathrop Wells area to the southwest of Yucca Mountain, ended about 9,000
years ago.
DOE and NRC agree that a future occurrence of silicic volcanism is highly unlikely; the
consequences of such an event, therefore, do not need to be considered in assessments of a waste
repository at Yucca Mountain. The Department and the NRC have also recently reached
agreement on the likelihood of future basaltic volcanic events and their possible consequences.
One of the phenomena potentially associated with basaltic volcanism is sheet-like intrusion of
ES-20
-------
molten or liquid basaltic rock along fractures in the overlying rocks, i.e., the formation of dikes.
Given the history of volcanism in the Yucca Mountain region, there is some potential for magma
from a basaltic volcanic event to either intersect the repository footprint and directly affect the
waste or to form a nearby intrusive dike that might affect the waste isolation capability of the
natural system. If such intrusions occur, they could mobilize wastes and/or alter ground water
pathways. DOE and NRC are currently developing a mutually acceptable approach to estimating
the likelihood and consequences of such intrusions.
ES.5.1.4 Geologic Stability
The NAS report, "Technical Bases for Yucca Mountain Standards," recommended that
assessments of compliance with the Yucca Mountain standards be conducted for the time at
which the greatest risk occurs, within the limits imposed by long-term stability of the geologic
environment. The report also stated that long-term geologic stability for time periods on the
order of one million years can be expected; i.e., the contribution of geologic and hydrologic
features to overall repository system performance can be assessed for time periods of this
duration. The NAS report concluded that there is no technical basis for selecting a shorter
compliance period, such as 10,000 years. However, the NAS also stated that EPA may select a
shorter period based on policy considerations.
The concept of geologic stability does not imply absence of geologic activity or absence of
change in geologic processes. Rather, the concept implies that processes and events such as
climate change, tectonic movement, and earthquakes will occur as in the past, and that variations
within these processes and events will be boundable. The NAS report does not explicitly justify
the assertion of million-year stability by providing a synopsis and interpretation of the
documented geologic record. Some of the references cited in the report contain information
about the geologic record, but none of the cited references interprets the record to indicate a
million-year stability of the geologic regime or the processes associated with it. Until recently,
DOE documents containing information about the geologic features of the Yucca Mountain site
anticipated that performance assessments for a disposal system at the site would be evaluated in
terms of EPA's 40 CFR Part 191 regulations, which require evaluation of performance for a
period of 10,000 years.
ES-21
-------
The 10,000-year time frame for compliance with EPA's 40 CFR Part 191 regulations, which
applied to Yucca Mountain until Yucca Mountain was exempted by the WIPP Land Withdrawal
Act of 1992, was selected by the Agency because it was relatively brief compared to the time
frame for long-term factors, such as tectonic motion, that might affect the natural environment
and are not reasonably predictable over that period. On the other hand, the time period was long
enough to bring into consideration factors such as degradation of engineered barriers and
earthquakes that might affect disposal system performance and allow radionuclides to reach the
accessible environment.
Available information generally supports the NAS assertion that the fundamental geologic
regime at Yucca Mountain will remain stable over the next one million years. The overall
picture that emerges from available information is that the site and region had a highly dynamic
period of volcanism, seismicity, and tectonic action during the past, but that this very dynamic
situation has matured into one where the magnitudes, frequencies, locations, and consequences
of such phenomena relevant to long-term future disposal system performance can be bounded
and projected with reasonable confidence.
Performance assessments define the expected behavior of the waste isolation system over time.
Within the framework of expected repository performance, it is convenient to characterize future
repository conditions over three time periods. A similar breakdown was presented by DOE in its
1998 Viability Assessment. In the first, short-term period, lasting about 100 to 1,000 years, the
repository is characterized by intact waste canisters, high temperatures, and temperature
gradients which serve as driving forces for transients such as chemical reactions, and the
retention of short-lived and long-lived radioactivity in the canisters. Percolation water may or
may not contact the canisters, depending on local conditions determined by the arrangement of
waste packages in the repository and the pattern of percolation into the repository.
In the intermediate period, with a duration between 1,000 and 10,000 years, temperature
gradients are diminished or gone and the engineered features of the repository start to degrade.
During this time, canisters begin to corrode and only long-lived radioactivity remains; some of
the radioactivity is released from a few canisters which are penetrated by water, but most is
retained within the repository. Percolation water contacts and transports radioactive waste.
Releases are dominated by technetium-99 and iodine-129.
ES-22
-------
In the long-term period, from 10,000 years to 1,000,000 years, the repository gradually evolves
into an assemblage of the oxides, hydroxides, or carbonates of waste-package and waste-form
materials at ambient conditions. Percolation water seeps through the repository level and
transports radionuclides that can be mobilized to the environment, where the radionuclide
concentrations are diluted and dispersed by ground-water flow processes. Potential for radiation
doses is dominated by neptunium-237 released from the repository.
ES.5.2 Hydrologic Features of the Yucca Mountain Site
The proposed repository depth in Yucca Mountain (about 300 meters or 1,000 feet) would locate
it in a geologic formation not fully saturated with water (the unsaturated zone). The unsaturated
zone depth to the water table beneath the repository horizon is variable but is on the order of 300
meters. Water that infiltrates into the mountain, percolates through the repository, and moves
through the matrix of the geologic formations in the unsaturated zone will travel slowly, thereby
delaying entry of radionuclides released from the repository into the saturated zone and ground-
water system. Fractures in the rocks within the unsaturated zone can act as conduits for relatively
rapid movement of ground water through Yucca Mountain. Some radionuclides may be
chemically trapped in rock formations in the unsaturated zone..
ES.5.2.1 Characteristics of the Unsaturated Zone
Water flows slowly through the pore space in the matrix of partially saturated rock (the degree of
saturation in the Yucca Mountain formations is on the order of 80 to 90 percent) because there is
little area! recharge. If the pore space becomes saturated, the water will flow more quickly under
the existing hydrologic conditions. Also, water may flow quickly and preferentially through
fractures in the rock matrix. There is experimental evidence of "fast paths" for flow in some
rock fractures at Yucca Mountain. The fraction of total flow through these fast paths is uncertain,
may be episodic, and may be a small percentage of the total ground water moving through the
repository host rock. There is also evidence that some faults and fractures are barriers to flow
because of solids deposited along the fractures which block potential flow paths.
The complexity of the geologic structures in the unsaturated zone and the complexity of flow in
partially saturated media make it difficult to develop accurate models to predict flow rates and
flow paths in the unsaturated zone below the proposed repository location. Water flow and
ES-23
-------
storage in the unsaturated zone is three-dimensional and is controlled by structural, stratigraphic,
thermal, and climatological features of the system. The presence of features such as fractured
porous media, layered geologic units with widely varying hydrologic properties, tilted rock units,
and bounding faults can be expected to result in phenomena such as flow in both the fractures
and the matrix, diversion of flow by capillary barriers, lateral flow along discontinuities, perched
ground water zones, and vapor movement.
Water quantities that enter the mountain from precipitation and that percolate through the
geologic structures are spatially and temporally variable. The amount of water that percolates
along different paths is highly variable. Infiltration pathways depend on variations in the
properties of geologic units, the intersections of faults with the surface, and the presence of local
fracturing in individual rock units. Variations in the time it takes water to infiltrate are related to
the seasonality and relative infrequency of precipitation at Yucca Mountain. Over long time
frames, variations will occur because of climate changes. The interplay of all of these factors
may act to even out downward movement of ground water in the unsaturated zone with
increasing depth from the surface. There is evidence of rapid movement of infiltrating waters
along fracture zones in the rock.
Quantities of water that percolate through the mountain at the proposed repository depth cannot
be measured directly. Recent estimates, based on analysis of site characterization data, place the
percolation rate in the range of one to 10 mm/yr. Base-case performance assessments for the
TSPA-VA used a range of three to 23 mm/yr, with an expected value (60% probable) of 7.7
mm/yr. Values hi this range are as much as two orders of magnitude higher than values
previously estimated using more limited data. The TSPA-VA also used a model of future climate
involving "long-term average" conditions, with an expected infiltration rate of 42 mm/yr, and
"superpluvial" conditions with an expected infiltration rate of 110 mm/yr.
Models of water flow in the unsaturated zone take into account potential for flow in both the
matrix and fractures, the relative distribution depending on the quantities of ground water
available. For example, at high percolation rates, a larger fraction may be transported laterally
and/or transported in fractures, including fast-path fractures. The models also take into
consideration the possibility that radionuclides may be removed from water that is intercepted by
geologic media having a high capacity to chemically absorb and retain some radionuclides, such
as the zeolite materials in the Calico Hills formation.
ES-24
-------
Using uncertainty distributions for the flow parameters, models are used to estimate values for
performance factors, such as the time necessary for water to move through the unsaturated zone.
Results of studies to date show that radionuclides, carried by water through fractures, cross the
unsaturated zone much more rapidly than those in water that travels through the rock matrix.
Similarly, radionuclides strongly sorbed on rocks, such as the Calico Hills zeolites, have transit
times through the unsaturated zone 50,000 times longer than for radionuclides that are soluble
and travel with the water. The conceptual models and transport parameters for water flow and
radionuclide transport in the unsaturated zone, and the results obtained from use of the models,
will be refined by DOE as additional data concerning the unsaturated zone are obtained from
future site characterization work.
ES.5.2.2 Characteristics of the Saturated Zone
Water that percolates through the repository and the unsaturated zone below will enter the
saturated zone where ground water fills the pore spaces and fractures within these rocks. The
saturated zone at Yucca Mountain is located at depths on the order of 300 m below the repository
horizon. Radionuclides transported to this zone will move toward the environment away from
Yucca Mountain through ground water. Radionuclide concentrations in the saturated zone will
be reduced by dilution caused by dispersion as radionuclides are transported away from the
repository at rates and in directions according to the flow characteristics of the hydrologic
regime. The saturated zone is, like the unsaturated zone, composed of numerous layers of rocks
with widely different characteristics and complex structures resulting from the dynamic geologic
history of the region. Flow rates and directions are of interest for evaluating compliance with
EPA's standards, as are the locations at which radionuclides would be accessible to human use
and the radionuclide concentrations at those locations.
The sequence of volcanic rocks within and below Yucca Mountain has been described
hydrologically in terms of four hydrologic units characterized by their ability to transmit water.
Beneath the volcanic rocks, at depths on the order of 2,000 meters at some locations, are older
rocks which contain the Lower Carbonate Aquifer. The volcanic hydrologic units and the lower
carbonate aquifer may all have a role in transporting radionuclides from the repository to the
surrounding areas.
ES-25
-------
The thicknesses of the rock formations and the depth to water in the saturated zone vary
significantly with distance and direction from the proposed repository location. For example, the
volcanic rock units are believed to thin out and disappear to the south of Yucca Mountain where
they are covered by the alluvial deposits of the Amargosa Desert. In this region, as illustrated by
Figure ES.4-1, the formations containing the Lower Carbonate Aquifer are near the surface.
Depths to ground water currently used for human consumption and activities such as irrigation
are shallow in this area, i.e., on the order of a few tens of meters. Consequently, human
habitation and water supply wells are currently located in this area.
Available data indicate that much of the outflow from the volcanic aquifers moves laterally into
the alluvial aquifer as the volcanic rock formations thin out below it. The alluvial aquifer may
also be receiving water from the carbonate aquifer. The data are not sufficient to indicate where
and how these flow transitions occur. Comparison of recent water-level altitude maps with those
completed in the 1950s indicates that aquifer development may have had a significant impact on
water levels and flow directions. Pumping of the alluvial aquifer may have induced upward flow
from the lower carbonate aquifer into the alluvial system.
Discharge from the alluvial aquifer system can occur by interbasin flow, leakage to underlying
units, evaporation, and extraction for human use. Available data indicate that the major
discharge area for the alluvial aquifer system is Alkali Flat, known also as the Franklin Lake
Playa. The estimated discharge rate in this area is 10,000 acre-feet per year, primarily based on
bare-soil evaporation. Some of the alluvial aquifer flow may also move further to the southwest
and discharge hi the Death Valley region, but the extent of this is unknown.
Estimates of rates and quantities of ground water flow in the saturated zone are based on
estimates of values for hydraulic conductivity, hydraulic gradient, and effective porosity of the
formations through which the water is flowing. Hydraulic gradient (i.e., the change in water
level between two locations) is generally the parameter best known and most easily measured.
In the Yucca Mountain region, three regions with distinct hydraulic gradients, designated as
small, moderate, and large, have been identified. Their extent and characteristics are governed
by the complexity and characteristics of the geologic formations.
ES-26
-------
YUCCA MOUNTAIN
Center of
dispose) 3iie CHOCOLATE MOUNTAIN
North
Lower Clostic /quitard
fie. Sedtm«ntory)
Upper Clastic Aquttord
(ie. Sedimenlory)
Figure ES.5-1.
Schematic North/South Cross-Sectional Illustration of Thinning of Volcanic Units Beneath the Amargosa
Desert [U.S. Geological Survey, Structure of Pre-Cenozoic Rocks in the Vicinity of Yucca Mountain, Nye
County, Nevada, USGS Survey Bulletin 1647, 1985]
ES-27
-------
Of particular interest to repository performance at Yucca Mountain is the high-gradient area,
located about two miles north of the mountain. The cause of the gradient, in which water levels
decline by more than 900 feet over a distance of about a mile, is unknown. Possible causes
include a flow barrier, a fault, an intrusive dike from volcanic rock flow, or changes in the
detailed structure of the rocks. If the gradient is caused by a flow barrier of some type, a loss of
this barrier due to future geologic movements could cause a rise in the water table in the area of
the proposed repository. A rise hi the water table would not be expected to intercept the
repository, but it would decrease the thickness of the unsaturated zone and decrease the
radionuclide travel time from the repository to the environment.
Ground-water flow rates, like flow directions and quantities, are at present highly uncertain
because of limited data and the complexity of the geologic structures that create the hydrologic
regime. Flow rates in the alluvial aquifer, the volcanic rock aquifers, and the lower carbonate
aquifer will differ because of the different rock characteristics for these geologic regimes.
Ground-water movement hi the volcanic rocks of the saturated zone was estimated by DOE in
1993 to be in the range of 5.5 to 12.5 meters per year. A more recent estimate concluded that a
flow rate of five meters per year is in the middle of the range of reasonable estimates. However,
recent data from the Exploratory Studies Facility tunnel into Yucca Mountain suggest the
existence of "fast paths" through the unsaturated zone that can allow water to move from the
surface to depths as far as 300 m in 50 years. At present, DOE believes that only a small fraction
of percolating water is transported to the-repository level through these pathways. Flow rates in
the lower carbonate aquifer have been estimated to be in the range of three to 3,000 meters per
year, depending on location. Pressure gradients are such that water flow from the volcanic
aquifers to the lower lying carbonate aquifer presumably does not occur. While reliable
estimates of flow rates in the alluvial aquifers are not available, flow rates in these strata are
believed to be lower than in the carbonate strata. This has the effect of preventing radionuclide
from moving into the higher flow rate paths hi the carbonate aquifer. The area! extent of the
region where upward flow comes from the carbonate aquifer is highly uncertain.
If ground water containing radionuclides flows at a rate of five meters per year, it would take
1,000 years for the ground water to travel a distance of five kilometers and 4,000 years to travel a
distance of 20 kilometers. Concentrations of soluble radionuclides in the ground water at these
distances from the repository would depend on the initial concentration at the boundary of the
repository, the dilution that occurs as a result of mixing of water from various sources, and the
ES-28
-------
dispersion of radionuclides. Overall, the mixing, dilution, and dispersion processes have been
estimated potentially to reduce radionuclide concentrations at distances on the order of 20
kilometers from the repository by a factor of 10 in comparison with the initial concentration.
Estimation of dispersion on a kilometer scale is difficult. The DOE used expert elicitation to fix
this parameter. The experts estimated that the parameter could vary from 1 to 100 with an
average assumed value of 10. The amount of concentration reduction that occurs may depend on
the direction of flow (i.e., as a result of dispersion being controlled by rock structures along the
flow path) as well as the distance over which flow has occurred.
Nye County, NV is currently drilling a series of 20 deep and shallow wells south of the Yucca
Mountain site and in Amargosa Valley to monitor the behavior of the saturated zone. Some of
the wells will measure hydraulic parameters of the alluvial and tuff aquifers. Other deep
monitoring wells will be installed to measure the properties of the carbonate aquifer and to define
how this aquifer connects with the shallower tuff and alluvial aquifers. These data will support
modeling of the saturated zone flow and transport on both site-scale and regional-scale. Results
to date indicate that the alluvium is complex and layered.
ES.5.3 Climate of the Yucca Mountain Region
The region surrounding Yucca Mountain currently has an arid climate, with total annual
precipitation on the order of 170 mm (six inches) of water. Precipitation rates vary throughout
the year, averaging about 18 mm/month during the fall and winter months and nine mm/month
during the spring and summer months. Current climate conditions have apparently prevailed
during the past 10,000 years, i.e., since the last ice age. Prior to the ice age, the climate cycled
between wet and dry; during the wet periods, many of the valleys that are now dry contained
lakes.
Future variations of precipitation and temperature are climate factors of considerable interest for
predicting the performance of a repository at Yucca Mountain. These factors influence the
percolation rate of water through the repository and the transport of radionuclides released from
the repository to the environment.
Current arid conditions are expected to persist well into the future. These conditions are
associated with the rain shadow caused by the Sierra Nevada Mountains to the west, which are
ES-29
-------
still rising. In addition, increases in greenhouse gases and global warming may affect general
atmospheric circulation and local climate conditions at Yucca Mountain. A panel of experts,
convened by the CNWRA, estimated that an enhanced greenhouse effect would probably
produce warmer conditions than have been experienced during the past few thousand years, with
a likely increase in the upper limit of temperature in the Yucca Mountain region on the order of
two to three degrees Celsius. The time period during which these elevated temperatures would
persist would depend on assumptions about future human use of fossil fuels. In general,
increased temperatures would be accompanied by lower precipitation rates and, therefore, lower
rates of percolation through the repository. Opposite changes could occur, however, especially
in connection with any future glacial periods.
Performance assessments in the Department's TSPA-VA assumed that the climate alternates
between the present (dry) climate, a long-term average climate during which the precipitation
rate is twice the current rate, and a superpluvial climate during which the precipitation rate is
three times the current rate. The expected duration of the initial dry climate was 5,000 years.
Subsequent dry periods would have an average duration of 10,000 years. The expected duration
of the long-term average periods was 90,000 years. Two superpluvial periods of 10,000 years
each were assumed to occur over the 1,000,000-year model period. About 90% of the 1,000,000-
year model period was characterized as having long-term average climate. Climatic fluctuations
were predicted to have virtually no impact on repository performance assessments over a 10,000-
year time frame. Over the longer term, climate assumptions affect the time at which the peak
dose rate occurs but not its magnitude.
ES.5.4 Repository Design Concepts Under Consideration for Yucca Mountain
Design concepts for a potential waste repository at Yucca Mountain have evolved significantly in
response to information from sources such as site characterization data, repository system
performance assessments, external technical reviews, and refinement of a waste isolation
strategy. The original design concept envisioned vertical emplacement of simple steel canisters
in individual boreholes; current plans call for end-to-end horizontal emplacement of large,
complex waste packages in parallel, excavated drifts. Design details are expected to continue to
evolve until a final design is selected for the License Application, if the site is approved for
disposal.
ES-30
-------
The repository design can be characterized as a multi-barrier system that functions to delay the
failure of the waste package, delay the release of radionuclides from the waste package, and
mitigate the effects of radionuclide release. Key design factors important to repository
performance and radionuclide release potential include the corrosion resistance of the waste
package wall material; the use of techniques to deflect or delay contact of percolating water with
the waste packages; and the use of techniques to stop or delay migration of releases to the
environment.
One technique to delay waste package failure is to emplace the packages so that heat from the
wastes will keep temperatures in the repository high enough to vaporize percolation water, for as
long as possible. Corrosion by liquid water is thereby delayed until the heat emissions decrease
to levels such that water can enter the repository. Another technique for delaying waste package
failure is to use shields that deflect water dripping into the emplacement tunnels from contact
with the packages. In general, various technologies and concepts are available for each of the
basic functions for delaying waste package failure and decreasing radionuclide releases.
Key features of the design used by DOE in the recently completed Viability Assessment include
the following:
• Horizontal emplacement of 7,642 commercial SNF and 2,858 HLW canisters
positioned end-to-end in parallel, excavated, concrete-lined drifts, with an initial
thermal loading (to vaporize percolation water) on the surroundings
corresponding to emplacement of 85 MTHM/acre of reference spent commercial
reactor fuel
• Emplacement of waste canisters only between the Ghost Dance fault and the
Solitario Canyon fault
Disposal of 63,000 MTHM of spent commercial fuel and 7,000 MTHM
equivalent of defense HLW in 120 miles of tunnels and drifts, over 840 acres of
emplacement area, at depths on the order of one-eighth to one-quarter of a mile
• Construction of 29 surface buildings encompassing 800,000 square feet of floor
space and serving the operational needs of 300 underground drift excavation
personnel and 600 surface and subsurface operational personnel
ES-31
-------
• Use of commercial spent fuel waste packages that are two meters in diameter, six
meters in length, with an outer shell of 10-cm thick A 516 carbon steel and an
inner shell of 2-cm thick Alloy 22 corrosion-resistant alloy.
Waste types to be disposed would include commercial SNF fuel in bare assemblies; canistered
commercial SNF; canisters of vitrified defense HLW; SNF from nuclear defense programs; and
other DOE-owned SNF such as unprocessed fuel from Hanford's production reactors.
A plan view of the repository layout for the TSPA-VA is shown in Figure ES.5-2. In this
diagram, the dense array of parallel lines in the subsurface emplacement block represents the
drifts where the waste canisters would be emplaced.
In the TSPA-VA, DOE evaluated several design options as variants to the base case. The options
considered included use of backfill; use of drip shields to preclude water from impinging on the
waste packages; use of a ceramic coating on the waste packages to defer corrosion; and whether
or not to take credit for fuel rod cladding as a barrier to radionuclide mobilization and transport.
During early 1999, DOE evaluated alternative repository designs based on, and evolved from, the
Viability Assessment design. These alternative designs were intended to reduce uncertainties in
performance identified in the TSPA-VA analysis. DOE has selected, as the reference design for
the Site Recommendation, a design whose key features include an area mass loading of 60
MTU/acre, drift spacings of 81 m, waste packages with 2 cm of Alloy 22 over 5 cm of 316L
stainless steel, use of steel ground support, and use of drip shields. These design features are
expected to significantly reduce uncertainties and technical issues associated with the VA
reference design. Engineered design concepts may continue to evolve to the design selected for
the LA.
ES-32
-------
N
ta 30000
Softtarw
Conyon
Fault \
Foutfi - I/town on urfoc* oxcop/
m 4Mng oma wnwv IMy on oAo
Figure ES.5-2. Repository Layout for the TSPA-VA Design
ES-33
-------
INNER BARRIER
(ALLOY 22)
SIDE GUIDE (A 516)
INTERLOCKING PLATES
(CUTAWAY VIEW)
(STAINLESS STEEL BORON)
INNER BARRIER LID
(ALLOY 22)
OUTER BARRIER LID
(A 516
OUTER BARRIER LID s
(A 516)
CORNER GUIDE
(A 516)
CORNER STIFFENER (A 516)
SIDE COVER (A 516)
TUBE (A 516)
INNER BARRIER LID
(ALLOY 22)
OUTER BARRIER
(A 516)
Figure ES.5-3. Waste Package for 21 PWR Uncanistered Fuel Assemblies
ES-34
-------
ES.5.5 Repository System Performance Assessments
Assessments of future repository system performance are currently used by DOE to aid
repository design and, in association with the license application, will be used to demonstrate
compliance with regulatory standards. The assessments are done using analytical models of
factors that affect performance, such as the waste package corrosion rate, and computer codes
that combine the models of performance factors with performance parameter values and
modeling assumptions.
The NRC is developing independent capability to perform performance assessments in order to
be able to review DOE's license application. In addition, the Electric Power Research Institute
has developed performance assessment methods that are used in its oversight of the government
program, and EPA has developed methods that are used in support of its promulgation of the
Yucca Mountain standards.
Prior to the TSPA-VA, DOE issued reports on its total system performance assessments in 1991,
1993, and 1995. None of the DOE's total system performance assessment (TSPA) results to
date, including those reported in the Viability Assessment documentation, are considered to be
sufficient for licensing a repository at Yucca Mountain. However, they do serve to aid evolution
of the DOE program and the eventual selection of the engineered design for the License
Application if the site is found suitable for disposal. They also serve as precursors of the TSPA
results that will be submitted with the License Application to demonstrate compliance with
regulatory requirements. The expected performance of the repository will be strongly dependent
upon the combination of engineered barriers finally selected by DOE.
The performance assessment models and codes address design features that affect repository
system performance, such as those identified in Figure ES.5-5. They also consider geologic and
hydrologic features that can affect performance, such as those discussed in Sections 4.1 and 4.2
of this BID, and uncertainties in these factors that affect uncertainties in demonstrating
compliance with regulatory requirements. Types of uncertainties that are considered include
uncertainties in measured values of performance parameters such as corrosion rates and
hydrologic parameters; spatial variability of parameters such as percolation rate and temperatures
around the repository; temporal, variability of factors such as annual variation of precipitation and
ES-35
-------
future climate change; and uncertainties in the analytical models as a result of simplifications or
imperfect knowledge of the processes simulated by the models.
These uncertainties are taken into consideration by the codes used in TSPAs through use of
probabilistic techniques in selecting the model parameters for the calculations. The numerical
values for uncertain performance parameters used in the TSPA codes are characterized using a
range of possible values. Calculational techniques are used to sample values from the
distributions to produce a large number of individual TSPA results which collectively
characterize the uncertainty in the overall system performance as a result of uncertainties in the
individual factors. DOE uses the Repository Integration Program (RIP) as the integrating shell to
link the various component codes. RIP includes parameter sampling capability to assess
uncertainty in dose rates as a function of uncertainty in the component models.
During the first 10,000 years, based on the TSPA-VA base case assumptions, one juvenile waste
package failure occurs at 1,000 years because of manufacturing defects and 17 additional waste
packages fail from corrosion, beginning at 4,200 years. The only radionuclides to reach the
biosphere are nonsorbing nuclides with high inventories — technetium-99 and iodine-129. These
produce a peak dose rate of 0.04 mrem/yr at 10,000 years. At 100,000 years, predicted dose rates
are dominated by neptunium-237 and reach a peak value of 5 mrem/yr. The base case dose rate
for TSPA-VA evaluations reaches a maximum value of about 300 mrem/yr at 300,000 years and
then declines to about 50 mrem/yr at one million years. Uncertainty in the base case results, as
measured by the spread between the 5th and 95th percentile dose rates, was about three orders of
magnitude at 10,000 years (i.e., from 0.001 to 1 mrem/yr) and about four orders of magnitude at
one million years (i.e., from 0.01 to 100 mrem/yr).
In the TSPA-VA, DOE found that the most important modeling assumption contributing to
uncertainty in peak dose rate for all time periods was the seepage fraction (i.e., the total number
of waste packages contacted by water seepage into the drifts). Other important factors
contributing to uncertainty during the first 10,000 years were the mean corrosion rate of the
waste package, the number of juvenile waste package failures, and the saturated-zone dilution
factor.
In the TSPA-VA, DOE considered the impact of four disruptive events on repository
performance: basaltic igneous intrusion, seismic activity, nuclear criticality, and inadvertent
ES-36
-------
human intrusion. The peak dose from volcanism was many orders of magnitude lower than for
the base case. Seismic events were determined to have no effect on the repository over
1,000,000 years. Nuclear criticality external to a waste package was determined not to be a
credible event and criticality within a waste package would have insignificant consequences.
Inadvertent drilling intrusion as a result of searching for water was assumed to occur at 10,000
years. Before this time, it would not be possible for a drill to penetrate a waste package because
significant degradation had occurred. The intrusion event was assumed to release 550 to 2,700
kg of spent fuel, which fell to the bottom of the borehole in the saturated zone. For an intrusion
event releasing 550 kg of fuel (i.e., 5% of the waste package inventory), the peak dose rate was
found to occur at 12,000 years and to be 3.7 times the base case dose rate, 0.04 mrem/yr. The
peak dose rate for this generic intrusion scenario, which adds intrusion release to base case
release, is calculated to be no more than 1 mrem/yr under the assumptions of the analysis.
The NRC is developing its own performance assessment models and codes for use in pre-
licensing technical exchanges with DOE, and, ultimately, for performing reviews of DOE's
license application for a repository at Yucca Mountain. The NRC models are similar in concept
and content to the DOE models in that they include models of the various factors relevant to
disposal system performance and have the capability to address uncertainties. NRC's recent
modeling has shown results similar to those developed by DOE in the TSPA-VA, even though
significant differences in underlying assumptions exist between the two approaches.
The NRC has estimated that the 10,000-year dose rate is about 0.003 mrem/yr as compared to the
equivalent TSPA-VA dose rate of 0.04 mrem/yr. At 100,000 years the NRC calculated dose rate
is 0.2 mrem/yr while the equivalent value in the TSPA-VA is 5 mrem/yr. Reasons for the
differences are not readily apparent because the parameters and modeling approaches used by the
two agencies differed markedly. For example, the NRC did not assume credit for fuel rod
cladding as an engineered barrier while in the TSPA-VA DOE did take credit for Zircaloy
cladding. The NRC assumed that dilution of ground water radionuclide concentrations occurs
during pumping by the dose receptor; the DOE assumed that this dilution did not occur.
The Electric Power Research Institute has also conducted performance assessments using models
and codes which address the same general features and processes as those modeled by DOE.
However, the codes differ in approach and detail from those used in the TSPA-VA. Differing
ES-37
-------
parameters are also used by the two organizations. In spite of these differences, 10,000 -year
performance assessment results were in reasonable agreement. The 10,000-year dose calculated
by EPRI was 0.08 mrem/yr as compared to the DOE TSPA-VA estimate of 0.04 mrem/yr. After
100,000 years the EPRI dose rates were about two orders of magnitude lower than the TSPA-VA
results due, at least in part, to differing assumptions about the available inventories of iodine -
129 and technicium-99.
ES-38
-------
ES.6 BIOSPHERE PATHWAYS LEADING TO RADIATION EXPOSURE
In order to evaluate compliance of the repository system performance with regulatory
requirements, potential radiation doses to humans from repository releases must be calculated.
This evaluation requires estimating radionuclide releases; modeling their movement through the
environment; selecting and characterizing the person(s) for whom the potential radiation dose is
to be evaluated; and characterizing the pathways by which the person(s) receives the dose.
This estimation also requires assumptions concerning the location and exposure scenarios of an
individual or group of individuals likely to be at greatest risk from potential radionuclide releases
after repository closure and removal of institutional controls. Prior to closure, such assumptions
are unnecessary because possible contamination levels can be measured with considerable
accuracy both within and outside the repository.
Releases of radionuclides from the repository are not expected to occur sooner than several
thousands of years in the future; the start of release might be deferred much longer if certain
repository design features are used (i.e., those aimed at delaying the start of release, such as
corrosion-resistant drip shields and waste packages). After release from the repository, the
radioactivity would migrate through environmental pathways until it reaches the location of the
person(s) selected for the evaluation of potential doses. Thus, radiation doses might first be
incurred many thousands of years in the future, when locations and lifestyles of humans in the
vicinity of Yucca Mountain might differ from those of the present. Human locations and
lifestyles far in the future cannot, however, be reliably estimated. Therefore, evaluations of
future potential radiation doses are based on an understanding of current patterns of human
habitation, physiology, and activities as well as current technology. This approach to addressing
future states was affirmed by the NAS who concluded in their study mandated by the EnPA that:
• ...based on our review of the literature we believe that no scientific basis
exists to make projections of the nature of future human societies to within
reasonable limits of uncertainty
• ...it is not possible to predict on the basis of scientific analyses the
societal factors required for an exposure scenario. Specifying exposure
scenarios therefore requires a policy decision that is appropriately made
in a rulemaking process conducted by EPA.
ES-39
-------
ES.6.1 Current Demographics and Land and Water Use
The boundaries of the unincorporated town of Amargosa Valley (the closest population center to
the repository site) encompass almost 500 square miles of the Amargosa Desert. The boundaries
of the town include all of the area in which the highest potential doses from a repository at Yucca
Mountain are anticipated. The remoteness and arid climate of the area are reflected by its
population of only about 1,000 residents. Only about 11 percent of the land is held privately; the
remainder is under Federal control.
Currently, agricultural activities in the Yucca Mountain region are restricted to the Ash Meadows
area and the southern portion of Amargosa Valley. Two commercial alfalfa farms, a dairy farm,
and one commercial sod farm operate full-time in the Valley; most other farms in the area
operate on a part-time basis. Despite some difficulties, a wide range of crops and livestock can
be raised. Alfalfa, hay and grass, wheat, fruits and melons, vegetable, cotton, nuts, poultry, beef
cattle, dairy cattle, and fish are being or have been grown on farms and ranches in Amargosa
Valley. However, because of local conditions, the population in the region does not currently
grow significant quantities of leafy vegetables, root vegetables, and fruit and grain crops for its
own use. Presently, no farming occurs closer than about 23 kilometers south of the repository
site.
Primary uses of water in the Amargosa Valley include domestic, industrial, agricultural, mining,
and recreational. Most residences are supplied by individual wells, though some trailer parks,
public facilities, and commercial establishments are served by small private water companies. A
number of springs also supply water, primarily to the resort area in Death Valley.
Water use data for Hydrographic Basin 230 (Amargosa Desert) in 1997 was 940 acre-feet for
domestic, quasi-municipal, and commercial uses exclusive of mining and irrigation. As such, the
usage is typical of a small rural residential community. The average per person use rate was 0.8
acre-feet per year. Since no major demographic changes are expected, these values should be
representative of future communities in the regions.
At the present time nine farms varying in size from 65 to 800 acres are cultivating alfalfa in the
area. It is estimated that a total of 2,500 acres is being cultivated in 1999 and that water usage
for alfalfa irrigation is, as limited by current allocations, 5 acre-feet per acre. The nine alfalfa-
ES-40
-------
growing operations have an average size estimated to be 255 acres. This results in an average
annual water use for irrigation of 1,275 acre-feet per year. The domestic use of water by a small
farming community of 25 people is estimated to be 10 acre-feet per year, so the average volume
of water that would supply the annual water needs of a hypothetical future agricultural small
community would be 1,285 acre-feet.
ES.6.2 Radiation Protectionj>fIndividuals
According to current understanding, contaminated ground water is the principal pathway by
which a release of radionuclides from a repository at Yucca Mountain could cause radiation
exposures to humans. Figure ES.6-1 illustrates the ground-water pathway leading to human
exposure from an undisturbed repository at Yucca Mountain. The major reservoirs (source
terms) containing radionuclides at various times following closure are depicted as rectangles.
Solid arrows between reservoirs represent the probable processes by which radionuclides are
transported from one reservoir to another hi an undisturbed repository. Major processes and
events with the potential to modify normal behavior or drastically alter the physical integrity of
reservoirs are shown hi the figure as diamonds. These modifiers are connected by dashed lines to
those reservoirs most likely to have the most significant impact.
Individuals in a human population may have greatly different responses to radiation exposure
reflecting differences hi factors such as age, life style, and family history. In addition, their
potential exposure to radiation released from a repository at Yucca Mountain will depend on
factors such as where they live and what they eat and drink. A wide range of radiation exposures
and effects is therefore possible. Because of these variations, some specification of the exposure
conditions to be considered hi measuring compliance must also be part of the regulations.
Specifying some variables in the compliance evaluations would provide a means to narrow and
characterize the range of conditions for which evaluations of compliance are to be made. More
than one approach is possible for assessing potential radiation doses to individuals down the
hydrologic gradient from the repository.
ES-41
-------
Critical group for
population dos«
Critical group?
Figure ES.6-1. Schematic Illustration of the Major Pathways from a Repository at
Yucca Mountain to Humans (copied from the NAS report, 1995)
ES-42
-------
ES.6.3 Dose Estimation Approaches
To determine the risk to exposed individuals resulting from contaminated ground water requires
the development of a comprehensive exposure scenario that specifies discrete pathways and
quantifies the intake of individual radionuclides. Pathways for human exposure from
contaminated well water include internal exposure from the ingestion of drinking water,
vegetables, fruits, dairy products, and meats. For persons engaged in agricultural activities,
internal exposure may also result from the inhalation of airborne contaminants resuspended from
soil irrigated with contaminated water. Over time, the buildup of soil contaminants could reach
levels that also give significant external doses.
The implementation of an exposure scenario appropriate for a specified population requires a
complex array of pathway parameter values that define potential radionuclide concentrations in
various media to which individuals may be exposed. Exposure scenarios must also provide
quantitative descriptions that include where individuals live, what they eat and drink, and what
their sources of food and water are. Many key parameters needed to model human exposures at
Yucca Mountain are highly site-specific and reflect the desert conditions of the sparsely
populated Amargosa Valley. For example, the combined impacts of low rainfall, desert
temperatures, and soil quality mandate extensive irrigation of farm crops and use of local ground
water for cattle. Under these conditions, contaminated well water has the potential for
developing unusually high radionuclide concentrations in all locally grown food products.
The Cricial Group Approach to Chatacterization of the Dose Receptor
The NAS report recommended the use of the critical group concept for the development of
environmental standards. The critical group concept was first introduced by the International
Commission on Radiation Protection (ICRP) in order to account for the variation in dose in a
given population which may occur due to differences in age, size, metabolism, habits, and
environment. This concept was adopted in total by the NAS panel, although the Academy differs
from ICRP in the implementation of the concept. The ICRP defines the critical group in dose
terms, while the NAS adapted the concept to individual risk. The critical group is defined by the
ICRP as a relatively homogeneous group of people whose location and lifestyle are such that
they represent those individuals expected to receive the highest doses (or be at highest risk) as a
ES-43
-------
result of radioactive releases. As part of the critical group definition, the ICRP specifies the
following additional criteria (also adopted by the NAS panel):
* Size - The critical group should be small in number and typically include a few to
a few tens of persons.
• Homogeneity among members of the critical group - There should be a relatively
small difference between those receiving the highest and the lowest doses. It is
recommended that the range between the low and high doses not differ by more
than a factor often or a factor of about three on either side of the critical group
average.
• Magnitude of dose/risk - It is suggested that the regulatory limit defined by a
standard exceed the calculated average critical group dose by at least a factor of
ten.
• Modeling assumptions - In modeling exposure for the critical group, the ICRP
recommends that dose estimates be based on cautious, but reasonable
assumptions.
The ICRP does not, however, prescribe the lifestyle, habits, or conditions of exposure that may
define a critical group into the future. Its generic recommendations suggest use of current
knowledge and cautious, but reasonable, assumptions for characterizing future exposure
scenarios.
To account explicitly for future uncertainties, the NAS report offered two probabilistic modeling
approaches. The first, described hi Appendix C of the NAS report, A Probabilistic Critical
Group Approach, uses statistical methods and probability values to characterize members of the
critical group. The second, The Subsistence-Farmer Critical Group, described in Appendix D of
the report, also employs a probabilistic method, but identifies the subsistence farmer as the
principal representative of the critical group.
The NAS Subsistence Farmer Critical Group model is quite similar to the RMEI approach
(described below) that is used by EPA to characterize the dose receptor for purposes of
rulemaking. The model described in Appendix D of the NAS report specifies a priori one or
more subsistence farmers and makes assumptions designed to define a highly exposed farmer as
representative of the critical group. Subsistence farming does not exclude commercial farmers
who raise food for personal consumption, in addition to cash farm products. The NAS assumed
ES-44
-------
the subsistence farmer of the future would have nutritional needs consistent with those of a
present-day person. Like the subsistence farmer of today, most or all drinking water would be
obtained from an on-site well also used in the production of all consumed food. The subsistence
farmer is also assumed to live his/her entire life at the same location. Thus, the magnitude of the
dose to a subsistence farmer will largely be defined by the radionuclide concentrations in ground
water at the point of water withdrawal.
EPA's Reasonably, Maximally Exposed Individual as the Dose Receptor
EPA has developed a method for estimating potential radiation doses based on the concept of the
reasonably, maximally exposed individual (RMEI). The RMEI concept, which involves
estimating the dose to a person assumed to be at high risk based on reasonable (i.e., not overly or
insufficiently conservative) assumptions, has been used in previous agency programs and
guidance.
The total population that might be exposed from ground-water pathways is very small. There are
only about ten people living in community of Lathrop Wells in Amargosa Valley about 20 km
from the Yucca Mountain site. If this small population was defined as the critical group, the
exposure to the group would likely be on the same order as if the exposure was defined based on
an RMEI living at that location. Thus, hi the Yucca Mountain setting, there is no significant
difference between the critical group and the RMEI.
The basic approach for estimating doses to be incurred by the RMEI is to identify and
characterize the most important exposure pathway(s) and input parameters. By using maximum
or near-maximum (e.g., 95th percentile) values for one or a few of the most sensitive parameters,
while assuming average values for others, the resulting dose estimates should reasonably
correspond to the near-maximum exposures to any member of the exposed population. The
ultimate objective of the approach is to define an exposure well above average exposures, but
within the upper range of possible exposures. The RMEI is not intended to represent the most
extreme case.
ES-45
-------
ES.6.4 Exposure Scenarios
The EPA has considered four basic scenarios for estimating potential exposures of the RMEI in
the Yucca Mountain area. The scenarios involve characteristics of the region and represent
potential human habitation patterns and lifestyles in the Yucca Mountain region based on local
climatic, geologic, and hydrologic conditions.
(1) Subsistence (low technology) Farmer. In this scenario, the farmer is assumed to live hi the
Yucca Mountain area and to be exposed chronically (both indoors and outdoors) to residual
concentrations of radionuclides in soil through all exposure pathways. Contaminated water from
the aquifer is the only source of water for these individuals. The location and habits of this
individual will be consistent with historical locations, and easily accessible water (approximately
30-40 km from the disposal system). All the individual's food and water would come from
contaminated sources.
(2) Commercial Farmer. Based upon economic factors and current technologies, certain areas
around Yucca Mountain are suitable for commercial crop production. These areas are either
currently being farmed (approximately 30 km from the Yucca Mountain disposal system) or
could be economically viable based upon reasonable assumptions, current technology, and
experience in other parts of the arid west. In addition, some parts of the region could possibly
support emerging technologies such as hydroponic applications and fish fanning. Exposure
pathways in this scenario are the same as those described for the subsistence farmer.
(3) Rural-Residential Person. In this scenario, individuals are assumed to live closer to Yucca
Mountain and to be exposed through the same pathways described for the subsistence farmer in
Scenario 1. However, in this case the residents are not assumed to be full-time agricultural
workers. Instead, these individuals work primarily out of the area and engage only in light
fanning and recreational activities within it. Furthermore, it is assumed that all of the drinking
water ( 2 liters/day) and some of the food production will involve use of water contaminated with
radionuclides. This lifestyle is typical of most of the people currently living in the Amargosa
Valley.
ES-46
-------
(4) Domestic Use of an Underground Drinking Water Supply. Based upon current water usage in
the arid West, there could be an hypothetical water supply which could serve a community living
north of Interstate 95 closer to the repository site (inside the Nevada Test Site).
For each of these four scenarios, there are eight exposure pathways to be evaluated:
• External radiation from radionuclides in soil
• Inhalation of resuspended soil and dust containing radionuclides
• Inhalation of radon and radon decay products from soil containing radium
• Incidental ingestion of soil containing radionuclides
• Ingestion of drinking water containing radionuclides transported from soil to
potable ground water sources
• Ingestion of home-grown produce contaminated with radionuclides taken up from
soil
• Ingestion of meat (beef) or milk containing radionuclides taken up by cows
grazing on contaminated plants (fodder)
• Ingestion of locally-caught fish containing radionuclides
ES.6.5 Compliance Evaluation
The above discussion of receptor groups and exposure scenarios illustrates the factors involved
in assessing compliance with radiation protection standards. In practice, the critical group and
exposure scenarios to be used in assessing compliance with EPA's standards for Yucca Mountain
will be implemented under regulations to be developed by the NRC in conformance with the
EPA standards. The NRC regulations will be the basis for review of DOE's License Application.
The License Application from DOE will include assessments of potential radionuclide releases
from the repository and assessment of compliance with regulatory standards under specified
exposure conditions. Because of the long time frames and uncertainties involved hi predicting
repository performance, DOE will be required to demonstrate "reasonable expectation" of
compliance with the standards. The term "reasonable expectation" conveys the concept that
ES-47
-------
absolute numerical proof of compliance with the standards is neither necessary nor likely to be
obtainable.
One of the key factors in evaluating compliance with EPA's Yucca Mountain standards is the
radiation dose potential associated with each of the exposure pathways used by the receptor. The
dose potential is characterized in terms of dose conversion factors, which relate radionuclide
concentrations in the pathways for exposure, such as water and food consumed, to dose received.
The dose consequence of radionuclides in the environment therefore depends on the relative
importance of the various pathways for the exposed individual, which depends, in turn, on the
lifestyle of the exposed individual. It is to be expected, for example, that the pathways and dose
factors for a farmer residing in an arid environment, such as Yucca Mountain, will differ from
those for an urban resident.
Dose conversion factors for assessing compliance with regulatory standards have been evaluated
by DOE, EPA, and CNWRA for a wide variety of environmental conditions and receptor
lifestyles. The DOE is currently acquiring data to enable characterization of dose factors
specifically for environmental conditions and human activities in the Yucca Mountain region.
The DOE plans to use site-specific dose conversion factors in the License Application.
ES-48
-------
CHAPTER 1
INTRODUCTION
The U.S. Environmental Protection Agency (EPA) is responsible for developing and issuing
environmental standards and criteria to ensure that public health and the environment are
adequately protected from potential radiation impacts. The EPA is proposing in 40 CFR Part
197 site-specific environmental standards to protect public health from releases from radioactive
materials disposed of or stored in the potential repository to be constructed at Yucca Mountain in
Nevada.1 These standards provide the basic framework to control the long-term storage and
disposal of three types of radioactive waste:
1. Spent nuclear fuel, if disposed of without reprocessing
2. High-level radioactive waste from the reprocessing of spent nuclear fuel
3. Other radioactive materials that may be placed hi the potential repository
The other radioactive materials that could be disposed of hi the Yucca Mountain repository
include highly radioactive low-level waste, known as greater-than-Class-C waste, and excess
plutonium resulting from the dismantlement of nuclear weapons. However, the plans for
placement of these materials are very uncertain and therefore, for the purpose of the present
rulemaking, the information presented in this Background Information Document (BID) is
limited to spent nuclear fuel and high-level radioactive waste. More details about the current and
projected inventories of these wastes can be found in Chapter 5 of the BID.
1 -1 PURPOSE AND SCOPE OF THE BACKGROUND INFORMATION DOCUMENT
This document presents the technical information used by EPA to understand the characteristics
of the Yucca Mountain site and to develop its proposed rule, 40 CFR Part 197. The scope of the
BID encompasses the conceptual framework for assessing radiation exposures and associated
health risks. In general terms, this assessment discusses the radioactive source term
characterization, movement of radionuclides from the repository at Yucca Mountain through the
1 No decision has been made regarding the acceptability of Yucca Mountain for storage or disposal. In
this document, the characterization of the Yucca Mountain repository as "potential" is often omitted but always
intended.
1-1
-------
appropriate environmental exposure pathways, and calculations performed to date of doses
received by members of the general public.
The significant alternative models for site and engineered barrier performance are presented in
the BID to the extent necessary to portray the current understanding of the site and the major
uncertainties in that understanding. Most of the technical information discussed in the BID is
derived from investigations sponsored by the Department of Energy (DOE). DOE has conducted
years of research; most of what is known about Yucca Mountain and the performance of an
underground radioactive waste repository is the result of this research. However, where
appropriate, information from other sources is presented to supplement the DOE data base, to fill
data gaps, and to illustrate alternative conceptualizations of geologic processes and engineered
barrier performance.
The BID is not intended to be a technical critique of the investigations conducted by DOE and
other parties. Nor is it a regulatory compliance or criteria document. The BID is a summary of
the technical information considered by EPA in developing the rationale for and specifics in 40
CFR Part 197.
In addition, the BID discusses only those issues related to the disposal of radioactive wastes in a
geologic repository. Although additional disposal strategies have been examined by the U.S. and
other countries, a geologic repository continues to be the most promising. Technologies to
separate and transmute long-lived radionuclides hi the waste to a stable form were examined
recently by the National Research Council. The Council concluded that such technologies do not
obviate the need for a geologic repository. The use of other disposal environments, such as the
seabed or natural or artificial islands, is fraught with political issues and therefore considered
infeasible. A final alternative of placing the waste into earth's orbit and accelerating it toward
the sun may be theoretically possible, but would require decades of technological development
and is likely to be much more costly than placing the waste in a geologic repository (NOR97).
Chapter 1 of the BID discusses EPA's regulatory authority for the current rulemaking and
summarizes the recommendations of the National Academy of Sciences report to Congress
entitled Technical Bases for Yucca Mountain Standards (NAS95). A summary of key events in
the history of EPA's rulemaking is also included. Chapter 2 provides a brief history of the
evolution of radiation protection activities in the United States as well as current U.S. regulatory
programs and strategies. A summary of key international programs for high-level waste disposal
1-2
-------
is presented in Chapter 3. Chapter 4 describes U.S. programs for the management and disposal
of high-level radioactive waste and spent nuclear fuel. Current and projected inventories of spent
nuclear fuel and DOE defense high-level radioactive waste are presented in Chapter 5. Chapter 6
describes the methodology used by EPA for dose and risk estimation. Chapter 7 provides
descriptions of the natural features of the Yucca Mountain site, the concepts under consideration
for the engineered features of a potential repository at the site, and analyses to date concerning
safety performance of a disposal system at the site. Chapter 8 describes the environment in the
Yucca Mountain region, current conditions of human radiation exposure in the region, and
concepts that could be used to evaluate the consequences of radioactivity release from a
repository at Yucca Mountain. Chapter 9 discusses Yucca Mountain exposure scenarios and
compliance assessment issues, and finally, Chapter 10 provides a literature review of radiological
risks from alternatives to geologic disposal of high-level radioactive waste.
1.2 EPA'S REGULATORY AUTHORITY FOR THE RULEMAKING
The proposed standards governing environmental releases from the Yucca Mountain repository
have been developed pursuant to the Agency's authorities under the Energy Policy Act (EnP A) of
1992 (Public Law 102-486). Section 801 of this Act directed EPA to promulgate standards to
ensure protection of public health from releases from radioactive material in a deep geologic
repository to be built at Yucca Mountain (EnPA92). EPA must set standards to ensure protection
of the health of individual members of the public. The EnPA also required EPA to contract with
the National Academy of Sciences (NAS) to advise the Agency on the technical bases for the
Yucca Mountain standards. These standards will apply only to the Yucca Mountain site and are
to be developed based upon and consistent with the findings and recommendations of the NAS:
... the Administrator shall, based upon and consistent with the findings and
recommendations of the National Academy of Sciences, promulgate, by rule,
public health and safety standards for protection of the public from releases from
radioactive materials stored or disposed of in the repository at the Yucca
Mountain site. Such standards shall prescribe the maximum annual effective dose
equivalent to individual members of the public from releases to the accessible
environment from radioactive materials stored or disposed of in the repository
(EnPA92).
1-3
-------
1.3 THE NATIONAL ACADEMY OF SCIENCES RECOMMENDATIONS
In the EnPA, the Congress asked the Academy to address three issues in particular:
• Whether a health-based standard based upon doses to individual members
of the public from releases to the accessible environment mil provide a
reasonable standard for protection of the health and safety of the general
public;
• Whether it is reasonable to assume that a system for post-closure
oversight of the repository can be developed, based upon active
institutional controls, that -will prevent an unreasonable risk of breaching
the repository's engineered or geologic barriers or increasing exposure of
individual members of the public to radiation beyond allowable limits;
and
• Whether it will be possible to make scientifically supportable predictions
of the probability that the repository's engineered or geologic barriers will
be breached as a result of human intrusion over a period ofl 0,000 years
(EnPA92).
To address these questions, the Academy assembled a committee of 15 members representing a
range of scientific expertise and perspectives. The committee conducted a series of five technical
meetings; more than 50 nationally and internationally known scientists and engineers were
invited to participate. In addition, the committee received information from the Nuclear
Regulatory Commission (NRC), the Department of Energy (DOE), EPA, Nevada State and
county agencies, and private organizations, such as the Electric Power Research Institute.
The committee's conclusions and recommendations are contained in its final report, entitled
Technical Bases for Yucca Mountain Standards, which was issued on August 1,1995 (NAS95).
In this report, the committee offered the Agency several general recommendations as to the
approach EPA should take in developing 40 CFR Part 197. Specifically, the NAS recommended
(NAS95,p.2):
The use of a standard that sets a limit on the risk to individuals of adverse
health effects from releases from the repository. 40 CFR Part 1912
* In 1985, EPA promulgated 40 CFR Part 191, "Environmental Standards for the Management and
Disposal of Spent Nuclear Fuel, High-Level and Transuranic Radioactive Wastes" (EPA85a). These are generally
1-4
-------
contains an individual-dose standard, and it continues to rely on a
containment requirement that limits the releases ofradionudides to the
accessible environment. The stated goal of the containment requirement
was to limit the number of health effects to the global population to 1,000
incremental fatalities over 10,000 years. We do not recommend that a
release limit be adopted.
• That compliance with the standard be measured at the time of peak risk,
whenever it occurs. (Within the limits imposed by the long-term stability
of the geologic environment, which is on the order of one million years.)
The standard in 40 CFR Part 191 applies for a period of 10,000 years.
Based on performance assessment calculations provided to us, it appears
that peak risks might occur tens or hundreds of thousands of years or even
farther into the future.
• Against a risk-based calculation of the adverse effect of human intrusion
into the repository. Under 40 CFR Part 191, an assessment must be made
of the frequency and consequences of human intrusion for purposes of
demonstrating compliance with containment requirements. In contrast,
we conclude that it is not possible to assess the frequency of intrusion far
into the future. We do recommend that the consequences of an intrusion
be calculated to assess the resilience of the repository to intrusion.
The NAS committee also recommended that policy issues be resolved through a rulemaking
process that allows opportunity for wide-ranging input from all interested parties (NAS95).
The committee also addressed each of the specific questions posed to it by the Congress in the
EnPA. With regard to the first issue, protecting human health, the NAS committee
recommended (NAS95, pp. 4-7):
• ... the use of a standard that sets a limit on the risk to individuals of
adverse health effects from releases from the repository.
• ...the critical-group approach be used in the Yucca Mountain standards.
applicable environmental standards promulgated under EPA's authority under the Atomic Energy Act of 1954 as
amended. As a result of court action, these standards were remanded back to EPA and were subsequently
repromulgated in 1993. (See Sections 1.3.1 and 1.3.4 for more detail.)
1-5
-------
• ... compliance assessment be conducted for the time when the greatest risk
occurs, \vithin the limits imposed by long-term stability of the geologic
environment.
The NAS also concluded that an individual-risk standard would protect public health, given the
particular characteristics of the site, provided that policy makers and the public are prepared to
accept that very low radiation doses pose a negligibly small risk. A necessarily important
component in the development of a standard for Yucca Mountain is the means of assessing
compliance. The NAS committee concluded the following (NAS95, p. 9):
.. .physical and geologic processes are sufficiently quantifiable and the
related uncertainties sufficiently boundable that the performance can be
assessed over time frames during which the geologic system is relatively
stable or varies in a boundable manner. The geologic record suggests
that this time frame is on the order of JO6 years. The Committee farther
concluded that the probabilities and consequences of modifications by
climate change, seismic activity, and volcanic eruptions at Yucca
Mountain are sufficiently boundable that these factors can be included in
performance assessments that extend over this time frame.
...it is not possible to predict on the basis of scientific analyses the societal
factors required for an exposure scenario. Specifying exposure scenarios
therefore requires a policy decision that is appropriately made in a
rulemakingprocess conducted by EPA.
With respect to the second and third questions posed by the Congress in Section 801 of the
EnPA, the NAS Committee concluded (NAS95, p. 11):
• ...it is not reasonable to assume that a system for post-closure oversight of
the repository can be developed, based on active institutional controls,
that will prevent an unreasonable risk of breaching the repository's
engineered barriers or increasing the exposure to individual members of
the public to radiation beyond allowable limits.
• ...it is not possible to make scientifically supportable predictions of the
probability that a repository's engineered or geologic barriers will be
breached as a result of human intrusion over a period of J 0,000 years.
1-6
-------
1.4 HISTORY OF EPA'S RULEMAKING
Many significant events have occurred in the past 50 years concerning the management of high-
level radioactive waste and spent nuclear fuel. Table 1-1 provides a timeline of these events.
The following sections describe them in detail.
Table 1 -1. Significant Events in the History of High-Level Radioactive Waste and Spent
Nuclear Fuel Disposal
Yew i
1944
1949
1955
1957
1962
1965-1967
1968
1970
1970
1971
1974
1974
1976
1976
1976
1976
1978
' "• s ,%•••• , Jj-VSlflil ••-. "" t *^e ? ">
Construction of first storage tanks for high-level radioactive waste (HLW).
The Atomic Energy Commission (AEC) initiates work to convert high-level liquid waste into a
stable form.
The National Academy of Sciences (NAS) Advisory Committee is established to consider disposal
of HLW in U.S.
The NAS suggests geologic disposal be investigated, particularly in naturally occurring salt
formations.
The AEC determines waste management to be technically feasible.
Project Salt Vault demonstrates the safety and feasibility of handling and storing waste in salt
formations.
The AEC requests NAS to establish a Committee on Radioactive Waste Management (CWRM).
The CWRM concludes that the use of bedded salt is satisfactory for the disposal of radioactive
waste.
The AEC announces tentative selection of a site at Lyons, Kansas, for the establishment of a
national radioactive waste repository.
The AEC pursues alternative sites for repository.
The AEC publishes its first analysis of methods for long-term management of HLW.
Congress passes the Energy Reorganization Act which abolishes AEC and creates a developmental
agency, the Energy Research and Development Agency (ERDA-now DOE) and an independent
regulatory commission, the Nuclear Regulatory Commission (NRC), which has authority to
regulate DOE facilities used for receipt and storage of HLW.
The Office of Management and Budget (OMB) establishes an interagency task force on
commercial HLW.
The Federal Energy Regulatory Commission (FERC) publishes a status report on the management
of commercial radioactive waste.
President Ford issues a major policy statement on radioactive waste which includes a charge to the
EPA to issue general environmental standards governing releases of radioactive material to the
biosphere.
The EPA announces its intent to develop environmental radiation protection criteria for radioactive
waste.
The EPA proposes criteria for management and disposal of radioactive wastes.
1-7
-------
Table 1-1 (Continued)
Year
1978
1979
1980
1981
1982
1982
1985
1987
1987
1992
1992
1993
1996
1998
Swwr
President Carter establishes the Interagency Review Committee.
The DOE publishes a draft GETS and decides to concentrate on mined geologic repositories as a
means for waste disposal.
President Carter outlines a national radioactive waste management program. The President decides
to investigate four to five sites in a variety of environments before a license application is
submitted to NRC.
The EPA withdraws its proposed "Criteria for Radioactive Wastes."
Congress enacts the Nuclear Waste Policy Act which requires characterization of three sites and
construction of a geologic repository available to receive spent nuclear fuel and HLW by 1998.
The EPA proposes 40 CFR Part 191, "Environmental Standards for the Management and Disposal
of Spent Nuclear Fuel, High-Level and Transuranic Radioactive Wastes."
The EPA issues a final rule under 40 CFR Part 191.
Congress passes the Nuclear Waste Policy Amendments Act which identifies Yucca Mountain as
the single site for characterization.
The EPA's 40 CFR Part 191 is remanded by the Court.
Congress enacts the Waste Isolation Pilot Plant Land Withdrawal Act which reinstated sections of
40 CFR Part 191 and exempted Yucca Mountain from the generic disposal standards set forth in
Subpart B of 40 CFR Part 191.
Congress enacts the Energy Policy Act and directs EPA to develop regulations for Yucca
Mountain.
The EPA issues amendments to 40 CFR Part 191 .
The DOE acknowledges it cannot proceed directly to License Application, but only to a
determination of site viability, by 1998.
The DOE publishes a "viability assessment" concluding that Yucca Mountain is a promising site
"or a geologic repository and that work should proceed toward a site recommendation in 200 1 .
1-4.1 Legislative History
EPA has the authority to set generally applicable environmental standards for radioactive
releases under the Atomic Energy Act (AEA) of 1954, as amended, and the EPA Reorganization
Plan No. 3 of 1970 (NIX70). The basic authority under the AEA, as transferred to the EPA by
Reorganization Plan No 3, includes the mandate of:
... establishing generally applicable environmental standards for the
protection of the general environment from radioactive materials. As used
herein, standards mean limits on radiation exposures or levels, or
concentrations or quantities of radioactive material, in the general
1-8
-------
environment outside the boundaries of locations under the control of
persons possessing or using radioactive materials (AEA54).
In 1982, the Nuclear Waste Policy Act (NWPA) (Public Law 97-425) established formal
procedures regarding the evaluation and selection of sites for geologic repositories, including
procedures for the interaction of State and Federal Governments. The Act established provisions
for the selection of at least two independent repository sites. Further, the NWPA limited the
quantity of spent nuclear fuel to be disposed of in the initial repository to 70,000 metric tons of
heavy metal (MTHM)3, or a quantity of solidified high-level radioactive waste resulting from the
reprocessing of such a quantity of spent nuclear fuel, until a second repository is in operation
(NWP83). The NWPA also reiterated the existing responsibilities of the Federal agencies
involved in the national program and provided a timetable for several key milestones to be met
by the Federal agencies. As part of this national program, the EPA, pursuant to its authorities
under other provisions of law, was required to:
• by rule, promulgate generally applicable standards for the protection of
the general environment from off-site releases from radioactive material
in repositories (NWP83).
In September 1985, EPA published 40 CFR Part 191, "Environmental Standards for the
Management and Disposal of Spent Nuclear Fuel, High-Level and Transuranic Radioactive
Wastes" (EPA85a). These standards were to apply to all sites for the deep geologic disposal of
high-level radioactive waste. In 1987, the U.S. Court of Appeals for the First Circuit responded
to a legal challenge by remanding Subpart B of the 1985 standards to the Agency for further
consideration.
In December 1987, Congress enacted the Nuclear Waste Policy Amendments Act (NWPAA).
The 1987 Amendments Act redirected the nation's nuclear waste program to evaluate the
suitability of the Yucca Mountain site as the location for the first high-level waste and spent
nuclear fuel repository (NWP87). Activities at all other potential sites were to be phased out. If
the Yucca Mountain site is found to be suitable, the President is required to submit a
recommendation to Congress to develop a repository at this location. In the event that site
characterization activities indicate that Yucca Mountain is an unsuitable site for the repository,
3 This is a measure of the uranium content of the spent nuclear fuel to be emplaced in the repository.
1-9
-------
the Secretary of Energy is required to inform Congress and the State of Nevada of its findings.
The NWPAA prohibits DOE from conducting site-specific activities for a second repository
unless authorized to do so by Congress. However, the NWPAA does require a report from the
Secretary of Energy on the need for a second repository no later than January 1,2010.
Finally, the Act established a commission to study the need and feasibility of a monitored
retrievable storage facility to complement the nation's nuclear waste management program. The
commission submitted to Congress (required under the original Act, as amended by Public Law
100-507) a report outlining its recommendations on November 1, 1989 (NWP88, RMR89).
In October 1992, the Waste Isolation Pilot Plant Land Withdrawal Act (WIPP LWA) was
enacted. While reinstating certain sections of the Agency's 1985 disposal standards, the Act
exempted the Yucca Mountain site from these generic disposal standards (WIP92). However, the
EnPA directed the EPA to set site-specific radiation protection standards for the Yucca Mountain
disposal system (EnPA92).
As part of the Fiscal Year 1997 appropriation action, the Congress required EPA to perform a
comparative assessment of risks associated with management of commercial spent nuclear fuel
for three circumstances: permanent storage at the site where it is now stored; one or more
centralized storage sites; and deep geologic disposal at Yucca Mountain. This requirement was
established in Senate Report 104-320 at page 98 and was retained by conference committee
action on the FY 1997 Energy and Water Appropriation Bill which stated that "The language and
allocations set forth in House Report 104-679 and Senate Report 104-320 should be complied
with unless specifically addressed to the contrary in the conference report and statement of the
managers" (Congressional Record, House, September 12, 1996, page HI 0244).
The requirement was stated in Senate Report 104-320 as follows:
• Any radiation protection standard proposed by the Environmental
Protection Agency for the Yucca Mountain repository should consider
specific alternatives to deep geologic disposal at Yucca Mountain and
should include an analysis of the comparative risk to the public from each
alternative. The alternatives considered should include the permanent
storage of nuclear waste at the site where it is now stored and one or more
centralized storage sites recommended by the administration for the
above-ground, managed storage. The Agency shall evaluate each of these
alternatives against the standards proposed for deep geologic disposal at
Yucca Mountain.
1-10
-------
1.4.2 The Development of EPA's Role in the Federal Program
Since the inception of the nuclear age in the 1940s, the Federal government has assumed ultimate
responsibility for the disposal of spent nuclear fuel and high-level radioactive waste, regardless
of whether it is produced by commercial or national defense activities. In 1949, the Atomic
Energy Commission (AEC) initiated work aimed at developing systems for converting high-level
liquid waste into a stable form. Then, in 1955, at the request of the AEC, an NAS Advisory
Committee was established to consider the disposal of high-level radioactive waste within the
United States. Its report, issued in 1957, recommended that:
1. The AEC continue to develop processes for the solidification of high-level
radioactive liquid waste
2. Naturally occurring salt formations be used as the medium for the long-
term isolation of the solidified waste (NAS57)
Project Salt Vault, conducted from 1965 to 1967 by the AEC in an abandoned salt mine near
Lyons, Kansas, demonstrated the safety and feasibility of handling and storing waste in salt
formations (McC70).
In 1968, the AEC again requested the NAS to establish a Committee on Radioactive Waste
Management (CRWM) to advise the AEC on its long-range radioactive waste management plans
and to evaluate the feasibility of disposing of solidified radioactive waste in bedded salt. The
CRWM convened a panel to discuss the disposal of radioactive waste in salt mines. Based on
the recommendations of the panel, the CRWM concluded that the use of bedded salt was
satisfactory for the disposal of radioactive waste (NAS70).
In 1970, the AEC announced the tentative selection of a site at Lyons, Kansas, for the
establishment of a national radioactive waste repository (AEC70). During the next two years,
however, in-depth site studies raised several questions concerning the safe plugging of old
exploratory wells and proposed expanded salt mining activities. These questions and growing
public opposition to the Lyons site prompted the AEC in late 1971 to pursue alternative sites
(DOU72).
1-11
-------
In 1976, the Federal government intensified its program to develop and demonstrate a permanent
disposal method for high-level radioactive waste. The Office of Management and Budget
(OMB) established an interagency task force on commercial wastes in March 1976. The task
force defined the scope of the responsibility of each Federal agency's activities on high-level
management, including the preparation of environmental standards for high-level waste by the
EPA (LYN76, ENG77a, ENG77b).
Shortly after the interagency task force was formed, the Federal Energy Regulatory Commission
(FERC) published a status report on the management of commercial radioactive waste. The
report, issued in May 1976, emphasized the need for coordination of administration policies and
programs relating to energy and called for an accelerated comprehensive government radioactive
waste program plan. The report also recommended that an interagency task force be formed to
coordinate activities among the responsible Federal agencies.
Subsequent to its findings, FERC established a nuclear subcommittee to coordinate Federal
nuclear policy and programs. The EPA was given the responsibility of establishing general
environmental standards governing waste disposal activities, including standards for high-level
radioactive waste to be delivered to Federal repositories for long-term management (FER76).
In October 1976, after the OMB interagency task force proposed its plan for spent nuclear fuel
and high-level waste management, President Ford issued a major policy statement on radioactive
waste. As part of his comprehensive statement, he announced new steps to assure that the United
States had the facilities for the long-term management of nuclear waste from commercial power
plants. He also reported that experts had concluded that the most practical method for disposing
of spent nuclear fuel and high-level radioactive waste would be in geologic repositories located
in stable formations deep underground. The EPA was charged with the responsibility of issuing
general environmental standards governing releases of radioactive material to the biosphere
above natural background radiation levels (FOR76). These standards were to place a numerical
limit on long-term radiation releases outside the boundary of the repository.
1.4.3 Early Federal Action
In December 1976, the EPA announced its intent to develop environmental radiation protection
criteria for radioactive waste to assure the protection of public health and the general
environment (EPA76). These efforts resulted in a series of radioactive waste disposal
workshops, held in 1977 and 1978 (EPA77a, EPA77b, EPA78a, EPA78b). Based on issues
1-12
-------
raised during workshop deliberations, EPA published a Federal Register Notice on November 15,
1978 (43 FR 53262) (EPA78c) of intent to propose criteria for radioactive wastes and to solicit
public comments on possible recommendations for Federal Radiation Guidance. In this notice,
EPA presented a set of criteria to address six key waste control decision issues: 1) the types of
materials to be categorized as radioactive wastes and subject to control; 2) the efficacy of
engineered controls and natural barriers to isolate wastes; 3) the usefulness of social institutions
in providing control, especially their viability over time; 4) the potential health risks of wastes
(over various time intervals and with differing levels of control); 5) the unacceptability of various
levels of risk; and 6) other considerations such as retrievability and communication of waste
disposal sites to succeeding generations to ensure continued isolation. As proposed, EPA
intended that the initial set of six criteria—each addressing one of the six key issues—would
serve collectively as the basis for developing environmental standards for different radioactive
waste sources.
During this time, President Carter established the Interagency Review Group (IRG) to develop
recommendations for an administrative policy to address the long-term management of nuclear
waste and supporting programs to implement the policy. The IRG report re-emphasized EPA's
role in developing generally applicable standards for the disposal of high-level waste, spent
nuclear fuel, and transuranic waste (DOE79). In a message to Congress in February 1980, the
President outlined the content of a comprehensive national radioactive waste management
program based on the IRG recommendations. The message called for an interim strategy for
disposal of spent nuclear fuel and high-level and transuranic wastes that would rely on mined
geologic repositories. The message reiterated that the EPA was responsible for creating general
criteria and numerical standards applicable to radioactive waste management activities (CAR80).
In March 1981, the EPA withdrew the proposed "Criteria for Radioactive Wastes" because it
considered the implementation of generic disposal guidance too complex given the many
different types of radioactive waste (EPA81).
In 1982, Congress enacted the NWPA, which established the current national program for the
disposal of spent nuclear fuel and high-level waste. The Act assigned DOE the responsibility of
siting, building, and operating an underground geologic repository for the disposal of these
wastes and directed the EPA to "promulgate generally applicable standards for the protection of
the general environment from off-site releases from radioactive material in repositories"
(NWP83). In that same year, under the authority of the AEA, the EPA proposed a set of
1-13
-------
standards under 40 CFR Part 191, "Environmental Standards for the Management and Disposal
of Spent Nuclear Fuel, High-Level and Transuranic Radioactive Wastes" (EPA82).
After the first comment period on the proposed rule ended in May 1983, the EPA held two public
hearings on the proposed standards—one in Washington, DC, and one in Denver, CO. During a
second public comment period, EPA requested post-hearing comments (EPA83a, EPA83b).
More than 200 comment letters were received during these two comment periods, and 13 oral
statements were made at the public hearings. Responses to comments received from the public
were subsequently published and released in August 1985 (EPA85b).
In parallel with its public review and comment effort, the EPA conducted an independent
scientific review of the technical bases for the proposed 40 CFR Part 191 standards through a
special subcommittee of the Agency's Science Advisory Board (SAB). The subcommittee held
nine public meetings from January to September 1983 and released a final report in February
1984 (SAB84). Although the SAB review found that the Agency's analyses in support of the
proposed standards were comprehensive and scientifically competent, the report contained
several findings and recommendations for improvement. The report was publicly released in
May 1984, and the public was encouraged to comment on the findings and recommendations
(EPA84). Responses to the SAB report were subsequently presented and released in August
1985 (EPA85c).
In February 1985, the Natural Resources Defense Council, the Environmental Defense Fund, the
Environmental Policy Institute, the Sierra Club, and the Snake River Alliance brought suit
against the Agency and the Administrator because they had failed to comply with the
January 1984 deadline mandated by the NWPA for promulgation of final standards. A consent
order was negotiated with the plaintiffs that required the standards to be promulgated on or
before August 15, 1985. The EPA issued the final rule under 40 CFR Part 191 on
August 15, 1985 (EPA85d, EPA85e).
1.4.4 40 CFR Part 191
The 1985 EPA standards for the management and disposal of spent nuclear fuel and high-level
and transuranic waste were divided into two main sections, Subparts A and B (EPA85a).
Subpart A, which addressed the management and storage of waste, limited radiation exposure to
1-14
-------
any member of the general public to 25 millirem (mrem) to the whole body and 75 mrem to any
critical organ for disposal facilities operated by the Department of Energy, but not regulated by
the NRC or an Agreement State. For facilities regulated by the NRC or an Agreement State, the
standards endorsed the annual dose limits given in the environmental standards for the uranium
fuel cycle (40 CFR Part 190): 25 mrem to the whole body, 75 mrem to the thyroid, and 25 mrem
to any critical organ (EPA77c).
Subpart B imposed limits associated with the release of radioactive materials into the
environment following closure of the repository. The key provisions of Subpart B were:
• Limits on cumulative releases of radioactive materials into the environment
during the 10,000 years following disposal
• Assurance requirements to compensate for uncertainties in achieving the desired
level of protection
• Individual exposure limits based on the consumption of ground water and
any other potential exposure pathways for 1,000 years after disposal
• Ground water protection requirements in terms of allowable radionuclide
concentrations and associated doses for 1,000 years after disposal
(EPA85a)
Under sections 191.15 and 191.16 of Subpart B, the annual dose to any member of the general
public was limited to 25 mrem to the whole body and 75 mrem to any critical organ. The ground
water concentration for beta or gamma emitters was limited to the equivalent yearly whole body
or organ dose of 4 mrem. The allowable water concentration for alpha emitters (including
radium-226 and radium-228, but excluding radon) was 15 picocuries/liter (pCi/L). For radium-
226 and radium-228 alone, the concentration limit was 5 pCi/L. Appendix A of the standards
provided acceptable radionuclide-specific cumulative release limits.
In March 1986, five environmental groups led by the Natural Resources Defense Council and
four States filed petitions for a review of 40 CFR Part 191 (USC87). These suits were
consolidated and argued in the U.S. Court of Appeals for the First Circuit in Boston. The main
challenges concerned:
1-15
-------
1. Violation of the Safe Drinking Water Act (SDWA) underground injection section
2. Inadequate notice and comment opportunity on the ground water protection
requirements
3. Arbitrary standards, not supported in the record, or not adequately explained
In July 1987, the Court rendered its opinion and noted three findings against the Agency and twc
favorable judgments. The Court's action resulted in the remand of Subpart B. The Court began
by looking at the definition of "underground injection." In the view of the Court, the method
envisioned by DOE for disposal of radioactive waste in underground repositories would "likely
constitute an underground injection under the SDWA."
Under the SDWA, the Agency is required to assure that underground sources of drinking water
will not be endangered by any underground injection. With regard to such potential
endangerment, the Court supported part, but not all, of the Agency's approach. Inside the
controlled area, the Court ruled that Congress—through the EPA—had allowed endangerment of
ground water. However, the Court accepted EPA's approach of using the geological formation as
part of the containment.
Outside the controlled area, the Court found that Section 191.15 would allow endangerment of
drinking water supplies. In the context of the SDWA, "endangerment" was considered when
doses higher than those allowed by the Primary Drinking Water Regulations could occur.
Section 191.15 permitted an annual dose of 25 mrem to the whole body and 75 mrem to any
critical organ from all pathways. Existing EPA regulations promulgated under the SDWA
allowed an annual dose of 4 mrem from drinking water. Although the Court recognized that an
exposure level less than 4 mrem could result from the ground water pathway, it rejected this
possibility because the Agency stated that radioactivity could eventually be released into the
ground water system near the repository and that substantially higher doses could result.
Therefore, the Court decided that a large fraction of the 25 mrem limit could be received througb
the ground water exposure pathway. Accordingly, the Court found that the Part 191 standards
should either have been consistent with the SDWA or the Agency should have justified the
adoption of a different standard.
1-16
-------
The Court stated that the Agency was not necessarily incorrect in promulgating the proposed
standards. However, it noted that the Agency never acknowledged the interrelationship of the
SDWA and the Part 191 standards nor did it present a reasonable explanation for the divergence
between them. The Court also supported the petitioner's argument that the Agency had not
properly explained the selection of the 1,000-year limit for individual protection requirements
(Section 191.15). The Court indicated that the 1,000-year criterion was not inherently flawed,
but rather that the administrative record and the Agency's explanations did not adequately
support this choice. The criterion was remanded for reconsideration, and the Agency was
directed to provide a more thorough explanation for its basis.
Finally, the Court found that the Agency did not provide sufficient opportunity for notice and
comment on Section 191.16 (Ground Water Protection Requirements), which was added to
Subpart B after the standards were proposed. This section was remanded for a second round of
notice and comment. There were, however, no rulings issued on technical grounds about Section
191.16.
In August 1987, the Department of Justice petitioned the First Circuit Court to reinstate all of 40
CFR Part 191 except for Sections 191.15 and 191.16, which were originally found defective.
The Natural Resources Defense Council filed an opposing opinion. The Court then issued an
Amended Decree that reinstated Subpart A, but continued the remand of Subpart B.
In 1992, the WIPP LWA reinstated Subpart B of 40 CFR Part 191, except Sections 191.15 and
191.16, and required the Administrator to issue final disposal standards no later than six months
after enactment. On December 20,1993, EPA issued amendments to 40 CFR Part 191 which
eliminated section 191.16 of the original rule; altered the individual protection requirements; and
added Subpart C on ground water protection (EPA93). The amended standards represent the
Agency's response to the above legislation and to the issues raised by the court pertaining to
individual and ground water protection requirements. In so doing, EPA did not revisit any of the
regulations reinstated by the WIPP LWA.
The WIPP LWA also exempted Yucca Mountain from the generic disposal standards set forth
under 40 CFR Part 191, Subpart B. Pursuant to specific provisions in the EnPA, EPA was
charged with setting site-specific environmental radiation standards for Yucca Mountain. The
proposed rule, 40 CFR Part 197, is responsive to this mandate.
1-17
-------
REFERENCES
AEA54 Atomic Energy Act, Public Law 83-703, as amended, 42 USC 2011 et seq., 1954.
AEC70 Atomic Energy Commission Press Release No. N-102, dated June 17,1970.
CAR80 The White House, President J. Carter, The President's Program on Radioactive
Waste Management, Fact Sheet, February 12, 1980.
DOE79 U.S. Department of Energy, Report to the President by the Interagency Review
Group on Nuclear Waste Management, Report No. TID-29442, March 1979.
DOU72 Doub, W.O., U.S. Atomic Energy Commission Commissioner, Statement before
the Science, Research and Development Subcommittee for the Committee on
Science and Astronautics, U.S. House of Representatives, U.S. Congress,
Washington, D.C., May 11 and 30,1972.
ENG77a English, T.D. et al., An Analysis of the Back End of the Nuclear Fuel Cycle with
Emphasis on High-Level Waste Management, JPL Publication 77-59, Volumes I
and II, Jet Propulsion Laboratory, Pasadena, California, August 12,1977.
ENG77b English, T.D. et al., An Analysis of the Technical Status of High-level
Radioactive Waste and Spent Fuel Management Systems, JPL Publication 77-69,
Jet Propulsion Laboratory, Pasadena, California, December 1, 1977.
EnPA92 Energy Policy Act of 1992, Public Law 102-486, October 24, 1992.
EPA76 U. S. Environmental Protection Agency, Environmental Protection Standards for
High-Level Wastes - Advance Notice of Proposed Rule making, Federal Register,
41 FR 53363, December 6, 1976.
EPA77a U. S. Environmental Protection Agency, Proceedings: A Workshop on Issues
Pertinent to the Development of Environmental Protection Criteria for
Radioactive Wastes, Reston, Virginia, February 3-5, 1977, Office of Radiation
Programs, Report ORP/SCD-77-1, Washington, D.C., 1977.
EPA77b U. S. Environmental Protection Agency, Proceedings: A Workshop on Policies
and Technical Issues Pertinent to the Development of Environmental Protection
Criteria for Radioactive Wastes, Albuquerque, New Mexico, April 12-17, 1977,
Office of Radiation Programs, Report ORP/SCD-77-2, Washington, D.C., 1977.
EPA77c U.S. Environmental Protection Agency, Environmental Radiation Protection
Standards for Nuclear Power Operations, 40 CFR Part 190, Federal Register, 42
FR 2858-2861, January I3t 1977.
1-18
-------
EPA78c
EPA81
EPA82
EPA78a U.S. Environmental Protection Agency, Background Report - Consideration of
Environmental Protection Criteria for Radioactive Wastes, Office of Radiation
Programs, Washington, D.C., February 1978.
EPA78b U. S. Environmental Protection Agency, Proceedings of a Public Forum on
Environmental Protection Criteria for Radioactive Wastes, Denver, Colorado,
March 30-April 1, 1978, Office of Radiation Programs, Report ORP/SCD-78-2,
Washington, D.C., May 1978.
U. S. Environmental Protection Agency, Recommendations for Federal Guidance,
Criteria for Radioactive Wastes, Federal Register, 43 FR 53262-53268
November 15,1978.
U. S. Environmental Protection Agency, Withdrawal of Proposed Regulations,
Federal Register, 46 FR 17567, March 19, 1981.
U. S. Environmental Protection Agency, Proposed Rule, Environmental
Standards for the Management and Disposal of Spent Nuclear Fuel, High-level
and Transuranic Radioactive Wastes, 40 CFR Part 191, Federal Register 47 FR
58196-58206, December 29,1982.
U. S. Environmental Protection Agency, Environmental Standards for the
Management and Disposal of Spent Nuclear Fuel, High-level and Transuranic
Radioactive Wastes, Notice of Public Hearings, Federal Register, 48 FR 13444-
13446, March 31,1983.
EPA83b U. S. Environmental Protection Agency, Environmental Standards for the
Management and Disposal of Spent Nuclear Fuel, High-level and Transuranic
Radioactive Wastes, Requests for Post-Hearings Comments, Federal Register 48
FR 23666, May 26,1983.
U. S. Environmental Protection Agency, Environmental Standards for the
Management and Disposal of Spent Nuclear Fuel, High-level and Transuranic
Radioactive Wastes, Notice of Availability, Federal Register, 49 FR 19604-19606,
May8, 1984.
EPA85a U. S. Environmental Protection Agency, Final Rule, Environmental Standards for
the Management and Disposal of Spent Nuclear Fuel, High-level and
Transuranic Radioactive Wastes, Federal Register, 50 FR 38066-38089,
September 19,1985.
EPA83a
EPA84
1-19
-------
EPA85b
EPA85c
EPA85d
EPA85e
EPA93
FER76
FOR76
LYN76
McC70
NAS57
NAS70
U. S. Environmental Protection Agency, High-Level and Transuranic Radioactive
Wastes - Response to Comments for Final Rule, Volume I, Office of Radiation
Programs, EPA 520/1-85-024-1, Washington, D.C., August 1985.
U. S. Environmental Protection Agency, High-level and Transuranic Radioactive
Wastes - Response to Comments for Final Rule, Volume II, Office of Radiation
Programs, EPA 520/1-85-024-2, Washington, D.C., August 1985.
U. S. Environmental Protection Agency, High-Level and Transuranic Radioactive
Wastes - Background Information Document for Final Rule, Office of Radiation
Programs, EPA 520/1-85-023, Washington, D.C., August 1985.
U. S. Environmental Protection Agency, Final Regulatory Impact Analysis - 40
CFR Part 191: Environmental Standards for the Management and Disposal of
Spent Nuclear Fuel, High-Level and Transuranic Radioactive Wastes, Office of
Radiation Programs, EPA 520/1-85-027, Washington, D.C., August 1985.
U. S. Environmental Protection Agency, 40 CFR Part 191, Environmental
Radiation Protection Standards for the Management and Disposal of Spent
Nuclear Fuel, High-Level and Transuranic Radioactive Wastes; Final Rule,
Federal Register, 58 FR 66398-66416, December 20,1993.
Federal Energy Resources Council, Management of Commercial Radioactive
Nuclear Wastes - A Status Report, May 10, 1976.
The White House, President G. Ford, The President's Nuclear Waste Management
Plan, Fact Sheet, October 28,1976.
Memorandum from J.T. Lynn, OMB to R. Train, EPA; R. Peterson, CEQ; R.
Seamans, ERDA, and W. Anders, NRC; Concerning the Establishment of an
Inter agency Task Force on Commercial Nuclear Wastes, March 25, 1976.
McClain, W.C., and R.L. Bradshaw, Status of Investigations of Salt Formations
for Disposal of Highly Radioactive Power-Reactor Wastes, Nuclear Safety,
11(2):130-141, March-April 1970.
National Academy of Sciences - National Research Council, Disposal of
Radioactive Wastes on Land, Publication 519, Washington, D.C., 1957.
National Academy of Sciences - National Research Council, Committee on
Radioactive Waste Management, Disposal of Solid Radioactive Wastes in Bedded
Salt Deposits, Washington, D.C., November 1970.
1-20
-------
NAS95 National Academy of Sciences - National Research Council, Committee on
Technical Bases for Yucca Mountain Standards, Technical Bases for Yucca
Mountain Standards, National Academy Press, Washington, D.C., 1995.
NIX70 The White House, President R. Nixon, Reorganization Plan No. 3 of 1970,
Federal Register, 35 FR 15623-15626, October 6,1970.
NWP83 Nuclear Waste Policy Act of 1982, Public Law 97-425, January 7,1983.
NWP87 Nuclear Waste Policy Amendments Act of 1987, Public Laws 100-202 and 100-
203, December 22,1987.
NWP88 Nuclear Waste Policy Amendments Act of 1988, Public Law 100-507, October 18,
1988.
RMR89 Nuclear Waste: Is There A Need For Federal Interim Storage ?, Monitored
Retrievable Storage Review Commission, November 1,1989.
S AB84 Science Advisory Board, Report on the Review of Proposed Environmental
Standards for the Management and Disposal of Spent Nuclear Fuel High-level
and Transuranic Radioactive Wastes (40 CFR Part 191), High-Level Radioactive
Waste Disposal Subcommittee, U.S. EPA, Washington, D.C., January 1984.
USC87 United States Court of Appeals for the First Circuit, Natural Resources Defense
Council, Inc., etal, v. United States Environmental Protection Agency,
Docket No.: '85-1915, 86-1097, 86-1098, Amended Decree, September 23,1987.
WIP92 Waste Isolation Pilot Plant Land Withdrawal Act, Public Law 102-579,
October 20,1992.
1-21
-------
CHAPTER 2
HISTORY OF RADIATION PROTECTION IN THE UNITED STATES
AND CURRENT REGULATIONS
2.1 INTRODUCTION
Radiation from cosmic rays and naturally occurring radioactivity contained in the earth make up
the natural radiation background environment in which all life forms have evolved. Society's
recognition of radiation began in 1895 with the discovery of X-rays; naturally occurring
radioactivity was observed in 1896. These discoveries marked the beginning of the study and use
of radioactive substances in science, medicine, and industry.
The discovery of radioactivity led rapidly to the development of medical radiology, industrial
radiography, nuclear physics, and nuclear medicine. By the 1920s, the use of X-rays in
diagnostic medicine and Industrial applications was widespread. Radium was being routinely
used in luminescent dials and by doctors in therapeutic procedures. By the 1930s, biomedical
and genetic research scientists were studying the effects of radiation on living organisms, and
physicists were beginning to understand the mechanisms of spontaneous fission and radioactive
decay. In the 1940s, research hi nuclear physics had advanced to the point where a self-
sustaining fission reaction was demonstrated under laboratory conditions. These events led to
the construction of the first nuclear reactors and the development of atomic weapons.
Today, the use of radiation, be it naturally occurring or man-made, is widespread and reaches
every segment of our society. Common examples include:
Nuclear reactors used: 1) to generate electricity, 2) to power ships and
submarines, 3) to produce radioisotopes used for research, medical,
industrial, space and national defense applications, and 4) as research tools
for nuclear engineering and physics
• Particle accelerators used to produce radioisotopes and radiation and to
study the structure of matter, atoms, and common materials
Radioisotopes used in nuclear medicine, biomedical research, and medical
treatment
2-1
-------
• X-rays and gamma rays used as diagnostic tools in medicine, as well as in
diverse industrial applications, such as industrial radiography, luggage X-
ray inspections, and nondestructive materials testing
• Common consumer products, such as smoke detectors, luminous-dial wrist
watches, luminous markers and signs, cardiac pacemakers, lightning rods,
static eliminators, welding rods, lantern mantles, and optical glass
It was soon recognized that the use of radioactive materials would have to be controlled to
protect the public, workers, and the environment from radiation exposures. The following
sections present a brief history of the evolution of radiation protection activities, their principles
and concepts, and U.S. regulatory programs and strategies. Included in this discussion is the
influence that certain international advisory bodies, such as the International Commission on
Radiological Protection (ICRP), have had on the development of U.S. radiation protection
policies. Chapter 3 presents a summary of spent nuclear fuel and high-level waste disposal
programs in other countries.
2.2 THE INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, THE
NATIONAL COUNCIL ON RADIATION PROTECTION AND MEASUREMENTS,
AND THE INTERNATIONAL ATOMIC ENERGY AGENCY
Initially, the dangers and risks posed by X-rays and radioactivity were poorly understood. By
1896, however, "X-ray burns" were being reported in the medical literature, and by 1910, it was
understood that such "burns" could be caused by radioactive materials. By the 1920s, sufficient
direct evidence (from radium dial painters, medical radiologists, and miners) and indirect
evidence (from biomedical and genetic experiments with animals) had been accumulated to
persuade the scientific community that an official body should be established to make
recommendations concerning human protection against exposure to X-rays and radium.
In 1928, at the Second International Congress of Radiology meeting in Stockholm, Sweden, the
first radiation protection commission was created. Reflecting the uses of radiation and
radioactive materials at the time, the body was named the International X-Ray and Radium
Protection Commission. It was charged with developing recommendations concerning radiation
protection. In 1950, to better reflect its role in a changing world, the Commission was
reorganized and renamed the International Commission on Radiological Protection.
During the Second International Congress of Radiology, the newly created Commission
suggested to the nations represented at the Congress that they appoint national advisory
2-2
-------
committees to represent their viewpoints before, the Commission and to act in concert with the
Commission in developing and disseminating recommendations on radiation protection. This
suggestion led to the formation of the U.S. Advisory Committee on X-Ray and Radium
Protection in 1929. In 1964, the Committee was congressionally chartered as the National
Council on Radiation Protection and Measurements (NCRP).
Throughout their existence, the ICRP and the NCRP have worked closely together to develop
radiation protection recommendations that reflect the current understanding of the risks
associated with exposure to ionizing radiation (ICR34, ICR38, ICR51, ICR60, ICR65). Neither
organization has official status, in that they do not have authority to issue or enforce regulations.
However, their recommendations often serve as the basis for the radiation protection regulations
adopted by the regulatory authorities in the United States and most other nations.
The International Atomic Energy Agency (IAEA) was chartered hi July 1957 as an autonomous
intergovernmental organization under the aegis of the United Nations. The IAEA gives advice
and technical assistance to Member States on nuclear power development, health and safety
issues, radioactive waste management, and on a broad range of other areas related to the use of
radioactive material and atomic energy in industry and government. As is the case for ICRP and
NCRP, Member States do not have to follow IAEA recommendations. However, funding for
international programs dealing with the safe use of atomic energy and radioactive materials can
be withheld if Member States do not comply with IAEA recommendations. In addition, in
matters related to safeguarding special nuclear material, the full weight of the UN can be brought
to bear to "enforce" UN resolutions pertaining to the use of nuclear materials for peaceful
purposes. Many of the IAEA recommendations adopt ICRP recommendations with respect to the
Commission's radiation protection philosophy and numerical criteria.
In 1977, the ICRP released recommendations that are in use today. ICRP Publication No. 26
(ICR77) adopted the weighted, whole-body dose equivalent (defined as the effective dose
equivalent) concept for limiting occupational exposures. This approach reflected the increased
understanding of the differing radiosensitivities of various organs and tissues and was intended to
sum exposures from external sources and from internally deposited nuclides. (Note: The
concept of summing internal and external exposures to arrive at total dose had been mentioned as
early as ICRP Publication No. 1 PCR60].)
ICRP No. 26 defined the goal of radiation protection as the prevention or limitation of effects
from radiation exposure and the assurance that practices involving radiation exposure are
justified. The concept of collective dose equivalent for populations was also discussed. The
2-3
-------
ICRP No. 26 recommendations represented the first explicit attempt to relate and justify
permissible radiation exposures with quantitative levels of acceptable risk. The ICRP concluded
that "...the mortality risk factor for radiation-induced cancers is about 10"4 per rem, as an average
for both sexes and all ages...." The risks of average occupational exposures (about 0.5 rem/year)
are roughly comparable to risks experienced in safe industries, 10"4 annually. At the permissible
limit of 5 rem/year, the risk is comparable with that experienced by some workers in occupations
having higher-than-average risk.
For members of the public, the ICRP considered that an annual risk in the range of 10"6 to 10~5
would likely be acceptable (ICR77). The ICRP recommended an annual individual dose limit of
100 mrem (1 mSv) from all radiation sources. However, the Commission also recognized that an
annual individual dose limit of 500 mrem (5 mSv) may be permissible, provided that the average
annual effective dose equivalent over a lifetime does not exceed the principal limit of 100 mrem
(1 mSv) (ICR85a). No dose limits for populations were proposed.
In 1979, the ICRP issued Publication No. 30 (ICR79) establishing the Annual Limit on Intake
(ALI) system for limiting the intake of radionuclides by workers. The ALI is the activity of a
given nuclide that would irradiate a person to the limit set in ICRP No. 26 for each year of
occupational exposure. It is a secondary limit, based on the primary limit of equivalent whole-
body irradiation, and applies to intake by either ingestion or inhalation. The recommendations of
ICRP No. 30 applied only to occupational exposures. In 1983, the ICRP issued a statement
(ICR84) clarifying the use of ALIs and Derived Air Concentrations (DACs) for members of the
public.
In 1985, the ICRP issued a statement (ICR85a) refining dose limits for members of the public.
ICRP No. 26 had endorsed an annual limit of 500 mrem, subject to certain conditions. In making
this endorsement, it was assumed that the conditions would, in practice, restrict the average
annual dose to about 100 mrem. In its 1985 statement, the Commission stated that the principal
limit was 100 mrem, while occasional and short-term exposures up to 500 mrem were thought to
be acceptable.
The Commission has also published guidance for waste disposal (ICR85b) and for general
radiological protection (ICR91). The first of these, "Radiation Protection for the Disposal of
Solid Radioactive Waste," emphasizes an individual-risk approach that considers both the
probability of a breach of a disposal site and its consequence upon the critical group.
2-4
-------
In 1987, the NCRP issued Report No. 91 (NCR87), which acknowledged the assumptions and
the basic thrust of the recommendations in ICRP Reports 26 and 30. In discussing risk estimates,
the NCRP noted in its report that new data were becoming available that might require changes
in the current estimates. However, the value recommended in ICRP No. 26 of 10"4 per rem was
retained for a nominal lifetime somatic risk for adults.
The NCRP also noted that continuous annual exposure to 100 mrem gives a person a mortality
risk of about 10"5 annually, or approximately 10'3 in a lifetime (NCR87). Similar to the 1985
ICRP statement, annual limits of 500 mrem were recommended for infrequent exposures and 100
mrem for continuous (or frequent) exposures. These limits do not include natural background or
medical exposures.
In 1989, the IAEA issued reports 96 and 99 in its Safety Series (IAE89a, IAE89b). These
documents presented criteria and guidance for the underground disposal of nuclear waste. Safety
Series No. 99, "Safety Principles and Technical Criteria for the Underground Disposal of High-
Level Radioactive Wastes," set out basic design objectives to ensure that "humans and the human
environment will be protected after closure of the repository and for the long periods of time for
which the wastes remain hazardous." The report went on to state that for releases from a
repository due to gradual processes, the dose upper bound should be less than an annual average
dose value of 1 mSv (i.e., 100 mrem/yr)4 for prolonged exposures for individuals in the critical
group (defined as the members of the public whose exposure is relatively homogeneous and is
typical of individuals receiving the highest effective dose equivalent or dose equivalent from a
given radiation source). Finally, it suggested a risk upper bound of 10"5 per year for an individual
for disruptive events.
The ICRP has adopted the international system of units (SI). Under this system, 1 Sv equals 100 rem. As
i, 1 mSv equals 100 mrem.
such, 1 mSv equals 100 mrem
2-5
-------
In 1990, the ICRP issued Publication 60, which broadened its recommendations to include a
wider range of exposure scenarios than had been previously addressed. Publication 60 also gave
support to new concepts in the field of radiation exposure protection, most notably the ALARA
(as low as reasonably achievable) concept of worker protection optimization. The ALARA
principle suggests dose limits should be set at the lowest levels reasonably possible for a given
scenario. In recent years, several international organizations, including the Council of the
European Communities (CEC) and the Organization for Economic Cooperation and
Development/Nuclear Energy Agency's (OECD/NEA's) Committee on Radiation Protection and
Public Health (CRPPH), have worked to interpret this principle and develop guidelines for its
practical use (NEA94). The formality with which the ALARA principle has been adopted varies
widely internationally. In many cases, the ALARA principle is being applied only as part of a
nonqualified conceptual framework within which protection measures are implemented; in
other countries, the application of the ALARA approach to worker safety is becoming
increasingly formalized (OEC95a).
In recent years, the IAEA has been developing new international safety standards and guidance
documents. Foremost among these is "International Basic Safety Standards for Protection
Against Ionizing Radiation and for the Safety of Radiation Sources," known as BSS (Basic
Safety Standards, Safety Series 115-1). The BSS was approved by the IAEA Board of Governors
in 1994 and published as an interim document in December 1995. A joint effort of the Food and
Agricultural Organization of the United Nations, the International Labor Organization, the
OECD/NEA, the Pan-American Health Organization, and the World Health Organization, the
BSS is notable primarily for its movement toward an integrated approach to managing exposure
risk in which potential but unlikely events (such as accidents) are evaluated along with
comparatively normal, likely scenarios for exposure. Previously, safety assessment had focused
only on comparatively normal, likely scenarios (OEC95a). IAEA has also been developing a
comprehensive set of safety standards for radioactive waste management called Radioactive
Waste Safety Standards (RADWASS). RADWASS includes a safety fundamentals document
entitled "The Principles of Radioactive Waste Management" and a safety standard document
entitled "Establishing a National Safety Standard for Radioactive Waste Management." Both of
these were approved by the IAEA Board of Governors and published in October 1995. Three
other safety standards (S-2, S-3, and S-6) addressing predisposal management of radioactive
waste, near-surface disposal of radioactive waste, and decommissioning are under review
(OEC95b). The entire RADWASS series is currently under review to ensure harmonization with
Safety Series Publications and BSS documents.
2-6
-------
Criteria development is also continuing through an IAEA Working Group on Principles and
Criteria for Radioactive Waste Disposal. The working group's focus includes post-closure
monitoring, optimization, retrievability, dose vs. risk, and safety indicators under different time
frames. The group's first report, entitled Safety Indicators in Different Time Frames for the
Safety Assessment of Underground Radioactive Waste Repositories, was published in 1994
(SNI95).
In recent years, the CEC has been developing directives on radiation safety standards for
radiation exposures established under European Atomic Energy Community (EURATOM)
agreements. In accordance with ICRP recommendations, the CEC suggested in 1993 that doses
to members of the public be limited to 100 mrem per year from all sources except medical and
that occupational doses be limited to 2,000 mrem annually. The CEC is also expected to propose
criteria for the shipment of radioactive waste among member countries and for the export of
radioactive waste to nonmember countries (OEC93).
Finally, in 1989, Radiation Protection and Nuclear Safety authorities in Denmark, Finland,
Iceland, Norway, and Sweden developed a set of safety criteria for the disposal of high-level
radioactive waste. Revised in 1993 after international review, the Nordic Principles are largely
consistent with other criteria developed on the international level. The Principles outline a
radiation protection approach employing the concept of optimization and an individual dose limit
of 0.1 millisievert (10 mrem) per year. Basic guiding objectives for HLW disposal programs
include reduction of burden for future generations, long-term environmental protection, and the
use of specific safety assurance measures. Finally, the Principles contain technical
recommendations for repository design, site geology, and closure (SNI95).
2-3 FEDERAL RADIATION COUNCIL GUIDANCE
The Federal Radiation Council (FRC) was established in 1959 by Executive Order 10831. The
Council arose as a result of new information that became available in the 1950s on the effects of
radiation. Before that time, only nongovernmental radiation advisory bodies (i.e., ICRP and
NCRP) existed, and their recommendations were not binding on users of radiation or radioactive
Materials. The FRC was established as an official Government entity and included
representatives from all Federal agencies concerned with radiation protection. The Council
served as the primary coordinating body for all radiation activities conducted by the Federal
Government (FRC60a) and was responsible for:
2-7
-------
...advising the President with respect to radiation matters, directly or indirectly
affecting health, including providing guidance to all Federal agencies in the
formulation of radiation standards and in the establishment and execution of
programs of cooperation with States....
The Council's first recommendations concerning radiation protection guidance for Federal
agencies were approved by President Eisenhower in 1960 (FRC60b). The guidance established
exposure limits for members of the general public. These included the yearly radiation exposure
of 0.5 rem per year for the whole body of individuals in the general population and an average
gonadal dose of 5 rem in 30 years for the general population (exclusive of natural background
and the purposeful medical exposure of patients).
The guidance also established occupational exposure limits, which differed only slightly from
those recommended by the NCRP and ICRP at the time (NCR54, NCR59). The guidance
included:
• Whole body, head and trunk, active blood-forming organs, gonads or lens
of the eyes are not to exceed 3 rem in 13 consecutive weeks, and the total
accumulated dose is limited to 5 rems multiplied by the number of years
beyond age 18, expressed as 5(N-18), where N is the current age
Skin of the whole body and thyroid are not to exceed 10 rem in
13 consecutive weeks or 30 rem per year
• Hands, forearms, feet, and ankles are not to exceed 25 rem in
13 consecutive weeks or 75 rem per year
• Bone is not to exceed 0.1 microgram of radium-226 or its biological
equivalent
• Any other organs are not to exceed 5 rem in 13 consecutive weeks or
15 rem per year
In addition to the formal exposure limits, the guidance also established as Federal policy that any
radiation exposure should be justified and that "...every effort should be made to encourage the
maintenance of radiation doses as far below this guide as practicable...." Both of these concepts
had previously been proposed by the ICRP. The inclusion of the requirements to consider
benefits and keep all exposures to a minimum was based on the possibility that there is no
2-8
-------
threshold for radiation. The linear, nonthreshold, dose-response relationship was assumed to
place an upper limit on the estimate of radiation risk. However, the FRC explicitly recognized
that it might also represent the actual level of risk.
Following the issuance of this initial guidance, the FRC continued to provide guidance on a
number of radiation protection matters. In 1970, the Council was dissolved, and its functions
were transferred to the Environmental Protection Agency under authority of Reorganization Plan
NO. 3 (NIX70).
2-4 ENVIRONMENTAL PROTECTION AGENCY
Since its creation in 1970, the EPA has issued regulatory standards regarding radiation hazards
from a number of different sources, including underground mining (EPA71), the uranium fuel
cycle operations (EPA77), uranium and thorium mill tailings (EPA83), radionuclide air
emissions (EPA89a), and management and disposal of spent nuclear fuel and high-level and
transuranic radioactive wastes (EPA93). Recently, EPA issued compliance criteria for the WIPP
(EPA96). EPA is currently developing a standard for the disposal of contaminated soil at
decommissioned sites, including Federal facilities.
gency has also exercised its authority to issue Federal guidance to limit radiation exposures
to workers (EPA87), as well as to the general public. In December 1994, EPA issued proposed
Federal guidance to update the previous Federal Radiation Protection Guidance for Exposure to
toe General Public which was originally adopted in 1960 and 1961 (EPA94). The Agency is now
nnalizing these new recommendations.
EpA has also provided extensive technical information regarding the assessment of risk from
radiation hazards. Specific examples of such information include radionuclide intake limits,
Occupational radiation doses, biological parameters, and dose conversion factors (EPA88). This
^formation has been used extensively in the development of EPA standards and guidance, as
Well as specific site assessments.
h addition to its responsibility to provide Federal guidance on radiation protection, the EPA has
various statutory authorities and responsibilities for regulating exposure to radiation. The
standards and regulations that EPA has promulgated and proposed with respect to controlling
2-9
-------
radiation exposures are summarized in the following paragraphs. Their applicability to EPA's
proposed standards under 40 CFR Part 197 is also discussed.
2.4.1 Environmental Radiation Exposure
The Atomic Energy Act (AEA) of 1954, as amended, and Reorganization Plan No. 3 granted the
EPA the authority to establish generally applicable environmental standards for exposure to
radiation (AEA54, NIX70). The AEA is the cornerstone of current radiation protection activities
and regulations. In 1977, pursuant to this authority, the EPA issued standards limiting exposures
from operations associated with the light-water reactor fuel cycle (EPA77). These standards,
under 40 CFR Part 190, cover normal operations of the uranium fuel cycle. The standards limit
the annual dose equivalent to any member of the public from all phases of the uranium fuel cycle
(excluding radon and its daughters) to 25 mrem to the whole body, 75 mrem to the thyroid, and
25 mrem to any other organ. To protect against the buildup of long-lived radionuclides in the
environment, the standards also set normalized emission limits for krypton-85, iodine-129, and
plutonium-239 combined with other transuranics with a half-life exceeding one year. The dose
limits imposed by the standards cover all exposures resulting from radiation and radionuclide
releases to air and water from operations of fuel-cycle facilities. The development of these
standards took into account both the maximum risk to an individual and the overall effect of
releases from fuel-cycle operations on the population, and balanced these risks against the costs
of effluent control.
2.4.2 Environmental Impact Assessments
In 1969, Congress passed the National Environmental Policy Act (NEPA), which declared a
national policy that encouraged a productive and enjoyable harmony between the public and the
environment (NEP70). The Act recognized the profound impact of human activity on the
interrelations of all components of the natural environment and sought to promote efforts to
prevent or eliminate damage to the environment. To this end, the national policy is geared
towards increasing the understanding of the ecological systems and natural resources important
to the United States. In addition, the Act established a Council on Environmental Quality to
assist the President in determining the state of the environment and developing environmental
policy initiatives.
2-10
-------
The Act also directed all Federal agencies to use a systematic, interdisciplinary approach to
ensure the integrated use of natural, social, and environmental sciences in support of plans and
decisions that have a potential impact on the environment. Specifically, it mandated that a
detailed Environmental Impact Statement (EIS) be submitted for any major action proposed by a
Federal agency or for legislation that would significantly affect the quality of the environment.
The EIS must describe any adverse environmental effects that the proposal would cause,
alternatives to the proposed action, effects of the project on the long-term productivity of the
environment, and any irreversible and irretrievable commitment of resources involved in the
proposed action. The EIS must also be prepared through consultation with any Federal agency
having jurisdiction or special expertise regarding the project and its environmental impact.
The Final EIS prepared by the Department of Energy for the Yucca Mountain site must comply
with NEPA requirements.
4-3 Ground-Water Protection
The Safe Drinking Water Act (SDWA) was enacted to assure safe drinking water supplies and to
Protect against endangerment of underground sources of drinking waters (USDWs). Under the
authority of the SDWA, the EPA issued interim regulations (40 CFR Part 141, Subpart B)
covering the permissible levels of radium, gross alpha, man-made beta, and photon-emitting
contaminants in community water supply systems (EPA76). Similar to hazardous chemical
substances, limits for radionuclides in drinking water are expressed as Maximum Contaminant
Levels (MCLs). The current MCL for radium-226 and radium-228 combined is 5 picoCuries per
Ut* (5 pCi/L), and the MCL for gross alpha particle activity (including radium-226, but
excluding radon and uranium) is 15 pCi/L. For man-made beta particle- and photon-emitting
radionuclides (except tritium and strontium-90), individually or in combination, the MCL is set
at an annual dose limit of 4 millirem to the total body or any internal organ. For tritium and
strontium-90, the MCLs are 20,000 pCi/L and 8 pCi/L, respectively.
In 1991, the EPA issued a Notice of Proposed Rulemaking (NPRM) under 40 CFR Parts 141 and
142 to update the 1976 interim regulations for radionuclide water pollution control (EPA91).
^ NPRM, under the SDWA, proposed the establishment of Maximum Contaminant Level
°oals (MCLGs) and Maximum Contaminant Levels (MCLs). The MCLGs and MCLs target
*diurn-226, radium-228 natural uranium, radon, gross alpha, gross beta, and photon emitters.
As Proposed, MCLGs are not enforceable health goals. In contrast, MCLs are enforceable
2-11
-------
standards. The EPA concluded that radionuclide MCLGs should be set at zero to avert known or
anticipated adverse health effects while providing an adequate margin of safety. In setting the
MCLGs, the EPA also committed itself to evaluating the feasibility, costs, and availability of
water treatment technologies, as well as other practical considerations. The proposed regulations
state the following MCLs: radium-226,20 pCi/L; radium-228,20 pCi/L; radon-222, 300 pCi/L;
uranium, 20 micro g/L; adjusted gross alpha, 15 pCi/L; and beta and photon emitters, 4 mrem
ede/yr.
Over the past 20 years, the EPA has used three different methods to calculate radioactivity
concentrations for beta particle and photon emitting radionuclides in drinking water
corresponding to the MCL of 4 mrem/yr. Each method incorporates successive improvements to
the risk models and dose conversion factors for ingested radioactivity recommended by national
and international advisory committees on radiation protection and adopted by the Agency.
The first method is a requirement (§ 141.6(b)) of EPA's 1976 Interim Regulations. It specifies
that, with the exception of tritium and strontium-90, the concentration of beta/photon emitters
causing 4 millirem (mrem) total body or organ dose equivalent shall be calculated on the basis of
a 2 liter per day drinking water intake using the 168 hour data listed in Handbook 69 of the
National Bureau of Standards (NBS63). The dose models used in preparing Handbook 69 are
based on earlier recommendations of the International Commission on Radiological Protection
(ICR60). For tritium and strontium-90 s the EPA provides derived activity concentrations in
Table A of § 141.6(b) based on specific dose models for these nuclides.
The second method is presented in EPA's 1991 proposed rule on final drinking water standards
for radionuclides (EPA91), This method is based primarily on the updated dosimetric data in
ICRP Publication 30 (ICR79) and uses the Agency's own risk assessment methodology
formalized in the RADRISK computer code (DUN80). Under this approach, concentration
levels are calculated for each radionuclide individually by limiting the dose to the total body
(i.e., the effective dose equivalent or ede) to 4 mrem/yr ede, rather than on a dose rate of
4 mrem/yr to the critical organ. Similar to the first method, the second method assumes
continuous intake of activity over a lifetime at a rate of 2 liters of drinking water per day.
The third and final method is currently being developed by the EPA as a substitute for the first
and second methods. Under this proposed method, concentrations corresponding to 4 mrem/yr
ede will be calculated using the dose conversion factors for ingestion in Federal Guidance
2-12
-------
No. 11 (EPA88), assuming a 2 liter per day intake rate, 365 days per year. Similar to the second
method, the proposed method is also based primarily on the updated dosimetric data in ICRP
Publication 30 (ICR79), but will not use EPA's RADRISK code. Instead, it will use a simple
algebraic equation and the Federal Guidance Report No. 11 dose factors.
2-4.4 Radionuclide Air Emissions
In December 1979, the EPA designated radionuclides as hazardous air pollutants under Section
112 of the Clean Air Act (CAA) Amendments of 1977 (Public Law 95-95) (EPA79). In April
1983, the EPA proposed standards regulating radionuclide emissions from four source categories,
one of which included U.S. Department of Energy (DOE) facilities. The rule established annual
airborne emission limits for radioactive materials and specified that annual doses resulting from
such emissions should not exceed 25 mrem to the whole body and 75 mrem to any critical organ
for members of the general public. The EPA also proposed nojt to regulate several other
categories of facilities, including high-level radioactive waste disposal facilities. EPA based its
decision with respect to high-level waste disposal facilities on estimated releases from conceptual
repositories that indicated that the airborne exposure pathway would not cause doses high enough
to warrant regulation.
In October 1984, following a court order, the EPA withdrew the proposed emission standards
based on the findings that the control practices already in effect protected the public from
radionuclide releases with an ample margin of safety. The Agency also affirmed its position not
to regulate other categories of emission sources, including uranium fuel facilities and high-level
radioactive waste.
fa December 1984, a U.S. District Court found the EPA in contempt of its order and directed the
EPA either to issue final radionuclide emission standards or make a finding that radionuclides
are not hazardous air pollutants. The EPA complied with the court order in 1985 by issuing
standards for selected sources (EPASSa, EPA85b). As a result of the decision in National
Resources Defense Council Inc. v. EPA, November 1987, the Agency submitted a motion to the
court requesting a voluntary remand of its national emission standards for the four original
categories of emission sources proposed in April 1983. In December 1987, the Court granted the
EPA's motion for voluntary remand and established a schedule to propose new regulatory
standards within one year. The Court decision also defined the analytical process under which
the EPA was to re-evaluate its standards. Two steps were identified: 1) determine what is safe,
2-13
-------
based exclusively on health risk, and 2) adjust the level of safety downward to provide an ample
margin of safety.
In March 1989, the EPA issued a proposed rule for regulating radionuclide emissions under the
CAA following the re-examination of the regulatory issues associated with the use of Section
112 (EPA89a). The rule proposed four policy alternatives to control emissions and
risks from 12 categories of sources. Each of the four approaches considered the acceptable risk
criterion differently. The four approaches were:
• Case-by-Case Approach — Acceptable risk considers all health
information, risk measures, potential biases, assumptions, and quality of
the information. The maximum individual lifetime fatal cancer risk must
not exceed 1 x 10"4.
• Incidence-Based Approach — Based on the best estimate of the total
incidence of fatal cancer. The proposed acceptable level of incidence must
not exceed one fatal cancer per year per source category.
Maximum Individual Risk Approach (10"4 or less) — Only risk indicator
considered is the best estimate of the maximum individual lifetime risk of
fatal cancer. The maximum individual lifetime risk must not exceed 1 x
10-4.
• Maximum Individual Risk Approach (10~6 or less) — This approach is
similar to the previous one. The maximum individual lifetime risk,
however, must not exceed 1 x 10"6.
Consistent with the two-step process established by the Court, the Agency determined an ample
margin of safety after ascertaining a safe level based solely on health risks. In reaching its final
decision, the EPA considered all health risk measures, as well as technological feasibility, costs,
uncertainties, economic impacts of control technologies, and any other relevant information.
In its radionuclide emission standards, EPA considered a lifetime risk to an individual of
approximately 1 in 10,000 as acceptable. The presumptive level provides a benchmark for
judging the acceptability of maximum individual risk, but does not constitute a rigid line for
making that determination.
2-14
-------
In its final rule, EPA concluded that there was no need to establish air emission standards for
high-level waste disposal repositories since anticipated operations at the site would be governed
by 40 CFR Part 191. Radioactive materials received at such facilities are sealed in containers.
Normal operations do not require additional processing or handling because spent nuclear fuel or
high-level waste is received and emplaced into the ground in its original containers. Operations
at the disposal site, which may require additional waste processing or repackaging before the site
!s declared a disposal facility, are covered by 40 CFR Part 191 and must comply with Subpart I
°f the National Emission Standards for radionuclides5 (EPA89b). Consequently, the Agency
believed there is an ample margin of safety since the likelihood of releases, and attendant risks, is
very low.
Disposal of High-Level Radioactive Waste and Spent Nuclear Fuel
Congress passed the Nuclear Waste Policy Act (NWPA) of 1982 to provide for the development
°f repositories for the disposal of high-level radioactive waste and spent nuclear fuel, and to
establish a program of research, development, and demonstration regarding this disposal
(NWP83). The Act established a schedule for the siting, construction, and operation of
repositories that would provide a reasonable assurance that the public and environment would be
adequately protected from the hazards posed by high-level radioactive waste. The Secretary of
Energy was charged with nominating candidate sites for a repository and following a number of
steps through a process of Presidential and Congressional approval, site characterizations, public
Participation, and hearings. The Act also required the Secretary to adhere to NEPA in
considering alternatives and to prepare an EIS for each candidate site.
Irutially the Act called for the development of two mined geologic repositories. The first
repository was to be selected from nine candidate sites in western states; the second repository
was to be located in the eastern United States in crystalline rock. EPA was charged with the
responsibility of promulgating generally applicable standards for the protection of public health
^d the environment from off-site releases from radioactive material in repositories. The NRC,
111 turn, was responsible for promulgating technical requirements and criteria consistent with
standards to serve as the basis for approving or disapproving applications regarding the
Subpart I of the National Emission Standard can be found in 40 CFR Part 61.101 and is entitled "National
Standard for Radionuclide Emissions from Facilities Licensed by the Nuclear Regulatory Commission
and Federal Facilities Not Covered by Subpart H." Subpart H of the National Emission Standard
addresses radionuclide standards for DOE facilities.
2-15
-------
use, closure, and post-closure of the repository. The Act also discussed interim waste storage
requirements, as well as the payment of benefits to affected States and tribal groups to allow
them sufficient resources to participate fully in the process.
In 1987, the NWPA was amended to reflect a redirection of the nuclear waste program. The
generic nature of the original act was changed to reflect the selection of the Yucca Mountain site
in Nevada as the only candidate site for the repository (NWP87). The State of Nevada was also
identified as the affected community. All site-specific activities at other candidate sites were
phased out, and the Final EIS, necessary for compliance with the NEPA, was to be prepared
specifically for the Yucca Mountain site without further consideration of alternative sites. The
redirection charged DOE with reporting to Congress on the potential social, economic, and
environmental impacts of locating the repository at Yucca Mountain.
2.4.5.1 Generic Disposal Standards for High-Level and Transuranic Wastes
As discussed in Chapter 1, the First Circuit Court of Appeals remanded Subpart B of EPA's
standards for the management and disposal of spent nuclear fuel and high-level and transuranic
waste (40 CFR Part 191) in 1987. (See Section 1.3.4 for additional detail regarding the Court's
action on 40 CFR Part 191.) The Waste Isolation Pilot Plant Land Withdrawal Act (WIPP LWA)
of 1992 reinstated all of the disposal standards remanded by the First Circuit Court of Appeals in
1987 except the three aspects of the individual and ground-water protection requirements that
were the subject of the court remand (WIP92). It then put the Agency on a schedule for issuing
the final disposal standards. They were published in December 1993. The law also provided an
extensive role for EPA in reviewing and approving various phases of DOE activities at the WIPP
and required EPA to certify whether the WIPP repository would meet the final 40 CFR Part 191
standards. Finally, and of greatest importance to the current rulemaking, the WIPP LWA
exempted radioactive waste disposal activities at Yucca Mountain from compliance with the
generic standards set forth under the 40 CFR Part 191 standards.
2.4.5.2 Site-Specific Disposal Standards for High-Level Radioactive Waste
The Energy Policy Act (EnPA) of 1992 addressed energy efficiency throughout the United States
in different situations and for various types of fuel. Title VIII of the Act dealt specifically with
high-level radioactive waste. Section 801 of the EnPA assigned EPA the responsibility of
promulgating public health and safety standards for protection of the public from releases from
radioactive materials stored or disposed of hi the repository at the Yucca Mountain site. EPA is
2-16
-------
to prescribe a maximum annual effective dose equivalent to individual members of the public
from releases to the accessible environment from radioactive materials stored or disposed of in
the repository (EnPA92). The Act also requires that the standards developed be based upon and
consistent with the findings and recommendations of the NAS. Specifically, the NAS was
charged with considering: the use of a dose-based standard, the reasonableness of post-closure
oversight in preventing breaches, and the predictability of human intrusion over a period of
10,000 years. NAS's findings and recommendations were published on August 1, 1995, in its
report Technical Bases for Yucca Mountain Standards (NAS95). These standards will apply
only to Yucca Mountain.
^•4-6 Evaluation of Radiation Dose
The radiation dose incurred by an exposed individual is evaluated using the "committed effective
dose equivalent" (CEDE) concept. The CEDE is the weighted sum of the "committed dose
equivalent" to specified organs and tissues. The committed effective dose equivalent is the total
effective dose equivalent, averaged over a given tissue or organ, that is deposited in the 50-year
Period following the intake of a radionuclide.
The CEDE approach to dose evaluation therefore takes into account the differing dose effects of
various radionuclides hi specific parts of the body over time, and the differing dose effects of
Eternal exposure to ionizing radiations of different types and energy levels. It accounts, for
example, for the fact that some radionuclides that are taken into the body will be rapidly excreted
^fter ingestion or inhalation, so that the dose effect is small. Other radionuclides may be retained
indefinitely in specific organs so that if the decay rate is low and exposure continues over time,
the body burden of the dose source, and therefore the dose committed to the organ, will
continually increase with time. In general, the dose incurred will depend on the types and
concentrations of radionuclides present, the conditions and duration of exposure, the bilogical
half-life of the radionuclide in the body, and the effects of exposure on organs and tissues of the
Ability to apply the CEDE approach to dose evaluation is the result of a decades-long
evolutionary process which has developed a data base for, and an understanding of, the
Physiological effects of radiation exposure. A brief history of the evolution of information and
Methodology for radiation dose evaluation, and a description of the CEDE methodology, are set
forth in EPA's Federal Guidance Report No. 1 1 (EPA88). This document also contains tables of
2-17
-------
values for the committed dose equivalents per unit uptake for various radionuclides taken into the
body and for various body organs and tissues.
In 1993, EPA issued a companion report, Federal Guidance Report No. 12 (EPA93a), which
tabulates dose coefficients for external exposure to photons and electrons emitted by
radionuclides distributed in air, water, and soil. The dose coefficient values provided in this
document are, like those in Federal Guidance Report No. 11, intended to be used by government
agencies to calculate the dose equivalent to organs and tissues of the body for given exposure
conditions.
2.5 NUCLEAR REGULATORY COMMISSION
The NRC was created as an independent agency by the Energy Reorganization Act (ERA) of
1974 (ERA74), which abolished the AEC and moved the AEC's regulatory function to the NRC.
This Act, coupled with the AEA, as amended, provided the foundation for regulation of the
nation's commercial nuclear power industry, NRC regulations are issued under the U.S. Code of
Federal Regulations Title 10 Chapter 1.
The mission of the NRC is to ensure adequate protection of public health and safety, the national
defense and security, and the environment in the use of nuclear materials in the United States.
The NRC's scope of responsibility includes regulation of commercial nuclear power reactors;
nonpower research, test, and training reactors; fuel cycle facilities; medical, academic, and
industrial uses of nuclear materials; and the transport, storage, and disposal of nuclear materials
and waste. In addition to licensing and regulating the use of byproduct, source, and special
nuclear material, the NRC is also responsible for assuring that all licensed activities are
conducted in a manner that protects public health and safety. The NRC assures that none of the
operations of its licensees expose an individual of the public to more than 100 mrem/yr from all
pathways (NRC91).
The dose limits imposed by the EPA's standards for uranium fuel-cycle facilities (40 CFR Part
190) apply to the fuel-cycle facilities licensed by the NRC. These facilities are prohibited from
releasing radioactive effluents in amounts that would result in doses greater than the 25 mrem/yr
limit imposed by that standard. Currently, NRC-licensed facilities are also required to operate &
2-18
-------
accordance with the requirements of the CAA (40 CFR Part 61), which limits radionuclide
emissions into the air (EPA89b).6
The NRC exercises its statutory authority over licensees by imposing a combination of design
criteria, operating parameters, and license conditions at the time of construction and licensing. It
assures that the license conditions are fulfilled through inspection and enforcement activities.
2-5-1 Fuel Cvcle Licensees
The NRC licenses and inspects all commercial fuel cycle facilities involved in the processing and
fabrication of uranium ore into reactor fuel. NRC regulations require an analysis of probable
radioactive effluents and their effects on the population near fuel cycle facilities. The NRC also
assures that all exposures are maintained as low as reasonably achievable (ALARA) by imposing
design criteria for effluent control systems and equipment. After a license has been issued, fuel-
cycle licensees must monitor their emissions and set up an environmental monitoring program to
assure that the design criteria and license conditions have been met.
2-5-2 Radioactive Waste Disposal Licenses
The NWPA, as amended, specifies a detailed approach for high-level radioactive waste disposal.
°OE has operational responsibility and the NRC has licensing responsibility for the
transportation, storage, and geologic disposal of the waste. The disposal of high-level
radioactive waste requires a determination of acceptable health and environmental impacts that
may occur over a period of thousands of years. Current plans call for the ultimate disposal of
Waste in solid form in a licensed, geologic disposal system. The NWPA, as amended, designates
ountain, Nevada, as the candidate site for the high-level waste repository.
The EnPA provides additional direction to the NRC as to its role in the licensing of a specific
disposal site at Yucca Mountain. Section 801 of the EnPA requires the Commission to modify
its technical requirements and criteria under section 121(b) of the NWPA of 1982, as necessary,
to be consistent with EPA's standards for the Yucca Mountain site. The NRC's requirements
6 Pursuant to Section 112(d)(9) of the CAA Amendments of 1990, EPA is proposing to rescind Subpart I as it
ies to NRC-licensed facilities. The NRC is proposing to adopt a constraint level rule which would limit
radionuclide airborne emissions to 10 mrem/yr.
2-19
-------
and criteria shall assume that engineered barriers and post-closure oversight provided by the
DOE will be sufficient to: 1) prevent any activity at the site that poses an unreasonable risk of
breaching the repository's engineered or geological barriers and 2) prevent any increase in the
exposure of individual members of the public to radiation beyond allowable limits (ErtPA92).
NRC's original generic regulations governing deep geologic disposal (which were largely
developed prior to the EnPA) are contained in 10 CFR Part 60 entitled Disposal of High-level
Radioactive Wastes in Geologic Repositories (NRC81, NRC83). However, since the EnPA
specifies that sites for consideration be limited to Yucca Mountain and since the legislation
specifies the types of standards the Commission is to implement, NRC decided to promulgate
site specific standards for Yucca Mountain at 10 CFR Part 63. The proposed rule is entitled
Disposal of High-level Radioactive Wastes in a Proposed Geologic Repository at Yucca
Mountain, Nevada (Federal Register, February 22,1999). The proposed rule applies only to
Yucca Mountain. The generic rule at 10 CFR Part 60 will be modified to indicate that does not
apply to Yucca Mountain nor can it be used as a basis for litigation in NRC's Yucca Mountain
licensing procedures. The proposed 10 CFR Part 63 regulations are summarized below. In
addition, the NRC promulgates (under 10 CFR Part 71) packaging criteria for the transportation
of spent nuclear fuel and high-level and transuranic radioactive wastes. Under 10 CFR Part 72,
the NRC licenses independent spent nuclear fuel storage facilities (NRC88).
Under the proposed 10 CFR Part 63, DOE is required to conduct site characterization activities
prior to submitting a license application and to regularly report on these activities to NRC.
When DOE submits the license application it must contain certain prescribed general information
and a Safety Analysis Report. The license application must be accompanied by an environmental
impact statement. The prescribed general information includes:
• A general description of the proposed geologic repository
• Proposed schedules for construction, receipt of waste, and emplacement of wastes
• A detailed plan to provide physical protection of the waste
A description of the material control and accounting program
• A description of the site characterization work
The Safety Analysis Report is a comprehensive document with 22 prescribed elements including
such items as a description and discussion of the engineered barriers system, an assessment of the
expected performance after closure, an explanation of how expert elicitation was used, and a
description of the quality assurance program.
2-20
-------
After review of the license application and the environmental impact statement, NRC may
authorize construction of the geologic repository operations area. In deciding whether to provide
such authorization to DOE, NRC will examine safely, common defense and security and
environmental values in making its determination that construction can begin.
The NRC may subsequently issue a license to DOE to receive nuclear waste if it finds that
construction has been substantially completed, that the proposed activities in the operations area
are in conformity with the application, that the issuance of a license is not inimical to common
defense and security and will not constitute and unreasonable risk to public health and safety, and
that adequate protective measures can be taken hi the event of a radiological emergency at any
time before permanent closure. The NRC license will contain a variety of conditions relating to:
• Restrictions on the physical and chemical form and radioisotopic content of the
waste
• Restrictions on the size, shape, and materials and methods of construction of the
waste packages
• Restrictions on the volumetric waste loading
• Testing and inspection requirements to assure that any restrictions are met
• Controls to limit access and prevent disturbance of the site
• Administrative controls to assure that site activities are conducted in a safe
manner and in accordance with license requirements
e waste has been emplaced, DOE is required to file an application to amend the license
permanent closure. The DOE submission shall include, inter alia, a updated performance
^sessment of the geologic repository, and a detailed plan for post-closure monitoring of the site
including land use controls, construction of monuments and preservation of records. Upon
completion of permanent closure activities and D&D of surface facilities, DOE can then apply
for an amendment to terminate the license.
^•5-3 Repository Licensing Support Activities
The current NRC repository licensing program consists of both proactive and reactive activities.
Proactive activities include developing and reviewing regulatory requirements and guidance to
identify and resolve regulatory and technical uncertainties. Regulatory uncertainties exist where
2-21
-------
regulatory requirements are ambiguous and could be subject to various interpretations. Technical
uncertainties are related to demonstrating compliance with a particular regulation.
The NRC staff is currently developing and implementing performance assessment models using
Yucca Mountain site data. The models will assist the NRC in performing a technical assessment
of the site, as well as identifying areas of regulatory and technical uncertainty during the license
application review process. The uncertainties identified must be addressed in a timely fashion so
that the NRC can meet the three-year license review schedule mandated by the NWPA.
Additional details are provided in Chapter 7.
These activities have produced licensing review plans in anticipation of the DOE submissions.
They include review of the SCP, Study Plan, and Quality Assurance Plan (QAP).
The major focus of pre-licensing activities has been on 10 key technical issues (KTIs) that NRC
has identified as being most important to repository performance. NRC's objective is to seek
staff-level resolution of these issues during pre-licensing consultstion with DOE although the
procedure does not preclude rasing the issues during the licensing porcess. These issues are:
Total system performance assessment
• Unsaturated and saturated flow under isothermal conditions
• Rvolution of the near-field environment
• Container life and source term
• Repository design and thermal-mechanical effects
Thermal effects on flow
• Radionuclide transport
* Structural deformation and seismicity
• Igneous activity
Activities related to NRC high-level radioactive regulations
NRC periodically publishes Issue Resolution Staus Reports (IRSRs) which provide DOE with
feedback on KTI subissues. For example, NRC published IRSR Revision 1, on total system
performance assessment and integration, in November 1998 (NRC98). The report documents the
acceptance criteria NRC proposes to use for addressing each identified KTI subissue and the
review method NRC plans to use in determing whether or not the each acceptance criterion has
been met. As of the November date, 18 subissues relating to total system performance
assessment and integration had been resolved and 13 remained open.
2-22
-------
Reactive activities of the repository licensing program consist of pre-licensing reviews that
follow DOE's sequence and schedule of activities. To date, the NRC has reviewed a number of
the QAPs proposed by DOE and its contractors for Yucca Mountain. Any quality assurance
issues identified must be resolved before significant data collection activities are performed at the
Yucca Mountain site.
The NRC has also provided formal comments to DOE on the 1998 TSPA Viability Assessment
(NRC99).
As site characterization activities proceed, the NRC will review DOE's semiannual progress
reports on the site characterization program. The review will focus on the resolution of
Previously identified concerns and will evaluate new information about the site and repository
design. In addition, the NRC will review selected DOE study reports and position papers that
document the results of work performed to date, and topical and issue resolution reports that
summarize the site characterization work for specific licensing topics. These reviews will be
used to evaluate compliance with NRC regulations.
All concerns identified by the NRC will be tracked by its staff. The tracking system now being
implemented will focus not only on the issues identified, but also on DOE's progress towards
their resolution. The system also provides a licensing record of all NRC and DOE actions related
to resolving specific issues.
2-6 DEPARTMENT OF ENERGY
operates facilities for the production and testing of nuclear weapons; for the management
^d disposal of radioactive waste generated in national defense activities; for research and
development; and for the storage of spent nuclear fuel. In addition, DOE is conducting several
remedial action programs, such as the program for the management of uranium mill tailings and
the cleanup of sites formerly used for nuclear activities. These facilities and activities are not
Kcensed by the NRC. However, to protect public health and the environment, DOE has
implemented orders and procedures that are consistent with NRC regulations under 10 CFR Part
20 (NRC60), standards promulgated by the EPA, and other applicable Federal regulations and
guidelines.
D°E is also responsible for the disposal of spent nuclear fuel from the generation of electricity
by commercial nuclear reactors and high-level radioactive waste from defense activities. The
2-23
-------
facilities developed by the DOE for the management and disposal of these wastes must be
licensed by the NRC. The Yucca Mountain site in Nevada is the candidate location for disposal
of these wastes.
DOE is responsible for operating its facilities in a manner that is environmentally safe and sound,
as stated hi DOE Orders 5400.1 (DOE88) and 231.1 (DOE95a). In meeting this mandate, DOE
has issued a number of orders specifying environmental standards and procedures. Many of
these orders are currently under review to determine their conformance with NRC and EPA
regulations and standards and will be revised in accordance with the applicable NRC or EPA
guidance. Key DOE orders pertaining to the management of radioactive and hazardous
materials include:
DOE Order 460.1A (DOE96b), which establishes administrative procedures for
the certification and use of radioactive and other hazardous materials packaging
by the DOE.
DOE Order 460.2 (DOE95b), which specifies DOE's policies and responsibilities
for coordinating and planning base technology for radioactive material and
transportation packaging systems. (Cancels DOE Orders 1540.1 A, 1540.2, and
15403A-Change 1.)
DOE Order 451.1A (DOE97), which establishes procedures for implementing the
requirements of NEPA (NEP70). The order requires new facilities and existing
facilities with proposed modifications to submit EISs with then- proposed facility
design or design modification. In addition, the facilities are subject to extensive
design criteria reviews to determine compliance. (Cancels DOE Order 451.1.)
In addition to the above orders, in March 1993, DOE published a Notice of Proposed
Rulemaking for 10 CFR Part 834, entitled Radiation Protection of the Public and the
Environment (58 FR 16268) (DOE93). The proposed rule contains DOE's internal primary
standards for the protection of the public and environment against radiation. The requirements
would be applicable to control of radiation exposures from normal operations under the authority
of DOE and DOE contractor personnel. In December 1996, DOE proposed revisions to its siting
guidelines in 10 CFR Part 960 which were specific to the Yucca Mountain site (DOE96a). DOE
has not yet taken final action on its proposal.
2-24
-------
REFERENCES
AEA54 Atomic Energy Act, Public Law 83-703, as amended, 42 USC 2011 et seq., 1954.
&OE88 U.S. Department of Energy, General Environmental Protection Program, draft,
DOE Order 5400.1, November 9,1988.
DOE93 U.S. Department of Energy, Radiation Protection of the Public and the
Environment, 10 CFR Part 834, Federal Register, 58 FR 16268, March 25,1993.
DOE95a U.S. Department of Energy, Environmental Safety and Health Reporting, DOE
Order 231.1, September 30,.1995.
&OE95b U.S. Department of Energy, Departmental Materials Transportation and
Packaging Management, DOE Order 460.2, October 26,1995.
E>OE96a U.S. Department of Energy, General Guidelines for the Recommendation of Sites
for Nuclear Waste Repositories; Proposed Rule and Public Hearing, 10 CFR Part
960, Federal Register. 61 FR 66158, December 16,1996.
E>OE96b U.S. Department of Energy, Packaging and Transportation Safety, DOE Order
460.1A, October 2,1996.
DOE97 U.S. Department of Energy, National Environmental Policy Act Compliance
Program, DOE Order 451.1 A, June 5,1997.
&UN80 Dunning, D.E. Jr., R.W. Leggett, and M.G.Yalcintas, A Combined Methodology
for Estimating Dose Rates and Health Effects From Exposure to Radioactive
Pollutants, ORAL/TM-7105, 1980.
£nPA92 Energy Policy Act of 1992, Public Law 102-486, October 24,1992.
El>A71 U.S. Environmental Protection Agency, Radiation Protection Guidance for
Federal Agencies: Underground Mining of Uranium Ore, Federal Register, 36
FR 12921, July 9, 1971.
EJ)A76 U.S. Environmental Protection Agency, National Interim Primary Drinking Water
Regulations, EPA 570/9-76-003,1976.
EPA77 U.S. Environmental Protection Agency, Environmental Radiation Protection
Standards for Nuclear Power Operations, 40 CFR Part 190, Federal Register, 42
FR 2858-2861, January 13,1977.
2-25
-------
EPA79
EPA83
EPA85a
EPA85b
EPA87
EPA88
EPA89a
EPA89b
EPA91
EPA93
U.S. Environmental Protection Agency, National Emission Standards for
Hazardous Air Pollutants, ANPRM, Federal Register, 44 FR 46738, December 27
1979.
U.S. Environmental Protection Agency, Health and Environmental Protection
Standards for Uranium and Thorium Mill Tailings, 40 CFR Part 192, Federal
Register, 48 FR 602, January 5,1983.
U.S. Environmental Protection Agency, National Emission Standards for
Hazardous Air Pollutants, Standards for Radionuclides, Federal Register 50 FR
5190-5200, February 6,1985.
U.S. Environmental Protection Agency, National Emission Standards for
Hazardous Air Pollutants, Standards for Radon-222 Emissions from
Underground Uranium Mines, Federal Register, 50 FR 15386-15394 April 17
1985.
U.S. Environmental Protection Agency, Radiation Protection Guidance to
Federal Agencies for Occupational Exposure, Federal Register, 52 FR 2822-2834
January 27, 1987.
U.S. Environmental Protection Agency, Limiting Values ofRadionuclide Intake
and Air Concentration and Dose Conversion Factors for Inhalation, Submersion,
and Ingestion, Federal Guidance Report No. 11, Office of Radiation Programs
EPA 520/1-88-020, Washington, D.C., September 1988.
U.S. Environmental Protection Agency, National Emission Standards for
Hazardous Air Pollutants: Regulation of Radionuclides, 40 CFR Part 61
Proposed Rule and Notice of Public Hearing, Federal Register 54 FR 9612 9668
March 7, 1989.
U.S. Environmental Protection Agency, National Emission Standards for
Hazardous Air Pollutants: Regulation of Radionuclides, 40 CFR Part 61 Final
Rule and Notice of Reconsideration, Federal Register, 54 FR 51695 December
15,1989.
U.S. Environmental Protection Agency, 40 CFR Parts 141 and 142, Proposed
Rule, National Primary Drinking Water Regulations; Radionuclides Federal
Register, 56 FR 33050, July 18, 1991.
U.S. Environmental Protection Agency, 40 CFR Part 191, Proposed Rule
Environmental Radiation Protection Standards for the Management and Disposal
2-26
-------
of Spent Nuclear Fuel, High-level and Transuranic Radioactive Wastes, 58 FR
7924 - 7936, February 10,1993.
EPA93a U.S. Environmental Protection Agency, External Exposure to Radionuclides in
Air, Water and Soil, Federal Guidance Report No. 12, Office of Radiation and
Indoor, EPA402-R-93-081, September 1993.
EPA94 U.S. Environmental Protection Agency, Federal Radiation Protection Guidance
for Exposure of the General Public; Notice, Federal Register, 59 FR 66414,
December 23,1994.
EPA96 U.S. Environmental Protection Agency, Criteria for the Certification and
Recertification of the Waste Isolation Pilot Plant's Compliance with the 40 CFR
Part 191 Disposal Regulations; Final Rule, Federal Register, 61 FR 5224-5245,
February 9, 1996.
EPA98 U.S. Environmental Protection Agency, Health Risks from Low-Level
Environmental Exposure to Radionuclides, Federal Guidance Report No. 13, Part
I - Interim Version, Office of Radiation and Indoor Air, EPA 402-R-097-014,
January 1998.
ERA74 Energy Reorganization Act, as amended, 1974.
FRC60a Federal Radiation Council, Radiation Protection Guidance for Federal Agencies,
Federal Register, 25 FR 4402-4403, May 18,1960.
FRC60b Federal Radiation Council, Staff Report No. 1, Background Material for the
Development of Radiation Standards, May 13,1960.
!AE89a International Atomic Energy Agency, Guidance for Regulation of Underground
Repositories for Disposal of Radioactive Wastes, Safety Series No. 96, Vienna,
Austria, 1989.
!AE89b International Atomic Energy Agency, Safety Principles and Technical Criteria for
the Underground Disposal of High-Level Radioactive Wastes, Safety Series No.
99, Vienna, Austria, 1989.
ICR34 International X-Ray and Radium Protection Commission, International
Recommendations for X-Ray and Radium Protection, British Journal of Radiology
7, 695-699,1934.
2-27
-------
ICR38 International X-Ray and Radium Protection Commission, International
Recommendations for X-Ray and Radium Protection, American Journal of
Roentgenology and Radium, 40,134-138,1938.
ICR51 International Commission on Radiological Protection, International
Recommendations of Radiological Protection 1950, British Journal of Radiology
24,46-53,1951.
ICR60 International Commission on Radiological Protection, Recommendations of the
International Commission on Radiological Protection Report of Committee II on
Permissible Dose for Internal Radiation (1959), ICRP Publication 2, Pergamon
Press, Oxford, 1960.
ICR65 International Commission on Radiological Protection, Recommendations of the
ICRP 1965, ICRP Publication 9, Pergamon Press, Oxford, 1965.
ICR77 International Commission on Radiological Protection, Recommendations of the
ICRP, ICRP Publication 26, Pergamon Press, Oxford, 1977.
ICR79 International Commission on Radiological Protection, Limits for Intakes of
Radionuclides by Workers, ICRP Publication 30, Part 1, Ann. ICRP 2 (3/4),
Pergamon Press, Oxford, 1979.
ICR84 Annals of the ICRP, Vol 14, No. 1,1984, Statement from the 1983 Washington
Meeting of the ICRP.
ICR85a Annals of the ICRP, Vol. 15, No. 3, 1985, Statement from the 1985 Paris Meeting
of the ICRP.
ICR85b International Commission on Radiological Protection, Radiation Protection
Principles for the Disposal of Solid Radioactive Waste, ICRP Publication 46,
Pergamon Press, Oxford, 1985.
ICR91 International Commission on Radiological Protection, 1990 Recommendations of
the International Commission on Radiological Protection, ICRP Publication 60
Pergamon Press, Oxford, 1991.
NAS95 National Academy of Sciences - National Research Council, Committee on
Technical Bases for Yucca Mountain Standards, Technical Bases for Yucca
Mountain Standards, National Academy Press, Washington, D.C., 1995
NBS63 National Bureau of Standards (1963), Maximum Permissible Body Burdens and
Maximum Permissible Concentrations of Radionuclides in Air and Water for
2-28
-------
Occupational Exposure, NBS Handbook 69 as amended August 1963, U.S.
Department of Commerce. Revised and republished in 1963 as NCRP Report No.
22 by the National Committee on Radiation Protection and Measurements.
NCR54 National Council on Radiation Protection and Measurements, Permissible Dose
from External Sources of Ionizing Radiation., National Bureau of Standards
Handbook 59,1954.
NCR59 National Council on Radiation Protection and Measurements, Maximum
Permissible Body Burdens and Maximum Permissible Concentrations of
Radionuclides in Air and in Water for Occupational Exposure, National Bureau of
Standards Handbook 69, 1959.
NCR87 National Council on Radiation Protection and Measurements, Recommendations
on Limits for Exposure to Ionizing Radiation, NCRP Report No. 91, June 12,
1987.
NEA94 Nuclear Energy Agency, NEA Annual Report: 1994 Activities, 1994.
NEP70 National Environmental Policy Act of 1970, Public Law 91 -190, January 1,1970.
NIX70 The White House, President R. Nixon, Reorganization Plan No. 3 of 1970,
Federal Register, 35 FR 15623-15626, October 6,1970.
NRC60 U.S. Nuclear Regulatory Commission, Standards for Protection Against
Radiation, 10 CFR Part 20, Federal Register, 25 FR 10914, November 17,1960,
and as subsequently amended.
NRC81 U.S. Nuclear Regulatory Commission, Disposal of High-level Radioactive
Wastes in Geologic Repositories: Licensing Procedures, 10 CFR Part 60, Federal
Register, 46 FR 13971-13988, February 25, 1981.
NRC83 U.S. Nuclear Regulatory Commission, 10 CFR Part 60, Disposal of High-Level
Radioactive Wastes in Geologic Repositories, Technical Criteria, Federal
Register, 48 FR 28194-28229, June 21,1983.
NRC88 U.S. Regulatory Commission, 10 CFR Part 72, Licensing Requirements for the
Independent Storage of Spent Nuclear Fuel and High-Level Radioactive Waste,
53 FR 31658, August 19, 1988.
NRC91 U.S. Nuclear Regulatory Commission, 10 CFR Part 20 et al., Standards for
Protection Against Radiation, Final Rule, Federal Register, 56 FR 98, May 21,
1991.
2-29
-------
NRC98 U.S. Nuclear Regulatory Commission, Issue Resolution Status Report, Key
Technical Issue: Total System Performance Assessment and Integration, Revision
J, Division of Waste Management, Office of Nuclear Material Safety and
Safeguards, November 1998.
NRC99 U.S. Nuclear Regulatory Commission, Staff Review of the U.S. Department of
Energy Viability Assessment for a High-Level Waste Repository at Yucca
Mountain, Nevada, Letter to Lake H. Barnett, OCRWM/DOE from Carl J.
Paperiello, ONMSS/NRC, June 2,1999.
NWP83 Nuclear Waste Policy Act of1982, Public Law 97-425, January 7,1983.
NWP87 Nuclear Waste Policy Amendments Act of '1987, Public Law 100-203, December
22,1987.
OEC93 Organization for Economic Cooperation and Development/Nuclear Energy
Agency, Nuclear Waste Bulletin: Update on Waste Management Policies and
Programs, No. 8, July 1993.
OEC95 a Organization for Economic Cooperation and Development/Nuclear Energy
Agency, Radiation Protection Today and Tomorrow, 1995.
OEC95b Organization for Economic Cooperation and Development/Nuclear Energy
Agency, Nuclear Waste Bulletin: Update on Waste Management Policies and
Programs, No. 10, June 1995.
SNI95 Snihs, Dr. J.O., "Radioactive Waste Disposal: Radiological Principles and
Standards," IAEA Bulletin, 1995.
WIP92 Waste Isolation Pilot Plant Land Withdrawal Act, Public Law 102-5 79, October
20,1992.
2-30
-------
CHAPTERS
SPENT NUCLEAR FUEL AND HIGH-LEVEL WASTE DISPOSAL PROGRAMS
IN OTHER COUNTRIES
As in the United States, other countries that use nuclear power are establishing long-term
programs for the safe management and disposal of spent nuclear fuel and high-level radioactive
waste. Such programs include adopting a national strategy, assigning the technical responsibility
for research and development activities to designated agencies, selecting disposal strategies and
development activities, and setting the appropriate regulatory standards to protect public health
and the environment. Management strategies may include spent nuclear fuel storage at and away
from reactor sites, spent nuclear fuel reprocessing, high-level waste vitrification and storage,
partitioning and transmutation of the waste into short-lived or stable forms, and disposal in deep
geologic media. Typically, the objective of such geologic disposal programs is to immobilize
and isolate radioactive waste from the environment for a sufficient period of time under
conditions such that any radionuclide releases from the repository will not result in unacceptable
radiation exposure of the public. This strategy takes advantage of the geology surrounding the
disposal site to act as a passive barrier to radionuclide releases and eliminates many surface
factors, such as sabotage, hurricanes, theft, and flooding, which could compromise an above
ground facility.
As discussed in Chapter 1 of the BID, deep geologic disposal is considered by many in the
scientific community to be the most promising method for disposing of long-lived nuclear waste.
Consequently, several nations have begun activities associated with disposal of spent nuclear fuel
and high-level waste by isolation in deep geologic formations. These countries envision
emplacing solidified high-level waste in a deep geologic formation located within their borders.
Only the United States and Germany have identified candidate locations for disposal of high-
level waste, i.e., the Yucca Mountain site in Nevada and the Gorleben site in Germany1. Other
countries are, to varying degrees, engaged in technical evaluations of the potential suitability of
indigenous geologic formations for disposal. Some nations, such as France, have several
geologic formations, such as clay and granite, that might be used for disposal, and each
alternative is being evaluated. Others, such as Canada, have focused on one type of geologic
1 As will be discussed in Section 3.5.2, the suitability of the Gorleben site has been questioed by the new
German government which was elected in 1998. .
3-1
-------
formation. (Canada is evaluating a crystalline rock formation in a setting with low seismic
activity.) In addition, several countries, such as Canada and Sweden, have established
underground research laboratories (URL's) and extended their research programs to include
participation by other nations with similar candidate geologies. For example, the Swedish
research facility is in a crystalline rock formation and its research program has included
participation by Japan, Spain, Finland, Switzerland, the U.S.7, and Canada.
The disposal strategies for all nations assume that waste isolation will be maintained by reliance
on a combination of engineered and natural barriers between the emplaced waste and the
environment. Currently, the United States is considering a potential repository site at Yucca
Mountain, Nevada where the disposal facility would be located in an arid environment and
wastes would be emplaced in an unsaturated geohydrologic regime. The geohydrologic features
of the Yucca Mountain site allow potential to use a thermal loading strategy in which heat
emissions can deter water from contacting waste packages for some period of time. This type of
repository environment and disposal strategy is uniquely available to the Yucca Mountain site in
comparison with other current national programs.
Other countries, because of the characteristics of their available geologic formations, are placing
greater emphasis on engineered barrier systems and are designing these systems to ensure their
long-term performance as a barrier to radionuclide release. For example, in response to a
mandate from the national government in the 1970s, the Swedish commercial nuclear waste
program developed an engineered barrier concept involving emplacement of spent nuclear fuel in
a copper matrix contained within a highly-robust copper canister. The viability of this concept to
maintain wastes in isolation for one million years in Sweden's geologic formations was then
demonstrated, as required by the governmental directive. Sweden has not, however, committed to
the use of this engineered barrier concept.
Various nations and international agencies, in addition to the United States, have begun to give
consideration to regulations and regulatory standards for high-level waste disposal. Some
nations have developed broad risk or dose criteria, and some have supplemented such criteria
with additional qualitative technical criteria concerning features of the disposal system.
International organizations, such as the Nuclear Energy Agency in Paris, France, provide
7The United States no longer actively participates in cooperative R&D programs with other countries.
However, the U.S. continues to exchange information with other countries through its bilateral agreements and its
representation in international agencies.
3-2
-------
opportunities for discussion of regulatory criteria and also provide programs of common interest,
such as comparison of performance assessment computer codes. There are, however, no
international standards for high-level waste disposal accepted by all nations.
Although the performance standards and criteria for the various national regulations are similar,
each nation has established specific requirements to meet its needs. Current information
concerning the provisions of national and international criteria and objectives for the safety of
long-lived radioactive waste disposal is presented in Table 3-1. As will be clear from the
ensuing discussion, regulatory requirements are still evolving and the information Table 3-1 is
subject to change.
Characteristics of programs in ten nations with major commitments to nuclear power and
existing activities concerning disposal of high-level wastes are summarized below. The
descriptions address nuclear power utilization, waste disposal, and regulatory programs.
Discussions are provided for programs in Belgium, Canada, Finland, France, Germany, Japan,
Spain, Sweden, Switzerland, and the United Kingdom.
3.1 BELGIUM
3.1.1 Nuclear Power Utilization
In 1994, Belgium met about 56 percent of its electrical needs through nuclear power (EIA95).
The Belgian nuclear power program relies on seven pressurized light water reactors, all of which
are operated by Electrabel, a privately-owned company.
From 1966 to 1974, Belgium reprocessed spent nuclear fuel at its Eurochemic facility. The
company Belgoprocess was created as a consortium with foreign firms to reactivate the
Eurochemic plant, but these efforts failed in the mid-1980s. Belgoprocess is now responsible for
decommissioning the plant. Belgium had previously shipped spent nuclear fuel to France for
reprocessing, but stores current inventories in reactor pools and in dry storage facilities pending
the results of a future parliamentary debate on whether to continue reprocessing. France will
soon begin returning to Belgium the high-level vitrified waste created from the reprocessing of
Belgian fuel. It is estimated that by the year 2000, Belgium will have produced about 2,500
MTHM of spent nuclear fuel.
3-3
-------
Table 3-1. National and International Criteria and Objectives for the Disposal of
Long-Lived Radioactive Wastes (OEC95a)
f&mwzaisnfif ,
Gtmttyf*
NBA (1984)
ICRP
(Pub. 46, 1985)
IAEA
(Safely Series 99,
1989)
CANADA
(Reg. Document
R-104, 1987)
FINLAND
Decision of the
Council
of State, 1991)
FRANCE
(Basic Safety
Rule,
RFS ni.2.f, 1991)
GERMANY
Section 45, para.
1 of Radiation
>rotection
Ordinance, 1989)
NORDIC
COUNTRIES
Basic Criteria
Document, 1993)
;- ,$&&$$»* '
^^..jDfelecafeyCd^na'^ -
For HLW:
max. indiv. risk < lO'Vy
(all sources)
For HLW, for individuals
(all sources):
1 mSv/y (normal
evolution scenarios);
10'Vy (probabilistic scenarios)
Idem ICRP Publication 46
For HLW:
max. indiv. risk objective:
< 10"*/y
For LLW and ILW:
max. indiv. dose
< 0.1 mSv/y, with max.
indiv. dose < 5 mSv/y from
accident conditions caused by
possible natural events or
human actions
For ILW and HLW:
max. indiv. dose
< 0.25 mSv/y for normal
evolution scenarios;
for altered evolution scenarios,
risk may be considered
(probability
of scenario times effect
of exposure)
For all waste types:
max. indiv. dose
< 0.3 mSv/y for all reasonable
scenarios
For all waste types:
max. indiv. dose
< 0.1 mSv/y (normal
scenarios):
max. indiv. risk
< 10"*/y (disruptive events)
*•>•.""• '
" v J -:>. "
Both probability and dose
should be taken into
account in ALARA
Period of time for
demonstrating compliance:
10V
No sudden and dramatic
increase for times > lOfy
Beyond 10* y, dose limit
is considered as a
"reference" level
Calculation of individual
doses limited to 10*y
but isolation potential
beyond 10*y may be
assessed
For HLW, additional
criterion on "total
activity inflow" limiting
releases to biosphere,
based on inflow of natural
alpha radionuclides
\ Criteria for &<%XB§
BOttmfc»^iM-(BI)
'•.,- 'Scenarios ;, * -
Indiv. risk/dose - best
criterion to judge long-
term acceptability
Future human activities
should be treated
probabilistically
Future human activities are
randomly disruptive events
that usually are examined
probabilistically
Main criteria applicable to
all exposure scenarios;
no criteria specific to HI
scenarios
Max. indiv. dose < 5
mSv/y from possible human
actions
Assumptions (French Basic
Safely Rule, Appendix 2):
Date of HI occurrence
>500y;
Existence of repository and
location forgotten;
Level of technology same
as present day
* , *<*> , '.
' ,"3' - ,
<•* • v;O6inineats * <
No consensus on
ALARA/
optimization
ALARA useful, notably
to compare alternatives,
but may not be the most
important siting factor
Also includes qualitative
technical criteria on
disposal system features
and role of safety
analysis and quality
assurance
Additional qualitative,
non-prescriptive
requirements and
guidelines in regulatory
documents
For spent nuclear fuel or
HLW, proposed criterion
for max. indiv. dose <
0.1 mSv/y
Technical criteria for
siting established in 1987
Additional qualitative
technical criteria in
guidelines and
regulatory document
Includes other
qualitative criteria
3-4
-------
Table 3-1 (Continued)
OjgatB:23tlOIJ/
'•CouHatey/
Jta&renGe
SWITZERLAND
(Reg. Document
R-21, 1993)
UNITED
KINGDOM
(OECD/NEA
Doc.
66-94-041,1995)
•. .• •. '*
- 1. f>
Main Objective/ "x
OMe&iv$C*fc$da - f
For all waste types:
max. indiv. dose
< 0.1 mSv/y at any time for
reasonably probable scenarios;
max. indiv. risk
< lO*/y for unlikely scenarios
For L/ILW:
10~Vy target for indiv, risk from
a single facility
For HLW:
no specific criteria but likely
application of principles similar
to existing objectives for L/ILW
O.^ %A/ J f
V^^MJ^v
- *n
I, • ' ,, *
OthBir M»m -Bealares -
Repository must be
designed in such a way
that it can at any time be
sealed within a few years
without the need for
institutional control
No time frame for
quantitative assessments
specified
^Crifectfttetatytai^ I
Hmnain Intrusion (Hi) ;
- Scenarios
No criteria for HI scenarios
except that for high
consequences, probabilities
can be
taken into account
Main criterion for HI
scenarios currently indiv.
risk
< . , :
<•.' -.^ '
\'* ' A \ ^
Comments ..•
ALARA to be used to
the extent practical and
reasonable
In 1985, a vitrification plant, PAMELA, began processing high-level waste from the Eurochemic
plant. Vitrified high-level waste will be stored in an intermediate storage facility (recently
constructed by the National Agency for Radioactive Waste and Fissile Materials, (ONDRAF))
for 50-70 years. Characterization of a potential site for a repository located in a clay formation at
the Mol-Dessel site in the northeast corner of the country is progressing.
3.1.2 Disposal Programs and Management Organizations
The Belgian program to establish a radioactive waste repository was initiated in 1974 with the
establishment of a government-sponsored research and development initiative. In 1982, the
National Agency for Radioactive Waste and Fissile Materials, ONDRAF, was established to
implement and manage a multi-year national program addressing the long-term management and
disposal of radioactive wastes, including spent nuclear fuel, high-level waste, and other
reprocessed waste returned from French facilities. SYNATOM, an agency privatized in 1994, is
responsible for uranium procurement, reprocessing spent nuclear fuels, off-site waste
management, and disposal of packaged irradiated fuel assemblies. The Nuclear Research Center
(CEN), under the Ministry of Economic Affairs, provides technical assistance in basic and
applied R&D in nuclear energy and technology.
Belgium's waste disposal program takes a multi-barrier approach, relying significantly upon the
low permeability of the Boom Clay formation and the assumption that there are no structural fast
paths through the clay. Engineered barriers are anticipated to last no more than a few thousand
years, after which the geologic barriers will provide primary containment. For this reason,
3-5
-------
engineered barriers have been designed to minimize their impact on surrounding geology, and to
provide interim public health protection (NWT94).
ONDRAF intends to begin operation of a shallow land-burial facility for low-level waste by the
year 2000 and has established an underground laboratory in a clay formation at Mol-Dessel to
evaluate the site's suitability as a high-level waste repository. Twenty years of research at Mol-
Dessel has led to significant evolution of the design for the planned repository. The clay
formation in which the site is situated is the only suitable geological medium that has been
identified in Belgium. In 1980, a repository conceptual design was developed for a clay site, and
an underground research laboratory at Mol-Dessel (Project HADES) began operation by 1985.
In recent years, the repository's design has been altered. The original HADES design was
conceived according to the perceived thermal tolerance of the host rock and was intended to
allow for retrieval of containers of vitrified waste for a long period of time. The new design does
not permit easy retrieval and allows for more homogeneous dispersion of heat. The new design
is also believed to be simpler to construct and less damaging to the surrounding clay layer
(NWT95).
The underground research laboratory at Mol-Dessel extends to a depth of 224 meters. Research
conducted there includes experiments in corrosion properties of containers and engineered
barriers, geochemistry and radionuclide migration, backfilling and sealing technology, and near-
field effects of heat and radiation on clays. Over the next few years, a preliminary demonstration
test for clay disposal (PARCLAY) will be launched at the Mol-Dessel site. PARCLAY will be
used to investigate the thermal effects of final disposal on clay and will include the construction
of a new 1:1 scale gallery. Based on the outcome of these studies, a larger underground facility
might be constructed for a full-scale demonstration project. Assuming that the results of
investigations at Mol-Dessel are favorable, repository construction could begin around 2025 and
operation around 2030 (OEC95b).
3.1.3 Regulatory Organizations and Their Regulations
Belgium does not currently have specific regulatory requirements or criteria governing the
disposal of spent nuclear fuel or high-level waste. In 1994, the Federal Nuclear Inspection
Agency (AFCN) was created to oversee inspection and surveillance of Belgium's nuclear
facilities under guidance of the Ministry of Employment and Labor and the Ministry of Public
3-6
-------
Health and the Environment. The King of Belgium has the authority to grant, suspend, reject or
withdraw authorization for the construction and operation of nuclear facilities (OEC95c).
3.2 CANADA
3,2.1 Nuclear Power Utilization
In 1994, Canada produced about 19 percent of its electrical needs through nuclear power
(22 pressurized heavy-water cooled and moderated "CANDU" reactors (EIA95)). Canadian
utilities currently produce a surplus of electric power, and only one new nuclear facility is
planned before the year 2005 (OEC95a). Canada's nuclear power is produced by three provincial
utilities, Ontario Hydro*, Hydro Quebec, and New Brunswick Power As of February 1, 1999
Ontario Hydro had 12 reactors in operation, seven under extended shutdown and one in a laid-up
decommissioning mode. The reactors under extended shutdown may be brought back on line in
the 2000-2009 timeframe depending on economic conditions, The Hydro Quebec reactor and the
New Brunswick Power reactor are both operational (KIN99).
Canada relies on the CANDU reactor design, which uses natural uranium in a once-through fuel
cycle. Currently, the program considers only direct disposal of spent nuclear, fuel without
reprocessing, although the reprocessing option has not been completely ruled out. Until a
repository is available, spent nuclear fuel will initially be stored at each reactor site and, later,
possibly at a central facility. Estimates indicate that Canada will have produced about 34,000
MTHM of spent nuclear fuel by the year 2000. The country's five existing sites (the Ontario
Hydro utility has three sites) have adequate storage facilities for spent nuclear fuel, and there is
little urgency to dispose of waste. Current time lines suggest that a disposal facility could be
established by about 2025 (OEC95c).
Ontario Hydro fuel has been stored at the Pickering Waste Management Facility since 1995
(KIN99). Dry storage utilizes concrete-filled, steel-shelled vessels which contain 384 fuel
bundles each. Currently 600 metric tons of uranium (3 0,000 fuel bundles) are in storage at
Pickering. Ontario Hydro has filed an application to construct the Bruce Waste Management
8 Ontario Hydro was split into several companies on April 1, 1999 with nuclear power production assigned
to Ontario Power Generation Inc.
3-7
-------
Facility with planned operation in 2002, and a similar facility is planned for Darlington in 2005.
New Brunswick Power has operated a dry storage facility at its Pt. Lepreau reactor since 1992
where storage is done in concrete steel-lined vessels each containing 540 fuel bundles. Hydro
Quebec uses a concrete vault for spent fuel storage at its Gentilly-2 reactor site.
3.2.2 Disposal Programs and Management Organizations
Responsibility for the management and disposal of Canada's nuclear waste was allocated in 1978
under the Canadian Nuclear Fuel Waste Management Program. Ontario Hydro (owner of 20 out
of the country's 22 total nuclear power units) assumed the responsibility for interim storage and
transportation of nuclear fuel waste9. The Federal corporation, Atomic Energy of Canada
Limited (AECL), took responsibility for research and development on deep repository disposal,
with support given by Ontario Hydro.
The Canadian disposal concept involves siting a repository at a depth of 500-1000 m in a
granitic formation located in the Canadian Shield, which is a large region of geologically old
rocks that is tectonically quiescent and centered around the Hudson Bay. It stretches east to
Labrador, south to Lake Ontario, and northwest to the Arctic Ocean. The repository would be
located at a depth between 500 and 1,000 meters. Spent nuclear fuel canisters would be inserted
into floor cavities located hi excavated disposal rooms and surrounded with a mix of bentonite
and silica sand. A mix of glacial rock clay and crushed granite aggregate, along with engineered
barriers, would be used to seal most of the remainder of the vault (OEC93).
AECL submitted an Environmental Impact Statement (EIS) evaluating the planned disposal
program to the Federal Environmental Assessment Review Panel in 1994. AECL estimated that
siting, licensing, and construction of a disposal facility will take 25 to 30 years and that the
facility could, therefore, be in operation by 2025. The Environmental Assessment Panel reported
in early 1998 that the AECL concept for deep geologic disposal did not enjoy broad public
support and that a number of steps were required to achieve such support (EAP98). Until revised
management and public participation procedures were put in place and broad public support was
obtained, no work to search for a specific site should proceed. In the course of seeking public
9 One additional reactor is owned by Hydro Quebec and one by New Brunswick Power.
3-8
-------
support, the AECL repository concept may emerge as the most acceptable approach but that
issue must remain open until other policy issues have been resolved.
In response to the report of the Environmental Assessment Panel, the Canadian government set
forth its position in December 1998 (KIN99). In its response the government asserted that it
expects producers and owners of nuclear fuel to establish, as a separate legal entity, an
organization to manage and coordinate the full range of activities relating to the management of
the fuel waste including disposal. The producers and owners of the fuel are also expected to
establish a fund to develop and compare waste management options, design and site the preferred
approach for long-term management, implement the preferred approach, and decommission the
waste management facilities. Ontario Hydro has assumed the lead hi investigating how best to
establish and separately fund the waste management entity.
In 1986, AECL established the Whiteshell underground research laboratory hi undisturbed
granitic rock at a depth of 240 meters at Lac du Bonnet in the Province of Manitoba. The AECL
has since deepened the facility to 440 meters. The purpose of the laboratory is to conduct large-
scale, in-situ experiments in the type of rock envisioned under the Canadian disposal concept,
demonstrating some of the components of the disposal concept (the facility is not a candidate
repository site). AECL is developing methodologies and analytical techniques to evaluate the
geomechanical and geohydrological properties of granitic rock. The underground research
laboratory was also recently used to study the possible effects of microbial activity in a disposal
system. Other studies conducted at the laboratory include large-block radionuclide migration
studies, container corrosion studies, and an alternate post-closure assessment case study.
As the result of a government-wide Federal Program Review begun by Ottawa in 1995, hi
response to a large national budget deficit, AECL's responsibilities have been pared back
considerably. The AECL will now focus on a core mission of developing and vending CANDU
reactor technology. The AECL's Whiteshell facilities, including the Underground Research
Laboratory, may be privatized. In 1997, the Canadian government was in final negotiations to
sell these facilities to Canadian Nuclear Projects Ltd., a consortium led by British Nuclear Fuels,
Ltd. and Wardrup, a large Canadian engineering firm. However, the deal fell through and the
government is looking for other economic opportunities for this site. In the meantime, AECL
continues to operate the Underground Research Laboratory.
3-9
-------
3.2.3 Regulatory Organizations and Their Regulations
Regulation of nuclear matters in Canada is handled under the Atomic Energy Control Act The
Atomic Energy Control Board (AECB) is currently the lead regulatory agency in Canada for
assessing the long-term performance of the disposal facility. The AECB also develops and
issues policy statements and regulatory guidance for the eventual licensing of the high-level
waste repository. Other provincial and Federal agencies operate under AECB in the regulation of
some activities in the nuclear fuel cycle.
In 1987, the AECB issued a policy statement containing objectives, requirements, and guidelines
on nuclear waste disposal and high-level waste repository siting (AECB87). The overall
objective expressed in these documents is to ensure that there is only a small probability that
radiation doses to the public associated with the repository will exceed a small fraction of natural
background radiation doses. Under this policy, predicted radiological risk of death to
individuals from a waste repository must not exceed 1 x 10'6 annually. For the purpose of
demonstrating compliance with the individual risk requirement, the time period need not exceed
10,000 years (AECB87). However, Canada does not have any nuclear-waste-specific regulation
at the present time (KIN99).
Canada is in the process of replacing the Atomic Energy Control Act with the Nuclear Safety and
Control Act (NSC A) and the new law is expected to be in place in mid-1999. Under the terms of
the NSCA, the AECB will be replaced with an expanded Canadian Nuclear Safety Commission
(CNSC). The NSCA will provide the framework under which licenses for site preparation,
construction, operation, decommissioning, and abandonment of nuclear waste facilities are
obtained. One of the tasks allocated to the CNSC is the preparation of regulatory guidance
documents. Any application to build a nuclear waste facility would initiate a requirement to
prepare an environmental assessment under the Canadian Environmental Assessment Act
(KIN99).
3-10
-------
3.3 FINLAND
3.3.1 Nuclear Power Utilization
Currently, Finland has four nuclear power plants in operation. These facilities are owned and
operated by two separate utilities. Two 445 MWe PWR units, located at Loviisa on the southern
coast of Finland, are run by Imatran Voima Oy (IVO). Two 710 MWe BWR units, located at
Olkiluoto on the western coast of Finland, are run by Teollisuuden Voima Oy (TVO). The
possibility of a fifth reactor has been under consideration for a number of years in Finland. The
Technical Research Centre of Finland (VTT) also operates a small (250 kW) Triga research
reactor at its Reactor Laboratory (VTT93).
In the past, the two utilities had different waste management strategies. Until recently, IVO has
shipped its spent fuel to Russia with no return of reprocessing wastes to Finland. Spent fuel
generated was allowed to cool for a period of four to five years at the IVO reactor site before
being shipped. However, this practice is no longer allowed. TVO's strategy involves disposing
of its spent fuel in Finnish bedrock. No plans exist for reprocessing the spent fuel from TVO's
reactors, although this option remains open (VTT93).
In May 1995, TVO and IVO agreed to cooperate for the final disposal of spent fuel and a new
company, Posiva Oy, was established. This company, which is jointly owned by TVO and IVO,
took over TVO's program on spent fuel disposal. The total amount of spent fuel to be disposed
of in Finland is now estimated to be about 1,700 MTHM of BWR fuel from TVO's reactors at
Olkiluoto and 740 MTHM of PWR fuel from IVO's reactors at Loviisa (OEC96).
3.3.2 Disposal Programs and Management Organizations
In 1983, the Finnish government set general targets and schedules for research and development
of a nuclear waste management program. Based on these guidelines, TVO conducted
preliminary site investigations for a spent fuel repository at five sites: Olkilouto in Eurajoki;
Kivetty in Konginkangas (now a part of Aanekoski); Romuvarra in Kuhmo; Syyry in Sievii; and
Veitsivaara in Hyrynsalmi. Technical plans for managing and disposing spent fuel were also
developed. In addition to these activities, extensive research has been conducted on the
phenomena and processes affecting the safety of long term disposal of spent fuel. Several
3-11
-------
Finnish research institutes, universities, and companies made R&D contributions, including the
Technical Research Centre of Finland, the Geological Survey of Finland, and the Department of
Radiochemistry at the University of Helsinki. Finland has also drawn upon the expertise of other
countries, particularly the Swedish Nuclear Fuel and Waste Management Company (8KB)
(YJT92).
The decision, in principle, made by the Finnish government in 1983 required TVO to propose
suitable areas for more detailed site investigations by the end of 1992. As discussed above,
Posiva Oy was established in 1995 and began operation in early 1996. Since its creation, Posiva
Oy has continued investigating the three candidate sites originally selected in 1992: Olkiluoto,
Kivetty, and Romuvaara. In 1994, TVO also conducted a preliminary feasibility study in
Kannonkoski, near Aanekoski. Another feasibility study has been initiated for the island of
Hastholmen in Loviisa (OEC96).
Finland plans to select the site for the repository in the year 2000 based on updated technical
information on encapsulation and facility design, as well as site-specific safety analyses
(OEC96). Detailed construction plans for the encapsulation facility and repository must be
presented in 2010 and the repository is to go into operation in 2020. Current plans call for
sealing the repository in the year 2050 (YJT92).
3.3.3 Regulatory Organizations and Their Regulations
Finland's Ministry of Trade and Industry is responsible for nuclear power in the country. The
Finnish Centre of Radiation and Nuclear Safety is responsible for nuclear safety, including
nuclear waste management. This latter organization reviews technical and safety-related license
applications (VTT93).
The principles of Finland's waste management policy were originally outlined in the
Government's policy decision of 1983. The nuclear energy law of Finland includes specific
directives concerning nuclear waste management. Each nuclear waste producer is responsible for
the safe handling and management of its waste, including final disposal. This responsibility
extends to the financing of such operations. No governmental organizations are envisioned for
waste management operations. The utilities contribute to future waste management activities
3-12
-------
through the Nuclear Waste Management Fund established by the Finnish Ministry of Trade and
Industry (VTT93).
In developing its radiological and safety criteria for nuclear waste disposal systems, Finland is
closely following the international efforts of the International Atomic Energy Agency (IAEA),
the International Commission on Radiological Protection (ICRP), and the Nuclear Energy
Agency of the Organization for Economic Cooperation and Development. In addition, the
Nordic governments published joint recommendations in 1993 hi a document entitled Disposal
of High Level Radioactive Waste, Consideration of Some Basic Criteria (OEC96).
3.4 FRANCE
3.4.1 Nuclear Power Utilization
In 1994, France met approximately 75 percent of its electrical needs through nuclear power,
having the highest per capita installed capacity in the world (OEC95c). The French nuclear
power program relies on 56 units, the vast majority of which are light water reactors. Older gas
cooled reactors are being phased out, while research and development activities and
demonstration projects focus on an alternate reactor designs (liquid metal fast breeder reactor) for
power production. France plans to continue building nuclear power plants, although some will
serve as replacements for old facilities. The overall contribution of nuclear power to the
country's electricity production is not expected to exceed 80 percent (GAO94).
The French radioactive waste disposal program is based on a closed fuel cycle involving spent
nuclear fuel reprocessing and recovery and re-use of plutonium in breeder and light water
reactors. From 1976 through 1990, France reprocessed over 20,000 MTHM of metallic and
oxide fuel. France has already begun to solidify high-level waste in glass and, ultimately,
intends to dispose of it—as well as alpha-emitting transuranic waste—in deep geological
formations. Vitrification plants for France's two reprocessing plants, UP2 and UP3, entered
service in 1990 and 1992. France also provides reprocessing services to foreign customers; in
1993, the international component comprised an estimated one-third of the country's
reprocessing business.
3-13
-------
3.4.2 Disposal Programs and Management Organizations
The French nuclear waste program has been entrusted to the National Radioactive Waste
Management Agency (ANDRA). ANDRA was formed in 1979 as an arm of the French Atomic
Energy Commission (CEA), but 1991 legislation made it an independent entity. ANDRA is
responsible for all radioactive waste disposal activities and long-term waste management. Other
organizations with key roles in the management of the country's high-level waste include
Electricite de France (the national electric utility) and COGEMA (operator of spent nuclear fuel
reprocessing and high-level waste immobilization and storage facilities).
In 1987, ANDRA identified four geological media for potential high-level waste disposal—clay,
salt, granite, and schist—and began investigative work at a site in each medium. An
underground research laboratory was to be established at one or more of the candidate sites; if
found suitable, and one of these sites was to have been converted to an operating repository to
receive transuranic waste by 2000 and high-level waste by 2010. However, in light of the serious
public protests at three of the sites under investigation, former Prime Minister Michel Rocard
declared a one-year moratorium on siting activities to allow a reassessment of France's overall
waste management strategy. The moratorium began in February of 1990, and, in January 1991,
the Parliamentary Office for the Assessment of Technological Options published a report that
recommended major changes to the program.
On December 30,1991, the Parliament enacted a new Law on Radioactive Wastes. The 1991
law requires the government to submit a report to Parliament after 15 years that assesses the
results of studies on partitioning and transmutation of actinides, the retrievable or permanent
storage of high-level waste in deep geologic formations including the use of underground
laboratories, and the technologies for waste conditioning and surface storage. (Work on the first
and third options is coordinated by CEA while work on the second option is the coordinated by
ANDRA.) The report must also propose a bill authorizing an underground waste repository. At
this time, no schedule has been set for developing such a repository. Instead, the Parliament will
reassess the program based on the results of the 15-year research phase. The law states that, once
the underground research laboratory is built, only research-level quantities of waste may be
emplaced into it until the Parliament votes to convert the laboratory into a repository. While no
direct disposal of spent nuclear fuel is envisioned, the law also requires that the government
perform research on direct spent nuclear fuel disposal options. Annual progress reports and the
3-14
-------
final report on the three technological areas are to be prepared by a National Evaluation
Commission composed of eminent scientists (JOR99).
The new law allowed the government to resume site selection efforts for underground research
laboratories. A waste "negotiator" or 'mediator" was appointed to discuss proposed
investigations with local and regional officials. About 30 localities subsequently expressed
interest in hosting a laboratory. In 1994, the government's Bureau of Geological and Mineral
Research (BRGM) investigated these regions and eliminated those with adverse geology. In
early 1994* the negotiator announced the selection of four new regions as candidates to host a
repository: (1) the southern region of the Vienne "departement,11 in west-central France; (2) the
area surrounding Marcoule in the Gard departement; (3) the Meuse departement, bordering
south/eastern Belgium; and (4) the northern Haute Marne departement, north of Dijon. Two
other localities were selected as secondary choices because their local governments had not voted
on their candidacy; the four primary localities all voted in favor of their candidacy.
The Meuse and the Haute Marne sites were subsequently merged and designated as the East site.
Since two years of geophysical examination and drilling revealed no prohibitive factors at any of
the sites, ANDRA proposed in 1997 to proceed with underground laboratories at each location.
Whereas ANDRA had previously considered four types of potential host rock, the selected
regions represent only two: clay and granite. The National Evaluation Commission supported
selection of the East site and the Gard site which are in clay but noted confinement problems
with the granitic host rock at the Vienne site (JOR99). On December 9,1998 the French
government authorized construction of underground laboratories at the East site in clay and at a
new site in a granitic formation to be located by ANDRA.
The 1991 law includes additional provisions designed to ease public concern about France's
high-level waste management program, including the creation of a policy of openness concerning
the country's high-level waste disposal program and a requirement that government grants and
jobs for the host municipality accompany the underground research laboratories.
3-15
-------
3.4.3 Regulatory Organizations and Their Regulations
Agencies with regulatory responsibilities include the Directorate for the Safety of Nuclear
Installations (DSIN) within the Ministry of Industry; the CEA and its subsidiary, the Institute for
Nuclear Protection and Safety (IPSN); the BRGM; and SON (architect and engineering services).
DSIN, France's principal nuclear regulatory authority, issued "Fundamental Safety Rule III.2.f."
(DSI91) pertinent to high-level and alpha waste disposal, on June 10, 1991. The rule requires
that:
• The impact of a deep geologic disposal facility on radiation exposures be as low
as reasonably achievable
Individual dose equivalent due to the facility be limited to 0.25 mSv (25 mrem)
per year for likely events
The stability of geologic barriers be demonstrated for at least 10,000 years
• High-level waste packages prevent the release of radioactive contents during the
period when short- and medium-lived radionuclides dominate total radioactivity
In preparation for the underground laboratory phase, the IPSN, within CEA, is independently
preparing facilities to evaluate the long-term safety requirements of a repository on behalf of the
regulatory authority DSIN.
3.5 GERMANY
3.5.1 Nuclear Power Utilization
In 1994, Germany met about 30 percent of its electrical needs through nuclear power (EIA95).
The German nuclear power program relies primarily on pressurized light water reactors (14
units) and boiling water reactors (7 units), although research and development activities and
demonstration projects are also evaluating alternate reactor designs (high temperature gas-cooled
reactors and liquid metal fast breeder reactors) for power production. It is estimated that by the
year 2000, Germany will have generated about 9,000 MTHM of spent nuclear fuel. Germany
has historically planned to dispose of spent nuclear fuel in deep geological formations only after
3-16
-------
reprocessing, as stipulated in a 1976 amendment to Germany's Atomic Energy Law. Plans for a
domestic reprocessing facility were abandoned in 1989 and German utilities chose instead to ship
their spent nuclear fuel to France and Britain for reprocessing. Resulting vitrified waste is
currently returned to Germany and stored in metal casks for planned subsequent disposal. A
1994 amendment to its Atomic Energy Law, however, legalized the direct disposal of spent
nuclear fuel elements as well (Atomic Energy Law, Article 4, amendment of section 9a(l))
(GER94). Since then, German utilities have been considering both management options.
3.5.2 Disposal Programs and Management Organizations
The German government's Institute for Radiation Protection (BfS) is responsible for the design,
construction and operation of waste disposal facilities. Vitrified high-level waste returned from
foreign reprocessors was targeted for disposal at the Gorleben facility, a salt dome located in
Lower Saxony, if the site proves acceptable. Spent nuclear fuel would also be directly disposed
at Gorleben. The newly legalized option of direct disposal has required modification of plans at
the facility. Until a repository is in operation, vitrified waste will be stored at Gorleben and the
Ahaus facility in Northrhine-Westphalia. Expansions at both the Gorleben and Ahaus storage
facilities are currently proposed. A former salt mine at Asse, which served until 1978 as a
repository for low-level (125,000 containers) and intermediate-level (1,300 drums) radioactive
wastes, now serves as an underground research laboratory for high-level waste disposal.
The Gorleben salt dome ranges in depth from 250 meters to 3000 meters. Construction of an
underground research laboratory was initiated in 1986, but all work was stopped for over a year
in 1987 because of a construction fatality. As of 1995, two shafts had been sunk to depths of 600
and 620 meters (emplacement at approximately 870 meters is anticipated (LOM95)). Current
areas of emphasis include hydrogeological investigation and seismic measurements (OEC95b).
Construction of the repository could start at the turn of the century, and the facility is scheduled
to remain operational for about 50 years. It is anticipated that the site will receive about 550
MTHM of vitrified waste, 200 metric tons of directly disposed spent nuclear fuel, and about
6,690 containers of low-level and intermediate-level waste per year (LOM95).
Repository design emphasizes the role of the surrounding geology as a barrier. It is anticipated
that the salt dome's formations will creep over tune in response to radiogenic heating and
pressure gradients to encapsulate the waste. The use of steel and iron canisters is intended
3-17
-------
primarily to contain waste in the short-term. The possibility of direct spent nuclear fuel disposal
has required additional research and design development.
Federal agencies have been generally positive about the suitability of Gorleben as a repository
site. However, while research appears to generally support the suitability of the Gorleben site,
the project has faced increasing opposition from the government of Lower Saxony. As a
precautionary measure, in 1995 the Federal Institute for Geosciences and Natural Resources
prepared two reports identifying other potential sites for a HLW repository in the event that the
Gorleben site is found unsuitable (OEC96).
Subsequently, Federal elections in 1998 dramatically altered the direction of German nuclear
program. A coalition agreement signed by the Greens and the Social Democrats on October
20,1998 stated that use of nuclear energy would be phased out (BRE99). With regard to waste
management, the agreement noted that a single repository in a deep geologic formation is
sufficient for all types of radioactive waste and the time-dependent target for HLW disposal is
2030. The agreement further stated that the suitability of the Gorleben salt dome is questionable
and work there should be interrupted while potential sites in other host rocks are examined.
Emplacement of radioactive wastes at Morsleben (in former East Germany) is to be terminated
and the site decommissioned..
3.5.3 Regulatory Organizations and Their Regulations
Key German agencies include the Federal Ministry for Environment, Protection of Nature and
Reactor Safety (BMU), the Federal Ministry for Research and Technology (BMFT), the BfS, the
Federal Institute for Geosciences and National Resources, and the host state's ministry for
environmental protection. As the primary federal supervisory authority, BMU receives advice
from two committees of independent experts, the Reactor Safety Commission (RSK) and the
Committee on Radiological Protection (SSK).
In Germany, the institutional and legal framework for the regulation of nuclear facilities is based
on the joint participation of Federal and state governments. State governments serve as licensing
authorities for all nuclear waste facilities, although the Federal government has the authority to
override these decisions. The Federal government retains primary responsibility for waste
disposal; the Atomic Energy Law and the Radiation Protection Ordinance (GER94) establish the
3-18
-------
principles and requirements regarding the safe utilization and application of atomic energy and
radioactive materials, including the disposal of radioactive waste. Under the Radiation
Protection Ordinance, dosage limits are set at 0.3 mSv (30 mrem) per year for "all reasonable
scenarios" (OEC95a).
German regulators had been developing safety regulations that the Gorleben facility would be
required to meet through a site-specific safety assessment. It is expected that this safety
assessment will be required to demonstrate that potential exposure to radiation from disposed
waste will be kept within the range of natural radiation for a period of about 10,000 years and
that integrity of the repository system will be maintained over a longer period of time (GAO94).
3.6 JAPAN
3.6.1 Nuclear Power Utilization
In 1994, Japan produced about 31 percent of its electrical needs through nuclear power provided
by 49 reactors (EIA95). It is anticipated that this figure will increase to approximately
33 percent by the year 2000 and to about 42 percent by the year 2010 (AEC94). The Japanese
nuclear power program currently relies primarily upon light water reactors, although research and
development activities and demonstration projects are also evaluating alternate reactor designs
(gas cooled reactor, heavy-water moderated reactor, and liquid-metal fast-breeder reactor). By
1994, the country's first prototype fast breeder reactor, Monju, had reached criticality (OEC95b),
but a December, 1995 coolant leak dealt a setback to the project.
Japan's spent nuclear fuel is currently reprocessed hi France and England. However, both
countries have exercised their option to return vitrified residue to Japan; the first return delivery
from France took place in February, 1995. Domestically, the Power Reactor and Nuclear Fuel
Development Corporation (PNC) has operated a small reprocessing plant since 1977, where
roughly 720 tons of spent nuclear fuel had been reprocessed as of 1993. Furthermore, at the
Rokkasho site in Aomori prefecture, a private utility consortium, Japan Nuclear Fuel Services
Limited (JNFL), plans to begin operating a large commercial-scale plant shortly after the year
2000 (AEC94). It is estimated that by the year 2000, Japan will have discharged about 20,000
MTHM of spent nuclear fuel from its reactors. Vitrified high-level waste will be stored 30 to 50
years for cooling before disposal hi a geologic repository.
3-19
-------
3.6.2 Disposal Programs and Management Organizations
As noted above, Japan's current waste management strategy includes spent nuclear fuel
reprocessing using domestic and foreign facilities, on-site spent nuclear fuel storage, waste
solidification followed by long-term storage, and eventual disposal in a suitable deep geological
formation. Japanese nuclear utilities are responsible for storing high-level waste and funding its
disposal; JNFL is responsible for low-level waste disposal activities at Rokkasho (OEC95c).
Two government-sponsored organizations—PNC10 and the Japan Atomic Energy Research
Institute (JAERI)—are responsible for research and development addressing the fuel cycle, waste
management, and disposal. In 1993, the Steering Committee on High-Level Radioactive Waste
(SHP) was created to spearhead planning for disposal of the country's high-level waste.
Radioactive waste is managed in accordance with Japan's Long Term Program for the
Development and Utilization of Nuclear Energy (AEC94). In 1994, the Atomic Energy
Commission (AEC) issued an update to the long-term disposal plan, placing particular emphasis
on the disposal of high-level waste and adding new details to the country's plans and timetables
for this effort. The 1994 update established a procedure for implementation of a deep geologic
repository and provided guidelines on storage, vitrification, and geologic disposal. The plan also
added clarity to the roles of Japan's nuclear-related organizations, which can be summarized as
follows:
is the lead organization implementing the research and development program in
various areas of the fuel cycle and geologic disposal, while JAERI performs
research in support of the government's safety evaluation of geological disposal,
as well as research on advanced waste management technologies.
• Utilities and their Consortia: Utilities are responsible for funding high-level waste
disposal programs and for contributing to related research and development work.
• Government Agencies: Government agencies are responsible for oversight and
overall coordination of disposal. While it has not yet been decided what entity
will implement or license the disposal project, the AEC's 1994 update of the
country's long-term disposal plan suggests that this duty will be delegated by the
10 On October 1, 1998 PNC was succeeded by the Japan Nuclear Cycle Development Institute (JNC).
3-20
-------
year 2000 and that SHP is responsible for studying the matter (SHP has not been
designated the implementing entity).
The 1994 plan also laid out a five-step process to develop a high-level waste repository. The first
phase, selection of effective formations, was completed in 1984. Subsequent steps as established
by the AEC include: 1) establishment of an implementing organization by the year 2000; 2)
selection of candidate disposal sites, subject to government cooperation and community
acceptance; 3) demonstration of disposal technology at the candidate site, followed by license
application; and 4) establishment of necessary laws and policies for the disposal implementation
and safety. The plan called for the repository to be operational by 2030 to 2045. Japanese
authorities have determined that high-level waste disposal should be possible in any geologic
formation excluding unconsolidated media (e.g., soil and sand). Because of geological
heterogeneities in Japan, geological characterization is expected to be difficult, causing
uncertainties in predicting the performance of natural barriers. Thus, Japan is assigning a major
role to the engineered barrier system, while defining a small number of critical natural
characteristics for the site which are expected to be achievable in various geological settings.
In April 1997 the AEC issued "Guidelines on Research and Development Relating to Geological
Disposal of High-Level Radioactive Waste hi Japan." The Guidelines describe the level of
technical reliability which must be demonstrated for a geologic repository. The Guidelines also
identify issues which must be addressed to establish the technical basis for selection of potential
disposal sites and for the formulation of safety standards. R&D activities to be conducted after
2000 are also identified (MAS99).
Considerable research and development has been underway in recent years. Most research is
conducted by PNC (now JNC) and regular plans are submitted to AEC; the most recent progress
report (designated H3) was submitted in 1992, and a subsequent report (tentatively designated
H12) is expected before the year 2000 (OEC95b)(MAS99). The H12 report (a draft of which
was issued in 1998) is expected to more rigorously address the feasibility of the specified
disposal concept than did the H3 Report. The HI2 report is also expected to provide input for
regulatory and siting processes which will be set in motion after 2000 (MAS99).
PNC operates an underground test facility in both sedimentary and crystalline rock
environments. The test facility is located in the Tono Uranium Mine in central Japan. Major
experiments in the mine include ground water flow investigation; studies on the effects of
excavation on the mechanical and hydraulic behavior of the'repository; natural analogue studies
3-21
-------
and evaluations of the chemical durability of simulated waste glasses; and the corrosion rates of
candidate overpack materials. Since 1988, PNC has also conducted major tests in the Kamaishi
iron ore mine in northern Honshu. Work at this mine is currently guided by a 5-year research
plan, submitted by PNC in 1993 and characterized by work in a deeper gallery. Major
investigations at Kamaishi have included detailed fracture mapping, cross-hole hydraulic and
geophysical testing, drift excavation-effect studies, in-situ stress measurements, single-fracture
flow tests, and observations of seismic activity. In December 1995, PNC signed an agreement
with local governments to build an underground research laboratory near the city of Mizunami.
The laboratory will be used to research ground-water and rock mass characteristics for geologic
disposal (OEC96).
Several new facilities were started in recent years, including: (1) the Tokai Vitrification Facility
(the first vitrification facility in the country, where operation began in 1995); (2) the Nuclear
Fuel Cycle Engineering Facility (NUCEF), where construction was completed in 1995; (3) the
Recycling Equipment Testing Facility (RETF), begun in 1995 to develop reprocessing
techniques for spent nuclear fuel from fast breeder reactors (OEC95b); and (4) the ENTRY
facility at the Tokai Research Lab to conduct full-scale engineering tests and non-radioactive
simulations of the performance of natural and engineered barriers.
3.6.3 Regulatory Organizations and Their Regulations
The Atomic Energy Basic Law of 1955 established the AEC and the principles and requirements
for the safe utilization and application of atomic energy and radioactive materials, including the
disposal of radioactive waste. In addition to the AEC, other key agencies or organizations
include the Nuclear Safety Commission (NSC), the Ministry of International Trade and Industry
(MITI), and the Science and Technology Agency (STA). Regulatory requirements for the high-
level waste repository have not yet been established, nor have formal individual dose limits been
issued.
3.7 SPAIN
3.7.1 Nuclear Power Utilization
As of December 31,1990, Spain had a total of nine light water nuclear power plants in operation.
A tenth reactor, the graphite-gas Vandellos plant, was expected to begin decommissioning in
3-22
-------
1996. Spent fuel from the light water reactors is stored on site in pools specifically designed for
this purpose. Spent fuel from the Vandellos 1 plant was sent to France for reprocessing. Spam's
radioactive waste is currently managed by the Empresa Nacional de Residues Radioactivos, S.A.
(The Spanish National Radioactive Waste Company - ENRESA) which was established by
Royal Decree 1522 on July 4,1984. Eighty percent of the company is owned by the Spanish
Centre for Energy, Environmental and Technological Research (CIEMAT) and 20 percent is
owned by the National Institute for Industry (INI) (SMI91).
Two types of high-level wastes will have to be managed in Spain: spent fuel from light water
nuclear power plants and the vitrified wastes from reprocessed fuel from the Vandellos 1 plant.
At the end of 1990, 974 MTHM of spent fuel were stored at the sites of Spain's nine nuclear
power plants. It is projected that approximately 5,200 MTHM of spent fuel will ultimately be
managed in Spain. In addition, 180 m3 of vitrified wastes from the Vandellos 1 plant will require
final disposal (SMI91).
3.7.2 Disposal Programs and Management Organizations
ENRESA manages all radioactive wastes in Spain, including low-level, intermediate level, and
high-level wastes. The high-level waste program in Spain considers both intermediate storage
and final disposal of these wastes. Intermediate storage, which allows the radioactive elements
of the waste to cool down and decay, includes both dry storage hi casks and vaults and liquid
storage in pools. Waste has been stored on-site, although Spain has contemplated a Temporary
Centralized Storage option (SMI91).
The goals and objectives of Spain's program to permanently dispose of high-level waste and
spent fuel were outlined in its Third General Radioactive Waste Plan. This strategy, which was
defined and initiated in 1987, included three areas of work:
• Search for a site for facility construction. Spam considered granites, salts, and
clay media.
• Acquisition of technology and training of teams required for characterization of
the chosen site and construction of the disposal facility.
Development of the basic design for a deep geological disposal facility (SMI91).
3-23
-------
Progress is being made in terms of defining the conceptual design of a geological repository in
Spain. The development of a preliminary conceptual design for granite and clay was completed
in 1992 and for clay in 1994. A non-site specific conceptual design for salt has also been
completed. More recently, a probabilistic performance assessment in a generic granite formation
was done in 1997 and a similar study in a generic clay formation was done in 1998 (SAN99).
Spain has also adopted its Third Research and Development Plan which covers the period from
1995 through 1999. This plan, which includes all types of radioactive wastes, has as its main
objective the support required for the performance of the high-level waste program. The Plan
primarily emphasizes verification of site characterization methodologies and preliminary
repository designs, application of numerical models, and acquisition of specific data for the
performance of long-term safety assessments (OEC96).
Public pressure caused a cessation of field work in 1996 and, in response, the Spanish Senate
created an Inquiry Committee to provide recommendations to the government on how to develop
a radioactive waste management policy (SAN99). The Committee noted that, while deep
geological disposal is the basis for most international programs, there is growing interest in
partitioning and transmutation of long-lived radionuclides. They stated that decisions needed to
have a high and broad level of socio-political consensus which could be obtained only through
involvement of a wide range of institutions and administrations. Subsequently, the Spanish
government provided further guidance on waste management policy hi early 1998 which
included the following:
• No decision on disposal of high-level waste will be made before 2010
No further siting activities will be undertaken before 2010 (After that date siting
activities must be voluntary.)
• Deep disposal will continue to be studied but other technologies such as
partitioning and transmutation should also be analyzed
ENRESA is modifying its strategy to be congruent with the government guidance and the
changes will be reflected a new R&D Plan (1999-2003).
3-24
-------
3.7.3 Regulatory Organizations and Their Regulations
The Spanish Nuclear Safety Council (Consejo de Seguridad Nuclear) has officially adopted a
level of individual risk below 10"6 per year for the long term disposal of radioactive wastes. This
risk value equates to a dose to individuals in the critical group of less than 0.1 mSv/year (10
mrem/year) (SMI91).
3.8 SWEDEN
3.8.1 Nuclear Power Utilization
Following a 1980 national referendum, the Swedish Parliament decided to phase out nuclear
power plants by the year 2010. Although the Swedish government maintains this commitment,
the country remains dependent on nuclear fuel for approximately 51 percent of its electrical
power needs (as of 1994). Sweden's nuclear power is produced with nine boiling water reactors
and three pressurized water reactors (EIA95).
By 2010, Sweden will have produced nearly 8,000 MTHM of spent nuclear fuel. In the 1970s,
Swedish utilities had entered into agreements with other countries to reprocess foreign sources of
spent nuclear fuel; however, this approach was abandoned following the 1980 referendum and
the utilities have since sold their reprocessing contracts or traded high-level waste from
reprocessing for other spent nuclear fuel. A joint utility consortium, the Swedish Nuclear Fuel
and Waste Management Company (8KB), manages the disposal of radioactive waste. In 1985,
8KB began operating a centralized spent nuclear fuel storage facility (CLAB) that will
eventually hold all of Sweden's spent nuclear fuel for about 40 years. As of 1994, this facility
was filled to about 45 percent capacity (SKB94). The facility is situated in an underground
granite cavern at a depth of 30 meters, near an existing nuclear power plant (Oskarshamn). In
1998 the Swedish government authorized expansion of CLAB and the work is expected to be
completed in 2004.
3.8.2 Disposal Programs and Management Organizations
Nuclear waste management activities in Sweden are guided by the Act Concerning Nuclear
Activities and the Act Concerning the Management of Natural Resources. Every three years,
3-25
-------
SKB is required to provide Swedish regulators with a research and development plan for
activities related to the management and disposal of the country's radioactive waste. The 1992
plan was approved contingent upon additional details regarding deposition and canister design
(SKB94).
As outlined in the 1993 plan, Sweden's reference disposal concept for spent nuclear fuel is to
encapsulate it in high-integrity copper canisters and emplace the canisters in a repository built in
crystalline rock at a depth of about 500 meters, backfilling the deposition holes with highly-
compacted bentonite and the tunnels and shafts with a mixture of sand and bentonite. SKB is
evaluating alternative concepts such as deep boreholes and tunnel emplacement, as well as
alternative canister designs. Canisters are expected to consist of a steel insert (for mechanical
protection) inside of a copper sleeve (for corrosion protection).
SKB's 1995 R&D Programme (SKB95) set forth the following schedule for establishing a deep
geologic repository:
SKB is currently conducting feasibility studies, planned at a total of five to ten
municipalities
Feasibility studies are expected to be completed by 1997, after which time two
municipalities, and locations within both, will be selected for site investigation
• One site will be selected around 2001, and deposition of encapsulated fuel is
planned for 2008, when a small portion (approximately 800 tons) of Sweden's
nuclear fuel will be deposited
Feasibility studies were completed in Storumann in May, 1995, but a local referendum rejected
further research there. A feasibility study at a second municipality, Malaa, was completed in
March of 1996. The municipality of Malaa is conducting an independent review of the
feasibility study involving local stakeholders and independent experts, and the decision on
whether to organize a local referendum on further research will be based on the results of this
review (SKB96). Other sites where feasibility studies are underway or being considered include
four of the five municipalities with existing nuclear facilities (Varberg, Oskarshamn, Nykoping,
and Osthammer). The fifth site with an existing nuclear facility, Kavlinge, is currently not
considered a candidate.
3-26
-------
However, site selection has not progressed as rapidly as envisioned in the 1995 R&D
Programme. The 1998 RD&D Programme calls for feasibility studies at five sites to be
completed and two sites to be selected by 8KB for site investigations by 2001 (HED99).
The OECD/NEA conducted an international research project in an underground research
laboratory at Sweden's Stripa mine from 1980 to 1991 (OEC95b). 8KB has recently completed
construction of a second laboratory under the island of Aspo,2 km north of Oskarshamn, at a
depth of 450 meters. The Aspo' site will be used to test methods of site selection and
characterization, and to research disposal technologies for later use in Sweden's deep geologic
repository,
3-8.3 Regulatory Organizations and Their Regulations
Key government entities with direct responsibilities hi waste management include the Swedish
Nuclear Power Inspectorate (SKI), the National Institute for Radiation Protection (SSI), and the
Swedish Consultative Committee for Nuclear Waste Management (KASAM). All operate under
the supervision of the Ministry of the Environment and Natural Resources. The National Board
for Spent Nuclear Fuel (SKN), a former public entity with regulatory responsibilities, was
absorbed into SKI in the early 1990s.
As of 1995, safety requirements for management and disposal of high-level waste were in
development. These requirements are the responsibility of SSI, hi cooperation with SKI. Both
SSI and SKI favor a total systems approach, without specifying detailed sub-system quantitative
criteria in early phases of repository development. Proposed guidelines for the deep repository
would require that:
• Radiation doses to individuals be limited to 0.1 mSv/yr for a reasonably
predictable period of time (one million years), after which radionuclide fluxes are
to be limited to a level corresponding to naturally occurring fluxes of
radionuclides
• A passive multi-barrier approach be used
• Future safety of the facility requires no further controls after the facility is sealed
3-27
-------
• The repository be designed to not restrict future attempts to change the repository
or retrieve the waste (SKB95)
Established general principles for the management of nuclear waste state that:
• Radiation protection take into consideration issues of biodiversity and natural
resource use in addition to human health
• Radiation protection be independent of whether doses arise today or in the future,
or whether they originate within or outside the country
• The disposal of nuclear waste pose a risk no greater than that of other portions of
the nuclear fuel cycle
• All activities must be justified, protection must be optimized, and the individual
must be protected by dose limits (SKB95 j
3.9 SWITZERLAND
3.9.1 Nuclear Power Utilization
In 1994, Switzerland's five nuclear power plants supplied about 37 percent of the country's
electrical power needs (EIA95). The Swiss nuclear power program relies on a mix of pressurized
and boiling light water reactors (three PWRs and two BWRs). Although there is currently a
moratorium on construction of new nuclear plants, capacity increases at existing plants have kept
supply high, and a 10 percent increase by the year 2000 is planned (OEC95c),
The Swiss estimate that, by the year 2000, they will have produced about 1,800 MTHM of spent
nuclear fuel. Switzerland currently ships its spent nuclear fuel to France and Britain for
reprocessing but maintains the options of spent nuclear fuel management both with and without
reprocessing in the future.
3.9.2 Disposal Programg and Management Organizations
The responsibility for establishing radioactive waste disposal facilities in Switzerland lies with
the National Cooperative for the Storage of Radioactive Waste (NAGRA), a joint government
and utility cooperative agency. NAGRA was established in 1972 to manage the disposal of
3-28
-------
radioactive wastes, including spent nuclear fuel, high-level waste and other reprocessed waste
returned from the French and British reprocessing facilities. Waste conditioning and interim
storage of reprocessed waste, high-level waste and spent nuclear fuel is the responsibility of
ZWILAG, a cooperative comprised of nuclear utility operators.
Overall nuclear policy is governed by the Swiss Atomic Law, to which two major changes were
proposed in 1994. These proposed revisions were subsequently dropped by the parliament in
favor of drafting an entirely new national- nuclear energy law. A draft of this law is in
development (OEC96). The overall goal of the Swiss program is to establish the viability of a
repository in Switzerland by the year 2000, although commissioning of a repository will not
occur before 2020 to allow a 40-year spent nuclear fuel/high-level waste cooling period.
Participation in any international repository projects that may develop is also under
consideration.
The Swiss have historically considered two rock types, crystalline rock and sedimentary rock, as
potential host media for a high-level waste repository. In 1984, NAGRA launched studies in
crystalline rock by drilling seven deep boreholes into the crystalline basement of northern
Switzerland and, subsequently, conducted geological and safety assessment studies. In 1994,
NAGRA released a synthesis of this research (Kristallin I), expressing optimism about the use of
crystalline rock as a host rock; specifically, NAGRA is considering crystalline rock formations in
northern Switzerland as viable sites for a repository and is planning additional field work there
(OEC95b). In support of crystalline rock studies, a new three-year Phase IV study was launched
at the Grimsel Rock Laboratory in 1994. The Swiss have also considered two sedimentary rock
types, Opalinus clay and freshwater molasse, and have conducted field research on both
formations. In January 1994, the Safety Inspectorate, NAGRA, and representatives of the
relevant government agencies identified Opalinus clay as the preferred sedimentary host rock
option (OEC95b). A site in Benten, just north of Zurich, has been identified for seismic survey
and the construction of an 800-meter borehole to further examine the feasibility of a repository in
clay.
By the year 2000, NAGRA must submit a program—the Siting Feasibility Project—for
government approval that demonstrates the feasibility of siting a repository in one or more of the
crystalline or sedimentary media under consideration. NAGRA intends to select which medium
or media by 1997.
3-29
-------
3.9.3 Regulatory Organizations and Their Regulations
Key organizations or agencies with direct regulatory responsibilities in waste management
include: the Nuclear Safety Division (HSK) of the Federal Energy Office (BEW) within the
Federal Department of Transport, Communications, and Energy (EVED); the Federal
Commission for the Safety of Nuclear Installations (KSA); the Federal Department of Interior
(EDI); and the Institute for Reactor Research (EIR). An interagency working group (AGNEB)
was also established to coordinate activities in support of government decisions on the licensing
of nuclear waste facilities.
In November 1993, HSK released the current guidelines for management of nuclear waste in the
country, entitled Radiation Protection for the Disposal of Nuclear Waste (HSK93). Dosage is
limited to 0.10 mSv (10 mrem) per year for reasonably probable scenarios, and annual risk is
limited to 10"6 for unlikely scenarios. Candidate repositories must produce a system capable of
meeting these requirements in order for their application to be considered (GAO94); furthermore,
all repositories must be designed to be sealed at any time within a few years, after which it must
be possible to dispense with institutional controls.
3.10 UNITED KINGDOM
3.10.1 Nuclear Power Utilization
In 1994, the United Kingdom (UK) met about 26 percent of its electrical needs through nuclear
power (EIA95). During the 1960s and 1970s, the UK depended primarily on a series of Magnox
(magnesium-clad, uranium metal-fueled) reactors (20 units in operation as of 1995), but began to
use advanced gas-cooled reactors (AGRs) during the 1970s and 1980s (14 units in operation as
of 1995). One pressurized water reactor (PWR) was commissioned in 1995, Use of a fast
reactor was explored as well, but a prototype facility in Dounreay was closed in 1994.
British Nuclear Fuels, pic. (BNFL), a government-owned corporation, reprocesses spent nuclear
fuel at its Sellafield facility on behalf of both domestic and foreign utilities. Spent metallic fuel
from the country's Magnox reactors is reprocessed at the Sellafield facility at a rate of
approximately 400 cubic meters annually (NIR94). In March 1994, BNFL began operating the
Thermal Oxide Reprocessing Plant (THORP) at Sellafield to reprocess spent nuclear fuel
3-30
-------
produced by the country's AGR and PWR reactors and by international customers. THORP is
the country's first commercial-scale reprocessing plant for oxide fuels.
In 1994, the Board of Trade and the Secretary of State for the Environment placed portions of the
British nuclear program under review. In May 1995, at the conclusion of the review, it was
announced that the country's comparatively modern facilities (7 AGR stations and the Sizewell
PWR) and all future facilities were to be privatized (nuclear power had been excluded from the
1990 privatization of the electric utility industry). Under privatization, all AGR and PWR
stations have been grouped under a new company called British Energy pic. Magnox stations
have been transferred to a new company called Magnox Electric pic (Magco), a government-
owned company responsible for operation of and liabilities resulting from these stations. It is
expected that Magco will ultimately become a subsidiary of BNFL.
Since 1952, over 30,000 MTHM of metal fuel from the Magnox reactors have been reprocessed
in the UK. It is estimated that by the year 2000, Britain will have about 4,000 cubic meters of
high-level waste destined for storage or disposal due to the reprocessing of some 60,000 metric
tons of spent nuclear fuel. High-level waste is currently stored in an air-cooled facility at
Sellafield.
3.10.2 Disposal Prograjns and Management Organizations
The responsibility for managing the storage and disposal of radioactive waste lies with its
producers. BNFL has the lead responsibility for management of high-level waste from
reprocessing. In 1990, BNFL began operating a vitrification plant at Sellafield. In 1982, the
government established the Nuclear Industry Radioactive Waste Executive (NIREX) to develop
and operate intermediate- and low-level radioactive waste disposal facilities. NIREX was
originally established as a partnership consisting of private firms and governmental agencies. In
1985, NIREX was restructured as an independent legal entity, UK NIREX Ltd.
Historically, the UK's radioactive waste disposal strategy has postponed the development of a
high-level waste disposal facility, considering deep disposal of low- and intermediate-level
wastes a higher priority. NIREX had been researching a potential disposal site near Sellafield
and plans suggested that a repository for low- and intermediate-level wastes could be operational
there by the year 2010. A recent governmental review, however, has comprehensively rejected a
3-31
-------
proposed environmental impact study by NIREX that was necessary to proceed with construction
of an underground research laboratory at Sellafield. This rejection has caused NIREX to refocus
its program onto more generic issues while continuing to condition and package intermediate-
level wastes for eventual disposal (HOL99).
Eventual deep disposal is planned for high-level waste. Current plans call for continued
reprocessing of spent nuclear fuel, solidification of high-level waste, and surface storage for
about 50 years. Under this schedule, the need for a high-level waste repository is not expected
before the year 2040. Vitrified high-level waste would then be disposed in deep geologic media.
The UK Department of Environment, Transport and the Regions instituted, in 1997, a two-year
review of options for management and eventual disposal of spent fuel. The study, which is
nearing completion, is designed to formulate a program for development of a deep repository for
HLW and to define the key elements of the required R&D work.
The UK has also adopted a policy of monitoring the results of research activities being conducted
by other countries. Depending on the outcome of research being conducted abroad, Britain
would then develop a high-level waste disposal and repository strategy using concepts that best
fit British needs.
3.10.3 Regulatory Organizations and Their Regulations
The Atomic Energy Act of 1946 establishes the authority and responsibility to control and
regulate the development of nuclear power in Britain. The Act has been amended several times
to establish new requirements, including those addressing the management and disposal of
radioactive waste.
The regulatory functions are performed by the Nuclear Installations Inspectorate, which is part of
the Health and Safety Executive; the Radiochemical Inspectorate of the Department of the
Environment; the Ministry of Agriculture, Fisheries, and Food; the UK Atomic Energy
Authority; and the Secretaries of State of Scotland and Wales. The government also takes advice
from several independent experts and advisory committees, including the Radioactive Waste
Management Advisory Committee and the National Radiological Review Board.
3-32
-------
Exposure limits are based on recommendations of the National Radiological Protection Board.
While no current limits pertaining to exposure from spent nuclear fuel and high-level waste have
been set, indications are that they will be similar to those set for low- and intermediate-level
wastes (10"6 per year for individual risk from a single facility (OEC95a)). The British waste
management philosophy favors the use of broad safety goals over prescriptive regulatory
approaches, placing the burden of compliance upon the operator.
3-33
-------
REFERENCES
AEC94 Atomic Energy Commission (Japan), Long-Term Program for Research,
Development, and Utilization of Nuclear Energy, 1994.
AECB87 Atomic Energy Control Board (Canada), AECB Regulatory Document R-l 04,
June 1987.
BRE99 Brennecke, P.W. et al., Realization of the German Repository Concept - Current
Status and Future Prospects, WM '99 Conference, Tucson AZ, February 28-
March4, 1999.
DSI91 Directorate for the Safety of Nuclear Installations (France), Rule No. III.2.f., June
10,1991.
EAP98 Environmental Assessment Panel, Report of the Nuclear Fuel Waste Management
and Disposal Concept Environmental Assessment Panel, Canadian Environmental
Assessment Agency, February 1998.
EIA95 Energy Information Administration, World Nuclear Outlook, 1995, DOE/EIA-
0436(95), October 1995,
GAO94 General Accounting Office, Nuclear Waste: Foreign Countries'Approaches to
High-level Waste Storage and Disposal, GAO/RCED-94-172, August 1994.
GER94 Article 4, Section 9a(l) of German Atomic Energy Law, amended to the law on
June 19,1994.
HED99 Hedman, T., Management of Spent Nuclear Fuel in Sweden, WM '99 Conference,
Tucson AZ, February 28-March 4, 1999.
HSK93 Nuclear Safety Division (Switzerland), Radiation Protection for the Disposal of
Nuclear Waste, Regulatory Document R-21, 1993.
JOR99 Jorda, M. and I. Forest, The Radioactive Waste Management Program in France,
WM '99 Conference, Tucson AZ, February 28 - March 4, 1999.
HOL99 Holmes, John and John Mathieson, Recent Developments in the United Kingdom
Programme for the Deep Disposal of Radioactive Wastes, WM '99 Conference,
Tucson AZ, February 28-March 4, 1999.
3-34
-------
KIN99 King, Frank, Recent Developments in Nuclear Waste Management in Canada,
WM '99 Conference, Tucson AZ, February 28-March 4, 1999.
LOM95 Lommerzheim, A., Repository Planning for the Gorleben Working Model,
Proceedings of the Fifth Annual Conference on Radioactive Waste Management
and Environmental Remediation, Volume 2, New York: American Society of
Mechanical Engineers, 1995.
MAS99 Masuda, S. et al., Key Aspects of the Second Progress in the Japanese R&D
Programme for HLW Disposal, WM 499 Conference. February 28-March 4,1999.
NIR94 United Kingdom Nirex Limited, Going Underground: An International
Perspective on Radioactive Waste Management and Disposal, 2nd Edition,
September 1994.
NWT94 United States Nuclear Waste Technical Review Board, Report to the U.S.
Congress and the Secretary of Energy: January to December 1993, May 1994.
NWT95 United States Nuclear Waste Technical Review Board, Report to the US Congress
and Secretary of Energy: 1994 Findings and Recommendations, March 1995.
OEC93 Organization for Economic Cooperation and Development/Nuclear Energy
Agency, Nuclear Waste Bulletin: Update on Waste Management Policies and
Programs, No. 8, July 1993.
OEC95a Organization for Economic Cooperation and Development/Nuclear Energy
Agency, Future Human Action at Disposal Sites, OECD/NEA Document 66-94-
041,1995.
OEC95b Organization for Economic Cooperation and Development/Nuclear Energy
Agency, Nuclear Waste Bulletin: Update on Waste Management Policies and
Programs, No. 10, June 1995.
OEC95c Organization for Economic Cooperation and Development/Nuclear Energy
Agency, Nuclear Energy Programs of OECD/NEA Member Countries, 1995.
OEC96 Organization for Economic Cooperation and Development/Nuclear Energy
Agency, Nuclear Waste Bulletin: Update on Waste Management Programs,
No. 11, June 1996.
SAN99 Santiago, J. and J. Astudillo, Overview of the Spanish R&D Program for High-
level Waste Disposal, WM '99 Conference, Tucson AZ, February 28-March 4,
1999.
3-35
-------
SKB94 Swedish Nuclear Fuel and Waste Management Company, Activities 1994, 1994.
SKB95 Swedish Nuclear Fuel and Waste Management Company, RD&D Programme 95:
Treatment and Final Disposal of Nuclear Waste, September 1995.
SKB96 Swedish Nuclear Fuel and Waste Management Company, SKB Annual Report
1995, May 1996.
SMI91 Spanish Minesterio de Industria, Comercio Y Turismo, Third General
Radioactive Waste Plan, July 1991.
VTT93 Technical Research Centre of Finland, Final Report of the Project, Performance
Assessment and Economic Evaluation of Nuclear Waste Management, 1993.
YJT92 Nuclear Waste Commission of Finnish Power Companies, TVO-92. Safety
Analysis of Spent Fuel Disposal, December 1992.
3-36
-------
CHAPTER 4
U.S. PROGRAMS FOR THE MANAGEMENT AND DISPOSAL OF
SPENT NUCLEAR FUEL AND HIGH-LEVEL RADIOACTIVE
WASTE AND THE EVALUATION OF YUCCA MOUNTAIN
4.1 INTRODUCTION
The Department of Energy (DOE), the U.S. Nuclear Regulatory Commission (NRC), and the
U.S. Environmental Protection Agency (EPA) each have legislatively defined roles in
management and disposal of spent nuclear fuel and high-level radioactive wastes (HLW) at the
proposed Yucca Mountain disposal site. As stated in the Nuclear Waste Policy Act of 1982
(NWP83), DOE is responsible for developing, constructing, and operating repositories for
disposal of these wastes. The NRC has responsibility to license the repository and related
facilities, and the EPA is to promulgate radiation protection standards which the NRC is to adopt
as basis for their licensing actions. The Nuclear Waste Policy Amendments Act of 1987
(NWP87) designated the Yucca Mountain site in Nevada as the only site to be evaluated by DOE
as a potential location for disposal of spent fuel and HLW. The Energy Policy Act of 1992
(EnPA92) directed EPA to promulgate site-specific radiation protection standards for the Yucca
Mountain site.
The legislative framework also prescribes roles for state governments, local governments, and
Indian tribes in the waste management and disposal program, and establishes the Nuclear Waste
Technical Review Board which provides oversight of the DOE program. This chapter presents
an overview of the responsibilities and program activities of the DOE, NRC, and these groups.
Responsibilities and activities of the EPA are described in Chapter 1 of this BID.
4.2 THE DEPARTMENT OF ENERGY
As noted above, DOE is responsible for the management and disposal of high-level radioactive
waste, which includes spent nuclear fuel and other waste generated by nuclear reactors and
reprocessing plants.8 Disposal of these wastes would occur at the Yucca Mountain site if it is
found suitable and approved for this function. Other radioactive waste categories defined by
DOE are transuranic (TRU) and low-level waste (LLW). TRU, consisting of material with
8 DOE typically separates spent nuclear fuel from other high-level waste by definition, although NRC
includes spent fuel as part of its high-level radioactive waste category.
4-1
-------
atomic numbers greater than 92, is generated as a result of defense production operations. DOE
began disposal of TRU wastes at the Waste Isolation Pilot Plant (WIPP), in New Mexico, in
1999. LLW is buried at DOE sites where it is generated or, if commercially generated, at sites
operated by private firms in several locations.
Fulfillment of its responsibility for radioactive waste management and disposal involves four
principal program activities in DOE: 1) receipt, transport, interim storage, and disposal of spent
nuclear fuel from commercial nuclear power operations, 2) management and disposal of DOE
spent nuclear fuel, which originates from DOE production and research operations and from
naval propulsion reactors, 3) solidification and disposal of high-level waste generated by
reprocessing operations for spent nuclear fuel from DOE's production reactors at Hanford and
Savannah River, and 4) storage and disposition of fissile materials from dismantled nuclear
weapons. Materials generated by the dismantling of weapons may be treated and disposed of like
high-level radioactive waste or they may be used as reactor fuel. In either case, such materials
will eventually become part of the disposal inventory.
In addition to commercial spent nuclear fuel and high-level waste from DOE's production
operations, other radioactive wastes that have been considered for disposal in a repository at
Yucca Mountain include DOE spent nuclear fuel, fissile materials from dismantled nuclear
weapons, and low-level radioactive wastes known as Greater-Than-Class-C (GTCC). The
radioactivity levels of wastes in this latter category exceed the NRC's limits for Class C wastes
as established in the 10 CFR Part 61 regulations. Decisions concerning disposition of these
radioactive materials have not been made. The NWPA limits the contents of a repository at
Yucca Mountain to "...70,000 metric tons of heavy metal or a quantity of solidified high-level
radioactive waste resulting from the reprocessing of such a quantity of spent nuclear fuel until
such time as a second repository is in operation" (NWPA, Section 114(d)). As detailed in
Chapter 7 of this BID, DOE currently plans that a repository at Yucca Mountain would contain
approximately 63,000 metric tons of spent commercial reactor fuel, and defense high-level
wastes, DOE spent fuel, and Navy spent fuel that are the equivalent of 7,000 metric tons of heavy
metal.
4.2.1 DOE'S Office of Civilian Radioactive Waste Management CQCRWMt
The DOE's Office of Civilian Radioactive Waste Management (OCRWM) was established by
Congress specifically to provide management and disposal of spent nuclear fuel from commercial
nuclear power reactors. Under a 1985 Presidential Executive Order, the repository established by
4-2
-------
OCRWM is also to be used for disposal of high-level waste from DOE operations. The
OCRWM charter includes responsibility for receipt of spent nuclear fuel from reactors at the
reactor sites, interim storage of spent nuclear fuel as necessary prior to disposal, transport of
spent nuclear fuel to the site(s) for interim storage and disposal, and siting, design, licensing, and
operation of a central interim storage facility. DOE has developed alternative designs for a
central interim storage facility (known historically as a Monitored Retrievable Storage (MRS)
facility), but, as of mid-1999, the Department has not established a site for such a facility.
Since passage of the Nuclear Waste Policy Amendments Act in 1987, OCRWM activities have
been focused on evaluating Yucca Mountain as the disposal site for spent nuclear fuel and high-
level waste. In accordance with the Site Characterization Plan (DOE88a), characterization of the
Yucca Mountain site is proceeding with surface-based and sub-surface activities. Recently, DOE
has focused on the "Viability Assessment" (VA), which is intended to allow a greatly unproved
appraisal of the prospects for geologic disposal at the Yucca Mountain site. The VA consists of:
• A reference engineered design for the repository and the waste package
A total system performance assessment describing the probable behavior of the
repository based on available data and the reference engineered design
A plan and cost estimate for completing a License Application (LA)
Cost estimates for constructing and operating the repository.
The VA was published hi December 1998. It was the basis for continued evolution of the
engineered design for the repository and for future data acquisition activities. Future program
documentation is planned to include an Environmental Impact Statement (EIS), planned to be
released in final form in August 2000; the Site Recommendation, planned to be submitted to the
President in July 2001; and the License Application (LA), to be submitted to the NRC in March
2002 if the site is approved for disposal. To date, principal program accomplishments include:
Completion of excavation of the north-south Exploratory Studies Facility (ESF)
tunnel at Yucca Mountain and the east-west Cross-Drift; both excavations have
been mapped and will be used as sources of in-situ data at the repository horizon
Initiation of various types of testing in alcoves and niches in the ESF and the
Cross-Drift
4-3
-------
• Development of a market-driven approach to storage and transportation of
commercial spent nuclear fuel
• Completion of the VA, followed by an analysis of enhanced design alternatives
aimed at resolving some of the issues identified in the VA, and selection of the
preferred engineered design for the Site Recommendation.
The OCRWM program has produced thousands of technical documents concerning its mission
and activities. Future technical documents are expected to support the Environmental Impact
Statement, the Site Recommendation, and the License Application if the site is approved for
disposal.
4.2.2 DOE Management and Disposal of Defense Wastes
The DOE's defense programs have produced significant amounts of high-level waste mat may
eventually be disposed of hi a repository at Yucca Mountain (see Chapter 5). Other wastes
produced by these defense programs (e.g., TRU waste) will be managed and disposed of
separately.
During the last 40 years, DOE and its predecessor agencies generated, transported, received,
stored, and reprocessed spent nuclear fuel at facilities throughout its nationwide complex. Spent
nuclear fuel was generated by nuclear weapons production reactors; U.S. Navy nuclear
propulsion program power reactors; government, university, and test reactors; special-case
commercial reactors; and research reactors. The DOE operated production reactors at the
Hanford and Savannah River Sites to provide special nuclear materials and other isotopes. These
production reactors are no longer operating. However, the Naval Nuclear Propulsion Program
and some test and research reactors are still in operation.
The DOE has reprocessed more than 100,000 metric tons of heavy metal (MTHM) of spent
nuclear fuel at the Idaho National Engineering and Environmental Laboratory (INEEL), the
Hanford Site, and the Savannah River Site to recover fissile material (uranium-235 and
plutonium-239) and other nuclides needed for national defense or research and development
programs. These reprocessing operations generated large quantities of high-level radioactive
waste. This waste exists as liquid, sludge, solids, and calcine and is stored primarily at its
reprocessing sites.
4.4
-------
In April 1992, the DOE began to phase out defense spent nuclear fuel reprocessing. As a result,
approximately 2,700 metric tons of unreprocessed spent nuclear fuel exist today in the DOE
inventory. This spent nuclear fuel is in a wide range of enrichments and physical conditions and
is stored at several locations throughout the United States. In addition to this inventory, the DOE
estimates that over the next 40 years it will generate another 100 MTHM from defueling DOE
and naval reactors.
From 1990 to 1995, the DOE made a number of decisions concerning the management of its
spent nuclear fuel inventories.9 In 1993, a Federal judge enjoined shipments of spent nuclear fuel
to the INEEL. In May 1995, the DOE issued a Record of Decision (ROD) concerning spent
nuclear fuel management at the INEEL (DOE95a). The ROD documents decisions made as a
result of information and analyses contained in a Final Environmental Impact Statement (FEIS)
for the management of spent nuclear fuel at the INEEL (DOE95b).10 The ROD, as well as the
Spent Nuclear Fuel FEIS, details the management of spent fuel and high-level waste at other
DOE facilities.
As documented in the FEIS published in April 1995, the DOE decided to regionalize spent
nuclear fuel management by fuel type at the INEEL, Hanford, and Savannah River reprocessing
sites (DOE95b). This decision calls for the following future fuel type distribution:
• The Hanford production reactor fuel will remain at the Hanford Site
• Aluminum clad fuel will be sent to the Savannah River Site
• Non-aluminum clad fuels, including spent nuclear fuel from the Fort St. Vrain
reactor and naval spent nuclear fuel, will be sent to the INEEL
4.3 THE NUCLEAR REGULATORY COMMISSION
The NRC is responsible for licensing and regulating the receipt and possession of high-level
waste, including spent fuel, at privately owned facilities and at certain facilities managed by
DOE. This responsibility will extend to a repository at Yucca Mountain. The NRC currently
9 See, e.g., ID Dept. Health and Welfare v. U.S., F.2nd 149 (9th Cir. 1992).
10 Hereinafter referred to as the Spent Nuclear Fuel FEIS.
4-5
-------
licenses temporary storage facilities at reactor sites, as well as commercial storage facilities at
West Valley, New York, and Morris, Illinois.
4.3.1 legislative Requirement? andJRegulatory Framework
The NWPA specifies that licensing of a geologic repository will occur in three phases. In the
first phase, which follows site characterization and approval of the site for disposal, DOE will
submit a License Application (LA) for the repository to NRC. After the LA is submitted, NRC
will have three years to perform its review, conduct a public hearing, and reach a construction
authorization decision by an independent licensing board. To comply with this schedule, NRC is
already reviewing DOE's site characterization, repository design, and performance assessment
activities to identify and resolve potential licensing issues. However, during the licensing
proceeding itself, all issues, including those previously resolved, can potentially be re-opened by
the licensing board.
In the second phase, as construction of the repository nears completion, DOE will request a
license to receive high-level waste and spent nuclear fuel. Only after that license is granted will
DOE begin placing waste into the repository. In the third phase, when all waste is in place, DOE
will apply for a license amendment to decommission and permanently close the disposal facility.
The NWPA directed both EPA and NRC to publish standards and criteria for the storage and
disposal of high-level waste. In response to the NWPA, NRC developed a generic regulation for
geologic disposal at 10 CFR Part 60. Although the regulation has been amended several times,
the technical criteria date to 1983. As previously noted, the Energy Policy Act of 1992 directed
EPA to develop new individual dose standards for the Yucca Mountain site and for NRC to
conform its standards to the new EPA standards. In light of the requirements of the EnPA, NRC
has elected to develop additional regulations specific to Yucca Mountain. To that end the
Commission has proposed a new rule at 10 CFR Part 63 entitled "Disposal of High-Level
Radioactive Waste in a Proposed Geologic Repository at Yucca Mountain Nevada". Additional
discussion of the proposed rule is included hi Chapter 2 of this BID.
The existing NRC 10 CFR Part 60 regulation does not spell out specific criteria for site
suitability. Under its own 10 CFR Part 960 regulations, if DOE identifies potentially adverse
conditions, the Department must demonstrate that the conditions can be compensated for by the
repository design or favorable site conditions. The DOE siting guideline regulation is written to
4-6
-------
favor the selection of a candidate site with strong waste isolation capabilities by natural features.
However, it is DOE's responsibility to select an appropriate site and design an adequate
repository.
4.3.2 Status of NRC's Program
The NRC's Prelicensing High-Level Waste Repository Program is currently part of the NRC's
Office of Nuclear Material Safety and Safeguards (ONMSS). This program was refocused in FY
1996 based on three events: 1) a reduction in congressional funding, 2) a reorganization of
DOE's high-level waste program, and 3) the publication of the National Academy of Sciences'
report, Technical Bases for Yucca Mountain Standards (NRC97).
The NRC program is now focused on the following ten issues which the Commission believes
are most important to repository performance:
• Igneous activity
• Structural deformation and seismicity
• Evolution of the near-field environment
• Container life and source term
• Thermal effects on flow
• Repository design and thermal-mechanical effects
• Total system performance assessment and integration
• Activities related to the development of the NRC high-level waste regulations
• Unsaturated and saturated flow under isothermal conditions
• Radionuclide transport
The status of resolution of these Key Technical Issues (KTIs) will be periodically re-evaluated
based on new information, performance assessments, and technical interactions with DOE.
The NRC and DOE have agreed on the potential significance of eight of the ten KTIs (excluding
igneous activity and structural deformation and seismicity). Both agencies continue to work
toward the resolution of these issues and meet regularly for DOE/NRC Technical Exchanges.
NRC documents its current positions in a series of Issue Resolution Status Reports (IRSRs)
which are periodically revised.
4-7
-------
4.4 NUCLEAR WASTE TECHNICAL REVIEW BOARD
The NWPAA established the Nuclear Waste Technical Review Board comprising 11 members
recommended by the National Academy of Sciences and appointed by the President. These
individuals are experts in the fields of science, engineering, or environmental sciences and
represent a broad range of scientific and engineering disciplines, including hydrology,
underground construction, hydrogeology, and physical metallurgy. No member of the Board may
be employed by DOE, its contractors, or the National Laboratories. The current Board is
composed of individuals with academic and public and private sector experience.
As defined in Section 503 of the NWPAA,
The Board shall evaluate the technical and scientific validity of activities
undertaken by the Secretary [of Energy].,., including-
(1) site characterization activities, and
(2) activities related to the packaging or transportation of high-level radioactive
waste or spent nuclear fuel
The NWTRB meets four times a year in open public meetings. Two of these meetings are held
in Nevada. In addition, the Board reports to Congress and to the Secretary of Energy at least
twice a year on scientific issues associated with the high-level waste and spent fuel disposal
program. The Board also publishes a periodic newsletter and other information about its views
and activities. Information concerning the Board's membership, activities, and links to NRC and
DOE activities can be found at the Board's website, www.nwtrb.gov.
4.5 STATE AND LOCAL AGENCIES
Congress provided for active State participation in both the NWPA and the NWPAA. The
NWPAA provides for financial assistance to the State of Nevada and any affected unit of local
government to allow for participation in activities related to the establishment of a repository at
Yucca Mountain. Specific activities include:
• Reviewing all work done at the Yucca Mountain site to determine any potential
economic, social, public health and safety, and environmental impacts of a
repository on a State or local government and its residents
4-8
-------
• Developing an impact assistance request
• Monitoring, testing, or evaluating site characterization programs
• Providing information to State residents
• Requesting information from and making comments or recommendations to the
Secretary of Energy
The State of Nevada and any affected unit of local government may also request assistance to
mitigate any economic, social, public health and safety, and environmental impacts that are likely
to result from site characterization activities at Yucca Mountain. The NWPAA specifies that this
financial assistance shall continue until "such time as all such activities, development, and
operation are terminated at such site."
The Nevada legislature created the State's Nuclear Projects/Nuclear Waste Project Office
(NWPO) in 1985 to oversee Federal high-level nuclear waste activities hi the State. Since then,
the NWPO has dealt primarily with the technical and institutional issues associated with DOE's
efforts to characterize the Yucca Mountain site.
Yucca Mountain lies in Nye County, Nevada. This county and ten others that are contiguous
have been designated "affected" and are therefore eligible to receive financial assistance under
the NWPAA. Nye County sponsors a year-round on-site representative. The ten other counties
include: Churchill County, Clark County, Esmeralda County, Eureka County, Lander County,
Lincoln County, Mineral County, Pahrump County, and White Pine County, all in Nevada; and
Inyo County, California.
4.6 NATIVE AMERICAN TRIBES
Native American tribes have a unique sovereign status in U.S. law which was recognized by the
NWPA and the NWPAA. This government-to-govemment relationship between the Federal
Government and the tribes obligates the Federal Government to interact directly and specifically
with tribes in areas where repository or MRS siting activities will occur. The NWPA, as
amended, under Section 2(2), defines an affected tribe as any tribe:
(A) within -whose reservation boundaries monitored retrievable storage facility,
test and evaluation facility, or a repository for high-level waste or spent nuclear
4-9
-------
fuel Is proposed to be located; or (B) whose federally defined possessory or usage
rights to other lands outside of the reservation's boundaries arising out of
congressionally ratified treaties may be substantially and adversely affected by
the locating of such a facility. Provided, That the Secretary of the Interior finds,
upon the petition of the appropriate governmental officials of the tribe, that such
effects are both substantial and adverse to the tribe... (NWP83)
As noted above, specific provisions of the NWPA, as amended, that delineate the participation
activities and rights of affected States in repository and MRS siting decisions also apply to
affected tribes. The means for an affected tribe to disapprove of the site selection and
designation process is given in Section 118(a). An affected tribe is also eligible to receive the
same grants, financial and technical assistance, and payments equal to taxes for which a State is
eligible under Section 116(c). Since the passage of the NWPAA, no tribes have been designated
as affected tribes. However, to ensure compliance with the American Indian Religious Freedom
Act (AIRFA), the National Historic Preservation Act (NHPA) and related statutes, the Native
American Graves Protection and Repatriation Act (NAG90) and the National Environmental
Policy Act (NEPA), the DOE is cooperating with Indian tribes that have current or traditional
religious or cultural ties to the Yucca Mountain site or that may be located near the transportation
routes to or around the site (DOE88b).
In 1985 and in keeping with the NHPA, the Advisory Council on Historic Preservation (ACHP)
issued guidelines for discussing which tribes should be involved in the Yucca Mountain cultural
resource study (STO90). The guidelines contributed to the Yucca Mountain Project's
Programmatic Agreement (PPA), which was jointly produced by DOE and the ACHP. The PPA
requires that DOE consult with tribal groups having traditional cultural ties to the Yucca
Mountain area prior to land-disturbing activities to assure that cultural or religious values are
preserved to the extent practicable. The PPA further stipulates that when such activities are
thought to have a negative effect that cannot be avoided, the DOE will consult further with the
tribal groups and others to identify ways to mitigate those effects.
DOE has established the Yucca Mountain Site Characterization Project, which led to the Cultural
Resources Program to meet resource preservation requirements set forth in the PPA. The
preliminary site characterization (DOE87) identified the ethnic and tribal affiliations of the tribal
groups most likely to have traditional ties to cultural resources located in the Yucca Mountain
region. These groups consist of Southern Paiute, Western Shoshone, and Owens Valley
Paiute/Shoshone people from Nevada, Utah, Arizona and California. Extensive ethnographic
4-10
-------
research led to the identification of 16 tribes, containing members of these three ethnic groups,
that potentially could be involved in the Yucca Mountain Cultural Resources Program:
1. Benton Paiute Indian Tribe, California
2. Timbisha Shoshone Tribe, California
3. Bishop Paiute Indian Tribe, California
4. Big Pine Indian Tribe, California
5. Fort Independence Indian Tribe, California
6. Lone Pine Indian Tribe, California
7. Yomba Shoshone Tribe, Nevada
8. Duckwater Shoshone Tribe, Nevada
9. Pahrump Paiute Indian Tribe, Nevada
10. Las Vegas Paiute Indian Colony, Nevada
11. Las Vegas Indian Center, Nevada
12. Chemehuevi Tribe, California
13. Colorado River Indian Tribes, Arizona
14. Moapa Paiute Tribe, Nevada
15. Paiute Indian Tribe of Utah
16. Kaibab Paiute Tribe, Arizona
All 16 tribes requested that they be included in the project. The DOE informs tribes of the status
of the project through a cooperative agreement with the National Congress of American Indians.
Through this group, DOE and the tribal governments have established a consulting relationship
through which the concerns of the tribal peoples can be expressed.
4-11
-------
REFERENCES
DOE87 U. S. Department of Energy, Native Americans and Nuclear Waste Storage at
Yucca Mountain, Nevada: Potential Impacts of Site Characterization Activities,
Ann Arbor: Institute for Social Research, University of Michigan, 1987.
DOE88a U. S. Department of Energy, Site Characterization Plan, Yucca Mountain Site,
Nevada Research and Development Area, DOE/RW-0199, December 1988.
DOE88b U. S. Department of Energy, Draft 1988 Mission Plan Amendment, DOE/RW-
0187, June 1988.
DOE95a U. S. Department of Energy, Record of Decision, Department of Energy
Programmatic Spent Nuclear Fuel Management and Idaho National Engineering
Laboratory Environmental Restoration and Waste Management Programs, May
30, 1995.
DOE95b U. S. Department of Energy, Department of Energy Programmatic Spent Nuclear
Fuel Management and Idaho National Engineering Laboratory Environmental
Restoration and Waste Management Programs Final Environmental Impact
Statement, DOE/EIS-0203-F, April 1995.
EnPA92 Energy Policy Act of 1992, Public Law 102-486, October 24,1992.
NAG90 Native American Graves Protection and Repatriation Act, Public Law 101 -601,
November 1990.
NRC97 U. S. Nuclear Regulatory Commission, NRC High-Level Radioactive Waste
Program Annual Progress Report: Fiscal Year 1996, NUREG/CR-6513, No. 1,
January 1997.
NWP83 Nuclear Waste Policy Act of 1982, Public Law 97-425, January 7, 1983.
NWP87 Nuclear Waste Policy Amendments Act of 1987, Public Law 100-203, December
22, 1987.
4-12
-------
STO90 Stoffle, Richard W., David B. Halmo, John E. Olmsted, and Michael J. Evans,
Native American Cultural Resource Studies at Yucca Mountain, Nevada, Ann
Arbor: Institute for Social Research, University of Michigan, ISBN 0-877944-
328-6, 1990.
4-13
-------
CHAPTER 5
QUANTITIES, SOURCES, AND CHARACTERISTICS OF SPENT NUCLEAR
FUEL AND HIGH-LEVEL WASTE IN THE UNITED STATES
5.1 INTRODUCTION
This chapter presents current and projected inventories of spent nuclear fuel and DOE defense
high-level radioactive waste. Current plans call for both of these waste forms to be disposed of
in the Yucca Mountain repository. The waste inventories cited are from the most recent and
reliable sources of Federal Government information publicly available (DOE94a, DOE95a,
DOE95b, DOE95c, DOE95d, DOE95e, DOE95f). The waste forms are inventoried by mass or
volume and radioactivity content.
5.2 SPENT NUCLEAR FUEL
Spent nuclear fuel is defined as fuel that has been withdrawn from a nuclear reactor following
irradiation and whose constituent elements have not been separated by reprocessing (EPA85).
Generators of spent nuclear fuel include: 1) commercial Light Water Reactors (LWR), which
consist of Pressure Water Reactors (PWR) and Boiling Water Reactors (BWR); 2) government-
sponsored research and demonstration programs, universities, and industry; 3) experimental
reactors, e.g., liquid-metal, fast-breeder reactors (LMFBR) and high-temperature gas-cooled
reactors (HTGR); 4) U.S. Government nuclear weapons production reactors; and 5) Department
of Defense (DOD) reactors.
Approximately 98 percent of the spent nuclear fuel from commercial power reactors is stored at
the reactor sites where it was generated; the rest is stored at central commmercial storage
facilities. Spent nuclear fuels from one-of-a-kind reactors are currently stored at the Hanford Site
and the Idaho National Engineering and Environmental Laboratory (INEEL). Spent nuclear fuels
from the Fort St. Vrain HTGR, the N-reactor, the Fast Flux Test Facility (FFTF), the
Shippingport reactor, and the damaged Three Mile Island Unit 2 reactor are stored at INEEL.
Other types of special spent nuclear fuel are stored at the Savannah River Site (DOE95a). The
fuels at these DOE facilities are Government-owned and are not scheduled for reprocessing in
support of DOE defense activities.
5-1
-------
The fuel for LWRs consists of uranium dioxide pellets encased in zirconium alloy (Zircaloy) or
stainless steel tubes. During reactor operation, fission of the uranium-235 produces energy,
neutrons, and radioactive isotopes known as fission products. The neutrons produce further
fission reactions and thus sustain the chain reaction. The neutrons also convert a portion of the
uranium-238 into plutonium-239, which can also undergo fission. In time, the fissile uranium-
235, which originally constituted some 3 to 4 percent of the enriched fuel, is depleted to such a
level that power production becomes inefficient. Once this occurs, the fuel bundles are deemed
"spent" and are removed from the reactor. In the United States, reprocessing of commercial
spent nuclear fuel to recover the unfissioned uranium-235 and the plutonium for reuse as a fuel
resource is currently not taking place, nor is it expected to occur in the future.
The radioactive materials associated with spent nuclear fuel fall into three categories: 1) fission
products; 2) actinide elements (atomic numbers of 89 and greater); and 3) activation products.
Typically, fresh spent nuclear fuel contains more than 100 radionuclides as fission products.
Fission products are of particular importance because of the quantities produced, their high
radiological decay rates, their decay-heat production, and their potential biological hazard. Such
fission products include: strontium-90; technetium-99; iodine-129 and -131; cesium isotopes,
such as cesium-134,.-135, and -137; tin-126; and krypton-85 and other noble gases.
Activation products include tritium (hydrogen-3), carbon-14, cobalt-60, and other radioactive
isotopes created by neutron activation of fuel assembly materials and impurities in cooling water
or in the spent nuclear fuel. The actinides include uranium isotopes and transuranic elements,
such as plutonium, curium, americium, and neptunium. The exact radionuclide composition of a
particular spent nuclear fuel sample depends on the reactor type, the initial fuel composition, the
length of time the fuel was irradiated (also known as "burnup"), and the elapsed time since its
removal from the reactor core.
5.2.1 Commercial Spent Nuclear Fuel Inventory and Projection
By the end of 1998, there were 37,700 MTIHM" of spent nuclear fuel in inventory from
commercial reactor operations. Approximately 37,000 MTIHM is stored at reactor sites. The
remainder is stored at the West Valley Demonstration Project (WVDP) (27 MTIHM) in West
1' Commercial spent nuclear fuel reported in DOE95e is in units of metric ton (tonne) of initial heavy
metal (MTIHM) to avoid difficulties arising from the need to estimate ranges of varied heavy-metal content that
result from different levels of enrichment and reactor fuel burnup. A metric ton is 1,000 kilograms, corresponding
to about 2,200 pounds.
5-2
-------
Valley, New York; the Idaho National Engineering and Environmental Laboratory (43 MTIHM)
in Idaho Falls, Idaho; and the Midwest Fuel Recovery Plant (MFRP) (744 MTIHM) in Morris,
Illinois (DOE95e). The historical (1970-1994) and projected (1995-2030) spent nuclear fuel
inventories and accumulated radioactivities are given in Table 5-1. Projections of nuclear
capacity are based on the DOE/EIA Low Case assumptions, which forecast an increase in the
installed nuclear capacity from 99.1 gigawatts-electric (GW(e)) in 1994 to a peak of 100.3
GW(e) in 2000, then a decrease to 2.3 GW(e) by 2030, as shown in Table 5-2 (DOE95f). The
Low Case scenario is based on these assumptions: 1) reprocessing of spent nuclear fuel will n6t
occur; 2) currently licensed reactors will be retired when their initial license terms expire; and 3)
new advanced LWRs will not be available before 2015. The DOE/EIA projections also assume
that bumup levels of spent nuclear fuel will increase from their current average of 33,065 and
39,989 Megawatt days (MWd) per MTIHM to 42,000 and 54,000 MWd/MTIHM for BWRs and
PWRs, respectively. This increase is predicted over the period 1994 to 2020. Based on
currently-mandated limits, only 63,000 MTHM can be accommodated at the Yucca Mountain
site.
Table 5-1. Historical and Projected Mass and Radioactivity of Commercial Spent Nuclear Fuel
(DOE94a, DOE95e, DOE95f, DOE96a)
f V
1970
1975
1980
1985
1990
1994
1995°
2000C
2005*
2010C
2015C
2020C
2025*
2030C
••"• f *• \ %
*" '* f f f '•
v <• ^^ifc j|*y *t1CSffcjJf \*^* "* *"* °*
v ^lt*"l 3t fjrjff*ijr T
55
1,567
6,558
12,684
21,547
29,811
32,022
43,100
53,500
63,600
73,900
80,000
85,500
87,900
•. ^
215
3,315
10,137
14,228
22,910
26,661
25,600
32,600
36,900
39,800
36,700
34,700
32,100
24,700
* Metric tons initial heavy metal refers to the original mass of the actinide elements of
the fuel.
b A curie of radioactivity corresponds to 3.7 x 1010 disintegrations per second.
c Projections beyond 1994 are based on the DOE/EIA Low Case.
5-3
-------
Table 5-2. Historical and Projected* Installed Nuclear Electric Power Capacity (DOE951)
>f { ': -
/ ^Eilidof jQalafttfar YK^f
1960
1965
1970
1975
1980
1985
1990
1994
T<&mwffi -..I..
0.3
0.4
5.8
38.3
51.9
78.5
99.6
99.1
-. /
.Bifid ^f t^jttBdigr Yfflafji
1995*
2000*
2005*
2010*
2015*
2020*
2025*
2030*
^jT^IGW{$
99.1
100.3
100.3
91.1
61.4
46.7
22.0
2.3
* Lower Reference Case projected capacity includes all existing reactors, completed or
under construction, plus additional new reactors beyond the year 2005.
5.2.2 DOE Spent Nuclear Fuel
The DOE reprocessed most of its spent nuclear fuel in the facilities at INEEL, the Hanford Site,
and the Savannah River Site. However, some spent nuclear fuel remains because of U.S.
Government decisions to stop reprocessing. Most of this fuel came from the Hanford Site N-
Reactor, a dual-purpose reactor designed to produce plutonium for use in nuclear weapons and to
generate electricity for commercial use. Smaller amounts of spent nuclear fuel associated with
nuclear weapons production are stored at the Savannah River Site. Spent nuclear fuel from the
Naval Nuclear Propulsion Program is stored at INEEL and, for short time, at some naval nuclear
shipyards. The DOE will also assume responsibility for fuel from some special-case commercial
nuclear reactors, foreign research reactors, and certain domestic research and test reactors. The
following sections discuss the nature and quantity of this spent nuclear fuel and DOE's plans to
manage it. Most of the discussion that follows is derived from the Spent Nuclear Fuel FEIS
(DOE95c).
Hanford Site
The Hanford Site produced plutonium for use in nuclear weapons from the start of the Manhattan
Project until DOE halted production in 1989. Hanford's production reactors generated 2,100
MTHM of the existing DOE inventory of spent nuclear fuel. There is a total of 2,096 metric tons
5-4
-------
of spent N-reactor fuel at Hanford, which comprises all but about 1 percent of the spent nuclear
fuel inventory at the site. This fuel is stored in three facilities; DOE's interim plans for
management of this fuel include possible relocation to a single storage facility. Sources of the
other spent nuclear fuel at the site included single-pass Hanford production reactors, the Fast
Flux Test Facility, Shippingport Core H, and miscellaneous test facilities.
Idaho National Engineering and Environmental Laboratory
Six major facility areas at the INEEL store spent nuclear fuel: Argonne National Laboratory-
West; Idaho Chemical Processing Plant; Naval Reactors Facility; Power Burst Facility; Test Area
North; and the Test Reactor Area. Spent nuclear fuel is kept in a variety of dry and wet
configurations. The INEEL stores about 10 percent of DOE's current inventory of spent nuclear
fuel, i.e., about 300 metric tonnes.
Savannah River Site
The DOE has 200 MTHM, or about 8 percent of its system-wide spent nuclear fuel inventory, in
storage at the Savannah River Site. This fuel is stored in the Receiving Basin for Off-site Fuels
(RBOF), in three reactor disassembly basins, and in basins in the F- and H-Area Canyons.
The F- and H-Area Canyons are among the only remaining operable chemical separation
facilities of their kind in the DOE complex. Each canyon has an associated storage basin that
serves as an interim staging area where spent nuclear fuel awaits chemical separation. The basins
contain 13 reactor fuel assemblies (H-Area) and aluminum-clad targets (F-Area).
The DOE has stored most aluminum-clad spent nuclear fuel from Savannah River Site reactors in
water-filled concrete basins. Reactor disassembly basins for the K-, P-, and L-reactors contain
spent nuclear fuel and target material. These basin structures were built in the 1950s and were
not intended for prolonged storage of radioactive materials.
The RBOF has been receiving fuels of U.S.-origin since 1964, including fuel manufactured in the
United States but irradiated in foreign reactors. About 30 percent of the fuels in the RBOF
consist of uranium clad in stainless steel or zircaloy which Savannah River Site facilities cannot
process without modification.
5-5
-------
Other Generator/Storage Locations
The DOE has in its possession, or has title to, a small amount of spent nuclear fuel in many other
locations throughout the United States. These locations include both DOE and non-DOE
facilities. For example, the Oak Ridge National Laboratory (ORNL) stores less than 1 MTHM of
spent nuclear fuel. This fuel is left over from research on fuel elements removed from
commercial or demonstration reactors, as well as fuel removed from reactors that operated at
ORNL. Under the Spent Nuclear Fuel FEIS, this fuel will be transferred to either INEEL, the
Hanford Site, or the Savannah River Site.
Besides ORNL, DOE is responsible for spent nuclear fuel from research and test reactors at the
Brookhaven, Los Alamos, Sandia, and Argonne-East Laboratories. These facilities have a total
of about two MTHM in storage. Other DOE sources include:
• Non-DOE Research Reactors - The DOE has title to the spent nuclear fuel that is
stored at or is generated by 57 small research reactors. These reactors operate at
universities, commercial establishments and other government agencies, such as
the Department of Defense. These reactors have a current inventory of less than 5
MTHM and will generate very little additional spent nuclear fuel by 2035.
• Commercial Power Reactors - The DOE has possession of 125 spent nuclear fuel
assemblies and 20 complete or sectioned spent nuclear fuel rods from several
commercial power reactors that supported DOE-sponsored research and
development programs. This fuel is stored at the West Valley Demonstration
Project and at the Babcock and Wilcox Lynchburg Technology Center in
Campbell County, Virginia. Other commercial spent nuclear fuel is already stored
at the INEEL, the Hanford Site, and the Savannah River Site.
• Foreign Research Reactors - The DOE has accepted limited amounts of spent
nuclear fuel from foreign reactors (WCM95a). In some cases, this fuel was
manufactured by the DOE. The DOE will, under the Spent Nuclear Fuel FEIS,
continue to receive and store spent nuclear fuel from foreign sources (DOE95c).
All the spent nuclear fuel listed above, if not already transferred, will be shipped to INEEL, the
Hanford Site, or the Savannah River Site, according to the Spent Nuclear Fuel FEIS.
5-6
-------
Spent Nuclear Fuel Management Options
The Spent Nuclear Fuel FEIS says that most spent nuclear fuel in the possession of the DOE will
be stored until a geologic repository is available. Some of this storage will be wet, but some
fuels, such as those with aluminum cladding, may be placed in dry storage. The DOE is also
considering options to stabilize some of its corroding spent nuclear fuel (WCM95b). One of
these options is chemical separation at the Savannah River Site.
A summary inventory of DOE spent nuclear fuel is given in Table 5-3. Spent nuclear fuel in this
listing includes material from fuels other than those discharged from production reactors.
Table 5-3 includes nuclear fuel that has been withdrawn from or resides in storage at a reactor
following irradiation but that has not been reprocessed. Also included are some defective fuel
elements and special nuclear forms, as well as some commercially generated nuclear fuels and
fuels from foreign reactors and university research reactors.
The estimates hi Table 5-3 have been recently updated by DOE (DOE98, vol. 3, Table 3-13).
The more recent estimate is 2,496 MTHM. Based on regulatory limits, only 2,333 MTHM is
scheduled for disposal at Yucca Mountain.
5.3 DEFENSE HIGH-LEVEL RADIOACTIVE WASTE
High-level radioactive wastes are the highly radioactive materials resulting from the reprocessing
of spent nuclear fuel, including liquid waste produced directly in reprocessing, and any solid
material derived from such liquid waste (EPA85, NWP83). Commercial high-level radioactive
waste currently stored at the West Valley Demonstration Project will be converted to a solid form
(glass) prior to disposal (NRC88). Substantial quantities of this waste have been vitrified.
High-level waste is generated by the chemical reprocessing of spent research and production
reactor fuel, irradiated targets, and naval propulsion fuel. The fission products, actinides, and
neutron-activated products of particular importance are the same for high-level waste as they are
for spent nuclear fuel assemblies (DOE88, DOE95e).
5-7
-------
Table 5-3. DOE Spent Nuclear Fuel Inventory (DOE95a)
J,
f =
'
OeuB3atarOr$««^Site ; "
c
DOE Sites
Hanford Site
Idaho National Engineering and
Environmental Laboratory
Savannah River Site
Oak Ridge Reservation
Other DOE Sites
Naval Nuclear Propulsion
Reactors
Foreign Research Reactors
Non-DOE Domestic
Domestic Research and Test
Reactors0
Special-Case Commercial SNF at
non-DOE locations'"
Total6
Percent of 2035 total
<&dMte&0&&
f > »f CJ
ifn&i* ^MIVWI
2132.44 80.6
261.23 9.9
206.27 7.8
0.65 <0.1
0.78 <0.1
0.00" 0.0
0.00 0.0
2.22 <0.1
42.69 1.6
2646.27
96.5
Fafi^£ leases
(tarouglfaa^
Mimi Fstfcstt
0.00 0.0
12.92 13.5
0.00 0.0
1.13 1.2
1.50 1.6
55.00 57.6
21.70 22.7
3.28 3.4
0 0
95.53
3.5
f»feJ{io35)
MfHM $omt(
2132.44 77.8
274.14 10.0
206.27 7.5
1.78 <0.1
2.28 <0.1
55.00 2.0
21.70 0.8
5.50 0.2
42.69 1.6
2741.80
100.0
MTHM = metric tons of heavy metal.
b The existing inventory of Naval Nuclear Propulsion Program spent nuclear fuel (10.23 MTHM) stored at the
INEEL is included in the INEEL total.
c Includes research reactors at commercial, university, and government facilities.
" The total inventory of spent nuclear fuel from special case commercial reactors is 186.41 MTHM. The 42.69
MTHM listed here is that stored at the Babcock & Wilcox Research Center, Fort St. Vrain Reactor, and West Valley
Demonstration Project. The remaining special-case commercial spent nuclear fuel is stored at the INEEL, the Oak
Ridge Reservation, and the Savannah River Site, and is included in the totals for those locations.
' Numbers may not sum due to rounding.
Weapons program reactors were operated mainly to produce plutonium. Reprocessing to recover
the plutonium was an integral part of the weapons program. Naval propulsion reactor fuel
elements were also reprocessed to recover the highly enriched uranium that remained after use.
DOE decided in 1992 to phase out the domestic reprocessing of irradiated nuclear fuel of defense
program origin, so minimal amounts of high-level waste will be added to the current inventory.
High-level radioactive waste that is generated by the reprocessing of spent reactor fuel and
targets contains more than 99 percent of the nonvolatile fission products produced in the fuel or
targets during reactor operation. It generally contains about 0.5 percent of the uranium and
plutonium originally present in the fuel. Most of the current high-level waste inventory, which is
5-8
-------
the result of DOE national defense activities, is stored at the Savannah River Site, INEEL, and
the Hanford Site. A limited quantity of high-level waste Is stored at the West Valley
Demonstration Project. These high-level wastes have to date been through one or more treatment
steps (e.g., neutralization, precipitation, decantation, evaporation). It is currently planned that
this HLW will be solidified, using a vitrification process, for disposal. Vitrification is well
underway at the Savannah River She and the West Valley Demonstration Project.
The DOE defense high-level waste at INEEL results from reprocessing nuclear fuels from naval
propulsion reactors and special research and test reactors. The bulk of this waste, which is
acidic, has been converted to a stable, granular solid (calcine). At the Savannah River and
Hanford Sites, the acidic liquid waste from reprocessing defense reactor fuel is or has been made
alkaline by the addition of caustic soda and stored in tanks. During storage, this alkaline waste
separates into three phases: liquid, sludge, and salt cake. The relative proportions of liquid and
salt cake depend on how much water is removed by waste treatment evaporators during waste
management operations.
Both alkaline and acidic high-level wastes were generated at West Valley. The alkaline waste
was generated by reprocessing commercial power reactor fuels and some Hanford N-Reactor
fuels. Acidic waste was generated by reprocessing a small amount of commercial fuel containing
thorium.
Projected volumes and total radioactivity for high-level waste stored at the Hanford Site, INEEL,
the Savannah River Site, and WVDP are given in Table 5-4. Projected inventories for each site
are based on specific assumptions and are subject to change. New treatment methods and waste
forms are possible and may affect the future projections. Since all sites are progressing toward
closure, there should be minimal amounts of waste added to the current inventory. Interim
storage of DOE high-level waste will be required and will most likely be at the site where the
waste is produced. Current DOE policy states that DOE high-level waste will not be accepted at
the geologic repository until six years after initial receipt of commercial spent nuclear fuel
(DOE94b).
DOE currently estimates that 10,110 MTHM will be available for disposal (DOE98, vol. 3,
Section 3.5.1.5). Of this total, only 4,667 MTHM are actually scheduled for disposal based on a
regulatory limit for the repository of 70,000 MTHM.
5-9
-------
Table 5-4. Historical and Projected Cumulative Volume and Radioactivity of High-Level Waste Stored in Tanks,
Bins, and Capsules By Site (DOE95d, DOE95e)
: * * "• V-^A v
« - Year * ,
1980
1981
1982
1983
1984
1985
1986
1987
198&
1989
1990
1991
1992
1993
1994
1995
2000
2005
2010
2015
2020
2025
2030
x"s "™4 "* ^
/ M«iQltj&* '"'' "1J
219.4
219.4
213.3
229.4
225.6
222.1
226.4
239.7
243.4
244.8
253.6
256.4
258.7
261.7
238.9
237.3
232.1
229.6
197.9
134.4
70.8
26.3
2.4
;:ttS
^IfiL-
11.4
12.0
11.5
9.7
10.1
10.1
9.5
11.9
11.0
12.0
12.0
10.4
11.2
10.5
11.0
11.4
9.8
9.5
9.8
10.4
7.2
5.8
4.1
j$$$g;fflj&
^W^
96.7
105.7
115.0
111.4
125.6
122.7
127.8
127.6
128,4
122.0
131.7
127.9
126.9
129.3
126.3
122.1
98.7
75.2
51.7
28,2
4.7
0
0
^../lI.J
<>$v0FY:':'
2.2
2.2
2.2
2.2
2.2
2.2
2.2
2.2
2.1
2.4
1.2
1.7
1.6
2.0
2.2
0.9
0
0
0
0
0
0
0
pr;;
£|&aL >',
329.7
339.3
342.0
352.7
363.5
357.1
365.9
381.4
384.9
381.1
398.5
396.5
398.3
403.5
378.4
371.7
340.6
314.2
259.4
173.0
82.8
32.0
6.5
>;>
?3fefi&'
576.7
550.2
437.1
427.5
470.2
519.0
534.6
478.2
447.4
419.3
399.3
384.2
372.1
361.4
348.0
339.9
302.4
269.1
232.2
128.9
45.5
15.2
0
\;^
SS9S£....\....^ .•:.
53.4
63.6
71.6
64.8
58.6
69.4
60.6
62.5
67.0
68.4
63.2
59.4
50.8
52.5
51.6
50.5
44.1
38.9
34.6
30.8
26.9
13.8
3.9
- '"-ftto&iTigS
Vsife '
699.0
982.0
828.8
776.2
795.9
841.4
794.7
734.0
664.4
598.9
561.6
537.6
632.4
606.0
534.5
502.2
352.7
239.5
147.0
71.7
10.7
0
0
j&t •% +
'wv»pL
33.5
32.7
31.9
31.2
30.5
29.7
29.1
28.4
27.9
27.3
26.7
26.2
25.9
25.3
24.7
24.1
0
0
0
0
0
0
0
v ''"'
•f , •••.
"tfc&SU
1,362.6
1,628.5
1,369.4
1,299.7
1,355.2
1,459.5
1,419.0
1,303.1
1,206.7
1,113.9
1,050.8
1,007.4
1,081.2
1,045.3
958.8
916.7
699.2
547.5
413.8
231.3
83.1
29.0
3.9
5-10
-------
5.3.1 High-Level Waste Inventories at the Hanford Site
The alkaline high-level waste (239,000 m3) located at Hanford is stored in underground carbon-
steel tanks. Currently 155,800 m3 is solid (salt cake and sludge) and 83,200 m3 is liquid; waste
volumes change with time because of on-going waste management activities. There are
approximately 350 million curies of total radioactivity contained in the waste, which has been
accumulating since 1944. The high-level waste was generated by reprocessing production
reactor fuel for the recovery of plutonium, uranium, and neptunium for defense and other Federal
programs.
AH of the fuel reprocessing methods generated acidic waste streams. Sodium hydroxide or
calcium carbonate was added to the waste before it was transferred to the tanks to neutralize the
acid and minimize tank corrosion. The tanks currently contain moderate to strong alkaline
solutions. Additional post-processing of the waste to recover plutonium and uranium, or to
reduce the volume of high-level waste, has resulted in the addition of ferrocyanide and some
organic compounds listed as hazardous. Fuel reprocessing was suspended from 1972 until
November 1983. Most of the high-heat-emitting isotopes (strontium-90, cesium-137, and their
decay products) have been removed from the old waste, converted to solids as strontium fluoride
and cesium chloride, placed in double-walled capsules, and stored in water basins. A total of
2,217 capsules were manufactured and 1,933 remain. (A portion of these capsules have been
used outside the facility or have been dismantled.)
Double-shell tanks continue to receive waste generated by decommissioning and cleanup of
Hanford Site facilities. This includes: effluents associated with the deactivation program for the
PUREX Plant; waste from B-Plant maintenance activities; laboratory waste; and miscellaneous
waste streams from ion-exchanger resin regeneration.
The tanks now contain a mixture of salt cake, liquid, and sludges with both radioactive and
hazardous components. Sludge consists primarily of solids (hydrous metal oxides) precipitated
from the neutralization of acid waste. Salt cake consists of the various salts formed from the
evaporation of water from the waste. Liquids exist as supernatant (liquid above solids) and
interstitial liquid (liquid filling the void between solids) in the tanks.
5-11
-------
The tank waste is mostly inorganic, containing sodium hydroxide; salts of nitrate, nitrite,
carbonate, aluminate, and phosphate; and hydrous oxides of aluminum, iron, and manganese.
The radioactive components consist primarily of long-lived fission products and shorter-lived
radionuclides, such as strontium-90 and cesium-13 7, and isotopes of uranium, plutonium, and
arnericium. Some tanks contain the chelating agents EDTA and HEDTA. Some contain
halogenated and nonhalogenated organic contamination, while others contain mixed waste with
detectable levels of lead, chromium, and cadmium.
DOE has in place a program to treat and remediate some of this tank waste. In August 1998
DOE awarded a contract to BNFL Inc. to undertake tank waste remediation. Under the contract
BNFL will spent the initial two years in facility design. Assuming that DOE provides approval, a
facility will then be constructed and remediation will begin. Facility operation is expected to
begin in 2005 or 2006 and, during the initial ten-year operational period, DOE expects to process
waste from 11 storage tanks. The material treated during this initial phase is estimated to
constitute about 10% of the total waste mass and 20 to 25% of the total radioactivity. DOE plans
to separate tank contents into high-level and low-level components, thereby reducing the amount
of high-level radioactive waste. All remaining high-level liquid waste would then be vitrified
and placed in stainless steel canisters for storage on site until a geologic repository is available
for disposal. Vitrification is also planned for the low-level (low activity) waste. The Hanford
Waste Vitrification Plant is currently scheduled to begin operation on HLW in 2007 (90%
confidence date) (DOE98a).
5.3.2 High-Level Waste Inventories at INEEL
About 11,000 m3 of high-level waste, containing approximately 50 million curies of total
radioactivity, is currently stored at INEEL; this volume consists of 7,200 m3 of acidic liquid
waste (1,306 m3 is high-level waste; the remainder is sodium waste that was treated as high-level
waste) and 3,800 m3 of solid materials. Liquid high-level waste was generated at INEEL
primarily by the reprocessing of spent nuclear fuel from the national defense (naval propulsion
nuclear reactors) and reactor testing programs; a small amount was also generated by
reprocessing fuel from non-defense research reactors. This acidic waste is stored underground in
large, high-integrity, stainless steel tanks. Waste that has been converted to a calcine is stored in
retrievable stainless steel bins housed in reinforced concrete vaults. Greater than 90 percent of
the total radioactivity is contained in the calcine.
5-12
-------
5.3.3 High-Level Waste Inventories at the Savannah River Site
Approximately 126,300 m3 of alkaline high-level waste that has accumulated at the Savannah
River Site over the past three decades is currently stored underground in carbon-steel tanks. The
current inventories consist of alkaline liquid, sludge, and salt cake that were generated primarily
by the reprocessing of nuclear fuels and targets from plutonium production reactors. The sludge
is formed after treatment with caustic agents. Salt cake results when the supernatant liquor is
concentrated in waste treatment evaporators. The high-level waste consists of 58,100 m3 of
liquid and 68,200 m3 of solid material having a total radioactivity of approximately 500 million
curies.
Tank farms at the Savannah River Site contain 24 single-shell and 27 double-shell tanks for
storing high-level waste. The DOE plans to remove the liquid waste from these tanks by 2035
(DOE95d). The removal process includes these process steps involved in vitrifying the waste:
• The salt solution is removed from the tanks and chemically treated in the In-Tank
Precipitation Facility to precipitate radionuclides. The decontaminated filtrate is
stripped of benzene and transferred to the Saltstone Facility for disposal. The
concentrated precipitate is treated to remove nitrites and then stored. (This
technology was abandoned in 1999.)
• Sludge from the tanks is transferred to a tank at the Extended Sludge Processing
Facility and washed in water and sodium hydroxide to remove salts and
aluminum. The washed sludge is stored in a tank until it is transferred to the
Defense Waste Processing Facility.
• At the Defense Waste Processing Facility, which recently began operation, the
washed sludge and the precipitate from the In-Tank Precipitation Facility are
combined with glass frit and vitrified to form glass logs. The process removes
organics from the precipitate and mercury from both precipitate and sludge. The
vitrified waste is contained hi stainless steel canisters.
5.3.4 High-Level Waste Inventories at the West Valley Demonstration Project
About 2,180 m3 of high-level waste is stored at the WVDP facility and consists of 2,040 m3 of
liquid alkaline waste and 140 m3 of solid waste (consisting of alkaline sludge and inorganic
zeolite ion-exchange medium). The alkaline waste is stored in an underground carbon-steel tank,
and the zeolite waste is stored in an underground carbon-steel tank covered by an aqueous
5-13
-------
alkaline solution. Reprocessing was discontinued at the WVDP in 1972. No additional high-
level waste has been generated since.
In June 1996, the vitrification of HLW into glass logs was initiated at the WVDP. The glass logs
are two feet in diameter by 10 feet long. As of mid-1999, more than 680 glass logs have been
made.
5.4 SIGNIFICANT RADIONUCLIDES CONTAINED IN SPENT NUCLEAR FUEL AND
HIGH-LEVEL WASTE
Of the 70,000-tonne capacity limit for Yucca Mountain, about 40,785 MTHM and 22,210
MTHM represent spent PWR and spent BWR fuel, respectively (DOE95g). About 7,000
MTHM of vitrified high-level waste represents the balance of the total repository inventory. For
the Yucca Mountain Site, radionuclide-specific activity levels are estimated by assuming that all
spent fuel had been removed from the reactors 30 years before emplacement with burnups of
39,651 MWd/MTHM for PWR fuel and 31,186 MWd/MTHM for BWR fuel12. Although the
bumup of spent fuel producing HLW is generally unknown, this uncertainty is thought to affect
the adjustment for decay only marginally.
Table 5-5 lists some radionuclide inventories for PWR and BWR reactor fuels, based on the
DOE assumptions concerning burnup and cooling time as cited above. These values are
generated from the ORIGEN2 computer code (ROD86), which calculates depletion, buildup, and
decay of isotopes for given fuel initial conditions and utilization histories. Also shown in Table
5-5 are estimated nuclide inventories for the defense high-level waste, based on assumptions
comparing burnup and fissile material contents for fuel from defense production reactors and
commercial power reactors. The values shown in Table 5-5 demonstrate that the radionuclide
inventories in a repository at Yucca Mountain stemming from defense high-level wastes are
expected to be much less than those from commercial spent fuel.
The radionuclide inventory of the repository will change with time due to radioactive decay and
ingrowth of radioactive decay products. For example, inventories of the initially-prominent
fission products Cs-137 and Sr-90, which have approximately 30-year half lives, will decay to
insignificant levels within 1,000 years, while some decay products, such as Pb-210 and Ra-226,
will not achieve peak values until about 100,000 years after repository closure. Activity levels
12 Inventory and burnup values were slightly revised in the 1998 Viability Assessment.
5-14
-------
for very long-lived radioisotopes will maintain low but nearly constant levels for periods on the
order of a million years. Overall, the radioisotope inventory of the wastes placed in the
repository will decrease by about five orders of magnitude during the first 100,000 years after
closure, and remain virtually constant thereafter.
Table 5-5. Radionuclide Inventories in Spent Nuclear Fuel and High-Level Wastes
Expected to be Disposed in a Yucca Mountain Repository*
4
sotope
Ac-227
Ag-108m
Ajn-241
\m-242m
\m-243
C-14
Cl-36
Cm-243
Cm-244
Cm-245
Cm-246
Cs-135
Cs-137
-129
Mo-93
"Jb-94
4i-59
•ii-63
ty-237
'a-231
>b-210
Pd-107
»u-238
»u-239
>u-240
Pu-241
*u-242
la-226
5e-79
Sra-151
>n-121m
W&fevi&teiy
1.70 xlO'5
9.82x10°
2.50 xlO3
8.38x10°
1.00 x 10'
1.43x10°
1.04xlO-2
7.21x10°
3.52 xlO2
6.24 x lO'2
l.OSxlO'2
4.32 x 10"'
5.19x10"
2.75 x lO'2
5.56 x 1C"4
2.95 x 10'2
1.01x10°
1.23 xlO2
2.87 x lO'1
3.53 x lO'5
5.34 x lO'7
8.70 x 10'2
1.61 x 103
2.97 xlO2
4.64 xlO2
3.09xl04
1.18x10°
2.19 xlO-6
3.73 x 10'1
2.80 xlO2
9.52 xlO'1
PWR^eO&Ky--
1.85xlO-5
1.14xlO'2
3.67 x 10'
1.03x10'
2.09x10'
1.46x10°
1.14xlO'2
1.61x10'
9.86 xlO2
2.20 xlO'1
4.39 xlO'2
5. 04 xlO'1
6.96 xlO4
3.74xlO'2
2.76 xlO'2
1.41x10°
4.21 x 10°
5.44 xlO2
4.27 x 10'1
3.82 x lO'5
5.37 x lO'7
1.28x10-'
2.87 xlO3
3.53 x 102
5.34 xlO2
4.61 x 10"
2.05 x 10°
2.24 xlO'6
4.96 xlO'1
3.66 xlO2
6.08 xlO'1
•f s %
• BkWlGve&toty
6.17 xlO"4
0
1.05 xlO3
1.70x10-'
2.36 xlO'1
0
0
4.29 xlO'2
2.77x10'
5.47 xlO"1
6.19 xlO-5
2.98 x 10-'
2.97x10"
3.44x10-*
0
6.93 x 10-5
1.12x10-'
8.13x10°
6.69 x lO'2
1.44xlO'2
0
2.74 x 10'2
9.27 x 102
1.13x10'
7.77x10°
3.49 xlO2
1.16 xlO'2
0
1.88x10-'
4.66 xlO2
5.91 x lO'2
Combined Weighted
Aver^g^^
7.79 xlO'5
9.76 xlO'3
3.04 x 103
8.68 x 10°
1.54x10'
1.30x10°
9.94 xlO'3
1.17x10'
6.89 x 102
1.48x10-'
2.90 xlO'2
4.61 x 10-'
6.00 x 10"
3.05xlO-2
1.62xlO'2
8.30x10-'
2.78 x 10°
3.57 x 102
3.46x10''
1.47x ID'3
4.82 xlO'7
1.05x10-'
2.28 x 103
3.01 x 102
4.59 x 102
3.67x10"
1.57x10°
2.00 x 10-6
4.26 xlO'1
3.49 xlO2
6.62x10''
5-15
-------
Table 5-5 (Continued)
'&&)$&
Sn-126
Sr-90
Pc-99
Fh-229
Fh-230
LJ-232
J-233
LJ-234
LJ-235
U-236
LJ-238
Zr-93
; "••
^B'WR5lfeveiilorsf
6.26 x ID'1
3.79 x 104
1.22x10'
2.07 x ID'7
3.53 x 10-"
1.79 xlO'2
4.30 x 10's
1.41 x!0°
2.25 x 10'2
2.66 x 10-'
3.18x10-'
1.98x10°
P$^^0f!P
9.01 x ID'1
4.91 x 104
1.55X101
3.66 x ID'7
3.72 x 10^
4.16 xlO'2
6.57 x lO'5
1.56x10°
2.30 x ID'2
3.37 xlO'1
3.13 xlO'1
2.36 x 10°
$„>£
WUW$mjfowtf -
5.10x10''
2.93 x 104
6.33 x 10°
2.04 xlO"4
2.90 xlO"4
2.12xlO'2
2.41 x 10'2
3.40 x 10-2
2.22 xlO"4
1.11x10°
8.98 xlO'3
1.55x10°
'pmliloedWeij^fed - :
- ^ - Avferi^* ' '"f
7.75 x 10-'
4.36x10"
1.35x10'
2.07 x lO'5
3.58 x 10"4
3.20 xlO'2
2.46 xlO'3
1.36x10°
2.06 x 10'2
2.81 x 10-'
2.84 x 10'1
2.16x10°
Inventories for spent BWR and PWR fuel are in curies per initial metric ton of heavy metal. Inventories for
HLW are for estimated equivalent metric tonnes of heavy metal. Values are based on burnup and cooling
histories assumed by DOE.
5-16
-------
REFERENCES
DOES 8 U.S. Department of Energy, Site Characterization Plan, Yucca Mountain Site,
Nevada Research and Development Area, DOE/RW-0199, December 1988.
DOE94a U.S. Department of Energy, Energy Information Administration, Nuclear Fuel
Data Form RW-859, December 1994.
DOE94b U.S. Department of Energy, Waste Acceptance System Requirements Document,
DOE/RW-0351P, Rev. 1, March 1994.
DOE95a U.S. Department of Energy, Department of Energy Programmatic Spent Nuclear
Fuel Management and Idaho National Engineering Laboratory Environmental
Restoration and Waste Management Programs Final Environmental Impact
Statement, DOE/EIS-0203-F, April 1995.
DOE95b U.S. Department of Energy, Office of Scientific and Technical Information,
Nuclear Reactors Built, Being Built, or Planned: 1994, DOE/OSTI-8200-R58,
August 1995.
DOE95c U.S. Department of Energy, Draft Waste Management Programmatic
Environmental Impact Statement for Managing Treatment, Storage, and Disposal
of Radioactive and Hazardous Waste, DOE/EIS-0200-D, August 1995.
DOE95d U.S. Department of Energy, Integrated Data Base Report-1994: U.S. Spent
Nuclear Fuel and Radioactive Waste Inventories, Projections, and
Characteristics, Revision 10, September 1995.
DOE95e U.S. Department of Energy, Integrated Data Base Report-1994: U.S. Spent
Nuclear Fuel and Radioactive Waste Inventories, Projections, and
Characteristics, Revision 11, September 1995.
DOE95f U.S. Department of Energy, Energy Information Administration, World Nuclear
Outlook 1995, DOE/EIA-0436(95), October 1995.
DOE95g U.S. Department of Energy, Total System Performance Assessment -1995: An
Evaluation of the Potential Yucca Mountain Repository, TRW Environmental
Safety Systems, Inc., BOOOOOOO-01717-2200-00136, Revision 01, November
1995.
DOE96a U.S. Department of Energy, Integrated Data Base Report - 1995, Revision 12,
December 1996.
5-17"
-------
DOE98 U.S. Department of Energy, Viability Assessment of a Repository at Yucca
Mountain, DOE/RW-0508, December 1998.
DOE98a U.S. Department of Energy, Report to Congress: Treatment and Immobilization of
Hanford Radioactive Waste, available on Hanford website at www.hanford.org.
EPA85 U.S. Environmental Protection Agency, Draft Environmental Impact Statement
for 40 CFR Part 191: Environmental Standards for Management and Disposal of
Spent Nuclear Fuel, High-level and Transuranic Radioactive Wastes, EPA
520/1-85-023, August 1985.
NRC88 U.S. Nuclear Regulatory Commission, Code of Federal Regulations, Title 10, Part
60, Disposal of High-Level Radioactive Wastes in Geologic Repositories, as
amended, October 1988.
NWP83 Nuclear Waste Policy Act of J982, Public Law 97-425, January 7, 1983.
ROD86 Roddy, J.W., H.C. Claiborne, R.C. Ashline, PJ. Johnson, and B.T. Rhyne,
Physical and Decay Characteristics of Commercial LWR Spent Fuel, ORNL/TM-
9591/V2&R1, Oak Ridge National Laboratory, Oak Ridge, TN, 1986.
WCM95a Spent Fuel Goes to Idaho Following State-DOE-Navy Agreement, Weapons
Complex Monitor, October 26,1995.
WCM95b OE Chooses Reprocessing for Selected Spent Fuel, Weapons Complex Monitor,
December 13, 1995.
5-18
-------
CHAPTER 6
DOSE AND RISK ESTIMATION
6.1 INTRODUCTION
Ionizing radiation emitted by the radioactive decay of nuclides released into the environment
poses a risk of inducing excess cancers or heritable genetic effects in exposed humans. The curie
(Ci) and becquerel (Bq) are units used to measure the activity of radioactive material, i.e., the rate
atoms are giving off radiation or disintegrating. The curie is equal to 37 billion disintegrations
per second, while the becquerel is equal to one disintegration per second. Exposure can occur
through several "pathways," including inhalation, ingestion, or external irradiation by
radionuclides in the air or deposited on the ground (see Chapter 8).
The risk of a health effect being induced in an exposed individual by a given exposure is
calculated by first estimating the radiation dose to sensitive tissues in the individual, as a function
of age. Depending on the radionuclide in question, its chemical form, and the exposure pathway,
its distribution will vary within the body and with time, leading to a variation in radiation dose
with organ and across time. The dose per unit exposure is referred to as a "dose conversion
factor" (DCF). From the tissue-specific doses, the risks of a radiation-induced cancer, cancer
death, or genetic effect are calculated using age- and organ-specific "risk factors." The dose
conversion and risk factors are generally calculated from models, as outlined below. The number
of excess cancers in a population is projected using a life-table calculation (BUN81, EPA94),
which corrects for competing causes of death.
6.2 DOSE ESTIMATION
The risk of inducing a cancer in a specific tissue or organ increases with the absorbed dose, i.e.,
the amount of ionization and excitation energy per unit mass deposited in that tissue or organ.
The risk of inducing a genetic effect increases with dose to the testes or ovaries. The absorbed
dose, D, is expressed in gray (Gy) or rad, where 1 Gy = 100 rad. The risk also depends on the
density of ionizations (the number of ionizations per unit path length) produced by the radiation.
The density of ionizations is directly related to the "linear energy transfer" (LET), which is a
measure of the amount of energy per unit path length deposited by a charged particle track in
traversing a material. When the density of ionizations is high, the radiation is referred to as
"high-LET"; conversely, "low-LET" radiation refers to that which is sparsely ionizing.
6-1
-------
Accordingly, a derived quantity called the effective dose is introduced, which is expressed in
units of sie verts (Sv) or rera. The effective dose in a tissue is given by QXD, where Q is a quality
factor (unitless) defined for a specific type of radiation.
Note that the absorbed dose is a physical quantity, but that the effective dose is a regulatory
concept determined in part by the choice of Q. Values for Q are assigned based on
radiobiological information on the relative biological effectiveness (RBE) of different types of
radiation. Since the RBEs of different types of radiation are not known precisely, the assignment
of Q rests heavily on the subjective judgments of experts on the ICRP. This document is
concerned only with: (1) low-LET radiation from beta particles, gamma rays, or energetic X-rays,
for which Q is taken to be unity and (2) high-LET alpha particles for which Q is taken to be 20
(ICR91, EPA94). In the case of low-LET radiation, 1 Sv = 1 Gy, and 1 rem = 1 rad. It follows
that 1 Sv= 100 rem.
For regulatory purposes, it is useful to introduce certain other measures of "dose". First, there is
the concept of the effective dose equivalent (EDE), which allows one to combine the dose
equivalents to different organs into a single quantity. In this connection, each target organ, /', is
assigned a weighting factor, wt, which roughly represents the estimated proportion of the risk
from a uniform, whole-body irradiation occurring in that particular organ. The effective dose
equivalent is then the weighted sum of doses to the individual organs (ICR77):
Second, in dealing with internally deposited radionuclides that remain in the body and irradiate
tissues for extended periods of time, the concept of" committed dose" is introduced (ICR77). For
example, the 50-yr committed effective dose equivalent (CEDE) from a given intake is the
calculated total EDE received over a 50-yr period following that intake. Finally, the annual
committed effective dose equivalent (annual CEDE) refers to the CEDE resulting from one year's
exposure or intake.
When the exposure is external, the dose calculation is a straightforward application of radiation
physics. The radiation doses to target organs in an idealized "reference man" are calculated from
the decay properties of the radionuclides and the well-understood interactions of radiation with
matter (ICR79, EPA89).
6-2
-------
For ingested or inhaled radionuclides, the dosimetry modeling is more complex. It is necessary
to incorporate biokinetic information to describe the distribution and retention of the
radionuclide (and any radioactive decay products) in the body as a function of time after intake.
The irradiation of target tissues by internally deposited radionuclides is further complicated by
the need to consider the cross irradiation of one tissue by radionuclides deposited in another
tissue. Dosimetry models for internally deposited radionuclides have been developed by the
International Commission on Radiological Protection (ICR79, ICR80, ICR81, ICR88). Dose
conversion factors for internal and external radionuclide exposures are tabulated hi EPA's
Federal Guidance Reports Nos. 11 and 12, respectively (EPA88, EPA93).
6.3 CANCER RISK ESTIMATION
EPA's current model for estimating radiogenic cancer risks incorporates age- and organ-specific
risk coefficients for low-LET radiation based on data obtained from the Japanese atomic bomb
survivors up through 1985, supplemented by organ-specific data from other sources (e.g., breast
cancer induction in fluoroscopy patients). For most cancer sites, EPA's methodology involves an
averaging of two sets of coefficients, reflecting two different ways of projecting risk from the
atomic bomb survivors to the U.S. population, which have significantly different baseline rates of
specific cancers (LAN91, EPA94, EPA98, EPA99).
Aside from breast cancer, for which there is good epidemiological evidence that the dose
response is approximately linear and independent of fractionation (NAS90), it was assumed that
the risks at low doses and dose rates are reduced by a "dose, dose rate effectiveness factor"
(DDREF) of 2 compared to the acute high dose exposures experienced by the bomb survivors.
The value of 2 for the DDREF is consistent with ICRP recommendations (ICR91). For low dose
(or dose rate) conditions, the calculated risk of a premature cancer death attributable to uniform,
whole-body, low-LET irradiation is about 5.75><10"2/Gy. Neglecting nonfatal skin cancers, which
are usually not serious, the corresponding incidence risk estimate is 8.5xlO~2/Gy (EPA99).
High-LET (alpha particle) risks are presumed to increase linearly with dose and to be
independent of dose rate. Except for leukemia and breast cancer, a relative biological
effectiveness (RBE) factor of 20 is adopted for estimating the risk of high-LET radiation relative
to that for low-LET radiation at low dose or dose rate conditions. Again the RBE value of 20 is
consistent with the recommendations of the ICRP (ICR91). In view of epidemiological data on
people ingesting or being injected with alpha-emitting radionuclides that deposit in bone, an
6-3
-------
effective RBE of 1 was adopted for leukemia; for breast cancer, the high-LET RBE of 10 is used
to be consistent with the DDREF of 1 adopted for this site.
The lifetime excess risks of cancer incidence and mortality, for constant exposure rates to over
100 different radiomiclides, are tabulated in the Interim Version of EPA's Federal Guidance
Report No. 13 (EPA98). The dosimetry models employed in deriving these risk estimates reflect
new ICRP recommendations and incorporate age-specific biological parameters (EPA98).
6.4 GENETIC EFFECTS
Genetic effects of radiation exposure are defined as stable, heritable changes induced in the germ
cells (eggs or sperm) of exposed individuals, which are transmitted to and expressed only in their
progeny across future generations.
The genetic risk of radiation exposure is more subtle than the somatic risk since it does not affect
the persons exposed, but only their progeny. Somatic effects are expressed in the exposed
individual over the person's remaining lifetime, while about 30 subsequent generations (nearly
1,000 years) are needed for near complete expression of genetic effects. Genetic risk is incurred
by fertile people when radiation damages the DNA of the germ cells. The damage, in the form of
a mutation or a chromosomal change, is transmitted to, and may be expressed in, a child
conceived after the radiation exposure. However, the damage may also be expressed in some
subsequent generation(s) or never.
Estimates of the genetic risk per generation are conventionally based on a 30-year reproductive
generation. That is, the median parental age for conception of children is defined as age 30
(approximately one-half the children are produced by persons less than age 30, the other half by
persons over age 30). Thus, the radiation dose accumulated from birth to age 30 is used to
estimate the .genetic risks. A basic assumption in assessing radiation genetic risk is that, at low
doses and low dose-rates of low-LET radiation, there is a linear relationship between dose and
the probability of occurrence of the genetic effect.
In the EPA Background Information Document for Radionuclides (EPA84), direct and indirect
methods for obtaining genetic risk coefficients are described, and some recent estimates based on
these methods are tabulated. Briefly, the direct method takes the frequency of mutation or
occurrence of a heritable defect per unit dose observed in animal studies and extrapolates to what
6-4
-------
is expected for humans. These direct estimates are usually used for first generation effects
estimates.
The EPA assessment of risks of genetic effects includes both first generation estimates and total
genetic burden estimates. In developing risk coefficients for genetic effects, EPA has employed
traditional definitions of genetic effects and dose-response relationships. Although the newly
recognized mechanisms of genetic change listed above have future implications for genetic risk
assessment, there are no data upon which to base radiation risk coefficients for these kinds of
damage at this time.
In the NESHAPs Environmental Impact Statement (EPA89), the EPA estimated the low dose-
rate, low-LET doubling dose for genetic effects to be 1.0 Gy (100 rad). That is, 1.0 Gy per
reproductive generation (considered to be 30 years) would double the rate of occurrence of
congenital defects (a defect existing at birth but not hereditary). However, at that time, the
Agency indicated, based on limited human data, that the true doubling dose might be about 3
times greater. There is still no consensus on this point.
Neel and Lewis reviewed untoward pregnancy outcomes (UPOs) in the Japanese A-bomb
survivors and compared them to mouse genetic effects data (NEE90a). The gametic doubling
dose for low dose-rate, low-LET radiation in man, in this case, would be 400 rad (NEE90a). In a
companion analysis of mouse genetic data, they estimated a gametic doubling dose in mice of
135 (16-400) rad. The gametic doubling dose for a study where only one sex was irradiated
provides an analog of the "conjoint" parental gonadal dose for comparison purposes. However,
for mice, they recommended a dose-rate factor of 3 for low dose-rate, low-LET radiation, so the
doubling dose would also be 400 rad in mice (NEE90a).
UNSCEAR reviewed the recommendations listed above and concluded that the doubling dose in
humans is most likely between 1.7 and 2.2 Sv (170 and 220 rad) for acute exposure to low-LET
radiation, but 4.0 Sv (400 rad) for chronic exposure (UNS93). However, the UNSCEAR report
also continued to estimate the hereditary effects of exposure to ionizing radiation using a
doubling dose of 1.0 Sv (100 rad), just as in earlier UNSCEAR reports (UNS86, UNS88).
The. EPA assumes a doubling dose of 100 rad (1 Sv) in this document, but again notes that some
estimates of the doubling dose is about four times greater. The EPA estimate for equilibrium
effects is about twice that of recent estimates by BEIR V and UNSCEAR because EPA included
6-5
-------
a value for equilibrium multifactorial effects where these others did not. The EPA estimates
incorporate a dose-rate factor of 3 for low-LET radiation as reported in the 1977 UNSCEAR
Report (UNS77).
The projected genetic effects attributable to a given population exposure depend on the
population dynamics of future generations. However, if a stationary population is assumed, the
number of effects can be derived from Table 6-1. The dose in the table is that received by
parents in the first 30 years of life, the assumed generation period. Since the average lifetime of
a person in the 1980 stationary population is about 75 years, 40 percent (30/75) of the population
dose is considered to be genetically significant Thus, to calculate genetic risk coefficients
comparable to the cancer risk coefficients cited above, the values in Table 6-1 should be
multiplied by 0.4. On this basis, eight serious heritable disorders are expected in the first
generation following a 104 person-Gy population exposure of low dose (or dose rate), low-LET
radiation, and 104 such effects would be expected over all generations. The number of serious
genetic effects projected over all generations is then about 20 percent of the excess fatal cancers
projected in the exposed population.
Table 6-1. Estimated Frequency of Genetic Disorders in a Birth Cohort Due to Exposure of
Each of the Parents to 0.01 Gy (1 rad) per Reproductive Generation (30 yr)
Low Dose Rate, Low-LET
High Dose Rate, Low-LET
6.5 DEVELOPMENTAL EFFECTS
6.5.1 In Utero Carcinogenesis
Studies of the effects of in utero X-ray exposures in the U.K. in the 1960s showed increased
childhood cancer as a sequela. The BEIR III Committee reviewed the data and estimated that
there was a risk of 25x10"* excess fatal leukemias per year per Gy exposure (25 x 10"6 per rad) and
6-6
-------
28x lO"4 excess fatal cancers of other types (28x 10'6 per rad) (NAS80). The risk starts at birth and
continues for 12 years for leukemias and 10 years for solid tumors (NAS80). Having reviewed
additional data, the BEIR V Committee estimated that the risk was "... about 200 to 250 excess
fatal cancer deaths xiO"4 per Gy [200 to 250 x 10'6 per rad] in the first 10 years of life...." It also
estimated one-half would be leukemias and one-quarter tumors of the nervous system (NAS90).
UNSCEAR estimated a risk of leukemia and solid tumors expressed during the first 10 years of
life of 2x10"4 per rad (UNS86). The NRPB estimated a cancer risk of 2.5 xlO"4 cases of leukemia
and 3.5 x 10"4 cases of solid tumors per rad of in utero exposure (STA88). The NRPB hi 1993
retained the same cancer risk estimates but concluded about one-half the cases would be fatal and
they would be expressed in the first 15 years of life (NRP93). However, the NRPB also
estimated the lifetime risk would be four times greater than that of the first 15 years (NRP93).
6.5.2 Brain Teratology
The ICRP published an excellent review of the biology and the possible mechanisms of
occurrence of radiation-induced brain damage in utero (ICR86). ICRP estimates: 1) for
exposures from the 8th through the 15th week after conception, the risk of severe mental
retardation is 4x10'' per Gy (4xlO'3 per rad), with a confidence interval of 2.5x10"' to 5.5x10''
per Gy (2.5xlO'3 to 5.5x 10"3 per rad) and 2) for exposures from the 16th through the 25th week
after conception, the risk of severe mental retardation is 1 xlO'1 per Gy (1 x 10'3 per rad).
However, a threshold below 50 rad could not be excluded (ICR86).
Effects other than mental retardation and microcephaly have been noted in the Japanese A-bomb
survivors. Schull et al. (SCH88) reported that in individuals exposed prenatally between weeks 8
and 25 of gestation there is a progressive shift downward in IQ score with increasing exposure
and that the most sensitive group is between 8 and 15 weeks gestational age at time of exposure.
The BEIR V Committee estimated a 30 point loss in IQ per Gy exposure (0.3 points per rad)
consistent with a linear nonthreshold relationship (NAS90). However, even if the effect is linear-
nonthreshold, the response would be too small to be detectable at environmental exposure levels.
Much the same pattern was reported for average school performance, especially in the earliest
years of schooling (OTA88). Finally, a linear-nonthreshold relationship between exposure and
incidence of unprovoked seizures hi later life has been found to be consistent with the data for
individuals exposed between 8 and 15 weeks gestational age (DUN88).
6-7
-------
In 1986, the United Nations Scientific Committee on the Effects of Atomic Radiation also
reviewed the question of mental retardation as a part of the overall review of the biological
effects of prenatal radiation exposure (UNS86). UNSCEAR, like the ICRP, concluded there was
a risk of severe mental retardation of 4* 10"3 per rad over the period of 8 to 15 weeks after
conception and of 1 x 1O'3 per rad over the period 16 to 25 weeks after conception (UNS86).
The question of a threshold for central nervous system effects, particularly for the 8 to 15 week
period of gestation, is unresolved. Apparent thresholds in the human data may merely reflect the
statistical uncertainty due to the small number of cases. If, as has been suggested, the effects are
due to improper synaptogenesis in the brain (temporal or spatial) (ICR86, OTA87), it should be
noted that significant prolongation of cell cycle in matrix cells of the developing telencephalon in
mice (exposed on day 13 of gestation) has been reported following exposures as low as 10 R
(KAM78). Exposure of mice to 1 R on day 13 of gestation resulted in an increase in eye and
brain abnormalities, but the increase was not statistically significant (MIC78).
6.5.3 Other Effects of Prenatal Irradiation
UNSCEAR estimated: (1) a pre-implantation loss of 1 x 10"2 per rad during the first two weeks
after conception and (2) a malformation risk of 5* 10"3 per rad during weeks 2 to 8 after
conception (UNS86).
For many of the teratologic effects observed, no threshold has been demonstrated. If a
teratogenic effect of radiation is due to cell-killing effects, then a threshold for that effect is
probable. While early studies of radiation as a teratogen used high exposures and probably
induced effects through cell killing, cell killing may not be required. Patrick cites Zwilling as
follows: "... developmental anomalies appear to be caused by 'failure of proper tissue interaction
to occur'" (PAT78, ZWI63). For example, a somatic mutation in a single cell, perhaps through
clonal expansion, could cause improper tissue interaction with no loss of cells; or, killing a single
cell could cause release of a toxicant that causes an improper local interaction (RUS54, WEI54).
Jacobsen exposed pregnant mice to 0, 5R, 20R, or 100R on day 8 of gestation and scored skeletal
abnormalities on day 19. He interpreted the dose-effect curve as linear or nearly so and saw no
evidence of a threshold for the types of damage studied (JAC70). He stated: "The observations
made, and in particular that concerning the apparent absence of a threshold dose, indicate that it
is not justified to assume that irradiation with doses of 5 R and less is entirely without effect on
6-8
-------
the human embryo in early developmental stages" (JAC70). In another study, exposure of mice
to 1 R on day 8 of gestation resulted in a significantly higher incidence of malformed and
retarded fetuses compared to controls (MIC78). A 1981 review of data on the effects of ionizing
radiation on the developing embryo/fetus reached essentially the same conclusions as Jacobsen
(HHS81). Given the large number of experimental animals that would be required, direct
evidence for a threshold below 5 rad will be difficult to provide.
6.5.4 Summary of Developmental Effects
EPA risk coefficients for estimating prenatal carcinogenic, teratologic, and nonstochastic effects
in man (see Table 6-2) are, with one exception, the same as those published in the 1989
NESHAPs BID (EPA89). The first entry in the corresponding table in the NESHAPs BID lists
"Fatal Cancer" as 6.0* 10"4. The entry should be for "Cancer Incidence." The fatal cancer risk is
about half as great, 3 x 1 (T4.
Table 6-2. Possible Effects of In Utero Radiation Exposure
T^fffKi^^Conc^m *
Cancer Incidence
Mental Retardation" (exposure at 8-15 weeks)
Mental Retardation11 (exposure at 16-25 weeks)
Malformation1" (exposure at 2-8 weeks)
Pre-implantation Loss (exposure at 0-2 weeks)
Iti* ported H '
6x10-"
4x10°
IxlO'3
5x10°
lxl(r2
A threshold for mental retardation following exposure at 8-15 weeks of gestational age may depend on the
mechanism of action.
b A threshold is expected for mental retardation following exposure during the 16-25 week period of gestation and
for many types of malformations following exposures at early gestational age.
6-9
-------
REFERENCES
BUNS 1 Banger, B., JR. Cook and M.K. Barrick, Life Table Methodology for Evaluating
Radiation Risk: An Application Based on Occupational Exposure, Health Phys.,
40:439-455, 1981.
DUN88 Dunn, K., H. Yoshimaru, M. Otake, J.F. Annegers and W.J. Schuil, Prenatal
Exposure to Ionizing Radiation and Subsequent Development of Seizures,
Technical Report RERF TR 13-88, Radiation Effects Research Foundation,
Hiroshima, 1988.
EPA84 U.S. Environmental Protection Agency, Radionuclides, Background Information
Document for Final Rules, Volume /, Office of Radiation Programs, EPA Report
520/1-84-022-1, 1984.
EPA88 U.S. Environmental Protection Agency, Limiting Values of Radionuclide Intake
and Air Concentration and Dose Conversion Factors for Inhalation, Submersion,
andIngestion, Federal Guidance Report No. 11,EPA-520/1-88-020,1989.
EPA89 U.S. Environmental Protection Agency, Risk Assessment Methodology,
Environmental Impact Statement for NESHAPs Radionuclides, Volume I,
Background Information Document, Office of Radiation Programs, EPA Report
520/l-89-005,Washington, DC 1989.
EPA93 U.S. Environmental Protection Agency, External Exposure to Radionuclides in
Air, Water, and Soil, Federal Guidance Report No. 12, EPA-402-R-93-081,1993.
EPA94 U.S. Environmental Protection Agency, Estimating Radiogenic Cancer Risks,
Office of Radiation and Indoor Air, EPA Report 402-R-93-076, Washington, DC,
1994.
EPA98 U.S. Environmental Protection Agency, Health Risks From Low-Level
Environmental Exposure to Radionuclides. Federal Guidance Report No. 13 -
Part 1-Interim Version, EPA Report 402-R-97-014, 1998.
EPA99 U.S. Environmental Protection Agency, Estimating Radiogenic Cancer Risks,
Addendum: Uncertainty Analysis, EPA Report 402-R-99-003, 1999.
HHS81 Health and Human Services, Effects of Ionizing Radiation on the Developing
Embryo and Fetus, A Review, Bureau of Radiological Health, Public Health
Service, Food and Drug Administration, HHS Publication FDA 81-8170,
Rockville, MD, 1981.
6-10
-------
ICR79 International Commission on Radiological Protection, Limits for Intakes of
Radionuclides by Workers, ICRP Publication No. 30, Part 1, Annals of the ICRP,
2(3/4), Pergamon Press, Oxford, 1979.
ICR80 International Commission on Radiological Protection, Limits for Intakes of
Radionuclides by Workers, ICRP Publication No. 30, Part 2, Annals of the ICRP,
4(3/4), Pergamon Press, Oxford, 1980.
ICR81 International Commission on Radiological Protection, Limits for Intakes of
Radionuclides by Workers, ICRP Publication No. 30, Part 3, Annals of the ICRP,
6(2/3), Pergamon Press, Oxford, 1981.
ICR86 International Commission on Radiological Protection, Developmental Effects of
Irradiation on the Brain of the Embryo and Fetus, ICRP Publication 49, Annals
of the ICRP, !£(4): 1-43, Pergamon Press, Oxford, 1986.
ICR88 International Commission on Radiological Protection, Limits for Intakes of
Radionuclides by Workers: an Addendum, ICRP Publication No. 30, Part 4,
Annals of the ICRP, 12(4), Pergamon Press, Oxford, 1988.
ICR91 International Commission on Radiological Protection, 1990 Recommendations of
the International Commission on Radiological Protection, ICRP Publication 60,
Annals of the ICRP, 21(1-3), Pergamon Press, Oxford, 1991.
JAC70 Jacobsen, L., Radiation Induced Fetal Damage, Adv. Teratol. 4, 95-124, 1970.
KAM78 Kameyama, Y., K. Hoshino and Y. Hayashi, Effects of Low-Dose X-Radiation on
the Matrix Cells in the Telencephalon of Mouse Embryos, pp. 228-236, in:
Developmental Toxicology of Energy-Related Pollutants CONF-771017, DOE
Symposium Series 47, Pacific Northwest Laboratories, Richland, WA, 1978.
LAN91 Land, C.E. and W.K. Sinclair, The Relative Contributions of Different Organ
Sites to the Total Cancer Mortality Associated-with Low-Dose Radiation
Exposure, in: Risks Associated with Ionizing Radiations, Annals of the ICRP
22( 1), Pergamon Press, Oxford, 1991.
MIC78 Michel, C. and H. Fritz-Niggli, Radiation-Induced Developmental Anomalies in
Mammalian Embryos by Low Doses and Interaction with Drugs, Stress and
Genetic Factors, pp. 397-408, in: Late Biological Effects of Ionizing Radiation
Vol. II, International Atomic Energy Agency, Vienna, 1978.
6-11
-------
NAS80 National Academy of Sciences/National Research Council, The Effects on
Populations of Exposure to Low Levels of Ionizing Radiation (BEIR III), National
Academy Press, Washington, DC, 1980.
NAS90 National Academy of Sciences/National Research Council, Health Effects of
Exposure to Low Levels of Ionizing Radiation (BEIR V), National Academy Press,
Washington, DC, 1990.
NEE90a Neel, J.V. and S.E. Lewis, The Comparative Radiation Genetics of Humans and
Mice, Annual Review of Genetics, 24, 327-362,1990. [reprinted pp. 451-486 in:
The Children of Atomic Bomb Survivors, A Genetic Study, J.V. Neel and WJ.
Schull, eds., National Academy Press, Washington, DC, 1990.]
NEE90b Neel, J.V., W.J. Schull, A.A. Awa, C. Satoh, H. Kato, M. Otake and Y.
Yoshimoto, The Children of Parents Exposed to Atomic Bombs: Estimates of the
Genetic Doubling Dose of Radiation for Humans, Am. J. Hum. Genetics, 46,:
1053-1072, 1990. [reprinted pp. 431-450 in: The Children of Atomic Bomb
Survivors, A Genetic Study, J.V. Neel and W.J. Schull, eds., National Academy
Press, Washington, DC, 1991.]
NRP93 National Radiological Protection Board of the UK, Estimates of Late Radiation
Risks to the UK Population, in: Documents of the NRPB, Volume 4, Number 4,
Chilton, England, 1993.
OTA87 Otake, M., H. Yoshimaru and W.J. Schull. Severe Mental Retardation Among the
Prenatally Exposed Survivors of the Atomic Bombing of Hiroshima and
Nagasaki: A Comparison of the T65DR and DS86 Dosimetry Systems, Technical
Report RERF TR16-87, Radiation Effects Research Foundation, Hiroshima,
1987.
OTA88 Otake, M., W.J. Schull, Y. Fujikoshi, and H. Yoshimaru, Effect on School
Performance of Prenatal Exposure to Ionizing Radiation: A Comparison of the
T65DR and DS86 Dosimetry Systems, Technical Report RERF TR 2-88,
Radiation Effects Research Foundation, Hiroshima, 1988.
PAT78 Patrick, C.H., Developmental Toxicology as Input to the Methodology for Human
Studies of Energy-Related Pollutants, pp. 425-440, in: Developmental Toxicology
of Energy-Related Pollutants, CONF-771017, DOE Symposium Series 47, Pacific
Northwest Laboratories, Richland, WA, 1978.
RUS54 Russell, L.B. and W.L. Russell, An Analysis of the Changing Radiation Response
of the Developing Mouse Embryo, pp. 103-149, in: Symposium on Effects of
6-12
-------
Radiation and Other Deleterious Agents on Embryonic Development, J. Cell.
Comp. Physiol., 43_, Supplement 1, May 1954.
SCH88 Schull, W.J., M. Otaki, and H. Yoshimaru, Effects on Intelligence Test Score of
Prenatal Exposure to Ionizing Radiation in Hiroshima and Nagasaki: A
Comparison of the T65DR andDS86 Dosimetry Systems, Technical Report RERF
TR 3-88, Radiation Effects Research Foundation, Hiroshima, 1988.
STA88 Stather, J.W., C.R. Muirhead, A.A. Edwards, J.D. Harrison, D.C. Lloyd, and N.R.
Wood, Health Effects Models Developed, in: 1988 UNSCEAR Report, NRPB-
R226, National Radiation Protection Board, Chilton, England, 1988.
UNS77 United Nations Scientific Committee on the Effects of Atomic Radiation, Sources
and Effects of Ionizing Radiation, Report to the General Assembly, with Annexes,
Sales No. E.77.IX.L, United Nations, New York, 1977.
UNS86 United Nations Scientific Committee on the Effects of Atomic Radiation, Genetic
and Somatic Effects of Ionizing Radiation, 1986 Report to the General Assembly,
Sales No. E.86.IX.9., United Nations, New York, 1986.
UNS88 United Nations Scientific Committee on the Effects of Atomic Radiation,
Sources, Effects and Risks of Ionizing Radiation, 1988 Report to the General
Assembly, Sales No. E.88.IX.7., United Nations, New York, 1988.
UNS93 United Nations Scientific Committee on the Effects of Atomic Radiation, Sources
and Effects of Ionizing Radiation, 1993 Report to the General Assembly, Sales
No. E.94.IX.2., United Nations, New York, 1993.
WEI54 Weiss, P., Summarizing Remarks, pp. 329-331, in: Symposium on Effects of
Radiation and Other Deleterious Agents on Embryonic Development, J. Cell.
Comp. Physiol., 43_, Supplement 1, May 1954.
ZWI63 Zwilling, E., Cell Differentiation and Embryogenesis, pp.75-90, in: Birth Defects.
M. Fishbein, ed., J.B. Lippincott Co., Philadelphia, 1963 (cited by Patrick in
PAT78).
6-13
-------
CHAPTER?
CURRENT INFORMATION CONCERNING A POTENTIAL WASTE REPOSITORY AT
YUCCA MOUNTAIN
7.1 PRINCIPAL FEATURES OF THE NATURAL ENVIRONMENT
This section describes the principal features of the natural environment at Yucca Mountain and
the surrounding area. This information is based primarily on the site characterization work of the
Department of Energy (DOE). Particular emphasis is given to those aspects of the geology,
mineralogy, structure, hydrology, and climate of the site that are most likely to affect the
performance of a high-level waste repository. The glossary of technical terms at the end of this
BID should be helpful to the reader.
7.1.1 Geologic Features
A description of the important features of Yucca Mountain and the surrounding area provides a
picture of the geologic setting that serves as the context for understanding the repository design.
Important aspects of the geology around the site, such as the presence of faults, seismicity, and
the nature and distribution of rock types, are discussed.
7.1.1.1 Location and Principal Physical Features of the Site (Adapted from DOE95a)
The Yucca Mountain site is located hi Nye County, Nevada approximately 150 kilometers (km)
northwest of Las Vegas, Nevada (Figure 7-1). The site is at the southwestern boundaries of the
Nevada Test Site and the adjoining Nellis Air Force Base and about 50 km east of Death Valley
National Monument. The Yucca Mountain Region includes the southern Great Basin in southern
Nevada and an adjacent area in California (Figure 7-2). The Great Basin, which is in the
northern portion of the Basin and Range physiographic province, is bounded geologically by the
margins of the Colorado Plateau to the east and southeast, by the Sierra Nevada and Transverse
Ranges to the west and south, and by the Snake River Plain and flood basalts of the Columbia
Plateau to the north. Typical Great Basin topography consists of north-south mountain ranges
separating narrow structural valleys with internal drainages. The Colorado River, flowing along
the margin of the Colorado Plateau and topographically isolated from Yucca Mountain, provides
the only external drainage. Yucca Mountain is situated in the southern section of the Great
7-1
-------
Basin, in the Southwest Nevada Volcanic Field (SNVF). This area is bounded on the south by
the Death Valley region and the Mojave Desert of California. Yucca Mountain is a narrow ridge
which trends north-south and extends approximately 20 km from the southern margin of the
Timber Mountain caldera complex. The area is mapped on the following U.S. Geological
Survey 7.5-minute topographic quadrangles: Amargosa Valley, Big Dune, Busted Butte, Crater
Flat, East of Brady Mountain, and Pinnacles Ridge (formerly Topopah Spring NW).
Figure 7-1. Location of Yucca Mountain (DOE94a)
7-2
-------
,03"
ttl'
Figure 7-2. Boundaries and Larger Subdivisions of the Basin and Range Physiographic
Province. Province boundary is indicated by heavy solid line (HUN74)
Yucca Mountain is an irregularly shaped upland, six to 10 km wide and about 40 km long.
Uplands in the Yucca Mountain area are composed of ridge crests, valley bottoms, and
intervening hill slopes (DOES 8) with dominantly north-trending echelon ridges and valleys
controlled by high-angled faults. The fault blocks, composed mostly of welded fine-grained
volcanic rocks, are tilted eastward. As a result, the fault-bounded west-facing slopes are
generally high, steep, and straight, whereas the east-facing slopes are more gentle and usually
deeply dissected. Except where protected by a resistant rock layer capping the lip slopes, the
ridge crests are mostly angular and eroded. Valleys range from shallow, straight, steeply sloping
gullies and ravines to relatively steep, bifurcating, gently sloping valleys and canyons. Hill
slopes are typically narrow and moderately steep near the crest, with progressively gentler slopes
toward the valley floor. The crest elevation of Yucca Mountain ranges between 1,500 and 1,930
meters (m) above sea level. The summit is about 650 m above the floors of adjacent washes in
Crater and Jackass Flats.
7-3
-------
The main drainage system for the Yucca Mountain area, including the Timber Mountain area, the
Calico Hills, and the mesas lying to the south of Timber Mountain, is in the Amargosa Valley.
This drainage, east of Beatty, Nevada, carries runoff from the region south through the Tecopa
basin into the southern part of Death Valley. The Amargosa Valley carries significant runoff
only after extraordinarily heavy precipitation. There are no perennial streams or natural bodies
of surface water on or adjacent to the Yucca Mountain, The major drainages, Solitario Canyon
on the west, Forty Mile Wash on the east, and tributary drainages are primarily on the east flank
of the mountain and flow only briefly immediately after rainstorms (Figure 7-3).
Bedrock exposures are common at higher elevations in the Yucca Mountain Region. Many of
the hill slopes have a discontinuous veneer of blocky talus and wedges of colluvium cover the
lower hill slopes. The rates of erosion in the Yucca Mountain area are lower than in similar arid
areas in the southwestern U.S. and other parts of the world. Conditions contributing to these low
erosion rates include existence of fine-grained volcanic rocks which are relatively erosion-
resistant, insufficient runoff during interpluvial periods to remove hillslope colluvium, and
topography that has not been significantly affected by Quaternary tectonic activity (WHI93).
Regional erosion projections over 10,000 years are less than one meter of down cutting in
canyons above the potential repository block, and less than 0.02 m of slope retreat (DOE95a).
7.1.1.2 Geologic History of the Region (Adapted from DOE95a)
The physiography and geomorphic features in the Yucca Mountain area influence the
characteristics of the surface water system, and to some extent, the ground-water system as well.
The flow of water into, within, and around a repository at Yucca Mountain would directly affect
its ability to contain the waste over time. The composition and chemical behavior of ground
water at Yucca Mountain will be affected by the type, size, and abundances of primary and
secondary mineral phases in the contacting rock formations. Furthermore, the geologic processes
and events important to repository performance and design can only be understood within the
broader context of the geologic history of the region. Current and future geologic processes and
events are a direct product of the area's geologic history; projecting their effect on repository
performance requires an understanding of causes, frequencies, durations, and magnitudes over
time. For example, projecting the potential frequency and magnitude of earthquakes is based on
the historical record of past seismic activity. This information has been developed from records
7-4
-------
Wlp^%v' t*F%
Figure 7-3. Physiographic Features in the Yucca Mountain Site Area (DOE88)
7-5
-------
of past seismicity and geologic studies on the effects of faulting (displacement of strata across
faults, topographic features, etc.) in the vicinity of the site.
In general terms, the Yucca Mountain Region is characterized by a thick section of Precambrian
and Paleozoic sedimentary rocks overlain by a sequence of Tertiary silicic volcanic rocks (see
Figure 7-4). The older rocks have been folded and faulted by a compressional tectonic process
and the entire stratigraphic section subsequently deformed by extensional basin-and-range
tectonics. Uplifted ranges, such as Yucca Mountain, are separated by basins partially filled with
alluvial deposits.
A basement complex of older Precambrian metamorphic and younger Precambrian igneous rocks
is presumed to underlie the area. The basement rocks are overlain by a westward-thickening
accumulation of shallow marine late Precambrian and early Cambrian marine sediments,
quartzite, siltstone, shale, and carbonate rocks. These deposits are interpreted as a rifted
continental margin miogeosyncline, shown hi Figure 7-5, formed seaward of the highlands area.
These rocks are locally fossiliferous. Deposition that continued through the Devonian Period is
represented by carbonate and shale with interbedded quartzite and sandstone, thickening from up
to 500 meters in western Utah to at least 6,100 meters in central Nevada.
In late Devonian and early Mississippian time, the Antler Orogeny, a mountain-building event,
formed a north-northeast trending highland area adjacent to the Roberts Mountains Thrust.
Large volumes of sediments eroded from the highlands into a foreland basin in the eastern half of
the Great Basin, forming thick flysch14 deposits adjacent to the highlands and shallow-water shelf
carbonates to the east (Figure 7-6). Erosion of the highlands and deposition into the basin
continued through the Permian Period, decreasing as the mountain-building waned. In Mesozoic
and early Cenozoic tune, these rocks were folded and displaced along thrust faults with extensive
fracturing of the brittle rocks in the upper thrust plates. This faulting was accompanied by
intrusion of granitic stocks, uplift, and erosion of the land surface (DUD90).
'"Flysch deposits are typified by the widespread sandstones, marls, shales, and clays exemplified by deposits
occurring at the northern and southern borders of the Alps.
7-6
-------
'HYDROGEOLOGIC
UNITS
PERIOD
GEOWG/C FORMATIONS
LAVA FLOW &
VALLEY FILL
WELDED - TUFF
AQUIFERS
TUFF fc LAVA
FLOW
AOUITAROS
UPPER
CARBONATE
AQUIFER
UPPER
CLASTIC
AQUITARO
LOWER
CARBONATE
AQUIFER
LOWER
CLASTIC
ADUITARD
&
ruaccMC fMUtr
THfarr CMmm ruff
OtSMTS OF KMUCSA * SO/U. * OOUE HTK.
laairtm rurr
ruff
IUHUOHK romum*
FOUUITIOH
aea-co HMKX ruff
cuff em rur rurr
otecn rum
noes of fMirs
PUOflfSTOHCEHE
MID YOUHGER
(0 -5 He)
UOCEN£
(5 -24 Ua
OUOOCEHE
(24 -3? Uai
TlfflftH IMESTOHf.
CLCA** MftUO-IOM
oeyiis we uuEsmtc
HCVJOA romuroH
nr
cotemre
Mtrcwrc vMtff lutcsroue
wKEmf fomuroH
mam* UKCSTONE
HOPJIH mttuenoH
DUMIXABOG SHM£
OOHAZA ton roMurKm
POGOHIP GROUP
iMMSiae auMrzrre
tiaoo (
STOOMO cvHirzirc
oounarc
OUCOUS - UCTMOnfMK:
CetlPlEX
Figure 7-4. Generalized Regional Stratigraphic Column Showing Geologic Formations and
Hydrological Units in the Nevada Test Site Area (Modified from DOE95a). The
repository host rock at Yucca Mountain is in the Tertiary age Paint Brush Tuff.
7-7
-------
121
115"
109°
N
- EASTERN LIMIT OF LOWER PALEOZOIC EUCEOCL1NAL ROCKS
. UPPERMOST PROTEROZOIC AMD LOWER CAMBRIAN MIOGEOCLINAL ROCKS '
- MIDDLE CAH6RIAN THROUGH UPPER DEVONIAN MIOCEOCUNAL ROCKS
Figure 7-5. Late Precambrian Through Mid-Paleozoic Paleography of the Great Basin
(Modified from DOE95a)
7-8
-------
121*
115°
109
Figure 7-6. Late Devonian and Mississippian Paleogeography of the Great Basin
(Modified from DOE95a)
7-9
-------
Middle and late Cenozoic crustal uplifting and extension in the region occurred over an area
1,500 km long and 500 to 1,000 km wide. The stretching, estimated at 10 to 50 percent of the
original width and locally as great as 100 percent, resulted in large north-northeast fractures with
sliding and tilting of large crustal blocks, forming the characteristic structure and topography of
the Great Basin.
Accompanying these crustal adjustments, volcanic eruptions in the vicinity of Yucca Mountain
formed a series of calderas and deposited numerous thick beds of pyroclastics, tuff, and lava,
aggregating up to three km in thickness near Yucca Mountain. The major episodes of silicic
volcanism ceased about 7.5 million years ago (mega annum; Ma); however, smaller basaltic
eruptive centers formed in the basins adjacent to Yucca Mountain perhaps as recently as 4,000
years ago.
7.1.1.3 Stratigraphy of the Yucca Mountain Area (Adapted from DOE95a)
An understanding of the stratigraphy of the rocks at Yucca Mountain and the surrounding area is
important to: 1) designing and constructing the repository, 2) assessing the potential of the
natural barrier to retard the movement of radionuclides from the repository, and 3) describing the
expected behavior of ground water movement through these rocks. For example, the physical
properties of the rocks at the repository horizon determine the effects of heat generated by the
radioactive waste on the near-field environment in the postclosure time period. They can also
determine the speed at which radionuclides can be transported through the repository.
The stratigraphy of the southern Great Basin is highly varied, with formations ranging hi age
from Precambrian to Holocene, that is, from 500 million to less than 400,000 years. These rocks,
briefly described in Table 7-1, are divided into eight general groups based on age, lithology, and
history.
At Yucca Mountain, the stratigraphy is dominated by mid-Tertiary rocks of volcanic origin that
erupted from the southwestern Nevada volcanic field. The stratigraphic sequence can be divided
into four general categories based on similarities in lithology, age, and history of deposition or
emplacement: 1) pre-Cenozoic rocks, 2) mid-Tertiary pyroclastic rocks, 3) younger basalt, and
4) late Tertiary to late Quaternary surficial deposits (Figure 7-7). These categories are discussed
in the following sections.
7-10
-------
Table 7-1. Stratigraphy of the Southern Great Basin
Older Precambrian
Crystalline Rocks
These include extensive exposures of older Precambrian schist and gneiss and younger
Precambrian igneous rocks in eastern Clark and southeastern Lincoln Counties. Outcrops of
Precambrian granite, pegmatite, amphibolite, and gneiss exist in southern Lincoln County.
Schist, gneiss, and gneissic quartz monzonite, possibly as young as late Proterozoic, are
exposed in the Bullfrog Hills and Trapman Hills of southern Nye County.
Precambrian and
Lower Cambrian
Rocks
Late Precambrian and early Cambrian strata include a westward-thickening prism of quartzite,
siltstone, shale, and carbonate interpreted as a rifted continental margin miogeosyncline. This
prism has been divided into two depositional systems in Nevada: an eastern quartzite and
siltstone system and a western siltstone, carbonate, and quartzite province.
Middle Cambrian
through Devonian
Middle Cambrian through Devonian rocks exposed in the southern Great Basin consist of
carbonates and shales, with interbedded quartzite and sandstone with thicknesses from up to
500 m in western Utah to at least 6,100 m in central Nevada. Strata of middle Cambrian
through Devonian age comprise the Lower Carbonate Aquifer.
Mississippian through
Permian Sedimentary
Rocks
Thick flysch* deposits result from erosion of the north-northeast trending highland formed
during the Antler Orogeny in late Devonian and early Mississippian time. This sedimentation
continued through Permian time, declining as the orogeny waned.
Mesozoic Rocks
Mesozoic sedimentary rocks, locally present only in Clark County, consist of continental and
marine sandstone, siltstone, and limestone of the Triassic and Jurassic Aztec Sandstone, Chinle
Formation, and Moenkopi Formation. Approximately 30 separate Mesozoic to Tertiary granitic
plutons are exposed in Esmeralda County, west of Yucca Mountain. These range in size from
less than one km2 to the 1,000 km2 Inyo Batholith.
Tertiary Sedimentary
Rocks
Tertiary sedimentary rocks, such as the Esmeralda and Horse Spring Formations, crop out
throughout the southern Great Basin. These consist of poorly to moderately consolidated
alluvial deposits and fresh water limestones in variable thicknesses of up to 1,000 m. They are
commonly found interbedded with volcanic deposits.
Tertiary and
Quaternary Igneous
Rocks
The most prevalent Tertiary igneous rocks of the southern Great Basin are pyroclastic deposits
of rhyolitic to trachytic composition. Eruptions from four calderas at Yucca Mountain between
approximately seven and 16 Ma produced a complex mixture of pyroclastic flow and fall
deposits, epiclastic deposits, and subsidiary lavas approximately 3050 m in thickness at Yucca
Mountain. This was followed by scattered, small-volume basaltic or bimodal basaltic-andesitic
lava and scoria eruptions.
Tertiary and
Quaternary Surficial
Deposits
Late Tertiary to Quaternary surficial deposits occur throughout the region as unconsolidated
alluvial fan, pediment, and basin fill deposits of highly variable thickness and character.
* Deposits largely of sandy and calcareous shales.
7-11
-------
NEVADA
TEST
SITE
1 0 ' 2 3
KEY MAP
LEGEND
Miocene felslc volconlcs
Kolnltr Utso
Paintbrush Group
tuffactous roe/4
Polrtttrush Group
-------
Pre-Cenozoic Rocks
Pre-Cenozoic rocks, believed to consist primarily of Paleozoic sedimentary strata, underlie the
volcanic rocks at Yucca Mountain. Little detailed information is available as to their thickness,
lithology, and contact with overlying stratigraphic units. Exposures of highly deformed
Paleozoic rocks occur at scattered localities in the vicinity of Yucca Mountain, including the
Calico Hills to the east, Bare Mountain to the west, and Striped Hill to the south. Carbonate
rocks have been detected at a depth of 1,244-1,807 m in a borehole two km east of Yucca
Mountain (DOE95a).
In the Calico Hills, exposures of carbonate rocks occur in the upper plate of a gently dipping
thrust fault over a black shale sequence containing minor amounts of siltstone, sandstone,
conglomerate, and limestone. These strata are locally highly folded, making correlation with
stratigraphic units elsewhere in the region uncertain.
At Bare Mountain, there is a varied sequence of pre-Cenozoic sedimentary and meta-sedimentary
rocks, totaling about 6,650 m in thickness and ranging from Precambrian to Mississippian in age.
Fourteen Paleozoic and two Proterozoic formations are represented. Dolomite and limestone
dominate, with minor stratigraphic units of clastic rocks (quartzite, sandstone, and siltstone).
Paleozoic rocks found at a depth of 1,244 to 1,807 m in a borehole two km east of Yucca
Mountain are almost entirely dolomites and have been identified as related to the Lone Mountain
Dolomite and the Roberts Mountains Formation. Seismic reflection data are inconclusive as to
the thickness and extent of pre-Cenozoic rocks underlying Yucca Mountain, but the thickness is
believed to be substantial.
Mid-Tertiary Pyroclastic Rocks
These rocks, resting unconformably on older pre-Cenozoic rocks, compose the portion of Yucca
Mountain most important to the design and performance of the repository because they are the
host rocks for the repository and define the pathways for ground-water flow into and out of the
repository. Volcanic rocks ranging in age from about 11.4 to 15.2 Ma form the bulk of the
volcanic sequence, including the host rock of the potential repository, known as the Topopah
Spring tuff (Figure 7-8). The volcanic sequence consists of welded and nonwelded silicic
Pyroclastic flow, fallout tephra deposits, and volcanic breccias erupted from nearby calderas in
southwestern Nevada volcanic field. Non-welded tuffs typically have large primary porosity.
7-13
-------
West
|
1500-1
1000-
SOUTW10 CANYON YUCCA MOUNTAIN
YUCCA CREST
GHOST DAHCE FAULT
East
DM.1 HOLE WASH
£ SOO-
'"//// "/"/
WOO 3000 HORIZONTAL SCALE
FEET
[Tpc] TWA CANYON MCMBER ffcpl MOW PASS UEMER
PETI BEDOCO TUFF EH BULLFROG MEMBER
ffpil TOPOfAH SPM40 MCMKR E5J TRAM ICU6CR
03 TUFFMXOUS Bf OS OF CALICO MILS 153 FLOW BRECCIA
I HCti.Y FAULTED AW
I BRCCOATEO ZONE
ALLUVIUM I COLLUVUI
FAULTS WITH MINOR OF-SLF CNSPLACCMEMTS
POSITIONS KNOWH OR CONCEALED AT SURFACE
. FAULTS WITH MAJOR OP-SLP WSPLACCUENTS
POSITOHS KMOWN OR COHCEALEO AT SURFACE
? STRATIGRAPHY UNCERTAM
WATER TABLE
.&* ARROWS SHOW DIRECTION
^ OF RELATIVE DISPLACEMENT
_ UNLMTED i MFCRRED FAULTS
WITH SMALL OtSPlACEVENT
Figure 7-8. East-West Geologic Cross Section for the Yucca Mountain Site
This figure shows the relative positions of various rock units at the site,
including the unit proposed for the potential repository (Topopah Spring
Member of the Paintbrush Tuffs) and the fault zones that are closest to the site
(USG88a)
-------
However, the large porosity is poorly interconnected resulting in low permeability. The harder,
welded tuffs are commonly more highly fractured and, consequently, have significant bulk
permeability. The principal stratigraphic units are listed in Table 7-2, in order of increasing age
(adapted from DOE94a).
Table 7-2. Principal Stratigraphic Units
ll&it *,
Younger Post-caldera Basalts
Older Post-caldera Basalts
Shoshone Rhyolite Lava
Timber Mountain Group
Ammonia Tanks Tuff
Rainier Mesa Tuff
Post-Tiva/pre-Ranier Rhyolites
Paintbrush Group
Tiva Canyon Tuff
Yucca Mountain Tuff
Pah Canyon Tuff
Topopah Spring Tuff
Calico Hills Formation
Crater Flat Group
Prow Pass Tuff
Bullfrog Tuff
Tram Tuff
Dacite Lava and Flow Breccia
Lithic Ridge Tuff
Older Tuffs - Pre-Lithic Ridae
Agb$ft$
0.27-3. 8«
8.5-10.5
9
11.45
11.6
12.5
12.7
-
-
12.8
12.9
13.1
13.25
13.45
14.0
14-16
(a) Based on information from DOE95a to be discussed subsequently in Section 7.1.1.7. The age of the older
post-caldera basalts ranges from 10.4 to 6.3 Ma; for the younger post-caldera basalts, the age ranges from
4.9 to 0.004 Ma.
Many of these formations, particularly those in the Prow Pass Tuff, Calico Hills Formation, and
the Paintbrush Group, are further subdivided into members or units. The formations are
summarized below, from oldest to youngest, with an emphasis on thickness, general composition
and minerals important to radionuclide retardation along potential ground water transport
pathways.
a. Pre-Lithic Ridge Volcanics. The oldest known volcanic rocks in the area were deposited
approximately 15 million years ago and are represented in site boreholes by 45 to 350 m
of bedded tuffaceous deposits, pyroclastic flow deposits, and quartz-latitic to rhyolitic
7-15
-------
lavas and flow breccia. Correlation of these rocks with other rocks in the area is difficult
because of their heterogeneous character and varying degrees of alteration.
b. Lithic .Ridge Tuff. This thick, massive pyroclastic flow deposit overlying the older tuffs
appears to represent several eruptive surges and ranges in thickness from 185 m north of
the site to 304 m at the south end of the site. This unit is nonwelded to moderately
welded and has been extensively altered to smectites and zeolites.
c. Dacitic Lava and Flow Breccia. Dacitic lava and flow breccia overlie the Lithic Ridge
Tuff in deep boreholes at the northern and western parts of Yucca Mountain but are
absent elsewhere. Observed thicknesses in boreholes range from 22 m to 249 m. Much
of the unit has been moderately to intensely altered to smectite clays and zeolites.
d. Crater Flat Group. This group, overlying dacitic lavas and flow breccias in the northern
part of Yucca Mountain and the Lithic Ridge Tuff in the southern part, includes three
rhyolitic, ash-flow-tuff sheets—the Tram, Bullfrog, and Prow Pass Tuffs, in ascending
order. The Crater Flat Group is distinguished from other pyroclastic units at Yucca
Mountain by the relative abundance of quartz and biotite phenocrysts.
• Tram Tuff, The Tram Tuff appears to comprise at least 28 separate magmatic
pulses and includes two subunits distinguished on the basis of the relative
abundance of lithic fragments. The lower subunit is rich in these fragments
throughout, while the upper unit is poor in lithic clasts. The upper subunit, 126 to
171m thick, is partially welded and has a microcrystalline ground mass.
There are six to 22 m of ash-fall and reworked tuff, primarily comprising zeolitic
pumice clasts, between the Tram and the overlying Bullfrog Tuff.
Bullfrog Tuff. The Bullfrog Tuff is 68 to 187 m thick, consisting mostly of
pyroclastic flow deposits with thin-bedded tuffaceous deposits. North of borehole
USW G-4 (see Figure 7-8), this tuff consists of a moderately to densely welded
core enclosed by nonwelded to partially welded zones. To the south, the tuff is
composed of two welded zones separated by a one-meter-thick bed of welded
fallout tephra.
Prow Pass Tuff. The Prow Pass Tuff is a sequence of variably welded
pyroclastic deposits that erupted from an unidentified source between 13.0 and
13.2 Ma. The formation, 90 to 165 m thick across the repository area, consists of
four pyroclastic units overlying a variable sequence of bedded tuffs. These units,
designated Unit 1 through 4 by decreasing age, are characterized by
orthopyroxene pseudomorphs and the abundance of siltstone and mudstone lithic
clasts. Unit contacts are defined by fallout tephra horizons and abrupt changes in
sizes and amounts of pumice and lithic clasts.
7-16
-------
A bedded tuff unit at the base of the Prow Pass Tuff consists of unwelded, altered
tuffaceous deposits with a total thickness ranging from less than one meter to 11
m in boreholes.
Unit 1, a pumiceous pyroclastic flow deposit with an aggregate thickness of 25 to
70 m in cored boreholes, consists of three subunits separated on the basis of their
lithic clast content.
Unit 2 consists of nonwelded to partially welded lithic-rich pyroclastic flow
deposits with an aggregate thickness of three meters to 34 m in cored sections.
The unit has not been subdivided since distinguishing characteristics are lacking;
however, locally preserved ash horizons and abrupt changes in the amount and
size of pumice and lithic clasts suggest at least three flow deposits.
Unit 3 consists of 40 m to nearly 80 m of multiple welded pyroclastic flow
deposits, either separated by thin fallout tephra horizons or defined by abrupt
changes in the amount and size of pumice and lithic clasts. Two of three flow
deposits have been identified in most core holes but have not been correlated.
Unit 4 is distinguished by comparatively abundant pseudomorphic pyroxene in
pumice clasts and rock matrix and by a comparatively low ratio of flesic to mafic
phenocryst minerals. This unit includes three irregularly distributed subunits.
The aggregate thickness in cored sections ranges from about 4 m to as much as
20.5 m.
Calico Hills Formation. The Calico Hills Formation, a series of rhyolite tuffs and lavas,
includes five pyroclastic units overlying a bedded tuff unit and a local basal sandstone
unit in the Yucca Mountain area. The formation thins southward across the site area,
declining from about 290 m in the north to 43 m in the south. Basal beds of the Calico
Hills Formation include two units. One unit consists of a nine- to 39-meter-thick bedded
tuff unit containing coarse-grained fallout, primary and reworked pyroclastic-flow
deposits, and fallout-tephra deposits. The other unit consists of a 0- to 5.5-meter-thick
volcaniclastic sandstone unit with abundant lithic clasts and swarms of altered (to clay
minerals) pumice clasts, interbedded with rare pyroclastic-flow deposits.
The pyroclastic units are composed of one or more pyroclastic-flow deposits separated by
pumice- and lithic-fallout tephra deposits included with the unit lying above. Five units,
designated Units 1 through 5 by decreasing age, can be distinguished on the basis of
textural characteristics (percentages of various clastic material). In the northern part of
Yucca Mountain (below the proposed repository horizon) the formation is high in
zeolites, which compose 60 to 80 percent of the rock. In the southern portion of Yucca
Mountain, the rock remains vitric.
7-17
-------
Unit 1 is a nonwelded, lithic rich, pyroclastic-flow deposit ranging from 0 to 58 m thick
in cored sections. Pumice clasts constitute 10 to 15 percent of the unit and lithic clasts
increase from three to seven percent at the top to 15 to 20 percent at the base; phenocrysts
compose seven to 12 percent of the rock.
Unit 2,0 to 54 m thick, is a nonwelded, pumiceous, pyroclastic-flow deposit composed of
20 to 40 percent pumice clasts and up to five percent lithic clasts. Fallout deposits at the
base are ash-rich, have a porcelaneous appearance, and are less than one meter thick.
Unit 3 is a nonwelded lithic-rich pyroclastic flow deposit 22 m to 100 m thick in cored
sections. The unit is generally composed of 10 to 40 percent pumice clasts and five to
10 percent lithic clasts.
Unit 4 is a 0 to 57 m thick nonwelded, pumiceous pyroclastic flow deposit, with pumice
clasts and lithic clasts constituting 10 to 30 percent and one to five percent, respectively.
Thinly bedded ash-fall deposits, reworked pyroclastic-flow tuffs, and tuffaceous
sandstone form a thin basal subunit.
Unit 5 is a nonwelded to partially-welded pyroclastic-flow deposit ranging from 0 to 20
m thick in cored sections. The unit is characterized by a bimodal distribution of pumice
clast sizes—larger, slightly flattened clasts of 20 to 60 mm and smaller equidimensional
clasts of two to 12 mm. The unit is composed of 20 to 30 percent pumice clasts and two
to five percent lithic clasts,
f. Paintbrush Group. This group—one of the most widespread and voluminous caldera-
related assemblages in the southwestern Nevada volcanic field—consists of primary
pyroclastic flow and fallout tephra deposits, lava flows, and secondary volcaniclastic
deposits from eolian and fluvial processes.
Eruptive centers for the Topopah Spring and Pah Canyon Tuffs are uncertain, but the
Claim Canyon caldera (see Figure 7-7) is identified as the source of the Tiva Canyon and
perhaps the Yucca Mountain Tuffs.
The Topopah Spring Tuff (Figure 7-8) is the host rock for the proposed Yucca
Mountain repository. The tuff has a maximum thickness of about 350 m in the
vicinity of Yucca Mountain, The unit is divided into two members—an upper
crystal-rich member and a lower crystal-poor member—each of which is
subdivided based on variations in crystal content, phenocryst assemblage, pumice
composition, distribution of welding and crystallization zones, depositional
features, and fracture characteristics.
The upper, crystal-rich member is characterized by greater than 10 percent
phenocrysts, with a basal transition zone where the percentage increases from five
7-18
-------
to 10 percent. The member is divided into vitric, nonlithophysal, and local
lithophysal zones.
The lower, crystal-poor member is characterized by less than three percent
phenocrysts and is divided into devitrified rocks of the upper lithophysal, middle
nonlithophysal, and lower lithophysal zones and a vitric zone. Below the vitric
zone (the vitrophyre), concentrations of clay and zeolites increase significantly
from alteration of the volcanic glass.
• The Pah Canyon Tuff, a simple cooling unit composed of multiple flow units,
reaches its maximum thickness of 70 m in the northern part of Yucca Mountain
and thins southward. This tuff varies from nonwelded to moderately-welded.
Throughout much of the area, vitric pumice clasts are preserved in a sintered or
lithified nondeformed matrix.
• The Yucca Mountain Tuff, a simple cooling unit in the Yucca Mountain area,
varies in thickness from 0 to 30 m. Generally nonwelded, the unit is
nonlithophysal throughout Yucca Mountain but contains lithophysae where
densely welded in northern Crater Flat.
• The Tiva Canyon Tuff (Figure 7-8) is a large-volume, regionally extensive,
compositionally-zoned (from rhyolite to quartz latite) tuff sequence that forms
most of the exposed surface rocks exposed at Yucca Mountain. The tuff ranges in
thickness from 100 to 150 m. Separation into crystal-rich and crystal-poor
members and into zones within these members is based on similar criteria and
characteristics discussed above for the Topopah Spring Tuff.
g. Post-Tiva Canyon. pre-Rainier Mesa Tuffs. A sequence of pyroclastic flow and fallout
tephra deposits occurs between the Tiva Canyon Tuff and the Rainier Mesa Tuff hi the
vicinity of Yucca Mountain. The sequence ranges from 0 to 61 m thick and is
intermediate in composition between Tiva Canyon and Rainier Mesa Tuffs.
h. Timber Mountain Group. This group includes all of the quartz-bearing pyroclastic flow
and fallout tephra deposits that erupted from the Timber Mountain caldera complex about
11.5 Ma (see Figure 7-7). The complex consists of two overlapping, resurgent
calderas—one formed by eruption of the Rainier Mesa Tuff and a younger, nested one
formed by eruption of the Ammonia Tanks Tuff.
• The Rainier Mesa Tuff is one of the most widespread pyroclastic units of the
Yucca Mountain area. It is a compositionally-zoned unit consisting of high-silica
rhyolite tuff overlain by a considerably thinner quartz latite tuff restricted to the
vicinity of the Timber Mountain caldera. Exposed thicknesses along the west side
of the caldera are as great as 500 m. The formation is absent across much of
Yucca Mountain, but appears in down-thrown blocks of large faults in valleys on
7-19
-------
either side. The tuff is nonwelded at the base, grading upward into partially- to
moderately- welded devitrified tuff.
The Ammonia Tanks Tuff consists of welded to nonwelded rhyolite tuff with a
highly variable thickness of up to 215 m. It is absent across Yucca Mountain, but
is exposed in the southern part of Crater Flat.
Hydrostratigraphy
The formal geologic stratigraphy for those rocks near the repository horizon has been
reorganized into four major hydrostratigraphic units for ground-water modeling and performance
assessment. The groupings are based primarily on the degree of welding of the tuffs. These
units and their relationship to formal geologic stratigraphy are as follows (descriptions taken
from DOE95b):
• Tiva Canyon welded CTCw) unit: Consists of the moderately- to densely-welded
zones of the Tiva Canyon geologic member. This unit is characterized by low
matrix porosity (-10 percent), low matrix saturated hydraulic conductivity (-10"
um/s), and high fracture density (10-20 fractures/m3),
Paintbrush nonwelded fPTnt unit: Consists of the lower partially-welded to
nonwelded zones of the Tiva Canyon geologic member, partially-welded to
nonwelded Yucca Mountain and Pah Canyon members, the porous interlayers of
bedded tuffs, and the upper partially-welded to nonwelded part of the Topopah
Spring member. This unit is characterized by high matrix porosity (-40 percent),
high matrix saturated hydraulic conductivity (-10'7 m/s), and low fracture density
(~ 1 fracture/m3).
• Topopah Springs welded (TSw) unit: Consists of the welded zones of the
Topopah Spring member. This unit is characterized by low matrix porosity (-10
percent), low matrix saturated hydraulic conductivity (-10'7 m/s), and high
fracture density (8-40 fractures/m3). The basal vitrophyre of the Topopah Spring
member (TSv) is generally identified as a subunit because of its lower porosity as
compared to the TSw unit.
• Calico Hills nonwelded (CHn) unit: consisting of the moderately-welded to
nonwelded zones of the Topopah Spring member underlying the basal vitrophyre,
the partially-welded to nonwelded tuffs of the Calico Hills formation, and other
partially-welded to nonwelded tuffs located below the Calico Hills formation (i.e.,
the Prow Pass, Bullfrog and Tram members of the Crater Flat Unit). Portions of
the lower Topopah Spring member are vitrified and zeolitic alteration appears in
both the lower part of the Topopah Spring member and in the tuffaceous beds of
7-20
-------
the Calico Hills. This leads to a further division of this unit into vitric (CHnv)
and zeolitic (CHnz) subunits. The fracture density (2-3 fractures/m3) is similar in
both zones, and the porosity in the vitric tuffs (-30 percent) is marginally higher
than that of the zeolitic tuffs. However, matrix saturated hydraulic conductivity of
the CHnv subunit (~ 10'9 m/s) is roughly two orders of magnitude higher than that
of the CHnz subunit.
In some discussions of Yucca Mountain stratigraphy, the stratigraphic column is divided into
thermal/mechanical units, rather than the more formal geologic formations or the
hydrostratigraphic units (see, for example, Figure 6-7 in DOE94a). The boundaries between the
thermal/mechanical units tend to be defined by the interface between welded and non-welded
lithologies and the units are very similar to the hydrostratigraphic groupings.
Younger Basalt
The youngest volcanic rocks in the Yucca Mountain area are the basalts at Lathrop Wells, where
multiple eruptions occurred over a period of about 120,000 years with the latest event occurring
less than 10,000 years ago.
Surficial Deposits
Surficial deposits in the area reflect the effects of erosive processes and affect the surficial
recharge of water to the underlying rocks. Numerous Quaternary/Tertiary surficial deposits have
been defined in the Yucca Mountain area. These include alluvial, colluvial, and eolian deposits.
The alluvial deposits range hi age from late Tertiary (probably late Miocene) to late Holocene
and generally consist of sandy gravel (granules to boulders), often with interbedded sands. These
deposits occur along the washes, drainage channels, and valley slopes. The colluvial deposits are
Primarily of Quaternary age and generally consist of a thin mantle of angular gravels on slopes
and highlands.
Two deposits of eolian sand ramp are defined, both formed of massive to poorly-bedded sand
with five to 50 percent fine angular gravel. One deposit (late and middle Pleistocene) forms
Partially-dissected aprons between gullies on lower hill slopes. The other deposit (Holocene and
late Pleistocene) forms undissected and poorly-exposed sand ramps along Forty Mile Wash.
7-21
-------
Summary
The most important rocks affecting the design and performance of the proposed Yucca Mountain
repository are the sequence of Miocene volcanic rocks that overlie, underlie, and are the host
rocks for the repository. These silicic rocks consist of ash-fall and air-fall tuffs produced by
eruptions from the Timber Mountain-Oasis Valley caldera complex. Most of the exposed surface
rock over the repository is the 100-150 m thick Tiva Canyon Tuff. Below this, is the Yucca
Mountain Tuff, which is largely nonwelded and up to 30 m thick. The Claim Canyon caldera
segment lying to the east of the proposed repository site is a possible source for rocks in these
units. The repository horizon is in the Topopah Spring Tuff which has a maximum thickness of
350 m in the vicinity of Yucca Mountain, These units are all part of the Paintbrush Group.
Next, in descending sequence, is the Calico Hills Formation consisting of rhyolite tuffs and lavas
which, in turn, is underlain by the Prow Pass Tuff in the Crater Flat Group. The Prow Pass Tuff
is 90 to 165 m thick under the potential repository location. The surface of the water table lies
near the base of this unit. Lower lying units, generally in the saturated zone, include the 68 to
187 m thick Bullfrog Tuff and the Tram Tuff. These two tuffs are separated by six to 22 m of
ash-fall and reworked tuff comprised mainly of zeolitic pumice clasts.
7.1.1.4 Major Fault Features of the Yucca Mountain Area (Adapted from DOE95a)
The faults present in the site area are important for several reasons. To avoid adverse effects of
fault movement, areas of active fault movement should be avoided when deciding on the location
of surface waste handling facilities for the repository, as well as when designing the underground
waste emplacements locations. The fractured rocks in fault zones can also act as preferential
pathways for ground-water movement and radionuclide migration. Their location and hydrologic
properties are important for developing an understanding of the flow system and performing
quantitative calculations of ground-water movement essential to assessing the repository's
performance.
Faulting and the Structural Setting Around Yucca Mountain
The location of faults, and the extent of recent movement along these faults, is important to the
location and design of surface facilities and the layout of the underground repository at the Yucca
Mountain site. Seismic conditions in the area show at least some degree of correlation with the
7-22
-------
faults observed. Seismic activity could affect surface facilities of the repository. In addition, the
fractured rock zones typical of fault zones often serve as preferential pathways for the movement
of ground water. Rapid flow of ground water along fractures in the site area has been observed
and DOE's current layout of the repository has been designed to avoid emplacing wastes in areas
where the host rock is prominently fractured (e.g., the Ghost Dance Fault zone).
Yucca Mountain consists of a series of north-trending, eastwardly tilted structural blocks that
were segmented by west-dipping, high-angle normal faults during a period of major extensional
deformation. The site is situated near the southern end of the northwest trending Walker Lane
Belt, a zone of northwest-directed shear about 700 km long and 100 to 300 km wide. This Belt
absorbs part of the transform motion of the regional plates and the strain from the extension of
the Great Basin. It parallels the San Andreas fault and the Sierra Nevada Mountains and is
truncated on the south by the east-west Garlock fault (Figure 7-9).
BETTLES WELL FMJLT
M32 CXDM) UTN. EMTHOIUKE
TONOPAN
EXCELSIOR FJULT ZONE
Figure 7-9. The Walker Lane Belt and Major Associated Faults (DOE88)
7-23
-------
Cenozoic deformation probably took place on preexisting structures and is characterized by
strike-slip faulting, regional folding, and large-scale extension (see, for example, STE90). The
current type of deformation in the Walker Lane Belt probably began about five million years ago
as an overlap between the right-lateral shear caused by the North American and Pacific plates
and the gravity-driven extension of the regional uplift in the Great Basin. In the modem stress
field, northwest-striking faults move with left-lateral strike-slip or oblique-slip along the fault
planes.
In the Walker Lane Belt, right angle-shear totaling 4.27 to 7.35 millimeters per year (mm/yr) is
distributed along three major faults: the Owens Valley, Panamint Valley-Hunter Mountain, and
Death Valley-Furnace Creek faults. This, along with lesser amounts of slip on other fault
systems to the east, correlates well with the approximate 10 mm/yr of slip estimated from field
measurements.
The major north-trending faults transecting or close to Yucca Mountain are, from west to east,
the Crater Flat, Windy Wash, Fatigue Wash, Solitario Canyon, Stagecoach Road, Ghost Dance,
Bow Ridge, Midway Valley, and Paintbrush Canyon faults (Figure 7-10). Bedrock has been
displaced downward and to the west along these faults, which show predominantly dip slip, with
varying amounts of left-oblique slip, along the faults. Estimates of bedrock displacement over
the past 12 million years range from less than 100 m to as much as 600 m, with the displacement
increasing southward along each fault. The faults are projected up to 25 kilometers, but surface
exposures can usually be traced only one kilometer or less. Dips of the fault planes are generally
70 to 75 degrees.
Several northwest-trending faults have been identified along valleys, the most prominent being
the Yucca Wash, Sever Wash, Pagany Wash, and Drill Hole Wash faults. A northwest-trending
shear zone, the Sundance Fault, crosses the potential repository site (Figure 7-11). These faults
are thought to be strike-slip faults, with nearly horizontal slickenside lineations and vertical
displacements generally less than five to 10m.
7-24
-------
-« MXESSIBLE fWIROHUEHT BOUNDMf
K3 POTENTM REPOSrrofir SITE
Figure 7-10. Major North-Trending Faults in the Vicinity of Yucca Mountain (DOE95k)
7-25
-------
116'37'30" *
IK-JO'.00" W
"6-22-30-
Voih
2 Croltr riot
3 WMy Won.
roliout Wmn
5 S«CIWM Con;
6 Iron Rxlo*
7
8 Choil Donc<
9
•
!l
u
O Vueeo '
Omw *
15 Bo. Hi
Conyon
Poqon, WOK
17
a SloaKOOCti Rood
PorcHTin Ptmsffoar yri
Figure 7-11. Index Map of Faults at and near Yucca Mountain (Modified from DOE95k)
7-26
-------
Quaternary Faulting in the Yucca Mountain Area
Of particular concern for the Yucca Mountain site are faults considered to be Type I faults, as
classified by the U.S. Nuclear Regulatory Commission (NRC). Type I faults or fault zones are
those subject to displacement and are sufficiently long or located such that they may affect
repository design and/or performance. Evidence of movement during the Quaternary Period (the
past 1.6 million years) is the primary criterion for identification of these faults
Studies to identify and characterize faults that may be of concern to the Yucca Mountain facility
have focused on evaluating the potential Type I faults within 100 km of the site, as well as a few
major faults at greater distances. Some 82 known or suspected Quaternary faults and fault
rupture combinations have been identified within 100 km of the Yucca Mountain site (Figure 7-
12). DOE reports that 38 of these are capable of generating a peak acceleration of 0.1 g (the
force of gravity) or greater at the ground surface of the proposed repository site; these are
classified as relevant earthquake sources.15 An updated compilation of faults has been prepared
by the U.S. Geological Survey (USGS) which identifies 67 faults with demonstrable or
questionable evidence of Quaternary movement and the capability of accelerations of at least 0.1
g at an 84 percent confidence limit (WHI96). Significant known or suspected Quaternary faults
located within 20 km of the Yucca Mountain site are briefly described in Table 7-3.16 The more
distant major fault zones include: the Garlock Fault (125 kilometers south), the Owens Valley
Fault (140 kilometers west), the Stewart-Monte Cristo Valley Fault (200 kilometers northwest),
and the Dixie Valley Fault (see page 3.1-8 et seq, DOE95a).
15The NRC-supported program of the Center for Nuclear Waste Regulatory Analyses has identified 52 Type I
faults within a 100-km radius of Yucca Mountain (NRC97a).
16NRC-supported studies have identified 24 Type I faults within a 10-km radius of Yucca Mountain capable of
generating peak accelerations of greater than 0.3 g (NRC97a).
7-27
-------
™8rB6SBMi3
-
Figure 7-12 Index Map of Known or Suspected Quaternary Faults in the Yucca Mountain
Legion (Modified from DOE95a). Circles are 50 and 100 km radii from Yucca
Mountain (YM). Faults are identified as follows:
AM - Ash Meadow
AR - Amargosa River
AT - Area Three
- Bonnie Claire
»H - Buried Hills
BI.K -Belted Range
BM - Hare Mountain
BUL - Bullfrog Hills
CB - Carpetbag
CF - Cactus Flat
CFML - Cactus Flat-Mellan
CGV - Crossarain Valley
CHV -Chicago Valley
CLK - Chalk Mountain
- Checkpoint Pass
CRPL - Cockeyed Ridge-Papoose
Lake
CRWH - Cactus Range-Wellington
Hills
- Cane Spring
DV - Death Valley
EPR - East Pintwater Range
- Eleana Range
EVN - Emigrant Valley North
- Emigrant Valley South
- Furnace Creek
FLV - Fish Lake Valley
- Grapevine Mountains
- Groom Range Central
ORE - Groom Range East
GV - Grapevine
IIM - Hunter Mountain
- Indian Springs Valley
JUM - Jumbled HilTs
KRW - Kawich Range West
KV - Kawich Valley
KW - Keane Wonder
LM - La Madre
MER - Mercury Ridge
MM - Mine Mountain
NDR - North Desert Range
OAK - Oak Spring Butte
OSV - Oasis Valley
PAH - Pahranagat
PEN - Penoyer
- Pahute Mesa
PSV - Pahrump-Stewan Valley
- Panamim Valley
PVNH - Plutonium Valley-North
-Haltpint Range
- Ranger Mountains
RTV - Racetrack Valley
RV - Rock Valley
RWBW - Rocket Wash-Beatty Wash
- Sarcobatus Flat
SOU - South Ridge
SPR - Spoiled Ransie
SIM - Stumhle
SWI -Stonewall Hat
SWM - Sinncwall Mountain
IK - Tikahoo V:illc>
'I'M - Tin Mountain
TO1. - Toledia Peak
TP - To\\ne Pass
WAH - Wahmonie
WPR - West Pintwater Range
WSM - West Springs Mountain
YF - Yucca Flat
YL - Yucca Lake
7-28
-------
Table 7-3. Known or Suspected Quaternary Faults within 20 km of the Proposed Repository Site
v. -. -.
; ' ^ ,V C ' * *»
! ,- x t&itff Haitte , t . ,
* ? -. * , . . ' ... •• .
Bare Mountain
Crater Flat
Windy Wash
Fatigue Wash
Solitario Canyon
Stagecoach Road
Ghost Dance
Dune Wash
Bow Ridge
Midway Valley
Paintbrush Canyon
foetid
N
N
NE
N-NE
N
N
N-NE
N
N-NW
N
N
N
£gpr$»*
Length
20km
14-20 km
25km
17km
20km
10km
3.5km
8km
10-19km
1-4 km
25-32 km
®V,
E50-70
W70
W63
W73
W72
W73
W80-90
W
W65-75
W
W41-71
- pisjamtf
froiftSite
15kmW
5kmW
3kmW
2kmW
atW
boundary
SE corner
of area
center of
area
at E side
2kmE
3kmE
E side of
Yucca Mm.
o
y
'•'• v "•
' :,, l&feStAtffrtft
1 ff f : j- %
Most recent surface rupture 16 to 2 1 thousand years ago (ka); one to 1.5m
displacement; recurrence interval 100 ka; slip rate 0.01 mm/yr
Quaternary deposits (17 to 30 ka) displaced less than one m
At least four events in past 300 ka; recurrence interval 75 ka;
Pleistocene displacement approximately one m
Five late Quaternary events; cumulative displacement 2.2 m
Multiple mid- to late-Quaternary events;
1.7 to 2.5 m displacement of Quaternary deposits
Three to seven events during late Quaternary; displacement one to 2.3 m;
recurrence interval five to 70 ka; slip rate 0.01 to 0.06 mm/yr
No offset or fracturing of late Pleistocene or Holocene noted except for a
single fracture in one trench. Fracture zone varies up to 213 m across.
No evidence of Quaternary activity found
Most recent event 48±20 ka; cumulative displacement 0.3 to 0.7 m; likely
recurrence interval 60 to 100 ka; slip rate 0.002 to 0.01 mm/yr
No recognizable ruptures of Quaternary deposits
Six to eight events evident;
Midway Valley excavation: most recent event at 38±6 ka: cumulative
displacement 1.7 to 2.7 m; recurrence interval 20 to 80 ka, slip rate 0.007 to
0.02 mm/yr;
Busted Butte exposure: Quaternary displacement 4.8 to 7.8 m: recurrence
interval 40 to 125 ka; slip rate 0.006 to 0.01 mm/yr
7-29
-------
Several of the north-trending faults show evidence of activity during Quaternary time; the total
displacements on the most active of these is estimated to be less than 50 meters over the past 1.6
million years. Since the late Quaternary Period (< 128,000 years), displacements have been as
much as six m but are more commonly in the one to 2.5 m range. Recurrence intervals on the
faults showing movement in the Quaternary Period fall in the range of tens of thousands of years,
commonly between 30-80 thousand years with slip rates typically in the range of 0.01-0.02
mm/yr. The northwest-trending faults do not appear to have been active.
The three major faults in the immediate region of Yucca Mountain are the Ghost Dance fault,
which passes through Yucca Mountain and the proposed repository; the Bow Ridge fault, just to
the east of Yucca Mountain; and the Solitario Canyon fault, just to the west of Yucca Mountain.
According to DOE's interpretation of available data, the Solitario Canyon fault has shown no
significant movement over the last 40,000 to 110,000 years. No movement has occurred during
the last 10,000 years. The most recent surface-rupturing motion on the Bow Ridge fault is
estimated to have occurred 48,000 ±20,000 years ago, with a recurrence interval most likely in
the range of 60,000 to 100,000 years. There has been no offset or fracture on the Ghost Dance
fault for the past 20,000 years.
7.1.1.5 Tectonics and Seismicity (Adapted from DOE95a)
The fault systems and the seismic history of the Yucca Mountain area must be considered hi the
larger context of regional tectonics. By so doing, predictions of future seismic hazards and their
potential effects on the repository, as well as the performance of natural barriers, can be made
with reasonable certainty, within the limits of the available data. This section discusses what is
currently known about the tectonic setting of the region encompassing the repository site. Data
concerning the seismicity of the area and historic earthquake activity are also presented.
Regional Plate Tectonic Setting
The plate tectonic setting of the southwestern United States is dominated by the interaction of the
North American and Pacific Plates. In the Yucca Mountain Region, particularly west of Yucca
Mountain, this interaction is complicated by the overlap of right-lateral plate boundary stress
from these plate movements and extensional stress from the Basin and Range tectonics.
Based on geologic and geodetic measurements, the Pacific plate appears to be moving northwest
at approximately 50 mm/yr relative to the North Atlantic plate. The stresses generated from this
7-30
-------
movement are distributed to structural features on the North American Plate and contribute to the
tectonic processes (extension or compression of the crust, folding and faulting, etc.) in the region.
About 35 mm/yr of the motion from the Pacific Plate is absorbed by the San Andreas fault
system; another 5 mm/yr may be absorbed by coastal strike-slip faults parallel to and west of the
San Andreas fault. The eastern edge of the Sierra Nevada microplate (composed of the Sierra
Nevada Mountains and the Great Valley of California) appears to move northwest at
approximately 10 mm/yr. This latter movement, between the eastern edge of the Sierra Nevada
Mountains and the western edge of the Colorado Plateau, is most likely to contribute to the
seismicity and tectonic processes around the Yucca Mountain site (Figure 7-13). Uncertainties in
the understanding of the regional tectonic processes include: the amount of compression normal
to the San Andreas fault induced by Pacific plate motion (N36°W ±2°), the rate of relative
motion between plates, and the amount of motion taken up within the Sierra Nevada microplate.
The timing and mechanisms for producing the crustal extension which characterizes the
structural and physiographic features of the Great Basin are a subject of debate. Several
mechanisms have been proposed for the extensional tectonic processes that produced the major
land forms of the Great Basin. Relatively high-angle, planar, normal faults cutting brittle crust
can accommodate up to 10 or 15 percent of the crustal extension. Normal faults at a high angle
at the surface and curving to lower angles at depth (listric faults) may accommodate much greater
extension. Modeling of very low angle detachment faults suggests extensive crustal thinning that
may accommodate extension of the crust by 200 percent or more.
The typical Basin and Range structures were developed by about 11 Ma. They are tilted fault
block ranges with relatively large displacement, high-angle normal faults exposed at the surface
bounding one or both sides of each range. Scott (SC090) suggests that rates of fault movement
were highest between 13 -11.5 Ma and thereafter decreasing over time.
This crustal extension varied across the region in time and space. One thought is that rapid
Miocene extension migrated westward from Yucca Mountain after about 11.5 Ma and may also
have been nonuniform from north to south. Pliocene and later extension, accompanying a
Postulated region-wide uplift starting about five million years ago, is more evenly distributed and
is taken up by movement on high-angle normal faults at depth which are coincident with the
Miocene faults expressed at the surface. This belief is consistent with the evidence of the
existence of faulting to depths of 15 km or more indicated by the pattern of hypocenters for the
current seismicity in the region.
7-31
-------
IwC^/
\ Yva.ii.r-' I/
Southern Basfn and Range
Figure 7-13. Sketch Map of the Western United States Showing Some Major Structural Features.
Symbols (©) at the latitude of Las Vegas give approximate motions toward the NW in mm/yr relative
a "stable North America." This interpretation suggests that 10 mm/yr of NW movement occurs
between the Colorado Plateau and the crest of the Sierra Nevada Range, 35 mm/yr occurs on the Safl
Andreas Fault, and five mm/yr occurs west of the San Andreas Fault. This is consistent with the
paleoseismic data and historic observations of strike slip faulting in this region. (Modified from
DOE95a)
7-32
-------
Structural Features and Seismicity
The relationship between specific structural features, particularly faults, and seismicity in the
Basin and Range Province is not entirely clear. The Central Nevada Seismic Belt (CNSB), for
example, is clearly associated with major faults or fault systems showing historic surface rupture.
However, other zones of seismic activity and areas of diffuse activity show no evidence of
historic surface faulting. One example is the east-west seismic belt, which includes the Nevada
Test Site.
The apparently poor correlation between earthquakes and faults may be attributable, at least hi
part, to several factors: 1) the short historical record relative to the long recurrence intervals for
earthquakes, 2) the difficulty of accurately locating epicenters in this remote area, and 3) the
unknown geometry of faults at depth. Study of the paleoseismic record for the Quaternary
Period suggests that, in the Yucca Mountain Region, recurrence intervals for surface rupture are
on the order of thousands to tens of thousands of years.
Seismology of the Yucca Mountain Area
In the region around the site, there are several zones in which seismicity is concentrated: the
Sierra Nevada-Great Basin Boundary Zone (SNGBZ), the CNSB, the Southern Nevada
Transverse Zone (SNTZ), the Garlock Fault, and the Mojave Block. All of the zones, except the
Mojave Block, are wholly or partially in the Walker Lane Belt, a major tectonic element of
southwestern Nevada. In addition, there is a broad distribution of seismic activity that is not
associated with any known major tectonic feature throughout much of the Great Basin.
The Walker Lane Belt tectonic element (Figure 7-9) consists of nine structural blocks acting
more or less independently. The belt is defined by a style of faulting within and bounding the
blocks which ranges from northwest-trending right-lateral slip (the Pyramid Lake, Walker Lane,
and Inyo-Mono blocks) to northeast-trending left-lateral slip (the Carson, Spotted Range-Mine
Mountain, and Lake Mead blocks) to east-west trending left-lateral slip (Excelsior-Coaldale
block). Cumulative lateral offset on individual major faults ranges from a few kilometers up to
!00 kilometers and faults rarely extend to adjacent blocks.
The Walker Lane Belt probably developed in the Mesozoic Period and is still active. Most of the
faults show evidence of Cenozoic movement and numerous zones exhibit Quaternary and
Holocene offset (STE90). Although the recurrence interval for the late Quaternary faulting is
7-33
-------
generally thousands to tens of thousands of years, recurrence may be on the order of decades in
some sections of the seismic zone, e.g., the CNSB.
Of the four seismic zones identified in the Walker Lane Belt, the SNTZ is nearest to the Yucca
Mountain site and is the most significant to repository performance. Although the other zones
exhibit recent seismic activity, they are further removed from the Yucca Mountain site and are
less likely to affect the repository.
The Southern Nevada Transverse Zone, which includes Yucca Mountain, is an arcuate belt of
seismicity about 150 kilometers wide, extending from the southern region of the Intel-mountain
Seismic Belt (in southwestern Utah) to the Mammoth Lakes area in California. Historic
earthquakes in this zone have been of moderate magnitude with no documented surface rupture.
Earthquake events include the 1902 Pine Valley, Utah (ML 6.3)17, the 1966 Caliente-Clover
Mountain, Nevada (ML 6.0), and the 1992 Little Skull Mountain, Nevada (ML 5.6) near the
proposed site (see Table 7-3).
Seismic Distribution
Studies of the large Great Basin earthquakes suggest faulting on steeply dipping fault planes that
penetrate the upper 15 kilometers of crust as the focal mechanism for many of the earthquakes
observed. In general, mainshock hypocenters for earthquakes of magnitude seven or greater hi
this region can be located on the down-dip projection of the surface rupture observed along faults
identified in the field, suggesting that large Great Basin events occur on steeply dipping planar
faults at depths less than about 15 kilometers.
Three—with perhaps two additional possible—seismic gaps (areas of no recent seismic activity)
have been identified in the western Great Basin. These gaps occur between the rupture zones of
major historic earthquakes and contain structures that show evidence of prehistoric activity.
Seismic gaps are generally considered to be significant in plate-boundary regions but their
relevance for interplate regions such as the Great Basin is not clear. These gaps may represent
areas of prolonged low or no seismic activity or areas where stresses are not being released by
fault movements.
17ML is a measurement of the magnitude of the seismic event. See Table 7-4 for a definition of this and other
magnitude measures.
7-34
-------
Table 7-4. Significant Earthquakes within 320 km of Yucca Mountain Site Since 1850
Owens Valley, CA,
1872
March 26, 1872; estimated at M» 7.8 to Ms 8.0**; considered largest historic event of the
Basin and Range; surface raptures along 90 to 110 km on Owens Valley fault; average
net oblique slip of 6.1 ±2.1 m and up to four m vertical displacement; liquefaction of
unconsolidated sediments.
Wonder, NV, 1903
Fall 1903; estimated magnitude 6.5; rupture of the Gold King fault; ruptures of five to
16 km with fissures up to 1.5 m wide and ! .5 m deep in alluvium; in the same area as
the 1954 Fairview Peak-Dixie Valley earthquakes.
Cedar Mountain,
NV, 1932
December21, 193 2 ;M, 72; about 61 km of discontinuous faulting in a be It six to 14
km wide; displacements up to 1.8 m horizontal and 0.5 m vertical; analysis indicated
main shock was two sources occurring about 20 seconds apart; an M» 6.7 event and a
second M,, 6.6 event; series of seven moderate events in this part of the CNSB from
1932 to 1939.
Excelsior Mountains,
NV, 1934
January 30, 1934; ML 6.3 (Mw 6.1); on Excelsior-Coaldale section of the Walker Lane
belt; about 60 km west-southwest of the 1932 event; foreshock of ML 5.6 preceded
mainshock by 45 min.; surface rupture 1.4 km in length and less than 13 cm vertical
displacement An ML 5.5 earthquake occurred on August 9,1943, approximately 40 km
southeast.
Rainbow Mtn.-
Stillwater, NV, 1954
July 6,1954; two events of M 6.6 and M 6.4 in Rainbow Mountain area were followed
on August 24 by the Slillwater M 6.8 event initiating a six-year period of 10 events
greater than M 5.5 in the CNSB.
Fairview Peak-Dixie
Valley, NV, 1954
December 16, 1954; an ML 7.3 event on the Fairview fault followed four minutes later
by an ML 6.9 event rupturing the Dixie Valley fault; diffuse fracture zone covering an
area 100 km by 30 km from Mount Anna to the northern part of Dixie Valley;
displacements four m right lateral and three m vertical on Fairview Peak fault and over
two m vertical in Dixie Valley.
Caliente-Clover
Valley, NV, 1966
On August 16,1966; ML 6.0; near Caliente, Nevada, about 210 km east-northeast of
Yucca Mountain. The source depth is estimated at 6 km; with the focal mechanism a
strike-slip motion on steeply dipping plates oriented either north-northeast or west-
northwest.
Mammoth Lakes,
CA, 1978-1980
An ML 5.8 earthquake midway between Bishop and Mammoth Lake in October, 1978,
was followed 18 months later (May, 1980) by a swarm-like sequence of four events (ML
6.5, ML 6.0, ML 6.7, ML 6.3) within two days. This sequence was accompanied by
inflation of the resurgent dome in the Long Valley caldera. Activity continued with
moderate earthquake swarms in the southern part of the caldera with spasmodic tremor
sequences usually associated with magma injection at depth. The Chalfant sequence,
discussed below, occurred to the east in 1986.
7-35
-------
Table 7-4. Significant Earthquakes within 320 km of Yucca Mountain Site Since 1850
(Continued)
ChaJfant Valley, CA,
1986
On July 21, 1986, an ML 6.6 earthquake occurred in the Chalfant Valley in eastern
California about 15 km north of Bishop with about 10 km of rupture along the White
Mountains fault zone. The source-depth was located 11 km below the surface and the
focal mechanism indicates right lateral slip on a plane oriented north-northwest dipping
70° southwest.
Landers, CA, 1992
The Landers sequence began April 23rd with the ML 6.2 Joshua Tree earthquake,
followed by a sequence of 6000 events. On June 28,1992, an Ms 7.6 earthquake near
Landers, California, ruptured sections of several mapped north- to northwest-trending
faults and several concealed unmapped north-trending faults in the south-centra! portion
of the Mojave block. An extensive aftershock sequence followed, extending 85 km
north of the mainshock and 40 km to the south. The sequence included the Mj 6.7 Big
Bear earthquake three hours after and 30 km west of the mainshock. Surface rupture
extended for 85 km, with displacement averaging two to three meters across the rupture
zone, up to 6.7 m on the Emerson fault, and minor rupture of faults within 30 km of
either side of the main rupture zone. The Lander event was followed by a sudden
increase in seismic activity in the western U.S. up to 1250 km from the mainshock, with
an intense cluster of events in the Walker Lane belt. This included the ML 5.6 Little
Skull Mountain earthquake on June 29, 1992, approximately 20 km SE of Yucca
Mountain.
Eureka Valley, CA,
1993
On May 17, 1993, an ML 6.1 earthquake occurred 30 km southeast of Bishop,
California. The hypocenter was located nine kilometers below the surface in the
southern part of Eureka Valley. Preliminary analysis indicates normal faulting on a
northeast striking plane, perhaps paralleling a north-northwest trending inferred
Quaternary fault in the area.
*a Terms used for earthquake magnitude in the table above include:
ML Local magnitude; this is the original Richter scale, developed in California for earthquakes with
epicentral distances less than 600 km and focal depths less than 15 km; uses waves with periods of
about 1 s; saturates at M = 7.25;
Ms Surface-wave magnitude; suitable for global distance; uses waves with 20 s periods; saturates at
about M = 8.6;
Mw Moment magnitude; based on seismic moment (Mo = uAD), where u = shear modulus, A = area
of fault rupture, and D = fault displacement; Mw = 2/3 log MO-10.7; does not saturate;
M This is assumed to be local magnitude.
7-36
-------
Significant Historical Earthquakes
Figure 7-14 depicts the epicenters for earthquakes of magnitude 3 and greater occurring within
320 kilometers of the proposed site from 1850 through 1992. These data show a clustering of
seismicity in the CNSB and the SNGBZ, as well as in the southern Mojave Desert and along the
San Andreas fault zone. In addition to those identified in the figure, numerous small magnitude
earthquakes have occurred in clusters or as isolated events throughout much of Nevada. The
Garlock Fault and a large portion of the southern Great Basin appear to show relatively little
seismic activity during this period.
Earthquakes occurring since 1850 within 320 km of the Yucca Mountain site with magnitudes
greater than 6 are summarized in Table 7-3. These either resulted in surface rupturing or
represent the largest event in a particular seismic-source zone. The most recent strong
earthquake (ML =5 or greater) in the vicinity of Yucca Mountain was the Little Skull Mountain
(ML = 5.6) event in June 1992, associated with the Landers, California earthquake earlier that
year.
Studies of ground motion from recorded seismic activity around Yucca Mountain and of surface
features susceptible to ground motion effects, suggests that Yucca Mountain has not been subject
to ground accelerations at the surface in excess of 0.2 g for over several tens of thousands of
years. At the depth at which waste is likely to be emplaced in the repository, the effects of
ground motion would be expected to be significantly less. These ground accelerations do not
present excessive demands on seismic facility design requirements for the repository or its
associated surface facilities.
The largest seismic event in the immediate area of Yucca Mountain since 1978 was an ML 2.1
event on November 18,1988, centered 12 km northwest of the proposed repository location. An
earthquake of magnitude Mw 5.7 occurred on June 29,1992, beneath Little Skull Mountain
approximately 20 km southeast of Yucca Mountain. This earthquake is the largest ever recorded
(in about 100 years of records) hi the vicinity of the site. It caused minor structural damage to
the Yucca Mountain project field office near Yucca Mountain but had no apparent effect on
geologic features near the mountain.
7-37
-------
124"
122- W
WO' W
TW W
m* w
Figure 7-14. Magnitude 3 or Greater Earthquakes Within 320 Km (200 Miles) of Yucca
Mountain from 1850 to 1992 (Modified from DOE95a)
Based on a return period of 12,700 years, Bechtel Nevada estimates that for the adjacent Nevada
Test Site there is a 0.55 probability of at least one earthquake of magnitude 6.8 or greater
occurring in the next 10,000 years (SHO97).
DOE has not considered seismicity to be a significant factor in repository safety performance.
Seismic effects are not considered in previous total system performance assessments (DOE94a,
DOE95b) because DOE believes that they will have virtually no effect underground. Dowding
and Rozen (DOW78) examined empirical evidence of damage to 71 rock tunnels in Alaska,
7-38
-------
California and Japan from earthquake shaking. From this analysis, the authors concluded that,
for peak surface accelerations which would cause heavy damage to above ground structures,
there was only minor damage to tunnels. No tunnel damage was observed for peak surface
accelerations of less than approximately 0.2g and only minor tunnel damage occurred when the
peak surface acceleration was less than 0.5g.
DOE quantitatively analyzed the variation of ground motion with depth using both stochastic and
empirical methods (DOE94e). Peak surface accelerations were shown to be reduced by a factor
of two at a depth of about 400 m.
DOE considered tectonism in the TSPA-VA released in 1998, including the effects of parameter
variability (DOE98). NRC included the effects of fault displacement impacts and seismic
rockfall impacts on waste packages in TPA 3.1 (NRC97c).
In its 1996 Phase 3, Yucca Mountain Total System Performance Assessment, the Electric Power
Research Institute (EPRI) did not include consideration of earthquakes since it was concluded
that "...tectonic activity is not expected to significantly impact repository integrity" (EPR96).
The National Academy of Sciences (NAS) supports DOE's view that seismic effects on
underground excavations are usually less severe than on surface facilities (NAS95, p. 93). In
addition, NAS states that while the timing of seismic effects is unpredictable, the consequences
of such events are boundable for performance assessment purposes (Ibid., p. 94). The NAS
further notes that it is possible for the hydrologic regime to be affected either adversely or
favorably by seismic events.
The technical community did not agree with DOE's position on structural deformation and
seismicity presented in TSPA-95. Subsequently, in May 1996, a meeting of involved groups was
held to review and seek agreement on defensible tectonic models based on available data. The
group included DOE, NRC, the Advisory Committee on Nuclear Waste (ACNW), the Nuclear
Waste Technical Review Board, the USGS, the State of Nevada, the EPRI, and the Center for
Nuclear Waste Regulatory Analyses (CNWRA) (NRC97a). Of 11 proposed models, the group
agreed that only five were supported by existing data. Agreement on the five supportable models
not unanimous nor was agreement on the relative importance of the five models. In
7-39
-------
addition, some of the models may be independent and some may be subsets of others. The five
viable alternative models are:
Deep detachment fault (12-15 km)
• Moderate detachment fault (6-8 km)
• Planar faults with block deformation
• Pull-apart basin18
• Amargosa shear
The pull-apart basin model proposed by the USGS and the Amargosa shear model proposed by
the State of Nevada are based on buried or blind seismic sources at Crater Flat and involve the
greatest seismic risk. These seismic sources are not included in DOE's Probabilistic Seismic
Hazards Analysis which was used as a partial basis for the conclusions reached in TSPA-95.
Depending on proximity to the repository, the Amargosa shear could result in an earthquake with
magnitude Mw^7.8 and accelerations exceeding 1 g (NRC97a). More recently, CNWRA stated
that apatite-fission-track dating from Bare Mountain and Striped Hills does not support the
USGS reconstruction of the Amargosa shear model (McK96). CNWRA believes that the pull-
apart basin model is more tenable but requires additional direct observations of basin-bounding
and cross-basin strike-slip faults.
Additionally, DOE argued that future tectonic events are unlikely to significantly alter the
hydrologic characteristics of the Yucca Mountain site. This argument is based on the position
that the current state of faults and fractures at the site is the result of cumulative tectonic events.
However, CNWRA posits that a single tectonic event can cause significant changes in hydrologic
characteristics. The DOE argument is valid only for characteristics resulting from cumulative
events and not for the most recent single tectonic event (NRC97a).
7.1.1.6 Fractures {Adapted from DOE95a)
Closely allied with tectonic issues is the consideration of fractures in the rocks surrounding the
repository. An extensive fracture network can provide fast paths both for influx of water into the
repository for overlying strata and egress of water potentially contaminated with radionuclides
through underlying strata. To develop an understanding of fractures, studies have been
18 A pull-apart basin is a structural depression formed by localized extension along strike-slip fault zones. The
basin is formed in the brittle upper crust above a horizontal detachment in the lower crust (NRC97a).
7-40
-------
conducted to examine the age and connectivity of fractures primarily in a portion of the Tiva
Canyon Tuff. Outcrop studies were conducted for a number of units. The studies were designed
to define the general orientations of fracture sets over all of Yucca Mountain and to establish the
relationship of fracture sets to regional tectonic history. A few studies of the vertical continuity
of fractures have been conducted in the Paintbrush nonwelded unit. These are designed to
examine changes in fracture pattern as a function of stratigraphy (DOE95a).
Four sets of tectonic fractures with consistent orientation were identified within the Paintbrush
Group. In addition, a set of sub-horizontal joints with variable strikes and dips of less than 10
degrees exists. These fracture sets may have originated as extension joints, many of which have
been subsequently been reactivated. It has been postulated that the fractures developed as a
mountain-wide response to far-field stresses rather than local movement of structural blocks.
However, data to support this hypothesis conclusively are limited (DOE95a).
Fracture widths are defined both by rock wall separation and actual fracture aperture. Rock wall
separation is the distance between the fractured surfaces without reference to any infilling with
secondary minerals. Aperture includes the effects of any infilling and is the amount of open
space remaining. Wall separations are typically one to 10 mm from the surface to a depth of
about 200 m. Surface fractures are 50 to 75 percent filled with caliche which reduces the
aperture to one to two mm. Below about 10 m from the surface, the fractures are 40 to 50
percent filled, primarily with quartz and calcite (DOE95a).
Studies of surface fractures have led to the following general conclusions (DOE97c, SWE96):
• Fracture intensity is a function of lithology, variation in the degree of welding in
the tuffs, and, to a lesser extent, proximity to faults
Connectivity of the fracture network also depends largely on the degree of
welding and the lithology
• Width and intensity of fractured zones vary around faults and are related to fault
complexity
The degree of welding within the Paintbrush Group has the greatest effect on the overall
character of the fracture network with fracture intensity and network connectivity being least in
nonwelded or poorly-welded units.
7-41
-------
Subsurface studies have indicated that correlation with surface features diminishes as the depth
increases because:
• Some faults which displaced units in the Topopah Spring Tuff became inactive
before the overlying Tiva Canyon Tuff was deposited
• Many faults are discontinuous so that the displacement may die out between
observation points
• Faults commonly spread upward resulting in differing surface and subsurface
geometries (DOE97c)
7.1.1.7 Volcanism (Adapted from DOE95a)
To assess the possibilities of disruptive volcanic events, the nature and history of volcanism hi
the area must be understood. Yucca Mountain consists of silicic volcanic rocks originating from
the Timber Mountain caldera complex to the north. A resurgence of silicic volcanism is unlikely
since the activity that formed the rocks at Yucca Mountain ceased millions of years ago.
However, basaltic volcanism has taken place more recently. Basaltic volcanism is commonly
accompanied by the intrusion of dikes into the surrounding rocks and could pose the potential for
intrusion into the repository itself if such volcanism occurred close to the repository. Magmatic
intrusions could mobilize waste and/or alter ground-water pathways. The volcanic history of the
Yucca Mountain area is discussed below.
Yucca Mountain is composed of Miocene volcanic rocks erupted from the overlapping Silent
Canyon, Claim Canyon, and Timber Mountain calderas between 11 and 15 million years ago.
The silicic volcanic tuffs that comprise Yucca Mountain are typical of mid-Tertiary basin and
range extensional tectonics in southern Nevada. Yucca Mountain, at the depth of the proposed
repository, is comprised of units of the Paintbrush Tuff, a major outflow ignimbrite of the Claim
Canyon caldera segment of the Timber Mountain caldera complex (Figure 7-15). During the late
Neogene (two to 10 Ma) and Quaternary (0 to two Ma) Periods, small-volume, mostly
polygenetic, basaltic centers produced lava flows, air falls, and cinder cones hi the area. The
silicic and basaltic volcanism are described below.
7-42
-------
IW 45- W
v^fpt'* ":''&-,.
MEMBER;'1
-SLEEPS
,",i/'BUTTE;;v
TOPOPAJH-
$%ffi%S$j?t -r^-.'^
^fl
.*,
&
••»'*-'• • ~ '' 'A '
^*i.-*££.* ^
«: 1«
pli^,.; •'"\s-^^^^^^- ^^^^'
/>
T1VA CANYON MEMBER DASHED WHERE INFERRED
Figure 7-15. Index Map Showing Outlines of Calderas in the Southwestern Nevada Volcanic
Field and the Extent of the Tiva Canyon and Topopah Spring Tuffs of the
Paintbrush Group (Modified from DOE95a)
7-43
-------
Silicic Volcanism
The silicic volcanism in the Yucca Mountain area is part of an extensive, time transgressive pulse
of mid-Cenozoic volcanism that occurred throughout much of the southwestern United States.
Yucca Mountain is in the south-central part of the SNVF, a major Cenozoic volcanic field
covering an area of over 11,000 km2. Magmatism in the region was distributed hi linear belts
parallel to the convergent plate margin during the Mesozoic Era. In the southwestern United
States, a pause or disruption in the belts about 80 Ma formed the Laramide magmatic gap or
hiatus, which lasted until renewed silicic magmatism began in the northeastern part of the Great
Basin about 50 Ma. Sites of eruptive activity migrated south and southwest across parts of
Nevada and Utah, with eruptive centers distributed along arcuate east-west trending volcanic
fronts. The most intensive eruptions were at the leading edge of the migrating front, with the
most voluminous silicic volcanic activity in the Yucca Mountain area occurring between 11 and
15 Ma. Silicic magmatic activity in the area ceased about 7.5 to 9 Ma. The Yucca Mountain
area marks the southern limit of time-transgressive volcanic activity.
Between 10 and 13 Ma, there were two significant changes in the regional volcanic and tectonic
patterns: the southern migration of volcanism halted and the composition of the volcanic activity
changed. Diminished silicic-eruptive activity migrated in less systematic patterns to the
southwest and southeast, leaving a conspicuous amagmatic gap from the southern edge of the
Nevada Test Site south to the latitude of Las Vegas.
Should volcanism occur in the future, the type of volcanism (basaltic or silicic) is potentially
significant, since silicic eruptions are more explosive. The DOE claims that there has been no
silicic volcanism in the Yucca Mountain Region since about 7.5 Ma at the Stonewall Mountain
caldera more than 100 km northwest of Crater Flat and since nine Ma at the closer Black
Mountain caldera (60 km northwest of Crater Flat). Consequently, DOE has concluded that the
potential for future silicic volcanism is negligible (DOE96e). However, work by NRC suggests
that silicic pumice with an age of 6.3 ±0.8 Ma (based on zircon fission track data) existed
beneath basalts in Crater Flat. This is at odds with the DOE position that post-caldera silicic
eruptions had not occurred near the proposed repository site (NRC97a). Subsequently, NRC
reported that, based on argon isotope dating, the age of the silicic material was 9.1 ±3 Ma, which
correlates with the eruptions from the Black Mountain caldera (NRC97b). On the basis of this
information, NRC concluded that silicic volcanism did not need to be considered in evaluating
the probability and consequences of igneous activity at Yucca Mountain.
7-44
-------
Basaltic Volcanism
Two episodes producing basaltic-volcanic rocks have been defined in the Yucca Mountain area,
both occurring after the majority of the silicic volcanism ended. The first, marked by basalt of
the silicic episode (BSE), consists of basalt-rhyolite volcanism postdating most silicic eruptions
of the Timber Mountain-Oasis Valley (TM-OV) complex. The second episode is comprised of
spatially-scattered, small-volume centers marked by scoria cones and lava flows of alkali basalt,
ranging in age from about 10 Ma to less than 10,000 years. These post-caldera basalts of the
Yucca Mountain Region are divided into older post-caldera basalts (OPB) and younger post-
caldera basalts (YPB). The locations of basalts in the Yucca Mountain Region with ages of less
than 12 Ma are shown in Figure 7-16 (NRC96). (The cited ages of some of the occurrences
reported by NRC differ slightly from those reported by DOE. The differences are not
substantive.)
The BSE crops out throughout the Yucca Mountain area and is identified by several
characteristics: 1) a close association (in time and space) with activity of the TM-OV complex,
2) all centers of the BSE are large-volume eruptive units (<3km3 dense-rock equivalent—the
largest centers are in the ring-fracture zone of the Timber Mountain caldera), and 3) a wide range
of geochemical composition. The BSE occurs in three major groups:
• Mafic Lavas of Dome Mountain (age 10.3 ±0.3 Ma) are exposed in the moat
zone of the Timber Mountain caldera and comprise the largest volume of basaltic
rocks
• Basaltic Rocks of the Black Mountain Caldera overlap some units of the
caldera in age
Basaltic Volcanic Rocks, Yucca Mountain Area include the basaltic andesite of
Skull Mountain (dated 10.2 ±0.5 Ma), the basalts of Kiwi Mesa, and Jackass Flats
The second episode of basaltic volcanism, marked by the post-caldera basalt of the Yucca
Mountain Region, occurred at sites either well removed from the eruptive centers of the TM-OV
complex or younger than the silicic-magmatic activity. These sites generally consist of small
volume (<1 km3) centers marked by clusters of scoria cones and lava flows.
7-45
-------
&*3£ f r'€£\ **
wk«-.
• vtrwKe
/ xf.^(»
Figure 7-16. Distribution of Basalts in the Yucca Mountain Region with Ages of Less Than 12
MA (NRC96). Dotted line defines boundary of Yucca Mountain/Death Valley
isotopic province where basalts have same relatively unique isotopic structure.
7-46
-------
The OPB were produced along either north-northwest trending Basin and Range faults or at the
intersection of Basin and Range faults with the ring-fracture zone of older calderas. These range
in age from 10.4 to 6.3 Ma and are represented at four localities:
• Rocket Wash, thin, basalt lava flows (8.0 ±0.2 Ma) occur at the edge of the ring-
fracture zone of the Timber Mountain caldera
• Pahute Mesa, three separate but related basalts (with ages ranging from 8.8 ±0.1
to 10.4 ±0.4 Ma) occur at the intersection of faults with the ring-fracture zone of
the Silent Canyon caldera
• Paiute Ridge, dissected scoria cones and lava flows (8.5 ±0.3 Ma) are associated
with intrusive bodies occurring at the interior of northwest-trending graben; the
related Scarp Canyon basalt (8,7 ±0.3 Ma) crops out west of Nye Canyon
Nye Canyon, three surface basalts (6.3 ±0.2 Ma, 6.8 ±0.2 Ma, and 7.2 ±0.2 Ma)
and a buried basalt (8.6 Ma) occur in the Canyon.
The second eruptive cycle, resulting in the YPB, usually occurred at clusters of small-volume
centers aligned along predominantly northeast structural trends. These eruptions occurred from
4.9 Ma to as recently as 0.004 Ma and are represented at the following localities (in decreasing
age):
• Thirsty Mesa, a thick accumulation of fluidal lava and local feeder vents erupted
onto a pre-existing Thirsty Canyon Group ignimbrite (welded tuff) plateau (ages
of 4.6, 4.68 ±0.3, and 4.88 ±0.4 Ma are reported for various samples)
• Amargosa Valley, cuttings from a buried basalt gave ages of 3.85 ±0.05 and 4.4
±0.07 Ma
Southeast Crater Flatbasalt lavas (4.27 to 3.64 Ma) are the most areal-extensive
of the YPB
Buckboard Mesa basaltic andesite (3.07 ±0.29 to 2.79 ±0.10 Ma) erupted from a
scoria cone in the northeast part of the ring-fracture zone of the Timber Mountain
caldera and from nearby fissures
Quaternary Basalt of Crater Flat consists of a series of four northeast trending
basalt centers extending along the axis of Crater Flat including the Little Cones
(0.76 ±0.20 to 1.1 ±0.3 Ma), the Red and Black Cone centers (1.55 ±0.15 to 0.84
7-47
-------
±15 Ma and 1.09 ±0.3 to 0.80 ±0.06 Ma, respectively), and the Makani Cone
(1.66 ±0.519 to 1.04 ±0.03 Ma)
• Sleeping Butte Centers are two small volume (<0.1 km3) basaltic centers about
2.6 km apart with an estimated age of 0.38 Ma based on recent argon isotope
dating measurements
Lathrop Wells Center, the youngest and most thoroughly studied center of
basaltic volcanism, involved multiple eruptions over more than 100,000 years
Three alternative models involving various chronologies of volcanic events have been proposed
by DOE to explain the eruptive history of the Lathrop Wells volcanic center. These include a
four-event eruption model (eruption at >0.13,0.08 to 0.09,0.065, and 0.004 to 0.009 Ma), a
three-event eruption model (eruptions at 0.12 to 0.14, 0.065, and 0.004 to 0.009 Ma), and a two-
event eruption model (eruptions at 0.12 to 0.14 and 0.004 to 0.009 Ma). Exact dating of the
eruptions has been problematic and the exact number and timing of the eruptions is not certain,
but the youngest eruption is believed to be less than 10,000 years old. This most recent activity
was restricted to minor ash deposits (TRB95).
Summary
The majority of the silicic volcanic rocks that form the most important units in the Yucca
Mountain stratigraphic section were deposited about 11 to 15 Ma. This silicic volcanism ceased
about 7.5 Ma. Silicic volcanism was followed by two subsequent episodes of basaltic volcanic
rock formation. In the first episode, basalts of the silicic episode were deposited about 10 Ma.
In the second or post-caldera episode, smaller eruptions occurred beginning 8 to 10 Ma and
continuing to near present time. The youngest basaltic rocks at the Lathrop Well volcanic center
have ages between 4,000 and 9,000 years.
Both DOE and NRC agree that a future occurrence of silicic volcanism is highly unlikely and
therefore the consequences of such an event need not be considered in system performance
assessment. However, DOE and NRC have not reached agreement on the treatment of igneous
activity associated with possible future basaltic volcanic events.
19 This value appears to be an anomaly and will be investigated further.
7-48
-------
Given the history of volcanism in the Yucca Mountain Region, there is some probability that a
volcanic event can either intersect the repository footprint and directly affect the waste or that a
nearby intrusive dike can indirectly affect the natural and engineered barriers. In TSPA-93
(DOE94a), DOE used available data to estimate the impact of indirect magmatic effects, such as
heating or attack by aggressive volatiles on waste packages, when contact of the waste packages
with magma does not occur. Assuming that the waste packages were vertically emplaced, such
that the thermal loading they produced was 57 kW/acre, the magmatic effect on peak drinking
water doses is virtually indistinguishable from a case in which magmatic effects are not
considered.
In subsequent activities to address the stochastic uncertainty associated with the possibility that a
future magmatic event may intersect the repository, DOE convened a panel of 10 experts and
used a formal elicitation process to develop disruption20 probability estimates (DOE96f). Results
of the elicitation include (DOE97a):
• A mean annual disruption probability of 1.5x10'*
A 95 percent confidence interval of 5.4X10'10 to 4.9x10'8
Upper and lower bounds of 10'10 to 10'7
The NRC has taken a different tack in establishing the probabilities of volcanic disruption. The
NRC approach considers spatial patterns of basaltic volcanism, regional recurrence rates of
volcanic activity, and structural controls on volcanism in the Yucca Mountain Region (NRC96).-
Using two different measures to assess the impact of structural controls on volcanism (density of
high dilation-tendency faults and horizontal gravity gradients), two methods to assess spatial-
temporal distributions (near-neighbor and Epanechnikov kernel methods) and regional recurrence
rates varying from two to 10 volcanoes per million years, calculated probabilities based on
NRC's bounding approach ranged from 1x10'8 to 2x10"7 volcanic disruptions per year (NRC96).
Based on a homogeneous Poisson model (i.e., with a time invariant rate), the probability of at
least one volcanic disruption event occurring in 10,000 years, using DOE's estimated maximum
(95 percent confidence) disruption rate of 4.9x10"8/y, is 0.0005. Based on the maximum
disruption rate estimated by NRC of 2xlO-7/y, the probability of at least one disruption is 0.002 in
10,000 years.
20 Disruption is the physical intersection of magma with the potential repository volume (DOE97a).
7-49
-------
In its 1996 Phase 3, Yucca Mountain Total System Performance Assessment, EPRJ did not
include consideration of volcanism (EPR96). This position was based on an assessment made by
one member of the expert panel — one of 10 volcanologists sponsored by DOE — who
estimated that the annual probability of a magmatic intrusion into the proposed repository is 1.0
x lO"8.
Scientists at UNLV, supported by the State of Nevada, have considered a number of alternative
modeling approaches to volcanism. (See, for example, H096 and HO95.) Using a non-
homogeneous Poisson model (i.e., with a time varying rate), Ho estimated the probability of at
least one disruption in 10,000 years to lie between 0.0014 and 0.03.
DOE plans to conduct further analyses related to igneous activity hi the TSPA-VA scheduled for
publication in 1998 (DOE97b).
7.1.1.8 Geologic Stability Issues
The NAS Panel report states that the Yucca Mountain site will exhibit long-term geologic
stability on the order of one million years (NAS95). This implies that the contribution of
geology to overall system performance can be assessed for that time period. The Panel therefore
concludes that there is no need to arbitrarily select a shorter compliance evaluation period, such
as 10,000 years. The Panel recommends "...that compliance assessment be conducted for the
time when the greatest risk occurs, within the limits imposed by long-term stability of the
geologic environment."
This section examines the Panel's assertion of long-term geologic stability and related issues.
Factors addressed include characteristics of the geologic and hydrologic systems implied by the
Panel's concepts of "stable" and "boundable;" validity of the assertion of stability; and the
significance of stability to the occurrence, magnitude, and evaluation of peak dose. Geologic
stability does not imply absence of geologic activity or absence of changes in geologic processes,
but rather that any changing characteristics of the system do not introduce uncertainties of
sufficient magnitude to compromise the ability to perform credible analyses of future repository
performance.
7-50
-------
Characterization of Geologic Stability by the NAS Panel
The NAS report (NAS95) does not specifically define geologic stability. The existence of
stability is discussed six times in the report, in different ways:
The geologic record suggests that [the time frame during which the geologic system is
relatively stable or varies in a boundable manner] is on the order of one million years.
(Executive Summary, page 9)
...the long-term stability of the fundamental geologic regime [is] on the order of one
million years at Yucca Mountain, (page 55)
The long-term stability of the geologic environment at Yucca Mountain... is on the order
of one million years, (page 67)
The time scales of long term geologic processes at Yucca Mountain are on the order of
one million years, (page 69)
The time scale for long-term geologic processes at Yucca Mountain is on the order of
approximately one million years, (page 72)
The geologic record suggests that [the time frame over which the geologic system is
relatively stable or varies in a boundable manner] is on the order of about one million
years, (page 85)
These characterizations of geologic stability are quite similar, although some are expressed in
terms of the geologic regime itself and others are described in terms of the processes that operate
on or within that regime. These two assertions are not necessarily the same. For example,
characteristics of the geologic regime that are important to peak dose evaluation might remain
stable while tectonic and other natural processes and events continue in the future, even varying
from past characteristics. Alternatively, natural processes and events may continue in the future
as they have occurred in the past (i.e., the processes and events exhibit stability), while the
effects they produce may change the features of the geologic regime that are important to peak
dose evaluation. Conditions in which past and continuing tectonic movement produces
differential movement of deep geologic structures might cause changes in the hydrologic regime
important to the occurrence of the peak dose. The various expressions of stability used in the
Panel's report imply no significant change in either the geologic regime or in the processes and
events that affect the characteristics of that regime.
7-51
-------
The Panel's report does not explicitly justify the assertion of million-year stability by providing a
synopsis and interpretation of the geologic record. Some of the references cited in the report
contain information about the geologic record (e.g., DOE's Site Characterization Plan for the
Yucca Mountain site (DOE88)), but none of the cited references interprets the record to indicate
a million-year stability of the geologic regime or the processes associated with it.
Existing Documentation Related to Stability
Existing documentation does not directly address long-term stability of the natural features of
Yucca Mountain and its environs. Until quite recently, the DOE documents containing
information about the geologic features of the Yucca Mountain site anticipated that evaluations
of site suitability would be made in accord with DOE's 10 CFR Part 960 Site Suitability
Regulations and anticipated safety performance of a repository at the site would be evaluated in
terms of EPA's 40 CFR Part 191 regulations and NRC's 10 CFR Part 60 regulations. Under this
regulatory framework, the time period of concern is 10,000 years.
The 10,000-year tune frame for compliance with EPA's 40 CFR Part 191 regulation was selected
by the Agency because it was short compared to long-term factors, such as tectonic motion, that
might affect and change in ways that could not be characterized, the natural environment
conditions important to regulatory compliance evaluations. On the other hand, the time period
was long enough to bring into consideration, at least in principle, factors such as seismicity that
are important in geologic time scales and might affect repository performance.
The DOE has, in many Yucca Mountain project documents, implied geologic stability or the
equivalent for time periods of 10,000 years. The State of Nevada believes, however, that the
record does not justify such a conclusion. For example, the State asserts in its comments
(NEV85) on DOE's draft Environmental Assessment (DOE84) for the Yucca Mountain site, that
DOE's conclusion that "neither major tectonic activity nor the resumption of large-scale silicic
volcanic activity in the area near Yucca Mountain is likely in the next 10,000 years" is
premature, based on existing evidence. The State also asserts that "possible hydrovolcanic
activity at Yucca Mountain has not been sufficiently evaluated" (NEV85, Volume II, page 125).
In general, the documents of record show controversy over the stability of the geologic regime
and associated natural processes and events at the Yucca Mountain site. The controversy stems
7-52
-------
both from opposing interpretations of the available data by DOE and the State of Nevada and by
differing definitions of geologic stability. To some extent, the opposing viewpoints reflect the
institutional positions of the parties involved; nonetheless, the uncertainties in the data permit
alternative interpretations to be made and controversy to persist.
Interpretation of the Geologic Record Related to Stability
The geologic history of the area provides the basis for assertions concerning the stability of the
geologic regime for Yucca Mountain and its vicinity. Site characterization activities for DOE's
Yucca Mountain project, and other activities unrelated to the Yucca Mountain project (e.g.,
commercial characterization of natural resource potential), have yielded an extensive data base
concerning geologic features and the geologic record of the region. The most comprehensive
data available for judging the geologic stability of the Yucca Mountain site are presented in
DOE's Site Characterization Plan (DOE88).
Such data do not, however, definitively resolve the question of the long-term stability of the
geologic regime. Such issues can be resolved only in context, through the expert judgment of the
involved parties. The NAS Panel's assertion of long-term geologic stability at Yucca Mountain
for the next million years is an example of such judgment.
The basis for the Panel's judgment of the geologic stability of Yucca Mountain over the next one
million years is the conclusion that the properties and processes of the geologic regime
important to repository performance "... are sufficiently understood and stable over the long tune
scales of interest to make calculations [of repository performance] possible and meaningful"
(NAS95, page 68). The relevant properties and processes include the radionuclide inventory of
the waste, the influx of water to the repository, migration of the water and its contained waste
materials from the repository to the ground water, and subsequent dispersion and migration of
contaminated ground water to the regional biosphere. The Panel considers it possible, for
example, to estimate, with acceptable uncertainty, concentrations of wastes hi ground water at
various locations and times for the purpose of a bounding safety assessment.
The assertion of geologic stability implies a judgment that the basic features of the geologic
regime that affect waste release and transport will remain as they are, or change in a limited and
reasonably predictable fashion, over the next million years. In other words, phenomena that
7-53
-------
would substantially and unpredictably change the current, relevant geohydrologic regime are not
expected. Such phenomena would include tectonic motion, seismicity, and volcanism sufficient
to change the features of the geologic regime that govern radionuclide release and transport.
The Panel's assertions also imply that the geologic and hydrologic features of the site and region
can and will be characterized in a way that allows repository performance to be reliably projected
on the basis of current conditions. Two of the parameters cited by the panel as important to
predicting the performance of the repository—water influx to the repository and dispersion and
migration of ground water in the biosphere—are demonstrated by DOE modeling studies
(DOE95b, herein also termed TSPA-95) to be highly important to estimating potential health
effects from the repository. However, these two parameters are currently among the least well-
known of the parameters related to repository performance.
The DOE performance assessment reports indicate that these hydrologic parameters will be
extremely difficult to evaluate reliably. As DOE notes hi TSPA-95, direct observation of water
infiltration rates is not possible. Consequently, TSPA-95 treats the infiltration rate to the
repository as an uncertain parameter. Bounding values, consistent with the NAS Panel's concept
of bounding, can be established, but the bounds may have to be narrowed considerably from
present ranges to be meaningful to the process of determining compliance.
This situation raises an issue not addressed directly by the NAS Panel: Can key performance-
related parameters be adequately characterized? The long-term geologic stability of the Yucca
Mountain site may be less important to evaluating repository performance than the actual values
of those parameters most significant to its performance. As the example given above
demonstrates, the variability of a parameter such as infiltration rate presents an obstacle to
characterizing reliably the long-term risks to the critical group. In addressing the overall
question of long-term repository performance, the uncertainty associated with these factors may
be much more significant than the uncertainty associated with the long-term geologic stability of
the site.
Summary of Evidence for Stability
The information presented in this chapter generally supports the NAS Panel's assertion that the
fundamental geologic regime at Yucca Mountain will remain stable over the next one million
7-54
-------
years. The overall picture that emerges from the data is that the site and region had a highly
dynamic period of volcanism, seismicity, and tectonic adjustment in the past, but these processes
and events have matured into a system in which the magnitudes, frequencies, locations, and
consequences of such phenomena can be bounded with reasonable confidence relative to
assessing the long-term repository performance.
The possible exception to this finding is the chance that on-going processes and events are
producing differential changes to the geologic and hydrologic regimes that are currently
unrecognized but could affect repository performance and potential radiation risks for affected
populations in the future. For example, on-going tectonic processes and movements could
potentially have different effects on the geologic and hydrologic regimes near the surface and at
depth, and the at-depth changes may not be readily recognizable. At present, tectonic movement
in the area varies by location but falls generally within the range of four to 10 mm/year
(DOE95a). Over one million years, an annual tectonic movement of 10 mm/year will produce a
total translation of location of about 5 miles.. If all of the elements of the geologic and
hydrologic regime important to repository performance and dose estimation do not move
together in space and time, the differential movement could invalidate the results of performance
and exposure assessments. The potential for differential movement and its consequences are not
yet addressed.
Perspective on the Significance of Stability of the Geologic Regime
A judgment that the geologic regime at Yucca Mountain will be stable for one million years
enhances confidence in the results of model-based assessments of the effects of natural processes
and events over that time frame on repository performance. Long-term natural phenomena may
not, however, control repository performance or uncertainties in performance assessment results.
Uncertainties in other factors involved in performance projections may ultimately control the
reliability of the projections.
The existence of long-term geologic stability can assure reliable estimation of long-term peak
doses only if stability-related issues are confirmed to dominate repository performance and
numerical values of relevant parameters have been established with confidence. As discussed
subsequently hi Section 7.3, DOE's total system performance assessments indicate that the rate
°f infiltration of water to the repository and the dilution and dispersion characteristics of ground
7-55
-------
water containing radioactive contamination released from the repository are the dominant factors
in repository performance and dose assessment. If the repository design is altered, these may no
longer be the dominant factors.
These findings suggest that geologic stability is not significant unless it affects these water-flow
parameters (e.g., through differential tectonic displacement). However, the scope of DOE's
performance assessments is to date highly limited. In addition, DOE carefully notes (Chapters 9
and 10, DOE95b) that the approach to dose estimation used in its TSPA evaluations does not
correspond to that considered by the NAS Panel. Overall, DOE's performance assessments to
date have not attempted to establish a perspective on geologic stability and other factors that
might affect repository performance and radiation doses for one million years.
The DOE's performance assessments to date for Yucca Mountain have emphasized release of
nuclides from the repository over a 10,000-year time frame, in response to the requirements of
EPA's 40 CFR Part 191 regulations, which were applicable until enactment of the WIPP Land
Withdrawal Act. Experience in evaluating repository performance over a 10,000-year time frame
(DOE94a, DOE95b) has shown that repository conditions must be assessed at, or near, the time
when key performance parameters, such as temperature, may be at their peak values. The
10,000-year time frame encompasses the time of highest uncertainty in the effect of repository
design factors important to waste isolation and safety performance. These uncertainties may
have a greater effect on predicting long-term repository performance and regulatory compliance
than a natural process or event, such as an earthquake or a volcanic eruption. This is due to the
high degree of uncertainty in the "nominal" dynamics and performance of the repository's
barriers and the low probability of a major natural process or event occurring.
Beyond 10,000 years, however, the technical factors associated with repository design features
that dominate performance issues earlier may become less important to determining regulatory
compliance at the time of peak dose. If the engineered barrier system is likely to have failed in
the long term, radionuclides will be available for transport to the environment. The DOE
performance assessment report by Intera, Inc. (DOE94b) states that variations in assumptions and
conditions for waste package degradation produce less than a 20 percent variation in results for a
10,000 year assessment period and less than a 10 percent variation in results for a 100,000 year
period. Supplemental calculations in DOE94c show that peak doses and releases at the
accessible environment boundary over a one million-year period are generally unaffected by
7-56
-------
waste package lifetimes up to 100,000 years. It is in the time period beyond 10,000 years that
the issue of long-term geologic stability becomes more important to repository performance.
In summary, three periods of repository conditions in the future can be characterized, with
geologic stability being maintained throughout all three. In the first, short-term period, lasting
about 100 to 1,000 years, the repository is characterized by intact waste canisters, high
temperatures and temperature gradients which serve as driving forces for transients such as
chemical reaction, and the retention of short-lived and long-lived radioactivity in the canisters.
Infiltrating water may or may not contact the canisters.
In the intermediate period, with a duration between 1,000 and 10,000 years, gradients are
diminishing or gone and the engineered features of the repository are degrading. During this
time, canisters are corroding; only long-lived radioactivity remains; some of the radioactivity
from the waste is released from the canisters, but most is retained within the repository.
Infiltrating water contacts and transports radioactive waste.
In the long-term period, from 10,000 to 1,000,000 years, the repository is essentially an
isothermal ore body of the oxides, hydroxides, or carbonates of waste-package materials at
ambient conditions. Infiltrating water seeps through the bed of oxides and transports long-lived
radioactivity to the environment, where the radioactive contamination is diluted and dispersed by
ground-water flow processes.
Given this perspective, the transitional processes associated with the engineered features and
heat-emission characteristics of the repository will essentially be complete in one percent of the
elapsed time of a regulatory period of 1,000,000 years. Therefore, the physical state of the
repository at 10,000 years can serve as the initial condition for the assessment of repository
performance and dose assessment under conditions of geologic stability for a period of 1,000,000
years.
7-1.2 Hydrologic Features
7.1.2.1 Unsaturated Zone Hydrology
The region beneath the surface of Yucca Mountain in the vicinity of the proposed repository is
characterized by a very thick unsaturated zone, ranging in thickness from about 500 to 750 m.
The variable thickness is produced by the combined effects of rugged topography and a sloping
Water table. The presence of a thick unsaturated zone is desirable for siting an underground
7-57
-------
waste repository because ground water, and any contaminants it might carry, generally travels
more slowly through the unsaturated zone than through the saturated zone. The thicker the
unsaturated zone, the longer contaminants will take to reach the water table.
In this document, and in the literature generally, the term unsaturated flow actually means
partially-saturated flow, since by definition there can be no water flow through a totally dry
medium. Unsaturated ground-water flow is more complex than fully-saturated flow because it
involves the simultaneous movement of water, air and water vapor. For unsaturated media, the
measure of permeability is called the effective hydraulic conductivity. The effective hydraulic
conductivity, and hence the rate of fluid flow, through any given partially-saturated porous
medium depends on the degree of saturation of that medium. The higher the saturation, the
greater the quantity of water that can flow through it, all other factors (saturated hydraulic
conductivity, hydraulic gradient, etc.) being equal. As the degree of saturation reaches
100 percent, the effective hydraulic conductivity approaches fully-saturated hydraulic
conductivity. The dependency between degree of saturation and effective hydraulic conductivity
is complex, due to the nonlinearity of the relationship.
The dependence of unsaturated flow on the degree of saturation is important to understand when
reading the following sections of this document because some of the phenomena described are
not intuitively obvious. An example of this is described later, where it is stated that water
moving downward in the partially-saturated zone encounters zones of increased effective
porosity, which may act as barriers to further downward flow. It may at first seem
counterintuitive that a zone of increased porosity could act as a flow barrier until one considers
that a geological zone with a high porosity possesses a low capillary suction potential. If this
zone is overlain by a zone which has a lower porosity and thus a higher capillary potential, water
entering the upper zone will be retained there as a result of capillary equilibration. These
conditions will prevail until the gravitational force overcomes the capillary force in the upper .
zone as more water enters, which usually happens when the bottom of the upper zone becomes
nearly saturated, allowing water to flow into the lower zone.
A sequence of nonwelded porous tuffs that overlies the Topopah Spring Member (Section 7.1.1)
may act as a natural capillary barrier to retard the entrance of water into the fractured tuffs. A
similar sequence of nonwelded tuffs underlies the Topopah Spring Member. These underlying
nonwelded tuffs locally contain sorptive zeolites and clays that could be an additional barrier to
the downward transport of some radionuclides from a repository to the water table.
7-58
-------
The proposed repository is surrounded by and crossed by numerous strike-slip and normal faults
with varying amounts of offset (LBL96). The repository would be located largely, if not entirely,
within what is known as the "central block" as described below (see Figure 7-8). The structural
geology of this block is less complex than in the surrounding area, although one extensive, nearly
vertical normal fault has been mapped in the block (Ghost Dance Fault). The central block of
Yucca Mountain is a large block beneath the center of the Yucca Mountain ridge and is bounded
on its west side by the Solitario Canyon fault, a major north-striking normal fault with greater
than 100 m of offset. West of this fault is a chaotic, brecciated and faulted west-dipping zone
caused by drag on the fault. A zone of imbricate normal faults forms the eastern boundary of the
central block. These faults are west-dipping and have vertical offsets of about two to five m.
Northwest striking strike-slip faults also occur in the area, such as the one forming the northern
boundary of the central block, beneath Drill Hole Wash. The concept of a central block should
not, however, be taken to imply that the central block or the proposed repository area is free of
faults (USG84a).
Unsaturated Zone Hydrogeologic Units
The detail of the layered volcanic rock sequence beneath Yucca Mountain is very complex. The
various rock units can be separated into a small or large number of units depending upon the
scale and aims of a particular study. For the purposes of this document, the unsaturated zone is
considered to consist of six hydrogeologic units, based on their physical properties. This
grouping and the description of the six units are based primarily on USG84a, except where
otherwise referenced. Additional data regarding matrix and fracture properties are presented in
the hydrogeologic database developed in DOE95c.
The physical properties within each formation vary considerably, largely due to variation in the
degree of welding of the tuffs. In most cases, physical property boundaries do not correspond to
rock-stratigraphic boundaries. However, it is the physical properties that largely control water
occurrence and flow; the hydrogeologic subunits into which the volcanic sequence is separated
are different than the lithological units outlined in Section 7.1.1.3. The hydro-geologic units are,
in descending order, Quaternary Alluvium (Qal), the Tiva Canyon welded unit (TCw), the
Paintbrush nonwelded unit (PTn), the Topopah Spring welded unit (TSw), the Calico Hills
nonwelded unit (CHn), and the Crater Flat unit (CFu). Figure 7-17 illustrates these
hydrogeologic units and some of their characteristics. They are described in detail in the
following paragraphs.
7-59
-------
1 Thicknesses from geologic sections of Scott and Bonk US84).
' Scott and others (11831.
> Inferred from phtpieal properties.
Figure 7-17. Unsaturated Zone Hydrogeologic Units (USG84a)
Structural features, although they are not hydrogeologic units in the same sense as stratigraphic
units, are mappable, have certain measurable hydraulic characteristics, and may have a
significant effect on unsaturated zone flow. Because these structural features are regarded as
important components of the unsaturated hydrologic system, they are described later in this
section.
Qal. Unconsolidated alluvium underlies the washes that dissect Yucca Mountain and forms the
surficial deposit in broad inter-ridge areas and flats nearby. Thickness, lithology, sorting, and
permeability of the alluvium are quite variable; particles range in size from clay to boulders, and
7-60
-------
in places the unit is moderately indurated by caliche. Alluvial and colluvial deposits generally
have small effective hydraulic conductivity, large specific retention, and large effective porosity
as compared to the fractured rocks. Therefore, a large proportion of the water infiltrated into the
alluvial and colluvial material is stored in the first few meters of the soils and is lost to
evaporation during dry periods. The saturated permeability of alluvium generally is substantial
compared to the tuff units.
TCw. Lying immediately beneath the Qal is the Tiva Canyon welded unit, consisting of
devitrified ash-flow tuffs ranging from 0 to 150 m in thickness across the site. The TCw is the
densely to moderately-welded part of the Tiva Canyon Member of the Paintbrush Tuff. This unit
is the uppermost stratigraphic layer that underlies much of Yucca Mountain; it dips 5 ° to 10°
eastward within the central block, resulting in a relatively planar eastward-sloping, dissected land
surface. The unit is absent in some washes and is about 150 m thick beneath Yucca Crest. This
unit has a fracture density of 10 to 20 fractures/m3 and small matrix permeability. Saturated
matrix hydraulic conductivity has been estimated at about 2x10-6 m per day (m/d); the effective
hydraulic conductivity is thought to be lower, as saturation is estimated to range from 60 - 90
percent. Neither bulk rock nor fracture hydraulic conductivities are well characterized for this
unit.
pTn. The Paintbrush nonwelded unit is situated below the TCw unit and consists of the
nonwelded and partially welded base of the Tiva Canyon Member, the Yucca Mountain Member,
toe Pah Canyon Member, the nonwelded and partially-welded upper part of the Topopah Spring
Member, and associated bedded tuffs. All are part of the Paintbrush Tuff. The unit consists of
thin, nonwelded ash-flow sheets and bedded tuffs that thin to the southeast from a maximum
thickness of 100 m to a minimum thickness of about 20 m. The unit dips to the east at 5° to 25°;
toe dip at any location depends on the tilt of the faulted block at that site. In the central block,
toe dip rarely exceeds 10°. In the vicinity of the central block, this unit crops out in a narrow
band along the steep west-facing scarp along Solitario Canyon.
Tuffs of this unit are vitric, nonwelded, very porous, slightly indurated, and in part, bedded. The
utit has a fracture density of about one fracture/m3. Saturated hydraulic conductivities of five
c°re samples of the matrix have a geometric mean of about 9.0x10'3 m/d. Porosities average
about 46 percent, but some porosities are as much as 60 percent. The rocks of this unit are
Moderately saturated, with an average value of about 61 percent. However, water contents are
relatively large; the mean volumetric water content is about 27 percent and the mean water
7-61
-------
content by weight is about 19 percent. The maximum values reported are: saturation, 80 percent;
volumetric water content, 42 percent; and water content by weight, 36 percent.
TSw. The Topopah Spring welded unit consists of a very thin upper vitrophyre, a thick central
zone consisting of several densely welded devitrified ash-flow sheets and a thin lower vitrophyre
of the Topopah Spring Member of the Paintbrush Tuff. The unit, which varies from 290-360 m
in thickness, is densely- to moderately-welded and devitrified throughout its central part. The
TSw contains several lithophysal cavity zones that generally are continuous, but vary appreciably
in thickness and stratigraphic position. The TSw is also intensely fractured.
The Topopah Spring Member is the thickest and most extensive ash-flow tuff of the Paintbrush
Tuff. The central and lower densely-welded, devitrified parts of the Topopah Spring welded unit
are the candidate host rock for a repository. This part of the unit contains distinctive subunits
that have abundant lithophysal gas cavities within the central block. The saturated hydraulic
conductivity of the matrix of this unit generally is small and has a mean of about 3.0xlO'6 m/d.
Because of the densely fractured nature of this unit, bulk hydraulic conductivity is substantially
greater than matrix hydraulic conductivity. Saturated horizontal hydraulic conductivity of the
rock mass is about one m/d for a 120-meter interval of the TSw that was packed off and tested at
Well J-13 (see Figure 7-18 for bore hole locations), about six km east of Yucca Mountain.
Because of the marked contrast between the matrix and the bulk hydraulic conductivities in this
unit, values of the bulk hydraulic conductivity from Well J-13 (USG83) and borehole UE-25a#4
probably represent the hydraulic conductivity of the fractures in this unit. The large bulk
hydraulic conductivity of this unit probably promotes rapid drainage of water. The amount of
flow carried in the fractures with respect to the matrix has been estimated to range between 10 -
95 percent (GEO97).
The effect of lithophysal cavities on the hydrologic properties of the TSw is not well understood.
Total porosity is much greater where lithophysal cavities are more abundant than in those
sections that are free of these cavities. Overall unsaturated hydraulic conductivity probably is
decreased by the presence of these cavities. These cavities commonly are several centimeters in
diameter, filled with air, and form capillary barriers with the fine grained matrix. In effect, the
cavities decrease the transmissive cross-sectional area, decrease effective porosity, and
consequently, decrease the effective hydraulic conductivity.
7-62
-------
Figure 7-18. Locations of Deep Boreholes in the Vicinity of Yucca Mountain (USG96a)
n. Beneath the TSw unit is a series of non- to partially-welded ash-flow tuffs called the
Calico Hills nonwelded unit. Locally, these may be vitric (CHnv) or zeolitized (CHnz). The
includes the following components, in descending order:
1. A nonwelded to partially-welded vitric layer, locally zeolitic, that is the lowermost
part of the Topopah Spring Member of the Paintbrush Tuff.
2. Tuffaceous beds of Calico Hills.
3. The Prow Pass Member of the Crater Flat Tuff, which is nonwelded to partially-
welded where it occurs in the unsaturated zone beneath the central block.
7-63
-------
4. The nonwelded to partially-welded upper part of the Bullfrog Member of the
Crater Flat Tuff where it is above the water table.
In the vicinity of the central block, this unit crops out in a narrow band along the steep west-
facing scarp along Solitario Canyon. Both vitric and devitrified facies occur within the CHn. As
described below, the permeability of the vitric facies is substantially greater than that of the
devitrified facies. Alteration products in the devitrified facies include zeolites (most abundant),
clay, and calcite (rare). Because this facies is mostly zeolitic, it is hereafter referred to as the
zeolitic facies. Thickness of the zeolitic facies generally increases from the southwest to the
northeast beneath Yucca Mountain. Beneath the northern and northeastern parts of the central
block, the entire unit is devitrified and altered.
Both the vitric and zeolitic facies of the CHn are very porous, with a mean porosity of about 37
percent for the vitric facies and 31 percent for the zeolitic facies. Saturations in this unit
generally are greater than 85 percent, with a mean value for the zeolitic facies of about
91 percent.
A significant difference exists in values of vertical hydraulic conductivity of the matrix between
the vitric and zeolitic facies of the CHn. The mean vertical hydraulic conductivity of the matrix
of the vitric facies is 4.0xlO~3 m/d. The geometric mean of the vertical hydraulic conductivity of
the matrix of the zeolitic facies is about S.OxlO*6 m/d. The marked contrast in vertical hydraulic
conductivities of the two facies probably is the result of extensive argillization in the zeolitic
facies, which tends to decrease permeability.
CFu. In approximately the southern half of the central block, the lowermost unit in the
unsaturated zone is the Crater Flat unit. This unit consists of the unsaturated welded and
underlying nonwelded parts of the Bullfrog Member of the Crater Flat Tuff. No differentiation is
made between the welded and nonwelded components of the Crater Flat unit because of the
limited extent of the unit in the unsaturated zone beneath the central block, and therefore, its
probable limited effect on the unsaturated flow system. Beneath the central block, the thickness
of the CFu ranges from 0 to 160 m. Little is known about the unsaturated hydrologic properties
of the unit, but it is assumed that the properties are similar to those of the nonwelded and welded
counterparts higher in the section.
7-64
-------
Structural Features
As previously described, the central block of Yucca Mountain is bounded on three sides by
faults. Because these major faults and fault zones transect the full thickness of the unsaturated
zone, they may by hydrologically significant either as flow barriers or as flow pathways. The
variation in unsaturated hydraulic properties of these features have in most cases not been
measured. However, some inferences can be made, based on the physical properties of the
welded and nonwelded tuff units and on observations of drill cores.
The welded units are relatively brittle. Open faults have been observed in cores even from below
the water table. Conversely, the nonwelded units generally are more ductile than the welded
units and more readily produce a sealing gouge material. Fault zones are less common in the
Calico Hills nonwelded unit In general, hydraulic conductivity varies greatly along the faults
and is greater in welded units than in nonwelded units (USG84a),
Knowledge of the permeability of the numerous faults which cross Yucca Mountain is important
because some faults may act as conduits for rapid vertical flow in the unsaturated zone. This
possibility is especially critical in areas in which such faults may intercept large amounts of
lateral flow and divert this flow downward, potentially into the repository. Evidence for the
permeability of the faults in and around the proposed repository area is mixed. Studies
performed to date indicate that particular faults are barriers, while other faults are more
permeable (LBL96). It is also possible that a particular fault may be relatively impermeable in
some areas of the fault plane, and relatively permeable in others. Factors which may reduce
permeability of faults include development and alteration of fault gouge, deposition of fracture
coating materials on fault surfaces, and the juxtaposition of permeable and nonpermeable units
by movement along the fault plane. Faulting can also create zones of enhanced permeability
where the rock around the faults is highly fractured or brecciated.
Studies in the Exploratory Studies Facility (ESF) indicate that the permeability of the Bow Ridge
fault is about the same as measured with air permeability testing of highly permeable bedded tuff
formations or highly fractured welded units. Also, the geothermal profile in borehole ONC#1
shows that the geothermal profile is offset by several degrees as the borehole passes through the
Bow Ridge fault zone. This indicates that the fault may be highly permeable to gas or moisture
flow which decreases the temperature in that region (LBL96).
7-65
-------
Evidence from other faults indicates that they may act as low permeability barriers. For instance,
the water body observed at borehole SD-7 is thought to be perched over a zeolitic layer and
prevented from moving laterally by the presence of the Ghost Dance fault. A similar hypothesis
has been invoked to explain perched water in a borehole intersected by a splay of the Solitario
Canyon fault. This conclusion is corroborated by pneumatic pressure data taken in borehole UZ-
7a, which appear to show a degree of anisotropy in the fault which is consistent with a
permeability barrier, at least in the horizontal direction (LBL96).
Another indication that some faults at the site may act as permeability barriers is obtained from
potentiometric surface measurements. For instance, the potentiometric surface elevation on the
western side of the Solitario Canyon fault is approximately 40 m higher than on the eastern side
of the fault. This gradient could only be maintained if the Solitario Canyon fault is somehow a
permeability barrier to flow (LBL96).
The ESF has provided data and observations regarding the structural features within Yucca
Mountain. Prior to the construction of the ESF, detailed geological and structural cross-sections
were prepared. As-built cross sections prepared from data and observations from the ESF show
that geologic sections drawn prior to construction compare favorably with results from tunneling.
These findings indicate that the lithostratigraphy, and to a lesser extent structure, of this are well-
characterized and predictable. Detailed information on the results of ESF geological mapping is
available in BOR96 and BOR96a. These publications provide detailed fracture pattern analysis
including measurements of trace length, orientation, continuity, roughness, aperture, and mineral
infilling. From ESF studies, three main fracture sets are reported; two are approximately vertical
and strike north-south, and east-west, while the third fracture set is close to horizontal. BOR96
reports that the open distance between fracture faces averages 2.3 mm over the entire fracture
population. The largest aperture is 91 mm, although this is anomalously large in this population;
67 percent of the fractures are closed (0 mm). For fractures with an aperture greater than zero,
the average is 7.2 mm. The fracture population includes measurements from the Tiva Canyon
Tuff, the Paintbrush Tuff, and the Topopah Spring Tuff. The repository horizon is generally
more fractured, containing an average of about four fractures per meter, but typically ranges from
about two to six fractures per meter (LLNL96).
A common feature in some horizons in the volcanic rocks are lithophysal cavities, which are
voids in the rock presumably created by gases exsolved from cooling lavas and pyroclastic
7-66
-------
deposits. In the Tiva Canyon and Topopah Spring Tuffs, lithophysae are mostly concentrated
into stratiform zones, but they also occur adjacent to lithophysal zones and sporadically in
nonlithophysal zones. The cavities range in size from less than one centimeter (cm) to greater
than 1.4 m. Fractures demonstrate several different relationships with lithophysal cavities.
Fractures that intersect and terminate in lithophysal cavities are common. This, and other
evidence, suggest that lithophysal cavities may locally influence fracture propagation (BOR96,
BOR96a).
Ground Water Flow In The Unsaturated Zone
Water flow and storage in the unsaturated zone is three-dimensional and is controlled by the
structural, stratigraphic, thermal, and climatological setting. The dynamics of water-air-vapor
flow in the layered, fractured rock unsaturated zone beneath Yucca Mountain are complex and
highly uncertain at this time. In the unsaturated zone, water is present both in liquid and vapor
phases within the interstitial, fracture, and lithophysal openings. Hydrogeologic features that
probably affect flow significantly in the unsaturated zone include the presence of fractured
porous media, layered units with contrasting properties, dipping units, bounding major faults,
and a deep water table. These features probably result in the occurrence of phenomena such as
flow in both fractures and matrix, diversion of flow by capillary barriers, lateral flow, perched
ground water zones, and vapor movement.
Infiltration Rates
The ultimate source of water in the unsaturated zone at Yucca Mountain is precipitation on the
mountain. The spatial and temporal relationships between infiltration and recharge are complex,
because of the hydrogeologic variability of Yucca Mountain. Some water that infiltrates returns
to the surface by interflow; another part is returned to the atmosphere by evapotranspiration. A
small quantity that is not evaporated, or discharged as interflow, percolates deep into the
unsaturated zone and becomes net infiltration or percolation. The terms "infiltration" and
"percolation" are used frequently, sometimes interchangeably, in literature about the Yucca
Mountain unsaturated zone. For the purposes of this report, "infiltration" is used to describe the
amount of water which enters Yucca Mountain at the ground surface, while "percolation" is used
to describe the amount of water which actually penetrates deep enough into the mountain to
reach the repository horizon and below. The difference between the two terms lies mainly in the
7-67
-------
partitioning of part of the infiltration flux into the vapor phase, which may then be recirculated to
the atmosphere.
At Yucca Mountain, the infiltration rate is both spatially and temporally variable. Because the
quantity of net infiltration that percolates through different paths is quite variable, estimated
average recharge rates do not represent percolation rates through specific flow paths. Spatial
variations of infiltration depend mostly on variations in the properties of surficial units,
topography, the intersection of faults with the surface, and the presence of local fracturing.
Temporal variations in infiltration rate are related to the seasonality and relatively infrequent
precipitation events in the arid climate of Yucca Mountain. Temporal variations in the
infiltration rate have also occurred over a much larger time span, reflecting long term climate
changes.
Knowing the temporal and spatial variability of the percolation rates is crucial to modeling
efforts because of the importance of the relationship of infiltration rate to horizontal and vertical
permeabilities of the various units and the effect this has on whether or not significant lateral
flow occurs in the unsaturated zone. The higher the actual infiltration rate, the greater the
likelihood of significant lateral flow. Such lateral flow could result from a combination of two
factors. The first factor is that infiltrating water may encounter zones of lower relative
permeability as it moves downward. The second factor is that in many of the units, the relative
permeability is far greater in the direction parallel to bedding than the direction perpendicular to
it The anisotropic permeability may cause lateral flow of mounded water away from the area in
which it accumulates. Lateral flow is important because it could transmit water to structural
features which would then move the water downward, possibly acting as a conduit to divert large
amounts of water flowing downward through a small area. Such flow paths could direct water
into and through the repository or away from it.
The actual quantity of net infiltration or percolation beneath the surface of Yucca Mountain has
not been accurately determined. The percolation flux is a difficult parameter to determine for
low flux regions such as Yucca Mountain. There are currently no reliable direct measurements
that can be made to determine this important parameter (LBL96). Existing estimates have been
obtained from a mixture of indirect methods involving field testing and modeling of various
processes at different scales. Data exist to suggest that the flux reaching the repository horizon
through the matrix is relatively small. Relatively low matrix saturations measured in the upper
portion of the TSw suggest that much of the moisture which infiltrates into the TCw does not
7-68
-------
reach the TSw (LBL96), Data from the ESF show that no weeping fractures were found, even in
the region where perched water is found in boreholes. (Note, however, that because of
ventilation equipment inside the ESF, much of any such moisture might be removed from the
ESF as water vapor). Furthermore, no moisture was observed infiltrating into the radial
boreholes of Alcove 1 of the ESF after storm events, even though the boreholes are located close
to the land surface in the highly fractured and broken TCw formation (LBL96). However, other
data suggest that the percolation flux may reach the repository level mainly through episodic
fracture flow. These data include observation and testing of extensive bodies of perched water
located below the repository horizon, as well as measurements of bomb-pulse isotope levels from
atmospheric nuclear testing which show that some water in the unsaturated zone is relatively
young (LBL96).
Estimates of net infiltration vary from slightly negative (net loss of moisture from the mountain)
to about 10 rnrn/yr (LBL96). USG84a reports that net infiltration flux probably ranges from 0.5
to 4.5 mm/year, based on estimates of earlier workers for various localities in the Yucca
Mountain area. Flint and Flint (FLI94) provide preliminary estimates of spatial infiltration rates
that range from 0.02 mm/yr, where the welded Tiva Canyon unit outcrops, to 13.4 mm/yr in
areas where the Paintbrush nonwelded unit outcrops. The bulk of the area above the repository
block is underlain principally by the Tiva Canyon member. The DOE's 1995 Total System
Performance Assessment (DOE95b) concludes that, if the predominant flow direction is vertical,
then the average infiltration through the repository block, using the average infiltration rates of
Flint and Flint (FLI94), would be 0.02 mm/yr. If, on the other hand, the predominant flow
direction has a significant lateral component due to material property heterogeneity and/or
anisotropy and the sloping nature of the hydrostratigraphic unit contacts, then the average net
infiltration rate over the repository block could be as high as some weighted average of the
infiltration rates inferred from FLI94. The 1995 TSPA (DOE95b) also reports that the average,
spatially-integrated infiltration rate is about 1.2 mm/yr; most of this infiltration occurs along the
Paintbrush outcrop in the washes north of the repository block.
Recently, several lines of evidence have converged to alter the prevailing view regarding the
magnitude of infiltration/percolation rates beneath Yucca Mountain, with the most recent
estimates being revised upward from previous work. The newer estimates of percolation are
around five mm/yr, with a range of one to 10 mm/yr (LANL96, LBL96). Recent isotopic
analyses of rock samples from the ESF are consistent with a percolation rate of five mm/yr
(LANL96, LBL96). Profiles of temperature vs. depth of water in boreholes are consistent with a
7-69
-------
range of infiltration rates from one to 10 mm/yr (LBL96). Three-dimensional modeling results
of percolation flux at the repository horizon using the latest available spatially varying
infiltration map indicate percolation fluxes on the order of five to 10 mm/yr. The expert
elicitation panel estimates for mean infiltration rates range from 3.9 to 12.7 mm/y (GEO97). The
effect of uncertainty in infiltration and percolation flux rates is examined in the discussion of the
unsaturated zone conceptual model.
Conceptual Model(s)
The first detailed conceptual model of unsaturated zone flow at Yucca Mountain was proposed in
USG84a. Since then, the majority of the data collected has been in general agreement with these
ideas and concepts (LBL96). Most subsequent conceptualizations of unsaturated zone behavior
are largely refinements of this model, revised to accommodate newly-acquired data (Figures 7-19
and 7-20). Newly-acquired data include isotopic analyses, concentration ratios of ions dissolved
in matrix rocks and perched water zones, calcite fracture fillings, and thermal modeling of
vertical temperature gradients. Perhaps the most significant change from early conceptual
models has been the recent acquisition of new isotopic data which indicate the presence of "fast
paths" for water moving through the unsaturated zone. This topic is discussed in more detail in a
subsequent section.
The following presentation of the unsaturated zone flow conceptual model is taken primarily
from USG84a. Where appropriate, the published literature is referenced when describing
refinements or revisions that have been made to the USG84a model. The following conceptual
model is presented as if it were an established physical reality. Bear in mind, however, that the
proposed model is probably not the only reasonable description that could be made of the
system. Following the description of the conceptual model is a discussion of critical unknowns,
their effects on unsaturated zone flow, and results of numerical modeling studies.
Percolation of infiltrated water through the exposed fractures of the Tiva Canyon welded unit is
relatively rapid because of the large fracture permeability and small effective porosity of this unit
compared to the alluvial material. Therefore, a large proportion of the infiltrated water normally
is percolated sufficiently deep within the fractured tuff to be unaffected by the evaporation
potential that exists near the surface. Depending on the intensity of the infiltration, percolation
downward through the Tiva Canyon welded unit may occur without a significant change in rate.
7-70
-------
A ALLUVIUM
TC TIVA CANYON WELDED UNIT
P PAINTBRUSH NONWELDEO UNIT
TS TOPOPAH SPRING WELDED UNIT
CH CALICO HILLS NONWELDED UNIT
NOTE: NOT TO SCALE
CF CRATER FLAT UNIT
*• DIRECTION OF LIQUID FLOW
DIRECTION OF VAPOR MOVEMENT
PERCHED WATER
Figure 7-19. Early Conceptual Model of Ground-water Flow in the Unsaturated Zone at Yucca
Mountain (USG84a)
7-71
-------
Cat Preiure
Figure 7-20. Current Conceptual Model of Ground-water Flow in the Unsaturated Zone at
Yucca Mountain (LBL96)
A small proportion of the water percolating through the fractures slowly diffuses into the matrix
of the Tiva Canyon welded unit. Downward flow in the matrix is very slow because of the small
effective hydraulic conductivity of the matrix. During dry periods, some of the diffused water
flows back into the fractures and probably reaches the land surface by vapor diffusion. The mass
of water involved during this process is likely to be negligible compared to the percolating water.
The densely fractured Tiva Canyon unit, with small matrix porosity and permeability, overlies
the very porous, sparsely fractured Paintbrush unit. A marked contrast in material properties
exists at the contact between these two units; depending on the magnitude of the infiltration flux,
this contrast could impart a significant lateral component of flow. Flow of water through
7-72
-------
fractures of the Tiva Canyon unit occurs rapidly until it reaches the contact. At this point, the
velocity is significantly decreased because of the greater effective porosity and lesser hydraulic
conductivity of the Paintbrush unit. As a result, lateral, unsaturated flow of water above this
contact can occur. Perched water may occur above this unit if displacement along faults has
created significant differences in permeability on opposite sides of the fault.
The saturated hydraulic conductivity of the Paintbrush nonwelded unit in the direction of dip is
10 to 100 times greater than saturated hydraulic conductivity in the direction normal to the
bedding plane. The combination of dipping beds and differences in directional permeability
creates a downdip component of flow. The magnitude of this component depends on the
magnitude of the principal hydraulic conductivity ratio. The permeability contrast may be
sufficient to decrease vertical percolation into the underlying Topopah Spring welded unit to
almost zero. In this case, water would flow laterally downdip until structural features are
encountered that create perching conditions or provide pathways for vertical flow.
As water moves downward through the PTn, the effect of high porosity and low fracture density
progressively moves water from fractures into the matrix. Except for areas where fast paths may
exist (such as faults), beyond a certain depth in the PTn, flow may be almost entirely in the
matrix. Travel times through the matrix of the PTn are thought to be relatively long because the
matrix of this unit appears to act as a "sponge" which dampens out episodic infiltration pulses.
Water flows from the matrix of the Paintbrush nonwelded unit into the fractures or matrix of the
underlying Topopah Spring welded unit. Owing to the thickness of this unit, it is hypothesized
by ROB96 that water moving through the fractures eventually diffuses into the matrix and moves
very slowly downward. An exception is the second subunit of the TSw (ROB96). In contrast to
this conceptualization, the unsaturated zone expert evaluation panel estimated that up to 95
percent of the flow in the Tsw could remain hi the fractures (GEO97).
Flow enters the Calico Hills nonwelded unit either from the matrix of the Topopah Spring
welded unit or through structural flowpaths. How much flow occurs in the fractures of the lower
part of the Topopah Spring unit is unknown, and therefore their potential to contribute to flow
into the Calico Hills unit is also uncertain.
The nature of flow at the contact between the Topopah Spring welded unit and the Calico Hills
nonwelded unit depends on whether the vitric or zeolitic facies of the Calico Hills unit is present.
The permeability and effective porosity of the vitric facies are much greater than those of the
7-73
-------
matrix of the Topopah Spring unit, which may result in a capillary barrier where those units are
in contact. Conversely, the permeability of the zeolitic facies is about the same as for the matrix
of the Topopah Spring unit, resulting in continuity of matrix flux across the contact.
Flux within the Calico Hills unit may occur with some lateral component of downdip flux,
because of the existence of layers with contrasting hydraulic conductivity in the unit. A large
scale anisotropy probably is caused by intercalation of ruffs with alternately large and small
permeability and by compaction.
Water that flows downdip along the top of the Calico Hills unit slowly percolates into this unit
and slowly diffuses downward. Fracture flow is known to occur near the uppermost layers of the
Calico Hills unit, but diffusion into the matrix may remove the water from the fractures deeper in
the unit and thereby limiting flow mostly to within the matrix, except along the structural
flowpaths. It is possible, however, that fractures provide significant avenues for rapid flow
through this unit. Beneath the southern part of the block, the Crater Flat unit occurs between the
Calico Hills unit and the water table. Included are the welded part and underlying nonwelded
part of the Bullfrog Member of the Crater Flat Tuff.
Fluxes along many structural flowpaths are probably larger than within the units they intersect.
The Calico Hills unit is more ductile than the overlying Topopah Spring unit, which may give
the Calico Hills unit fracture sealing properties. In addition, because of the lesser shear strength
of this unit compared to that of the Topopah Spring, gouge formation along faults and shear
zones is more common. These properties may result in a smaller fracture conductivity in the
Calico Hills unit. In the case where the structural flowpaths are hydraulically continuous across
the upper contact of the Calico Hills unit, water would be more likely to flow downward without
a significant change in its path until it reaches the water table. In cases where the structural flow
paths are discontinuous across the upper contact, flow may be diverted downdip along this
boundary. Intermediate conditions between the two extreme cases are also possible. Recent
numerical modeling (LBL96, ROB96) of flow through the unsaturated zone has provided
important insights into the possible characteristics of flow in each subunit of the unsaturated
zone. Some of these insights are discussed in the following paragraphs.
Discussion of Unsaturated Zone Conceptual Flow Model and Modeling of the Unsaturated Zons
Under current conceptualizations the net infiltration rate through the unsaturated zone beneath
Yucca Mountain is one of the most critical parameters for determining the nature of flow in the
unsaturated zone, yet it is one of the least well characterized. Numerous modeling studies, based
7-74
-------
on varying conceptual models, have been performed to simulate unsaturated flow beneath Yucca
Mountain (e.g., DOE94a, DOE95b, LBL96, ROB96). Sensitivity analyses performed in these
studies indicate that uncertainty in the amount of net infiltration accounts for as much as 90
percent of the variability in the results.
The magnitude of infiltration flux has a significant bearing on the potential for lateral unsaturated
flow beneath Yucca Mountain. In the Paintbrush nonwelded unit, the overall hydraulic
conductivity parallel to bedding is 10 to 100 times greater than that in the direction normal to the
bedding plane. At higher flux rates, the potential vertical flow rate of some units is exceeded,
thereby inducing a significant lateral component of flow to the infiltration flux. Some authors
have examined the possibility of "focused recharge," a phenomenon in which surface rainfall
runoff is directed to areas where faults intersect the surface. Significant amounts of recharge
may infiltrate into these zones, which may induce lateral unsaturated flow in the underlying units
(LEH92). One obvious area where this may be occurring is the northern extension of Solitario
Canyon fault, which bounds Yucca Mountain on the west. As previously described, lateral flow
could direct water to structural flow paths, which may then redirect the flow vertically
downward, providing a "fast path" and potentially reduced travel times to the saturated zone.
There is growing evidence to suggest episodic water flow at Yucca Mountain may take place
along "fast paths" (LBL95, FAB96, LBL96). Data obtained from recent sampling conducted
within the ESF runnels drilled into Yucca Mountain provide compelling evidence that not only
does flow occur along "fast paths," but that such flow is capable of moving considerable
distances over a relatively short time frame. The amount of water which may be infiltrating by
fast paths is obviously of critical importance to predicting repository performance. Samples
taken in the ESF tunnel show elevated concentrations of some radionuclides, principally
chlorine-36, as well as lesser amounts of tritium and technetium-99 (FAB96). Chlorine-36 is a
radioactive isotope produced in the atmosphere and carried underground with percolating ground
water. High concentrations of this isotope were added to meteoric water during a period of
global fallout from atmospheric testing of nuclear devices, primarily in the 1950's. This "bomb-
pulse" signal can be used to test for the presence of fast transport paths (FAB96).
Testing for bomb-pulse radionuclides was conducted by collecting and analyzing rock samples
from the ESF. Systematic samples were collected every 200 m, and feature-based samples were
collected whenever a structural feature such as the intersection of the tunnel with a fault, was •
recognized. The results of the testing indicate that most of the samples had 36C1 ratios ranging
from 400e-l 5 to 1300e-l 5. The analysis in LANL96 indicates that although many samples
7-75
-------
showed 36C1 ratios above present day atmospheric levels, it is believed that they represent
Pleistocene water which entered the system when the 36C1 ratios of infiltrating water were higher
than they are today. Samples with 36CI ratios above 1500e-15 were interpreted as containing a
component of bomb pulse water, indicating that at least a small proportion of die water at those
locations is less than 50 years old. Locations at which multiple samples showed indications of
bomb-pulse 36C1 ratios appear to be associated with the Bow Ridge fault zone, the Drill Hole
Wash fault zone, and the Sundance fault zone (ROB96). The most significant result of the 36C1
testing is that some water travels to the repository horizon in less than 50 years. It is important to
recognize, however, that these results do not indicate that all water travels this quickly in the
unsaturated zone. The 36C1 data do not indicate what fraction of the water now in the unsaturated
zone has traveled by fast paths, nor do they by themselves indicate the magnitude of infiltration
fluxes. Age dating, numerical modeling, and other lines of evidence suggest that travel times for
most of the unsaturated zone are on the order of thousands to tens of thousands of years
(LBL96).
Recent numerical modeling studies (LBL96, LANL96, ROB96) suggest two important
requirements for rapid (less than 50 years) transport of 36C1 to the ESF: 1) a continuous, high
permeability pathway must exist to depth, and 2) a means of focusing infiltration and
maintaining flux to the pathway must exist for a sufficient time. The eastward dip of the highly
permeable PTn unit allowed strong lateral flow which was subsequently diverted downward at
faults in these simulations. The strong lateral, down dip flow in the PTn was subsequently
channeled into local permeability highs. In both the Paintbrush and Calico Hills units several
vertical "fast paths" developed in response to these conditions. The recent modeling suggests
that where the PTn is relatively thick, it was necessary to modify fracture properties to represent
greater fracture densities and/or fracture apertures in order for bomb-pulse 36C1 to migrate to the
ESF in less than 50 years (ROB96).
The presence of perched water has implications for travel tunes, flow paths, and fluxes of water
through the unsaturated zone. Analysis of water from several perched water zones documents a
number of important findings, including perched water compositions that are out of equilibrium
with pore water, showing little fracture/matrix interaction (DOE96d). This indicates that the
perched water probably reached its present location without extensive travel through and
interaction with the rock matrix, thus suggesting that this water had traveled relatively quickly
through the unsaturated zone. Recently-measured tritium concentrations in perched water are at
background levels, therefore suggesting that perched water is older than thermonuclear weapons
testing. Also, preliminary data from isotope testing of perched water samples from boreholes
7-76
-------
UZ-14 and SD-7 indicates an apparent residence time of about 10,800 years, with corrected ages
ranging from 5,000 to 10,800 years (LBL96). A detailed conceptual model of perched water is
presented in LBL96.
ide Transport in the Vnsalurated Zone
The travel time of radionuclides beneath Yucca Mountain is a function of both physical and
chemical processes and interactions between fluid and rock. In terms of physical processes,
radionuclides travel by gas phase and liquid phase advection, dispersion, and diffusion.
Radionuclide travel times to the accessible environment are a function of the percolation flux
distribution in the unsaturated zone and the advective flux distribution in the saturated zone, as
well as the hydrostratigraphy along the ground-water flow paths between the repository and the
accessible environment. The percolation flux distribution within the Topopah Spring
hydrostratigraphic unit (and other unsaturated zone units below it) is a function of the infiltration
rate and the complex mechanism of ground-water flow in the unsaturated zone. Chemical
influences on radionuclide travel times include retardation processes involving liquid and gas
phase diffusion, ion-exchange, adsorption on solids, surface complexation, colloidal suspension,
chemical reactions, mineral alteration and dehydration reactions, radioactive decay, and
Precipitation/dissolution reactions.
In particular, the key conceptual uncertainty in the transport of radionuclides through the
unsaturated zone at Yucca Mountain is the presence of fracture flow and transport which might,
Jf fracture pathways are continuous and interconnected, lead to the formation of "fast paths" to
the underlying saturated zone.
Uncertainties in chemical retardation mechanisms and the lack of rock/radionuclide interaction
data also lead to considerable uncertainty in predicting future repository performance. For
instance, in TSPA (DOE95b), modeling efforts have simulated fluid/rock interactions that can
serve to chemically retard the transport of radionuclides with a simple equilibrium (infinite
capacity) distribution coefficient (Kj) model. Generally, values for distribution coefficients are
related to both the chemical nature of the individual hydrostratigraphic unit and to the properties
of the radionuclide. Since distribution coefficients are used to model such a wide variety of
Phenomenological processes, they are modeled in TSPA-95 as stochastic parameters with a high
degree of uncertainty. This process results in a broad range of predicted times it would take •
radionuclides to travel from the repository to the water table. Radionuclides that are little
affected by chemical retardation (e.g., I, Tc) could reach the water table within the same time
7-77
-------
frame as the ground water. Alternatively, KjS used in TSPA-95 for a number of radionuclides
(i.e., Am, Ra, Cs, Sr) result in travel times to the water table that are 50,000 times greater than
those for the ground water. Plutonium exhibits significant sorption on all types of Yucca
Mountain tuffs, with sorption coefficients often in excess of 100 cubic centimeters per gram
(cc/g) (ROB96). Detailed analysis of laboratory data for 237Np showed that a nominal sorption
coefficient of 2.5 cc/g could be used in the clinoptilolite-rich zeolitic rocks, with a value of 0 cc/g
elsewhere. Measured K^, values for "Se are on the order of one cc/g. Sorption of uranium,
similar to 237Np, is significant only for zeolitic tuffs (ROB96).
Recent numerical modeling of the role of rapid transport through fractures was studied for 237Np
(ROB96). For peak dose criteria, the model indicates that the peak may be a result of rapid
radionuclide transport through fractures. However, this does not mean that most of the
radionuclides travel through fractures. According to this model, 10 percent of the source
radionuclides typically travel rapidly in the fracture system, while 90 percent traveled much
slower in the matrix material. (Other conceptualizations suggest that up to 95 percent of flow is
in the fractures.) These results must be interpreted with the realization that the distribution of the
simulated flux between the fractures and matrix is entirely the result of the parameters used to
characterize the system. The Calico Hills, the primary unit through which radionuclides must
travel to get to the water table, is poorly characterized; nothing is known of its fracture hydraulic
properties.
Simulations of 36C1 ratios and 14C in the unsaturated zone indicate that infiltration rates between
one and five rnrn/yr are more consistent with the field measurements than infiltration rates on the
order of 0.1 mm/yr (ROB96). The environmental isotope simulations also helped provide a
reasonable explanation for the bomb-pulse 36C1 ratios measured in the ESR This explanation
involves disturbance of the PTn (e.g., faulting) which led to increased bulk fracture
permeabilities and provided a local hydrologic environment conducive to rapid fracture flow of a
small fraction of the total infiltrating flux. The flow in the fractures associated with these
disturbances is rapid enough to transport solutes from the ground surface to the ESF in less than
50 years.
When flow and transport in fractures is simulated using a particle tracking method, a bimodal
distribution of travel times is obtained — an early arrival through fractures, followed by a much
delayed breakthrough of radionuclides that traveled through the matrix (ROB96). Although
ROB96 predicts that the percentage of the total radionuclide inventory that travels rapidly to the
7-78
-------
water table is small, the radionuclide flux entering the saturated zone is at its greatest level
during this period, and thus the peak dose is controlled by fracture transport. Migration of
radionuclides through fractures is likely to be retarded by diffusion and in some cases adsorption.
ROB96 noted that there is an inverse relationship between infiltration rate and arrival time of
first breakthrough peak.
Due to sparse data and limited or nonexistent testing of the CHn, characterization of fracture
hydrologic properties in this unit is based on speculation and application of theoretical
relationships (ROB96). Model simulations indicate that the nature of fracture flow in the Calico
Hills is critical to characterizing the performance of the site. Changes in estimated hydrologic
property values estimated for these units have considerably altered the simulated flow and
transport behavior through the unsaturated zone natural barrier.
?•1.2.2 Hydrologic Characteristics of Saturated Zone Units
In contrast to the unsaturated zone in which the flow of water is considered to be primarily
vertical, ground-water flow in the saturated zone at Yucca Mountain is principally in the
horizontal direction. This consideration, coupled with the fact that it is the saturated zone in
which most downgradient radionuclide transport from a repository would occur, requires the
description of saturated zone hydrology to cover an area much greater than Yucca Mountain
itself. Thus, while the discussion of unsaturated zone hydrology is conveniently limited to the
Tertiary volcanic rocks beneath the proposed repository, this section broadens in scope to include
not only the saturated volcanic rocks, but also the adjacent Paleozoic carbonates and the alluvial
basin fill deposits. Because of the complex three-dimensional geometric relationships of these
geologic materials, the BID breaks the description of saturated zone hydrology into two parts.
Section 7.1.2.2 is restricted to a description of each of the three individual geologic materials
(volcanic rocks, alluvium, and Paleozoic carbonates) and their hydrogeologic properties; Section
7.1.2.3 attempts to describe the geometric and hydrologic relationships of the various units to one
another and to present an integrated picture of regional ground-water flow.
Before beginning a detailed description of the hydrologic properties of the individual aquifer
units, it will be helpful for the reader to keep in mind the following information while reading
this section. As previously described, Yucca Mountain is composed of a thick sequence of
Tertiary volcanic rocks. Beneath Yucca Mountain, the thickness of these rocks is more than
1,800 m (SPE89). The Tertiary volcanic sequence is underlain by complexly folded and faulted
7-79
-------
Paleozoic sedimentary rocks, including thick sections of carbonate rocks (SPE89), The
Paleozoic rocks beneath the volcanic section are water-saturated and capable of transmitting
ground water, probably over great distances. Bounding Yucca Mountain on three sides are
downdropped basins filled with alluvial deposits eroded from the surrounding mountains. Water
recharged in the higher altitude areas north of Yucca Mountain flows generally southward
through the volcanic, carbonate, and alluvial aquifers toward discharge areas located in the
southern Amargosa Desert and in Death Valley.
Volcanic Aquifer
At Yucca Mountain, where the volcanic rocks may or may not be fractured and where the
hydrologic properties can change significantly in a single stratigraphic unit, stratigraphic units
are useful only in a very general sense for defining hydrogeologic units. The volcanic rock
section beneath Yucca Mountain has been divided informally into the four hydrogeologic units
shown in Figure 7-21: 1) the upper volcanic rock aquifer, 2) the upper volcanic confining unit, 3)
the lower volcanic aquifer, and 4) the lower volcanic rock confining unit. Note that the
boundaries of these hydrogeologic units do not correspond necessarily to stratigraphic or
thermal/mechanical units as defined by other studies. Ground water flows through all of these
units to some degree (where saturated); these hydrogeologic unit designations serve primarily to
distinguish between zones which transmit relatively large quantities of ground water ("aquifers")
and zones which transmit lesser, but not necessarily insignificant, amounts of ground water
("confining units") (DOE95e; USG94a).
The largely nonwelded and intensely altered lower volcanic section, the Lithic Ridge Tuff and
older tuffs, is a confining unit The variably-welded Crater Flat Tuff constitutes an aquifer of
moderate yield. The tuffaceous beds of Calico Hills are largely nonwelded and are zeolitized
where saturated; however, this unit is significantly less altered than the lower volcanic section.
Where saturated, it is generally a confining unit, but locally parts of the formation are permeable.
The Topopah Spring Member of the Paintbrush Tuff is predominantly densely welded and has
abundant lithophysal horizons. It contains the zones of greatest primary and secondary
permeability and constitutes the most productive aquifer in the tuff section, where it is saturated
(FRI94). Units of the lower volcanic aquifer generally are completely or mostly in the saturated
zone. Because it is deeper, increased lithostatic load probably accounts for part of the difference
between the two aquifers, but the lower aquifer also tends to be less fractured than the upper
7-80
-------
TY
Dl
ME
O
200
400
600
800
1000
1200
1400
1600
1800
3000
i
NCAL scLtcno a
fW GEOLOGIC 1
IN UNITS
TCR3
Pointbruth Croup 1
c*
t
is
ib
i
7!
"-TW1"
.MM.
ft*. Poa Ml
Mb** luir
Iiom TW(
Flow breccias
OftQ lOVCS
UWe Rkiae Tuff
Older tuffs, lavas,
and brecoos
? '
Cleano Formation
Cambrian to
Devonian
formations
Proterotoic rocks
ATURATIO VWC ' TYP
lYDftOQBOLOGir fl
UNITS
'
Ugper
oaufer
ry
Lower
volcanic
aquifer
Lower
VOcCQftC
confining
irit
?
Upper dosfc
confining unft
Lower
carbonate
ooufer
Proterazoic
confring
unit
1
un
UNO SUWACE 0-
100
200
300
\ 400
\
\ 500
\
i
\ 600
"s
V
I 700
i
\BOO
. \
\
8dp
\
\
1000 '
ICAL StLECTEO TBBMIAL/ SATURATED ZONE
TH CKJUWIC MECHANICAL. KYOROCGOLOCIC
N . UNITS UNITS UNITS
DO lOrUs tnd othen,
18891
PohtbrMh Group
Aluvbm
***"iSr**
-fir
Topopoh
Spring
Tuff
CafcoMb
Formation
1
0
1
S
0
Pro* POM
. Tull
Trim Wf
EXPAND
EO I
UO
PTn
TSwl
TSw2
Tbl
CHnl
CHn2
CHn3
PP«
CFUn
8Fw
aua
ana
CRMaJ
TRw
IPPER UMTS
Upper
volcanic
aquifer
Opp«r
volcanic
confining
unit
Cower
volcanic
aquifer
Figure 7-21. Saturated Zone Hydrostratigraphy of Volcanic Rocks (USG96a)
7-81
-------
volcanic aquifer. The lower volcanic aquifer is also more altered, which accounts for (he
decreased permeability (USG96a).
The physical properties within each formation van' considerably, largely due to variation in the
degree of welding of the tuffs. The nonwelded tuffs are characterized by having a relatively
large primary porosity, but low permeability. This low permeability results from small pore
sizes and the presence in many nonwelded units of secondary alteration minerals (primarily
zeolites and clays). The welded tuffs are typically very hard and densely welded. The welded
tuffs are commonly more highly fractured than the nonwelded units. The fractures in the welded
tuffs endow them with a significant bulk permeability. For this reason, many of the welded tuff
units are capable of transmitting greater quantities of water than their nonwelded counterparts
(USG84a).
The occurrence of the water table is not restricted to any one hydrogeo logic unit. Directly
beneath Yucca Mountain, the water table occurs primarily within the Calico Hills Formation and
toward the southern end of Yucca Mountain in the underlying Crater Flat Tuff. To the east of
Yucca Mountain, in the vicinity of Forty Mile Wash, the water table occurs in the Topopah
Spring member of the Paintbrush Tuff. The occurrence of the water table in different
hydrostratigraphic units is attributable to three factors: 1) the vertical displacement of
hydrostratigraphic units by the numerous faults that dissect the area, 2) the eastward dip (five to
10 degrees) of the volcanic units, and 3) the variable elevation of the water table. See USG93a
and USG84b for graphical depictions of the relationship of the water table to stratigraphic units
and FRI94 for a map of the geology at the water table.
Aquifer Geometry
The thickness of the volcanic units is greatest to the north of Yucca Mountain toward the
eruptive centers of the Timber Mountain Caldera Complex (USG85a; USG90a), diminishing
gradually from the eruptive centers to zero thickness at the limits of the southwest Nevada
volcanic field. The thickness of the volcanic deposits also varies considerably for two reasons.
First, these units were deposited on a topographic surface of considerable relief. Second, erosion
and postdepositional structural events have significantly modified their original distribution and
thickness (USG85a, p. 8). In the vicinity of Yucca Mountain, the only direct measurement of the
thickness of the volcanic sequence has been at Well UE-25p#l, where the thickness was
measured to be 1,244 m. Seismic reflection studies have not yielded definitive data, owing to
absorption of reflected energy by the thick volcanic cover (USG85a). Drill hole USW H-l,
located immediately north of the proposed repository boundary, was drilled to a depth of 1,829 m
7-82
-------
entirely in volcanic rocks. Thus, the thickness of the volcanic sequence at the north end of Yucca
Mountain may exceed 2,000 m.
The saturated thickness of the volcanic unit has been measured only at Well UE-25p#l. At this
location, the water table is 752.6 m above mean sea level (MSL) and the bottom of the volcanic
sequence was encountered at 129.1 m below MSL, giving a saturated thickness of the volcanic
rocks of approximately 881.7 m (USG84c). Other information can be used to provide a crude
approximation of the saturated thickness of the volcanic units. For example, the elevation of the
water table beneath Yucca Mountain ranges from 1029 m above MSL at the northern part of
Yucca Mountain to 729 m above MSL at the southern end of Yucca Mountain, a difference of
300 m (USG94a). Assuming that the bottom of the volcanic sequence beneath Yucca Mountain
is 129 m below sea level everywhere (which it assuredly is not), the saturated thickness of the
volcanic sequence would range from about 1,158 to 858 m.
The subsurface extent of the volcanic units south of Yucca Mountain is not reliably known
because the volcanic rocks dip under and are covered by alluvial deposits of the Amargosa
Desert. See Figure 7-15 for an illustration of the generalized extent of the volcanic rocks in
southern Nevada and Figure 7-22 for a schematic cross-section showing the southward thinning
of the volcanic units. Aeromagnetic maps suggest that the volcanic rocks pinch out at about the
latitude of Lathrop Wells, and therefore, alluvial deposits constitute most or all of the cover hi
the Amargosa Desert (USG85a). Further evidence for the disappearance of the volcanic rocks is
provided by two oil exploration wells drilled in the Amargosa Valley (DRI94). These two wells
were drilled through alluvium into the underlying carbonate aquifer without encountering any
volcanic rocks. USG85a, p. 12, notes that the "southward thinning of the volcanic rocks has
been placed in question by recent north-south unreversed seismic refraction measurements.
Preliminary profiles suggest that some highly magnetized volcanic rocks may indeed thin as
Proposed but that an underlying rock sequence of less magnetized volcanic rocks may continue
southward far beyond Lathrop Wells." USG9la notes the presence of rhyolitic volcanic units
7-83
-------
South
YUCCA MOUNTA»
Center ql
dspotdule CHOCOLATE UOUNTAK
North
Lowef Clastic Aquilord
(ie. Sedrnetitofy)
Upper doslic Aqu'tard
Tie. Sedmenlary)
Figure 7-22. Schematic North/south Cross-sectional Illustration of Thinning of Volcanic Units
Beneath the Amargosa Desert
(USG85a)
7-84
-------
within the Amargosa Basin, although the genetic relationship of these units, if any. to the
volcanic rocks that comprise Yucca Mountain is not clear.
Bare Mountain, located approximately nine kilometers to the west of Yucca Mountain across
Crater Flat, consists of Paleozoic rocks. Tertiary volcanic rocks are known to lie beneath the
area may be located at the eastern bounding fault of Bare Mountain. To the north and east of
Yucca Mountain, the volcanic sequence continues for several to several tens of kilometers.
Hydraulic Conductivity
Rock properties largely control the characteristics of watei1 occurrence and flow in the saturated
zone. Rock properties, in turn, are dependent on eruptive history, cooling history, post-
depositional mineralogic changes, and structural setting. Permeability of ash-flow tuffs is in part
a function of the degree of fracturing, and thus, the degree of welding. Densely-welded tuffs
fracture readily; airfall tuffs do not. Therefore, the distribution of permeability is affected by
irregular distribution of different tuff lithologies and is a function of proximity to various
eruptive centers. Permeability is also a function of proximity to faults and fracture zones
(USG82a).
The most reliable method for determining aquifer hydraulic properties are pumping tests,
especially those in which drawdowns are measured and analyzed in wells other than those being
pumped. More than 150 individual aquifer tests have been conducted at and around Yucca
Mountain since the 1980s. Most hydraulic data were from tests conducted in the lower volcanic
aquifer and in the lower volcanic confining unit. Very few data were available for the upper
confining aquifer and the upper volcanic confining unit. Almost all the tests were single-
borehole tests in specific depth intervals and included constant-discharge, fluid-injection,
Pressure-injection, borehole flow meter, and radioactive tracer tests. Multiple-borehole tests
have been conducted only at the C-well complex (USG96b, DOE96a). Most reported values of
hydraulic conductivity available in the published literature were calculated from transmissivity
values calculated from single-borehole pumping tests and should be regarded as "apparent
hydraulic conductivity." Single-borehole tests do not record drawdown data from a large enough
sample of the aquifer to be considered reliable. Drawdown data in the pumped well may be
affected by a variety of factors such as fractures, well efficiency, borehole storage, gravity
drainage, and borehole plumbing. USG96b reported that transmissivity and apparent hydraulic
conductivity values determined using multiple-borehole hydraulic tests tend to be much
7-85
-------
higher—about two orders of magnitude—than values reported for single-borehole tests
conducted at the same borehole.
Laboratory permeameter testing has been conducted on core samples taken during drilling of
boreholes at Yucca Mountain. Welded units were reported to have matrix hydraulic
conductivities with geometric means ranging from 2.0xIO"6 to 3.0X10"6 m/day and bulk-hydraulic
conductivities of 0.09 to 10.1 m/day. The nonwelded units have variable hydraulic
conductivities, with geometric means ranging from 2.6xlO'5 to 3.0xlO"2 m/day (USG84a).
USG91b reports that, for Well USW H-6, water production during pumping tests was coincident
with fractured, partially, and partially- to moderately-welded tuff units. The reverse was not
necessarily true; that is, not all fractured partially-welded tuff units produced water. USG91b
also states that for Well USW H-6 "porosity and permeability of these rocks is generally
inversely related. Porosity is greatest near the top and bottom of ash flow tuff units and is the
least near the center. Permeability, as indicated by water production, is greatest near the center
of units, where the degree of welding is greatest."
Hydraulic conductivity of the Topopah Spring Member, as determined from aquifer testing of a
120 meter interval of Well J-13, located about five miles east of the crest of Yucca Mountain, is
about one m/d. Below the Topopah Spring Tuff Member, tuff units are confining beds.
Hydraulic conductivities of units tested below the Topopah Spring Member at Well J-13 range
from 0.0026 to 0.15 m/d (USG83).
Beneath Yucca Mountain, the Topopah Spring Member is above the water table. Wells installed
in Yucca Mountain are open to the upper volcanic aquitard (Calico Hills Formation) and the
lower volcanic aquifer (Crater Flat Tuff). Pumping tests conducted in these wells derived water
primarily from the Bullfrog and Tram Members of the Crater Flat Tuff (USG91b). Hydraulic
conductivities calculated from single-borehole pumping test data are shown in Table 7-5.
Table 7-5. Hydraulic Conductivities Calculated from Pumping Test Data
'«- ' w*» -i
UE-2SW1
USWH-4
USW H-6
USWG-4
^ ^K^fa*y>>y ';
0.46
0.3-1.1
0.85
1.34
USG84d /
USGSSc
USG91b
USG86
7-86
-------
In addition to the cautions expressed above regarding the accuracy of single-borehole pumping
test analyses, it is important to recognize that the values of hydraulic conductivity presented here
are average values for the entire pumped interval in the well. Borehole flow surveys, in
conjunction with acoustic televiewer logging, indicate that the volcanic rocks are highly
inhomogeneous in the vertical direction and that the majority of water yielded from the wells
derives from a few highly fractured water-bearing zones of limited thickness. The hydraulic
conductivities shown above are likely to significantly underestimate the actual horizontal
hydraulic conductivity of the water-bearing zones and to overestimate the hydraulic conductivity
of the less transmissive zones. USG9Ib estimates hydraulic conductivities for specific intervals
within the volcanic section. The authors calculated a hydraulic conductivity of about 9.1 m/d for
a 15.2-meter section of the Bullfrog Member and 6.7 m/d for a 10.4-meter section of the Tram
Member.
As previously stated, multiple-borehole tests have been conducted only at the C-well complex
(USG96b, DOE96a). The pumping tests at this location involved pumping of selected horizons
isolated by inflatable packers. In this way, transmissivity and hydraulic conductivities can be
calculated for individual members of an aquifer or confining unit. The following description of
transmissivity and hydraulic conductivity data is taken directly from DOE96a.
The results of four pumping tests conducted from June 1995 to May 1996 indicate that the
transmissivity of the Calico Hills interval typically is 100-200 ft2/d; the transmissivity "of the
Prow Pass interval typically is 400-700 fWd; the transmissivity of the Upper Bullfrog interval
typically is 400-1,000 ft2/d; and the transmissivity of the Lower Bullfrog interval typically is
18,000-20,000 ft2/d. The pumping tests conducted in 1996 indicate that transmissivity is about
the same from UE-25 c#l to UE-25 c#3 as it is from UE-25 c#2 to UE-25 c#l (DOE96a).
Horizontal hydraulic conductivities were calculated from computed transmissrvities by dividing
the transmissivity by the thickness of the transmissive rocks in the interval. Horizontal hydraulic
conductivity typically is one to five ft/d in the Calico Hills interval and five to 10 ft/d in the Prow
Pass interval. The horizontal hydraulic conductivity of the Upper Bullfrog interval typically is
two to three ft/d from UE-25 c#I to UE-25 c#3 and eight to 10 ft/d from UE-25 c#2 to UE-25
c#3. The horizontal hydraulic conductivity of the Lower Bullfrog interval typically is 70-90 ft/d
from UE-25 c#l to UE-25 c#3 and 150-210 ft/d from UE-25 c#2 to UE-25 c#3. Composite
horizontal hydraulic conductivity from UE-25 c#2 to UE-25 c#3 consistently was found to be
twice the composite value from UE-25 c#l to UE-25 c#3. Ratios of vertical to horizontal
hydraulic conductivity were determined to range downward from 0.08 to 0.0008 in the Calico
7-87
-------
Hills, Prow Pass, and Upper Bullfrog intervals. Note that the anisotropy in calculated hydraulic
conductivities between UE-25 c#2/#3 and UE-25 c#l/#3 is opposite of that predicted on the basis
of prevalent fracture orientations. The layout of the three boreholes to form a triangular pattern,
with boreholes UE-25 c#l/#3 located along a line estimated to be the major semiaxis of the
permeability tensor and UE-25 c#2/#3 along a line estimated to be the minor semiaxis of the
permeability tensor (USG96a, p. 48). One possible explanation for this can be found in the
relative distances of the wells from each other. Well #1 is twice the distance from #3 {pumped
well) than is well #2; the apparent anisotropy may result from fracture/channeling effects
associated with sampling the aquifer at different scales.
Porosity
In terms of bulk porosity, the volcanic sequence may be considered to consist of two different
types of tuffs: welded and nonwelded (or bedded). The welding process generally reduces the
matrix porosity. Therefore, the welded tuffs typically have a lower porosity than the non-welded
tuffs (USG75, USG84a). The welded tuffs are also more highly fractured than their nonwelded
counterparts. USG84a reports that welded units have a mean fracture density of eight to 40
fractures per cubic meter and mean matrix porosities of 12 to 23 percent. The nonwelded units
have a mean fracture density of one to three fractures per cubic meter and mean matrix porosities
of 31 to 46 percent. In both rock types, however, matrix porosity probably comprises the
majority of bulk porosity because fracture porosities, even in the more highly fractured units, are
reportedly quite small (USG85d). USG85d, using a theoretical model to calculate fracture
porosity, reports a fracture porosity of tuffs penetrated by Well USW H-4 ranging from 0.01 to
0.1 percent. Matrix porosities probably decrease with depth due primarily to lithostatic loading
and formation of secondary minerals (SPE89).
Effective Porosity
Effective porosity is that portion of the total porosity that contributes to saturated flow. Many of
the volcanic rocks are characterized by relatively small pore sizes and lack of inter-
connectedness of pores; thus, the effective porosity is normally significantly less than the total
porosity. USG84a, p. 18, reports that preliminary laboratory studies of the v^tric facies of the
Calico Hills unit show that only about five percent of the pore space is large enough to contribute
significantly to flow under saturated conditions. USG85d, p. 28, considers that fracture porosity
7-i
-------
is a reasonable estimate of effective porosity. USG83, p. 13, reports that effective porosities in
samples of welded tuff, vitrophyre, and zeolitized clayey pumiceous tuff range from 2.7 to 8.7
percent.
Storage Properties
Numerous pumping tests have been conducted in water wells completed in the volcanic rocks at
Yucca Mountain and may be used to estimate storage properties. However, most calculations of
storage coefficients for the volcanic rocks are based on single well pumping tests which
generally do not produce reliable estimates of storage properties. The ground-water storage
characteristics of the fractured tuffs at Yucca Mountain are complex (USG85d). Estimates of
storage properties of the volcanic rocks vary widely, depending partly upon the lithology and the
degree of hydraulic confinement of the unit being tested. A particular hydrostratigraphic unit
roay be under unconfined conditions at one location and under confined conditions at another.
USG91b calculates a storage coefficient of about 0.2. USG93a, p. 78, calculated storage
coefficients for the more densely welded units that ranged from 1x10'5 to 6x10~5; for nonwelded
to partially-welded ash flow tuff zones storage coefficients were estimated to range from 4x10"5
to 2xlO*4. Composite storage coefficients calculated from the multiple-borehole C-well tests
ranged from 0.001 to 0.004 (DOE96a).
The degree of confinement of the volcanic aquifers and confining units varies in ways that are
consistent with the geology of the intervals and their distance below the top of the saturated zone
(USG96b, DOE 96a). Beneath Yucca Mountain, the water table is within or just above the
Calico Hills interval (upper volcanic confining unit); this interval typically responds to pumping
as an anisotropic, unconfined aquifer. The underlying Prow Pass and Upper Bullfrog intervals
(part of the lower volcanic aquifer) respond to pumping as either a leaky, unconfined or fissure-
block aquifer. The Lower Bullfrog, isolated by intervals of nonfractured rock, typically responds
to pumping as a nonleaky, confined aquifer.
Egcharge and Discharge
Precipitation is the primary source of recharge to the volcanic aquifer (USG8)6; USG83).
Snowmelt in the Timber Mountain area to the north of Yucca Mountain, as well as on Yucca
Mountain itself, provides some of the precipitation-derived recharge. The occasional intense
7-89
-------
rainstorms experienced in the area also provide a source of recharge to ground water. However,
because so much of the water that falls either evaporates immediately or is directed into steep
channels along the flanks of the mountains to the permeable talus and alluvial deposits at the
base of the mountain, the extent of this contribution is less certain.
Various methods have been employed to estimate the amount of precipitation that recharges the
saturated zone beneath Yucca Mountain (NDC70; USG84e; USG82b). The most frequently
employed approach is to divide the recharge area into a number of zones by altitude and to
assume higher precipitation at the higher altitude zones. Some fraction of this precipitation,
usually less than 10 percent, is then assumed to recharge the underlying saturated zone.
Enhancements of this method allow for variable infiltration fractions to account for factors such
as topography, rock type, and vegetation. In the volcanic system, recharge is more easily
quantified than discharge, and discharge is usually calculated by assuming that outflows are
equal to inflows. This assumption is necessary, but questionable. Some researchers have raised
the possibility that the volcanic aquifer may still be equilibrating to a long term pulse of higher
recharge during the wetter climate of the Pleistocene (about 10,000 years ago) (USG85f,
USG96a). This possibility is not inconsistent with apparent ground-water ages of 9,000 to
15,000 years calculated for the volcanic aquifer (USG93a; USG83). NDC70 estimated that the
maximum recharge for Crater Flat and Jackass Flats is three percent of the precipitation rate, or
about 4.5 mrn/y. USG84a considers this the upper bound for the recharge rate that may be
occurring in certain parts of the saturated zone beneath Yucca Mountain, estimating that recharge
ranges from approximately 0.5 to 4.5 mm/year. Recent evidence, discussed previously, indicates
that the percolation flux through the unsaturated zone probably ranges from one to 10 mm/yr,
and averages approximately five mm/yr. Most of this percolation flux would be expected to
recharge the saturated zone.
An upward hydraulic gradient from the underlying Paleozoic carbonate unit to the volcanic units
(measured in Well UE-25p#l) indicates the potential for flow in the carbonate rocks to move into
the overlying volcanic units. Additional evidence of upwelling flow from the carbonate aquifer
includes zones of elevated ground-water temperature and carbon isotopic relationships. Elevated
temperature measurements from the upper saturated zone indicate the possibility of upwelling
from the carbonate aquifer along the Solitario Canyon fault and in the area between the Bow
Ridge and Paintbrush Canyon faults (USG96a, FRI94). Stuckless et al. (STU91) used the
relationship of the I3C/I2C ratio to the 5I4C of the ground water to argue for at least three sources
7-90
-------
°f water under the mountain. They tentatively identified the three sources as: 1) lateral How
from the tuff aquifer to the north; 2) local recharge, probably introduced dominantly by flow in
flash-flood watercourses on the eastern side of Yucca Mountain (Forty Mile Wash); and 3) water
that upwells from the deep carbonate aquifer into the tuff aquifer. Savard (SAV94) has
documented recharge to the volcanic aquifer from intermittent streamflow in Forty Mile Wash.
In a saturated zone ground-water model developed by the USGS, areal recharge had to be
specified along Forty Mile Wash for the model to adequately simulate measured potentiometric
levels in the vicinity of Yucca Mountain (USG84e).
Potential pathways by which ground water leaves the volcanic units include downgradient
outflow, pumping, outflow to the carbonate aquifer, and flow into the unsaturated zone. Of the
four pathways, flow into the unsaturated zone, where it occurs, is probably among the least
significant (USG96a). There is no direct evidence that water from the volcanic units flows into
the carbonate aquifer. Vertical hydraulic gradients, where measured, indicate the potential for
flow is from the carbonate aquifer to the volcanic aquifer. The DOE states that the "current
conceptual model for the regional ground-water flow system considers that ground water in the
volcanic rocks beneath Yucca Mountain moves generally southward and discharges in the
subsurface into the valley fill alluvium as the volcanic section thins and ultimately pinches out
south of Yucca Mountain" (DOE95f). Currently, water is pumped from the volcanic aquifer
from two wells, J-12 and J-13, located in Jackass Flat near Forty Mile Wash. These wells supply
water for part of the Nevada Test Site, as well as for all site characterization activities at Yucca
Mountain, including human consumption.
Paleozoic Carbonate Aquifer
Thick sequences of carbonate rock form a complex regional aquifer system or systems that are
largely undeveloped and not yet fully understood. Secondary permeability in this sequence has
developed as a result of fracturing and enlargement of existing fractures by solution. The area
underlain by carbonate rocks is characterized by relatively low volumes of runoff. Flow can be
complex and may include substantial interaction with volcanic and basin fill aquifers (USG75).
to the extensive, thick cover of volcanic rocks and alluvium in the vicinity of Yucca
Mountain, the local characteristics of the Paleozoic sequence are not well lowwn. In eastern
Nevada, the Paleozoic sequence of sedimentary rocks is commonly divided into four general
hydrogeologic units: the lower clastic aquitard, the lower carbonate aquifer, the upper clastic
aquitard, and the upper carbonate aquifer. Evidence from drill hole data and geologic mapping in
7-91
-------
surrounding mountain ranges indicates that only the lower carbonate aquifer may be present in
the vicinity of Yucca Mountain and to the south.
Aquifer Geometry
Evidence suggests that the tower Carbonate aquifer underlies the entire area. Exposures of
Paleozoic rocks at the perimeter of the study area include Bare Mountain to the west of Yucca
Mountain, the Funeral Mountains south of the Amargosa Desert, and the Specter Range to the
east and southeast. Further evidence comes from three drill holes which have penetrated the
overlying units to reach saturated carbonate rocks — borehole UE-25p#l on the eastern flank of
Yucca Mountain, which penetrated through Tertiary volcanic rocks into the underlying carbonate
sequence, and two oil wildcat wells drilled near Amargosa Valley. Additional information
regarding these wells is provided in Table 7-6.
Examination of the altitudes of the top of the carbonate aquifer in Table 1-6 indicates that the
buried surface of the buried carbonate aquifer is quite irregular. This variability is probably a
combination of relief of the original erosional surface of the carbonate units coupled with
structural offsets produced by faulting.
Table 7-6. Borehole Location and Depth Data for Wells Drilled to the Lower Carbonate
Aquifer in the Vicinity of and Downgradient of the Yucca Mountain Area
Wc*iH>*
UE-25 p#l
Federal-
FederboffS-
1
Federal-
Federhoff
25-1
>
ljsiitK$e&. •••
LocgJtede
36°49'38"/
116C25'21"
36°35'32"/
116°22'54"
36°37'07"/
116°24'26"
: "'gnrfite ' '
AffitajJefm)
1,114.9
772.9
783.9
^&b£»rt«M»*te i
v> Aqw[er(m)
1,244
259
671
AMibideCMSl^ofTt^
tftitttim^AytftH?
' (m) > '
-129.1
513.9
112.9
"Note: Information for well UE-25 p#l obtained from USG84c. Information on oil exploration wells
obtained from DRI94.
7-92
-------
Saturated thickness of this aquifer is largely unknown; USG75 indicates that water circulates
freely to depths of at least 1,500 feet beneath the top of the aquifer and up to 4,200 feet below
land surface. The effective flow thickness of the aquifer depends, in part, upon the lithostatic
pressure at depth, which in turn depends on the thickness of the column of rock overlying the
carbonate aquifer.
Hydraulic Conductivity
Interstitial permeability of the carbonate rocks is negligible; essentially all of the flow
transmitted through these rocks is through fractures. Permeability measurements of the
carbonate rocks are reported as transmissivity values, as opposed to hydraulic conductivity
values, because the thickness of the carbonate unit through which water is flowing is not well
known. Estimates of fracture transmissivity range from 1,000 to 900,000 gallons per day per
foot (USG75). USG75 reports the results of six pumping tests in the lower carbonate aquifer.
The average calculated transmissivity was 13,000 gallons per day per foot.
USG75 reports that total porosity determinations were made for 16 samples of the lower
carbonate rocks. Total porosities ranged from 0.4 to 12.4 percent with an average of 5.4 percent.
Fracture porosity of the rock is estimated to range from 0 to 12 percent of rock volume.
fifieetive Porosity
£*ue to the extremely low matrix permeability of the carbonate rocks, effective porosity can be
aPproximated as the effective porosity of the fractures. Many of the fractures in the carbonate
units are partially filled with clay or other materials which reduce both fracture permeability and
effective porosity. USG75 reports that effective porosity values determined for 25 samples of
the lower carbonate rocks ranged from 0.0 to 9.0 percent, with an average of 2.3 percent.
S&>mge Properties
USG75 reported that, based on examination of rock cores, the effective fracture porosity of the
l°wer carbonate aquifer is probably a fraction of one percent; accordingly, the storage coefficie
storage coefficient
7-93
-------
under unconfmed conditions is not likely to exceed 0.01. Because of the extremely low effective
porosity of the carbonate rocks, the specific storage under confined conditions probably ranges
between 10"5 and 10"6 per foot. Where the aquifer is several thousand feet thick the storage
coefficient may be as large as 10"3.
Recharge and Discharge
Direct areal recharge to the carbonate aquifer occurs where these rocks are exposed at the
surface. The highest amounts of areal recharge are expected to occur in highland areas where
precipitation levels are highest and where the highly fractured rocks are exposed at the surface.
Recharge to the carbonate units may also derive from downward infiltration through overlying
volcanic or alluvial deposits. The relationship of flow potential in the carbonate aquifer to that in
the overlying units is not well known. No downward gradients have been measured between the
carbonate aquifer and overlying units in the study area. This would seem to indicate that the
recharge areas for the carbonate aquifer are located relatively far away from Yucca Mountain.
North of the proposed repository area is an area of relatively high hydraulic gradient, measured
in the saturated volcanic rocks. One proposed explanation for this high hydraulic gradient is an
inferred east-west striking graben which provides a conduit for ground water flowing in the
volcanic aquifer to drain into the underlying carbonate aquifer (FRI94). If this is the case, then
the carbonate aquifer is being recharged by flow from the overlying volcanic units at this
location.
The only measurements of potential in the carbonate aquifer made near Yucca Mountain indicate
vertically upward hydraulic gradients over wide areas of the carbonate unit. Over at least part of
the study area (in borehole UE-25 p#l) and beyond (specifically in the Amargosa Desert east of
the Gravity and Specter Range Faults), upward hydraulic gradients have been measured between
the carbonate aquifer and overlying units. These upward hydraulic gradients indicate the
potential for upward flow, but do not demonstrate that such flow is occurring in these areas.
Discharge from the carbonate aquifer would occur in those areas where such flow actually
occurs. FRI94 describes anomalously high ground-water temperatures measured beneath Yucca
Mountain in the saturated volcanic aquifer which indicates upward flow (discharge) from the
carbonate aquifer into the overlying volcanic units may be occurring in the vjcinity of the
Solitario Canyon Fault.
7-94
-------
One major discharge location for flow in the regional carbonate aquifer is at Ash Meadows,
located southeast of Yucca Mountain. It is not clear, however, whether discharge at Ash
Meadows includes any ground water that has flowed beneath Yucca Mountain (this point is
discussed in more detail in Section 7.1.2.3). Additionally, Death Valley, located about 60
kilometers south-southwest of Yucca Mountain, is regarded by many researchers as the base
level or terminus for the entire regional system and, as such, accommodates discharge from the
carbonate aquifer (USG88a). There are also numerous small, relatively low flow springs located
throughout eastern Nevada, though to a lesser extent in the study area, which represent discharge
Points from the carbonate aquifer(s) (USG75).
Alluvial Aquifer
Valleys, topographic basins, and other topographic and structural lows are filled with variable
thicknesses of unconsolidated, often poorly-sorted sand and gravel deposits. Lacustrine and
eolian deposits are found locally. Basin-fill deposits are generally 2,000 to 5,000 feet thick, but
in some basins exceed 10,000 feet in thickness. Basin-fill ground- water reservoirs are restricted
in area! extent, generally being bounded on all sides by mountain ranges. Beneath the central
Parts of the deeper valleys, the water table is encountered in the alluvium. At and near the valley
Margins, the alluvium is relatively thin and the water table occurs in the underlying consolidated
rocks.
In the Yucca Mountain area, several basin-fill aquifers or potential aquifers exist. These are:
Crater Flats, west of Yucca Mountain; Jackass Flats, east of Yucca Mountain; and Amargosa
Valley, located south of Yucca Mountain. The Amargosa Valley aquifer is substantially larger
and more significant as an aquifer than the Crater Flats and Jackass Flats basins (USG91a).
Farther to the south, across the Funeral Mountains, lies the Death Valley alluvial aquifer.
Geometry
intermontane alluvial basins tend to be elongated in a north-south direction and are of
r°ughly the same dimensions as the mountain ranges that separate them (FIE86). The alluvial
fill thickens toward the center of the basins. The Crater Flats and Jackass Flats alluvial basins
ar« bounded on their northern sides by mountainous areas at approximately the latitude of the
north end of Yucca Mountain. Crater Flat is bounded at its southern end by a small, southeast
7-95
-------
trending ridge of rock outcrops. Topographic map patterns and satellite photographs (DOE95g)
suggest that the Crater Flat Basin may be closed. The Jackass Flats basin does not have a well-
defined southern terminus; it appears to have an outlet at its southwestern end which merges into
the larger, northwest trending Amargosa Desert Basin. The Amargosa Basin is bounded on its
northwest end by the Bullfrog Hills and on its southwestern boundary by the Paleozoic carbonate
sequences of the Funeral Mountains. Both the Crater Flats and Jackass Flats alluvial basins are
bounded below by their contact with Tertiary volcanic rocks (USG88b; USG83). South of
Yucca Mountain, the volcanic sequence thins and probably pinches out (USG85a). If so, alluvial
deposits may rest directly on top of Paleozoic carbonate units in the southern part of the basin.
As previously described, two oil exploration wells drilled in the Amargosa Desert, near the town
of Amargosa Valley, went through sedimentary (mostly alluvial) deposits into the carbonate
aquifer. The thickness of the alluvial deposits at these wells was 259 m and 671 m, respectively
(See Table 7-6). The exact nature of the sediments through which these wells were drilled is not
clear, as drilling logs were not examined. DRI94 refers to the sediments both as "alluvium" and
as "Neogene." Czamecki and Wilson (HST91; p. 22) refer to deep (600 m) boreholes in the
south-central Amargosa Desert which terminated in "Tertiary basin-fill sediments" underlying
the Quaternary alluvial fill, thus opening the possibility that the Quaternary alluvial basin-fill
sediments do not directly overlie the Paleozoic carbonate sequence, but are instead separated
from it by an unknown thickness of undifferentiated Tertiary sediments.
Thicknesses of the deposits hi the three alluvial basins in the study area are not well known due
to the scarcity of drill holes that penetrate the entire alluvial sequence. Two drill holes in Crater
Flat (USW VH-1 and USW VH-2) penetrate through the alluvial cover into volcanic rocks.
Thickness of the alluvium in drill hole USW VH-2 is approximately 305 m, with a depth to water
of 164 m. In Jackass Flats, Well J-13 penetrated approximately 137 m of alluvium prior to
entering Tertiary volcanic rocks; the alluvium was not saturated at this location (USG83). Most
of the wells drilled in the Amargosa Valley are water wells for irrigation and water supply. Since
most of these wells encountered sufficient water in the alluvium, drilling was not carried through
to the underlying units; thus, direct evidence for the thickness of the Amargosa Basin alluvial
deposits is lacking. Indirect evidence (geophysical methods) indicates that the thickness of the
alluvial cover in the southern Amargosa Desert may be as much as 1,585 m (USG89).
Saturated thickness and depth to water varies considerably among basins and within a given
basin. In basins where significant discharge areas exist (typically manifested as dry lakes or
7-96
-------
playas), depth to water may be only a fraction of a meter to a few meters. Other alluvial basins
may have no saturated zone at all. In the Amargosa Basin, south of Yucca Mountain, the water
table in some irrigation wells is about 56 m deep. Considering that the basin may be over 1 500
m deep, the thickness of the saturated zone in the Amargosa Basin could be over 1500 m. A
study conducted in the Amargosa Basin area (USG89) concluded that at least 85 percent of the
Pluvial thickness in the Amargosa Basin is saturated.
Hydraulic Conductivity
USG75 reports the results of several single well pumping tests in alluvial aquifers at the Nevada
Test Site. These wells are located outside of the area studied for the Yucca Mountain Project,
but the formations tested are broadly similar, and the results are generally applicable to alluvial
deposits within the immediate area of concern. These authors found the hydraulic conductivity
of the alluvial deposits to range from 0.020 to 2.84 m/d. Due to the discontinuous nature of
individual lenses or units within alluvial fill, hydraulic conductivity is expected to show wide
variations in magnitude.
The sediments which comprise the alluvial fills are typically coarse grained and poorly sorted,
roost of them having been deposited by flash flood conditions over many thousands of years.
Although sediments such as these characteristically have relatively large total porosities,
Measured porosities tend to be highly variable due to their poorly sorted nature. USG75 reports
that the total interstitial porosity of 42 samples of valley fill range from 16 to 42 percent and
averaged 31 percent. Caliche, where present, would reduce porosity, perhaps significantly.
^SG75, p. 37, reports that caliche is a common cementing material at all depths in a shaft sunk
in alluvium in the northwestern part of Yucca Flat to a depth of 550 feet.
Effective Pnrnsitv
Poorly sorted sediments often have values of effective porosity that are substantially less than
their total porosity. Given the grain size and poorly sorted nature of the alluvium, effective
Porosity values may range from a few percent to perhaps as much as 25 to 30 percent.
7-97
-------
Storage Properties
NDC63 estimated specific yield for the alluvial deposits in the Amargosa Basin using grain size
distribution methods. The estimated average specific yield for this basin is 17.34 percent; actual
values ranged from not less than 10 percent to not greater than 20 percent (NDC63).
Recharge and Discharge
There are several potential sources of recharge for the alluvial aquifers in the vicinity of Yucca
Mountain, One source is direct recharge from precipitation falling on the alluvial areas.
Recharge is also derived to some extent from infiltration of intermittent surface waters of the
Amargosa River and washes draining off the mountains (SAV94). A third source of recharge to
alluvial aquifers is infiltration or leakage from underlying bedrock aquifers. Human activity may
also provide a source of recharge to the aquifers, chiefly by return infiltration of irrigation and
percolation of sewage or wastewater. The primary method of estimating recharge in the alluvial
aquifers is to calculate discharge from the aquifer, most of which occurs as evapo-transpiration at
playas, and to assume inflows are equal to outflows. NDC63 and USG85e provide details of
calculation methods and estimates of recharge for the Amargosa Basin; values are discussed in
Sections 7.1.2.3 and 7.1.2.4.
The nature and relative importance of potential recharge sources to the Amargosa Desert alluvial
aquifer is a matter of some debate. Perhaps the major source of recharge to the alluvial aquifer is
lateral flow into the alluvial deposits from the thinning volcanic aquifer to the north (USG86).
This is contradicted by USG85f, which uses ground-water geochemical data to argue that
"ground water in the west-central Amargosa Desert ....was recharged primarily by overland flow
of snowmelt in or near the present-day stream channels, rather than by subsurface flow from
highland recharge areas to the north," and that "much of the recharge in the area occurred during
Late Wisconsin time" (USG85f, p Fl), This conclusion fails to account for the eventual fate of
water in the volcanic units to the north and is probably too restrictive.
The upward hydraulic gradients measured in the lower carbonate aquifer support the idea that
much of the outflow from the volcanic aquifer moves into the alluvial aquifer,. Although this
outflow presumably occurs somewhere between Yucca Mountain and Amargosa Valley, the
potentiometric surface, at the scale at which it is currently mapped, provides little indication as to
7-98
-------
how or where this transition occurs. A recent study, using streamflow data and a modified
version of the HYMET model for the Amargosa River, suggests that the alluvial aquifer may also
be receiving recharge via upward flow from the carbonate aquifer (INY96).
USG91a shows water level altitude maps for 1950's (predevelopment) conditions in the
Amargosa Desert. Comparison of this map with more recent (1987) water level altitude maps
indicates that aquifer development may have had a significant impact on water levels and flow
directions. Pumping of the alluvial aquifer may have induced upward flow from the underlying
lower carbonate aquifer into the alluvial system. The extent to which areal recharge occurs via
infiltration of present-day precipitation falling directly onto the alluvial valleys is thought to be
minimal. This is because of the infrequent rainstorms and the shallow depths to which rainfall
soaks into the desert soil during such events. After a rainstorm, much of this water rapidly
evaporates back into the atmosphere (USG85f).
Several potential modes for natural discharge from alluvial basins exist, including interbasin flow
to other alluvial basins; leakage to the underlying units, either volcanic or carbonate; and
evapotranspiration (NDC63). Discharge from the alluvial aquifers also occurs in the form of
ground-water withdrawals by pumping. In the Amargosa Valley alluvial basin, ground water is
Pumped for domestic and irrigation purposes (USG91a). Quantitative estimates of recharge and
discharge from the Amargosa alluvial basin are discussed in more detail in Section 7.1.2.4.
Potentiometric and hydrochemical data indicate that the Alkali Flat (also known as the Franklin
Lake Playa), located in the southern end of the Amargosa Desert, is a major discharge area for
the alluvial aquifer system. Estimated discharge at Alkali Flat is about 10,000 acre-feet per year
0^0163). Discharge at the playa occurs primarily through evapotranspiration, the principal
c°mponent of which is bare-soil evaporation (USG90b). Some ground water may flow beneath
the mountain at the south end of the playa and continue southward (USG96a). Regional water
table maps of the alluvial aquifer (see USG91a) also suggest that a portion of the flow in the
Pluvial aquifer may be moving southwest through the abutting carbonate rocks of the Funeral
fountains, and discharging into Death Valley. The extent to which this occurs is unknown.
7-99
-------
7.1.2.3 Regional Ground-Water Flow and Hydrology
The nature of regional ground-water flow in the Yucca Mountain area is governed by the
complex three-dimensional nature of the geological and structural units through which it flows.
As previously described, the geological setting in this area involves a basement of Paleozoic 1
sedimentary rocks which have been complexly folded and faulted. The Paleozoic sequence is
overlain in many areas by a thick section of volcanic rocks and/or alluvial basin fill deposits.
The Paleozoic and volcanic sequences have been disrupted by faults which have juxtaposed
various units against one another and created the basin and range structure. The resulting
geological and stratigraphic complexity creates a correspondingly complex regional ground-
water flow system.
Key to understanding regional ground-water flow in this area is the concept that the large-scale
flow system may comprise up to three coexisting ground-water flow subsystems: local,
intermediate, and regional. These subsystems exist one on top of the other, as well as side by
side. This concept is illustrated in Figure 7-23. The coexistence of such subsystems means that
deep regional flow can pass beneath shallow local areas of high permeability and that the
presence of hydraulic barriers or variations in permeability can cause appreciable discharge
upgradient from the hydraulic terminus of the system. Major flow systems in the Great Basin are
defined by the dominant flow system, whether it be local, intermediate or regional. Where
consolidated rocks are permeable enough to afford significant identifiable hydraulic continuity
on a regional scale, the local and intermediate types of systems are considered to be subsystems
with major regional flow systems. Boundaries between systems are only generally defined; some
may represent physical barriers to flow, such as masses of intrusive rocks, while others represent
ground-water divides or divisions where an area of parallel flow ultimately diverges
downgradient.
Regional Ground-Water Flow Systems in the Yucca Mountain Area
The Great Basin is considered to consist of 39 "major flow systems" (USG93b). The study area
is located within the Death Valley Ground-Water Flow System (DVGWS) which covers an area
of 15,800 square miles (40,100 km2) in Nevada and California (Figure 7-24). The boundaries of
the DVGWS are not precisely known; traditional lateral boundaries are topographic divides that
7-100
-------
Gaining reach, net gain from ground-water inflow although in InnBrrri
_ _ . *nan may recharge wet meadows along flood pbin.
Hydraulic continuity is maintained between stream and groundwatcr reservoir. Pumping an affect ttramflow by inducing stream
recharge or by diverting groundwater inflow which would hive contributed ID ttreamflow.
Minor tributary streams, may be perennial in the mountain! but become losing ephemeral streams on the alluvial Ems. Pumping will not
affect the flow of these* streams brr'lrr' hydraulic mntimiify is not maintained between streams and the principal groundwatcr reservoir.
TKcsc streams arc the only ones present in arid rrarint
.Losing reach, net loss in flow due to surface water divcniofu and seepage to groundwatrr. Local sections may lose or gain depending on
hydraulic gradient between stream and graandwatcr lacrvoir. Gradia* may reverse during certain Dmcs of die year. Hydraulic con-
tinuity b maintained between stream and groundwater reservoir. Pumping eta affect strcamflow by inducing recharge or by diverting
irrigation return flows.
Wgated area, tome return flow from irrigation water recharges groundwater.
Flood plain, hydrologic regimen of this area dnrninslrd by die river. Water table fluctuates in response to charges in river stage and
diversions. Area commonly coveted by phreatophytcs (shown by random dot pattern).
Approximate point of maximum stream flow.
Figure 7-23. Schematic Illustration of Ground-water Flow System in the Great Basin (USG76a)
7-101
-------
118* W
38* N
115* W
37* N-
36° N
35° N
L •> • . •- /ft*--- •• • i£v~; -; ; '••-*-. ^jri?: J?
^;;|P'>
^ -i
..-;--.,c.-/'".-:-.,/. • .. :,*^'--. -•••^•^f^'.^y.^-''-•-: • •. *.^*>£*.' •••-,-«
^wWv:;c^^iaia£w.:..^^^^^Si|^^5^^ : • ?'->^
I I
YUCCA MOUNTAIN
50
25 50
100
KILOMliTIJRS
75
MILES
Figure 7-24. Death Valley Ground-Water Flow System (USG96a)
7-102
-------
may be physical barriers to ground-water flow or may coincide with ground-water mounds
formed by local recharge. Rarely, however, are these boundaries true hydraulic barriers.
The DVGWS is further subdivided into a small number of hydrogeological subareas or basins.
Yucca Mountain is located within the Alkali Flats-Furnace Creek Ranch subbasin (Figure 7-25).
Definition of the hydrologic boundaries of the basins is greatly hindered by the complexity of the
geologic structure, the limited potentiometric data, and most critically, the interbasin movement
of ground water through the thick and aerially extensive lower carbonate aquifer (USG75). The
basin covers an area of about 2,800 mi2 and was named after the two major discharge areas near
its southern end (USG82c). The principal aquifers in the northern part of the subbasin are
volcanic aquifers; valley-fill and carbonate rock aquifers dominate in the southern part. The
subbasin receives water from recharge within its boundaries and probably also receives water as
underflow from adjoining subbasins. Ground water leaves the subbasin as evapotranspiration at
discharge areas or as interbasin outflow (USG96a). Alkali Flat is an area where ground-water
discharge occurs almost entirely through evapotranspiration. The other major discharge is
thought to be from springs near Furnace Creek Ranch, near the headquarters of the Death Valley
National Monument. A 1984 study (USG84g) estimated discharge from the subbasin at about
15,600 acre-ft/yr; of this total, about 10,000 acre-ft/yr discharges at Alkali Flat and the remainder
discharges from springs and as evaporation near Furnace Creek Ranch in Death Valley. More
recent work (HST91) developed a conceptual model that excluded the Furnace Creek Ranch
discharge area from the shallow flow system that includes Yucca Mountain. HST91 reported
that a ground-water divide could exist in the Greenwater and Funeral Ranges between the
southern Amargosa Desert and Death Valley. Such a divide, if it exists, could limit discharge
from the shallow flow system in the Amargosa Desert to the Furnace Creek Ranch area, although
it would not necessarily affect the deeper flow system that may also contribute discharge to the
Furnace Creek Ranch area.
Adjoining the Alkali Fiats-Furnace Creek Ranch subbasin to the east is the Ash Meadows
subbasin. These subareas are separated by an irregular north-south line which runs east of Yucca
Mountain. In general, ground-water flow in these basins is considered to originate from recharge
in the upland areas of the basin and to move in a southerly direction toward discharge points in
Alluvial basins located in the southern parts of the basins. The southern portipn of the boundary
between the Alkali Flat-Furnace Creek Ranch subbasin and the Ash Meadows sub-basin is
located along a line of springs (Ash Meadows) which coincides with the trace of a buried fault.
7-103
-------
36'00'N
N
'\ YUCCA MOUNTAIN
NOTE: NORTHERN BOUNDARY Of ALKALI FLAT FURNACF SU88ASIN IS PROeABLY
FLOW BOUNDARY WITH UNO£RTLOW FROM PAHUTE AND RAINtR MESAS.
Figure 7-25. Alkali Flat-Furnace Creek Ranch Ground-Water Subbasin (USG96a)
7-104
-------
This fault causes water to rise to the surface by juxtaposition of permeable and impermeable
units of the Paleozoic rocks. Subsurface outflow into the Alkali Flat-Furnace Creek Ranch
subbasin is probable, especially in the vicinity of the buried fault. Geochemical and
potentiometric data suggest leakage of water from the carbonate aquifer into the alluvial aquifer
east of the fault line (USG85f)- The degree of connectedness of the two subbasins may be more
significant than localized leakage across the bounding fault. USG96a suggests that "deep
hydraulic connection through the carbonate aquifer may connect the Ash Meadows area on the
east side of the Amargosa Desert to the Furnace Creek Ranch area of Death Valley. This
possible connection is consistent with the observation that the hydrochemistry of water from
springs that discharge at Furnace Creek Ranch is similar to the hydrochemistry of water
discharging at some springs in the Ash Meadows area. This similarity in hydrochemistry allows
the possibility of westward ground-water flow through deep aquifers beneath the Amargosa
Desert, whereas flow through the shallower aquifers seems to be predominately southward"
(USG96a).
Ground-Water Flow Directions and Potentiometric Surfaces
Within the DVGWFS, recharge from precipitation probably occurs at Timber Mountain, Pahute
Mesa, Ranier Mesa, Shoshone Mountain, and the Spring Mountains. In the vicinity of Yucca
Mountain, infiltration of runoff in Forty Mile Canyon and Forty Mile Wash probably contributes
to recharge. On a regional and subregional scale, ground water is generally considered to flow
from these recharge areas to discharge areas located at the southern end of the flow system
(USG75). Much of the ground water which travels beneath Yucca Mountain probably discharges
at Alkali Flat (Franklin Lake) in the southern Amargosa Desert and/or in the springs on the
eastern side of Death Valley. Death Valley is the ultimate ground-water discharge area and is a
closed basin; no water leaves it as surface or subsurface flow (USG96a). Numerous workers
have constructed potentiometric surface maps for this area, including USG75, USG82c, USG84f,
USG91a, and USG94a. Availability and quality of potentiometric data for the subbasin are
highly variable. Wells are irregularly distributed throughout the subbasin; the greatest density of
^elis is on Yucca Mountain itself and in the Amargosa Valley. Data are almost entirely lacking
*n the mountainous recharge areas north of Yucca Mountain. In the immediate vicinity of Yucca
Mountain itself, numerous wells have been drilled to the saturated zone and the potentiometric
surface is well-characterized. The potentiometric surface in Amargosa Valley and in the vicinity
°f Alkali Flat is also relatively well defined by numerous irrigation and monitoring wells. There
are almost no potentiometric data available in the Greenwater and Funeral Ranges, which bound
7-105
-------
the Amargosa Desert on its southwestern side. Figure 7-26 shows the regional potentiometric
surface for the DVGWFS. The following sections discuss in detail the nature of the
potentiometric surfaces in each of the three main aquifer types.
Volcanic Aquifer
The lateral extent of the volcanic rocks that make up Yucca Mountain is not well defined,
primarily because the volcanic units are buried beneath alluvial deposits in the topographically
low areas. South of Yucca Mountain, the volcanic section is believed to thin and pinch out
somewhere in the vicinity of Lathrop Wells (USGSSa, DOE94b). Where the volcanic unit is not
present, alluvial deposits presumably directly overlie Paleozoic sedimentary rocks. Where the
volcanic units thin south of Yucca Mountain, ground water flowing in the volcanic aquifer
discharges horizontally into the adjoining alluvial deposits and continues to flow in a southerly
direction beneath the Amargosa Desert.
At the scale of Yucca Mountain, there are significant variations from the regional flow pattern,
resulting in local ground-water flow with a strong easterly component. The potentiometric
surface beneath Yucca Mountain has been relatively well-characterized. Potentiometric surface
maps are presented in USG95a, USG94a, and USG84f, among others. The potentiometric
surface can be divided into three regions: 1) a small-gradient area (0.0001) to the southeast of
Yucca Mountain, 2) an area of moderate-gradient (of about 0.015) on the western side of Yucca
Mountain, where the water level altitude ranges from 775 to 780 m and appears to be impeded by
the Solitario Canyon Fault and a splay of that fault, and 3) a large-gradient area (0.15 or more) to
the north-northeast of Yucca Mountain, where water level altitudes range from 738 to 1,035 m
(USG94a). Numerous theories have been proposed to explain the presence of the three domains
and especially the cause of the large gradient area, where water levels decline by more than
900 feet over a distance of slightly greater than one mile. The position of the large gradient area
does not correlate well with any observed geologic feature in the upper 1,500 feet of the
mountain (FRI91). The area where the gradient has been defined is about 1.7 miles north of the
design repository. If the gradient is caused by a barrier to ground-water flow, it could be of
particular importance to the design and performance of the repository; an increase in the
permeability of such a barrier could cause a substantial rise in water table altitude in the area of
the proposed repository. A rise in the water table would decrease the thickness of the unsaturated
zone beneath the repository and decrease ground-water travel time from the repository to the
accessible environment (SIN89).
7-106
-------
dveUf* V, •••e*MMr'%.- •*XV5
>|p?W?in
,*'•':' 5s$<^5s«agy£; ^
Bf •-• •-^,- IsBggt:; .fe %
g- " -^K^?MW"i
I -*:imi
36* N-
J5* N
\
YUCCA MOUNTAIN
BOUNDARY OF DEATH VALLEY
GROUNDWATER BASIN
100
MILES
Figure 7-26. Potentiometric Surface in the Death Valley Ground-Water Flow Syst
(USG96a)
;em
7-107
-------
Possible causes of the large gradient other than the flow barrier include, but are not limited to: a
fault or fault zone; an intrusive dike; a change in lithologic facies or a pinch-out; a change in
fracture orientation, density, aperture, or fracture fillings; perched water zones; or some
combination of the above phenomena. Fridrich et al. (FRI94) have proposed two models for the
large gradient zone, integrating geologic, geophysical and geochemical evidence to support their
analysis. These and other authors interpret a northeast trending gravity low and drill hole data to
indicate the presence of a buried northeast striking graben (a downdropped block of rock
bounded on both sides by faults) immediately south of the water table decline. The large
gradient zone is coincident with the northern bounding fault of the proposed graben. The
presence of the northern bounding graben fault, which is not exposed at the surface and is not
known to have been encountered in any drill holes in Yucca Mountain, is central to both models
proposed.
Briefly, the first conceptual model proposes that the buried fault zone provides a permeable
pathway through the volcanic section into the underlying deep carbonate aquifer. The second
model has the buried fault acting as the northern boundary for a much thicker and more
transmissive volcanic section south of the buried fault. These authors also suggest that rapid
draining of water in the large gradient zone may cause the low gradient area to the south and
southeast. In this model, the small gradient zone may result partly from a reduced ground-water
flux in the volcanic rocks due to the capture of flow by the underlying deep carbonate aquifer.
Carbonate Aquifer
The lower carbonate aquifer has a maximum thickness of about 8,000 m. Because the carbonate
aquifer in the study area is overlain by thick deposits of volcanic rocks or alluvium, flow
directions and gradients are not well-defined. Regional ground-water flow through the lower
Paleozoic aquifer is considered to be generally southward. Small-scale potentiometric surface
maps are presented in USG75. The lower carbonate aquifer is present below Yucca Mountain at
a depth of about 1,000 m and extends southward below the Amargosa Desert into Death Valley.
There are a very limited number of holes that penetrate the lower carbonate aquifer beneath the
valley fill. Much of the physical knowledge of the system is based upon studies of the outcrop
areas, most of which are in the mountain ranges. The best interpretation of available geological
data indicates that the lower carbonate aquifer is continuous from beneath Yucca Mountain to
7-108
-------
Death Valley and is a potential pathway for radionuclide transport in what appears to be the
ultimate discharge point for the aquifer in Death Valley.
The extent of hydraulic communication between the volcanic and underlying Paleozoic sequence
is not well characterized. In the only well (UE-25p#l) at Yucca Mountain which penetrated into
the Paleozoic sequence, an upward hydraulic gradient (from Paleozoic to the Tertiary) "was
measured. Analysis of earth-tide response of water levels in this well have been interpreted to
indicate that the carbonate aquifer is well-confined by an overlying low-permeability confining
layer and has a relatively high transmissivity (INY96). Additional evidence, including isotopic
composition and temperatures of ground water beneath Yucca Mountain, supports the concept
that ground water may be flowing from the Paleozoic aquifer into the volcanic aquifer (USG88c;
STU91).
Alluvial Aquifer
Significant amounts of ground water occur in the alluvial aquifer beneath the Amargosa Desert.
In the Amargosa Valley area, irrigation activity derives all of its water from wells completed in
the alluvial aquifer, some of which yield water at rates of several hundred gallons per minute.
Static water levels are less than 55 m below the surface in some locations. Figure 7-27, taken
from USG91a, shows a map of the water table in the Amargosa Desert. USG91a also provides a
fttep of depth to water in the Amargosa Desert. Ground-water flow in the alluvial aquifer is
generally perpendicular to the potentiometric contours. The potentiometric contours shown in
Figure 7-27 indicate that the predominant flow direction is to the south. The ground-water flow
direction is also roughly parallel to the surface drainage direction. At the southern end of the
Amargosa Desert, low permeability playa and lake bed deposits create locally-confined
conditions. The potentiometric surface at Alkali Flat is in some locations above the ground
surface (USG90b).
The potentiometric surface shown in Figure 7-27 is drawn from 1987 data. Comparison of this
map with water level altitude maps for 1950's (predevelopment) conditions (USG91a) in the
Amargosa Desert indicates that irrigation pumping has had a significant impact on water levels
^d local flow directions. Pumping of the alluvial aquifer may also have induced upward flow
from the underlying lower carbonate aquifer into the alluvial system.
7-109
-------
117*00-«•
37*00'N
JS'JO'N -
36'OO'N
»5a •>-.—. -7*i-» .;*•«:
t-f'sr^&F.--'•- IT,-*,'-':.
^^t^^H^i-m^^i^^&^JK&i^^
POTENTIOMCTRIC COMTOOR - Sho-» oHilirfe ol wt^cti wale' levrf-ould hove itooa
«d *e«i. Do«h«d »h«fe Mtnt
-------
- Water Travel Times and Radionuclide Transport
The transport of radionuclides in the saturated zone away from a repository depends on a wide
variety of factors including, but not limited to, ground-water and host rock geochemistry;
advective ground-water velocities; radionuclide concentrations and retardation properties; flux
rates of radionuclides from the unsaturated zone; the presence of sorbing materials such as
zeolites and clays; rock fracture density; fracture-matrix interaction; future climate changes; and
anthropogenic influences. Knowledge of the transport properties in the site-scale and regional
flow systems would allow researchers to more completely address four of the most important
questions surrounding repository performance and regional ground-water flow issues in the area
around Yucca Mountain:
1. What path would radionuclides from the repository follow?
2. How fast and how far would radionuciides travel in the saturated zone?
3. Where would radionuclides become accessible to the biosphere?
4. What will the concentrations of radionuclides be when they become accessible to
the biosphere?
The answer to all of these questions is uncertain. The ability to know or predict the answers to
these questions depends on performing sufficient scientific investigations over the study area in
order to reduce the associated uncertainties to acceptable levels. Some level of uncertainty will
always remain, as it is not possible to completely characterize any underground system.
Recent testing activities conducted at the C-well complex have been designed to provide more
formation regarding contaminant transport properties in the saturated zone (DOE96a,
DOE96b). Tracer testing at the C-wells complex has included the injection of both conservative
(non-sorbed/non-decaying) and nonconservative tracers (sorbed). All tracer tests were performed
by establishing a quasi-steady convergent flow field and hydraulic gradient by pumping from
borehole UE-25 c#3 for several days prior to the injection of tracer compounds. Test results are
collected by analyzing samples taken at regular intervals from the pumped well and preparing
breakthrough curves" which plot the concentration of the tracer in the pumped well versus time.
After the first detection of tracer compound, breakthrough curves typically show an initial rapid
nse in tracer concentration, which then peaks and tails off gradually.
7-111
-------
The first tracer test performed at the C-wells used sodium iodide, a conservative solute. Because
it is negatively charged, sodium iodide does not sorb to zeolites and clays, and has an average
matrix retardation coefficient of 0.93. The retardation coefficient should be less than one
because of a process known as anion expulsion, wherein anions are repelled by negatively
charged grain surfaces and arrive at the recovery well prior to neutrally-charged tracers. Test
conditions were negatively impacted by decreasing pump discharge and the resulting nonsteady
hydraulic gradient and flow rates. Tracer recovery data were analyzed to determine effective
porosity and longitudinal dispersivity using an analytical solution. The analytical method
employed has a high uncertainty and calculated parameters do not represent a unique solution to
the breakthrough curve data. Test data were analyzed using several different sets of assumptions
including a single-porosity solution, a weakly dual-porosity solution, and a moderately dual
porosity solution.
In a single-porosity solution, calculated fracture porosity was 0.036 and longitudinal dispersivity
was 17.00 ft. In a weakly dual-porosity solution, calculated matrix porosity was 0.032, fracture
porosity was 0.0068, and longitudinal dispersivity was 20.75 feet. In a moderately dual-porosity
solution, good matches were obtained using a matrix porosity of 0.0778, a fracture porosity of
0.0237 and a longitudinal dispersivity of 13.64 feet. It is important to recognize that parameters
used in analyzing tracer recovery data have a high degree of uncertainty and that because the
ground-water flow field at the C-wells is anisotropic, the transport field is most likely anisotropic
as well.
Subsequent to performing the conservative tracer test, two additional pilot tracer tests were
performed. Both tests were conducted hi the 100 meter thick isolated interval within the
Bullfrog member of Crater Flat Tuff. This interval has the largest hydraulic conductivity of any
interval at the C-holes. The objectives of these tests were to determine: 1) which injection well
(c#l or c#2) would result in a higher peak concentration of a conservative tracer, and thus be a
better injection well for a reactive tracer test, and 2) what minimum mass of lithium bromide
would have to be injected to conduct a successful reactive tracer test. Both pilot tests were
successful in that they clearly identified that Well c#2 is the preferred injection hole for a
reactive tracer test and that at least 80 kilograms (kg) of lithium bromide would be needed to
ensure a successful test. The analysis of these tracer tests and any subsequent tests for transport
parameters is not currently available.
7-112
-------
The current state of knowledge suggests that ground water beneath the proposed repository
moves laterally downgradient until the volcanic aquifer pinches out, at which point it discharges
laterally into the alluvial aquifer. Radionuclides dissolved in ground water would potentially
follow a similar path. Much of the ground water that enters the alluvial aquifer currently moves
southward to the primary discharge location at Alkali Flat. Other actual or potential points of
discharge for the system include water wells in the Amargosa Desert and springs in the Furnace
Creek Ranch area of Death Valley.
Ground-water travel times to any of these locations are not well known. Estimates of ground-
Water travel times can be developed by simple calculations or by more sophisticated numerical
modeling. In either case, travel times calculations are based on hydraulic gradient, hydraulic
conductivity, and effective porosity of the formation through which the water is flowing. Of
these three parameters, hydraulic gradients are probably the best known and most easily
measured. A range of ground-water travel tunes in the Tertiary volcanic aquifer has been
developed in support of DOE's Total System Performance Assessment conducted in 1993.
TSPA93 predicted a range in advective velocities between 5.5 and 12.5 m/yr. These velocities
represent average velocities in the Tertiary volcanic aquifer between the footprint of the potential
repository and a 5 km "accessible environment" located to the south and east of the potential
repository (DOE95f). Performance assessment parameters and results are more fully described
in Sections 7.3 and 7.4.
A more recent study on radionuclide transport in the saturated zone (DOE96c) concluded that an
advective travel time of five m/yr is in the middle of the range of reasonable estimates. At this
velocity, unretarded radionuclides would take approximately 1,000 years to travel five km from
the repository and 5,000 years to travel 25 km from the repository. This study also documents
the results of preliminary, highly simplified radionuclide transport modeling work performed
Using advective velocities of five m/yr. The nature of downgradient breakthrough curves and
resulting peak dose calculations were highly dependent on assumed values of dispersivity. The
study also found that the breakthrough curves, travel times, and peak dose results were strongly
dependent on the retardation properties of individual radionuclides, the presence of sorbing
materials such as zeolites, and the possibility of fracture transport bypassing sorptive horizons
within the volcanic aquifer.
No reliable estimates of advective velocity in the alluvial aquifers have been made downgradient
°f the potential repository.
7-113
-------
An important unresolved issue is the extent of interaction between the volcanic aquifer and the
underlying carbonate aquifer. The possibility that radionuclides might enter the regional lower
carbonate aquifer, with its higher permeability, raises concerns that radionuclides could be
transported as far as Death Valley. Current evidence, such as hydraulic head measurements in
UE-25 p#l, isotopic data, and saturated zone temperature anomalies suggests that the lower
carbonate aquifer has a higher hydraulic head than the overlying units. This upward gradient
indicates that it is unlikely that radionuclide contaminants will be transported into the carbonate
aquifer in the vicinity of Yucca Mountain. Velocities through the lower carbonate aquifer range
from an estimated 0.02 to 200 feet per day, depending upon geographic position within the flow
system (USG75). It should be noted that the figures given above are for an area of carbonate
rocks outside, and much larger, than the study area. No data are available regarding actual
ground-water flow velocities in the study area. Carbonate rocks with solution-widened fractures,
cavities, and caves typically exhibit an extremely large variation in ground-water velocities.
Ground-water age dating (WIN76) using carbon-14 methods in the springs of Ash Meadows
suggested ages of ground water in the majority of the springs ranging from 19,000 to 28,000
years. INY96 describe more recent studies which indicate that water may move through the
lower carbonate aquifer in times less than 1000 to 2000 years.
7.1.2.4 Ground-Water Resources and Utilization
Many of the studies performed in the Yucca Mountain characterization process have thus far
focused narrowly on the immediate area in and around the proposed repository. Few studies to
date have attempted to present a regional picture of ground-water resources for the areas
downgradient from Yucca Mountain. This section presents a summary description of water
resources in the area downgradient (generally south) of Yucca Mountain.
Water Quality
Volcanic Aquifer
The chemistry of water flowing through the volcanic aquifers exhibits complex dependency upon
rock composition, residence time in the aquifer, and position along a flow line (USG75).
Ground-water chemistry in a volcanic rock is controlled by primary glass, pumice fragments, and
the diagenetic minerals (NAN89). Water samples from wells drilled in Yucca Mountain indicate
that the water is predominantly a sodium bicarbonate water containing small concentrations of
7-114
-------
silica, calcium, magnesium, and sulfate (USG83). Sodium levels are generally elevated in these
rock types due to the presence of volcanic glass, which is not stable in the presence of water and
contains appreciable sodium. Two water wells, J-12 and J-13, currently supply water for site
characterization activities at Yucca Mountain and have been pumped extensively for decades
with no signs of deteriorating water quality (USG83; USG94b). (Additional sources of
information regarding ground-water chemistry can be found in USG86, USG84d, USG91b,
USG91c, and USG93a.)
With the exception of substances deliberately introduced into wells during drilling and testing,
such as drilling fluids (including diesel fuel at Well J-13 (USG83)) and radioactive tracers
(Iodine-131; USG93a), no anthropogenic effects on water quality are observed in the volcanic
rocks. This is attributed to the relatively low levels of human activity and the presence of a thick
unsaturated zone with long travel times for infiltration to reach the saturated volcanic rocks.
Alluvial Aquifer
The chemical quality of the ground water in the saturated alluvial deposits varies from place to
place. In general, ground water in wells closer to Yucca Mountain is of better quality than near
the ultimate discharge areas of the system, such as the southern Amargosa Desert and Death
Valley. Ground water near these latter areas contains higher concentrations of dissolved
constituents and is less suitable for most purposes (NDC63). NDC63 states that "although the
chemical quality of ground water in the Amargosa Desert may be suitable generally for
irrigation, water of median salinity is common and water of high salinity occurs locally."
Ground water in the alluvial aquifers in many cases contains excessive concentrations of
fluoride; a dental examination of school children in Beatty found that 19 out of 20 children who
lived in Beatty since birth were affected with dental fluorosis (NDC63). (See USG94b and
USG91d for additional ground-water chemical quality data for the alluvial aquifer.)
Carbonate Aquifer
In general, water occurring in the carbonate rocks is a calcium and magnesium carbonate water.
Where water in the carbonate aquifer has moved through the overlying volcanic rocks, analyses
show increased levels of sodium and potassium (USG75). See USG84c for chemical analyses of
water from Well UE-25 p#l completed in the carbonate aquifer beneath Yucca Mountain.
7-115
-------
7.1.3 Climate Considerations
For the purposes of this document, climate is defined as the ensemble of weather conditions over
time. Precipitation and temperature variability are the aspects of climate that are most significant
to the long-term performance of a high-level waste repository at Yucca Mountain. These
parameters influence, directly and indirectly, water infiltration rates in the area of the proposed
repository.
"Variability" means the timing, rates of change, magnitude, and persistence of conditions.
Inferences about variability are based on studies of past conditions in the region, as recorded by
both geological and biological paleo-environmental indicators. Computer models of the
atmospheric circulation are used to simulate both past and future climatic regimes. Modelling
results are compared to paleo-data. The better their simulations of past climatic conditions, the
more confidence scientists and policy makers will have in the ability of models to predict future
climate. Thus, paleo-data are considered essential in assessing future climates.
The impact of human interference with naturally-occurring climate variations must also be
considered. Large-scale changes in atmospheric composition have occurred and are almost
certain to continue for the next several thousand years (HOU92). General circulation models
may be used to anticipate the consequences of such changes and to help chart the future course of
climate change. Since the concentration of greenhouse gases in the 21st century will likely
exceed anything the world has experienced for millions of years, the paleo-record may not fully
define the climate of the future. Unknown feedbacks or abrupt, rare changes in the climate
system may occur in the future. Nevertheless, the paleo-record, combined with realistic
computer models of existing and future climate, provide the best set of tools currently available
to define the potential limits of climate variability in the Yucca Mountain area.
7.1.3.1 Past Climate Conditions and Variations
Global climate has evolved over glacial to interglacial time scales in response to changes in
orbital forcing (the relative position of the earth to the sun, with consequent changes in the
geographical and seasonal distribution of incoming solar radiation). In simple terms, these
changes altered the Pole-Equator temperature gradients, which led to changes in atmospheric
circulation and the overall hydrological balance of the earth. These changes caused ice sheets to
7-116
-------
accumulate on the continents at high latitudes, the sea level to fall, global temperatures to
decrease, and rainfall patterns in the tropics to shift.
Changes hi incoming solar radiation alone were insufficient to bring these environmental
changes about; they were amplified by internal feedbacks of the climate system itself, most
probably through changes in atmospheric composition and the albedo (reflectivity) of the earth's
surface. Such feedbacks led to reduced levels of carbon dioxide and methane (both greenhouse
gases); a higher overall albedo for the earth, due to more extensive snow and ice cover; and more
extensive deserts. However, at other times in the cycle of orbital changes, feedback mechanisms
brought about increases in greenhouse gases and other changes in the climate system, eventually
leading to rapid destruction of the ice sheets and abrupt deglaciations. The growth and decay of
ice sheets affected the atmospheric circulation, displacing jet streams equatorward and causing
massive increases in rainfall in previously dry areas.
Southern Nevada and the Great Basin experienced such dramatic changes, which, together with
lower temperatures, led to aquifer recharge and the filling of many closed basins with extensive
lakes. Such changes are evident in geologic features of the region. Variations in lake levels
extending back into the last glaciation are best known; they are generally well-dated and have
been studied in many areas of the western United States. Observed changes are well supported
by a variety of biological evidence, particularly that obtained from the analysis of packrat
middens, which contain discrete samples of local vegetation in the vicinity of the packrat nests
from particular time periods in the past. For example, when lake levels were high, vegetation
was generally more extensive; some areas that are arid today were forested. This can be seen
from the packrat middens, where vegetation can be related to past time periods.
Hydrological changes in the arid western United States do not coincide in detail with the record
of continental ice volume changes. However, it is clear that high lake levels were present when
the Laurentide Ice Sheet was extensive and that water levels fell in association with deglaciation.
As noted by Smith and Street-Perrott, "more than a hundred closed basins in the western United
States contained lakes during the Late Wisconsin [the last episode of the ice ages], 25,000 to
10,000 yr B.P. [before present], but only about 10 percent of the lakes are perennial and of
substantial size today...." Even in today's hyperarid Death Valley, there is evidence that an
extensive lake occupied the basin between 21,500 and 11,900 years ago (SMI83; HOO72).
7-117
-------
The longer term record of hydrological variability is much harder to document, given the
problems of dating water levels and precipitation. In addition, it is possible that some paleo-
lakes may have been caused by slight tectonic changes or other geomorphological factors.
Furthermore, rapid changes in ice sheet size, as postulated from sedimentary records in the North
Atlantic and elsewhere, may have resulted in very abrupt changes in the hydrological regime in
the western United States.
If jet stream displacement, due to ice sheet growth and decay, is the principal factor in
hydrological change in the western United States, there is good reason to suspect that a quite
variable hydrological regime has influenced the region over glacial-interglacial timescales.
Nevertheless, the more prolonged glacial episodes were dominated by cooler, wetter conditions,
associated with higher infiltration rates, more vegetation, and the presence of many freshwater
lakes in the Great Basin. Quantifying such changes is difficult, but Spaulding et al. estimate the
limit at the last glacial maximum as approximately 6°C colder, with precipitation levels double
those of today (SPA83).
7.1.3.2 Potential Future Climate Conditions
Orbital variations clearly have driven the broad-scale variations of global climate over the last
several million years, at least. These orbital variations are likely to be a dominant influence hi
the future. Since the orbital variations are periodic and predictable, their occurrence in the past
and in the future can be calculated. Variations over the past million years have occurred within a
fairly limited envelope; predicted variations for the future show that, for at least the next 250,000
years, the expected orbital changes will stay well within this envelope. How such changes will
affect climate can be assessed by using the solar radiation changes to force a global climate
model to simulate both past and potential climate variations in the future.
Most studies attempt to reconstruct past changes where the simulations can be verified by
observation, but a few attempts have been made over the past 25 years to forecast future changes,
at varying levels of sophistication. Figure 7-28 shows the results of these efforts, with the
overall parameter describing the output expressed (on the righthand side) in terms of global
temperature. Obviously, the sophistication of such calculations has increased over the years, but
most studies consistently predict that global climates over the next 60,000 years or so will
gradually shift towards a full glacial mode, similar to that experienced 20,000 years ago during
7-118
-------
«ST (*» BPi o FUTURE IM
STUOV »
Of
WINTER
MftAOIATIOM
atniprar
tee
VOLUME
STUDY 9
ICE
SHEET
IICMI. * IUU*. 1»U
cruor «
ice
VOLUME
STUSYS
<»o
c%*>
STUDY a
ACUM 1
MOEX
K£
SHEET
WARM
COLO
WAHU
eoto
coto
WAWU
COCO
WAAM
COLD
WAAM
COLO
M no
1M
WARM
COLD
Figure 7-28. Future Climates, Expressed in Terms of Overall Global Temperature Change
Future climates, expressed in terms of overall global temperature change, as predicted by seven
different models driven by changes in orbital forcing. The boxes on each diagram delimit the last
glacial and interglacial extremes. Dates are in years x 103. (GOO92)
7-121
-------
the most recent glacial period. Indeed, the trend towards such a state began a few thousand years
ago, in the mid-Holocene Period.
The trend towards a glacial extreme is not monotonic, but involves minor oscillations on a
generally downward trend in temperature. Following the temperature minimum, there is some
indication that conditions like those of today will not return again until about 120,000 years into
the future. It also appears that the "saw-tooth" nature of past climate variations-slow declines to
cold glacial conditions, followed by abrupt "terminations" of glacial conditions-will also
continue into the future.
In general, the present arid climate conditions are expected to be maintained in the future. The
Sierra Nevada Mountains, which lie to the west of Yucca Mountain, have a strong rain-shadow
effect on the Yucca Mountain Region. This effect is expected to be maintained or enhanced in
the future because the Sierra Nevada range is still increasing in elevation (DeW93).
These are very broad conclusions that do not allow for the high-frequency oscillations,
superimposed on longer term trends, which have been seen in the Greenland ice cores and in
some marine sedimentary records from the North Atlantic. High-frequency oscillations have
most recently been seen in the Santa Barbara Basin (BEH96). Such changes would be expected
to occur in any future glaciation, since they appear to be integrally linked to the dynamics of ice
growth and decay and their impact on ocean circulation (BRO94).
What these models do not consider is the potential additional effects of greenhouse gas increases
on the radiative balance of the earth and, consequently, on the general atmospheric circulation. It
is generally believed that the small insolation changes brought about by orbital changes are
insufficient by themselves to bring about glaciation, or indeed to terminate glaciations. The
critical issue is the feedbacks, which may amplify the small radiative signal, with the ice sheets
themselves playing a major role (via albedo effects, sea-level change, topographic influences on
atmospheric circulation, effects on ocean thermohaline circulation, etc.). What is not clear is
whether any near-term increase in greenhouse gases (in the next few decades to centuries) would
eventually be overwhelmed by the orbitally-induced shift toward future glaciation or if the
warmer climate would preclude such a development by rninimizing the necessary feedback
mechanisms. Broecker (BRO75) termed this near-term warm episode a "super-interglacial"
because it may involve temperatures higher than in any recent interglacial period. As such, it is
7-122
-------
difficult to predict what the overall consequences of such a unique state might be for the future
evolution of climate.
One study of such a scenario used a 2.5D general computer model to assess both anthropogenic
effects and orbital forcing (BER91). The model assumes that the Greenland Ice Sheet will be
entirely consumed in the near term, but that the general direction of long-term climate change
towards glaciation is not changed. The peak timing of the next glaciation is delayed by about
5,000 years (Figure 7-29). However, this model is still fairly crude and does not incorporate
many of the feedbacks that may be critical in the evolution of future climate. More experiments
with transient climate simulations, using the next generation of coupled ocean-atmosphere
general circulation models, will be needed to obtain a more sophisticated answer to this question.
GLOBAL
PAST
FUTURE
NORTHERN
HEMISPHERE
030
O50
100
-140-120-100-80 -60-40-20 0 2040 6080
BP TME (ka) AP
Figure 7-29. Model Simulations of Past and Future Climate Conditions
Model simulations (solid line) of past and future climate conditions, expressed in terms of
changing ice volume on the continents, and including anthropogenic greenhouse effects in the
immediate future. Dashed line gives past global ice volume changes as registered by oxygen
isotope ratios in benthic foraminifera from the oceans (BER91).
7-123
-------
At this stage, there is no compelling evidence that the world of the next million years will not be
subjected to the same range of climate variations experienced over the last million years.
However, in the near term (from the next few decades to several thousand years), an enhanced
greenhouse effect will very probably bring about warmer conditions than have been experienced
for thousands, perhaps even hundreds of thousands of years. This was the general conclusion of
experts who were asked to assess the magnitude and direction of future climate change (Figure 3-
11 in DeW93). They estimate that the likely upper limit of a temperature increase in the mean
annual temperature of the Yucca Mountain Region would be about two to three degrees Celsius.
Whether this effect will persist for hundreds or thousands of years depends greatly on
assumptions made about future energy consumption patterns and the overall availability of fossil
fuels. If society eventually limits fossil fuel consumption, this warmer episode may come to a
close, with the naturally-occurring trends then becoming dominant. Nevertheless, the possibility
that a greenhouse gas-induced "super-interglacial" may lead to unanticipated pathways in the
climate system and new climate states can not be entirely ruled out (BRO87).
The potential changes of greatest concern at Yucca Mountain are those associated with the
"glacial climate mode" rather than with an "interglacial mode." Past history indicates that wetter
conditions in the region have generally been associated with globally cooler climates, or with
transitions to such climates. Interglacial periods have been arid- Currently, no evidence suggests
that this basic pattern is likely to be different hi the future. Hence, the immediate future climate
of Yucca Mountain, dominated by anthropogenic effects, is likely to be as dry or drier than the
present Eventually, however, cooler and wetter conditions will dominate the area during
persistent glacial climate-modes.
7.1.3.3 Summary Regarding Climate
The climate in the Yucca Mountain region is currently warm and semi-arid, with a mean annual
average temperature of 16°C (61 °F) and mean annual precipitation of 170 mm/yr (6.7 in/yr).
Precipitation varies throughout the year, averaging about 18 mm/month in the fall and winter,
and about 9 mm/month in the spring and summer.
Physical evidence of past climates shows that climate conditions previously cycled between cold
glacial climates and warm interglacial climates such as the present. Fluctuations averaged about
100,000 years in length. Present climate conditions have prevailed since the last glacial period
about 10,000 years ago.
7-124
-------
Infiltration, into Yucca Mountain, of water from precipitation is a factor of primary importance
to performance of a potential repository at the site. Projections of future climate conditions,
precipitation rates, and infiltration rates are therefore key factors in total system performance
assessments such as are discussed in Section 7.3.
The historical record of climate conditions and climate changes in the Yucca Mountain region
was interpreted quantitatively by DOE for modeling of future climate conditions in the Total
System Performance Assessment for the Viability Assessment (TSPA-VA; see Section 7.3.2).
For these performance evaluations, DOE assumed that there would be three characteristic climate
conditions in the future: the present-day dry climate, a long-term-average (LTA) climate, with
precipitation at levels twice the present, and a superpluvial climate, with precipitation three times
the current rates. The climate conditions were assumed to alternate in sequence, with average
durations of 10,000, 90,000, and 10,000 years for the present-day, long-term-average, and
superpluvial conditions, respectively. For the base-case TSPA-VA evaluation of future
repository performance, the present day climate was assumed to continue for 5,000 years into the
future, and the first superpluvial climate period was assumed to occur about 300,000 years in the
future.
For the TSPA-VA performance evaluations, the average annual precipitation rates were assumed
to be 170, 340, and 510 mm/yr, for the present-day, LTA and superpluvial climates respectively.
These precipitation rates were assumed to result in average infiltration rates of 7.7,42, and 110
flun/yr. The three-fold increase in precipitation rate for the superpluvial climate, in comparison
with the present-day climate, was therefore assumed to result in a factor of 14 increase in water
infiltration into the mountain.
7.2 REPOSITORY CONCEPTS UNDER CONSIDERATION FOR YUCCA MOUNTAIN
7-2.1 Conceptual Repository Systems
Design concepts for a repository at Yucca Mountain have changed and evolved significantly
during the 20 years of site evaluation work to date. Changes have been made in response to
information from sources such as site characterization data, repository system performance
assessments, external technical reviews, and evolution of a waste isolation strategy. Changes
have occurred in fundamental concepts as well as in design details. For example, the Site
Characterization Plan issued in 1988 (DOE88) envisioned vertical emplacement of waste
7-125
-------
packages in individual boreholes in the floor of tunnels; current plans call for end-to-end
horizontal emplacement hi long, excavated drifts. The 1988 waste package design concept was a
simple steel canister approximately two feet in diameter with an expected lifetime of 1,000 years
or less; the current design concept is a container about six feet in diameter with two-layer,
corrosion-resistant walls and a lifetime objective of more than 10,000 years. Other changes have
evolved as a result of acquisition of site and laboratory data and from consideration of the results
of total-system performance assessments.
In response to requirements of the Fiscal Year 1997 Energy and Water Appropriations Act (PL
104-782), the DOE performed a Viability Assessment (VA)2' for development of a repository for
disposal of highly radioactive wastes at Yucca Mountain. The purpose of the VA was to provide
policy makers with an estimate of the viability of a repository at the Yucca Mountain site in the
time frame required for decision making.
The five-volume VA report was released by the DOE in December 1998 (DOE98). The
Department found "... that Yucca Mountain remains a promising site for a geologic repository
and that work should proceed to support a decision in 2001 on whether to recommend the site to
the President for development as a repository" (DOE98, Overview).
The design concepts used for the VA are described below. DOE considers the VA, and its
repository design features, to constitute a snapshot in time of an evolutionary process leading
potentially to a finding that the site is suitable for disposal and subsequently to a License
Application. Further development of the repository design features and performance evaluation
methodology will be needed for the Site Recommendation and for a License Application if the
site is found to be suitable for disposal.
Design concepts used by the DOE in the Viability Assessment were as follows:
• Horizontal emplacement of waste packages hi parallel excavated drifts.
• An initial thermal loading on the surroundings corresponding to 85 MTU/acre.
• Emplacement of waste packages only between the Ghost Dance fault and the
Solitario Canyon fault.
21 The terms Total System Performance Assessment-Viability Assessment and Viability Assessment and
the acronyms TSPA-VA and VA are used interchangeably throughout this report.
7-126
-------
Disposal of 63,000 MTU of commercial spent fuel and 7,000 MTU equivalent of
various types of defense wastes. A total of 10,500 waste packages would be
emplaced, consisting of 7,642 packages of commercial spent fuel and 2,858
packages of defense wastes.
• Disposal in excavated drifts 5.5 m in diameter, with a total of about 107 km of
tunnels and drifts in an emplacement area of 740 acres. Drifts would be spaced 28
m apart.
Packages of commercial spent fuel would contain 21 PWR fuel rod assemblies or
44 BWR assemblies.
Waste package design features which include, for the commercial spent fuel
packages, dimensions of 2-m diameter and 6 m length, with an outer shell of A
516 carbon steel 10 cm thick and an inner shell of corrosion-resistant Alloy 22
that is 2 cm thick.
• Temperature limits of 200 °C for the drift walls and 350 °C for the commercial
spent fuel cladding.
Waste types to be disposed would include uncanistered and canistered commercial spent fuel
assemblies; canisters of vitrified defense high-level wastes; navy spent fuel; other DOE-owned
spent fuel, such as from the Hanford N-reactor; and surplus plutonium from dismantled nuclear
weapons. Most of the commercial SNF is clad with zirconium alloys (Zircaloy-2 and Zircaloy-
4); about 1.15% is clad with stainless steel. In the VA, the DOE assumed that the Zircaloy
cladding would act as a significant barrier to radionuclide release. No credit was taken for
stainless steel cladding.
7.2.2 Design Concepts for Engineered Features of the VA Repository
7.2.2.1 Repository and Surface Facility Layouts
The VA reference design for excavation of tunnels and drifts for emplacement of wastes is
shown in Figure 7-30. The repository footprint, which covers about 740 acres, is offset from
both the Ghost Dance and Solitario Canyon faults. The footprint is about 1 km wide and 3 km
long. This layout resulted from consideration of factors such as potential for fault movement,
location of dominant fracture systems in the geologic formations, ease of access during
operations, and the heat emissions and temperature limits assumed as the basis for establishing
7-127
-------
N234000
N734000
N232COO
Solitario
Canyon,_
Fault \
LEGEND
Available repository sttlng area
lOO-mater cover limit
fouffi - shown on jurfocs eicepf
Jn iWng area when they an alto
of repository level.
Drill Hole
Wash Fault
Dune '
Wash
Fault
Bow
Ridgs
Fault
Figure 7-30. Repository Layout for the VA Reference Design (DOE98)
7-128
-------
design parameters. The location of the repository within Yucca Mountain is shown in cross
section in Figure 7-31.
The VA plan for functions and layout of the North Portal facilities is shown in Figure 7-32.
Plans for South Portal operations and facilities are still under development and were not
addressed in the VA.
Because of their initial high heat and radiation emissions, emplacement of the waste packages
will be done remotely. As previously noted, the VA design temperature limit for the drifts is 200
°C; radiation field levels at the surface of the packages would be on the order of 35-60 rem/hour.
A perspective view of the VA design concept for the emplacement transfer dock is shown in
Figure 7-33. Design considerations include recovery from off-normal conditions.
7.2.2.2 Waste Package Design
Waste package designs will be tailored to the characteristics of the waste type (commercial spent
PWR and BWR fuel; U.S. Navy spent fuel; other DOE-originated spent fuel; vitrified high-level
waste; and immobilized surplus plutonium from nuclear weapons). The dominant types of waste
packages in the repository will be those for commercial spent PWR and BWR fuel; in the VA
reference design, there would be about 7,600 commercial spent fuel packages, two-thirds of
which would contain PWR spent fuel and one-third BWR spent fuel. Most of the PWR packages
would contain 21 spent fuel assemblies; the BWR packages would contain 44 assemblies (the
BWR assemblies are about half the size of the PWR assemblies). Both types of waste packages
contain about 10 MTHM.
The reference waste package design used in the Viability Assessment for the 21-PWR container
is shown in Figure 7-33 (the BWR package is similar), and the design concept for the defense
high-level waste container is shown in Figure 7-34. A key feature of the designs is use of two
materials to form the walls of the package. The outer material, designated as a Corrosion
Allowance Material (CAM), is A 516 carbon steel. The inner material, designated as a Corrosion
Resistant Material (CRM), is a high-nickel alloy, Alloy 22, which is highly resistant to corrosion.
The CAM is intended principally to provide strength and radiation shielding for the package; the
CRM is intended to serve as the.principal barrier to contact of water with the waste form within
the package.
7-129
-------
m
(D
s
I
E172000
SECTION IN
•*•
•*-
DEX MAP
-------
Radlolofllcalry Controlled Ana
WHB Wwto HandBng Building
Waata Treatment Bidding
Cantor Preparation BuBdlng
Traraportar Ktolntonanot Building
Airlock Building
Bulltfing &
Electrical
Balance of ;
Plant Area Facilities
Warehouse
Shape
Admlnbtntion
Moekup
Utiltty
Storm
Water
Rftantfoo
Pond<
Bwurttv-J
Station #3
Scrvtc* Station
Security Station* V2
Fir* Station
Figure 7-32. North Portal Facilities Layout for the VA Reference Design (DOE98)
7-131
-------
OUTER BARRIER LID
(A 516)
INNER BARRIER
(ALLOY 22)
SIDE GUIDE (A 516)
INTERLOCKING PLATES
(CUTAWAY VIEW)
{STAINLESS STEEL BORON)
INNER BARRIER LID
(ALLOY 22)
OUTER BARRIER LID
(A 516!
INNER BARRIER LID
(ALLOY 22)
OUTER BARRIER
(A 516)
CORNER GUIDE
(A 516)
CORNER STIFFENER (A 516)
SIDE COVER (A 516)
TUBE (A 516)
Figure 7-33. 21 -PWR Waste Package Design for the VA Reference Design (DOE98)
-------
INNER BARRIER
(ALLOY 22}
DOE SNF CANISTER
OUTER BARRIER LID
(A 516)
INNER BARRIER LID
(ALLOY 22)
INNER BARRIER LID
(ALLOY 22)
OUTER BARRIER
(A 516)
5 POUR CANISTERS
(304L)
DOE SNF CANISTER LID
OUTER BARRIER LID
(A 516)
Figure 7-34. Defense HLW Package Design for the VA Reference Design (DOE98)
7-133
-------
In the VA reference design, the waste packages were emplaced horizontally on concrete inverts
in excavated drifts that were 5.5 m in diameter and lined with concrete. A cross section diagram
of this reference design is shown in Figure 7-35. The drifts were spaced 28 m apart and the
waste packages were spaced about 19 m apart in the drifts. Under this design concept, each
waste package acts as a point source of heat emissions for repository performance evaluation
purposes. An alternative design concept is to emplace the packages so that they touch each other
end-to-end, in which case the performance evaluations treat the packages as a line source of heat
emissions.
The VA also considered other engineered design concepts that were not included in the VA
reference design. These design options included use of drip shields to aid in delaying and
deflecting water from contact with the waste package, use of backfill, use of ceramic coatings on
the waste packages, and use of waste package designs with the CRM on the outside or with use
of two CRM materials. After the VA report was issued, the DOE began detailed evaluation of
alternative designs with the objective of selecting design features that would be used in the Site
Recommendation (SR) and the License Application (LA) if the Yucca Mountain site is found to
be a suitable location for disposal. The design that will potentially be used in the SR and the LA
is discussed in Section 7.2.2.5.
7.2.2.3 Thermal Management Strategy
Thermal management strategy is concerned with using the heat emitted by decay of the
radioactive isotopes in the waste to control the temperature and the temperature gradients in and
around the repository, thereby controlling or affecting access of water to the repository, contact
of water with the waste packages, and the timing and rate of corrosion or degradation of the
waste packages and other components of the engineered barrier system.
The thermal management strategy used for the VA was to impose a high heat load on the rocks
surrounding the drifts so that water contained in the pore spaces would boil and be driven away
from the drifts for as long as possible before the waste package heat emissions are too low to
sustain this phenomenon. The heat load selected for the VA reference design was 85 MTU/acre,
which was estimated to sustain temperatures at levels which would vaporize the percolation
water for about 2,000 years (DOE98, Vol. 3, Figure 3-14).
7-134
-------
5.5 m DIA. EMPLACEMENT DRIFT
WASTE PACKAGE (WP)
EMPLACED ON
TOP OF SUPPORT
GROUND SUPPORT
SEGMENTS
WP SUPPORT
PIER
CONCRETE INVERT
Figure 7-3 5. Drift Cross-Section for the VA Reference Design (DOE98)
ermal loading of the geohydrologic regime surrounding the drifts has potential to produce
a variety of effects on and within the regime, including opening or closure of fractures,
mineralization, and changes in the composition of solid and dissolved species hi the percolation
water. The occurrence of such phenomena, and the impacts on long-term performance of the
repository, are highly uncertain and will be difficult to model reliably for repository performance
evaluations. These effects could lessen or improve repository performance. The geohydrologic
regime would undergo a temperature transient in which the temperatures near the drifts would
peak at about 150 °C a few tens of years after emplacement, and would not return to pre-disposal
ambient conditions for about 100,000 years. However, the temperature will have decayed to
levels where liquid water can impinge on the waste packages in no more than 2,000 years.
7-135
-------
The Electric Power Research Institute has provided comprehensive analyses and discussions of
these complex issues and has developed models to characterize water/package contacts for
alternative engineered designs and geohydrologic regime characteristics (EPR96). Their
analyses demonstrate the wide range of conditions that can exist in the repository, and they also
demonstrate the dependence of performance on interactions between the heat transfer regime, the
hydrologic regime and repository thermal loading. They developed a five-dimensional matrix of
scenarios and packages-wetted fractions which "...provides a method for capturing the
correlations among heat transfer, water flow, waste package performance, and radionuclide
migration in a performance assessment model." DOE and EPRI performance assessment
methods and results are discussed in Section 7.3.
7.2.2.4 Data Sources
Characterization of the Yucca Mountain site has spanned more than 20 years to date. Both
surface-based and underground investigations have been and are being performed to characterize
the natural features of a repository at the site.
Surface-based studies have included mapping of geological structures; monitoring of seismic
activity; use of gravitational, magnetic, and other non-invasive methods to infer geologic
characteristics at depth; monitoring of current weather and climate conditions; collection of data
to characterize past climates; heating of a large block of rock to determine the effects of heat on
hydrologic and geochemical properties; and drilling of numerous boreholes to obtain data on
geologic and hydrologic conditions at depth. Several hundred deep and shallow boreholes have
been drilled at the proposed repository site and within the region.
Underground data have been obtained from tunnels excavated specifically to obtain in-situ data
at the proposed repository horizon. The Exploratory Studies Facility (ESF), which is a north-
south tunnel 8 m in diameter and 7.9 km in length and parallels what would be the eastern
boundary of the repository and terminates at the North and South portals (see Figure 7-30). The
Cross-Drift is an east-west tunnel which was excavated at a depth approximately 17m above the
proposed depth of the waste emplacement drifts and at about the mid-point of the north-south
axis of the proposed repository. The surfaces of both of these tunnels have been mapped to
obtain data on the geologic units, faults, and fractures at the repository horizon.
7-136
-------
Alcoves and niches have been constructed at various locations along these tunnels to serve as
facilities for a variety of experiments. Phenomena and physical properties being characterized
include water flow characteristics in the unsaturated zone; drift-scale seepage; effects of high
precipitation rates on flow; effects of heating on rock characteristics; fracture mineralization;
characteristics of small-scale fractures; and the presence and characteristics of fluid inclusions.
In addition to these site characterization activities at the repository horizon, other data acquisition
activities are in process. These include:
Experiments are being performed in the tunnel facilities and at the Sundance fault
zone and the Drillhole Wash fault zone to extend the data base of "bomb-pulse"
Cl-36. This isotope can serve as a tracer to characterize the existence and
characteristics of potential "fast paths" for water and radionuclide transport
through the unsaturated zone.
• Pilot scale tests of backfill and drip shield performance are being conducted.
The Nye County drilling program is providing data on the geologic and
hydrologic characteristics of the alluvial deposits in the vicinity of Lathrop Wells.
These data will be used to refine or revise the saturated zone flow and transport
models.
• A multi-phase, multi-purpose test program concerning radionuclide transport hi
the unsaturated zone is being conducted at Busted Butte. Phases I and II are
currently underway; Phase III of the program would be conducted as part of the
performance confirmation program, i.e., after licensing if the site is approved for
disposal.
The site data acquisition programs are augmented with laboratory programs to obtain other types
of data. An extensive program to obtain corrosion data for candidate waste package materials is
underway, involving a variety of corrosion environments and conditions expected potentially to
exist in the repository. Laboratory investigations also use rock samples to characterize chemical,
mechanical, and hydrologic properties of the geologic structures. Laboratory measurements also
characterize radionuclide solubilities and sorption properties using water with chemical
compositions expected to be characteristic of the repository.
These data acquisition activities have two broad purposes: to assure an adequate data base for
licensing reviews if the site is approved for disposal, and to reduce reliance on the results of
formal expert elicitations as a basis for performance models and performance parameter values.
7-137
-------
To establish values for parameters used in the Viability Assessment, the DOE made extensive
use of recommendations produced from formal expert elicitations conducted in accordance with
guidelines established by the NRC. Process models subjected to expert elicitation included
unsaturated zone flow, near-field environment, waste package degradation, waste form alteration
and radionuclide mobilization, saturated zone flow and transport, probabilistic volcanic hazard
assessment, and probabilistic seismic hazard assessment (DOE98, Vol. 3, Table 2-1). Reviewers
of the VA, including the NRC, noted that the data base would have to be improved for a License
Application, so that there would be less reliance on expert opinion. Present activities are
intended to produce a data base that will be a sufficient foundation for performance models and
parameter values to be included in the License Application,
7.2.2.5 Alternative Repository Design Concepts Under Consideration
The DOE considered the repository design concept used in the Viability Assessment to be a
snapshot in time of the design evolution process. Within the VA documentation, the DOE
identified, and provided preliminary characterizations of, alternative design features not included
in the VA reference design. These included drip shields, backfill, alternative waste package wall
materials, ceramic coatings on the waste packages, alternative thermal loadings, and alternative
waste package emplacement configurations. The intent of these additional changes is to improve
the performance of the engineered barrier system or reduce uncertainities in assessing its
performance. Since issuance of the VA report hi December 1998, the DOE has identified and
characterized six alternative engineered repository designs incorporating these options (DOE99).
As outlined below, one of these Enhanced Design Alternatives (EDA) has been selected to be the
reference design concept for the Site Recommendation. If considered necessary, further
evolution of the design may occur for the License Application if the site is approved for disposal.
The ED As considered had common and variable features. Common features include use of drip
shields; use of carbon steel ground support, use of a steel invert with granular ballast, instead of
the concrete used in the VA reference design; use of a drift diameter of 5.5 m; use of pre-closure
forced ventilation; and emplacement of 70,000 MTHM of radioactive wastes.
Design features that varied for the ED As considered were the thermal loading and temperature
objectives; use of backfill; selection of waste package wall materials; use of thermal blending to
even out waste package heat emissions; drift spacing; waste package spacing; and repository
location within the characterized area. Constraints imposed on the options were to maintain the
7-138
-------
temperature of cladding on commercial spent nuclear fuel at less than 350 °C; allow personnel
access for off-normal events; and allow repository closure 50 or more years after start of waste
emplacement. The thermal goals for the EDA options, which influence many design features,
were:
• EDA I: Maintain drift wall temperature below boiling
EDA II: Keep centers of pillars between drifts below boiling
• EDA III: Cool waste package surface to 80 °C before relative humidity reaches
90%
EDA IV and V: Keep drifts dry for thousands of years
The design parameters for the EDAs considered are shown in Table 7-7. Note that EDA III
includes two options for the waste package wall materials.
Analyses of these options produced the results shown hi Table 7-8. Comparison of these results
produced a recommendation by the M&O contractor to the DOE, which was accepted, that EDA
H be used as the initial, reference design for the Site Recommendation. Principal features of the
EDA II design are compared with those of the VA reference design in Table 7-9.
In comparison with the VA reference design, the EDA II design is expected to reduce
uncertainties that could be of concern during licensing reviews. Uncertainties that are expected
to be less significant as licensing issues are those concerning coupled thermal, hydrologic,
mechanical, and chemical processes; alteration of the natural system as a result of the heat load
on the geologic units surrounding the drifts; processes and phenomena that affect radionuclide
transport; and potential for localized corrosion of waste package wall materials. The EDA II
design is also expected to provide improved defense-in-depth and overall performance. One of
the principal features of the design is that the time-temperature history of the waste packages is
expected to avoid conditions in which the Alloy 22 outer wall would be vulnerable to crevice
corrosion.
Repository performance assessment models and parameter values (see Section 7.3) will be
revised from those used in the VA in accord with the EDA II design parameters and the
information emerging from the data acquisition program described in Section 7.2.2.4. One of the
Principal performance assessment issues for the Site Recommendation, using the EDA II design,
7-139
-------
DESIGN ELEMENT
Thermal Goals
• Cladding
• Waste package
surface
• Drift wall
• Drifl environment
• Pillar temperatures
• Other goals
Area) Mass Loading
(MTHM/acre)
Area (acres) (or 70.000 MTHM
Line/Point Load
Waste Package Size (PWR)
Drift Diameler (m)
Drift Spacing (m)
Preclosure Ventilation
Waste package heat output at
emplacement
Maximum
Average (PWR waste
package)
[CRWMSM&O 1999fat>L
Waste Package Material
Fillers
Backfill
Drip Shield
Total Waste Packages
EDA I
350°C
96°C
45
1,400
Point
12
5.5
43
50 years 0 2 to10 m'/s
20% blending used to
reduce maximum
6.7 kW
5-6 kW
2-cm Alloy-22 over 5-cm
stainless steel
No
No
Yes
15.903
EDA It
350 °C
200°C
Keep centers of pillars
below boiling {96°C)
60
1.050
Line
.21
5.5
81
50 years 3 2 to 10 m3/s
20% blending used lo
reduce maximum
M .8 kW
9.6 kW
2-cm Alloy-22 over 5-cm
stainless steel
No
Yes
Yes
10.039
EDA 111
3SO"C
Cools to 60"C before
relative humidity
reaches 90%
200'C
65
740
Line
21
5.5
56
50 years 9 2 tolO m*/s
Limited blending
18.0 kW
9. 5 kW tor PWR
a) 2-cm Alloy-22 over
5-cm stainless
steel
b) 2-cm Alloy-22 over
1.5-cm Ti-7 over
4 -cm stainless
steel
No
No
Yes
10.213
EDA IV
350°C
200"C
Keep drifts dry for
thousands of years
Limit gamma dose a I
waste package surface
to 200 mrem/hr
85
740
Line
21
5.5
56
50 years 9 2 tolO n\3la
Limited blending
18.0kW
9.5 kW
30-cm carbon sleel
Integral filler
Yes
Yes
10,213
EDAV !
1
350*C
225«C
Keep drifts dry for
several thousand years
ISO
420
Line
21
5.5
32
50 years 9 2 to 10m9 /s
20% blending used to
reduce maximum
11.8KW
9.8 kW
2-cm Alloy-22 over
5-cm stainless steal
No
No
Yes
10.039
Table 7-7. Design Parameters for the Enhanced Design Alternatives (DOE99)
T-Y40
-------
Performance Categories
Performance
Factors
Licensing
Probability/Safety
Factors
Construction,
Operations, and
Maintenance
Factors
Flexibility Factors
Cost
Margin
Time to 25 mrem
Peak Annual Dose
Rock Temperatures
Waste Package
Corrosion
Number of Waste
Packages
Length of
Emplacement Drifts
Key Construction,
Operations, and
Maintenance Issues
Emplacement area
lor 70,000 MTHM
Ability to Change to
Lower Temperature
Ability to Change to
Higher Temperature
Repository Life
Cycle Cost
Net Present Value
EDAI
2,500
290.000 years
85 mrem
Always below 96*C
Does not enter
aggressive corrosion
range
15,903
132km
Operational impacts of
more packages and
longer drifts; blending
1,400 acres
N/A
Requires development
of larger packages and
coupled models for PA
$25.1 billion
$13.4 billion
EDA II
3,550
310.000 years
85 mrem
>96eC several m's
into drift for
hundreds of yrs.
Does not enter
aggressive
corrosion range
10,039
54km
Blending;
emplacement of
backfill
1 ,050 acres
Requires longer
ventilation
Requires devel. of
coupled models for
PA
$20.6 billion
$11.0 billion
EDAs Illa/lltb
1,500
290.000/31 0.000 years
215/100 mrem
>96*C across most of
repository
Some WPs In
aggressive coir, range
for 1000s of years
10.213
55km
Fabrication of dual
corrosion- resistant
material package in Illb
740 acres
Requires changes In
drift spacing
N/A
$20.1 billion/
$21. 3 billion
$10.7 billion/
$11. 4 billion
EDA IV
160,000
100.000 years
1.200 mrem
>96*C across most
of repository
Humid air corrosion
of WPs begins as
early as 100 years
10,213
60km
Fabrication,
welding, and
handling thick WPs;
empl. of backfill
740 acres
High temp, integral
to WP performance
N/A
$21.7 billion
$11. 3 billion
EDAV
1.250
300,000 years
200 mrem
>96"C across
essentially all of
repository
Some WPs in
aggressive corrosion
range >10,000 years
10,039
54km
Blending
420 acres
Requires changes in
drift spacing
N/A
$20.0 billion
$10.8 billion
Table 7-8. Principal Results of Enhanced Design Alternative Analyses (DOE99)
7-141
-------
Design Characteristics
Areal Mass Loading
Drift Spacing
Drift Diameter
Total Length of Emplacement Drifts
Ground Support
Invert
Number of Waste Packages
Waste Package Material
Maximum Waste Package Capacity
Peak Waste Package Power (blending)
Drip Shield
Backfill
Preclosure Period
Preclosure Ventilation Rate
EDA II
60 MTU/acre
81m
5.5m
54km
Steel
Steel with sand or gravel
ballast
10,039
2-cm Alloy 22 over
5 cm stainless steel 316L
2 1PWR assemblies
20% above average PWR
waste package power
2 cm Ti-7
Yes
50 years *
2 to 10 cubic m /s
Viability Assessment
Design
85 MTU/acre
28m
5.5 m
107km
Concrete lining
Concrete
10,500
10 cm carbon steel over 2 cm
AIloy-22
21 PWR assemblies
95% above average PWR
waste package power
none
none
50 years
0. 1 cubic m/s
Table 7-9. Comparison of EDA II and Viability Assessment Design Features (DOE99)
-------
will be the potential for early ("juvenile") waste package failures that allow seepage water to
enter the package, mobilize radionuclides, and transport them to the environment. For EDA II
design conditions, waste package failures would not be expected for 100,000 years or more,
except as a result of manufacturing or handling defects. One of the principal potential
manufacturing defects is imperfect welds, characteristics of which have been under investigation
by the NRC for over 20 years in connection with potential for failures in nuclear power reactors.
The DOE has developed the RR-PRODIGAL code, based on research results, to model flaw
occurrences in welds, and is developing probability and consequence estimates for flaw types not
addressed by this code. These methods for estimating potential for juvenile waste package
failures in a repository design based on the EDA II design features will be an important element
of performance assessments for the Site Recommendation and the License Application.
7.3 REPOSITORY SYSTEM PERFORMANCE ASSESSMENTS
The post-closure safety performance of a geologic repository for radioactive wastes is evaluated
using a Total System Performance Assessment (TSPA). A TSPA involves use of models of the
physical characteristics of the repository system, in a suite of linked computer codes, to forecast
the longterm performance of the system in terms of factors, such as waste package degradation,
which lead to release of radionuclides from the repository and their transport in the environment.
The TSPA tskes into consideration the features, processes, and events that can affect radionuclide
release and transport.
Features that affect performance include factors such as the corrosion rate of the waste package.
Processes that affect performance include factors such as the rate at which water seeps into the
drifts, and events important to performance include factors such as earthquakes, volcanic
eruptions, and intrusion of the repository by human action. A TSPA takes all of these factors
into account, consistent with the engineered and natural features of the repository system.
Evaluations of total system performance for potential repositories at Yucca Mountain have been
Performed by DOE, EPRI, and the NRC. As discussed below, the DOE has performed a series
°f TSPA evaluations, for purposes of helping to guide design evolution and site characterization
^vork. EPRI has also performed a series of independent evaluations, using models and methods
significantly different from those of the DOE. The NRC has performed evaluations to
demonstrate their capability to perform licensing reviews of TSPA results that would be provided
by the DOE in a License Application.
7-143
-------
DOE's historic TSPA efforts are discussed in Section 7.3.1, and the TSPA-VA is described in
Section 7.3.2. NRC's performance assessments are discussed in Section 7.4, and EPRI's efforts
are described in Section 7.5. Results of recent assessments by DOE, NRC, and EPRI are
compared in Section 7.6.
7.3.1 DQE's Historic Performance Assessments
DOE's TSPA process began with the PACE-90 project (DOE91). PACE-90 was not a total-
system evaluation; it focused on numerical modeling of the hydrologic regime and simulated
ground-water flow and aqueous transport of radionuclides. Because data were sparse at the time,
models were simplistic and many performance factors were not considered. The PACE-90
analyses served to demonstrate the TSPA concept, and it laid the foundation for future TSPA
evaluations.
The DOE subsequently has conducted TSPA evaluations in 1991 (DOE92), 1993 (DOE94a,
DOE94b), 1995 (DOE95b), and, most recently, for the Viability Assessment (TSPA-VA,
DOE98). Each assessment built on the insights and results of prior assessments, and on the
evolving data base and design concepts. Each successive TSPA evaluation added details and
features to the models and parameter values in accord with progress enabled by the evolving
information base.
During the period of evolution of TSPA analyses to date, the regulatory basis for standards,
against which repository performance is to be evaluated, was revised. As discussed in Section
12 of this BID, the Energy Policy Act of 1992 directed the EPA to develop site-specific
radiation protection standards for Yucca Mountain, consistent with the findings and
recommendations of the National Academy of Sciences. Accordingly, the Agency has developed
the proposed 40 CFR Part 197 regulations supported by this BID. These standards propose dose
limits as a basis for radiation protection. The prior standards, contained in 40 CFR Part 191, also
included individual protection requirements (Section 191,15; see Section 1.4.4 of this BID) but
established cumulative release of radionuclides across an accessible environment boundary as the
basis for regulatory compliance.
Because of the difference in the type of radiation protection standards, the results of the TSPA-
VA analyses are expressed differently from those of prior analyses. Consistent with a dose-limit
standard, the TSPA-VA results are expressed as potential doses to receptors, for time periods up
7-144
-------
to one million years. In contrast, results for the TSPA 1991,1993, and 1995 analyses were
expressed in terms of a Complementary Cumulative Distribution Function (CCDF), which is an
appropriate representation of results for comparison with the cumulative release standards
established in the 40 CFR Part 191 regulations.
Key features of DOE's TSPA evaluations in 1991,1993, and 1995 are summarized below.
TSPA-91
The TSPA-91 analyses were designed to develop the framework for probabilistic total-system
performance characterizations. They built upon the PACE-90 analyses by modeling nominal
conditions and disturbances from basaltic volcanism, human intrusion, and climate change. They
included the first set of stochastic analyses, in which hydrologic parameters were represented by
probability distribution functions based on site and analog data. Gaseous flow of C-14 was
modeled, the saturated zone was modeled for the first time, and results were, for the first time,
obtained at the accessible environment boundary as defined by EPA's 40 CFR Part 191
regulations. Future changes in climate were represented by a range of percolation flux values at
the repository horizon.
XSPA-93
The TSPA-93 analyses were aimed at providing guidance for site characterization work and
engineered designs. In comparison with TSPA-91, the models of physical features and processes
were more sophisticated and the data base for selection of models and parameter values was
larger. Important features of the analyses included:
• A three-dimensional stratigraphy for the unsaturated zone which was based on site
data.
A saturated zone model in which each geohydrologic unit was discretely modeled.
Assessment of the effect of thermal loading (at levels of 57 and 114 kW/acre) on
performance.
Waste package failure models which included aqueous and dry oxidation
corrosion, and waste form degradation models which included dissolution and
oxidation.
7-145
-------
• Consideration of two types of waste packages: the thin-walled, small-capacity
containers emplaced in boreholes, as envisioned in the Site Characterization Plan
(DOE8S), and, for the first time, the large-capacity packages emplaced
horizontally in drifts.
In anticipation of changes in regulations as a result of requirements of the Energy Policy Act of
1992, the TSPA-93 analyses included assessments of potential doses to humans as well as results
based on cumulative radionuclide releases from the repository, consistent with the 40 CFR Part
191 disposal standards. These results were illustrative, and were not intended hi any way to
represent the actual potential performance of a repository at the Yucca Mountain site. At that
time the observation was made that more-representative models and data were needed to improve
the realism of the analyses.
TSPA-95
As a result of studies of design options and guidance for site characterization work provided by
the results of the TSPA-93 analyses, the data basis for the TSAP-95 evaluations was significantly
unproved over that which had previously been available. TSPA-95 sought to be as realistic as
possible on the basis of available information and the evolved repository and waste package
designs.
The focus of the TSPA-95 analyses was those components of the system that had been
determined by prior analyses to be most important to the waste isolation capability of the
repository. Emphasis was therefore placed on the engineered components and the near-field
environment in which they would reside. In comparison with TSPA-93, the TSPA-95
evaluations used improved and more realistic models of the drift-scale thermal-hydrologic
environment and also of waste package degradation. Models describing the transport of water in
the near-field engineered barrier system were included, and flow in the unsaturated zone was
modeled. Disruptive events and gaseous release were not considered because they had been
shown in TSPA-93 not to be significant to overall performance.
Some of the models and parameter values used in TSPA-95 were based on judgments derived
from expert elicitations, because experimental data were limited or non-existent. Data
acquisition programs, such as corrosion testing and site characterization, are continuing and are
expected in the future to enable replacement of expert elicitation judgments with experimental
data.
7-146
-------
The TSPA-95 analyses evaluated waste package lifetime, the peak BBS release rate, the
cumulative release at the boundary of the accessible environment, assumed to be 5 km from the
repository, and the peak dose rate, at 10,000 and one million years, to the maximally exposed
individual located at the boundary of the accessible environment. Evaluations were done using
alternative models and a range of alternative values for performance parameters, such as the
repository thermal loading, infiltration rate, and climate change. The DOE noted that, at the time
TSPA-95 was conducted, there were no documented models with substantiation adequate for use
with confidence in performance assessments. Never-the-less, TSPA-95 laid the foundation for
future TSPA evaluations using unproved models and an expanded data base.
According to the DOE, the principal findings derived from the TSPA-95 analyses can be
summarized as follows:
Percolation flux at the repository horizon (and attendant seepage into the drifts) is
a dominant factor hi repository system performance. This flux affects the
potential for water to drip into the drifts, the magnitude of radionuclide release
from a penetrated waste package, and the movement of radionuclides through the
unsaturated zone.
Radionuclides that dominate dose potential for the 10,000-year time frame are Tc-
99 and 1-129. Long-term doses are dominated by Np-237.
Assumptions about dispersion and dilution in the UZ and SZ will have a strong
effect on peak dose rates.
• Excluding juvenile waste package failures from manufacturing defects, if waste
packages using the TSPA-95 design are not penetrated as a result of highly
aggressive corrosion conditions such as crevice corrosion, the EBS can by itself
provide complete containment of radionuclides for 10,000 years. Similarly, if the
percolation flux is low the natural-barriers system will provide complete isolation
for 10,000 years.
7.3.2 DOE's TSPA for the Viability Augment fTSPA-VA^
The TSPA-VA was part of the comprehensive assessment of the viability of the Yucca Mountain
Project that was mandated by Congress in the Energy and Water Appropriations Act of 1997. In
comparison with prior TSPA efforts, the TSPA-VA was much more comprehensive and detailed.
Some previously used models were revised; models of repository features that affect
7-147
-------
performance and had not been included in previous TSPA efforts were added to the computer
code configuration; waste package design features were revised; and data that had been
developed since TSPA-95 was prepared were used to provide details such as the spatial
distribution of infiltration rates.
The discussion in this section of the BID is specific for the VA repository design, the TSPA-VA
models and assumptions, and the data base used in the TSPA-VA. As noted by DOE in the VA
report, the VA data base, reference design, and TSPA results constitute a step in an evolutionary
process. Further design revisions and data additions are expected, leading to design features and
TSPA methods and results for the Site Recommendation and for a License Application if the site
is found to be a suitable location for disposal.
Comprehensive discussion of the TSPA-VA is included in this BID because it is the most
recently available detailed information concerning DOE performance assessments for Yucca
Mountain. Although revisions to TSPA-VA methods and results are expected, only limited
information on future repository designs and TSPA methods is currently available.
Documentation of the first draft of the TSPA for the Site Recommendation is currently planned
to be available in July 2000; documentation of a revised TSPA-SR is currently planned for
February 2001.
7.3.2.1 Repository Design Features for the TSPA-VA
Repository design concepts have evolved significantly over the years of site evaluation. As
previously noted, for example, the design concept used in the Site Characterization Plan issued in
1988 was vertical emplacement of canisters with small capacities into the floors of the tunnels
and with expected lifetimes on the order of 300-1,000 years. The basic concept used for the
TSPA-VA was to emplace large, highly robust waste packages with design lifetimes on the order
of tens of thousands of years horizontally in excavated drifts. This concept is similar to that used
in TSPA-95, but the waste package wail materials were different.
This section summarizes the engineered features of the VA repository that are of importance to
safety performance and TSPA results. In general, these are design features that are specifically
selected to aid waste isolation by delaying and diminishing opportunities for water to enter the
drifts, to contact the waste form, leach out radionuclides, and transport the radioactivity to the
environment.
7-148
-------
In the reference Engineered Barrier System (EBS) design that served as the basis for the TSPA-
VA analyses, the principal design features that contributed to waste isolation were use of high
waste package emplacement density so that repository temperatures would be high enough to
boil water in the rocks and drive it away from the repository for as long as possible; use of a drift
liner to help keep out seepage water for as long as the liner lasts; and use of a highly corrosion-
resistant waste-package wall material which would be expected not to be penetrated by corrosion
for very long periods of time. The TSPA-VA also characterized the potential performance of
supplemental engineered features (use of backfill, drip shields over the waste packages, and
ceramic coatings on the packages), but these features were not included in the VA reference
design.
Assumptions That Provide the Basis for Design Parameter Values
within the framework of the waste isolation strategy outlined above, assumptions were necessary
as a basis for selecting design parameters. Key assumptions included the following:
The Nuclear Waste Policy Act of 1982 limits the repository to a total capacity of
70,000 metric tonnes of uranium (MTU) as spent fuel or equivalent. The
repository for the TSPA was assumed to contain 63,000 MTU of commercial
spent fuel and 7,000 MTU equivalent of defense wastes, including vitrified high-
level waste from defense production operations and spent fuel from naval
reactors.
• Spent nuclear fuel assemblies from pressurized-water reactors will be, on average,
25.9 years out-of-reactor, with a 3.69 weight percent initial enrichment and a
burnup value of 39.56 gigawatt-days per MTU. Spent fuel assemblies from
boiling water reactors will be, on average, 27.2 years out-of-reactor, with 3.00
weight percent initial enrichment and a burnup value of 32.24 gigawatt-days per
MTU.
Commercial spent nuclear fuel (CSNF) will be emplaced in the repository in
packages containing 21, 12, or 24 PWR assemblies per package and 44 BWR
assemblies per package each containing about 10 MTHM. There will be a total of
7,642 CSNF packages in the repository. There will be a total of 2,858 packages
of defense wastes, for a repository total of 10,500 waste packages.
The surface facilities, subsurface facilities, and waste package designs will be
based on a reference areal mass loading range of 80 to 100 MTU/acre.
The temperature of the drift walls will be limited to no more than 200 °C (392 °F).
7-149
-------
The temperature of the CSNF fuel cladding will be limited to 350°C (662 °F).
• The repository's western and eastern boundaries will be between the Solitario
Canyon fault and the Ghost Dance fault
The reference repository and waste package designs that emerged from these and other
assumptions important to safety for handling and emplacement operations are summarized
below.
Repository Footprint
The repository layout that resulted from the assumptions concerning standoff from the faults,
temperature limits, and the areal emplacement density is shown in Figure 7-30. The repository
east-west width is about 1 km and the north-south length is about 3 km. The repository would be
located at a depth about 300 m (1,000 feet) below the crest of the mountain and 300 m above the
water table. The main emplacement drifts would be 5.5 meters (18 feet) in diameter; 104 drifts,
totaling 107 km (67 miles) of length, would be excavated to emplace the 70,000 MTU of wastes.
The drifts would be spaced 28 meters (90 feet) apart, and the extraction ratio (fraction of the
volume excavated) for the emplacement region of the repository would be 19.6%.
Waste Package Emplacement Configuration
Given the assumptions about waste-package capacity, each package would be about 6 feet (2
meters) in diameter and about 6 meters (18 feet) long to accommodate the dimensions of the
intact CSNF assemblies. Details of the package dimensions will vary because of variations in
assembly dimensions.
A cross-section diagram of a typical waste package emplaced in a drift is shown in Figure 7-35.
The package will be emplaced horizontally on steel V-shaped supports, which in turn are set on a
concrete invert and pier. The drift is lined with concrete. The invert completes a concrete ring
around the perimeter of the drift and also provides a roadbed for construction and emplacement
operations.
7-150
-------
Wasie Package Design
A perspective diagram of the waste package design for disposal of 21 PWR spent fuel assemblies
is shown in Figure 7-34. Packages for disposal of BWR spent fuel assemblies and for disposal of
defense wastes are conceptually similar in design. As previously indicated, the packages for
disposal of PWR and BWR spent fuel would be about 6 feet in diameter and 18 feet long.
Packages for disposal of defense wastes would be about 6 feet in diameter and 10 feet long.
The design features of most importance to the TSPA-VA are the materials selected for the waste
package walls, identified in Figure 7-34 as the inner and outer barriers. Each package has an
inner barrier of Alloy 22, which is a high-nickel, corrosion-resistant alloy intended in the design
to provide the principal barrier to penetration of water into the interior of the package. The outer
barrier, which in the reference design is a 516 steel, is intended primarily to provide shielding
and package strength. The reference design thickness of the outer barrier is 100 mm (4 inches);
the inner barrier is 20 mm (0.7 inches) thick.
Design Options
Many other possible design concepts and parameter values are identified and discussed in some
detail in the VA documentation (see, for example, Volume 2, Section 8 of DOE98). The options
include alternative design features, such as use of drip shields or ceramic coatings to defer the
time at which water can contact the waste package wall and begin to penetrate it, and alternative
design strategies. Although not part of the VA reference design, the effects of backfill, drip
shields, and ceramic coatings on repository performance were evaluated in the TSPA-VA.
Alternative strategies include use of a low emplacement density or long-term cooling before
eniplacement, either of which would reduce the areal thermal loading and would be intended to
reduce performance issues and uncertainties arising from the high temperatures associated with
the VA reference design. DOE is proceeding to characterize and evaluate some of the options,
^hich might be implemented in the design for the Site Recommendation and the License
Application if the site is found suitable for use for disposal.
7-151
-------
7.3.2.2 TSPA Concepts and Methodology
This section presents an overview of TSPA concepts and methodologies that are the basis for
DOE's implementation of performance assessment in the TSPA-VA. As previously noted, the
TSPA-VA is a snapshot in time of performance evaluation for the VA reference design, data
base, and models that were available for the purpose. If the Yucca Mountain project proceeds to
the stage of preparing a License Application for a repository at Yucca Mountain, the details of
the TSPA for the application would likely be different from those of the TSPA-VA.
The basic TSPA principles used for the TSPA-VA have been adopted in radioactive waste
disposal programs throughout the world as the means for forecasting the post-disposal
performance of a repository. For any given repository natural setting and engineered design, the
process involves five basic steps:
• Develop and screen scenarios of conditions and factors important to performance.
Scenarios address features, processes, and events that can affect repository
performance, such as average annual precipitation rates and changes therein.
• Develop analytical models to represent the factors important to performance. The
models are usually implemented as computer codes
Assign values to performance parameters in the models. Some parameters will be
single-valued, such as the density of water at a given temperature; others will have
uncertainty ranges because of inherent variability or lack of certain knowledge of
the value.
• Implement the models by operating the computer codes
• Interpret and apply the results for purposes such as identification of additional
data needs or assessment of compliance with regulatory standards
For a proposed repository at Yucca Mountain with its particular geohydrologic setting, DOE
selected four basic performance strategy factors:
• Limit the potential for water to contact the waste packages
• Design the waste package for a long lifetime
Seek a low rate of release from breached waste packages
7-152
-------
• Seek radionuclide concentration reduction during transport through the
environment to the location of the dose receptor
This strategy was implemented by identifying principal performance factors and components of
the TSPA modeling configuration as shown in Table 7-10. As indicated in this table, the model
components are aligned with the Key Technical Issues that NRC has identified as the basis for
review of DOE's assessments of repository performance. Parameter values and subsystem
models were developed for each of the 19 principal performance factors listed in Table 7-10.
Each of the performance factors listed in Table 7-10 can be characterized as a driver or an
inhibitor of radionuclide release and transport. For example:
Precipitation, infiltration, seepage, and dripping are drivers for radionuclide
release that bring water to the waste packages
• Waste package humidity, temperature, and chemistry drive the rate of attack on
the inner and outer waste package barriers
• The waste package wall is a principal inhibitor of radionuclide release; inhibition
of release is also accomplished by the integrity of the spent fuel cladding,
resistance to dissolution of the waste forms, and the limited solubility in water of
Np-237
Radionuclide mobility during transit from the repository to and through the
environment is aided if the radionuclides are attached to colloids but inhibited if
they become sorbed onto surfaces along the flow path
Transport of radionuclide-bearing water from breached packages brings the
radionuclides to the dose receptor location through pathways in the unsaturated
and saturated zones
• Dilution during transit and pumping will reduce the radionuclide concentrations in
water used by the dose receptor
• Biosphere transport will bring radionuclides into contact with the dose receptor in
accord with his/her life style and practices
*he specific characteristics of each of these drivers or inhibitors of radionuclide release and
transport are represented in the parameters and models used in the TSPA.
7-153
-------
Table 7-10. Principal Performance Factors for TSPA-VA Modeling (DOE98)
Attribute* oUh*
Repository
Safety Strategy
Limited water
contacting, waste
packages
Long waste
package lifetime
Low file of
release of radio-
midic'is from
breac'wd waste
packages
• Radip'iuclide
concentration
reduction during
transport from
the wfste
packages
Principal Factor*
Prsclpitalion and infiltration of water into the
mountain
Percolation to depth
Seepage into drifts
Effects of heat and excavation on flow
Dripping onto waste package
Humidity and temperature at waste package
Chemistry on waste package
Integrity of waste package outer barrier
Integrity of waste package inner barrier
Seepage into waste package
Integrity ol spent nuclear luel cladding
Dissolution of UOj and glass waste form
Solubility of neptunium-237
Formation of radionudide-bearing conoids
Transport within and out ol waste package
Transport through unsaturaled zone
Transport in saturated zone
Dilution from pumping
Biosphere transport
TSPA Model Component*
Unsaturated Zone Flow
Seepage
Thermal Hydrology - Mountain Scale
Thermal Hydrology-Drift Scale
Near-Field Geocnemical Environment
Waste Package Degradation
Waste Form Degradation
Radionuclide Mobilization and
Engineered Barrier System Transport
Unsaturated Zone Transport
Saturated Zone Flow and Transport
Biosphere Transport and Uptake
NRC Key
Technical tasue
Unsaturated and
Saturated Flow
under Isothermal
Conditions
Repository Design
and Thenmome-
chanicat Effects
Thermal Effects on
Flow
Evolution of the
Near-Field
Environment
Container Life and
Source Term
i
Unsaturaled and
Saturated Flow
under Isothermal
Conditions and I
Radionuclide
Transport
As noted in Section 7.2, one of the features of the repository design used in the TSPA-VA was an
initial high thermal loading, i.e., 85 MTU/acre, with a drift wall temperature of 200 degrees C.
The performance objective for this design concept is to drive the water in the geologic
formations around the repository away from the drifts for as long as possible, while radionuclides
in the wastes decay and heat emissions from the waste packages decrease. An adverse
consequence of the concept is that it produces high temperature levels and temperature gradients,
which will accelerate degradation processes and can change the characteristics of the geologic
formations. The thermal, chemical, hydrologic, and mechanical factors associated with the high
temperatures are coupled in highly complex ways that are difficult to model and characterize
with reliable parameter values. The modeling approach used in the TSPA-V^. uncoupled these
factors, thereby adding to the uncertainty of the TSPA-VA results.
7-154
-------
The computer codes and their configuration used in the TSPA-VA are shown in Figure 7-36. As
indicated in this diagram, thermal hydrology factors and UZ flow were modeled at both
mountain (large) and drift (small) scales. The Repository Integration Program (RIP) code
receives input from the codes for the individual performance factors and processes the inputs to
calculate radiation doses to the dose receptor(s). Many of the codes shown in Figure 7-36 were
developed or adapted specifically for use in the TSPA-VA; details are provided in the VA
documentation (DOE98) and supporting documents (DOE98a).
The codes used in the TSPA-VA include considerations of uncertainty and produce
characterizations of uncertainty in the assessment results. Four types of uncertainty are
considered: parameter value uncertainty, conceptual model uncertainty, numerical model
uncertainty, and uncertainty in the occurrence of future events such as earthquakes or human
intrusion into the repository. For the TSPA-VA, there was considerable uncertainty in most of
the component models and in parameters that represent performance factors that are inherently
variable or had a sparse data base. Techniques such as Monte Carlo sampling are used to
characterize uncertainty in the results of the assessments; uncertainties in the peak dose rate
results of the TSPA-VA evaluations spanned four to five orders of magnitude.
Nine radionuclides were considered in the TSPA-VA evaluations: C-14,1-129, Np-237, Pr-231,
Pu-239, Pu-242, Se-79, Tc-99, and U-234. These are the nuclides that prior TSPA work has
shown to have the most potential to produce dose effects in the future because of their long half-
hves, their high dose consequences (e.g., Np and Pu), or their high mobility in the environment
(e.g., Tc-99, and 1-129). As discussed below, the highly mobile Tc-99 and 1-129 were found to
he the source for doses in the 10,000 year time period; Np-237 dominated doses in the period
tens-of-thousands to about 300,000 years; and Np-237 and Pu-242 were dominant hi the period
from 300,000 to one million years.
7-3.2.3 Key Features of the TSPA-VA Base Case Models
This section summarizes key features of the performance factors and computer codes that were
used to implement the TSPA-VA. The descriptions are based on information contained in
^OE98, Volume 3, Section 4. Highly detailed discussions of the performance factors were
Provided in the chapters of the Technical Basis Document for the VA (DOE98a), and in topical
reports that were discussed as references in the Technical Basis Document chapters.
7-155
-------
near-field^
geochemfcal
thermal
hydrology
igWAPDEG \*.y.
-/waste-package
^degradation... -.-.
^drift-scale
thermal
[infiltrations;*
•-»* t • «!5" , —,•
•".-..-•.-.v* s».'.- -. •*•-.'-
CLAD-DEG
drift-scale ~\
^unsaturated^
v* zone flow
waste-form
degradation,
EBS transport
::TOUGH2
mountain-scale
• unsaturated
: zone flow
unsaturated
zone transport
Final
Performance
Measure
SZ CONVOLUTE
FEHM
saturated
zone flow,
transport
saturated zone
transport
GENII-S
biosphere
dose
calculation
Biosphere
Run within RIP -'
OUTPUT Parameters
Legend
T Temperature
RH Relative humidity
S( Liquid saturation
Xt Air mass fraction
qg Gas flux
. Between RIP Cells and
External Corf«
Figure 7-36. Computer Code Configuration for the TSPA-VA (DOE98)
7-156
-------
Climate
The TSPA-VA assumed there would be three characteristic climate regimes in the future at
Yucca Mountain, with periodic recurrence intervals: dry (current conditions), long-term average,
and superpluvial. Present conditions were assumed to prevail for the next 5,000 years. Long-
term average conditions were assumed to persist for 90,000 years each time they occur, and
superpluvial periods were assumed to last for 10,000 years.
Average precipitation rates in the long-term average and superpluvial periods were assumed to be
two and three times, respectively, higher than present rates, which average about 170 mm/yr.
Two superpluvial periods, in which glaciation is at a maximum and temperatures are a minimum,
were assumed to occur in the next million years: one at about 300,000 years and the other at
700,000 years. Between the superpluvials, the 5,000-year dry periods and the 90,000-year long-
term average periods alternate. Under these assumptions, about 90% of the next million years
experiences the long-term average climate..
The water-table level was assumed to respond to the changes in precipitation, rising by 80 meters
from present levels during long-term average climates and 120 meters during the superpluvial
periods. One of the modeling consequences of the water-table rise is that the UZ flow path
length is shortened.
Unsaturated Zone Flow and Infiltration
On the basis of site characterization data, the repository footprint was divided into six UZ flow
and infiltration zones. Three-dimensional steady-state flow models were developed for fracture
and matrix flow under current climate conditions and were extrapolated to the wetter climate
conditions. Average infiltration rates for the present, long-term average and superpluvial climate
conditions were assumed to be 7.7, 42, and 110 mm/yr, respectively. The infiltration rates were
therefore assumed to increase by factors of about 6 and 14 from the present rate, even though the
precipitation rate increases only by factors of 2 and 3.
Drift Scale Seepage
Characterization of seepage into the drifts was based on modeling of a three-dimensional,
heterogeneous fracture continuum surrounding the drifts. The seepage flow rate and fraction of
7-157
-------
the packages that are affected by seeps were modeled in terms of percolation flux, i.e., the water
flux that arrives at the repository horizon after infiltration at the surface and flow through the UZ
above the repository. Percolation flux was characterized for each of the six regions of the
repository footprint and the three climate conditions, based on site data and the climate model.
The modeling showed that about 10% of the waste packages would be exposed to seeps during
the dry-climate period, 30% would be exposed to seeps during the long-term average climate
conditions, and 50% would be exposed during the superpluvial periods. The estimates of the
fraction of the packages exposed to seeps had a very high uncertainty range in the TSPA-VA
evaluations.
Thermal Hydrology
Thermal hydrology addresses the temporal and spatial impact of the spent fuel heat output on the
natural system geologic and hydrologic characteristics and on the performance of the engineered
features of the repository. Thermal hydrology models are used to calculate temperatures (waste
package surface, waste form, drift wall) and relative humidities in the drifts. Values for these
parameters provide information needed for other models such as the waste package degradation
model and the near-field geochemical environment models. Standard models of heat transfer,
and data concerning the physical properties of repository system materials, are used to
characterize the thermal parameters.
Near Field Geochemical Environment
The near-field geochemical environment models calculate the time-dependent evolution of the
gas and water compositions that interact with the waste package, the waste form, and other
materials in the drift. The evolution of changes in gas and water composition is modeled as a
sequence of steady-state conditions. The chemical, thermal, hydrologic, and mechanical factors
important to the near field environment are in reality coupled, but an integrated model of the
coupling and its effects was not developed for the TSPA-VA.
Five separate but interacting models were used in the TSPA-VA to characterize the near field
geochemical environment:
• Gas, water, and colloid compositions as they enter the drift
Composition of the in-drift gas phase
7-158
-------
• Chemistry of in-drift interactions of water with the solids and gases in the drift
• In-drift colloid compositions
• In-drift microbial communities
The near-field geochemical environment models are connected to other component models see
Figure 7-37), The near-field models receive input from the UZ and thermal hydrology models
and from design parameters; they provide outputs to the waste package corrosion model, the
waste form model, the UZ radionuclide transport model, and the nuclear criticality model.
Waste Package Degradation
Modeling of waste package degradation was based on waste type contained in the package,
whether the packages were dripped on or not dripped on, and their location in the repository.
Seepage into the drifts is modeled as a function of the infiltration rate of water and the fracture
Properties of the rock. With the expected percolation flux, only about one-third of the waste
packages are dripped for most of the one million year modeling period. If water seeps onto the
surface of a waste package, 100% of the surface is assumed to be wetted. Uncertainty in the
corrosion rate of the Alloy 22 corrosion-resistant barrier in the waste package wall was also
modeled, and the expected-value base case assumed that a single juvenile waste package failure
occurs 1,000 years after disposal. Corrosion of waste package materials was assumed to occur
pits and patches that always encounter seeping water.
Cladding Degradation
Mechanisms included in models for degradation of fuel rod cladding on commercial spent
nuclear fuel included some pre-disposal failures, creep failure of zircaloy at high temperatures,
total failure of rods clad with stainless steel, fuel rod fracture from falling rocks, and long-term
general corrosion failure. Breaching of cladding was assumed to expose all of the waste-form
surface in the rod to water that had entered the waste package.
Waste Form Degradation and Mobilization
Dissolution of CSNF was modeled to be a function of pH, temperature, and total dissolved
carbonate; model parameters were based on experimental data. Dissolution of vitrified high-
level defense waste was modeled as a function of surface temperature and water pH, and a
dissolution rate constant for metals was used for degradation of the defense spent fuel from the
7-159
-------
N-Reactor. Under the assumption that all spent fuel is exposed and wetted for rods with
breached cladding, the spent fuel would be totally dissolved in about 1,000 years. Dissolution of
uranium dioxide fuel is known to result in formation of secondary minerals which can trap
species such as Np-237 and reduce their release, but credit for this phenomenon was not taken in
the TSPA-VA modeling.
Engineered Barrier System Transport
Transport in the BBS was modeled as a series of connected mixing cells, with one cell combining
the waste form and waste package, and three pathway cells representing the invert, in order to
reduce numerical dispersion in model calculations. The models did not include factors that could
defer and decrease radionuclide release after a waste-package wall is breached, such as low
seepage rates and partial seepage into the package interior, and in-package dilution. Sorption and
diffusional transport was assumed for radionuclide movement through the concrete invert.
Consistent with data which indicated rapid transport of plutonium from the Benham weapon test
location on the Nevada Test Site, a small fraction of the plutonium mobilized was assumed to be
attached to mobile colloids.
Unsaturated Zone Transport
The radionuclide transport model for the unsaturated zone was based on the flow model for that
zone. Three flow fields, corresponding to the three climate conditions, and a dual-permeability
geologic regime were assumed, Radionuclide movement was modeled using a three-dimensional
particle tracking model. Sorption was assumed to occur for Np-237, Pu-239, and Pu-242.
Matrix diffusion and dispersion were also assumed to occur,
Saturated Zone Flow and Transport
Flow in the saturated zone was simulated using a coarsely discretized three-dimensional model
which establishes the general plume direction and flow path in the geologic media. Radionuclide
transport was assumed to occur in six one-dimensional stream tubes corresponding to the six area
regions defined for the repository footprint. Based on the recommendations of the saturated zone
expert elicitation panel, the specific discharge in all stream tubes was assumed to be 0.6 m/yr,
and a dilution factor probability range, with a mean value of 10, was assumed to apply to all of
the stream tubes.
7-160
-------
Biosphere Transport
Water used by the dose receptor was assumed to be drawn from a well 20 km (12 miles) down
gradient from the repository. Dilution was assumed not to occur during pumping, so the
radionuclide concentration in the water emerging from the well is the same as the stream tube
concentration at the withdrawal location. The dose receptor was assumed to receive doses from
all biosphere pathways in accord with site-specific dose conversion factors and the water use and
life style habits assumed for the receptor. For the TSPA-VA, DOE assumed the dose receptor is
a current-day average adult living in Amargosa Valley. A survey was conducted to obtain
lifestyle and dietary data for the dose evaluations.
7.3.3 TSPA-VA Results
DOE produced the following categories of TSPA-VA results:
Deterministic results for the TSPA-VA base case
• Results of uncertainty analyses using Monte Carlo techniques
• Results of analyses to assess the sensitivity of performance to uncertainties in
parameter values
Assessments of the effect of disruptive events on performance
Assessment of the effect of design options on performance
Collectively, these assessment results address the expected performance of the repository, the
role of the various performance factors in producing the expected performance, factors that could
alter expected performance, and the uncertainty in expected performance. The repository
Performance forecasted for the base case is discussed in Section 7.3.3.1. Uncertainties in the
TSPA-VA result are discussed in Section 7.3.3.2.
7-3.3.1 Base Case Expected Repository Performance
The deterministic results for the TSPA-VA base case are responsive to the Congressional
mandate for assessment of "...the probable behavior of the repository in the Yucca Mountain
geological setting...". These results were a forecast of the dose rate to the average individual
located 20 km from the repository, for time periods up to one million years. Graphs showing
7-161
-------
forecasts of peak doses throughout the million-year time period were produced, and specific
dose-rate values were identified and discussed for time periods of 10,000, 100,000 and one
million years.
DOE described the results for the deterministic evaluation in which values for all uncertain
parameters were set at their expected values as follows (DOE98, Volume 3, p. 4-21):
"1. Within the first 10,000 years, the only radionuclides to reach the biosphere are the
nonsorbing radionuclides with high inventories, technetium-99 and iodine-129, and the
total peak dose rate is about Q. 04 mrem/year.
2. Within the first 100, 000 years, the weakly sorbing radionudide neptuniwn-23 7 begins
to dominate doses in the biosphere at about 50,000years, with the total dose rate
reaching about 5 mrem/year.
3. Within the first million years, neptunium continues to be the major contributor to peak
dose rate, which reaches a maximum of about 300 mrem/year at about 300,000 years
after closure of the repository, just following the first climatic superpluvial period. The
radionudide plutonium- 242 is also important during the one million-year time frame
and has two peaks, at about 320,000 and 720,000 years, closely following the two
superpluvial periods. There are regularly spaced spikes in all the dose rate curves (more
pronounced for nonsorbing radionuclides such as Tc-99 and 1-129) corresponding to the
assumed climate model for the expected value base-case simulation... these spikes are a
result of assumed abrupt changes in water table elevation and seepage through the
packages."
As shown in Figure 7-37, doses to the receptor 20 km from the repository, as a result of the
mobile Tc-99 and 1-129 radionuclides, first occur about 3,500 years after disposal. These fission
products are dominant because of substantial inventory in CSNF, high solubility in seepage
water, relatively low decay rate relative to 10,000 years, and neglible sorption on tuff rocks. The
scenario presented hi Figure 7-38 results from the assumption that a single juvenile waste-
package failure occurs at 1,000 years; the "blip" in the curve at about 5,500 years is the result of
the change of climate conditions from dry to long-term average at 5,000 years, which causes a
major rise in the water table. During the 10,000-year period, 17 additional packages are modeled
to fail at various times, beginning at about 4,200 years. These failures contribute to the dose at
10,000 years in accord with the TSPA-VA model assumptions concerning package failure times
and conditions.
7-162
-------
10,000-yr Dose Rate
10'3
0 2,000 4,000 6:000 8.000 10,000
Time (years)
Figure 7-37. TSPA-VA Base Case Dose Rates for Periods Up to 10,000 Years (DOE98)
100,000-yr Dose Rate
20,000 40,000 60,000 80.000 100,000
Time (years)
Figure 7-38 TSPA-VA Base Case Dose Rates for Periods Up to 100,000 Years (DOE98)
7-163
-------
Dose rate histories for times up to 100,000 years are shown in Figure 7-38. Tc-99 continues to
dominate the dose rate up to about 50,000 years, after which the Np-237 dominates the dose rate
out to 100,000 years. There is a relatively large inventory of Np-237 in CSNF resulting from the
decay of Am-241. The Np-237 does not begin to appear at the dose location until after about
30,000 years, because its release from the waste form is solubility limited and it exhibits some
sorption on the rock surfaces along the transport pathway. The Pu-239 does not begin to appear
at the dose location until more than 80,000 years have elapsed because it is more strongly sorbed
than the Np-237. A small fraction of the Pu-239 is assumed, however, to be attached to colloids
that are not sorbed onto the rock surfaces.
As with the 10,000-year results, the dose rate forecasts for periods to 100,000 years are
dominated by climate change assumptions and waste package failure history. The jagged
appearance of the Tc-99 curve is the result of individual package failures; each small peak
corresponds to a failure. This illustrates one of the key features of the TSPA-VA modeling
scheme: because features such as slow drip entry to the package interiors and in-package
dilution, which provide storage capacity along the transport path, were not included in the
models, the nonsorbing species such as Tc-99 directly track release behavior, and concentrations
are simply attenuated by dilution along the pathway. The sorbing and solubility-limited species,
such as Np-237 and Pu-239, have the capacity for storage along the transport path because of
these properties, but the effects would have been more exaggerated if factors such as in-package
dilution had been included in the TSPA models.
As shown in Figure 7-39, Np-237 continues to dominate the dose rate from 100,000 years all the
way to the end of the million-year dose evaluation period. At about 300,000 years, Pu-242
becomes the second most important contributor to dose and remains in this role, at a level about
a factor often less than that of the Np-237, to the end of the dose evaluation period. The
contribution of other radionuclides to dose during the long-range time frame is insignificant.
The dose rate after about 300,000 years is seen in Figure 7-39 to be essentially constant. This is
because, in the TSPA-VA modeling scheme, the repository as a source term for radionuclides
released to the environment goes into essentially steady state. All of the packages that are
modeled to fail have failed, the seepage fluxes into the repository and into the packages have
become virtually the same and constant, and the rate of change in exposure of waste form has
become constant.
7-164
-------
1.000,000-yr Dose Rate
0 200.000 400,000 600,000 800,000 I.000,000
Time (years)
Figure 7-39. TSPA-VA Base Case Dose Rates for Periods Up to One Million Years (DOE98)
dominant effect of waste package failure history and climate conditions on dose rates
continues to the end of the million-year dose evaluation period. At about 200,000 years, cladding
degradation begins to contribute to the exposed waste form area, and at times greater than about
700,000 years, waste packages that are never dripped on, which total about 55% of the package
uwentory, begin to fail as a result of low corrosion rates in a non-wetted condition over a very
l°ng time frame.
The base case TSPA results for the VA repository show that the performance of the highly
complex and multi-element system is strongly dominated by very few factors. In brief:
• Performance is dominated by assumptions concerning waste package failure
history and climate, and the effect of these factors on predicted doses is primarily a
consequence of the assumptions concerning juvenile package failures and climate
change.
Three nuclides dominate the forecast doses: Tc-99 and I-129 in the shorter time
frames and Np-237 in the longer time frames. The dose levels associated with the
Np-237 are higher than those associated with the technetium and the iodine, in
large measure because the health consequences of a unit quantity of Np-237 are
much greater than those for the technetium and iodine.
7-165
-------
The fact that the dose results clearly reflect the occurrence of climate changes and
individual package failures shows that the TSPA-VA modeling system is
fundamentally simple. Factors in performance that would serve to smooth and
smear the consequences of phenomena that change system conditions were
omitted from the models.
7.3.3,2 Uncertainty in the TSPA-VA Results
The Monte Carlo type of analyses that were done to assess the uncertainty in the TSPA-VA
deterministic base-case results showed an uncertainty range spanning about four to five orders of
magnitude throughout the million-year period, as shown in Figure 7-40. These results were
obtained by using statistical methods to select values from the distributions for the uncertain
parameters used in the TSPA-VA models. For each of the three time frames (i.e., 10,000,
100,000, and one million years) one hundred such runs were done, and a few 1,000-run studies
were done to demonstrate that the uncertainty ranges found for the 100-run studies were
representative.
The large uncertainty range, i.e., spanning four to five orders of magnitude, is in part due to the
many uncertain parameters in the TSPA-VA computer codes. The RIP code alone, for example,
contains 177 uncertain parameters, and there are many more in the codes that have inputs to RIP.
Another possible cause of the wide uncertainty range is that many of the uncertain parameters
themselves have wide uncertainty ranges, either as a result of use of a broad range of possible
values because the actual value of the parameter is poorly known, or because the parameter is
inherently highly variable. It would be difficult, if not impossible, to sort out the sources and
principal causes of the uncertainty range. The uncertainty range for the TSPA-VA results is
therefore a consequence of the specific way uncertainty was used in assigning numerical value
distributions to parameters in the TSPA-VA models and codes.
Another source of uncertainty, not reflected in the results of the TSPA-VA studies, is the
possibility that some of the models used in the codes may not be correct, e.g., because of a sparse
data base, or, as in the case of modeling of the near-field geochemical environment, because
coupled phenomena were uncoupled to simplify modeling. This type of uncertainty should be
regarded as uncertainty in the conceptual models for the waste containment and isolation
systems. In translating conceptual models into calculational models, conservative assumptions
are typically made about processes which should be included and how the processes would
operate. This is done, in part, for modeling convenience, and, in part, because the level of
7-166
-------
10'8
2000 4000 6000
Time (years)
8000
10000
200000 400000 600000 800000 1000000
Figure 7-40. Uncertainties in the TSPA-VA Base Case Results (DOE98)
7-167
-------
process complexity cannot be handled manageably. These assumptions can have significant
implications for interpreting the results of performance assessments, and should be understood
when interpreting the results. (See Section 7.3.3.5 for additional discussion of conservatism in
the TSPA-VA modeling.)
In evaluating the status of knowledge and uncertainty as a prelude to selecting further work to
improve the TSPA methodology for a License Application (DOE98, Volume 4), DOE often
noted that the models used in the TSPA-VA might not adequately capture the full range of
possibilities. If this is indeed the case, and the uncertainty in parameters or models has to be
expanded hi order to embrace the full range of possibilities (as opposed to simply revising the
model in response to better information), the uncertainty ranges for future TSPA results might
actually be broader.
DOE used a technique known as Stepwise Regression Analysis to determine which of the
performance factors were most important to the uncertainty results. These evaluations showed,
for the 10,000-year tune period, that the fraction of the packages contacted by seepage, the mean
Alloy 22 corrosion rate, the number of juvenile failures, and the saturated zone dilution factor are
the most important performance parameters. For the 100,000-year period, the most important
parameters were the seepage fraction, the mean Alloy 22 corrosion rate, and the variability in the
Alloy 22 corrosion rate. For one million years, the most important factors were found to be the
seepage fraction, the saturated zone dilution factor, the mean Alloy 22 corrosion rate, and the
biosphere dose conversion factors. The fraction of waste packages contacted by seepage water
was the dominant performance factor for all three time periods. It is the dominant factor for
TSPA modeling of repository system performance because it has a direct effect on the number of
waste packages that fail, and it has a very large uncertainty.
Additional sensitivity studies were done to determine the performance factors of secondary
importance to the TSPA-VA results. In these analyses, the performance factor of primary
importance were held constant, and Monte Carlo runs were done for the other uncertain
parameters. The performance factors that were held constant at then: baseline values were the
infiltration rate and mountain-scale saturated zone flow rates, the fraction of the waste packages
contacted by seepage, the seepage flow rate, the Alloy 22 mean corrosion rate, and the Alloy 22
corrosion rate variability.
7-168
-------
With the above parameters held constant, the parameters of principal secondary importance at
10,000 years were found to be the saturated zone dilution factor, the biosphere dose conversion
factors, the solubility of technetium, and the fraction of seepage contacting a package that enters
a failed package. The factors that were important for 10,000 years were found to be also
important for 100,000 years, except that the solubility of neptunium replaced the solubility of
technetium as an important factor, and the fraction of saturated zone flow in alluvium was added
to the list. At one million years, the most important factors were the saturated zone dilution
factor, the cladding failures by corrosion and by mechanical disruption, the biosphere dose
conversion factors, the saturated zone longitudinal dispersivity, and the saturated zone alluvium
fraction. In all time frames, the most important of these secondary factors was the saturated zone
dilution factor.
All of these sensitivity findings reflect the fundamentals of repository system performance: the
potential doses depend primarily on the fraction of waste packages intercepted by seepage, the
amount of waste form available to be a source of radionuclides, the amount of water available to
Pick up the radionuclides and to transport them to the environment, the amount of water available
to dilute radionuclide concentrations, and the extent and means of interaction of the dose receptor
with the contaminated water.
7.3.3.3 Effects of Disruptive Events on Performance.
The TSP A-VA evaluated the effects of four types of disruptive events on repository
performance: basaltic igneous activity, seismic activity, nuclear criticality, and inadvertent
human intrusion. The basis for inclusion of evaluations of the effects of disruptive events on
repository performance includes the probability of occurrence of the event, the consequences of
occurrence, and any regulatory requirements that mandate or exclude consideration of
disturbances.
The igneous activity evaluations considered events in which molten igneous material is cooled
within the earth or on the surface. In the case where magma reaches the surface, explosive
releases may carry radioactive materials directly into the atmosphere. Cooling of magma within
the earth may involve destruction of waste packages so that radionuclides in the waste form are
more accessible for release and transport.
7-169
-------
Results of the direct-release igneous activity evaluations showed that the maximum dose rate
from this volcanism would be about two million times less than for the base case ground water
contamination scenario. The underground-cooling scenarios showed that dose rate peaks would
occur tens of thousands of years after the actual magma intrusion event.
The seismic activity studies considered phenomena such as rockfall onto waste packages as a
result of earthquakes, and the effects of seismicity on the hydrologic regime in the near field and
in the saturated zone. These studies showed that rockfalls could not contribute significantly to
waste package degradation until after at least 100,000 years and that changes to the hydrologic
regimes would be negligible. Overall results of the analyses showed that seismic events would
have almost no effect on repository performance over one million years.
The potential for nuclear criticality within waste packages and external to the packages after
transport of fissionable material from the package was investigated within the TSPA-VA. The
evaluations were done assuming that criticality occurs 15,000 years after emplacement, which is
when the commercial spent fuel is most reactive. The analyses determined that criticalities
external to the waste packages are not a credible event, and that criticality within a package is
extremely unlikely and would have insignificant consequences. Criticality within a waste
package is extremely unlikely because only 8% of the commercial fuel waste packages contain
sufficient fissile material to acheive a critical mass and only 10% of the waste packages are
expected to be breached in 40,000 years. Breached waste packages must retain sufficient water
to act as a moderator for the nuclear chain reaction to be sustained and DOE has estimated that
only 25% of the breached waste packages will hold water for a period sufficient to flush out
boron which is included hi the waste package as a neutron absorber. Even if criticality did occur
within the waste package, the incremental radioactivity is less than the normal radioactivity from
most waste.
In keeping with the recommendations of the NAS panel that developed the technical basis for the
Yucca Mountain standards, a stylized human intrusion scenario was characterized and evaluated.
Intrusion of a waste package by an 8-inch drill bit, as a result of search for water, was assumed.
The bit was assumed to penetrate the package and the mountain stratigraphy to the water table,
with large quantities of pulverized fuel being transported to the bottom of the bore hole, which
was never sealed. Water would then dissolve the fuel inventory at the bottom of the bore hole
and transport radioactive material to the dose receptor location. The intrusion was assumed to
occur at 10,000 years, which is the first time at which it is estimated the drill bit could penetrate
7-170
-------
the package wail. The NAS panel did not feel it would be useful to assess hazards to drillers or
to the public from radioactive materials transported directly to the surface since these risks would
be the same for all geologic repositories.
The total amount of fuel deposited at the bottom of the bore hole was assumed to range between
550 and 2,700 kilograms (1,200 and 6,000 pounds), which corresponds to about 5 to 22% of the
total spent fuel inventory in the package. The actual mass of fuel that would be intercepted by
the 8-inch drill would be about 160 kg, so the analyses assumed that large quantities of fuel
would be entrained by the bit as it passed through the package.
The analyses for this intrusion scenario showed that the consequent radionuclide releases for the
2,700 kg release would produce a blip in the dose rate curve, in comparison with the base case,
that starts at about 11,000 years, peaks at 12,000 years, at levels about 145 times higher than the
base case dose rate at that time (i.e., 1 mrem/yr), and returns to base case levels at about 14,000
years. The 550-kg spent fuel release from intrusion produces a dose rate at 12,000 years that is
3.7 times the base case dose rate. All effects of the intrusion on dose rate are gone by 150,000
years. The TSPA-VA observed that the effects of the intrusion on dose rates are significant only
for times near the occurrence of the intrusion, and that the maximum resulting dose is 1 mrem/yr.
7.3.3.4 Effects of Design Options on Performance
The TSPA-VA included evaluation of the effect, on repository performance, of design features
that were not included in the VA reference design. The three features considered were emplaced
drift backfill, drip shields, and ceramic coating of the disposal containers, with backfill. The
objective for use of these design options would be to reduce and defer liquid water contact with
the waste package.
7.3.3.4.1 Effects of Backfill
The backfill was assumed to be crushed tuff, emplaced 100 years after the end of emplacement
operations. The backfill will initially perform as a thermal blanket for the waste packages, and
cause a temperature spike of as much as 80-90 °C. The temperature spike might cause a slight
increase in the waste package corrosion rate, but it would also delay the rate of increase of
relative humidity as the heat emissions from the waste packages decrease and the repository
system cools. A potentially major effect of backfill would be to change the potential for, and
7-171
-------
patterns of, seepage water contacting the waste packages. This effect was not modeled in the
TSPA-VA analyses.
The analyses for the assumed backfill effects showed that the use of backfill would defer
corrosion of the Corrosion Allowance Material, but corrosion of the Corrosion Resistant Material
would be virtually unaffected based on the modeling assumption that corrosion of this material is
driven by whether or not dripping occurs and the same dripping conditions are assumed for the
case of backfill and no backfill. Use of backfill would therefore have little effect on repository
performance if the backfill does not reduce or defer contact of seepage water with the waste
packages. The backfill might actually have effects such as diverting the seepage water around
the waste packages or reducing the amount of seepage that gets to the package as a result of
evaporation, but a basis for modeling such effects was not available for the TSPA-VA.
7.33.4.2 Effects of Drip Shields
The drip shields were assumed to be made of Alloy 22 and to be 2 cm (0.8 in.) thick. The shields
would be shaped like a Quonset hut, shrouding the waste packages but not touching them. The
dripshields would be covered with backfill, emplaced 100 years after emplacement of the waste
packages was completed. The shields upper surfaces were assumed to be totally wet in dripping
regions of the repository, and they were assumed to fail only by general corrosion. After drip
shield failure, 10% of the waste package area under the failed shield was assumed to be wetted
(in contrast, the base case analyses assumed 100% of the package surface area would be wetted)
because only a small fraction of the drip shield surface area was modeled to fail.
TSPA-VA results based on the above assumptions showed that the drip shields enhanced the
overall waste package lifetime by more than 100,000 years. Dose rates for the first 300,000
years are reduced by one to two orders of magnitude in comparison with base case results. After
500,000 years, the drip shield dose projections become the same as those for the base case. The
results were interpreted to indicate that the life span if the drip shield is the key determinant of
improved performance.
As a result of these findings, drip shields are included as a design feature for the repository
design expected to be selected as the reference design for the Site Recommendation and the
License Application (see Section 7.2.2.5).
7-172
-------
7.3.3.4.3 Use of Ceramic Coating of Disposal Containers with Backfill
This design option involves coating the waste packages with a ceramic material in order to delay
corrosion of the outer wall of the packages (in the VA design, A 516 carbon steel). Backfill is
added to the repository to protect the ceramic coatings.
Performance of this design concept was modeled assuming that the ceramic coating functions as
a barrier to oxygen transport to the carbon steel package wall. For the assumed conditions, the
analyses determined that the ceramic coatings would not be breached for more than 300,000
years. Dose rates would not begin until about 500,000 years, and at one million years the dose
rates would be nearly two orders of magnitude less than those for the TSPA-VA base case.
If ceramic coatings perform as modeled for the TSPA-VA, they would have a profound effect on
repository system performance. At this time, however, there are uncertainties and concerns
associated with potential for defects and flaws in the coatings, differential thermal expansion
between the coating and the substrate that could result in cracks in the coating, and dissolution of
the coating over long time periods. Analysis of these effects is needed before the potential
benefits of use of ceramic coatings can be verified.
7.3.3.5 Conservatism In The TSPA-VA Base Case Results
The TSPA-VA base-case results (an expected (average) value dose rate of 0.04 mrem/yr 10,000
years after disposal, to a reference person 20 km downstream) are a consequence of choices that
were made concerning performance parameter values, performance models, and assumptions.
This section discusses conservatism that was exercised in making the TSPA-VA choices, and the
effects of conservatism on the base case results. Similar discussions are provided for the NRC
performance assessments (Section 7.3.5.3) and the EPRI assessments (Section 7.3.6.4).
Performance Parameters
The TSPA-VA base-case evaluations used expected values of performance parameters, based on
available information. Expected values for some of the parameters, such as the dilution factor for
the saturated zone and corrosion rates of Alloy 22, were based primarily on results of expert
elicitations because of limited availability of data at the time that the TSPA-VA analyses were
Performed. The parameter values developed by the expert elicitations may be conservative
7-173
-------
because the experts are, themselves, working with limited information. Expected values of
parameters, and the uncertainty ranges for parameters that are inherently variable, may change in
the future as a result of data additions, but the TSPA-VA analyses sought to be as realistic as
possible, rather than conservative, in their choices of performance parameter values.
Performance Models
Conservatism in the suite of performance models and computer codes used for the TSPA-VA
analyses was introduced by using simplified models and by omitting from the suite of models
some performance factors that could have significant impact on predicted doses. Examples of
this type of conservatism include:
Dilution and transport delay for radionuclides released from the waste form but in
water still within the failed package were not considered. Under realistic package
failure conditions during the first 10,000 years, when disruptive failure scenarios
are insignificant, water will fill the package interior very slowly from a
penetration in the top. By the time that radionuclide release and in-package
transport occurs, temperature gradients will be too low to drive advective
transport processes, and temperature levels will be too low for inside-to-outside
corrosion of the Alloy 22 to occur and create an exit at the bottom of the package.
Radionuclide transport rates within the package will therefore be low, the package
interior will have to fill with water in order to enable radionuclides to exit through
the same penetration that provides water ingress, and the volume of water to fill
the package interior will be available to provide dilution. Radionuclide releases to
the exit of the package may therefore be greatly delayed, and concentrations at the
package exit would be much lower than for the no-dilution assumption.
• Release of radionuclides from a breached waste package was assumed in the VA
models to begin immediately after the waste package was breached, i.e., an exit
hole in the metal container was assumed to be created as soon as the container
wall was breached by corrosion. In reality there would be a time delay before an
exit hole at another location on the container was developed. This time delay
could be relatively short if exterior corrosion was taking place concurrently at
opposite sides of the container, or it could be very long if, as indicated above, the
exit pathway had to develop from inside the container. By delaying the exit of
radionuclides the actual containment time of the waste containers would be
significantly increased and doses during the regulatory time frame would be
consequently decreased.
• Dilution of radionuclide concentrations during transit of the unsaturated zone
from the repository to the water table was not considered. When few packages are
7-174
-------
failed and releasing radionuclides (in the TSPA-VA, only 18 of 10,000 packages
are failed at 10,000 years), uncontaminated percolation water adjacent to
contaminated streams emanating from the failed waste packages could provide
extensive dilution as a result of mixing of contaminated and uncontaminated
water in the fracture and matrix flow paths. This mixing would lower the
radionuclide concentrations at the start of saturated zone transport and result in
lower predicted doses to the receptor.
• A simple, one-dimensional model of radionuclide transport along the saturated
zone flow paths from the repository to the dose receptor location was used, and
dilution of initial SZ radionuclide concentrations under the repository was
assumed to occur at the end of the path, in accord with dilution factors
recommended by experts (for the base case, a dilution factor of 10 was used).
Processes that could delay and disperse radionuclide transport along the pathway,
and therefore would reduce the predicted dose rates to the receptor, were not
included in the modeling.
• Dilution during well pumping by the dose receptor was assumed not to occur.
This expected dilution process, which is included in NRC modeling of repository
performance, would reduce predicted doses to the receptor.
These processes and phenomena were omitted from TSPA-VA modeling of repository
Performance because at the time the data base for characterizing the relevant performance
parameters and their uncertainties was limited or non-existent. Also, the magnitude of these
effects is difficult to quantify with high confidence even with site characterization and laboratory
'work focused on them. However, these processes would be expected to function in the actual
repository environment, and reasonable but cautious estimates could be made to support
assessments, through a combination of data collection and expert judgment.
Rather than choosing to incorporate models for these processes in the TSPA-VA assessments,
"with estimated values of the parameters used in the calculations, they were omitted from the suite
of TSPA-VA models. This approach had the consequence of producing a spectrum of
performance results that are an assessment of a potentially very conservative performance
scenario, incorporating some unrealistic modeling assumptions. Omission of these modeling
features introduces a significant level of conservatism in the assessment results whereas better
Performance would reasonably be expected.
Additional data (e.g., additional characterization of the SZ geology and hydrology), may enable
inclusion of at least some of these performance factors in the TSPA for the LA. Their omission
7-175
-------
introduces conservatism to the TSPA results, but also avoids licensing issues that may be
difficult to resolve unless a data base adequate to support their use is available.
Conservative Assumptions
The TSPA-VA evaluations included conservative assumptions for some of the key performance
factors, as follows:
• In the base case, early failure of a waste package was assumed to occur at 1,000
years as a result of an imperfection such as a poor weld. Performance parameters
selected in association with this assumption (e.g., the size of the hole on the
package wall) were such that nuclide releases from this single package were a
dominant factor in the predicted base case dose rate at 10,000 years.
• The Corrosion Resistant Material for the waste package wall, Alloy 22, was
assumed to be penetrated rapidly by crevice corrosion as a result of being under
carbon steel in the VA waste package design. This assumption was derived from
the waste package expert elicitation, which conservatively interpreted the highly
limited data base for the corrosion performance of Alloy 22.
• In characterizing corrosion processes, the TSPA-VA assumed that all ground
water seeping into the emplacement drifts contacts the waste packages, even
though the package width is only one-third the width of the drift, thereby
overstating the amount of water available to cause corrosion. In addition, the
entire surface of a waste package wetted by seepage water dripping onto the
package was assumed to be wetted, and all seepage water contacting the package
was assumed to enter the package wall penetration(s) when they occur. The
TSPA-VA support analyses (DOE98a) recognized that only a small fraction of the
waste package surface would be wetted (the total amount of water contacting the
package each year is estimated to be on the order of 20 liters), and that only a
fraction of the seepage water contacting the package would enter the wall
penetration (e.g., because corrosion products would block entry). Because of
uncertainties in placing values on the relevant performance parameters, these
factors, which could greatly defer and diminish radionuclide release from the
waste form, were omitted from the TSPA-VA evaluations and the bounding
conservative assumptions were used.
The TSPA-VA assumed that 0.1 % of the Zircaloy-clad commercial spent fuel
rods emplaced in the repository will be "failed" at the time of emplacement, that
the spent fuel contents of each penetrated waste package will include 1.15%
stainless-steel-clad fuel rods, all of which fail completely and immediately when
the package wall is penetrated, and that all waste form area in failed fuel rods is
7-176
-------
exposed and contacted by water that enters the package. Overall, therefore,
1.25% of the waste form area in a failed package was assumed to be exposed and
wetted. In the context of the TSPA-VA evaluations this was considered by some
(i.e., NRC staff and the NWTRB) to constitute cladding "credit" because only a
small fraction of the waste form in a failed package was assumed to be exposed
and wetted. The TSPA-VA assumptions may in fact greatly overstate the extent
of exposed waste form area. An extensive data base shows that "failures" of spent
fuel cladding are predominantly hairline cracks which would expose only a small
waste form area. In addition, the Zircaloy cladding is not susceptible to
significant degradation after disposal, and there are only about 2,100 stainless-
steel-clad subassemblies, which could be packaged together in less than 100 of the
10,000 waste packages. These segregated packages could be made more failure
resistant by using some of the design options assessed in the VA, such as drip
shields. With a greatly prolonged waste package lifetime the level of assumed
cladding "failure" at emplacement would be lowered by an order of magnitude
with consequent lowering of the dose to the receptor. In summary, if only the
penetrations of Zircaloy cladding that exist at emplacement allow water to contact
the waste form, and if extreme assumptions concerning stainless-steel-clad spent
fuel are avoided, the DOE assumptions could overstate the waste form area
available for radionuclide release by as much as three orders of magnitude.
7-3.4 Reviews of the TSPA-VA
Formal reviews of the DOE Viability Assessment and the TSPA-VA were documented by the
Nuclear Regulatory Commission, the TSPA-VA Peer Review Panel, and the Nuclear Waste
Technical Review Board. Their comments are summarized below.
7-3.4.1 NRC Review of the TSPA-VA
In a March 1999 letter to the NRC Commissioners, the NRC Staff provided comments on the
TSPA-VA (NRC99c). In addition, the NRC provided some informal feedback to DOE during
the May 25-27, 1999 DOE/NRC Technical Exchange (NRC99b). The NRC's feedback was
based primarily on a comparison of the TSPA-VA with NRC's TPA 3.2 performance assessment.
Details of TPA 3.2 are presented in Section 7.3.5. As discussed in that section, there are
substantive differences in the models and parameters used by the two agencies. The purpose of
this section is not describe the differences between the TSPA-VA and TPA 3.2 but rather to
summarize some of the key NRC comments on the TSPA-VA.
7-177
-------
The NRC Staff review covered: (1) the preliminary design concept for the critical elements of the
repository and the waste packages; (2) the TSPA based on this design concept and data available
as of June 1998; and (3) the license application (LA) plan. The Staff did not review the DOE
cost estimates to construct and operate the Yucca Mountain repository. The review focused on
those issues that needed to be addressed before the LA is issued (scheduled for,2002) to insure
that the application will be complete and minimize the need for a protracted license review. The
NRC agreed with DOE's position that work should proceed toward a decision on recommending
the Yucca Mountain site as a repository for high-level waste.
There were a number of areas where the NRC Staff did not have major comments at the time of
its review based on general agreement with DOE on the particular issues. These included:
mechanical disruption of the waste packages; radionuciide release rates and solubility limits;
spatial and temporal distribution of flow in the unsaturated zone (UZ); distribution of mass flux
between fractures and matrix in the unsaturated zone; retardation in the UZ fractures; retardation
in the water-production zones and alluvium; dilution of radionuclides in the ground water from
well pumping; airborne transport of radionuclides; dilution of radionuclides in the soil; and
location and lifestyle of the critical group. This is not to say that these processes are
insignificant; rather, there were no significant issues in these areas at the time of the reviews.
Areas where the Staff had significant comments included:
Repository design
• Waste package corrosion
• Quantity and chemistry of water contacting waste packages and waste forms
Saturated zone flow and transport
Volcanic disruption of the waste packages
Quality assurance
With regard to repository design, NRC expressed concern as to whether adequate time was
available before the LA is scheduled for submittal to address all the design options under
consideration, select a reference design, develop data and models, and conduct the analyses
required to produce an LA which is complete and of high quality.
Doses received by down gradient receptors are highly sensitive to the corrosion performance of
the waste packages. The DOE is exploring several alternatives to the waste package design
proposed in the TSPA-VA, which was a 10-cm outer layer of carbon steel corrosion allowance
material and a 2-cm inner layer of Alloy 22 corrosion resistant material. It was not clear to the
7-178
-------
NRC that the DOE would be able to gather adequate long-term corrosion data in time to
definitively support the LA. The TSPA-VA relied heavily on expert elicitation rather than long-
term test data and this is a significant weakness.
The amount and chemistry of the water which contacts the waste packages is of critieal
importance not only to waste package lifetime but also to release of radionuclides once the waste
package is breached. The NRC concluded that "the range of activities outlined in the LA Plan
are unlikely to provide an adequate licensing basis for assessing the quantity and chemistry of
water contacting waste packages and waste forms. *--..v. Additional data and analysis of seepage
under both isothermal and thermal conditions will be required for a complete LA."
The NRC was not satisfied that flow and transport in the saturated zone from beneath the
repository to a receptor 20 km down gradient had been adequately characterized. Additionally,
the NRC did not concur with the DOE's view that saturated zone uncertainties were a
"moderate" contributor to receptor dose uncertainties. This descriptor was inappropriately
optimistic based on sensitivity studies conducted by both organizations. The Staff expressed
concerns that the location where ground water enters the alluvium (which delays radionuclide
migration) was not well documented. High permeability features between the repository and the
receptor could alter the flow direction away from the alluvium and confine the flow to the
fractured tuffs.
Based on Staff review, the NRC concluded that the consequences of volcanism were understated
in the TSPA-VA. The DOE assumptions on physical conditions were not representative of
basaltic volcanism at Yucca Mountain. In addition, the DOE's models did not consider the
impact of the dynamic forces produced by the volcanism on waste packages in a volcanic
conduit.
Implementation of an appropriate Quality Assurance (QA) program has been an on-going
problem. The NRC has reviewed and accepted the DOE's QA program on procedural basis.
However, audits and surveillances have identified deficiencies in implementing the program.
Some data in the technical data bases are not traceable. The NRC is concerned that the LA Plan
did not recognize these implementation deficiencies and provide for remedies.
The NRC staff provided some additional reactions to the TSPA-VA in the May 1999 Technical
Exchange (NRC99b). The TSPA-VA documentation included several features which facilitated
7-179
-------
the NRC's understanding of the DOE performance assessment. These included extensive use of
plots of intermediate outputs such as time-dependent Tc-99 release from a waste package. Plots
of the performance of sub-systems such as the number of waste packages which failed as a
function of time were also valuable as were dose rate plots which showed the mean, median, 5th,
and 95th percentiles over time. The DOE's presentation of the results of sensitivity .analyses and
the dose rates expected with alternative conceptual models also enhanced the NRC's
understanding of the TSPA-VA. On the other hand, the NRC felt that there were areas where
transparency and traceability could be improved. The NRC staff noted that the flow of key
information between the RIP computer code and external process models was difficult to trace.
The NRC also concluded that there was inadequate sampling of parameters potentially important
to repository performance and they could not determine whether correlations between sampled
parameter had been properly addressed. The Staff suggested that a table listing all important
parameters and their assigned distributions would significantly facilitate review.
The NRC felt that both agencies needed to have a better technical basis for establishing the initial
waste package failure levels. Improved linkage was required between initial defects and waste
package failure rates. This would involve consideration of the detectability of initial defects and
consideration of the expected performance of the defective waste packages. Further, with regard
to long-lived waste packages, the NRC averred that there were potential failure processes such as
stress corrosion, microbial activity and exposure to alternating wet/dry cycles which could
accelerate failure. These processes were not considered by either organization.
The NRC concluded that there were no major performance-affecting differences in the
approaches taken by the two organizations with regard to ground-water infiltration and deep
percolation. However, the modeling approaches taken for unsaturated zone flow and transport
differed markedly.
In near-field modeling the DOE did not consider that penetration of the boiling isotherm in the
drift wall could occur by water flowing down a fracture. The NRC concluded that the DOE's
assumption that water will not contact a waste package until the waste temperature drops below
the boiling point was not conservative.
The NRC observed that the TSPA-VA methods for calculating biosphere dose conversion factors
(DCF) were consistent with the NRC approach, but the Commission raised some questions as to
whether the procedures used for sampling the DCF distributions created modeling
7-180
-------
inconsistencies. The NRC also felt that the documentation on dose parameters used in the
TSPA-VA needed to be improved.
The NRC concluded that the model for igneous activity used in the TSPA-VA was inadequate.
Additional work would be required to develop acceptable models. However, based on -
discussions between the DOE and the NRC subsequent to publication of the TSPA-VA, the NRC
was of the opinion that acceptable modeling approaches can be developed before the License
Application is submitted.
7.3.4.2 Review by the TSPA Peer Review Panel
DOE created the TSPA-VA Peer Review Panel to provide the Civilian Radioactive Waste
System Management and Operating Contractor with a formal, independent review and critique of
the TSPA-VA (PRP99). In its review of the Viability Assessment, the Panel was charged with
considering both the analytical approach used and its traceability and transparency in assessing
the probable behavior of the repository. Factors evaluated in assessing the analytical approach
included:
Physical events and processes included in the assessment
Use of appropriate and relevant data
• Assumptions made
Abstraction of process models used in total system models
• Application of accepted analytical methods
Treatment of uncertainties
The Panel concluded that, due to the complexity of the system and the nature of the current or
reasonably obtainable data, it may be impossible for any technical team to develop the analytical
capabilities to prepare a credible assessment of the probable future behavior of the repository.
The long time scales which must be considered, coupled with the complexity of the geologic
setting, compound the analytical problems. The Panel suggested that dealing with these complex
coupled processes can best be handled through bounding analyses or by incorporation of
engineered features which minimize the effects of these processes.
In the Panel's words, a credible assessment "would have needed to include:
• Component subsystem models that capture important and relevant phenomena;
• Adequate databases;
7-181
-------
• Proper coupling between the subsystem models; and
• Tests of modeled behavior ".
Although the TSPA-VA offers many examples of partial, even substantial, success in each of
these four areas, the Panel has also observed examples of important deficiencies in each.
Concerning subsystem models, the final dose estimates within the TSPA-VA rest
in large part on potentially optimistic, or at least undemonstrated, assumptions
about the behavior of certain barriers in the system (for example, performance of
the cladding and the waste package).
Concerning databases, some of the important analyses are not supported by an
adequate database, (for example, databases for corrosion of spent fuel alteration
products and the saturated zone analysis).
• Concerning coupled processes (that is, thermohydrological, thermomechanical,
and thermochemical effects) and the data and models that support them, the Panel
believes that it may be beyond the capabilities of current analytical methodologies
to analyze systems of such scale and complexity. For this reason, the effects of
coupled processes can probably best be dealt with through a combination of
bounding analyses and engineered features designed to minimize the effects of
such processes.
• Concerning tests of modeled behavior, the TSPA-VA does not contain the
convincing direct measurements or confirmation of the modeled behavior of
components or subsystems for which testing is feasible. This testing should be
part of the analyses of such a complicated system."
The Panel concluded that the sensitivity analyses in the TSPA-VA did not provide sufficient
insights to overcome these deficiencies and uncertainties.
The Panel expressed concern over the lack of data relating to the performance of the waste
packages and reliance on instead on expert elicitation. The Panel stated that DOE must define
the environmental extremes to which the Alloy C-22 corrosion resistant liner will be exposed and
establish experimentally the critical temperature for crevice corrosion in these aggressive
environments. The need to obtain more and better data to enhance performance assessment
credibility was a repeated theme throughout the Panel's report.
The behavior of the waste packages is strongly dependent on the extent to which contact with
infiltration water seeping into the drifts is minimized. The Panel was not convinced that the
7-182
-------
TSPA-VA base case correctly captured seepage into the drifts over long periods of time. The
Panel concluded that "Better characterization of the hydrologic properties near the drifts,
improved modeling, consideration of coupled effects, and additional experimentation at the drift
scale would add confidence to the approach taken."
The Panel reviewed the impacts of five potentially disruptive processes on the Yucca Mountain
repository. The Panel concurred with DOE findings in the TSPA-VA that impacts of
earthquakes would be minor as would the impacts of volcanism on offsite groups. The Panel
also agreed with DOE's analysis that nuclear criticalitywas^iighly improbable and, if it occurred,
only modest increases in offsite doses would be expected. However, the Panel was not satisfied
with DOE's analysis of human intrusion. They stated that the scenario in which the waste
generated from an intruding borehole was driven downward into the SZ was not realistic and
analytical treatment of transport within the saturated zone was potentially non-conservative. The
particular concern with the transport model was the assumption that radioactive material was
distributed over a wide area at the top of the SZ. This would not be the case with the selected
drilling intrusion scenario. The Panel noted that a regulatory basis for analyzing human intrusion
had not been established by either NRC or EPA at the time when the TSPA-VA calculations
were made. The approach taken on the climate change in the TSPA-VA was judged to be
reasonable, in-so-far as temporal variations in precipitation are concerned. The Panel noted that
the U.S. Geological Survey disputed the manner in which the variation in precipitation was
translated into infiltration rates into the repository but the Panel took no position on that issue.
Two potentially non-conservative approaches used in the TSPA-VA were identified by the Peer
Review Panel, namely:
Long-term performance of Zircaloy cladding on spent fuel
• Buildup of radionuclides in soil irrigated with contaminated groundwater
With regard to cladding performance, the Panel stated that additional failure mechanisms
including (1) pitting and crevice corrosion, (2) hydride-induced embrittlement and cracking, and
(3) unzipping of the cladding due to secondary phase formation when the UO2 fuel is converted
to various alteration products in a moist, oxidizing environment all need to be experimentally
investigated. Until such work is completed and the expected cladding longevity can be
substantiated, the TSPA-VA assumptions about the ability of the cladding to act as a significant
barrier are not defensible.
7-183
-------
The Panel observed that irrigation water was assumed in the TSPA-VA to be deposited on the
soil for only one year prior to intake by the receptor via various soil-related pathways. In reality
irrigation can continue for thousands of years and an equilibrium concentration for each nuclide
will be established in the soil which is higher than that based on only a one-year exposure period.
In addition, the assumption that iodine is rapidly washed through a soil column is not supported
by field observations which show considerable holdup in the surface layers.
The Panel also identified three factors which were believed to treated with significant
conservatism in the TSPA-VA including:
Transport through penetrations in the waste package
Retention of radionuclides in spent fuel alteration products
• Potential sorption of technetium and iodine in the UZ and SZ
The Panel felt that the modeling of the transport of radionuclides from failed waste packages
through pits, cracks or crevices was not realistic since no significant retardation was included.
Since this assumption is not consistent with expected physical reality, better methods are
required to analyze the movement of radionuclides within and from the failed waste packages.
Any UO2 in spent fiiel packages which is exposed to moist air is expected to be converted to
secondary uranium minerals such as schoepite within a few hundred years after waste package
and cladding failure. It has been experimentally established that neptunium would be
incorporated into the alteration products and, consequently, Np release would be controlled by
the dissolution rate of these alteration products. While this process was not included in the
TSPA-VA base case, it was cursorily examined in a sensitivity analysis (DOE98, Volume 3,
Section 5.5.3). No impact was shown over the first 10,000 years or after about 700,000 years
because releases are dominated by other nuclides for those time periods. However, at 100,000
years, the dose rate is reduced by about a factor of 10 when solubility of Np from the alteration
products is considered.
No sorption of technicium or iodine (the major contributors to dose over the first 10,000 years)
on geologic materials was considered in the TSPA-VA. However, the Peer Review Panel cited
field observations, such as those of Straume et al, (STR96), taken near the site of the Chernobyl
nuclear power plant accident suggesting that radioiodine may be retarded in soil surface layers.
The Panel did not cite any instances where technetium was retarded but suggested that the issue
should be reviewed on the basis, for example, of measurements near the Chernobyl site.
7-134
-------
to addition to these general conclusions, the Panel provided detailed comments on all of the
component models used in the TSPA-VA including the UZ flow, thermohydrology, near-field
geochemical environment, waste package degradation, fuel cladding as a barrier, waste form
degradation, radionuclide mobilization, UZ ^transport, SZ flow and transport, biosphere, and
disruptive events. Recurring themes were the need for additional data and improved models to
produce a credible and defensible LA.
7.3.4.3 Review by the U.S. Nuclear Waste Technical Review Board
The Nuclear Waste Technical Review Board (NWTRB; see Section 4.4 of this BID) also
critiqued the TSPA-VA (TRB99). The Board stated that they had identified no features or
processes which would disqualify the Yucca Mountain site but felt that DOE should give serious
attention to replacing the high-temperature design evaluated in the TSPA-VA with a ventilated
low-temperature design where waste package surface temperatures were maintained below the
boiling point of water. Such a change should significantly reduce the uncertainties involved in
attempting to analyze complex coupled thermal-hydraulic and thermal-mechanical, and thermal-
geochemical interactions within the repository.
The NWTRB also expressed concerns as to whether the amount of work required to support a
technically defensible decision on Yucca Mountain could be completed on DOE's proposed
schedule which calls for a site recommendation decision by 2001. This is a matter of particular
concern, since the Board stated that expert elicitation should not be used as substitute for data
gathering at the site or in the laboratory. Areas where additional factual input is required include
waste package performance (e.g., resistance to stress-corrosion cracking), and the magnitude and
distribution of seepage into the repository.
The Board also stressed the need for long-term scientific studies assuming the site is ultimately
found to be suitable and construction is approved. These scientific studies should include
selected aspects of both natural and engineered barriers.
to summary, the Board agreed with DOE "that Yucca Mountain continues to merit study as the
candidate site for a permanent geologic repository and that work should proceed to support a
decision on whether to recommend the site to the President for development. ... The Board
supports continuing focused studies of both natural and engineered barriers at Yucca Mountain to
7-185
-------
attain a defense-in-depth repository design and to increase confidence in predictions of repository
performance."
7.3.5 NRC Total SystemJPerformance Assessments
7.3.5.1 Background
To support it licensing responsibilities, the NRC is developing the capability to review DOE's
TSPA in support of a License Application, if the Yucca Mountain site is found to be suitable for
disposal. The Commission staff, like DOE, is iteratively developing TSPA modeling capability
based on evolving information and insights concerning factors that affect repository system
performance. Development of the TSPA methodology is independent of DOE's effort, and the
DOE and NRC TSPA models and codes differ in detail.
The NRC's strategic planning calls for early identification and resolution, at the staff level, of
TSPA issues before receipt of an LA, if the Yucca Mountain site is found to be suitable for
disposal. The principal means for achieving this goal is on-going, informal, pre-licensing
consultation in which performance issues are identified and discussed, and issue resolution is
sought. Resolution of issues is sought at the staff level before formal licensing reviews, but
issues may be raised and considered again in the licensing process.
To implement its goals, the NRC has focused its pre-licensing work on issues most critical to the
post-closure performance of the proposed repository; these have been designated as Key
Technical Issues (KTI). To facilitate dialog with DOE concerning resolution of the KTIs, the
NRC has established Issue Resolution Status Reports (IRSR) to serve as the primary mechanism
through which feedback to DOE concerning KTIs and KTI subissues will be expressed and
documented. The IRSRs address acceptance criteria for issue resolution and the status of
resolution. Updating revisions of the IRSRs will be issued periodically as progress is made in
resolution of the KTIs and their subissues.
One of the Key Technical Issues identified and discussed in an IRSR is Total System
Performance Assessment and Integration (TSPAI). The NRC has, to date, issued the original
version of the IRSR on this topic in April 1998 and Revision 1 in November 1998 (NRC98). As
basis for its review of the DOE TSPA and development of its own TSPA methodology, the staff
has adopted the hierarchical structure of performance assessment factors shown in Figure 7-41-
7-186
-------
This performance factor structure was used to develop the NRC TSPA code structure (e.g., TSP
3.x.y) illustrated in Figure 7-42. This code structure can be compared to DOE's TSPA-VA code
structure shown in Figure 7-36.
The IRSR on total system performance assessment and integration identifies and describes the
key subissues for this topic as follows:
"Demonstration of the Overall Performance Objective. This subissue focuses on
the role of the performance assessment to demonstrate that the overall
performance objectives have been met with reasonable assurance. This subissue
includes issues related to the calculation of the expected annual dose to the
average member of the critical group and the consideration of parameter
uncertainty, alternate conceptual models, and the results of scenario analysis.
Demonstration of Multiple Barriers. This subissue focuses on the demonstration
of multiple barriers and includes: (1) identification of design features of the
engineered barrier system and natural features of the geologic setting that are
considered barriers important to waste isolation; (2) description of the capability
of barriers to isolate waste; and (3) identification of degradation, deterioration, or
alteration processes of engineered barriers that would adversely affect the
performance of natural barriers.
Model Abstraction. This subissue focuses on the information and technical needs
related to the development of abstracted models for TSPA. Specifically, the
following aspects of model abstraction are addressed under this subissue: (i) data
used in development of conceptual approaches or process-level models that are
the basis for abstraction in a TSPA, (ii) resulting abstracted models used to
perform the TSPA, and (iii) overall performance of the repository system as
estimated in the TSPA. In particular, this subissue addresses the need to
incorporate numerous features, events, and processes into the performance
assessment and the integration of those factors to ensure a comprehensive analysis
of the total system.
7-187
-------
TOTAL
SYSTEM
REPOSITORY )
PERFORMANCE|
v Individual .
Dose or Risk) )
SUBSYSTEMS
(Intermediate calculations
of key contributors to
system-level performance)
COMPONENTS
OF SUBSYSTEM
ENGINEERED
SYSTEM
GEOSPHERE
BIOSPHERE
Engineered
Barriers
UZ
Flow and
Transport
SZ
Flow and
Transport
: Direct
Release
and
Transport^
Dose
Calculation]
^ J
KEY ELEMENTS
OF SUBSYSTEM
ABSTRACTIONS
'wP cor/oilon
(temperature, humidity
.and ctiemlalry)
/"meehtnlcit
I of WPi
I tovlllnp. roekfill
V«lk« Intruilon)
/quantity and chemistry^
of water contacting WPs
\artd waste forms )
radlonuclide
release rates and
solubility limits
I spalfal and temporal
I distribution or fJow
;distribution of ]
mass flux between
fracture and matrixy
/relardatfon In
1 fracluros In the
y^unsaturated zone
(low rates In
wfller-produclion
: retardation in
water-production
zones and alluvium
volcanic disruption j
of wasis packages 1
aifborna
transport of
radio nuc)id«s
(*
r
^
dilution of radio-
nuclides 'in ground-
water (well pumping)
/dilution of ^
radionuclldas in soil
'vjsuftacft processes)^
s -N
location and lifestyle]
Of critical group I
-41. Structure of Performance Factors for "NRC Performance Assessments (NRC98)
-------
Seismic/Rock Fall Impacts
onWT>
Fault DispUcemenc Impacts
onWP(FAULTO)
Igneous Event Impacts
on WP (VOLCANO)
o
1
Percolation Flux in UZ
(UZFLOW)
Near-Field Flow
and Chemistry (NFENV)
S
n
.0
EBS Public
(EBSFA1L)
EBS Release
(EBSREL)
Unsani
id Zone Flow
ind Transport (LEFT)
Sanvated Zone Flow
and Transport (SZFT)
Dose Cottvcnioii tor
Grouodwater (DCAGW)
AsfaOispcaal
(ASHFUMO)
(ASHRMOVO)
Dote OaoveniaB for
GcaaadSiB&oe(DCACS)
Figure 7-42. Structure of NRC Computer Codes for Performance Assessments (NRC98)
7-189
-------
Scenario Analysis. This subissue considers the process of identifying possible
processes and events that could affect repository performance; assigning
probabilities to categories of events and processes; and the exclusion of processes
and events from the performance assessment. This is a key factor in assuring the
completeness of a TSPA.
Transparency and Traceability of the Analysis. This subissue emphasizes staff
expectation of the contents of DOE's TSPA to support an LA. Specifically, it
focuses on those aspects of the TSPA that will allow for an independent analysis
of the results."
Details of acceptance criteria and review methods for the subissues related to demonstration of
overall performance and demonstration of multiple barriers will be provided in the next revision
of the IRSR for TSPAL Details of criteria and review methods for model abstraction, scenario
analysis, and transparency are included in NRC98.
7.3.5.2 NRC Development and Use of TSPA Models
The content and characteristics of NRC's TSPA models have, like DOE's, evolved over time as
information and insights as basis for the models have developed. Current models, also like
DOE's, are considered to be a snapshot in time from an on-going model-development process.
Under its Iterative Performance Assessment (IPA) program, NRC has adopted a phased approach
to its TSPA modeling capability. Phase 1 used relatively simplistic models and was designed
primarily to demonstrate capability to perform TSPA reviews as part of the licensing reviews.
Phase 2 used significantly enhanced modeling methods to identify and assess factors of primary
importance to repository system performance. Phase 3, which is still underway, uses more
general and versatile computer codes to perform TSPA evaluations analogous to those performed
by DOE.
Three versions of the Total-system Performance Assessment (TPA) code have been developed in
Phase 3 of the IPA program. TPA 3.1.3 has been used to calculate mean doses for alternate
conceptual models, and TPA 3.1.4 has been used for system-level sensitivity and uncertainty
studies. The most recent version of the TPA code, 3.2, was used to provide feedback to DOE on
the results of NRC's review of the TSPA (see Sections 7.3.2. and 7.3.3).
7-190
-------
The most recently documented description of the NRC TPA codes is provided in NUREG 1668,
which describes the characteristics and use of the 3.1.3 and 3.1.4 codes to perform sensitivity and
uncertainty analyses for a proposed repository at Yucca Mountain (NRC99a). Characteristics of
the TPA 3.2 code have not yet been documented, but results of its use were presented and
discussed at the May 1999 DOE/NRC Technical Exchange (NRC99b) in which NRC staff
provided feedback to DOE concerning results of their review of the TSPA-VA.
The TPA 3.1 and 3.2 codes have capability and flexibility comparable to those of the DOE codes
for the TSPA-VA. As previously noted, the DOE and NRC codes differsignificantly in detail,
but both have capability to evaluate performance for alternative repository design features.
natural system features, and disruptive scenarios, at a level of detail and characterization of
uncertainty commensurate with the available information base. At present, the principal
difference between the NRC and DOE performance assessment codes is that the NRC codes give
considerable attention to disruptive events associated with seismicity and volcanism, while the
DOE approach considers these phenomena to be unlikely to occur in ways that could affect
repository performance. These differences are expected to be resolved as part of the issue
resolution process.
Principal features of the NRC's Phase 3 performance assessment codes include the following:
• Water infiltration into the subsurface. Calculation of percolation flux takes into
account the time history of climate change, variation of shallow infiltration with
climate change, and the areal-average percolation flux at the repository horizon.
Near-field environment. The near-field environment, which affects the waste
package corrosion rate, is characterized in terms of drift wall and waste package
surface temperatures, relative humidity, water chemistry, and water reflux during
the thermal pulse phase.
Waste package depradatjnp and EBS release. Waste package failures depend on
near-field conditions, corrosion mechanisms and rates, and mechanical effects
such as rockfall. Radionuclide release from the EBS is calculated in terms of rate
of release from the waste form, solubility limits, and transport mechanisms out of
the EBS. No credit is taken in the base case for cladding performance as a barrier.
Transport in the ITZ and SZ. Time-dependent flow velocities in the UZ are
calculated using the hydrologic properties of the major hydrostratigraphic units.
Matrix and fracture flow are modeled, Radionuclide retardation on fracture
surfaces is assumed not to occur, but sorption in the rock matrix is modeled. The
7-191
-------
conceptual hydrologic model for flow in the SZ assumes fracture flow in the tuff
aquifer and matrix flow in the alluvial aquifer.
• Airborne transport for direct releases. NRC performance assessments include
consideration of airborne releases from low-probability intrusive igneous events
which cause direct release of waste package-materials into the air. Factors
considered include number of packages failed and quantities of radionuclides
released, ash deposition patterns, and degradation of deposited, contaminated ash.
Biosphere dose exposure scenarios. Dose evaluations are done for the average
person in a designated receptor group. Two types of groups are considered: a
farming community 20 km downgradient from the repository, and a residential
community. The farming community is assumed to use contaminated ground
water for drinking and agriculture; the residential community uses it only for
drinking. Dilution of radionuclide concentrations in the ground water as a result
of pumping is considered.
NUREG-1668 (NRC99a) reports the results of dose evaluations in which the base case TSP 3.1.4
model and 11 alternative conceptual models (such as including cladding credit) were used to
calculate doses at 10,000 and 50,000 years for a receptor 20 km from the repository. The
repository system conceptual design was similar to that used by the DOE in the TSPA-VA, but
the corrosion-resistant inner package barrier was assumed to be Alloy 625. The annual base case
mean peak total effective dose equivalent (TEDE) was projected to be 2.3 mrem at 10,000 years.
Annual results for the alternative conceptual models ranged from a low of 0.012 mrem when
cladding credit was taken to a high of 12.5 mrem when no radionuclide retardation was assumed.
The range of results is shown as a bar chart in Figure 7-43.
As previously noted, the NRC presented its more recent TSP 3.2 results evaluations at the
DOE/NRC Technical Exchange in May 1999 (NRC99b). Results presented for the ground-water
dose using the NRC's mean-values data set are shown in Figure 7-44, for 10,000 and 100,000-
year dose rates. As can be seen, the 10,000-year dose rate is forecasted to be about 0.002
mrem/yr, and the 100,000-year dose is about 0.2 mrem/yr. These results can be compared to
DOE's TSPA-VA results, which indicated a 10,000-year dose rate of 0.04 mrem/yr and a
100,000-year dose rate of about 5 mrem/yr (see Figures 7-37 and 7-38). Reasons for differences
in the NRC and DOE results are not readily apparent because parameter values and modeling
approaches used by the two agencies differed markedly. For example, the DOE assumed
cladding credit while the NRC did not; the NRC assumed an average of 32 juvenile waste
package failures while the DOE assumed one; the DOE used three-dimensional modeling of UZ
below the repository which suggested significant lateral diversion while the NRC used one
7-192
-------
Clad Cladding protection factor - 0,5%
C22base Alloy C-22
C22mult Alloy C-22 with less flow diversion
Natan Natural analog release
Flwthru Flowthrough model with 1/100 flow
Bkfill Backfill emplaced at 100 yr
Base Alloy 625, no backfill, no matrix diffusion,
bathtub model, no cladding
Focflow Flow focused to 1/4 the WP
Flwthru2 Flowthrough with carbonate dissolution
Totdef WPs fail at time - 0
NoRd No contaminant retardation in geosphere
Matdif Matrix diffusion in geosphere
8
Mean Annual Peak Dose (mrem)
Figure 7-43 NRC TSPA Results for Alternative Conceptual Models (NRC99a)
7-193
-------
,
10
icr'
10'a
8
Q
iff4
10"
2000
Outputs Using Mean-values Data Set
—T—
Total Dose
• ->'-Subarea1
—o-subarea2
Subarea 3
—'—Subarea 4
--*--Subarea 5
- - Subdrea 6
*— Subarea 7
4000 6000
Time (yr)
10000
^tr-!
10-'
iov
o
vt
-i—i—i—r
Total Dose
Subarea 1
—e—Subarea 2
—*•— Subarea 3
—i—Subarea 4
--»--Subarea 5
-A—Subarea S
• •(—Subarea 7
i- • i-
20000 40000 60000
Time (yr)
80
100000
10kyr
100kyr
Figure 7-44. NRC TSPA Results for Mean-Values Data Set (NRC99b)
-------
dimensional modeling with seven stream tubes and no lateral diversion. In addition, the NRC
assumed dilution during pumping of contaminated ground water by thedose receptor, while DOE
assumed this dilution did not occur.
7.3.5.3 Conservatism In The NRC Performance Assessments^
As noted in Section 7.3.5.2, the NRC staff are independently developing performance assessment
capability in order to be able to be able to perform comprehensive reviews of DOE's TSPA in
the License Application. The NRC performance assessment capabilities and methods are, like
DOE's, continuing to evolve. Documentation of NRC's parameter values, models and
assumptions are not yet as comprehensive as DOE's; the most recent description of the NRC
models and the results of their use was provided in the NRC/DOE Technical Exchange of May
27-29, 1999 (NRC99b). As reported during the Exchange, NRC's base-case performance
evaluations using VA design parameters projected a 10,000-year dose rate of about 0.003
mrem/yr; DOE's base-case 10,000-year dose rate projection was 0-04 mrem/yr. Conservatisms
in NRC's performance parameters, models, and assumptions, as indicated by information
Provided at the Technical Exchange, are summarized below.
Performance Parameters
NRC presentations at the May 1999 Technical Exchange indicated that "mean values" of the
Performance parameters were used for the base case performance assessments. Values of some
of the parameters were presented, but comparisons with DOE are difficult because of differences
in modeling approaches and parameters used. In general, NRC's use of "mean values" appears
to correspond in concept to DOE's use of "expected values". Values of parameters used by NRC
for precipitation and infiltration were, for example, similar to those used by DOE.
Performance Models
Key features of NRC's performance assessment modeling approach that are indicative of
conservatism include the following:
• Impacts of igneous events, seismic rock falls, and fault displacements on waste
packages were included in the models. Seismicity impacts were included in the
base case evaluations; volcanism and faulting impacts were treated separately.
7-195
-------
• No credit was taken for spent fuel cladding as an engineered barrier. Half of the
spent fuel in a failed waste package was assumed to be exposed, wetted, and a
source for release of radionuclides.
Transport of radionuclides in the unsaturated zone from the repository to the water
table was assumed to occur vertically, with no effeet-of matrix diffusion or
sorption on fracture surfaces. This assumption is similar to that made by DOE in
theTSPA-VA.
• Radionuclide transport in the saturated zone was assumed to occur via four
pathways through fractured tuff and alluvium. Transport in the tuff occurred only
via fractures, with flow rates between 50 and 500 m/yr. Flow velocities in the
alluvium were assumed to be between 3 and 5 m/yr, and radionuclide retardation
was assumed to occur.
Dilution of radionuclide concentrations in ground water as a result of pumping by
the dose receptor was assumed to occur (the dilution factor was not stated). This
is a non-conservative modeling feature in contrast with DOE's assumption that
such dilution does not occur.
Conservative Assumptions
Conservative assumptions in the NRC performance assessments described at the May 1999
Technical Exchange (NRC99b) included the following:
• Thirty-two waste packages were assumed to be defective at the time of
emplacement. Rates and mechanisms of degradation and radionuclide release for
these and other packages that fail were not described, however.
• The mean value of the localized corrosion rate for the Alloy 22 corrosion resistant
material in the waste package was stated to be 2.5 E-4 m/yr. This is a factor of
100 higher than experimental values cited in EPRI's IMARC-4 report (EPR98)
and in DOE's VA Technical Support Document (DOE98a).
Detailed comparison of NRC and DOE performance assessment conservatisms is not possible
because the modeling approaches and parameters used differ significantly. In general, it appears
that, in comparison with DOE, NRC's approach produces a larger radionuclide source term (e.g.,
as a result of assuming no cladding credit), but compensates for it by assuming that dilution
occurs during pumping. The net result is that the results of NRC's performance assessments
reported at the May 1999 NRC/DOE Technical Exchange agree with DOE's TSPA-VA results
within an order of magnitude.
7-196
-------
7-3.6 EPRI Total System Performance Assessments
7.3.6.1 Background
The nuclear power utilities have for many years maintained oversight of the OCRWM program
in DOE because of their contracts with the Department concerning its responsibilities for receipt
and disposal of commercial spent fuel. Technical contributions to the oversight are provided by
EPRI in programs that are selected and guided by the utilities. EPRI maintains peer capability to
review and comment on DOE's program activities and to independently perform performance
assessments and other analyses of the type done by the Department within the OCRWM
Program.
EPRI has performed independent total system performance assessments in parallel with DOE's
efforts. A report on EPRI's TSPA concepts and methods was first issued in 1990 (EPR90), and
TSPA reports were subsequently issued in 1992, 1996, and 1998 (EPR92, 96, 98). The EPRI
studies have kept pace with the DOE efforts, making use of the evolving repository design
concepts, data bases, and modeling methods. The EPRI Phase 4 report, issued in November
!998 (EPR98) parallels the DOE's TSPA-VA report (DOE98) and uses the VA design.
The overall goal of the EPRI assessments is to provide an "...independent assessment of the
Performance of the potential repository site, identifying fatal flaws in the site itself, in the
engineering design, or in the licensing program, so that the decision makers in the utility industry
can judge the likelihood of potential outcomes of the licensing process and take appropriate
action" (EPR96).
7-3.6.2 EPRI's TSPA Technical Approach
EPRI uses a logic tree approach to performance assessment modeling. The EPRI TSPA code is
termed the Integrated Multiple Assumptions and Release Calculations code (IMARC), The logic
tree approach, illustrated in Figure 7-45, represents uncertain inputs to the TSPA calculations as
nodes in a tree, with branches from a node indicating alternative models or parameter values for
that input and the weight associated with that model or parameter value. In contrast, the DOE
TSPA code structure (Section 7.3.2.2) and the NRC approach (Section 7.3.4) use a central
processor (e.g., the RIP code for DOE), which is fed information from codes for the various
repository performance factors.
7-197
-------
CLIMATE
SCENARIO
FOCUSED SOLUBILITY
INFILTRATION FLOW AND ALTERATION RETARDATION
FACTOR TIME
Figure 7-45. EPRTs IMARC Logic Tree (EPR98)
All TSPA methods include models for essentially the same performance factors, e.g., climate,
infiltration, waste package performance, etc. They differ, however, in the details of how they
model the performance factors and in their assignment of values for uncertain performance
parameters. For example, the DOE assumed three climate conditions for the TSPA-VA, with
precipitation spanning the range 170 to 540 mm/yr; in contrast, the EPRI interpreted the historic
climate data to indicate two future climate conditions, with precipitation spanning the range 150
to 220 mm/yr.
Other key features of the EPRI Phase 4 TSPA modeling approach are outlined below. As for
DOE, details of models and parameter characterization have evolved in accord with evolution of
the data bases for performance assessment. Because the modeling approaches used in IMARC-4
were similar to those used in IMARC-3, the IMARC-4 report (EPR98) did not repeat technical
details of modeling that were discussed in EPR96.
Climate
As indicated above, EPRI's interpretation of available data concerning past and possible future
climate conditions led to an estimate that the long-term average precipitation should be between
150 and 220 mm/yr, a much narrower range than used by DOE in the TSPA-VA. EPRI believes
7-198
-------
the DOE precipitation values are based on ostracode species assemblages found in Minnesota
and Washington, rather than on specific plant taxa calibrated near Yucca Mountain.
Infiltration
The basic IMARC net infiltration model is a one-dimensional finite difference code that
incorporates source and sink terms for surface infiltration, uptake of water by plants, and
drainage from the root zone to the deep subsurface (which is net infiltration). For Phase 4, the
runoff features of the model were revised as a result of recent data. As a result, the~net -
infiltration for current climate conditions increased from the Phase 3 (1996) value of 1.2 rnrn/yr
to 7.2 mm/yr. The full glacial climate value increased from 2.9 to 19.6 mm/yr. (DOE's TSPA-
VA values showed similar increases in comparison with TSPA-95 values.) The TSPA-VA
results are higher than the Phase 4 results because the DOE assumed a precipitation rate of 300
mm/yr as compared to EPRTsassumption of 195 mm/yr for a full glacial climate.
Near Field Conditions
For IMARC-4, EPRI developed a model and analytic solution which describes heat transfer and
fluid flow in the near field in terms of a uniform disk-shaped heat source located in a moist,
unsaturated, porous medium. Large-scale convective gas flow and countercurrent flow of water
and vapor were assumed to occur. Heterogeneity of the repository's geohydrologic regime was
represented by what was termed "focused flow, and "hot" and "cool" zones of the repository
were characterized. The objective of the modeling was to estimate that fraction of the waste
package inventory that is wetted; results indicated that the maximum fraction of the waste
packages that are wetted is 0.24. In contrast, DOE's expected values in the TSPA-VA for waste
packages with seeps were about 0.5 during superpluvial conditions and about 0.33 during the
extended periods associated with long-term average climate (DOE98, Volume 3, Figure 4-3).
Waste Package Performance
The waste package performance model used in IMARC-4 differed significantly from that used in
ttvlARC-3 because of improved understanding of the repository environment and corrosion
processes, and because the reference corrosion resistant material (CRM) was changed from Alloy
825 to Alloy 22. The basis for characterizing corrosion rates was changed from Weibull
7-199
-------
distributions' to recently-obtained corrosion data and the results of DOE's expert elicitation on
waste package performance. Corrosion rates were characterized for various environmental
conditions, e.g., humid air or water dripping onto the package, and for various corrosion
mechanisms, including crevice corrosion of the Alloy 22, which is anticipated to represent the
mechanism for most-rapid penetration of the CRM. Results for the VA waste package design
(see Section 7.2) show that, in the absence of drips onto the package, penetration would not
occur for more than one million years. When drips do contact the packages, penetration by
general corrosion is predicted not to occur for about 30,000 years. Under adverse conditions, the
carbon steel outer wall could be penetrated in only 300 years, and the Alloy 22 inner wall could
be penetrated by crevice corrosion, which is conservatively assumed to occur during the time
period during which the waste package temperatures are greater than about 80°C. The EPRI
estimates that "hot" waste packages would remain above the 80°C threshold for crevice
corrosion for about 3,000 years. For "cold" waste packages this period would be reduced to
about 200 years. The EPRI notes in IMARC-4, as did the DOE in the TSPA-VA, that the data
base for estimating Alloy 22 corrosion rates is currently quite limited.
Source Term Parameters
Source term parameters discussed in IMARC-4 include radionuclide sorption, solubility, release
from the waste form, and waste form alteration. Values for these parameters were changed in
IMARC-4 in comparison with IMARC-3 because of recent data additions. The computer code
COMPASS, Version 2.0, which is a compartment model for predicting radionuclide release rates
from the engineered barrier system (BBS) into the near-field rock, was used in IMARC-4. The
Compass 2.0 code models EBS features, such as waste form, canister corrosion products,
backfill, and rock fractures, as compartments. It accounts for time-dependent cladding
degradation, modes of water contact with the waste package, and modes of water transport
through the waste package interior (overflow or through-flow).
Discussions of source term parameters in IMARC-4 addressed the following:
• New values of sorption coefficients for sorption of radionuclides on corrosion
products (principally iron oxides) were presented for cases where recent data
differ from results of a prior expert elicitation by more than a factor of five.
1 A Weibull distribution is a function used to describe the fraction of waste packages which have failed as
function of time based on mean container lifetime, threshold failure time and failure rate at the mean lifetime.
7-200
-------
Median values for the actinides are in the range 5-10 m3/kg; the median value for
Npis0.1m3/kg.
• Extensive discussion was presented on the validity of the two-orders-of-
magnitude reduction in the solubility of Np in the TSPA-VA in comparison with
TSPA-95. The EPRI analyses basically concurred with the action, which was
based on re-assessment of prior data and additions to the data base for solubility
values. The solubility of neptunium is important to prediction of doses after
10,000 years, when neptunium is the principal contributor to dose.
• Extensive discussion was provided concerning thin films surrounding spent fuel
undergoing dissolution. The EPRI concluded that the TSPA-VA approach was a
"sensible, but non-unique first step in attempting to derive more realistic
radioelement solubility constraints from laboratory tests." The EPRI
recommended additional modeling and laboratory tests to establish lower, more
realistic solubility constraints.
and Transport in the Unsaturated and Saturated Zones
The flow and transport models used in IMARC-4 were the same as those used in Phase
Values used for parameters were revised, however, as a result of recent insights concerning
conceptual modeling of the UZ and SZ and continuing integration of field and theoretical studies.
The IMARC-4 UZ hydrology model accounts for transient, variably-saturated flow and
advective-dispersive transport in a coupled dual-porosity-dual-permeability regime, from the
of the repository to the water table. Radionuclide sorption can occur both in the fractures
in the rock matrix. In the SZ, the model takes into account three-dimensional advective-
dispersive transport of the radionuclides during down-gradient migration. The SZ model can
handle matrix diffusion, radionuclide sorption and daughter-product ingrowth.
The repository footprint can be divided into subregions, each of which constitutes the top of a
UZ hydrologic column. Input variables such as infiltration rates can therefore be varied over the
fcrea of the repository. The model assumes that there is no lateral coupling between the columns
and that the system is isothermal, so that no coupling to the energy equations is needed.
Once the radionuclides reach the water table, they can. advect, disperse, sorb, diffuse into or out
°f the matrix, and decay within the three-dimensional SZ. Ground water flow in the SZ is
assumed to be representative of long-term steady-state conditions. The bulk hydraulic
7-201
-------
conductivity of the fractured rock mass is assumed to be representative of an equivalent porous
medium, which may be anisotropic.
IMARC-4 discusses the impact of recent determinations that the net infiltration rate is much
higher than originally believed, and the discovery of bomb-pulse Cl-36 at repository depths, on
conceptual modeling of the UZ. It also discusses the impact of current lack of data for the SZ on
uncertainty in the flow paths and dilution factors for the SZ. It notes that IMARC-3 asserted that
overall dilution for the SZ was about a factor often, and that this value is retained in IMARC-4
and corresponds to the base case value used by DOE in the TSPA-VA. It also discusses dilution
for a small radionuclide plume, such as would result from a single package failure, and asserts
that the dilution factor for this situation would be on the order of 100,000.
Biosphere
The EPRI's IMARC analyses use a probabilistic model to estimate radiation doses. The model
has three basic parts: probabilistic modeling of releases from the repository, characterization of
dose conversion factors for the biosphere pathways and the nuclides of interest, and
characterization of the dose receptor. In IMARC-4, EPRI used a farming critical group and the
water-only pathway for their base case. Other possible dose circumstances (e.g., all pathways)
were also evaluated. The critical group was assumed to be located 5 km from the boundary of
the repository, i.e., at the boundary for release to the accessible environment as defined by 40
CFR Part 191.
The hypothetical critical group was assumed to extract ground water from the point of highest
contamination in the contaminant plume, and to use this contaminated water for all of their food
and water needs for their entire lifetime. Dose conversion factors were based on ICRP
definitions of dose established in 1991 and on IAEA recommendations for metabolism of the
elements established in 1994.
7.3.6.3 Results of IMARC-4 Dose Evaluations
The EPRI's IMARC-4 analyses produced base case results for conditions and assumptions
outlined above, and also produced results for a wide range of sensitivity analyses. The EPRI
base case results are shown in Figure 7-46. These results were obtained assuming that 0.01% of
the waste packages had failed at emplacement (i.e., one package) and that 0.1% failed soon after
7-202
-------
emplacement (i.e., 1,000 yr). These early failures may be caused by manufacturing defects,
construction errors, or emplacement mishandling. The EPRI modeling assumes no corrosion
failures during the initial 10,000 years while the DOE modeling assumes that 17 waste packages
will fail by corrosion during this period. Thus, the EPRI assumption for total waste package
failures (juvenile plus corrosion) is 11 while the equivalent DOE assumption is 18.
The dose receptor was assumed to be an average member of a farming community located 5 km
from the repository, and the doses are the result of exposure only via the ground water pathway.
When all exposure pathways were included, the dose rate variations as a function of time were
similar to those shown in Figure 7-46, but about a factor often higher. This indicates that, for
the EPRI modeling approach for the critical group, the drinking water contribution to dose is
minor in comparison with the agricultural and other pathways.
Comparison of Figure 7-46 with the results of the DOE TSPA-VA analyses, Figure 7-39, shows
that the dose rates at various times are generally similar (e.g., DOE projects a dose rate at 10,000
years of 0.04 mrem/yr; EPRI projects 0.08 mrem/yr), and the sources of dose are similar, i.e., Tc-
99 and 1-129 are dominant in the near term and Np-237 is dominant in the long term. In the
EPRI results, Figure 7-46, the decrease in dose rate over the interval 60,000 to 100,000 years is
the result of depletion of the Tc-99 and 1-129 inventories for release from the repository.
EPRI IMARC-4 results are compared to DOE's TSPA-VA results and NRC's TSP 3.2 results in
Section 7.3.7.
7.3.6.4 Conservatism In The EPRI Performance Assessments
As indicated in Section 7.3.6.2, the EPRI approach to total system performance assessments
differs markedly from those used by DOE and NRC. As a result, direct comparison of EPRI
conservatism with that of DOE and NRC is neither possible nor appropriate. In general, the
IMARC-4 report (EPR98) suggests that EPRI seeks to be as realistic as possible in all aspects of
its assessment efforts. For example, EPR98 criticizes the DOE interpretation of data concerning
Past climates as being too conservative, observes that the assumption of an early package failure
is arbitrary, and notes that the EPRI and TSPA-VA approaches to modeling of fracture /matrix
interactions in the saturated zone differ markedly.
7-203
-------
10
Time (years)
Figure 7-46. Results of EPRI's IMARC-4 Dose Evaluations (EPR98)
In contrast to DOE's adoption of expert opinion as the basis for waste package material corrosion
rates, EPR98 includes a comprehensive effort to develop parametric models of corrosion
behavior on the basis of available data. Like NRC, the EPRI IMARC-4 analyses take no credit
for spent fuel cladding as a barrier. However, in contrast to NRC's bathtub model, EPRI uses a
flow-through model for water entry to and exit from the interior of a failed waste package. This
is similar in concept to DOE's approach, which assumed that radionuclides are instantaneously
released to the BBS from the wetted waste form.
The IMARC-4 report, EPR98, includes a discussion which compares the IMARC-4 and TSPA-
VA results. The report states:
"We observe that the magnitude of the doses estimated by IMARC Phase 4 are in
general agreement with those in the TSPA-VA (within an order of magnitude for
all time periods). This agreement can be considered quite close, given that the
models, level of abstraction, and input parameters for particular FEPs [features,
events, and processes] are considerably different between the two analyses.
Whether this is simply fortuitous or speaks to the robustness of the combined
7-204
-------
analyses is not altogether clear. It may be that one particular combination of
conservatisms (and potential non-conservatisms) in one TSPA effort were, on the
•whole, balanced by a different combination ofconservatisms/nonconservatisms in
the other TSPA analysis. There is certainly some evidence for this.
In the end, this independent comparison of TSPA approaches for the proposed
Yucca Mountain repository provides further confidence that the major FEPs
controlling the overall safety of the facility have been identified. "
7.3.7 Comparison of DOE. NRC. *H ^PRT TSPA Results for the VA Repository
Although the TSPA models, assumptions, and parameter values used by DOE, NRC and EPRI
differed greatly, each of the TSPA evaluations discussed above (DOE's TSPA-VA, NRC's TSP
3-2, and EPRI's IMARC-4) has as its basis the VA repository design concept, key features of
which are the waste package design (an outer wall of carbon steel and an inner wall of Alloy 22),
and an areal heat loading of 85 MTU/acre. Despite widely different modeling concepts, and with
only the principal design features of the repository and the existing data base as the basis for
commonality of the analyses, the results of the three TSPA efforts are quite similar, as shown in
Figure 7-47.
In Figure 7-47 the EPRI results are decreased by a factor often in comparison with the actual
results because the EPRI dose receptor was assumed to be located only 5 km from the repository.
This location, in comparison with the 20 km distance assumed by DOE and NRC, would not
have achieved the SZ radionuclide concentration reduction as a result of dilution that was
assumed for the DOE and NRC analyses. Decreasing the EPRI results by a factor of 10 therefore
Puts all results on essentially the same basis with respect to the SZ dilution factor.
The similarity of the three sets of TSPA results may be the fortuitous consequence of offsetting
assumptions. For example, DOE's TSPA-VA took credit for cladding performance as a barrier
took no credit for dilution during pumping; NRC's assumptions were the opposite of these.
Conversely, the similarity may be due to the dominant influence on results of performance
factors for which the three analyses made similar assumptions, e.g., those concerning future
climate conditions and early waste package failures. For all analyses, the dose rate results at
1 0,000 years are dominated by radionuclide releases from packages that were assumed to fail
relatively soon after repository closure, and by the highly mobile Tc-99 and 1-129 isotopes whose
7-205
-------
10*
TSPA-VA
result
2000
4000 6000
Time (yr)
10,000 yr
8000 10000
•"f
10*
0
TPA
base case
20000 40000 60000
Time (yr)
BOOOO 100000
100,000 yr
Figure 7-47. Comparison of DOE, NRC, and EPRI Performance Assessment Results (derived from NRC99b)
-------
arrival at the dose receptor location is not significantly affected by assumptions concerning
Phenomena along the UZ and SZ pathway.
After EPA and NRC post-closure radiation protection standards for a possible repository at
Yucca Mountain are established, opportunities for differences in assumptions concerning the
dose receptor and biosphere pathways will be narrowed. Similarly, the need for assumptions
concerning performance parameter values will be reduced by future additions to the data base.
However, alternative TSPA modeling approaches can and will be maintained.
REFERENCES
BEH96 Behl, R.J., and J.P. Kennett, Brief Interstadial Events in the Santa Barbara Basin,
NE Pacific, During the Past 60 kyr, Nature, 322:243-246, 1996.
&ER91 Berger, A., H. Gallee, and J.L. Melice, The Earth's Future Climate at the
Astronomical Timescale, Proc. Int. Workshop on Future Climate Change and
Radioactive Waste Disposal, Nirex Safety Series NSS/R257, pp. 148-165,1991.
&OR96 U.S. Bureau of Reclamation/U.S. Geological Survey, Report: Geology of the
North Ramp, Exploratory Studies Facility Stations 4+00 to 28+00, 1996.
7-207
-------
BOR96a U.S. Bureau of Reclamation/U.S. Geological Survey, Report: Geology of the
North Ramp, Exploratory Studies Facility Stations 0+60 to 4+00, 1996.
BRO75 Broecker, W.S., Climate Change: Are We on the Brink of Pronounced Global
Warming?, Science, 460-463, 1975.
BR087 Broecker, W.S., Unpleasant Surprises in the Greenhouse?, Nature, 328: 123-126,
1987.
BRO94 Broecker, W.S., Massive Iceberg Discharges as Triggers for Global Climate
Change, Nature, 372: 421-424, 1994.
DeW93 DeWispelare, A.R., et al. (eds.), Expert Elicitation of Future Climate in the
Yucca Mountain Vicinity, Report 93-016, Center for Nuclear Waste Regulatory
Analyses, San Antonio, Texas, 1993.
DOE84 U.S. Department of Energy, Draft Environmental Assessment: Yucca Mountain
Site, Nevada Research and Development Area, Nevada, December 20, 1984.
DOE88 U.S. Department of Energy, Site Characterization Plan: Yucca Mountain Site,
Nevada Research and Development Area, Nevada, DOE/RW-0199, 1988.
DOE91 U. S. Department of Energy, Technical Summary of the Performance Assessment
Calculational Exercises for 1990 (PACE-90), Volume 1: Nominal Configuration
Hydrogeologic Parameters and Calculational Results, Sandia National
Laboratories, SAND90-2726,1992.
DOE92 U.S. Department of Energy, TSPA 1991: An Initial Total-System Performance
Assessment for Yucca Mountain, Sandia National Laboratories, SAND91-2795,
1992.
DO£94a U.S. Department of Energy, Total System Performance Assessment for Yucca
Mountain - SNL Second Iteration (TSPA-1993), SAND93-2675, April 1994.
7-208
-------
DOE94b U.S. Department of Energy, Total System Performance Assessment -1993: An
Evaluation of the Potential Yucca Mountain Repository, Intera, Inc., B 00000000-
01717-2200-00099, Revision 01, March 1994
DOE94c U.S. Department of Energy, Calculations Supporting Evaluation of Potential
Environmental Standards for Yucca Mountain, B-0000000-01717-2200-00094,
Revision 01, April 1994.
DOE94e U.S. Department of Energy, Seismic Design Inputs for the Exploratory Studies
Facility at Yucca Mountain, BABOOOOOO-01717-5705-00001, Revision 00,
Management and Operating Contractor, April 29, 1994.
E>OE95a U. S. Department of Energy, License Application Annotated Outline, Predecisional
Preliminary Draft, YMP/94-05, Revision 01, Chapter 3, December 21,1995.
E>OE95b U.S. Department of Energy, Total System Performance Assessment -1995: An
Evaluation of the Potential Yucca Mountain Repository, TRW Environmental
Safety Systems, Inc., BOOOOOOO-01717-2200-0013 6, Revision 01, November
1995.
DOE95c U.S. Department of Energy, Stochastic Hydrogeologic Units and Hydrogeologic
Properties Development for Total System Performance Assessments, Sandia
National Laboratories, SAND94-0244, Albuquerque, New Mexico, 1995.
E>OE95e U.S. Department of Energy, Engineered Barrier Design Requirements Document,
Office of Civilian Radioactive Waste Management, YMP/CM-0024, Revision 01,
Las Vegas, Nevada, October 1995.
DOE95f U.S. Department of Energy, Predecisional Preliminary Draft - Responses to
Questions from the Environmental Protection Agency Concerning Water
Resources and the Hydrologic Regime in the Yucca Mountain Vicinity, December
7, 1995.
E>OE95g U.S. Department of Energy, Site Atlas 1995, Yucca Mountain Site
Characterization Project, U.S. DOE Remote Sensing Laboratory, July 1995.
7-209
-------
DOE95k U.S. Department of Energy, Environmental Management Programmatic
Environmental Impact Statement: Appendix E, Nevada Test Site, 1995.
DOE96a U.S. Department of Energy, Letter Report: Results of Hydraulic & Conservative
Tracer Testing at C-Wells Complex, WBS 1.2.3.3.1.3.1, 3GWF660M, August
1996.
DOE96b U.S. Department of Energy, Results of Reactive-Tracer Testing at the C-Wells,
WBS 1.2.3.3.1.3.1, 4270M, August 1996.
DOE96c U.S. Department of Energy, Saturated Zone Radionuclide Transport Model,
WBS 1.2.3.4.1.5.1, 3624, August 1996.
DOE96d U.S. Department of Energy, Presentation to Nuclear Waste Technical Review
Board, October 9, 1996.
DOE96e U.S. Department of Energy, Study Plan for Study 8.3.1.8.1. L, Probability of
Magmatic Disruption of the Repository, Revision 3, Washington, DC, U.S.
Government Printing Office.
DOE96f U.S. Department of Energy, Probabilistic Volcanic Hazards Analysis for Yucca
Mountain, Nevada, BAOOOOOOO-01717-2200-00082 Rev. 0, Civilian Radioactive
Waste Management System, Management and Operating Contractor, June 1996.
DOE97a U.S. Department of Energy, Overview ofTSPA-VA Plan, Eric Smistad,
Presentation to DOE/NRC Technical Exchange Workshop on Total System
Performance Assessment, San Antonio, Texas, July 21-22, 1997.
DOE97b U.S. Department of Energy, Yucca Mountain Project Update, Susan Jones,
Presented to Nuclear Waste Technical Review Board, October 22-23,1997.
DOE97c U.S. Department of Energy, Site Characterization Progress Report: Yucca
Mountain, Nevada, DOE/RW-0498, April 1, 1996 - September 30,1996, Number
15, April 1997.
7-210
-------
DOE98 U.S. Department of Energy, Viability Assessment of A Repository at Yucca
Mountain, DOE/RW-0508, December 1998.
DOE98a CRWMS Contractor, Total System Performance Assessment - Viability
Assessment (TSPA-VA) Analyses Technical Basis Document, BOOOOOOOO-01717-
4301-00001 Rev 01, November 13, 1998.
DOE99 U.S. Department of Energy, Presentations to the Nuclear Waste Technical Review
Board Summer Meeting, June 29 and 30,1999.
DOW78 Dowding, Charles H, and Arnon Rozen, Damage to Rock Tunnels from
Earthquake Shaking, Journal of Geotechnical Engineering Division - ASCE,
v.104, pp. 175-191,1978.
DRI94 Desert Research Institute, Potential Hydrologic Characterization Wells in
Amargosa Valley, Lyles, B., and Mihevic, T., Publication #45129,1994.
DUD90 Dudley, W.W. Jr., Multi-Disciplinary Hydrogeological Investigation at Yucca
Mountain, Nevada: High Level Redioactive Waste Management, Volume 1,
American Nuclear Society, 1990.
EPR90 Electric Power Research Institute, Demonstration of a Risk-Based Approach to
High-level Waste Repository Evaluation, EPRINP-7057, 1990.
EPR92 Electric Power Research Institute, Demonstration of a Risk-Based Approach to
High-level Water Repository Evaluation: Phase 2, EPRI TR-10084, Palo Alto,
California, 1992.
EPR96 Electric Power Research Institute, Yucca Mountain Total System Performance
Assessment, Phase 3, EPRI-TR-107191, 3055-02, Final Report, December, 1996.
EPR98 EPRI, Alternative Approaches to Assessing the Performance and Suitability of
Yucca Mountain for Spent Fuel Disposal, TR 108732, November 1998.
7-211
-------
FAB96 Fabryka-Martin, J.5 and Wolfsberg, A., Hydrologic Flow Paths and Rates Inferred
from the Distribution ofChlorine-36 in the ESF, Nuclear Waste Technical
Review Board, My 1996.
FIE86 Fiero, B., Geology of the Great Basin, University of Nevada Press, Reno, 1986.
FLI94 Flint, A.L., and L.E. Flint, Spatial Distribution of Potential Near Surface
Moisture Flux at Yucca Mountain, Proceedings of the Fifth Annual International
Conference on High Level Radioactive Waste Management, Las Vegas, Nevada,
pp. 2352-2358, 1994.
FRI91 Fridrich, C J., D.C. Dobson, and W. W. Dudley, A Geologic Hypothesis for the
Large Hydraulic Gradient Under Yucca Mountain, EOS, TRANS AGU, 22:121,
121,1991.
FRI94 Fridrich, C.J., W.W. Dudley, Jr., and J.S. Stuckless, Hydrogeologic Analysis of
the Saturated-Zone Ground Water System, under Yucca Mountain, Nevada,
Journal ofHydrology, 15.4:133-168,1994.
GEO97 Geomatrix Consultants, Inc. and TRW, Unsaturated Zone Flow Model Expert
Elicitation Project, Las Vegas, Nevada, May, 1997.
GOO92 Goodess, C.M, J.P. Palutikof and T.D. Davies, The Nature and Causes of Climate
Change, Belhaven Press, 1992.
HO95 Ho, C.-H., Sensitivity in Volcanic Hazard Assessment for the Yucca Mountain
High-level Nuclear Waste Repository Site, Mathematical Geology, Vol. 27, pp-
239-258,1995
HO96 Ho, C.-H., A Report Summarizing the Statistical Modeling of Volcanic Risk
Studies at the Yucca Mountain Nuclear Waste Repository Site, University of
Nevada Las Vegas, submitted to NWPO, State of Nevada, December 1996.
7-212
-------
H0072 Hooke, R. LeB., Geomorphic Evidence for Late Wisconsin andHolocene
Tectonic Deformation, Death Valley, California, Geological Society of America
Bulletin, 83:2073-2098, 1972.
HOU92 Houghton, J.T., GJ. Jenkins and J.J. Ephraums, Climate Change: the IPCC
Assessment, Cambridge University Press, 1992.
HST91
HUN74
INY96
LANL96
LBL95
LBL96
Czamecki, J.B., and Wilson, W.E. Conceptual Models of the Regional Ground-
Water Flow and Planned Studies at Yucca Mountain, Nevada, Hydrological
Science and Technology, V. 7, Nos. 1-4, pp. 15-25,1991.
Hunt, C.B., National Regions of the United States and Canada, 1974.
Inyo County, CA and Esmeralda County, NV, An Evaluation Of The Hydrology
At Yucca Mountain: The Lower Carbonate Aquifer AndAmargosa River, The
Hydrodynamics Group, 1996.
Los Alamos National Laboratory, Summary Report
-------
LLNL96 Lawrence Livermore National Laboratory, Volume II: Near-Field and Altered-
Zone Environment Report, UCRL-LR-124998, August 1996.
McK96 McKague, H.L., J.A. Stamatakos, and D.A. Ferrill, Type I Faults in the Yucca
Mountain Region, CNWRA 96-007, Region 1, Center for Nuclear Waste
Regulatory Analyses, San Antonio, Texas, November 1996.
NAN89 Nevada Agency for Nuclear Projects, The Relationship of the Yucca Mountain
Repository Block to the Regional Ground Water System: A Geochemical Model,
Nuclear Waste Project Office, NWPO-TR-011-89,1989.
NAS95 National Academy of Sciences - National Research Council, Committee on
Technical Bases for Yucca Mountain Standards, Technical Bases for Yucca
Mountain Standards, National Academy Press, Washington, DC, 1995.
NDC63 Nevada Department of Conservation and Natural Resources, Geology and Ground
Water ofAmargosa Desert, Nevada-California, Water Resources-Reconnaissance
Series Report 14, 1963.
NDC70 Nevada Department of Conservation and Natural Resources, Regional Ground
Water System in the Nevada Test Site Area, Nye, Lincoln, and Clark Counties,
Nevada, Reconnaissance Series Report 54,1970.
NEV85 State of Nevada, Comments on the U.S. Department of Energy Draft
Environmental Assessment for the Proposed High-Level Nuclear Waste Site at
Yucca Mountain, March 1985.
NRC96 U.S. Nuclear Regulatory Commission, Presentation to the Advisory Committee on
Nuclear Waste Concerning Duration of the Regulatory Period, T. McCartin,
March 27, 1996.
NRC97a U.S. Nuclear Regulatory Commission, NRC High-Level Radioactive Waste
Program Annual Progress Report Fiscal Year 1996, NUREG/CR-6S13, No. 1,
Budhi Sagar (editor), 1997.
7-214
-------
NRC97b U.S. Nuclear Regulatory Commission, CNWRA Investigations ofYMR Geologic
Setting Relevant to Igneous Activity, Dr. Brittain Hill, presentation at DOE/NRC
Technical Exchange - Igneous Activity Program, February 25-26,1997.
NRC97c U.S. Nuclear Regulatory Commission, Overview ofNRC's TSPA Methodology,
Presentation by R.G. Baca, CNWRA, to DOE/NRC Technical Exchange on Total
System Performance Assessment, San Antonio, Texas, July 21-22,1997.
NRC98 U.S. Nuclear Regulatory Commission, Issue Resolution Status Report, Key
Technical Issue: Total System Performance Assessment and Integration, Revision
1, November 1998.
NRC99a U.S. Nuclear Regulatory Commission, NUREG-1668, Vol.2, NRC Sensitivity and
Uncertainty Analyses for a Proposed HLW Repository at Yucca Mountain,
Nevada, Using TPA 3.1, March 1999.
NRC99b U.S. Nuclear Regulatory Commission, DOE/NRC Technical Exchange of Total
System Performance Assessments for Yucca Mountain, May 25-27, 1999.
NRC99c U.S. Nuclear Regulatory Commission, Staff Review of the U.S. Department of
Energy Viability Assessment for a High-Level Radioactive Waste Repository at
Yucca Mountain, Nevada, SEC-99-074, Letter from William D. Travers,
Executive Director for Operations, to the Commissioners, March 11,1999.
PRP99 TSPA-VA Peer Review Panel, Peer Review of the Total System Performance
Assessment- Viability Assessment, Final Report, February 1999.
ROB96 Robinson, B.A., A.V.Wolfsberg, H.S. Viswanathan, C.W. Gable, G.A. Zyvoloski,
and H.J. Turin, Site-Scale Unsaturated Zone Flow and Transport Model-
Modeling of Flow; Radionuclide Migration, and Environmental Isotope
Distributions at Yucca Mountain, 1996.
7-215
-------
SAV94 Savard, C.S., Groundwater Recharge in Forty Mile Wash Near Yucca Mountain,
Nevada, 1992-1993, High Level Radioactive Waste Management, Proceedings of
the 5th Annual International Conference, 4: 1805-1813, 1994.
SCO90 Scott, R.B., Tectonic Seating of Yucca Mountain, Southwest Nevada, in: Basin
and Range Extension TectonicsNear the Latitude of Las Vegas, Nevada, Chapter
12, Geologic Society Memoir 176, Boulder, Colorado, 1990.
SIN89 Sinton, P.O., Characterization of the Large Hydraulic Gradient Beneath the
North End of Yucca Mountain, Nevada, EOS, TRANS AGU, Abstract, 70
(15):321, 321, 1989.
SHO97 Short, G.J., et al., Performance Assessment for the Area 5 Radioactive Waste
Management Site at the Nevada Test Site, Nye County, Nevada, Revision 2.1,
prepared by Bechtel Nevada for U.S. Department of Energy, Nevada Operations
Office under Contract DE-AC08-96NV11718, February 1997.
SMI83 Smith, G.I., and A. Street-Perrott, Pluvial Lakes of the Western United States, in:
Late Quaternary Environments of the United States, University of Minnesota
Press, pp. 190-212,1983.
SPA83 Spaulding, W.G., E.B. Leopold, and T.R. Van Devender, Late Wisconsin
Paleoecology of the American Southwest, in: Late Quarternary Environments of
the United States, Vol. 1, University of Minnesota Press, 1983.
SPE89 Spengler, R. W., and K.F. Fox, Jr., Stratigraphic and Structural Framework of
Yucca Mountain, Nevada, Radioactive Waste Management and the Nuclear Fuel
Cycle, 13:21-36, 1989.
STE90 Stewart, J.H., Tectonics of the Walker Lane Best, Western Great Basin: Mesozoic
and Cenozoic Deformation in a Zone of Shear, Metamorphism and Crustal
Evolution of the Western United States, W.G. Ernst, editor, Vol. VII, pp. 683-713,
1990.
7-216
-------
STR96 Straume, T.,et. al., The Feasibility of Using 1-129 to Reconstruct 1-131 Deposition
from the Chernobyl Reactor Accident, Health Physics, 71(5): 733-740, 1996.
STU91 Stuckless, J.S., J.F. Whelan, and W.C. Steinkampf, Isotopic Discontinuities in
Groundwater Beneath Yucca Mountain, Nevada, American Nuclear Society,
High-Level Radioactive Waste Management, 2nd International Yucca Mountain
Conference, 2:1410-1415, La Grange Park, Illinois, 1991.
SWE96 Sweetkind, D.S., and S.C. Williams-Stroud, Characteristics of Fractures at Yucca
Mountain, Nevada: Synthesis Report, U.S. Geological Survey, Denver, Colorado.
TRB95 U.S. Nuclear Waste Technical Review Board, Report to the U.S. Congress and
the Secretary of Energy: 1994 Findings and Recommendations, March 1995.
TRB99 U.S. Nuclear Waste Technical Review Board, Moving Beyond the Yucca
Mountain Viability Assessment - A Report to the U.S. Congress and the Secretary
of Energy, April 1999.
USG75 U.S. Geological Survey, Hydrogeologic andHydrochemical Framework, South-
Central Great Basin, Nevada-California; With Special Reference to the Nevada
Test Site, Professional Paper 712-C, 1975.
USG76a U.S. Geological Survey, Summary Appraisals of the Nation's Ground- Water
Resources - Great Basin Region, Professional Paper 813-G, 1976.
USG82a U.S. Geological Survey, Two-Dimensional, Steady-State Model of Ground-Water
Flow, Nevada Test Site and Vicinity, Nevada-California, Water Resources
Investigations Report 82-4085,1982.
USG82b U.S. Geological Survey, Preliminary Interpretation of Thermal Data from the
Nevada Test Site, Open File Report 82-973,1982.
7-217
-------
USG82c U.S. Geological Survey, Two-Dimensional, Steady-State Model of Ground-Water
Flow, Nevada Test Site and Vicinity, Nevada-California, Water Resources
Investigations Report 82-4085,1982.
USG83 U.S. Geological Survey, Geohydrologic Data and Test Results from Well J-13,
Nevada Test Site, Nye County, Nevada, Water Resources Investigations Report
83-4171, 1983.
USG84a U.S. Geological Survey, Conceptual Hydrologic Model of Flow in the
Unsaturated Zone, Yucca Mountain, Nevada, Water Resources Investigations
Report 84-4345, 1984.
USG84b U.S. Geological Survey, Preliminary Geologic Map of Yucca Mountain, Nye
County, Nevada with Geologic Sections, Open-File Report 84-494, 1984.
USG84c U.S. Geological Survey, Geohydrologic Data for Test Well UE-25p#l, Yucca
Mountain Area, Nye County, Nevada, U.S. Geological Survey Open File Report
84-450, 1984.
USG84d U.S. Geological Survey, Geohydrology of Volcanic Tuff Penetrated by Test Well
WE-25b#l, Yucca Mountain, Nye County, Nevada., U.S. Geological Survey Water
Resources Investigations Report 84-4253,1984.
USG84e U.S. Geological Survey, Finite-Element Simulation of Ground Water Flow in the
Vicinity of Yucca Mountain, Nevada-California, U.S. Geological Survey Water
Resources Investigations Report 84-4349,1984.
USG84f U.S. Geological Survey, Ground Water Level Data and Preliminary
Potentiometric Surface Maps, Yucca Mountain and Vicinity, Nye County, Nevada,
Water Resources Investigations Report 84-4197, 1984.
USG84g U.S. Geological Survey, Hydrology of Yucca Mountain and Vicinity, Nevada-
California - investigative results through mid-1983. Water Resources
Investigations Report 84-4267,1984.
7-218
-------
USG85a U.S. Geological Survey, Structure of Pre-Cenozoic Rocks in the Vicinity of Yucca
Mountain, Nye County, Nevada-A Potential Nuclear-Waste Disposal Site, U.S.
Geological Survey Bulletin 1647,1985.
USG85d U.S. Geological Survey, Identification and Characterization ofHydrologic
Properties of Fractured Tuff Using Hydraulic and Tracer Tests, Test Well USW
H-4, Yucca Mountain, Nye County, Nevada, U.S. Geological Survey Water
Resources Investigations Report 85-4060,1985.
USG85e U.S. Geological Survey, Simulated Effects of Increased Recharge of the
Ground-Water Flow System of Yucca Mountain and Vicinity, Nevada-California,
Water Resources Investigations Report 84-4344,1985.
USG85f U.S. Geological Survey, Sources and Mechanisms of Recharge for Ground Water
in the West-Central Amargosq Desert, Nevada - A Geochemical Interpretation,
Professional Paper 712-F, 1985.
USG86 U.S. Geological Survey, Geohydrology of Rocks Penetrated by Test Well USW H-
6, Yucca Mountain, Nye County, Nevada, U.S. Geological Survey Water
Resources Investigations Report 86-4015,1986.
USG88a U.S. Geological Survey, Major Ground-water Flow Systems in the Great Basin
Region of Nevada, Utah, and Adjacent States, Hydrologic Investigations Atlas
694-C, 1988.
USG88b U.S. Geological Survey, Volcano-Tectonic Setting of Yucca Mountain and Crater
Flat, Southwestern Nevada, Bulletin 1790,1988.
USG88c U.S. Geological Survey, Temperature, Thermal Conductivity, and Heat Flow
Near Yucca Mountain, Nevada: Some Tectonic and Hydrogeologic Implications,
Open File Report 87-649, 1988.
7-219
-------
USG89 U.S. Geological Survey, Hydrogeologic Inferences from Drillers'Logs and from
Gravity and Resistivity Surveys in the Amargosa Desert, Southern Nevada, Open
File Report 89-234, 1989.
USG90a U.S. Geological Survey, Stratigraphic Correlation and Petrography of the
Bedded Tuffs, Yucca Mountain, Nye County, Nevada, U.S. Geological Survey
Open File Report 89-3, 1990.
USG90b U.S. Geological Survey, Geohydrology andEvapotranspiration at Franklin Lake
Playa, Inyo County, California, Open-File Report 90-356, 1990.
USG91a U.S. Geological Survey, Ground Water Conditions in Amargosa Desert, Nevada-
California, 1952-1987, U.S. Geological Survey Water Resources Investigations
Report 89-4101, 1991.
USG91b U.S. Geological Survey, Geohydrology of Rocks Penetrated by Test Well USWH-
6, Yucca Mountain, Nye County, Nevada, U.S. Geological Survey Water
Resources Investigations Report 89-4025,1991.
USG91c U.S. Geological Survey, Geohydrology of Rocks Penetrated by Test Well USW
H-5, Yucca Mountain, Nye County, Nevada, Water Resources Investigations
Report 88-4168,1991.
USG9 Id U.S. Geological Survey, Chemical Analyses of Water from Selected Wells and
Springs in the Yucca Mountain Area, Nevada and Southeastern California, U.S.
Geological Survey Open File Report 90-355,1991.
USG93a U.S. Geological Survey, Preliminary Hydrogeologic Assessment of Boreholes
UE-25c#l, UE-25c#2, and UE-25c#3, Yucca Mountain, Nye County, Nevada,
U.S. Geological Survey Water Resources Investigations Report 92-4016,1993.
USG93b U.S. Geological Survey, Major Ground-water Flow Systems in the Great Basin
Region of Nevada, Utah, and Adjacent States, Hydrologic Investigations Atlas
694-C, 1988.
7-220
-------
USG94a U.S. Geological Survey, Revised Potentiometric Surface Map, Yucca Mountain
and Vicinity, Nevada, U.S. Geological Survey Water Resources Investigations
Report 93-4000,1994.
USG94b U.S. Geological Survey, Selected Ground-Water Data for Yucca Mountain
Region, Southern Nevada and Eastern California, Through December 1992, Open
File Report 94-54, 1994.
USG95a U.S. Geological Survey, Potentiometric Surface Map, 1993 Yucca Mountain and
Vicinity, Nevada, Open File Report 95-4149, 1993.
USG96a U.S. Geological Survey, Status of Understanding of the Saturated-Zone Ground-
water Flow System at Yucca Mountain, Nevada, as of 1995, Water-Resources
Investigations Report 96-4077.
USG96b U.S. Geological Survey, Results and Interpretation of Preliminary Aquifer Test in
Boreholes UE-25c #1, UE-25c #2, and UE-25c #3, Yucca Mountain, Nevada,
Water-Resources Investigations Report 94-4177, 1996.
WIN76 Wingrad, I.J., and Pearson, F.J., Jr., Major Carbon-14 Anomaly in the Regional
Lower Carbonate Aquifer, Possible Evidence for Mega-Scale Channeling,
Southern Great Basin., Water Resources Research, 12:1125-1143,1976.
WHI93 Whitney, J.W. and C.D. Harrington, Relic Colluvial Boulder Deposits as Paleo-
Climate Indicators in the Yucca Mountain Region, Southern Nevada, Geological
Society of America Bulletin, Vol. 105, pp. 1008-1018, August 1993.
WHI96 Seismotectonic Framework and Characterization of Faulting at Yucca Mountain,
Nevada, John W. Whitney, Report Coordinator, U.S. Geological Survey, Denver,
CO, 1996.
7-221
-------
CHAPTER 8
RADIOLOGICAL PATHWAYS THROUGH THE BIOSPHERE
8.1 INTRODUCTION
In order to evaluate the performance of a disposal system at Yucca Mountain, the potential
radiation dose to members of the public must be estimated. This estimation requires identifying
the potential pathways of radionuclides from the repository to the biosphere. These pathways
include "the air, water, food and other components of the landscape that are accessible to humans
as well as the humans themselves; estimates of the concentrations that will be present in air,
water, food, and other materials with which humans might come into contact; and estimates of
the probabilities that humans will be exposed to contaminated air, water, food, or other materials
leading to a radiation dose" (NAS95).
To estimate dose, assumptions must be made concerning the location and exposure scenarios of
an individual or group of individuals who are likely to be at greatest risk from potential releases
of radionuclides from the repository after closure and removal of institutional controls. Prior to
closure, such assumptions are unnecessary because possible contamination levels can be
measured with considerable accuracy both within and outside the repository footprint. This
chapter examines the key assumptions necessary to calculate doses associated with potential
post-closure release of radioactivity from a repository at Yucca Mountain.
Figure 8-1 illustrates the major radioactivity pathways from a repository at Yucca Mountain to
humans. For the Yucca Mountain repository, the doses and risks to the critical groups for the
atmospheric pathway (nearby, and world population) are not considered to be significant relative
to the doses and risks to critical groups from ground water pathways. The existing conditions
and potential changes in geologic, hydrologic, and atmospheric (climate) conditions in the Yucca
Mountain vicinity which affect radionuclide transport are described in Chapter 7, with climatic
changes taken into consideration to develop the range of hydrological parameters included in the
radionuclide transport assessments.
During the post-closure period, the ground water will transport radionuclides released from the
repository to the surrounding area. As currently envisioned, the repository at Yucca Mountain
will be located in the unsaturated zone, approximately 400 meters (m) above the aquifer that is
within the tuff strata underlying the site. A deeper aquifer is in the carbonate rocks underlying
8-1
-------
Cirodation
CrtUcal group for
population dose
Critical group?
Figure 8-1. Schematic Illustration of the Major Pathways from a Repository
at Yucca Mountain to Humans (Ref. NAS95)
8-2
-------
the tuff. Ground water flow in both aquifers appears to be to the south-southeast from the site,
with the tuff aquifer discharging to the alluvial aquifer which is under much of the Amargosa
Valley. The deeper carbonate rock aquifer initially discharges in surface springs in the area
known as Ash Meadows in the southeast portion of the Amargosa Valley and may also discharge
in Death Valley. Based on the current understanding of the ground-water flow in the vicinity of
the proposed repository, the areas of highest potential exposures are presumed to be to the east
and south of Yucca Mountain. Figure 8-2 shows the current land use in the area surrounding
Yucca Mountain.
This chapter presents information regarding the characteristics of the Yucca Mountain area,
including past, current, and potential use of the region. Information concerning the
demographics and ecosystems of Nye County and the Amargosa Valley are included in
appendices to this report. Based on this information, four scenarios involving human use of
potentially contaminated ground water in the area surrounding Yucca Mountain are discussed.
These scenarios can be used to define the critical groups from which EPA will determine the
reasonably, maximally exposed individual (RMEI). One particular scenario, that of the
subsistence farmer, is described in detail. This chapter concludes with a consideration and
discussion of a special exposure scenario in which future generations intrude unknowingly into
the repository in their efforts to locate resources, such as minerals or water.
8.2 PAST, CURRENT, AND POTENTIAL USE OF THE YUCCA MOUNTAIN REGION
8.2.1 Past Use of the Yucca Mountain Region
This historical review of land use in the vicinity of Yucca Mountain is intended to provide
background to the definition of reasonable possible exposure scenarios for post-closure dose
assessments. These exposure scenarios are based on current land use and a reasonable
extrapolation of these trends into the future. Such extrapolations must be consistent with
historical land uses to assure that a possible exposure scenario is not overlooked simply because
an historic land use is not currently practiced.
In defining the post-closure exposure scenarios, biosphere conditions are assumed to remain as
they are today, with the exception of the predictable effects of climatic conditions. Within these
defined variations in biosphere conditions, possible land uses are extrapolated from current use
considering historic factors and other constraints that limit such uses. In establishing the
exposure scenarios, institutional impediments to use (e.g., the denial of access to the NTS) may
be disregarded, but technological impediments (e.g., the costs of well drilling or farming on steep
8-3
-------
PuMelands
CZ3 totyatM National Fonei
Ck«h VaMey Nallooal Montment
NTS
Yucca MouiUki Land Wimdrawai
"I I Patantod Lands (Private)
:'r~l DoMrt National VWdHe Rano«
;.Bga Paluta Indian Resarvattoo
S»al«Par1c '
.saate Lancia . " '
*:E2, Federal Agancy Ptotacttve Wrthdr«wah
Pow«' Withdrawals and dasslficatioos
Figure 8-2. Yucca Mountain and Surrounding Land Use (Source: DOE96)
8-4
-------
slopes) must be factored into the definition. Without this assumption of a constant level of
technology and its associated cost constraints, the range of possible land use scenarios for any
location would be unlimited.
8.2.1.1 Historic Native American Settlement and Use
In its attempts to understand the socioeconomic impacts of siting a high-level nuclear waste
repository at Yucca Mountain, the State of Nevada's Agency for Nuclear Projects/Nuclear Waste
Project Office (NWPO) has conducted a number of studies relative to Native American concerns.
These studies are listed in the references at the end of this chapter.
The proposed repository site at Yucca Mountain would be located on a border between what
were Western Shoshone and Southern Paiute lands. There were two large Native American
cultural entities whose territories once covered what is now central and southern Nevada,
southern Utah, and the adjacent areas of California (NWP095). The extensive literature on these
people suggests that the groups had lived in the area since prehistoric times and survived by
hunting and gathering food from the region. They moved about in extended family groups from
base camps established near water, fuel, and food. During the winter, the base camps were hi
and near Oasis Valley, Death Valley, Kawich Valley, Ash Meadows, Pahrump and Lower
Amargosa Valleys, Indian Springs, Las Vegas, and Moapa. From these base camps, the groups
went to Yucca Mountain to hunt game and gather a variety of plant foods. Archaeological
studies in this area have located over 400 sites in the Yucca Mountain area and its immediate
vicinity. Based upon these studies, it appears that Native Americans inhabited the land from
12,000 years ago to the immediate past, including the drainage areas at the base of Yucca
Mountain (e.g., Forty Mile Wash). Around 6,000 years ago, camp sites appeared at the higher
elevations on Yucca Mountain, including the saddles and low passes, used mainly for hunting.
2,000 years ago the settlement pattern again shifted upward to small rock shelters at the top of
steep slopes on Yucca Mountain and outlying ridges. These sites were used mainly for seed
gathering rather than hunting (NWPO90a).
Prior to the American push westward, both the Shoshone and Paiutes were divided into smaller
subgroups, ranging in size from a few families to 100 or more persons. Each of these subgroups
occupied a region with permanent water and food which generally consisted of a valley and its
adjacent mountains. Historical data suggest that several Western Shoshone and Southern Paiute
subgroups lived in the immediate vicinity of Yucca Mountain. Included were six camps in Oasis
Valley, the present site of Beatty, Nevada, as well as several camps in the Belted Range, Ash
Meadows, and the Pahrump and Lower Amargosa Valleys. Several camp groups have also been
identified in the Indian Springs/Cane Springs area to the southeast. These are shown in Figures
8-5
-------
8-3, 8-4, and 8-5 (indicated by triangles). The winter village sites were homes to which people
returned while they were engaged in hunting and gathering activities. They were also sites of
permanent residence from approximately November to May (NWPO90a).
The area used by the Oasis Valley population for subsistence extended from the Grapevine
Mountains in the west to the Sarcobatus Flat in the north, and from the Belted Range in the east
to the middle of the Amargosa Desert in the south. Yucca Mountain is included in this area, its
apparent attraction being Bighorn Sheep and seed resources (NWPO90a).
Religious Significance
The Western Shoshone and Southern Paiute people are deeply religious; their beliefs are based
upon their relationships with the land and its resources. Like other Native Americans, they
believed that the earth is a living being, along with all other natural forces.
In addition to animating the universe, power could be focused everywhere—in
beings, such as humans, plants and animals, and in springs, rocks, mountains,
caves, and other features of the natural landscape. Animal progenitors, in the
myth-time 'when animals were people,' were, along with the Earth and others,
among the most powerful beings. They were considered to be 'bosses,' 'owners,'
'masters,' 'beautiful progenitors' of present-day species. Each set the course for
its species, and at the same time, set human customs through a series of
adventures and misadventures. Particularly active in this period were Coyote
and Wolf, often portrayed as dueling brothers, but also Mountain Lion, Badger,
water beings such as Frog, raptorial birds, and a host of others. Their activities,
myth-specific, were mapped onto the landscape in a myriad of place names, often
associated with individual features of the geography such as rock formations,
specific caves or springs, petroglyph andpictograph panels, trails, washes or
arroyos, and much more. People, even today if they have been properly
instructed, cannot move about the landscape without thinking of and feeling these
links to the past. They also feel the power emanating from these specific features
as well as more generally. (NWPO90a, p. 15,16),
The most apparent sources of power are associated with caves, springs and other water sources,
and especially mountains. Although the winter habitation area shown hi the previous figures
refer to lowlands in the vicinity of springs, each is also defined with reference to mountain peaks
in the area. The mountains around Yucca Mountain are a very sacred place, along with other
peaks in Death Valley and those around Beatty.
8-6
-------
Figure 8-3. Winter Sites Near Beatty and Belted Range (Contour Intervals 1000ft.)
8-7
-------
Figure 8-4. Major Winter Sites in Ash Meadows and Pahrump Valley
(Contour Intervals 1000ft.}
8-8
-------
Figure 8-5. Major Winter Sites in Northern and Central Death Valley
(Contour Intervals 2000ft.}
8-9
-------
In summary.
According to Native Americans, the whole earth is sacred because it is the source
of life, and there are many places on the landscape where communication with the
spirits and processes of renewal can take place. Of specific importance is the
belief that landmarks cannot be moved or altered. Consequently, land-altering
activities threaten not only sacred places but concepts of the entire natural order.
This more generalized view is well expressed in Western Shoshone and Southern
Paiute concepts of a living, breathing Earth with waters flowing uninterrupted
and interconnected through it (or her, in Western Shoshone, $ogobia. 'Mother
Earth'), as well as their concepts of 'power 'free in nature and unpredictably
localized. Native American tribal religions also have rituals and ceremonies that
are involved with continuing, or constantly renewing the creation process, and
keeping proper forces (such as again. Western Shoshone and Southern Paiute
'power') in balance (Federal Agencies Task Force, 1979). The sacred is
conceptually totally enmeshed with the natural, with the earth and other natural
phenomena seen as one with humans, plants and animals. The sacred is a force
in itself, and it calls for the harmonious integration of land and people (Deloria,
Jr., 1973; Curtis, 1988:3). (NWPO90a, pp. 35, 36).
An excellent description of these people and their religion is contained in NWPO90a, which is
the result of extensive research over a number of years. The remainder of this section is taken
from this report.
8.2.1.2 Early Non-Native Settlement of the Amargosa Valley
In the 1870s, the mining boom in the Death Valley area attracted the first non-native settlers to
Amargosa Valley. In 1873, Charles King established a ranch in the Ash Meadows area (near the
Devil's Hole Protected Withdrawal area shown in Figure 8-2) where he had 1,300 cattle grazing
on the extensive grasslands watered by the surface springs in the area. In 1874, the Lee brothers
staked a claim near King's ranch and also established a herd of cattle. By the end of the 1870s,
homesteaders had claimed most of the land from Beatty to the Pahrump Valley that was watered
by springs and seeps. The mining camps in the area provided the market for the vegetables and
beef raised on these farms and ranches. When mining declined in the early 1880s, most of the
homesteaders were forced to abandon their lands (MC92).
/
The next period of growth in the area occurred in the early 1900s. The discovery of gold and
silver in the Tonopah-Goldfield district to the Northwest of the Amargosa Valley, the founding
8-10
-------
of Las Vegas, and the continued exploitation of the borax resources in Death Valley to the south
brought railroads through the Amargosa Valley. The Las Vegas and Tonopah railroad
(LV&T) entered the Valley west of the present day town of Mercury and ran across the Valley
along the route now followed by Nevada Rt. 95. The LV&T operated from 1906 to about 1918.
The Tonopah and Tidewater railroad (T&T) was begun in 1904, with a route planned to connect
Tonopah, NV with Tecopa, CA. While the planned line was never completed, it did operate
between California and Rhyolite, west of Beatty. The route followed the gorge of the Amargosa
River along the southwest boundary of the Valley. During the 1920s and 1930s, the T&T
provided the major transportation corridor for products moving into and out of the Amargosa
Valley. While agriculture continued in the Ash Meadows area, the broad flat expanse of the
Valley to the northwest was unoccupied until officials of the T&T railroad decided to "prove" the
land in 1915, perhaps envisioning the revenues that would be generated once homesteaders
claimed the tens of thousands of available acres and began shipping their products on the T&T
(MC92).
The T&T ranch, established in the southern end of the valley east of the T&T right of way,
proved that the land was arable with irrigation, but it failed to attract homesteaders; the
conditions imposed by the Homestead Act were too difficult to meet in the desolate area (MC92).
Recognizing this problem, officials of the T&T railroad persuaded Senator Pittman of Nevada to
sponsor legislation in 1919 to make it possible for individuals to acquire 640 acres (one section)
of Nevada desert land. Under the terms of the 1919 legislation, a claim could be made on four
adjacent sections of public land in Nevada that was "unreserved, unappropriated, nonmineral,
and non-timbered" and that was "not known to be susceptible to successful irrigation at a
reasonable cost from any known source of water supply" (MC92). If within two years the
claimant could show that sufficient underground water had been developed to produce a
profitable agricultural crop on at least 20 acres of the land, rights to one-fourth of the claim (640
acres or one section) could be obtained (MC92).
Far from attracting an influx of homesteaders, only five claims were filed under the 1919
legislation, all by employees of the Pacific Coast Borax Company which owned the T&T
railroad. These claims were patented in 1927 and the homesteaders transferred their claims to
the company, creating a contiguous holding, centered on the T&T ranch, of the best agricultural
land in the Amargosa Valley. A number of wells were dug on the property, at depths of 72 to 88
feet, and crops including alfalfa, vegetables, grapes, fruits, and nuts were raised. A small dairy
herd was also established. Despite the success of the T&T ranch in showing that the soil could
be productive, the prospect of the Amargosa Valley becoming a productive agricultural center
8-11
-------
faded with the decline of the T&T railroad in the 1930s. The railroad ceased operations in 1940
(MC92).
The creation of the Nevada Test Site and the passage of the Desert Land Act (sometimes referred
to as the Desert Entry Act) in the early 1950s marked the next stage in the settlement of the
Amargosa Valley. The Nevada Test Site did not bring the influx of population to the Valley that
might have been expected. While Lathrop Wells (now part of the town of Amargosa Valley) was
considered as a possible location to house the workers at the Test Site, the Atomic Energy
Commission opted for a location closer to Las Vegas and established the town of Mercury
approximately 20 miles to the east of Lathrop Wells (NYE93a).
The Desert Land Act did lure more homesteaders to the Amargosa Valley. Under the terms of
the Act, a person could claim 320 acres of unreserved land (a half-section); if within three years
they developed sufficient ground water resources to cultivate the land and "proved the land" by
bringing 40 acres into production, they could purchase the "patented" land for $1.25 per acre.
Patents granted under the Desert Land Act almost tripled private ownership of the acreage in the
Valley (NYE93a).
It should be noted that all of the agricultural development that took place in the Amargosa Valley
up through the 1960s was on a modest scale; no large-scale commercial farms were created. This
changed in the late 1960s when the Spring Meadows Ranch was established in the Ash Meadows
area on 5,645 acres obtained through a land swap with the Bureau of Land Management.
Through additional purchases of private lands, the owners of Spring Meadows Ranch were able
to expand their holding to 12,000 acres. They also attained the rights to a majority of the water
allocated to the area. Wells were drilled, pumping began, and a cattle and alfalfa operation
employing about 100 persons was created. With their increased use, water levels fell in the wells
and springs in the Ash Meadows area, causing a heated controversy over the impact on
endangered species (MC92).
In 1978, following a 1976 Supreme Court decision in its favor, the government established a
minimum water level for Devil's Hole to protect the endangered Devil's Hole pupfish. With the
restriction placed on pumping, the owners of Spring Meadows Ranch sold out to Preferred
Equities, a land development company based in Pahrump, NV. Preferred Equities planned to
develop a residential community of 50,000 persons. Towards that end, the company purchased
additional property bringing its total holdings to about 17,000 acres. The plan was highly
8-12
-------
controversial; in the early 1980s, the Nature Conservancy purchased more than 12,600 acres of
land and the associated water rights to prevent development of the area. Subsequently, in 1984,
the U.S. Fish and Wildlife Service purchased the Nature Conservancy's interest and permanently
withdrew the land from development (MC92),
8.2.2 Current Demographics and Land Use
The boundaries of the unincorporated town of Amargosa Valley encompass almost 500 square
miles of the Amargosa Desert. The boundaries of the town include all of the area where the
highest potential doses from a repository at Yucca Mountain are anticipated, with the exception
of the lands to the east and southeast that are part of the Nevada Test Site. Located 90 miles
north of Las Vegas and 330 miles from Los Angeles, the remoteness and arid climate of the area
are reflected by its population of fewer than 1,000 residents (NYE93a). Only about 11 percent of
the land (about 35,000 acres) is held privately; the remainder is under Federal control.
In 1993, only slightly more than 1,000 acres of land were cultivated in the Amargosa Valley
(NYE93a). The assessed value of these lands for tax purposes in 1993 was slightly less than
$120,000, or about $120/acre (NYE93a). This is consistent with the average value of about
$230/acre for agricultural real estate (land and buildings) in Nevada (NV95). Although two
commercial alfalfa farms and one commercial sod farm are operating full-time in the Valley,
most farms in the Amargosa Valley are operated on a part-time basis with other employment
serving as the primary source of income. Fewer than 30 persons are employed in the "Farming
and Agricultural Services" sector of the economy (NYE93a). The lack of large-scale commercial
agricultural development in the Amargosa Valley is not surprising given the following factors
listed by the U.S. Department of the Interior: "primary soil deficiencies such as coarse textures,
low water-holding capacity, high infiltration rates, and poor inherent fertility combined with
extremely hot summers, high winds, and distances from markets and services" (NYE93a).
The difficulties in making a living off agriculture in the Amargosa Valley are also illustrated by
the experience under the Desert Land Act. Prior to 1954, there were only eight wells and 8,000
acres under patent in the Valley. Between 1954 and 1960,167 new wells were drilled and
17,700 acres were patented under the Act. However, the amount of land in actual agricultural
production remained small with fewer than 1,000 acres in production, and by 1973, only 17 wells
were still used for irrigation (NYE93a). While the Act attracted many potential settlers, most
arrived and departed in a very short period (NYE93a).
8-13
-------
Despite the difficulties, a wide range of crops and livestock can be raised. Alfalfa, hay and grass,
wheat, fruits and melons, vegetables, cotton, nuts, poultry, beef cattle, dairy cattle, and fish are
being or have been grown on farms and ranches in the Valley.
Both historically and currently, agricultural activities have been restricted to the Ash Meadows
area and the portion of the Amargosa Valley known as Amargosa Farms (the private lands
southwest of Amargosa Valley shown on Figure 8-2). Currently, no farming occurs closer than
about 23 kilometers (km) south of the site. Readily accessible water from springs and seeps are
sufficient to explain why the land in the Ash Meadows area was the first to be used for
agriculture; cattle could be grazed on existing grasslands and crops could be raised without
having to develop wells for irrigation. Similarly, proximity to the T&T railroad and relatively
shallow depths to the ground water are sufficient to explain why agriculture developed in the
Amargosa Farms area during the 1920s and 1930s. Yet, after examining Figure 8-2, it is
reasonable to ask whether or not the lack of agricultural activities along the current route 95 and
north towards Yucca Mountain simply reflects historical facts (e.g., the loss of rail transport with
the early demise of the LV&T railroad and the withdrawal of lands for the Nevada Test Site) or
fundamental differences in the quality of the lands and soils and/or the availability of water. This
issue is explored in the following section which addresses the economics of ground water
development and use; the topography and soil conditions in the areas south and southeast of
Yucca Mountain; and other factors which may affect the future use of the region.
8.2.3 FactQLsAffecting Future Use of the Region
8.2.3.1 Hydrologic Characteristics and Use
Data indicate that the overall ground water flow direction in the alluvial aquifer is to the south
and southwest, with local variations (DOE96). Hydraulic gradients in the alluvial basins vary
widely, both between different basins and within any given basin. In the central section of the
Amargosa Desert, the hydraulic gradient is approximately 0.002 (refer to Figure 7-26 for a map
of the potentiometric surface in the Amargosa Desert). These aquifers are recharged directly by
precipitation, by runoff from the surrounding mountains, by infiltration from the underlying
bedrock formations, and possibly by returns of irrigation water and percolation of wastewaters.
Water leaves the alluvial aquifers by flowing to other basins, percolation to trie volcanic or
carbonate aquifers, evapotranspiration, and pumping for domestic and irrigation uses. (See
Section 7.1.2 for more information.)
8-14
-------
Ground-water flow in the volcanic aquifer is generally to the south, with a strong tendency to the
east in some areas. In one area about three km upgradient from the proposed repository site,
water levels drop over 275 meters in slightly less than two km. The precise cause of this large
gradient is not known. Outside of this large-gradient zone, hydraulic gradients measured in the
volcanic units are quite low, around 0.0003.
The volcanic aquifer is recharged primarily by melting snow on uplands north of Yucca
Mountain (e.g., Timber Mountain), with occasional intense rainstorms adding to the infiltration.
There may also be some unqualified recharge from the underlying carbonate aquifer. The
location and amount of the volcanic aquifer discharges are not currently known, but water very
likely moves to the south to enter the alluvium as the volcanic layer pinches out south of the test
site boundary; the location of this pinchout is thought to be approximately at the latitude of
Lathrop Wells (DOE95) (see Figure 7-21). A few wells account for some discharge, supplying
water for the Nevada Test Site and Yucca Mountain characterization activities (FRI94).
Flow direction and gradients in the Paleozoic carbonate aquifer are not well-defined because very
few wells have penetrated this layer. However, regional flows are generally thought to be
southward. The velocities in an area of similar rock outside the study area have been estimated at
0.006 to 60 m per day. The carbonate aquifer can be recharged directly where highly fractured
rocks are exposed at the surface at higher elevations, where precipitation is greatest Recharge
also occurs by infiltration from the overlying volcanic and alluvial deposits. The carbonate
aquifer is known to discharge at Ash Meadows, southeast of Yucca Mountain, and probably in
Death Valley, about 100 km south-southwest of Yucca Mountain. Other discharge points may
include small, low-flowing springs, though most of these are not in the study area (USG75).
The chemical quality of ground water in the area varies considerably. Generally, ground water in
wells closest to the discharge area of the system (mainly Death Valley and the Amargosa Desert)
contains high concentrations of dissolved minerals and is unsuitable for most uses, though it is
generally useable for irrigation. Water quality as measured by total dissolved solids (TDS) is
highly variable, with values typically ranging from greater than 200 mg/L to less than 1,000
mg/L. Occurrences of TDS greater than 1,000 mg/L are not uncommon; in discharge areas, TDS
values can range from 10,000 to as high as 80,000 mg/L (see Table 8-1). Individual dissolved
constituents also occur over a wide range, as shown in the following tabulation of data obtained
from analyses of water in the Amargosa Desert.
8-15
-------
Table 8-1. Range in Concentration of Dissolved Constituents in
Ground Water in the Amargosa Desert (Source: after NDC63, p. 36)
- ^\
•; *&
^G>a$fita«tt\ *
^>^ ^
Calcium (Ca)
Magnesium (Mg)
Sodium (Na)
Potassium (K)
Bicarbonate plus Carbonate
(HCO3 + CO3)
Sulfate (SO«)
Chloride (Cl)
Fluoride (F)
Nitrate (NO3)
Total Dissolved Solids (IDS)'
Range
(In pants per raiHion) -•
Low
1.9
1.0
41
3.2
102
Bicarbonate plus
Carbonate
24
6.0
0.6
0.0
217
High
85
26
1060
88
778
102
484
1050
7.9
17
79,700
Note: IDS data taken from USG90.
The alluvial aquifers tend to have high concentrations of fluoride; water from the tuff aquifer is
dominated by bicarbonates of sodium and also contains small amounts of silica, calcium,
magnesium, and sulfate. The wells that supply water for the Nevada Test Site and for
characterization activities at Yucca Mountain draw from the tuff aquifer (including human
consumption) and have shown no deterioration in water quality despite decades of pumping.
Water from the carbonate aquifer shows elevated levels of calcium and magnesium carbonates.
This water also has increased levels of sodium and potassium if it has percolated through the tuff
formation.
In terms of water rights, Nevada is an appropriative state and limits the amount of water that may
be withdrawn from any hydrographic basin to the perennial yield for that basin, i.e., the yield that
reflects sustainable withdrawals given the natural recharge and discharge of the hydrographic
basin. Ground water "mining" is not allowed. Water for almost all human activities
(consumption, irrigation, ranching) is currently drawn from the alluvial and the lower carbonate
aquifers; water is only taken from the tuff aquifer for use at the Nevada Test Site and for
8-16
-------
characterization activities at Yucca Mountain. Access to the volcanic aquifer is currently limited
for two reasons. First, the volcanic aquifer is not believed to extend much farther south than the
southern boundary of the NTS, and access to the NTS is currently restricted. Second, productive
water bearing zones within the volcanic aquifer are sufficiently deep as to make drilling too
costly except for large organizations such as government agencies or large corporations.
The major users of ground water in the area are the town of Amargosa Valley and small rural
communities in the northeast Amargosa Desert (Figure 8-6). In Amargosa Valley, water is
supplied by wells into the alluvial aquifer. Primary uses are domestic, agricultural, mining
(specialty clays), recreation (e.g., golf courses), and industrial. Most residences are supplied by
individual wells, though some trailer parks, public facilities, and commercial establishments are
served by small, private water companies. A number of springs also supply water, primarily to
the resort area in Death Valley.
The hydrographic basin in which Amargosa Valley is located (Basin 230) currently is rated at a
perennial yield of 34,000 acre-feet per year. The 1993 population of Basin 230 was about 1,100.
While currently allocated usage rights for Basin 230 stand at a little more than 41,000 acre-feet
per year, the actual usage has not yet exceeded the estimated perennial yield.
8.2.3.2 Ground Water Use
Water rights in Nevada are strictly controlled by the state and appropriated to users on a case-by-
case basis. Ground water use in Nevada is regulated by the Department of Conservation and
Natural Resources through the State Engineer's Office. For purposes of water resources
administration, Nevada is divided into 253 hydrographic basins (see USG88a for details). The
state limits the amount of water that may be withdrawn from any hydrographic basin to the
perennial yield for that basin.
The five hydrographic basins listed in Table 8-2 are of principal interest to the consideration of a
repository at Yucca Mountain. Figure 8-7 shows the location of these basins.
8-17
-------
116*43- W
i6* 43" 00' N •
116* 30'W
116*13'
m'oo-w
36* 30- 00' N
36* 15'00'N - ;-
36*7' 30" N
SPRING
IRRIGATION WELLS
AREA WITH DOMESTIC WELLS
BOUNDARY OF PHREATOPHYTE VEGETATION
BOUNDARY OF BARE SOIL EVAPORATION
(Number is estimol* of yeorly flo». in ocre-feet.)
N
10
20
Mil-
LOB A ngclrm
Figure 8-6. Map Showing Boundaries of Ground Water Subbasins in the Study Area
(DOE95a)
8-18
-------
Table 8-2. Hydrographic Basins in the Vicinity of Yucca Mountain
Hydrog^phic Basin Name
Mercury Valley
Rock Valley
Forty Mile Canyon - Jackass Flats
Crater Flats
Amargosa Desert
225
226
227-A
229
230
The regional aquifers in the five hydrographic basins that are used for human activity include the
volcanic aquifer, the valley fill aquifer and the lower carbonate aquifer. The welded tuff aquifer
is locally important; it is developed only in the southwestern areas of the Nevada Test Site. This
water is withdrawn from two water wells (J-12 and J-13) located in Basin 227-A and is used for
all site characterization activities at Yucca Mountain, including human consumption. A
significant percentage of this water is used to wet local unpaved roads for dust suppression
(STE95). Well J-13 currently is being pumped at a rate of 550 gallons per minute and Well J-12
is being pumped at a rate of 800 gallons per minute, depending on demand (DOE95f).
Most of the water pumped from the Ash Meadows ground-water subbasin is pumped from the
lower carbonate aquifer (USG76). In 1971, ground-water withdrawals associated with the
planned development of a large agricultural enterprise caused a decline in the water level of the
pool at Devil's Hole. This natural pool, formed from the collapse of the limestone bedrock, is
the only habitat of the Devil's Hole pupfish, an endangered species. As a consequence of court
action, ground-water withdrawals hi this area are now restricted to a degree that is sufficient to
maintain the water level hi Devil's Hole (USG76a).
The Nevada Test Site (NTS) receives its water from wells drilled on the NTS. The NTS
accommodates a worker population of approximately 5,000 individuals, most of whom reside in
Las Vegas and other nearby communities; a very small percentage of this workforce resides in
Mercury on an intermittent basis. There are 12 NTS wells that currently withdraw water from the
Ash Meadows ground water subbasin for construction, drilling, fire protection, and consumption
uses: Some of the water requires treatment before distribution (DOE95a).
8-19
-------
MS* W
HYDROCRAPHIC AREAS
147 GOLD PLAT
157 KAWICH VALLEY
158A CROOK LAKE
I56B PAPOOSE LAKE
159 YUCCA FLAT
180 FRENCHMAN FLAT
161 INDIAN SPRINGS VALLEY
168 THREE LAKES VALLEY IN)
169A TIKABOO NORTH
169B TIKABOO SOUTH
211 THREE LAKES VALLEY IS)
285 MERCURY VALLEY
226 ROCK VALLEY
227A JACKASS FLATS
227B BUCKBOARD MESA
22e OASIS VALLEY
229 CRATER PLAT
230 AMARG03A DESERT
D
YUCCA
SIATC
BOUNDARY
A«tA
OCSCMATON
Of OCATH VAUtY
C«OUW>*ATCR BASM
BOIXOAflY
o c jo jo
36 N
Figure 8-7. Gropund Water Usage in the Amargosa Desert (USG91a)
8-20
-------
Table 8-3 indicates that ground-water usage rights are over-allocated in Basin No. 230 by
approximately 17,000 acre-feet per year. However, actual usage in Basin No. 230 has thus far
not exceeded the estimated perennial yield. Available usage figures demonstrate that annual
basin-wide withdrawals have not been in excess of approximately 12,000 acre-feet per year. A
1993 pumpage inventory (State of Nevada, Division of Water Resources, Las Vegas Office) for
the Amargosa Desert Basin shows that the actual ground-water usage for 1993 was 11,300 acre-
feet. In 1993, water actually pumped and used for irrigation was 8,559 acre-feet, or about 30
percent of the amount allocated for that purpose; water for mining operations stood at 44 percent
of the allocated amount; and water for community and municipal uses was little more than four
percent of the allocation (NYE93a). Table 8-4 provides a breakdown of 1993 ground water
usage in Basin 230 by category.
In 1997, the total ground water use in Basin 230 was 13,902 acre feet (AV197). Water pumped
and used for irrigation was 9,349 acre-feet, or about one-third of the allocation for this purpose.
Other uses in 1997 can be compared with those of 1993 in Table 8-4. The increase in use
between 1993 and 1997 does not necessarily represent a net growth over time; water usage in
Basin 230 fluctuates significantly, depending primarily on the level of irrigation and mining
activities in a given year.
Table 8-3. Water Appropriations by Hydrographic Basin in the Study Area
Municipal ;
Ofcer
225
8,000
0
0
226
8,000
227-A
4,000
56
39
17
0
229
900
2,995
0
61
2,934
230
34,000*
41,093
28,600
85
2,486
4,255
5,667
The perennial yield is a combined total for all of the above basins and Basin #228. (Source: Hydrographic
area summaries, State Engineer's Office Nevada Department of Conservation and Natural Resources.)
8-21
-------
Table 8-4. 1993 Ground Water Pumpage Inventory for Basin No. 230
^ - • \
'-* .••""- *
* ,- -v &
Ground W»ter!Userft&«
Irrigation
Irrigation (no permits or certificates)
American Borate (314 acre-feet pumped from CA side)
Industrial-Mineral Ventures
St. Joe Bull Frog
Commercial, Quasi-Municipal, Domestic
Pumpage (acre-feet)
1993
8,559
150
512
495
1,474
no
1997
9,349
1,105
666
251
1,589
942
The 1993 ground-water withdrawals for the Jackass Flats Basin (No 227-A) and Mercury Valley
Basin (No. 225) were 205 and 338 acre-feet, respectively (USG95b), The pumpage from Basin
227-A reflects withdrawals from the J-12 and J-13 wells described earlier. Data for 1997 for
these basins are not available.
Two mineral production operations are located in the Amargosa Desert. One operation, owned
by the American Borate Corporation and located between Amargosa Valley, Nevada and Death
Valley Junction, California, was decommissioned in July 1986. The facility consisted of a large
mineral processing plant and a housing development for its employees. Water for the community
was pumped from a shallow well and was treated by a reverse osmosis process to reduce total
dissolved solids before distribution. The other operation is owned by the IMV Division of
Floridin, Inc. and is also located between Amargosa Valley, Nevada and Death Valley Junction,
California. As of 1995, the operation employed approximately 53 people to mine specialty clays
(DOE95a).
In addition to well production, a number of springs supply water to the region. The main
concentration of springs is in Death Valley in the vicinity of Furnace Creek Ranch,
approximately 60 km southwest of Yucca Mountain. The water supply for the National Park
Service facilities is derived principally from three groups of springs: Travertine Springs, Texas
Springs, and Nevares Springs. The population served by this water supply varies during the year.
From October through April, approximately 800 persons live in the area on a semipermanent
basis and an additional 2,000 persons live in the area as visitors. From May through September,
the number of semipermanent residents decreases and there are few visitors (DOE95a).
8-22
-------
Three resorts are located within the boundaries of the Death Valley National Monument: the
Stovepipe Wells Hotel, Furnace Creek Inn, and Furnace Creek Ranch. Water for the Stovepipe
Wells Hotel is trucked in from Nevares Spring. Water for Furnace Creek Inn and Furnace Creek
Ranch is reportedly conveyed from an excavated sump lined with drainage tile in the Furnace
Creek Wash (DOE95a).
Crater Flat (Basin No. 228) is currently overdrawn because of an appropriation made to Saga
Exploration, Inc. for development of the Panama-Sterling Mine, located on the east side of Bare
Mountain. The mine uses its own well for its heap-leach operation and relies on municipal water
for its potable water. The mine employs approximately 40 individuals and is expected to be in
operation until 1997 or 1998 (DOE95a).
The proposed repository at Yucca Mountain would be'about 400 meters above the aquifer that
occurs in the tuff members underlying the site. The tuff aquifer appears to discharge to the
alluvial aquifer that underlies much of Amargosa Valley (Y072). At the northern end of the
Amargosa Valley, on Jackass Flats, approximately five to seven km south-southeast of the site,
the depth to the ground water (tuff aquifer) is approximately 300 m. In the Amargosa Farms
area, between 30-40 km south- southwest of Yucca Mountain, the (alluvial) aquifer is at a depth
of 10-40 m. A deep aquifer in carbonate rock underlies the tuff aquifer and portions of the
alluvial aquifer. This carbonate rock aquifer lies at a depth of more than 1,000 m at the northern
end of the valley (Jackass Flats) and discharges at or near the surface in the Ash Meadows area
approximately 45 km southeast.
The feasibility of using a ground water resource depends on the economic value of the water to
the user, the costs of drilling the well, and the costs of pumping water from the well. While
much has been written on the theory and practice of determining the economic value of water to
the user, it is sufficient for present purposes to recognize that: 1) the marginal value of water
varies greatly depending on its use, and 2) agricultural use for irrigation generally has the lowest
marginal value, while domestic use for drinking and hygiene has the highest marginal value.
Preliminary estimates of the marginal value of water for irrigation in the Amargosa Valley, based
on the economics of raising alfalfa (the major cash crop), suggest a marginal value of about
$40.00 per acre-foot (DOE91). Based on long-range plans for providing water to users in Las
Vegas, a marginal value of about $800 per acre-foot can be assumed for domestic uses (MIK92).
Marginal values for other agricultural uses or industrial and mining uses would be intermediate
between these values and would depend upon the specific crop, process or resource being
8-23
-------
produced. When the costs of drilling and pumping water are less than the marginal value for the
intended use, the ground water resource can be economically developed.
Figure 8-8 shows the depths to ground water in the vicinity of Yucca Mountain and the locations
of existing wells and boreholes (DRI94). Table 8-5 lists the wells and boreholes that are
designated as being privately owned and provides information on their exact locations; depth to
water, and well depth. Examination of the well data in Table 8-5 identifies 34 wells for which
the use is shown as "irrigation" or "domestic/irrigation" and for which the depth of the well and
the depth to ground water is known. The averages for these wells (excluding surface springs and
seeps) include an average depth of less than 300 feet and a depth to water of less than 100 feet,
which is consistent with the heavy concentration of wells depicted in Figure 8-6 at locations
where the depth to ground water is less than 100 feet. The deepest wells used for human
consumption other than J-12 and J-13 in the NTS are those near Lathrop Wells. At 23 km from
Yucca Mountain, these vary from 90 m to 120 m depth to water. Figure 8-9 shows the depth to
water versus the distance from Yucca Mountain in graphical form.
Water Availability/Perennial Yields
In order to estimate the population that may be supported at some time in the future by the water
resources available in the Yucca Mountain area, the following analysis was performed.
Perennial yield is defined as the maximum amount of water that can be withdrawn from the
ground-water system for an indefinite period of time without causing a permanent depletion of
the stored water or causing a deterioration in the quality of the water (NDC63). It is ultimately
limited by the amount of water annually recharged to or discharged from the ground-water
system through natural processes in addition to that which might become available by artificial
recharge and water returned to the ground-water system by infiltration of irrigation or waste
water.
In estimating perennial yields, the effects mat ground-water development may have on the natural
circulation in the ground-water system should be considered. The location of the withdrawal
centers in the ground-water system may permit optimum utilization of available supply.
Alternately, the location of withdrawal centers may be ineffective in the utilization of the
available water supply. The location of the wells may favor improving the initial quality with
time or may result in deterioration of quality under continued withdrawals. Development by
8-24
-------
CL
• L»r4li»n .f WfH. ft Sv.ni.
K il Ini'l P n
Figure 8-8. Locations of Water Wells in the Amargosa Farms Area (DRI94)
8-25
-------
Table 8-5. Wells and Boreholes in the Amargosa Valley
USW Diane
USW H-5
USW STN2
USW UZ-N70
USW UZ-N64
USWUZ-N98
USWSTN3
UE-25 NFCW
USW H-1
USW UZ-N75
USW UZ-N40
USW UZ-N88
UE-25UZN#18
USWUZ-N95
UE-25 UZN#2
USW UZ-N52
USWUZ-N15
USWUZ-N16
UE25WT#04
USW UZ-N66
UE-25UZMM3
USW GA-1
USW H-3
UE-25 UZN#60
UE-25 WTO16
UE-25 STN1
UE-25 WX Station 1
UE-25 STN6
USWUZ-N57
USW Carolyn
USWUZ-N67
36:51:21
36:51:22
36:51:19
36:51:48
36:51:13
36:51:35
36:51:17
36:51:16
36:51:58
36:50:31
36:51:17
36:50:24
36:51:20
36:50:15
36:51:41
36:50:25
36:53:15
36:53:16
36:51:40
36:50:01
36:51:35
36:53:28
36:49:42
36:50:14
36:52:39
36:50:34
36:50:06
36:53:40
36:49:28
36:49:06
36:49:13
116:27:56
116:27:55
116:27:56
116:27:40
116:27:49
116:27:16
116:27:06
116:27:01
116:27:12
116:27:53
116:26:50
116:28:24
116:26:37
116:28:04
116:26:36
116:27:06
116:27:47
116:27:46
116:26:03
116:27:19
116:26:00
116:27:51
116:28:01
116:26:21
116:25:34
116:25:49
116:26:04
116:26:45
116:27:28
110:27:56
116:26:55
1219
1829
482
1219
519
704.2
' 572
439
750.8
473
Meteorological
Hydrologic Testing
Meteorological
Precip. Gauge
Precip. Gauge
Precip. Gauge
Meteorological
Precip. Gauge
Hydrologic Testing
Precip. Gauge
Precip. Gauge
Precip. Gauge
Precip. Gauge
Precip. Gauge
Precip. Gauge
Precip, Gauge
Precip. Gauge
Precip. Gauge
Water Level Monitoring
Precip. Gauge
Precip. Gauge
Precip. Gauge
Hydrologic Testing
Precip. Gauge
Water Level Monitoring
Meteorological /AQ
Meteorological
Meteorological /AQ
Precip. Gauge
Meteorological
Precip. Gauge
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
Government
0.5
: 0.7
0.8
0.9
1
1.4
1.8
1.9
2
2.1
2.2
2.4
2.5
2.6
2.7
2.8
3.1
3.1
3.2
3.3
3.4
3.5
3.6
3.8
4
4.1
4.3
4.4
4.4
4.7
4.9
8-26
-------
Table 8-5. Wells and Boreholes in the Amargosa Valley (Continued)
UE-25D 1 PTH
USWUZ-13
UE-25 c#1
UE-25 D#1
UE-25 STN8
PLUG HILL
UE-25 WT#14
UE-25 STN4
UE-25 UZ#4
UE-25 WTO15
Fran Ridge
UE-25WT#17
UE-25 UZ#5
UE-25WTCM3
UE-25 JF#1
UE-25 JF#2
40-mile No.. 2
J-13
UE-25 UZN#85
UE25WT#03
UE-25 Robin
UE-25 WT*12
CF2. USWVH-1
USWVH-1
J-12
UE-25 JF#3
USWCF1,Gexa4
CF3.Cind-r-liteWell
CF1a. GexaS
#22 NECO
J-11
#34LathropWell
36:49:38.
36:48:57
36:49:47
36:49:38
36:49:42
36:48:43
36:50:32
38:51:51
36:51:42
36:51:16
36:49:17
36:48:22
36:51:41
36:49:43
36:51:16
36:49:43
36:54:15
36:48:28
36:48:44
36:47:58
36:48:55
36:46:56
36:47:32
36:47:32
36:45:54
36:45:27
36:55:20
36:41:05
36:54:45
36:46:00
36:47:06
36:38:27
116:25:21
116:28:01
116:25:43
116:25:21
116:25:35
116:29:26
116:24:35
116:24:15
116:26:26
116:23:38
116:24:55
116:26:26
116:26:26
116:23:51
116:23:38
116:23:51
116:23:57
116:23:40
116:24:06
116:24:58
116:23:42
116:26:16
116:33:07
116:33:07
116:23:24
116:23:23
116:37:03
116:30:26
116:38:39
1ia-41:30
116:17:06
116:26:23
1115
1084
1033
1084
1033
1011
964
954
945
1199
832
1245
850
1050
796
1806
914
1805
399
415
443
352
1063
348
399
763
762
347
488
140
214
175
405
NA
1298
401
384
346
354
394
303
354
303
283.2
300
346
278
184.2
226.2
217
244
98
63
86
317.4
90
Water Level Monitoring
Precip. Gauge
Hydrologic Testing
Water Level Monitoring
Meteorological
Meteorological
Water Level Monitoring
Meteorological
Infiltration Monitoring
Water Level Monitoring
Meteorological
Water Level Monitoring
Infiltration Monitoring
Water Level Monitoring
GW Monitoring
GWM
Meteorological
Precip. Gauge
Water Level Monitoring
Meteorological
Water Level Monitoring
Water Level Monitoring
Water Level Monitoring
GWM
INO
IND
IND
Industrial
GW Monitor/Dom
Dom/lrrigation
DOE
Government
Government
Government
Government
Government
' Government j
Government
Government
DOE
Government
Government
Government
DOE
DOE
DOE
Government
DOE
Government
Government
Government
Government
DOE
Government
DOE
DOE
NV Gold
Cind-r-Lite
NV Gold
Private
DOE
Private
5
5
5.1
5.3
5.4
5.7
5.9
6
6
6.3
6.6
6.6
7
7.1
7.5
7.6
8
8
8.2
8.4
8.5
9.3
10.5
10.5
14
14
15
16.5
19
21
21
23
8-27
-------
Table 8-5. Wells and Boreholes in the Amarosa Valley (Continued)
iiiiiiHi^^
HHBBMBJBHHBHHI|HfflHH^
H^B^^^^BI^^^^^^^^^^^B»I^^^^M^l^^^BB^HB^KHBHKiMilmiiHi^MBIIHjBBttHH{HilB
H^g^|p|^BaaMaaHaMM^BMMH|8BBamlHtt«HlB»|«B^^
|#37 Lathroo 36:38:36, i 116:23:57 I 812 1 120
#38 Lathrop
#35 Lathrop 15s/49e
#36 Amarqosa
USWAD17
#57A. SasseWell
USWAD1
USWAD2
#59 Well 16s/49e OSacc
#60 K. GareyWell
#63 School house 15aaa
RV-1 (TW-5)
#39 Private Well
#41 Private Well
#43KirkerWell
#44 Bob Nichols Well
#45WelM6s/48eAmarflo
#46 Amaraosa Well 15dda
#53 Amargosa Well 24aaa
#58 K. Finical Weil
#61 School Well 09dcc
#62 Amargosa 12ddd
USW RV1
#42 Sullivan Wen
#64 Amargosa 16ccc
#65 Amarqosa 18dc..
#67 Amargosa 23add
#71 Well 07bcd Cook
#52 Amargosa Well 23da
#66 Jacob's #2
#40 Private Well
#47 Amaraosa Well 17abb
36:38:32
36:37:40
36:37:14
36:38:35
36:35:28
36:41:30
36:38:25
36:34:35
36:34:18
36:34:00
36:38:15
36:34:55
36:34:50
36:34:25
36:34:03
35:33:58
36:33:25
36:33:13
36:34:56
36:34:10
36:34:20
36:38:16
36:34:15
35:33:1 1
36:33:23
36:33:10
36:34:25
36:32:44
36:32:49
36:34:25
36:34:00
116:23:48
116:26:40
116:26:45
116:23:47
116:28:42
116:41:12
116:24:33
116:28:40
116:27:42
116:26:00
116:17:59
116:36:40
116:35:30
116:33:20
116:32:31
116:33:18
116:32:35
116:30:25
116:28:41
116:27:35
116:24:50
116:17:59
116:35:20
116:28:09
116:29,44
116:25:10
116:23:50
116:31:35
116-29:19
116:33:50
116:35:10
812
784
111
746
802
805
738
741
744
932
725
727
726
725
724
719
722
739
739
750
722
726
723
732
756
713
720
722
722
**
148
467
87
293
229
62
94
120
38
80
46
56
50
**
146
60
58
**
**
*•*
105
116
60
140
94
46
85
[BR&HBHSaBiHlBHi^B^B^^IHHBS^HRHifiH^H^^^IH^H^Ri^l^^^l^^^^^^^^^^^^H
ffl™8BBiBmlB8ifflK8^^H8Hffi"ffl™""'""^^""B^^^^^^^^^^E
W^Ka9^^XSi^^^^^^^>^^SySKmvKS^SnaUmK^
I 105 I Public Supply | Water Co.
120
78
73
21
82
99
45
46
51
207
0
34
32
29
30
**
29
45
49
44,
40
**
33
32
42
24
30
23
31
Unused
IND
Observation
Water Level Monitoring
**
Water Level Monitoring
DOM
Irrigation
DOM
GW Monitoring
DOM
Irrigation
DOM
Dom/lrrigation
**
**
Irrigation/DOM
Domestic
Public Supply
*•
Water Level Monitoring
DOM
**
Unused
Irrigation
Dom/lrrigation
Irrigation
**
Dom / Irrigation
Irriqation
Water Co.
DOM
Private
Government
Private
NA-6 USGS
NVDOT
Private
Private
Private
DOE
Private
Private
Private
Private
Private
Private
Private
Private
Muni
M
Government
Private
*4
• *
Private
Private
Private
Private
Private
Private
||y>y£i9UjE^y^Xffl
^EB^^RH
HRBffll
I " 23
23
24
25
25
27
27
27.2
28
28
28
28
29
29
29
29
29
29
29
29
29
29
29.6
30
30
30
30
30
31
31
32
32
8-28
-------
Table 8-5. Wells and Boreholes in the Amarosa Valley (Continued)
j^BimmHHH^
#48 Amargosa Well 17ccc
#49AmaraosaWell18bcc
#50 Amargosa Well 18dad
#51 Lathrop Well 23bdb
#69 Well 35baa Amargosa
USWAD5
USW AD6
#21 Mathev/s Well
#55 Smith's Well 36aaa
USWAD3
#56 John Mills Well
#68 Well 35aaa Amargosa
#70 Well 36aba Amargosa
#73 Well17s/48e lab
USWAD4
#74 LvleRecs. #2
#76Well 09aa Copeland
#72 Lvle Recs.#2
#75 Well oSddb
#77 Well Mecca Club
#78 Well 15bbd Amargosa
#1 1 Fairbanks Spring
#27 Soda Spring
#79 Well 15 be Amargosa
#02 Amargosa Tracer #1
#03 Amargosa Tracer#2
#04 Amargosa Tracer #3
#26 Rogers Sp.
#84 Well 19aab Amargosa
#20 Longstreet Sp.
#80Well 28bcd Amargosa
^•H
36:33:09
36:34:00
36:33:32
36:33:00
36:31:27
36:33:10
36:23:13
36:31:32
36:31:28
36:35:26
36:30:35
36:31:10
36:31:20
36:30:28
36:34:28
36:29:38
36:29:40
36:29:20
36:29:04
36:29:36
36:28:39
36:29:26
36:29:22
36:28:32
36:32:13
36:32:11
36:32:13
36:28:40
36:28:00
36:28:04
36:26:50
HHHHHHI
116:35:47
116:35:20
116:35:49
116:32:10
116:25:37
116:23:40
116:13:38
116:24:00
116:30:24
116:35:29
116:30:50
116:25:10
116:24:20
116:30:25
116:23:47
116:30:01
116:26:58
116:31:10
116:28:08
116:25:15
116:26:37
116:20:30
116:20:10
116:26:43
116:13:37
116:13:39
116:13:80
116:19:20
116:22:30
116<19:30
116:27:40
BHHBHBH
718
720
722
716
714
725
732
707
709
730
701
708
712
702
756
698
695
697
693
694
690
695
695
690
733
732
732
695
698
701
689
HjHHSi
**
110
100
100
106
207
91
73
124
52
41
82
152
6
99
56
110
NA
NA
157
202
252
246
NA
30
NA
tt
27
29
26
37
13
21
40
13
30
0
16
36
12
5
15
20
17
0
15
13
12 ^
12
5
0
j^HHIHHHBHHI^HIIHH|HHIHIIIIIIH
umirriiirrvii "-^BHHahttMMfllulHMMi^mHtw
**
Irrigation
Irrigation
DOM/Irrigation
Water Level Monitoring
Water Level Monitoring
Irrigation
DOM/Irrigation
Irrigation
Irrigation
Unused
Irrigation
Unused
unused
DOM
Unused
Irrigation
Irrigation
Irrigation
Observation
Observation
Observation
Irrigation
Irrigation
Irrigation
BMH
Private
Private
Private
Private
Private
BLM
uses
Private
Private
Davidson, Rober
Private
Private
Private
Private
Cook. L.C.
Private
'rivate
'rivate
'rivate
'rivate
Private
'rivate
'rivate
Private
NV State
NV State
NV State
Private
Private
Private
Private
^HMMunuu^H
BB^^^BHl^K
32
32
32
32
32
32.8
32.8
33
33
33
34
34
34
34
34.4
35
35
36
36
37
37
38
38
38
39
39
39
39
39
40
40
8-29
-------
Table 8-5. Wells and Boreholes in the Amarosa Valley (Continued)
#81 Welt 29acc Amargosa
#96 Well 27bbb
USWAD9
#05 Armyl
#23 Pt. Rocks Hwy Well
#28 Spring 17s/50e
#83 Well14cac Flowing
#06 Ash Tree Spring
#29 Soring 18s/49e Clay C
#85 WeU23bb2 Flowing
#89 Tenneco #3
USWAM1 a. Fairbanks Sp.
USWAD7
#82 Well 36ccd
#88Well 02caa Tenneco#2
#90 Well! 1bbb Amargosa
#91 Well 06dac
USWAD10
USWSP1
#87 Well 08c1
#92 07bbb
#94 Ash Meadows
USWAM1, Rogers Sp.
#09 Crystal Spring
#10 Devils Hole
#86 Well 23b Flowing.
#93 Spring Meadows
USWAM2. Five Springs
#24 Pt. Rocks Sp.
#25 Pr. Rocks So. Rock
#19 J. Rabbit Sp.
36:26:50
36:27:00
36:28:48
36:35:30
36:33:33
36:27:36
36:28:20
35:25:35
36:25:30
36:27:50
36:24:51
36:29:26
36:30:09
36:25:30
36:24:59
36:24:35
36:24:54
36:25:25
36:29:26
36:29:00
36:24:33
36:24:03
36:28:56
36:25:16
36:25:32
36:27:40
36:24:00
36:27:65
36:24:02
36:24:05
36:23:24
116:28:17
116:32:15
116:26:46
116:02:14
116:06:42
116:19:04
116:18:55
116:24:42
116:23:50
116:19:05
116:25:41
116:20:28
116:30:27
116:24:25
116:25:10
116:25:15
116:22:37
116:27:43
116:20:28
116:09:10
116:16:57
116:16:08
116:19:53
116:19:19
116:17:27
116:12:10
116:16:05
116:19:04
116:16:26
116<16:15
116:16:41
683
685
691
961
859
715
713
664
664
713
658
691
703
671
664
658
658
668
729
707
707
691
671
720
710
704
722
701
707
692
*«
91
121
593
244
NA
28
NA
NA
46
224
34
213
114
**
**
332
122
152
86
62
Spring
*»*
7
91
38
NA
NA
NA
**
14
22
239
138
»»
0
0
0
0
22
0
20
**
20
tt
**
3
10
7
0
31
0
0
0
5
0
0
0
0
**
Irrigation
Observation
Destroyed
Unused
Irrigation
Domestic
Unused
Irrigation
Industrial
Industrial
Industrial/Public Supply
»*
**
Water Level Monitoring
Spring-Discharge Monitoring
DOM/Irrigation
IND
Irrigation
Irrigation
Public Supply
Domestic
Irrigation
Irrigation
Irrigation
Irrigation
Private
Private
Gilgan's No.
Army
Private
**
Private
Private
*4
Private
Private
FWS
Blackman
Private
Private
Private
Private
NA-9 USGS
Government
Private
Private
Private
FWS
Private
Water Co.
Private
Private
FWS
Private
Private
Private
i^^OBBHBHHBBi
40
40
40
41
41
41
41
42
42
42
42
42
42.5
43
43
43
43
43
43.2
44
44
44
44
45
45
45
46
46.5
47
47
48
8-3O
-------
Table 8-5. Wells and Boreholes in the Amarosa Valley (Continued)
m^H^^^^^^^m
^H^HBB^H^^^l
#07 Big Spring
#08 Bore Spring
USWAD11
USWAM3, Gamer's Well
AM5a. Crystal Pool
USWAD8
USW SP2
USWAM4. Devils Hole. (A.M.
USW AM5. Devils Hole Well
#33WeN15F.LDVJ
#95Well 14c1 DVJ
USWAM6. Pt. Rocks No.
#54 Jacob's Well #1
USWAM7. Pt. Rocks So.
#18GS-8F.L DVJ
DV-1. Texas Spring
USW SP5
USW SP3
#14GS-12F.L DVJ
USWAD12
#17GS-4F.L DVJ
USW SP4
#12 Franklin Lake
#13 Franklin Lake
#15GS15F.L DVJ
#16GS-18F.L DVJ-.
AM-8
AM8, Big Spring
DV-2
DV-3
l#30 Well 05 Franklin L DVJ
BfflfflBiHiHHlffii
BB
36:22:29
36:21:47
36:19:57
36:25:55
36:25:13
36:29:29
36:25:13
36:25:32
36:25:30
36:18:33
36:18:15
36:24:32
36:32:19
36:24:20
36:17:00
36:27:28
36:27:28
36:22:29
36:16:27
36:20:21
36:15:53
36:22:52
36:15:15
36:15:15
35:15:16
36:14:44
36:22:29
36:22:29
36:22:52
36:22:30
36:14:15
^^HBni
116:16:26
116:16:21
116:17:52
116:20:53
116:19:27
116:08:57
116:19:27
116:17:27
116:17:15
116:22:00
116:24:46
116:16:57
116:30:24
116:16:37
116:22:02
116:50:11
116:50:11
116:17:15
116:22:12
116:13:30
116:21:21
116:42:53
116:22:08
116:22:08
116:22:01
116:21:57
116:16:25
116:16:25
116:42:53
116(39:29
116:22:21
BBjpgi
ffmspi
683
683
717
658
669
730
720
733
622
662
707
714
712
611
613
741
611
611
611
611
610
683
683
634
832
610
HIHI
NA
NA
610
62
66
61
5
45
152
50
179
10
9
482
7
41
102
7
8
NA
11111$
0
0
69
43
Surface
10
15
0
1
42
26
40
0
1
24
1
0
0
2
4
Surface
Surface
Surface
198
2
HHHBiBBB
Irrigation
Irrigation
Water Level Monitoring
Spring-Discharge Monitoring
TEST
**
Irrigation
Testing
DOM
Spring-Discharge Monitoring
Spring-Discharge Monitoring
Testing
Water Level Monitoring
Testing
Spring-Discharge Monitoring
Testing
Testing
Testing
Testing
Domestic
Industrial
Industrial
TEST
jjjHflttjGftBtKam
mm
Private
Private
GS-3 USGS
Gamer, G.
FWS
Cherry Patch
Government
NPS
FWS
Water Co.
Private
FWS
Private
FWS
Water Co.
NPS
Government
Government
Water Co.
GS-1 USGS
Water Co.
Government
Water Co.
Water Co.
Water Co.
Water Co.
FWS
FWS
US Borax
US Borax
Water Co.
mwajnKBttjgjgm
•HgSm^B
•BaSSOi
49
! 49
49
50
51
51
51 '
51.5
52
53
53
53.6
54
54.5
55
56
56
56.8
57
57
58
58
59
59
59
59
59
59
60
60
61
8-31
-------
Table 8-5. Wells and Boreholes in the Amarosa Valley (Continued)
Qso^^H^^I^HHI^HHB
#32WelM3F.LDVJ
#31 Well 10 Franklin L. DVJ
USWAD15
USWAD14, DVJ Well+A297
USWAD16
USWAD13
GS-10
GS-12
GS-02
GS-04
GS-17
Well 13
GS-18
36:14:43
36:14:12
36:19:54
36:18:17
36:20:14
36:17:24
36:17:00
36:16:27
36:16:05
36:15:53
36:15:16
36:14:43
36:14:44
IBB^BBRSH^HIB
116:23:31
116:22:30
116:18:12
116:24:47
116:13:39
116:32:42
116:22:02
116:22:12
116:21:27
116:21:21
116:22:01
116:23:31
116:21:57
611
609
623
824
617
613
613
612
612
608
611
BBBiHHBSBi
10
69
610
7.25
8.84
3.57
6.83
10.67
5.55
8.2
HiiiiBli
2
1
117
-0.73
0.82
0.4
1
2.42
3.02
3.4
H^^^Hl^^lHHflHl^fl^^nH^^H^^^^H
TEST
TEST
Water Level Monitoring
Water Level Monitoring
Water Level Monitoring
Piezometer
Piezometer
Piezometer
Piezometer
Piezometer
Water Level Monitoring
Piezometer
mmmm
Water Co.
Water Co.
Government
Ettie, Lee
Government
S-1 USGS
it
H
n
M
ii
Government
11
^•KMUUWUBH
. 61
62
62
62.5
63
64
76.3
77.9
78
78.4
79.1
79.9
80.1
8-32
-------
1200
1000
800 •
I
200
0
.200 J
**r
•*
* • *«* *i
•
10 20 30 40 60 60 70 60 00
Dlit {km) from Yucci MounUtn
Figure 8-9, Wells and Boreholes in the Amargosa Valley. Only 10 Persons
Currently Live at the 20 km Distance.
8-33
-------
wells may or may not induce recharge in addition to that received under natural conditions. Part
of the water discharged by wells may re-enter the ground-water reservoir by infiltration of excess
irrigation or waste water and thus be available for re-use. Ground water discharged by wells
eventually reduces the natural discharge. In practice, decreasing natural discharge by pumping is
difficult, except when the wells are located where the water table can be lowered to a level that
eliminates evapotranspiration in the natural area of discharge or underflow from the basin.
There are a number of means by which the perennial yield can be calculated. The State of
Nevada accepts the method proposed by NDC63, which estimates the perennial yield of
hydrologic basins by assuming that perennial yield is equal to the volume of water that would
naturally discharge through evapotranspiration and lateral outflow (underflow). In other words,
perennial yield is considered equal to total natural basin discharge.
An alternative method to that presented by NDC63 for the determination of perennial or safe
aquifer yields is presented by Linsley et al. (LIN82), in which the perennial yield is expressed as
a function of the quantity of water available. This hydraulic limitation is often expressed by the
equation:
G = P-Qs-ET + Qg-Sg-Ss (1)
where G is safe yield (i.e., perennial yield); P is precipitation on the area tributary to the aquifer;
Qs is surface streamflow from the same area; ET is evapotranspiration; Qg is net ground water
inflow to the area; Sg is the change in ground-water storage; and Ss is the change in surface
storage. If the equation is evaluated on a mean annual basis, Ss will usually be zero. All terms in
Eq. (1) are subject to artificial change. G can be computed only by assuming the specific
conditions for each item. For example, artificial recharge operations can reduce Qs. Irrigation
diversion from influent streams may increase evapotranspiration. Lowering the water table by
pumping may increase ground-water inflow (or reduce ground water outflow) and may make
gaining streams into losing streams.
The factors that control the assumptions on which Eq. (1) is evaluated are primarily economic.
The feasibility of artificial recharge or surface diversion is usually determined by economics. If
water levels in the aquifer are lowered, pumping costs are increased. Theoretically, there is a
water-table elevation at which pumping costs equal the value of the water pumped and below
which water levels should not be lowered. Practically, the increased cost is often passed on to
the ultimate consumer. The minimum water level is determined after excessive lowering of the
water table results in contamination of the ground water by upcoming and inflow of undesirable
waters.
8-34
-------
The permanent withdrawal of ground water is called mining. If the storage in the aquifer is
small, excessive mining may be disastrous to any economy dependent on the aquifer for water.
Oa the other hand, many ground-water basins contain vast reserves of water and planned
withdrawal of the water at a rate that can be sustained over a long period may be a practical use
of this resource. The annual increment of mined water, Sg in Eq. (1), increases the yield. Thus,
Eq. (1) cannot properly be considered an equilibrium equation or evaluated in terms of mean
annual values. It can be evaluated correctly only on the basis of specified assumptions for a
stated period of years. The following discussion presents a methodology by which the various
parameters in Eq. (1) were determined for the Yucca Mountain area.
The hydrographic areas (HA) that are most relevant to the determination of perennial yields
downgradient of Yucca Mountain are basin numbers 225,226,227-A, 229 and 230. Table 8-6
presents the water budget information for these hydrographic areas, obtained from the State of
Nevada's water planning report (NDC71). Each of the column entries are discussed below.
Table 8-6. Ground Water Budget for Hydrographic Basins in Study Area (Source: NDC71)
Basin Number
Basra
Water
-------
Column 2 - Hydrographic Basin Name
Almost all of the current ground-water use downgradient of Yucca Mountain is derived from the
Amargosa Desert Hydrographic Basin (ADHB). Those basins within the study area that are
hydraulically connected to the ADHB via ground water include Mercury Valley (HB 225), Rock
Valley (HB 226), Jackass Flats (HB 227-A), and Crater Flat (HB 229).
Column 3 - Ground-Water Recharge from Precipitation
Ground-water recharge from precipitation represents the volume of precipitation that moves
vertically through the unsaturated zone (region above the water table) and becomes available for
pumping. Other sources of recharge (e.g., irrigation return flow) are thought to be insignificant
and are not included in this column.
Column 4 - Ground-Water Inflow
The ground-water inflow is the volume of ground water that enters the hydrologic area from
other hydrologic basins. In the case of the ADHB, ground water enters from Hydrologic Basins
225, 226,227-A and 229. The volumes derived from each of these basins are presented hi acre-
feet/year and total 43,800 acre-feet/year. As noted in the table, these values should total ground-
water inflow into ADHB (44,000 acre-feet/year). Apparently either an error was made in the
data entries or the value was rounded to 44,000.
Column 5 - Ground-Water Surface Discharge
The ground-water surface discharge volumes represent the volume of water that is discharged to
the surface via streams and seeps, in addition to water that is removed from the aquifer by
evaporation arid the transpiration of plants. In the perennial yield calculations performed below,
this discharge is actually treated as ground-water outflow and is assumed to be available for
consumption. The rationale for this assumption, presented in DOI63, is that once the water table
has sufficiently dropped below some point, significant transpiration and surface discharges will
no longer occur.
All of the 24,000 acre-feet/yr that is discharged to the surface in the ADHB is removed from the
system by evapotranspiration. Furthermore, almost all of this water is attributed to spring
discharges at Ash Meadows.
8-36
-------
Column 6 - Ground-Water Outflow
The ground-water outflow is the volume of ground water that flows out of the hydrologic basin
into adjacent basins. The table indicates that the outflow from the ADHB of 19,000 acre-feet/yr
flows into the Death Valley Hydrographic Basin. Note that ground-water outflow (19,000 acre-
feet/yr) added to evapotranspiration (24,000 acre-feet/yr) should be equal to ground water inflow
(43,000 acre-feet/yr) for Basin # 230. However, it is unclear why a discrepancy of 800 acre-
feet/yr exists. This discrepancy will not significantly affect perennial yield estimates.
The site for the proposed Yucca Mountain repository lies primarily within hydrographic basin
227A (Figure 8-6). For this basin, as shown in Table 8-3, the perennial yield cited in Nevada's
water planning report (NDC71) is 4,000 acre-feet per year. The water planning report also
indicates that the storage volume for this basin is 7,400 acre-feet per foot (Table 8-7). In
contrast, the perennial yield for Basin 230, Amargosa Desert, is 34,000 acre-feet per year, and the
storage volume is stated in NDC71 to be 35,000 acre-feet per foot. The perennial yield for Basin
230 is therefore seen to be nearly the full amount of the storage volume per foot of depth, while
the perennial yield for Basin 227A is only about 50% of the storage volume per foot.
Because of data limitations, the perennial yield estimate of 4,000 acre-feet per year for Basin
227A is not derived from water budget relationships such as Equation 1. It is, instead, an
estimate of the water that can be removed annually without significantly altering the ground
water regime. Habitation of Basin 227A and direct use of its water resources in domestic or
irrigation wells is not expected because of the large well depth that would be required across
most of the basin (Figure 8-9). The Basin 227 A water may, however, be extracted in the future
and exported to locations such as Pahrump, where water supplies are already oversubscribed, as
part of the county-wide water utilization strategy.
Size of Potential Populations
The following paragraphs examine the size of the potentially-affected population that could be
sustained by the ground water available in the Yucca Mountain region. The available ground
water has been defined as that ground water which is contained within Hydrographic Basins 225,
226,227-A, 229 and 230. These hydrologic basins are considered to be the most relevant to the
analysis because they are located downgradient of Yucca Mountain to a distance of
approximately 50 miles. Since Basins 225,226,227-A, and 229 discharge into Hydrographic
Basin 230, this basin (HB 230) is used in the calculations.
8-37
-------
In Eq. (1), precipitation (P) minus evapotranspiration (ET) is assumed to equal ground-water
recharge. Table 8-6 indicates that Basin #230 receives 600 acre-feet/yr of recharge from
precipitation. There is no significant surface stream flow (Qs) or change in surface storage (Ss).
As mentioned previously. SK represents the annual increment of mined water and should be set to
zero for perennial yield determinations. This suggests mat Eq. (1) may be written as:
G = 600 + Qg (2)
Table 8-6 indicates that 44.00022 acre-feet/yr enters Basin #230 as lateral ground-water inflow
(Qg) from other hydrographic areas (225,226,227-A, 229). Therefore, based on Eq. (2), the total
volume of yearly sustainable water under current conditions would be 44,600 acre-feet/yr. A
ground water modeling study performed in USG95c, and an alternative analysis (NDC63),
indicate sustainable yields may be closer to 24,000 acre-feet/yr. Furthermore, the State of
Nevada assumes a perennial yield of 24,000 acre-feet/yr for Basin #230. The State's estimates
are based on work reported in NDC63 in which the authors estimated that discharge via
evapotranspiration is 23,500 acre-feet/yr and ground-water outflow is 500 acre-feet/yr for a total
perennial yield of 24,000 acre-feet/yr. However, USG88a indicates that ground water outflow
could be as high as 19,000 acre-feet/yr. In this case, NDC63's method of determining perennial
yields (i.e., evapotranspiration plus lateral ground water discharge) would result in a perennial
yield of 42,500 acre-feet/yr. Therefore, the estimate of 44,600 acre-feet/yr appears to represent a
reasonable upper bound maximum for the water available. This value would also tend to
maximize the estimates of the potentially-affected population size.
In 1993, there were approximately 1,100 people residing within Basin #230. The water
withdrawal for the same year from the underlying aquifer was 11,300 acre-feet (Table 8-4). This
translates to a yearly per capita withdrawal rate of 10.27 acre-feet (this value is relatively large
and reflects water use primarily for irrigation). If it is assumed that future water consumption in
the area is proportional to current per capita water consumption rates, the total population that
could be sustained by a perennial yield of 44,600 acre-feet/yr is 4,342 people.
A scenario that provides a reasonable upper bound on the number of people that could be
supported by the ground water in this area can be made by assuming that all water use in the
basin would be consumed entirely by domestic use, possibly exported to Las,Vegas. Van der
"Although there may be a slight error in the reported value, it is used, rather than the corrected value,
because its use will provide higher population estimates.
8-38
-------
Leeden et al. (VAN90) indicate that the average person in the United States utilizes 86.5 gallons
per day (gal/day) of water for domestic use (0.097 acre-feet/yr). Van der Leeden et al. (VAN90)
also indicate that the average individual in the State of Nevada utilizes 141 gal/day (0.16 acre-
feet/yr), which is somewhat higher than the national average. In order to maximize the size of
the potentially-affected population, the lower value for domestic water use (0.097 acre-feet/yr) is
used in conjunction with an assumed sustainable yield of 44,600 acre-feet/yr. This results in a
potentially-affected population size of 459,794; this value is expected to be a reasonable
maximum.
The water use data of 1997 provide a basis for estimating the per capita water use for a
community large enough to have water uses beyond strictly domestic consumption. As shown in
Table 8-4, water use for Basin 230 in 1997 for domestic, quasi-municipal, and commercial uses
totaled 942 acre-feet. This usage encompasses all demands except irrigation, mining, and other
commercial uses, and can be considered representative for the ranges of activities of a typical
small rural residential community in the region. As shown in the details of the 1997 report,
AVI97, these uses and activities include typical household uses such as drinking water; watering
of lawns and windbreaks; and small commercial operations such as gas stations and fast-food
stores. Because of local conditions, the population in the region does not grow significant
quantities of leafy vegetables, root vegetables, and fruit and grain crops for its own use.
The non-commercial use of 942 acre-feet was consumed by a total of 1,143 (estimated) residents
of the region. The average per-person use for the range of activities engaged in by this
population was therefore about 0.8 acre-feet per year. Dramatic changes in the demographic
characteristics of the population in the region is not expected in the future, so this value might be
taken as representative for non-commercial uses by future communities in the region.
The largest water use in the area down gradient from the repository is for irrigation, particularity
for the cultivation of feedstock (primarily alfalfa cultivation). Feedstock cultivation in the recent
past has varied in response to change in demand from local users, particularily the local dairy
industry, but has shown a general increase over the last ten years. The extent of cultivation is
expected to peak in the near future due to water limits on the water available for allocation.
Estimates of the number of acres under cultivation for feedstock production (largely alfalfa) are
given in Table 8-7. Water consumption for alfalfa cultivation varies as a function of soil and
weather conditions, and the number of harvests through the year. Records of historical water use
for alfalfa cultivation indicate a range of 2.7 to 5.0 acre-feet/acre, with 5 acre-feet/acre as the
8-39
-------
current allocation limit for this type of farming. Currently there are nine farms cultivating
feedstock, with acreage sizes ranging from approximately 65 to 800 acres.
Table 8-7. Estimates of Acreage Under Cultivation for Feedstock (DeL99, TRW96, TRW98)
YEAR
Acres Under Cultivation
1994
1,120
i£95
1,400
•' i*Sfc
1,750
-------
all of such future populations might be exposed to possible radionuclidcs in ground water coming
from the Yucca Mountain repository.
Table 8-8. Ground Water Storage Value's for Relevant Hydrographic Basins (USG88a)
HydrograpbicBasitt ;
Number * i
225
226
227-A
229
230
Hytfrographte Basin
Mercury V.
RockV.
Jackass Flats
Crater Flat
Amargosa Desert
Ground Water Storage in Upper
1 ft Saturation (AF> , ; * ^
minor
1,500
7,400
_
35,000
8.2.3.3 Soil and Topographic Constraints
The economics of well development and pumping costs are consistent with both the current
demographics and the historical development seen in Amargosa Valley. However, given the fact
that permanent settlement in the Valley has only occurred over the past century, it is important to
ascertain whether or not the inherent nature of the soils and/or the topography are conducive to or
constrain further development of the area. To obtain insights into this question, soil and
topographic characteristics in the immediate vicinity of the proposed repository site have been
obtained from the U.S. Department of Agriculture (USDA) and the U.S. Geological Service
(USGS).
From the topographic maps, slopes in the immediate vicinity of the proposed repository typically
exceed 15 percent, which would preclude large scale agricultural activities as they are currently
practiced in the Valley. Not only are slopes of this magnitude not amenable to heavily-
mechanized fanning methods, the soil on these slopes is very rocky. Small scale plots might be
feasible in the fan skirts and insets in the immediate vicinity, but such lands are not extensive.
Characterization of the soils at Yucca Mountain is incomplete because the USDA's survey of the
area stops at the boundaries of the Nevada Test Site and Nellis Air Force Base. However, from
the quadrangle map that includes Yucca Mountain (Busted Butte), the dominant soil type is a
gravelly, sandy loam that is found on the 2 - 8 percent slopes on the tops of low hills. It accounts
for about 45 percent of the land area. Another soil type is a very gravelly, sandy loam found on
the 15 - 30 percent slopes of upper fan piedmonts. It encompasses about 30 percent of the land
area. The Upspring series is very cobbly, sandy loam found on the 8 -15 percent slopes of low
8-41
-------
hills and accounts for about 15 percent of the land area. The remaining land area is accounted lor
by contrasting inclusions which include the Lealandic series, lithic Haplargids, and rock outcrop.
(Note: Appendix III provides detailed descriptions of the soil types.)
In summary, the gravelly, sandy loam found throughout the relatively flat portions of the area
downgradient from Yucca Mountain offers no inherent obstacle to agriculture. Given adequate
water, the same soil*types in the Amargosa Farms area produce crops for both commercial and
domestic use. (See Appendix III.)
8.2.3.4 Field Survey Findings
Water Resources and Current Agricultural Activities
As previously discussed, water resources of Amargosa Valley are currently over-allocated.
However, State officials have acknowledged that only a fraction of the allocated water rights is
being used and a review is currently underway to rescind permits in cases where water has not
been used in the past five years. Water consumption in the Amargosa Valley appears to vary
considerably over time. Data from Nye County indicate that, from 1985 to 1990, water usage
declined from 9,672 acre-feet to 4,109 feet (NYE93b, NYE93c, NYE93d, and NYE93e). State
water records show that in 1993, water usage rose to 11,300 acre-feet. Despite these fluctuations.
the majority of water use is for irrigation, with the second largest use being mining.
A tour of the Amargosa Valley performed for the preparation of this BID focused on current
agricultural activities, inclusive of those with a limited operating history. The first farm visited
had grown barley and alfalfa in 1995 with a yield of about 1.5 tons per acre per cutting. With
five to six cuttings per year, the yearly yield was estimated at 10 tons per acre. Recently, several
pistachio trees have been planted, which are expected to bear nuts within a few years (Photo #1).
For the future, the farm owner anticipates raising cattle and estimates that his land could produce
60 head of cattle per year.
A second and much larger farm that was visited (Funeral Mountain Ranch) also raised alfalfa.
Alfalfa grown here commercially is utilized as "green chop" for consumption by local dairy
farmers, baled and shipped to California (Photo #2), and dried/pelletized for shipment to Japan.
A third farm visited was a large dairy farm with 2,800 cows of which 2,300 are milk producers.
This farm has only been in operation for a few years. Its milk is shipped unprocessed into
8-42
-------
California. Due to the size of the operation, cows do not graze but are fed locally grown "green
chop." Due to the success of this farm, a second and third dairy farm of comparable size are
under construction.
The study team visited four additional farms that included the following:
A small farm raising pigs, sheep, and ducks (Photo #3)
A farm growing primarily vegetables that are sold locally
• A small fruit-tree orchard that was originally planted as an experiment to
determine the feasibility of growing apricots, peaches, and figs (Photo #4)
A sod farm, which ships and sells its products outside the Valley
8.2.3.5 New and Unusual Fanning Practices
Both ostrich and catfish farming have been identified as current fanning enterprises in the
Amargosa Valley. Along with hydroponic farming, these new and unusual farming practices are
described and assessed in Appendix V to determine their potential impact(s) on human exposure
modeling. The following paragraphs summarize the findings for each of these farming practices.
For practical and economic reasons, ostrich farmers do not allow their birds to range freely.
Rather, they are restricted to a confined area and fed pelletized commercial feed. As a result, the
ptake of radionuclides is confined to the consumption of contaminated water. Using radionuclide
transfer rates derived for poultry (ostrich-specific values are not available), the concentrations of
19 radionuclides were calculated on a per-unit weight basis. A comparison of these data with
radionuclide concentrations in beef indicate that 12 of the 19 radionuclides would be present in
lower concentrations in ostrich meat than in beef. Therefore, substituting ostrich meat from
farm-raised birds for beef in the dose/risk assessments is not expected to have a significant
impact on the results.
Warm climates favor fish fanning due to the fact that fish metabolism (and therefore growth rate)
increases with increases in ambient water temperature. In arid areas, fish farming is usually
/
conducted in large tanks filled with ground water that is continually filtered and aerated. Food,
in the form of commercial pelletized floating feed, is introduced into the tanks daily. The
extensive literature on concentration factors for radionuclides in freshwater fish is not considered
applicable to the unique conditions of aquaculture. Uptake is limited to direct sorption of
8-43
-------
1 Pistachio Orchard
"Large" Bales of Alfalfa
Pig, Sheep, and Duck Farm
-------
radionuclides in the water. Based on bioaccumulation factors adjusted to reflect direct sorption
as the only uptake mechanism, it is concluded that substituting catfish for beef in the dose/risk
assessments will not significantly affect the results.
In arid regions such as Yucca Mountain, hydroponic farming is conducted in hot-houses to avoid
dehydration and damage to the plants and their root systems. A quantitative assessment of the
potential impact of substituting hydroponically-grown vegetables for soil-grown vegetables is not
possible at this time. However, given that hydroponically-grown vegetables would not be subject
to the buildup of radionuclides in soil, it is reasonable to conclude that they would have lower
radionuclide concentrations than vegetables grown in soil.
SJ RADIATION PROTECTION OF INDIVIDUALS
In order to evaluate compliance of the repository system with regulatory requirements, potential
radiation doses to humans as a result of releases of radioactivity from the repository must be
calculated. This evaluation requires estimating radioactivity releases from the repository;
characterizing movement of the radioactivity through the environment; selecting and
characterizing the person(s) for whom potential radiation dose is to be evaluated; and
characterizing the interaction between the potentially-affected person(s) and the radioactivity in
the environment.
Information which provides the basis for estimating potential of radioactivity releases from the
repository and their movement through the environment is presented in Chapter 7 of this BID.
Information concerning past and present human occupation and use of the environment into
which the radioactivity could be released is presented in Section 8.2. This section of the BID is
concerned with identifying those individuals for whom the potential radiation dose is to be
evaluated, as well as their interaction with any radioactivity released from the repository. This
latter information can be used to estimate potential radiation doses to compare these values with
established regulatory limits.
Releases of radioactivity from the repository are expected to occur no sooner than several
thousands of years in the future; the start of release could be deferred on the order often
thousand years or more if certain repository design features are used (i.e., those aimed at delaying
the start of release) (see Chapter 7). After release from the repository, the radioactivity may
transit environmental pathways for long periods of time until it reaches the location of the
persons selected for the evaluation of potential doses. Radiation doses might first be incurred
8-45
-------
many thousands of years in the future, when locations and lifestyles of humans in the vicinity of
Yucca Mountain might differ from those of the present. Human locations and lifestyles far in the
future cannot, however, be reliably estimated. Therefore, evaluations of future potential radiation
doses are based on current patterns of human habitation and activities, as described in Section 8.2
of this BID.
8.3.1 The CriticaiLGroup Concept
Individuals in a human population may have widely different responses to radiation exposure
given differences in factors such as age and heritage. In addition, their potential to encounter
radiation released from a repository at Yucca Mountain will depend on factors such as where
they live and what they eat or drink. A wide range of radiation exposures and effects is therefore
possible. Means are needed to narrow and characterize the range for which evaluations of
compliance with regulatory requirements are to be made. Specification of the exposure
conditions to be considered must also be part of the regulations.
The NAS report on the technical basis for EPA's Yucca Mountain standards (NAS95)
recommended use of the critical group concept for the development of environmental standards.
The critical group concept was first introduced by the International Commission on Radiation
Protection (ICRP) in order to account for the variation in dose in a given population which may
occur due to differences in age, size, metabolism, habits, and environment. The critical group is
defined by the ICRP as a relatively homogeneous group of people whose location and lifestyle
are such that they represent those individuals expected to receive the highest doses as a result of
radioactive releases (ICR77, ICR85). As part of the critical group definition, the ICRP specifies
the following additional criteria:
Size - The critical group should be small in number and typically include a few to
a few tens of persons.
Homogeneity among members of the critical group - There should be a relatively
small difference between those receiving the highest and the lowest doses. It is
recommended that the range between the low and high doses not differ by more
than a factor often or a factor of about three on either side of the critical group
average.
Magnitude of dose/risk - It is suggested that the regulatory limit defined by a
standard exceed the calculated average critical group dose by at least a factor of
ten.
8-46
-------
Modeling assumptions - In modeling exposure for the critical group, the ICRP
recommends that dose estimates be based on cautious, but reasonable
assumptions.
The ICRP does not, however, prescribe the lifestyle, habits, or conditions of exposure that may
define a critical group into the future. Its generic recommendations suggest use of current
knowledge and use of cautious, but reasonable, assumptions for characterizing future exposure
scenarios.
According to current understanding, contaminated ground water is the principal pathway by
which a release of radionuclides from a repository at Yucca Mountain could cause radiation
exposures to humans. To determine the risk resulting from contaminated ground water to
exposed individuals requires the development of a comprehensive exposure scenario that
specifies discrete pathways and quantifies the intake of individual radionuclides. Depending
upon the potential uses of contaminated well water, prominent pathways for human exposure
may include internal exposure from the ingestion of contaminated drinking water, vegetables,
fruits, dairy products, and meats. For persons engaged in agricultural activities, internal exposure
may also result from the inhalation of airborne contaminants resuspended from soil that has been
irrigated with contaminated water. Over time, the buildup of soil contaminants could reach
levels that also yield significant external doses.
The selection of an exposure scenario that is appropriate for a specified critical group requires a
complex array of pathway parameter values that define potential radionuclide concentrations in
various media to which individuals may be exposed. Exposure scenarios must also provide
quantitative descriptions that include where individuals live, what they eat and drink, and what
their sources of food and water are. Many key parameters needed to model human exposures at
Yucca Mountain are highly site-specific and reflect the desert conditions of the sparsely-
populated Amargosa Valley. For example, the combined impacts of low rainfall, desert
temperatures, and soil quality mandate extensive irrigation of farm crops and use of local ground
water for cattle. Under these conditions, contaminated well water has the potential for unusually
high activity concentrations in all locally-grown food products.
8.3.2 Probabilistic Scenario Modeling
The unique requirements for modeling repository performance and human exposure scenarios
over times far into the future highlight the limitations, as well as uncertainties, in dose
8-47
-------
assessment methodology. The need to provide numerical values lor parameters that define
human exposure pathways is a major source of uncertainty.
To account explicitly for uncertainties, the NAS (NAS95) offered two probabilistic modeling
approaches. The first, described in Appendix C of the NAS report, A Probabilistic Critical
Group Approach, uses statistical methods and probability values to characterize members of the
critical group. The second, The Subsistence-Farmer Critical Group, described in Appendix D of
the report, also employs a probabilistic method, but identifies the subsistence farmer as its
principal representative of the critical group. A brief description of these two modeling
approaches is presented below.
Approach #1: A Probabilistic Critical Group. In support of ICRP recommendations to use
current knowledge and cautious, but reasonable, assumptions to identify the potential future
critical group and conditions of exposure, the NAS report (NAS95) suggested the following steps
for the Monte Carlo method used to implement a probabilistic assessment:
Step 1: Identify general lifestyle characteristics of the larger population that includes the
critical group.
Step 2: Quantify important characteristics, distributions of characteristics, and geographic
locations of the potentially exposed population.
Step 3: Based on findings in Steps 1 and 2, model radionuclide transport for estimates of
exposures to members of the critical group.
The first and second steps serve to identify the larger exposed population of which the critical
group is a subset. As noted previously, human exposure to ground water contaminants may
involve several exposure routes. Some routes are likely to be more important than others and
reflect the way in which contaminated water is used. Thus, specific information on location,
living patterns, lifestyles, and economic activities of potential members of the exposed
population can lead to the identity and characterization of the critical group of individuals at
greatest risk. Based on current understanding, principal factors affecting the magnitude of
individual exposure include 1) distance of residence from the repository, 2) level of dependence
on local well water, 3) use of local well water for drinking, crop irrigation, liyestock, etc., and
4) personal habits that affect food and water consumption. Consequently, if current population
data were to show that individuals at greatest risk involved a cluster of residents whose potable
8-48
-------
water was supplied by a common well, the critical group might consist of individuals with a
variety of lifestyles.
An important component of Step 3 is linking the critical group to future area(s) that will use
water from the contaminated aquifer. Potential exposures may range from relatively high levels
at locations near the footprint of the repository, because of the lack of dilution of ground water
contamination, to lower levels of contamination and correspondingly lower doses at greater
distances. Actual exposures will depend on the rate of migration and dilution of contaminated
ground water as a function of distance and direction from the repository. For each human
habitation location, specific data should be sought to define the slope/topography of the land,
quality of soil, depth to ground water, well productivity, and other factors. Taken collectively,
data for each location can be used to determine its suitability and probabilistic future use for
farming, residential, commercial, industrial, and other purposes that may affect the exposure
levels of members of the critical group. To account for probabilistic land use, local ground water
dependence, and numerous model parameter uncertainties, the NAS (NAS95) recommended that
probabilistic distributions of doses/risks be based on Monte Carlo simulations. In this method,
data on the frequency distribution are sampled to provide input to generic model equations. In
effect, the Monte Carlo method produces a single predicted value for each set of randomly
selected parameter values. The results of numerous (hundreds to thousands) iterations of model
solutions are then statistically analyzed to determine their distribution. From the distribution of
predicted values, information is extracted that defines the best estimate of an average value (i.e.,
the most probable value), the range of potential values, and a measure of uncertainty of model
predictions that reflects the collective uncertainties of input parameters (HEN92).
Approach #2: The Subsistence Farmer Critical Group. The model described in Appendix D of
the NAS report specifies a priori one or more subsistence farmers and makes assumptions
designed to define the farmer at maximum risk as representative of the critical group.
Subsistence fanning does not exclude commercial farmers who, in addition to cash farm
products, raise food for personal consumption. The NAS assumed the subsistence farmer of the
future would have nutritional needs consistent with those of a present-day person. Like the
subsistence farmer of today, most or all drinking water would be obtained from an on-site well
that would also be used in the production of all consumed food. The subsistence farmer is also
assumed to live his/her entire life at the same location. Thus, the magnitude of the dose to a
subsistence farmer will largely be defined by the distribution of radionuclide concentrations in
ground water at the point of water withdrawal.
8-49
-------
Each of these two approaches to defining the critical group has its advantages and disadvantages.
For example, a standard that incorporates the probabilistic critical group would accommodate
variabilities, but this approach is relatively complex and difficult to implement. Moreover, the
assignment of probability values relating to land use, demands on natural resources, and human
activities to the probabilistic critical group may be viewed as subjective and potentially biased by
the limitations that define present-day society.
The subsistence farmer approach is relatively simple and easy to use and understand, but it may
be formulated to be unrealistically or insufficiently conservative. For example, the subsistence
farmer could be assumed to use a well at the repository boundary, where radioactivity
contamination levels are highest, even though this location is unsuitable for fanning.
Alternatively, the subsistence fanner could be located and characterized such that projections of
radiation dose potential are unrealistically low.
8.3.3 Exposed Individuals and Exposure Scenarios for Yucca Mountain
EPA has developed a method for estimating potential radiation doses based on the concept of the
reasonably, maximally exposed individual (RMEI). The RMEI concept, which involves
estimating the dose to a person assumed to be at greatest risk based on reasonable (i.e., not overly
or insufficiently conservative) assumptions, has been used in previous agency programs and
guidance (EPA 92). For example, the National Emission Standards for Hazardous Air Pollutants
(NESHAPS, 40 CFR Part 61), require estimation of dose to a person assumed to reside at a
location where the highest dose would be received.
The basic approach for estimating doses to the RMEI is to identify and characterize the most
important exposure pathway(s) and input parameters. By using maximum or near-maximum
(i.e., 95th percentile) values for one or a few of the most sensitive parameters, while assuming
average values for others, it can reasonably be assumed that the resulting dose estimates
correspond to the near-maximum exposures that could be received by any member of the
exposed population. The ultimate objective of the approach is to define an exposure that is well
above average exposures, but within the upper range of possible exposures.
The EPA expects to use the RMEI approach as the basis for radiation protection standards for
Yucca Mountain. The concept is consistent with distinct patterns of human lifestyles, locations,
and activities currently characteristic of the Yucca Mountain region, as described in Section 8.2
of the BID. The subsistence farmer approach, described in Appendix D of the NAS report and
8-50
-------
summarized above, is similar to the RMEI approach in that it utilizes specific exposure scenarios
and parameters, rather than the probabilistic approach described in Appendix C of the report.
The EPA has defined four basic scenarios for estimating potential exposures to the RMEI in the
Yucca Mountain area. These scenarios represent current human habitation patterns and lifestyles
in the Yucca Mountain region. They are considered to be scenarios characteristic of the region
based on local climatic, geologic, and hydrologic conditions. The four scenarios are summarized
below.
(1) Subsistence (low technology) Farmer. In this scenario, the farmer is assumed to live in the
Yucca Mountain area and to be exposed chronically (both indoors and outdoors) to residual
concentrations of radionuclides in soil through all exposure pathways. The location and habits of
this individual will be consistent with historical locations, easily accessible water, and new
locations based on our current state of knowledge.
(2) Commercial Farmer. Based upon economic factors and current technologies, certain areas
around Yucca Mountain are suitable for commercial crop production. These areas are either
currently being farmed or could be economically viable based upon reasonable assumptions,
current technology, and experience in other parts of the arid west. In addition, some parts of the
region could possibly support up and coming technologies such as hydroponic applications and
fish farming. Exposure pathways in this scenario are the same as those described for the
subsistence farmer.
(3) Rural/Residential Person. In this scenario, individuals are assumed to live around Yucca
Mountain and to be exposed through the same pathways described for the subsistence farmer in
Scenario 1. However, in this case the residents are not assumed to be full-time agricultural
workers. Instead, it is assumed that these individuals work primarily out of the area and engage
only in light farming and recreational activities within. Furthermore, it is assumed that at least
50% of the locally-grown produce, meat, milk, and fish consumed by these individuals comes
from the vicinity of the site.
(4) Domestic Use of an Underground Drinking Water Supply. Based upon current water usage
in the arid west, it appears to be reasonable to assume that there could be an hypothetical water
supply near or on the Nevada Test Site which could serve a community living near the repository
site. In this scenario, sites will be identified which, under reasonable assumptions, could provide
drinking water to support a future community.
8-51
-------
For each of these four scenarios, the following exposure pathways can be evaluated:
• External radiation from radionuclides in soil
• Inhalation of resuspended soil and dust containing radionuclides
• Inhalation of radon and radon decay products from soil containing radium
• Incidental ingestion of soil containing radionuclides
• Ingestion of drinking water containing radionuclides transported from soil to
potable ground water sources
• Ingestion of home-grown produce contaminated with radionuclides taken up from
soil
• Ingestion of meat (beef) or milk containing radionuclides taken up by cows
grazing on contaminated plants (fodder)
• Ingestion of locally-caught fish containing radionuclides
8.3.4 Details and Analyses for the Subsistence Fanner Scenario
This section provides a detailed discussion of the subsistence fanner scenario. It includes the
comparison of results of analyses of this scenario that were obtained by several sources, as well
as the comparison of the parameters, methodologies, and assumptions used to conduct the
analyses. The purpose of this section is to illustrate the range of factors involved in
characterizing a scenario and the variability of analysis input parameters and results for a given
scenario.
As noted above in Section 8.3.3, the Subsistence Fanner Scenario is one of four scenarios
selected by EPA to provide part of the basis for the Yucca Mountain standards. The NAS also
described a subsistence farmer scenario in Appendix D of its report (NAS95), and the Center for
Nuclear Waste Regulatory Analyses (CNWRA) performed analyses of the "resident farmer"
scenario for NRC. These scenarios are highly similar and can be summarily characterized as the
self-sufficient farmer scenario. Details of the assumptions used to characterize the scenario vary,
but its basic concept is that the farmer grows all of his/her own food and obtains all needed water
from a well on his/her property. This scenario corresponds to one type of current lifestyle in the
Yucca Mountain region and, depending on assumptions, such as the location of the farm relative
to the repository, may correspond to EPA's RMEI for the site.
t
8-52
-------
In EPA's version of the scenario, which is reflects current farming activities in the Yucca
Mountain region, the fanner is assumed to grow alfalfa which is used as feed for beef cattle and
milk cows. The farmer has a garden lot which grows vegetables, fruits, and grain sufficient for
home use. In sum, all food consumed by the farmer is assumed to be home-grown. In addition,
all water needed for personal consumption, crop irrigation, and drinking by livestock is assumed
to be obtained from a single on-site well.
The pathways by which radioactivity released from the repository to the environment can
produce radiation doses to this farmer are illustrated in Figure 8-10. As shown, the radioactivity
carried from the repository by ground water can distribute in the environment to produce
ingestion, inhalation, and external radiation doses to the exposed individual via a variety of
environmental pathways.
The relative importance of the three types of dose sources to the total dose incurred will depend
on the specific details of the individual's lifestyle in terms of interaction with the environmental
pathways, the quantities and types of radionuclides available in each of the pathways, and the
rates and means by which the radionuclides move within the pathways. Over many years,
extensive research has produced data on how radionuclides move in the environment and the
pathways for human exposure. These data are expressed in terms of parameters such as
concentration factors and transfer coefficients. These parameters are used as input to computer
codes which model radionuclide transport and predict radiation doses.
Quantitative characterization of the pathways and parameters represented in Figure 8-10, for the
purpose of estimating radiation dose for the subsistence farmer or any other scenario, has been
accomplished using the GENII and GENII-S code. Parameter values were obtained from various
sources. As originally developed, the GENII codes implemented the NRCs Regulatory Guide
1.109 and dosimetry models recommended by the ICRP. Parameter values were selected that
were appropriate for the Hanford site, but the codes were "...designed with the flexibility to
accommodate input parameters for a wide variety of generic sites" (NAP97). The codes have
been subjected to rigorous peer review and meet the benchmarking requirement as defined by the
American Society of Mechanical Engineers (ASM89a, ASM89b).
The GENII-S codes were originally developed by Sandia National Laboratories for use in
performance assessments for the Waste Isolation Pilot Plant (WIPP) (DOE93a). The GENII-S
system includes menu-driven programs that assist with scenario generation and input data
requirements. Values for parameters can be selected from appropriate sources. An important
8-53
-------
G.
R
0
ir
N
D
W
A
T
E
R
C
0
N
C
E
N
T
R
A
T
I
0
N
Drinking Water
(Well)
- Coniuff^lto
n RoU
Irrigation Woter
- IrriQOtioA Rate1
- Irrigation Our o lion1
livestock Uptake
C»Mumpli»A
*' Uv
Crop Uptok*
time
- lnie(«ptl»n
> tiunii«««Uon I'aetoi
- Irfx «l Ci'ip" Crown i
Liveslock
ConcenUolion
(meol ond milk)
Crop Conc«nirotton
N
0
0
s
E
0
N
.Aaltl*
radon
Soil Concenlrotion
- itoch Roii>
- sen O
/
Rejuspension
- Rttwtftintion roelcr-
- DepatlUan Velocity
Inhalolion
Oose
- 8'tClhlna, R«tt
- Ooit
T
0
T
A
L
D
0
S
E
. Identifies poromtlers Ihol'ore best defined by site-ipfrifh- »oiuos
Figure 8-10. Ground Water Pathway Model for Subsistence Farmer
8-54
-------
feature of GENII-S is its versatility in performing either deterministic or stochastic analyses in
which probabilistic distributions are assigned to the values of the input parameters. The
stochastic analyses provide opportunity to assess sensitivities and uncertainties for results in
terms of variations in input parameters. GENII-S is considered by the NAS Committee to be a
proven code for dose estimates. As described below in Section 8.3.4.1, it was used by CNWRA
to perform illustrative dose assessments for the NRC.
8.3.4.1 The CNWRA Quantification of a Subsistence Farmer Scenario
As part of its development of capability to review a license application from DOE for a
repository at Yucca Mountain, NRC, with technical assistance from the CNWRA, has performed
analyses of potential radiation doses to receptor groups in the vicinity of Yucca Mountain. The
analyses considered an Amargosa Valley resident farmer group, with life style similar to EPA's
RMEI subsistence farmer scenario, and a non-farmer receptor group located between five km and
20 km to the south of the proposed repository site. Analysis results were obtained for assumed
unit concentrations of radionuclides in ground water and soil (e.g., rem per pCi/1 of ground
water). Actual doses would depend on actual radionuclide concentrations in these media.
The principal objectives of the analyses were to: 1) summarize and document site-specific
characteristics and parameters used to model environmental pathways; 2) assess the relative
importance to dose of the pathways identified in Figure 8-10; and 3) provide values for dose
conversion factors (DCFs) for use in performance assessments for the Yucca Mountain site. The
scope of work also included calculation of the annual total effective dose equivalent (TEDE) for
individuals. Results were first reported in 1995 (CNW95) and an update report was issued in
1997 (CWN97).
The 1997 CNWRA report presented updated values for pathway parameters, calculated dose
conversion factors based on recent information, and a Monte Carlo-based sensitivity analysis
used to identify parameters having the greatest impact on dose evaluations. The GENII-S code
was used to make the calculations.
Brief descriptions of the parameters addressed in the CNWRA analyses are presented below.
These descriptions illustrate the scope of parameters considered, the means used to quantify the
parameters, and the levels of DCF-value results obtained for unit concentrations of the
radionuclides and assumed parameter values in the GENII-S code.
8-55
-------
AH numerical values presented should be considered to be only illustrative. The analyses were
done to demonstrate capability and methodology; changes in parameter values and results are to
be expected as a result of future updates to the information base for the calculations.
Agricultural Data Used
Subsistence farming, as described in the CNWRA reports, is consistent with present-day
agricultural practices for Nye County and incorporates all of the exposure pathways shown in
Figure 8-10. Crops likely to be grown in a local vegetable garden were selected (MLS93) and
grouped into categories of leafy vegetables, fruits, and root vegetables, as directed by the GENII-
S code. Growing times for each group of crops were selected from Chambers and Mays
(CHA94) and reflect climatic conditions consistent with those at the Yucca Mountain site.
Conversion ofAreal Deposition into Mass Concentration
Contamination of feed and food crops is directly related to the areal deposition of contaminants
by water irrigation. To convert the amount deposited by irrigation per unit area (areal
concentration, CA) into the mass concentration, CM (per units mass) of soil or vegetation, it is
necessary to divide the areal concentration by the soil density per unit area or by the mass of
vegetation per unit area (termed the vegetation density or, in the case of farm products, the yield
or weight per unit area). In soil, it is necessary to specify the depth of interest.
The areal soil density is given by pA = pz, where p is the soil density (typically 1.6 to 2.6 g/cm3
or 1600 to 2600 kg/m3) and z is the depth of interest (cm or m). For determining uptake by
plants from soil, root depths of 0.15 to 0.2 m are common, so that areal soil densities of 240 to
520 kg/m2 are reasonable. This calculational process assumes a uniform distribution of the
radionuclide with depth, which would be typical of tilled soil used for agriculture.
In the case of vegetation, the mass concentration CM is obtained by dividing the areal
concentration by the vegetation density YD (kg/m2). As the amount of water in vegetation is
highly variable and dependent upon collection and storage techniques, use of the dry-weight yield
or dry vegetation density [kg(dry)/ni2] is preferable.
For Yucca Mountain, site-specific soil density values suggest a range of 1.2 to 1.8 g/cm2, which
for a 15 cm depth corresponds to a surface areal density of 180 to 270 kg/m2. While this range of
values appears appropriate for most vegetative pathways, alfalfa's tap root can grow to depths of
8-56
-------
*veral feel (ST191). The GEN11-S code uses a two-compartmenl soil model thai accounts for
iflercnces in densities between the tilled layer and the denser lower layer. The higher density of
Ac lower layer has less pore space available to hold water available for root uptake. For allalfa,
iconservative decision was made. A single-compartment soil model represented exclusively by
lhe upper layer was selected.
Crop Yields
Estimates of crop yields were based on data provided by the Nevada Agricultural Statistics
Service, which reflect state-wide yields for wheat, barley, potatoes, garlic, and onions. From
fee data, crop yields for leafy vegetables and root vegetables were assumed to have a range of
1618 to 6.47 kg/m2 and 0.769 to 20.8 kg/m2, respectively. Because fruits are not commercially
produced in this region, estimates of yields between 0.3 and 2.0 kg/m2 were obtained from
Snyderetal. (DOE94b).
>o/7 Interception Fraction
foe crop interception fraction refers to the fraction of contaminants in irrigation water deposited
n the plant surface. The interception fraction varies among crops and with geographic location.
Based on laboratory and field study data that included the Yucca Mountain site (LLL87), values
were assumed to range from 0.06 to 1.0 with a best estimate of 0.40. While a value of 1.0
represents a theoretical upper limit, it is not considered excessively conservative in instances of
high-density vegetation which is characteristic of home gardens; instances in which irrigation is
employed judiciously; and instances of high evaporation rate.
In summary, parameters selected for characterizing Yucca Mountain soil, crop growing times,
crop interception fractions, and crop yields are representative of site-specific values or fall within
the range of values cited in the scientific literature.
Food Transfer Factors
Food transfer factors quantify the amount of contaminants that may sequentially be transferred
from soil to vegetation. When vegetation is used for animal feed, transfer factors must also be
used to estimate contamination levels in meat, milk, and other food products. Radionuclide
uptake by plants from soil has generally been described by an empirical concentration ratio, CR,
which is defined as:
8-57
-------
CR =
Radionuclide activity per unit mass of plant
Radionuclide activity per unit mass of soil
The radionuclide soil concentrations are generally expressed in terms of oven-dried soil weight.
However, the radionuclide concentrations in crops are reported both in terms of fresh (or wet)
weight and dry weight. The relationship between the fresh and dry concentration ratios is:
5, = CR (fresh) =
FWIDW
where FW and DW correspond to the fresh and dry weight, respectively. Due to variations in
water content, the ratio of fresh to dry weights among leafy vegetables may vary from a low of
about seven for Brussels sprouts to a high of 20 for lettuce. For root vegetables, ratios range
from four (potatoes) to 18 (radishes) (NRC83).
The feed-to-beef and feed-to-milk transfer factors are also empirically derived constants that
establish a relationship between the amount of an element (or radionuclide) that cattle and milk
cows ingest daily and the concentration of that element in edible meat or milk at equilibrium.
The standard units are expressed as a ratio of pCi/kg of meat or milk to pCi/day of chronic intake
contained in feed.
Concentration ratios and food transfer factors are affected by many processes and factors, some
of which are site-specific. Empirical data for the area around the Yucca Mountain site were
reviewed in the CNWRA 1995 analyses, but are considered to be incomplete. Element-specific
transfer factors used in the CNWRA 1997 analyses (Table 8-9) were obtained from data
published by the International Atomic Energy Agency (IAE94), the International Union of
Radioecologists (IUR89), and by Oak Ridge National Laboratory (ORN82).
In EPA's reviews of the CNWRA reports, transfer factors selected for analysis by CNWRA were
compared to values that are employed by EPA NESHAPs, cited in the literature, or used as
default values in other computer codes. Based on this review, it appears that except for iodine,
the plant transfer values used by CNWRA are consistent with other values and appear
appropriate for modeling the Yucca Mountain site. For iodine, CNWRA's plant transfer factors
appear low by one to two orders of magnitude. With one exception, beef and milk transfer
factors values cited for analysis also appear consistent with commonly-used values. There
8-58
-------
appears to be a substantial discrepancy in the milk transfer factor for technetium. The cited value
of 1.4E-4 is nearly 100-fold lower than those recommended by the NRCs NUREG/CR-5512,
EPA's NESHAPs, and the computer code PRESTO.
Table 8-9. Concentration Ratios and Transfer Coefficients By Element (Source: CNW97)
i
f s <:
Element
Ac
Am
Cs
Cm
I
Pb
Mo
Np
Ni
Nb
Pd
P
Pu
Po
Pa
Ra
Sm
Ag
Se
Sr
Tc
Th
Sn
U
Zr
1 " Concentration Itafio \
I*af;p
Vegetables
3.5E-03
1.2E-03
1.1E-OI
l.IE-03
3.4E-03
1.1E-03
8.0E-01
6.9E-02
I.8E-01
5.0E-02
1.5E-01
4.0E-00
3.4E-04
l.OE-02
2.5E-03
8.0E-02
l.OE-02
2.7E-04
2.5E-02
1.1E-00
7.6E+01
I.1E-02
3.0E-02
2.3E-02
l.OE-03
Otber
V&eta&fes
3.5E-03
4.7E-04
7.2E-02
5.8E-04
2.0E-02
6.4E-03
8.0E-01
2.7E-02
3.0E-02
1.7E-02
1.5E-01
4.0E-00
2.3E-04
l.OE-02
2.5E-03
I.3E-02
l.OE-02
1.3E-03
2.5E-02
S.6E-01
1.1E+01
3.IE-04
3.0E-02
1.1E-02
l.OE-03
#r«&
3.5E-03
4.7E-04
7.2E-02
5.8E-04
2.0E-02
6.4E-03
8.0E-01
2.7E-02
3.0E-02
1.7E-02
1.5E-01
4.0E-00
2.3E-04
l.OE-02
2.5E-03
1.3E-02
l.OE-02
8.0E-04
2.5E-02
2.0E-01
1.1E+01
3.1E-04
3.0E-02
1.1E-02
l.OE-03
Grain*
3.5E-03
2.2E-05
l.OE-02
2.1E-05
2.0E-02
4.7E-03
8.0E-01
2.7E-03
3.0E-02
1.7E-02
1.5E-01
4.0E-00
8.6E-06
l.OE-03
2.5E-03
1.2E-03
l.OE-02
1.5E-01
2.5E-02
1.2E-01
7.3E-01
3.4E-05
3.0E-02
1.3E-03
l.OE-03
v. •* * ••
Transfer Coefficient
Betf
2.5E-05
4E-05
5EO-2
3.5E-06
4E-02
4E-04
1E-03
1E-03
5E-03
3E-07
4.0E-03
5.0E-02
1E-05
4E-03
l.OE-05
9E-04
5.0E-03
3E-03
1.5E-02
8E-03
1E-04
6.0E-06
8.0E-02
3E-04
1E-06
^ V
^^^ ;
. wm. \
2.0E-05
1.5E-06
7.9E-03
2.0E-05
l.OE-02
2.5E-04
2.0E-03
5.0E-06
1.6E-02
4.1E-07
l.OE-02
I.5E-02
1.1E-06
12E-04
5.0E-06
1.3E-03
2.0E-05
5.0E-05
4.0E-03
3.0E-03
1.4E-04
5.0E-06
l.OE-03
4.0E-04
6.0E-07
B^g
2.0E-03
4E-03
4E-01
2.0E-03
3E+00
8.0E-01
9E-03
2.0E-03
l.OE-01
1E-03
4.0E-03
l.OE+01
5E-04
7.0E-K)0
2.0E-03
2.0E-05
7.0E-03
5.0E-01
9E+00
2E-01
3E+00
2.0E-03
8.0E-01
1E+00
2E-04
8-59
-------
Water Consumption Rales for Drinking and Irrigation
A search for site-specific or representative values failed to yield useful information regarding
water consumption rates for hot, dry regions, such as southwestern Nevada. To model potential
exposure from drinking contaminated water, CNWRA selected values from a nationwide Food
and Drug survey (ROS92). Results indicated that water consumption rates in the United States
are distributed log normally, yielding a geometric mean of 349 liters per year (L/yr) and a
geometric standard deviation of 1.78.
These values are low when compared to generic values recommended by EPA (EPA89), which
assume an average value of 511 L/yr (1.4 liters per day) and a 90th percentile value of 730 L/yr,
or 2 liters per day. A 2-liter per day consumption rate is, at the 90th percentile, conservative for
most conditions. The consumption rates for inhabitants of arid environments may be nearer to
the conservative 2-liters per day rate than the overall average of 1.4 liters per day.
Contaminated water used for irrigation is also a potentially important pathway for human
exposure. Previously, it was noted that the Amargosa Valley region south of the Yucca
Mountain site currently uses ground water for agricultural irrigation. This supports the
likelihood of its future use by a subsistence farmer. However, suitable data do not currently exist
for quantifying irrigation rates associated with small-scale or subsistence farming. For surrogate
values, the CNWRA relied on water irrigation required for lawn maintenance in Nye County
(MLS93). Data suggest a range of values between 26 and 84 inches per year, which corresponds
to a growing season of six to 12 months per year.
A comparison of physiological parameters between lawn grasses and edible crops (moisture
content, surface to volume ratio, root depth, etc.) suggests that these surrogate values provide a
reasonable approximation. The appropriateness of these values is also supported by the fact that
the range of irrigation values generally approximates the natural precipitation rates in parts of the
country where irrigation is not required.
Buildup of Soil Contaminants
The chronic irrigation of farm land with contaminated water may result in a buildup of
contaminants over time. A buildup of soil contaminants affects the ingestion pathways involving
8-60
-------
edible crops and animal feed, the inhalation exposure from resuspended soil particles, and the
external exposure from contaminated ground surfaces.
Estimates of soil buildup are complex and reflect the rate of deposition of a contaminant and its
rate of removal. Soil contaminants may be removed by several concurrent processes that include
leaching (or washoff), crop uptake (and subsequent harvesting of crop), wind erosion, and
radionuclide decay. These biophysical processes can be combined and represented by a simple
removal rate constant. Removal rate constants, however, are not easily determined since they are
radionuclide-specific and affected by a host of site-specific parameters.
Soil buildup was not modeled in the unit concentration dose estimates for the 20 radionuclides
analyzed by CNWRA. Dose estimates, in effect, reflect the combined annual external exposure
and committed internal exposures associated with soil irrigation for a period of one year. The
potential impact of neglecting soil buildup is likely to vary depending on the radionuclide. When
radionuclides are assumed highly soluble and exhibit high leach rates, buildup may be
insignificant. For this condition, contamination of food crops may be dominated by external
deposition on vegetative (leaf) surfaces.
The EPA evaluated the lack of consideration of soil buildup in the CNWRA analyses. Test runs
were performed using leaching removal terms provided by the GENII code. For radionuclides
with high leach rates, such as 1-129 and Tc-99, dose estimates were unaffected by long-term
irrigation, indicating that soil buildup was insignificant. For other radionuclides, doses were not
significantly affected for irrigation times of 100 and even up to 1,000 years. With irrigation
periods lasting 10,000 years, however, soil buildup for a limited number of radionuclides yielded
a ten-fold increase in dose estimates. An irrigation period of 10,000 years is extremely
conservative and unrealistic. In addition, the agricultural potential of irrigated land in the arid
environment around Yucca Mountain would decrease markedly with time due to build up of salts
in the soil from the high evaporation rate.
An independent assessment by the Electric Power Research Institute (EPRI) cited provisional
calculations for 1-129 and Np-237 (EPR94). The EPRI results confirm that for the ground water
release scenario, soil buildup from irrigation is insignificant. EPRI concluded that the need to
assess the impact of long-term soil accumulation is limited when radionuclides can be assumed
to be relatively soluble.
8-61
-------
Inhalation and Soil Exposure Times
In the CNWRA analyses, inhalation and soil exposure times are assumed to be the same. The
maximum exposure time of 7,117 hr/yr is based on an individual spending 15 hours per day,
seven days per week outdoors in the contaminated area (i.e., a 1.0 exposure factor). The rest of
his/her time is spent indoors, exposed to contamination 50 percent of the time (i.e., an exposure
factor of 0.5). The minimum exposure time of 5,548 hours per year is based on spending 73
percent of the time indoors with a 0.5 indoor exposure factor and the remaining portion of the
time outdoors in the contaminated area. A triangular distribution was assumed, sloping to the
minimum value from the maximum value, based on the assumption that the fanner is likely to
spend much of the day outdoors.
The CNWRA Selection of Dose Conversion Factors
Estimates of radiation doses that result from the ingestion, inhalation, or external exposure to
radioactivity are based on dose conversion factors (DCFs) that make use of contemporary
metabolic models and dosimetry methods. A critical choice in selecting DCF-values for dose
calculations relates to the solubility of the radionuclide contaminant in aqueous fluids. When
ingested, a radionuclide that is highly soluble is readily absorbed from the intestinal tract to the
blood stream where it may be metabolized and retained for long periods of time. Similarly, a
soluble contaminant that is inhaled may also be quickly removed from the lung and enter the
blood stream where its fate is essentially that of an ingested radionuclide. For ingestion and
inhalation pathways, DCFs are generally defined in terms of solubility by means of the f,
(fractional uptake of nuclides from small intestine) and lung clearance class. Lung clearance,
designated as D,W, or Y, refers to "days, weeks, or years" for the radionuclides to be removed
from the pulmonary region of the lung.
For insoluble contaminants, the potential for internal exposure through inhalation versus
ingestion is quite dissimilar. Inhaled insoluble radionuclides are not readily removed from the
lung and may, therefore, result in long-term exposure of the lung and other tissues. Internal dose
is minimized, however, when the insoluble contaminant passes through the digestive system
without being absorbed.
The GENII code offers a choice of DCFs that correspond to very soluble, soluble, and insoluble
states of individual radionuclides. A comparison of GENII DCF values with those in EPA
8-62
-------
Federal Guidance Reports No. 11 and No. 12 (EPA88, EPA93) shows that, when matched for
solubility, there is generally good agreement.
As previously noted, the CNWRA 1997 report (CWN97) presented revisions to some of the
pathway parameter values used in the 1995 report and described the basis for the revisions. The
parameters were then used to evaluate dose conversion factors for several exposure scenarios,
with unit concentration values assumed for the radionuclides. The scenarios considered were the
current-day and pluvial-biosphere resident farmer for ground water and soil sources, and the non-
farming resident for the same types of scenarios. The GENII-S code was used to make the
calculations. Table 8-10 reproduces the set of the results obtained for the resident farmer in the
current biosphere with a contaminated ground water source.
The CNWRA 1997 report states that the farmer scenario is conservative in that it is unlikely that
there would be a group that would be expected to receive higher exposures. In the CNWRA
analyses, the parameter values for calculation of the DCFs were set to average values, so the
DCF results obtained, such as those shown in Table 8-9, represent the average member of the
most highly-exposed critical group as defined by the ICRP. The approach is described by
CNWRA as a shortcut to the computationally-intensive methods that would be required by the
concepts put forth by the NAS (NAS95).
To the extent that the details (e.g., lifestyle and pathway parameters) of the CNWRA subsistence
farmer scenario correspond to those used by EPA for its subsistence farmer RMEI, the unit-
concentration DCF results obtained from the CNWRA analyses correspond to those obtained by
EPA. Actual doses predicted to be incurred by the RMEI would depend on actual biosphere
concentrations predicted from performance assessments (Chapter 7 of this BID) compared to the
unit concentrations used to evaluate the DCFs.
8.3.4.2 Summary of CNWRA Analysis
As previously noted, CNWRA performed an analysis of the subsistence farmer scenario using the
GENII-S code. Input parameter values were taken from available sources that compile and
interpret past research to develop such values. Variations in input parameter values, which lead
in part to variations in analysis results, are also discussed in the CNWRA report.
Table 8-11 shows the CNWRA results for evaluating the arithmetic and geometric means for
TEDEs for selected nuclides.
8-63
-------
Table 8-10. Dose Conversion Factors for a Resident Fanner in Current Biosphere By Exposure
Pathway and Radionuclide for Ground Water Source (rem per pCi/1 in ground water) (CN WRA)
1 "
Radtanudide
C14
C136
Ni59
Ni63
Se79
Sr90
Zr93
Nb94
Mo 93
Tc99
Pd 107
AgllOM
Sn 121M
Snl26
I 129
Cs 135
Csl37
SmlSl
Pb210
Ra226
Ac 227
Th229
Th230
Pa 231
U232
U233
U234
U235
U236
U238
Np237
Pu239
Pu 240
Pu 242
Am 241
Am 242M
Am 243
Cm 243
Cm 244
Cm 245
|| Cm 246
Atoimai:;,||ll Drinking
Frodftct^x Water
tngestion H Ingcstion
3.2E-06
2.0E-05
6.6E-07
1.8E-06
1.1E-05
1.2E-04
3.2E-10
6.7E-10
5.9E-07
1.4E-07
3.1E-07
3.7E-06
6.9E-06
6.4E-05
8.5E-04
2.3E-05
1.6E-04
7.3E-08
4.0E-04
2.7E-04
7.9E-05
2.9E-05
6.4E-07
1.4E-05
6.8E-06
2.4E-06
2.4E-06
2.5E-06
2.2E-06
2.3E-06
2.0E-04
2.9E-08
3.0E-08
2.8E-08
6.6E-06
6.3E-06
6.5E-06
9.9E-06
7.9E-06
1.5E-05
1 5E-05
1.5E-06
2.2E-06
1.5E-07
4.0E-07
6.1E-06
8.8E-05
1.2E-06
5.3E-06
l.OE-06
1.6E-06
1.1E-07
2.4E-05
1.1E-06
1.4E-05
1.8E-04
5.0E-06
3.5E-05
2.8E-07
3.9E-03
7.0E-04
l.OE-02
2.6E-03
3.9E-04
7.8E-03
5.0E-05
1.9E-05
1.9E-05
1.9E-05
1.8E-05
1.7E-05
3.8E-03
3.6E-05
3.6E-05
3.4E-05
2.7E-03
2.5E-03
2.6E-03
1.8E-03
1.5E-03
2.7E-03
2.7E-03
External
jp]ume«nd
tiroundshine
O.OE-00
7.9E-09
O.OE-00
O.OE-00
7.3E-IO
9.3E-09
1.6E-09
5.3E-05
1.7E-07
6.8E-10
O.OE-00
5.6E-05
1.7E-07
6.7E-05
5.3E-07
1.2E-09
1.9E-05
1.8E-10
1.3E-07
5.6E-05
1.3E-05
1.1E-05
2.6E-08
1.4E-06
7.3E-06
2.5E-08
2.6E-08
5.8E-06
2.3E-08
8.4E-07
6.3E-06
1.3E-08
2.8E-08
1.1E-07
9.6E-07
7.0E-07
7.5E-06
4.4E-06
3.1E-08
3.0E-06
2.7E-08
iS\s %\
«><. ••* -
^, *^
Inhalation
O.OE-00
1.9E-12
2.4E-12
6.0E-I2
2.6E-11
5.3E-10
2.2E-10
l.OE-09
2.5E-12
5.6E-12
3.3E-11
1.5E-10
3.0E-11
2.5E-10
2.4E-10
1.2E-11
7.8E-11
7.8E-11
4.8E-08
2.2E-08
3.6E-06
4.6E-06
7.0E-07
2.3E-06
1.8E-06
3.5E-07
3.5E-07
3.2E-07
3.3E-07
3.1E-07
1.5E-06
8.2E-07
8.2E-07
7.9E-07
1.2E-06
1.1E-06
1.2E-06
8.0E-07
6.4E-07
1.2E-06
1.2E-06
Terrestrial
Crop
Ingcstton
8.9E-06
2.3E-05
2.0E-07
5.3E-07
8.0E-06
1.4E-04
1.6E-06
7.0E-06
1.6E-06
3.0E-06
1.5E-07
3.0E-05
2.1E-06
2.0E-05
2.4E-04
6.6E-06
4.6E-05
3.6E-07
5.3E-03
9.2E-04
1.4E-02
3.6E-03
5.2E-04
l.OE-02
7.9E-05
2.5E-05
2.5E-05
2.6E-05
2.3E-05
2.9E-05
5.0E-03
4.8E-05
4.8E-05
4.5E-05
3.5E-03
3.3E703
3.5E-03
2.4E-03
1.9E-03
3.6E-03
3.6E-03
N
Total EDE
1.4E-05
4.5E-05
l.OE-06
2.8E-06
2.5E-05
3.4E-04
2.8E-06
6.6E-05
3.4E-06
4.8E-06
5.7E-07
1.1E-04
l.OE-05
1.6E-04
1.3E-03
3.5E-05
2.6E-04
7.1E-07
9.6E-03
2.0E-03
2.4E-02
6.1E-03
9.1E-04
1.8E-02
1.4E-04
4.7E-05
4.6E-05
5.4E-05
4.4E-05
4.9E-05
9.1E-03
8.5E-05
8.5E-05
8.0E-05
6.1E-03
5.9E-03
6.1E-03
4.3E-03
3.4E-03
6.3E-03
6.3E-03
8-64
-------
Table 8-11. Summary of Mean TEDE Results From CNWRA
Unit Concentration Evaluations for Water
-
^:; ;*%?
\ • \ -
Radionucilde
C-14
Tc-99
1-129
Cs-137
Ra-226
Np-237
U-238
Pu-239
Am-241
Annual TEDE, rcm/yr pcrpCi/1
. Arithmetic I
• \Mean
CNWRA
1.9E-05
8.4E-06
3.1E-03
7.6E-04
2.8E-03
1.3E-02
7.2E-05
1.1E-04
7.9E-03
<3«oj««i|k
^ MMffii
1 cmv&A
1.8E-05
7.9E-06
2.7E-03
6.6E-04
2.6E-03
1.2E-02
6.8E-05
l.OE-04
7.4E-03
The results shown in Table 8-11 do not imply that the isotopes will actually contribute to dose at
these relative levels. Actual levels of concentration in ground water, and the actual relative
contributions of various nuclides to dose, will depend on many factors, such as mobility of the
nuclide in the environment, solubility in water, time of release of radionuclides from the
repository, distance of the exposed individual from the repository, flow rate of ground water in
the environment, and the lifestyle of the exposed individual, represented in these results by the
subsistence farmer. Performance assessments to date (Chapter 7 of the BID) indicate that Tc-99
and 1-129 will be the principal contributors to predicted doses.
Potential variability of input parameter values is illustrated by Table 8-12, which shows values of
water and food consumption rates used in a variety of dose studies. Included in Table 8-12 are
the parameter values used in the CNWRA analyses, values from NRC's Regulatory Guide 1.109,
and values used by DOE based on surveys conducted in the Yucca Mountain area.
The DOE values were developed based on a 1997 survey of over 1,000 households in the Yucca
Mountain region. Note that most of the DOE values are substantially lower than the values from
the other sources. This is at least partly due to the fact that the DOE values are for "Locally
Produced Food," whereas the other sources are reporting total consumption values. EPA 540/1-
89-002 reports that, on average, the fraction of food that is "home grown" is 0.20 for fruits, 0.25
8-65
-------
for vegetables. 0.44 for beef, and 0.40 for milk; worst case fractions are 0.30 for fruits, 0.40 for
vegetables, and 0.75 for beef and milk. Using these "home grown" fractions, DOE's leafy
vegetable "Total Population" consumption rate is consistent with the average leafy vegetable
values from the other sources. However, for the other pathways, the DOE values remain lower
than the rates from the other sources. This is consistent with the arid nature of the Yucca
Mountain region, which makes farming difficult, and the fact that what agriculture there is is
mostly commercial (e.g., alfalfa, milk) with products shipped out of the region.
Table 8-12. Comparison of Inhalation, Drinking Water and Food Consumption Rate
Parameter Values From Various Sources
, Jtatitttft.?
Inhalation
Drinking Water
Leafy
vegetables
Root vegetables
Fruit
Grain
Beef
Milk
Units
m3/yr
1/yr
kg/yr
kg/yr
kg/yr
kg/yr
kg/yr
1/yr
CmVRA*
NR
511/730
4.27/1 1/28.3
11.3/51/231
10.2/46/208
15.3/69/312
22.1/59/157
20.8/100/482
BIOMOV5S*
8400
730
62.2
235
NR
148
94.9
330
0OEVM1
NR
646.16/769.70/769.70
4.39/9.70/63.55
2.13/6.37/28.86
4.47/10.54/59.32
0.40/11.01/60.64
0.92/8.66/8.97
4.84/60.50/119.39
MRC*
8000/8000
370/730
23/64
80/217
42/114
46/125
95/110
110/310
1 Except for drinking water, values were taken from CNWRA 97-009, Table 2-4. Middle value is the
consumption rate from NUREG/CR-5512, side values are the low and high range values based on log-
normal distribution from Hoffman et al. (1982). Drinking water values are the average and 90th percentile
values from EPA 540/1-89-002, Exhibit 6-11.
2 Taken from BIOMOVS II, Technical Report No. 12, Appendix A, Table A-7. These are the values used in
the deterministic analysis. Appendix B presents values used in the stochastic analysis; however, stochastic
values are not provided for food consumption, drinking water or inhalation rates.
'Taken from October 12, 1997 presentation to the Nuclear Waste Technical Review Board. First value is
for "Total Population," middle value is for "Partial Subsistence" population, and last value is for
"Subsistence" population. All values are for "Locally Produced Food."
4 Taken from NRC's Regulatory Guide 1.109. First value is for an average adult from Table E-4, while the
second value is for die maximum exposed adult from Table E-5.
NR = not reported
8-66
-------
Of special interest in Table 8-12 is the column of data from Biosphere Model Validation Study
(BIOMOVS), which represents parameter values derived from data sources in nations other than
the United States. The BIOMOVS is an international cooperative study to test models designed
to quantify the transfer and bioaccumulation of radionuciides and other trace substances in the
environment. Participating nations include Sweden, the Netherlands, France, Belgium,
Switzerland, Canada, Spain, and the United Kingdom. Technical Report No. 12 (BI096)
specifically addresses biosphere modeling for dose assessment for radioactive waste repositories.
A detailed review of the BIOMOVS II Technical Report No. 12 was conducted as a means of
comparing BIOMOVS parameter values with those from other sources. Findings can be
summarized as follows:
The BIOMOVS data are "broadly representative of a valley in central
Switzerland," not Nevada's Amargosa Valley. Therefore, for parameters such as
the irrigation rate, for which CNWRA developed data specific to Amargosa
Valley, the CNWRA values would be more appropriate than the BIOMOVS data.
The only stochastic food transfer factor given in BIOMOVS are for 1-129 and the
Np-237 chain. For all other radionuciides, only deterministic values are given.
Both BIOMOVS and CNWRA used IAEA's Handbook of Parameter Values for
the Prediction of Radionuclide Transfer in Temperate Environments (Technical
Report Series No. 364,1994) as their primary source of food transfer factors.
• In BIOMOVS, the annual drinking water consumption is not a stochastic
parameter - its deterministic value is given as 0.73 m3/y. This deterministic
value is identical with the average water consumption rate used by CNWRA.
The BIOMOVS residential dust loading central value is given as 5 x 10~5 g/m3.
The BIOMOVS suggested using a log-triangular distribution with the
maximum/minimum values plus/minus one order of magnitude from the central
value. The lower/upper limit suggested by BIOMOVS is only a factor of two
lower/higher than the value used in the CNWRA analysis.
In addition to residential dust loading, BIOMOVS utilizes an occupational
(farming) dust loading central value of 10~2 g/m3, a central value exposure time of
300 hr/y, and the same distributions as the residential dust loading. No reference
is provided for these values, although the dust loading appears to be at the OSHA
Permissible Exposure Limit (PEL) for nuisance dust (note the upper limit of the
stochastic range is an order of magnitude above the OSHA PEL). The BIOMOVS
average (residential and occupational) dust loading central value is 3.88 x 10"4
g/m3. The CNWRA analysis did not include occupational dust loading.
8-67
-------
• The BIOMOVS docs not provide a range of values for consumption rates, only a
fixed deterministic value. Most of the BIOMOVS deterministic values lend to be
towards the high end of the stochastic values provided by other sources.
These findings indicate that BIOMOVS parameter values could be used in the Yucca Mountain
dose evaluations where appropriate. They also tend to confirm the validity of the parameter
values used in analyses to date. Values may be revised and refined (e.g., better characterization
of values and uncertainty distributions), as characterization of the Yucca Mountain region
continues in the future.
8.3.5 Alternative Exposure Scenarios For Consideration at Yucca Mountain
In Section 8.3.4, the subsistence farmer was characterized and modeled as the individual most
likely to receive the highest dose among the exposed population. It was noted that the results of
the CNWRA evaluations of DCFs might correspond to those for EPA's criteria for the RMEI. At
Yucca Mountain, however, the qualification of the subsistence farmer as the RMEI is
conditional.
One factor important to characterizing the Yucca Mountain RMEI is his/her location relative to
the repository. For example, a subsistence farmer who derives all drinking water and home-
grown food from contaminated ground water at a location 10 miles from the repository may be
exposed to lower doses than persons whose exposure pathways are limited to drinking water, or
fractional quantities of contaminated home-grown food products, residing close to the repository
boundary. This is due to the fact that radioactivity concentrations in ground water are expected
to decrease with increasing distance from the repository boundary as a result of dilution and
dispersion.
The RMEI could also be represented by a commercial farmer, a rural resident, or someone simply
using contaminated ground water for domestic uses. For example, slope of terrain and poor soil
quality are factors that could potentially preclude farming, but not rural residency, near the
repository boundary. Alternatively, commercial farming might economically justify the one-time
high cost of drilling a deep well near the repository boundary.
The availability and cost of extracting ground water may also impact the type of farming. While
some farming in a desert environment would require extensive use of ground water, certain
highly specialized farming may require only modest amounts of water.
8-68
-------
A discussion of the alternative RMEI scenarios identified in Section 8.3.3 is presented below.
8.3.5.1 Commercial Farming Scenario
To maximize efficiency and monetary profit, farmers tend to specialize within a particular sector
of agriculture. As a result of local conditions, commercial fanning in Nevada tends to be focused
in the following sectors:
• Dairy Products - Principal activities include milk production and/or breeding of
dairy cows. Feed for dairy cows may involve some grazing, but most is either
purchased or produced by the dairy farmer.
• Livestock - The largest percentage of livestock is represented by cattle ranchers.
Based on range and pasture conditions, grazing may be supplemented to varying
degrees by cattle feed that is either purchased or produced by the rancher. (A
smaller percentage of livestock farmers specialize in poultry, hogs, etc. In general,
feed for these animals is either obtained commercially or from source(s) not likely
to be affected by contaminated ground water.)
Field Crops - Primary field crops in Nevada include wheat, barley, alfalfa, and
hay.
Produce and Specialty Crops - Included in this category are leafy vegetables, root
crops, and a limited variety of fruits/nuts.
For the purpose of modeling potential exposure pathways, the commercial fanner is assumed to
utilize contaminated ground water for personal use and all activities associated with farming (i.e.,
irrigation of crop, animal feed, produce, etc. and watering of livestock). Accordingly, the basic
model parameters previously identified in the subsistence farm model also apply to the
commercial farmer.
There are significant differences in potential radiation dose estimates between the subsistence
fanner and the commercial farmer because the commercial fanner is expected to derive only a
fraction of consumed food from home-grown/contaminated food products; the specialized farm
activities of a commercial farmer may in some instances exclude pathways generically assumed
for the subsistence farmer. The variability and uncertainty for modeling a commercial farmer is
due to the unlimited number of combinations of contaminated food groupings and the variable
ratio or fractions of contaminated food products to the total quantity consumed. For most food
8-69
-------
categories, fractional quantities have the potential to range from zero to one. For illustration, the
following sample scenarios are cited:
Scenario #1: Contaminated Food Intake Approaches Zero. In this sample scenario, the
commercial farmer's activity is limited to growing alfalfa for animal feed that is shipped,
pelletized, and sold on die open market. Food ingested by the farmer is assumed not to be
contaminated and exposure pathways are limited to drinking water, soil/dust inhalation,
and external exposure. The latter two pathways are the result of field irrigation and
ground contamination.
Scenario #2: Low to Moderate Levels of Contaminated Food Intake. This scenario would
represent a commercial farmer who engages in a single activity that may involve
livestock, milk, field crops, produce or fruit production. Based on the farming activity
selected, the farmer may be reasonably assumed to consume his/her home-grown food
product. However, the home-grown/contaminated food category may represent a highly
variable fraction of the total quantity consumed. (While it is reasonable to assume that a
dairy farmer consumes milk that is 100% derived from personal milk production, it is not
reasonable to assume that a potato farmer's diet of root vegetables is limited to home-
grown potatoes.)
Scenario #3: Medium to High Levels of Contaminated Food Intake. This scenario would
represent a commercial farmer engaged in multiple activities. For example, a large dairy
farm, in addition to milk production, may also raise alfalfa/hay for its own animal feed.
To maintain or increase milk production, dairy farms commonly breed their own stock,
leaving aged-/poor-milk producers and male calves available for slaughter. If, in addition
to these activities, a sizeable home-garden is added for personal food production,
contaminated food consumption by this commercial farmer approaches that of the
subsistence farmer.
The above-cited examples illustrate the difficulty in characterizing a "typical commercial
farmer." No single set of model parameters that specify individual food categories and quantify
the fractions of contaminated food products can represent the range of conditions.
8-70
-------
8.3.5.2 The Rural Residential Scenario
Consideration must be given to the possibility that a rural resident may be the RMEI. The
possibility may occur if subsistence and commercial farming close to the repository side are
excluded by geophysical limitations (e.g., soil quality, slope of terrain, etc.), but rural residency is
not. As was previously noted, using a lower amount of contaminated water drawn from a well
close to the repository can lead to greater exposure than using a greater amount of contaminated
water drawn from a well farther away from the repository.
A rural residential scenario may reasonably include the drinking water pathway and fractional
intakes from ingestion of contaminated home-grown vegetables, produce, and other food
products. Consequently, at or near the footprint of the repository, where contamination levels are
likely to be greatest, the restricted or unrestricted use of well water for drinking and limited food
production may result in exposures greater than those to commercial, and even subsistence
fanners, residing at more distant locations. The scope of home-grown food production for rural
residents is highly variable and may range from a few kilograms for a single food category to
substantial quantities that represent nearly all major food categories.
8.3.5.3 Domestic Use of Contaminated Water Scenario
A community well is a common way to provide domestic water in instances where conventional
municipal water supplies are not available. Domestic use implies all normal household uses of
water such as bathing, washing, sewage, and human consumption.
Exposure for this scenario is, therefore, limited to the drinking water pathway that has been
previously defined. On average, for the 20 radionuclides considered, drinking water contributes
about one-fifth of the maximum total dose from all pathways.
8.4 THE REPOSITORY INTRUSION SCENARIO: A SPECIAL CASE
Inadvertent human intrusion into the repository could occur if future generations were to attempt
to extract minerals, oil and gas, water or other resources from the site. Such intrusion could
result in a breach of the repository's geologic and engineered barriers, thus releasing
radionuclides into the atmosphere or ground water. Such a release could pose a risk for future
residents of the area. This section examines the possibility of finding resources at the site and the
incentive for future exploration.
8-71
-------
With regard to the human intrusion issue, the National Academy of Sciences (N AS95) in its
report, "Technical Basis for Yucca Mountain Standards," reached the following conclusions:
• There is no technical basis for predicting either the nature or the frequency of
intrusions.
• There is no scientific basis for making projections over the long term of the social,
institutional, or technological status of future societies.
There is no scientific basis from which to project the durability of government
institutions over the period of interest, which exceeds that of all recorded human
history.
Some degree of continuity of institutions, and hence of the potential for effective
active institutional controls, into the future might be expected; but there is no
basis in experience for such an assumption beyond a time scale of centuries.
There is no scientific basis for assuming the long-term effectiveness of active
institutional controls to protect against human intrusion.
There is no technical basis for making forecasts about the reliability of passive
institutional controls.
There is no scientific basis for estimating the probability of inadvertent, willful, or
malicious human action.
Based on these findings, the NAS made the following observations:
Although it can not be proven, it is believed that a collection of prescriptive
requirements, including active institutional controls, record keeping, and passive
barriers and markers, will help to reduce the risk of human intrusion, at least in the
near term. The degree of benefit is likely to decrease over time.
Because it is not technically feasible to assess the probability of human intrusion
into a repository over the long term, it is not scientifically justified to incorporate
alternative scenarios of human intrusion into a fully risk-based compliance
assessment that requires knowledge of the character and frequency of various
intrusion scenarios. However, it is possible to carry out calculations of the
consequences for particular types of intrusion events. Such calculations might be
informative in the sense that they can provide useful insight into the degree to
which the ability of a repository to protect public health would, be degraded by
intrusion.
The NAS made the following recommendations for the approach to addressing the human
intrusion issue:
8-72
-------
• ... the repository developer should he required to provide a reasonable system of
active and passive controls to reduce the risk of intrusion in the near term.
• EPA should specify' in its standard a typical intrusion scenario to be analyzed for
its impact on the performance of the repository.
This section of the BID presents background information relevant to human intrusion scenarios
that can be developed for the Yucca Mountain Repository site. The assumptions made about the
intrusion scenarios follow the guidelines provided by the NAS. This discussion is organized as
follows:
• The potential causes of intrusion are first discussed by summarizing the current
resource potential in the vicinity of Yucca Mountain and how that may influence
future intrusion.
Possible intrusion scenarios are presented for each of the resources that are likely
to occur in the vicinity of Yucca Mountain.
The assumptions that apply to each scenario and the parameters that would be
used to calculate the consequences of each intrusion scenario are provided.
8.4.1 Site Resources as Potential Cause for Intrusion
Extensive exploration, development, and mining have occurred in the Great Basin of Nevada and
extensive histories and lists of mineral deposits are available. Predicting future economic
conditions and what materials in the vicinity of the site may be considered resources, or how they
may be explored or produced, is not feasible. However, the consequences of an intrusion
scenario based on exploration and/or production of a present-day resource can be evaluated using
current methods and technologies and by assuming similar types of intrusion in the future. The
information on current and historic natural resources in the vicinity of Yucca Mountain is
presented to establish plausible background data for scenarios based on current resource
exploration and/or production.
The discussion of mineral resources, based on information from Miklas and Fiero (MIK92;
FIE86), is focused on oil, natural gas, geothermal resources, and metallic ores. Other mineral
resources that are or may be present at or in the vicinity of the Yucca Mountain site, such as
gravel, building stone, and pumice, are excluded from this discussion. Such minerals are
abundantly present in other parts of the region. Moreover, since they have a low bulk value, they
can only be profitably extracted from large-scale, shallow surface workings.
8-73
-------
8.4.1.1 Petroleum and Natural Gas Resources
The Great Basin of Nevada, in which Yucca Mountain is located, has the potential for petroleum
deposits. Excellent reservoir rocks and structures (faults and folds) exist and source beds (rocks
in which petroleum might have formed) are also present. However, the complexity of the
geology, due to deformation resulting from tectonic forces, makes exploration difficult and
costly. Also, the potential size of such reservoirs is limited due to the high degree of faulting in
the region.
Oil and natural gas have been produced in Nye County (Railroad Valley) and Eureka County
(Pine Valley). Both sites are about 100 to 300 km northeast of the Yucca Mountain site. The
fields are relatively small and production is on the order of several hundred to a few thousand
barrels per day. Given that all production to date has come from Tertiary basins in the Sevier
mountainous belt between the Devonian/Mississippian Antler highland and the Paleozoic
continental shelf, the potential for petroleum resources in the vicinity of Yucca Mountain is rated
as low.
8.4.1.2 Geothermal Resources
In general, the Basin and Range Province, which contains the Great Basin subprovince, is an area
of elevated heat flow (about two heat flow units [HFU]) relative to other continental settings
(about one HFU). This is believed to be due to the thin crust and near melting conditions at the
crust/mantle boundary. In Nevada alone, there are nearly 300 thermal springs and warm water
wells. However, the hot spring activity is concentrated in the west-central and north-central parts
of the state. Yucca Mountain is located in an area of moderately elevated heat flow (1.5 to 2.5
HFU). The Eureka heat flow, on the order of 0.75 to 1.5 HFU, is immediately to the north of
Yucca Mountain and is thought to be below the average heat-flow values for the region due to
underflow of ihtrabasinal ground water. A geothermal test well drilled on Pahute Mesa,
approximately 40 km north of Yucca Mountain, found maximum water temperatures on the order
of 60 to 90°C, well below current geothermal resource values. Warmer temperatures (125°C)
were found at a depth of 3,700 m; however, this is below the 1,000 meter depth considered to be
economical for low temperature geothermal production (MIK92).
8-74
-------
8.4.1.3 Mineral Resources
Disseminated Gold/Silver and Uranium Deposits
Disseminated gold/silver deposits have fueled the Nevada precious metals boom over the past 15
years. These are low-grade deposits (0.01 to 0.1 ounce per ton (oz/ton) cutoff grade) worked by
open pit operations involving minimal milling and cyanide leaching technologies. Base metal
concentrations are generally low in these deposits, while mercury, arsenic, thallium, and
antimony concentrations are elevated. In the Great Basin subprovince, these deposits occur in
both sedimentary and volcanic host rocks.
Sedimentary rock that hosts disseminated gold/silver deposits is located predominately in the
northern and western portions of the Great Basin, primarily between the Sierra Nevada
mountains to the west and the Paleozoic eastern assemblage of the continental shelf. In addition
to clustered deposits in sedimentary rock, the Carlin, Getchall, and Cortez metallogenic trends
are recognized. Host rock is variable, ranging from calcareous through clastic sedimentary rocks,
with some preference for argillaceous or carbonaceous carbonates. The Roberts Mountain thrust
of the Antler orogeny, north of the Yucca Mountain site, marks a fairly sharp boundary between
gold-bearing deposits northwest of the thrust and barren mineral deposits southeast of the thrust.
While relatively few of the disseminated gold/silver deposits in Nevada are hosted in volcanic
rock, those that are have been associated with a magnetic anomaly along the Walker Lane Belt
that includes Yucca Mountain. However, the host rock is generally Tertiary andesites, silicic
tuffs, and volcaniclastic sedimentary rocks, none of which have been found in the vicinity of
Yucca Mountain.
Most of the uranium production in Nevada has been from disseminated tertiary deposits and
associated veins at the Apex mine in Lander county north of Yucca Mountain. Volcanic-hosted
uranium in sub-economic concentrations is found in silicic volcanics at the McDermitt Caldera in
the northwest corner of Nevada.
Porphyry Deposits of Copper, Molybdenum, and Gold/Silver
!
Porphyry deposits are igneous rocks in which large crystals are enclosed in a very fine-grained
matrix. Calc-alkaline porphyry deposits associated with fossil hydrothermal systems contain
many of the richest copper and molybdenum deposits in the Great Basin. While ore grades are
8-75
-------
generally low (0.5 to 1.0 percent copper, 0.01 to 0.1 percent molybdenum), there is typically a
high-grade enriched cap. Although concentrations of gold and silver are very low. byproduct
recovery from the large volumes of ore processed for copper and/or molybdenum is economic.
Porphyry intrusions are typically 0.5 to 3 km in diameter and lie at depths of one km. The
deposits in the province generally date from 50 to 70 million years, although the deposits at
Battle Mountain, NV and Btngham, UT are dated from 35 to 40 million years. The relatively few
copper and molybdenum deposits in the Great Basin lie far north of Yucca Mountain. The
richest porphyry deposits in the Basin and Range Province are south of the Great Basin in
southern Arizona and New Mexico. Given the relatively young age of Yucca Mountain, less than
17 million years, it is unlikely that porphyry deposits are to be found in the area.
Skarn and Carbonate-Hosted Deposits
Skam and carbonate-hosted deposits in the Great Basin have been exploited for a variety of base
and precious metals, including iron, tin, tungsten, copper, zinc, lead, molybdenum,
gold, and silver. However, these deposits are largely limited to the northern Great Basin and the
Porphyry Copper Block of Arizona/New Mexico.
Epithermal Vein Deposits
In the Great Basin, through-going normal faults and associated fracture sets are clearly correlated
with mineralization trends for a variety of metals. Veining is largely controlled by normal and
slip-strike faulting resulting from caldera formation, although thrust faulting has also played a
role.
Many of the historic mining districts in the Great Basin, including Comstock, Bodie, and
Tonopah, exploit polymetallic vein deposits. These deposits are usually mined for high-grade
gold and silver, but economic concentrations of antimony, lead, zinc, copper, manganese, and
uranium have also been developed. Near Yucca Mountain, gold and silver have been produced
from vein deposits in the Wahmonie District (25 km east), Bare Mountain (15 km west), and in
the Bullfrog Hills (30 km west).
8-76
-------
Breccias (Gold/Silver)
Breccia deposits (in pipes, stockwork fractures, and brecciated fault zones) of gold, silver, and
base metals are widely dispersed in the Great Basin. Such deposits have been identified at
Paradise Peak, Borealis, Victoria, and Ortiz in Nevada, northwest of the Yucca Mountain site.
While such deposits are widespread, it should be noted that they contain only a small fraction of
the total reserves of the region.
Massive Sulfide (Copper, Lead, Zinc)
Small deposits are found throughout the southern Basin and Range in Arizona and in the north-
central Great Basin at Big Mike and Mountain City in Elko County, NV. Of volcanic origin,
such deposits are believed to form at tectonic plate margins where seawater circulates near the
vents of submarine hydrothermal systems.
Roll-Front (Silver, Uranium)
While roll-front uranium deposits are associated with much of the uranium that has been
discovered and mined on the Colorado Plateau, no such deposits have been discovered in the
Great Basin subprovince. Indeed, only a single sandstone roll-front deposit of silver (at Silver
Reef, UT) has been discovered in the subprovince.
Placer (Gold, Platinum)
Placer deposits in the Great Basin are fairly common, and, due to limited lateral transport, are
almost always found in close proximity to the parent deposit. Exceptions include the Snake
River, ID and Spring Valley, NV placers, which are not associated with a lode deposit Gold and
platinum placers in Nevada are found north of the Yucca Mountain site at Round Mountain,
Battle Mountain, and Manhattan.
8.4.1.4 Other Materials
Other materials found in the Great Basin include barite, manganese, borax, mercury, beryllium,
gallium/germanium, zeolites, and fluorspar. Of these, only zeolites and fluorspar are believed to
occur in significant deposits near the Yucca Mountain site.
8-77
-------
The unique ion exchange and sorptive properties of zeolite minerals find numerous practical
applications, including molecular sieves and water softeners. While thick zeolite beds are
present in the vicinity of Yucca Mountain, they are found only at great depth. Because of the low
value of the resource, economic recovery currently relies on surface-mining techniques.
Fluorspar also has a number of industrial applications, primarily in the chemical, ceramic, and
metallurgical industries. The largest fluorspar-producing region of Nevada is the Bare
Mountains, about 15 km west of Yucca Mountain.
8.4.1.5 Ground Water
Ground water is currently the only source of water in the area and is used for domestic,
agricultural, and industrial purposes. Water for site investigation requirements is obtained from
two wells (J-12 and J-13) located approximately five km from the proposed repository footprint.
Currently, the J-12 and J-13 wells are the closest production wells to Yucca Mountain; there are
no production wells situated on Yucca Mountain itself. These wells are completed in the welded
Tertiary volcanic rocks (Topopah Spring Member of the Paintbrush Tuff). Wells in the tuff
aquifer range in depth from 850 to 3,500 feet and are capable of producing from 370 to 770
gallons per minute (gpm) based on testing performed in 1967 and reported in U.S. Geological
Survey Water Supply Paper No. 1938. The water table beneath the proposed repository site is
located within the Calico Hills and Crater Flat formations. The Crater Flat hosts the lower
volcanic flow system, with the Calico Hills acting as an aquitard between the Crater Flat
formation and the upper volcanic flow system within the Topopah Spring. Ground water quality
in the volcanic aquifers is variable, being a complex function of many factors. The primary
factor governing water quality in the volcanic rocks is the residence time of the water within the
aquifer. Wells completed near recharge areas are likely to produce better quality water than those
completed in areas with a long residence time. The two existing water supply wells, J-12 and J-
13, are completed within the recharge area of Forty Mile Wash and thus produce water of good
quality. Water beneath Yucca Mountain is generally found to be older and of poorer quality.
(See Section 8.2.3.1.) The latest available data compiled by Lyles and Mihevc in 1994 (DRI94)
identify over 500 domestic, agricultural, and monitoring wells in the Amargosa Valley. Past
hydrogeologic studies were conducted to: 1) evaluate the water resources potential of the area; 2)
evaluate the impact of ground water pumping; 3) estimate the ground water recharge; and 4)
evaluate regional ground water flow. These studies are referenced in the Lyles report.
8-78
-------
8.4.1.6 Resource Summary
Within a 30-km radius of the Yucca Mountain site, there are six active gold- and silver-
producing properties in the Bullfrog and Bare Mountain mining districts to the west of Yucca
Mountain. Fluorspar is also produced from the Daisy Mine in the Bare Mountain district. A
small amount of mercury has been produced, at both the Thompson Mine on the north end of
Yucca Mountain and at Bare Mountain. Borax is produced in the Amargosa Valley due south of
die site near the California border. Uranium, geothermal, and hydrocarbon resources have not
been exploited in the vicinity of Yucca Mountain; however, hydrocarbon exploration (wildcat)
wells have been drilled within 20 km of the site.
Ground water in the vicinity of Yucca Mountain is currently produced from the Tertiary volcanic
welded tuff units. Yields from wells constructed in fractured volcanic rocks are generally high.
Ground water is also found within the underlying Paleozoic carbonate unit, but the relatively
great depths and poor quality of this water preclude it from being utilized as a resource at present.
The resource value of ground water in this area depends on the depth to the water (as reflected in
drilling costs) and water quality. Beneath Yucca Mountain the resource value of ground water is
considerably lower than in the adjacent valleys, due to the increased depth to water and
somewhat poorer ground water quality.
8.4.2 Types of Human Intrusion
The NAS recommended that EPA require that the consequences of human intrusion on repository
performance be analyzed and that the Agency's standards specify a typical intrusion scenario to
be used for this purpose. Selecting a scenario entails judgment. To provide for the broadest
consideration of what scenario or scenarios might be most appropriate, the NAS recommended
that EPA make this determination in its rulemaking. As a starting point, the NAS suggested a
stylized intrusion scenario consisting of one borehole of a specified diameter drilled from the
surface through a canister of waste into the underlying aquifer.
As background for selecting a scenario, the types of human intrusion scenarios that may be
considered, based on current knowledge of the area's resource potential and exploration
technologies in use today, are discussed below. Table 8-13 lists likely scenarios for each type of
resource currently found in the vicinity of Yucca Mountain. Further discussions of these likely
scenarios follow the table.
8-79
-------
Table 8-13. Likely Human Intrusion Scenarios for DiiTerenl Types of Resources
Nature of Busman Intrusion
(1) Borehole completed into repository
(2) Borehole completed into tuff aquifer beneath
repository
(3) Borehole completed into carbonate aquifer
and tuff aquifer beneath repository
(4) Aquifer testing with uncased borehole or test
well
(5) Production well completed and placed into
service
Petroleum or
GeotheoruU
X
•; '^s
^ V
Minerals
X
X
TulT
Aquifer
X
X
X
Carbonate'^
Aquifer \
X
X
8.4.2.1 Petroleum/Geothermal-Related Intrusion
Human intrusion resulting from the exploration for petroleum or geothemial resources would be
comparable due to the depth at which the resources are expected to be found in the area around
Yucca Mountain. Petroleum typically is found in the Paleozoic carbonates and the geothermal
temperatures that make recovery economic are in the Precambrian basement rocks. Both
resources are at depths that would require drilling through the repository horizon elevation and
the tuff aquifer, and into or through the Paleozoic carbonate aquifer (Scenario No. 3, Table 8-14).
Typically, petroleum and geothermal exploration holes are cased into competent rock
(unfractured and minimal porosity) to provide a seal at the surface in the event that high pressure
gases or liquids are encountered during drilling. The seal is expected to withstand the pressures
anticipated. It is usually a 14- to 30-inch diameter pipe, depending on the largest drill bit
expected to be used, which is set and cemented in the initial borehole advanced into unfractured
rock. Once the cement has set, a drill bit slightly smaller than the surface casing, typically 12 to
24 inches in diameter, is lowered to the bottom of the casing and the drill string advanced.
Drilling is usually continued in the open hole (no casing) with cuttings or core samples being
collected to identify the rock type being penetrated and to evaluate resource content or potential.
Excess cuttings are collected in a pit adjacent to the drill rig and the fluid recirculated to flush
more cuttings to the surface. In the case of air drilling methods, which are common in
geothermal exploration, the air from the cutting return stream is discharged directly to the
atmosphere above the cuttings pit.
8-80
-------
There are two ways in which release of radionuclides could result from petroleum and
geothermal exploration. The first involves potential releases resulting from the drill passing
through the repository and associated waste, entraining or dissolving radioactive waste products,
and carrying waste products to: 1) the surface in the cuttings return; 2) the tuff aquifer, once it is
reached, through the drilling fluid circulation path; and 3) to the deeper carbonate aquifer, once it
is reached, through the drilling-fluid circulation path. The primary mechanism for contaminating
the aquifer would be the circulation of contaminated drilling fluid. Petroleum exploration
drilling commonly uses the direct rotary drilling method, which pumps the drilling fluid down
the center of the drill pipe, exits the bit, and flushes the cuttings up the annular space between the
hole and the drill pipe to the surface. The cuttings are collected on the surface in pits.
Direct rotary air drilling methods, used to drill geothermal wells, would also discharge the
cuttings to a pit, but would not recirculate contamination as readily because they do not
recirculate the returning fluid. In both types of drilling, contamination can also be spread when
the drill string is removed from the hole to change bits, test the formation, or abandon the hole.
The second possibility of releasing radionuclides occurs when the borehole is being abandoned
and is not being sealed. In this instance, material from the breached waste area could fall through
the open borehole to the aquifer zones, where it can be dissolved and transported to the
environment. If the borehole fills with water, material from the repository horizon could still
sink through the water column; contamination also could be circulated by density and/or thermal
effects.
8.4.2.2 Mineral Exploration-Related Intrusion
Mineral exploration drilling has the potential to result in more than one intrusion scenario, as
illustrated in Table 8-14. The difference among the scenarios is the depth drilled, due to the high
degree of uncertainty with respect to what might be considered a mineral resource in the future.
Completion of an exploratory borehole into the repository horizon (Scenario No. 1, Table 8-14)
is conceivable because the radioactivity released from the repository might be detected using
remote sensing instruments and be mistaken as an indication of mineral deposits. Further
exploration that would require drilling might thus be undertaken.
Typically, mineral exploration drilling is performed using relatively small diameter (nominally
three- to seven-inch) drills or coring bits with air or rotary wash. Mineral exploration holes are
not cased, except when the near-surface materials are very unstable which could preclude
8-81
-------
keeping the top of the hole open. The potential pathways to the environment are very similar to
those discussed for petroleum or geothermal exploration, with the primary difference being the
size of the borehole and the associated quantity of material potentially removed and circulated.
Coring is a frequently used method which provides direct visual identification of the material
being penetrated and permits the evaluation of ore grade. If coring were done when the drill is
penetrating the repository horizon, it would be possible for a sample of the waste material or
contaminated materials from the repository to be brought to the surface.
Considering the known occurrences of mineral resources in the vicinity of Yucca Mountain, It is
likely that an exploration borehole would be completed in the Paleozoic and older rocks that are
beneath the volcanics that contain the repository horizon and the tuff aquifer zone (Scenario No.
3, Table 8-14). Improper abandonment (e.g., a borehole left open with no backfilling) of a
borehole could create a contamination circulation route similar to that described for an
abandoned petroleum or geothermal exploration hole.
8.4.2.3 Ground-Water Resource-Related Intrusion
The intrusion scenarios developed for ground water resources, shown in Table 8-14, relate to
exploration, aquifer testing, and well development and production.
Ground Water Exploration Drilling
Table 8-14 shows the possible borehole scenarios for the tuff aquifer (Scenario No. 2) and
carbonate aquifer (Scenario No. 3). The exploration for ground water resources would probably
involve a direct rotary-air or water-configured rig using a drill bit or pneumatic hammer on the
order of six to eight inches in diameter. In arid regions, like the Yucca Mountain area, air
drilling is commonly used to minimize the amount of water required. The exploration borehole
would be advanced as rapidly as possible until the first water is noted in the return air, or an
increased flow rate is identified from the annular space if water or drilling mud were used. In
this case, where the repository horizon is above the aquifer, the repository materials could be
penetrated and circulated to the surface for several minutes and would probably not be noticed
until ground water began to be emitted from the return line or annulus. Even at this time, it is not
likely that the waste material would be recognized unless some type of radiation detector were
being used at the drilling site. As described in Section 8.5.2.1, the contamination would be
circulated to the surface with the flow up the annular space and, if water were used, returned to
the aquifer by the mud pump taking water from the settling pit. If air were used, the recirculated
8-82
-------
air would not be as contaminated, unless the compressor intake were in close proximity to the
discharge line (bloey line), which would be discharging the contaminated return air and cuttings.
Additional releases of the repository waste material and mixing with the aquifer fluids could
result from removing the drill string from the hole and reinserting it (tripping). This is done
when drill collars need to be added or the bit must be changed. This random action depends on
the depth of drilling, bit wear, and rock type and could be exacerbated by drilling through
repository waste containers (creating the need for a bit change). A more common tripping of the
drill string is done to recover core when a fixed core barrel is used. In this case, the drill string is
removed every 30 to 48 inches of drilling, depending on the length of the core barrel, to recover
the cored rock. In some instances, a wire-line coring device is used to preclude the necessity of
removing the entire drill string from the hole. The core barrel is lowered inside the drill rod,
attached to the bit and, once the core barrel is full, pulled to the surface using a wire line on a
hoist. Wire-line coring is most frequently used in mineral exploration drilling due to the smaller
core diameters (typically less than 2.5 inches) needed for mineral identification.
An improperly abandoned borehole would have consequences similar to those described in
previous sections.
Aquifer Testing
Scenarios which entail aquifer testing are shown in Table 8-14 (Scenario No. 4) for the tuff and
carbonate aquifers. Aquifer testing could be performed during drilling in the uncased borehole,
typically done when drilling is performed using an air rotary rig. More extensive aquifer testing
would be performed in a well constructed in the exploratory borehole.
Testing in the open (uncased) borehole is referred to as drill-stem testing. It is performed using
the air flow from the compressors) to lift the water from the aquifer zone, by inserting the drill
pipe near the bottom of the hole and injecting air, causing the fluid column to rise to the surface
and/or entraining the water in the air stream to remove it from the borehole. This method creates
a scouring action in the open borehole due to up-hole air/water mixtures reaching velocities on
the order of 3,000 feet per minute (ft/min). A fluid stream moving this fast would produce
erosion in the repository zone penetrated, increasing the material carried to the surface or falling
into the borehole. Testing in this manner is usually of shorter duration (several minutes to a few
hours) than aquifer tests performed in cased holes. In some cases, an air-lift pumping system (a
8-83
-------
pipe or drill rod with an internal air line suspended to beneath the water table but not to the end
of the pipe) can be lowered into the open hole and used to test the flow.
The second testing-related scenario consists of a well constructed in the exploratory borehole for
testing the potential yield and evaluating the storage capacity of an aquifer. Aquifer depths in the
vicinity of Yucca Mountain are currently in excess of 244 m and would require a well of at least
12 inches in diameter to set a pump that would be capable of testing the aquifer adequately.
Exploratory boreholes are not typically drilled large enough to facilitate constructing a well of
this diameter, therefore, conducting a test would require increasing the size of the borehole. This
action would remove more material from the breached repository area and allow it to circulate,
dissolve, or slough into the borehole. Once the borehole is enlarged, a casing with well screen in
the aquifer zone would be set into the borehole, gravel packed in the aquifer zone, a cement plug
placed on top of the gravel pack, a bentonite slurry placed around the casing to the surface, and a
cement plug placed around the upper few feet of casing to form a surface seal. This well would
be constructed in the same manner as a production well, which is discussed in the next scenario.
Testing would be performed by placing a pump in the screened zone of the casing and varying
the pumping rate to evaluate the aquifer parameters and, after an optimum rate is selected,
pumping the aquifer at that rate for several hours or days. During testing, the only release of
radioactive contaminants from repository materials to the aquifer would be prior to or during
well construction. After the well is constructed, the breached repository horizon would be cased
with solid pipe and isolated from the fluid stream.
Ground Water Production
As shown in Table 8-14, the ground water production scenario (Scenario No. 5) is identified only
with the tuff aquifer because of the depth and reported poor water quality of the carbonate
aquifer. If the production of the carbonate aquifer were to be considered, the scenario would
differ only in terms of drilling depth.
The construction of a production well is similar to the process described above for test-well
construction, except that a production well is larger in diameter. A reverse rotary drill rig may be
used for drilling production wells. In reverse rotary drilling, the fluid flows into the annular
space at the surface and maintains a static head of water in the hole. The mud pump draws a
suction on the drill rod and the fluid is pumped from the drill rod to the mud pit, where the
cuttings settle out and the fluid flows back into the hole. In this scenario, a potential release of
radionuclides to the environment could occur during the drilling of the well as contaminated fluid
8-84
-------
is circulated from, and later past, the repository horizon, into the mud pit, and ultimately into the
aquifer zone. Once the well is constructed, the repository horizon is isolated by packings and the
primary source of contamination would be residual fluids in the well and aquifer.
After well construction is completed, the casing is pumped and surged to remove residual drilling
fluids from the aquifer and to develop a loose aggregate at the pumping level. This removes the
residual fluids from the casing. Well testing may be conducted to confirm a welFs performance
prior to placing it into production. This will also flush the aquifer zone, with the fluids from all
testing typically being discharged to a natural drainage feature, the mud pit, or the ground surface
in the vicinity of the well. After all testing is completed, the well is connected to a distribution
system and placed into service. Well water could be a sole source supply for a commercial
lication or combined with several other wells in a large facility or municipal supply system.
8.4.3 Parameters and Assumptions Associated with Ground Water Withdrawal
The potential exposure to radiation associated with ground water withdrawal results from the
contamination of the aquifer being pumped. The aquifer considered for ground water withdrawal
in the vicinity of Yucca Mountain is the tuff aquifer. The primary parameters necessary to assess
the consequences of intrusion are the aquifer pumping rate, the duration of pumping, aquifer
properties, the degree to which the aquifer has been contaminated, and the nature of the
contaminants.
Production wells are typically large in diameter (16 to 36 inches) to accommodate multistage
turbine pumps that can lift water from the aquifer zone at the optimum flow rate, which can
range from 500 to 1,500 gpm. For example, the intrusion scenario used by Sandia National
Laboratories in TSPA-93 (DOE94a) assumed that a production well drilled using a 24-inch bit
intersected the repository.
Pumping rates ranging from 300 to 700 gpm were used for USGS tests at the Nevada Test Site.
The tuff aquifer was pumped at 370 gpm for four days at one well location and 697 gpm for four
days at another well (USG72).
The assumed pumping duration for a production well would be based on how many gallons per
day would be necessary to supply the user. This value would determine the duty cycle of the
pump required. For example, pumping a well at 770 gpm continuously would produce one
million gallons per day. In a production or test well, the screened zone would be the only source
of contamination because the repository horizon above the aquifer would be cased with pipe to
8-85
-------
facilitate transporting water to the surface. Assumptions would be required for the well-drilling
scenarios (air or water) to assess the amount and nature of residual contamination in the aquifer
zone, if the well were eventually used as a supply. Significant factors would include the nature
of the permeability (i.e., primary porosity or fractures), physical and chemical properties of the
tuff aquifer (i.e., adsorption and redox potential), and secondary mineralization and its influence
on radionuclide transport. Assumptions regarding the contaminants that have been introduced
into the aquifer either as solids or in solution would be equally important.
8.4.4 Parameters and Assumptions Associated with Human Intrusion
There are three categories of future human intrusion events:
• Inadvertent intrusion in which the intruder does not recognize that a hazardous
situation has been created. This category has been the focus of discussion in the
context of standard-setting and licensing.
• Inadvertent intrusion in which the intruder recognizes that a radioactive waste
repository has been disrupted and takes corrective actions. On the assumption that
the corrective measures taken are effective and the repository is sealed, this
category is not of concern. If, however, corrective actions are not taken or are
ineffective, this type of intrusion is operationally the same as the above category.
• Intentional intrusion for either beneficial or malicious purposes. The NAS report
considers it presumptuous to try to protect against the risks arising from the
conscious activities of future human activities. However, given the potential
energy value of the wastes intended for Yucca Mountain, this category of intrusion
scenarios might be likely.
Two broad categories of risk could result from the release of radioactive material due to an
intrusion into the repository of the type characterized by borehole scenarios. These categories
are:
Risks from materials brought directly to the surface by the intrusive activity.
• Risks that arise from improper abandonment of an exploratory borehole that could
compromise the integrity of the repository's engineered or geologic barriers.
Radioactive materials brought directly to the surface by intrusive activity would likely pose
hazards to the intruders themselves and to the public. The NAS concluded that analyzing these
risks is unlikely to provide useful information about a specific repository site or design and,
8-86
-------
therefore, should not provide a basis for judging the resilience of the proposed repository to
intrusion (NAS95). Accordingly, the NAS recommended that these risks not be considered in
the compliance analysis. For these reasons, discussions of parameters and assumptions
associated with these types of scenarios are not presented in the BID.
Long-term consequences would result from the abandonment of a borehole that had intersected
repository waste without plugging it with impermeable material. The importance of this
scenario, as suggested by die NAS, is that it could create enhanced pathways to the environment
(both air and ground water). The fact that the scenario could also breach a waste canister is less
significant because this will happen eventually even without human intrusion (NAS95).
8.4.4.1 Factors of Consideration
To evaluate the human intrusion scenarios, the following factors or parameters must be evaluated
and the associated assumptions made.
Institutional Controls
According to the NAS report, there is no scientific basis for making projections over the long-
term of either the social, institutional, or technological status of future societies. There is no
scientific basis from which to project the durability of government institutions over the period of
interest, which exceeds that of all recorded human history. Some degree of continuity of
institutions, and hence of the potential for active institutional controls, into the future might be
expected, but there is no basis in experience for such an assumption beyond a time scale of
centuries. Similarly, there is no scientific basis for assuming the long-term effectiveness of
active institutional controls to protect against human intrusion.
Furthermore, according to the NAS report, there is no scientific basis for making forecasts about
the reliability of passive institutional controls. The likelihood that markers or barriers would
persist, be understood, and deter intrusion cannot be assessed from a technical basis (NAS95).
Drill Depth and Hole Size
As noted by the NAS, it is not feasible to predict which natural resources will be discovered or
will become valuable enough to be the object of an intruder's activity. The characteristics of
8-87
-------
future technologies for resource exploration and extraction also cannot be predicted (NAS95).
The availability of such information would affect the assumptions of drill depth and hole size.
Based on current practice, typical diameters of exploration boreholes and depths of penetration
are as shown in Table 8-14.
Table 8-14. Typical Borehole Characteristics (Source: CNW96)
.III v
fy$#kitf $x$l o*atw«i
Petroleum/Geothermal
Mineral
Ground Water
HoleSizeflndi)
12-24
3-7
6-8
* % Briauepifc \
carbonate aquifer
carbonate aquifer
tuff or carbonate aquifer
Number of Boreholes and Borehole Location
Generally, resource exploration utilizes remote sensing, topographic, and geologic information to
select drilling locations. However, when investigating a broad area like the Yucca Mountain
region, the spacing of exploration boreholes will vary for the various types of resources.
Petroleum and geothermal resource exploration is performed to detect regional or structural
trends that can extend for tens or hundreds of miles. Thus exploration drilling typically involves
a single hole in a region or within a geologic structural trend. Mineral exploration is carried out
in an orderly manner, usually employing a grid. The initial grid size, when regional resources are
being evaluated (instead of localized vein-type deposits), may be a mile or more on center for
boreholes. The grid spacing is decreased only if economic levels of target minerals are detected,
which is not expected to be the case in the immediate vicinity of Yucca Mountain. Borehole
locations could be on mountain tops or in the low areas.
Exploration for ground water resources can be focused based on surface features or convenience
to a user and, in such case, the exploration wells are typically clustered or linearly spaced a mile
or more apart. For regional investigations of ground water resource potential, randomly and
widely-spaced boreholes are commonly used. In such case, a density of one well per 2,000 km2
is reasonable; this provides adequate information on the nature and presence of a ground water
resource. In the Yucca Mountain area, the most likely locations for ground water exploration are
the drainage basins that surround Yucca Mountain.
-------
In terms of the number of boreholes to be assumed in the scenario, the NAS report suggests a
stylized intrusion scenario consisting of only one borehole. A single borehole scenario holds the
promise of providing considerable insight into repository performance. Under many conditions,
the effect of multiple boreholes presumably would be the sum of the effects of each taken
separately, but circumstances when this assumption is invalid can also be conceived. Because
construction of the scenario is arbitrary, the NAS report argues for the simplest case that
evaluates repository performance (NAS95).
In determining the location of the borehole, the stylized single borehole scenario suggested by the
NAS postulates drilling from the surface through a canister of waste to the underlying aquifer.
The emphasis would be on the creation of enhanced pathways to the environment as opposed to
breaching the canister, which will happen eventually even without human intrusion.
Borehole Sealing
According to the NAS report, the characteristics of future technologies for resource exploration
and extraction, and whether future practice will include sealing of physical intrusions such as
boreholes, cannot be predicted (NAS95).
A common practice in current exploration drilling is to leave the borehole open and allow it to
backfill naturally or assume the mud drilling fluid will act as a sealant. For air rotary drilling,
there is no drilling fluid filler. For mud rotary drilling, the mud drilling fluid may lose its
effectiveness as a sealant if the mud shrinks excessively as it dehydrates.
In an open abandoned borehole, the most likely materials to cause natural backfilling are the
loose granular surface materials or friable or loose tuffaceous formations. The only way that the
tuffaceous material could be loosened to fall into the open borehole would be by erosion (running
water), mechanical impact (scraping, etc.), or shock (seismic waves). If loose surficial materials
were washed or ran into the open hole, backfilling of an abandoned borehole could take place
relatively quickly. On the other hand, if the loose surface materials or materials from the
borehole wall were too large to fall freely, they could plug or bridge (stick together) in the
borehole. In such case, the top of the hole could be plugged, precluding backfilling.
8-89
-------
Time of Drilling
According to the NAS report, the predictions for how long into the future institutional controls
might survive and remain effective are arguable. The probability that an intrusion would occur in
a given future time period, such as in any one year, cannot be assessed from a technical basis
(NAS95).
Detection of Repository
Two drilling companies were contacted to determine the likelihood that an intact waste canister
could be penetrated with a drill being used in a conventional drilling operation. Mr. Leroy
Jochum (VIC96) stated that, irrespective of bit type (carbide, diamond, etc.) the drill would not
penetrate the canister but would most likely be deflected. If the driller wanted to penetrate the
canister, tools could be fabricated to cut the steel, but deliberate effort would be needed and it
would take a long time.
Mr. John Horton (LAY96) also indicated that special effort would be required to penetrate the
canister. It would require a concerted effort by the driller, possibly involving modification of the
bit and a considerable amount of time. He mentioned laser/plasma drilling technology that is
being developed by companies involved with DOE's Hanford Site in Washington State, and
stated that future technology might be able to penetrate a waste canister if adequate energy could
be applied to the drilling face.
These professional opinions indicate that present-day drilling technology could only penetrate a
waste canister if the driller was dedicated to doing so. Conventional drilling methods, without
special tools or spending an inordinate amount of time and effort, would not be able to breach the
canister. The possibility does exist that future technologies could do so, but it is not likely to be
easy. It is conceivable that the radioactivity produced by the repository could be detected using
remote sensing instruments and prompt further investigation.
In summarizing this issue, the NAS report concluded that the probability that a future intrusion
would be detected and remediated, either when it occurs or later, cannot be assessed from a
technical basis (NAS95).
8-90
-------
Mechanism for Wasle to Reach the Aquifer
In addition to the assumptions regarding the borehole, the contamination conditions intercepted
by the borehole mechanism by which the contamination is transported to the aquifer must also be
assumed in order to assess the source term.
8.4.4.2 Scenario Examples
As mentioned previously, the NAS report suggested a stylized intrusion scenario consisting of
one borehole of a specified diameter drilled from the surface through a canister of waste to the
underlying aquifer (NAS95). Two examples of such a scenario are described below.
Example I
The NAS report provided an example of a scenario which postulates current drilling technology,
but assumes sloppy practice, such as not plugging the hole carefully when abandoning it. It is
assumed that the intrusion occurs during a period in which some of the canisters will have failed,
but the released materials would not otherwise have had time to reach the ground water. In this
example, the original hole size, the modification of hole size by natural processes, and the
mechanism and processes for waste to reach the aquifer must be assumed or analyzed.
Example 2
A hypothetical, non-mechanistic scenario is another example. In this scenario, the entire content
of a single waste canister is emptied through the abandoned borehole into the aquifer. Evaluation
of drilling technology, drill size, modification of hole size by natural processes, and the
mechanism and processes for waste to reach the aquifer is unnecessary for this scenario.
8.4.4.3 Consequence Analysis
Having defined the reference scenario, the principal questions remaining are: 1) What
consequences should be assessed? and 2) How should the results be interpreted?
According to the NAS, the consideration of human intrusion cannot be integrated into a fully
risk-based standard because the results of any analysis of increased risk as a consequence of
intrusion events would be driven mainly by unknowable factors. The numerical value of the risk
of adverse health effects due to intrusion is the product of the frequency of an intrusion scenario
8-91
-------
and the measure of consequences. However, the frequency of an intrusion scenario in the distant
future cannot be determined in a technically-rigorous and defensible manner.
The NAS recommended that the Yucca Mountain standard require a consequence-only analysis,
without attempting to determine an associated probability for the analyzed scenario. The
calculations of consequences would provide useful information about how well a repository
might perform after an intrusion occurs. Such an analysis would evaluate whether the repository
would continue to be able to isolate wastes from the biosphere, or if its performance would be
substantially degraded as a consequence of an intrusion of the type postulated.
According to the NAS report, the performance of the disturbed repository should be assessed
using the same analytical methods and assumptions (including those about the biosphere and
critical groups) as those used in the assessment of the performance for the undisturbed case. This
analysis should be carried out to determine how the hypothesized intrusion event affects the risk
to the appropriate critical groups. The results of this calculation, however, constitute a
conditional risk, that is, one based on the occurrence of a hypothetical intrusion.
Because the probability of intrusion is inherently unknowable, the most useful purpose of this
type of analysis is to evaluate the incremental consequences resulting from an assumed scenario.
Since human intrusion of some type may be likely to occur in the future, the design of a
repository should be resilient to at least modest inadvertent intrusions. In other words, a
repository that is suitable for safe, long-term disposal of waste should be able to provide
acceptable waste isolation despite some type of intrusive event.
The NAS report recommends that EPA require that the conditional risk resulting from the
assumed intrusion scenario should be no greater than the risk levels that would be acceptable for
the undisturbed repository case. It is further recommended that compliance analysis not include
risks to the intruder or those arising from the material brought directly to the surface as a
consequence of the intrusion.
The following sections discuss the potential pathways of exposure that could occur using a
borehole scenario, as well as the consequences of intrusion occurring during specific time
periods of repository performance.
8-92
-------
Ground-Water Pathway from Abandoned Borehole
The location of the assumed borehole is a very important factor in evaluating its effects on
repository performance and radionuclide transport through ground water. The closer the
borehole location is to the boundary of the repository footprint and the location of the critical
group, the less time would be required for radioactive materials to travel to the critical group.
Consequently, a specific exposure scenario, with an appropriate critical group, would be required
for evaluating the ground water pathway from an abandoned borehole.
Air Pathway from Abandoned Borehole
In addition to the ground water pathway, an uncapped, abandoned borehole that penetrates into
the repository could provide a path for waste materials to be released to the atmosphere. The
radionuclide of primary concern for this air release pathway is carbon-14 (I4C). The travel time
for gaseous releases would depend on the location of failed waste canisters in relation to the
abandoned borehole and the manner in which the repository's openings have been backfilled.
Maximum exposures would occur from HC released from waste canisters that fail relatively
early. In comparison with natural pathways that exist at the Yucca Mountain site, an uncapped,
abandoned borehole would have an insignificant incremental effect on gaseous transport routes
to the surface. Although the presence of a borehole would not change the total release of I4C to
the surface, it could affect the routes used. Consequently, the effect of the borehole on potential
exposures to the public would be highly dependent on the assumed location of an exposed
individual(s) relative to the borehole and the tune at which nearby waste canisters failed.
For these reasons, it is concluded that the air pathway need not be considered as a measure of
repository performance in evaluating human intrusion scenarios.
Waste Materials Brought to the Surface by Human Action
The radioactive materials brought directly to the surface by the intrusive activity would pose
hazards to the intruders themselves and to the public. According to the NAS report, whenever
highly dangerous materials are gathered into one location and an intruder inadvertently breaks in,
that intruder runs an inevitable risk. Therefore, it would not be feasible to take regulatory actions
today to protect future intruders from the risk of their actions. However, requirements for active
8-93
-------
or passive institutional controls may provide some protection by decreasing the likelihood of
inadvertent intrusion.
The DOE containment and isolation strategy defines three post-emplacement time periods: the
containment period, in which the waste canisters remain intact; the transition period, during
which canister failure and waste mobilization are gradually increasing; and the peak dose period,
in which the canisters have failed and seepage of water into the repository is mobilizing the waste
radioactivity and transporting it to the environment. The physical condition of the repository will
change throughout these time periods and affect the circumstances of an intrusion scenario, as
outlined below.
Intrusion During the Containment Period
During the containment period, the waste containers remain intact. An intrusion of the repository
by drilling will either intercept an intact container or miss completely. If there is no interception
of a container, there will be no evidence to the drillers that a waste repository has been
penetrated. If an intact container is intercepted, it is unlikely that the drill bit will be able to
penetrate the container easily. Advance of the drill bit will be stopped or severely slowed,
leading to investigation of the cause for the resistance. If drilling persists, metal will be evident
in the cuttings and it will suggest that something unusual has occurred. Drilling may continue as
part of an investigation of the circumstances encountered, in which case portions of the container
and the waste form intercepted by the drill bit will be brought to the surface and exposure of
workers may occur. Alternatively, the drilling effort at that location may be abandoned with no
radiological consequences.
Intrusion After Initiation of Container Degradation
During the transition period, containment degradation is occurring as a result of corrosion of the
container and waste form caused by infiltration of water to the repository. A drilling intrusion oi
the repository might encounter an intact container, a partially degraded container, materials
between the containers that contain no radioactivity and thus give no evidence of the existence of
the repository, or materials between containers that contain radioactive waste material that has
been mobilized and has migrated some distance from the emplacement location.
This latter type of encounter would give no indication of the existence of the repository unless
the drilling cuttings were being monitored for radioactivity. An encounter with an intact or
8-94
-------
partially-degraded container would produce circumstances such as those described above for the
containment period, i.e., an effect on drilling progress, metal in the cuttings, and investigation of
the situation or abandonment of the drill hole.
Intrusion During the Peak Dose Period
In the peak dose period, it can be assumed that all containers have failed and all metals have
oxidized. The repository conditions will be similar to that of an ore body, with pockets of
radioactive materials at locations where containers used to be and radioactivity dispersed
throughout the repository. Depending on the extent of lateral migration and dispersion of
mobilized radioactivity from the waste, there may still be areas between the emplacement
locations where no radioactivity is present.
Under these circumstances, there may be nothing to suggest to a drilling operation that a
radioactive waste repository has been penetrated. Evidence might be available if cuttings are
being monitored for radioactivity or if it is noticed that some of the cutting materials are
composed of oxides of waste package materials. If neither of these pieces of evidence is
recognized, drilling operations will proceed as planned.
In summary, the consequences of intrusion, as an incremental effect on expected repository
performance, will depend on when the intrusion is assumed to occur. If, for example, intrusion is
assumed to occur late in the containment period, the effect on expected waste isolation
performance could be relatively significant because no releases are otherwise expected to occur.
However, the risks (probabilities and consequences) associated with such a scenario could be
extremely small. If the intrusion is assumed to occur in the peak dose period as defined by
DOE's waste isolation strategy, it may have an incrementally insignificant impact on repository
performance because radionuclide release is already occurring as a result of ongoing degradation,
release, and transport processes.
8-95
-------
REFERENCES
ASM89a American Society of Mechanical Engineers, Quality Assurance Program
Requirements for Nuclear Facilities, Report No. ASME NQA-1-I9S9. New York,
1989.
ASM89b American Society of Mechanical Engineers. Quality Assurance Program
Requirements for Nuclear Facilities Applications, Report No. ASME NQA-2y-
1989, New York, 1989.
AVI97 "Ground Water Pumpage Inventory, Amargosa Valley, No. 230, 1997," State of
Nevada.
BIO96 Biosphere Modeling for Dose Assessment for Radioactive Waste Repositories,
BIOMOVS II Technical Report No. 12, September 1996.
CHA94 Chambers, D. and L. Mays (eds.), The American Garden Guides: Vegetable
Gardening, Pantheon Books, New York, 1994.
CNW95 Center for Nuclear Waste Regulatory Analysis, Initial Analysis of Selected Site-
Specific Dose Assessment Parameters and Exposure Pathways Applicable to a
Ground Water Release Scenario at Yucca Mountain, CNWRA 95-018, San
Antonio, Texas, 1995.
CNW96 Center for Nuclear Waste Regulatory Analysis, Use ofGroundwater in the Arid
and Semi-Arid Western United States: Implications for Yucca Mountain Area,
Contract NRC-02-93-005, San Antonio, Texas, August, 1996.
CNW97 Center for Nuclear Waste Regulatory Analysis, Information and Analyses to
Support Selection of Critical Groups and Reference Biospheres for Yucca
Mountain Exposure Scenarios, CNWRA 97-009, San Antonio, Texas, August
1997.
DeL99 Michael DeLece, Amargosa Water Authority, personal communication, 1999.
DOE91 U.S. Department of Energy, Water Resource Assessment of Yucca Mountain,
Nevada, Revision 0, Study Plan for Study 8.3.1.9.2.2, Office of Civilian Nuclear
Waste Management, 1991.
DOE93a U.S. Department of Energy, User's Guide for GENII-S: A Code for Statistical and
Deterministic Simulation of Radiation Doses to Humans from Radionuclides in
the Environment, Sandia National Laboratories, SAND 91-2375, 1993.
8-96
-------
DOE94a U S Department of Energy, Total System Performance Assessment for Yucca
Mountain - SNL Second Iteration (TSPA-J993), SAND93-2675, April 1994.
DOE94b US Department of Energy, Parameters Used in the Environmental Pathways and
Radiological Dose Modules (DESCARTES, CIDER, and CRD Codes) of the
Hanford Environmental Dose Reconstruction Integrated Codes (HEDRIC),
Pacific Northwest Laboratory, PNWD-2023 HEDR, Revision 1, Richland,
Washington, 1994.
DOE95a U S. Department of Energy, License Application Annotated Outline,
Predecisional Preliminary Draft, YMP/94-05, Revision 01, Chapter 3, December
21, 1995.
DOE95f U.S. Department of Energy, Predecisional Preliminary Draft - Responses to
Questions From the Environmental Protection Agency Concerning Water
Resources and the Hydrologic Regime in the Yucca Mountain Vicinity, December
7,1995.
DOE96 U.S. Department of Energy, The Environmental Impact Statement for the Nevada
Test Site and Off-site Locations in the State of Nevada (Draft), Volume 1,
DOE/EIS 0243, Las Vegas, Nevada, 1996.
DOI63 U.S. Department of the Interior, Geology and Ground Water ofAmargosa Desert,
Nevada-California, Ground Water Resources - Reconnaissance Series, Report 14,
1963.
DRI94 Desert Research Institute, Potential Hydrologic Characterization Wells in
Amargosa Valley, Brad Pyles and Todd Miheve, September 1994.
EPA88 U.S. Environmental Protection Agency, Limiting Values of Radionuclide Intake
and Air Concentration and Dose Conversion Factors for Inhalation, Submersion,
and Ingestion, Office of Radiation Programs, Federal Guidance Report No. 11,
EPA 520/1-88-020, Washington, D.C., September 1988.
EPA89 U S Environmental Protection Agency, Risk Assessment Guidance for Superfund,
Volume I, Human Health Evaluation Manual (Part A), Office of Emergency and
Remedial Response, EPA 540/1-89-002, Washington, D.C., 1989.
EPA92 U.S. Environmental Protection Agency, Guidance on Risk Characterization for
Risk Managers and Risk Assessors, Deputy Administrator F. Henry Habicht II,
February 1992.
8-97
-------
EPA93 U.S. Environmental Protection Agency, External Exposure to Radionuclides in
Air, Water, and Soil, Office of Air and Radiation, Federal Guidance Report No.
12, EPA 402-R-93-081, Washington, D.C., 1993.
EPR94 Electric Power Research Institute, A Proposed Public Health and Safety Standard
for Yucca Mountain, APR TR-104012, Palo Alto, California, 1994.
FIE86 Fiero, B., Geology of the Great Basin, University of Nevada Press, Reno, Nevada
1986..
FRI94 Fridrich, C.J., W. W. Dudley, Jr., and J.S. Stuckless, Hydro geologic Analysis of
the Saturated-Zone Ground Water System, under Yucca Mountain, Nevada,
Journal of Hydrology, 154:133-168,1994.
HEN92 Henley, E.J., and H. Kumamoto, Probabilistic Risk Assessment: Reliability,
Engineering, Design, and Analysis, Institute of Electrical and Electronics
Engineering Press, New York, 1992.
IAE94 International Atomic Energy Agency, Handbook of Parameter Values for the
Prediction ofRadionuclide Transfer in Temperate Environments, Technical
Report Series No. 364, Vienna, Austria, 1994.
ICR77 International Commission on Radiological Protection, Recommendations of the
ICRP, ICRP Publication 26, Pergamon Press, 1977.
ICR85 International Commission on Radiological Protection, Principles of Monitoring
for the Radiation Protection of the Population, Publication 43, Annals of the
ICRP 15(1), Pergamon Press, 1985.
IUR89 International Union of Radioecologists, Sixth Report of the Working Group on
Soil-to-Plant Transfer Factors, Biltoven, The Netherlands: RIVM, 1989.
LAY96 Layne-Northwest, W229N5005 DuPlainville Road, Pewaukee, Wisconsin, 53072,
personal communication with Mr. John Horton, April 1996.
LLL78 Lawrence Livermore National Laboratory, Assessment of the Dose Commitment
from Ingestion of Aquatic Foods Contaminated by Emissions from a Proposed
Nuclear Fuel Reprocessing Plant, UCID-17882,1978.
LIN82 Linsley, R.K., M.A. Kohler, and J.L.H. Paulhus, Hydrology for Engineers,
McGraw-Hill, New York, 1982.
8-98
-------
LLL87 Lawrence Livermore National Laboratory, Retention by Vegetation of
Radionuclides Deposited in Rainfall - A Literature Summary, UCRL-53810,
Livermore, California, 1987.
MC92 McCracken, Robert D., The Modern Pioneers of the Amargosa Valley, Nye
County Press, Tonopah, Nevada, 1992.
MIK92 Miklas, M.P. Jr., et ah, Natural Resource Regulatory Requirements: Background
and Consideration of Compliance Methodologies, Center for Nuclear Waste
Regulatory Analyses, San Antonio, Texas, September 1992.
MLS93 Mills, L., Beginning Desert Gardening, University of Nevada Cooperative
Extension, Reno, Nevada, 1993.
NAP97 Napier, B.A. et. Al., GENII World Wide Web Site,
http://www.pnl.gov/health/health_prot/genii.html, 1997.
NAS95 National Academy of Science - National Research Council, Committee on
Technical Bases for Yucca Mountain Standards, Technical Bases for Yucca
Mountain Standards, National Academy Press, Washington, D.C., 1995.
NDC63 Nevada Department of Conservation and Natural Resources, Geology and Ground
Water of Amargosa Desert, Nevada-California, Water Resources-Reconnaissance
Series Report 14,1963.
NDC71 Nevada Department of Conservation and Natural Resources, Water for Nevada,
1971, Division of Water Resources Water Planning Report No. 3, 1971.
NRC83 U.S. Nuclear Regulatory Commission. Radiological Assessment: A Textbook on
Environmental Dose Analysis, NUREG/CR-3332, Office of Nuclear Reactor
Regulation, Washington, D.C., 1983.
NV95 Nevada Agricultural Statistics Service, Nevada Agricultural Statistics - 1995,
Reno, Nevada, September 1995.
NWP090a State of Nevada, Agency for Nuclear Projects/Nuclear Waste Project Office,
NWPO-SE-026-90, Native Americans and Yucca Mountain: A Summary Report,
by Catherine S. Fowler with data contributed by Maribeth Hamby, Elmer Rusco,
and Mary Rusco, September, 1990.
NWP095 State of Nevada Agency for Nuclear Projects/Nuclear Waste Project Office, State
of Nevada Socioeconomic Studies Biannual Report, 1993-1995, by Nevada
Socioeconomic Study Team: James Flynn (Project Director), James Chalmers,
Doug Easterling, Catherine Fowler, Richard Krannich, Howard Kunreuther,
8-99
-------
Ronald Little, C.K. Mertz, Alvin Mushkalel, K. David Pijawka, Paul Slovic, and
James Williams.
NYE93a Nye/Esmeralda Development Authority, Nye County Overall Economic-
Development Plan: Appendix L " Amargosa Valley Background Information",
(Draft), July 23, 1993.
NYE93b Nye/Esmeralda Economic Development Authority, Nye County Overall Economic
Development Plan (Draft), 1993.
NYE93c Nye County Board of Commissioners, 1990 Census Data Profiles Nye County.
Nevada and Its Communities; Social, Labor Force, Income and Poverty, and
Housing Characteristics, Planning Information Corporation, 1993.
N YE93d Nye County Board of Commissioners, Baseline Economic and Demographic
Projections: 1990-2010 Nye County and Nye County Communities, Planning
Information Corporation, May 25,1993.
NYE93e Nye/Esmeralda Economic Development Authority, Nye County Overall Economic
Development Plan; Appendix I 'Beatty Background Information,' Appendix J'
Pahrump Background Information, 'Appendix K Tonopah Background
Information' (Drafts), 1993.
ORN82 Oak Ridge National Laboratories, A Review and Analysis of Parameters for
Assessing Transport of Environmentally Released Radionuclides Through
Agriculture, ORNL-5786, Oak Ridge, Tennessee, 1982.
ROS92 Roseberry, A.M., and D.E. Burmaster, Lognormal Distributions for Water Intake
by Children, Risk Analysis, 12:99,1992.
STE95 Stellovato, Nick, Personal Communication to David Back 1995.
STI91 Stichler, C., Texas Alfalfa Production, B-5017, Texas Agricultural Extension
Service, College Station, Texas, 1991.
TRW96 TRW Environmental Safety Systems, Inc., Draft Summary of Socioeconomic
Data Analyses Conducted in Support of the Radiological Monitoring Program
During Calendar Years 1990 Through April 1996 (April '91, April '92, May '93,
June '94, June '95, June '96).
TRW98 TRW Environmental Safety Systems, Inc., Yucca Mountain Sike Characterization
Project: Summary of Socioeconomic Data Analyses Conducted in Support qflhe
Radiological Monitoring Program, April 1997 to April 1998, June 1998.
8-100
-------
USG72 U.S. Geological Survey, Water Supply for the Nuclear Rocket Development
Station at the U.S. Atomic Energy Commission's Nevada Test Site, Geological
Survey Bulletin 1938, Washington, D.C., 1972.
USG75 U.S. Geological Survey, Hydrogeologic and Hydrochemical Framework, South-
Central Great Basin, Nevada-California; With Special Reference to the Nevada
Test Site, Professional Paper 712-C, 1975.
USG76 U.S. Geological Survey, Summary Appraisals of the Nation's Ground-Water
Resources - Great Basin Region, Professional Paper 813-G, 1976.
USG88a U.S. Geological Survey, Major Ground-water Flow Systems in the Great Basin
Region of Nevada, Utah, and Adjacent States, Hydrologic Investigations Atlas
694-C, 1988.
USG9la U.S. Geological Survey, Ground Water Conditions in Amargosa Desert Nevada-
California, 1952-1987, U.S. Geological Survey Water Resources Investigations
Report 89-4101,1991.
USG95b U.S. Geological Survey, Selected Ground Water Data for Yucca Mountain
Region, Southern Nevada and Eastern California Colander Year 1993, U.S.
Geological Survey Open File Report 95-158, 1995.
USG95c U.S. Geological Survey, Simulated Effects of Proposed Ground Water Pumping in
17 Basins of East-Central and Southern Nevada.Waler Resources Investigation
SReport 95-4173, Nevada, 1995.
VAN90 Van der Leeden, F., F.L. Troise, O.K. Todd, The Water Encyclopedia - Second
Edition, Geraghty & Miller Groundwater Series, 1990.
VIC96 Vickery Drilling, P.O. Box 9013, Evansville, Indiana, 47724, personal
communication with Mr. Leroy Jochum, April 1996
Y072 Young, R.A., Water Supply for the Nuclear Rocket Development Station at the
U.S. Atomic Energy Commission's Nevada Test Site, Geologic Survey Water-
Supply Paper 1938, U.S. Department of the Interior, Washington, D.C., 1972.
8-101
-------
CHAPTER 9
YUCCA MOUNTAIN EXPOSURE SCENARIOS
AND COMPLIANCE ASSESSMENT ISSUES
9.1 INTRODUCTION
Chapter 7 described the proposed Yucca Mountain repository in terms of its site characteristics,
engineering designs, and current performance assessments. In Chapter 8, derived unit-
concentration exposure and risk estimates were presented. These estimates were based on a
conceptual model in which radionuclides were released into the biosphere from a repository
through the following sequence: degradation and failure of the waste canister(s) through
corrosion; release of radionuclides from the waste package into host rock; migration of
radionuclides through the unsaturated zone into the aquifer (saturated zone); and dissemination
of contaminated ground water to wells used for drinking and agricultural purposes.
This scenario, Involving a gradual release from an undisturbed repository, characterizes the most
probable events and conditions of future human exposure. It also conforms with the primary
objective of deep geological disposal which is to provide long-term barriers that isolate wastes
and limit the release of radionuclides into the biosphere by virtue of siting and engineering
design. Deep geologic disposal isolates the wastes for a sufficiently long period of time to allow
most of the radionuclides to decay to natural background levels. While estimates of dose and
risk for this gradual release process cannot be calculated with complete precision, there is a
substantial scientific basis for modeling the various processes that take into account parameter
variabilities. By means of statistical processes, such as the Monte Carlo method (see Section
8.5), these uncertainties can be rninimized, thereby yielding dose/risk estimates that are
reasonable.
Figure 9-1 illustrates the major release pathway leading to human exposure which involves
ground water from an undisturbed repository at Yucca Mountain. The major reservoirs (source
terms) containing radionuclides at various times following closure are depicted as rectangles.
These reservoirs do not have discrete physical boundaries, but rather form a continuum. Solid
arrows between reservoirs represent the probable processes by which radionuclides are
transported from one reservoir to another in an undisturbed repository.
9-1
-------
Critical group for
population dose
Critical group?
Figure 9-1. Schematic Illustration of the Major Pathways from a Repository at
Yucca Mountain to Humans (NAS95)
9-2
-------
Major processes and events23 with potential to modify normal behavior or drastically alter the
physical integrity of reservoirs are shown in Figure 9-1 as diamonds. These modifiers are
connected by dashed lines to those reservoirs upon which they are likely to have the most
significant impact.
To ensure maximum public protection, a standard for a repository at Yucca Mountain must also
consider: 1) release pathways other than ground water and 2) improbable conditions that may
lead to individual doses and risks well in excess of those specified for the undisturbed repository.
A demonstration of compliance with such a standard, therefore, requires estimating potential
doses resulting from secondary release pathways and predicting improbable events and processes
that may disturb the repository and their corresponding outcomes.
This chapter summarizes issues involved hi developing a repository standard that addresses the
important exposure pathways and related performance issues that have been identified,
9.2 GASEOUS RELEASES: A SECONDARY PATHWAY FOR HUMAN EXPOSURE
The primary pathway of radionuclide releases from an undisturbed repository involves the
introduction of radionuclides into the underlying aquifer, contaminating wells used for drinking
or agricultural irrigation. However, as shown in Figure 9-1, humans could also be exposed to
radiation as a result of gaseous emissions from the repository. Due to the ease with which
gaseous contaminants are distributed in the atmosphere, human exposure would not be limited to
the nearfield population but could extend to the world at large. The radionuclide with the highest
potential for gaseous release and human exposure is carbon-14.
This section provides a brief overview of the primary parameters affecting the timing and
magnitude of gaseous releases and assesses bounding values for human doses.
23The difference between an event and a process is the time interval over which the phenomenon occurs
relative to the time frame of interest; events occur over relatively short time intervals and processes occur over
relatively long time periods. For example, a disruptive seismic event may occur over minutes, hours, or days. Even
a volcanic eruptive cycle that may have time frames extending over several years must be considered an event when
judged in context with the lifespan of humans and/or a million-year repository assessment period. Phenomena that
exceed human life expectancy or occur over a significant portion of the period of regulatory concern are considered
to be processes.
9-3
-------
9.2.1 Production and Early Containment of Carbon-14
Carbon-14 is produced in nuclear fuel as a result of neutron absorption by the following
reactions: (1) N-14 (n, p) C-14 and (2) O-17 (n, He-4) C-14 (DAV79). Thus, the quantity of C-
14 produced is governed by the amount of nitrogen and oxygen contained within the fuel core.
Carbon-14 in oxide fuels is assumed to exist as either CO2 or low molecular weight hydrocarbons
that in time are oxidized to CO2. The total inventory of C-14 for the 63,000 tons of spent nuclear
fuel is estimated to be about 91,000 Ci (DOE94).
When estimating gaseous releases from the repository through the unsaturated zone to the
accessible environment, the following parameters should be considered: (1) container
performance and (2) the bulk permeability and retardation capability of the tuffs. The retardation
of gaseous CO2 flow is due to its exchange with the relatively immobile bicarbonate (HCO3~) in
the pore water of the unsaturated zone.
9.2.2 Impacts of Thermal Loading on Gaseous Releases and Transport
The emplacement configuration of heat-generating waste containers is likely to disturb the
ambient environment of the repository in a number of ways. Waste-generated heat is expected to
enhance vaporization of water within the tuff matrix and, at temperatures above 96 °C,
completely "dry out" the adjacent host rock and move the water into the surrounding rock.
The impact of thermally-displaced water in the vicinity of the repository has a dual effect on
gaseous releases. Since most of the waste-container corrosion processes are known to be
temperature and moisture dependent inclusive of: 1) general aqueous corrosion, 2) steam
corrosion, 3) pitting corrosion, 4) dry oxidation corrosion, and 5) stress corrosion, the potential
impacts of waste emplacement and thermal loading on container failure are highly critical for
modeling the time and release fraction of gaseous C-14.
Estimates of travel time for C-14 released from a container into the unsaturated zone are strongly
affected by the moisture content. Under conditions of 100 percent humidity, C-14 is assumed to
exist for the majority of the time as bicarbonate (HCCV) in the slow-moving aqueous phase
(ROS93). Conversely, within the dry-out zone, C-14 can be assumed to exist almost exclusively
in the fast-moving gaseous form (CO2).
9-4
-------
9.2.3 Estimates of Travel Time
Estimates of travel time for C-14 released from a failed container to the accessible environment
are complicated by the fact that the radionuclide is likely to exist only a small portion of the time
in gaseous form as 14C02; the majority of the time, it will exist as bicarbonate (HC03~) in the
aqueous phase. In the bicarbonate form, C-14 moves more slowly than in the fast moving
uncondensible gaseous form (ROS93). This "slowing," or retardation, must be incorporated into
the travel-time calculations by dividing the short-lived gas velocity at each point along the flow
path by a retardation factor that accounts for the longer tune and limited movement of C-14 in
the aqueous bicarbonate phase. Travel-time probability distributions can be determined by
coupled calculations of gas and heat flow (i.e., time-dependent temperature distributions in the
repository environs).
Because estimates of early waste-container failure and release of C-14 are currently highly
uncertain, travel times for release of C-14 have been estimated at 1,000-year intervals following
waste emplacement. For example, Figures 9-2 and 9-3 show travel-time histograms for a thermal
loading of 57 kW/acre and welded-tuff bulk permeability of 10'11 m2 at 1,000 and 10,000 years
(DOE94):
• At 1,000 years, temperature gradients in the vicinity of the repository are high due to the
large heat output. Correspondingly, gas velocities in the nearfield ("dry-out zone") are
larger than in the far field. Calculated C-14 travel times range from 200 to 600 years.
• At 10,000 years, heat has been conducted outward and temperature gradients have been
reduced, resulting in estimated travel times that range from 500 to 1,200 years.
Given this, it appears that the magnitude of potential atmospheric releases would be greatest if
containment failure were to occur early. At early times (i.e., 1,000 years), transport velocities
can be expected to be maximal and the reduction of C-14 by natural decay is minimal.
It is reasonable, however, to expect that individual container failures will occur over a long
period of time. This could substantially broaden the range of C-14 travel times in the unsaturated
zone from as little as 200 years to as much as 1,800 years. The period of C-14 release into the
accessible environment is further delayed by the fact that, at time of canister failure, only a small
percentage of the C-14 inventory has leaked from the fuel matrix into the void spaces of the
container for instantaneous release. Barnard et al. (BAR92) estimate that this quick release
9-5
-------
20 n
&
I
H
tw
1 n
500 1000 1500
Trawl time (yeors)
2000
Figure 9-2. Retarded Travel Times of C-14 from the Repository to the Atmosphere for
Particles Released at 1,000 years. Source: DOE94.
30
S
20-
O-i
C
1
-
Mi
1
M
5C
—
)0
••
-. j] ] ; )
1000 '1500" 2000
Travel time (yean)
Figure 9-3. Retarded travel time of C-14 particles from the repository to the atmosphere for
particles released at 10,000 years. Welded-tuff bulk permeability of 10'11 m2.
Source: DOE94.
fraction is likely to represent between 1.25 and 5.75 percent of the total inventory. The slow
release of the larger, remaining fraction of C-14 from the fuel matrix to the container and
subsequently into the repository environs is likely to further extend the time period during which
C-14 can be expected to enter the accessible environment. With a physical half-life of 5,730
9-6
-------
years, it is reasonable to conclude that the quantity of C-14 that will reach the accessible
environment will be less than the 91,000 Ci inventory existing at the time of waste emplacement.
9.2.4 Dose Modeling and Exposure Estimates
For practical reasons, estimates of human exposure to C-14 assume that the specific activity of
C-14 in the atmosphere in gaseous CO2 form is equal to that of organically bound carbon
contained in all plant and animal products that may be ingested as food. Thus, pathways for
internal exposure may involve inhalation and ingestion.
For all practical purposes, under a steady-state distribution of C-14 in the environment, the
inhalation of 14C02 contributes insignificantly when compared to the ingestion pathway and may,
therefore, be eliminated from dose consideration.
Upon the ingestion of organically bound C-14, the uptake, retention, and excretion by the body
involve numerous pathways that correspond to biologic half-times ranging from less than one
hour to several years. Even for a specific category of organic molecules such as proteins,
turnover times are highly variable. While structural proteins show relatively long turnover times,
other proteins such as enzymes, plasma albumin, and hemoglobin have relatively short turnover
times. When all protein compartments are considered, the half-time of carbon (and therefore C-
14) is estimated to be 119 days (NCR93). For fats, which are largely stored in the body as
adipose tissue, the biological half-time of carbon is estimated to be 99 days; for carbohydrates,
the half-time is estimated to be one day. For a daily dietary intake of 300 grams of carbon, a
weighted biologic half-tune of about 39 days is obtained. For dosimetric purposes, the ICRP has
suggested a biological half-time of 40 days for C-14 (ICR82).
For steady-state environmental conditions, estimates of individual organ and whole body doses
from ingestion have been derived by Killough and Rohwer (KIL78). Their model assumes that
the specific activity of C-14 (i.e., pCi/g carbon) in the human body will, in time, be the same as
that observed in environmental media, inclusive of all plant and animal food products.
Correspondingly, the model takes into account the carbon content of individual tissues and
organs that will be subject to the beta-ray exposure of C-14. At the present specific activity of C-
14 in the atmosphere of seven pCi/g carbon, C-14 is estimated to contribute an annual dose of
about 1.5 mrem to humans throughout the world.
9-7
-------
9.2.5 Dose Estimates from Repository Releases
Global Doses. Any gaseous release of C-14 from the proposed Yucca Mountain repository will
disperse itself globally and, therefore, lead to relatively constant exposures among individuals
within the world community. The global distribution model yields a population dose estimate of
399 person-rem per curie of C-14 for a world population of 12.2 billion over a 10,000-year
period (EPA96). Using this dose-conversion factor and assuming that the entire repository
inventory of 91,000 Ci of C-14 is released, an average individual dose of about 0.0003 mrem/yr
is estimated. If this release is scaled down to the C-14 release from the 18 failed packges used in
the TSPA-VA analyses (18 packages over 10,000 years), where each package has an inventory of
11.7 Ci, the global dose estimates become extremely small, particularily in comparison with the
above estimate for doses from atmospheric C-14.
Local Doses. Estimating potential doses from gaseous releases from the repository is a very
complicated assessment, particularily since any potential doses received would be strongly
influenced by wind direction and population distributions. In addition, estimating the amount of
C-14 that would be released at the ground surface in gaseous form rather than contained as
biocarbonate ion dissolved in water is also difficult. Some insight relative to the potential
magnitude of doses through the gaseous release pathway can be gleaned from looking at dose
estimates from gaseous C-14 emissions from a nuclear power plant. In both the repository and
nuclear power plant situations, gaseous C-14 is released into the atmosphere and dispersed
downwind of the source.
Doses Within 50 Miles of a Nuclear Power Plant. In another study, air concentrations of 14CO2
were modeled out to 50 miles for the Dresden Nuclear Power facility (NCR93). A standard
diffusion model, as defined above, and local meteorological data were used to calculate
concentrations for all sectors out to 50 miles from the plant. Figure 9-4 identifies isolines of the
average annual 14CO2 air concentrations.
The numbers not in parentheses in Figure 9-4, multiplied by the continuous source activity per
second (e.g., Ci/s), yield the predicted 14C concentrations in Ci/m3. However, to estimate uptake
by growing vegetation, the concentration of 14C should be given as a ratio to stable carbon. To
make the example more relevant, a continuous emission rate of 100 Ci/y of 14C was used as the
source strength. With the additional assumptions listed on the figure, the isolines are also
labelled, in parentheses, in units of specific activity (pCi l4C/g ' C).
9-8
-------
Parenthetical values may also be converted to dose based on the relationship that on average
seven pCi 14C per gram of carbon is estimated to result in an annual dose of about 1.5 mrem (see
Section 9.2.5). Based on the model defined in Figure 9-4, annual doses from a continuous
release of 100 Ci of I4C can be estimated at various distances as defined in Table 9-1. The
isolines can be scaled linearly to source strengths other than 100 Ci/yr.
Nuabtri IB p*riothtili eorrupond to:
1) tourct 100 Ci/jr«*r
I) CO. crac«atr« tleo 3JO vL/L
J) «lt feaiity ef 1.2 x 10Jg a'1 »d
•r« |lTtn lo unit* ef pCl/jC
U.10-S,
12.7110-2
. .
(2.TilO'z)
Figure 9-4. Annual Average Concentration for Uniform Continuous Source and Specific
Activity (in parentheses) for 100 Ci/year Source (NCR93)
9-9
-------
Table 9-1. Annual Average Doses Resulting from the Release of 100 Ci I4CO2
for Distances Out to 50 Miles
!-«^, •
10
20
30
40
50
.;
l.SxlO'1 -
9.1 x 10'2
5.5 x lO'2
3.6 xlO'2
2 7 x 10'2
,Aaa«al D^se- CfjiDiE
3.8 x 10'2
2.0 xlO'2
1.2 x lO"2
7.7 x 10'3
5.8 x lO'3
Numbers were based on atmospheric CO2 concentration of 350 nL/L and air density of
1.2xlO-3g/cm3.
For the Yucca Mountain situation, the assumed 100 Ci/yr source release assumed in the power
plant assessment can be approximately scaled to estimate potential doses from failed waste
packages. The quick release fraction of C-14 in the waste packages may vary from 1.25 to 5.75%
of the inventory. Using the high end estimate and a C-14 inventory of 11.4 Ci/waste package
(DOE98, Vol. 3 p. 3-96), 0.673 Ci/waste package would be released. Assuming that this
instantaneous release is transported completely to the ground surface and released, scaling the
doses calculated for the power plant (for a location of 20 km from the source) gives a dose
estimate on the order of 1.3 x 10"4 mrem/yr for gaseous releases from a single waste package.
This estimate represents a conservative upper bound estimate in that actual gaseous releases from
the repository would be attenuated during upward transport through the unsaturated zone by
dissolution in pore waters in the overlying rocks, and by partitioning between fractures and the
matrix porosity voids in the overlying rocks. Gaseous C-14 releases after initial breaching of the
waste package are likely to be less than the instantaneous release fraction since ground waters
interacting with the wastes would dissolve the C-14, with its subsequent transport through the
ground water pathway rather than gaseous release.
9.2.6 Potential Non-Radiological Impacts of C-14
A concern uniquely associated with some contaminant radionuclides involves the transmutation
effect and its potential for inducing molecular disorientation. The potential impact of chemical
transmutation is of particular concern for genetic macro-molecules of DNA and RNA. Chemical
transmutation refers to the fact that when a radioactive isotope emits a beta particle, it also
undergoes chemical transformation due to the change in atomic number. For example, when C-
14 undergoes radioactive decay, it becomes nitrogen. When such atoms are incorporated in
critical molecules such as DNA, the resulting change in atomic number, recoil, or excitation may
9-10
-------
give rise to biologic effects, including mutation, beyond those induced by the attendant ionizing
radiation. At issue, therefore, is whether or not dose-response values, involving
cytogenetic/genetic effects for absorbed radiation energy, might underestimate the hazards
presented by these potential radionuclide contaminants. Potential impacts of transmutation have
been reviewed by the National Academy of Sciences (NAS). In their first report, the NAS
Committee on the Biological Effects of Ionizing Radiation (BEIR) concluded:
...that the genetic effects of decays ofH-3, C-14, andP-32 can, in fact, be
attributed almost entirely to their beta radiation and that the contribution from
transmutation is so small in comparison that it is justified to consider the main
effect to come from the radiation emitted-when the isotope disintegrates (NAS72).
However, in the Committee's subsequent report (BEIR III), evidence was acknowledged which
indicated a modest transmutation effect when C-14 (and H-3) occupied highly specific locations
within DNA (NAS80). The Committee concluded that it still seems unlikely that neither H-3 nor
C-14 decay are significantly underestimated by considering only the ionizing radiation dose
accumulated by germ-line cells .
9.3 DEVELOPMENT OF PERFORMANCE SCENARIOS AND COMPLIANCE ISSUES
9.3.1 Identification of Improbable Phenomena
For a regulatory time frame that can extend to thousands of years, it is reasonable to conceive of
circumstances defined by various natural and human-induced events and processes that may
result in some persons at some time being exposed to levels well in excess of anticipated levels
considered acceptable for an undisturbed repository. In recognition of the need to address
repository performance under disturbed conditions, the NAS Committee on Technical Bases for
Yucca Mountain Standards stated the following:
... the probabilities and consequences of modifications by climate change, seismic
activity, and volcanic eruptions at Yucca Mountain are sufficiently boundable that
these factors can be included in performance assessments that extend over this
time frame. ...The challenge [therefore] is to define a standard that specifies a
high level of protection but that does not rule out an adequately sited and well-
designed repository because of highly improbable events (NAS95). (Emphasis
added.)
Substantial difficulties are likely to be encountered in making these predictions. Both the NRC
and EPA have explicitly recognized that no analyses of compliance will ever constitute an
absolute proof; the objective instead is a reasonable level of confidence in analyses that indicates
9-11
-------
whether limits established by the standard will be exceeded. Thus, in 40 CFR Part 191
(Appendix B), the EPA stated the following for a disturbed disposal system:
In making these various predictions, it will be appropriate for the implementing
agencies to make use of rather complex computational models, analytical
theories, and prevalent expert judgement relevant to the numerical predictions.
Substantial uncertainties are likely to be encountered in making these predictions.
In fact, sole reliance on these numerical predictions to determine compliance may
not be appropriate; the implementing agencies may choose to supplement such
predictions -with qualitative judgement as well (EPA85). (Emphasis added)
Similarly, hi 10 CFR Part 60, the NRC acknowledged that for performance assessment "...it is
not expected that complete assurance that they [performance objectives and criteria] will be met
can be presented." (NRC81)
Events and processes that may require consideration are not limited to those identified by the
NAS and shown hi Figure 9-1. Over the years, numerous reports have identified generic events
and processes that do not consider geographical or site-specific features (DOE74, DOE79,
BUR80, IAE83, AND89, and DOE90a). Table 9-4 represents a consolidated listing that was
used as a starting point in the development of disruptive scenarios for the Waste Isolation Pilot
Plant (WIPP).
9.3.2 Screening of Events and Processes
Not all events and processes cited in Table 9-2 need necessarily be considered for Yucca
Mountain. Phenomena such as erosion, sedimentation, etc. are certain to occur during extended
tune periods such as the NAS-suggested one million-year time frame. This suggests that these
phenomena should be part of the base-case scenario. The effects of other events (e.g., sea-level
variations, hurricanes, seiches, and tsunamis) are restricted to coastal areas.
To analyze the potential relevance of events and processes to a specific repository site, three
criteria must be considered:
• Probability of occurrence
• Physical reasonableness
• Consequence
9-12
-------
Table 9-2. Potentially Disruptive Events and Processes (DOE91)
<-eiestial Bodies:
Meteorite Impact
Surficial Events and Processes:
Erosion/Sedimentation
Glaciation
Pluvial Periods
Sea-Level Variations
Hurricanes
Seiches
Tsunamis
Regional Subsidence or Uplift
Mass Wasting
Flooding
inadvertent Intrusions
Explosions
Drilling
Mining
Injection Wells
Withdrawal Wells
Hvdrologic Stresses:
Irrigation
Damming of Streams and Rivers
Repository- and Waste-Induced Events and Processes-
Caving and Subsidence
Shaft and Borehole Seal Degradation
Thermally-Induced Stress Fracturing in Host Rock
Excavation-Induced Stress Fracturing in Host Rock
Gas Generation
Explosions
Nuclear Criticality
To analyze the likelihood of a given event, it is most desirable to express its probability of
occurrence in quantitative terms that draw on scientific data. Physical reasonableness as a
screening criterion is a qualitative estimate of low probability that reflects subjective judgment.
For subjective probability, the ICRP states:
...a number is assigned to the likelihood of an event occurring in a defined period of time,
as a measure of the degree of belief that the event will actually occur during that time....
The assignment can be made on the sole basis of subjective judgement, no statistical
experience being needed. The result is conceptually identical to a traditional probability
and can be used in the same -way (ICR85a). (Emphasis added)
In instances where events are assigned subjective probabilities of occurrence, the ICRP offers an
additional note of caution:
It is important to distinguish between the degree of belief and the idea of confidence
limits applicable to an estimate of probability, which itself has some associated
uncertainty.
The third screening criterion is consequence. An assessment of consequence determines whether
the event or process either alone or in combination with other phenomenon may adversely affect
performance of the repository.
9-13
-------
On the basis of these criteria, a proposed future standard may, for example, specify that events
and processes with less than a specified chance of occurring within the regulatory period do not
have to be considered in scenarios used to demonstrate compliance with the standard.
Conversely, physically reasonable events and processes with significant impacts and probabilities
greater than a threshold value would be considered for scenario development.
The likelihood of a disruptive event and its consequence must also be defined temporally. For
some events (e.g., meteorite impact), the probability of occurrence over time is a constant. For
these cases, the probability of events occurring within a year's time interval can be assessed from
Poisson statistics. For other types of events, the probability of occurrence will vary with time
after repository closure, or it may be co-dependent on the occurrence of other time-dependent
events. This second and more complex event scenario is described in ICRP Publication 46
(ICR85b) and is illustrated in Figure 9-5. For this type of event-induced scenario, the
probabilistic annual individual dose rate is a function of both the time of occurrence of the
initiating event, t, and the time elapsed since its occurrence, (T-t).
9.3.3 Compliance With a Standard
It is the responsibility of the Nuclear Regulatory Commission (NRC) to assure compliance with
EPA's environmental radiation protection standards for Yucca Mountain. Accordingly, the
following discussion is for illustrative purposes only and not indicative of how the NRC will
discharge this responsibility.
A total system performance assessment employs a quantitative approach that characterizes the
releases and health impacts of the disposal system. Key questions that must be addressed in the
performance assessment are: "How reliable are the models employed in the performance
assessment?" and "What is the uncertainty in the results of the performance assessment?"
Preceding portions of this document have acknowledged uncertainties associated with all major
elements affecting repository performance. Important sources of uncertainties pertain to the
appropriateness of selecting scenarios representing conditions far into the future; the variability
and/or lack of knowledge regarding many parameter values employed by the models; the
reliability of historical data in predicting the time-dependent probability of future events that may
disrupt the repository; and the complex, but uncertain, interaction of independent variables on
repository performance. While the uncertainty for some of the sources can be reasonably
quantified (e.g., quantities of food and water ingested by humans), others are considerably more
9-14
-------
difficult (e.g., the probability of human intrusion). While there are no rigorous techniques for
quantifying or eliminating uncertainties, several techniques for mitigating then- impacts have
been proposed by Bertram-Howery and Hunter (BER89), as summarized in Table 9-3.
1
4
c
1 *
S c
.£ o
T»
Tim« (t)
Figure 9-5. An Illustration of Hypothetical Individual Dose Rates Associated with A
Disruptive Event Happening at Two Different Times after Disposal of
Radioactive Waste
The EPA has acknowledged that performance assessments will contain uncertainties and that
many of these uncertainties cannot be eliminated. Accordingly, the EPA has previously stated
that:
...standards must accommodate large uncertainties, including uncertainties in our
current knowledge about disposal system behavior and the inherent uncertainties
regarding the distant future. (EPA85)
Uncertainty and sensitivity analyses are, therefore, important aspects of performance assessment.
Uncertainty analysis involves determining the uncertainty in model projections that results from
imprecisely known (or variable) model input parameters. Sensitivity analysis involves
determining the contribution of individual input parameters to the uncertainty in model
predictions
9-15
-------
Table 9-3. Techniques for Quantifying or Reducing Uncertainty in the Performance
Assessment
scenarios
'Completeness, Logic, and
probabilities)
Expert Judgment and Peer Review
Quality Assurance
Conceptual Models
Expert Judgement and Peer Review
Sensitivity Analysis
Uncertainty Analysis
Quality Assurance
Computer Models
Expert Judgment and Peer Review
Verification and Validation*
Sensitivity Analysis
Quality Assurance
Darameter Values and Variability
Expert Judgement and Peer Review
Data-Collection Programs
Sampling Techniques
Sensitivity Analysis
Uncertainty Analysis
Quality Assurance
* To the extent possible.
Source: BER89
Because of the many uncertainties associated with the events and processes affecting repository
performance, probability distributions of human exposure (and risk) are likely to vary over
several orders of magnitude within the 5th and 95th percentile range. An important limitation of
such a probability distribution is that no single value is correct in predicting future exposures.
The probability distribution, however, does identify mean and median values, which represent
expected values of dose (or risk) most likely to be received by individuals considered at
maximum risk. To that extent, EPA (EPA85) has previously acknowledged that the most
probable (or expected) value of a probabilistic distribution of estimated radiation exposure may
be used to demonstrate compliance:
...the implementing agencies need not require that a very large percentage of the range of
estimated radiation exposures.. .fall below limits [of the standard]. The Agency assumes
that compliance can be determined based on best estimate predictions (e.g. the mean or
the median of the appropriate distribution, whichever is higher).
9-16
-------
9.3.4 Development of Site Performance Issues
The subject of defining performance issues for a geologic repository has been addressed on a
generic basis both in this country and abroad. Processes and events that could potentially affect
repository integrity and performance have been identified so that they can be critically examined
to develop site-specific performance scenarios for a particular candidate repository site. The
challenge in developing site-specific scenarios for a candidate repository location is to credibly
incorporate all the relevant processes and events that significantly affect repository performance.
To develop site-specific scenarios that will be addressed in licensing assessments, an iterative
process of site characterization and evolving performance assessments has been used since the
initiation of extensive site characterization work. Chapter 7 describes the sequence of total
system performance assessments for the Yucca Mountain site that have been performed by DOE
as site characterization has proceeded. These assessments have identified which processes and
events have most relevance to sub-system and total system performance, and indicated the areas
where more rigorous assessment capability was needed and areas where more extensive
laboratory and field studies are necessary. As a result of this iterative site characterization -
performance assessment process, a number of technical issues have been defined through
interactions between the DOE and the NRC staff. Resolution of these issues is critical in
developing a credible assessment of the site's performance that can be carried into the licensing
process. These issues are discussed in more detail later in this section.
The progress of DOE's Yucca Mountain site characterization efforts is documented in semi-
annual progress reports prepared as required by provisions of the Nuclear Waste Policy Act of
1982 (NWP83). Descriptions of the laboratory and field studies conducted by the DOE and other
parties are reported in numerous reports published in the open technical literature. In Chapter 7,
descriptions of the total system performance assessments carried out by the DOE, the NRC, and
EPRI are given. These descriptions illustrate the iterative and evolving nature of the site
characterization and performance assessment efforts carried out by these organizations. In
developing the Yucca Mountain standard, the Agency has relied heavily on the information
provided from these sources as well as the reviews of these efforts by experts outside the
Program.
The NRC staff have been in dialog with DOE concerning technical issues for the repository
since inception of site characterization and repository design work. As a result of this dialog, the
staff has identified 10 Key Technical Issues (KTIs) to be resolved as part of demonstration
9-17
-------
of regulatory compliance, and has established Issue Resolution Status Reports (IRSRs) as the
means by which the requirements for, and the status of, issue resolution are documented (see
Chapter 4 of this BID).
The IRSRs are updated periodically as progress is made in issue resolution as a result of, for
example, data additions or improvements in performance models. Resolution of the KTIs is
sought before submission of the LA, but any issues resolved to the satisfaction of DOE and NRC
staff during pre-licensing interactions are subject to being opened again during licensing reviews,
e.g., by the licensing board.
9.3.4.1 Reviews of Recent Yucca Mountain Performance Assessments
DOE's Viability Assessment (DOE98) provided the most recent basis for NRC expression of
issues of current concern. Although NRC had no statutory or other official role in review of the
VA, staff performed a review of the total system performance assessment (TSPA) and licensing
plan elements of the VA (i.e., cost estimates were not reviewed) as an aid to DOE's development
of a complete and high-quality LA. The NRC comments were officially transmitted to DOE in a
June 2, 1999 letter (NRC99). This transmittal of the results of the staff review had been
preceded by informal expression of comments and an NRC/DOE Technical Exchange (NRC99a)
in which NRC staff provided detailed feedback to DOE concerning issues associated with the
TSPA-VA.
Comments on the VA were also documented by the NWTRB (TRB99) and DOE's independent
Peer Review Panel (PRP99). In general, the NRC, NWTRB, and Peer Review Panel comments
(summarized in Section 7.2.4 of this BID) were consistent with DOE's own assessment of issues
associated with the VA's reference engineered design and the status of data and models used in
the TSPA-VA (documented in Volume 4 of the VA, DOE98). Overall, there was consensus that
the data base and performance models available for the VA were inadequate for the LA, and that
there were technical issues stemming from the VA reference repository design that would be
difficult to resolve in licensing reviews. These findings were consistent with DOE's assertion
that the VA was a snapshot in tune of evolution of repository design, performance modeling,
and the technical data base.
9-18
-------
9.3.4.2 Compliance Issues for Licensing Reviews
On the basis of their VA reviews, NRC staff identified, in NRC99, the following current issues,
in order to help DOE "...to focus its program and develop a high-quality LA." Broad issues
identified by the NRC staff are identified below along with associated issues more specific to the
performance of the natural and engineered barriers.
While the growing emphasis on the engineered barriers is apparent in the evolution of DOE's
efforts, there are some data gaps in the characterization of the natural barrier system that have
important performance implications. Hydrologic characterization data are sparse for the down
gradient portion of the ground water flow system at distances greater than 5 km from the
repository. The absence of hydrologic data seriously limits the reliability of radionuclide
transport calculations for projected repository releases. Some specific technical issues involved
include:
• The range of infiltration estimates for water moving through the unsaturated zone
and entering waste emplacement drifts in the repository
The nature of the flow regime in the alluvial deposits lying beyond the Yucca
Mountain location southward into the Amargosa Valley area, i.e., where and how
flow from the tuff aquifer enters the alluvial sediments
• The extent of down gradient dispersion expected for the contamination plume
from ant repository releases and the extent of mixing that can be assumed when
releases from the repository enter the saturated zone beneath the repository
The areal extent of the upward hydrologic gradient assumed to exist between the
lower carbonate aquifer and the overlying tuff and alluvial flow system
These concerns, and planned efforts to address them and others, are discussed below.
Timely Selection of a Preferred Repository Design
NRC staff noted that the TSPA-VA identified, and provided preliminary evaluations of, design
enhancements in comparison with the VA reference design that could provide defense-in-depth
and improved performance for the engineered features of the repository system. The staff
expressed concern that continued retention of flexibility and analysis of design options might not
permit DOE to meet its schedule for preparation of a high-quality LA focused on a specific
design.
9-19
-------
DOE's selection of the EDA II design as the basis for future evolution of engineered barrier
system design provides focus for future design development efforts. It will be necessary,
however, that evolution of the data base and performance models needed to support use of the
EDA II design (or its progeny) in the LA proceed at a rate, and with content, sufficient to produce
a high-quality LA.
Waste Package Corrosion
Under present repository design concepts and waste isolation strategy, waste package corrosion
resistance (i.e., lifetime before penetration which allows water to contact the waste form) is one
of the most important factors in repository system performance. The waste package design for
the VA used a so-called Corrosion Allowance Material (CAM), as the outer package wall, with a
primary purposes of providing structural strength and radiation shielding. The CAM was A 516
steel. The VA waste package design also included a high-nickel alloy, Alloy 22, as the inner
package wall material. This element of the waste package design served as the Corrosion
Resistant Material (CRM) and was intended to be the principal basis for VA-design waste
package lifetime.
The waste package design for the VA posed two major licensing issues. One was potential for
rapid crevice corrosion of the Alloy 22 CRM, as a result of its being under the A 516 steel CAM-
Rapid penetration of the CRM would negate its effectiveness as the means for achieving a long
package lifetime. The other concern was the lack of a data base for corrosion of the Alloy 22
CRM under repository conditions. As noted in NRC99, the corrosion rate values used in the
TSPA-V A, especially those for the CRM, were based primarily on results of expert elicitations
rather than experiments under representative service conditions.
A key current licensing issue concerning waste package performance is therefore the sufficiency,
and applicability to the repository environment, of data for waste package corrosion performance.
NRC staff are concerned that, under current schedules, there is not enough time to gather quality
data for the LA that are sufficient in quantity and duration of test conditions. NRC99 notes that it
is appropriate for DOE and NRC to take into consideration more long-term data at later times
(e.g., at the time disposal is approved), but sufficient data must be available to support the LA.
In sum, corrosion parameter values used in the TSPA-V A that were based on results of expert
elicitations must be adequately supplanted, for the LA, by data derived from measurements under
environmental conditions expected for a repository at Yucca Mountain.
9-20
-------
Quantity and Chemistry of Water Contacting Waste Packages and Waste Forms
NRC99 states that DOE and NRC analyses both indicate that the fraction of waste packages
contacted by water is the most important factor affecting dose for the ground water pathway. It
also notes that the quantity and chemistry of water contacting the waste packages are the major
factors determining waste package lifetime, and that these water characteristics also affect
radionuclide release from the waste forms and waste packages.
NRC99 indicates that DOE recognizes the need for additional data concerning water quantities
and chemistry in the repository, and also recognizes that performance models used in the TSPA-
VA do not adequately capture the effects of coupled processes (thermal; chemical, hydrologic,
and mechanical) on the quantity and chemistry of water that contacts the waste packages.
NRC99 also states, however, that the range of licensing plan activities outlined by DOE in the
VA are unlikely to provide, for the LA, an adequate basis for assessing the quantity and
chemistry of water that contacts the waste packages and waste forms.
Two additions to DOE's suite of pre-LA activities, which could be completed before submission
of the LA, are suggested in NRC99, based on a Drift Seepage Peer Review held in January 1999
(HUG99). One action would be to conduct systematic measurements of air permeability in
horizontal boreholes in the repository host rock units. These measurements would provide data
on the scales of variability in rock properties that are needed to adequately describe seepage. The
other action would be to expand model development efforts to focus on explanation of patterns
of seepage observed in the niche experiments.
The amount of additional data and analyses needed for an adequately supported LA will depend
on how important the quantity and chemistry of water contacting the waste package is to the
DOE's safety case.
Saturated Zone Flow and Transport
The saturated zone (SZ), which is the principal pathway for radionuclide transport from the
repository to the dose receptor location, has been shown by DOE and NRC sensitivity studies to
be an important factor in repository system performance. Potentially important SZ performance
factors include low flow rates, dilution along the SZ pathway, and radionuclide holdup in
alluvium between the repository and the dose receptor location. Dilution effects expected during
9-21
-------
pumping of contaminated water by the dose receptor may also help reduce the potential dose, but
DOE did not take credit for this performance factor in the TSPA-VA.
A major issue for assessing the SZ contribution to performance in the TSPA-LA is the fact that
the SZ is at present poorly characterized, and current DOE plans to improve characterization
before submittal of the LA are limited. Flow and transport characterization data between about 5
and 20 km from the repository are at present highly limited; Gelhar (Gel98) described the
situation as a "data hole". Nye County is implementing an "Early Warning Drilling Program"
(NYE99) in which shallow and deep wells will be drilled and tested down gradient of the
repository; these wells, which are basically arrayed along Highway 95 at a radius of about 20 km
south of the proposed repository location, will provide data concerning SZ flow and transport
properties, but the data field will be limited under current DOE and Nye County plans. Well log
data to date (NYE99a) indicate that the geologic features of the alluvial deposits are highly
heterogeneous.
NRC99 expresses concern about availability of sufficient SZ characterization data for the LA. It
notes that DOE's planned activities are of low priority and will extend beyond current cutoff
dates for refinement and update of SZ flow models, and that DOE might supplement the Nye
County program with additional field work that could produce meaningful data for the LA, but
has no present plans to do so. NRC99 also notes that DOE's licensing plan in the VA
documentation characterized the SZ flow and transport uncertainties as "moderate", but that this
designation appears to be inconsistent with the results of DOE and NRC/CNWRA sensitivity
studies which show an important role for the SZ in overall repository system performance.
Defensible demonstration in the LA of the SZ contributions to the natural system component of a
defense-in-depth repository design will be essential.
Volcanic Disruption of the Waste Package
NRC99 asserts that the TSPA-VA analyses may underestimate the contribution to risk associated
with future igneous activity. It states that NRC calculations show a small but finite risk from
volcanic disruption of the proposed repository; DOE concluded in the TSPA-VA that there are
no risks from volcanism during a 10,000-year post-closure period. The NRC staff review of the
TSPA-VA concluded that the DOE analyses are based on assumptions of physical conditions not
representative of basaltic volcanism; that the data base was insufficient for evaluation of waste
package performance under appropriate physical conditions; and that the modeling assumptions
were incongruent with those used elsewhere in the TSPA-VA.
9-22
-------
NRC99 asserts that models used by DOE in their volcanism evaluations are nonconservative, and
that key assumptions are not supported by available data. It also notes that, although the VA
licensing plan does not indicate planned activities to resolve these issues, supplemental work
plans are being developed which could resolve them if implemented. NRC staff positions will be
well documented in future versions of the Igneous Activity IRSR.
Quality Assurance
NRC99 notes that DOE has consistently had problems in implementing its Quality Assurance
(QA) program, which was reviewed and accepted by NRC. Audits have determined that some
data in the technical data base are not traceable to their origins and could not be ensured to be
applicable, correct, and technically adequate. The Technical Basis Document supporting the VA
(DOE98a) indicated that a major portion of the data supporting the VA is not qualified in accord
with QA program requirements.
NRC99 states that DOE must be able to demonstrate in its LA that the data, analyses, and designs
of barriers and systems important to safety or waste isolation meet QA requirements of Appendix
BtoCFRPartSO.
DOE has recognized the need to meet QA requirements and has committed resources to
development of an overall data qualification strategy and to resolution of QA issues. Plans for
resolving QA issues address identification of unqualified data sets approved for qualification;
methods for qualification and their rationale; technical disciplines required to achieve
qualification; data evaluation criteria to be used; and criteria for changing data status from
"unqualified" to "qualified". The NRC has formed a QA Task Force which will conduct an
independent and objective review of the DOE program and its implementation.
9.3.4.3 New Repository Design Concepts
In the DOE TSPA-VA a number of alternative engineered barrier design elements were evaluated
relative to their effects on the performance of the reference repository design in the VA. These
alternatives included such design options as drip shields and backfilling the emplacement drifts.
Reviews of the VA assessments pointed out a number of uncertainties in DOE's VA performance
assessments concerning the performance of engineered and natural barriers as well as site
characteristics. These uncertainties might be reduced hi two ways: by more extensive site
characterization work to more defensibly quantify the range of expected natural barrier
9-23
-------
performance conditions and by introducing additional engineering features that compensate for
the uncertainties in site conditions and in performance of the engineered repository system
components.
Subsequent to issuance of the VA report in December 1998, DOE assessed alternative
engineered repository designs intended to reduce issues associated with uncertainties in
performance of the VA reference design. Ads described in Section 7.2.2.5 of this BID, DOE
characterized and evaluated six design options and selected, as basis for the Site
Recommendation, the so-called EDA II design, whose key design features are compared with
those of the VA reference design in Table 7-9 of this BID.
The EDA II design features are in part responsive to repository performance and regulatory
compliance issues identified by NRC staff, the NWTRB, and the Peer Review Panel as a result of
their review of the TSPA-VA. NRC staff also, however, identified and characterized issues that
are important to DOE's effort for timely development of a high-quality LA, and for pre-licensing
resolution of KTIs (NRC99) so that the basis for LA-review findings concerning reasonable
assurance of compliance with regulatory standards is as clear and well-established as possible.
These issues were identified and discussed in Section 9.3.3.
Key performance issues and uncertainties associated with the reference VA design that are
intended to be mitigated by the EDA II design include:
• The reduced area! mass loading and increased drift spacing for the EDA II design
are intended to reduce uncertainties associated with the effect of heat emissions on
the movement and chemical characteristics of water in the geologic formations
adjacent to the repository, and issues concerning coupling of thermal, hydrologic,
chemical, and mechanical phenomena such as the areal extent of the "dry out"
zone and its changes over time.
• Elimination of the concrete lining and invert is intended to eliminate uncertainties
about the effect of concrete materials on the chemical characteristics of water that
can contact and corrode the waste packages and dissolve the waste form.
• Revision of the waste package design to use Alloy 22 as the outside wall material
is intended to eliminate the potential for crevice corrosion of this material when it
is under carbon steel, as was the case for the waste package design for the VA.
This change would significantly increase waste package lifetimes by significantly
reducing or eliminating the potential for rapid corrosion of the Corrosion
Resistant Material. The EDA II waste package design concept uses, in addition to
. Alloy 22 as the outer wall material, 316L stainless steel as the inner wall material-
9-24
-------
As described by DOE at the summer meeting of the NWTRB (DOE99), this
design concept, in conjunction with the planned heat load and expected repository
temperatures, is expected to keep corrosion conditions for the Alloy 22 outside the
"window of vulnerability" to rapid crevice corrosion, which spans the range of 80-
100 °C and 50-100% relative humidity. If sufficiently supported by data and
analyses, this strategy could eliminate crevice corrosion as a potential cause of
early waste package failure.
Use of drip shields and backfill is intended to add defense-in-depth to the
engineered design and to defer the time at which seepage water could contact the
waste packages and initiate aqueous corrosion. Drip shields would significantly
delay the start of waste package corrosion, and would also dramatically reduce the
uncertainty in projecting the effects of premature failures due to manufacturing
defects. The occurrence of premature failures is a difficult performance factor to
characterize because there is no directly-transferrable empirical data base from
industrial experience that matches expected repository conditions. DOEVs base
case VA assessments show that the projected dose rates for 10,000 years are
strongly influenced by radionuclides releases from the single package assumed to
be prematurely failed at 1,000 years. If more than one package is assumed to fail,
dose estimates would increase in proportion. Use of drip shields is an example of
an additional engineered measure to compensate for uncertainties that can not be
reduced in any other way. By shielding the waste packages with the drip shield,
ground water infiltrating into the drift has to first cause corrosion of the shield
rather than directly contacting a prematurely-failed waste package.
Blending of spent fuel subassemblies to reduce variations in waste package heat
emissions is intended to reduce temperature levels and gradients that could
stimulate corrosion and other degradation processes, and to reduce uncertainties in
modeling of EBS performance. By eliminating "hot spots" in the repository, the
averaging approach needed to model at the repository scale becomes more
defensible.
9-25
-------
REFERENCES
AND89 Andersson, J., T. Carlsson, T. Eng, F. Kautsky, E. Soderman, and S. Wingefors,
The Joint SKI/SKB Scenario Development Project, TR89-35, Stockholm, Sweden,
1989.
BAR92 Barnard, R.W., M.L. Wilson, H.A. Dockery, J.H. Gauthier, P.O. Kaplan, R.R.
Eaton,,R.W. Bingham, and T.H. Robey, TSPA 1991: An Initial Total-System
Performance Assessment for Yucca Mountain, SAND91-2795, Sandia National
Laboratories, Albuquerque, NM, 1992.
BER89 Bertram-Howery, S.G. and R.L. Hunter, Editors, Preliminary Plan for Disposal-
System Characterization and Long-Term Performance Evaluation of the Waste
Isolation Pilot Plant, SAND89-0178, Sandia National Laboratories, Albuquerque,
NM, 1989.
BUR80 Burkholder, H.C., Waste Isolation Performance Assessment: A Status Report,
Scientific Basis for Nuclear Waste Management, Volume 2, Plenum Press, pp.
689-702,1980.
DAV79 Davis, W., "Carbon-14 Production in Nuclear Reactors", page 151 in
Management of Low-Level Radioactive Waste, Carter, M.W., Moghissi, A.A., and
Kahn, B., editors, Vol. I, Pergamon Press, New York, NY 1979.
DOE74 U.S. Department of Energy, Potential Containment Failure Mechanisms and
Their Consequences at a Radioactive Waste Repository in Bedded Salt in New
Mexico, ORNL-TM-4639, 1974.
DOE79 U.S. Department of Energy, Scenarios for Long-Term Release of Radionuclide
From a Nuclear-Waste Repository in the Los Medanos Region of New Mexico,
SAND78-1730, 1979.
DOE90a U.S. Department of Energy, Risk Methodology for Geologic Disposal of
Radioactive Waste: Scenario Selection Procedure,NUREG/CR.-\667, SAND80-
1429,1990.
9-26
-------
DOE91 U.S. Department of Energy, Preliminary Comparison with 40 CFR Part 191,
Subpart Bfor the Waste Isolation Pilot Plant, December 1991, Volume 1:
Methodology and Results, prepared by Sandia National Laboratories,
Albuquerque, NM, under Contract DE-AC04-76DP00789, 1991.
DOE94 U.S. Department of Energy, Total-System Performance Assessment for Yucca
Mountain - SNL Second Iteration (TSPA-1993), SAND93-2675, April 1994.
DOE98 U.S. Department of Energy, Viability Assessment of A Repository at Yucca
Mountain, DOE/RW-0498, December 1998.
DOE98a CRWMS Contractor, Total System Performance Assessment - Viability
Assessment (TSPA-VA) Analyses Technical Basis Document, BOOOOOOOO- 01717-
4301-00001 Rev 01, November 13,1998.
DOE99 U.S. Department of Energy, Presentations at the Summer Meeting of the
Nuclear Waste Technical Review Board, Beatty, Nevada, June 29 and 30 1999.
EPA85 Environmental Protection Agency, Final Rule, Environmental Standards for the
Management and Disposal of Spent Nuclear Fuel, High-level and Transuranic
Radioactive Wastes, 40 CFR Part 191, Federal Register, 50 FR 38066-38089,
September 19,1985.
EPA96 Environmental Protection Agency, Summary of EPA Office of Radiation
Programs Carbon-14 Dosimetry as used in the Analysis for High-level and
Transuranic Wastes (Draft), prepared for the HLW/Carbon-14 Release
Subcommittee of the EPA Science Advisory Board's Radiation Advisory
Committee, 1996.
GEL98 Gelhar, Lynn W., Report on U.S. Technical Review Board Winter Meeting,
January 20-21,1998, Amargosa Valley, Nevada.
HUG99 Hughson, D., Drift Seepage Peer Review, Trip Report, Las Vegas, Nevada,
January 11-13, 1999, Center for Nuclear Waste Regulatory Analyses, San
Antonio, Texas.
9-27
-------
IAE83 International Atomic Energy Agency, Concepts and Examples of Safety Analyses
for Radioactive Waste Repositories in Continental Geological Formations, Safety
Series No. 50, Paris, 1983.
ICR82 International Commission on Radiological Protection, Limits for Intakes of
Radionuclides by Workers, Part 3, ICRP Publication 30, Pergamon Press, New
York, NY, 1982.
ICR85a International Commission on Radiological Protection, Principles of Monitoring
for the Radiation Protection of the Population, Publication 43, Annals of the
ICRP, 15(1), Pergamon Press, 1985.
ICR85b International Commission on Radiological Protection, Radiation Protection
Principles for the Disposal of Solid Radioactive Waste, ICRP Publication 46,
Pergamon Press, 1985.
KIL78. Killough, G.G. and P.S. Rohwer, A New Look at the Dosimetry of14C Released to
the Atmosphere as Carbon Dioxide, Health Physics 34:141,1978.
NAS72 National Academy of Sciences - National Research Council, The Effects on
Populations of Exposure to Low Levels of Ionizing Radiations (BEIR I), Report of
the Biological Effects of Ionizing Radiation, Washington, D.C., 1972.
NAS80 National Academy of Sciences - National Research Council, The Effects of
Populations of Exposure to Low Levels of Ionizing Radiation (BEIR III), Report
of the Biological Effects of Ionizing Radiation, Washington, D.C., 1980.
NAS95 National Academy of Science - National Research Council, Committee on
Technical Bases for Yucca Mountain Standards, Technical Bases for Yucca
Mountain Standards, National Academy Press, Washington, DC, 1995.
NRC81 U.S. Nuclear Regulatory Commission, Disposal ofHigh-Level Radioactive
Wastes in Geologic Repositories, 10 CFR Part 60, Federal Register, 46 FR 13980,
1981.
9-28
-------
NCR93 National Council on Radiation Protection and Measurements, Carbon-14 in the
Environment, NCRP Report No. 81,1993.
NRC99 U.S. Nuclear Regulatory Commisssion, Letter to DOE-Viability Assessment, June
2,1999.
NRC99a U.S. Nuclear Regulatory Commission, NRC/DOE Technical Exchange on Total
System Performance Assessment (TSPA) for Yucca Mountain, May 25-27,1999.
NWP83 Nuclear Waste Policy Act of 1982, Public Law 97-425, January 7,1983.
NYE99 Nye County, Nevada, Presentation by T. Buqo to the Nuclear Waste Technical
Review Board Summer Meeting, Beatty, Nevada, June 29 and 30, 1999.
NYE99a Nye County, Nevada, Well Logs for the Early Warning Drilling Program,
available on the Nye County website, www.nyecounty.com.
PRP99 TSPA-VA Peer Review Panel, Peer Review of the Total System Performance
Assesment- Viability Assessment, Final Report, February 1999.
PRU93 Pruess, K., and Y. Tsang, "Modeling of Strongly Heat-Driven Flow Processes at
Potential High-Level Nuclear Waste Repository at Yucca Mountain, Nevada," pp.
568-575 in High Level Radioactive Waste Management, Proceedings of the
Fourth Annual International Conference, Las Vegas, Nevada, April 26-30, 1993,
American Nuclear Society, La Grange Park, IL, 1993.
ROS93 Ross, B., Y. Zhang, and N. Lu, "Implications of Stability Analysis for Heat
Transfer at Yucca Mountain," in High Level Radioactive Waste Management,
Proceedings of the Fourth Annual International Conference, Las Vegas, Nevada,
April 26-30, 1993, American Nuclear Society, La Grange Park, IL, 1993.
TRB99 U.S. Nuclear Waste Technical Review Board, Moving Beyond the Yucca
Mountain Viability Assessment - A Report to the U.S. Congress and the
Secretary of Energy, April, 1999.
9-29
-------
CHAPTER 10
RADIOLOGICAL RISKS FOR DEEP GEOLOGICAL DISPOSAL
AND SURFACE STORAGE OF SPENT NUCLEAR FUEL
10.1 BACKGROUND INFORMATION
In September 1996, the United States Senate (Senate Report 104-320, p.98, September 12, 1996)
requested an evaluation by EPA of alternatives to the disposal of radioactive materials in a deep
geologic repository at Yucca Mountain, as well as an evaluation of public health risks of these
alternatives against standards proposed for deep geologic disposal. Alternatives to be considered
included: 1) storage of nuclear wastes at each site where it is currently stored and 2) one or more
centralized above-ground storage sites.
Spent nuclear fuel (SNF) from the operation of commercial nuclear power reactors is currently
stored at more than 70 nuclear generating sites around the country. It is expected that existing
nuclear power plants will produce approximately 87,000 metric tonnes of spent fuel during their
operational lifetimes. Approximately 28,000 metric tonnes of spent nuclear fuel were stored at
commercial nuclear power reactors as of 1993. By the year 2003, this amount is expected to
increase to 48,000 metric tonnes (NPJ97).
To date, most SNF is stored in water-filled pools at the reactor sites where it was generated.
However, space is not available in existing pools to store all of the spent fuel expected to
accumulate over the lifetime of the reactors. When the pool capacities were established, it was
expected that the SNF would be removed from the reactor site for reprocessing about five years
after discharge from the reactor. After national plans for reprocessing were terminated, removal
of SNF from the reactor sites for central interim storage or disposal was expected to begin in
January 1998, but these programs have been delayed. Consequently, additional SNF storage
capacity is therefore needed.
Facilities for interim storage of SNF prior to disposal have been under consideration since 1972.
In February 1972, the Atomic Energy Commission (AEC) began consideration of surface storage
facilities at the Hanford site in the State of Washington. The facility would be used "...for high-
level commercial wastes and low-level wastes from both commercial and AEC activities
(HEW87)." In June 1972, the AEC revealed plans to develop a Retrievable Surface Storage
10-1
-------
Facility (RSSF), which would be an array of mausolea or vaults where waste or spent fuel
canisters would be stored (CAR87).
The decision to choose the surface storage option "...was a response to the dilemma of
irretrievability" and seemed a "practical answer to a difficult political and technical problem
(HEW87)". The AEC concluded that such storage would be satisfactory for decades or centuries
(USC72, USC75). The Congress' Joint Committee on Atomic Energy reported in 1972 that the
radioactive waste management program "...now includes the conceptual design of manmade
surface facilities.of an expected lifetime of several centuries (USC72)."
Three technical concepts for the RSSF were considered by the AEC: 1) stainless steel canisters in
water basins for heat removal and shielding; 2) canisters in concrete basins, cooled by circulating
air; and 3) a canister within a two-inch thick container with doubly-contained waste in a three-
foot thick concrete cask cooled by circulating air.
Recently, members of Congress, other public officials, environmental groups, and private
citizens have expressed concern that a surface storage facility might be regarded as a de facto
repository, thereby reducing the impetus for building a geologic repository as expeditiously as
possible. To allay these concerns, proposed Monitored Retrievable Storage (MRS) facilities
have consistently limited the total storage capacity to well below total SNF quantities projected
for permanent deep geologic disposal. MRS designs have been proposed from a few thousand up
to 15,000 metric tonnes uranium (MTU). This chapter compares the potential impacts of
continued storage to those associated with disposal at a geologic repository at Yucca Mountain.
The comparison is presented in terms of applicable regulatory limits (Section 10.2) and
anticipated (estimated) risks (Sections 10.3 and 10.4).
10.2 REGULATORY LIMITS
The Energy Policy Act of 1992 (EnPA) directs the Administrator to establish, after consultation
with the National Academy of Sciences (NAS), a maximum individual dose standard for the
proposed repository at Yucca Mountain. The NAS found that such an approach would provide
protection for all exposed individuals and suggested that levels of a few millirems/year to a few
tens of millirems per year (mrem/yr) would provide a reasonable point of departure for
rulemaking. Therefore, for the purposes of comparison, it is assumed that the Yucca Mountain
standard will establish a maximum individual dose standard. Further, absent the Administrator's
final decision on the level of the standard, this comparison uses the upper end of the
NAS'suggested range (i.e., a few tens of millirems per year) to characterize the allowable
10-2
-------
exposure limits for the proposed repository. This limit, which encompasses all exposure
pathways, will apply to releases from the repository over an extended period of time. The NAS
has recommended that it apply during the period of geologic stability for the site, a time frame
that could extend to one million years.
The regulatory limits that would apply to continued surface storage depend upon the specific site
at which the wastes are stored. For continued storage at the sites at which the spent fuel and high
level wastes were generated or are currently stored, different limits apply to the following types
°f facilities:
• Power reactors
• Research reactors
Independent spent fuel storage installation (ISFSI)
DOE facilities
If the government opts to store wastes at a centralized facility such as an an ISFSI or an MRS
facility, the applicable limits would likely be those for an ISFSI. However, if the ISFSI or MRS
were constructed on a site owned by DOE prior to 1983, the limits applicable to DOE facilities
would likely apply.
-2.1 Power Reactors
For waste stored at power reactors, the applicable regulatory limits would be those established by
EPA for the nuclear fuel cycle (40 CFR Part 190), and the NRC (10 CFR Parts 20, 50, and 100).
40 CFR Part 190 establishes exposure limits from normal facility operations of 25 mrem/yr to the
whole body or any organ (except the thyroid, which is allowed 75 mrem/yr). These limits,
established under the old "whole body/critical organ" protection concept, are roughly equivalent
to a 10 mrem/yr limit under the current "effective dose equivalent (EDE)" protection concept.
The 40 CFR Part 190 limits consider all exposure pathways and require the site to consider
Potential exposures from all facilities that are part of the nuclear fuel cycle. Under 10 CFR Part
20, power reactors are required to maintain exposures as low as is reasonably achievable
(ALARA), consistent with maximum individual exposures of 100 mrem/yr "total effective dose
equivalent" (TEDE) at an exposure rate not to exceed 2 mrem/hr. The limits for normal
operations at power reactors can be characterized as ALARA, with maximum individual
exposures limited to less than two millirem per hour (mrem/hr) TEDE and not to exceed 75
ttirem/yr to the thyroid or 25 mrem/yr to the whole body or any other organ. The EPA's recent
evaluation of the airborne emissions from power reactors during its reconsideration of the
10-3
-------
National Emission Standards for Hazardous Air Pollutants (NESHAPS) (40 CFR 61 Subpart I)
found that the ALARA design objectives and limiting conditions of operations set forth in
Appendix I to 10 CFR Part 50 are being met by power reactor licensees and that actual exposures
from the air pathway are a fraction of the regulatory limits.
In addition to the operating limits on power reactor licensees, 10 CFR Part 100 establishes siting
criteria that assure that no member of the public will receive an exposure greater than 25 rem to
the whole body or 300 rem to the thyroid over a two-hour period of exposure to a fission product
release associated with a hypothetical major accident at the facility.
10.2.2 Research Reactors
The regulatory limits for research (non-power) reactors are established by 10 CFR Part 20.
Exposures via all pathways to any member of the public from normal operations are also required
to be as low as reasonably achievable, consistent with a maximum individual exposure limit of
100 mrem/yr TEDE at an exposure rate not to exceed 2 mrem/hr. Additionally, under the
recently adopted "constraint rule," corrective actions must be initiated should exposures via the
air pathway exceed 10 mrem/yr TEDE. Given this and a 50/50 split of the 100 mrem/yr limit
between the air and liquid pathways outlined in 10 CFR Part 20, the effective maximum
individual dose limit for research reactors from normal operations is 60 mrem/yr TEDE. No
quantitative limits are imposed on research reactors for demonstrating ALARA for non-airborne
exposure pathways, nor are quantitative criteria given for exposures from accidental releases.
However, site suitability is considered during licensing.
10.2.3 Independent Spent Fuel Storage Installations TISFSIsI
The regulatory requirements for ISFSIs are those established by the NRC in 10 CFR Parts 20 and
72. The limits established in Part 20 require exposure to be maintained as low as reasonably
achievable, consistent with a maximum individual exposure limit of 100 mrem/yr TEDE at an
exposure rate not to exceed two mrem/hr. Additionally, Part 72 imposes the nuclear fuel cycle
exposure limits of 40 CFR 190. Therefore, like power reactors, the limits for normal operations
of an ISFSI can be characterized as ALARA with maximum individual exposures limited to less
than two mrem/hr TEDE and not to exceed 75 mrem/yr to the thyroid or 25 mrem/yr to the whole
body or any other organ.
10-4
-------
10 CFR Part 72 also establishes an exposure limit of five rem to the whole body or any organ in
the case of an accident. This limit is applied at the boundary of the controlled area and cannot
be exceeded by any design basis accident.
It should be noted that the limits for an ISFSI currently apply to dry storage of spent fuel at
operating nuclear power plants, but not the original wet fuel pools. However, as reactors reach
the end of their operating lives and are decommissioned, it is likely that long-term storage of all
spent fuel would be conducted under a license issued pursuant to 10 CFR Part 72.
10.2.4 DOE Facilities
Relevant exposure limits for DOE facilities are those established by EPA for airborne releases of
radionuclides (40 CFR 61, Subpart H) and by DOE Order 5280. 40 CFR 61, Subpart H, limits
airborne releases of radioactive materials (excluding radon and its decay products) to quantities
that do not cause any member of the public to receive an exposure greater than 10 mrem/yr EDE.
The limits established by DOE Order 5280 mirror 10 CFR Part 20; i.e.,, exposures are to be as
low as reasonably achievable, consistent with a maximum individual exposure limit of 100
mrem/yr TEDE from all pathways at an exposure rate not to exceed two mrem/hr. Given the
constraint rule for the airborne pathway and a 50/50 split of the 100 mrem/yr limit between the
air and liquid pathways, the limits for DOE facilities can be characterized as ALARA with
maximum individual exposures not to exceed 60 mrem/yr TEDE at a dose rate of less than two
mrem/hr (10 CFR Part 20).
No quantitative exposure criteria for accidental releases from DOE facilities are established by
DOE Order 5280 or 40 CFR Part 61.
10.2.5 Summary of P^i^tnrv Limits
As the above subsections have detailed, different regulatory limits would apply at existing or
future storage sites, depending upon the specific use of the site. However, given the ALARA
requirement imposed on all sites and the various limits on maximum annual exposures, it is not
Unreasonable to expect that exposures from undisturbed storage would be on the order of a few
tens of millirems per year at any of these facilities. This level of exposure is consistent with the
limits that will likely be promulgated for Yucca Mountain.
10-5
-------
Such comparability cannot be assumed in the event of accidents. The Yucca Mountain standards
explicitly require that natural phenomena be evaluated and factored into the design of the facility-
Thus, a maximum individual exposure limit of a few tens of millirems per year would apply to
exposures caused by both accidents and natural phenomena. By contrast, no explicit criteria exist
for accidents at research reactors or DOE facilities. Consequently, the limits for storage at power
reactors and ISFSIs could potentially allow individual exposures of up to five rem EDE in the
case of an accident.
10.3 REPORT BY THE MONITORED RETRIEVABLE STORAGE REVIEW
COMMISSION
Information presented in this and the following section addresses the risks associated with
storage of SNF at reactor sites and at a central interim storage facility, characterized as a
Monitored Retrievable Storage Facility. These risks are compared to those associated with
disposal at the proposed Yucca Mountain repository site in Nevada. A review of literature on the
risks of SNF storage and disposal revealed a large body of information. However, no studies to
date have specifically addressed the scope of the directives of the Senate Report. Past studies
have focused on dry storage at reactors and on an MRS facility as part of a dynamic total waste
management system configuration. Despite its limitations, this information was used as the basis
for the data presented in this section of the BID.
In 1987, the Nuclear Waste Policy Amendments Act established the Monitored Retrievable
Storage Review Commission. The Commission's charter was to compare storage of spent fuel at
a Federal MRS facility to storage of spent fuel at the reactors at which it was generated. Through
public hearings, the Commission solicited the views of private citizens, technical experts, and
utility and government representatives. In addition, several contractors performed specific tasks
to augment the Commission's technical work. The Commission's final report, issued in
November 1989, examined each alternative's merits, including an assessment of radiological
doses and risks to members of the general public. The report's principal findings are
summarized below (MRS 89).
10.3.1 At-Reactor Storage Options
Water-filled pools have been used for SNF storage since the earliest days of nuclear reactor
operation and are universally used for storage of commercial Light Water Reactor (LWR) fuels
today. Spent fuel pools employ a large amount of water for heat removal and radiation shielding.
10-6
-------
Pools remain a proven method for cooling LWR fuels for periods lasting from five to ten years.
Thereafter, the fuel can be removed from the pool to make room for new inventories of hot fuel.
As stated above, existing pool capacities were not designed to accommodate SNF quantities
generated from the 40-year operating life of a reactor. Expansion of storage capacity at the
reactor sites includes options that are broadly categorized as wet storage and dry storage. (Rod
consolidation for expanding on-site storage capacity is currently not considered a viable option
was therefore not considered.)
Two wet storage options commonly considered are spent fuel reracking and new pool
construction:
Spent Fuel Reracking. This option entails a reconfiguration for high density
storage. Typically, this is done by manufacturing fuel racks that bring fuel
elements closer together hi order to create additional storage space. Most utilities
have reracked their pools at least once. From the original typical design of 1-1/3
reactor core storage capacity, utilities have frequently increased storage to four to
six reactor cores. However, even with these measures, pool-storage capabilities at
most reactors will not be considered adequate.
New Pool Construction. This option entails the construction of a pool for long-
term storage of SNF. Due to high costs and operational/maintenance factors, new
pool construction is not considered competitive to dry storage.
Since the early 1980s, demonstration projects at several utilities and research by the Electric
Power Research Institute have demonstrated the viability of dry storage methods for SNF. Many
different dry storage methods have been proposed and/or tested including metal or concrete
storage casks, air-cooled vaults, and universal multipurpose casks. (Note: In 1990, the NRC
amended its regulations to authorize licensees to store spent fuel at reactor sites in storage casks
approved by the NRC. To date, seven cask designs have received certificates of compliance
(NPJ97).) A generic description of dry storage methods includes:
Modular COT**** Vaults (UCV\ MCVs consist of sealed metal tubes inside an
above surface concrete structure. Inside the sealed metal tubes, the spent fuel is
kept under an inert cover gas or ah*. The tubes are typically made of carbon steel
and each tube contains a single fuel assembly or a single element. The MCV has
received site-specific NRC licensing.
Horizontal C^rete Vaults nvtodulesX Horizontal concrete modules keep the
fuel inside a sealed stainless steel canister back-filled with an inert gas. The
10-7
-------
canister is protected and shielded by an above-surface concrete module. The heat
generated by the spent fuel is removed by thermal radiation, conduction, and
natural convection through air channels in the concrete module. The canister
contains a basket for holding the fuel in place. The horizontal concrete vault has
received site-specific NRC licensing.
• Metal Drv Casks CMDCV Metal dry casks are the most mature of the methods
available for dry interim storage. Their use was successfully tested and
demonstrated in 1984. The casks are large heavy vessels (100 to 125 tonnes
loaded). They are equipped with an internal basket for holding the spent fuel
assemblies or elements. The body is made from forged steel, modular cast iron, or
lead and stainless steel with a double seal lid. The MDC has received site-specific
NRC licensing.
• Concrete Dry Storage Casks. Concrete dry storage casks are similar to metal dry
casks except that the body of the cask is made of steel-reinforced concrete with an
inner metal liner for containment. Concrete dry storage casks have received site-
specific NRC licensing.
• Dual Purpose Casks (DPC). The DPC was derived from the metal dry cask
concept. The design and manufacture are very similar to that of the metal dry
cask, but it would also be used to transport the fuel to a Federal spent fuel
management facility. The fuel would be removed upon arrival at the spent fuel
management facility, but would not be disposed of in a DPC. The DPC remains
in development.
• Multi-Purpose Casks fMPO. This dry cask combines storage, transportation and
disposal hi one container. The MPC would potentially allow spent fuel to be
stored, transported and disposed of in the same container in which it was
originally placed. The use of the MPC would not require the fuel to be extracted
from it prior to being placed in a repository,
10.3.2 Radiation Exposure Modeling Assumptions for At-Reactor Storage of SNF
To model public radiation exposures associated with at-reactor storage of SNF, the MRS
Commission Report assumed that there were to be no new orders for nuclear power plants
beyond those operating or being constructed as of December 1987. For post-1988 SNF, burnup
rates of 36,600 megawatt-days per metric ton uranium (MWd/MTU) and 42,000 MWd/MTU
were assumed for BWRs and PWRs, respectively.
It was also assumed that all fuel would be stored at the reactor sites until DOE was ready to
accept the waste and ship it to the repository for disposal. The analysis assumed utilities would
10-8
-------
select from a number of currently available options to provide at-reactor storage that includes
fuel-pool reracking supplemented by dry-storage. Since most utilities have already reracked their
pools at least once, the Commission concluded that future reracking will be limited. Adding a
second tier of racks was not considered a practical way to expand the current pool storage
capacity. It was further assumed that there would be no transhipment of spent fuel from one
reactor site to another to alleviate storage problems. Every utility would maintain enough pool
storage capacity so that the full core of the reactor could be unloaded into the spent fuel pool, if
necessary.
For the balance of life-of-plant SNF, dry storage, either at the reactor sites or on utility-owned
land contiguous to the reactor, was assumed to involve metal or concrete casks. The typical
metal dry storage cask is made of stainless steel or nodular cast iron that may hold from a few to
25 PWR fuel assemblies per module. For concrete casks with an inner metal liner, the
unventilated type will hold nine PWR or 25 BWR fuel elements, while a ventilated type may
hold 17 PWR or 50 BWR elements.
Although a few reactors have already been permanently shut down, the majority of currently
licensed and operating facilities will reach their end-of-operating life between the years 2009 and
2030. Table 10-1 shows the amounts of five-year-old SNF that are expected to be stored in fuel
pools and in dry storage.
Table 10-1. Spent Fuel Accumulation at Shutdown Commercial Light Water Power Reactors
(Source: MRS89)
\ ..
:-- ' Year - *
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
i* -i^$la»i»:^"*
0
0
0
500
7,500
12,000
19,500
27,000
29,000
30000
\fjt \f y*. ^!*y pn.'WBfc V.-M** * w
**l's|^i^(fc-^"*^ :
500
500
1,500
4,000
21,000
29,000
40,000
54,000
55,500
57,000
" \\ -Tcftai^5 '-;
500
500
1,500
4,500
28,500
41,000
59,500
81,000
84,500
87.000
10-9
-------
10.3.3 Model Assumptions for MRS Storage of SNF
For the MRS alternative analyzed by the Commission, spent fuel is assumed to be stored at the
reactors until an MRS facility becomes available. At this time, spent fuel from some reactors
would be transported to and stored at an MRS until a repository is available for permanent
disposal. The MRS would continue to operate until all the spent fuel has been emplaced in the
repository.
The MRS facility as defined in the Nuclear Waste Policy Amendments Act of 1987 was
analyzed. Given the MRS capacity limitation of 15,000 MTU, most spent fuel would be stored at
the reactor site. The MRS would, therefore, only supplement at-reactor storage and reduce the
need for dry storage at reactor facilities, as defined in Table 10-2. Dry modular storage, using
technologies similar to those described in Section 10.3.1, was assumed for the MRS facility.
Table 10-2. Reduction in Dry Storage Needs At Reactor Facilities
with Linked MRS (Source: MRS89)
NO-MRS
Linked MRS*
7,693
3,562
4,131
*MRS schedule linked to repository schedule.
10.3.4 Transportation Models for SNF With and Without MRS
The MRS Review Commission's dose assessment for members of the public also included
radiation exposure that would result from transportation of wastes if an MRS facility were part of
the spent fuel management system and if it were not. In the absence of an MRS facility, transit
doses could result to members of the public along the paths of travel between individual reactor
sites and a repository assumed to be at the Yucca Mountain Site. With an MRS facility, transit
exposures could occur between: 1) reactor sites and the MRS facility, 2) reactor sites and the
repository, and 3) the MRS facility and the repository. For SNF shipments originating from
reactors, 54 percent would be shipped by rail and 46 percent by truck; 100 percent of the
shipments between the MRS facility and the repository were assumed to be made by rail. These
assumptions will hold despite the location of the MRS, since all waste that could be moved by
10-10
-------
rail (rail head at reactor site) were assumed to be moved by rail. The assumed MRS location was
actually in the Eastern United States.
Table 10-3 identifies primary parameters that define transportation risk for each of the base
cases. In effect, these parameters serve as surrogate measures to approximate risk. For example,
to compare the relative radiological risk of the three base cases, the total number of shipment
miles or cask miles was added. The underlying assumption was that public exposure was
Proportional to the number of casks and distances traveled.
Table 10-3. Life-Cycle Transportation Risk Measures*
(Source: MRS89)
* Repository in 2013 (54% rail/46% truck from reactor; 100% rail from MRS facility)
" MRS facility to begin operations in 2010
10.3.5 Public Exposur Frrtm SNF Storage
Radiation doses to members of the public were assessed for spent fuel management operations at
reactor sites, the MRS facility, and the repository. The computer model used to evaluate
radiation doses for different system configurations was MARC: MRS Review Commission's
Analysis of System Risk and £ost MARC is a network model that incorporates DOE's
computer code TRICAM. TRICAM describes how spent fuel moves through the system and is
used for modeling transportation. The code also uses facility-specific data such as reactor spent
fuel discharges, population data, reactor rail accessibility, repository capacity and demographics,
and acceptance schedules.
Table 10-4 summarizes population doses predicted by MARC for members of the public.
Population dose estimates were 130 person-rem for individuals living within a 50-mile radius of
the 70 reactor sites; 4 person-rem for individuals living within a 50-mile radius of an MRS
facility located in the eastern United States; and 0.125 person-rem for individuals living within a
50-mile radius of a deep geologic repository assumed to be located at Yucca Mountain.
10-11
-------
The Commission report found that public exposures from spent fuel were small at all locations
associated with SNF management. By far, the largest source of public exposures was estimated
to result from the transportation of SNF between reactor sites and the MRS facility and the
repository and between the MRS facility and the repository. Table 10-4 shows that public
transportation exposure is reduced by more than a factor of two (i.e., from 7,900 to 3,400 person-
rem) when an MRS facility is included in the management of SNF. This is almost exclusively
due to reduced shipping miles and its attendant shift from truck to rail services when an MRS
facility is employed.
Table 10-4. Total Life-Cycle Doses in Person-Rem from Spent Nuclear
Fuel Management With and Without MRS
(Source: MRS89)
All Reactors
MRS Facility
Repository
Transportation
TOTAL
130
Not applicable
0.1
7,900
~ 8,000
13
4
0.1
3,400
3,500
Results reported by the MRS Review Commission (as summarized hi Table 10-4 above) cannot
be used directly to compare potential public risks from SNF stored at the proposed Yucca
Mountain Site (YMS) with the alternatives of At-Reactor Storage and MRS storage. The dose
estimates given in Table 10-4 correspond to different and changing SNF inventories at these
facilities over different time periods. The variations in SNF quantities and locations with time
associated with the MRS Review Commission evaluations are shown in Table 10-5.
By interpolation, average SNF quantities can be defined for each time period that, when added,
yield time-weighted quantities of stored SNF at the reactor facilities, MRS, and repository in
terms of metric tonnes of uranium-years (MTU-years):
• For At-Reactor Storage, which spans the 50-year period between 1995 and
2045, fuel pool and dry storage correspond to 1,672,067 MTU-years
10-12
-------
MRS Storage that is assumed to begin in 2010 and ends in 2045 represents
a time-weighted SNF storage value of 106,400 MTU-years
depository Stnf?ffe ^d Disposal that is assumed to begin in 2013 yields a
cumulative value of 1,302,431 MTU-years
Table 10-5. Location of Spent Fuel With MRS in 2010 and Repository in 2013 (MTUs)
(Source: MRS89)
•• .• :
f f "^Y^jttfrt
1995
2000
2005
2010
2015
2020
2025
2030
2035
2040
28,680
36,807
42,026
46,362
49,914
43,857
36,799
28,037
18,949
9,831
0
1,286
3,711
8,019
11,532
7,907
5,819
4,208
459
0
0
0
—
—
—
2,400
12,100
15,000
15,000
15,000
9,800
3,000
1,000
0
0
0
0
1,149
12,798
27,715
45,644
57,596
72,436
86.756
From the MRS Review Commission's previous estimates of public exposures, normalized public
dose estimates can be derived that provide a fair comparison for these three modes of SNF
storage (Table 10-6). Based on normalized values, public exposures that would result from SNF
stored at reactor sites and a designated MRS are nearly the same. For storage at a repository,
however, public exposure is projected to be lower by two to three orders of magnitude.
The projected similarity of public doses for at-reactor and MRS storage is to be expected if it is
assumed that: 1) release fractions of stored SNF for these two alternatives are either identical or
very similar and 2) the population density and distribution for the hypothetical MRS facility are
similar to the 0-50 mile populations that characterize each of the 70 reactor facilities expected to
store SNF onsite.
10-13
-------
Table 10-6. Comparison of Public Exposures Resulting from Three SNF Storage Alternatives
1)0005
At-Reactor Storage
(130person-rem)*
MRS
(4 person-rem)'
Repository
1,672
106
1,302
7.8E-02
3.8E-02
<9.6E-05
5.5E+00
2.7E+00
6.7E-03
"MRS Review Commission's public dose estimates (see Table 10-4).
The much lower collective population exposure estimated for repository storage is also to be
expected. For deep geologic disposal, the release fraction from the waste package to the
biosphere is likely to be reduced by at least two to three orders of magnitude, as suggested by the
reduced normalized population dose estimates in Table 10-6. It can also be assumed that cited
population dose estimates will decline to even lower levels when the repository progresses from
its operational phase to post-closure that includes backfilling of all access shafts and repository
penetrations (MRS89).
10.4 OTHER INFORMATION SOURCES
A comprehensive literature review was performed to identify other potential sources of
information concerning radiation exposures associated with SNF management and disposal.
Additional data were needed to: 1) confirm and/or compare dose estimates cited by the
Commission, and 2) supplement Commission's data that were limited to collective population
doses from routine facility operation. Lacking in the Commission Report were dose data for the
reasonably maximally exposed individual (RMEI) and doses linked to accidental releases.
Estimates of offsite doses that result from accidental releases of radioactivity to the environment
are complex and require predictive risk analyses that include: 1) a facility-specific
characterization, 2) identification of potential accident scenarios, 3) estimation of accident
probability, and 4) pathway modeling that incorporates site-specific data on weather, population
distribution, land use, hydrology, etc.
10-14
-------
It was found that, for the studies reviewed, a simple comparison of reported dose estimates is
made difficult by variations in the studies' scope and objectives, selection of accident scenarios,
and model-parameter values. To provide a common basis for comparison, secondary
information, when provided within each study, was used to convert reported data to a common
normalized value that would permit comparison. Summarized below are the most relevant
studies and their estimates for doses to members of the public.
10.4.1 "An Assessment of LWRS Spent Fuel Disposal Options" (BEC79^
This study was conducted by the Bechtel Corporation for the DOE National Waste Terminal
Storage Program and provided background documentation that dealt with three treatments of
SNF prior to disposal at a repository:
Case 1: Simple encapsulation and disposal of spent fuel at the repository following storage
at an ISFSI for nine years.
Case 2: Encapsulation of fuel, end fittings, and secondary wastes after chopping the fuel
bundle and removal of volatile materials.
Case 3: Encapsulation of fuel, end fittings, and secondary wastes after chopping, removal
of volatile materials, calcination, and vitrification.
These base case scenarios assumed a spent fuel throughput of 5,000 MTU per year at a
Processing and encapsulation facility before the SNF was shipped to a repository for final
disposal.
Risk analysis at the repository was further limited to the pressure period, which defines the
Period of emplacement of the processed/encapsulated SNF. Estimates of public exposures were
defined for normal operations and "shaft drop" accident conditions as summarized in Table 10-7.
10-15
-------
Table 10-7. Public Doses for Normal Repository Operation
and From Shaft-Drop Accident (Based on data from BEC79)
Casel
Normal Operation
Accident
Case 2
Normal Operation
Accident
Case 3
Normal Operation
Accident
l.OE-06
1.1E-02
1.5E-06
1.1E-02
2.0E-06
1.1E-02
10.4.2 "Generic Environmental Impact Statement. Management of Commercially Generated
Radioactive Waste" (DOE80X
This study, known as the GEIS, was issued as a basis for reexamining the strategy for disposing
HL W in a mined geologic repository with several alternative configurations. Reference
repositories located in salt, granite, shale, and basalt were analyzed at a "reference" location in a
midwestern state. Preclosure facility risk categories analyzed included routine exposures to the
regional population and maximum individual exposure from potential worst-case accidents.
Routine doses to the regional population from chronic radiological releases to the atmosphere
were derived from the standard Gaussian dispersion model. Of a total of 207 potential accident
scenarios considered, 116 had the potential for offsite exposure. Dropping a spent fuel canister
was considered the most serious radiological event, with an estimated frequency of occurrence of
l.OE-05 per year.
Exposure to the 50-mile population from routine repository preclosure operation was considered
"negligible," with no quantitative estimate given.
For all accidents considered, public exposure was estimated at 5.0E-05 person-rem per year; and
for any single worst-case accident, a maximum individual exposure of 1.1E-04 rem was
estimated.
10-16
-------
10.4.3 "Review of Drv Storage Concepts Using Probabilistic Risk Assessment" rORV84>)
This study provided a comparative risk analysis for dry storage of SNF at reactor sites. Assessed
storage designs included drywall, storage cask, and vault. The reference reactor facility was a
1,000-MWe PWR that was assumed to discharge 60 spent fuel assemblies (~i9 MTU) to the fuel
Pool. After five years of cooling, SNF was transferred to dry storage with a capacity of 2,400
foel assemblies (-62 MTU).
To model accidental release fractions, a starting assumption was that one percent of the fuel was
failed prior to any of the accident scenarios. Principal elements modelled included long-lived
radionuclides of Kr, I, Cs, Sr, Cs, and actinides.
Conservative model parameters were used to estimate population doses out to a distance of 200
miles from the reference reactor site. For example, primary model parameters used included the
relatively high population density distribution for the Zion nuclear power plant, one meter per
second wind velocity, and stability class D. Table 10-8 defines population dose risks for 12
accident scenarios. For cask and drywall storage, the annual population doses of 23 and 14
Person-rem, respectively, were estimated.
10.4.4 "Requirement far th* TndepepH^t Storage of Spent Fuel and HJgh-Level Radioactive
Waste" rNRC84\
This environmental assessment study performed by the Nuclear Regulatory Commission
analyzed a fuel storage installation (i.e., MRS) that was intended to accommodate 70,000 MTU
for a period of 70 years using dry store technology. It was further assumed that the facility would
receive 3,500 MTU per year for a period of 20 years, with an additional 50-year storage period at
Maximum capacity.
The NRC assumed MRS construction designs that effectively reduce potential air emissions to
near zero levels. Public doses resulting from routine operation/storage were, therefore, assumed
to be "insignificant."
For accidental fuel canister failure containing 1.7 MTU, the Commission estimated a maximum
individual dose of about 1 x 10"6 rem per year per event. -
10-17
-------
Table 10-8. At-Reactor Storage Accidents: Summary of Results (ORV84)
Aeci&at Scenario- ' '
Fuel Assembly Drop During Loading
Drop of Transport Cask During
fading
Cask
Drywall
Venting of Cask During Transport
Cask
Drywall
Collision During Transport
Cask
Drywall
Collision with Fire During Transport
Cask
Drywall
Canister Drop During Emplacement
Drywall
Canister Shear During Emplacement
Drywall
Cask Drop During Emplacement
Cask
Tornado Missile Penetration
Cask
Drywall
Plane Crash Topples Cask with Fire
Cask
Plane Crash Plus Fire
Cask
Drywall
Earthquake
Cask
Drywall
Total Risk: Cask
Drvwall
•.
feyeat&vrl ..
1E-01
4E-03
7E-02
2E-03
3E-02
2E-04
2E-05
2E-06
2E-07
1E-06
2E-06
IE-OS
6E-06
1E-04
6E-09
9E-09
2E-07
2E-08
4E-06
4E-08
8E-06
8E-07
2E-08
1
10
10
24
1
24
1
24
1
1
1
24
10
10
24
24
I
10
24
2,400
1
10
2,400
.Dose Coaaep.eace -f
Vi'^PH.-rTgtBiitegeBtl-. . '
4E-01
4E+00
4E+00
1E+03
4E+01
1E+03
4E+01
5E+03
2E+02
4E+01
4E+01
1E+03
4E+02
4E+02
5E+03
5E+03
2E+02
2E+03
1E+03
1E+05
4E+01
4E+02
2.4E-K)4
4E-02
2E-02
3E-01
-
2E+00
1E+00
2E-01
8E-04
1E-02
4E-05
4E-05
8E-05
1E-05
2E-04
4E-02
3E-05
4E-05
4E-05
4E-05
4E-03
4E-03
3E-04
3E-04
5E-04
2.3E+00
1.4E+00 J
10-18
-------
10.4.5 "Environmental Assessment Related to the Construction 'and Operation of the Surrv Dry
Cask Independent Spent Fuel Storage Installation" (NRC85)
Virginia Electric and Power Company conducted a study as part of an application for a license to
construct and operate an onsite Dry Cask ISFSI. The function of the Dry Cast ISFSI was to
Provide on-site interim storage for about 420 MTU of spent fuel from its reactors, Surry 1 and 2.
Based on cask design and specifications, no liquid and gaseous releases were assumed and only
direct irradiation was considered for an exposure pathway under normal facility operations. The
estimated maximum annual dose to the nearest actual offsite person located at 2.5 km from a
direct radiation source was estimated to be 6E-05 mrem/yr.
For a hypothetical worst case accident, an upper-bound individual dose of 1.35 mrem/yr was
estimated at the site boundary.
10-4.6 "Envii
.1 ^ggftgfitnent n^f Smith County Site. Texas" (DOE86a>)
This DOE study evaluated the suitability of an HLW repository site in salt as specified in the
NWPA of 1982. The reference repository was assumed to have a capacity of 36,000 MTU SNF
and 3,510 metric tonnes of DOE-HLW. SNF was assumed to be 6.5 years old with an average
annual receipt rate of 634,000 rods per year.
Estimated offsite radiological effects for routine and accidental releases are cited in Table 10-9.
Table 10-9. Preclosure Exposure Associated with
a Reference Salt Repository
(Based on data from DOE86a)
^Formal Operation
50-Mile Population
RMEI
LAccJdeQj
50-Mile Population
RMEI
390 person-rein/yr
5.6E-03 rem/yr
3,000 person-rem/event
4.7E-02 rem/event
10-19
-------
10.4.7 "Preliminary Assessment of Radiological Doses in Alternative Waste Management
Systems Without an MRS Facility" (SCH86)
This study analyzed the effects of nine waste management system alternatives, excluding an
MRS facility, specifically dealt with at-reactor facilities and the surface facilities of a deep
geologic repository site. The nine alternatives largely involved transportation modes and options
between reactors and the repository and resulted in nominal differences in public exposures.
Public doses from routine waste management activities at reactor facilities were estimated to be
at less than one person-rem per year per 1,000 MTU. The potential drop of a fuel assembly was
cited as the typical accident scenario, with an occurrence frequency estimated at 0.006 per year
and an estimated population exposure of 0.1 person-rem per event.
Public doses under normal operating conditions at the surface facility of a repository would likely
be due to effluents associated with cask venting and fuel consolidation. On a 1,000 MTU basis,
public exposure was estimated to be 6 person-rem. Public doses from accidental fuel-assembly
and shipping cask drop were estimated to vary between 0.03 and 0.006 person-rem per year.
Table 10-10 summarizes the public-dose estimates cited in this study.
Table 10-10. Public-Dose Estimates for Reference Reactor and Repository Surface Facility
(Based on data from SCH86)
SNF Handling At-Reactor
SNF Handling/Consolidationat
Repository Surface Facility
Transportation Between
Reactor and Repository
At-Reactor
At Reositor Surface Facility
l
6
164
0.0006
0.03 - 0.006
10-20
-------
10.4.8 "Monitored Retrievable
Submi^on to Congress
This comprehensive environmental assessment study was prepared by DOE and submitted to
Congress under Section 141 of the Nuclear Waste Policy Act. The study assessed impacts to
humans and the environment from construction, operation, and decommissioning of an MRS
facility at three potential sites (Clinch River, Oak Ridge, and Hartville, Tennessee) and two
Possible designs for a storage system: storage casks and field dry wells. Radiological doses from
routine emissions from the MRS facility were estimated, along with doses resulting from specific
accident scenarios. The facility was assumed to have a 26-year operating life with a total
throughput of 62,000 MTU.
For routine operations that assume a "store only MRS" and a repository that containerizes intact
the estimated population doses are given in Tables 10-1 la and 10-1 Ib.
Tables 10-1 la and 10-1 Ib indicate that: 1) exposures to members of the public are significantly
greater for routine operations than for accidental releases, 2) exposures are consistent with values
cited by the MRS Review Commission, and 3) exposures for the RMEI are well below all
regulator}' limits.
Table 10-1 la. Public Doses From Routine Operations at MRS and Repository (DOE86a)
MRS-Routine Release
Repository-Routine
Release
Transportation-Normal
- Reactors to MRS
- MRS to Repositor
0.1*
-------
Table 10-1 Ib. Public Doses from Accidental Releases at MRS and Repository (DOE86a)
•* ; * •*
! - V 'V '
i Aedfal Events ""
MRS Facility
- Fuel Assembly Drop
- Shipping Cask Drop
- Storage Cask Drop
• Repository
>-Aim«ai O&ftef f -- ' :
-l"op«ktiDni>!Osev ,
f (pei^fl-rert^yr)
0.03
0.006
0.006
N/A
-, ^ 5
,, JRMEI ''
Xt«m/K)
0.004
0.0009
0.0009
N/A
10.4.9 "The Safety Evaluation of Tunnel Rack and Drv Well Monitored Retrievable Storage.
Concepts" (LIG&3}
This study evaluated the safety of dry rack storage of SNF at an MRS facility. Only the
radiological risks to members of the public due to select accidents during MRS operations were
addressed. The MRS facility was assumed: 1) to have a throughput of 900 MTU per year; 2) to
receive SNF for 36 years; and 3) to store the fuel for 100 years before it was shipped to a
repository for permanent disposal.
In total, the study analyzed 15 different accident scenarios that included transportation collision
during emplacement/retrieval of SNF, canister drops during emplacement/retrieval, an airplane
crash with and without fire, and pin failure resulting from seismic events. Accident frequencies
for these scenarios ranged from a low of 4E-10 per year (plane crash) to 1.4E-01 per year
(canister drop during retrieval).
For all 15 accident scenarios combined, the annual population dose risk was estimated to be 17
person-rem for the 0-50 mile population.
10.4.10 Summary Assessment of Available Data
This chapter of the BID has presented information from a variety of studies that provided
quantitative estimates of population doses from future fuel management activities that may be
conducted at individual reactor sites, at a centralized monitored retrievable storage facility, and at
a deep geologic repository.
10-22
-------
Because the MRS Review Commission Study provided data for all three facility types, it was
regarded as the reference study. Table 10-12 presents population doses for a 50-mile area around
a given facility based on normalized Commission data.
Table 10-12. Normalized Population Doses (Based on data from MRS89)
At-Reactor Storage
MRS (Eastern U.S. location)
'ository (at Yucca Mountain)
7.8E-02
3.8E-02
< 9.6E-05
While acknowledging the much lower population dose associated with deep geologic disposal,
the Commission concluded that short-term spent fuel storage at reactor facilities or at an MRS
facility, albeit higher, yielded population doses that were safe. In association with this
conclusion, the Commission offered the following summary statements:
spent fuel management operations have been safely carried out at
reactors for many years under NRC regulatory control and by trained
personnel.... Although the inventory of spent fuel at reactors is increasing
there is no reason to believe that safe management of fuel cannot continue
or that the fuel will interfere significantly with safe reactor operations.
It appears that most, if not all, reactor sites can safely store all of the spent
fuel that would be generated during the reactor's 40-year operation life....
This storage can be expanded as necessary to meet life-of-plant storage
requirements.... At most sites, life-of-plant storage can be accomplished by
reracking spent fuel pool and using dry storage.
From a technical perspective, both the No-MRS and MRS options are
safe Although neither option is completely without risk, ...the risks are
expected to be small and within regulatory limits, and the degree of
difference in risks is so small that the magnitude of difference should not
affect the decision as to whether there should be an MRS.
The views of the MRS Review Commission are generally supported by other data. Radiological
dose values from various other studies were reviewed and are summarized in Table 10-13.
10-23
-------
As indicated by the brief overviews presented, the studies varied widely in scope, facility
designs, primary assumptions regarding SNF processing, waste packaging, accident scenarios,
modeling approaches, and other factors.
Table 10-13 illustrates the risk categories that have been addressed and the forms or units hi
which the data were reported. Although an attempt was made to present results in common units
for unbiased comparison, necessary data were frequently lacking. For example, some studies
presented accident risk in terms of resultant dose without specifying the probability of occurrence
for a given accident scenario. In other instances, 50-mile population doses were cited for the
entire preclosure period of a facility with full awareness that SNF inventories were
variable/accumulating throughout that period. In the absence of detailed inventory data, a
representative annual population dose could not be determined.
10-24
-------
Table 10-13. Summary Data of Public Doses Associated with SNF Storage At-Reactor, MRS, and Repository
SB*-*
Bechtel(1979)
DOE (1980)
Orvis(1984)
NRC (1984)
DOE (1985)
DOE (1986a)
Schneider (1986)
DOE(1986b)
Ligon (1983)
MRS Review
Commission (1989)
" /',< ~-'s , „
iattW^
negligible'
negligible'
l.OE+00*
,
7.8E-02*
'-3»H
Accident. 1
2.3E+000
6.0E-04C
i* <•••.••>
tor ,s -; "f", "-;
'/ V",Pl
6.0E-08C
Jf /;*J^
Ace/dent
1.0E-06e
^
$w**&\
yfa&M?-.
i.oE-or
3.8E-02'
>Huj»;^r :»;/":"
,"p^'>
j^feni'^
3.0E-02C
1.7E+01C
}V;f *&#/'-/-.
^^i^>
l.OE-030
'Mdl^
4.0E-03'
' ' kfepasltoty
ii^tot-
;WatJ«e -
1.5E-06"
negligible
3.9E+02"
6.0E+00'
l.OE-01'
9.6E-05'
=•-- "-
" Pm*
-MM* '
1.1E-02'
3.0E+03f
negligible
5.6E-03e
l.OE-030
Accident :
1.1E-041"
4.7E-02f
Population Doses:
a = person-rem/1,000 MTU-yr
b = person-rem/SNF inventory-yr (see text)
c = person-rem/yr
RMEI Doses: d = Rem/1,000 MTU-yr
e = Rem/yr
f = Rem/event
10-25
-------
REFERENCES
BEC79 Bechtel National, Inc., 1979, An Assessment ofL WR Spent Fuel Disposal Options,
ONWI-39, prepared for the Office of Nuclear Waste Isolation, Columbus, OH.
CAR87 Carter, Luther, Nuclear Imperatives and Public Trust, Washington, D.C.:
Resources for the Future, 1987 p. 46.
DOE80 U.S. Department of Energy, 1980, Generic Environmental Impact Statement,
Management of Commercially Generated Radioactive Waste, DOE/EIS-0046F,
Vol. 1 of 3, Office of Nuclear Waste Management, Washington, D.C.
DOE86a U.S. Department of Energy, 1986, Environmental Assessment Deaf Smith County
Site, Texas, DOE/RW-0069, Office of Civilian Radioactive Waste Management,
Washington, D.C.
DOE86b U.S. Department of Energy, 1986, Monitored Retrievable Storage Submission to
Congress, DOE/RW-0035, Office of Civilian Radioactive Waste Management,
Washington, D.C.
EnPA92 Energy Policy Act of 1992, Public Law 102-486, October 24, 1992.
HEW87 Hewlett, Richard G., Federal Policy for the Disposal of Highly Radioactive
Wastes from Commercial Nuclear Power Plants, Department of Energy,
Washington, D.C., March 9,1987, p. 21.
LIG83 Ligon, D.A., et al., 1983, The Safety Evaluation of Tunnel Rack and Dry Well
Monitored Retrievable Storage Concepts, GA-A16955, GA Technologies, Inc.,
San Diego, CA.
MRS89 Monitored Retrievable Storage Revew Committee, Nuclear Waste: Is There a
Need for Federal Interim Storage?, November 1,1989.
NPJ97 Nuclear Plant Journal, U.S. Rad Waste Disposal Update, Vol. 15, No. 1, Jan.-
Feb. 1997.
10-26
-------
NRC84 U.S. Nuclear Regulatory Commission, 1984, Licensing Requirements for the
Independent Storage of Spent Fuel and High-Level Radioactive Waste, NUREG-
1092, Washington, D.C.
NRC85 U.S. Nuclear Regulatory Commission, 1985, Environmental Assessment Related
to the Construction and Operations of the Surry Dry Cask Independent Spent Fuel
Storage Installation, Virginia Electric and Power Company, Docket No. 72-2,
Washington, D.C.
ORV84 Orvis, D.D., 1984, Review of Proposed Dry Storage Concepts Using Probabilistic
Risk Assessment, EPRI/NP-3365, NUS Corporation, San Diego, CA
SCH86 Schneider, K. J., et al., 1986, Preliminary Assessment of Radiological Doses in
Alternative Waste Management Systems Without an MRS Facility, PNL-5872,
Pacific Northwest Laboratory, Richland, WA.
USC72 U.S. Congress, Joint Committee on Atomic Energy, Report on Authorizing
Legislation for the Atomic Energy Commission for Fiscal Year 1973, JCAE
Report 92-1066,92nd Congress, 2d. Session, May 16, 1972, p. 11.
USC75 U.S. Congress, Hearings on ERDA Authorizing Legislation for Fiscal Year 1976
Before Subcommittee on Legislation, Joint Commission on Atomic Energy, Part 2,
94th Congress, 1st Session, February 27,1975, pp. 1260-1261,1297.
USC96 Senate Report 104-320, p. 98, September 12,1996.
10-27
-------
GLOSSARY OF TERMS
alluvial fan
alluvium
anisotropic
Antler Orogeny
aquifer
aquitard
argillhe
andesite
ash
-fall tuff
ash-flow tuff
bar
An outspread, gently sloping mass of alluvium deposited by a stream.
A general term for clay, silt, sand, gravel, or similar material that is not
compacted and has been deposited in fairly recent geologic time by
streams, rivers, or floods.
Having some physical property that varies with direction.
A mountain building episode that extensively deformed Paleozoic rocks of
the Great Basin in Nevada during the late Devonian and early
Mississippian time.
A formation, group of formations, or part of a formation that contains
sufficient saturated permeable material to yield quantities of water to wells
and springs.
A geologic formation, group of formations, or part of a formation through
which virtually no water moves.
A compact rock, derived from either mudstone or shale, that has
undergone a somewhat higher degree of induration than mudstone or shale
but is less clearly laminated than shale and without its fissility, and that
lacks the cleavage distinctive of slate.
A dark colored, fine-grained, extrusive igneous rock.
Pyroclastic material less than 4.0 mm in diameter. This term refers to both
unconsolidated detritus and the consolidated deposit.
(1) A tuff deposited by volcanic ash settling out of the atmosphere and
forming a blanketing deposit of relatively uniform thickness regardless of
the underlying topography. (2) A deposit of volcanic ash resulting from
such a fall and lying on the ground surface.
A tuff deposited by a volcano-derived hot density current. It can be either
welded or unwelded and often fills in channels making the thickness of the
resulting deposit a function of the underlying topography.
A measurement of pressure, one bar being approximately equal to standard
sea level atmospheric pressure, or 14.7 pounds per square inch.
G-l
-------
basement rock
Basin and Range
basalt A dark to medium dark igneous rock usually formed from lava flows and
composed chiefly of calcic plagioclase and clinopyroxene minerals in a
glassy or fine-grained ground mass.
Undifferentiated rocks that underlie the stratified rocks of interest in an
area.
Physiographic province in the southwest United States characterized by a
province series of tilted fault blocks forming longitudinal, asymmetric
ndges or mountains and broad, intervening basins.
Basin-Range faulting Faulting characterized by normal (extensional) fault movements. Regional
geologic structure dominated by generally subparallel fault-block
mountains separated by broad alluvium-filled basins.
Rock consisting of sharp fragments cemented together or embedded in a
fine-grained matrix.
breccia
bulk modulus
byproduct
material
caldera
caliche
Cambrian
A modulus of elasticity which relates a change in volume to the
hydrostatic state of stress. It is the reciprocal of elasticity.
Any radioactive material (except special nuclear material) yielded in or
made radioactive by exposure to the radiation incident to the process of
producing or utilizing special nuclear material and the tailings or wastes
produced by the extraction or concentration of uranium or thorium from
any ore processed primarily for its source material content.
A volcanic collapse structure, generally on the order of tens of kilometers
in diameter, formed during the eruption of volumetrically large (tens to
hundreds of cubic kilometers of rock) ash-flow and ash-flow tuff deposits.
A term applied broadly in the southwest United States to a reddish-brown
to buffer white calcareous material of secondary accumulation, commonly
found in layers on or near the surface of stony soils of arid and semiarid
regions, but also occurring as a subsoil deposit in subhumid climates. It is
composed largely of crusts of soluble calcium salts in addition to such
materials as gravel, sand, silt, and clay. Caliche appears to form by a
variety of processes, e.g. capillary action, in which soil solutions rise to the
surface and on evaporation deposit their salt content on or in the surface
materials.
The oldest of the periods of the Paleozoic Era, which lasted from 570
million to 500 million years ago.
G-2
-------
carbonate
carbonate rocks
Cenozoic
clastic
colluvium
conceptual model
c°nglomerate
containment
inheritance
dacite
Devonian
dike
design earthquake
earthquake
A sediment formed by the organic or inorganic precipitation from aqueous
solution of carbonates of calcium, magnesium, or iron.
A rock consisting chiefly of carbonate minerals, such as limestone and
dolomite.
The latest of the eras into which geologic time, as recorded by stratified
rocks of the earth's crust, is divided; this era is considered to have begun
about 65 million years ago.
Pertaining to a rock or sediment composed principally of broken fragments
that are derived from pre-existing rocks or minerals and that have been
transported some distance from their places of origin.
A general term applied to the accumulation of loose incoherent soil and
rock material at the base of a slope.
A physical description of a system devised to show property variations as
based on field and laboratory measurements and best technical judgments.
A sedimentary rock composed of rounded fragments of pebble to gravel
size cemented in a finer-grained matrix.
The confinement of radioactive waste within a designated boundary.
A property of genes located in mitochondria or chloroplasts (or possibly
other extranuclear organelles).
A fine-grained extrusive igneous rock with the same general composition
as andesite but with less calcic feldspar.
The period of the Paleozoic Era lasting from 408 to 360 million years ago.
A tabular body of igneous rock that cuts across the structure of adjacent
rocks or cuts massive rocks.
A hypothetical earthquake against which protective measures are taken,
commonly used to design seismic protection features in engineered
facilities.
A sudden motion or trembling in the earth caused by tjie release of slowly
accumulated strain.
G-3
-------
en echelon
eolian
epicenter
evapotranspiration
extrusive
fault
fault block
flow breccia
flow unit
fluvial
formation
fracture
gene amplification
Geologic features that are in an overlapping or staggered arrangement "
(e.g., faults). Each is relatively short, but collectively they form a linear
zone in which the strike of the individual features is oblique to that of the
zone as a whole.
Term applied to deposits arranged by the wind.
The point on the earth's surface directly above the exact subsurface
location (hypocenter) of an earthquake
Loss of water from a land area through transpiration of plants and
evaporation from the soil.
Igneous rock that has been erupted onto the surface of the earth.
A fracture or zone of fractures along which there has been displacement of
the sides relative to one another, parallel to the fracture or zone of
fractures.
A structural unit in the earth's crust that is formed by faulting and is
bounded completely or in part by faults. This structure behaves essentially
as a unit during tectonic activity.
A breccia that is formed contemporaneously with the movement of a lava
flow; the cooling crust becomes fragmented while the flow is in motion
and is either incorporated into the flow, or falls in front of the moving flow
and is overtaken.
A group of stacked pyroclastic deposits that were emplaced as separate
ash-flow tuffs during the same or closely associated eruptive event(s).
Of or pertaining to rivers; growing or living in a stream or river; produced
by the action of a stream or river.
The basic rock-stratigraphic unit in the local classification of rocks. It
consists of a body of rock generally characterized by some degree of
internal lithologic homogeneity or distinctive features.
A general term for any break in a rock, whether or not it causes
displacement, due to mechanical failure by stress. Fractures include
cracks, joints, and faults.
Refers to the production of additional copies of a chromosome sequence,
found as intrachromosal or extrachromosomal DNA.
G-4
-------
genomic imprinting Describes a change in a gene that occurs during passage through a sperm
or egg with the result that the paternal and maternal allels have different
properties in the very early embryo. May be caused by methylation of
DNA.
The genetic constitution of an organism.
The geologic, hydrologic, and geochemical systems of the region in which
a geologic repository operations area is or may be located.
The branch of geology that deals with the general configuration of the
earth's surface; specifically the study, classification, description, nature,
origin, and development of landforms.
A usually elongated depression of the earth's crust between two parallel
faults.
A foliated rock formed by regional metamorphism, in which bands of
granular materials alternate with bands of minerals with elongate prismatic
habit.
A medium- to coarse-grained intrusive igneous rock consisting primarily
of feldspar and quartz.
A subdivision of the Basin and Range physiographic province, located in
southern Nevada in a broad desert region. The Yucca Mountain site is
located in the Great Basin.
The highly radioactive waste material that results from the reprocessing of
spent nuclear fuel, including liquid waste produced directly in
reprocessing and any solid waste derived from the liquid, that contains a
combination of transuranic waste and fission products in concentrations
requiring permanent isolation.
historical seismicity Earthquakes that occurred during recorded history, including those
reported before the existence of seismographs and those recorded by
seismographs.
genotype
geologic setting
geomorphology
graben
gneiss
granite
Great B
>asin
high-level waste
Hoi
locene
An epoch of the Quaternary Period, from the end of the Pleistocene to the
present.
G-5
-------
host rock
hypocenter
igneous activity
igneous rock
ignimbrite
imbricate
indurated
intermontane
internal drainage
intrusive
latite
liquefaction
The rock in which the radioactive waste will be emplaced; specifically, the
geologic materials that will directly encompass and will be in close
proximity to the underground repository.
The focus or specific point at which initial rupture occurs in an
earthquake.
The emplacement (intrusion) of molten rock (magma) into material in the
earth's crust or the expulsion (extrusion) of such material onto the earth s
surface or into its atmosphere or surface water.
A rock that solidified from molten or partially molten material (i.e., from
magma). Igneous rock is one of the three main classes into which rocks
are divided, the others being metamorphic rock and sedimentary rock.
A silicic volcanic rock forming thick, massive, compact, lavalike sheets
that cover a wide area. The rock is chiefly a rhyolitic tuff and the deposits
are believed to have been produced by eruption of dense clouds of
incandescent volcanic glass in a semimolten or viscous state from groups
of fissures.
A tectonic structure displayed by a series of nearly parallel faults,
characterized by rock slices, sheets, plates, blocks, or wedges that are
approximately equidistant and have the same displacement and that are all
steeply inclined in the same direction.
Used to describe a rock or soil hardened or consolidated by pressure,
cementation or heat.
Situated between or surrounded by mountains, mountain ranges, or
mountainous regions.
Surface drainage in which the water does not reach the ocean, such as
drainage toward the central part of an interior basin.
Of or pertaining to the emplacement of magma in preexisting rock.
A porphyritic extrusive rock with nearly equal amounts of plagioclase and
potassium feldspar, little or no quartz, and a fairly finely crystalline
groundmass.
In cohesionless soil, the transformation from a solid to a liquid as a result
of increased pore pressure and reduced effective stress.
G-6
-------
lithology
lithophysae
lithostatic pressure
low-level waste
magma
magmatism
magnitude
metamorphism
miogeosyncline
Mississippian
mixed low-level
radioactive and
hazardous waste
The study of rocks. Also the description of a rock on the basis of such
characteristics as structure, color, mineral composition, grain size, and
arrangement of its component parts.
Bubble-like structures in rocks, generally hollow, composed of concentric
shells of finely crystalline alkali feldspar, quartz, and other minerals.
The vertical pressure at a point in the earth's crust, equal to the pressure
caused by the weight of a column of the overlying rock or soil.
Waste that contains radioactivity and is not classified as high-level,
transuranic, or spent nuclear fuel or byproduct material. Test specimens of
fissionable material irradiated for research and development only, and not
for the production of power or plutonium, may also be classified as low-
level waste, provided the concentration of transuranic is less than 100
nanocuries per gram.
Naturally occurring mobile rock material, generally within the earth and
capable of extrusion and intrusion, from which igneous rocks are formed
through solidification and related processes.
The development and movement of magma within the earth.
The measure of the strength of an earthquake; related to the energy
released in the form of seismic waves. Magnitude is quantified by a
numerical value on the Richter scale.
The mineralogic, chemical, and structural adjustment of solid rocks to
physical and chemical conditions imposed at depth below the surface
zones of weathering and cementation which differ from the conditions
under which the rocks were formed.
A geosyncline in which volcanism is not associated with sedimentation.
The first period of the Carboniferous Era lasting from 360 to 320 million
years ago.
Mixed low-level waste is defined as waste that satisfies the definition of
low-level radioactive waste in the Low-Level Radioactive Waste Policy
Amendments Act of 1985 and contains hazardous waste that either (1) is
listed as a hazardous waste in Subpart D of 40 CFR Part 261 or (2) causes
the low-level waste to exhibit any of the hazardous waste characteristics
identified in Subpart C of 40 CFR Part 261.
G-7
-------
model
mosaicism
natural barrier
near field
normal fault
orogeny
Paleozoic
phenocryst
phenotype
physiography
physiographic
porphyritic
Proterozoic
A conceptual description and associated mathematical representation of a
system, component, or condition. It is used to predict changes in the
system, component, or condition in response to internal or external stimuli
as well as changes over time and space. An example is a hydrologic
model to predict ground-water travel or radionuclide transport from the
waste emplacement area to the accessible environment.
The condition in which tissues of genetically different types occur in the
same organism.
The physical, mechanical, chemical, and hydrologic characteristics of the
geologic environment that, individually and collectively, act to minimize
or preclude radionuclide transport.
The region where the natural geohydrologic system has been significantly
perturbed by the excavation of the repository and the emplacement of the
waste.
A fault in which the hanging wall (the strata above the fault plane) appears
to have moved down relative to the foot wall (the strata below the fault
plane). The angle of the fault is usually 45 to 90 degrees measured from
the horizontal.
The process of forming mountains, particularly by folding and thrusting.
The era of geologic time, from the end of the Precambrian to the beginning
of the Mesozoic or from about 570 million to 225 million years ago.
A term applied to any large, conspicuous crystal in an igneous rock.
The appearance or other characteristics of an organism, resulting from the
interaction of its genetic constitution with the environment.
The descriptive study of landforms as opposed to geomorphology, which
is the interpretive study of landforms.
A region in which all parts are similar in geologic structure and climate
province and which consequently have a unified geomorphic history.
A texture of igneous rock in which large crystals are set in a finer
groundmass that may be crystalline or glassy or both. •
The entire Precambrian period.
G-8
-------
Precumbrian'
pumice
pyroclastic
quartizite
Quaternary faults
Quaternary Period
retardation
recurrence interval
ring-fracture zone
rhyolite
All geologic time, and its corresponding rocks, that elapsed before the
beginning of the Paleozoic Era (the Paleozoic Era began about 570 million
years ago).
A light-colored, cellular, glassy rock commonly having the composition of
rhyolite.
Pertaining to clastic rock (rock composed of fragments of preexisting
material) material formed by volcanic explosion or aerial expulsion from a
volcanic vent. Also pertaining to rock texture of explosive origin.
Used to refer to either sedimentary or metamorphic rocks composed
chiefly of quartz. In sedimentary petrology, a very hard but
unmetamorphosed sandstone, consisting chiefly of quartz grains that have
been so completely and solidly cemented with secondary silica that the
rock breaks across or through the grains rather than around them. In
metamorphic petrology, a granoblastic metamorphic rock consisting
mainly of quartz and formed by recrystallization of sandstone or chert.
Faults that formed or experienced movement during the Quaternary
Period.
The second part of the Cenozoic Era (after the Tertiary), beginning about
1.8 million years ago and extending to the present.
The process that causes the time required for a given radionuclide to move
between two locations to be greater than the ground-water travel time
because of physical and radionuclide interactions between the radionuclide
and the geohydrologic unit through which the radionuclide travels.
The average time interval between occurrences of a hydrologic or geologic
event of a given or greater magnitude.
A steep-sided fault pattern cylindrical in outline and associated with
caldera subsidence.
A group of extrusive igneous rocks, generally porphyritic and exhibiting
flow texture with crystals of quartz and alkali feldspar in a glassy to
cryptocrystallme groundmass.
Richter magnitude See "Richter scale'
G-9
-------
Richter scale
sandstone
scoria
seismic
seismicity
shale
silicic
siliceous
siltstone
slickenside
source material
special nuclear
material
specific storage
specific yield
A scale for measuring the energy released by an earthquake. It was
derived in 1935 by the seismologist C.F. Richter.
A sedimentary rock composed of cemented or otherwise compacted
material of sand size particles predominantly of quartz but may contain
other mineral components.
Pyroclastic ejecta of basic composition characterized by a high content of
vesicles (cavities) and a texture that is part glassy and part crystalline.
Pertaining to, characteristic of, or produced by earthquakes or earth
vibrations.
The occurrence of earthquakes or the spatial distribution of earthquake
activity. Also the phenomenon of earth movement.
A sedimentary rock in which the particles are of clay size.
A chemical classification of igneous rocks in which silica (SiO2) exceeds
66%.
Of or pertaining to silica (Si02) in various forms.
A sedimentary rock in which the constituent particles are predominantly of
silt size.
Polished or striated (scratched) surface that results from friction along a
fault plane.
Uranium, thorium, or any other material which is determined by NRC to
be source material; or ores containing one or more of the foregoing
materials in concentrations the NRC may by regulation determine from
time to time.
Plutonium, uranium enriched in the isotope 233 or 235, and any other
material which NRC determines to be special nuclear material, but does
not include source material.
The volume of water yielded from a unit volume of a confined aquifer per
unit decline hydraulic head.
The ratio of the volume of water that a given mass of saturated rock or soil
will yield by gravity to the volume of that mass. This ratio is stated as a
percentage.
G-10
-------
spent nuclear
fuel
strike-slip faulting
strike
storativity
surface facilities
talus
tectonic
tectonics
tephra
Tertiary
thrust fault
transmissivity
Fuel that has been withdrawn from a nuclear reactor following
irradiation, the constituent elements of which have not been separated by
reprocessing.
A fault in which the net slip is horizontal or parallel to the strike of the
fault plane.
The direction or trend of a structural surface (e.g., a bedding or fault plane)
as it intersects the horizontal.
The volume of water an aquifer releases from or takes into storage per unit
surface area of the aquifer per unit change in head.
All repository operations and support facilities located on the surface of
the site.
Loose rock fragments of any size or shape derived from, and lying at the
base of, a steep slope.
Of, or pertaining to, the forces involved in tectonics or the resulting
structures or features.
A branch of geology dealing with the broad architecture of the outer part
of the earth; that is, the regional assembling of structural or deformational
features, a study of their mutual relations, their origin, and their evolution.
A collective term for all clastic volcanic materials which are ejected
through the air from a crater, includes volcanic dust and ash, cinders,
lapilli, scoria, pumice, bombs and blocks.
The earlier of the two geologic periods that make up the Cenozoic Era,
extending from 65 million to 1.8 million years ago.
A fault with a dip of 45 degrees or less in which the hanging wall (the rock
mass above the fault plane) appears to have move upward relative to the
foot wall (the rock mass below the fault plane).
The rate at which water is transmitted through a unit width of an aquifer
under a unit hydraulic gradient. In the English Engineering system,
transmissivity values are given in gallons per minute through a vertical
section of an aquifer one foot wide and extending the full saturated height
of an aquifer under a hydraulic gradient of 1; in the International System,
G-ll
-------
transmissivity is given in cubic meters per day through a vertical section of
an aquifer one meter wide and extending the full saturated height of an
aquifer under a hydraulic gradient of 1.
transposable Refers to the movement of a DN A sequence to a new site in the genome.
elements The DNA sequence that is transposable is refered to as a transposon.
transuranic An element with an atomic number greater than that of uranium.
materials
transuranic waste Without regard to source or form, waste containing more than 100
nanocuries (37 becquerels) of alpha-emitting transuranic isotopes, with
half-lives greater than twenty years, per gram of waste, excluding (1) high-
level waste, (2) wastes that DOE and EPA have determined do not need
the degree of isolation required by 40 CFR Part 191, or (3) wastes that the
NRC has approved for disposal on a case-by-case basis with 10 CFR Part
61.
tuff A compacted pyroclastic deposit of volcanic ash and dust that may contain
rock and mineral fragments incorporated during eruption and transport.
uniparental disomy A form of uniparental inheritance, when the genotype of only one parent is
inherited and that of the other parent is permanently lost. Usually it is the
mother whose genotype is inherited in uniparental inheritance.
vitophyre Any porphyritic igneous rock with a glassy groundmass.
vitric Term describing igneous material that is characteristically glassy (i.e.,
contains more than 75% glass).
volcanism The process by which magma and its associated gases rise into the crust of
the earth and are extruded onto the earth's surface and into the
atmosphere.
vug A small cavity in a vein or in rock, usually lined with crystals of different
mineral composition from the enclosing rock.
welded tuff Indurated volcanic ash in which the constituent glassy shards and other
fragments have become welded together, apparently while still hot and
plastic after deposition. Where the distinction between nonwelded and
partly welded tuff is necessary, the boundary should b£ placed at or close
to that point where the deformation of glassy fragments becomes visible.
The transition from partly to densely welded tuff is one of progressive loss
G-12
-------
of pore space accompanied by an increase in the deformation of the shards
and pumiceous fragments.
zeolites A SrouP of hydrous aluminosilicale minerals containing sodium, calcium,
barium, strontium, and potassium, and characterized by their ease of
exchange of these ions.
G-13
-------
APPENDIX I
DEMOGRAPHY AND ECOSYSTEMS
-------
Land Use
The environment in and around Yucca Mountain (YM) is characterized by desert valley and
Great Basin mountain terrain and topography. Its climate, flora, and fauna are typical of the
southern Great Basin deserts. Access is restricted to the Yucca Mountain area due to its
remoteness and its location near the Nevada Test Site (NTS) and adjacent U.S. Air Force
lands. The predominant land use surrounding Yucca Mountain is open range for livestock
grazing, with scattered mining, farming, and recreational areas.
Figure 1-1 delineates the variety of land uses within 180 miles (300 km) of NTS/YM. The
area southeast of Yucca (shown on the southwest border of the NTS) is relatively uniform,
since the Mojave Desert ecosystem comprises most of this part of Nevada and California.
The area directly south is the Amargosa Valley, which has limited but locally intensive
tanning and ranching activity. In the relatively barren area north of Yucca Mountain, the
Bttjor agricultural activity is the grazing of cattle and sheep.
£oj)ulation
As shown in Figure 1-2, eight counties in Nevada and one county in California border Nye
County, Nevada. The county population levels shown in this figure are from the 1990
census and are shown for consistency. However, estimates for Nye County and its
communities were updated in 1994 and will be used in the remainder of this section.
Excluding Clark County, the major population center in Nevada (about 750,000 persons), the
Population density of counties adjacent to Yucca Mountain is about 0.7 people per square
(0.4 per square km).
comparison, the population density of the 48 contiguous states is 70.3 persons per square
(27 per square km). The average population density of Nevada is 10.9 persons per
square mile, or 3.1 per square km. The only region in Nye County with a density greater
than three people per mile is in the extreme southern portion, in and around the community
°f Pahrump, which is 60 miles west of Las Vegas.
1-1
-------
Camping &
Recreational
Areas
Q Hunting
• Fishing
O Mines
A On Reids
^r
LakeHavasu
Figure 1-1. General Land Use Within 180 Miles (300 km) of the Nevada Test Site(EPA93a)
1-2
-------
-I
Figure 1-2.
Population of Arizona, California, Nevada, and Utah Counties Near the
Nevada Test Site (EPA93a)
1-3
-------
The primary area of interest (based on DOE95a) is within an 80-km radius of Yucca
Mountain, shown approximately in Figure 1-2. This region also includes several small
communities that are not shown here, but are generally located southeast to west of the site.
The largest of these communities, Pahrump, is a growing rural community with a 1994
estimated population of 10,892, located about 80 km southeast of Yucca Mountain. Other
communities in the immediate area are Beatty (25 km west) and Amargosa Valley (20 km
south) in Nye County, Nevada; Indian Springs (70 km east) in Clark County, Nevada; as
well as Death Valley Junction in Inyo County, California (55 km south). Also contained in
this area are portions of Death Valley National Park (DVNP). The socioeconomic
characteristics of the Nevada communities are summarized in Table 1-1 and in NYE94.
The Mojave Desert of California, which includes DVNP, lies along the southwest border of
Nevada. Population within the park ranges from a minimum of 200 residents in the summer
to 5,000 tourists per day in winter (excluding major holidays, when as many as 30,000 could
be present). The largest populated area in the region is Ridgecrest, California (160 km
southwest), with a population of 28,000. The Owens Valley, beginning 50 km west of Death
Valley, contains many small towns, the largest of which is Bishop, California, with a 1990
population of 3,475. As shown in Figure 1-2, based on the population levels, the area of
southwestern Utah, due east from Yucca Mountain, is more developed than the adjacent parts
of Nevada. St. George (200 km east) is the largest community, with a 1990 population of
28,500. The extreme northwestern part of Arizona (Mojave County) is mostly range land
except for the portion containing Lake Meade and other small communities along the
Colorado River.
Employment
The NTS, which is adjacent to Yucca Mountain, accounts for a high concentration of
employment in southern Nye County. Many of the workers live in group quarters during the
week. In December 1994, NTS employment was reported to be 3,000, down significantly
from more than 5,000 workers in the mid-1980s. At the same time, the employment at
Yucca Mountain increased twelve-fold, from 38 workers in 1988 to 478 by December 1994.
Table 1-1 shows establishments in the Nye County communities by Standard Industrial
Classification (SIC) Groups and demonstrates potential employment concentrations in the
region.
1-4
-------
Table 1-1. Summary of Socioeconomic Characteristics Compiled by Community for the
First Quarter of 1994 (DOE95a)
Soaoecpnoniic^ » '"
Characteristics " ""
Square Miles
Acreage (1)
lotal Occupied
Housing Units:
single Family (2)
MUlti Family
Group Quarters (3)
lotal Estimated
Population:
bugle Family (2)
Group Quarters (3)
tStahli«hrrv.pK hy
Standard Industrial
Classification
Group:
Ag/For/Fishing (4)
Manufacturing
iCEGSS (5)
wholesale & Retail
Trade
FISE<6)
services
Government
*V< p^1™mP;^ *
?•>
298
190,720
4,879
4,692
187
2
10,892
10,463
417
12
1350
20
98
21
42
178
67
209
25
, ^ -Beattyx „ ";,
692.5
443,200
788
719
69
4
1,947 .
,747
hies
32
134
_,
15
2
8
29
17
47
10
,- - Anpsss'sa ,-•, ».
Valley
499
319,360
352
344
8
909
TJSS
21
60
/
6
1
7
13
20
4
,-. :• ' .s'lhdian • ^
J« -k < V S
Springs
18
11,520
492
492
1,200
1,200
—
37
2
5
7
13
10
(1)
(2)
0)
(4)
* Tax boundaries specified by the Nye County Board of Commissioners are used to delineate the
boundaries for Pahrump, Beatty, and Amargosa Valley. For Indian Springs, the legal description
specified by the Clark County Commissioners for the unincorporated town is used.
Please note: Community boundaries encompass many whole, as well as some partial, cells. Therefore,
information within mis table is not directly comparable to the information presented in the Appendix.
For Pahrump, the information included in this table is for the entire community both inside and outside of
the RadMP grid.
Acreages for the communities hi Nye County were supplied by the Nye County Assessor's Office, and
are the best estimate of the actual acreages encompassed within the taxation boundaries (Nye County
Assessor's Office. 1988)
This category was refined to include all single-family dwellings and mobile homes, due to the new
method of data collection. Units housing persons visiting or residing hi the area on a "short-term"
temporary basis, such as in RV parks, are not included.
This category includes the group quarters in Pahrump and the employee housing in Beatty reported as
the number of facilities hi the housing section (not included in total) and number ol residents in the
population section (included hi total and not used to calculate the PPH).
Agnculture/Forestry/Fishing
TCEGSS refers to Transportation, Communications, Electric, Gas, and Sanitary Services
FIRE refers to Finance, Insurance, Real Estate
1-5
-------
Agriculture
Within the 80 km radius around Yucca Mountain, agricultural activity appears to be holding
steady, with increases hi alfalfa being offset by decreases in acreage planted hi barley and
oats. The Amargosa Valley primarily grows sod/turf, alfalfa, barley, and oats, and relatively
small amounts of fruits. The area west of Pahrump grows primarily alfalfa. The majority of
livestock in the region consists of bee colonies in Pahrump (honey production), catfish
farming hi Amargosa Valley, dairy cows in Pahrump and Amargosa that produce milk
snipped to southern California, pigs raised for commercial consumption locally, and range
cattle. Recent openings of new dairies hi Amargosa represent additional local demand for
these products.
Mining and Construction
Within the 80-km radius, the areas west and south of Yucca Mountain near the communities
of Beatty, Amargosa Valley, and Pahrump contain 12 mining and open pit operations and 71
construction and drilling operations. The activities associated with these businesses include
rnining, sand and gravel operations, construction, drilling, and landfills. Other active mines
and oil and gas wells are widely dispersed throughout the state.
Ecosystems
As described in previous sections, the diverse topography, geology, and climates of the
southern Nevada desert create a complex variety of plantlife. Vegetation ranges from sparse
desert scrub in the lowest valleys to well-developed woodland on highlands above 2,000 m.
Only sheer cliffs and playa floors are devoid of plants. Even the apparently barren hills of
the Amargosa Desert support widely spaced shrubs and succulents.
Table 1-2 shows plant types and associations found hi the regions hi and around Yucca
Mountain. As described hi the table, plant associations classified as Great Basin conifer
woodlands are distinguished by dominance of single-needle pinyon pine and Utah juniper. In
south-central Nevada, these pygmy conifer communities are restricted to elevations above
1,800 meters. Dominant plant taxa in Great Basin desert scrub communities are flowering
plants such as shadscale. These desert scrub associations usually occur below the tree line,
but above Mojave Desert vegetation. The plant species typical of lower elevation Mojave
1-6
-------
desert scrub vegetation, like creosote bush and white bufsage, have their center of
distribution south of Yucca Mountain. Exceptions include plants dominated by species
endemic to the northern Mojave, such as box thorn and grcasewood. The vegetation
classifications at Yucca Mountain are dominated by representatives of Mojave desert scrub,
but many Mojave desert scrub species are at the northern limits of their distribution in
southwest Nevada, and most are restricted to elevations below 1,800 meters.
According to the Nye County Overall Economic Development Plan (NYE93a), the county,
like most of the state, hosts a number of threatened and endangered species, as shown in
Table 1-3. According to the Nye/Esmeralda Economic Development Authority, however,
only two of these have affected growth and development.
The Devil's Hole Pupfish has limited development in the Ash Meadows area of Amargosa
Valley. Protection of the Pupfish and several other threatened and endangered'species
resulted in the creation of the Ash Meadows National Wildlife Refuge. Thus, casual use of
this region is restricted to existing roads, trails, and washes. The ground water level is
Protected in spring flows in Ash Meadows, which is managed by the Fish and Wildlife
Habitat Management Program of the BLM.
Also, like much of southern Nevada, certain areas in Pahrump and Amargosa Valley have
been classified as desert tortoise habitat. Land within this classification requires a biological
assessment before it can be developed. If necessary, the U.S. Fish and Wildlife Service may
require a second environmental assessment and a site-specific habitat conservation plan.
1-7
-------
Table 1-2. Principal Plant-Community Types and Examples of Representative Plant
Associations on Rock Slopes (SPA85)
- f*$aawv&^
Great Basin Conifer Woodland
Pinus monophylla-Quercus gambelii-Juniperus osteosperma
Pinus monophylla-Artemisia tridenta-Juniperus osteosperma
Volcanic highlands in the northern test site;
generally at elevations above 1,950 meters (such
as the eastern Pahute Mesa, Timber Mountain).
Associates include Artemisia tridentata
Symphoricarpos longiflorus, Purshia tridentata,
and Lupinus argenteus.
Highlands at elevations above 1,770 meters;
restricted to xeric habitats at elevations above
2,100 meters. Common associates include
Artemisia nova. Cowania mexicana.
Haplopappus nanus. Brickellia microphylla.
Great Basin Desettscrub
Atriplex canescens-mixed scrub
Atriplex confertifolia-mixed scrub
The flanks of hills and rocky mesas, usually of
volcanic substrate; at elevations from about 1,500
to 2,000
On limestone and dolomite slopes; at elevations
from about 850 to 1,700 meters. Common
associates are usually Mojave Desert shrubs such
as Amphipappus fremintii; Ephedra torreyana,
Larrea divaricata. Gutierrezia microcephala.
Mojave Desertscrub
Lepidium fremontii-mixed scrub
Gutierrezia microcephala-mixed scrub
Ambrosia dumosa-Larrea divaricata
On the talus slopes and ridges of calcareous
mountains: at elevations from about 1,050 to
1,700 meters. Associates include a diverse
complement to upper elevation Mojave
desertscrub species such as Coleogyne
ramosissma, Ephedra torreyana, Buddleja
utahensis, and Lycium andersonii.
Talus slopes, cliff bases, and ridges; generally at
elevations below 1,400 meters on calcareous
substrates. Common associates include Larrea
divaricata. Ambrosia dumosa, Ephedra spp.,
Amphipappus fremontii, Lycium pallidum,
On talus slopes, ridges, and mesas; generally at
elevations below 1,200 meters. Normally
occurring with lower-elevation Mojave Desert
species such as Peucephyllum schottii. Eucnide
urens, Gutierrezia microcephala, Echinocactus
polycephalus. Atriplex confertifolia is common at
some sites. ,
1-8
-------
Table 1-3. Threatened and Endangered Species in Southern Nevada (NYE935)
Common Nanie-
Endangered
American Bald Eagle
Haliaeetus leucocephalus
Ash Meadow speckled dace
Thinichthys osculus nevadensis
Ash Meadow speckled pupfish
Cyprinodon nevadensis mionectes
Cui-ui
Chasmistes cujus
Devil's Hole pupfish
Cyprinodon diabolis
Hiko White River springfish
Chrenichthys baileyi grandis
Moapadace
Moapa coiaicea
Pahnunp poolfish
Empetrichthys latos
Peregrine falcon
Falco peregrihus
Wann Springs pupfish
Cyprinodon nevadensis pectoralis
White River spinedace
Lepidomeda albivallis.
White River springfish
Chrenichthys baileyi baileyi
Threatened
Ash Meadows naucorid
Ambrysus amargosus
Big Spring spinedace
Lepidomeda mollisinis pratensis
Desert dace
Eremichthys acres
Desert tortoise
Xerobates agassizzi
Lahontan cutthroat trout
Oncorhynchus clarki henshawi
[Railroad Valley springfish
Crenichthys nevadae
Sources of Human Radiation Exposure
All members of the public are exposed to ionizing radiation from a variety of sources.
Exposure to some sources is not only inevitable but life-long, while exposure to others may
be episodic and influenced by numerous factors. For convenience, sources of public
exposures are commonly categorized as (1) of natural origin and unperturbed by human
activities, (2) of natural origin but affected by human activities (termed enhanced natural
1-9
-------
sources), and (3) man-made sources.Natural'radiation and naturally occurring radioactivity in
the environment are by far the major sources' of .human radiation exposure. Consequently,
these sources have been extensively studied and are commonly compared with various man-
made sources of ionizing radiation.
Natural sources include cosmic radiation from outer space; terrestrial radiation from
radionuclides in soil, rocks, and other materials; and radionuclides within our bodies. Each
of these natural sources has specific characteristics that cause variations in individual
exposures, which are influenced by geographic location, dietary habits, lifestyles, and other
factors.
Enhanced natural sources include those for which human exposures have increased due to
deliberate or inadvertent behavior. For example, extensive air travel may significantly
increase exposure to cosmic radiation, and tailings from phosphate mining^ when used for
construction fill, can increase terrestrial exposure. Another example involves the combustion
of fossil fuels (coal, oil, natural gas) by industry and electric utilities, which results in the
localized release of naturally occurring radionuclides in stack gases .released into the air.
Even indoor exposure to radon might be considered "enhanced," because air concentrations
of radon and radon daughters are significantly affected by the design, construction, and use
of a home or building.
In addition to natural sources, most individuals are also exposed to radiation from numerous
man-made sources, materials, and devices. The largest category among man-made sources is
classified as medical and refers to a variety of diagnostic and therapeutic procedures (e.g.,
x-rays, fluoroscopic examinations, CAT-scans, radioactive Pharmaceuticals and implants,
exposure to teletherapy units). The public is also exposed to a variety of consumer products,
such as televisions, smoke detectors, nuclear weapons production and testing, and nuclear
power and the associated fuel cycle.
The scientific literature abounds with information on public exposure to natural and man-
made sources. Among the most comprehensive reports are those issued by the EPA
(EPA72a, EPA77) and several prominent scientific committees, including the United Nations
Scientific Committee on the Effects of Atomic Radiation (UNS88, UNS93, UNS94), the
Committee on the Biological Effects of Ionizing Radiation of the National Academy of
Sciences (NAS80, NAS88, NAS90), and the National Council on Radiation Protection and
Measurements (NCR87a, NCR87b, NCR89).
1-10
-------
This section summarizes average exposures received by the general public in the United
States, as well as estimates of exposures to individuals currently residing in the vicinity of
Yucca Mountain.
Natural Radiation Sources and Exposures
Cosmic Radiation
Cosmic radiation refers to primary energetic particles originating from the sun and from
outside the solar system, and to secondary radiation generated by their interaction with the
earth's atmosphere. In the absence of the earth's atmosphere, the biosphere's dose of cosmic
radiation would be about 1,000 times greater than current levels. The intensity of cosmic
radiation is also affected by the earth's geomagnetic field, which varies with latitude, and by
the sun's activity, which follows a cycle of about 11 years." However, at ground level,
variations in cosmic radiation intensity within the continental United States, due to the
geomagnetic field effect, are less than 2 percent (CAR69), while the 11-year variations due
to solar activity are less than 10 percent of the mean level (ERD77).
At sea level, the average cosmic-ray annual dose equivalent is about 26 mrem. This annual
dose rate essentially doubles with each 2,000 meter increase in, altitude in the lower
atmosphere. Accordingly, inhabitants of Denver at 1,600 meters and Leadville, Colorado, at
3,200 meters have estimated annual external exposures from cosmic radiation of 50 mrem
and 125 mrem, respectively (NCR87b). Considering the distribution of altitudes for the U.S.
population, an average annual cosmic-ray dose of 27 mrem is generally assumed, although
the dose to a specific individual is affected by the amount of time spent outdoors and the
shielding provided by indoor environments.
The cosmic-ray exposure rates in aircraft are considerably higher. At normal commercial jet
aircraft altitudes of about 11-12 km, average dose rates of 0.5 mrem/hour have been
estimated (NAS86). A single transcontinental flight from New York to Los Angeles would
be expected to result in an average dose of 2 to 3 mrem, so crew members working ordinary
schedules on high-altitude, long-distance routes would likely receive average doses in excess
of 500 mrem per year (BAR95). Solar flares, although infrequent, can yield dose rates in
excess of 1,000 mrem/hour at these altitudes (UNS82).
1-11
-------
Naturally Occurring Radionuclides
Several dozen naturally occurring radionuclides exist with half-lives of the same order of
magnitude as the estimated age of the earth (about 4.5 billion years). These include
potassium-40, rubidium-87, and radionuclides belonging to the decay chain series of
uraniurn-238, uranium-235, and thorium-232 (such as radium). These naturally occurring
radionuclides contribute to exposure that is external to the body, internal to the body, or
(when inhaled) to tissues of the lungs and respiratory tract.
External Terrestrial Radiation
Potassium-40 and several decay-chain members in each of the uranium and thorium series
emit penetrating gamma radiation. These radionuclides exist in low, but varying,
concentrations in virtually all types of rocks and soil. Since many building products, such as
cut stone, brick, cement, and gypsum are derived from natural stone, they also contribute to
external radiation. For most individuals, external exposure indoors to radionuclides derived
from the terrestrial environment is nearly equivalent to terrestrial exposure outdoors (Table I-
4).
Table 1-4. Comparison of External Indoor to Outdoor Radiation Dose
s)
1. Mostly Woodframe (single homes)
2. Brick (apartment building)
3. Stone (apartments & houses)
4. Steel and Concrete (office building)
70-82
96
80- 100
87- 106
Source: EPA72b
In EPA72b, two distinct major regions of the United States were observed that differed in
average terrestrial dose rates by a factor of about two; the lower dose rate of 16 mrem/year
corresponded to the Atlantic and Gulf Coastal Plain and the remaining major portion of the
country (referred to as Non-Coastal Plain) yielded average dose rates of about 30 mrem/year
(Figure 1-3). The Denver area showed average terrestrial dose rates of 63 mrem/year. (It is
assumed that other areas on the eastern slopes of the Rocky Mountains would show similar
levels.) It should be further noted that for each average value cited above, the range in
values between the 10th and 90th percentile differed by at least a factor of two.
1-12
-------
Midde
(Noa-doostal
Range:
Range:! 50 to 100 nr
Averaae: t53 mra<
Figure 1-3. Terrestrial Dose Rates in the United States
Eadionuclides in the Body
Naturally occurring radionuclides in rocks and soil also give rise to internal radiation
exposure because they are present in drinking water and foods that are consumed by humans.
Upon ingestion, their distribution in the body is complex and is governed by then- chemical
Properties and the physiological regulatory mechanisms of the body. For example,
radioactive potassium-40 exists in a fixed ratio to its non-radioactive form of potassium.
Potassium is an essential dietary element to most living systems, and it is distributed nearly
uniformly in lean soft tissue.
1-13
-------
The concentrations of other radionuclides associated with the decay chains of uranium and
thorium are potentially highly variable. For humans, internal exposure to these naturally
occurring radionuclides is not only determined by dietary composition but also by the total
quantities of foods consumed and the geographic location/origin of the food products.
Maximum body burdens would, therefore, be expected for individuals who consume foods
(and water) 1) that are derived from areas of high terrestrial background radioactivity, 2) that
are higher in radionuclide content than foods of normal diets, and 3) in large quantities.
Internal Doses. Reports of unusual exposure hi the United States are contained in the
literature. High exposures have been documented for Eskimo and select Indian tribes whose
diet includes reindeer caribou, moose, elk, and deer that subsist largely on lichen during
winter months. Here the lichen-animal-human food chain leads to high concentrations of
natural Pb-210 and Po-210 (as well as Cs-137 from fallout) in the diet (HOL66, ECK86
MAT75). - -
In general, however, there are insufficient data at this time to define the extent of variability
of internal exposures among individuals. Factors cited above that would potentially yield
significant differences in exposure are largely eliminated by a food distribution system that is
nationwide and offers a wide range of foods.
Dose estimates from internally deposited naturally occurring radionuclides are principally
based on limited post-mortem measurements of the nuclide content of various organs.
Estimates of average doses to specific tissues of the body are given in Table 1-5. By
multiplying annual tissue doses by their corresponding weighing factors, it is estimated that
the average individual in the United States receives an internal dose of 39 mrem per year
from naturally occurring radionuclides.
Inhaled Radionuclides
Except for radon and its short-lived decay products, the inhalation pathway yields doses that
are small relative to the ingestion pathway. For this reason, reference to inhaled
radionuclides
1-14
-------
Table 1-5. Average Annual Dose from Internal Exposure to Naturally
Occurring Radionuclides
, jrr -< - -; *
Tissue ,
Lung
Gonads
Bone Surfaces
Bone Marrow
Other Tissues
Total Body
. , ti&vSab&tM
33
36
100
50
35
&%s1ul^g^ng *
««><$&***
0.12
0.25
0.03
0.12
0.48
1.00
fifff^ttv^'' T^nptt *"•'"''• f •
Equivalents (mrem)
4
9
3
6
17
39
is limited to gaseous radon and its solid radioactive daughter products. This group of
radionuclides belongs to the uranhim-238 and thorium-232 decay chain series and is,
therefore, present in all soils and rocks, as previously discussed.
As a gas, radon is able to diffuse through soil pore spaces and escape into surrounding
media. In outdoor air, radon quickly diffuses and results in relatively small exposures.
Significant exposures, however, result when radon gas enters homes and other buildings
where it is concentrated and contained for periods of time that allow the buildup of its solid
(i-e., non-gaseous) radioactive daughters. When inhaled, paniculate radon daughters deposit
onto the mucous layer of the airways, some of which emit energetic alpha particles and
impart substantial doses to cells that line the upper respiratory tract and lungs.
Estimates of Radon Doses. The magnitude of human exposure to radon and radon daughters
is not only influenced by the amount of radon formed in the soil, but also by numerous other
factors that affect its 1) migration and movement through soil, 2) rate of entry into a home or
building, and 3) containment and daughter buildup in indoor air. For example, factors
affecting exposure include soil porosity and moisture content, the design of the structure, and
building operating variables related to heating, air conditioning, and ventilation.
A large number of localized survey data exist of indoor radon or decay-product
concentrations. However, most of these measurements have been made hi single-family
houses, and little data exist for high rise apartments, commercial and industrial buildings,
and other structures hi which various population subgroups spend significant amounts of
time.
1-15
-------
Further complicating estimates of average exposure and the distribution of individual
exposures is the fact that existing data exhibit very large variations. For example, clusters of
unusually high indoor radon concentrations have been found in parts of northeastern
Pennsylvania, New Jersey, and southeastern New York. This area, known as the Reading
Prong, is characterized by soil concentrations of uranium series radionuclides that are about
100 tunes greater than the national average.
Best estimates suggest an average indoor radon concentration of about 1.25 pCi/L, which
yields an average individual dose (i.e., effective dose equivalent) of about 200 mrem per
year (NCR87a, NCR87b).
Summary of Natural Background Exposures"
Table 1-6 summarizes the average individual exposures in the United States to natural
background radiation. On average, cosmic radiation, external terrestrial gamma radiation,
and radionuclides within the body contribute nearly equally, and yield a total dose of about
100 mrem per year. However, this is only about one-half of the annual inhalation exposure
resulting from indoor radon and its decay products, estimated to be 200 mrem. It is
important to note, however, that the dose resulting from exposure to indoor radon has a very
large amount of variability, with a significant number of people experiencing doses that are
several times that of the average value.
Table 1-6. Estimated Average Annual Dose* To Members of the Public in
the United States from Natural Background Radiation
Source
Cosmic
Terrestrial
In the Body
Inhaled (Radon)
Rounded Total
Dose* (mrem/yr)
28
28
39
200
300
* All doses are expressed as effective dose equivalents.
1-16
-------
Public Exposures to Artificial and Other Sources of Radiation
Members of the public are also exposed to a variety of radiation sources categorized as
"artificial," those that involve human technological activities. Among artificial sources, the
largest exposures involve medical x-rays and radioactive Pharmaceuticals used for diagnostic
purposes or the treatment of various diseases. There are also a number of consumer
products that contain radioactivity or emit radiation.
It is estimated that on average, artificial sources contribute an annual dose of 60 mrem. An
important aspect of human exposure to most of these sources is that the exposure is 1)
episodic, 2) voluntary, or 3) has an .associated benefit to the individual or society at large.
Medical Sources
Radiation has broad application in the diagnosis and treatment of various diseases that affect
humans and is widely employed by physicians, dentists, podiatrists, chiropractors, and other
health-care professionals. Diagnostic radiation principally involves x-rays, fluoroscopic
examinations, and specialized medical imaging procedures such as computerized tomography
and scans involving radiopharmaceuticals. Therapeutic uses of radiation, such as in cancer
therapy, involve similar sources of radiation but deliver larger individual doses.
Table 1-7 provides summary data on medical radiation exposures in the United States
(NCR89). It is estimated that on average, medical radiation contributes an annual dose of
about 54 mrem to individuals living in the United States. While such an average value is an
important and useful statistic, it must be pointed out that the distribution of medical
exposures among individuals is highly variable.
Consumer Products
Many commercial and consumer products either emit radiation or contain radioactive
materials In 1977, the NCRP issued a comprehensive report, NCRP Report No. 56
(NCR77) that estimated population exposures from consumer products. Because of revised
Federal regulations and newly introduced technologies, the NCRP updated its earlier
estimates in a revised report issued in 1987 (NCR87c). Scaling the population estimates
contained in the 1987 report to the current 260 million, it is estimated that the average annual
exposure in the United States to consumer products is in the range of 6 to 13 mrem.
1-17
-------
Table 1-7. Estimated Total Number and Frequencies of Diagnostic Procedures
in the United States and Associated Radiation Doses
f
<5roup
Diagnostic X-Rays2
• Physicians/Osteopaths
• Chiropractors
• Podiatrists
• Dentists
RadioDharmaceuticals3
Total
•• f
vTotalNo. of
Exams
181,000,000
164,000,000
9,800,000
5,900,000
101,000,000
7,400,000
™".'. **^TTrT~, . il. '"-'-,
•. *• f
No*
f f
. Range of
Doses per
Exam ".
(mrem)1
6 > 400
—
—
150-1,200
••\ ''t'J*'
. ' - ^i
AfrgiDdSfeV
Perfixam
-------
nature of these facilities/sources suggests that public exposures are highly variable and
primarily affect near-field residents. Doses as high as 50 mrem/year, for example, have
been assigned to near-field residents of the Oak Ridge Reservation (NCR87c). The total
estimated population dose of 16,000 person-rem from all such sources yields an average
annual dose of about 0.6 mrem to individuals in the near-field. Since the exposed near-field
population represents only about 10 percent of the total U.S. population, these miscellaneous
sources are thought to contribute about 0.06 mrem/year on average to members of the total
population.
Between 1945 and 1962, atmospheric tests conducted by the United States involved nuclear
weapons with an explosive yield of about 140 megatons of TNT equivalence, which
represents about 27 percent of the world's total, estimated to be 510 megatons (UNS82).
The radiation dose commitment from fallout has changed significantly over the years due to
natural decay and depletion/removal mechanisms from environmental media, which limit
further biological uptake. The current dose rate to members of the general public is
estimated to be 1 mrem per year (NCR87b), derived largely from residual radionuclides
contained in our bodies.
1-19
-------
APPENDIX H
RADIONUCLIDE EXPOSURES TO PERSONS IN THE VICINITY OF THE
NEVADA TEST SITE/YUCCA MOUNTAIN SITE
-------
APPENDIX H
RADIONUCLIDE EXPOSURES TO PERSONS IN THE VICINITY OF THE
NEVADA TEST SITE/YUCCA MOUNTAIN SITE
For populations living near the Nevada Test Site (NTS), exposure to normal background
radiation levels may be modified by NTS activities. Past NTS activities have introduced
radioactivity into air, soil, and ground water. Atmospherically-released radioactivity has
contributed to local and global exposure in the past and will continue to do so in the future.
Radioactivity introduced on-site into soil and proximally to ground water have the potential to
expose future persons living in the vicinity of the NTS. For purposes of this discussion, past
NTS activities relevant to human exposure may be grouped into several discrete categories that
are discussed below. In addition, risks and consequences from these activities are
summarized.
Atmospheric Weapons Testing. Over 100 events on the Nevada Test Site have resulted in the
release and deposition of radionuclides on the soil surface outside the test-site boundary
(HIC90). Atmospheric weapons testing at NTS began in 1951 and continued into 1962.
Atmospheric testing included weapons that were dropped by airplanes, those detonated from
towers at heights ranging from 30 to 213 meters (m), tests conducted on land surfaces, and
tests in which helium balloons lofted weapons 137 to 457 m above ground.
It is estimated that for near-surface detonations, about 12 percent of the fission products were
distributed locally, with the remaining 10, 76, and 2 percent introduced into the troposphere,
stratosphere, and very high altitudes, respectively (UNS82). In addition, the large number of
neutrons released at the time of the detonation results in significant quantities of activation
products in the bomb's structural components, as well as ambient surface materials.
The primary radionuclides deposited locally were americium, plutonium, cobalt, cesium,
strontium, and europium. Based on the most recent estimates, about 20 Curies (Ci) remain in
surface soils at or near the original testing area(s) (MCA91).
Safety Tests. Between 1954 and 1963, more than 30 tests were conducted to investigate safety
issues regarding nuclear weapons in accident scenarios. The safety tests used mixtures of
Plutonium and uranium that were detonated using conventional explosives. These tests also
assessed the disposal and transport of these isotopes in the environment, including plant and
n-i
-------
animal uptake. In the 3,500 acres originally contaminated, the inventory of radionuclides is
estimated to be between 34 and 39 Ci.
The primary isotopes at test locations are plutonium, uranium, and americium, with lesser
amounts of cesium, strontium, and europium. Currently, these long-lived radionuclides are
contained in surficial soils and are relatively immobile. They are, however, potentially
available to be transported off-site by wind erosion.
Nuclear Rocket and Related Tests. Between 1959 and 1973, the Nuclear Rocket Development
Station area was used for a series of open air nuclear reactor, nuclear engine, and nuclear
furnace tests, and for the High Energy Neutron Reactions Experiment. The total estimated
inventory of soil contaminants that include strontium, cesium, cobalt, and europium has been
estimated to be one Ci (MCA91).
Waste Disposal Activities. Since the early to mid-1960s, NTS Areas 3 and 5 were established
for disposal of low-level waste from on-site and off-site DOE waste generators and include
landfill cells (pits and trenches) and greater confinement disposal (GCD) boreholes.
Approximately one-half of the buried waste represents atmospheric testing debris generated
during cleanup activities of above-ground nuclear test areas, with the remaining half from other
defense-related facilities.
Currently, NTS operates Areas 3 and 5 as a LLW repository and receives waste from on-site
activities and off-site defense generators. Approximately 500,000 Ci of low-level waste are
disposed hi shallow pits and trenches. Approximately 9.3 million Ci of high-specific-activity
waste containing primarily tritium have been disposed in GCDs. In addition, both areas also
contain smaller inventories of mixed waste.
Underground Testing. In August of 1963, the United States and the former Soviet Union
signed the Limited Test Ban Treaty (LTBT), which effectively banned weapons testing in the
atmosphere. Approximately 800 underground nuclear tests have been conducted that include
shallow borehole tests (< 60 m) and deep underground tests (about 600 m). Of the total
inventory (estimated to be 300 million Ci), about 112 million Ci are considered a potential
hydrological source term. About 90 percent (or 100 million Ci) of this radioactivity is
represented by tritium.
n-2
-------
Table II-1 provides a summary of residual radionuclide source terms at NTS for the
aforementioned activities.
Table II-l. Summary of Remaining Radioactivity on the NTS
f
Atmospheric &
Tower Tests
Safety Tests
Nuclear Rocket
Development
Station
Shallow Land
Disposal
Greater
Confinement
Disposal
Shallow
Borehole Tests
Deep
Underground
Tests
.!*'*', -- «
Aboveground
Nuclear Weapon
Proving Area
Aboveground
Experimental
Areas
Nuclear Rocket
Motor, Reactor,
and Furnace
Testing Area
Waste Disposal
Landfills
Monitored
Underground
Waste Disposal
Borehole
Underground
Nuclear Testing
Areas
Underground
Nuclear Testing
Areas
,"• -, % %
* <\ '•, %
•, "* f •""
Surficial Soils
&Test
Structures
Surficial Soils
Surficial Soils
Soils &
Alluvium
Soils &
Alluvium
Soils &
Alluvium
Soils,
Alluvium, &
Consolidated
Rock
^ JSOSDPJS ^
*s "* "*
Americium
Cesium
Cobalt
Europium
Strontium
Americium
Cesium
Cobalt
Plutonium
Strontium
Cesium
Strontium
Dry-packaged
low-level &
mixed wastes
Tritium
Americium
Americium
Cesium
Cobalt
Europium
Plutonium
Strontium
Tritium,
fission, and
activation
products
s
Above land
surface
At land
surface
Less than 10
feet
Less than 200
feet
60 meters
Less than 200
feet
" 600 meters
Remaining
-_ Inweatoxy ..
-20
-35
-1
-500,000*
" 9.3 million*
C 10,000
feet2)
' 2,000 at
land surface,
unknown at
depth
112 million1'
Inventory at time of disposal (not corrected for decay).
The 112 million Ci represents that fraction of the total underground source term (estimated to be 300
million Ci) which is within 100 m of the water table. It is this fraction that is available to the ground
water regime and is, therefore, referred to as the hydrological source term.
n-s
-------
Potential Impacts on Surrounding Populations
Population impacts from NTS activities are most effectively discussed in terms of activities that
have resulted in the introduction of radioactivity: 1) into the atmosphere, 2) into surficial soils,
and 3) at subsurface depths, where radioactivity could in the future become available to the
ground-water regime.
While maximum human exposure from atmospheric releases essentially coincided with peak
periods of past nuclear detonation, human exposures to radioactivity introduced below ground
is primarily a concern of the future.
Past Impacts Associated with Atmospheric Releases
In 1979, the Department of Energy (DOE) launched a major effort, called the Off-site
Radiation Exposure Review Project (ORERP). The principal objective of the ORERP was to
collect, organize, and analyze all relevant documents and data pertaining to fallout and
resultant exposure to off-site population groups in the vicinity of the NTS. Since that time,
more than 200,000 documents have been amassed and exposure estimates for discrete fallout
events have been derived from empirical measurements and computer-projection models
(ANS90).
External exposure estimates were originally published hi 1986 (ANS86) and updated in 1990
(ANS90). The total collective external exposure from 1951 through 1975 for all communities
was estimated to be 86,000 person-R, with the greatest exposures occurring in Saint George,
Utah; Ely, Nevada; and Las Vegas, Nevada. Summaries of the distribution of individual
cumulative external exposures are provided in Tables n-2 and II-3, which identify three
discrete time periods. By far, the largest collective and individual exposures occurred between
1951 and 1958. During the period from 1961 to the time of the Limited Test Ban Treaty, no
individuals are known to have received cumulative external exposures greater than 0.5 R. The
480 individuals who received exposures between 0.1 and 0.5 R lived hi small ranch
communities just north and northeast of the NTS. From 1963 to 1975, cumulative external
exposures were small, with only six individuals (at the Diablo Maintenance Station) receiving
more than 0.1 R.
H-4
-------
The contribution of dose resulting from inhalation and ingestion of radionuclides was not
considered in earlier exposure estimates. Investigators from the Desert Research Institute
(DRI), Colorado State University (CSU), and the Lawrence Livermore National Laboratory
(LLNL) are now systematically reconstructing the internal dose to individuals for all locations
and test events at the NTS. The computer code PATHWAY was developed to predict
radionuclide ingestion by residents in the arid regions around the NTS following radioactive
fallout deposition (WHC90). PATHWAY simulates the transport of approximately 21 fallout
radionuclides through agricultural ecosystems to humans and accounts for agricultural
conditions of the southwestern United States during the 1950s. Outputs can be generated that
are specific to age, sex, and radionuclides. For the inhalation pathway, estimates will be based
on empirical air sampling measurements, fallout data, and meteorologic records.
NTS Health Studies
Numerous population groups exposed to fallout from NTS weapon tests have been studied for
health effects. Those studied include civilian populations in the Utah - Nevada area and
military participants hi weapons testing. Most of these studies assessed the incidence of
leukemia and thyroid disorders among the exposed populations.
Table II-2.
Exposure Summary by Major Time Period of the Locations with Recorded
External Gamma Exposures, the Mean Location Exposure, and the Population
Weighted Exposure*
;^,-,' v t %^\,,v ^°$;i
.,/- ^ v x */^/> ;- '---^
& > :- * ,{ - vv\
Collective exposure (Person-R)
Number of locations with recorded exposure
Mean location exposure (R)
Population weighted exposure (R)
*&. J^*\/t'v ••"!.. **>'
\*v^t -:V^i
% < "&• *" -. v 3 ^ :• •*'" *•
^"W$tmim'«-
84,400
260
1.3
0.47
* "" *«•' ^" «. •.'"'* "•
£.; -•.Tteae|»i^oa^ ; V-
^4«at^fflnr:-
610
74
0.048
0.031
•.
^'LT5M*"^I9I»
320
72
0.017
0.002
* Source: ANS90
" Limited Test Ban Treaty signed August 5, 1963
n-s
-------
Table II-3. Distribution of Individual Cumulative External Gamma Exposure by Exposure
Range During the Three Major Time Periods
Bij«warea«stgq:^3i
< 0.01 to 0.1
0.01 to 0.5
0.5 to 1.0
1.0 to 5.0
5.0 to 10.0
10.0 to 15.0
Total
•" % "**
• jfttt&lft* 'r
61,000
80,000
19,000
20,000
520
45
180,000
•.
*.J*Qi TO T.y J.-itf J.
180,000
480
0
0
0
0
180,000
*rr«r to 19-75
180,000
6
0
0
0
0
180,000
Source: ANS90
The results of key leukemia and thyroid studies involving NTS population groups are
summarized below.
• A 1979 study reported an apparent twofold increase in the rate of leukemia
mortality among Utah residents born between 1951 and 1958 in "high exposure
counties" (LYO79). Some scientists view this finding with skepticism because
of a possible misinterpretation of the dose distribution and the paradox that the
rates of cancer at other anatomical sites were lower in "high exposure" areas
than those in "low exposure" areas.
• Machado et al. (MAC87) reported similar findings of an excess of childhood
leukemia deaths in three "high exposure" southwestern Utah counties among
individuals younger than 15 years of age who were bora before the tests ended.
These authors suggested the possibility that the transient increase of radiation-
induced childhood leukemias followed the peak fallout deposition between 1953
and 1957.
• Johnson (JOH87) identified radiation-induced cancers among Mormon families
in southwestern Utah exposed to fallout between 1951 and 1962 and venting of
underground nuclear detonations between 1962 and 1979. This study was found
to suffer from methodological deficiencies related to the selection of study
E-6
-------
subjects, the methods of obtaining medical information and cancer diagnosis,
and the interpretation of data (ICR91).
• Caldwell et al. (CAL83) reported an excess incidence of leukemia, but no
overall excess of other cancers, among the 3,224 military personnel who
participated in the 1952 Smokey nuclear test. Through 1977, nine cases of
leukemia had occurred, compared with 3.5 cases expected. The recorded
average external dose was 520 mrem. A similar study of 5,000 other
individuals who had participated in 24 detonations found no leukemia excess
(ROB83).
• Another population group studied since 1965 for thyroid disorders includes a
cohort of about 2,600 public school students who as infants lived hi proximity to
the Nevada Test Site in Utah and Nevada. The prevalence of thyroid
abnormalities in these children has been compared to that in a control group of
2,219 children selected from a county in Arizona that was presumed to have
received little or no fallout from the Nevada Test Site. Thyroid doses occurred
primarily as the result of ingesting milk contaminated with radioiodine.
Cumulative thyroid doses among study subjects were estimated to range from 30
to 700 rad (MAY66). Incidence of thyroid neoplasms was first reported in 1974
and 1975 (RAL74, RAL75). Although the rate of thyroid neoplasms among the
Utah/Nevada subjects of 5.6 per 1,000 was higher than that of Arizona control
subjects (3.3 per 1,000), the difference was statistically insignificant. In a
follow-up study conducted in 1985-1986, hi which 3,122 of the original 4,819
subjects were reevaluated, the rate of thyroid neoplasms hi the Utah/Nevada
subjects of 24.6 per 1,000 was again slightly but insignificantly higher than the
Arizona subjects (20.2 per 1,000) (RAL90). The authors previously concluded
that living near the Nevada Test Site in the 1950s had not resulted in a
statistically significant increase hi thyroid neoplasms among exposed subjects
when compared to control subjects of the same age and gender.
It is now generally accepted that a fundamental limitation hi all previous NTS studies was that
individual radiation exposures were uncertain or lacking because individual residence histories
for study subjects were unknown. In addition, reliable exposure rates for many locations were
not available at the time of the study.
In response to DOE's previously cited Off-site Radiation Exposure Review Project that
amassed exposure data on a county-by-county basis for all or part of seven western states, the
National Cancer Institute (NCI) sponsored two major studies to determine whether there were
any effects of fallout on the public near the NTS (WAC90).
H-7
-------
The first NCI-sponsored study was intended to examine whether leukemia in the state of Utah
was related to radiation fallout. Dose estimates for the Utah leukemia case-control study were
recently reported by Simon et al. (SIM95). The primary objective of the dosimetry task was to
estimate the total observed dose from all pathways to the active marrow by summing exposure
from each event at each location where the individual resided. External exposure from
radionuclides deposited on the ground presented by far the most significant dose contribution
to the active marrow.
The second NCI-sponsored study was a reevaluation of the earlier thyroid study. This study
reassessed exposures to the same cohort of subjects identified in the 1965-1970 study and
reexamined subjects for thyroid neoplasia. Results of this study were reported by Kerber et al.
(KER93) and more recently by Till et al. (TIL95). Their reassessment of the study cohort
demonstrated a statistically significant dose-response relationship between exposure to
radioiodines from open-air weapon tests at the NTS and the occurrence of thyroid neoplasms
(carcinomas and benign neoplasms). It should be noted, however, that the association was not
statistically significant for. thyroid carcinomas alone.
In summary, most of the airborne radioactivity released during the detonation to which
nearfield residents were exposed has been widely dispersed in the atmosphere, greatly diluted
hi the terrestrial biosphere, or decayed in the more than 30 years since the last atmospheric
test. Therefore, future human exposures from past atmospheric tests can be assumed
negligible.
Potential Future Exposures Associated with Current Soil Contaminants
The potential for significant future exposures to area residents of the NTS is limited to those
soil contaminants that in time may migrate down through the unsaturated zone and encounter
ground water that may subsequently be withdrawn for human use and consumption.
Radionuclide inventories residing in surficial or shallow strata are unlikely to reach an aquifer.
DOE considers only radionuclides from deep underground tests that were deposited beneath
the water table or within 100 m of the top of the water table as a potential hydrological source
term (DOE96).
H-8
-------
As previously noted, the hydrological source term available to the ground-water regime is
estimated to be 112 million Ci, of which about 100 million Ci is represented by tritium. There
is considerable uncertainly about the actual quantity of tritium that can enter the ground-water
regime. Uncertainties involve the extent to which radioactivity is securely trapped in the melt
glass matrix formed in the detonation cavity and the nearfield impact of the detonation on
ground permeability.
The shock wave and compressive forces from the tests can, on one hand, enhance permeability
by creating fractures nearby; on the other hand, these forces may decrease permeability by
closing pre-existing fractures.
Tritium, as water, is considered by far the most mobile radionuclide present in the subsurface
environment surrounding the underground test cavity. With its half-life of about 12 years, the
estimated 100 million Ci hydrologic source term of tritium represents the major radionuclide of
concern for the next 200 years.
Risks Associated with Tritium Migration
Proposed changes in NTS operations, as well as DOE's policy of reviewing sitewide impacts
under the National Environmental Policy Act (NEPA), have prompted the need for a new
Environmental Impact Statement (EIS) for the NTS (DOE93a). The draft EIS (DOE96),
issued hi January 1996, assessed doses and risks from past activities and future operations
under each of the following four alternatives:
• Alternative 1: No Action. The DOE would continue to support ongoing
program operations, but no new initiatives would be pursued.
• Alternative 2: Discontinue Operations. Under this option, only services
required to continue the protection of human health and safety would be
performed, inclusive of environmental monitoring.
• Alternative ?; Expanded Use. Implementation of this alternative would involve
expansion of many current activities and programs, including current
remediation and waste management activities.
• Alternative 4- Alternate Use of Withdrawn Land. While defense programs
would be discontinued, there would be increased activities for waste
n-9
-------
management, remediation, and nondefense research activities (e.g., solar
energy).
The proposed NTS EIS alternatives, however, are not expected to change the current inventory
or configuration of subsurface contamination. Thus, an assessment of future radiological
impacts to off-site residents is considered identical among the proposed alternatives. The
migration of tritium from discrete underground NTS test areas to locations outside the current
site boundary and accessible to members of the public are of primary concern and have been
evaluated hi the draft EIS.
Table II-4 provides summary data regarding doses and risks to hypothetical individuals.
Individuals are assumed to ingest contaminated well water for a period of 70 years from the
nearest accessible location. The 70-year lifetime exposure scenario coincides with the time of
peak concentrations of tritium in ground water for each of the three underground test sites:
• Yucca. Fist. Tritium concentrations migrating from Yucca Flat to Mercury,
Nevada are not expected to reach the minimum detectable level of one pCi/L.
Lifetime doses and risks are, therefore, negligible.
• Project Shoal Area. At the closest accessible location (the eastern boundary of
the Project Shoal Area), tritium is expected to reach a maximum concentration
of about 280 pCi/L in about 206 years, yielding a lifetime dose of 1.6 mrem.
At the nearest existing public well, maximum concentrations are not expected to
occur for 278 years, resulting hi doses and risks that are nearly four orders of
magnitude lower.
•• Central Nevada Test Area. At the nearest existing public well, the tune of
maximum tritium concentration is not expected for more than 400 years at
concentrations that are small fractions of one pCi/L. Associated doses and risks
at this location are essentially non-existent. Near the southern boundary, tritium
concentrations as high as 1.2 x 10s pCi/L had been predicted for 1983 (or 15
years after testing), yielding a lifetime dose of about 8,000 mrem, or an average
annual dose of 114 mrem. In 1996, these concentrations would be reduced by
more than a factor of two due to natural decay. However, there has been no
confirmation of these concentrations by ground water sampling and assessment
at this location.
H-10
-------
Radiological Surveillance Around the Nevada Test Site
Since 1970, the EPA's Characterization Research Division (formerly named the Environmental
Monitoring Systems Laboratory - Las Vegas or EMSL-LV) has assumed responsibility for the
Off-site Radiological Safety Program (ORSP) at NTS and other U.S. nuclear test sites. Among
ORSP's primary objectives are to systematically measure and document levels and trends of
environmental radiation and radioactive contaminants in the vicinity of the test sites.
Off-site levels of radiation and radioactivity are assessed by gamma-ray measurements using
highly -ensitive pressurized ion chambers (PICs) and thermoluminescent dosimeters (TLDs);
by sampling ah*, water, soil, milk, meats, food crops, and indigenous flora and fauna; and by
in-vivo/-vitro bioassays of off-site population groups. Results of these measurements are
collated and made available to the public in an annual report (DOE93b). Provided below is a
brief description of the major elements of the ORSP and summary data for 1993, the most
recent year of published data.
Table II-4. Doses and Health Risks to Exposed Individuals3 from Subsurface Radioactivity
Y < ' -
testtocatitts
, jv • N.; -
Yucca Flat
Project Shoal
Area0
Project Shoal
Area0
Central
Nevada Test
Area*
Central
Nevada Test
Area*
U^^F^
^tae&pte*?
\ "- ideals" "^
;'dnwakCdfliu#
^:.km>..?
100
206
278
15
410
f/?$^Ed&wa.
•. N * "Cis> >'•>
k^dfcM&H ,
< 1
280
<\
1.8x10*
« 1
^J *^i'
«xUteiii*$lQ!i6
^Xfoa»tnJ^ s "'
3.0 x 10'5
1.6x10*°
2.0 x 10-4
8.0xlO+3
l.SxlO'17
;-
:i&y&afe0atfcf
.•" ^ " Cancer
1.5 x 10'11
8.0 x lO'7
l.OxlO'10
4.0 x 10'3
9.0 x 10'24
The maximally exposed individual is a hypothetical person who is assumed to obtain drinking water from a well at the receptor
location for a lifetime of 70 years, centered around the time of peak tritium concentration in the well water.
Time period from the underground test date to the arrival of the peak tritium concentration in well water at the receptor's location.
Results based on analysis performed by Chapman et al. 1995 (CHA95).
No public well currently exists at these locations.
Results based on analysis performed by Pohlmann et al. 1995 (POH95).
n-n
-------
External Ambient Gamma Monitoring at the NTS
External ambient radiation levels are measured independently by a network of 27 pressurized
ion chambers and 127 thermoluminescent dosimeters located in various communities
surrounding the NTS (Figure H-l). Ambient dose and dose rates measured by these devices
represent the combined sources of cosmic and terrestrial radiation. Ambient air dose levels
ranged from 66 mR/yr at Pahrump, Nevada, to 166 mR/yr at Austin, Nevada, with an average
absorbed tissue dose value of 97 mrem/yr. Observed variations in ambient dose rates reflect
differences hi altitude, soil composition, and meteorological factors. This average of 97
mrem/yr is considerably higher than the combined national average value of cosmic and
terrestrial radiation level of 56 mrem.
Atmospheric Monitoring
A network of 30 continuously-operating stations monitor airborne paniculate radionuclides and
radioiodines. An additional 14 sampling stations sample for atmospheric tritium and noble
gases. Data indicate that airborne radioactivity from diffusion, evaporation of effluents, or
resuspension of radionuclides from past releases are currently below detection limits at off-site
locations.
Using the CAP88-PC model and NTS radionuclide emission data, an effective dose equivalent
of 0.004 mrem/yr was calculated for the off-site maximally exposed individual.
Monitoring of Local Food Products
A large variety of local foods that included milk, meat, vegetables, fruits, and wild game were
obtained from specified locations to distances of up to 200 miles. Food products were
analyzed for various radionuclides including H-3, Sr-89/90, and Pu-238/-239/-240. The Sr-90
levels in samples of animal bone remained very low, as did Pu-239/-240 in both bone and liver
samples of domestic and game animals. Although a few milk samples contained measurable
levels of Sr-90 and several fruit and produce samples contained measurable levels of Pu-239/-
240 and Sr-90, their potential contribution to human internal exposures was considered
insignificant.
H-12
-------
1 . .
I NEVADA FUTAH" """"""*•"
«
1
/<
i
PYRAMID
I
i
I
Austin
\
Ely
\\
Nvata
Storw Twin "uhaldatRn.
Cabin Ra
Tonopah* •
Raa». <**
-------
Population Monitoring by Bioassay
Since 1970, the ORSP has been assessing representative members of the off-site populations
for potential internal exposure from fallout. The off-site internal dosimetry program is
designed to measure radionuclide body burdens among persons who were subjected to fallout
during the early years of weapons testing, as well as to provide a monitoring system for
present-day NTS activities and environmental conditions.
In 1993, this program included 158 individuals representing 54 families. Evaluation of
participants includes a biannual whole-body count, lung count, and urinalysis. At 18-month
intervals, participants also receive a comprehensive medical examination.
No transuranics were detected in any lung counts. In general, body burdens of participants
were representative of any normal population when matched for age and sex distribution.
Ground-Water and Long-Term Hydrological Monitoring
Since 1972, a Long-Term Hydrological Monitoring Program (LTHMP) has been implemented
at the NTS. Routine monitoring is conducted at specified on-site wells and at wells, springs,
and surface waters in the off-site area around the NTS.
Because tritium is a product of nuclear testing that was found in significant quantities in
underground test cavities and is highly mobile, it is expected to be the first radionuclide to
migrate. Therefore, tritium serves as a warning indicator of other potential radionuclide
migration and was the primary radionuclide analyzed hi the LTHMP. Off-site sampling
locations include 23 wells, seven springs, and two surface water sites, which are sampled on a
monthly basis.
In 1993 and over the past decade, detectable levels of tritium have been found in a limited
number of samples obtained from surface water. In all cases, the tritium activities fall within
the range of environmental levels and are thought to be the result of rainfall containing
scavenged atmospheric tritium.
H-14
-------
Summary of ORSP Results and Off-site Dose Estimates Pertaining to NTS Activities
For 1993, EPA's comprehensive off-site environmental surveillance program around NTS
measured no levels of radiation that would contribute significant exposure to any member of
the public.
Potential exposure from all pathways to members of the public due to NTS activities are
estimated annually by two separate methods. The first calculates annual dose by means of
computer effluent modeling (CAP88-PC), meteorologic, and demographic data. The second
approach uses measurement data from the ORSP with conservative assumptions and standard
dose conversion factors.
Based on computer modeling, the committed effective dose equivalent to the maximally
exposed off-site resident for 1993 was estimated to be 0.004 mrem. Environmental sampling
data estimate a comparable dose of 0.05 mrem/yr from NTS and non-NTS fallout. For the 80-
km (50-mile) radius population of 21,750 individuals, a collective population dose of
1.2 x 10"2 person-rem was estimated for 1993. These doses are considered negligible when
compared to the average ambient external gamma dose rate of 97 mrem/yr contributed by
natural cosmic and terrestrial radiation alone.
H-15
-------
APPENDIX m
SOIL TYPES FOUND IN THE YUCCA MOUNTAIN AREA
-------
Rock outcrop
LJpsprlno-Rubblelond
Rock outcrop
St. Thomas
SonwflH gravelly
fine sondy toom
Zaido-Greyeagte
Upsprtng
Cave gravelly
sandy loam
Shomock gravelly
fine sandy loom
'o'd Arlzo gravelly
-0 g sondy loam
Bkjepolnt loamy
fine sand
•"] Sonwell-SonweH
J worm-Yermo
Nopoh-Woda
Guliled land
Kilometers
Miles
Valley Soil Classification
Nye County, Nevada
ey of Nye Counttj. Nevada. Southwovt Part, bosed on US Geological Survey Orthophotoquods. Ju
-------
APPENDIX IV
WELL DRILLING AND PUMPING COSTS
-------
APPENDIX IV
WELL DRILLING AND PUMPING COSTS
Drilling Costs
The cost of drilling and finishing a well depends on a number of factors, including the yield of
water that is required, the geologic media through which the drilling is made, the depth to the
water table, the yield of the aquifer, and whether or not the borehole needs to be cased to provide
long-term stability.
Estimates for drilling and completing five wells typical of those that are currently in use or might
be envisioned in the area under consideration are given in WI96 and reproduced (with minor
corrections) in Tables IV-1 through IV-5. Wells 1 and 2 are intended to show the costs that
would be incurred in duplicating the two production wells (J13 and J12, respectively) that DOE
currently maintains on Jackass Flats. Both are presumed to be drilled in the tuff aquifer and, like
the two actual wells, the boreholes are fully cased and screened. (Note: The actual construction
of Well J-13 was accomplished by using a telescoping diameter to overcome the caving
difficulties that were encountered while drilling (WI96)). Well 3 represents an agricultural well
drilled in the alluvial aquifer, capable of providing sufficient water for a 1/4 section, center-pivot
irrigation system. Well 4 represents a typical well for domestic use, sized to provide about 10
gallons of water per minute to satisfy the needs of one or two dwellings. Like Well 3, it is drilled
into the alluvial deposits that comprise the alluvial aquifer. Both Wells 3 and 4 are fully cased
and screened to provide stability. Finally, Well 5 is an uncased well drilled in welded and
bedded tuff. At its design capacity of two gallons per minute, it would be sufficient to provide
water to stock. Only the first 150 feet of Well 5 is cased.
On a $/foot basis, these five wells range from $97 to more than $500 to drill and finish, with an
average cost of about $300/foot. An estimate derived from the costs cited by BCI Geonetics for
the development of a deep well field hi a remote desert environment indicates a cost of about
$165/foot (BCI85). As few details are given in either reference for the rationale used in sizing
the boreholes, the length of screens needed, or the bases of the unit costs used, these estimates
should only be considered as very approximate values. Both references do include costs for
extensive logging and testing of the wells.
IV-2
-------
Given the sparsity of supporting data in the two references, it is difficult to evaluate their
reasonableness. In particular, the cost of $524/foot for Well 3, designed for irrigation use, is
difficult to reconcile with the estimate of $40/foot quoted by a driller who has actual experience
in providing well-drilling services in Amargosa Valley (D096). Likewise, the estimated cost of
$97/foot for a domestic well is difficult to reconcile with costs of about $15/foot currently
charged for drilling residential wells in a mountainous area of Colorado (GO96).
However, some insight into the significance of drilling costs on the overall cost of water can be
derived by estimating the costs of various wells (different uses and depths) from the data
available and then calculating the capital cost per acre-foot. The well costs shown in Table IV-6
are for a private domestic well sized to provide sufficient water for about 10 persons (4
acre-feet/yr), a communal well sized to serve the domestic needs of about 300 persons (120
acre-feet/yr), and a large-scale agricultural well sized to provide irrigation for a 1/4 section (625
acre-feet/yr). Costs are computed for water depths of 100 feet, 300 feet, 600 feet, 900 feet, and
1,200 feet. In all instances, the depth of the wells are assumed to be 200 feet deeper than the
water depth.
Unit drilling costs ($/foot) for the wells are based on $15/foot for a 4" diameter casing and scaled
to the area of the casings using a 0.7 power function (e.g., an 8" casing has an area four times
that of a 4" casing; thus the scaling factor is 4A°7 = 2.64 and cost = $40/ft.). Unit costs ($/foot) of
drilling wells in the tuff aquifer were increased by 25 percent, to reflect the greater difficulty this
media presents when compared to drilling in the alluvial aquifer. Pump costs are based on
$8,000 for a five horsepower pump and scaling functions of 0.7 for pumps less than 20 horse-
power, 0.6 for pumps between 20 and 100 horsepower, and 0.5 for pumps greater than 100
horsepower. The costs per acre-foot were computed by amortizing the costs shown over a 30
year period at seven percent interest and dividing by the yields of the wells in acre-feet per year.
Adapted from WI96 IV-3
-------
Table IV-1. Well 1 - 3,385' Yielding 700 gpm w/ static head of 1,000 ft
Modeled on DOE Well J-13
Borehole Depth
Well Depth
Borehole Diameter
Casing Diameter
Screen Length
Item
Install 30" Conductor Casing
Drill Pilot Hole
E-log
Ream Pilot Hole to 26"
Caliper Log
Install Blank Casing
Install Screen
Install Gravel Pack
Gravel Tube
Grout Seal
Plumb & Alignment Test
Surge/Airlift Development
Pumping Development
Step Test
Constant Q Test
Pump Cost
Install Pump
Electric & Wellhead Finish
Total Cost
Cost per Foot
3,450 ft
3,385 ft
26 in
14 in
2,162ft
Quantity Units
50ft
3,450 ft
1 ea
3,450 ft
1 ea
1,223 ft
2,162 ft
2,515 ft
990ft
985ft
1 ea
24 hr
24 hr
10 hr
40 hr
1 ea
1 ea
lea
Unit
Cost ($)
175.00
45.00
7,000.00
60.00
4,000.00
120.00
160.00
45.00
6.00
55.00
5,500.00
275.00
150.00
150.00
150.00
20,000.00
6,500.00
20,000.00
II
Total
Cost ($)
8,750.00
155,250.00
7,000.00
207,000.00
4,000.00
146,760.00
345,920.00
113,175.00
5,940.00
54,175.00
5,500.00
6,600.00
3,600.00
1,500.00
6,000.00
20,000.00
6,500.00
20,000.00
1,117,670.00
330.00
Adapted from WI96
IV-4
-------
Table IV-2. Well 2 - 887' Yielding 800 gpm w/ static head of 800 ft
Modeled on DOE Well J-12
900ft
887ft
22 in
12.75 in
75ft
Borehole Depth
Well Depth
Borehole Diameter
Casing Diameter
Screen Length
Item
Install 22" Conductor Casing
Drill Pilot Hole
E-log
Ream Pilot Hole to 22"
Caliper Log
Install Blank Casing
Install Screen
Install Gravel Pack
Gravel Tube
Grout Seal
Plumb & Alignment Test
Surge/Airlift Development
Pumping Development
Step Test
Constant Q Test
Pump Cost
Install Pump
Electric & Wellhead Finish
Total Cost
Cost per Foot
Quantity Units
50ft
900ft
1 ea
900ft
1 ea
812ft
75ft
117ft
125ft
783ft
1 ea
24 hr
24 hr
lOhr
40 hr
1 ea
lea
1 ea
Unit
Cost ($)
125.00
40.00
4,000.00
50.00
2,000.00
55.00
75.00
25.00
6.00
45.00
2,500.00
275.00
150.00
150.00
150.00
20,000.00
6,500.00
20,000.00
=====
Total
Cost ($)
6,250.00
36,000.00
4,000.00
45,000.00
2,000.00
44,660.00
5,625.00
2,925.00
750.00
35,235.00
2,500.00
6,600.00
3,600.00
1,500.00
6,000.00
20,000.00
6,500.00
20,000.00
249,145.00
280.00
Adapted from WI96
IV-5
-------
Table IV-3. Well 3 - 320' Yielding -2,400 gpm w/ static head of 150 ft
Modeled on typical 1/4 section irrigation well
Borehole Depth
Well Depth
Borehole Diameter
Casing Diameter
Screen Length
Item
Install 30" Conductor Casing
Drill Pilot Hole
E-log
Ream Pilot Hole to 28"
Caliper Log
Install Blank Casing
Install Screen
Install Gravel Pack
Gravel Tube
Grout Seal
Plumb & Alignment Test
Surge/Airlift Development
Pumping Development
Step Test
Constant Q Test
Pump Cost
Install Pump
Electric & Wellhead Finish
Total Cost
Cost per Foot
!^^EHa^S55S^=5SSS=H5^^5;
320ft
320ft
28 in
16 in
150ft
Quantity Units
50ft
320ft
1 ea
320ft
lea
175ft
150ft
180ft
145ft
140ft
lea
24hr
24hr
lOhr
40 hr
lea
1 ea
lea
=====
Unit
Cost ($)
175.00
40.00
3,000.00
50.00
2,000.00
65.00
85.00
35.00
6.00
55.00
2,500.00
275.00
150.00
150.00
150.00
40,000.00
6,000.00
20,000.00
=^== ._.——— _
============
Total
Cost ($)
8,750.00
12,800.00
3,000.00
16,000.00
2,000.00
11,375.00
12,750.00
6,300.00
870.00
7,700.00
2,500.00
6,600.00
3,600.00
1,500.00
6,000.00
' 40,000.00
6,000.00
20,000.00
167,745.00
524.00
Adapted from WI96
IV-6
-------
Table IV-4. Well 4 - 600' Yielding 10 gpm w/ static head of 300 ft
Modeled on typical residential well
Borehole Depth
Well Depth
Borehole Diameter
Casing Diameter
Screen Length
Item
Install 16" Conductor Casing
Drill Pilot Hole
E-log
Ream Pilot Hole to 19"
Caliper Log
Install Blank Casing
Install Screen
Install Gravel Pack
Gravel Tube
Grout Seal
Plumb & Alignment Test
Surge/Airlift Development
Pumping Development
Step Test
Constant Q Test
Pump Cost
Install Pump
Electric & Wellhead Finish
Total Cost
Cost per Foot
600ft
600ft
19 in
8 in
200ft
Quantity Units
50ft
600ft
lea
600ft
1 ea
400ft
200ft
260ft
345ft
340ft
1 ea
24 hr
24 hr
lOhr
40 hr
lea
lea
1 ea
Unit
Cost ($)
100.00
35.00
3,000.00
45.00
2,000.00
41.00
60.00
20.00
6.00
40.00
2,500.00
275.00
150.00
150.00
150.00
8,000.00
6,000.00
20,000.00
Total
Cost ($)
5,000.00
21,000.00
3,000.00
27,000.00
2,000.00
16,400.00
12,000.00
5,200.00
2,070.00
13,600.00
2,500.00
6,600.00
3,600.00
1,500.00
6,000.00
8,000.00
6,000.00
20,000.00
161,470.00
269.00
Adapted from WI96
IV-7
-------
Table IV-5. Well 5 -1,500' Yielding 2 gpm w/ static head of 1,000 ft
Modeled on simple stock water well
Borehole Depth
Well Depth
Borehole Diameter
Casing Diameter
Screen Length
Item
Install 16" Conductor Casing
Drill Pilot Hole
E-log
Ream Pilot Hole
Caliper Log
Install Blank Casing
Install Screen
Install Gravel Pack
Gravel Tube
Grout Seal
Plumb & Alignment Test
Surge/Airlift Development
Pumping Development
Step Test
Constant Q Test
Pump Cost
Install Pump
Electric & Wellhead Finish
Total Cost
Cost per Foot
1,500ft
1,500ft
Sin
NAin
NAft
Quantity Units
50ft
1,500ft
1 ea
Oft
1 ea
150ft
Oft
Oft
Oft
Oft
Oea
24 hr
24 hr
lOhr
40 hr
1 ea
1 ea
1 ea
Unit
Cost($)
100.00
45.00
5,000.00
NA
3,000.00
41.00
NA
NA
NA
NA
NA
275.00
150.00
150.00
150.00
15,000.00
6,000.00
20,000.00
Total
Cost($)
5,000.00
67,500.00
5,000.00
0.00
3,000.00
6,150.00
0.00
0.00
0.00
0.00
0.00
6,600.00
3,600.00
1,500.00
6,000.00
15,000.00
6,000.00
20,000.00
145,350.00
97.00
Adapted from WI96
IV-8
-------
Table IV-6: Estimated Capital Costs of Wells - Yucca Mountain Area
Private Wells - Domestic Use
10 gpm/4 acre-feet per year
100'Depth to Water
Alluvial Aquifer
4" Casing
1.25 HP Pump
Drilling Costs i
Pump Cost
Total Cost
Cost/acre-foot
$15/foot=
$4,500
3,000
7,500
$150
300' Depth to Water
Alluvial Aquifer
8" Casing
5 HP Pump
Drilling Costs @ $40/foot= $20,000
Pump Cost 8,000
Total Cost 28,000
Cost/acre-foot
$565
600' Depth to Water
Tuff Aquifer
8" Casing
10 HP Pump
Drilling Costs @ $50/foot= $40,000
Pump Cost 13,000
Total Cost 53,000
900' Depth to Water
Tuff Aquifer
8" Casing
15 HP Pump
Drilling Costs @ $50/foot= $55,000
Pump Cost 17,000
Total Cost 72,000
Cost/acre-foot $1,070
1,200'Depth to Water
Tuff Aquifer
8" Casing
20 HP Pump
Drilling Costs @ $50/foot=$70,000
Pump Cost 21,000
Total Cost 91,000
Cost/acre-foot
$1,450
Cost/acre-foot
$1,830
CAUTION: Values shown in this table are deemed reliable only for the purposes of this report and only within the
contexts used or implied by the calculational methodologies, assumptions, and data sources presented in the text.
IV-9
-------
Table IV-6: Estimated Capital Costs of Wells - Yucca Mountain Area - continued
Communal Wells
700 gpm/120 acre-feet per year
100'Depth to Water
Alluvial Aquifer
14" Casing
15 HP Pump
Drilling Costs @ $85/foot= $25,500
Pump Cost 15,000
Total Cost 40,500
Cost/acre-foot $30
600' Depth to Water
Tuff Aquifer
14" Casing
100 HP Pump
Drilling Costs @ $105/foot=$ 84,000
Pump Cost 36,000
Total Cost 120,000
300' Depth to Water
Alluvial Aquifer
14" Casing
50 HP Pump
Drilling Costs @ $85/foot=$42,500
Pump Cost 24,000
Total Cost 66,500
Cost/acre-foot $45
900' Depth to Water
Tuff Aquifer
14" Casing
150 HP Pump
Drilling Costs @ $105/foot=$115,500
Pump Cost 46,000
Total Cost 161,500
Cost/acre-foot $80
1,200'Depth to Water
Tuff Aquifer
14" Casing
200 HP Pump
Drilling Costs @ $105/foot=$ 147,000
Pump Cost 55,000
Total Cost 202,000
Cost/acre-foot
$110
Cost/acre-foot
$135
CAUTION: Values shown in this table are deemed reliable only for the purposes of this report and only within the
contexts used or implied by the calculational methodologies, assumptions, and data sources presented in the text.
IV-10
-------
Table IV-6: Estimated Capital Costs of Wells - Yucca Mountain Area - continued
Irrigation Wells
2,400 gpm/625 acre-feet per year
100' Depth to Water 300' Depth to Water
Alluvial Aquifer Alluvial Aquifer
16" Casing 16" Casing
60 HP Pump 200 HP Pump
Drilling Costs @ $100/foot= $30,000 Drilling Costs @ $100/foot= 50,000
Pump Cost ' 40,000 Pump Cost . 55,000
Total Cost 70,000 Total Cost 105,000
Cost/acre-foot $9 Cost/acre-foot $14
600' Depth to Water 900' Depth to Water
Tuff Aquifer Tuff Aquifer
16" Casing 16" Casing
400 HP Pump 600 HP Pump
Drilling Costs @ $125/foot=$ 100,000 Drilling Costs @ $125/foot=$137,500
Pump Cost 78,000 Pump Cost 95,000
Total Cost 178,000 Total Cost 232,500
Cost/acre-foot $23 Cost/acre-foot $30
1,200'Depth to Water
Tuff Aquifer
16" Casing
800 HP Pump
Drilling Costs @ $125/foot=$ 175,000
Pump Cost 110,000
Total Cost 285,000
Cost/acre-foot $37
CAUTION: Values shown in this table are deemed reliable only for the purposes of this report and only within the
ontexts used or implied by the calculational methodologies, assumptions, and data sources presented hi the text.
IV-11
-------
Pumping Costs
Pumping costs for water can be computed based on the following equation:
CAT = (K • L • PKwh)/E
where:
CAP = cost Per acre-foot ($)
K = number of kilowatts hours need to lift one acre-foot of
water one foot at 100 percent pump efficiency
L = lift of water (feet)
E = overall pumping efficiency
PKwh = price per kilowatt hour of electricity ($)
Based on the density of water of one gram per cubic centimeter (g/cm3), K can be calculated to be
about 1.024. The lift of water, which is specific for any given well, is simply the depth to the
water from the land elevation at the head of the well. Overall pumping efficiency depends on
both the efficiency of the pump and frictional losses that depend on the diameter of the well
casing and the lift of the well. Current pumps range from 65 to 80 percent efficient. The
following calculation assumes an overall efficiency of 70 percent. The cost of electricity in the
Amargosa Valley is currently about $0.05 per kilowatt hour, which is significantly lower than the
national average. Using these values, the cost of a one foot lift of an acre-foot of water is $0.073.
In Table IV-7, this is rounded to $0.075/foot of lift.
Using a marginal value of $800 per acre-foot for domestic use water and a marginal value of $40
per acre-foot for irrigation use water as guides, the capital costs of private wells for domestic use
become prohibitive at depths between 300 and 600 feet. For communal domestic use and
irrigation use, the capital costs do not become prohibitive even at depths of 1,200 feet. However,
the capital costs do not include pumping costs or any of the costs for distribution facilities that
would be needed for a community water supply or an irrigation system.
Pumping costs are not likely to be sufficient to affect the observations made on the economic
feasibility of wells for private or communal domestic use. Indeed, the wells at Lathrop Wells
and on Jackass Flats demonstrate that private and communal water systems have already been
deemed economically justified in those locations. This conclusion is consistent with the finding
IV-12
-------
made by the Center for Nuclear Waste Regulatory Analyses in their study of the use of ground
water in arid and semi-arid parts of the United States (WI96). They state "[t]he well water data
suggest that water use practice in the immediate vicinity of YM [Yucca Mountain] may have
included a small cluster of homes supplied by one or more small-diameter, low discharge,
high-lift wells or a community or suburb supplied by wells similar in construction to J-13 had the
land not been withdrawn by the Federal government."
However, because of the very large volumes of water needed for irrigating field crops,
particularly in the climate of Amargosa Valley, pumping costs are very significant for such
agricultural applications. Combining the pumping cost estimates in Table IV-7 with the capital
cost estimates in Table IV-6, the marginal value of water for irrigation is exceeded at depths to
water greater than 300 feet. In fact, since these estimates do not consider the distribution costs
for the irrigation system or any maintenance costs over the 30-year amortization period, it is not
surprising to see that commercial agricultural activities in Amargosa Valley have been restricted
thus far to areas where the depth to water is generally less than about 200 feet.
Table IV-7: Calculated Pumping Costs for Various Lifts
100' Lift = $ 7.50/acre-foot
300' Lift = $22.50/acre-foot
600' Lift = $45.00/acre-foot
900' Lift = $67.50/acre-foot
1,200' Lift = $90.00/acre-foot
IV-13
-------
APPENDIX V
NEW AND UNUSUAL FARMING PRACTICES
-------
APPENDIX V
NEW AND UNUSUAL FARMING PRACTICES
Ostrich Farming
Ostrich Facts. The ostrich, Struthis camelus, is the largest living bird. It may stand up to ten
feet tall and attain a weight in excess of 300 pounds. Although flightless, the bird can run at
sustained speeds of 30 mph and peak speeds of 45 mph when escaping predators. Its body is
covered with large loosely structured feathers whose function is to protect the bird from heat
and cold (WAL80).
Today, its natural range is limited to Africa, where it lives in open country feeding on plants,
fruits, grasses, leaves, and on occasion, insects, lizards, rodents, and small birds. The bird
can go for long periods without water, which allows it to survive in arid regions.
Domestic Farming
Ostriches may reach full size in as few as six months but do not attain sexual maturity until
about three years of age. They may live for 30 years or more. Females may lay several eggs
weighing up to 3.5 pounds each with an incubation period of 42-43 days.
The first commercial ostrich farm was established in South Africa hi 1838. In recent years, a
limited number of farmers in the United States have begun to raise ostriches for their meat,
leather, feathers, and other byproducts. At the typical age of slaughter (i.e., 10-14 months),
the average bird yields about 75-90 pounds of consumable meat, 12-14 square feet of leather,
and 3-4 pounds of feathers. The high protein red meat has a taste much like that of beef but
has a fat and cholesterol content that is even lower than turkey. Ostrich leather is regarded as
among the best and most durable of leathers. Ostrich feathers have a wide range of
commercial uses (AOA96).
For practical and economic reasons, ostrich farmers generally do not allow birds to
forage/graze naturally for food but restrict them to a confined area where they are provided
pelletized commercial feed, similar to that used in poultry farming. Adult ostriches eat about
3-4 pounds of commercial pelletized feed per day.
V-l
-------
Human Exposure
The principal pathway for human exposure associated with ostrich fanning is the consumption
of ostrich meat. Traditionally, for meat consumption such as beef, human exposure is the
result of two mechanisms: (1) soil-crop-animal-human, and (2) water-animal-human. Thus,
the activity of a given radionuclide hi meat that may be consumed by humans is determined by
the following generic equation:
Activity in Meat (pCi/kg) = [(Feed consumed-kg/d)(Activity in feed-pCi/kg) +
(water consumed-L/d)(Activity in water-pCi/L)] x transfer coefficient (d/kg)
To determine the activity in beef, for example, it is generally assumed that the adult cow
consumes about nine kg of dry feed and 50 liters of water per day.
Because ostriches are assumed to be raised on commercial, pelletized feed that is unaffected by
contaminated ground water in the vicinity of the Yucca Mountain Site, contamination of ostrich
meat is limited to the bird's consumption of contaminated water. For ostrich meat derived
from commercial fanning, the above equation is, therefore, reduced to:
Activity (pCi/kg)ostrich = (water consumed-L/d)(Activity in water-pCi/L)(transfer coefficient-d/kg)
Information regarding water consumption rates by ostriches was provided by Bud Aldrich,
DVM (ALD96). Dr. Aldrich, a veterinarian, is not only affiliated with the American Ostrich
Association, Fort Worth, TX, but raises ostriches personally. Based on knowledge and
personal observation, he stated that water consumption is highly variable and reflects ambient
temperature and the water content of feed. Low water consumption would be expected for
temperate climates and feed consisting of succulent vegetation. Conversely, high water
consumption rates would be expected for areas with high temperatures/low humidity and feed
with low water content (e.g., pelletized feed). Based on the fact that ostriches, at time of
slaughter, are between 10 and 14 months of age, they will experience temperatures that reflect
seasonal changes at Yucca Mountain for a full calendar year. On average, the daily water
consumption for an adult ostrich is estimated at 12 liters (ALD96).
A review of the scientific literature reveals a total absence of data regarding the uptake and
retention of ingested elements in the ostrich from which radionuclide-specific transfer
V-2
-------
coefficients are derived. This is not surprising since historically the ostrich has not posed a
significant link in the food chain leading to human exposure. In spite of acknowledged
anatomical and physiological features that are unique to the ostrich (e.g., it is the only bird that
eliminates its urine separately from its feces), its metabolism of food products is generally
considered equal or similar to that of the chicken, turkey, and other domestic poultry
(ALD96). Radionuclide transfer coefficients derived for poultry have, therefore, been applied
to the ostrich in deriving meat activity levels for drinking water contaminated at one pCi/L and
compared to those of beef (Table V-l). On a per unit weight basis, seven of the 19
radionuclides assessed in Table V-l are estimated to be present hi ostrich meat at a higher level
than those of beef (conversely beef is estimated to exhibit higher levels for 12 of the 19
radionuclides).
i
In summary, ostrich farming and the substitution of ostrich meat for beef is not likely to have a
significant impact on dose/risk estimates.
Catfish Farming
The growing interest in aquaculture and its expansion into various geographic areas, including
the desert Southwest, is due to several factors:
• Overfishing and environmental factors have steadily reduced harvests of marine fish.
• There is an increased demand for fish that is influenced by population growth and
dietary concerns regarding animal fats/cholesterol.
• Aquaculture is based on proven methods and has the support of an established
infrastructure (i.e., how-to information, equipment, fish feed, processing, and
wholesale/retail outlets).
Aquaculture is currently the most lucrative sector within U.S. agriculture (GEL94).
Aquaculture Facts. Data regarding aquaculture were obtained by personal communication
from Arid Lands Fish Production, located hi Chino Valley, Arizona. Aquaculture facilities
can be characterized as either "warm" or "cold" water operations. Cold water fish farms are
generally located in areas that are suitable for maintaining water temperatures below 60°F and
principally involve various species of trout.
V-3
-------
Warm water facilities in southern states and in desert regions of the southwest take advantage
of a climate that allows for water temperatures above 70°F. For example, Arizona fish
farmers currently produce about 500,000 pounds of fish per year. Fish farming in Arizona
consists mainly of family operations. Warm water fish include catfish, large-mouth bass, and
tilapia. Warm climates favor fish farming due to the fact that fish metabolism (and therefore
growth rate) and increases with ambient water temperature.
Table V-l. Comparative Data Pertaining to Ostrich Farming
Radionuclides
U-234
U-238
Th-230
Tc-99
Se-79
Ra-226
Pu-239
Pu-240
Pb-210
Np-237
Ni-59
Nb-94
1-129
Cs-137
Cs-135
Cm-246
Cm-245
Am-243
Am-241
Ostrich Data
Transfer Coeff.
<
-------
Farming methods vary depending on the availability and cost of water. In southern states
where water can be readily diverted from proximal bodies of surface water, fish production
commonly employs rectangular, levee-style ponds constructed on flat land that are similar to
cranberry bogs. Alternatively, natural depressions in the valleys of hilly terrain may be used
for the construction of watershed ponds that rely on runoffs from rain as their primary water
source.
For arid areas that lack available surface water and have limited ground water resources, fish
fanning is generally conducted in large tanks filled with ground water that is continuously
filtered and aerated.
Independent of whether fish are raised hi levee-style ponds, natural depression ponds, or in
tanks, then- food is limited to commercial pelletized floating feed that is introduced daily.
Catfish are generally harvested at around 200 days when they attain a body weight of about
one pound. At time of harvest, the catfish will have consumed about two pounds of feed
yielding a feed to body weight ratio of two.
Like other terrestrial human food chains, the aquatic food chain also consists of multiple
trophic levels. Trophic levels represent individual steps hi the food chain and for aquatic
systems are generally more complex than those of the terrestrial world. This is due to the fact
that aquatic species often consume several types of prey that represent different trophic levels
(Figure V-l). To further complicate matters, physicochemical parameters of radionuclides
and, thus, their transfer from one organism to another are generally more variable hi aquatic
ecosystems than those in terrestrial ecosystems. Important parameters that affect a
radionuclide's distribution hi aquatic ecosystems include its tendency for colloid formation, co-
precipitation, and adsorption-desorption on sediments and suspended solids (NRC83).
Over the past several decades, a significant number of studies have been conducted hi which
radionuclides have been introduced into the water medium of a natural ecosystem or under
controlled laboratory conditions (FRE67, LLL68, ORN76, LLL78, BLA82, PNL86 and
POS88).
V-5
-------
f>; «an (3, 4, S, 5)
(2)
seed plents (l)
Hutrtents in solution
Nutrients tn substrate
Figure V-l. A Simplified Lake Ecosystem (parenthesized numbers note the trophic level)
V-6
-------
A common and primary objective of these studies was the determination of concentration
factors of radionuclides that pose environmental risks. The concentration factor (L/kg) for
aquatic species is the ratio of a given element/radionuclide concentration hi the organism to
that in water.
Radionuclides with high concentration factors are those with established biological significance
or chemical similarity to biologically active elements. Biological significance, however, varies
among species within and among different trophic levels. While the majority of radionuclides
in the environment do not increase in concentration with trophic level, a limited number of
highly soluble radionuclides with mineral nutrient value do in fact increase with trophic level
(PEN65). For example, the concentration of plutonium and strontium generally decreases at
higher trophic levels due to decreased efficiencies in assimilating the ingested radionuclides
(Figure V-2). In contrast, measurements of the concentration of cesium-137 in freshwater fish
show that larger predacious fishes tend to have a markedly higher cesium concentration than
smaller fish (lower trophic level), zooplankton, algae, and other components of the aquatic
food web inclusive of waterfowl, raptor, and terrestrial species (Figure V-3). Concentration
factors are also high among fishes like the catfish that are bottom-feeders, since cesium has a
strong affinity for clay-containing sediments/mud.
In an extensive review of the scientific literature, the International Atomic Energy Agency
recently compiled bioaccumulation factors for the edible portions of freshwater fish (IAE94).
Table V-2 summarizes the range of reported values and cites a best value for specific elements
and their associated long-lived radionuclides.
The applicability of concentration factors cited hi Table V-2 to catfish farming, however, is
highly dubious for the following reasons:
(1) Derived concentration factors generally represent ecological conditions in which the
. radionuclide was assumed to exist in a steady-state or equilibrium condition in all
trophic levels and compartments that define the ecosystem.
V-7
-------
«xu TROUT.
CMWOO* SALMON
COMO SALMON1
SMELT
ALlWIfE
ILOAUR
MIXED
t'OO
ALGAEl
Pu-239/-240
iiiuij—i 111uii)—i i .mil!—i i mm
TT
Cs-137
"•'4 i '
Sr-90
••' ' ' "'nil < i .mill 1 I mit'l 1,1 H1"ll ' ' '»•
10"
I04
f AESMWATf A CONCCNTRATION FACTO*
•o'
Figure V-2. The Concentration Factors for Pu-239/240, Cs-137, and Sr-90
in a Freshwater Ecosystem (WAH75)
MAN
1100% upufctl
MEATS MILK WATERfOWt KESH SUNMSM FLESH FROG MUSCLE CAW MUSCLE
9-IZ* «1 WOO-w. 6000 __ SOOO «000 »00
ortl oou
« oru oou y
X.1 //
wtLC CAME LIVESTOCK
bnoou. dMf.
fMMUtC «C->
SUBMERGED SEED PLANTS
EMERGENT SEED HANTS 400-1000
50-600
DETRITUS AND MUO
10.000
t w« coMMtwtiion tocMrc
Figure V-3. A Freshwater Food Web Illustrating the Pathway to Human for
Cesium-137 in Aquatic Environment (PEN58)
V-8
-------
Table V-2. Concentration Factors for Edible Portions of Freshwater Fish (L/kg)
Element
H-3
He
Be
C
N
0
Na
P
S
Sc
Cr
Mn
Fe
Co
Ni
Cu
Zn
Br
Rb
Sr
Y
Zr
Mb
Mo
Tc
Ru
Rh
Ag
Sn
Sb
Te
Recommended Value
1
1
IxlO2
5xl04
2X105
1
2x10'
5xl04
8xl02
IxlO2
2xl02
4xl02
2xl02
3X102
IxlO2
2xl02
IxlO3
4xl02
2X103
6x10'
SxlO1
3 x 102
3X102
10
2x10'
10
10
5
3xl03
IxlO2
4X102
Range
6 x 10-1 - 1
5 x 103 - 5 x 104
2 x 101 - 1 x 102
3 x 103 - 1 x 10s
2 - 1 x 102
4 x 101 - 2 x 103
5 x 101 - 2 x 103
5 x 101 - 5 x 102
. 10 - 3 x 102
5 x 101 - 2 x 102
1 x 102 - 3 x 103
2 x 102 - 9 x 103
1 - 1 x 103
3 - 3 x 102
1 x 102 - 3 x 104
2 - 8 x 10'
10 - 2 x 102
2 x 10'1 - 10
1 - 2 x 102
4 x 102 - 1 x 103
V-9
-------
Table V-2. Concentration Factors for Edible Portions of Freshwater Fish (L/kg) (Continued)
Element
I
Cs
Ba
La
Ce
Pr
Nd
Pm
Eu
Ta
W
Hg
Pb
Bi
Po
Ra
Th
Pa
U
Np
Pu
Am
Cm
Recommended Value
4x10'
2xl03
4
3x10'
3x10'
IxlO2
IxlO2
3x10'
5x10'
IxlO2
10
IxlO3
SxlO2
10
SxlO1
5x10'
IxlO2
10
10
3x10'
3x10'
3xlOl
3x10'
Range
2 x 101 - 6 x 102
3 x 101 - 3 x 103
4 - 2 x 102
3 x 101 - 5 x 102
3 x 101 - 1 x 102
3 x 10' - 1 x 102
10 - 2 x 102
10 - 2 x 102
1 x 102 - 3 x 104
10 - 1 x 103
1 x 102 - 3 x 102
10 - 5 x 102
10 - 2 x 102
3 x 101 - 1 x 10*
2 - 5 x 101
10 - 3 x 103
4 - 3 x 102
3 x 101 - 3 x 102
3 x 101 - 3 x 102
Source: IAE94
Highlighted elements represent radionuclides under consideration in this report.
V-10
-------
(2) In freshwater fish, the uptake of biologically significant elements (and radionuclides)
that leads to bioaccumulation occurs principally through the ingestion of food and .not
through direct sorption from water (FLE70, KIN61, WIL61, HAS63, EHS63, RIC66).
For this reason, reported concentration factors observed in natural environments
commonly exceed those of laboratory conditions by several orders of magnitude
(IAE75).
(3) For catfish provided pelletized feed and raised in tanks, exposure to radioactivity is,
therefore, limited to water that is assumed to be contaminated at activity levels of one
pCi per liter. (Since water is continuously aerated and mechanically filtered, tanks are
assumed to contain insignificant amounts of sediment or suspended particulates.)
In the absence of scientific data regarding concentration factors that are limited to the direct
sorption of radionuclides contained hi water and applicable to the unique conditions of
aquaculture, ^pert opinion was sought from individuals associated with SENES Oak Ridge
Inc., Center for Risk Analysis (APO96). Although reluctant to suggest specific
bioaccumulation values, experts did not object to EPA using traditional concentration factors
like those hi Table V-2 and applying an adjustment factor that reduces bioaccumulation by one-
hundred fold with a lower-limit concentration factor of one. For example, direct sorption of
Cs-137, Pb-210 and Th-230 would yield concentration factors of 20, 3, and 1, respectively
(Table V-3).
On the basis of bioaccumulation factors that are derived for direct sorption, it is concluded that
activity levels in catfish raised under controlled conditions are low. Previous estimates of
individual dose/risk that may result from substituting catfish for beef consumption are,
therefore, not significantly affected.
Hydroponic Farming'
Hydroponics is the science of growing plants without soil. Nutrient solution alone provides a
more direct and efficient way to provide the essential constituents for plant growth. No soil
means no weeds that compete for nutrients or soil-born parasites that requires pesticides. By
controlling nutrient concentrations near optimal levels, the root systems are proportionately
smaller than plants grown in soils with varying nutrient contents. This implies that plants not
only grow faster, but can channel growth on the edible plant mass rather than on an extensive
root system.
General Information was obtained from InterUrban Water Farms, Riverside, California.
V-ll
-------
Table V-3. Assumed Concentration Factors Limited to Direct
Sorption for Edible Portions of Freshwater Fish
Radionuclide
U-234
U-238
Th-230
Tc-99
Se-79
Ra-226
Pu-239
Pu-240
Pb-210
Np-237
Ni-59
Nb-94
1-129
Cs-137
Cs-135
Cm-246
Cm-245
Am-243
Am-241
Cone. Factor :
(L/kg)
1
1
1
1
1
1
1
1
3
1
1
3
1
20
20
1
1
1
1
Activity .in
Catfish Meat*
(pCi/kg)
1
1
1
1
1
1
1
1
3
1
1
3
1
20
20
1
1
1
1
Activity in catfish corresponds to a water concentration of one pCi/L.
V-12
-------
Aside from nutrients, the growing media in hydroponic farming is totally inert. There are
various methods of growing plants hydroponically that include highly aerated water, moist
humid air, or a solid, but hydroscopic, inert medium:
Water culture: Narrow open troughs commonly fashioned from rain gutters or bisected
PVS pipes are commonly used to hold plants. An aerated nutrient solution is circulated
around the root system submerged in the trough.
Aeroponics: Humid air provides the environment in which the plant roots grow.
Troughs or bags are commonly used to hold/support the plant while nutrient solution is
sprayed to keep roots moist.
Media culture: A number of different inert media may be used to provide support for
roots. Common media include rockwool (a fibrous sponge-like material made from
molten rock), or geolite (a ceramic kiln-fired pebble). When placed in troughs or bags,
their porosity and/or particle size allows for free circulation of nutrient-containing
water.
Depending on climatic conditions, hydroponic farming can be conducted hi greenhouses or
outdoors and is suitable for a variety of plants that includes tomatoes, sweet peppers, snow-
peas, bean-sprouts, etc.
The limiting factor for outdoor hydroponic farming would be the relative humidity. The threat
of rapid plant/root dehydration in the hot and arid climate of Yucca Mountain would limit
hydroponic farming to hot-houses (for commercial production) and indoor gardens (for
personnel production).
Currently, there are commercial vendors who market a variety of equipment and supplies for
both large-scale and small-scale production. Relative to conventional fanning, the cost of
hydroponic farming is low. It is estimated that the yearly cost of fertilizer and pH control
products for a personal-use system that produces about 200 pounds of tomatoes annually
averages around $60 to $80. This is about three to four cents per pound.
V-13
-------
RadionuclideJlontenLQlUydroponically Grown Vegetables
The principal method by which plants, inclusive of vegetables and fruits, incorporate
radionuclides contained in soil is by root uptake. Root uptake requires that the contaminant is
dissolved in water that occupies soil pore spaces within the plant's root zone. For modeling
and calculational purposes, it would, therefore, appear appropriate for bioaccumulation factors
to be defined as the ratio of the radionuclide activity per unit mass of plant to the radionuclide
activity per unit volume of soil water.
However, the water content of most soils is highly variable with time due to the episodic
nature of precipitation. Additionally, pore space water represents a small fraction of the total
soil mass and is difficult to remove under normal conditions. For these reasons, the traditional
method for modeling food pathways is to base the expected radioactivity levels in plants to that
of the soil in which they are grown.
In Section 8.3.4.1, food transfer factors are discussed in terms of a concentration ratio (CR)
where:
_ Radionuclide activity per unit mass of plant
CR — ———————^—^^—^—
Radionuclide activity per unit mass of soil
It should be noted, however, that CR values, even for a specific radionuclide, are not constant,
but reflect the complex physical/chemical behavior of the contaminant in soil.
For Yucca Mountain, the vegetable exposure scenario is based on the fact that when soil is
repeatedly irrigated with contaminated ground water, there is a steady buildup of soil
contaminants over time. At some point, however, an equilibrium condition is reached where
further irrigation is off-set by the removal rate of contaminants by the combined effects of
radioactive decay and various removal mechanisms such as of soil leaching. Plants grown in
soil at equilibrium soil conditions can be expected to have the highest radionuclide
concentration.
This complex relationship of contaminant buildup/removal in soil and plant uptake is strongly
influenced by the partitioning coefficient of the radionuclide contaminants previously defined
as its K value where:
V-14
-------
_ concentration of radionuclide adsorbed on soil particles (pCilkg)
d concentration of contaminant in water occupying soil pore space (pCUL)
The impact of IQ values on soil buildup and plant uptake do not parallel each other for the
following reasons:
(1) When soil conditions yield a high IQ value, the radionuclide can be expected to strongly
adhere to soil particles and resist removal by leaching. This leads to a higher
equilibrium soil concentration value.
(2) Increased K,, values, however, imply that radionuclides are not readily removed by
pore-space water, which reduces the opportunity for plant uptake.
From the combined values of concentration ratios for plants grown in .soil and their assumed
soil partitioning coefficient, a concentration ratio value can be derived for plants grown hydro-
ponically by the following relationship:
radionuclide activity per unit mass of plant
Plant CR
Hydroponicallygroy>n radionuclide acuity per unit volume of water in root zone
activity per unit masss of plant ( activity per unit mass of soil particles
activity per unit mass soil activity per unit volume -water in root zone
Table V-4 cites the concentration ratios of plants grown hydroponically (CP^a,,^^ for
radionuclides under assessment for Yucca Mountain. These data were derived from the Kj
values in Table 8-3 and concentration ratios for plants grown in soil cited hi Tables 8-4 and
8-6. These latter tables appear in Section 8.3.4 of this chapter.
V-15
-------
Table V-4. Inferred Concentration Ratios for Plants Grown Hydroponically
Radio-
nuclides
U-234
U-238
Th-230
Tc-99
Se-79
Ra-226
Pu-239
Pu-240
Pb-210
Np-237
Ni-59
Nb-94
1-129
Cs-137
Cs-135
Cm-246
Cm-245
Am-243
Am-241
Leafy Vegetables
6.0E+0
6.0E+0
1.1E+0
9.5E+0
4.8E-1
4.5E+1
1.7E-2
1.7E-2
8.1E-1
6.9E+0
1.3E+1
6.0E+0
3.4E-4
3.3E+1
3.3E+1
2.9E+0
2.9E+0
1.2E-1
1.2E-1
Concentration Ratio
epCtlkg planf^.
I )
pCitL water
Other Vegetables
2.9E+0
2.9E+0
3.4E-2
1.1E-1
4.8E-1
7.3E+0
1.1E-2
1.1E-2
4.7E+0
2.7E+0
2.1E+0
2.0E+0
2.0E-3
2.2E+1
2.2E+1
1.5E+0
1.5E+0
4.7E-2
4.7E-2
Fruit
2.9E+0
2.9E+0
3.4E-2
1.1E-1
4.8E-1
7.3E+0
1.1E-2
1.1E-2
4.7E+0
2.7E+0
2.1E+0
2.0E+0
2.0E-3
2.2E+1
2.2E+1
1.5E+0
1.5E+0
4.7E-2
4.7E-2
V-16
-------
Concentration ratios for plants grown hydroponically in a contaminated water medium clearly
show which elements appear in higher concentrations in plant matter than in the water from
which they were removed. For example, when Am-243 is present in water at one pCi/L of
water, leafy vegetables grown hydroponically would be expected to have an activity level of
about 0.12 pCi/kg. Similarly, leafy vegetables would be expected to exhibit about 33 pCi/kg
Cs-137 when grown hydroponically in a contaminated water medium.
A quantitative assessment of the potential impact of substituting hydroponically-grown
vegetables for soil-grown vegetables in dose assessments is not possible at this time. This is
due to the fact that previous dose/risk models associated with soil-grown plants did not
consider soil buildup (See Section 8.3.4.1). It is reasonable, however, to conclude that soil-
grown vegetables grown hydroponically would be expected to have lower activity levels than
those grown in soil.
V-17
-------
REFERENCES
(Appendices I-V)
ACS95 .- American Cancer Society, Cancer Facts & Figures - 1995, Atlanta, Georgia,
1995.
ALD96 Aldrich, B. American Ostrich Association, Fort Worth, Texas, personal
communications, Nov. 6, 1996.
ANS86 Anspaugh, L.R., and B.W. Church, Historical Estimates of External Exposure
and Collective External Exposure from Testing at the Nevada Test Site. Vol. I.
Test Series through HardtackII, 1958, Health Physics, 51:35, 1986
ANS90 Anspaugh, L.R., et al., Historical Estimates of y Exposure and Collective
External y Exposure From Testing at the Nevada Test Site. Vol. II. Test Series
after Hardtack II, 1958, and Summary, Health Physics, 59:525, 1990.
AOA96 American Ostrich Association, Fort Worth, Texas, personal communications,
Nov. 6, 1996.
APO96 Aposoaei, J., SENES Oak Ridge, Inc., Oak Ridge, TN, personal communications,
Nov. 8, 1996.
ASM89a American Society of Mechanical Engineers, Quality Assurance Program
Requirements for Nuclear Facilities, Report No. ASME NQA-1-1989, New York,
1989.
ASM89b American Society of Mechanical Engineers, Quality Assurance Program
Requirements for Nuclear Facilities Applications, Report No. ASME NQA-2y-
1989, New York, 1989.
BAR95 Barish, R.J., Health Physics and Aviation: 1990-1994, Health Physics, 69:538,
1995.
BCI85 BCI Geonetics International, Inc., Ground Water Exploration in Northwest
Somalia: Technical Report I - Regional Ground Water Potential and
Identification of Potential Test Drilling Sites in Fractured Bedrock Aquifers,
November 1985.
BLA82 Blaylock, E.G., Radionuclide Data Bases Available for Bioaccumulation Factors
for Freshwater Biota, Nuclear Safety 234:427, 1982.
CAL83 Caldwell, G.G., et al., Mortality and Cancer Frequency among Military Nuclear
.App. Ref. 1
-------
CAR69
CHA94
CNW95
DO96
DOE88a
DOE92
DOE93a
DOE93b
DOE93c
DOE93d
DOE94a
DOE94b
Test (Smokey) Participants, 1957 through 1979, Journal of the American Medical
Association, 250:620, 1983.
. Carmichael, H. and Bercovitch, M., Analysis oflQSY Cosmic Ray Survey
Measurements, Canadian Journal of Physics, 47:2073, 1969.
Chambers, D. and L. Mays (eds.), The American Garden Guides: Vegetable
Gardening, Pantheon Books, New York, 1994.
Center for Nuclear Waste Regulatory Analysis, Initial Analysis of Selected Site-
Specific Dose Assessment Parameters and Exposure Pathways Applicable to a
Ground Water Release Scenario at Yucca Mountain, Contract NRC-02-93-005,
San Antonio, Texas, 1995.
Doyle, J., personal communication with Mr. David Rau, Amargosa Valley,
Nevada, September 18,1996.
U.S. Department of Energy, GENII: The Hanford Environmental Radiation
Dosimetry Software System, Volumes 1, 2 and 3: Conceptual Representation,
Users Manual, Code Maintenance Manual, Pacific Northwest Laboratory,
PNL-6584, Richland, Washington, 1988.
U.S. Department of Energy, Preliminary Performance Assessment for the Waste
Isolation Pilot Plant, Volume 3, SAND92-0700/3, Sandia National Laboratories,
Waste Isolation Pilot Plant Performance Assessment Department, 1992.
U.S. Department of Energy, Recommendations for the Preparation of
Environmental Assessments and Environmental Impact Statements, Office of
NEPA Oversight, 1993.
U S Department of Energy, Nevada Operations Office Annual Site
Environmental Report, Report No. DOE/NV/10630-66, 1993.
U S Department of Energy, User's Guide for GENII-S: A Code for Statistical
and Deterministic Simulation of Radiation Doses to Humans from Radionuclides
in the Environment, Sandia National Laboratories, SAND 91-2375, 1993.
U S. Department of Energy, Preliminary Total-System Analysis of a Potential
High-Level Nuclear Waste Repository at Yucca Mountain, Pacific Northwest
Laboratories, PNL-8444,1993.
U S Department of Energy, Total System Performance Assessment for Yucca
Mountain - SNL Second Iteration (TSPA-1993), SAND93-2675, April 1994.
U S Department of Energy, Parameters Used in the Environmental Pathways
App. Ref. 2
-------
and Radiological Dose Modules (DESCARTES, CIDER, and CRD Codes) of the
Hanford Environmental Dose Reconstruction Integrated Codes (HEDRIC),
Pacific Northwest Laboratory, PNWD-2023 HEDR, Revision 1, Richland,
.. Washington, 1994.
DOE94c U.S. Department of Energy, Total System Performance Assessment -1993: An
Evaluation of the Potential Yucca Mountain Repository, Intera, Inc.,
BOOOOOOOO-01717-2200-00099-Rev.01, 1993.
DOE95a U.S. Department of Energy, Yucca Mountain Site Characterization Project:
Summary of Socioeconomic Data Analyses Conducted in Support of the
Radiological Monitoring Program During Calendar Year 1994, TRW
Environmental Safety Systems, Inc., Contract DE-AC01-91RW00134, Las
Vegas, 1995.
DOE95b U.S. Department of Energy, Total System Performance Assessment -1995: An
Evaluation of the Potential Yucca Mountain Repository, TRW Environmental
Safety Systems, Inc, BOOOOOOOO-01717-2200-00136, Revision 01, November
1995.
DOE96 U. S. Department of Energy, The Environmental Impact Statement for the Nevada
Test Site and Off-site Locations in the State of Nevada (Draft), Volume 1,
DOE/EIS 0243, Las Vegas, Nevada, 1996.
ECK86 Eckl, P., et al., Uptake of Natural and Man-made Radionuclides by Lichens and
Mushrooms, Radiation Environmental Biophysics, 25:43, 1986.
EHS63 Environmental Health Sciences, Distribution ofCo-60, Zn-65, Sr-85, and Cs-137
in a Freshwater Pond, Public Health Service Publication No. 999-RH-24, 1963.
EPA72a U.S. Environmental Protection Agency, Estimates of Ionizing Radiation Doses in
the United States 1960-2000, Office of Radiation Programs, Report ORP-CSD
72-1, Washington, D.C., 1972.
EPA72b U.S. Environmental Protection Agency, Natural Radiation Exposure in the
United States, Office of Radiation Programs, Report ORP/SID-72-1, Washington,
D.C., 1972.
EPA77 U.S. Environmental Protection Agency, The Radiological Quality of the
Environment in the United States 1977, EPA 420/1-77-0009, Washington, D.C.,
1977.
EPA88 U.S. Environmental Protection Agency, Limiting Values of Radionuclide Intake
and Air Concentration and Dose Conversion Factors for Inhalation, Submersion,
andlngestion, Office of Radiation Programs, Federal Guidance Report No. 11,
App. Ref. 3
-------
EPA 520/1-88-020, Washington, D.C., September 1988.
EPA89 U.S. Environmental Protection Agency, Risk Assessment Guidance for Superfund,
.- Volume I, Human Health Evaluation Manual (Part A), Office of Emergency and
Remedial Response, EPA 540/1-89-002, Washington, D.C., 1989.
EPA92 U.S. Environmental Protection Agency, Guidance on Risk Characterization for
Risk Managers and Risk Assessors, Deputy Administrator F. Henry Habicht II,
February 1992.
EPA93a U.S. Environmental Protection Agency, Offsite Environmental Monitoring
Report: Radiation Monitoring Around U.S. Nuclear Test Areas, EMS Laboratory,
Las Vegas, Nevada, 1993.
EPA93b U.S. Environmental Protection Agency, External Exposure to Radionuclides in
Air, Water, and Soil, Office of Air and Radiation, Federal Guidance Report No.
12, EPA 402-R-93-081, Washington, D.C., 1993.
EPA94 U S Environmental Protection Agency, Estimating Radiogenic Cancer Risks,
Office of Air and Radiation, EPA 402-R-93-076, Washington, D.C., 1994.
EPR94 Electric Power Research Institute, A Proposed Public Health and Safety Standard
for Yucca Mountain, APR TR-104012, Palo Alto, California, 1994.
ERD77 U S. Energy Research and Development Agency, Measurement of Cosmic-ray
Charged Particle lonization and Flux Densities in the Atmosphere, USERDA
Report HASL-324, 1977.
FIE86 Fiero B., Geology of the Great Basin, University of Nevada Press, Reno, 1986.
FLE70 Fleishman, D.G., Modern Problems in Radioecology, Nauk, Moscow, p. 11,
1970.
FRE67 Freke, A.M., A Model for the Approximate Calculation of Safe Rates of
Discharge of Radioactive Wastes into Marine Environments, Health Physics
13:743, 1967.
GO96 Goldin, D., personal communication with Mr. S. Goldin, Boulder, Colorado,
November 6,1996.
HAS63 Hasanen, E. and J.K. Miettinen, Cs-137 Content of Freshwater Fish in Finland,
Nature 200:1008, 1963.
HEN92 Henley, E.J., and H. Kumamoto, Probabilistic Risk Assessment: Reliability,
Engineering, Design, and Analysis, Institute of Electical and Electronics
App. Ref. 4
-------
Engineering Press, New York, 1992.
HIC90 Hicks, H.G., Additional Calculations of Radionuclide Production Folio-wing
. Nuclear Explosions and Pu Isotopic Ratios for Nevada Test Site Events, Health
Physics, 59(5):515, 1990.
HOL66 Holtzman, R.B., Natural Levels ofLead-210, Polonium-210, and Radium-226 in
Humans and Biota of the Arctic, Nature, 210:1094, 1966.
IAE75 International Atomic Energy Agency (IAEA), Field and Laboratory
Comparability of Radioecological Studies, In: Design of Radiotracer
Experiments in Marine Biological Systems, Technical Report Series No. 167,
Vienna, Austria 1975.
IAE94 International Atomic Energy Agency, Handbook of Parameter Values for the
Prediction of Radionuclide Transfer in Temperate Environments, Technical
Report Series No. 364, Vienna, Austria, 1994.
ICR77 International Commission on Radiological Protection, Recommendations of the
ICRP, ICRP Publication 26, Pergamon Press, 1977.
ICR85 International Commission on Radiological Protection, Principles of Monitoring
for the Radiation Protection of the Population, Publication 43, Annals of the
ICRP 15(1), Pergamon Press, 1985.
ICR91 International Commission on Radiological Protection, Risks Associated with
Ionizing Radiations, Volume 22, No. 1, Pergamon Press, 1991.
IUR89 International Union of Radioecologists, Sixth Report of the Working Group on
Soil-to-Plant Transfer Factors, Biltoven, The Netherlands: RTVM, 1989.
JOH87 Johnson, C. J., A Cohort Study of Cancer Incidence in Mormon Families Exposed
to Nuclear Fallout Versus an Area-Based Study of Cancer Deaths in Whites in
Southwestern Utah, American Journal of Epidemiology, 125:166, 1987.
KER93 Kerber, R. A.,et al., A Cohort Study of Thyroid Disease in Relation to Fallout
from Nuclear Weapons Testing, Journal of the American Medical Association,
270:2076, 1993.
KIN61 King, S.F., Uptake and Transfer ofCs-137 by Chlamydomonas, Daphnia and
Bluegill Fingerlings, Ecology 45( 4):852, 1961.
LAN91 Land, C.E., and W.K. Sinclair, The Relative Contributions of Different Organ
Sites to the Total Cancer Mortality Associated with Low-Dose Radiation
App. Ref. 5
-------
LAY96
Exposure, in: Risks Associated with Ionising Radiations, Annals of the ICRP
22(1), Pergamon Press, 1991.
-Layne-Northwest, W229N5005 DuPlainville Road, Pewaukee, Wisconsin, 53072,
personal communication with Mr. John Horton, April 1996.
LLL68 Lawrence Livermore National Laboratory, Concentration Factors of Chemical
Elements in Edible Aquatic Organisms, UCRL-50564, Livermore, California,
1968.
LLL78 Lawrence Livermore National Laboratory, Assessment of the Dose Commitment
from Ingestion of Aquatic Foods Contaminated by Emissions from a Proposed
Nuclear Fuel Reprocessing Plant, UCID-17882,1978.
LLL87 Lawrence Livermore National Laboratory, Retention by Vegetation of
Radionuclides Deposited in Rainfall - A Literature Summary, UCRL-53810,
Livermore, California, 1987.
LYO79 Lyon, J.L., et al.5 Childhood Leukemias Associated with Fallout from Nuclear
Testing, New England Journal of Medicine, 300:397, 1979.
MAC87 Machado, S.G., C.E. Land, and F.W. Mckay, Cancer Mortality and Radioactive
Fallout in Southwestern Utah, American Journal of Epidemiololgy, 125:44, 1987.
MAT75 Mattson S.L.J., Cs-137 in the Reindeer Lichen Cladonia Alpestris: Deposition,
Retention and Internal Distribution, 1961-1970, Health Physics 28:233, 1975.
MAY66 Mays, C.W., Thyroid Irradiation in Utah Infants Exposed to Iodine-131, Scientist
Citizen, 8:3, 1966.
MCA91 McArthur, R.D., Radionuclides in Surface Soil at the Nevada Test Site,
DOE/NV/10845-02, U.S. Department of Energy, Las Vegas, NV, 1991.
MIK92 Miklas, M.P. Jr., et al, Natural Resource Regulatory Requirements: Background
and Consideration of Compliance Methodologies, Center for Nuclear Waste
Regulatory Analyses, San Antonio, TX, September 1992.
MLS93 Mills, L., Beginning Desert Gardening, University of Nevada Cooperative
Extention, Reno, Nevada, 1993.
NAS80 National Academy of Sciences - National Research Council, The Effects of
App. Ref. 6
-------
Populations of Exposure to Low Levels of Ionizing Radiation (BEIR HI), Report
of the Biological Effects of Ionizing Radiation, Washington, D.C., 1980
NAS86 ..National Academy of Sciences - National Research Council, The Airliner Cabin
Environment: Air Quality and Safety, National Academy Press, Washington,
D.C., 1986.
NAS88 National Academy of Sciences - National Research Council, Health Risks of
Radon and Other Internally Deposited Alpha-Emitters (BEIR IV), Report of the
Biological Effects of Ionizing Radiation, Washington, D.C., 1988.
NAS90 National Academy of Sciences - National Research Council, Health Effects of
Exposure to Low Levels of Ionizing Radiation (BEIR V), Report of the Biological
Effects of Ionizing Radiation, Washington, B.C., 1990.
NAS95 National Academy of Science - National Research Council, Committee on
Technical Bases for Yucca Mountain Standards, Technical Bases for Yucca
Mountain Standards, National Academy Press, Washington, D.C., 1995.
NCR77 National Council on Radiation Protection and Measurements, Radiation Exposure
from Consumer Products and Miscellaneous Sources, NCRP Report No. 56,
Bethesda, Maryland, 1977.
NCR87a National Council on Radiation Protection and Measurements, Ionizing Radiation
Exposure of the Population of the United States, NCRP Report No. 93, 1987.
NCR87b National Council on Radiation Protection and Measurements, Exposure of the
Population in the United States and Canada from Natural Background Radiation,
NCRP Report No. 94, 1987.
NCR87c National Council on Radiation Protection and Measurements, Radiation Exposure
of the U.S. Population from Consumer Products and Miscellaneous Sources,
NCRP Report No. 95, 1987.
NCR89 National Council on Radiation Protection and Measurements, Exposure of the
U.S. Population from Diagnostic Medical Radiation, NCRP Report No. 100,
1989.
NRC83 U.S. Nuclear Regulatory Commission, Radiological Assessment: A Textbook on
Environmental Dose Analysis, NUREG/CR-3332, Office of Nuclear Reactor
Regulation, Washington, D.C., 1983.
NYE93a Nye/Esmeralda Economic Development Authority, Nye County Overall
Economic Development Plan (Draft), 1993.
App. Ref. 7
-------
NYE93b Nye/Esmeralda Economic Development Authority, Nye County Overall
Economic Development Plan; Appendix C 'Threatened and Endangered Species
Considered in the Analysis of the Effects of Defense-Related Activities on Wildlife
..in Nevada,' July 23, 1993.
NYE93c Nye County Board of Commissioners, 1990 Census Data Profiles Nye County,
Nevada and Its Communities; Social, Labor Force, Income and Poverty, and
Housing Characteristics, Planning Information Corporation, 1993.
NYE93d Nye County Board of Commissioners, Baseline Economic and Demographic
Projections: 1990-2010 Nye County and Nye County Communities, Planning
Information Corporation, May 25, 1993.
NYE93e Nye/Esmeralda Economic Development Authority, Nye County Overall
Economic Development Plan; Appendix I 'Beatty Background Information,'
Appendix J' Pahrump Background Information,' Appendix K 'Tonopah
Background Information'(Drafts), 1993.
NYE93f Nye/Esmeralda Economic Development Authority, Nye County Overall
Economic Development Plan; Appendix L 'Amargosa Valley Background
Information,' Appendix N 'Round Mountain/Smoky Valley Background
Information'(Drafts), 1993.
NYE94 Nye County Board of Commissioners, Nye County, Nevada Socioeconomic
Conditions and Trends 1993, Planning Information Corporation, 1994.
ORN76 Oak Ridge National Laboratories, BIORA2 Bioaccumulation Factors for
Freshwater Biota, Rep. ORNL-4992, Oak Ridge National Laboratory, Oak
Ridge, Tennessee, 1976.
ORN82 Oak Ridge National Laboratories, A Review and Analysis of Parameters for
Assessing Transport of Environmentally Released Radionuclides Through
Agriculture, ORNL-5786, Oak Ridge, Tennessee, 1982.
OWE95 Owen, M.J., & si., Nevada Agricultural Statistics 1995, Nevada Agricultural
Statistics Services, 1995.
PEN58 Pendleton, R.C., C.W. Mays, R.D., Lloyd, and B.W. Church, Absorption of
Cesium-137 by Components of an Aquatic Community, Proceedings of the Second
United National International Symposium qn the Peaceful Uses of Atomic
Energy, Vol 18, p. 419, 1958.
PNL86 Pacific Northwest Laboratories, A Literature Review of the Concentration Ratios
of Selected Radionuclides in Freshwater and Marine Fish, PNL-5484, Richland,
App. Ref. 8
-------
Washington, 1986.
POH95 Pohlmann, K., J.B. Chapman, and R. Andricevic, Exposure Assessment of
. - Ground-water Transport of Tritium from the Central Nevada Test Area,
Publication #45233, Desert Research Institute, Las Vegas, NV, 1995.
POS88 Poston, T.M. and D.C. Klopfer, Concentration Factors Used in the Assessment of
Radiation Dose to Consumers of Fish: A Review of27Radionuclides, Health
Physics 55(5):751, 1988.
PUS95 Puskin, J.S., and C.B. Nelson, Estimates of Radiogenic Cancer Risks, Health
Physics, 69(1):93, 1995.
RAL74 Rallison, M.L., Thyroid Disease in Children. A Survey of Subjects Potentially
Exposed to Fallout Radiation, American Journal of Medicine, 56:457, 1974.
RAL75 Rallison, M.L., et al., Thyroid Modularity in Children, Journal of the American
Medical Association, 233:1069, 1975.
RAL90 Rallison, M.L., et al., Cohort Study of Thyroid Disease near the Nevada Test Site:
A Preliminary Report, Health Physics, 59:739,1990.
RIC66 Rickard, W.H., Cs-137 in Cascade Mountain Vegetation, In: Symposium on
Radioecology. Editors: D.J. Nelson andF.C. Evans, 1966.
ROB83 Robinette, C.D., and S. Jablon, Studies of Participants at Testing of Nuclear
Weapons: I. The Plumbob series, in Somatic and Genetic Effects, Proceedings of
the Seventh International Congress of Radiation Research, Martinus Nijhoff
Publishers C8-13, Amsterdam, 1983.
ROS92 Roseberry, A.M., and D.E. Bunnaster, Lognormal Distributions for Water Intake
by Children, Risk Analysis, 12:99, 1992.
SIM95 Simon, S.L., et al., The Utah Leukemia Case-Control Study: Dosimetry
Methodology and Results, Health Physics, 68:460, 1995.
SPA85 Spaulding, W.J., Vegetation and Climates of the Last 45,000 Years in the Vicinity
of the Nevada Test Site, South Central Nevada, 1985.
STI91 Stichler, C., Texas Alfalfa Production, B-5017, Texas Agricultural Extension
Service, College Station, Texas, 1991.
TIL95 Till, J.E., et al., The Utah Thyroid Cohort Study: Analysis of the Dosimetry
App. Ref. 9
-------
Results, Health Physics, 68:472,1995.
UNS82 United Nations Scientific Committee on the Effects of Atomic Radiation,
Ionizing Radiation: Sources and Biological Effects, New York, 1982.
UNS88 United Nations Scientific Committee on the Effects of Atomic Radiation,
Sources, Effects and Risks of Ionizing Radiation, New York, 1988.
UNS93 United Nations Scientific Committee on the Effects of Atomic Radiation, Sources
and Effects of Ionizing Radiation, New York, 1993.
UNS94 United Nations Scientific Committee on the Effects of Atomic Radiation,
Sources and Effects of Ionizing Radiation, New York, 1994.
USG72 U.S. Geological Survey, Water Supply for the Nuclear Rocket Development
Station at the U.S. Atomic Energy Commission's Nevada Test Site, Geological
Survey Bulletin 1938, Washington D.C., 1972.
VIC96 Vickery Drilling, P.O. Box 9013, Evansville, Indiana, 47724, personal
communication with Mr. Leroy Jochum, April 1996.
WAC90 Wachholz, B.W., Overview of the National Cancer Institute's Activities Related to
Exposure of the Public to Fallout from the Nevada Test Site, Health Physics,
59:511, 1990.
WAL80 Walters, M., Birds of the World, 1980.
WAH75 Wahlgren, M. A., and J. S. Marshall, The Behavior of Plutonium and Other Long-
lived Radionuclides in Lake Michigan, In Impact of Nuclear Releases into the
Aquatic Environment, p. 227, Proceedings IAEA Symposium, Otaniemi, Finland,
30 June - 4 July, 1975.
WHC90 Whicker, F.W., et al., Estimation of Radionuclide Ingestion: The "PATHWAY"
Food-chain Model, Health Physics, 59:645,1990.
WI96 Wittmeyer, G.W., et. al. Use of Ground water in the Arid and Semi-Arid Western
United States: Implications for Yucca Mountain Area, Center for Nuclear Waste
Regulatory Analysis, San Antonio, Texas, August 1996.
App. Ref. 10
-------
APPENDIX VI
CURRENT INFORMATION REGARDING GROUND-WATER FLOW AND
RADIONUCLIDE TRANSPORT IN THE UNSATURATED AND SATURATED ZONES
-------
APPENDIX VI
CURRENT INFORMATION REGARDING GROUND-WATER FLOW AND
RADIONUCLIDE TRANSPORT IN THE UNSATURATED AND SATURATED ZONES
VI.l UNSATURATED ZONE
VI.1.1 T Tnsatiirated Z™"* ^yHrolopv Model
The unsaturated-zone (UZ) flow analysis of Yucca Mountain comprises four components
(climate, infiltration, mountain-scale UZ flow, and seepage into drifts) that are believed to play
an important role in the performance of the potential repository. Climate and infiltration
influence the amount of water percolating toward the repository. Subsequently, water seeping
into the drifts and onto the waste packages can accelerate waste-package degradation, and rapid
pathways from the repository to the water table via fractures can decrease the transit time of
radionuclides to the accessible environment. Prediction of these events relies on process models
of UZ flow that have been tested or calibrated against available data at Yucca Mountain. This
section describes the development of these models and the important processes and relevant
parameters used in the TSPA-VA.
VI. 1.1.1 Synopsis of TSPA-VA Treatment and Changes from Prior Efforts
A primary difference in the treatment of UZ flow between TSPA-VA and previous TSPAs is the
use of a fully three-dimensional UZ flow model developed at LBNL by Bodvarsson et al.
(BOD97). Their goal was to synthesize all of the available data into a coherent, predictive model
of water and air flow in the UZ using the dual-permeability continuum formulation or fracture
and matrix flow and interaction. The numerical formulation of be model is implemented with
the TOUGH2 computer code (PRU91). Model calibration is performed using an inverse method
implemented in ITOUGH2 (FIN93) to optimize the model parameters against available data. A
new feature of the current LBNL model (BOD97) is that a fracture-matrix coupling parameter,
related to the wetted contact area between fractures and matrix, can be used as an inversion
parameter in each hydrogeologic unit. The use of the three-dimensional UZ flow model
precludes the need for simplified or abstracted flow fields, as has been required in previous
TSPAs. However, because of the size and computational requirements of the three-dimensional
VI-1
-------
UZ flow model, the emphasis in TSPA-VA has been on using selected conceptual models and
parameter sets that span the range of uncertainty in UZ flow modeling, rather than randomly
sampling large combinations of parameters.
VI. 1.1.2 Infiltration
The ultimate source of water in the unsaturated zone at Yucca Mountain is precipitation on the
mountain. The spatial and temporal relationships between infiltration and recharge are complex,
because of the hydrogeologic variability of Yucca Mountain. Some water that infiltrates returns
to the surface by interflow; another part is returned to the atmosphere by evapotranspiration. A
small quantity that is not evaporated, or discharged as interflow, percolates deep into the
unsaturated zone and becomes net infiltration or percolation. The terms "infiltration" and
"percolation" are used frequently, sometimes interchangeably, in literature about the Yucca
Mountain unsaturated zone. For the purposes of this report, "infiltration" is used to describe the
amount of water which enters Yucca Mountain at the ground surface, while "percolation" is used
to describe the amount of water which actually penetrates deep enough into the mountain to
reach the repository horizon and below. The difference between the two terms lies mainly in the
partitioning of part of the infiltration flux into the vapor phase, which may then be recirculated to
the atmosphere.
At Yucca Mountain, the infiltration rate is both spatially and temporally variable. Because the
quantity of net infiltration that percolates through different paths is quite variable, estimated
average recharge rates do not represent percolation rates through specific flow paths. Spatial
variations of infiltration depend mostly on variations in the properties of surficial units,
topography, the intersection of faults with the surface, and the presence of local fracturing.
Temporal variations in infiltration rate are related to the seasonality and relatively infrequent
precipitation events in the arid climate of Yucca Mountain. Temporal variations in the
infiltration rate have also occurred over a much larger time span, reflecting long term climate
changes.
Knowing the temporal and spatial variability of the percolation rates is crucial to modeling
efforts because of the importance of the relationship of infiltration rate to horizontal and vertical
permeabilities of the various units and the effect this has on whether or not significant lateral
flow occurs in the unsaturated zone. The higher the actual infiltration rate, the greater the
VI-2
-------
likelihood of significant lateral flow. Such lateral flow could result from a combination of two
factors. The first factor is that infiltrating water may encounter zones of lower relative
permeability as it moves downward. The second factor is that in many of the units, the relative
permeability is far greater in the direction parallel to bedding than the direction perpendicular to
it. The anisotropic permeability may cause lateral flow of mounded water away from the area in
which it accumulates. Lateral flow is important because it could transmit water to structural
features which would then move the water downward, possibly acting as a conduit to divert large
amounts of water flowing downward through a small area. Such flow paths could direct water
into and through the repository or away from it.
The actual quantity of net infiltration or percolation beneath the surface of Yucca Mountain has
not been accurately determined. The percolation flux is a difficult parameter to determine for
low flux regions such as Yucca Mountain. There are currently no reliable direct measurements
that can be made to determine this important parameter (LBL96). Existing estimates have been
obtained from a mixture of indirect methods involving field testing and modeling of various
processes at different scales. Data exist to suggest that the flux reaching the repository horizon
through the matrix is relatively small. Relatively low matrix saturations measured hi the upper
portion of the TSw suggest that much of the moisture which infiltrates into the TCw does not
reach the TSw (LBL96). Data from the ESF show that no weeping fractures were found, even in
the region where perched water exists in boreholes. It should be noted, however, that because of
ventilation equipment inside the ESF, much of any such moisture might be removed from the
ESF as water vapor. Furthermore, no moisture was observed infiltrating into the radial boreholes
of Alcove 1 of the ESF after storm events, even though the boreholes are located close to the
land surface in the highly fractured and broken TCw formation (LBL96). However, other data
suggest that the percolation flux may reach the repository level mainly through episodic fracture
flow. These data include observation and testing of extensive bodies of perched water located
below the repository horizon, as well as measurements of bomb-pulse isotope levels from
atmospheric nuclear testing which show that some water in the unsaturated zone is relatively
young (LBL96).
Estimates of net infiltration vary from slightly negative (net loss of moisture from the mountain)
to about 10 mm/yr (LBL96). It is reported in USG84 that net infiltration flux probably ranges
from 0.5 to 4.5 mm/year, based on estimates of earlier workers for various localities in the Yucca
Mountain area. Flint and Flint (FLI94) provide preliminary estimates of spatial infiltration rates
VI-3
-------
that range from 0.02 mm/yr, where the welded Tiva Canyon unit outcrops, to 13,4 mm/yr in
areas where the Paintbrush nonwelded unit outcrops. The bulk of the area above the repository
block is underlain principally by the Tiva Canyon member. The DOE's 1995 Total System
Performance Assessment concludes that, if the predominant flow direction is vertical, then the
average infiltration through the repository block, using the average infiltration rates of Flint and
Flint (FLI94), would be 0.02 mm/yr. If, on the other hand, the predominant flow direction has a
significant lateral component due to material property heterogeneity and/or anisotropy and the
sloping nature of the hydrostratigraphic unit contacts, then the average net infiltration rate over
the repository block could be as high as some weighted average of the infiltration rates inferred
from FLI94. The 1995 TSPA (DOE95) also reports that the average, spatially-integrated
infiltration rate is about 1.2 mm/yr; most of this infiltration occurs along the Paintbrush outcrop
in the washes north of the repository block.
Recently, several lines of evidence have converged to alter the prevailing view regarding the
magnitude of infiltration/percolation rates beneath Yucca Mountain, with the most recent
estimates being revised upward from previous work. The newer estimates of percolation are
around five mm/yr, with a range of one to 10 mm/yr (LBL96). Recent isotopic analyses of rock
samples from the ESF are consistent with a percolation rate of five mm/yr (LAN96, LBL96).
Profiles of temperature versus depth of water in boreholes are consistent with a range of
infiltration rates from one to 10 mm/yr (LBL96). Three-dimensional modeling results of the
percolation flux at the repository horizon using the latest available spatially varying infiltration
map indicate percolation fluxes on the order of five to 10 mm/yr. The UZ expert elicitation panel
estimates for mean infiltration rates range from 3.9 to 12.7 mm/y (GEO97). The effect of
uncertainty in infiltration and percolation flux rates is presented in the TSPA-VA.
The conceptual model used in infiltration studies defines the physical processes determining net
infiltration, and is based on evidence provided from field studies at Yucca Mountain combined
with established concepts in soil physics and hydrology (FRE79; HOR33; HIL80; FLI96). The
overall framework of the conceptual model is provided by the hydrologic cycle, which includes
all the processes on the surface and in the shallow subsurface (0 to 6 m beneath the ground
surface) that affect net infiltration. These processes include precipitation, infiltration, run-off and
run-on, evapotranspiration, and the redistribution of moisture in the shallow subsurface.
Precipitation is the dominant hydrologic process at the site because it is the source of all moisture
for the surface and shallow subsurface (there are no permanent streams or bodies of surface water
VI-4
-------
affecting the site), excluding water introduced to the site by human activity (dust-control,
drilling, waste water). Precipitation is dependent mostly on meteorological factors, but
geographic location, elevation, and physiography are also important. Evapotranspiration is the
second most dominant hydrologic process (in terms of the total volume of water involved) at
Yucca Mountain and is dependent on vegetation, the distribution of available moisture stored in
the shallow subsurface, and potential evapotranspiration, which is determined by an energy
balance. Redistribution of moisture in the shallow subsurface occurs in response to gravity and
matric potentials and is strongly dependent on soil and bedrock properties. The removal of water
through evapotranspiration is dynamically integrated with the redistribution of moisture in the
shallow subsurface. The generation of run-off is dependent on a combination of soil depth, soil
porosity, soil permeability, bedrock permeability, and ground-surface slope. Run-on and the
routing of surface flow are dependent mostly on surficial material properties, topography, and
channel geometry.
The conceptual model of net infiltration has been developed from analysis of an 11-year record
of neutron logs from 99 boreholes on Yucca Mountain (FLI95). Relative changes in water
content profiles were compared against precipitation records and estimates of evapotranspiration
(HEV94). The measured changes in water content were also compared to physiographic setting,
bedrock geology, and soil cover. In general, field studies indicated that saturated soils are
established primarily during the winter in response to a series of medium to large storm events
which tend to occur more frequently during periods associated with an active El Nino Southern
Oscillation. The timing, intensity, and duration of precipitation, the storage capacity of the soil,
and evapotranspiration determine the availability of water for net infiltration. In the upland areas
of Yucca Mountain where the soil cover is shallow, the lower the effective conductivity of the
underlying bedrock, the longer moisture from precipitation is held in the soil profile where it is
potentially available for evapotranspiration. During winter, when potential evapotranspiration is
at a minimum, smaller amounts of precipitation are needed for developing and maintaining
saturated conditions at the soil-bedrock contact. When the storage capacity of the soil and the
effective conductivity of the underlying bedrock are exceeded, or when precipitation intensity
exceeds the infiltration capacity of the soil, runoff is generated and water is available for routing
down slopes and into channels. The significance of net infiltration beneath channels in washes
relative to sideslopes and ridgetops depends on the frequency and magnitude of runoff events.
In the current conceptual model, net infiltration at Yucca Mountain is characterized as an
episodic, transient process depending primarily on the length of time saturated or near-saturated
VI-5
-------
conditions are maintained at the soil-bedrock interface, or, at a depth of 6 m in deep alluvium,
and the effective conductivity of the underlying bedrock or alluvium. Net infiltration is
determined as the rate of water percolation into bedrock or below a depth of 6 m in alluvium, and
is limited by the effective permeability of the bedrock or alluvium. Evapotranspiration is
assumed to be negligible in bedrock or below a depth of 6 m because no evidence of plant roots
have been found beyond a 5 m depth. The potential for saturating the soil-bedrock interface is
determined by the timing, frequency and intensity of precipitation: the depth, field capacity, and
porosity of the soil cover; potential evapotranspiration, actual evapotranspiration (which is a
function of the available moisture in the soil profile), and the lateral re-distribution of surface
water. Lateral redistribution of soil moisture is assumed to be negligible because tho moisture-
retention potential in the soil to divert flow laterally is relatively small. A detailed description of
the actual determination of net infiltration is given in FLI96.
VI. 1.1.3 Unsaturated Zone Flow
The most important data for understanding unsaturated zone flow come from several surface-
based drill holes and from the ESF, which is an 8-km-long tunnel through Yucca Mountain.
Considerable amounts of data are available on rock-matrix saturations, water potentials, and
temperatures; on chemical composition and isotopic abundances of groundwater and mineral
deposits; on air permeability and air-pressure fluctuations; on rock types and mineralogy; on fault
locations and offsets; on fracture density and orientations; and on matrix permeability and
saturation/desaturation parameters. In addition, there is information on the upper boundary
condition (i.e..infiltration) from a series of weather stations and shallow drill holes instrumented
with neutron probes, and there is information on climatic effects from a variety of paleoclimate
studies and from analogues such as present-day Ranier Mesa.
The first detailed conceptual model of unsaturated zone flow at Yucca Mountain was proposed in
USG84. Since then, the majority of the data collected has been in general agreement with these
ideas and concepts (LBL96). Most subsequent conceptualizations of unsaturated zone behavior
are largely refinements of this model, revised to accommodate newly-acquired data. Newly-
acquired data include isotopic analyses, concentration ratios of ions dissolved in matrix rocks and
perched water zones, calcite fracture fillings, and thermal modeling of vertical temperature
gradients. Perhaps the most significant change from early conceptual models has been the recent
acquisition of new isotopic data which indicate the presence of "fast paths" for water moving
through the unsaturated zone. This topic is discussed in more detail in a subsequent section.
VI-6
-------
The following presentation of the unsaturated zone flow conceptual model is taken primarily
from USG84. Where appropriate, the published literature is referenced when describing
refinements or revisions that have been made to the USG84 model. The conceptual model is
presented as if it were an established physical reality. One must bear in mind, however, that the
proposed model is probably not the only reasonable description that could be made of the
system. Following the description of the conceptual model is a discussion of critical unknowns,
their effects on unsaturated zone flow, and results of numerical modeling studies.
Percolation of infiltrated water through the exposed fractures of the Tiva Canyon welded unit is
relatively rapid because of the large fracture permeability and small effective porosity of this unit
compared to the alluvial material. Therefore, a large proportion of the infiltrated water normally
is percolated sufficiently deep within the fractured tuff to be unaffected by the evaporation
potential that exists near the surface. Depending on the intensity of the infiltration, percolation
downward through the Tiva Canyon welded unit may occur without a significant change in rate.
A small proportion of the water percolating through the fractures slowly diffuses into the matrix
of the Tiva Canyon welded unit. Downward flow in the matrix is very slow because of the small
effective hydraulic conductivity of the matrix. During dry periods, some of the diffused water
flows back into the fractures and probably reaches the land surface by vapor diffusion. The mass
of water involved during this process is likely to be negligible compared to the mass of
percolating water.
The densely fractured Tiva Canyon unit, with small matrix porosity and permeability, overlies
the very porous, sparsely fractured Paintbrush unit A marked contrast in material properties
exists at the contact between these two units; depending on the magnitude of the infiltration flux,
this contrast could impart a significant lateral component of flow. Flow of water through
fractures of the Tiva Canyon unit occurs rapidly until it reaches the contact. At this point, the
velocity is significantly decreased because of the greater effective porosity and lesser hydraulic
conductivity of the Paintbrush unit. As a result, lateral, unsaturated flow of water above this
contact can occur. Perched water may occur above this unit if displacement along faults has
created significant differences in permeability on opposite sides of the fault.
The saturated hydraulic conductivity of the Paintbrush nonwelded unit in the direction of dip is
10 to 100 times greater than saturated hydraulic conductivity in the direction normal to the
bedding plane. The combination of dipping beds and differences in directional permeability
VI-7
-------
creates a downdip component of flow. The magnitude of this component depends on the
magnitude of the principal hydraulic conductivity ratio. The permeability contrast may be
sufficient to decrease vertical percolation into the underlying Topopah Spring welded unit to
almost zero. In this case, water would flow laterally downdip until structural features are
encountered that create perching conditions or provide pathways for vertical flow,
As water moves downward through the PTn, the effect of high porosity and low fracture density
progressively moves water from fractures into the matrix. Except for areas where fast paths may
exist (such as faults), beyond a certain depth in the PTn, flow may be almost entirely in the
matrix. Travel times through the matrix of the PTn are thought to be relatively long because the
matrix of this unit appears to act as a "sponge" which dampens out episodic infiltration pulses.
Water flows from the matrix of the Paintbrush nonwelded unit into the fractures or matrix of the
underlying Topopah Spring welded unit. Owing to the thickness of this unit, it is hypothesized
in ROB96 that water moving through the fractures eventually diffuses into the matrix and moves
very slowly downward. An exception is the second subunit of the TSw (ROB96). In contrast to
this conceptualization, the unsaturated zone expert evaluation panel estimated that up to 95
percent of the flow in the TSw could remain in the fractures (GEO97).
Flow enters the Calico Hills nonwelded unit either from the matrix of the Topopah Spring
welded unit or through structural flowpaths. How much flow occurs in the fractures of the lower
part of the Topopah Spring unit is unknown, and therefore their potential to contribute to flow
into the Calico Hills unit is also uncertain.
The nature of flow at the contact between the Topopah Spring welded unit and the Calico Hills
nonwelded unit depends on whether the vitric or zeolitic facies of the Calico Hills unit is present.
The permeability and effective porosity of the vitric facies are much greater than those of the
matrix of the Topopah Spring unit, which may result in a capillary barrier where those units are
in contact. Conversely, the permeability of the zeolitic facies is about the same as for the matrix
of the Topopah Spring unit, resulting in continuity of matrix flux across such a contact.
Flux within the Calico Hills unit may occur with some lateral component of downdip flux,
because of the existence of layers with contrasting hydraulic conductivity in the unit. A large
VI-8
-------
scale anisotropy probably is caused by intercalation of tuffs with alternately large and small
permeability and by compaction.
Water that flows downdip along the top of the Calico Hills unit slowly percolates into this unit
and slowly diffuses downward. Fracture flow is known to occur near the uppermost layers of the
Calico Hills unit, but diffusion into the matrix may remove the water from the fractures deeper in
the unit and thereby limit flow mostly to within the matrix, except along the structural flowpaths.
It is possible, however, that fractures provide significant avenues for rapid flow through this unit.
Beneath the southern part of the block, the Crater Flat unit occurs between the Calico Hills unit
and the water table. This includeds the welded part and underlying nonwelded part of the
Bullfrog Member of the Crater Flat Tuff.
Fluxes along many structural flowpaths are probably larger than within the units they intersect.
The Calico Hills unit is more ductile than the overlying Topopah Spring unit, which may give
the Calico Hills unit fracture sealing properties. In addition, because of the lesser shear strength
of this unit compared to that of the Topopah Spring, gouge formation along faults and shear
zones is more common. These properties may result in a smaller fracture conductivity in the
Calico Hills unit. In the case where the structural flowpaths are hydraulically continuous across
the upper contact of the Calico Hills unit, water would be more likely to flow downward without
a significant change in its path until it reaches the water table. In cases where the structural flow
paths are discontinuous across the upper contact, flow may be diverted downdip along this
boundary. Intermediate conditions between the two extreme cases are also possible. Recent
numerical modeling (LBL96, ROB96) of flow through the unsaturated zone has provided
important insights into the possible characteristics of flow in each subunit of the unsaturated
zone. Some of these insights are discussed in the following paragraphs.
Several conceptual models of unsaturated-zone flow at Yucca Mountain have been considered
for TSPA-VA. Past TSPAs have focused on the use of equivalent continuum models (ECMs).
The strength of the ECM is that it can describe observed matrix saturations at Yucca Mountain.
Two problems with ECM are (1) the forced-pressure equilibrium causes capillarity of the small
pores to overwhelm gravity-driven flow in the fractures, leading to inaccurate descriptions of
disequilibrium situations; and (2) it is computationally inefficient in solving time-variable flows.
The generalized-equivalent continuum model (GECM) is very similar to the ECM except that a
matrix saturation value less than one is prescribed to increase flow through the fractures. The
VI-9
-------
GECM solves the first ECM problem to some extent, but it suffers from lack of data to define the
fracture-flow threshold and the DOE indicates that it is not clear whether it is valid under
hydrothermal conditions.
The Weeps model, used in TSPA-93 (WIL94) is a simplified stochastic discrete fracture model
that only considers flow through fractures (similar to the discrete fracture conceptual model). It
apparently predicted the fast paths observed in the ECF (as indicated by elevated 36C1/C1
occurrences) and can describe observed flow-channel spacing at Rainier Mesa (CRW96).
The DOE cites two problems with the Weeps model: (1) it ignores the rock matrix and thus any
potential performance impact of flow in the matrix and (2) much of the data that it requires have
not been collected and might be difficult or even impossible to collect. This model was not used
in TSPA-VA because the DOE favored a process-based model that could the calibrated to
available site data such as borehole data and perched water data.
Another important flow conceptualization is the dual-permeability model (DKM). The DKM
allows computation of flow in pressure disequilibrium in matrix and fracture continua, and the
DOE believes it to be a reasonable compromise between the ECM and Weeps models. The
DKM conceptual model has the flexibility to represent almost the entire range of possible flow
behavior though variation of the fracture-matrix coupling parameter, allowing its behavior to
change continuously from the ECM (which is dominated by matrix flow) to a Weeps-type
flow almost entirely within the fracture network. Because of its flexibility and ability to model a
broader range of unsaturated flow problems, the DKM conceptual model was used in TSPA-VA.
However, the DOE points out that the DKM has its own problems, including (1) less
computational efficiency than the ECM and (2) lack of data describing the coupling term
between matrix and fractures.
In addition to the conceptual model for fracture-matrix partitioning, the TSPA-VA conceptual
model of flow in the UZ at Yucca Mountain includes an extensive perched water zone located
between the repository horizon and the water table. The perched water exists because of a low
permeability region that diverts flow laterally around the perched water region. Faults are also
incorporated into the conceptual model of unsaturated flow and are believed to be pathways of
fast flow from the surface down to the water table, giving rise to observed "bomb pulse" near
faults in the ESF.
VI. 1.1.4 TSPA-VA Abstraction Approach and Implementation
VI-10
-------
The DOE notes that the abstraction approach and implementation of UZ flow models for TSPA-
VA were motivated by components and issues of UZ flow and UZ-flow modeling that have been
identified as potentially important to PA calculations of Yucca Mountain. As noted previously,
important issues addressed in the abstraction and testing analyses fall into four areas: (1) climate,
(2) infiltration, (3) mountain-scale UZ flow, and (4) drift-scale seepage.
For TSPA-VA, uncertainty in future climates was represented by including uncertainty in the
infiltration rates. DOE used the UZ flow model to calculate flow fields, using the expected
present-day infiltration map, as well as to calculate variations on the expected map. The present-
day infiltration map was multiplied and divided by three in the base case to include uncertainty
in the infiltration values. The infiltration maps corresponding to the long term average and
superpluvial climates were similarly divided and multiplied by three in the base case to develop
lower-bound and upper-bound infiltration rates. The UZ flow model was used to generate flow
fields for these cases, and sampling of these flow fields in TSPA-VA calculations was weighted
the same as for the present-climate flow fields. Sensitivity studies also considered increased
uncertainty in the infiltration maps by using factors of five instead of three.
Because of the time scales involved, a measure of uncertainty in mean climate precipitation is not
presently available; in TSPA-VA, interannual precipitation variability is used as an estimate of
climate precipitation uncertainty. Interannual variability (defined as one standard deviation from
the mean) in precipitation in the State of Nevada over the last 100 years has been between 20%
and 30% (DEW93). Arid and semi-arid regions can have interannual variability of 50%
DEW93). A value of 50% uncertainty in mean climate precipitation was assumed for TSPA-VA.
Therefore, although the mean precipitation for the long term average climate is estimated to be
300 mm/yr, it could be as low as 150 mm/yr (0.5 x 300 = 150) and as high as 450 mm/yr (1.5 x
300 = 450). In TSPA-VA, the estimated range of uncertainty in mean precipitation is relatively
consistent with the estimated range of uncertainty in infiltration rate.
A quantitative characterization of the spatial and temporal distribution of net infiltration is
needed for defining upper boundary conditions for site-scale, UZ ground-water flow models in
TSPA-VA. Net infiltration is defined as the downward rate of water percolation immediately
below the zone of evapotranspiration; it is not necessarily equivalent to the rate of recharge to the
underlying SZ. A site-scale net infiltration model was developed to provide temporally and
VI-11
-------
spatially detailed estimates of net infiltration rates over the area of Yucca Mountain (BOD97).
The net infiltration model is primarily a deterministic model of surface and near-surface
hydrologic processes, although climatic input involves both deterministic and stochastic
processes. All major components of the water balance are solved in daily time increments.
Daily results are provided for specified locations for analyzing the temporal distribution of net
infiltration. Estimates of present-day and potential future net-infiltration rates are provided as
detailed mappings of temporally and spatially varying time-averaged fluxes, which can then be
applied to define upper-boundary conditions for the UZ.
VI. 1.1.5 Model Calibration
The initial version of the infiltration model was calibrated using the 1984-1995 record of
measured water-content profiles from approximately 80 neutron access boreholes and a
developed record of daily precipitation for 1980-1995 (FLI96). Records from boreholes
identified as potentially problematic because of the accelerated downward percolation of water
along the annular space were excluded. Model calibration consisted of both qualitative and
quantitative comparisons of measured versus simulated water-content changes for the soil
profile.
VI. 1.1.6 Unsaturated Zone Flow
The DOE would have preferred not to have had to abstract a model from the unsaturated zone
process level model (BOD97). The DOE initially expected that the model would be simplified
probably by reduction to two-dimensional or even one-dimensional geometry. The site-
characterization investigators and modelers strongly recommended that three dimensions were
important to represent Yucca Mountain flow adequately. This recommendation was based
primarily on the flow below the potential repository, where the DOE believes there is significant
nonvertical flow because of heterogeneity in the locations of the zeolitic layers and perched
water. With the direct use of the process flow model, there is no need for testing of abstractions
against the process model. The models are tested directly against the data as part of the
calibration procedure (i.e., each case must be calibrated).
An important consequence of using a complex three-dimensional flow model is that the number
of different cases that can be run is limited by computer-processing tune, but even more so by the
VI-12
-------
time needed for analysts to make necessary adjustments by hand for each case (to ensure proper
model calibration). However, the DOE's current approach uses several select conceptual models
and parameter sets that were determined (by sensitivity analyses) to have a significant impact on
performance in order to encompass the range of uncertainty hi UZ flow. The DOE has found
that aqueous travel times between the repository and water table can vary by several orders of
magnitude between different conceptual models.
The three-dimensional process flow model inherently contains several specific assumptions and
issues that are listed below:
• Dual-permeability flow modeling (i.e., coupled matrix and fracture continua) is
adequate. The DKM is capable of representing a large range of potential UZ-flow
behaviors (e.g., fracture-matrix interaction, fracture and flow-channel spacing and
geometry effective fracture apertures, and fracture- and matrix-flow velocities all vary
greatly as the model parameters vary). In future work, the DOE may consider
alternative models (e.g., discrete fractures, fractal fractures, etc.) to complement dual-
permeability models, but alternatives were not considered for TSPA-VA.
• Steady-state flow modeling is adequate. Climate changes have been included by
using a series of steady states; because of that, the flow could be said to be quasi-
steady state rather than steady state. Perturbations to flow caused by repository
heating were neglected. Such thermohydrologic perturbations were considered only
in sensitivity cases because the waste packages are expected to last through the period
when flow is strongly perturbed.
• Hydraulic properties of the matrix can be represented by the range of laboratory
measurements. In some cases matrix properties were adjusted to get better fits to
matrix-saturation measurements or other data.
• Fracture hydraulic properties can be derived from air permeabilities, fracture
frequencies, and fracture orientations measured in drill holes and in the ESF. In some
cases, the inferred fracture properties (van Genuchten alpha) were adjusted
significantly in order to get better fits to matrix-saturation measurements or other
data.
• The van Genuchten/Maulem functional form is satisfactory for use to represent the
saturation/desaturation behavior of both matrix and fractures.
• The fracture-matrix connection area (i.e. area available for flow between fractures and
matrix) is reduced below the geometric area implied by the fracture spacings used.
VI-13
-------
Physically, this reduction represents effects of channelization of flow in fractures.
The amount of reduction was chosen to optimize the fit to matrix-saturation
measurements or other data.
• Heterogeneity within a unit does not need to be included. Hydrogeologic units are
homogeneous.
• Infiltration at the surface is spatially variable, with the variability given by data in
FLI96. Sensitivity to infiltration was investigated by multiplying the infiltration
distribution by a constant factor keeping the same spatial variability.
• The lower boundary of the model is at the water table, which is fixed by drill-hole
observations. For future climates, the water-table elevation is increased by prescribed
amounts.
VI. 1.1.7 Base-Case Hydrologic Properties Used in TSPA-VA
Model calibration allowed the DOE to develop what they believed to be the most reasonable
estimate of parameters a be used with the UZ model for both liquid and gas flow. It was a
combination of the "matrix" and "fracture" parameter sets and was named the "preliminary base
case."
The base-case parameter sets used fracture-matrix multipliers that were calibrated to global
classifications of welded, nonwelded, and zeolitic stratigraphic units. Together with the
variations in present-day infiltration and ranges in fracture parameters, the base case consisted of
five calibrated parameter sets:
• Base infiltration -K3 and the van Genuchten air-entry parameter at a minimum for each
layer
• Base infiltration + 3 and the van Genuchten air-entry parameter at a maximum for
each layer
• Base infiltration and the van Genuchten air-entry parameter at the nominal "best
estimate for each layer
• Base infiltration x 3 and the van Genuchten air-entry parameter at a minimum for
each layer
VI-14
-------
• Base infiltration x 3 and the van Gemichten air-entry parameter at a maximum for
each layer
VI. 1.1.8 Recommendations for Development of Future Parameters
One of the DOE's contractors (LBNL) recommends that the project pursue three-dimensional
inversions using available parallel-processing capabilities to minimize the number of
assumptions in the inversions (such as the use of one-dimensional submodels that do not capture
the perched-water effects). The contractor also recommends that the three-dimensional
inversions add data, such as temperature and geochemical measurements, to explicitly constrain
infiltration rates, fracture/matrix equilibrium and travel times during the three-dimensional
inversions. Use of these three-dimensional inversions would increase the defensibility of the
current calibration process. In addition, the creation of a heterogeneous property set should take
advantage of the considerable data concerning parameter uncertainties developed in the one-
dimensional models and with the heterogeneous distributions provided by Rautman and
MCKenna (RAU97) to generate a stochastic representation of the parameter fields in the site-
scale UZ model.
VI. 1.1.9 Sensitivity Studies for Determining Important Hydrologic Properties
The DOE performed a series of UZ simulations to examine the sensitivity of water flow to matrix
and fracture permeabilities and van Genuchten properties in a one-dimensional, dual-
permeability system. These studies, in part, formed the basis for the DOE's choice of parameters
that were used in the base case, as described in a previous section. These simulations are divided
into two sets. In the first set, ranges in property values were defined using the properties and
mean and standard-deviation values determined from inverse modeling done at LBNL (BOD97).
Results from these simulations show that, for the ranges of values considered, the fraction of
infiltrating water that travels downward through the fracture continuum is primarily controlled by
fracture alpha and to a lesser extent by fracture permeability. In this first set of simulations, the
range of fracture alpha values considered was quite large relative to other property ranges and, as
a result, its impact on flow behavior was most significant. To reduce the uncertainty in fracture-
alpha values, the DOE conducted a subsequent study to derive more reasonable ranges of fracture
alpha values. For this study, The DOE relied on published fracture permeability and fracture
frequency data.
A second set of simulations based on the new fracture-alpha ranges was then performed. In this
set of simulations only two properties, fracture alpha and fracture permeability, were varied over
VI-15
-------
their ranges. There were two important differences between this set and the first set of
simulations: (1) a Weeps formulation was used; that is, the matrix-fracture conductance area was
reduced by the fracture relative permeability to water; and (2) in addition to using the new
fracture-alpha values, different minimum and maximum values of fracture permeability that are
consistent with the new fracture-alpha values were also used. Results of this study show that, for
the ranges of fracture permeability and fracture alpha values considered, the fraction of
infiltrating water that travels downward in the fracture continuum is controlled primarily by
fracture permeability.
VI. 1.1.10 Fracture-Matrix Interactions
Dual permeability models have been used to explicitly model unsaturated ground-water flow and
heat through both fractures and matrix. In the DKM, the fractures and matrix are treated as
separate discrete continua. Heat, gas, and liquid are allowed to flow between the fractures and
matrix, as well as through each continuum. While the DKM is generally more applicable to a
wider range of problems that the ECM (e.g., transient flows, high infiltration boundaries), the
DKM requires additional information about the coupling between the fracture and matrix
continua.
VI. 1.1.11 Analysis of Perched Water
Incorporation of perched water data is an important aspect of the UZ-flow-model calibration.
The presence of perched-water bodies implies that vertical water fluxes locally exceed the
saturated hydraulic conductivities of the perching layers. In order to capture the perched-water
phenomena, the UZ-flow model must have a representative geologic/conceptual/trapping model,
fracture/matrix properties, and sufficient net infiltration. The resulting model should reproduce
hydraulic responses in pumping rests and remain consistent with geochemical data and the area!
extent of the perched body.
Perched water can be defined as a SZ not directly connected to the static water table (FRE79).
Two criteria must be met for the model to accurately reproduce perched water at Yucca
Mountain. The first is that water saturation within a perched-water zone must be sufficiently
high to initiate substantial fracture flow (if any fractures are present). The second is that water
pressure within the perched-water volume must have values higher than the static atmospheric
VI-16
-------
gas pressure that would be expected at the same elevation. Under these conditions, water will
flow freely into a borehole intersecting a perched-water body. Perched water may accumulate
where large contrasts in hydraulic conductivity exist between adjacent formation units, where a
permeable layer overlies a relatively impermeable layer, or where a well-connected fractured unit
overlies a locally unfractured or poorly connected fractured unit. Relative fracture density and
fracture permeability have a strong influence on the accumulation of perched water Perched
water may also exist against a fault along a dipping horizontal plane if the fault acts as a barrier
to downdip water flow.
As the DOE points out, one of the implications of an existing perched-water body is that the flow
path may not be vertical through the UZ to the water table, but rather the water may be diverted
laterally to a fault zone or other type of higher permeability channel in order to reach the water
table. As a result, a nonuniform recharge rate to the water table is expected. The DOE notes that
this has important implications for waste isolation at Yucca Mountain. Existence of perched-
water zones alone the base of the Topopah Spring welded unit implies that water may partially
bypass the underlying zeolitic unit, and consequently some radionuclides may not be retarded by
the highly sorbing zeolites.
Bodvarsson et al. (BOD97) detail how the field observed, perched-water data at the Yucca
Mountain site were compiled, analyzed, and incorporated into the thee-dimensional UZ flow
model. A conceptual model of occurrences of perched water was discussed, and a series of
comprehensive computer modeling studies on perched water at the site was completed. A three-
dimensional UZ perched-water flow model was then developed to investigate the perched-water
phenomena at Yucca Mountain.
VI.1.1.12 Geochemical Analyses
Bodvarsson et al (ibid.) discuss the efforts made in calibrating the flow model to geochemical
data. These analyses also provide methods of developing bounds and ranges for percolation flux
and infiltration.
VI-17
-------
VI. 1.1.13 Interface Between Unsaturated Zone Flow and Unsaturated Zone Transport
TOUGH2 (PRU91) and FEHM (ZYV97) are two prominent codes for evaluating flow and
transport in the UZ for performance assessments. The application of TOUGH2 has focused on
site-scale UZ hydrology (BOD97), while the use of FEHM has focused on fluid flow and
radionuclide transport (ROB97) at Yucca Mountain. Both UZ hydrology and radionuclide
transport are critical components in performance assessment calculations, and methods of
coupling these components were investigated for TSPA-VA.
The DOE considered two methods to transfer the UZ flow results to the UZ radionuclide
transport model, which was chosen to be the particle-tracking method in the computer program
FEHM (ZYV97): (1) use the UZ flow fields calculated by TOUGH2 directly as input to the
FEHM particle tracker or (2) take the stratigraphy and calibrated hydrologic parameters and use
them as inputs to a combined flow and transport calculation within FEHM. According to the
DOE, the primary advantages of the first option are that preservation of the UZ-flow calibration
is assured, and it is not necessary to recalculate the flow and recheck the calibration. The
primary advantages of the second option are that the FEHM particle tracker is already set up to
use flow fields calculated by FEHM (the first option requires development of a linking program
to take TOUGH2 output and generate FEHM input), and there is additional flexibility to refine
the computational grid to make the transport calculations more accurate. The first option (flow
fields calculated with TOUGH2, radionuclide transport calculated with FEHM), has been
implemented in TSPA- VA.
The particle-tracking method used in FEHM is a cell-based model in which particles are routed
from grid block to grid block in a manner that preserves the overall residence time through any
portion of the model and probabilistically reproduces the migration of a solute through the
domain (ROB97). Flow calculation is based on a control volume in which fluid-flow rates into
and out of each cell are computed. Since TOUGH2 is an integrated, finite-difference code and
FEHM employs a control-volume, finite-element technique, the two codes are compatible from
the standpoint of implementation of the particle-tracking technique. The required inputs for
FEHM to use an externally developed flow field are: (1) grid-connectivity information and cell
volumes; (2) fluid-state variables for computing density, fluid saturation, and rock porosity at
each grid point; (3) internodal fluid-mass-flow rate for every connection in the numerical grid;
and (4) fluid source and sink flow rates for each grid block.
VI-18
-------
VI. 1.1.14 Grid and Model Domain for Three-Dimensional Site-Scale Unsaturated Zone Flow
Model
Design of the-three-dimensional grids was based on a geological framework model. The three-
dimensional numerical grids were designed in two steps: (1) using all available surface
information to create a horizontal, two-dimensional grid, and (2) integrating data from isopach
maps of hydrogeological units to vertically develop the horizontal grid between the ground
surface and the water table. This model was a further update of the existing three-dimensional
UZ model first developed by Wittwer et al. (WIT92), which was later updated in 1996 (BOD96).
The development of this new model, like the old one, started with the definition of the grid block
centers that defined, a single two-dimensional horizontal grid to use in all the three-dimensional
vertical layer sections. After all the nodal points had been located, a numerical grid generator
AMESH was used to develop the two-dimensional horizontal grid. These two-dimensional
horizontal grids and the sub-layering criteria were used to develop a new three-dimensional
numerical grid of the UZ at Yucca Mountain.
New boundaries for the site-scale model were developed as an update to the 1996 three-
dimensional LBNL UZ model. These boundaries were based on revised fault maps, the observed
shift in water level across the Solitario Canyon, the new infiltration and alluvium thickness maps,
the observed pneumatic signals that were observed during the construction of the ESF Tunnel,
and the requirements for thermal loading studies. The new boundaries take into account the
extensive high-gradient SZ to the north of G-2 and explicitly include the Solitario Canyon Fault
in the west, by extending the model boundaries to about 1 km west of Solitario Canyon fault.
The Bow Ridge fault forms the eastern boundary. The model extends from borehole G-3 in the
south to about 1.5 km north of G-2 in the north. These boundaries enclose most of the existing
and planned hydrology wells and the wells in which extensive moisture tension data and
lithology are used as calibration points for formation properties. The 1997 site-scale model was
based on the fault map that provided explicit offsets on Solitario Canyon Fault, Ghost Dance
Fault, Ironridge, and the Dune Wash Fault, defined by the base of the Tiva Canyon.
One primary objective in the selection of grid boundaries was to minimize boundary effects
resulting from thermal loading at the repository horizon, while investigating the influence of the
major faults on the hydrological and thermal-hydrological response of the UZ at ambient
conditions and during thermal loading. The modeled area covers nearly 43 km2 and is bounded
by Bow Ridge fault on the east; extends 1 km west of Solitario Canyon: is bounded by the
plateau of high pressure gradient about 1.5 km north of G-2; and extends about 1 km south of the
ESF south ramp. In this grid, the "East Block Repository" area is modeled as a locally refined
VI-19
-------
area with an average grid of 100 x 100 m and accounts for proposed extension of the potential
repository to the north. The resulting two-dimensional grid contains a maximum of 1,470 aerial
grid-block nodes in each layer. The two-dimensional grid extends about 1 km west of the
Solitario Canyon in order to explicitly model both the Solitario Canyon Fault and its associated
Iron Ridge fault branch to the south. Explicit modeling of the area west of the Solitario Canyon
also allows for specification of the 40-m shift in the water table west of this fault. This grid was
used to perform general site-scale UZ modeling and for detailed studies related to thermal
loading of the repository.
VI. 1.1.15 DOE Recommendations for Future Work
The DOE asserts that there is a need to harmonize the differences between the grids based on the
USGS geological framework model and the geological model in order to select a single
geological model for designing future numerical grids. Work is currently underway to integrate
the ISM 3.0 geological-framework model into the site-scale UZ-flow model. Integration of the
ISM 3.0 geological-framework model with the site-scale UZ-flow model will increase the
defensibility of the UZ modeling effort.
VI. 1.1.16 Summary of Implementation of the Base-Case Unsaturated Zone-Flow Model in
TSPA-VA
The previous sections have detailed the development of the UZ-flow model used in TSPA-VA
calculations. Sensitivity analyses have been performed to provide basis for parameters and
divided by 3 and multiplied by 3 to yield three present-day infiltration scenarios. In addition,
the DOE combined these present-day infiltration scenarios with variations of the fracture air-
entry parameter to yield five base-case parameter sets. For each set, two future-climate scenarios
were considered by using long term average infiltration maps. These were either divided by 3 or
multiplied by 3 to correspond to the value used in the present-day infiltration scenario. For all
UZ groundwater flow simulations, the EOS9 module of TOUGH2 has been used. This module
implements Richards' equation and assumes that the gas phase is passive.
Once the flow fields have reached steady state, as indicated by a global mass balance within 1%
error, the flow fields are used for mountain-scale transport calculations and near-field seepage
studies. All developed data that are fed to other TSPA components are submitted to the
Technical Data Base. These components are integrated for TSPA-VA calculations using the RIP
code.
VI-20
-------
VI. 1.1.17 Sensitivity to Mesh Resolution
All simulations aside from the mesh resolution study use a grid discretization of 0.5 m on a side
(i.e., the numerical flow grid has the same resolution as the generated geostatistical permeability
field). To study the sensitivity of the simulation results to the numerical grid resolution, the
DOE performed additional simulations, with refined grids for a two-dimensional vertical cross
section perpendicular to the drift center line. The original 0.5-m x 0.5-rn discretization is refined
by dividing each grid block first into 4 sub-blocks and then into 9 sub-blocks. The sub-blocks
are assigned the same permeability value as the original grid block (i.e., the heterogeneity
structure remains unchanged). Results indicate that there is some sensitivity to the grid design,
as the derived seepage rates increase when using a finer grid resolution. The DOE believes that
this is mainly because the gradients between elements of different permeability are steeper in a
simulation with fine sub-gridding, while a simulation with the original grid—identical resolution
of heterogeneity field and simulation grid—has smoother transitions as a result of the harmonic
weighting of the two neighboring permeability values at element interfaces. The DOE notes that
consideration must be given to the fact that an assumed step-change of permeability in the
generated random fields is only an approximation of the more smooth transition in natural
domains. Therefore, the DOp believes that the original grid design may actually allow for a
reasonable representation of natural heterogeneity. The DOE further asserts that if the
permeability values for the refined grids were derived by interpolation from the underlying
random field to smooth out the strong step-changes, the impact of the mesh refinement would
probably be small. DOE also notes that more work along this line will be performed in future
studies.
VI.1.1.18 Summary
Following is DOE's summary of important points from the base-case seepage model.
A. Heterogeneity in the flow domain is critical for the calculation of seepage. It causes
channelized flow and local ponding, so that the probability of seepage is much larger,
and the time required for seepage into the drift is much shorter, than for the case
where the flow domain is assumed to be homogenous.
B. The conceptual model for the interaction between fractures and matrix (i.e., X^) is
important for transient flow in the case of episodic percolation events. Field-scale
studies like the niche experiment can provide information on the "effective X^,"
VI-21
-------
value; however, in considering long-term climate changes, the primary interest is
seepage under steady-state conditions, so that fracture-matrix interaction need not be
considered.
C. In general, seepage time decreases and seepage rate increases with an increase in
percolation rate. The relationship between seepage and percolation is not linear,
because of the many nonlinear processes involved.
D. Variation among geostatistical realizations is significant (though within the same
order of magnitude) and dependent on details of local heterogeneity around the drift.
Since such details are not known, a stochastic approach is necessary.
E. Comparison between two-dimensional and three-dimensional runs indicates that the
probability of seepage and the seepage-initiation times are similar. The relative
seepage rate, however, appears to be different hi the three-dimensional runs, due to
the possibility of flow in the third dimension.
F. The three-dimensional runs offer the opportunity for evaluating the possible spatial
distribution of seepage along emplacement drifts, which cannot be achieved using
two-dimensional vertical cross sections. The spacing of seepage locations is
dependent on the correlation lengths of spatial heterogeneity.
G. Seepage is insensitive to the van Genchten p parameter of the fractures. Seepage is
sensitive to fracture a and permeability. For steady state, seepage is not sensitive to
the matrix hydrologic properties. For transient problems, and for percolation rates
lower than those considered here, matrix properties may be more important.
H. There are important questions about the effects of the discrete nature of fracture flow
especially the possible role of fractures that dead-end at the drift wall. The
preliminary niche-test results appear to fit the base-case conceptual model well, but
more analysis (and testing) is needed.
VI. 1.1.19 Recommendations
This chapter has detailed the DOE's development of four major components of UZ flow: (1)
climate, (2) infiltration, (3) mountain-scale UZ flow, and (4) drift-scale seepage. Issues
associated with each component have been presented, along with DOE's abstraction/testing plans
that address these issues. The following sections summarize each of the four major UZ-flow
components. The impact of each component on performance, along with the DOE's guidance
and recommendations for the license application, are also provided.
VI-22
-------
Climate
The primary purpose of climate modeling was to provide precipitation rates and water table
elevations that varied as a function of future climates. Future climate was modeled in TSPA-VA
as a sequence of discrete steady states. Only three discrete climate states were considered for
TSPA-VA: present-day, long-term average, and superpluvial. Present climate represented
relatively dry, interglacial conditions, while the long-term average represented an average pluvial
period at Yucca Mountain. The superpluvial represented periods of extreme wetness. The mean
annual precipitation (MAP) rates for the present, long-term average, and superpluvial climates
were estimated to be 150,300, and 450 mm/yr, respectfully. These values were used by
infiltration modelers to determine appropriate analog sites that had average precipitation rates
that were commensurate with the predicted future climate values. The water-table rise from the
present-day level (-730 m) was estimated to be 80 m and 120 m for the long-term average and
superpluvial climates, respectively. Sensitivity analyses have shown that the overall performance
is not sensitive to the duration of the climate cycles. The most significant impact was found to be
the abrupt changes in water-table elevation and groundwater flow rates that occurred at the
transition between climates.
Climate models strongly impact performance through their influence on precipitation and
evapotranspiration. These factors, in turn, influence the predicted infiltration in the UZ flow
model. Therefore, the magnitude and timing of the prescribed climate slates is important to
performance.
The DOE believes that additional work is needed to understand the natural variability of current
and future climates for Yucca Mountain. In particular the DOE feels that the adequacy of three
distinct climate states needs to be addressed further. If distinct climate states are used in future
analyses, their number, timing, duration, and the abruptness of the transition between them need
better support. Additional modeling is needed to determine how the dose-rate pulses depend on
the time of transition between climates, and whether noninstantaneous transitions would lead to
lower peak doses. Appropriate climate analogs need to be defined, based on temperature and
other factors in addition to precipitation. The superpluvial climate, especially, needs better
definition.
VI-23
-------
Infiltration
Infiltration modeling provides the spatial and temporal distribution of net infiltration as an upper
boundary for site-scale UZ-flow models in TSPA-VA. Distributed net infiltration rates were
determined for each of the three climate states using the YMP infiltration model. The infiltration
model simulated water movement at the ground surface by solving water mass balances using
precipitation, a model for evapotranspiration, and available water in the soil profile. Also
considered in the model were ground surface elevation, slope, bedrock geology, soil type, soil
depth, and geomorphology. The primary driver for the infiltration model was precipitation,
which was input using available records or, in some cases, a stochastic model. Daily
precipitation records from different locations were used to define the present-day, long-term
average, and superpluvial climates in the infiltration model. The sites were chosen based on how
well their MAP values matched the estimated values associated with each climate. General
results of the infiltration model are as follows:
• The modeled infiltration is highly heterogeneous and clearly correlated with
topographic features
• The highest net infiltration occurs along Yucca Crest
• Net infiltration is lower in washes.
The spatially distributed infiltration maps were then upscaled to the site-scale UZ-flow model by
averaging the simulated infiltration values over each surface element in the UZ-flow model. The
DOE used average infiltration for each climate over the UZ-flow model domain of 4.9, 32.5, and
118 mm/yr, respectively.
Sensitivity analyses were also conducted to determine the effects of episodic infiltration on the
percolation at the repository horizon. Results showed that the PTn unit effectively damped
episodic pulses that were simulated on a yearly cycle, preventing the transient pulses from
significantly impacting the percolation at the repository horizon.
Additional sensitivities that used infiltration to estimate the temperature profile in a borehole
indicated that infiltration rates that were greater than three times the average present-day
infiltration rate did not allow good matches with observed borehole temperatures because of
increased advective heat transfer. Therefore, DOE used a factor of three as the upper and lower
bounds for the range of infiltrations considered in each climate scenario.
VI-24
-------
Infiltration strongly affects repository performance because of its influence on mountain-scale
unsaturated-zone flow and seepage into drifts (which subsequently affects waste-package-
degradation models). The infiltration rates used in TSPA-VA are significantly higher than in
past TSPAs. Higher infiltration rates, in general, tend to adversely impact performance.
However, the increased infiltration must be considered in conjunction with other TSPA
components such as seepage to understand the overall impact on performance assessment.
The DOE believes that the greatest need for improvement in infiltration modeling is explicit
inclusion of processes that should be different for future climates, including effects of
temperature, cloudiness vegetation type, surface water runoff/run-on, and snow cover. Even for
current conditions, some experts on the UZ expert elicitation panel suggested that runoff and
on might be more important than is assumed in the infiltration model.
run-
To provide a more quantitative basis for the uncertainty distribution for infiltration, the DOE
notes that the infiltration model should be run in a stochastic mode (e.g., Monte Carlo
simulation) to derive the infiltration uncertainty from the uncertainties in the input parameters of
the model.
Finally, the DOE asserts that analogues with known infiltration such as Rainier Mesa and
Apache Leap, should be used to test and improve the infiltration models and methods.
Unsaturated Zone Flow
The three-dimensional UZ-flow model has been used to calculate unsaturated-groundwater flow
at Yucca Mountain for TSPA-VA. The model implements the dual-permeability formulation for
fracture-matrix interactions and consists of nearly 80,000 elements. Hydrologic properties were
determined using both direct measurements and calibration with field data, which, included core
samples, borehole log data, in situ water potential and temperature measurements, fracture
measurements from the ESF, in situ pneumatic data, air permeability tests, and geochemistry
data. A great deal of information on the calibration and details of the UZ-flow model
development was taken from Bodvarsson et al. (BOD97). In the calibration approach, a number
of vertical one-dimensional submodels that corresponded to borehole locations were extracted
from the three-dimensional model. The code ITOUGH2 was used to perform simultaneous
inverse simulations with these one-dimensional models to optimize hydrologic parameters by
VI-25
-------
matching predicted and observed matrix saturations and moisture potentials. The selection of
hydrologic parameters that were estimated by inverse modeling was influenced by sensitivity
studies that determined important parameters to UZ flow, including the fracture air-entry
parameter, and the fracture-matrix interaction parameter. The properties that were calibrated in
one dimension were then used in the three-dimensional site-scale model, which included
calibrations for perched water. Additional tests using geochemical data, infiltration data, and
alternative weighting schemes were also performed to improve the three-dimensional model and
increase confidence in the methods being used.
Results show that the flow through the UZ is predominantly in the fractures for the welded units
and predominantly in the matrix for the nonwelded units. High infiltrations resulting from
climate changes significantly increased the percolation flux in the vicinity of the repository and
decreased the travel time between the repository and the water table. Travel times between the
repository and the water table ranged from several days to hundreds of thousands of years. The
fastest transit times resulted from flow through-fractures, whereas the matrix contributed to
particle breakthrough at the water table at significantly longer times. Perched water, which has
been calibrated in the three-dimensional flow model, diverted vertical flow laterally in the three-
dimensional model, especially in the northern part of the repository. However, the total travel
time of the diverted water was not significantly altered due to the fast flow path through the
fractures. Sensitivity results using increased infiltration confirmed the importance of infiltration
rates in determining travel times between the repository and the water table. In addition,
sensitivity results using the DKM/Weeps alternative model showed that there was more
significant fracture flow than hi the base case, contributing to faster travel times. Finally,
sensitivity studies of the zeolitic matrix permeability showed that increased matrix permeability
can result hi slower travel tunes due to increased flow through the matrix. However, decreased
matrix permeabilities did not result hi significant changes from the base case.
The DOE believes that a significant result of the current TSPA-VA UZ flow calculations
relative to earlier TSPAs is that the higher estimates of current and future infiltrations can cause
percolation fluxes to be significantly greater and travel tunes to be significantly shorter. While
these effects have a negative impact on performance, the impact of UZ flow in general on the
performance calculations must be determined collectively with other system components. For
example, high infiltrations are thought to be adverse to performance, but performance
calculations have shown that for a period of tune, the high infiltration scenarios show a decrease
in dose. This counter-intuitive result occurs because the temperatures around the waste package
are reduced by the increased infiltration, and the corrosion of the waste packages is reduced.
VI-26
-------
The use of the DKM/Weeps model produced significantly shorter travel times between the
repository and water table because of increased partitioning of flow through fractures. As
demonstrated in PA calculations, the decreased travel times increase dose rates for periods less
than 10,000 years, but for longer times (100,000 and 1,000,000 years), the rapid transport does
not significantly impact performance. At these later times, the travel time becomes small relative
to the total simulation time, and the decreased travel in the DKM/Weeps model is less important.
However, colloid-facilitated transport can be enhanced by increased flow and partitioning in
fractures.
The uncertainty in matrix permeability in the zeolitic units resulted in travel times that could
differ by several thousand years. Sorbing tracers traveling through the zeolitic matrix were
retarded more if the permeability was increased, but little difference was observed if the
permeability was decreased. Because the matrix permeabilities in low-permeability units is
likely to be less than the reported values because of excluded "nondetect" values, the uncertainty
associated with matrix permeabilities may not significantly impact overall performance.
is a
in
The DOE believes that the most important need in the mountain-scale, UZ flow modeling
better representation of localized channeling of flow, and in particular, the effects of flow
discrete fractures. The current approach uses continuum models with very coarse spatial
discretization, and the adequacy of this approach is not fully established. There are indications
from geochemical and isotopic tracers (chloride concentration, 36C1-to-chlorine ratio and '^-to-
carbon ratio) that channeling of flow might be important. In addition, geochemical, isotopic and
temperature data should be integrated into be calibration procedure because such data provide
important information about flow through fractures.
The DOE also believes that more information is needed about the role of perched water in UZ
flow. The current model assumes that the water is perched on a very-low-permeability
underlying layer and flow is forced to go around it. The DOE notes that other interpretations are
possible, such as mixing within the perched water and matrix flow'out the bottom.
Thermal alterations of flow and thermal hydrology (TH)-chemical or TH-mechanical alterations
of hydrologic properties are potentially important. In the current TSPA structure, these effects
fall under the TH-component, but it is necessary to determine whether there should be a coupling
of TH effects on mountain-scale, UZ flow and transport.
VI-27
-------
The DOE technical recommendations for improvement of the current mountain-scale UZ-flow
model include the following:
• Incorporation of the most recent version of the integrated site model (site geologic
framework model)
• More refined numerical grid
• Additional data to gain better estimates of fracture-hydrologic parameters
• Additional inhibition tests of hydrologic properties of the matrix
• Additional measurements of permeability of the zeolitic hydrogeologic units and
properties of faults
• Additional studies to better characterize and understand the effects of perched-water
• Creation of heterogeneous property sets 'that take advantage of the heterogeneous
distributions provided by Rautman and McKenna (RAU97) and a stochastic
representation of the parameter fields in the site-scale UZ model.
Fracture-flow processes should be further investigated with alternative conceptual models and
additional field studies, such as niche and alcove studies, the planned east-west cross drift, and
the Busted Butte transport study. Other flow processes that need to be further characterized
include flow through faults, flow between disparate units (such as at the Paintbrush nonwelded—
Topopah Spring welded and Calico Hills vitric—Calico Hills zeolitic interfaces), and fracture/
matrix interactions. Finally, the process of model calibration can be further improved by
developing two-dimensional and three-dimensional calibrations against field data, which may
require using parallel computing techniques.
Seepage
The abstracted base-case seepage model was based on a large number of three-dimensional
process-model calculations. The process model consisted of a three-dimensional heterogeneous
fracture-continuum field. Three blocks of dimensions 20-m high. 15-m wide and 16.5-m long
were evaluated independently within this continuum. The drift was represented by a horizontal
cipen cylinder of diameter 5 m at the center of the lower part of the block. Simulation grid cells
were defined to be 0.5 x 0.5 x 0.5 m. Fracture properties were obtained from air-permeability
VI-28
-------
tests of the DST in the ESF Thermal Test Alcove 5 and from evaluation of fracture surveys in the
ESF. As a conservative estimate, matrix flow was neglected in seepage simulations used for
TSPA-VA, and sensitivity studies were performed to evaluate the effects of fracture-matrix
interactions. Sensitivity studies were also performed to determine the effects of episodic pulses,
variations to hydrologic properties, and grid refinement.
The process-model results were abstracted by fitting the calculated seepage-fraction and seep-
flow-rate distributions with beta probability distributions for which the mean and standard
deviation are functions of percolation flux in the fractures. The seepage process model has been
tested against recent preliminary data from the ESF niche liquid-release tests, and appears to fit
them reasonably well. Finally, seepage sensitivity studies were performed to investigate reduced
variance of fracture properties and variations to the fracture aperture.
Seepage into the drifts has a significant impact on performance for several reasons. Seepage
controls waste-package degradation because the waste-package corrosion resistance material
(Alloy 22) corrodes only in the presence of liquid water. Following the creation of openings
through the waste package wall, the seepage volume controls the amount of water that can enter
the waste package and dissolve the waste form. The flux of water into and through the waste
package in turn controls the release rate of the solubility-limited radionuclides from the waste
package. The impact of seepage on overall performance has been found to be important for
periods ranging from 10,000 years to 1 million years.
The DOE has identified a number of additional studies that can be addressed in the near future
(or are already underway) that the DOE believes will produce realistic and useful results for
TSPA-LA. They are listed below:
• One of the key factors that control the spacing between drip seepage locations is the
correlation lengths of spatial heterogeneity of the rock unit. The DOE recommends
that a further careful study of the fracture distribution along the ESF should be made
to provide estimates of these parameters. Field data from the ESF niche study can
also yield important information related to this factor.
• The DOE indicates that a more comprehensive parameter-sensitivity study should be
made including sensitivity of drift seepage to the width of the permeability
probability distribution function and the spatial correlation lengths of the
heterogeneous fields. The occurrence of special features, such as long fractures
VI-29
-------
intercepting the drift, should also be studied. The range of situations and property
values used should be representative of the three stratigraphic units in which the
potential repository will reside.
For reliable results, the DOE believes that the study needs to be performed with more
realizations in the sense of a stochastic analysis and with potentially finer grids.
Previous sensitivity studies have indicated that seepage may increase with finer grids.
• Gravity-driven flux hi near-field discrete fractures close to the drift wall may increase
the probability of drift seepage. Additional study of this possibility is needed.
• The DOE recommends that further study on successive percolation pulses should be
carried out, especially since the current calculations seem to indicate that the time
frame for the system to recover to its original initial state after the first pulse is very
long, perhaps as long as hundreds of years.
The ESF niche test is an important first step in verifying seepage models, but the DOE notes that
it is primarily a test of the overall conceptual model of the drift opening acting as a capillary
barrier. The test offers little validation of the calculated values of seepage fraction, which the
TSPA results show to be the most important aspect of seepage—indeed, the most important
aspect of repository performance. Seepage fraction, or the fraction of waste packages contacted
by seepage water, is related to the average spacing of seeps along the drift, which is presumably
related to quantities such as fracture and fault spacing, permeability distribution, and
permeability correlation length. Data on these quantities are needed, but the DOE notes that field
data relating them to seep spacing are required in order to gain confidence in the model.
Even more so than for mountain-scale flow, seepage into drifts is potentially strongly affected by
channeling of flow and discrete fracture effects. The DOE believes that the adequacy of the
current fracture continuum model to represent these effects must be examined, and the DOE
further asserts that the only real way to assess its adequacy is by testing it against field data. The
DOE suggests that the model could be tested against observed seep spacing at analogue sites
such as Rainier Mesa or Apache Leap. The DOE also recommends that additional niche tests
should be conducted in all three repository hydrogeologic units-Topopah Spring, Lower
Lithophysal, Topopah Spring Lower Nonlithophysal, and Topopah Spring Middle
Nonlithophysal (where the first niche test was conducted)--in the east-west cross drift. The main
ESF tunnel does not go through the Topopah Spring Lower Nonlithophysal, but the east-west
cross drift is designed to go through all three hydrogeologic units of the repository.
VI-30
-------
A potentially important issue identified by the DOE that was not addressed is the stability of seep
locations over time. In the present models, seeps are assumed to occur at the same locations
indefinitely, so that a fraction of the waste packages is always wet (the seepage fraction) and the
rest are always dry. If seep locations change with time, more waste packages would be contacted
by seeps, but only for a fraction of the time. This effect could result in more waste packages
failing, but over a longer period of time, which could be important for performance. Thus, the
DOE believes that the consequences of seep movement should be investigated.
Additional needs identified by the DOE are assessments of the effects of episodic percolation
pulses, the potential increase in seepage during drainage of thermally mobilized water, the effects
of chemical or mechanical alterations in hydrologic properties around the drifts, and the effects
of drift collapse or emplacement of backfill.
VI. 1.1.20 Unsaturated Zone Transport
Transport from a potential repository source is affected by the sorptive interactions with the rock
and the degree of contact between radionuclides and the rock matrix. Some radionuclides, such
as "Tc do not sorb. Other radionuclides, such as 237Np, move at a slower rate than a nonreactive
tracer due to moderate sorptive interaction with various rock types. Still others, such as aqueous
242Puare found to strongly interact with all rock and, therefore, are relatively immobile.
Nevertheless, sorptive interaction is only one part of the mechanism needed to retard the
movement of radionuclides. Radionuclides that are transported through fractures cannot sorb
onto the rock matrix without some mechanism that allows the radionuclides to contact the rock
matrix. (Although the DOE believes that sorption onto minerals along fracture surfaces is likely,
difficulty in characterization of this sorption mechanism has lead to the conservative assumption
used for the TSPA-VA of no sorption in the fractures.) For example, many radionuclides have
been found to strongly sorb to zeolitic rock. However, highly zeolitized rock generally has low
matrix permeability, and in some cases, low fracture permeability as well. The low-permeability
character will lead to transport pathways that bypass the zeolitic minerals, due to lateral diversion
or transport through fractures, severely limiting the degree of contact between radionuclides and
zeolitic minerals. Therefore, low levels of zeolitic alteration in the CHn vitric, which do not
severely reduce matrix permeability, are found to have more influence on the transport of sorbing
radionuclides. The effects of lateral diversion and focused flow in certain regions of the potential
repository also tend to reduce the degree of contact between radionuclides and both the Chn
VI-31
-------
vitric and zeolitic rocks. The more evenly distributed percolation flux in the TSw allows for
more intimate contact of radionuclides with rock matrix, through matrix diffusion and advection,
despite the low matrix permeabilities of the TSw.
Sorptive interactions may enhance radionuclide transport if the aqueous species sorbs to mobile
colloids. Colloid-facilitated transport enhances the movement of the aqueous species because the
sorptive interaction with matrix is reduced (hence reducing retardation in transport) and colloids
may tend to move preferentially through the higher-velocity fracture pathways. In addition to
reversible, sorptive type interactions with colloids, radionuclides may also be irreversibly
attached to colloids (e.g., coprecipitation during colloid formation). Isotopes of Pu have been
identified as radionuclides that are likely to be affected by colloid-facilitated radionuclide
transport.
Radionuclides that have little or no sorptive interaction with the welded tuff matrix are expected
to migrate through the TSw relatively quickly due to advective transport along fractures. The
nonsorbing radionuclides travel primarily through fractures except for transport through the CHn
vitric, where matrix flow and transport is expected to dominate. Therefore, transport times to the
water table for nonsorbing radionuclides such as "Tc, and 129I are primarily governed by
transport in the CHn vitric.
Another factor that affects travel time to the water table is the lateral diversion of flow above the
CHn. Although the lateral diversion increases the transport path length to the water table, the
transport pathways are primarily fracture pathways. The diverted percolating flow eventually
finds some pathway to the water table. Therefore, lateral diversion can lead to zones of focused
flow to the water table, where the flow rates may be locally magnified well beyond the flow rates
anticipated for uniform vertical percolation. The travel times for radionuclides transported
through a focused percolation zone will tend to be relatively short, including transport through
the CHn vitric.
The separation of UZ radionuclide transport from UZ flow is a process abstraction used by the
DOE for sensitivity studies and development of the TSPA-VA UZ radionuclide transport model.
The DOE combined the two processes in TSPA-VA calculations by using a set of pre-calculated,
three-dimensional, UZ flow fields computed with the site-scale UZ flow model. These flow
fields are incorporated directly into the 3-D, UZ radionuclide transport model for TSPA-VA.
VI-32
-------
The waste heat released in the potential repository influences UZ flow and temperature behavior.
As for UZ flow, the separation of thermal hydrology and UZ radionuclide transport is a process
abstraction. Since the DOE assumes that the thermal-hydrologic effects of the repository do not
result in any permanent changes to the mineralogic or hydrogeologic conditions of the UZ, then
the effects of the UZ temperature processes on radionuclide transport are assumed to the minor.
The DOE's modeling of this process suggests that the time period during which the temperature
and flow fields are significantly perturbed occurs prior to the release of most of the
radionuclides. Therefore, the DOE has assumed that the thermal-hydrologic effects on the UZ
temperature and flow fields have a negligible influence on radionuclide transport. However, the
DOE points out that the effects of thermal-hydrology are still important for defining the behavior
of the waste package and radionuclide releases from the engineered barrier system. Therefore,
thermal-hydrology is included for these subsystem models.
The DOE also notes that the effects of thermal perturbations can also potentially have long-term
consequences relative to minerals in the UZ and change both hydrogeologic and transport
properties of the system. These types of thermal-hydrologic changes to the system may affect
long-term radionuclide transport in the UZ. Although the DOE indicates that their present
evaluation of these coupled processes is not complete, they have conducted sensitivity
calculations concerning off-normal behavior to address this coupling.
The waste form mobilization process provides the radionuclide fluxes at the emplacement drift
boundary for UZ radionuclide transport calculations. The emplacement drift boundary represents
a spatial-domain abstraction interface between processes that affect radionuclide transport inside
the emplacement drift and radionuclide transport in the UZ. Therefore, the radionuclide fluxes
calculated at the emplacement drift wall are a source term for UZ radionuclide transport
calculations. The DOE expects this source term to provide radionuclide fluxes at the
emplacement drift boundary that will vary as a function of the location in the repository and
time.
Saturated zone flow and radionuclide transport calculations use the results of the UZ
radionuclide transport calculation to assess the migration to the accessible environment. The
water table in the vicinity of the potential repository represents a spatial-domain abstraction
interface between processes that affect radionuclide transport in the UZ and SZ. The
radionuclide fluxes calculated at the water table due to UZ radionuclide transport are a source
VI-33
-------
term for SZ radionuclide calculations. This source term is expected to provide radionuclide
fluxes at the water table that will vary as a function of position on be water table and time.
Although the DOE has used no feedback mechanisms between SZ radionuclide transport and UZ
radionuclide transport, the position of the water table is recognized as a feedback from SZ flow
to UZ radionuclide flow and transport.
Advection
Advection is the movement of dissolved or colloidal material because of the bulk flow of a fluid,
which in this case is water. This key transport mechanism can carry radionuclides through the
approximately 300 m of unsaturated rock between the potential repository and the water table.
Advection is also an important mechanism for radionuclide movement between fractures and
rock matrix. In many of the hydrogeologic units, advection through fractures is expected to
dominate transport behavior, primarily because the expected flow rates through these systems
exceed the matrix flow capacity under a unit, gravitational gradient. Advection through fractures
is fast because of high permeability and low porosity, with few opportunities for radionuclides to
contact rock matrix. A few of the hydrogeologic units have much larger matrix permeability and
are expected to capture most of the fracture flow by advection from the fractures to the matrix,
causing much slower transport velocities and closer contact of the radionuclides with the matrix.
Advective transport pathways result from and therefore follow the flow pathways, which are
predominately downward. However, lateral diversion is expected along hydrogeologic unit
contacts having strong contrasts in rock properties, particularly in areas of perched water. Flow
that is diverted laterally ultimately finds a pathway to the water table through more permeable
zones which may be faults.
The detailed geometry of fractures and matrix pore spaces at Yucca Mountain is far too complex
to be modeled explicitly. On the other hand, it is important to capture the larger-scale spatial
variability, such as differences between welded and nonwelded hydrogeologic units, and the
differences in fracture and matrix properties at the local scale. To show this variability, the DOE
uses a dual-permeability model for fractured rock. In the dual-permeability model, the fractures
and matrix are distinct interacting continua that coexist at every point in the modeling domain.
Each continuum is assigned its own hydrologic properties such as permeability and porosity,
which may also vary spatially. In general, the fractures are modeled as a highly permeable
continuum having low porosity while the matrix is modeled as a much less permeable continuum
VI-34
-------
having higher porosity. The dual-permeability model offers a bimodal approximation to the true
spectrum of fracture and matrix properties. More importantly, the dual-permeability model can
capture the effects of fast pathways for radionuclide transport from the repository to the water
table. This feature is an important improvement over single-continuum models.
The conceptual model for UZ transport is strongly tied to the conceptual model for UZ flow. As
described above, advective transport because of flow is the main transport mechanism that can
move radionuclides from the repository to the water table. Conceptual models for UZ flow are
commonly based on a continuum relationship, known as Darcy's law, which relates volumetric
flow rate and the gradient in hydraulic potential. In the case of fractured, porous rock such as the
volcanic rock that constitutes Yucca Mountain, these continuum relationships are extended to
embrace two coexisting continua, fractures and rock matrix, that interact according to the same
constitutive relationships that govern flow in a single continuum. From the standpoint of UZ
transport the need for explicit and separate representation of fracture and matrix flow is because
of the extreme disparity in transport velocities that can occur in the two continua. Travel times
for radionuclides transported to the water table exclusively in fractures are expected to be about
10 000 times faster than travel times for radionuclides moving exclusively in the matrix. In
addition the transport velocities in the fractures may be sufficiently rapid that radionuclide
concentrations in the fractures are in disequilibrium with the matrix. For example, a high
concentration of radionuclides entering fractures at the repository may penetrate the entire UZ
before establishing a uniform, equilibrated concentration in the rock matrix. Therefore, an
explicit and dynamic model of transport through fractures and matrix, as well as exchange
between the fractures and matrix, is needed to represent the system. The dual-permeability
model provides the necessary level of detail to capture the important differences between
transport through fractures and matrix. There are other possible approaches to modeling flow in
fractures and matrix rock. However, these other models either do not recognize the important
dynamic coupling between fractures and matrix, such as the ECM, or are impractical to
implement at the field scale, such as the discrete fracture model (particularly for systems with
advective transport in the fractures and matrix, such as a dual permeability system). For these
reasons, the flow and transport models for the UZ are based on dual permeability.
In general flow in the UZ is time dependent or transient. One mechanism responsible for this
time dependence is the time variations in the infiltration flux at the surface. The time variation of
the infiltration flux may be approximated as occurring over short intervals characterized by
VI-35
-------
changes in weather, resulting in episodic transient flows, or over much longer time periods
corresponding to climate changes. Existing information concerning episodic transient flow
seems to indicate that such flow may not be able to frequently penetrate through the UZ to the
level of the repository because of the dampening influence of the PTn hydrogeologic unit.
Episodic transient flow propagating through fractures tends to be absorbed by the PTn unit,
resulting in much slower drainage of the episodic flows in lower hydrogeologic units at and
below the repository level (ROB97, Section 6.13). For these reasons, the DOE has not
incorporated episodic transient flows into the TSPA-VA calculations.
Changes in unsaturated flow because of longer-term changes in climate have a more pronounced
influence on UZ flow than episodic transient flow. Sustained changes in infiltration associated
with climate change ultimately impact the entire flow field in the UZ. The actual transient period
during which the UZ flow responds to a climate change, however, has been found to be less
significant (ibid.). The reason is that the change in flow in the fractures, which dominates the
flux in most hydrogeologic units, responds relatively quickly to a change in infiltration.
Therefore, the quasi-steady flow model was used to estimate the effects of climate change on
radionuclide transport. In this model, infiltration rate is assumed to change abruptly when
climate changes from one steady flow field to another. Transport calculations simply re-start
when climate changes. A distributed source of radionuclides throughout the UZ is derived from
transport calculations using the flow field for the previous climate, and a new steady state flow
field is selected based on the new climate.
In addition to the change in the UZ flow field, the location of the water table is also assumed to
change abruptly at the time of climate change. The three climate states—present day, long-term
average, and super pluvial—have successively higher water table elevations to response to the
increasing infiltration. If the water table rises with the climate change, the radionuclides in the
UZ between the previous and new water table elevations are immediately available for transport.
In the TSPA-VA model, water table elevations change by 80 m from present-day to long-term-
average climates and by 120 m from present-day to super-pluvial climates.
The dominant flow direction is expected to be mainly downward over large scales. However,
local flow field variations are expected to be three dimensional. The importance of these
variations lies primarily in the kinds of rock units and fracture characteristics that dominate along
the 3-D flow paths. A secondary consideration is that 3-D flow paths from the repository to the
VI-36
-------
water table will necessarily be longer than strictly vertical flow paths. 3-D flow patterns are
expected along rock unit contacts with contrasting properties, particularly in zones where these
contrasts are believed to be features that create perched water. To capture these effects, the DOE
has used three dimensions to model the flow and transport. Spatial variability is captured in the
3-D relationships of the hydrogeologic units and structural features, for example, faults, and in
the variations in hydrogeologic and transport properties assigned to the hydrogeologic units.
Matrix Diffusion
Diffusion is the movement of dissolved or colloidal material because of random molecular
motion. It is not an effective mechanism for transport between the repository and the water table
because of the large distance involved (about 300 m). However, diffusion can play an important
role in radionuclide exchange between fractures and rock matrix. In this case, molecular
diffusion affects the persistence of a dissolved ion in the fracture flow stream. The relative
influence of this mechanism on overall transport through the UZ depends on the rate of
movement through fractures as well as the degree of fracture/matrix contact
Bulk diffusive flux occurs when concentration gradients are present because diffusion is driven
by random molecular motion. In addition, the matrix diffusion coefficient is a function of the
free water diffusion coefficient, temperature, radionuclide mass, atomic or molecular dimensions
and charge, as well as the matrix pore structure and water saturation. The temperature variations
are expected to be small over the time period for radionuclide releases to the UZ. The effects of
pore structure and water saturation have been shown to depend primarily on the volumetric water
content of the rock (LAN97). For rock in the UZ, the water content is relatively uniform
spatially. Therefore, as a simplification, variations in the matrix diffusion coefficient are
assumed to be primarily dependent upon the radionuclide type, that is, mass, size, and charge. In
this case, measurements indicate that the primary difference is between cationic and anionic
radionuclides (TRI97)- Anionic radionuclides have lower matrix diffusion coefficients than
cationic radionuclides; that is, they are transported more slowly by diffusion. Lower coefficients
for anionic radionuclides are believed to be a result of size and charge exclusion of the anionic
radionuclides from a portion of the pore structure. The surfaces of the pore minerals are
generally negatively charged under the chemical conditions of the undisturbed environment.
VI-37
-------
In general, the direction of radionuclide transport by matrix diffusion between the fracture
continuum and the matrix continuum depends on the direction of the concentration gradient.
However, the most important influence of matrix diffusion is expected to be on radionuclide
transport through fractures. Radionuclides traveling through fractures are always near the matrix
relative to the fracture spacing. This proximity is a result of the fracture and matrix geometry
where the ratio of fracture aperture to fracture spacing is small. Also, diffusive penetration of the
matrix is proportional to the square root of the fracture transport time. Because of relatively fast
transport through fractures, most radionuclides diffusing from fractures into the matrix do not
penetrate far, relative to the fracture spacing. For this reason, the DOE expects the effects of
fracture spacing on matrix diffusion to be negligible, and ignores the effects of finite fracture
spacing. Similarly, because of the relatively small fraction of matrix affected by matrix diffusion
(that is, diffusive movement from fractures into the matrix is relatively slow), the DOE makes an
approximate representation for fracture transport, including matrix diffusion, separately from the
advection of radionuclides through the matrix.
Dispersion
Dispersion is a transport mechanism caused by localized variations in flow velocity. These
variations cause dispersion of the radionuclides both along and transverse to the average flow
direction. Variations in both the magnitude and direction of the velocity contribute to the overall
dispersion. This dispersion, under certain limiting conditions, can act in a manner analogous to
diffusion, in which mass flux is proportional to the concentration gradient. Dispersion is most
important where concentration gradients are the largest—near the front of a propagating plume or
along the lateral edges of the concentration field. Dispersion smears sharp concentration
gradients and can reduce the breakthrough time, or arrival time at a specific point, for low
concentration levels of an advancing concentration front.
Dispersion is included in the transport conceptual model using a standard relationship based on
Pick's law between mass flux and concentration gradient (ROB97). The dispersion coefficient in
the Fickian relationship is expressed as the product of the mixing length scale or dispersivity, and
the average linear velocity. Dispersion is independently represented in both the fracture and
matrix continua. However, the DOE does not expect dispersion to play an important role in the
UZ transport. The repository emplacement area is very broad, relative to the distance to the
water table, and this geometrical arrangement tends to suppress dispersion effects. Longitudinal
VI-38
-------
dispersion becomes secondary to the explicitly modeled variations in transport velocities across
the repository because of variations in infiltration. Lateral dispersion is limited by the short
transport path from the repository to the water table compared with the size of the source. Also,
the explicitly modeled variations in transport velocity caused by the fracture/matrix system tend
to dominate dispersion.
Radionuclide Sorption
Sorption is the general term for describing a combination of chemical interactions between the
dissolved radionuclides and the solid phases (that is, either the immobile rock matrix or colloids).
Any given sorptive interaction is caused by a set of specific chemical interactions such as surface
adsorption precipitation, and ion exchange. However, the sorption approach does not require
identifying the specific underlying interactions. Instead, batch sorption experiments are used to
identify the overall partitioning between the aqueous and solid phase, characterized as a
"sorption" or distribution coefficient (Ka). The strength of the sorptive behavior is a function of
the chemical element, the rock type involved in the interaction, and the geochemical properties of
the water contacting the rock. Sorption reduces the rate of advance of a concentration front in
advective and diffusive transport, and amplifies the diffusive flux of radionuclides from the
fractures to the rock matrix through its influence on the concentration gradient.
Numerous rock-water chemical interactions may influence radionuclide transport (TRI97):
Ion adsorption —metal cations sticking to mineral surfaces due to London-van der
Waals forces and hydrogen bonding
Ion exchange —substitution of an aqueous cation for the anion in a mineral structure
Surface compilation —coordination of an aqueous cation with a deprotonized metal
hydroxide at the mineral surface
. Precipitation-g.eneiation of a bulk solid phase.
The nature and strength of these rock-water interactions are highly dependent on the chemical
composition of both the aqueous and solid phases. The conceptual model used to capture all
these interactions and sensitivities is the minimum Kd model. This model bounds the distribution
of radionuclides between the mobile, or dissolved in the aqueous phase, and immobile, or
VI-39
-------
attached to the solid phase, using a linear, infinite-capacity partitioning model. In this model, the
sorbed, or immobile, concentration is equal to the aqueous concentration times the partitioning
coefficient. Because of the numerous mechanisms and dependencies known to influence
sorption, using a linear partitioning model with a single coefficient provides only a minimum
bound for Kj, and hence the immobilized fraction.
Colloid Facilitated Transport
Colloids are potentially mobile in water flowing though the UZ. Since colloids are small solids
they can interact with radionuclides through sorption mechanisms. Unlike sorption of
radionuclides to the rock matrix, however, radionuclides sorbed on colloids are potentially
mobile. Therefore, colloidally-sorbed radionuclides can be transported through the UZ at a faster
rate than the aqueous species without colloids. Another form of colloidal radionuclide
movement occurs when the radionuclide is an integral component of the colloid structure. In this
case, the radionuclide is irreversibly bound to the colloid, as compared to the typically reversible
sorption mechanism. The different types of colloids considered in the performance assessment
model are clay colloids, iron oxy-hydroxy colloids, spent fuel waste form colloids, and glass
waste form colloids.
Thermal-Hydrologic, Thermochemical, and Thermomechanical Processes
The potential repository will perturb the natural system due to the introduction of waste heat and
foreign materials. Thermo-hydrologic modeling is done for TSPA-VA to better define the
temperature, humidity, and water contact conditions that will affect waste package corrosion,
waste form dissolution, and radionuclide releases from the engineered barrier system. The
thermal-hydrologic conditions also affect the flow field, and hence the radionuclide transport,
through the UZ. However, the thermal-hydrologic process is most significant over the first 2,000
to 3,000 years following waste replacement. Sensitivity studies indicate that most of the
radionuclide transport through the UZ will take place after thermal-hydrologic activity has
subsided. Thermal-hydrologic processes are not expected to substantially alter UZ radionuclide
transport as compared to transport through the undisturbed system. Therefore, the DOE has not
incorporated the effects of thermal-hydrologic processes in the UZ radionuclide transport model.
VI-40
-------
TSPA-VA Approach
Unsaturated zone radionuclide transport calculations are computed using the particle tracking
method available in the unsaturated flow and transport code FEHM. This computational method
for UZ radionuclide transport is dynamically linked with the TSPA's RIP code for these
calculations. A 3-D, dual-permeability model is used for computing radionuclide transport in a
fracture/matrix system. Three-dimensional, steady, UZ flow fields, computed with the
unsaturated flow code, TOUGH2, are used for the particle tracking transport calculations.
Changes in the flow field due to climate change are accounted for through a quasi-steady flow
and transport approximation. The transport model includes the effects of fracture/matrix
interaction driven by advective and diffusive exchange between the fracture and matrix continua.
The transport model includes the effects of radionuclide sorption in the rock matrix using a
bounding, minimum Kj modeling approach. Colloids are included as a mechanism for the
transport of plutonium. The model for aqueous plutonium transport allows for reversible
sorption of plutonium on colloids. Plutonium releases from the EBS also include a fraction
plutonium irreversibly sorbed to colloids.
Radionuclide mass flux at the EBS boundary is provided by RIP for the UZ radionuclide
transport calculations. The radionuclide mass flux from any source within a potential repository
region is distributed uniformly over the computational grids for fracture nodes in the transport
model that lie within the region. Radionuclide mass flux at the water table from the fracture and
matrix continua are mixed in a RIP cell and provided to the SZ radionuclide transport model.
The model for unsaturated radionuclide transport incorporates several assumptions. The primary
conservative modeling assumptions are no fracture sorption and no colloid filtration. These
assumptions clearly result in model predictions having more rapid migration of radionuclides
through the unsaturated zone. Assumptions about the influence of thermal-hydrologic-chemical
processes, fracture spacing in the matrix diffusion model, and fracture/matrix contact in the
matrix diffusion model, are potentially nonconservative.
Long-term Transient Flow and Climate Change
An approximate method was developed for computing UZ radionuclide transport under the long-
term transient flow conditions associated with climate change. The approximation was
developed due to the computational burden associated with the complete transient flow and
transport calculation. The method is based on a quasi-steady flow approximation, in which the
VI-41
-------
flow fields instantaneously change from one steady flow field to another in response to changing
infiltration rates caused by climate change. The transport calculations restart upon change in
climate using the new flow and water saturation distributions. Radionuclides present in the UZ
at the time of climate change are also restarted as a distributed source of radionuclide mass in the
UZ flow and transport grid. The method is shown to provide an excellent approximation to the
complete unsaturated transient flow and transport problem resulting from climate change.
Colloid-Facilitated Radionuclide Transport
An approximate computational method was developed to treat the effects of radionuclide
interactions with colloids. This method is applied in the TSPA-VA calculations to the problem
of plutonium transport. Previous TSPA analyses did not include the effects of colloid-facilitated
radionuclide transport. The treatment addresses two types of colloidal interactions with
plutonium: reversible sorption between aqueous and colloidal plutonium and plutonium
irreversibly attached to colloids. For reversibly sorbed plutonium, the approximate colloid
facilitated transport method assumes linear, equilibrium partitioning of aqueous and colloidally-
sorbed fractions. Colloid movement is represented by advective transport in fractures or matrix
with no diffusion of colloids between fractures and matrix, and no colloid filtration.
Unsaturated Zone Radionuclide Transport Sensitivity Investigations
The DOE performed additional sensitivity to provide a better understanding of transport
parameter sensitivity, identify the range of conditions under which a Kj sorption model may be
used to bound geochemical interactions, investigate the effects of episodic transient flow on
radionuclide transport, and investigate the effects of fine-scale mineralogic heterogeneity.
DOE Conclusions
The results of investigations into radionuclide transport through the UZ lead the DOE to
conclude the following:
• For present-day climate, the average travel time for nonsorbing radionuclides released
from the northern portion of the repository is on the order of a thousand years, but for
the wetter long-term average climate, the average travel times are tens of years.
VI-42
-------
• For present-day climate, the average travel time for nonsorbing radionuclides released
from the southern portion of the repository is several thousand years, but for the
wetter long-term average climate, the average travel time is hundreds of years.
• For most of the mass of strongly sorbing radionuclides, such as plutonium, arrival at
the water table is delayed tens to hundreds of thousands of years. Radionuclides that
reversibly sorb onto colloids are also delayed; however, irreversibly bound
radionuclides may behave as nonsorbing radionuclides.
• Matrix diffusion has little effect on transport of nonsorbing radionuclides. However,
combining matrix diffusion with sorption significantly retards radionuclides
compared with sorption and no matrix diffusion.
• The TSw unit is the primary barrier to radionuclide transport because of slower
fracture transport in this unit compared with the Calico Hills zeolitic unit, and its
greater thickness and continuity compared with the Calico Hills vitric unit.
• Alternative flow models like the dual-permeability model with Weeps modeling,
which have smaller coupling strengths for effective fracture/matrix contact area,
increase fracture-dominated transport, particularly for sorbing radionuclides.
However, the magnitude of this effect is not large for weakly sorbing radionuclides.
The following items are viewed by the DOE as the areas where additional work would provide
the greatest improvements in the comprehensiveness and credibility of the radionuclide transport
model in the UZ:
Effects of Thermal, Hydrologic-Chemical Alteration—The current model does
not account for alteration of the UZ because of thermal alteration of minerals,
chemical interactions of repository materials, mineral dissolution, and
precipitation. These effects are potentially important to transport behavior in the
UZ and need to be addressed. Thermal-chemical alteration could cause reduced
matrix sorption and fracture-matrix interaction. These effects could cause
increased release rates from the UZ for base case transport results; however, it is
expected that existing sensitivity studies bound most of these potentially
nonconservative interactions.
Colloid Filtration—The ability of colloids to facilitate radionuclide transport is a
function of their ability to migrate over large distances without being "filtered
out" by the host rock. This filtration effect was not included in the TSPA because
of inadequate information to bound the mechanism.
VI-43
-------
The filtration effect is particularly important for the fraction of radionuclides that
are irreversibly bound to colloids. However, the assumption that there is no
colloid filtration is conservative for transport in the UZ .
Fracture Sorption—The effects of higher infiltration determined in the TSPA-VA
guarantee that transport will be fracture-dominated in many of the unsaturated-
zone units. Minerals that line fractures are known to sorb radionuclides, but more
information is needed to define the distribution and character of the fracture
materials so that sorption on fracture surfaces may be included. However, the
assumption that radionuclides do not sorb onto fracture surfaces is conservative
for transport in the UZ. Although the present sensitivity studies have found that
transport is not highly sensitive to fracture sorption, this result requires further
investigation to better define the appropriate range of parameters for fracture
sorption.
VI. 1.1.21 Expert Elicitation and Peer Review Panel
The Yucca Mountain project conducted a UZ Flow Model Expert Elicitation, in which project
data and models were presented to a group of experts (mostly from outside the project) so that
they could provide an evaluation of the work being done (CRW97). Particular emphasis was
placed on estimates of surface infiltration, deep percolation, and the uncertainties associated with
each. Each expert estimated probability density functions (PDFs) of infiltration. The mean
values of their estimated net infiltration ranged from 3.9 mm/yr to 12.7 mm/yr, and the mean
values of their estimated percolation flux at the repository horizon ranged from 3.9 mm/yr to
21.1 mm/yr. These values are commensurate with the simulated average present-day infiltration
over the modeled repository (~8 mm/yr) and the simulated average percolation fluxes within the
six prescribed repository subregions for the present-day base case (3.9 mm/yr to 11 mm/yr) of
TSPA-VA.
Additional topics that were covered during the expert elicitation included spatial variability of
net infiltration and percolation flux, temporal variability of infiltration and percolation flux,
lateral diversion above the repository, partitioning of fracture and matrix flow, seepage into
drifts, modeling issues, and additional data and work requirements. The experts generally agreed
that the infiltration map that was used in TSPA-VA captures the general spatial variability of
VI-44
-------
infiltration, but several suggested that more infiltration could occur beneath washes with thin
alluvial cover. The experts also agreed that temporal behavior of net infiltration was
characterized by episodic events associated with major storm events, but that most of the
transient flux was attenuated within the system. With regard to lateral diversion above the
repository, most of the expert agreed that contrasts in the hydrologic properties between units
would likely cause lateral flow, particularly in the Tiva Canyon welded unit - Paintbrush Turf
nonwelded unit contact, but that the amount of diversion was limited to scales of several meters
to tens of meters. This is consistent with the results of the TSPA-VA UZ-flow simulations,
which show nearly vertical flow above the repository. The experts did not address the issue of
lateral diversion caused by perched water below the repository. The experts also estimated that,
based on the low matrix permeabilities in the Topopah Spring welded unit, partitioning of flow
in the TSw was predominantly in the fractures. Also, the fast-flow component likely represents
only a small part of the total flux. This is consistent with simulated results in TSPA-VA. With
regard to seepage into drifts, the experts agreed that a capillary barrier will exist around the
drifts, diverting a majority of the water around the area of the drifts. This behavior is also
consistent with the TSPA-VA simulations for seepage. Finally, the experts recommended a
number of additional modeling and data-collection activities that address estimates of infiltration,
percolation, and seepage.
In addition to the expert elicitation, the TSPA-VA Peer Review Panel provided input to the UZ
flow activities. They recognized the use of environmental tracers (36C1) as evidence of fast flow
paths between the surface and the repository and recommended its use in supporting the
conceptual models of UZ flow. The Peer Review Panel also recommended studies to better
understand fracture-matrix interaction. Finally, the Peer Review Panel emphasized the need to
better understand seepage into drifts.
VI.2 SATURATED ZONE
VI.2.1 Saturated Zone Modeling
*
VI.2.1.1 Saturated Zone Flow Model Construction
The transport of radionuclides in the saturated zone away from a repository depends on a wide
variety of factors including, but not limited to, ground-water and host rock geochemistry;
VI-45
-------
advective ground-water velocities; radionuclide concentrations and retardation properties; flux
rates of radionuclides from the unsaturated zone; the presence of sorbing materials such as
zeolites and clays; rock fracture density; fracture-matrix interaction; future climate changes; and
anthropogenic influences. Knowledge of the transport properties in the site-scale and regional
flow systems would allow researchers to more completely address four of the most important
questions surrounding repository performance and regional ground-water flow issues in the area
around Yucca Mountain:
1. What path would radionuclides from the repository follow?
2. How fast and how far would radionuclides travel in the saturated zone?
3. Where would radionuclides become accessible to the biosphere?
4. What will the concentrations of radionuclides be when they become accessible to the
biosphere?
The answer to all of these questions is uncertain. The ability to know or predict the answers to
these questions depends on performing sufficient scientific investigations over the study area in
order to reduce the associated uncertainties to acceptable levels. Some level of uncertainty will
always remain, as it is not possible to completely characterize any underground system.
In order to address these issues, the DOE has performed modeling of the saturated zone at Yucca
Mountain at a number of different scales, including: regional, site, and sub-site. The ground-
water flow modeling activities that are most closely linked to the TSPA-VA are a regional scale
model that was developed by D'Agnese et al. (DAG97), a site-scale model formulated by
Zyvoloski et al., (ZYV97), and a sub-site scale model that was completed by Cohen, et. al.
(COH97).
The regional-scale 3-D flow model was developed by D'Agnese et al. (DAG97) to characterize
the conditions of the present day ground-water flow in the Death Valley region. The modeled
area is 244 km long and 229 km wide. The numerical code used for the regional flow model was
MODFLOWP (HIL92). MODFLOWP is an adaptation of the U.S. Geological Survey 3-D,
finite-difference modular ground-water flow model, MODFLOW (MCD88).
A smaller site-scale TSPA three-dimensional flow model was formulated by Zyvoloski et al.,
(ZYV97) to determine the flowpath from the repository footprint, or outline, at the water table to
a distance 20 km downgradient, the approximate distance to the nearest domestic well where
VI-46
-------
ground water is extracted. The computer code FEHM (ZYV83) was used to model an area of 20
by 36 km. Originally, it was intended that the regional scale model be used to provide the
boundary conditions for the site scale model. Apparently, the authors of the site-scale modeling
(COH97), however, did not believe that the calibration of the regional model was sufficient to
provide reliable boundary conditions. Therefore, the regional modeling, from the perspective of
the TSPA-VA, was used primarily to determine a ground-water flux multiplier for the long-term
average and superpluvial climate conditions assumed for the base-case TSPA-VA analysis as
will be discussed later in this section.
The sub-site scale model constructed by Cohen et. al. (COH97) was also performed in order to
assess the effectiveness of the saturated zone as a barrier to radionuclide transport. However, the
sub-site-scale model more accurately captures the faulted and variable-thickness stratigraphy
than either the regional or site-scale modeling. Thus, allowing for more detailed modeling of
processes that cannot be obtained from larger scale models. As will be discussed later in this
section, the sub-site scale model was not as explicitly integrated into the TSPA-VA as the larger-
scale models.
Conceptual Model of Saturated Zone Flow
Regional Scale Conceptual Model
The current state of knowledge suggests that ground water beneath the proposed repository
moves laterally downgradient until the volcanic aquifer pinches out, at which point it discharges
laterally into the alluvial aquifer. Radionuclides dissolved in ground water would potentially
follow a similar path. Much of the ground water that enters the alluvial aquifer currently moves
southward to the'primary discharge location at Alkali Flat. Other actual or potential points of
discharge for the system include water wells in the Amargosa Desert and springs in the Furnace
Creek Ranch area of Death Valley.
Ground-water travel times to any of these locations are not well known. Estimates of ground-
water travel times can be developed by simple calculations or by more sophisticated numerical
modeling. In either case, travel time calculations are based on hydraulic gradient, hydraulic
conductivity, and effective porosity of the formation through which the water is flowing. Of
these three parameters, hydraulic gradients are probably the best known and most easily
measured.
VI-47
-------
D'Agnese et al. (DAG97) formulated a conceptual model of the Death Valley regional ground-
water flow system by integrating interpretations of flow system components. Their discussion of
the flow system dynamics includes a description of regional, subregional and sub-basin
boundaries, as well as the source, occurrence, and movement of ground water in the system. The
most pertinent aspects of that discussion are summarized below.
Flow System Boundaries. The Death Valley regional flow system consists of ground water
moving through a three-dimensional body of consolidated and unconsolidated materials. The
flow system boundaries may be either physical boundaries, caused by changes in bedrock
conditions, or hydraulic boundaries, caused by potentiometric surface configurations. The upper
boundary of the flow system is the water table. The lower boundary of the flow system is
located at a depth where ground-water flow is dominantly horizontal and moves with such small
velocities that the volumes of water involved do not significantly impact regional flow estimates.
The lateral limits of the regional flow system may be either no-flow or flow boundaries. No-
flow conditions exist where ground-water movement across the boundary is prevented by
physical barriers or divergence of ground-water flow paths. Flow boundaries exist where
ground-water potentiometric gradients permit flow across a boundary through fractures or higher
permeability zones.
Boundaries. The lateral boundaries selected for the flow system were modified from
those described by Waddell et. al. (WAD84). Most system boundaries are no-flow boundaries
that result from the presence of low-permeability bedrock. Flow boundaries occur where
bedrock has a high enough permeability to allow significant ground- water fluxes to enter the
system and where a hydraulic gradient exists across the boundary. Faulting and fracturing most
frequently cause the enhanced permeability, and ground-water flow may occur at various depths
through open regional fracture zones. Based on potentiometric and hydrogeologic framework
data, areas where inflow may occur from are Pahranagat Valley, Sand Spring Valley, Railroad
Valley, Stone Cabin Valley, Ralston Valley, Fish Lake and Eureka Valleys, Saline Valley,
Panamint Valley, Pilot Knob Valley , and Soda Lake Valley. Good estimates of flow across
these lateral flux boundaries do not exist except for Pahranagat Valley, which has been estimated
by Winograd and Friedman (WIN72) to be approximately 20,000 m3/d. The remaining areas
have very little data required to estimate flux volumes; however, the authors (DAG97) believe
that flux across these boundaries should not be dismissed without further investigation.
VMS
-------
The flow system boundary in northern Las Vegas Valley near Corn Creek Springs results from
the presence of a ground-water divide. Ground water recharging from the Sheep and Spring
Mountains forms a ground-water divide that extends across the valley and separates flow that
moves southeast toward Las Vegas Valley from flow that moves to the northwest toward Ash
Meadows in the Amargosa Valley.
For numerical simulation, the flow system was subdivided into three subregions that represent
the areas where regional ground-water flow moves from recharge areas in Nevada toward Death
Valley saltpan, the ultimate terminus of the system. Local recharge along the southern boundary
of the system and subsurface inflows along parts of the southeastern and southern boundary of
the system were not included in the simulation. These model boundaries are based on previously
defined flow system boundaries, the potentiometric surface developed for their modeling study,
and the hydrogeologic framework. The authors (DAG97) comment that few data exist that
would allow a precise definition of the western and southern extent of the flow system. The
western boundary of the flow system is placed to coincide with the eastern edge of the Death
Valley saltpan, which is interpreted as the terminal sink of the flow system. Although some
ground water that originates on the west side of Death Valley may discharge into the saltpan, this
discharge is mostly at Mesquite Flat and is a small volume compared to the contribution from the
east (PRU93).
£ubregional Boundaries. To define the subregional boundaries, the Death Valley regional
ground-water flow system was divided into three major subregional flow systems. The names of
the subregions reflect the part of Death Valley into which each discharges. For example, the
Northern Death Valley subregion discharges into the northern part of Death Valley at Grapevine
and Staininger Springs and Mesquite Flats. The Central Death Valley subregion predominantly
discharges into the Saratoga Springs area at the southern terminus of Death Valley.
Ground water is thought to flow across the subregional boundaries in three places: (1) across
the southeast border of the Central Death Valley subregion from the Amargosa Desert into the
Lower Amargosa Valley in the Southern Death Valley subregion; (2) from the Northern Death
Valley subregion across a boundary at Salt Creek Springs (just south of Mesquite Flat) into the
Central Death Valley subregion; and (3) at the southern end of Death Valley, ground water that
has not discharged in the Saratoga Springs area may continue to flow northward from the
Southern Death Valley subregion across the subregion boundary to discharge at Badwater Basin
in the Central Death Valley subregion.
VI-49
-------
Moddfap Difficulties aH Simp^np Assumptions. The authors (DAG97) pointed
out that previous studies by Prudic et al. (PRU93) and Waddell (WAD82) showed that it is
difficult to utilize computer models to effectively describe ground-water flow in an area as
geographically large and geologically complicated as the Death Valley region. Prudic et. al.
(PRU93) reiterated that many arguments can be invoked concerning the validity of the
assumptions and hydrologic values used in simulating ground-water flow when such complex
geology and hydrology are involved.
In any modeling investigation, it is inevitable that simplifications and assumptions must be used
to adapt the complex conceptual model for numerical simulation. The assumptions and
simplifications used to develop the Death Valley Regional Flow System (DVRFS) model include
the following:
1. Ground water in the region flows through fractured volcanic and carbonate rocks, as
well as in porous valley-fill alluvium. However, fracture flow simulation is
impractical at a regional scale, and, therefore, a porous medium simulation is used.
Zones of high hydraulic conductivity are used to account for highly faulted and
fractured regions.
2. Hydraulic conductivities within each model cell are assumed to be homogeneous and
horizontally isotropic. Thus, features smaller than the grid cells are not represented.
The authors (DAG97) believe that this approach is likely to produce reasonable
approximations to large-scale flow patterns. Small-scale flow paths, however, are not
represented.
The system can be represented adequately as steady state. Four conditions exist that
may violate this assumption. First, the regional flow system still may be undergoing
a drying-out sequence following a wetter climate cycle related to the late Pleistocene
(PRU93). As a result, current ground-water levels and discharge rates may not be in
equilibrium with present-day recharge and interbasinal flux rates. Second, and
perhaps more important, ground-water withdrawals by wells for domestic, municipal,
muiing and irrigation uses are imposing new stresses on the present-day system. This
pumpage is derived initially from ground water, from storage, and subsequently from
capture of natural discharge. Incorporating pumping in a steady-state model omits the
3.
VI-50
-------
possibility of deriving water from storage, so that water flowing to wells must be
offset by capture of natural discharge, that is, reductions in discharge or induced
inflow. Although a transient simulation beginning at predevelopment conditions
would avoid this assumption, additional assumptions would be needed to define
historic pumping levels. In addition, some current water-level data and some spring-
flow rates already reflect changes to the system resulting from development,
suggesting that the DVRFS may have already adapted to these changes. For example,
the springs at Pahrump Valley, including Manse and Bennetts Spring, have ceased to
flow in historic time. Third, the flow system experiences seasonal fluctuations that
are not simulated. A resulting annual average condition is simulated. Fourth,
hydraulic-head, spring-flow and other data used in model calibration were collected
over an interval of many years, and these data are affected by seasonal and yearly
changes to the ground-water flow system.
Site-Scale Conceptual Model
The major components of the site-scale conceptual model (ZYV97) are presented below.
Geologic Setting. The conceptualization of the geologic setting by Zyvoloski et al., (ZYV97)
which, briefly summarized, is that Yucca Mountain is located in the Great Basin section of the
Basin and Range physiographic province, and consists of a group of north-south-trending block-
faulted ridges that are composed of volcanic rocks of Tertiary age that may be several kilometers
thick. The basin to the west of Yucca Mountain is Crater Flat, which is comprised of a thick
sequence (about 2,000 m) of Tertiary volcanic rocks, Tertiary and Quaternary alluvium, and
small basaltic lava flows of Quaternary age. Crater Flat is separated from Yucca Mountain by
the Solitario Canyon Fault. West of Crater Flat is Bare Mountain, which is comprised of
Paleozoic and Precambrian rocks. Fortymile Wash, a structural trough, delimits the eastern
extent of Yucca Mountain. East of Yucca Mountain are the Calico Hills, a mottled assemblage
of Tertiary volcanic rocks and Paleozoic rocks. Yucca Mountain terminates to the south in the
Amargosa Desert, which consists of interbedded Quaternary and Tertiary alluvial, paludal, and
tuffaceous sediments.
These rocks and deposits in the vicinity of Yucca Mountain were classified by the authors
(ZYV97) into hydrogeologic units based on hydraulic properties. Where possible, hydrogeologic
VI-51
-------
units identified by previous investigators (LUC96, WIN75) were used. Many of the units are not
present in the model area and/or are not expressed at the land surface. In all, sixteen
hydrogeologic units are present in the model area.
In general, the hydrogeologic units at Yucca Mountain form a series of alternating volcanic
aquifers and confining units overlaying the regional carbonate aquifer. The volcanic aquifers and
confining units interbed with undifferentiated valley-fill and the valley-fill aquifer to the south,
while structural features delimit the eastern and western edges of Yucca Mountain.
Hydrogeologic Setting. The hydrogeologic framework briefly described by the authors (ZYV97)
is that Yucca Mountain is centrally located within the Death Valley ground-water basin and also
is centrally located within the Alkali Flat (Franklin Lake playa)-Fumace Creek subbasin. The
subbasin is assumed to receive water from areal recharge within its boundaries and the authors
(ZYV97) believe that it probably also receives water as underflow from adjoining subbasins.
Depths to water range from about 3 m beneath Alkali Flat (Franklin Lake playa) to about 750 m
beneath Yucca Mountain. Ground water beneath Yucca Mountain flows generally toward the
south through fractured volcanic rocks which interfmgers with Quaternary and Tertiary valley-
fill in the Amargosa Desert.
The climate is arid to semiarid, with Yucca Mountain receiving annual precipitation between 150
mm to 200 mm (HEV92). As a result, stream flow is infrequent and occurs following intense
precipitation events which can be very localized. There are no perennial streams.
Vertical gradients. As is discussed later in this section, the ability to reproduce and understand
the vertical flow gradients between hydrostratigraphic units will be an important aspect of model
calibration. Therefore, the authors (ZYV97) believe it is important to include them as an explicit
component of the conceptual model. The authors (ZYV97) remark that Luckey et. al. (LUC96)
examined the vertical relationship of hydraulic head at Yucca Mountain, and found "no
unambiguous areal patterns in the distribution of vertical hydraulic gradient around Yucca
Mountain." However, they also make the following generalizations as to the distribution of
potentiometric levels in the lower sections of the volcanic rocks. Potentiometric levels in the
middle volcanic confining unit are relatively high (altitude greater than 750 m) in the western
and northern parts of Yucca Mountain and are relatively low (altitude about 730 m) in the eastern
part of Yucca Mountain. Based on potentiometric levels that were measured hi borehole UE-25
VI-52
-------
p#l, the potentiometric levels in the middle volcanic confining unit in boreholes USW H-l, USW
H-3, USW H-5, and USW H-6 may reflect the potentiometric level in the carbonate aquifer. It is
also noted that boreholes UE-25 b#l and USW H-4 do not seem to fit the pattern established by
the other boreholes. Potentiometric levels generally are higher in the lower intervals of the
volcanic rocks than in the upper intervals, indicating a potential for upward ground-water
movement. However, for unknown reasons, at four boreholes (USW G-4, USW H-l, USW H-6,
and UE-25 b#l) potentiometric levels in the volcanic rocks are higher in the uppermost intervals
than in the next lower intervals.
The potentiometric levels in the Paleozoic carbonate aquifer at borehole UE-25 p#l are about 21
m higher than in the overlying volcanics. Therefore, a potential for upward ground-water
movement from the Paleozoic rocks to the volcanic rocks is indicated. Because of the large
difference in potentiometric levels in these two aquifers, Luckey et. al. (LUC96) conclude that
they seem to be hydraulically separate. This conclusion appears to be supported by
hydrochemical data. However, some of the analyses of hydraulic-test data at the C-hole complex
indicate a possible hydraulic connection between the volcanics and the carbonate aquifer at the
C-hole complex (GEL96). Hence, the vertical hydraulic gradients represent a complex three-
dimensional flow system that is not completely understood. Little information is available for
vertical gradients away from Yucca Mountain. A more detailed discussion of data limitations
and modeling needs will be presented later in this section.
Steady-state conditions. A comprehensive analysis of water levels from all observation wells at
Yucca Mountain (GRA97) shows the fluctuations of water levels for the period 1985 to 1995.
The authors (ZYV97) concluded that, in general, most wells at Yucca Mountain show less than 1
meter difference between the maximum and minimum values of water-level altitude during this
oeriod The authors (ZYV97) further conclude that the preponderance of wells with small water
level changes and the small fractional changes in saturated thickness at wells with greater
changes indicates that assuming that the flow system is at steady state at Yucca Mountain is a
reasonable approximation. It should be noted that the modelers simulating the regional scale
model (DAG97) expressed some concern with assuming that the system is or would remain at
steady state, and outlined four conditions that exist which may violate this assumption.
Zyvoloski et al., (ZYV97) do discuss, however, that water-level changes have declined by as
much as about 10 m (KIL91) in the Amargosa Farms area (southwest comer of the site model)
resulting from ground-water withdrawals for irrigation. The modelers made no attempt to
reconstruct the potentiometric surface for conditions prior to these ground-water withdrawals.
VI-53
-------
The potentiometric data dictate a complex three-dimensional flow system, but the authors
(ZYV97) have made the following generalizations. There appears to be a general upward
gradient from the regional carbonate aquifer into the volcanic rocks. In general, this upward
gradient persists in the volcanic rocks. Furthermore, the potentiometric data indicate that most of
the flow system is essentially at steady state.
Potentiometric surface. Because the potentiometric data dictate a complex three-dimensional
flow system, a number of different conceptual models of the flow system are possible. In
particular, the different conceptual models may result in different potentiometric surfaces.
Although the boreholes are open at different depths below the hydraulic head and are open to
different geologic zones, water levels in most of the wells appear to represent a laterally
continuous aquifer system. The well-connected system may result from the presence of many
faults and fractures (TUC95), and, at the scale of the site model, the authors (ZYV97) believe
that the ground-water flow system may behave as a porous medium. Flow in the volcanic rocks
occurs primarily in fractures and secondarily in the matrix of the rock. Therefore, the uppermost
aquifer may be unconfmed or confined depending upon the areal location of the point being
measured (TUC95).
Most of the wells only partially penetrate the hydrogeologic units. This can be important during
model calibration because it can lead to significant errors if vertical gradients are large. No
attempt was made by the modelers to segregate and analyze water-level measurements associated
with specific hydrogeologic units or fracture zones. The authors (ZYV97) believe, however, that
the potentiometric surface is probably a reasonable representation of the water table for the
following reasons: (1) At Yucca Mountain, water levels at most wells were obtained from the
uppermost part of the saturated zone (GRA97); (2) south of Yucca Mountain, wells penetrate a
significant thickness of the saturated zone, but hi this area most ground-water flow is believed to
be horizontal and all available data indicate that the vertical-head gradients are negligible; (3) for
the case of wells having multiple piezometers, only water levels from the uppermost saturated
interval were used in the construction of the potentiometric-surface map.
Large hydraulic gradient. Possible differences in conceptual models of the flow system pertain
to the representation of an apparent large hydraulic gradient (LHG) on the north end of Yucca
Mountain, an area where the altitude of the potentiometric surface appears to change by about
300 meters over a lateral distance of 2 kilometers (CZA94, CZA95). Prior to the construction of
VI-54
-------
borehole USW G-2 in 1981, no water-level data existed at Yucca Mountain on which to base the
LHG. As more boreholes were constructed at Yucca Mountain, particularly holes UE-25 WT#6
and UE-25 WT#16, a somewhat better definition of the LHG developed.
On a regional basis, other large hydraulic gradients are associated with a contact in the Paleozoic
rocks between clastic rocks and regional carbonate aquifer; however, the cause and nature of the
LHG near Yucca Mountain is not clear. Proposed explanations include: (1) faults that contain
nontransmissive fault gouge (CZA84); (2) faults that juxtapose transmissive tuff against non-
transmissive tuff (CZA84); (3) the presence of a different type of lithology that is less subject to
fracturing (CZA84); (4) a change in the direction of the regional stress field and a resultant
change in the intensity, interconnectedness, and orientation of open fractures on either side of the
area with the LHG (CZA84); or (5) the apparent large gradient actually represents a
disconnected, perched or semi-perched water body so that the high water-level altitudes are
caused by local hydraulic conditions and are not part of the saturated-zone flow system (CZA94,
ERV94). Fridrich et. al. (FRI94) suggest two hydrogeologic explanations for the LHG: (1) a
highly permeable buried fault that drains water from tuff units into a deeper regional carbonate
aquifer or (2) a buried fault that forms a "spillway" in the volcanic rocks. Their second
explanation, in effect, juxtaposes transmissive tuff against non-transmissive tuff, and is therefore
the same as (2) above. Explanation (5) differs from the others in that it does not require a
permeability contrast to represent the large gradient, because the LHG is absent, and actually
represents a disconnected, perched or semi-perched water body.
For the site-scale model, explanation (1): faults that contain nontransmissive fault gouge was
used to represent the LHG using reasonable permeability values by imposing a vertical barrier to
horizontal ground-water flow. Several of the other alternatives were tested by the modelers but
did not yield satisfactory results.
,iijP. properties. Knowledge of hydraulic properties is critical to understanding the
hydrogeology of Yucca Mountain and is required for numerical models. The authors (ZYV97)
obtained information on the hydraulic properties from the following sources: (1) previously
published hydraulic analyses for wells at Yucca Mountain conducted during the 1980s; (2)
published hydraulic properties for hydrogeologic units obtained beyond the immediate Yucca
Mountain area; and (3) recent (1995-97) hydraulic analysis of wells USW WT-10, UE-25
WT#12 and USW SD-7 (OBR97), UE-25 c#l, UE-25 c#2, UE-25 c#3, and USW G-2.
VI-55
-------
Aquifer tests. Several aquifer tests were conducted in Yucca Mountain boreholes during 1995
and 1996. Single borehole, composite interval tests resulted in transmissivity (i.e., hydraulic
conductivity multiplied by thickness) estimates in boreholes USW WT#10, UE-25 WT#12, and
USW G-2. The middle volcanic aquifer was the primary hydrogeologic unit tested in boreholes
USW WT#10 and UE-25 WT#12. Transmissivity in these boreholes ranged from 7 to 1,800
m2/day (OBR97). The upper volcanic confining unit was tested in USW G-2 and the mean
transmissivity was 9.4 m2/day. Transmissivity was reported for these boreholes because
composite intervals were tested and the thickness of water-producing intervals was unknown.
Hydraulic-conductivity estimates obtained from these transmissivity estimates would probably
understate the actual hydraulic conductivity because the entire interval thickness does not
contribute water to the borehole. Hydraulic properties obtained from single-borehole aquifer
tests generally represent flow conditions within tens of meters of the borehole. Given the large
degree of heterogeneity in the Yucca Mountain area, individual single-borehole aquifer-test
results are not directly appropriate for the scale represented by the site model (kilometers).
Preliminary aquifer tests were conducted at the C-well complex during 1984. Horizontal
hydraulic conductivity was about 0.15 m/d in the upper volcanic confining unit and ranged from
3 to 30 m/d within the middle volcanic aquifer (GEL96). Cross-hole aquifer tests during 1995-
96 in the C-well complex also resulted in transmissivity and hydraulic-conductivity estimates.
During these tests borehole UE-25 c#3 was pumped and boreholes UE-25 c#l, UE-25 c#2, UE-
25 ONC-1, USW H-4, and UE-25 WT#3 were used as observation wells. The lower Bullfrog
Tuff is the most transmissive interval within the middle volcanic aquifer and hydraulic
conductivity range from approximately IxlO'5 to TxlO"4 meters per second in the observation
boreholes (GEL97).
Hydraulic properties obtained from the cross-borehole aquifer tests at the C-hole complex
represent flow properties between the tested boreholes. As such, they may be more appropriate
for the scale of the site model than those obtained from single-hole tests. There is evidence that
this area has extensive fractures that enhance the transmissive properties of the aquifer system.
Northerly and northwesterly trending high angle faults such as the Paintbrush Canyon, Midway
Valley, and Bow Ridge Faults have brecciated, offset, and tilted the tuffaceous rocks in the
vicinity of the C-hole complex. Extensive tectonic and cooling fractures have been identified in
the C-hole complex boreholes (GEL96). The relatively high transmissive properties obtained
from C-hole testing are probably due to the intensity of fracturing in this area and may not be
representative of the entire Yucca Mountain area.
VI-56
-------
Recharge. The authors (ZYV97) assumed that recharge to the model area is from the following
sources: (1) downward and possible lateral recharge from episodic flooding of Fortymile Wash;
(2) throughflow from Pahute and Rainier Mesas, which is hypothesized to result in recharge
along the northern border of the study area; (3) throughflow from the northwestern part of the
Amargosa Desert; (4) minor recharge from episodic flooding of the Amargosa River channel;
and (5) net infiltration from precipitation events. Fortymile Wash is a major southward-draining
ephemeral channel located adjacent to Yucca Mountain and it is thought to contribute
intermittent recharge to the saturated zone. Water levels in UE-29 a#l and UE-29 a#2 are
affected periodically by streamflow events in Fortymile Wash and Pah Canyon Wash. In various
numerical ground-water flow models (CZA84, RIC84, CZA85, SIN87), recharge had to be
specified in Fortymiie Wash to replicate potentiometric levels. Czarnecki and Waddell (CZA84)
simulated a flux in Fortymile Wash of 22,140 mVd or 256 kg/s. Based on geomorphic/
distributed-parameter simulations, Osterkamp et. al. (OST94) estimated recharge along the entire
95-km length of Fortymile Wash to be about 4.22xl06 mVyear or 134 kg/s. Based on field
studies of stream loss, the total recharge in Fortymile Wash is estimated as 0.86 kg/s. Savard
acknowledges that this estimate would represent a minimum value based on the inability to
account for all reaches of Fortymile Wash, which may have received unobserved runoff and
recharge, coupled with the minimum period of streamflow observations.
Discharge. The authors (ZYV97) believe that no natural discharge occurs within the model
domain. The nearest natural discharge areas connected to the saturated-zone flow system
beneath Yucca Mountain are Franklin Lake playa (also known as Alkali Flat) and possibly the
major springs at Furnace Creek Ranch and the valley floor of Death Valley. Although most
models of the region (DAG97, RIC84, CZA84) require a ground-water flow path from Yucca
Mountain to Death Valley, Czarnecki and Wilson (CZA91) postulate that a ground-water flow
path from Yucca Mountain to Death Valley (by way of the Amargosa Desert and the Funeral
Mountains) was unsubstantiated (but not inconsistent with) with available data. They suggest
that ground water from Yucca Mountain ultimately discharges at Franklin Lake playa through
evapotranspiration (CZA90).
Discharge through ground-water withdrawals occurs within the model domain in the Amargosa
Desert for agricultural and domestic use. This discharge, which was estimated in the USGS
regional flow model at about 6,300 mVd, occurs mostly in the southwestern corner of the model
domain. The authors (ZYV97) believe that this discharge may be responsible for the
southwestwardly oriented gradient which appears to have persisted since the 1950s (KIL91).
VI-57
-------
Numerical Modeling Difficulties and Simplifying Assumptions. In the site-scale modeling, the
following assumptions are applicable (ZYV97):
1) The hydrogeologic framework is an appropriate description of the principal
hydrogeologic units and faults.
2) Permeability is invariant within each hydrogeologic unit.
3) Ground-water flow occurs in three dimensions and within the rock mass (which
includes both rock matrix and fractures).
4) Ground-water flow system is isothermal at 44°C (the effect of this assumption
was tested by simulating the system at 20°C).
5) Hydraulic heads of the potentiometric surface along the north, south, east, and
west edges of the modeled area are an appropriate data set for specifying
boundary conditions along the sides of the model.
6) The system is at steady state so that ground-water flow into and out of the flow
domain is invariant with time.
7) Volumes associated with the finite-element mesh are sufficiently large so as to
exceed the representative elementary volume necessary to simulate fracture flow
as porous-media flow.
8) A no-flow boundary at the base of the model approximates hydrologic conditions.
9) The large hydraulic gradient is part of the saturated zone and not an artifact of
perched-water occurrence.
10) Recharge is assumed to occur only at the top of the model along upper Fortymile
Wash; all other nodes on the top of the model are specified as a no-flow boundary.
Assumptions 5 and 8 are not and have not been supported by field data. The authors (ZYV97)
found, however, that they both represent an expedient means to assign boundary conditions,
which may have affected model calibration. Additional discussion of the site-scale model
calibration is presented in later in this section.
VI-58
-------
Sub-Site Scg'p rnnceptual Model
DOE's sub-site-scale conceptual model of the saturated zone at Yucca Mountain scale model
also places the saturated zone within a sequence of dipping and faulted tertiary volcanic rocks,
underlain by Paleozoic carbonates at depths greater than 1200 meters beneath the water table.
The sub-site scale conceptual model, however, places considerably more emphasis on the effects
that faults will have on ground-water flow than either the regional or site-scale models, as
discussed below.
The volcanic rock sequence is comprised of variable thickness tuffs, lava flows, and volcanic
breccias. Water levels in boreholes in the immediate area around Yucca Mountain are generally
either within or below the Topopah Spring Tuff. Borehole flow meter surveys show that discrete
fractured zones are the dominant pathways for saturated zone flow. More than 90% of pumped
flow commonly emanates from discrete zones within densely welded sections of tuff, typically
located in the central section of a particular unit. Fluid originates from fault zones and other
intervals in some boreholes. The high permeability of the densely welded tuff is due to the high
density of cooling joints, which generally form a fractured zone subparallel to dip. The Prow
Pass Bullfrog, and Tram Tuff each exhibit moderate to densely welded mid-sections, and are the
most permeable units within the saturated tuffs. Non-welded tuff has relatively few open
fractures, and is therefore less permeable. These sections are situated above and below the more
densely welded intervals within each geologic unit, thereby producing a layered permeability
heterogeneity. In addition, the dipping and faulted hydrostratigraphic units produce a
heterogeneous distribution of rock -units at the water table. The fate of potential radionuclides
reaching the water table is therefore a function of the infiltration source in addition to the flow
within the saturated zone.
Faults that displace stratigraphic units are pervasive at Yucca Mountain (DAY96). The dominant
fault set is composed of steeply dipping normal faults which have offsets on the order of
hundreds of meters and which are laterally continuous for kilometers. These faults are north-
trending and exhibit a surface spacing of approximately 1 km. When stratigraphic units are
displaced by faults, abutment of high permeability intervals against lower permeability intervals
can result. In this case, fluid pathways may diverge in three dimensions due to the presence of
low permeability structures. If fault zones exist, due to the presence of brecciated rock, for
example, they could provide the vertical pathways that link the displaced high permeability
zones. Conversely, mineralization within a fault zone may create a flow barrier if a high
permeability unit is not made discontinuous by the fault. In either case, fault displacements
VI-59
-------
produce discontinuities in stratigraphic units, which produce large scale permeability
heterogeneity. Northwesterly-striking faults with offsets on the order of meters to tens of meters
may introduce similar, yet smaller heterogeneities within the blocks bounded by normal faults.
Fluid between faults may be effectively isolated and have travel times orders of magnitude larger
than surrounding areas (flow stagnation), a feature suggested by some geochemical data. Such a
phenomenon would enhance the storage capability of portions of the saturated zone. The very
low horizontal gradient downstream of the repository may be the result of barrier faults or
divergence of flow upstream. West of Solitario Canyon Fault water elevations range from 775-
780 m above sea level and the hydraulic gradient ranges from 0.02-0.04 (TUC95). Ervin et al.
(ERV94) interpreted water to be mounded against the west side of the fault and a splay of the
fault. They attributed the influence of the fault to a low permeability fault gouge. However,
faults that simply displace units and that do not possess internal permeability may also have
considerable effect on flow geometry. Some fault zones may be infiltration sources at the water
table, and thereby influence the potential source locations of radionuclides at the water table.
Several researchers have proposed that the high temperature anomalies observed at the water
table near the Solitario Canyon and Paintbrush Canyon faults may be due to up welling of warm
water through permeable fault zones that extend into the Paleozoic formations (SZY89). This up
welling could result in dilution of radio nuclides because of the introduction of additional fluid
into the tuff aquifer. Conversely, the up welling could restrict vertical mixing. Vertical flows
may also be caused by the large-scale geologic heterogeneity described above. Geologic units
have different thermal conductivities, and these differences coupled with the faulted structure
could also be a contributing factor to the observed water table temperature anomalies (COH97).
In general, the large-scale heterogeneity created by superposition of the multiple sub-horizonal
flow zones, displacement of strata by faults of varying hydrologic properties, and variable
thickness and dipping geologic units will result in a complex 3-dimensional flow geometry.
Actual flow directions at depth are very likely different from those inferred from water table
elevations. In summary, saturated zone flow is affected by a) the presence of high- and low-
permeability faults which offset hydrogeologic units sub-vertically; b) potential up welling of
fluids from the Paleozoic carbonate formation via conductive faults; c) thermal conduction and
convection; d) heterogeneous permeability distribution within geologic units; e) variation of dip
and thickness of strata; f) distribution of infiltration from the unsaturated zone, and g) three-
dimensional channelization, hydrologic mixing and dilution produced by the above processes.
VI-60
-------
Computer Code Selection
Regional Flow Model
The computer code selected by D'Agnese et. al. (DAG97) to perform the regional flow modeling
is MODFLOWP (HIL92). As documented by Hill (HIL92), MODFLOWP is an adaptation of
the U.S. Geological Survey 3-D, finite-difference modular ground-water flow model,
MODFLOW (MCD88) in which nonlinear regression is used to estimate flow-model parameters
that result in the best fit to measured hydraulic heads and flows.
MODFLOWP is a block-centered finite-difference code that views a 3-D flow system as a
sequence of layers of porous material organized in a horizontal grid or array. The horizontal grid
is generated by specifying array dimensions in the x and y dimensions.
Flow between cells in each model layer is controlled by user supplied transmissivity values.
Similarly, flow between the layers is controlled by user supplied values of vertical transmission
or leakage.
The remainder of the model inputs describing boundary conditions, recharge, evapotranspiration,
spring flow and well discharge are specified using arrays or lists of row-column cell locations.
The model calculates ground-water heads based on the model boundary conditions, and
transmissivities.
Site-Seal? F1"W Model
To model the saturated-zone flow system at the site scale at Yucca Mountain, several simulation
capabilities were considered important, including the ability to: (1) simulate 3-D transient
ground-water flow and heat transport, including 3-D representation of spatially variable
permeability porosity, and thermal conductivity; (2) allow specification of constant pressure,
constant hydraulic head, constant fluid and heat flux boundary conditions; (3) represent
discontinuous, irregularly shaped 3-D hydrogeologic units; (4) perm,, spec,fica«on of dual
permeability and porosity representing both fiacture and matr* flow; (5) represent hydrauhc-
Lad and tempelre observation points where they occur u, 3-D space; 6) cahbrate the mode,
with respect to observations of hydraulic head and temperature through the use of automated
VI-61
-------
parameter estimation techniques; and (7) directly interface the resulting flow model with
radionuclide transport models used in Performance Assessment of the Yucca Mountain site. This
list includes features of the model not used in the present report, but important for anticipated
modeling efforts. The FEHMN simulation code was selected because it possessed these
capabilities when coupled with the mesh generation software, GEOMESH (described later in this
report), and with the model-independent parameter estimation software, PEST (also described
later in this report). The following section discusses the theory for many aspects of FEHMN.
The FEHMN (Finite Element Heat Mass Nuclear) computer code is capable of simulating flow
and transport through both the unsaturated and saturated zones. FEHMN is a non-isothermal,
multiphase flow and transport code. It can simulate the flow of water and air, and the transport
of heat and contaminants, in 2- and 3-D saturated or partially saturated, heterogeneous porous
media. The code includes comprehensive reactive geochemistry and transport modules and a
particle tracking capability. Fractured media can be simulated using an equivalent continuum,
discrete fracture, dual porosity or dual permeability approach. The basic conservation equations,
constitutive relations and numerical methods are described in Zyvoloski (ZYV83), Zyvoloski
(ZYV86), Zyvoloski and Dash (ZYV90), Reeves (REE94), and Zyvoloski et. al. (ZYV95).
Sub-Site-Scale Flow Model
Modeling at the sub-site scale uses the integral finite difference simulator TOUGH2 (PRU87).
TOUGH2 can simulate non-isothermal flow and transport in single and dual-porosity media.
TOUGH2 utilizes an integral finite difference formulation, which, in part, enables construction
of a mesh with highly variable gridblock geometries that are necessary to properly represent the
geologic structure.
Layering and Gridding
Regional Flow Model
DOE's regional flow model developed by D'Agnese et. al. (DAG97) used 163 rows, 153
columns and 3 layers. The 78,817-cell model is oriented exactly north-south. Grid discretization
along both rows and columns was set to 1,500 m. The three model layers represent
hydrogeologic units at 0-500 m, 500-1,250 m, and 1,250-2,750 below the interpreted water table;
which are 500,750, and 1,500 m thick. The first and second model layers are interpreted as
VI-62
-------
simulating local and subregional flow mostly within valley-fill alluvium, volcanic rocks and
shallow carbonate rocks; the third layer is interpreted as simulating regional flow in the volcanic,
carbonate and clastic rocks. The authors (DAG97) also point out that the use of only one model
layer to represent each of the local, subregional, and regional flow paths may potentially result in
model error in areas of significant vertical flow, particularly if the vertical flow component is
somewhat complicated.
-grate Flow Model
The site model area is 1,350 km2 and extends 30 km west to east, and 45 km north to south.
The model used a detailed three-dimensional hydrogeologic framework model (HFM) to
characterize the heterogeneous, porous, and fractured media beneath Yucca Mountain. This
approach is different from the regional model in that rather than inserting hydraulic properties
into directly into the numerical mesh, the HFM model was first developed so that it could be
converted into a tetrahedral mesh, using GEOMESH (GAB96).
The site-scale finite-element mesh is spaced at 1,500 m and consist of 9,279 nodes. The authors
(ZYV97) note that this is a very coarse resolution and is only suitable for initial calibration of a
preliminary flow model. For example, the upper volcanic confining unit is much more extensive
in the model than in reality. Because of the coarse grid increment, offsets across faults are also
much less abrupt in reality.
The authors (ZYV97) indicate that a 250-m sampled mesh is planned for the future with
improved error checking, which will improve the quality of both the framework model and the
numerical grid based on the framework model.
Suh-Site-Sca'« Elnw Model
The model covers an area of 150 km2 (10 x 15 km). The horizontal mesh has 1993 gridblocks.
There are currently 23 model layers that define different hydrostratigraphic units. Including fault
gridblocks, the total number of gridblocks is approximately 50,000. Gridblock sizes are highly
variable, mostly ranging in size from approximately 50 to 500 m on a side; several approach 1
km2 near model boundaries. The locations of boreholes correspond to the gridblock nodal points,
making conditions at the particular borehole correspond directly to conditions in the gridblock.
Horizontal dimensions of fault gridblocks are approximately 150 m x 150 m. These dimensions
VI-63
-------
do not represent the actual fault zone width, but rather are the gridblock dimensions over which
fault zone properties are averaged. (Note that for the pure displacement conceptualization of
modeled faults, the fault zone is given properties corresponding to the hydrostratigraphic units on
either side of the fault, and thus the fault effectively has no width whatsoever.)
The finely discretized region near the center of the sub-site-scale domain corresponds to the area
near the C-hole complex, borehole P#l, and borehole ONC#1. This area was finely discretized
because transient hydraulic tests can be simulated in this area to design future hydrologic tests
and simulate smaller-scale flow processes and effects.
The northernmost model area corresponds to the location of the large hydraulic gradient (FRI94).
Several conceptual models have been proposed to explain this feature (e.g., FRI94, BOD96).
The hydrogeologic structure beneath the water table in this area is poorly understood, and the
authors (COH97) believe that different hypothetical models should be tested by numerical
simulation to more fully constrain the possible causes.
The water table defines the top of the model. Use of this surface assumes that the water level
measurements in borehole G-2 and WT-6 represent a water table. The bottom of the model is
defined by the base of the Lithic Ridge Tuff. This unit is a thick confining structure that
separates the Crater Flat Tuff aquifer and the lower carbonate aquifer. The higher head anomaly
observed in UE-25p#l deemed indicative of the Paleozoic formation actually first appeared in
the older unnamed tuffs just beneath the Lithic Ridge Tuff and above the Paleozoic carbonates
(CRA84). The older tuffs lie beneath the Lithic Ridge Tuff. The isopachs (i.e., thickness) of
units beneath the Lithic Ridge Tuff and lower carbonate aquifer are unknown, as only p#l
penetrates the carbonate aquifer. Therefore, an isopach cannot be constructed for these intervals.
Using the base of the Lithic Ridge as a lower boundary preserves the lower confining structure
and still enables consideration of a higher head boundary condition. In addition, Lithic Ridge is
the lowermost unit for which an approximate isopach has been developed (CAR86).
Gridblocks are discretized in a manner that preserved the thickness, orientation, dip, and lateral
continuity of strata. The vertical dimension of the model is composed of gridblock layers with
variable thickness distributions, which are then concatenated to produce a sequence of layers that
correspond to the actual vertical distribution of units observed in boreholes. The lateral
continuity of hydrostratigraphic units is thus preserved, which is important since many strata
exhibit densely welded, high permeability zones parallel to dip. Displacement by faults is also
VI-64
-------
explicitly considered. The number of model layers in areas between faults remains constant at
23 Because not all geologic units are present beneath the water table, the total thickness of a
geologic unit that intersects the water table may be distributed over several thin model layers
near the top All gridblock nodes in the model are connected vertically to the adjacent node(s)
located directly on top and bottom of the same column of gridblocks. The nodes of gridblocks
located in regions between faults are also connected laterally to adjacent nodes of the same layer.
Fault gridblocks are connected differently, as described below.
The choice and rationale for representing particular faults for the model is described by the
authors in a previous report (COH97), a brief summary of which is presented below. Day et al.
(DAY96) divide the faults at Yucca Mountain into three main groups. The dominant set is
composed of north-trending normal faults that have steep down-to-the-west displacement over
most of their length. These faults generally have offsets on the order of hundreds of meters, and
define the boundaries of the relatively intact blocks of east-dipping Miocene volcanic strata.
They have therefore been termed block-bounding faults. The second set is composed of
northwesterly striking strike-slip faults located in the northwest section of the model region.
These faults have offsets on the order of meters to tens of meters only. Thirdly, intrablock faults
are those not continuous on scales greater than the defined fault blocks and not connected with
either set. There are very little data to indicate fault persistence and structure with depth,
especially below the water table. The linearity and dip of faults with depth, occurrence of fault
zones, and fault zone flow properties beneath the water table are mainly unknown. Given the
inherent uncertainty of fault properties beneath the water table and the intended modeling
objective of investigating the potential effects of the complex geologic structure on flow, only
north-trending faults, which have the greatest displacement of any of the fault sets, are
considered. Most of these are block-bounding faults. The location of faults comes from the
surface location of faults as defined in ISM2.0 (CLA97). Most faults at Yucca Mountain have
steep dips on the order of 80°; therefore the general fault structure in the model is preserved by
assuming vertical faults.
All faults can be assigned a fault zone with particular flow properties or can simply be treated
displacement-only faults. The gridding scheme enables hydrostratigraphic unit displacement
across the fault to be considered realistically, as described above.
as
VI-65
-------
Initial and Boundary Conditions
Regional Flow Model
Specified-head boundary conditions surrounding the entire model are based on interpolation of
measured head data. The slope of the potentiometric surface toward the southwest may be
indicative of ground-water withdrawals that were not specified in the model. Discharge by
pumping wells occurs in the Amargosa Desert in the southwest part of the model domain, but
this pumping was not explicitly represented in the model. However, the influence of these well
withdrawls is implicitly reflected in the specified-head lateral boundary conditions applied along
the southern boundary of the model. No-flow boundary conditions were specifed along the base
of the model. Specifically, the model boundaries used for the regional study are the following:
• In layer one, all lateral boundaries were designated as no-flow except along the
western boundary at Death Valley, where constant head boundaries are designated at
Cottonball, Middle and Badwater Basins, and Saratoga Springs. Head values along
this boundary were defined using the potentiometric-surface map developed for the
modeling study and reflect the nearly perennial ponds supported by ground-water flux
out of the system.
• In layer two, all lateral boundaries were designated as no-flow because at these depths
flow is believed not to cross the lateral boundaries.
• In layer three, the lateral boundaries were designated as no-flow except at flow
locations along the northern and eastern boundaries. These were assigned constant-
head values to reflect possible or perceived interconnections along buried high
transmissivity structural features with regional flow paths in adjacent valleys outside
the model domain. The head values were selected to correspond to measured water
levels.
Site-Scale Flow Model
A key concern addressed in the site-scale flow modeling is the compatibility of the regional flow
model and the site-scale flow model. Although, ideally, the output from the larger-scale regional
model would be used as direct input for the boundary conditions to the site model, the authors
(ZYV97) point out several factors, however, that make such an interface impractical. First, the
geologic model is defined in more detail at the site scale, so an exact piecing together of the
VI-66
-------
models is impossible. Second, the regional model, focusing on larger-scale issues, does not have
a detailed representation of geology near the site and does not attempt to include features such as
the large hydraulic gradient in the calibration. Because of these differences, the models are
compared with each other for consistency, rather than coupled more directly.
Hydraulic-head data are considered to be more accurate than flux data within the site model, and
for that reason were chosen for specifying boundary conditions for the model despite the
influence that such a constant-head boundary is likely to have on a model being calibrated to
hydraulic-head observations. Specified-head boundary conditions are based on the
potentiometric surface that includes the large hydraulic gradient. However, no measured vertical
head distributions at the boundaries of the model exist. The regional model (DAG97) does
provide coarse estimates of vertical hydraulic head, but these were not used in assigning the
boundaries at the site model. An appropriate set of hydraulic-head values on the outside nodes of
the model consistent with the potentiometric-surface data was computed for use in specifying
constant hydraulic-head boundary conditions.
Improving the representation of the lateral boundary conditions is considered by the authors
(ZYV97) to be of primary concern for future modeling efforts. Alternate ways to specify
boundary conditions within the site model exist These include but are not limited to: (1)
specifying constant heads only along the top edge of the model (this was not done because no
flow would be allowed at the remaining nodes along the sides); (2) specifying flux explicitly
(this was not done because of the difficulties in redistributing flux from the regional model onto
the sides of the site model); or (3) projecting hydrostatic head from the top edge down the
outside faces of the model (this was not selected because it forces flow to be horizontal).
As noted previously, one concern with specifying hydraulic heads on all model sides, while
calibrating using hydraulic heads within the model, is that the specified heads are likely to
dominate the simulated heads at the observation locations. The severity of this problem was
tested in independent numerical experiments using a model developed by Sandia National
Laboratory of a subdomain that included Yucca Mountain. The results indicated that specified
pressure (constant head) boundary conditions could be applied while still observing changes in
model simulated pressures as a result of changes in model permeability values. Because the site
model covers a substantially larger area than that of the Sandia model, application of specified
head boundary conditions was considered to be less of a constraint. However, the use of any
VI-67
-------
specified-head boundary condition will have some constraint on model calibration. As a result,
the fluxes in and out of the model will have to be checked against any available data.
Sub-Site-Scale Flow Model
The local-scale saturated zone model is situated such that the western and eastern sides are nearly
perpendicular to the hydraulic gradient. Constant head boundary conditions are therefore placed
along the western and eastern sides of the model. A uniform and constant head is assigned to the
entire vertical column of gridblocks that underlay a particular gridblock at the water table. In
addition, constant head conditions are assigned to gridblocks along the sides in the northwestern
corner in the area of the large hydraulic gradient where the potentiometric contours are not
perpendicular to the model sides. The latter condition assumes the high gradient area represents
a water table gradient, and it maintains the observed high gradients along the sides of the high
gradient area. Calibration still considers the high gradient area, however, since assignment of
different permeabilities in this area will produce different fluxes and hence head profiles
downstream. The northern side of the model immediately downgradient from the high gradient
area and the full southern side of the model are no-flux boundaries. The relative difference in
hydraulic head between gridblocks along head boundaries represents the relative elevation
change between gridblocks.
Model Parameterization
Regional Flow Model
Flow Parameters. The regional model lumped aquifer properties into 10 hydrogeologic units. In
reality, aquifer property variation is considerably greater; however, the properties were lumped
into a limited number of categories to facilitate the simulations. The subsurface materials of the
hydrogeologic framework model were initially classified into eight rock conductivity units
(RCUs). Each RCU represents mean hydraulic conductivity of several subsurface materials
whose interpreted characteristics, such as rock type, depth, and degree of fracturing resulted in
very similar hydraulic conductivity values.
Because each of the three model layers contained several hydrogeologic framework model units,
multiple RCUs were associated with each finite-difference model cell. The RCU occupying the
VI-68
-------
largest volume in the finite-difference flow model cell was assigned to each cell. To reduce the
number of parameters that would need to be estimated, the authors (COH97) reclassified the
layer maps by combining the eight RCUs into four hydraulic-conductivity zones (K-zones)
representing large (Kl), moderate (K2), small (K3) and very small (K4) hydraulic-conductivity
values. The resulting K-zones were not contiguous; each K-zone included cells distributed
throughout the model. The 50th percentile K-value for each of the zones was used for the initial
hydraulic conductivity values assigned to each K-zone. Transmissivity values for the model
layers were calculated by multiplying the applicable K-zone values by layer thickness.
F.vapntranspiration. The authors (COH97) expressed evapotranspiration (ET) in terms of a linear
function based on three variables: (1) land-surface altitude, (2) extinction depth, and (3)
maximum ET rate (MCD88). Each of these variables was specified from data sets including ET
area maps. Extinction depths were assigned for each unique ET area based on information about
plant type and ranged in value from 0 to 15 m. Each of these data sets was resampled to a 1,500
m grid. Since the Death Valley saltpan was simulated as a constant head boundary, it was not
assigned an ET rate.
Recharge. To define ground-water recharge, a recharge potential map was resampled to a
1,500 m grid and reclassified into an array for MODFLOWP containing four zones associated
with high (RCH3), moderate (RCH2), low (RCH1), and no (RCHO) recharge potential
parameters. Each parameter defines a percentage factor that represents the amount of average
annual precipitation that infiltrates. Average annual precipitation is defined by a multiplication
array in MODFLOWP. These zone and multiplication arrays, therefore, along with the
parameter values, define the recharge distribution for the model area. In initial parameter-
estimation runs, recharge rates based on fixed percentages were lumped into a single recharge
parameter (RCH) for simplicity.
gprings The regional springs data set was utilized to specify the row-column locations of spring
nodes. All but three groups of springs were thought by the authors (COH97) to discharge from
deep regional flow paths, and were, therefore, assigned to layer three. Sand Springs, Indian
Springs and Cactus Springs are interpreted as discharging from more localized flow paths in
model layer 2. Springs were specified using the general-head boundary package for which the
altitude and conductance of spring orifice are assigned. Because the conductance term is poorly
VI-69
-------
known, springs were grouped according to geographic location and a conductance parameter
(GHB) assigned for each of the groups of springs.
Pumping Wells. The MODFLOWP well package was used to simulate the amount of ground
water pumped from the system. Because most of the water pumped from the wells is relatively
shallow ground water, all pumping wells were located in the first model layer. The water-use
well data set was used to specify the grid-cell locations and approximate ground-water volumes
being removed from the model domain. The authors (COH97) believe that approximate
pumpage rates probably exceed actual values, so they assigned two parameters to make it easy to
modify simulated pumpage. The two parameters are multiplication factors representing the
percentage of pumpage included in the simulation. WEL2 represents the parameter applied to
the Pahrump Valley area, which bears the majority of ground-water withdrawal hi the region.
WEL1 represents the parameter applied to the remainder of the model domain. It is unclear
where the authors (COH97) obtained their estimated pumping rates.
Observation Data. Measured hydraulic heads and spring discharges were used by MODFLOWP
during parameter estimation (i.e., calibration) to provide values to define the objective function
for the model simulation (i,e., to see how well the model results matched actual field measured
values).
Faults and Geologic Structure. The regional model does not incorporate some structures
explicitly. In the regional model, northeast-southwest trending regional structures are identified
as zones of large permeability and northwest-south-east trending regional structures are identified
as zones of small permeability. Because of the large-scale of the regional model, hydraulic
properties of such features used in that model may not be appropriate at the scale of the site
model. The area underlying Fortymile Wash was also identified as a zone of large permeability
in the regional model. Because the site model does not explicitly consider many structural
features, the hydraulic conductivity ranges for these hydrogeologic units are much larger than
those defined for the regional flow model.
VI-70
-------
Site-Scale Flow Model
Potentiometric Data. Hydraulic-head values from eighty boreholes, located within the model
area, were used in model construction. Twelve of the boreholes (USW H-l, USW H-5, UE-25b
1, USW H-6, USW H-4, USW H-3, UE-25p 1, UE-25c 1, UE-25c 2, UE-25c 3, and two
unnamed boreholes) have multiple piezometers. Forty-five of the boreholes are either uncased or
have fifty percent or more perforated casing. Twelve boreholes are cased, while the presence or
absence of casing is unknown for eleven of the boreholes. Many of the boreholes are "dry" until
a fracture zone is intercepted, at which point the water level in the borehole rises to a static level.
Because of long open or perforated intervals, many boreholes intercept multiple permeable
zones. As a result, the hydraulic head in many of the boreholes represents a composite head.
Permeability Zones. In FEHMN, nodes are grouped into zones in which rock and hydraulic
properties, and boundary conditions may be specified. There are 16 zones used in the model that
define nodes pertaining to hydrogeologic units with specific permeability and porosity values.
Permeability values used in the model are considered preliminary. Only the nodes closest to
Fortymile Wash and Solitario Canyon Fault are represented explicitly as fault or fracture zones.
In the numerical model, Solitario Canyon is a separate permeability zone and forms a barrier to
flow.
The permeability values used in the model are derived partly from a sequence of parameter-
estimation simulations. Permeability specified for the middle volcanic aquifer (1.6 x 10'14 m2) is
about three orders of magnitude less than values reported by Geldon (GEL96, p. 70) for tests at
the C-wells. The authors (COH97) provide a possible explanation for this discrepancy which is
that the C-hole tests reflect hydraulic conditions in locally faulted and intensely fractured rock.
The possibility of such a condition was tested to a limited extent by specifying a vertical zone,
extending approximately 5 km southeast from the C-wells, with a larger permeability of 1 x 10'11
m2. The small increase in the resultant sum of squared residuals (23,262 m2) over that of
simulation 40 (23,163 m2) indicates that the model was insensitive to such a zone and that such a
zone might be possible. This zone would be consistent with northwest-southeast oriented faults
in the area. The small change could also be an artifact of the density of observation points near
this zone of large permeability coupled with the small horizontal hydraulic gradient. However,
because of the.non-unique nature of the model, an overall large permeability (1 x 10'" m2) for the
entire middle volcanic aquifer also is possible, but would require a considerably different
VI-71
-------
combination of permeability values for the other hydrogeologic units to achieve calibration.
Investigating the possibility of a zone of large permeability would be more appropriate using a
more finely sampled hydrogeologic framework model and associated finite-element mesh.
Large Hydraulic Gradient Zone. To reproduce the LHG on the north end of Yucca Mountain,
where the apparent water-table altitude changes about 300 meters in a distance of less than 2 km,
an additional zone was defined within the model as an east-west barrier to flow. Large head
residuals (i.e., errors during calibration) had occurred at the wells defining the LHG prior to the
definition of this zone. Because no independent geologic evidence for a structure exists, and
because the length of such a structure is in question, the coordinate defining the eastern extent of
this zone was selected as a parameter and allowed to vary from the western limit of the zone to
the eastern edge of the model during earlier scoping simulations. Model fit was best when the
zone was extended to the eastern edge.
A single zone extends from the top of the water table to the bottom of the model, and is one node
thick that, in essence, forms a 2-D plane. The present model zonation results in uniform
permeability changes over the entirety of the upper volcanic aquifer, the upper volcanic
confining unit, and the middle volcanic aquifer wherever they occur within the model.
The authors (COH97) note that an alternate approach to representing the LHG would be to
further subdivide the zones defining the upper volcanic aquifer, the upper volcanic confining
unit, and the middle volcanic confining unit along the east-west occurrence of the LHG. This
subdivision would then allow reduction of the permeability of these units where they occur to the
north of the gradient, producing a 'spillway' model (FRI94). The authors (COH97) also point
out that a 250-m resolution mesh would better represent the large hydraulic gradient as well as
the fault and hydrogeologic unit distribution, coincident with the LHG as portrayed by Fridrich
et. al. (FRI94).
Solitario Canvon Fault Zone. Based on hydrologic and hydrochemical data, the Solitario Canyon
fault appears to act as at least a partial barrier to ground-water flow. Currently, the authors
(COH97) have not specifically simulated the Solitario Canyon fault in the model. However, they
have included a hydraulic conductivity zone to better reproduce the approximately 50-meter
change hi hydraulic head across the Solitario Canyon fault system. Its exact correlation with
Solitario Canyon fault is approximate owing to the coarseness of the grid. This zone was
VI-72
-------
introduced after initial attempts to simulate the 50-m change in head resulted in large hydraulic-
head residual values (i.e., errors). No hydraulic-test data exist to provide information about the
permeability of the Solitario Canyon fault zone.
Fortymile Wash Recharge Zone. Many lines of evidence indicate recharge occurs in upper
Fortymile Wash. Therefore, the authors (COH97) placed a zone in this area to specify recharge
in upper Fortymile Wash. The zone consists of seven nodes each with a uniform mass recharge
rate of 0.22 kg/s.
Sub-Site-Scale Flow
Model
Hydrogeologic Units. Permeabilities of the Topopah Spring units for the sub-site-scale model
are derived from Bandurraga et al. (BAN97). These values are the averaged values from
inversion of pneumatic data from the unsaturated zone. The sources of data are described in
Bandurraga et al. (ibid.). The values shown are derived from analysis of fracture traces and
results of air-injection tests.
Luckey et al. (LUC96) provide apparent hydraulic conductivity values for saturated zone units, t
as obtained from pumping and other hydrologic tests conducted in the 1980's. The authors
(COH97) observe that no single value of permeability for each unit can be assigned, given the
inherent heterogeneity in the system and general paucity of data. The values chosen by the
authors (ibid.) for each unit fall within the range of values available, and also maintain the
contrast in unit permeabilities due to their geologic structure. Thermal properties of each unit are
derived from laboratory tests on rock core (BR097), and from analysis of temperature profiles in
boreholes (SAS88). These values are very well constrained relative to the hydrologic properties.
Porosity values are representative of the fracture porosity. An effective fracture porosity has
been difficult to determine for all flow models at Yucca Mountain, and simulations presented
later in this section illustrate the highly variable flow estimates obtained using different
porosities. The authors (COH97) apply a single-continuum model that considers fracture
porosity for most units since fractures are the dominant fluid pathways in the saturated zone.
The authors (ibid.) further believe that the fracture-dominated flow conceptualization is
supported by two lines of evidence: (1) borehole tracejector surveys have consistently shown
water derives from several discrete fractures or fractured zones (e.g., WAD84); and (2)
permeability calculations from pumping tests are many orders of magnitude larger than matrix
permeability (e.g., GEL96).
VI-73
-------
Drill-hole data for boreholes in the saturated zone were carefully reviewed by the authors
(COH97) to establish correlations, note discrepancies, and delineate hydrostratigraphic units for
the model. The geologic unit thickness values used to construct the model are defined by
lithologic logs (CLA97). These may be an apparent thickness, since faults may intersect the
borehole within a particular unit. For example, the thicknesses of the Tram Tuff and Lithic
Ridge Tuff in borehole p#l are very likely less than then- true values because a fault observed in
the borehole separates the units.
The Lithic Ridge Tuff located beneath the Tram Tuff and the lava flows and flow breccias
situated between the Tram and Lithic Ridge units are included in the model. A layer of lava flow
and flow breccias was penetrated in boreholes H 1, H 6, G-l, G-2, and H-5. The observed
thicknesses are 119,253,118,24, and at least 176 m, respectively. This unit is located between
the Lithic Ridge and Tram Tuffs of the Crater Flat Tuff, which are adjacent in the rest of the
model area. Most of the unit in each of these holes is altered to clays and zeolites (MAT96). The
unit therefore is considered a confining unit (i.e., low permeability). The isopach (i.e., thickness)
of the thin lava flows and breccias and the associated underlying thin bedded unit beneath was
constructed using the observed thicknesses and zero thickness constraint imposed by nearby
boreholes H-3, H-4, b-1, and WT-6, in which the unit was not observed (COH97).
The Lithic Ridge and lavas are lumped into a single lower confining structure due to their
common low permeabilities. This confining structure exhibits large variations in its thickness
over the model area, which may have implications for where large vertical gradients are
observed.
In the remaining upper stratigraphic units, an important hydrogeologic feature results because of
the intersection of the water table with the dipping units. Different units are located at the water
table in different areas. In addition, the actual thickness of each unit beneath the water table will
be less than its total thickness.
Selection and Representation of Faults in Model. The fault locations in the sub-site-scale model
are generally preserved as they occur at the surface. Due to the disparity in scales between
geologic mapping and model discretization, several closely-spaced faults observed in the field
VI-74
-------
are represented as one model fault, or one model fault accounts for the offset across two nearby
faults.
Faults are represented as continuous gridblock bands, which preserve the natural continuity of
faults. Fault displacement makes the contact area between different units on either side of the
fault vary, depending on the fault offset. At large fault offsets, hydrostatigraphic units may be
completely discontinuous at the fault and abut against different hydrostratigraphic units. Fault
displacement also varies along strike.
The column of gridblocks located along fault traces uses a special node-connection scheme that
enables proper representation of the large scale heterogeneity created by fault displacement. Due
to fault displacement, hydrostratigraphic units partially to fully abut against other units on the
adjacent fault side. The number of vertical gridblocks at fault traces is 46, which provides the
additional nodes needed to account for the fractional contract area that connects a particular unit
on one side of the fault to a unit on the other side. The rock property of the individual fault
blocks correspond to the rock of the adjacent layer node on one side. This assignment models a
displacement-only fault. The authors (ibid.) point out that a bulk fault zone property could later
be modeled by setting the properties of fault gridblocks to different values. This is done when
considering a fault as a zone of differing hydrologic character from that of the surrounding units.
Model Calibration
Regional flow Model
Calibration of the regional model by strictly trial-and-error methods was judged by the authors
(DAG97) to be both ineffective and inefficient; therefore, nonlinear regression methods are used
to estimate parameter values that produce the best fit to observed heads and flows.
Conceptual M"de1 Testing. During calibration, a number of conceptual models were evaluated
using the regression methods in MODFLOWP. A best fit to hydraulic-head and flow
observations was calculated for each conceptual model. Evidence of model error or data
problems were investigated after each model run. These analyses were used in conjunction with
independent hydrogeologic data to modify, and hopefully improve, the existing conceptual
model, observation data sets, and weighing. No modifications were made simply to improve
VI-75
-------
model fit; supporting independent hydrogeologic criteria were also needed before modifications
were made.
To perform the conceptual model testing, the authors (DAG97) adjusted the location and type of
flow system boundaries in the north and northeast parts of the model area to test the premise that
the regional flow system could be receiving interbasinal flux from adjacent basins. Although
water-level data exist adjacent to the model boundaries, considerable uncertainty remains
concerning the existence of such fluxes and their volumes.
Modification to the numerical model involved increasing or decreasing the size of the constant-
head boundaries, moving the locations of the boundaries to different model layers, and adding
new constant-head boundaries to the northeast. In general, these modifications provided very
clear results. The optimal location for the constant-head boundaries used to simulate interbasinal
flux conditions was the third layer. Locating the constant-head boundaries in the upper layers of
the model often led to extremely large deviations from observed heads.
The most appropriate boundary conditions between Sand Spring Valley and Emigrant Valley
were evaluated. Simulating a constant-head boundary in this region instead of a no-flow
boundary resulted in extremely large residuals for heads (100 m too high) in the northern part of
the model domain, and large residuals for spring flows (simulated flows 50 percent too large) at
both Ash Meadows and Furnace Creek Ranch. As a result, this boundary was redefined as no
flow boundary. The constant-head boundaries at Railroad, Stone Cabin and Ralston Valleys,
however, were needed to simulate spring flows close to measured flow at both Grapevine
Springs and Oasis Valley. The constant-head boundary at Pahranagat Valley was needed to
match to the measured domain and spring flows at Ash Meadows.
Hvdrogeologic Framework Testing. Four types of hydrogeologic framework variations were
considered during calibration of the regional model. These include: (1) adjustment of hydraulic
conductivity zones to improve numerical stability, (2) addition or refinement of hydraulic
conductivity zones to better define hydrogeologic units or geologic structures included in the 3-D
hydrogeologic framework model, (3) addition of hydraulic conductivity zones to better represent
interpreted geologic structures that were not included in the hydrogeologic framework model,
and (4) addition of new hydraulic conductivity zones required to better represent faulted terrains
supplying ground water to springs and discharge area.
VI-76
-------
Recharge Distribution Determination. The initial distribution of recharge areas was changed
during model calibration to determine the sensitivity to their extent and magnitude. Initially, a
single multiplication array was used to describe the rate of recharge, and a single multiplication
parameter defined to adjust this rate. Initial modeling indicated that the model results were very
sensitive to this single recharge parameter. The large residual errors suggested that the use of a
single recharge parameter was an oversimplification, and that parameters in multiple recharge
areas would have to be estimated with the available data.
A detailed evaluation that had been conducted by the authors (DAG97) to delineate various
zones of recharge potential was used to divide the single recharge parameter into four zones.
Each was assigned a parameter that represented a percent of average annual precipitation that
infiltrates. In the final model, RCHO and RCH1 were assigned values of 0 and 1 percent, RCH2
and RCH3 were estimated by regression and were variable, ranging from 1 to 10 percent and 10
to 30 percent, respectively.
Calibration Results. After calibration, the regional model was evaluated to assess the likely
accuracy of simulated results. This is accomplished by comparing measured and expected
quantities with simulated values. The quantities included hi the comparisons are (1) hydraulic
heads and spring flows, which were matched by the regression; (2) hydraulic conductivities,
vertical anisotropy, and percent of precipitation that infiltrates, all of which were represented by
parameters estimated in the regression; and (3) water budgets.
An advantage of calibrating the model using nonlinear regression is the existence of substantial
methodology by which to evaluate model results. As will be demonstrated, these methods
produce a more thorough evaluation than is normally accomplished and reported when
calibrating using trial-and-error methods. The thorough analysis produces a good understanding
of the strengths and weaknesses of the model, and the likely accuracy of simulated results and
associated confidence intervals and other measures of parameter and prediction uncertainty.
The authors (DAG97) believe that the model reproduces the measured hydraulic heads and
estimated water-budget components reasonably accurately. They also believe that the estimated
parameter values include all aspects of the system that are most important for steady-state
simulation, and the parameter values that produce the best match between simulated and
VI-77
-------
observed hydraulic heads and flows; that is, the parameter values estimated by the regression, are
all reasonable.
Because the weighted residuals are not entirely random, some model error may be indicated.
This is related to the occurrence of large positive weighted residuals for hydraulic heads and
large negative weighted residuals for spring flows. In addition, weighted residuals are not
normally distributed. The authors (DAG97) believe that additional calibration may significantly
improve model accuracy. They further conclude that the model is a reasonable representation of
the physical system, but evidence of important model error exists.
One of the more apparent factors contributing to model error is the vertical discretization of the
regional system into three layers. While a three-layer model is an improvement on previous 2-D
and quasi-3-D models, simplification of the complex 3-D hydrogeologic framework into three
layers inevitably results in model error, particularly in areas with significant vertical flow
components. The introduced model error may translate into model bias in computed parameters
and all quantities computed using them, particularly head and flux. Furthermore, this potential
bias may be contributing to closeness of fit calculated for the model. The authors (DAG97)
believe that an evaluation of the extent of model error should be conducted. This evaluation may
include a series of cross-sectional or subsystem models with varying degrees of vertical
discretization. A comparison of the levels of detail in vertical discretization with the model fit
and computed parameter values would give some indication of the potential for model error.
They also indicate that the calibrated site-scale model has a flux of 465 kg/s from the south face
versus 323 kg/s over the same region for the regional model. This agreement is acceptable
considering the compatibility issue raised above. Furthermore, the recharge and discharge flux
values themselves, as input into the regional-scale model, have uncertainties that are larger than
the computed difference of these two models. Because the difference is within the range of
uncertainty of the actual flux values, the authors (DAG97) consider the two models to be, for all
intents and purposes, in agreement with one another.
Site-Scale Flow Model
The calibration of this model was achieved through the use of the model independent parameter
estimation software, PEST (WAT94). PEST uses nonlinear least-squares regression to estimate
parameters. The benefits of using nonlinear regression include: (1) expedited determination of
VI-78
-------
best-fit parameter values; (2) quantification of the quality of the calibration; (3) estimates of the
confidence limits on parameter estimates; and (4) identification of the correlation among
parameters.
PEST was selected because of the ability to couple it with FEHMN without significantly
changing the FEHMN software. PEST is designed to be used with virtually any model, provided
that one can identify: a) model input files; b) model output files; c) commands that invoke the
model; d) observation data; and e) model parameters. Each of the required input and output files
needs to be in ASCII format.
PEST was used to run FEHMN and to vary user-specified model parameters prior to each run
such that the weighted sum of the differences between observed and simulated values of
pressure, hydraulic head, or temperature is minimized using nonlinear regression. The
optimization is accomplished using the Gauss-Marquardt-Levenberg method. The strength of
this method lies in the fact that it can generally estimate parameters using fewer model runs than
any other estimation method, a definite advantage for large models whose run times may be
considerable (WAT94).
As mentioned above, model calibration was attempted using nonlinear least-squares regression to
estimate parameter values. Permeability values were modified to achieve a close match to 94
measured hydraulic heads, all of which were equally weighted. Fluxes at the specified-head
nodes for the outside nodes were summed for each side of the model for comparison against
regional model values. The authors (ZYV97) believed that it may be advantageous to compare
flows for smaller parts of each side, but this was not done in the present work.
Several simulations using a pressure-based configuration instead of hydraulic heads, provided
experience regarding which parameters tended to be highly correlated, a condition which
indicates that the available data are not sufficient to estimate all parameters individually.
Hydrogeologic units with similar permeabilities were combined or "lumped" as parameters to
gain some insight about the hydrologic importance of areas of large and small permeability. For
example, the permeability parameter of the middle volcanic aquifer (mva) was observed to be
correlated to the upper volcanic aquifer (uva). Experience has shown that spatially connected
hydrogeologic units with similar permeability which are oriented approximately parallel to the
direction of ground-water flow tend to be highly correlated, preventing independent estimates of
VI-79
-------
their associated permeability values. An initial strategy of the modeling focused on optimizing
permeability in those units that appeared to have sufficient information provided by hydraulic-
head observation points. In addition, a determination of which potential model parameters were
highly correlated was done using PEST by assigning as many model variables of permeability
and flux as possible so that correlation among parameters could be evaluated. From these
correlations, parameters either could be lumped with other correlated parameters, or set so that
parameter estimation could be achieved.
Forty PEST parameter estimation runs were done for various combinations of fixed and
estimated parameters. Fixed parameter values are not modified during a run; estimated
parameter values are adjusted using nonlinear regression. In most of the runs, one or two
parameter values are estimated; at most, five are estimated. Because so few parameters are
estimated without a thorough evaluation showing that the other parameters are unimportant, the
authors (ZYV97) believe that the regression runs need to be considered as very preliminary. The
results of the PEST simulations include 95% confidence intervals for the adjustable parameters,
which the authors (ibid.) indicate may or may not be meaningful, depending on many factors in
the model construction and parameter estimation processes. A large range in the 95% confidence
interval generally indicates that the data contain little information about the parameter. In many
instances, minimum values of 95% confidence intervals were estimated as negative values. Use
of a log transformation of such a parameter typically would result in a minimum value with a
large negative exponent (or essentially a minimum value of zero), indicating that insufficient
information was available to provide a good estimate of the parameter.
Sub-Site-Scale Flow Model
The sub-site model was calibrated to the observed water table. The high gradient area is
conceptualized as a zone of uniform permeability. Initial simulations used a range of rock
properties, and all fault zones were assigned a permeability of 10'13 m2. Steady-state simulations
using these parameters showed that the medium gradient across the Solitario Canyon Fault could
not be sustained with the assigned fault permeability. Permeability values were adjusted
manually and simulations were performed in an interactive manner. Subsequent simulations
showed that the dominant areas that needed permeability adjustments were the Solitario Canyon
and Iron Ridge faults. The authors (COH97) found that in order to reduce the simulated heads on
the eastern side of the Solitario Canyon Fault, a permeability of 10'16 m2 was needed. In order to
VI-80
-------
reduce the heads east of the Iron Ridge Fault to their approximate measured values, the Iron
Ridge Fault and neighboring fault to the west were adjusted to a permeability of 10'15 m2. In
parallel to fault permeability adjustments, the gradient conditions to the north were observed, and
the authors (COH97) found that a permeability of 10'15 m2 for the entire high gradient area
enabled the lower heads immediately downgradient from this area to remain within observed
bounds.
Water-level matches for boreholes are generally within 1 m, except for those boreholes that are
situated immediately adjacent to the Solitario, Iron Ridge, and High Gradient areas. The
measured heads are generally taken from Ervin et al. (ERV94). The large discrepancy between
measured and simulated heads at borehole H-5 is due to the anomalously high head at H-5,
which may be due to an intersection of a Solitario Canyon fault splay with this borehole
(ERV94). The model does not consider this splay. The simulated gradient in the northwestern
corner of the model is 0.12, compared to the measured gradient of 0.11 (ERV94). A head
residual for borehole G-l results because the low permeability zone within the high gradient area
was not extended to this borehole. The authors (COH97) believe that a short extension of this
zone will yield a closer match. Gradients calculated from actual field measurements is the low-
gradient area range between 0.0003 and 0.0004 (ERV94). The simulated heads in the low-
gradient area yield a gradient that ranges between approximately 0.0005 immediately
downgradient from the repository, to as much as 0.003 in the area west of C-wells. The Midway
and Paintbrush Canyon faults are located in this area, and their displacement results in the
complete displacement of the Bullfrog Tuff, the most permeable unit in the model. This
discontinuity creates this localized higher gradient. The authors (COH97) believe that such a
gradient is possible, as it may occur between boreholes where measurements are not available.
The 2-D and 3-D simulations described later in this section show how the Bullfrog abutment at
these faults significantly alters the flow field, and reveal the possible fault effects in greater
detail.
Using Hvdrorhftmical Daf* for Calibration. The authors (COH97) reviewed the possibility of
using hydrochemical data to facilitate model calibration and concluded that the ground-water
composition in the region around Yucca Mountain is heterogeneous at the kilometer scale. They
also noted that separate domains define the ground-water composition of Jackass Flats, Fortymile
Wash, Crater Flat, and Yucca Mountain, but at the sub-site scale, it is difficult to distinguish
VI-81
-------
separate domains. All water samples from Yucca Mountain itself plot in a relatively coherent
region.
Another of their observations relates to the availability of data. Dozens of chemical analyses that
include rare earth and trace elements are now available from springs that sample discharge waters
(HOD96). These provide useful end components (although not end member compositions) to the
history of ground waters. Similar analyses would be very useful for recharge waters and for
wells in the Yucca Mountain area. Without additional samples at greater depths in the saturated
zone, the authors (COH97) believe that this assumption can only be tested on the basis of mixing
models using existing isotope data.
The authors (COH97) continued their discussion by outlining the data limitations with respect to
the ground-water chemistry and maintained that uncertainties regarding the saturated zone
hydrochemistry abound due to the limited number of boreholes that penetrate the water table near
Yucca Mountain and the uncertain quality of some of the existing data.
Predictive Simulations
Regional Flow Model
The predictive simulations of the regional model were all focused on determining the adequacy
of the calibration and the appropriateness of the conceptual model(s). Calibration of the regional
model using the techniques available in MODFLOWP allowed for estimation of a series of
parameters that provide a best fit to observed hydraulic heads and flows. Numerous conceptual
models were evaluated during calibration to test the validity of various interpretations about the
flow system. Conceptual model evaluations focused on testing hypotheses concerning the (1)
location and type of flow system boundaries; (2) extent and location of recharge areas; and (3)
configuration of hydrogeologic framework features. For each hypothesis tested, a new set of
parameters was estimated using MODFLOWP and the resulting new simulated heads and flows
were compared to observed values. Only those conceptual model changes contributing to a
significant improvement in model fit, as indicated by a reduction in the sum of squared errors,
were retained in the final optimized model.
VI-82
-------
The final model was evaluated to assess the likely accuracy of simulated results. This was
accomplished by comparing measured and expected quantities with simulated values. The
quantities included in these comparisons are (1) hydraulic heads and spring flows, which were
matched by regression; (2) hydraulic conductivities, which were represented by parameters that
were estimated in the regression; and (3) water budgets. Unweighted and weighted residuals for
hydraulic heads show a very good model match with observed conditions in flat hydraulic
gradient areas and a relatively good match in large hydraulic gradient areas. Weighted and
unweighted residuals for spring flows show somewhat of a bias in that simulated spring flows are
generally lower than observed. The difficulty in simulating these spring flows in previous
models of this area without imposing discharge by using a specified flux, however, suggests that
even the somewhat lower simulated discharges are an improved match with observed conditions.
Estimated parameters were evaluated to determine if reasonable values of hydraulic conductivity,
vertical anisotropy and recharge rates were obtained. All estimated parameter values are within
expected ranges. The MODFLOWP-calculated linear confidence intervals also were well within
the range of expected values. Water budgets were evaluated to determine if they were within the
range of expected values. The authors (COH97) believe that even with the limited understanding
effluxes in and out of the regional ground-water flow system, overall budgets are within the
expected ranges for the flow system.
Site Scale Flow Model
In a fashion similar to the regional scale model, the primary objective of the site-scale modeling
was also to achieve a calibrated model. Therefore, the predictive simulations are essentially an
evaluation of how well the model is calibrated. The authors (ZYV97) used several criteria
including how well the simulated hydraulic heads matched those observed in monitoring wells;
and how closely estimates of flux from the site model match those determined in the regional
modeling.
In general, the model fits the observation well data relatively closely in small gradient areas, but
fits more poorly in larger gradient areas. With respect to the model fluxes, a comparison of the
regional versus site model predictions indicates fairly large discrepancies. The authors (ibid.)
believe, however, that it is more of a problem with the regional model conceptualization than
with the site model.
VI-83
-------
Another flow issue addressed in the site-scale modeling is the effect of temperature on the flow
field and on the prediction of flux through the site-scale model. An analysis performed with the
site-scale model shows that the model calibration is not significantly affected by the selection of
20°C for model calibration purposes. Computed heads differed by less than 1% between the
system simulated assuming 40°C temperature and the system simulated assuming 20°C. The
difference in computed flux out of the south end of the model was somewhat larger but still well
within the range of uncertainty of the flux in the actual ground-water system. Therefore, the
authors (ibid.) conclude that the calibrated model at 20°C is acceptable for the present study but
recommend that, in future modeling, a more appropriate mean temperature be used to alleviate
this concern.
Sub-Site-Scale Flow Model
A number of different types of simulations were performed at the sub-site-scale; including both
two- and three-dimensional analyses, simulating matrix diffusion effects, and attempts to match
well pumping tests results at the C-hole complex with model predictions as discussed below.
Two-Dimensional Simulations. Two-dimensional cross-sections of the full model are used by
the authors (COH97) to study specific scenarios such as the coupled effects of up welling,
varying fault and hydrostratigraphic unit properties, and fracture-matrix interactions. Although
the 2-D cross sections are a larger abstraction from the actual system then the full 3-D model, the
results of 2-D simulations are often easier to understand. In addition, the faster execution time for
2-D simulations enables one to quickly investigate the effects of varying specific properties and
thereby guide the design of full 3-D simulations.
Simulations. The full 3-D saturated zone model is used by the authors
(COH97) as an investigative tool to explore and define saturated zone processes by way of
hypothesis testing and calibration against hydrologic, thermal, and geochemical data. This
approach yields answers to questions concerning the plausibility of various flow and transport
scenarios. In addition, unanticipated saturated zone flow processes can be demonstrated and
investigated. Three-dimensional simulation of hydraulic tests is also used for model calibration
by way of matching pumping test results.
VI-84
-------
Single and Dual Continuum Models. Although flow in the saturated zone may be dominated by
fractures, the interaction of fracture and matrix must be captured in order to understand dilution
and mixing. To this end, the authors (COH97) developed two saturated zone grids: (1) single
continuum: and (2) dual-porosity. Most of the simulations utilized the standard single continuum
model, which represents an effective porous medium as a fracture-dominated system. The single
continuum model is used for calibrating the model against measured water table elevations, and
for investigation of steady-state flow geometry and flow visualization. The dual-porosity model
is used to consider fracture-matrix interaction. The dual-continuum model enables simulation of
pumping and tracer tests in dual-porosity media. Matrix properties are derived from unsaturated
zone modeling analysis and fracture data.
Simulation of T.nnp-Term Pumping at the C-Hole Complex. Although more than 150 individual
aquifer tests have been conducted at 13 boreholes in the vicinity of Yucca Mountain, the C-hole
complex is the only multi-hole complex in the saturated zone at Yucca Mountain. The C-wells
were drilled to perform multi-well aquifer tests and tracer tests. The authors (COH97) used the
sub-site-scale model to simulate a long-term pumping test at the C-wells in order to use C-hole
data to calibrate and validate the model. The full three-dimensional model was used to study the
drawdown behavior at the six observation wells for the long-term pumping test.
The model uses the regional hydraulic gradient observed for the Yucca Mountain site as an initial
condition and the pumping test is then an additional stress on the system. Boundary conditions
are constant pressure corresponding to the observed heads (ERV94). Pumping for a total period
of 250 days is simulated and the final steady-state drawdowns are reported for the six
observation wells.
The matching of the simulated drawdowns to the observed drawdowns was done by a manual
trial and error process. A constant withdrawal rate of 9.461/s (150 gpm) was assigned to one
element (Lower Bullfrog unit at UE-25 c#3). A fine discretization in the vicinity of the C-hole
complex was judged by the authors (COH97) to provide sufficient resolution of the distance
between the pumping well c#3 and the two neighboring observation wells, UE-25 c#l and UE-25
c#2. However, the authors (ibid.) believe that a still finer discretization may be needed to fine-
tune the performance of the simulation results to the observed transients in the two wells, UE-25
c#2 and UE-25 c#3.
VI-85
-------
These simulations were intended to show what effects the faulted structure has on pumping test
results. This model capability is especially useful since the pumping test results at the C-hole are
a function of heterogeneities at the scale of tens to thousands of meters, and it is at this scale that
radionuclides may be dispersed within the immediate site area. Since the model has been
calibrated by matching results of long-term pumping tests at the C-wells, simulations of pumping
tests under different conditions and at different locations can show what data would be obtained,
as well as what hydrogeologic signatures could be subject to analysis. Proper test designs for the
proposed Second Testing Complex, for example, can also be determined by simulating different
pumping test scenarios around faults using the model.
VI.2.1.2 Saturated Zone Transport Model Construction
The saturated-zone flow modeling performed for the Yucca Mountain Project has focused on the
key controlling factors influencing the measured head distributions at the site. In this section,
issues related to the transport of radionuclides in the saturated zone are presented, which in
addition to flow issues, requires that processes specific to the migration of solutes be considered.
The saturated zone radionuclide transport model was constructed by Zyvoloski et. al., (ZYV97),
and although the transport modeling was not used directly in the TSPA-VA, the findings of the
modeling were used to guide the approach that was taken to address radionuclide transport in the
TSPA-VA as discussed in later in this section.
Conceptual Model of Saturated Zone Transport
The site-scale ground-water flow model provides the hydrologic framework for determining the
direction and rate of movement of radionuclides that reach the saturated zone beneath Yucca
Mountain. In addition to flow issues, the migration of radionuclides to the accessible
environment depends on transport processes and parameters distinct from the flow model itself.
This section presents the authors (ZYV97) conceptual model for transport that includes advective
transport of radionuclides, dispersion, diffusion of radionuclides from fractures into the rock
matrix and sorption. Within the fractured tuffs, the authors (ibid,) expect the migration of
radionuclides to be primarily through regions with higher bulk fracture permeability. They also
believe that flow within individual joints probably occurs through channels, rather than as sheet
flow through parallel-plate fractures. The authors (ibid.) note that at more distant downstream
VI-86
-------
locations, the migration is likely to be through alluvium, and a model for flow and transport
through a porous continuum, rather than a fractured rock, is likely to apply.
Fluid Flow In Fractures
The hydrologic evidence to date strongly supports the model of fluid flow within fractures in the
moderately to densely welded tuffs of the saturated zone (e.g., WAD84, WHI85). First, as
expected, the hydraulic conductivities measured for core samples in the laboratory are orders of
magnitude higher when the sample is fractured (PET84). Also, there is generally a positive
correlation between fractures identified using the acoustic televiewer of a borehole television
tool and the zones of high transmissivity.
Because the role of fractures is so important to the hydrology in the saturated zone, the
permeability distribution and principal flow directions depend strongly on the spatial distribution
and orientations of fractures. Karasaki et al.(KAR90), in an attempt to correlate the C-wells
fracture data with transmissivity measurements, tentatively concluded that the regions of high
transmissivity in the C-wells correlate with fractures oriented to the northeast, with the steeply
dipping fractures contributing most to the transmissivity.
Matrix Difflip^n - Laboratory and Field Evidence
Instead of simply traveling at the flow rate of the fluid hi the saturated zone, radionuclides will
potentially undergo physical and chemical interactions that must be characterized to predict
large-scale transport behavior. These interactions include molecular diffusion into the rock
matrix, sorption on the minerals along the fractures or within the rock matrix, or transport in
colloidal form. These phenomena are described below.
When a dissolved species travels with the fluid within a fracture, it may potentially migrate by
molecular diffusion into the stagnant fluid in the rock matrix. When a molecule enters the
matrix, its velocity effectively goes to zero until the Brownian motion carries it back into the
fracture or into an adjacent fracture. The result is a delay of the arrival of the solute at a
downgradient location from what would be predicted if the solute had remained in the fracture.
In hydrologic tests not involving tracers, the pore water velocity can often be estimated given
assumptions about the fracture porosity. For interpreting hydrologic tests, the fracture porosity is
VI-87
-------
usually the correct porosity value because aquifer properties (i.e., transmissivity) from pump
testing is controlled by fracture flow. However, the ground-water travel time is often computed
by dividing the flow path length by this velocity. This estimate is potentially a severe
underestimate of the time required for a water molecule to migrate along the flow path, A more
accurate definition of travel time for the purpose of predicting transport behavior would take into
account matrix diffusion.
There have been several theoretical, laboratory, and field studies performed to demonstrate the
validity of the matrix-diffusion model. Grisak and Pickens (GRI80) and Neretnicks (NER80)
first applied mathematical models to demonstrate the likely effect of matrix diffusion in flow in
fractured media. In these studies, transport was idealized as plug flow within the fracture with
diffusion into the surrounding rock matrix. Sudicky et al. (SUD85) applied a similar model to a
laboratory experimental apparatus which tracer was injected into a thin sand layer with
surrounding low permeability silt layers, showing that matrix diffusion was necessary to model
the conservative tracer data. Neretnicks et al. (NER82) reached the same conclusion in their
experiments of transport in natural fissures in granite. Rasmuson and Neretnicks (RAS86)
extended the concept of matrix diffusion to examine the coupling between matrix diffusion and
channel flow usually thought to occur within natural fractures.
Transport models incorporating matrix-diffusion concepts have also been proposed to explain the
often conflicting ground-water ages obtained from 14C data compared to ages predicted from
flow data. Sudicky and Frind (SUD81) developed a model of flow in an aquifer with diffusion
into a surrounding aquitard to show that the movement of 14C can be much slower than predicted
if only movement with the flowing water is considered. Maloszewski and Zuber (MAL85,91)
reach a similar conclusion with a model for 14C transport that consists of uniform flow through a
network of equally spaced fractures with diffusion into the surrounding rock matrix between the
joints. Their model also includes the effect of chemical exchange reactions in the matrix, which
further slows the migration velocity. Maloszewski and Zuber (MAL85) also present analyses of
several interwell tracer experiments that show that their matrix-diffusion model can be used to
provide simulations of these tests that are consistent with the values of matrix porosity obtained
in the laboratory and aperture values estimated from hydraulic tests. The results are, in all cases,
superior to previous analyses that did not include matrix diffusion effects. Finally, of greatest
relevance to the saturated zone beneath Yucca Mountain is the C-wells reactive tracer test
(REI97), which demonstrates that models incorporating matrix diffusion provide more
VI-J
-------
reasonable fits to the tracer-experiment data than those that assume a single continuum. They
showed that a suite of tracers with different transport characteristics (diffusion coefficient,
sorption coefficient) produced breakthrough curves that can be explained using a diffusion model
that assumes diffusion of tracers into stagnant or near-stagnant water. Finally, Waddell
(WAD97) recently reported a similar result for nonsorbing tracers with different diffusion
coefficients in a fractured-tuff tracer experiment at the NTS.
Thus, the theory of matrix diffusion is generally thought to be based on sound physical
principles, and demonstrations of its effect have been shown in both laboratory-sized specimens
and interwell tracer tests. The effect on transport under ground-water flow conditions could be
extremely large and, thus, should be incorporated into any realistic radionuclide migration
model. It is still unclear, however, how DOE will ultimately incorporate these processes into the
TSPA-LA since they are currently not considered in the TSPA-VA.
Dispersion. Dispersion is caused by heterogeneities at all scales from the pore scale to the scale
of the thickness of individual strata and the length of structural features such as faults. The
resulting spreading of radionuclides is important to performance in that it will lead to dilution
and should be captured in transport models. As will be discussed later in this section, the TSPA-
VA simply assigns a dilution factor, to account for dispersion, rather than modeling dispersion
explicitly.
Only the largest heterogeneities are represented as property zone values in the site-scale model;
all dispersion caused by smaller-scale features must be represented through the use of dispersion
values input into the model. Numerous ground-water transport studies have been conducted at a
variety of scales, and the results are compiled using the dispersivity as the correlating parameter.
It is well-known that dispersivity increases with the scale, or distance, for transport of a solute
(NEU90). The only site-specific data comes from the C-wells reactive tracer experiment of
Reimus and Turin (REI97) and accompanying tracer tests carried out by the USGS at the C-
wells. These experiments yielded estimated dispersivity values that fall in the range of
uncertainty of correlations to data collected and compiled at many sites. The authors (ZYV97)
believe that this result provides credibility that the dispersivity values used in the field
simulations are appropriate. Very little experimental information exists for assigning transverse
dispersivity. In the site-scale modeling study, however, the "rule-of-thumb" value of one-tenth
the longitudinal dispersivity is used for transverse dispersivity.
VI-89
-------
Other Transport Processes. Sorption of radionuclides on rock surfaces is another mechanism that
will result in retardation. These radionuclide-rock interactions can potentially occur on the
surfaces of fractures and within the rock matrix. This distinction is important because the
surface-area to fluid-volume ratio and the mineral distributions are probably different in the
fractures as compared to the matrix. The lithium tracer in the C-wells reactive tracer experiment
was modeled by Reimus and Turin (REI97) using a matrix-diffusion model with the sorption
coefficient as an additional adjustable parameter. The matrix sorption coefficient that fit the data
agreed quite well with the value determined hi laboratory sorption tests, thus providing an
additional degree of confidence in the matrix-diffusion model. The fact that the early lithium
response had the same timing as that of the nonsorbing tracers, but with a lower normalized peak
concentration, is consistent with matrix-diffusion coupled with sorption in the matrix.
Transport of radionuclides on colloidal particles or in colloidal form is a third mechanism that
may apply at the field scale. Colloidal particles have the potential to provide a direct pathway
through fractures. Strongly sorbing radionuclides such as plutonium may adhere to these
particles and move more rapidly than if the radionuclide were confined to the aqueous phase.
The size of the colloids may minimize diffusion into the rock matrix, thereby reducing one
possible retardation effect. On the other hand, filtration of colloidal material is likely to come
into play, thereby potentially resulting in large retardation factors. The key uncertainties
identified by the authors (ZYV97) in predicting colloid transport in the field are:
• whether a continuous pathway exists with large enough pore size to facilitate
transport over large distances without filtration or migration into stagnant fluid
storage; and
• the uncertainty regarding the relative amount of radionuclide on colloid surfaces
versus that in the aqueous phase.
For the first uncertainty, field demonstration of colloidal transport is necessary to prove that this
mechanism is important at the field scale. Furthermore, given the complexities of the
interactions of colloids with the surrounding rock and geochemical conditions, further laboratory
work to characterize these effects is necessary. Other key uncertainties not discussed by the
authors (ZYV97) include the relative stability of the colloid in various geochemical
environments and potential chemical and physical (i.e., filtration) retardation of the colloid itself.
VI-90
-------
Summary of Conceptual Model
Summarizing the discussion of fluid flow behavior given above, the saturated-zone transport
conceptual model is outlined below:
• Radionuclides enter the saturated zone via the fluid percolating through the
unsaturated rock above the water table. The exact nature of this transport is not
expected to exert a great effect on the subsequent saturated-zone model, and thus, the
saturated-zone transport model can be developed independently from a transport
model for the vadose zone, which acts as a boundary condition for the saturated-zone
model. Of course, for detailed predictions of radionuclide migration, the spatial and
temporal distributions of the input from the unsaturated zone are important.
• Flow occurs within the highly fractured portions of the tuffs near the water table.
There is probably not a continuous zone of high permeability to the accessible
environment. Assuming there is not a continuous zone, then the low-permeability
regions will effectively act as large-scale heterogeneities that give rise to large-scale
macroscopic dispersion due to the tortuous nature of flow over the scale of hundreds
of meters to kilometers.
• Although the vertical matrix permeability is assumed to be small over the length scale
of several hundreds of meters, within a fractured region, vertical permeabilities
should be as large as the horizontal permeabilities. Thus, the radionuclide will spread
vertically and be present within the entire thickness of a fractured zone. The
thicknesses of these fractured zones are difficult to estimate from the present data.
However, the extent of fracturing correlates reasonably well with the degree of
welding, which is one of the criteria used to define the submembers within a
lithologic unit. Therefore, it seems reasonable to assume that the heights of the
fracture zones are on the order of the thicknesses of the individual lithologic
members, namely 100 to 200 m. This possibility is in contrast to flow zones detected
in individual boreholes, where measurements reflect the intersection of specific
fractures with the well.
• Fluid flow occurs within the fracture with stagnant fluid residing in the rock matrix
blocks. All fractures have large contact areas due to the in situ stresses exerted on
them at these depths. The conductivity of an individual fracture is probably not a
strong function of its orientation because all are on the-flat portion of the aperture
versus effective-pressure curves. Therefore, the magnitude and direction of the
components of the hydraulic conductivity tensor should be controlled by the
distribution of joints of various orientations. Fractures detected from geophysical
logs are generally oriented in a north-south direction and are, within 30° of vertical.
VI-91
-------
Flow within individual joints probably occurs within channels. The fluid travels
preferentially within regions of large apertures with large sections of the fracture
surface containing stagnant fluid or no fluid where the faces are in contact.
The surrounding matrix material conducts no fluid under natural ground-water flow
conditions but is physically connected to the fracture fluid through the pore network.
Fluid is stored in this pore space and is important to radionuclide migration (see
below). The matrix porosities of interest are those of the rock within the fractured
regions. Fractures are generally found within the moderately to densely welded tuffs,
so the range of matrix porosities of these tuffs (0.06 to 0.09 for densely welded and
0.11 to 0.28 for moderately welded) more accurately reflect the fluid storage of
interest rather than the generally wider ranges of values found within a specific
lithologic member.
Computer Code Selection
The FEHMN code was selected for the transport simulations for many of the same reasons it was
chosen for the site-scale flow modeling discussed above. Using the same code for both ground-
water flow and contaminant transport modeling also eliminates model setup inconsistencies that
may otherwise arise (e.g., block centered vs. node centered grids).
Layering and Gridding
Grid Generation and Grid Resolution Studies
Several three-dimensional grids were generated in support of the site-scale flow and transport
modeling effort. Both structured and unstructured grids were developed, tested, and used for
sensitivity analyses and flow calibration studies. Structured grids, commonly referred to as
finite-difference grids, are easy to generate but their blocklike structure makes it difficult to
represent complicated geometries with all but the finest grid resolution. Unstructured grids, like
the common finite-element meshes, are more complicated and may consist of triangles and
tetrahedrals. They can, however, represent complicated hydrostratigraphy and topography
accurately at fairly low resolution. It should be noted that not all finite-element models are
formulated to accommodate unstructured grids, and recent developments now allow finite
differences to solve unstructured grids.
VI-92
-------
As was discussed in section VI.2.1.1, the hydrostratigraphy represented in the geologic model
consists of multiple layers of contrasting fluid flow and transport properties. The grid-generation
methods used in the transport study allow the stratigraphy to be honored in numerical grids of
different resolution so that comparison studies can be performed to test for grid quality and to
determine the resolution required for flow and transport simulations. All mesh generation and
manipulation is done with the GEOMESH/X3D toolkit (GAB95, GAB96; TRE96).
A series of six structured grids of increasing resolution was used to compare the flow through the
model domain to assess the point at which increasing resolution no longer influences the results.
This process identifies the resolution required for an accurate simulation of the flow field. A
similar study for two unstructured grids demonstrated that the coarser grid used for calibrating
the flow model was sufficiently resolved for accuracy. This section also reports on the
development of a technique for selectively refining the numerical grid for transport calculations.
The method uses the solute transport pathway determined on a coarse grid to identify regions of
the model where increased grid resolution is. required. The mesh-generation software refines the
grid in those areas. After two or more successive applications of this process, the grid is finely
resolved along the pathways of solute plume movement, and numerical error associated with
insufficient grid resolution is minimized. The main application of this technique is for refining
the revised site-scale model for transport calculations.
Approach This subsection addresses the construction of computational grids that reflect complex
geologic structure for saturated-zone flow and transport simulations. The importance of grids are
summarized by the authors (ZYV97) with the following questions:
• What resolution is required to represent geology?
• What resolution is required to represent flow and transport processes?
• Can a grid be optimized to represent features and flow processes while keeping the
size relatively small?
• If a flow model is calibrated on a coarse grid, is it calibrated on a fine grid?
The authors (ZYV97) point out that these grids must accurately represent the geologic structure
and be appropriate for numerically accurate simulations of flow and transport. That is, any error
associated with numerical grid discretization must be constrained within specified truncation
VI-93
-------
error tolerances. Understanding the quality of the grids and the effects of grid-related error is
necessary for assessing the quality of the simulations and generating a defensible model.
The process of developing flow and transport models for the saturated-zone studies are divided
by the authors (ZYV97) into three parts:
• Developing accurate conceptual models of the geology and hydrologic material
properties.
• Building the grid and prescribing boundary and initial conditions.
• Applying the computational physics models of flow, heat transport, and chemical
transport.
They provide further explanation by explaining that geologic interpretation, stratigraphic model
development, and material characterization are performed based on numerous field
measurements. The stratigraphic model populated with hydrologic material properties then
provides the basis for computational grid development. They also point out that the ability of the
numerical grid to represent the geologic complexity directly affects accuracy of the numerical
model's approximation of the actual physical system's response.
GEOMESH/X3D (GAB95, GAB96; TRE96) is described by the authors (ZYV97) as a mesh-
generation toolkit that can use the hydrostratigraphic model as input to create a numerical grid
that represents accurately complex structures and stratigraphy such as faults, pinchouts, and layer
truncations. As well as representing geologic structure, this software maintains and distributes
physical and chemical attributes such as porosity, permeability, or percent zeolite. The process
of populating the numerical grid with hydrologic and transport properties is described in
Robinson et al., (ROB97). This process of generating and populating numerical grids from
geologic framework models is automated, thus making the entire process easy to implement with
fewer user-induced errors. At any step in the process, numerical resolution can be added to a
particular subregion, new boundary conditions can be prescribed, or new attributes can be
incorporated.
The authors (ZYV97) believe that the grid quality control issues involved in this automated
process cannot be over-emphasized. They indicate that a system that uses only electronic
processing of the hydrogeologic data is desirable for a defensible model during the licensing
process because the interfaces can be qualified and metrics of goodness of representation can be
VI-94
-------
established. These metrics would likely include a volume comparison between the project
database hydrogeologic units and the model hydrogeologic units.
A primary objective of their study is to analyze how accurately grids of increasing resolution
capture this complex structure and simulate flow. At what resolution is the geometry of the
material distribution accurately represented with a structured grid in order to model fluid flow
through the aquifer? Because higher grid resolution is more computationally expensive, this
information can be used to determine what the desired accuracy of the result is and at what
resolution of grid that accuracy would be attained.
Saturated-Zone Unstructured Mesh from Stratamodel Geological Model. Software has been
developed to automate the steps required to create a finite-element mesh for flow and transport
calculations from a geological framework model.
The steps required to create a grid are:
• Convert Stratamodel Stratigraphic Framework Model (SFM) provided by USGS to
hexahedral finite-element grid.
• Remove zero volume elements from data structure.
• Remove elements with vertical height less than one meter.
• Convert hexahedral elements to tetrahedral elements.
• Add the points defining the potentiometric surface (surface provided by the USGS).
• Add nodes with specified xyz location for measured well-head calibration.
• Add nodes above and below well-head calibration nodes to enhance grid quality.
• Add buffer zones above potentiometric surface and below bottom boundary.
• Assign material values to all nodes.
• Optimize the grid connectivity to insure positive finite-element coefficients.
• Calculate Voronoi control volumes.
VI-95
-------
• Output node coordinates, node connectivity, nodal volumes, and material and
boundary lists.
All file conversion, gridding, node distribution, property interpolation, quality checking, and
output is carried out within a single integrated gridding tool kit. The authors (ZYV97) believe
that this approach insures that the steps taken are reproducible and traceable and modifications to
any step in the gridding process can be done without a great deal of labor.
Grid Generation and Pronertv Interpolation. Generation of grids and interpolation of properties
from the hydrostratigraphic model onto the numerical grids is performed with GEOMESH/X3D.
GEOMESH/X3D is also used for comparing results from different grids, each having different
resolution or grid structure. Interpolation with GEOMESH/X3D allows for the superposition on
mesh A of the attributes (material number, pressure, saturation, concentration, etc.) belonging to
mesh B. This utility is useful when refining a mesh or when comparing meshes of different
resolution. The meshes should generally be of the same volume but do not have to coincide
exactly. They do not need to have the same discretization, resolution, or element type. Either
mesh can be structured or unstructured. For interpolation, a node of mesh A is first located
within an element of mesh B. The attributes of the nodes defining the element in mesh B are
then linearly interpolated to the node in mesh A. Output of mesh A from GEOMESH then
contains the new attribute values associated with each node. If desired, the mesh-A node is also
assigned the material number of the mesh-B element. Elements of mesh A can then be assigned
material numbers based on the nodal material numbers that were interpolated from mesh B.
The saturated-zone model domain selected for their grid resolution study is defined by Nevada
State Coordinates of 533,340 m to 563,340 m in an east-west direction, 4,046,782 m to
4,091,782 m in the north-south direction, and -754 to 1332 m vertically above sea level.
Nineteen materials are assumed located within this volume.
Although a calibrated model must have appropriate heads at each inlet and outflow boundary, to
test the resolution of the various grids used, the actual pressure differential between the north and
south is not important. For simplicity, a pressure difference of 15 MPa is used for all simulations
in this grid resolution study. The isothermal site-scale flow model is run to steady state and the
flow out of the model is compared for each grid. The problem was designed to study the errors
made only in the representation of the hydrostratigraphy. This was achieved by specifying
pressures on the north and south faces of the grid. Thus, the solution is a linear variation in
VI-96
-------
pressure with a corresponding constant flux. Tests with homogeneous meshes confirmed this
result. The differences in reported fluxes are a result only of representation errors of the
hydrogeologic units. The outlet flux was chosen as the parameter to compare because this
parameter is likely to be a primary factor in assessing radionuclide transport to the accessible
environment. This flow was obtained by slimming the outflow at the south boundary of the
model. The flow volumes exhibit asymptotic behavior at higher resolution. Because higher grid
resolution is'more expensive, this type of analysis is used to determine what the desired
accuracy is and at what resolution that accuracy would be attained.
The results of the flux changes with grid resolution can be more easily understood hi terms of
average grid spacing. The range of grid spacings in the vertical direction is from 696 m to 70 m.
The results of the grid resolution study suggest that for saturated-zone flow computations, 100-m
vertical grid resolution may be sufficient if structured or finite-difference gridding is used. This
resolution is a finer grid than is commonly used in saturated-zone simulations of Yucca
Mountain. In the subsection below runs with unstructured-grid representations are presented to
determine the resolution necessary for unstructured grids. The authors (ZYV97) anticipate that
the restrictions of grid resolution will be somewhat relaxed if the stratigraphy is captured directly
hi the unstructured numerical grid. However, note also that this grid resolution study was
conducted using total outflow water flux. If a criterion based on radionuclide transport was used,
the result would most likely imply the need for higher grid resolution due to the nature of
simulating transport using the advection-dispersion equation. The authors (ZYV97) believe that
selective grid refinement hi the region near the repository and close to the water table in the
pathways to the accessible environment will be required for radionuclide transport.
Unstructured Ctrid Studies. As stated in above, it is anticipated that the unstructured grid would
produce closer flux agreement between grids of different resolution than the structured study. To
investigate this hypothesis, the authors (ZYV97) generated a finer-resolution model by refining
the stratigraphic framework model (SFM). This model has approximately 2X resolution in the x,
y, and z directions. The volume of each stratigraphic unit is identical to the lower-resolution
model. The boundary conditions used are the same as in the structured grid studies- that is, a 15
MPa difference between the north and south faces with no flow on the top, bottom, east, and
west. Again, the flux flowing out of the south face is compared. The results were 3290 kg/s for
the lower-resolution model (9279 nodes) and 3760 kg/s for the higher-resolution model (49,895
nodes). The flow rates differ by about twelve percent. The authors (ZYV97) believe that the
VI-97
-------
results indicate that the flow calibration with the lower-resolution model is sufficient but some
caution is warranted. The authors (ZYV97) also indicate that time constraints precluded
additional resolution studies, and conclude that these studies point to a need to systematically
study grid refinement and its relation to both flow and transport.
Grid Refinement Around Plumes. To adequately model contaminant transport through porous
media, a computational grid will require refinement beyond that required for flow modeling.
This requirement is true of the grid used for USGS/LANL site-scale saturated-zone flow
simulations of the regional aquifer beneath Yucca Mountain discussed in Section VI.2.1.1
(CZA97). The authors (ZYV97) of the transport modeling study developed an approach for grid
refinement that refines only that part of the grid where most contaminant migration occurs, thus
minimizing the number of nodes added to the grid to assure an accurate transport solution. The
refinement technique uses the results of a transport solution computed on a grid that is expected
to be too coarse for an accurate transport solution. The preliminary coarse grid is then refined in
the region where most contaminant migration occurs (e.g., within a specified isoconcentration
surface). The flow and transport solution is recalculated on the refined grid. This grid may then
be refined again within an isoconcentration surface that is generally of lower concentration than
used for the initial refinement. Refinement is complete when the breakthrough curves computed
at some compliance point no longer vary within a specified tolerance level. This approach is a
form of adaptive mesh refinement that uses the problem solution to guide the refinement process.
The technique, which uses the GEOMESH/X3D grid-generation system, was first tested on a
rectangular grid with a single porous medium. This case is discussed, as it is useful in explaining
the refinement methodology. The technique, tested below, was supposed to be employed in
FY98 to the 250-m sampled SFM currently being developed by the USGS. The EPA is unaware
of the current status of these activities.
For the single porous media test, a criterion is needed to determine when the grid is adequately
refined. The authors (ZYV97) chose to look for convergence in the breakthrough curve at a
downstream compliance point. The breakthrough monitoring location was chosen as the
uppermost, central node of the grid at the downstream boundary. The authors (ibid.) found that
the original constant-concentration source used for grid refinement results in different solute
mass-flux input rates with the different grids. Therefore, transport solutions were rerun with the
various grids using constant tracer flux so that each simulation would receive the same mass, and
the breakthrough curves could then be compared. As the grid is refined, breakthrough to the
VI-98
-------
monitoring point advances slightly in time and the steady-state breakthrough concentration
generally increases. This result occurs because the plume exhibits less numerical dispersion
around its centerline and with depth. There is little difference between the solutions for the grids
with 28,511 and 57,450 nodes indicating that the 28,511-node grid is adequately refined. It
should be noted that if values of dispersion or time steps are changed the grid may not be
sufficiently refined.
The authors (ibid.) indicate that future work will apply the refinement technique discussed above
to the USGS/Los Alamos multiple-material saturated-zone model. Refinement will be based on
the concentration of a steady plume generated with the calibrated flow field. Nodes added during
refinement are assigned material properties based on the original SFM discussed previously.
The original nodes remain in the grid and retain their material properties. Refinement of the
USGS/Los Alamos model was not pursued in this study. The authors (ibid.) indicate that the
coarse 1500-m spacing of the current SFM is inadequately refined to justify refinement for
transport simulations. They further note that an updated SFM with 250-m spacing will be
incorporated in future work. The authors (ibid.) also believe that this refined SFM has adequate
stratigraphic information to work ideally with the concentration adaptive mesh refinement
technique.
Initial and Boundary Conditions
The form of the flow boundary conditions employed in obtaining a calibrated model plays a large
role in the model results, especially with respect to subsequent transport predictions. A model
calibrated to measured head values is a nonunique solution that must be constrained with
measured or estimated flux values at boundaries. Simply put, if only head boundary conditions
are used, the total flux is directly proportional to the values of permeability in the model;
increasing or decreasing each permeability value by a fixed ratio changes the total flux by that
same fraction. At the site scale, the flux values into and out of the flow domain cannot be
directly tied to major recharge and discharge values because these occur outside the model
domain. Because this is the case, there is a need to tie the flux values into and out of model faces
to a larger-scale model. The authors (ZVY97) intend to further investigate the interplay between
the calibrated regional model and the site-scale model.
The original plan for the site-scale model was to use fluxes or head data from the regional model
as boundary conditions for the site-scale model. However, there are several problems in
VI-99
-------
applying fluxes directly. The first is that the site model has sixteen units represented within its
boundaries, whereas the regional-scale model represents only three units in this region. This
difference presents problems in assigning fluxes to nodes with widely varying permeabilities.
The second problem is that the calibrated heads for the regional model deviated the most from
well data near the northern boundary of the site-scale model. Thus, fluxes were apt to be most
inaccurate in that region. Because of these problems, the fixed head boundary conditions were
derived from measured head data. Because of the fixed head conditions, solutions are valid for
permeability ratios only. Fluxes must therefore be compared to those predicted by the regional-
scale model to insure consistency.
Because of the need for finer resolution (primarily in the z direction), the site-scale model was
developed with a smaller horizontal extent than previous site-scale models. The model
dimensions are approximately 30 km by 45 km. The compilation of all relevant data was done
by the USGS, and these data were then organized into a Stratigraphic Framework Model. The
SFM provided the basis for all grid generation. The grid building effort produced 15 grids for
testing. The grid used in the calibration was on a 1500 x 1500 m area! spacing that consisted of
5485 nodes, 29,760 elements, and 40,548 internode connections. The calibration of this model
is described in section VI.2.1.1. A good calibration on an initial modeling exercise was obtained,
but large flux values pointed to a systematic error in the geometric part of the flow terms. This
problem was subsequently corrected, and the flow results revealed more realistic fluxes. The
corrected model results were performed on a grid containing 9279 nodes, 52,461 elements, and
67,324 internode connections. The increase in the number of nodes was due to the inclusion of
model "buffer zones" that play no part in the solution but improve the definition of the
potentiometric surface. These zones are assigned values of zero for both porosity and
permeability and, therefore, do not enter into the fluid flow calculation. Only the layers
sandwiched between the upper and lower buffer zones are part of the flow model itself.
The source term boundary is treated as a mass-flux introduction to the saturated zone.
Model Parameterization
In section VI.2.1.1, a number of unwanted features in the geologic framework model of the
current site-scale model are described, that make it difficult to produce accurate transport results
in the vicinity of Yucca Mountain. These features do not significantly impact the calibration of
the flow model but make simulations problematic. In parallel to producing this calibrated site-
VI-100
-------
scale model, the USGS has been developing an updated geologic framework model with more
up-to-date geologic interpretation near Yucca Mountain. The authors (ZYV97) plan is to
develop grids and calibrated flow solutions on this revised model in FY98 and to use this revised
flow model as the platform for performing radionuclide transport calculation.
The main difference between the two framework models is that the discretization of the revised-
site SFM is finer (250 m) than the older-site SFM (1500 m). This difference leads to smoother
transitions between and more realistic representations of the hydrogeologic units. Additional
data from the ISM framework model (ZEL96) were incorporated into the 250-m framework
model. This approach gives better resolution to the volcanic units. The representation of the
upper volcanic confining unit (UVCU, Unit 14) is of importance due to its predominance in the
control of the flow beneath the potential repository site. The UVCU appears primarily in the
northern half of the 250-m framework model, with relatively isolated bodies in the southern part
of the modeled area. The UVCU does not appear as a large body hi the central part of the model,
as it does in the 1500-m framework model. It appears that the configuration of the UVCU in the
250-m framework model may produce the steep hydraulic gradients in the north, where it exists,
and have a less dominant control on the flow field in the southern and central parts, where it does
not exist.
With regard to transport, the authors (ZYV97) note that a revised flow model based on this
geologic interpretation will correct the problems in the current version and allow realistic
simulations from the potential repository to hypothetical accessible environment locations as far
as 25 km from the site.
Model Calibration
The transport model used the flow field of the calibrated sub-site model to predict contaminant
transport migration directions and velocities. Since there is no contaminant field data (e.g.,
existing contaminant concentrations in monitoring wells) on which to perform a transport
calibration the authors (ZYV97) did not conduct any further model calibration.
Predictive Simulations
VI-101
-------
Several models and numerical techniques are used to obtain results for radionuclide transport in
the saturated zone. First, a technique based on numerical convolution is developed to link the
unsaturated-zone breakthrough curves at the water table to the saturated-zone transport system.
In this method, the inputs are the mass flux of radionuclides reaching the water table versus time,
along with a generic breakthrough curve in the saturated zone, computed as the response at a
downstream location to a constant injection of radionuclides at the footprint of the proposed
repository. The numerical implementation of the method was verified by performing a full
calculation using the actual time-varying input from an unsaturated-zone calculation as input to
the saturated-zone model. The method allows a variety of input flux curves to be computed
quickly without recomputing the saturated-zone calculation each time. Assumptions inherent to
the convolution technique include steady-state flow and linear transport processes, which, for
sorption, implies that the linear-sorption isotherm model must be used. The authors (ZVY97)
believe that the linear-J^ sorption model is not overly restrictive because one can select a Kd
value that conservatively bounds the sorption behavior predicted by a nonlinear-sorption
isotherm model.
Before proceeding to more complex site-specific models, a simplified three-dimensional flow
and transport model is presented to examine the importance of several transport parameters that
are difficult to investigate fully with the current site-scale models. Dispersivity is established as
one of the key uncertain parameters that influence the concentration (and hence dose) at
accessible environment compliance points. The transverse dispersivity is actually a more
sensitive parameter for dilution because it governs the degree of lateral spreading of the plume.
Matrix diffusion, established as a valid process in the field test of Reimus and Turin (REI97),
affects the breakthrough times at a downstream location but, to a first approximation, does not
influence the peak concentration unless radioactive decay is significant. Another important
factor regarding matrix diffusion is that it allows radionuclides to contact minerals in the rock
matrix that potentially sorb radionuclides. Even small amounts of sorption have a large effect on
breakthrough times and peak concentrations. Finally, this simplified model was used to illustrate
that the saturated-zone transport system has the ability to dilute spikes of high concentration and
short duration that come from bypassing of the unsaturated zone through fractures. Thus, the
authors (ZYV97) believe that the saturated zone provides an important component in a "defense-
in-depth" strategy in which uncertainties leading to poor performance in one part of the
repository system are mitigated by the performance of another radionuclide transport barrier.
VI-102
-------
To simulate radionuclide transport from the footprint of the repository to a 5-km compliance
point, the sub-site-scale model was used. This model was chosen as an appropriate substitute to
performing these calculations using the site-scale model because of the more accurate
representation of the geology near Yucca Mountain. When the site-scale model is revised to use
the new hydrostratigraphic data based on a 250 x 250-m geologic grid, all calculations will be
performed with the site-scale model itself. The sub-site-scale flow model captures the large
hydraulic gradient and flow through the geologic strata of relevance downstream of the
repository footprint, including the Prow Pass, Bullfrog, and Tram Tuffs units. Radionuclides
travel to the east and south from the footprint to a 5-km compliance point. Releases into the
Prow Pass unit travel in a more easterly direction than releases in the other units; releases into the
Tram unit at the south end of the footprint travel almost due south. All releases follow the
dipping stratigraphy, indicating downward movement of radionuclides. This effect may be
important if upcoming field studies reveal more reducing conditions with depth, because sorption
coefficients of "Tc and 237Np are likely to be much higher and solubilities much lower under
reducing conditions.
Transport times to a hypothetical 5-km compliance point under conditions in which the effective
porosity is the matrix porosity are on the order of a few thousand years; much shorter transport
times result from an assumption of less matrix diffusion. However, the extent of matrix diffusion
itself only influences the arrival time rather than the concentration at the downstream location.
This effect is in contrast to differences in fluid flux that may occur due to future wetter climates,
which result in earlier travel times but also lower concentrations (greater dilution).
Sorption of radionuclides such as ^'Np onto zeolitic tuffs in the saturated zone also leads to
significant retardation and longer travel times to a 5-km compliance point.
The results of the sub-site-scale model are then used as input to transport calculations to a
hypothetical 20-km compliance point in the site-scale model. Simulations assuming both
fracture-like and matrix-like effective porosities in the fractured tuff were performed to
investigate the importance of this parameter. In both simulations, the alluvium present along the
transport pathways were assigned high porosity. In this set of calculations, the nature of
transport in the fractured tuffs is less important as the system becomes increasingly dominated by
long predicted travel times in the alluvium. Thus, even if the tuffs are presumed to have a low
effective porosity, travel times of 10,000 years or more are predicted in the alluvium alone.
Furthermore, the results show that even small amounts of sorption in the alluvium shift travel
VI-103
-------
times to values on the order of 50,000 years, and predicted concentrations at the compliance
point are lower than in the absence of sorption. Clearly, for more distant compliance points, the
flow and transport behavior of the alluvium becomes increasingly the controlling factor in
saturated-zone performance.
The influence of repository heat on saturated-zone flow and transport of radionuclides is also
studied using the subsite-scale model. Repository waste heat creates a zone of higher-than-
ambient temperature that extends vertically into the saturated zone and along the prevailing flow
pathway from the repository. However, the predicted impact on transport of "Tc to a 5-km
compliance point is very small. One outstanding issue related to repository heat is the possibility
of temporary or durable changes to the permeability and porosity due to temperature-dependent
rock-water interactions. If these effects turn out to be minor, then repository waste heat has
minimal influence on the migration of radionuclides through the saturated zone.
Integrated transport predictions are presented in which are linked to the unsaturated-zone
transport model of Robinson et al. (ROB97) with the subsite-scale model to predict the transport
of the key radionuclides "Tc, 237Np, and the isotopes of plutonium. In Robinson et al. (ROB97),
the authors investigated the performance of the unsaturated-zone system for different infiltration
rates that could result from changes to the present-day climate. These predictions of radionuclide
mass flux at the water table are input to convolution calculations to predict the combined
unsaturated/saturated-zone performance at 5 km. For the unsaturated-zone performance
predicted in Robinson et al. (ROB97), the integrated response in the saturated zone for 237Np and
"Tc is a direct consequence of dilution of percolating unsaturated-zone fluid with flowing
saturated-zone ground water. Therefore, poorer predicted unsaturated-zone performance under
wetter future-climate scenarios translates directly to higher predicted concentrations in the
saturated zone. Sorption onto zeolites in the saturated zone should provide a considerable delay
in arrival times for 237Np. For plutonium, rather than climate change or sorption to the host rock,
the key factor influencing concentrations is the propensity of plutonium to sorb to mobile
colloids.
Regarding the nature of flow and transport in the fractured tuffs, the authors (ZYV97)
experimented with different flow and transport models to investigate different methods of
simulating this dual-porosity system. A dual-porosity particle-tracking model for transport was
invoked, and the influence of matrix diffusion properties was examined. The results follow those
VI-104
-------
of the matrix-diffusion conceptual model, but several factors argue against its use in large-scale
transport model predictions. First, the particle-tracking module does not at present handle the
dispersion coefficient tensor formulated as longitudinal and transverse components.
Furthermore, although the diffusion model accurately simulates the case of diffusion into an
infinite matrix continuum, finite fracture spacings are not part of the model as currently
constituted. Therefore, particle tracking cannot be used in site-scale models. A finite-element
dual-porosity solution was also investigated. As expected, the dual-porosity flow simulation
yielded virtually identical steady-state results to the single-continuum model. Transport results
captured the two extremes (fracture-dominated and pervasive matrix diffusion) but failed to
produce accurate results for small but non-negligible diffusion into the rock matrix.
VI.2.1.3 TSPA-VA Implementation of Saturated Zone Modeling Analysis
The Saturated Zone Flow and Transport Technical Basis Document for the Total System
Performance Assessment - Viability Assessment provides a detailed description of the means by
which the saturated zone flow and transport modeling, as presented in sections VI.2.1.1 and
VI.2.1.2, has been integrated into the total system performance. The following sections outline
the basic components of the saturated zone modeling performed in the TSPA-VA.
Synopsis of TSPA-VA Approach
Past TSPAs focused on the biosphere interface located 5 km from the repository. Alternatively,
the TSPA-VA is now focused on calculating radionuclide doses 20 km downgradient from the
repository; the DOE indicates that this is due to changes in guidance that were based upon
recommendations from the National Research Council (NRC95).
As presented in the previous sections, numerical models were used for the purpose of
characterizing and understanding the flow and transport at the regional, site and sub-site scales
and to perform TSPA-VA calculations. The conceptual models, upon which the numerical
models were developed are based upon all available data and knowledge about the saturated
zone. Likewise, the TSPA flow and transport models were developed based upon site
knowledge, input from the workshops and expert elicitations, as well as insights provided by the
numerical modeling results.
VI-105
-------
To support TSPA-VA calculations a base-case numerical model was developed for the saturated
zone to evaluate the migration of radionuclides from their introduction at the water table below
the repository to the release point to the biosphere. In order to estimate the uncertainty
associated with the base case scenario, sensitivity analyses were also performed.
A hierarchy of models was used to simulate the movement of ground water and the transport of
radionuclides in the saturated zone. Explicit, two- and three-dimensional modeling was not used
to simulate radionuclide concentrations because it can generate numerical dispersion, which
artificially lowers the concentrations, particularly when matrix diffusion is occurring. The TSPA
3-D saturated zone flow model as described in the preceeding sections was used only to
determine flowpaths through the saturated zone and potential impacts due to climatic change
which are discussed later in this section. The TSPA 1-D saturated zone transport model was
developed based on the flow paths from the 3-D flow modeling and used to determine
concentration breakthrough curves at a distance of 20 km for unity release of radionuclides from
six streamtubes. The saturated zone transport component of the analysis was linked to the
transport calculations for the unsaturated zone through which contaminants migrate downward in
percolating ground water from the repository to the water table (e.g., the spatial and temporal
distributions of simulated mass flux at the water table). The linking was accomplished by using
the convolutional integral technique to combine the unit breakthrough curves calculated by the
TSPA one dimensional saturated zone transport model with the tune-varying radionuclide
sources from the unsaturated zone. Changes in the saturated zone flow and transport system in
response to climatic variations were incorporated for the three discrete climate states (dry, long-
term average, and superpluvial) considered in the other components of the TSPA-V A. Specific
discharge and volumetric ground-water flow rate in the saturated zone stream tubes were scaled
in transport simulations to reflect climate state. The saturated zone transport results were linked
to the biosphere analysis by the simulated time history, or system response as a function of tune,
of radionuclide concentration in ground water produced from a hypothetical well located at the
biosphere interface. The biosphere was assumed to be located 20 km from the repository.
Radionuclide concentrations in the hypothetical well water were then used in the biosphere
component to calculate doses received by the public.
For the base case, uncertainty in the saturated zone system was evaluated through Monte Carlo
variation in the input parameters used in the TSPA one dimensional saturated zone transport
model. Primarily, the uncertainty in radionuclide transport parameters was evaluated. The one-
VI-106
-------
dimensional saturated zone transport model was used to calculate 101 unit breakthrough curves
(100 Monte Carlo simulations and the expected value or base case). The results of the one
dimensional saturated zone transport calculation are located in a "family" of unit radionuclide
concentration breakthrough curves. For each TSPA-VA Repository Integration Program (RIP)
realization, a saturated zone unit breakthrough curve was randomly selected for use in the
convolution integral method.
The TSPA-VA sensitivity studies were designed to examine five of the key issues related to the
base case saturated zone analysis assumptions. The sensitivity studies were performed to provide
information about the importance of these issues with respect to repository performance. The
effect of dilution in the saturated zone and vertical transverse dispersivity was investigated to
address concerns from the Saturated Zone Expert Elicitation Panel. The impact of including
heterogenity and large-scale flow channelization in a three-dimensional flow and transport model
was also studied. A two-dimensional dual-porosity transport model was used to calculate
radionuclide concentrations to examine the effect of the base case assumptions of a single
continuum and using effective porosity as a surrogate for the matirx diffusion process. In
addition, alternative conceptual models of colloid-facilitated plutonium (Pu) transport were
developed and implemented for sensitivity analysis.
TSPA Flow System Conceptualization
Regional-Seal? Saturated Zone Flow Model. The regional-scale 3-D flow model for the TSPA-
VA was developed by D'Agnese et al. (DAG97) to characterize the conditions of the present day
ground-water flow in the Death Valley region. The numerical code used for the regional flow
model was MODFLOWP (HIL92). MODFLOWP is an adaptation of the U.S. Geological
Survey 3-D, finite-difference modular ground-water flow model, MODFLOW (MCD88).
The finite-difference mesh used for the USGS regional flow modeling consists of 163 rows, 153
columns, and three layers. The grid cells were oriented north-south and were of uniform size,
with side dimensions of 1,500 m. This results in a modeled area that is 244 km long and 229 km
wide. The layers represent conditions at 0-500 m, 500-1,250 m, and 1,250-2,750 m below the
estimated water table. The first and second layers were designed to simulate local and sub-
regional flow paths mostly within the valley-fill alluvium, volcanic rocks and shallow carbonate
VI-107
-------
rocks. The third (lowest) layer simulates deep regional flow paths in the volcanic, carbonate and
clastic rocks.
From the perspective of the TSPA-VA, the most important output from the USGS regional scale
model was information to determine a ground-water flux multiplier for the long-term average
and superpluvial climate conditions assumed for the base-case TSPA-VA analysis. The ground-
water flux multipliers were used to calculate a ground-water flux for each of the climate states
relative to present day conditions which is required as input to the TSPA one-dimensional
saturated zone model discussed later.
With respect to model calibration, the TSPA-VA states These results [calibration], suggest that
additional calibration may significantly improve model accuracy. This analysis suggests that the
model is a reasonable representation of the physical system, but evidence of model inaccuracies
exists. Inaccuracies in the simulated ground-water fluxes in the flow model are generally
proportional to uncertainty in the overall ground-water budget of the region. The model
continues to undergo development for Juture use by the YMP and the environmental restoration
program at the Nevada Test Site.
Site-Scale Satur*^ 7nn* Flow Model. Note: In a number of places the TSPA appears to be
inconsistent with the actual reference documents. For example, in Milestone SP25CM3A
(ZYV97) that describes the site scale modeling work it is indicated that the modeled area is 30 by
45 km, rather than 20 by 36 km as indicated in the TSPA.
The site-scale flow model developed by Zyvoloski et al., (ZYV97) and used to support the
TSPA-VA was developed using FEHM (ZYV83). The 3-D flow model that was developed
incorporatedanareaofabout20kmby36kmtoadepthof950mbelowthewatertable. The
model grid is a uniform orthogonal mesh with 500-m x 500-m x 50-m elements.
The results of the site-scale model calibration appear to be better than that for the regional scale
model. The results of the site-scale saturated zone flow model were used in the TSPA to
estimate flow path lengths and directions through each of the hydrostratigraphic units
downstream from the repository. This was done via a particle-tracking analysis, which only
simulates advective flow and does not account for dilution due to dispersion. The calculated
VI-108
-------
flow path lengths were incorporated into the 1-D saturated zone transport simulations discussed
below.
The TSPA states, with regard to limitations, in the site-specific flow model: The main concern is
related to the problem of large numerical dispersion inherent with the use of a relatively coarse
grid (cell size is 500 m x 500m x 50 m) when performing transport calculations. The desire to
minimize spurious transverse dispersion for the SZ analyses necessitated the use of 1-D
streamtubes for the transport calculations of the SZ analysis, which could be more finely
discretized (grids spaced at 5 m intervals).
One-Dimensional Saturated Zone Transport Model
The TSPA 1-D saturated zone transport model was developed to simulate the radionuclide
concentration breakthrough curves that form the basis of the TSPA-VA calculations performed
with the RIP computer program. Each radionuclide was transported separately in the analysis.
The DOE indicates that they chose the 1-D transport simulation method because of the desire to
eliminate spurious dilution of the radionuclide concentration resulting from numerical
dispersion, which can occur in coarsely gridded 3-D solute transport simulations.
The DOE goes on to state that Solute transport simulation using a 1-D numerical model
precludes dilution from transverse dispersion, by definition. The dilution from transverse
dispersion was explicitly specified in this modeling, as a post-processing step. This dilution
factor was treated as a stochastic parameter, as described in Section 8.4.2. Essentially, the DOE
dilutes the concentrations of the radionuclides reaching the down gradient well by a factor of 1 to
100, which they obtained from the expert elicitation panel estimates.
The six stream tubes for the 1-D transport simulations can be conceptualized as follows: They
are a combined width of 3,000 m and range between 10 and 20 m in vertical depth. This depth is
important as it will have a significant impact on dilution. The volumetric flow rate of the ground
water through each streamtube was determined at the water table from flow simulations using the
site-scale flow model developed by Bodvarsson et al. (BOD97) for the unsaturated zone. The
cross-sectional area of each streamtube was specified to be proportional to the volumetric
ground-water flow area. The specific discharge within the streamtubes in the saturated zone was
0.6 m/y for current climatic conditions.
VI-109
-------
Simulations of the radionuclide transport in the saturated zone for the TSPA-VA calculations
were performed using FEHM. The streamtubes were 20 km long, with regular grid spacing in
the tubes of 5 m. The radionuclide 1-D transport simulations were performed assuming a steady,
unit (1 g/y) radionuclide mass source at the upstream end of the streamtube (i.e., the water table
at the base of the unsaturated zone below the water table). Transport of each radionuclide was
simulated separately in the 1-D simulations. Transport simulations with the 1-D stream tube
approach implicitly assume complete mixing of the radionuclide in the volumetric ground-water
specific to each streamtube.
The convolution method was used in the TSPA-VA calculations to determine the radionuclide
concentration in the saturated zone, 20 km downgradient of the repository as a function of the
transient radionuclide mass flux at the water table beneath the repository.
Implementation with the RIP Computer Code
With the exception of the potential for colloidal transport of plutonium, all of the saturated and
unsaturated zone flow and transport parameters were sampled independently in RIP. At each
time step within the RIP simulation, the current climate state and the radionuclide mass flux at
the water table for each of the radionuclides from each of the six source subregions were passed
to the convolution integral subroutine. The convolution integral subroutine calculated the
concentration of each radionuclide for each of the six streamtubes at the 20 km distance from the
repository for that time step, including reduction of concentration from the dilution factor as
discussed in the next section. The simulated radionuclide concentrations were passed by the RIP
simulator to the biosphere component of the TSPA-VA at this point for dose calculation.
Development of Parameter Distributions apd Uncertainty
The DOE indicated that the key uncertain parameters in the analysis were the following: 1)
effective porosity in the volcanic units and the alluvium/undifferentiated valley-fill unit, 2)
distribution coefficients for sorbing radionuclides, 3) the ratio of the radionuclide mass in
aqueous and colloidal forms for colloid-facilitated transport of plutonium, 4) longitudinal
dispersivity, 5) the fraction of flowpath through the alluvium, and 6) the dilution factor.
VI-110
-------
A seventh uncertainty that should have been included is the degree to which fractures and matrix
are interacting (i.e., importance of matrix diffusion).
Uncertainty in radionuclide transport through the saturated zone was incorporated in the TSPA-
VA by varying key transport parameters for 100 realizations. All input parameters required for
the transport model were assumed to be either stochastic with an associated distribution or
constant. If the parameter was assumed to be stochastic, its distribution was determined and then
sampled using the Latin Hypercube Sampling module of the RIP code to obtain 100 sets of input
parameters. Since LHS tends to underestimate the variance on the mean, the DOE should ensure
statistical convergence.
The Saturated Zone Expert Elicitation Panel provided input for the input parameters and
distributions. For example, the panel estimated that the dilution factor ranged between 1 and 100
with a median value of about 10.
VI.2.1.4 Saturated Zone Information Needs
TSPA-VA Identified Needs
Regional Seal* Plnw Modeling
The Death Valley regional flow system consists of ground water moving through a 3-D body of
consolidated and unconsolidated materials. The 3D hydrogeologic framework model described
the characteristics of this saturated volume. The upper boundary of the flow system is the water
table. The lower boundary of the flow system is located at a depth where ground-water flow is
dominantly horizontal and moves with such small velocities that the volumes of water involved
do not significantly impact regional flow estimates. The lateral limits of the regional flow
system may be either no-flow or potential-flow boundaries. No-flow conditions exist where
ground-water movement across the boundary is prevented by physical barriers or divergence of
ground-water flow paths. Flow exists where ground-water potentiometric gradients permit flow
across a boundary.
For purposes of conceptualization and subsequent numerical simulation, the limits of the flow
system for the regional model were selected based on reevaluation of previously defined flow
VI-111
-------
system boundaries, the potentiometric surface developed for the regional modeling study, and the
hydrogeologic framework model. Very little hard data exist to support a precise definition of the
western extent of the flow system. The western boundary of the flow system is therefore placed
to coincide with the eastern edge of the Death Valley saltpan which is interpreted as the terminal
sink of the flow system.
The water budget for the Death Valley regional ground-water flow system is difficult to
compute, because inflow and outflow volumes are poorly defined for many areas. In addition,
the large size of this regional system precludes the comprehensive and accurate assessment of all
inflows to and outflows from the system. Previous attempts to estimate water budgets for
various parts of the flow system did not use consistent boundaries, so the budgets cannot be
readily compared. The regional model uses a lumped-budget approach; each component of the
ground-water budget is defined by a single lumped value even though it may have been
calculated originally for separate areas in the basin. This lumped-budget approach permits an
encapsulated view of the system, but the authors (ZYV97) point out that errors are inevitable in
the estimates. Short of physical measurements modeling is probably the best means of resolving
these errors.
Problems with the regional model are indicated by weighted residuals that are not entirely
random, indicating some model error. This is related to the occurrence of large positive
weighted residuals for hydraulic heads, where simulated hydraulic heads are distinctly lower than
the observed values, and large negative weighted residuals for spring flows, where simulated
flows are distinctly less than observed flows. The problem is also related to nonnormally
distributed less extreme weighted residuals. These results, combined with the previously
discussed observation that every model update considered thus far, significantly improved
model fit, suggests that additional calibration may significantly improve model accuracy. While
the authors (ZYV97) believe that the model is a reasonable representation of the physical system,
evidence of important model error exists.
Site-Scale Flow Modeling
motive* of the modeling. The authors (ZYV97) outlined the objectives of the site-scale model
as the following:
VI-112
-------
1. Provide a large-scale description of the hydrogeologic framework of the site saturated
zone flow system based on a sampling of 1500 m by 1500 m mesh;
2. Provide a mechanism to extend model calibration and sensitivity testing of parameters
used in the model;
3. Provide the flow field for doing preliminary transport simulations and estimates of
ground- water travel time through the use of additional transport related capabilities
within FEHM; and
4. Provide initial estimates of permeability for 16 hydrogeologic units from the SFM
and two additional zones of small permeability and recharge at Fortymile Wash.
Model Limitations
The authors (ZYV97) outlined the model limitations as follows:
1. Simulations are restricted to fully saturated conditions from the water table and below.
Although the model was built using a framework model that extended to land surface, the
unsaturated zone was not included as part of the flow model. The unsaturated zone was omitted
because of time constraints and the long execution times for forward simulation runs associated
with two-phase flow problems.
2. The model does not account for variations in temperature within the flow system.
Temperature varies within the ground-water flow system and may be a useful constraint in
identifying acceptable model representations of both temperature and hydraulic head. The
preliminary status of the model limited the extent to which temperature could be evaluated.
Furthermore, the temperature of the system was specified at a uniform 44 °C which may be too
high to represent the average temperature.
3. It is likely that the flow model is non-unique. Coordinated adjustments in permeability
values (either higher or lower by some multiplier) might lead to similar hydraulic head
distribution and calibration. Because fluxes were not specified explicitly at either the upgradient
or downgradient ends of the model, the model is less constrained as it would be with fluxes
included in the calibration. However, because some permeability values (of admittedly minimal
accuracy) were specified explicitly throughout the parameter estimation, the model was partially
constrained which likely caused the parameter estimation process to converge in many instances.
VI-113
-------
4. The large hydraulic gradient is poorly understood and greatly affects model calibration,
simulated permeability values, and flux. Additional data and testing are required to adequately
characterize this feature. Testing and reconfiguration of monitoring intervals within borehole
USW G-2 could be done to provide permeability, flow-survey, temperature, and hydraulic-head
data at different depths, particularly for the middle volcanic aquifer. Construction of additional
boreholes in the large hydraulic gradient area, such as a corehole into the middle volcanic aquifer
adjacent to drillhole WT-6, could provide useful vertical gradient, hydraulic-head, saturation, and
permeability data. The authors (ZYV97) believe that the site-scale model was successful in
representing the large hydraulic gradient through the incorporation of a vertical barrier to flow,
but other representations are possible.
5. Flux into the site model domain is poorly defined and remains one of the most elusive of
model variables. The quality of the model is in part a measure of the understanding of the
distribution and amount of recharge within the model domain. Comparison effluxes into and
out of the model is dependent on available data, which although greatly lacking will not likely be
improved substantially through additional field studies. Water levels within the flow system
could still be adjusting to recharge supplied during climatically wetter conditions. If such a
condition exists, the effect may be too subtle to observe with the available hydraulic data.
Adjusting water-level conditions could be evaluated using the regional model to replicate
conditions necessary to observe the effect of increased recharge under past wetter climates.
6. Limited hydraulic-test data exist for constraining permeability values used in the model.
Few hydraulic-test data are available that involve multiple observation wells within the model
domain from which huge-scale transmissivity or hydraulic conductivity values can be derived.
The exception to this condition is the C-hole-complex hydraulic testing which is optimally
located for conditions at Yucca Mountain and provides a test involving a large volume of the
middle volcanic aquifer. However, the C-hole testing is in a highly fractured area and might not
be representative of the entire area. In general, the model does not distinguish between the
permeability of the rock matrix, fractures, or faults. It is possible to add large-scale features such
as faults explicitly within the model by regridding, but hydraulic characteristics for faults in the
saturated zone are not presently available.
7. Definition of the hydrogeologic units within the model is limited by the sampling
interval used (1500 m). By sampling the framework model at a smaller interval (for example,
VI-114
-------
250 m) better resolution of the hydrogeologic units could be obtained, but resulting in a larger
computation mesh. Experience from the site-scale modeling exercise suggests that this approach
is warranted and likely would succeed. However, higher resolution sampling alone may be
insufficient to explicitly represent faults.
As noted above, comparisons of flux from the regional model showed almost twice the amount
discharging from the southern end of the site model, and substantially different amounts for the
north and east sides. The major flux differences between the two models occur in the northeast
corner where a large part of the recharge from the north is diverted east and discharges in part
because of the interaction of the constant-head boundaries and the imposed east-west barrier
needed to represent the large-hydraulic gradient.
The authors (ZYV97) observe that on initial inspection, model match to hydraulic-head data and
the resulting distribution of residuals have some problems. Although permeability values for all
of the hydrogeologic units used in the model lie within reported literature values, reported values
for individual units have large ranges. Furthermore, in the case of the middle volcanic aquifer,
values of permeability from large-scale hydraulic testing at the C-hole complex were three orders
of magnitude larger than those used in the model. This discrepancy may be indicative of model
error, or alternately, the possibility of a local, large-permeability zone not represented in the
present model. Finally, any model calibrated using hydraulic heads alone is subject to error in
simulated flux.
Improvements suggested by the authors (ZYV97) for future model developments include:
• Conduct sensitivity analyses with regard to which model variables have the greatest
effect when varied on the sum of squared residuals for hydraulic head. This would
provide a guide for additional field studies to reduce uncertainty in the model.
• Refine hydrogeologic framework model to better define the distribution of the
hydrogeologic units. In particular, the upper volcanic confining unit is currently over-
represented. This discrepancy substantially influences simulated flow and transport
simulations.
• Use higher resolution sampling of the hydrogeologic framework model to better
delineate unit offsets caused by faulting. This would result in a denser finite-element
mesh, resulting in longer execution times, but would provide a more realistic
portrayal of the flow system than is available in the model.
VI-115
-------
• Add major faults explicitly as surfaces within a refined version of the hydrogeologic
frame-work model, so that their potential as barriers to flow or as fast pathways to the
accessible environment may be evaluated.
• Decouple permeability parameters for the upper and middle volcanic aquifers to the
extent practical during model calibration. This separation of the two primary volcanic
aquifers at Yucca Mountain within the model would better represent the permeability
distribution.
• Recalibrate the existing model with larger values of permeability in the middle
volcanic aquifer and the upper volcanic aquifer.
• Incorporate additional data into the formal model calibration. This could include flux
data from the regional model for at least one face of the model and borehole-
temperature data to better constrain the solution.
• Fluxes should be extracted from a refined, improved version of the USGS regional
model of D'Agnese et. al. (in press) in which the topmost layer has been subdivided
to better represent the hydrogeologic units at Yucca Mountain and in the Amargosa
Desert.
• Include vertical flux through the bottom of the model based on regional model values.
• Use hydrochemical and isotopic data as a check against flow model results.
Snh-Site-Scale Flow Modeling
For the sub-site scale modeling the authors (COH97) reached a number of conclusions regarding
the adequacy of the existing data base. These conclusions/recommendations include:
• The properties of the Bullfrog need further characterization since (a) this unit is by far
the most important for flow in the saturated zone due to its large fracture
permeability, and (b) it underlies the repository at the water table.
• Fault properties need further characterization due to their obvious effects on flow at
the sub-site scale.
• Measurements of fracture porosity must be made so that models can better estimate
pore velocity.
VI-116
-------
• Geochemical sampling that considers vertical hydrochemical variations is needed to
understand the 3-D nature of flow in the saturated zone.
• Investigation by numerical simulation of the large hydraulic gradient should be
undertaken. The authors (COH97) believe that the existing unsaturated and saturated
zone models contain enough geologic detail and process modeling capability to make
a credible attempt in this direction. Coupling of the two models can be done to
facilitate these analyses.
• Because chemical components undergo mixing and dilution due to flow and fracture-
matrix interactions, the processes of flow and transport are coupled and should be
considered together.
• Additional hydraulic tests and well placement should be designed to focus on fault
properties and the Bullfrog unit.
• Extensions of the grid should be made to decrease boundary effects and better model
transient pumping tests.
• The grid should be extended spatially to model flow and mixing at greater distances
from the repository.
• Advanced visualization techniques should be brought to bear on the problem of
elucidating the complex 3-D flows observed in the simulations.
Contaminant Trnsort Modelin
Conclusions iH Recommendations. The authors (COH97) developed the following conclusions
and recommendations with respect to the contaminant transport modeling:
• Grid resolution was found to be critical for both flow and transport.
• The transport simulations are very sensitive to matrix diffusion.
• Continue close collaborative effort with the USGS site flow modeling. Continue to
update the Stratigraphic Framework Model with the best available data and establish
quantitative measures of grid fidelity to the SFM.
• Perform preliminary transport simulations during continued flow calibration efforts.
This effort will uncover early possible incompatibilities between the flow and
transport conceptual models and will eliminate any problems in sampling frequency
VI-117
-------
that may result in an inaccurate representation of units. This parallel computation
may also uncover potential fast paths for radionuclides that can be corroborated or
dismissed with supporting thermal or geochemical data. In this way, transport
modeling will provide a tool for improving the flow model. By establishing the
effect on the radionuclide flux at the compliance boundaries, flow model
improvements, such as creating a detailed inlet flux map, may be evaluated with
respect to their importance to performance assessment.
• Enlarge the modeling domain of the site-scale model north of the large hydraulic
gradient area where the calibration of the regional model is better and the fluxes of the
regional model are more defensible. This approach will also allow better
redistribution effluxes from the three-layer regional model to the sixteen-layer site-
scale model.
• Use the 250-m sampling of the SFM. This effort will provide suficient resolution, in
the vicinity of Yucca Mountain for transport calculations.
• Use stochastic approach for transport modeling to better characterize uncertainties in
the dispersive mixing process occurring at subgrid block scales.
NRC Identified Needs
The NRC is developing a strategy for reviewing the performance of the proposed repository. As
currently envisioned, the elements of this strategy necessary to determine acceptability of
repository performance are defined by the NRC as key technical issues of the subsystem
abstractions. As part of this process, the NRC has developed Acceptance Criteria for the key
issues of the DOE TSPA that will ultimately be used to determine the viability of the repository
at Yucca Mountain. This NRC evaluation, is very relevant in assessing whether the methods and
information presented by the DOE have the potential to produce results that are defensible under
regulatory reviews.
The two NRC status reports that are most pertinent to ground-water flow and contaminant
transport are entitled Issue Resolution Status Report - Key Technical Issue: Unsaturated and
Saturated Flow under Isothermal Conditions and Issue Resolution Status Report - Key Technical
Issue: Radionuclide Transport.
The following discussion is organized by key technical issues. Under each issue, the NRC's
Acceptance Criteria are presented and their assessment as to whether DOE has met the Criteria is
provided.
VI-118
-------
NRC's Issue/Subissue Statement
The NRC developed these Issue Resolution Status Reports with the primary objective to assess
all aspects of the ambient hydrogeologic regime at Yucca Mountain that have the potential to
compromise the performance of the proposed repository. The secondary objective was to
develop review procedures and to conduct technical investigations to assess the adequacy of
DOE's characterization of key site- and regional-scale hydrogeologic processes and features that
may adversely affect performance. The primary issues identified by NRC with respect to the
hydrologic regime are the following:
• Hydrologic Effects of Climate Change
• Present-Day Shallow Infiltration
• Deep Percolation (Present and Future)
• Saturated Zone Ambient Flow Conditions and Dilution Processes
• Matrix Diffusion
• Radionuclide Transport Through Porous Rock
• Radionuclide Transport Through Alluvium
• Radionuclide Transport Through Fractured Rock
The following sections discuss each of the above issues by presenting its relevance to PA, as well
as NRC's Acceptance Criteria and resolution status with respect to the TSPA-LA.
Issue Resolution Status
Hydrogeologic Effects of Climate Change
Relevance to PA. For the DOE to adequately demonstrate and quantify in its Total System
Performance Assessment (TSPA) the effects that climate change might have on repository
performance, the NRC believes that it should consider how these effects interplay with the other
factors within and between key elements in the engineered and natural subsystems of the
repository. Climate change and its hydrologic effects are important factors that need to be
abstracted into three of the key elements of the engineered and natural subsystems: (1) spatial
and temporal distribution of flow; (2) flow rate in water production zones; and (3) location and
lifestyle of critical group (includes consideration of water-table rise).
A description of the technical basis for review methods and acceptance criteria for the subissues
of climate change and hydrologic effects of climate change is presented in NRC97a. An
VI-119
-------
important new paper on Devils Hole was published in 1997; however, at this time the NRC has
not changed the previously developed acceptance criteria.
The NRC has previously recommended (NRC97a, p. 8) a pragmatic approach to address climate
change. Under this approach, global, enhanced, greenhouse warming would be presumed to last
no more than several thousand years, and that, about 3,000 years into the future, the climate at
Yucca Mountain will resume global cooling predicted by the Milankovitch orbital theory of
climate. Pluvial conditions should be expected to dominate at least several thousand years of the
next 10,000 years. According to NRC's analysis past climate conditions were cooler and wetter
than today, about 60 to 80 percent of the time.
NRC Acceptance Criteria. In the NRC's Technical Review of the TSPA-LA it will determine
whether DOE has reasonably complied with the Acceptance Criteria listed below:
. 1. Climate projections based primarily on paleoclimate data are acceptable for use in
performance assessments of the Yucca Mountain site. During its review, the staff
should determine whether the DOE has made a reasonably complete search of
paleoclimate data that are available for the Yucca Mountain site and region, and has
satisfactorily documented the results. Staff should determine that, at a minimum, the
DOE has considered information contained in Winograd et al. (WIN92); Szabo, et al.
(SZA94); and other reports that may become available.
2. The DOE's projections of long-term climate change are acceptable if these projected
changes are consistent with evidence from the paleoclimate data. Specifically, NRC
staff should determine whether the DOE has evaluated long-term climate change
based on known patterns of climatic cycles during the Quaternary, especially the last
500 ky. The current analysis indicates that these cycles included roughly 100,000
year cycles of glacial/interglacial climates, with interglacials lasting about 20,000
years.
3. The NRC will not require climate modeling to estimate the range of future climates.
If the DOE uses numerical climate models, NRC staff will determine whether such
models were calibrated with paleoclimate data before they were used for projection of
future climate, and that their use suitably simulates the historical record.
4. Values for climatic parameters (time(s) of onset of climate change; mean annual
precipitation (MAP); mean annual temperature (MAT); etc.) to be used in DOE's
safety case should be adequately justified. This includes determination of whether
appropriate scientific data were used, reasonably interpreted, and appropriately
VI-120
-------
synthesized into parameters such as MAP, MAT, and long-term climate variability.
The current knowledge about these parameters, coupled with past climate change,
will require that, as a bounding condition, a return to full pluvial climate (higher
precipitation and lower temperatures) be considered for at least a part of the 10,000
year period (current information does not support persistence of present-day climate
for a duration of 10,000 years or more). The current interpretations of paleoclimate
data indicate an increase in MAP by a factor of 2 to 3 and a lowering of MAT of 5-
10°C (9-18 °F) during the pluvial climate episodes.
5. If the DOE uses expert elicitation to arrive at values of climate parameters, the NRC
will determine whether the guidance hi the Branch Technical Position on Expert
Elicitation (NRC96) was followed by the DOE.
6. Bounding values of climate-induced effects (for example water-table rise) based
primarily on paleoclimate data will be acceptable. The NRC should determine
whether the DOE has made a reasonably complete search of paleoclimate data
pertinent to water-table rise and other effects (for example, changes hi precipitation
and geochemistry) of climate change that are available for the Yucca Mountain site
and region, and has satisfactorily documented the results. In evaluating the DOE's
analyses, NRC staff should determine whether, at a minimum, DOE has fully
considered information contained in Paces, et al, (PAC96), Szabo, et al. (SZA94),
and other reports that may become available.
7. It will be acceptable for the DOE to use regional and sub-regional models for the
saturated zone to predict climate-induced consequences if these models are calibrated
with the paleohydrology data. NRC staff should determine whether the DOE's
models of the consequences of climate change are consistent with evidence from the
extensive paleoclimate data base. Specifically, climate-induced water-table rise is
expected to occur in response to elevated precipitation during future pluvial climate
episodes, and the staff should determine whether the DOE's estimates of climate-
induced, water-table rise are consistent with the paleoclimate data. The current
estimate of water-table rise during the late Pleistocene is 120 m (394 ft). The NRC
should determine whether the DOE's assumptions about climate-induced, water-table
rise over 10,000 years, if different from 120 m (394 ft), are adequately justified.
8. Based on judgment and analysis, NRC staff will determine whether the DOE has
adequately incorporated future climate changes and associated effects in its
performance assessments. Current information does not support an assumption that
present-day climate will persist unchanged for 10,000 years or more. The NRC staff
should keep in mind that the consequences of climate change may be coupled to other
events and processes and therefore the projections of water-table rise that are used in
total system performance may be different from those based solely on climate change.
VI-121
-------
9. The collection, documentation, and development of data, models, and computer codes
have been performed under acceptable QA procedures. If they were not subject to an
acceptable QA procedure, they must be appropriately qualified.
Status of Issue Resolution at the NRC Staff Level. In Attachment E of the Issue Resolution
Status Report - Technical Issue: Unsaturated and Saturated Flow the NRC presents their current
concerns related to the potential influence of climate change on ground-water flow. The text
indicates that the NRC has identified no open items solely related to future climate change and
associated hydrologic effects.
Present-Day Shallow Infiltration
Relevance to PA. NRC believes that present-day shallow infiltration is a key hydrologic factor
in the isolation of high level wastes within the proposed geologic repository at Yucca Mountain.
Present day shallow infiltration should be reasonably understood to provide initial conditions for
projecting future hydrologic changes, because the Earth's climate could change significantly
during the time that wastes will remain hazardous. Climate controls the range of precipitation
that, in part, controls the rates of infiltration, deep percolation, and ground-water flux through a
geologic repository located in an unsaturated environment. Water flow through a geologic
repository and its environs depends on both surface processes (precipitation, evapotranspiration,
overland flow, and infiltration) and subsurface processes (deep percolation, moisture
recirculation, and lateral flow). Changes in infiltration will likely induce other changes, such as
regional fluctuations in the elevation of the water table. Water-table rise would reduce the
thickness of the unsaturated zone barrier. Therefore, future changes in climate could alter
infiltration from present-day rates and significantly influence the ability of a repository to isolate
waste.
The importance of ground-water flux as the key parameter for repository performance in an
unsaturated zone is well known, and has been further emphasized in DOE's 1995 Total System
Performance Assessment (TSPA). On page ES-30 of that report it is stated that:
...in the overall TSPA analyses, an over-arching theme comes back again and again as being
the driving factor impacting the predicted results. Simply stated, it is the amount of water
present hi the natural and engineered systems and the magnitude of aqueous flux through
these systems that controls the overall predicted performance.... Therefore, information
VI-122
-------
on...[this topic]...remains the key need to enhance the representativeness of future iterations
ofTSPA.
Sensitivity studies clearly showed the predominance of percolation flux in estimating cumulative
radionuclide releases and peak radiation doses over a 10,000-year period (DOE95).
The DOE's "Waste Containment and Isolation Strategy" (DOE96) likewise states that
"performance assessments have shown that seepage into the emplacement drifts is the most
important determinant of the ability of the site to contain and isolate waste." This conclusion
was reiterated in the DOE's recently published Repository Safety Strategy (DOE98). The
importance of infiltration as a hydrologic parameter was also recognized by the NRC in its
Iterative Performance Assessment Phase 2. The NRC (NRC95, p. 10-4) states that "Although
the flux of liquid water through the repository depends on...infiltration, hydraulic conductivity,
and porosity, performance correlates most strongly to infiltration."
In Section 5.1.2 of the DOE's 1998 TSPA-VA, the sensitivity to infiltration is investigated by
skewing the probabilities to the higher infiltration rates than used in the base case simulations.
The results of this analysis showed relatively small differences in the overall peak individual
dose rates largely because other factors such as seepage and waste package corrosion
uncertainties.
The NRC believes that, for the DOE to adequately demonstrate and quantify in its TSPA-LA the
effects that present-day infiltration might have on repository performance, it should consider how
these effects interplay with the other factors within and between key elements in the engineered
and natural subsystems of the repository.
NRC97b provides a description of the technical basis for review methods for the issue on
present-day shallow infiltration.
NRC Acceptance Criteria. In the NRC's Technical Review of the TSPA-LA it will determine
whether the DOE has reasonably complied with the Acceptance Criteria listed below:
(1) The NRC shall determine whether the DOE has estimated shallow infiltration for
use in the PA of Yucca Mountain using mathematical models that incorporate
site-specific climatic, surface, and subsurface information. The staff will also
VI-123
-------
determine whether the DOE provided sufficient evidence that the mathematical
models were reasonably verified with site data. These data would include
measured infiltration data and indirect evidence such as geochemical and
.. geothermal data. The DOE may choose to use a vertical one-dimensional (1-D)
model to simulate infiltration. However, in that case, the DOE should reasonably
show that the fundamental effects of heterogeneities, time-varying boundary
conditions, evapotranspiration, depth of soil cover, and surface-water runoff have
been considered in ways that do not underestimate infiltration.
(2) The NRC shall determine whether the DOE has: (1) appropriately analyzed
infiltration at appropriate time and space scales; and (2) has tested the abstracted
model against more detailed models to assure that it produces reasonable results
for shallow infiltration under conditions of interest. Recent studies by the NRC
(STO96) and the DOE (FLI94, FLI95, FLI96) suggest that shallow infiltration is
relatively high in areas where rocks are covered with shallow soils or channels
and relatively low hi areas where soil cover is deep. In addition, infiltration takes
place episodically in tune with areas having a shallow soil cover contributing
more frequently.
(3) The NRC shall determine whether the DOE has characterized shallow infiltration
in the form of either probability distributions or deterministic upper-bound values
for PA, and whether the DOE has provided sufficient data and analyses to justify
the chosen probability distribution or bounding value. The DOE's expert
elicitation on unsaturated zone flow (GEO97) resulted in various estimates of a
related parameter, the ground-water percolation flux at the depth of the proposed
repository. The estimated aggregate mean flux was approximately 10 mm/yr.
The panelists estimated the 95th-percentile percolation flux over a range from 10
to 50 mm/yr, with an aggregate estimate of 30 mm/yr. An independent NRC staff
assessment of an upper bound for yearly shallow infiltration under present
climatic conditions is about 25 mm, which is somewhat less than the aggregate
95th percentile flux estimated by the expert panel.
(4) The DOE's estimates of the probability distribution or upper bound for present-
day shallow infiltration need not be refined further if the DOE demonstrates
through TSPA and associated sensitivity analyses that such refinements will not
significantly alter the estimate of total-system performance.
(5) If used, expert elicitations should have been conducted and documented using the
guidance in the Branch Technical Position on Expert Elicitation (NRC96), or
other acceptable approaches.
(6) The NRC will determine whether the collection, documentation, and development
of data, models, and computer codes have been performed under acceptable QA
VI-124
-------
procedures. If they were not subject to an acceptable QA procedure, they have
been appropriately qualified.
Status of Issue Resolution at the NRC Staff Level. In Attachment F of the Issue Resolution
Status Report - Key Technical Issues: Unsaturated and Saturated Flow the NRC presents their
current concerns related to Present-Day Shallow Infiltration. The text indicates that the NRC
staff has identified no open items solely related to Present-Day Shallow Infiltration.
Deep Percolation (Present and Future)
Relevance to PA. The importance of ground-water flux as the key parameter for waste isolation
at Yucca Mountain is well known.
Deep percolation is related to two of the key elements of the engineered and natural subsystems:
(1) quantity and chemistry of water contacting waste packages and waste forms; and (2) spatial
and temporal distribution of flow.
The NRC's technical review of the DOE's treatment of deep percolation is to be based on an
evaluation of the completeness and applicability of the data and evaluations presented by the
DOE. The NRC expects that the DOE will summarize or document the results of all significant-
related studies that have been conducted in the Yucca Mountain vicinity.
NRC Acceptance Criteria and Resolution Status in the NRC's Technical Review of the TSPA-
LA it will determine whether the DOE has reasonably complied with the Acceptance Criteria
listed below. The results of the NRC's most recent analyses and issue resolutions are also
presented.
(1) It will be acceptable for the DOE to estimate present-day deep percolation by using (1) a
reasonable upper bound based on available data; or (2) through a demonstration in TSPA
and associated sensitivity analyses that further refinement of the estimate will not
significantly alter the estimate of total-system performance. In the latter case, the NRC
will conduct an independent analysis to judge the appropriateness of the estimate. In the
VA analysis, it will be acceptable to use the aggregate distribution for areally averaged
percolation flux estimated through the expert elicitation (i.e., GEO97). The DOE's
VI-125
-------
current infiltration map (e.g., FLI96) may be used to account for spatial variations in
percolation.
According to the NRC, the base-case percolation flux as described by DOE appears acceptable at
this time because it is similar to that estimated through expert elicitation (GEO97). If this base-
case flux is used by the DOE, this acceptance criterion will be met. The status of this issue is
open pending review of the DOE's VA.
(2) The DOE's estimate of future percolation will be acceptable if it provides a reasonable basis
for assumed long-term average net infiltration and percolation flux. It will be acceptable to
apply spatial- and temporal-average values of deep percolation through the use of an
abstracted deep percolation model in PA. In arriving at spatial- and temporal-average values:
variability is appropriately considered; model parameters are averaged over appropriate time
and space scales; and the abstracted model is tested against more detailed models and field
observations to assure that it produces reasonably conservative dose estimates. The current
understanding is that a vertical one dimensional model, capable of considering
heterogeneities and time-varying boundary conditions at the ground surface, may be
sufficient for such calculations above the repository, while a vertically oriented, two
dimensional model or three dimensional model may be necessary below the repository.
According to the NRC's analysis, the DOE currently (AND98) assumes that long-term average
precipitation at Yucca Mountain will be twice as high as present conditions and long-term
average percolation will be six times greater. The assumption about long-term average
precipitation at Yucca Mountain is reasonably consistent with that recommended in Attachment
E (NRC97a). It is not yet clear whether a six-fold increase in long-term average percolation is
reasonable. The staff will make that determination after review of DOE submittals.
The NRC staff considers that the LBNL 3-D site-scale model may be too coarse to provide more
than a general indication of subsurface processes at Yucca Mountain, but note that significant
model refinement may be computationally infeasible. Despite these reservations, the NRC staff
endorse the LBNL philosophy of using all available sources of information to calibrate the site-
scale model, and agree that, for many purposes, homogeneous effective properties for each layer
obtained through inverse modeling may be adequate.
VI-126
-------
The NRC staff supports the use of the DKM (i.e., dual permeability) approach for site-scale flow
modeling as long as the DOE demonstrates that the results bound the effect of episodic
infiltration and percolation pulses.
The NRC staff considers that approaches used by the DOE to estimate parameters for flow and
transport simulations generally use sound methods, particularly in the most recent work. The
NRC staff notes, however, that subgrid heterogeneity is not explicitly and transparently
addressed in the approaches, and caution that failure to consider subgrid heterogeneity may lead
to qualitatively incorrect results. Small scale modeling of heterogeneous zones is one approach
that may be used to support use of uniform properties in hydrostratigraphic units of the site-scale
UZ flow model.
The staff have reasons to believe that recharge and percolation in the Yucca Mountain region
may increase in the next few decades due to replacement of native shrubs by invading brome
grasses. The effect will likely be to replace the zero distributed recharge occurring in the alluvial
basins with, perhaps, 1 to 10 mm/yr under current conditions. Recharge in upland areas like
Yucca Mountain may also increase. The effect may be significantly greater during pluvial
periods. This point is based on infiltration simulations, and on observations of increased
streamflows where invasions of Bromus species have occurred in Nevada.
The status of this issue is open pending review of the DOE's VA.
(3) It will be acceptable for the DOE to conservatively assume that the fraction of deep
percolation that intercepts disposal drifts also drips onto waste packages. Technical bases
should be provided for deep percolation that is considered to bypass emplacement drifts.
These technical bases should use field observations, experimental data from the ESF,
calculations based on mass balance, tracer studies, and data from natural analog sites.
Likely changes in percolation rates and patterns due to climate change should also be
considered. Also, the abstracted model used in PA should be tested against more detailed
models and field observations to assure that it produces reasonably conservative dose
estimates. It is known that the amount of deep percolation into the waste emplacement
drifts is sensitive to fast flow in fracture zones. Such flow paths need to be considered in
the DOE's calculations
VI-127
-------
According to the NRC's analysis, the DOE is developing an approach for estimating seepage into
drifts. The current DOE approach for drift-scale modeling for isothermal flow is apparently to
represent the fracture system as an equivalent continuum with or without incorporating the
matrix continuum (NIT97). However, it is not clear that the fracture system can be represented
as a continuum at the scale of an emplacement drift based on the average fracture spacings in the
repository horizons and the grid size in the numerical model. The NRC believes that alternative
approaches will be needed to support estimates of seepage.
Although direct measurement of percolation flux at spatial and temporal scales relevant to
modeling at Yucca Mountain is difficult to accomplish, the NRC staff believes that efforts should
continue to identify pathways and measure percolation in the field at Yucca Mountain. A
number of field tests designed to investigate percolation and seepage rates are planned or
currently in progress, notably the alcove and niche infiltration tests and testing planned for the
east-west drift hi the TSw (WAN98). The direct measurement of percolation flux is encouraged,
and the DOE should consider, to the extent practicable, that the proposed east-west drift be
allowed to equilibrate with ambient conditions by closing down the tunnel for a period of time.
The east-west drift has a significant lateral extent for observing seepage and dripping into the
tunnel under ambient conditions, and will cross beneath what are expected to be areas of
relatively high infiltration.
Besides providing independent estimates of deep percolation rates, the NRC staff will review
whether or not the data used in the methods described in the following sections were extensively
incorporated, either directly or as constraints, into the calibration process for the LBNL site-scale
numerical model of the flow field (BOD97).
The NRC believes that another possible way that the DOE can demonstrate a reasonable
approach is to assume that the fraction of percolating water that contacts waste packages is at
least as great as the amount that intercepts disposal drifts. This means that most deep percolation
will bypass waste packages because the disposal drifts occupy a relatively small area! percentage
of the repository. This approach is probably reasonable and conservative given the tendency of
underground openings to divert UZ flow laterally. It may not be reasonable to assume that all
packages will receive equal amounts of dripping. Many may receive little or no dripping, while
others could experience greater than average dripping over long time periods, especially during
pluvial climate episodes.
VI-128
-------
The issue remains open pending review of DOE's VA and key supporting documents.
(4) It will be acceptable for the DOE to conservatively assume that all deep percolation
below the repository level bypasses the bulk of the units of the CHn formation, either by
lateral movement above the units or through vertical flow through fractures and faults.
Technical bases should be developed for any deep percolation considered to flow
vertically through the matrix of the nonwelded zone. Such technical bases should
consider spatial and temporal variability and the scales at which model parameters have
been-averaged. Also, the abstracted model has been tested against more detailed models
and field observations to assure that it produces reasonably conservative dose estimates.
According to the NRC's current understanding, flow will occur predominantly vertically as
matrix flow through the nonwelded vitric zones, including those that are slightly altered. Water
will tend to perch upon highly zeolitized horizons and move laterally until vertical structures are
encountered. Flow through fractures and fault systems will also occur. The NRC staff believes
that the heterogeneity of the hydraulic properties and the characteristics of the fractures cross-
cutting the units of the CHn are both poorly known. The field-scale UZ transport test at Busted
Butte will significantly improve the conceptual model of flow through the CHn, but it will also
contribute significant data for characterizing hydraulic properties, thus reducing uncertainty in
flow rates below the repository. The DOE is continuing to work on flow below the repository,
with one objective being to estimate how much bypass flux is reasonable.
The status of this issue remains open pending review of DOE's VA and reports on results from
the Busted Butte hydrologic test facility.
(5) If used, DOE's expert elicitations should have been conducted and documented using the
guidance in the Branch Technical Position on Expert Elicitation (NRC96), or other
acceptable approaches.
The NRC has concluded that the expert elicitation on DOE's unsaturated flow model (i.e.,
GEO97) was conducted and documented in an acceptable way. Consequently this issue is closed
and. the staff have no further questions at this time.
VI-129
-------
(6) The NRC will determine whether the collection, documentation, and development of data,
models, and computer codes have been performed under acceptable QA procedures. If they
were not subject to an acceptable QA procedure, they have been appropriately qualified.
The NRC has not yet analyzed this issue and determined the path to resolution.
Summary of Deep Percolation Topics That NRC Believes Warrant Further Analysis. Significant
variability of flow and transport pathways and travel times is expected to occur at Yucca
Mountain due to the natural heterogeneity, stratification, alteration, fracturing, and other
characteristics of the site. The extent to which such heterogeneities of the flow system should be
incorporated into the DOE site-scale UZ flow model depends on their importance for estimating
seepage into the repository and flow below the repository. Conceptualizations of flow in the UZ
at Yucca Mountain have ranged from single-continuum models, to equivalent continuum models,
to dual- and multiple- continuum models, to discrete-fracture models, as the importance of
particular components of the flow system was examined. Given the matrix permeability values
(FLI96) and assuming a unit hydraulic gradient, ground water flowing only in the matrix would
move sufficiently slowly that it would take many tens of thousands of years for shallow
infiltration to go through the repository horizon and arrive at the SZ. In contrast, both
geochemical evidence and transient-flow modeling have suggested that a significant amount of
ground-water flux occurs in the fracture system, and that these fluxes can travel at much faster
rates than in the matrix. Fluxes in the fracture systems may move sufficiently fast that some
component of shallow infiltration reaches the water table in tens to hundreds of years. Differing
conceptualizations of the link between the matrix and fracture systems and flow processes hi the
fractures cause important differences between alternative conceptual models. The differences in
the conceptualizations can have a strong impact on PA modeling and, as such, are the focus of
the discussion in this section.
The development of both the repository-scale and drift-scale conceptual models at Yucca
Mountain may be partitioned into:
(1) Percolation processes above the repository, which affect the spatial and temporal
distribution of water moving through the repository horizon
(1) Percolation processes at the drift scale, which affect the release of radionuclides from the
repository
VI-130
-------
(3) Percolation processes below the repository, which affect the transport of radionuclides
from the repository to the SZ
VI-131
-------
Saturated Zone Ambient Flow Conditions and Dilution Processes
Relevance to PA. This issue is important to repository performance because it constitutes an
important potential pathway for radionuclide transport from the repository to the environment
and receptor locations. Saturated zone characteristics will influence how future societies may
use ground-water resources in the Yucca Mountain region. The SZ also contributes to repository
performance through: (1) magnitude and direction of ground-water flow; (2) geochemical
retardation; and (3) dilution of radionuclides. The time of arrival and the concentration of
radionuclides at the receptor locations are based on the average ground-water fluxes and
velocities and the geochemical conditions encountered along the flow paths. Longer residence
times will provide opportunity for radioactive decay, and the ground-water pathways will affect
transport due to retardation and adsorption.
The concentration of radionuclides at the receptor locations is also affected by the dilution
processes during transport (dispersion and ground-water intrabasin mixing) and pumping. The
importance of dilution of radionuclides in the ground-water is a central issue for dose reduction
in the PA. The DOE TSPA-VA identifies dilution in the saturated zone below the repository as
one of the five major system attributes most important for PA.
The Repository Safety Strategy (DOE98) notes that "Significant flow must occur in the saturated
zone in order for the radionuclide-bearing flux that percolates to the water table to be diluted.
The magnitude of mixing and dispersion also must be established because certain conditions
have been noted to lead to persistence of contaminant plumes...However, even persistent
contaminant plumes may themselves be subject to significant dilution when mixed with other
water in a producing well."
Ambient flow conditions in the saturated zone are related to three of the key elements of the
engineered and natural subsystems: (1) flow rates in water-production zones; (2) dilution of
radionuclides in ground water (dispersion and well pumping); and (3) location and lifestyle of the
critical group.
NRC Acceptance Criteria and Resolution Status. In the NRC's Technical Review of the TSPA-
LA it will determine whether DOE has reasonably complied with the Acceptance Criteria listed
below. The results of NRC's most recent analysis of the issue are also presented.
VI-132
-------
(1) The staff shall determine whether the DOE considered conceptual flow and data
uncertainties. Uncertainties due to sparse data in some areas or low confidence in the
data interpretations (e.g., LUC96; also CZA97) should have been considered by
analyzing reasonable conceptual flow alternatives supported by site data, or by
demonstrating through sensitivity studies that the uncertainties have little impact on
repository performance.
According to the NRC's analysis, the reference Luckey, et al. (LUC96) does an excellent job of
describing various conceptual models of site-scale hydrology as they were known at that time.
The staff will exercise professional judgment in determining whether DOE has reasonably
treated the conceptual and data uncertainties in performance assessments or has shown that they
will not adversely impact performance.
This issue is open pending review of DOE's VA.
(2) The staff shall determine whether, based on site data, the DOE has reasonably delineated
approximate flow paths from beneath the repository to potential receptor locations. Flow
paths should consider: (i) aquifers (volcanic, alluvium, and carbonate) and continuity of flow
regimes; (ii) flow domains (matrix and fracture); (iii) flow directions; (iv) flow velocities
(approximate Darcy fluxes and average linear velocities); and (v) vertical hydraulic gradients,
including the potential flow direction between the Paleozoic carbonate aquifer and the
volcanic aquifers. Hydraulic and tracer testing along paths to potential receptor locations
should be conducted on a scale large enough to include a statistically representative
elementary volume in alluvium and in the fracture network in tuffs: A sufficient number of
tests should be conducted to reasonably reduce the uncertainty in hydraulic and transport
properties of the units downgradient from the proposed repository, including approximate
delineation of the southerly zone where the water table transitions from tuffs to alluvium.
These values, along with existing data such as that from the C-wells complex (e.g., GEL97),
should be used in ground-water flux calculations and mathematical models.
According to the NRC's analysis, the lack of hydrologic data for alluvium is a data gap in the
DOE's site characterization of saturated zone hydrology. Emphasis should be placed on
reasonable determinations of heads, transmissivity, hydraulic conductivity, effective porosity,
and dispersion coefficients. The hydraulic and geochemical characteristics of the likely flowpath
VI-133
-------
that exists south of well JF-3 have not been evaluated. It is unknown at which locations the
water table transitions from fractured tuff to overlying alluvium. The saturated thicknesses,
hydraulic properties, and geochemical properties of alluvium have not been determined for the
region that lies between well JF-3 and the Amargosa Desert. The DOE's cooperative well
drilling program with Nye County, Nevada could accomplish this if the wells are sited and tested
to characterize the hydrology along likely flow paths in a timely manner.
The staff believe that the three-phase SZ testing strategy described in Reimus, et al. (REI97)
could, if implemented, significantly improve understanding of the hydrogeologic system. New
wells may be needed, but possible locations for such testing using existing wells would include
(1) J-12, JF-3, and J-13; (2) H-4, SD-12, and WT-2; or (3) SD-6 and H-5. Other combinations
are also possible, and other wells could be expected to respond to long-term pumping tests.
Because fractures and faults have preferred orientations, and can act as preferred flow pathways,
quantitative studies require that more than one representative elementary volume of rock be
sampled.
Based on the available potentiometric head data, flow from the proposed repository is likely to be
in a southeasterly direction (i.e., along the natural hydraulic gradient) toward Fortymile Wash.
This is the general direction of flow that was interpreted by panelists in a recent expert elicitation
on the site saturated zone (GE098), and is also the flow pattern that is best supported by
hydraulic head data. Southeasterly flow is the direction used in the NRC/CNWRA performance
assessment model where saturated zone flow and transport are simulated in a series of stream
tubes. Radionuclides reaching the saturated zone from the repository along this southeasterly
flow path would migrate along fracture-dominated pathways hi the tuff aquifers in the general
direction of well J-12, and thence, southward in saturated alluvium toward the Amargosa Desert.
The southeasterly flow path assumes that the fractured tuff aquifer is an equivalent porous
medium at the site scale under isotropic conditions. Treating the aquifer as an equivalent
medium at a large scale is supported by the pervasiveness in the tuffs of faults and fractures
oriented in many directions, and by results of long-term testing at the C-wells. The NRC staff
plans to continue its analysis of the previous and ongoing C-wells testing.
As noted above, ground-water flow in the tuff aquifer is dominated by structural features. This
causes anisotropic conditions where structures may act as high- and/or low-permeability zones,
and this is most evident at small spatial scales. At larger scales the hydrologic properties of
VI-134
-------
interconnected fault and fracture networks are expected to dominate flow conditions. Because of
uncertainties about large-scale anisotropy, current DOE simulations assume that -10-percent of
transport pathways never come into contact with saturated alluvium. Data from aquifer pumping
tests in the C-well complex are now being analyzed to determine whether large-scale anisotropic
effects are evident. There are presently no data concerning the isotropy of saturated alluvium.
The staffs current model is subject to revision as new site data are collected and analyzed. Due
to sparse data in some areas and uncertainties in the interpretation, the staff continues to analyze
whether there are other viable SZ conceptual flow models which can be supported by available
data. For example, the staff is examining whether there is evidence of site-scale aquifer
anisotropy that could shift SZ flow patterns significantly away from the direction of the observed
southeasterly natural hydraulic gradient.
A promising new approach exists that may greatly improve the isotopic dating of ground waters
with the 14C technique, leading to better estimates of average ground-water residence times.
Residence time is related to average regional ground-water velocity. Thomas (THO96) describes
the separation of dissolved organic carbon from ground water using reverse osmosis and
ultrafiltration methods. The staff believes that this method should be applied to samples
collected at Yucca Mountain to provide an independent estimate of the apparent ground-water
velocities in the system and the average time of travel from principal recharge zones to Yucca
Mountain. This technique has been applied to ground water in the vicinity of Devils Hole, and
indicates that ground-water residence times in the carbonate aquifer feeding Devils Hole are
about 2000-3000 years (WIN97), significantly less than earlier estimates.
Information about flow conditions in the Paleozoic carbonate aquifer beneath Yucca Mountain is
based on only one well, USW p#l. Heads in this well are about 22 m higher in the Paleozoic
carbonate aquifer and lower volcanic confining units than in the Crater Flat tuffs (lower volcanic
aquifer), indicating a strong upward gradient. Likewise, heads in the lower volcanic confining
units in wells H-l and H-3 are also higher than in the Crater Flat Tuffs, providing evidence that
significant upward hydraulic potentials probably exist over most of the site east of Solitario
Canyon. This condition is favorable for waste isolation because an upward gradient, if
maintained in the future, would protect the deep Paleozoic carbonate aquifers from
contamination. The DOE's cooperative drilling program with Nye County, NV, should provide
timely additional data regarding the vertical gradients between the Paleozoic carbonate aquifers
VI-135
-------
and overlying tuffs or alluvium. It should also be noted that large differences in ground-water
chemistry between the carbonate aquifer system and the Crater Flat tuffs suggest that upward
fluxes are relatively small compared to those introduced by lateral flow within the tuffs.
The issue remains open pending review of the DOE submittals (e.g., VA, data for alluvium and
tuffs) and staff analysis of effects of large-scale anisotropy. The staff will determine what
adjustments, if any, to general flow paths are warranted.
(3) The staff should determine whether the DOE has provided a hydrologic assessment to
describe likely causes of the "moderate hydraulic gradient" and the "large hydraulic
gradient."
According to the NRC's analysis, at, or west of, Yucca Mountain is a zone of relatively low
permeability that tends to restrict flow from west to east. Based on current understanding, the
SCF and associated splays are the most likely cause of the 45-m head change known as the
"moderate hydraulic gradient." There is evidence that ground water crosses the fault, but actual
fluxes are not known. The tendency to restrict flow probably decreases toward the north as fault
displacement decreases. The fault displacement reaches a minimum at a hinge point, about one
km southwest of well G-2. When completed, well SD-6 located at the crest of Yucca Mountain
should be used to conduct pumping tests beneath the western repository block near the Solitario
Canyon fault, and to obtain estimates of transmissive properties beneath that part of Yucca
Mountain. Hydraulic testing at SD-6 should provide new insights about the nature of the so-
called "moderate hydraulic gradient."
Well WT-24 is currently being drilled to improve the DOE's understanding of the so-called
large-hydraulic gradient. The NRC believes that a sufficient understanding can be obtained
through the drilling and testing of WT-24. Preliminary data show that a perched zone is present
near the top of the Calico Hills in this well, and that the regional potentiometric surface is also
more than 100 m higher than in wells immediately to the south.
This issue remains open pending submittal and staff review of the DOE reports on the drilling
and testing of wells WT-24 and SD-6. Preliminary water-level elevations have been reported
(WT-24: 839.5 m; SD-6: 731.5). The data remain preliminary because the wells are still being
VI-136
-------
constructed, and staff await formal reports from the DOE on testing and data collection at these
wells.
(1) The staff shall determine whether the DOE has provided maps of approximate
potentiometric contours for an area that, at a minimum, includes wells J-l 1 on the east,
VH-1, VH-2, and the GEXA Well on the west, UE-29a #2 to the north, and domestic and
irrigation wells south of Amargosa Valley (Lathrop Wells). Maps of regional and site-
scale recharge and discharge should be provided, along with site-scale hydrostratigraphic
cross sections constructed along the paths to the accessible environment, and flow-net
analysis of the site-scale SZ.
The NRC has not completed its analysis of this issue which remains open pending review of
DOE's TSPA-VA.
(1) The staff shall determine whether the DOE has characterized key hydrologic parameters
in the form of either probability distributions or deterministic bounding values. These
parameters include transmissivity, hydraulic gradient, porosity (effective, matrix, and
fracture), and effective aquifer thickness. The DOE's parameters should be reasonably
consistent with site data.
Based on the NRC's analysis, the DOE is apparently using probability distributions to represent
key hydrologic parameters in TSPA, an approach that is acceptable to the staff. Staff will review
DOE submittals to determine whether the parameters used are reasonably consistent with site
data.
The issue remains open pending the NRC's review of DOE's VA.
(1) The staff shall determine whether the DOE has used mathematical ground-water model(s)
that incorporate site-specific climatic and subsurface information. Sufficient evidence
must be presented to show that the models were reasonably calibrated and that the
physical system is reasonably represented. The fitted aquifer parameters should compare
reasonably well with observed site data. Implicitly- or explicitly-simulated fracturing and
faulting should be consistent with the data in the 3-D geologic model. Abstractions
should be based on initial and boundary conditions consistent with site-scale modeling
VI-137
-------
(e.g., CZA97) and the regional models of the Death Valley ground-water flow system
(e.g., DAG97a,b). Abstractions of the ground-water models for use in PA simulations
should use appropriate spatial- and temporal-averaging techniques.
The NRC has not yet examined this issue in depth and it remains open pending review of DOE's
VA.
(1) It will be acceptable for the DOE to conservatively assume no wellbore dilution at a
receptor location. If wellbore dilution is used, a demonstration should be provided that
reasonable assumptions have been made about well design, aquifer characteristics, plume
geometry, withdrawal rates, and capture zone analysis for the receptor location.
The NRC has determined that currently the DOE is taking no explicit credit for wellbore
dilution. This is acceptable to the staff, but is inconsistent with a DOE-sponsored expert
elicitation (GEO98) which concluded that significant dilution can be expected through well
pumping. If the DOE takes credit for wellbore dilution in future submittals, the staff will
evaluate the information to determine if the acceptance criterion has been met.
This issue is resolved and the staff have no further questions at this time.
(1) It will be acceptable for the DOE to conservatively assume no dilution due to dispersion,
or no ground-water mixing below the repository footprint, and no mixing of the Yucca
Mountain water with water from the north in Fortymile Wash. If intra-basin mixing of
ground water is used, a demonstration should be provided that reasonable assumptions
have been made about spatial and temporal variations of aquifer properties and ground-
water volumetric fluxes. If dilution is simulated as dispersion in a numerical transport
model, scale-dependent dispersivities, constrained by the analysis in Gelhar, et al.
(GEL96), should be used.
The NRC notes that in the recent peer review of DOE's TSPA, panelists observed that the
saturated zone model used in the TSPA-VA is likely to result in a non-conservative estimate of
dilution due to mixing along the flow path. The model assumes that radionuclides reaching the
water table and transported in the ground water would be subjected to widespread and uniform
mixing in all of the stream tubes within the flow tube model. If only a small percentage of the
waste packages fail, and if the failures are confined to a small area of the repository, which is
probably one of the more likely scenarios, the radionuclides will more likely be confined to
VI-138
-------
specific stream tubes and not uniformly mix or spread over all of the stream tubes in the flow
tube model. Therefore, the presumed widespread and uniform mixing in the flow tube model is
not conservative because it would result in more dilution due to mixing in the flow tube than
would actually take place.
For 20-km flowpaths, the DOE appears to be using dilution factors that range from 1-100, with a
median value of 12. These estimates were derived from the conclusions of three members of a
five-member expert panel (GEO98), and consider dispersion effects. The other two panel
members did not estimate the dilution range. The estimates do not include the additional effects
of dilution within wellbores or intrabasin mixing. The range and median appear to be
conservative because they are reasonably low. The staff will assess the DOE's treatment of
dilution in the VA.
(1) The staff shall determine whether the DOE has incorporated key conclusions regarding
potential geothermal and seismic effects on the ambient SZ flow system (e.g., NRC92,
NWT98).
The NRC's analysis and proposed resolution of this issue remains open pending review of the
DOE's VA.
(1) It will be acceptable for the DOE to use estimates and recommendations provided by
expert elicitations (e.g., GEO98) as long as the expert elicitation is conducted and
documented using the guidance in the Branch Technical Position on Expert Elicitation
(NRC96) or other acceptable approaches.
The NRC has concluded that the expert elicitation on saturated zone flow and transport (i.e.,
GEO98) was conducted and documented in an acceptable way. Consequently, this issue is
resolved and the staff have no further questions at this time.
(1) The staff shall determine whether the collection, documentation, and development of
data, models, and computer codes have been performed under acceptable QA procedures.
If they were not subject to an acceptable QA procedure, they have been appropriately
qualified.
The NRC's analysis and proposed resolution of this issue are yet to be determined.
VI-139
-------
Matrix Diffusion in Saturated and Unsaturated Zones
Relevance to PA. Matrix diffusion is related to two of the key elements of the natural
subsystems: (1) distribution of mass flux between fracture and matrix; and (2) retardation in
water-production zones and alluvium. At Yucca Mountain, the process of matrix diffusion may
impact repository performance because ground-water flow, away from the repository, occurs
primarily in fractures that account for only a small fraction of total formation porosity. In such
hydrologic systems, matrix diffusion can attenuate migration of radionuclides in two ways: (1) it
can spread them physically from the flowing fractures into stagnant matrix pore water; and (2)
rock matrix can provide a vast increase in mineral surface available for geochemical surface
reactions (e.g., sorption) as compared to fracture surfaces alone. The extent to which matrix
diffusion can affect repository performance is controlled by the rate of solute diffusion from
fractures into rock matrix relative to the time scale for flow through the fracture system to the
receptor point. When diffusion is very slow relative to the transport time, the impact is negligible
in terms of solute arrival time, but there is a slight long-term attenuation of peak solute
concentration. If diffusion is fast relative to transport time, the impact is a significant delay in
solute arrival at the receptor point. At intermediate diffusion rates, the impact is a modest delay
in initial solute arrival tune with significant attenuation of solute concentration.
The Repository Safety Strategy (DOE98) noted that concentrations of radionuclides in ground
water can be reduced by matrix diffusion and sorption. If matrix diffusion is limited there can
still be sorption on fracture walls, but the depletion effect will be much smaller.
NRG Acceptance Criteria and Resolution Status. In the NRC's Technical Review of the TSPA-
LA it will determine whether the DOE has reasonably complied with the Acceptance Criteria
listed below. The results of the NRC's most recent analysis of the issue are also presented.
(1) It will be acceptable for the DOE to conservatively assume no credit for matrix diffusion
in the UZ. If credit is taken, then matrix diffusion predictions are consistent with
evidence for limited matrix diffusion in the UZ including: (i) geochemical data (e.g.,
YAN96) that provide evidence of geochemical disequilibrium between matrix and
fracture waters hi the UZ at Yucca Mountain; and (ii) 36C1 evidence for rapid transport
pathways to the repository horizon (e.g., FAB96).
VM40
-------
According to the NRC, the UZ radionuclide transport sub-model that is currently used in the
DOE TSPA model is described by Robinson, et al. (ROB97). From the model description it is
evident that effective diffusion coefficients are selected a priori and not correlated within the
model to fracture or matrix saturation. Although it is possible to simply reduce the value of the
selected diffusion coefficient to be consistent with reduced matrix and fracture saturation, there is
no analysis provided to show that selected diffusion coefficients are appropriate for the
conditions modeled. Additionally, no analyses of the resultant geochemical differences between
matrix and fracture water are provided. Thus it is not possible at this tune to assess whether the
DOE method of abstracting matrix diffusion into UZ radionuclide transport is suitable for
predictions of repository performance.
Available information for Yucca Mountain indicates that fracture-matrix interface area is limited
to the wetted surface area within fractures. Similarly, effective diffusion coefficients in the UZ
are saturation-dependent and should be proportional to the effective saturated cross-sectional area
through which solutes can diffuse. Furthermore, TSPA model predictions should be consistent
with UZ geochemical data (e.g., YAN96), which suggest that waters within rock matrix at Yucca
Mountain have different geochemical signatures than fracture waters; predictions should also be
consistent with 36C1 evidence (e.g., FAB96) for rapid transport pathways to the repository
horizon.
The DOE should clearly document the technical basis for assumptions used to estimate the
transfer term for fracture-matrix exchange in the dual permeability model for UZ transport. The
staff has concerns that the residence time transfer function for the dual continuum model is
overestimated, because assuming an immobile reservoir neglects the transfer function accounting
for particles moving from the matrix to the fracture.
The issue remains open pending review of DOE's VA.
(1) It is acceptable for the DOE to conservatively assume that no matrix diffusion will occur
in the SZ (i.e., that all solutes will remain in fractures) during transport through saturated
fractured rock aquifers. DOE's inclusion of matrix diffusion in SZ transport models for
. Yucca Mountain should be reasonably supported by both field and laboratory
observations. Acceptable field and lab observations include tracer tests that are
conducted over different distance scales and flow rates with multiple tracers of different
VI-141
-------
diffusive properties. Transport models should reasonably match the results of the field
tracer tests. Rock matrix and solute properties used to justify the inclusion of matrix
diffusion in TSPA models fall within a range that can be supported by laboratory data.
The staff believes there is much greater potential for radionuclide retardation in saturated alluvial
deposits than in fractured tuffs. From the proposed repository to potential receptors at a distance
of 20 km, flowpaths will probably include significant amounts of saturated alluvium. Matrix
diffusion in the Tertiary tuffs would then be of minor significance.
The DOE's current assumptions about matrix diffusion are supported to some extent by field and
laboratory results to date. However, the amount of matrix diffusion claimed by the DOE from
these results has been disputed by the staff and others. A clearer demonstration of the matrix
diffusion phenomenon can be made by (1) using tracers with more variation in physical
properties and by testing over different length scales and flow rates, and (2) by demonstrating the
degree to which matrix diffusion can actually occur within fractured welded and moderately
welded tuffs that are known to act as significant flow zones at Yucca Mountain. Zones that
contribute to flow in wells have been identified through borehole logging and hydrologic testing.
Samples from those zones should be subjected to visual testing techniques of the type described
by researchers like Tidwell, et al. (TID97).
The DOE is assuming that matrix diffusion in the saturated zone occurs over a range of porosity
described by a log-triangular distribution that ranges from 0.0001 to 0.2, with a mean of 0.02
(2%). The DOE is simplifying the treatment of matrix diffusion by assuming that the range and
mean are equivalent to that assumed for effective (advective) porosity and appear to be
reasonably conservative with respect to field and lab tests of matrix diffusion and porosity.
However, the DOE's basis for selecting of a log-triangular distribution needs to be clarified, and
will be examined in the staffs review of the VA.
In the DOE's evaluation of C-wells tracer tests, the staff is concerned that use of the 50% relative
solute concentration arrival times to derive the range of effective porosities has an inherent non-
conservatism that could be avoided by basing the effective porosity approach on relatively early
solute arrival times (e.g., about 10% relative solute concentration). Finally, it appears that,
although the matrix diffusion behavior is different for each solute, the DOE is using the same
effective porosity for all solutes. If the DOE intends to neglect this variation in solute behavior,
VI-142
-------
the NRC recommends using the effective porosity derived from the least diffusive solute likely
to influence performance.
This issue remains open pending review of DOE's VA.
(1) If used, DOE's expert elicitations should be conducted and documented using the
guidance in the Branch Technical Position on Expert Elicitation (NRC96), or other
acceptable approaches.
The NRC's analysis and resolution of this issue are yet to be determined.
(1) Staff shall determine whether the collection, documentation, and development of data,
models, and computer codes have been performed under acceptable QA procedures. If
they were not subject to an acceptable QA procedure, they have been appropriately
qualified.
The NRC's analysis and resolution of this issue are yet to be determined.
Radionuclide Transport through Porous Rock
Relevance to PA. When radionuclides pass through porous rock, the interactions between the
dissolved radionuclides and the rock surfaces (e.g., sorption) result in retardation of the velocity
of the radionuclides relative to the velocity of ground water. The large surface areas of the
porous media tend to enhance sorption and consequently retardation. Furthermore, for those
radionuclides whose sorption reactions may be kinetically inhibited, the slower average linear
velocities of ground-water flow in porous media promote the solid-liquid interaction. If the
radionuclides exist instead as particulates or as colloids, they may be filtered out as ground water
flows through the constricted pores of the matrix. Sorption of radionuclides on solids and
filtration of radiocolloids and particulates in the matrix reduces the radionuclide concentration in
the liquid. However, the low permeabilities of the matrix of some hydrostratigraphic units at
Yucca Mountain may make some rock inaccessible to radionuclide-contaminated water on the
timeframe of repository performance.
VI-143
-------
NRC Acceptance Criteria. The approach that recent performance assessment efforts have used to
simulate radionuclide transport in porous rock involves first establishing a ground-water flow
field. This flow field is generated as a result of hydrologic modeling of the Yucca Mountain
system using site-specific parameters, and can be one-, two-, or three-dimensional, depending on
the purpose of the modeling effort and the available data. The flow field representing the spatial
distribution of ground-water velocities is then adjusted by dividing the velocity vectors by the
retardation factor, Rf, for each radionuclide, to yield the radionuclide velocity fields. Current
approaches model the ground-water flow field as a dual continuum representing both fracture and
matrix flow at every point in the system.
The following Acceptance Criteria apply to evaluating the DOE estimates and consideration of
radionuclide transport through porous rock:
(1) For the estimation of radionuclide transport through porous rock, the DOE has
a. Determined, through performance assessment calculations, whether radionuclide
attenuation processes such as sorption, precipitation, radioactive decay, and colloidal
filtration are important to performance
b. (i) Assumed Kd is zero and radionuclides travel at the rate of ground-water flow, if it
has been found that radionuclide attenuation is unimportant to performance (in which
case, Acceptance Criteria 2 and 3 do not have to be met) or, (ii) demonstrated that
Criterion 2 or 3 has been met, if radionuclide attenuation in porous rock is important
to performance.
(2) For the valid application of the K<, approach, using the equation Rf = 1 + pKd/n, the DOE
has
a. Demonstrated that the flow path acts as an isotropic homogeneous porous medium.
b. Demonstrated that appropriate values for the parameters, K,,, n and p have been
adequately considered (e.g., experimentally determined or measured)
c. Demonstrated that the following assumptions (i.e., linear isotherm, fast reversible
sorption reaction, and constant bulk chemistry) are valid.
VI-144
-------
(3) For the valid application of process models such as surface complexation, ion exchange,
precipitation/dissolution, and processes involving colloidal material, the DOE has
a. Demonstrated that the flow path acts as an isotropic homogeneous porous medium.
b. Demonstrated that values for the parameters used in process models are appropriate
c. (i) Demonstrated that the three implicit assumptions (see 2.c.) are valid, if process
models are intended to yield a constant Kd for use in the retardation equation; or (ii)
determined transport in a fully coupled dynamic system (e.g., PHREEQC,
MULTIFLO, HYDROGEOCHEM, etc.)
(4) Where data are not reasonably or practicably obtained, expert judgement has been used
and expert elicitation procedures have been adequately documented. If used, expert
elicitations were conducted and documented in accordance with the guidance in NUREG-
1563 (NRC96) or other acceptable approaches.
(5) Data and models have been collected, developed, and documented under acceptable
Quality Assurance (QA) procedures or if data were not collected under an established QA
program, they have been qualified under appropriate QA procedures.
Status of Issue Resolution at the NRC Staff Level Most of the Yucca Mountain geochemical
work in the past twenty years has been directed toward determining the retardation of
radionuclides in porous rock. Significant progress has been made to address this issue that is
important to waste isolation and repository performance. However, in that time, there have been
major changes hi the conceptualization of the geologic setting of the repository that impact the
relative importance of this issue and the consideration of the point of compliance up to 20
kilometers away from the repository. The greater average infiltration results in a greater
proportion of the flux bypassing the sorptive porous rock by flow in fractures. A 20 km point of
compliance would result in the need to consider the alluvium along with porous and fractured
rock. These major changes reduce the relative importance of radionuclide transport in porous
rock on performance assessment.
The NRC considers that the subissue has been met for certain radionuclides but not for others.
Some of the radionuclides for which the issue has not been resolved on the staff level may be
important to performance.
VI-145
-------
The NRC finds that the approach adopted by LANL to determine minimum K,, values is logical
and defensible. By performing batch sorption tests using site-specific materials, followed by
confirmatory tests to establish the validity of the assumptions needed for the constant Kj
approach, and then selecting the minimum Kj from all the tests, an acceptable value can be
obtained.
In summary, the NRC chose three radionuclides as examples to highlight successes and areas
needing further work. They are neptunium, plutonium, and uranium. The minimum Kd approach
has worked well for neptunium. The staff recognizes that multiple tests have been performed to
establish reasonable Kj values for this radionuclide. Consequently, this issue is being resolved
for neptunium. On the other hand, although both batch sorption tests and flow-through column
tests have been performed to determine a minimum IQ for plutonium, significant inconsistencies
occurred. The NRC staff recognizes plutonium as problematic and encourages further work to
establish defensible K,, values. For uranium, geochemical modeling suggests that a uranyl
silicate phase, soddyite, could precipitate from solution, given the initial ground-water
composition. Eliminating the possibility that processes other than sorption (e.g., precipitation)
may be contributing to the removal of a radionuclide from solution is necessary for establishing a
valid Kd. On the other hand, the thermodynamic modeling could be in error based on parameter
uncertainties. To date, it does not appear that flow-through column tests were performed with
uranium. Consequently, the NRC does not believe that this issue has been resolved.
Radionuclide Transport Through Alluvium
Relevance to PA. Current conceptual models of the alluvium incorporated in performance
assessments reflect limited information concerning the physical and chemical conditions of
alluvium. For example, in the NRC's TPA 3.1 (NRC98), the alluvium is assumed to be crushed
tuff similar to the material used in batch sorption experiments. The DOE model abstraction
assumes flow as in a sand column driven by the hydraulic gradient. Furthermore, in the TSPA-
VA the DOE assumes there are no preferential pathways in the alluvium. This assumption has
not yet been tested. However, the occurrence of cut and fill structures formed in the alluvium by
braided streams as evident in the walls of Forty Mile Canyon may suggest that preferred
pathways exist in the alluvium with the potential to reduce mixing and dilution.
VI-146
-------
NRC Acceptance Criteria. The DOE considers radionuclide transport a key performance
attribute of the natural barrier system in the proposed repository. Retardation of radionuclides
through alluvium constitutes a key element of the DOE performance assessment. The NRC
requires that the DOE must adequately estimate the transport characteristics of the Yucca
Mountain site and appropriately consider radionuclide transport in their assessments of
repository performance. The NRC's review process is designed to determine which transport
processes have been addressed/assumed by DOE. The review will first identify, whether or not
the selected retardation processes are appropriate to the Yucca Mountain system, and second,
whether or not they are addressed adequately for those radionuclides of concern.
The following Acceptance Criteria, which are the same as those for radionuclide transport
through porous rock, apply to evaluating the DOE estimates and consideration of radionuclide
transport through the alluvium:
For the estimation of radionuclide transport through alluvium, the DOE has
a. Determined, through performance assessment calculations, whether radionuclide
attenuation processes such as sorption, precipitation, radioactive decay, and colloidal
filtration are important to performance
b. (i) Assumed K^ is zero and radionuclides travel at the rate of ground-water flow, if it
has been found that radionuclide attenuation is unimportant to performance, hi which
case, Acceptance Criteria 2 and 3 do not have to be met; or, (ii) demonstrated that
Criterion 2 or 3 has been met, if radionuclide attenuation in alluvium is important to
performance.
For the valid application of the Kj approach, using the equation Rf = 1 + pK^n, DOE has
a. Demonstrated that the flow path acts as an isotropic homogeneous porous medium
b. Demonstrated that appropriate values for the parameters, K<,, n and p have been
adequately considered (e.g., experimentally determined or measured)
c. Demonstrated that the three implicit assumptions (i.e., linear isotherm, fast reversible
sorption reaction, and constant bulk chemistry) are valid.
VI-147
-------
For the valid application of process models such as surface complexation, ion exchange,
precipitation/dissolution, and processes involving colloidal material, the DOE has
a. Demonstrated that the flow path acts as an isotropic homogeneous porous medium
b. Demonstrated that appropriate values are used in processes models
c. Demonstrated that the three implicit assumptions (as in 2.c.) are valid, if process
models are intended to yield a constant Kd for use in the retardation equation;
otherwise, determined transport in a fully coupled dynamic system (e.g., PHREEQC,
MULTIFLO, HYDROGEOCHEM, etc.)
Where data are not reasonably or practicably obtained, expert judgement has been used and
expert elicitation procedures have been adequately documented. If used, expert elicitations were
conducted and documented in accordance with the guidance in NUREG-1563 (NRC96) or other
acceptable approaches.
Data and models have been collected, developed, and documented under acceptable QA
procedures, or if data were not collected under an established QA program, they have been
qualified under appropriate QA procedures.
Status of Issue Resolution at the NRC Staff Level. The status of this issue is tied closely to that
of the previous issue. However, additional uncertainty is a result of the very limited information
collected to date on the mineralogy, ground water, chemistry, and physical flow systems of the
alluvium. Past efforts have focused on characterizing the geologic media within 5 kilometers of
the repository because of the provisions of the then applicable 40 CFR Part 191. With the
resultant increase in the length of the flowpath to the biosphere to 20 kilometers, now being
understood as consistent with draft 10 CFR Part 63, a significant portion of relatively
uncharacterized geologic media has been added to the system.
Although, like the DOE, the NRC has assumed in earlier modeling that the alluvium acts as a
homogeneous porous medium, the NRC also recognizes that little or no information is available
to support that assumption. Furthermore, it is recognized by the NRC that treating the alluvium
as a homogeneous porous medium may be nonconservative.
VI-148
-------
The NRC expects that the series of boreholes to be drilled by Nye County in the alluvium will
provide significant information concerning its geologic, and hydrologic characteristics. It is
expected that the mineralogy will reflect that used in batch sorption experiments for determining
sorption coefficients for radionuclides in tuff. If that is so, the NRC believes that the laboratory
work needed to address previous issue (i.e., retardation in porous rock) will also address the
retardation issues in the alluvium.
The NRC notes that the DOE will need to defend their conceptualization of the alluvium. If the
alluvium is a composite of cut and fill structures resulting from the accretion of braided streams,
preferred pathways limiting water-rock interaction may result. If on the other hand, the alluvium
is homogeneous, the application of experimentally determined K,jS to calculate retardation factors
would be appropriate. The NRC indicates that resolution of this issue will await the geologic and
hydrologic information to be collected.
Radionuclide Transport through Fractured Rock
Relevance to PA. Recent site characterization activities involving the radioisotopes 36C1 and 3H
provide evidence suggesting fast pathways of ground-water flow through the unsaturated zone
(FAB96). These fast pathways are proposed to occur as a result of flow down faults and
fractures. Also, responses from adjacent wells in large-scale hydrologic pump tests (C-Wells)
suggest that preferential pathways may exist in the saturated zone at Yucca Mountain (GEL97).
If preferential pathways exist from the repository to the critical group, performance may be
adversely affected, because portions of the geologic barrier would be bypassed.
In predicting flow and transport through the unsaturated zone, the TSPA-VA takes no credit for
retardation of radionuclides in fractures. The rationale for assigning no retardation in the
fractures is based on the hypothesis that there is limited capacity for sorption along fractures, and
average linear velocities in fractured rock are high, limiting time for interaction between the
dissolved radionuclides and the sorbing minerals lining the fracture walls. However, the
presence of specific fracture-lining minerals may provide significant opportunity for sorption of
specific radionuclides. For example, the manganese oxyhydroxides may strongly sorb
plutonium, uranium, and americium; calcite may strongly sorb or coprecipitate neptunium
(TRI96).
VI-149
-------
Since the TSPA-VA does not explicitly incorporate fractures in predicting flow and transport
through the saturated zone the potential sorption effects of the fractures and matrix are, in
essence, lumped into a single value. The DOE addresses the associated uncertainty by assigning
a range, mean and distribution to the sorption values (i.e., Kj's) for each of the hydrologic units.
NRC Acceptance Criteria. The DOE considers radionuclide transport a key performance
attribute of the natural barrier system in the proposed repository. Retardation of radionuclides hi
fractures in the unsaturated zone and in the saturated zone constitutes a key NRC consideration
for evaluating the DOE's performance assessment. The NRC notes that the DOE must
adequately estimate the transport characteristics of the Yucca Mountain site and appropriately
consider radionuclide transport in its assessments of repository performance. The NRC's review
process is designed to determine which transport processes have been addressed/assumed by the
DOE. The NRC will first identify whether or not the selected retardation processes are
appropriate to the Yucca Mountain system, and secondly, whether or not they are addressed
adequately for those radionuclides of concern.
The NRC indicates that Acceptance Criteria for evaluating the DOE estimates and consideration
of radionuclide transport through fractured rock will be developed in FY99.
Status of Tssue Resolution at the NRC Staff Level. The NRC has yet to develop the Acceptance
Criteria for this issue. The DOE has performed some experiments using fractured rock. Whereas
the retardation factor in fractures is assumed to be 1 (i.e., no sorption) in performance
assessments, due to the uncertainty with regard to radionuclide transport hi fractured rock,
preliminary experiments suggest that some retardation occurs. For example, neptunium
experiments have been performed and show reduced recovery and a delay in the breakthrough
relative to tritium and technetium.
Criticality in the Far Field
Relevance to PA. The Total System Performance Assessment-Viability Assessment Methods
and Assumptions Report (TRW97) states nuclear criticality scenarios will be evaluated. There
are some scenarios that could lead to criticality that may affect performance. For example,
Bates, et al, (BAT92) found that plutonium released from glass waste form exists predominantly
as colloids. If colloidal plutonium could be efficiently filtered hi nonwelded bedded units below
VI-150
-------
fractured strata of the repository horizon, it could accumulate sufficient mass for criticality.
Consideration of neutron sorbers or poisons, either contributed from the natural system or from
the repository may be important. Differences in the mobility may lead to chromatographic
separations of fissile material and poisons. A criticality occurring over a long tune could
produce increasing amounts of fission products and neptunium. Some of the radionuclides
generated hi a criticality event could be relatively mobile and, thus, could adversely affect
performance. Furthermore, criticality could: (1) generate additional colloids, capable of
transporting radionuclides unretarded; (2) affect the ground-water flow field; or (3) result in
gaseous release of volatile radionuclides.
NRC Acceptance Criteria: To be determined.
Status of Issue Resolution of the NRC StaffLevel- To be determined.
NWTRB IdentifiedNeeds
In their 1999 Report to Congress the Nuclear Waste Technical Review Board makes a number of
observations regarding the potential viability of the proposed repository. The following
discussion pertains to issues related to ground-water flow and contaminant transport in the
saturated zone. As a general consideration, the Board indicates that after reviewing the TSPA-
VA is has not identified any features or processes that would automatically disqualify the site.
The Board also notes that the TSPA-VA relies heavily in some cases on formal elicitation of
expert opinion. The Board maintains that this was necessary and extremely useful, given the
lack of field and laboratory data in certain areas and the equivocal nature of some of the data in
other areas. However, the Board expresses a concern that expert opinion should not be used as a
substitute for data that can be obtained directly from the site, laboratory and other investigations.
The Board also concludes that a significant amount of additional scientific and engineering work
will be needed to increase confidence in a site-suitability determination and license application.
The Board's expressed specific concerns with respect to the data needs of the near field (e.g.,
waste package) environment. However, with respect to ground-water flow and contaminant
transport hi the saturated zone the Board issued a general statement indicating that long-term
studies of the natural barriers also will be needed, primarily to verify projections of water
movement within the unsaturated and saturated zones near the repository.
VI-151
-------
The Board also expressed agreement with a DOE-commissioned peer review panel that two types
of additional data are needed to improve the credibility of the total system performance
assessment part of the TSPA-VA: (1) fundamental data that are essential to the development and
implementation of the models and (2) data sets designed to challenge conceptual models and test
the coupled models used in the TSPA-VA.
Peer Review Panel Identified Needs
The Total System Performance Assessment Peer Review Panel raised a number of concerns
related to ground-water flow and radionuclide transport in the saturated zone. A discussion of
their findings are presented below.
In their introductory remarks to this section the Panel states The current treatment of saturated
zone (SZ)flow and transport at Yucca Mountain is far from satisfactory. In part, this may reflect
a higher level of interest and activity in UZ processes during the earlier stages of the project.
This, in turn, may have resulted in less progress in SZ activities. Admittedly, the SZ encompasses
a much larger volume of the mountain than the UZ. Although it does not involve the
complexities of the UZ, it represents a much larger problem for site characterization, flow, and
transport. The Panel identified three main areas where they believe important weaknesses are
present in the current treatment:
• The lack of data for some important parameters;
• The incomplete nature or site characterization; and
• Continuing questions regarding the adequacy of the numerical models.
The Panel indicates that the first two areas of weaknesses have forced the DOE to rely primarily
on estimates of the expert panel that participated in the Saturated Zone Flow And Transport
Elicitation Project (GEO98), for guidance on selecting values for key parameters, including
dilution and retardation. As a result of comments and recommendations provided by these
experts, "Saturated Zone Flow and Transport Preliminary Draft Chapter 2.9 of TSPA-VA"
(CRW98), published on February 13,1998, has been replaced by a revised interpretation of the
SZ flow and transport process. However, it is the opinion of the Panel that inherent problems
remain. The Panel believes that additional work on this critical subject is needed. The Panel
offers the following specific comments.
VI-152
-------
Lack of Field Data
The Panel indicates that the lack of field data presents a major difficulty. There is a broad area
along the projected SZ flow path from Fortymile Wash to the Armagosa Valley, 10 km or more
in length, in which no boreholes have been drilled (A number of the Panel's concerns may be
alleviated by Nye County's well drilling/testing program as presented in the following section.)
The Panel maintains that, for this region, there is a resulting absence of data on key subjects such
as: subsurface geology, water table configuration, hydraulic parameters, etc. In other words, the
Panel is concerned that the characterization of the SZ flow path over about one half of its 20 km
length is currently not complete. In addition, the Panel notes that there is an apparent difficulty
in estimating vertical flow in the SZ, the location of the lower boundary, and the lack of account
for anisotropy and heterogeneity. Furthermore, the Panel references a more detailed discussion
of the serious uncertainties resulting from this lack of data is presented in a report submitted to
the U.S. Nuclear Waste Technical Review Board (GEL98).
The Panel also explains that the difficulty in evaluating the effects of retardation on radionuclide
transport, which is needed in determining dose rate, is another inherent problem. There are two
critical aspects to this problem: (1) the division of flow between the matrix and fractures in the
SZ zone, and (2) the magnitude of the Kj values to be used.
According to the Saturated Zone Expert Elicitation Panel, ground-water flow over the 20-km
path from the repository site occurs mostly in the volcanic units and alluvium, and flow occurs in
only 10% to 20% of the fractures. As indicated above, field data are needed to verify this picture
of the SZ zone. Because Kd values in the matrix (especially for Np) can be 10 to 100 times
higher than Kj values hi the fractures, it is necessary to know what percentage of the
radionuclides are in the matrix of the volcanics. Finally, Gelhar (GEL98) a member of the Peer
Review Panel has also indicated that Kj values cannot be used without knowing how
representative they are of field conditions.
Incomplete Characterization of the Site
The Panel believes that characterization of the site remains incomplete. In the TSPA-VA, SZ
site characterization affects primarily the description of the flow streamtubes, though the
estimation of the permeability field and the water fluxes (in both SZ and UZ). The current
VI-153
-------
approach for estimating the permeability field is based on the calibration of pressure heads. The
Panel also points out that, in addition to the problem of lack of data over a substantial region of
the SZ as mentioned above, it is known (as acknowledged by the DOE hi Chapter 8 of the
Technical Basis Document) that pressure data inversion does not guarantee uniqueness hi
parameter estimates. Thus, potential fast paths hi the SZ (such as permeability channels) may be
underestimated. The Panel maintains that the implications of such a possibility on the transport
of radionuclides are significant and cannot be dismissed.
In the same context, the Panel also raises an issue of numerical resolution in the modeling of
regional flow, where only 3 vertical layers (spanning 2,750 m) are used to represent the large-
scale hydrology and a typical grid has a linear (horizontal) size of the order of 1500 m. With
such limited resolution, the Panel believes that the intra-grid heterogeneity is seriously
misrepresented. Their same concern also applies to the site-scale model, which involves a grid
resolution of 200 m. The Panel indicates that, given the large range in permeabilities, which
spans 7 orders of magnitude, this limited resolution raises the issue of the relevance of numerical
predictions regarding the postulated flow fields.
The assumed water fluxes in the SZ and UZ, their variation with different climates, and the
recharge from the ground surface downgradient from the repository will also affect the
description of the streamtubes. The Panel asserts that the uncertainties pertaining to the
characterization of the site, the postulated flux multipliers for the future climates (four for the
Long-Term Average and six for the Super Pluvial) are also uncertain. Streamtubes are assumed
not to vary with tune, regardless of the changes in climate, which is assumed to affect only the
volume flux through them. The Panel indicates that this assumption is not consistent with the
change in the ratio of the water flux though the UZ and the SZ zones, as shown in Table 3.21,
Volume 3, of the TSPA-VA. The Panel further points out that instead of being constant, as
required by the assumption of a constant streamtube, this ratio is shown to increase more than
twofold as the climate changes from present day to super pluvial conditions.
In the current analysis, it is assumed that recharge along Fortymile Wash enters the ground water
to the east of the plume, but it does not enter on top of the contaminated water. Recharge on top
of the projected flow path would alter the streamlines significantly, resulting in a substantial
layer of clean water above the contaminated water. In his report to the NWTRB, Gelhar
(GEL98) suggests that such a layer could be 100 to 150 meters thick. The Panel believes that
VI-154
-------
this potentially conservative feature would call into doubt the basic biosphere model, in which a
farm family is assumed to pump contaminated water from the plume.
Streamtube Approach
In response to the criticism raised by the Expert Elicitation Panel on the SZ Flow and Transport
(GEO98), the DOE drastically revised the model of contaminant transport in the SZ in favor of a
new formulation based on flow streamtubes. While the Peer Review Panel believes that the
streamtube approach is better than the previous coarse-grid numerical models (200m x 200m, x
20m), it also believes that several issues need to be resolved.
The modeling of dispersion and dilution is treated quite empirically, using overall estimates of
dilution, provided by the Saturated Zone Flow and Transport Expert Panel. Since the DOE has
similarly provided overall dilution factors instead of a more detailed analysis, the net result is
that the interaction of plumes containing different radionuclide concentrations is also treated
inadequately, in a generally ad hoc manner. The Panel believes that a numerical approach based
on a streamtube formalism, well-resolved near the plume and with a correct representation of
dispersion and retardation, is feasible (provided that a good description of the heterogeneity from
field data is available). Development of such an approach would permit sensitivity studies to be
conducted or the effects various factors, including geostatistics, and would circumvent the
necessity to rely solely on estimates from an expert panel and/or empirical corrections. At the
same time, the Peer Review Panel raises a concern as to why the modeling of the transport
problem is treated differently in the UZ (using a particle-tracking method) and in the SZ (using
streamtubes with a dilution factor). The Panel maintains that a unified treatment should be
feasible and should be adopted.
On the positive side, the Panel notes the excellent analysis, described in the Technical Basis
Document, that relates the dilution factor to transverse dispersivity. The Panel asserts that the
proposed convolution approach is quite useful for abstraction, assuming that processes, such as
adsorption and retardation, remain in the linear regime, and the flow field is at steady state. The
Panel also notes the advances in the analysis of radionuclide sorption on colloids presented in
Chapter 8 of the Technical Basis Document, although these are not included in the current
TSPA-VA. The Panel believes that one of the most significant advances is that the process is
now correctly treated as being dynamic, rather than irreversible.
VI-155
-------
The Peer Review Panel raise a point regarding the fracture-matrix interaction. In the TSPA-VA
model, flow is assumed to occur only through the fractures; the water in the matrix being
stagnant. Instead of explicitly modeling mass diffusion from the fracture to the matrix, the
approach taken is to introduce an effective, time-independent porosity for the entire system, in
which low porosity values reflect limited diffusion, and high values reflect a more enhanced
diffusion. A problem with this representation is that the degree of fracture-matrix interaction is
fixed a priori, rather than being a time-dependent process as it is, in reality. Given that
retardation is associated with the matrix, this assumption will affect the transport predictions.
The Panel also brings forth an issue regarding the averaging the source concentrations over six
areas, which here lie at the interface between the UZ and the SZ. As in UZ transport, this
assumption introduces an artificial spreading which will lead to non-conservative estimates,
particularly at early times (e.g. within the first 10,000 years) when leakage of radionuclides from
waste packages is associated with isolated failures. For such failures, the Panel suggested in its
third interim report (WHI98) that it is unrealistic to assume that radionuclides will produce a
uniform concentration hi the ground water beneath the repository across a flow path that is
hundreds to thousands of meters wide. Even if multiple releases were to occur, the waste
packages that fail could be close to one another within the repository. Such a situation could
occur due to a locally aggressive corrosion environment or the fact that adjoining waste packages
share a common fabrication problem.
In response to the Panel's criticism, the DOE conducted a sensitivity analysis of the effect of the
source size on the dilution factor at the 20-km point. In this analysis, the degree of the non-
conservatism introduced by the approximation made in the TSPA-VA can be assessed. It was
found that, as a result of this approximation, the TSPA-VA underestimates the dose rates, for the
base case parameter values, by a factor of 3. A correction was not introduced in the TSPA-VA,
however. The Panel agrees with the general results of Arnold's analysis, and recommends that
the current TSPA-VA treatment be modified to correct the existing deficiency. In particular, the
Panel believes that the method described in the sensitivity analysis by Arnold and Kuzio
(ARN98) should be applied to the assessment of the exposures that would result from human
intrusion and from the juvenile failure of a waste package.
VI-156
-------
Soil Adsorption of Radionuclides
The TSPA-VA analysis of the performance during the first 10,000 years after repository closure
indicates that the estimated doses are due primarily to "Tc and I29I. The doses at later times from
237Np and 239Pu are projected to be larger. This is due to the fact that the transport of these two
actinides though the UZ and SZ will be delayed for extended periods by chemical sorption and,
therefore, they will not reach the accessible environment in significant concentrations until a
considerably later point in time. No retardation credit is taken for "Tc and I29I (or for three other
radionuclides), based on the lack of observed sorption in batch measurements of Kj values in the
laboratory. This decision is described in the TSPA-VA as being conservative. However, the
Panel points out that field measurements near the Savannah River Plant and in the vicinity of the
Chernobyl nuclear power plant, following the accident at that facility, indicate that radioactive
iodine deposited on the ground has been retained in the upper soil layer to about the same extent
as plutonium and cesium, (STR96; STR97). The Panel has not conducted a literature review on
this issue, but believes that it is likely that measurements taken of areas near the Chernobyl site,
for example, would also provide relevant data on the retention or lack of retention of technetium
in soil. Regarding the retardation of iodine, the Panel notes that additional data sets are likely to
be available from environmental measurements taken at the Hanford site, where radioactive
iodine was released during spent fuel reprocessing. Although it appears that some fraction of the
deposited radionuclides may be transported to ground water (e.g., as with cesium at the Hanford
tank farm), field data suggest that radionuclide does not move unretarded through the soil. It is
the Panel's view that the DOE, in preparing the TSPA-VA, unduly emphasized the results of
laboratory IQ measurements, and did not appropriately consider the results of field measurements
of radionuclide concentrations in the soil following releases that occurred as a result of nuclear
power plant accidents and past nuclear facility operations. The Panel further notes that the DOE
has now recognized this problem and it is being addressed.
State ofNevada/T-Reg, Inc. Identified Needs
The State of Nevada and their consultant T-Reg believe that structural controls on the flow
system have not been adequately addressed in the TSPA-VA. To support their position, they
have identified a number of items in the TSPA-VA which they believe are incompatible, not
well represented, or nonrepresentative of their structurally controlled conceptual flow field. Most
of the nonrepresentativeness occurs in the saturated zone, or in the areal distribution of recharge
to the water table, via the unsaturated zone. Each of these items is discussed separately.
VI-157
-------
Saturated zone
1 The TSP A conceptual flow model allows particles of water to move orthogonal to the
hydraulic gradient. Anisotropic effects due to structure are not considered. This causes
the flow path for releases from the repository to move initially eastward and then
southeastward to Forty Mile Wash, then curve back to the southwest to the Amargosa
Farms area at a 20 km radius. Utilizing anisotropic transmissivities, flow paths are
created which are directly south, then southeast and southwest, considerably shortening
the flowpath to the receptors.
2. Flow path properties used in the "six flow tubes" are not representative of the southerly
flow path. This is because the TSPA flow path would take the releases into alluvium at a
shorter distance than a more southerly flow path. Thus, out of the 20-km compliance
distance, less distance is assigned tuff properties and more is assigned alluvial properties.
(The TSPA flow path is actually longer than 20 km) The alluvial properties are generally
more favorable for retarding and dispersing the repository releases than the tuff properties
and, in fact, now constitute the most important barrier in the saturated zone. A shorter
flow path to the receptors would also be taken in the State's conceptualization.
3. Alluvial properties assigned may not be representative of valley- fill sediments.
According to drilling results of Nye County, presented at the Devils Hole Workshop, the
valley fill sediments south of Yucca Mountain are not primarily alluvium. Rather they
consist of coarse gravels, tufa, basalts, tuffs and lake bed sediments. The State asserts that
sorption, retardation, dispersion and effective porosity assumptions used to describe
transport through "alluvium" must be justified.
4. The State believes that fracture zone effective porosities or hydraulic apertures also need
to be reconsidered or verified. Porosities ranging up to 10% or more are used currently.
The TSPA sampled a distribution of porosities ranging from lO'5 to approximately 20%
but the mean value centered near 2-3%. The State points out that normally effective
porosities for fractured aquifers range on the order of 0.01 to 0.001. These changes would
work to increase the flow velocities inversely, making them higher than most base case
scenarios.
VI-158
-------
5. Eastward expansion of the water table receptor area appears inconsistent with channelized
flow through Ghost Dance fault zone. This eastward expansion could add up to about
25% more area over which to average repository releases from the unsaturated zone. The
State believes that this is inconsistent with results of Bodvarson, shown in the TSPA,
where his center of mass calculations show eastward movement cut off by the presence of
the Ghost Dance fault (If waste is placed east of the Ghost Dance, then some areas of
eastward expansion could be envisioned, but only at these positions, not uniformly across
the length of the repository.)
6. The eastward flow path is not consistent with chemistry data of the USGS, presented at
the Devils Hole Work Shop, April 1999, or earlier work. These data do not indicate an
eastward flow part, but rather a southerly one for numerous isotopes.
7. The eastward flow path may not be consistent with temperature data. Temperature
calculations must be a part of this flow path analyses. Both data sets (temperature and
pressure) must be matched before any flow paths can be believed.
8. The State also questions the NRC well bore dilution numbers and the DOE dilutions
based on the idea of rigid blocks separated by transmissive faults. Nye County drilling
results indicate three boundaries in one of their pump tests. These boundaries were not
distant and depict a situation where by smaller volumes of water would be available for
dilution. Their drilling results also show that pumping rates are highly non-uniform
ranging from a few gallons a minute to several hundred gallons per minute. To use huge
well bore dilution volumes (10s gallons per day) at this point is not justified. The DOE
flow path dilution numbers are much smaller than the NRC's but still not based on
channelized flow and therefore must be justified.
Infiltration
1. The State points out that the map of infiltration based on Flint is partially inconsistent
with their conceptual model of infiltration. While slope, depth of alluvium,
evapotranspiration and elevation are definitely important, so may be some other factors.
Flint assumes that where thick alluvium is present idle recharge occurs. The State
believes that this may not hold true on the steep western slope of Yucca Mountain. The
VI-159
-------
TSPA-VA infiltration model shows a dry area to the west of the crest of Yucca Mountain
rather than a wetter one which the State's conceptualization would predict.
2. The State asserts that it is possible for runoff to go under the alluvium and into fractures,
thus being blanketed from potential evaporation. Given that the Ptn unit is not present to
divert any infiltration in some areas to the west of the repository, then infiltration is
possible directly or nearly so into the Topopah Springs unit, up gradient of the repository.
As stated previously, in the State's conceptual model, the western side would be expected
to be wetter than the eastern side.
3. The State also notes that recent correspondence from Steve Brocum, DOE to the NRC,
Sandra Wastler, Monthly Progress Report, dated 03-26-99 indicates that water potentials
are higher in the East-West Cross Drift (ECRD) and indicate that the rock is wetter and
the moisture is more uniformly distributed than expected. The State points out that the
structure here is probably important and explains further that tensional north-south
trending smaller structures across the mountain block may be channeling the movement
of infiltrating water. By cutting across them in an east-west direction, the ECRD has
intercepted more pathways than when they bored north-south in the main drift section,
parallel to these features.
4. Flint et al. (FLI96) and the TSPA-VA infiltration model indicate that the net infiltration is
lower in the washes. The State generally concurs, except that they believe higher
infiltration occurs in the upper reaches of most washes, not along the lower reaches.
5 The State believes that infiltration at the water table surface may also not be
representative, nor consistent with their model. The TSPA-VA infiltration map shows
lower infiltration along the Ghost Dance fault area where the State would assume higher
infiltration based on the temperature distribution.
6. The State also notes that Zell Peterman, USGS has shown what appears to be a plume of
younger water along the northwest side of the mountain block. The State would expect
that this plume would be infiltrating or recharging in that area. The water table map in
the TSPA-VA shows it to be relatively dry.
7. Breakthrough curves simulating the dry (present day) climate conditions Figure 3-10 of
the TSPA-VA indicate less than 5% cumulative breakthrough for an unretarded tracer at
the water table in less than 800 years with 95% breakthrough between 800-12,000 years.
VI-160
-------
The State believes that it is reasonable that if in the north-south trending drifts, 36C1 is
seen within 50 years, that when sampled, the ECRD may also yield 36C1 and perhaps more
than in the north-south drift. The State asserts that these low percentages of ground-water
breakthrough are not justified.
8. Drift scale seepage assumes a 99.5% reduction of net infiltration. While the State would
expect some diversion they would also expect that the infiltration rate of water into the
drifts would be on the order of that calculated for G Tunnel, i.e., about 3% of the annual
average rainfall. This would allow for 4.5mm/vr into the tunnel. Instead, VA values are
orders of magnitude lower. The State notes that these low values need to be supported.
9. Seepage Fraction, or number of canisters hit by drips is surprisingly low. The State claims
that given their concept of fracture controlled drips and data from the DOE that the
ECRD is wetter than expected, no confidence can be assigned to this number. The State
further notes that testing in Alcove 1, though shallow with high infiltration applied, is
indicative of a potential for more wide spread dripping given that 45% of the roof area
catchments had contacted drips.
10. Drips onto packages are assumed stationary during their flow history. The State believes
that this may also not be the case for many reasons hydraulic, geochemical or tectonic,
and further suggests that moving them about over tune would tend to wet a larger number
of packages.
11. Sorption and matrix diffusion are assumed to always operate together on sorbing species.
The State questions whether this a valid assumption, and notes that it allows for more
retardation than may be justified if considering them separately.
12. Volumetric flux via the drift invert assumes 10% porosity, 99.8% saturation, sorption and
diffusion. The State contends that if the invert fails over time, or fractures develop due to
tectonic activity, then flow would be focused and radionuclides less retarded. The State
further notes that no provision for invert failure or degradation has been made in the
TSPA-VA.
State's Conclusions
Based on the concerns identified above, the State made the following three conclusions:
VI-161
-------
1. If the basic flow pathways and their characteristics are not correctly interpreted or
represented, when they possess qualities which can be measured or tested in the field,
then little confidence can be placed on analyses, interpretations or designs which have not
been or cannot be tested or verified.
2. The flow fields need to be calibrated against temperature or other independent variables
in order to support flow paths selected.
3. Future TSPAs will have to modified their flow model to be representative of a
structurally controlled flow field
VI.2.1.5 Nye County Well Characterization Program
An Early Warning Drilling Program (EWDP) has implemented as part of the Nye County
Nuclear Waste Repository Project Office (NWRPO) Yucca Mountain Oversight program. The
purpose of the EWDP is to establish a ground-water monitoring system to protect the residents of
Nye County hi Amargosa and Pahrump Valley against potential radionuclide contamination.
The program is also intended to provide geologic and hydrologic information to supplement
DOE's site characterization program. The focus area is located hi a very complex and least
understood hydrogeologic system in the vicinity of Yucca Mountain. The investigation intends
to address the following issues; the orgin of the spring deposits, the geology of the area and the
recharge and ground-water flow patterns.
An interim status report presented to the Nuclear Waste Technical Review Board in July of 1999
summarized the activities to date. Geophysical logging has been conducted on 5 of the wells and
includes temperature, magnetic and gamma logging. Interpretation of the data is providing
information on the rock types present, potential for vertical water migration and the geochemical
facies present. Ongoing studies involve additional petrographic and ground-water analyses.
A second phase of testing will also be conducted in which the drilling of the remaining wells will
be completed, high-discharge aquifer tests will be performed on selected wells and the spring
deposits in Crater Flat will be investigated.
VI-162
-------
REFERENCES
BAT92 . Bates, J.K., J.P. Bradley, A. Teetsov, C.R. Bradly, and M. Buchholtz ten Brink.
"Colloid Formation During Waste Form Reaction: Implications for Nuclear Waste
Disposal." Science, 256, 649-651. 1992.
BOD96 Bodvarsson, G.S. and T.M. Bandurraga, eds. Development and Calibration of the
Three-Dimensional Site-Scale Unsaturated Zone Model of Yucca Mountain,
Nevada for the Viability Assessment. Ernest Orlando Lawrence Berkeley National
Laboratory Report, LBNL-39315, 563. 1996.
BOD97 Bodvarsson, G.S., T.M. Bandurraga, C. Haukwa, E.L. Sonnenthal, and Y.S. Wu.
The Site Scale Unsaturated Zone Model of Yucca Mountain, for the Viability
Assessment, Yucca Mountain Project Milestone Report LBNL-40376, UC-814,
LB970601233129.001, Lawrence Berkeley National Laboratory, Berkeley, CA.
1997.
BRO97 Brodsky, N.S., M. Riggins, J. Connolly, and P. Ricci. Thermal Expansion
Thermal Conductivity and Heat Capacity Measurements for Boreholes UE25
NRC-4, UE25NRG-5, USWNRG-6, andUSWNRG 7/7a, SAND 95-1955,
SNL01 A05059301.005, Sandia National Laboratories, Albuquerque, NM. 1997.
CAR86 Carr, W.J., Byers, F.M.J. and Orkild, P.P. Stratigraphic and Volcano-Tectonic
Relations of Crater Flat Tuff and Some Older Volcanic Units, Nye County,
Nevada. U.S. Geological Survey Professional Paper 1323, GS900983117421.001,
NNA. 19930629.0005, U.S. Geological Survey, Denver, CO. 1986.
CLA97 Clayton, R.W., W.P. Zelinski, and C.A. Rautman. (CRWMS). ISM2.0:A3-D
Geologic Framework and Integrated Site Model of Yucca Mountain, ID
BOOOOOOOO-01717-5700-00004 Rev 0, MOL. 19970122.0053, Civilian
Radioactive Waste Management System Management and Operating Contractor.
1997.
COH97 Cohen, A.J.B., and A.M. Simmons. Development of the Sub-Site-Scale Three-
Dimensional Numerical Zone Saturated Zone Flow Model — Evaluation of Flow
Processes, Yucca Mountain Site Characterization Project Milestone SP33SM4,
Lawrence Berkeley National Laboratory, Berkeley, CA, 1997.
CRA84 Craig, R. W. and J.H. Robinson. Geohydrology of Rocks Penetrated by Test Well
UE-25p#l, Yucca Mountain Area, Nye County, Nevada, GS920408312314.009,
NNA. 19870519.0106, U.S. Geological Survey, Denver, CO. 1984.
CRW96 CRWMS. Calibration and Prediction of Pneumatic Response at Yucca
Mountain, Nevada Using the Unsaturated Zone Flow Model. Milestone Report
-------
OB12M, Las Vegas, Nevada: Author. MOL.19970714.0669. 1996.
CRW97 Unsaturated Zone Flow Model Expert Elicitation Project - May 1997, San
Francisco, California: Geomatrix Consultants Inc.; Las Vegas, Nevada: Author.
MOL.19971009.0582. 1997.
CRW98 Yucca Mountain Site Description," BOOOOOOOO-01717-5700-00019 REV 00. Las
Vegas, Nevada: Author. MOL. 19980729.0050 (Section 4); MOL 19980729.0051
(Sections). 1998.
CZA84 Czarnecki, J.B., and R.K. Waddell. Finite-Element Simulation of Ground-Water
Flow in the Vicinity of Yucca Mountain, Nevada-California, U.S. Geological
Survey Water-Resources Investigations Report, 84-4349,38 p., 1984.
CZA85 Czarnecki, J.B. Simulated Effects of Increased Recharge on the Ground-Water
Flow System of Yucca Mountain and Vicinity, Nevada-California: U.S.
Geological Survey Water-Resources Investigations Report 84-4344, 33 p., 1985.
CZA90 Czarnecki, J.B. Geohydrology and Evapotranspiration at Franklin Lake Playa,
Inyo County, California: U.S. Geological Survey Open-File Report 90-356, 96 p.,
1990.
CZA91 Czarnecki, J.B., and W.E. Wilson. "Conceptual Models of the Regional Ground-
Water Flow and Planned Studies at Yucca Mountain, Nevada," Hydrological
Science and Technology, v. 7, no. 1-4, p. 15-25,1991.
CZA94 Czarnecki, J.B., G.M. O'Brien, P.H. Nelson, J.H. Sass, J.W. Bullard, and A.L.
Flint "Is There Perched Water Under Yucca Mountain in Morehole USW G-2?,"
Trans. American Geophysical Union, EOS, vol. 75, no. 44, pp. 249-250,1994.
CZA95 Czarnecki, J.B., P.H. Nelson, G.M. O'Brien, J.H. Sass, Bhaskar Thapa, Yoshitaka
Matsumoto, and Osamu Murakami. "Testing in Borehole USW G-2 at Yucca
Mountain: The Saga Continues," Trans. American Geophysical Union, EOS, vol.
76,no.7,p.F190,1995.
CZA97 Czarnecki, J.B., C.C. Faunt, C.W. Gable, and G.A. Zyvoloski. Preliminary
Three-Dimensional Finite-Element Groundwater Flow Model of the Saturated
Zone, Yucca Mountain, Nevada. U.S. Geological Survey YMP milestone number
SP23NM3. 1997.
DAG97a D'Agnese, F.A., C.C. Faunt, A.K. Turner, and M.C. Hill. 1997. Hydrogeologic
Evaluation and Numerical Simulation of the Death Valley Regional Groundwater
Flow System, Nevada and California, Using Geoscientific Information Systems.
U.S. Geological Survey Water-Resources Investigations Report 96-4300.
-------
DAG97b D'Agnese, F.A., G.M. O'Brien, C.C. Faunt, and C.A. San Juan. 1997. Regional
Saturated-Zone Synthesis Report: Simulated Effects of Climate Change on the
Death Valley Regional Ground-Water Flow System, Nevada and California.
.. Denver, Colorado: U.S. Geological Survey. Draft. MOL. 19980224.0574.
DAY96 Day, W.C., CJ. Potter, D.S. Sweetkind, and R.P. Dickerson, mapping through
4/26/96 (in review). Preliminary Bedrock Geological Map of the Central Block
Area Yucca Mountain, Nevada, GS960708314221.001, U.S. Geological Survey
Preliminary Map (not for distribution).
DEW93 DeWispelare, A.R., et al. (eds.). Expert Elicitation of Future Climate in the
Yucca Mountain Vicinity, Report 93-016, Center for Nuclear Waste Regulatory
Analyses, San Antonio, Texas, 1993.
DOE95b U.S. Department of Energy, Total System Performance Assessment -1995: An
Evaluation of the Potential Yucca Mountain Repository, TRW Environmental
Safety Systems, Inc., BOOOOOOO-01717-2200-00136, Revision 01, November
1995.
Ervin, E.M., R.R. Luckey, and D.J. Burkhardt. Revised Potentiometric-Surface
Map, Yucca Mountain and Vicinity, Nevada, U.S. Geological Survey Water-
Resources Investigations Report, 93-4000, 17 p., 1994.
Fabryka-Martin, J., and Wolfsberg, A. Hydrologic Flow Paths and Rates Inferred
from the Distribution ofChlorine-36 in the ESF, Nuclear Waste Technical
Review Board, July 1996.
Finsterle, S. ITOUGH2 User's Guide Version 2.2, LBNL-34581. Berkeley,
California: Lawrence Berkeley National Laboratory. MOL. 19941026.0075.
1993.
FLI94 Flint, A.L., and L.E. Flint. Spatial Distribution of Potential Near Surface
Moisture Flux at Yucca Mountain, Proceedings of the Fifth Annual International
Conference on High Level Radioactive Waste Management, Las Vegas, Nevada,
pp. 2352-2358,1994.
FLI95 Flint, L.E. and A.L. Flint. Shallow Infiltration Processes at Yucca Mountain,
Nevada-Neutron Logging Data 1984-93, USGS Water-Resources Investigations.
Report 95-4035. Denver Colorado: U.S. Geological Survey.
MOL. 19960924,0577. 1995.
FLI96 Flint, A.L., J.A. Hevesi, and L.E. Flint. Conceptual and Numerical Model of
Infiltration for the Yucca Mountain Area, Nevada, Milestone Report 3GUI623M.
DTN: GS960908312211.003. Denver, Colorado: U.S. Geological Survey.
MOL. 19970409.0087. 1996.
ERV94
FAB96
FIN93
-------
FOR96 Forester, R.M., and J.P. Bradbury. Synthesis of Quaternary Response of the Yucca
Mountain Unsaturated and Saturated Zone Hydrology to Climate Change,
Milestone Report 3GCA102M, Denver, CO, U.S. Geological Survey, 1996.
FRE79 Freeze, R.A. and J.A. Cherry. Ground-water, p. 67. Englewood Cloffs, New
Jersey: Prentice-Hall. 240220. 1979.
FRI94 Fridrich, C.J., W.W. Dudley, Jr., and J.S. Stuckless. "Hydrogeologic Analysis of
the Saturated-Zone Ground Water System, under Yucca Mountain, Nevada,"
Journal of Hydrology, 154:133-168,1994.
GAB95 Gable, C.W., T.A. Cherry, H.E. Trease, and G.A. Zyvoloski. GEOMESH Grid
Generation. Los Alamos National Laboratory report LA-UR-95-4143. 1995.
GAB96 Gable, C.W., H. Trease, and T. Cherry. Geological Applications of Automatic
Grid Generation Tools for Finite Elements Applied to Porous Flow Modeling. In
Numerical Grid Generation in Computational Fluid Dynamics and Related
Fields, edited by B.K. Soni, J.F. Thompson, H. Hausser, and P.R. Eiseman,
Engineering Research Center. Mississippi State University Press. 1996.
GEL96 Geldon, A.L., Results and Interpretation of Preliminary Aquifer Tests in
Boreholes UE-25 c#I, UE-25 c#2, and UE-25 c#3, Yucca Mountain, Nye County,
Nevada, U.S. Geological Survey Water-Resources Investigations Report 94-4177,
119 p. 1996.
GEL97 Geldon, A.L., A.M. Umari, M.F. Fahy, J.D. Earle, J.M. Gemmell, and J. Darnell,
Results of Hydraulic and Conservative Tracer Tests in Miocene Tufaceous Rocks
at the C-Hole Complex, 1995 -1997, Yucca Mountain, Nye County, Nevada.
1997 Milestone Report SP23PM3. Denver, Colorado: U.S. Geological Survey -
Yucca Mountain Project Branch. MOL.19980122.0412. 1997.
GEO97 Geomatrix Consultants, Inc. and TRW. Unsaturated Zone Flow Model Expert
Elicitation Project, Las Vegas, Nevada, May, 1997.
GEO98 Geomatrix Consultants Inc. and TRW. Saturated Zone Flow and Transport
Expert Elicitation Project, January 1998, prepared for U.S. Department of Energy,
Yucca Mountain Site Characterization Office, WBW 1.2.5.7. 1998.
GRA97 Graves, R.P., Patrick Tucci, and G.M. O'Brien, Analysis of Water-Level Data in
the Yucca Mountain Area, Nevada, 1985-1995, U.S. Geological Survey Water-
Resources Investigations Report 96-4256,140 p., 1997.
GRI80 Grisak, G.E., and J.F. Pickens. "Solute Transport Through Fractured Media: 1.
The Effect of Matrix Diffusion." Water Resources Research 16(4): 719-730.
1980.
-------
HEV92 Hevesi, J.A., A.L. Flint, and J.D. Istok. "Precipitation Estimation in Mountainous
Terrain Using Multivariate Geostatistics: Part II: Isohyetal Maps: J." Applied
.- Meteorology, vol. 31, no. 7, p. 677-688, 1992.
HEV94 Hevesi, J.A., AX. Flint, and L.E. Flint. Verification of a One-Dimensional Model
for Predicting Shallow Infiltration at Yucca Mountain, High Level Radioactive
Waste Management, Proceedings of the Fifth Annual International, Las Vegas,
Nevada, May 22-26,1994,4,2323-2332. La Grange Park, Illinois: American
Nuclear Society, Inc.; New York, New York; American Society of Civil
Engineers. 238118. 1994.
HIL80 Hillel, D. "Applications of Soil Physics," New York, New York, Academic
Press. 217649. 1980.
HIL92 Hill, M.C. A Computer Program (MODFLOWP) for Estimating Parameters of a
Transient, Three-Dimensional, Ground-Water Flow Model Using Nonlinear
Regression. Open-File Report 91-484. Denver, Colorado: U.S. Geological
Survey. MOL.19961118.0122. 1992.
HOD96 Hodge, V.F., K.H. Johannesson, and K.J. Stetzenbach. Rhenium, Molybdenum,
and Uranium in Ground-waterfrom the Southern Great Basin, USA: Evidence for
Conservative Behavior, Geochimica et Cosmochimica Acta 60, pp. 3197-3214.
1996.
HOR33 Horton, R.E. The Role of Infiltration in the Hydrologic Cycle, Transactions,
American Geophysical Union, 14th Annual Meeting, June 1933, 14,446-460.
Washington, D.C.: American Geophysical Union. 237958. 1933.
KIL91 Kilroy, K.C., Ground-Water Conditions in Amargosa Desert, Nevada-California,
1952-1987, U.S. Geological Survey Water Resources Investigations Report 89-
4101,91pp., 1991.
LAN97 Summary Report, Geochemistry/Transport Laboratory Tests, Deliverable
Number SP23QM3. Los Alamos, New Mexico: Los Alamos National Laboratory.
MOL.19980122.0053. 1997.
LBL96 Lawrence Berkeley Laboratory. Development and Calibration of the Three-
Dimensional Site Scale Unsaturated Zone Model of Yucca Mountina, Nevada,
Berkeley, California, 1996.
LUC96 Luckey, R.R., P. Tucci, C.C. Faunt, E.M. Ervin, W.C. Steinkampf, F.A.
D'Agnese, and G.L. Patterson. Status of Understanding of the Saturated-Zone
Ground-Water Flow System at Yucca Mountain, Nevada, as of 1995. Water-
Resources Investigations Report 96-4077. Denver, Colorado: U.S. Geological
-------
Survey, 1996. MOL.19970116.0039. 1996.
MAL85 Maloszewski, P., and A. Zuber. "On the Theory of Tracer Experiments in
.- Fissured Rocks with a Porous Matrix." Journal of Hydrology 79: 333-358. 1985.
MAL91 Maloszewski, P., and A. Zuber. "Influence of Matrix Diffusion and Exchange
Reactions on Radiocarbon Ages in Fissured Carbonate Aquifers." Water
Resources Research 27(8): 1937-1945. 1991.
MAT96 Mattson, S.R. and R.W. Clayton. Yucca Mountain Project Stratigraphic
Compendium: Document Identifier BOOOOOOOO-01717-5700-00004 Rev. 2,
MOL.19970113.0088, Civilian Radioactive Waste Management System
Management and Operating Contractor, Las Vegas, Nevada. 1996.
MCD88 McDonald, M.G. and A.W. Harbaugh, A Modular Three-Dimensional Finite-
Difference Ground-Water Flow Model. Techniques of Water Resources
Investigations, Book 6, Chapter Al. Denver, CO: U.S. Geological Survey, 1988,
224714.
NER80 Neretnicks, I. "Diffusion in the Rock Matrix: An Important Factor in
Radionuclide Migration?" Journal of Geophysical Research 85(B8): 4379-4397.
1980.
NER82 Neretnicks, I., T. Eriksen, and P. Tahtinen. "Tracer Movement in a Single Fissure
in Granitic Rock: Some Experimental Results and Their Interpretation." Water
Resources Research 18: 849-858. 1982.
NEU90 Neuman S.P., 1990. "Universal Scaling of Hydraulic Conductivities and
Dispersivities in Geologic Media." Water Resources Research 26(8): 1749-1758.
1990.
NIT97 Nitao, J.J., "Preliminary Bounds for the Drift-Scale Distribution of Percolation
and Seepage at the Repository Level Under Pre-Emplacemnt Conditions",
Lawence Livermore National Laboratory, 1997.
NRC92 U.S. Nuclear Regulatory Commission, Initial Demonstration of the NRC 's
Capability to Conduct a Performance Assessment for a High-level Waste
Repository, WJREG-1327,1992.
NRC95 National Research Council. Technical Bases for Yucca Mountain Standards.
Committee on Technical Bases for Yucca Mountain Standards, Board on
Radioactive Waste Management. National Academy of Science. Washington,
D.C.: National Academy Press. 104273,217588. 1995.
NRC96 U.S. Nuclear Regulatory Commission, Presentation to the Advisory Committee on
-------
Nuclear Waste Concerning Duration of the Regulatory Period, T. McCartin,
March 27, 1996.
NRC97a U.S. Nuclear Regulatory Commission, NRC High-level Radioactive Waste
Program Annual Progress Report Fiscal Year 1996, NUREG/CR-6513, No. 1,
BudhiSagar (editor), 1997.
NRC97b U.S. Nuclear Regulatory Commission, CNWRA Investigations ofYMR Geologic
Setting Relevant to Igneous Activity, Dr. Brittain Hill, presentation at DOE/NRC
Technical Exchange - Igneous Activity Program, February 25-26,1997.
NRC98 U.S. Nuclear Regulatory Commission, 100th Advisory Commission on Nuclear
Waste Meeting, Rockville, Maryland, April 21-23,1998.
NWT98 Nuclear Waste Technical Review Board. 1998. Report to the U.S. Congress and
the U.S. Secretary of Energy, 52 pages.
OBR97 O'Brien, G.M. Analysis of Aquifer Tests Conducted in Boreholes USW WT-10,
UE-25 WT#12, and USWSD-7, 1995-96, Yucca Mountain, Nevada, U.S.
Geological Survey Water-Resources Investigations Report 96-4293, 36 p., 1997.
OST94 Osterkamp, W.R., L.J. Lane, and C.S. Savard. "Recharge Estimates Using a
Geomorphic/Distributed-Parameter Simulation Approach, Amargosa River
Basin," Water Resources Bulletin, 30 (3), 493-507. Minneapolis, Minnesota:
American Water Resources Association. 237428. 1994.
PAC96 Paces, J.B., L.A. Neymark, B.D. Marshall, J.F. Whelan, and Z.E. Peterman. Ages
and Origins of Subsurface Secondary Minerals in the Exploratory Studies Facility
(ESF), Milestone Report 3GQH450M. Denver, Colorado: U.S. Geological
Survey. MOL. 19970324.0052. 1996.
PET84 Peters, R.R., E.A. Klavetter, I.J. Hall, S.C. Blair, P.R. Heller, and G.W. Gee.
Fracture and Matrix Hydrologic Characteristics ofTufaceous Materials from
Yucca Mountain, Nye County, Nevada. Sandia National Laboratories Report
SAND84-1471. 1984.
PRU87 Pruess, K., TOUGH User's Guide, Nuclear Regulatory Commission, Report
NUREG/CR-4645. 1987. (also Lawrence Berkeley Laboratory Report, LBL-
20700, NNA.19890315.0010, Lawrence Berkeley National Laboratory, Berkeley,
California, June 1987).
PRU91 Pruess, K. TOUGH2 - A General Purpose Numerical Simulator for Multiphase
Fluid and Heat Flow, LBL-29400. Berkeley, California: Lawrence Berkeley
National Laboratory. NNA.19940202.0088. 1991.
-------
PRU93 Prudic, D.E., J.R. Harrill, and T.J. Burbey. Conceptual evaluation of regional
ground-water flow in the carbonate-rock province of the Great Basin, Nevada,
Utah, and adjacent States, U.S. Geological Survey Open-File Report 90-560,103
.-p., 1993.
RAS86 Rasmuson, A., and I. Neretnicks. "Radionuclide Transport in Fast Channels in
Crystalline Rock," Water Resources Research 22(8): 1247-1256. 1986.
RAU97 Rautman, C.A. and S.A. McKenna. Three-Dimensional Hydrological and
Thermal Property Models of Yucca Mountain, Nevada, SAND97-1730.
Albuquerque, New Mexico: Sandia National Laboratories. MOL.19980311.0317.
1997.
REE94 Reeves, M.. Review and Selection ofUnsaturated Flow Models. Intera
Document BOOOOOOOO-01 425-2200-00001.
REI97 Reimus, P.W., and H.J. Turin. 1997. Draft Report of Results of Hydraulic and
Tracer Tests at the C-wells Complex. Los Alamos National Laboratory YMP
Milestone SP23APMD.
RIC84 Rice, W.A. Preliminary Two-Dimensional Regional Hydrologic Model of the
Nevada Test Site and Vicinity: SAND83-7466, Sandia National Laboratories,
Albuquerque, New Mexico, 89 p., 1984.
ROB96 Robinson, B.A., A.V. Wolfsberg, H.S. Viswanathan, C.W. Gable, G.A.
Zyvoloski, and H.J. Turin. Modeling of Flow, Radionuclide Migration, and
Environmental Isotope Distributions at Yucca Mountain, Draft. Milestone Report
3672. Los Alamos, New Mexico: Los Alamos National Laboratory.
MOL.19970103.0042. 1996.
ROB97 Robinson, B.A., A.V.Wolfsberg, H.S. Viswanathan, G.Y. Bussed, C.W. Gable,
and A. Meijer. The Site-Scale Unsaturated Zone Transport Model of Yucca
Mountain, Milestone Report SP25BM3. Los Alamos, New Mexico: Los Alamos
National Laboratory. MOL. 19980203.0570. 1997.
SAS88 Sass, J.H., A.H. Lachenbruch, W.W. Dudly, Jr., S.S. Priest, and R.J. Munroe.
Temperature, Thermal Conductivity and Heat Flow Near Yucca Mountain,
Nevada Some Tectonic and Hydrologic Implications, U.S. Geological Survey
Report USGS-OFR-87-649; DE89 002697, NNA. 19890123.0010,
GS950408318523.001 Non-Q, U.S. Geological Survey. 1988.
SIN87 Sinton, P.O. Three-Dimensional, Steady-State, Finite-Difference Model of the
Ground-Water Flow System in the Death Valley Ground-Water Basin, Nevada-
California: Golden, Colo., Colorado School of Mines, unpublished Master's
Thesis, 145 p., 1987.
-------
SUDS 1 Sudicky, E.A., and E.O. Frind. "Carbon-14 Dating of Groundwater in Confined
Aquifers: Implications of Aquitard Diffusion." Water Resources Research 17(4):
1060-1064. 1981.
SUD85 Sudicky, E.A., R.W. Gillham, and E.O. Frind. "Experimental Investigation of
Solute Transport in Stratified Porous Media: 1. The nonreactive case." Water
Resources Research 21(7):1035-1041. 1985.
SZA94 Szabo, 1994. "Paleoclimatic Inferences from a 120,000-yr calcite record of water-
table fluctuation in browns room of Devils Hole, Nevada", Quaternary Research.
Vol, 41 59-69, 1994.
SZY89 Szymanski, J.S., Conceptual Considerations of the Yucca Mountain Groundwater
System -with Special Emphasis on the Adequacy of this System to Accommodate a
High-Level Nuclear Waste Repository, Las Vegas, Nev., 1989.
NNA.19890831.0152. 1989.
THO96 Thomas, J.M., and A.H. Welch, 1996. Geochemistry and Isotope Hydrology of
Representative Aquifers in the Great Basin Region of Nevada, Utah, and Adjacent
States, U.S. Geological Survey Professional Paper 1409-C, 102 pp. 1996.
TRE96 Trease, H., D. George, C.W. Gable, J. Fowler, A. Kuprat, and A. Khamyaseh.
1996. The X3D Grid Generation System. In Numerical Grid Generation in
Computational Fluid Dynamics and Related Fields, edited by B.K. Soni, J.F.
Thompson, H. Hausser, and P.R. Eiseman, Engineering Research Center.
Mississippi State University Press. 1996.
TRI97 Triay, I.R., A. Meijer, J.L. Conca, K.S. Kung, R.S. Rundberg, B.A. Strietelmeier,
and C.D. Tait. Summary and Synthesis Report on Radionuclide Retardation for
the Yucca Mountain Site Characterization Project, LA-13262-MS. Milestone
3784M. Los Alamos, New Mexico: Los Alamos National Laboratory.
MOL.19971210.0177. 1997.
TUC95 Tucci, Patrick and DJ. Burkhardt. Potentiometric-Surface Map, 1993, Yucca
Mountain and Vicinity, Nevada, U.S. Geological Survey Water Resources
Investigations Report 95-4149,15 p., 1995.
USG84 U.S. Geological Survey. Conceptual Hydrologic Model of Flow in the
Unsaturated Zone, Yucca Mountain, Nevada, Water Resources Investigations
Report 84-4345,1984.
WAD82 Waddell, R.K. Two-Dimensional, Steady-State Model of Ground-Water Flow,
Nevada Test Site and Vicinity, Nevada-California. Water-Resources
Investigations Report 82-4085. Denver, Colorado: U.S. Geological Survey, 1982.
NNA.19870518.0055.
-------
WAD84 Waddell, R.K., J.H. Robinson, and R.K. Blankennagel. Hydrology of Yucca
Mountain and Vicinity, Nevada-California: Investigative Results Through Mid-
1983. Water Resources Investigations Report, 84-4267, Denver, Colorado: U.S.
.. ... Geological Survey, 1984. NNA.19870406.0343.
WAD97 Waddell, R.K., 1997. Speech Presented at YMP Saturated Zone Expert
Elicitation, Denver, CO, July, 1997.
WAN98 Wang, L.M., S.X. Wang and R.C. Ewing. Radiation Effects in Zeolite: Relevance
to Near-Field Containment. Proceedings of the 8th International Conference on
High-Level Radioactive Waste Management, Las Vegas, Nevada, May 11-14,
1998, American Nuclear Society, 772-774.
WAT94 Watermark Computing. PEST Model-Independent Parameter Estimation: User's
Manual. Oxley, Australia: Watermark Computing. 240719.
WHI85 Whitfield, M.S., E.P. Eshom, W. Thordarson, and D.H. Schaefer. Geohydrology
of Rocks Penetrated by Test Well USWH-4, Yucca Mountain, Nye County,
Nevada. U.S. Geological Survey Water Resources Investigations Report 85-4030.
1985.
WIL94 Wilson, M.L., J.H. Gauthier, R.W. Barnard, G.E. Barr, H.A. Dockery, E. Dunn,
R.R. Eaton, D.C. Guerin, N. Lu, MJ. Martinez, R. Nilson, C.A. Rautman, T.H.
Robey, B. Ross, E.E. Ryder, A.R. Schenker, S.A. Shannon, L.H. Skinner, W.G.
Halsey, J.D. Gansemer, L.C. Lewis, A.D. Lament, I.R. Triay, A. Meijer, and D.E.
Morris. Total System Performance Assessment for Yucca Mountain - SNL Second
Iteration (TSPA-I993), SAND93-2675. Albuquerque, New Mexico: Sandia
National Laboratories. NNA. 19940112.01232. 1994.
WIN72 Winograd, I.J., and Irving Friedman. Deuterium as a tracer of regional ground-
water flow, southern Great Basin, Nevada-California, Geological Society of
America Bulletin, v. 83, no. 12, p. 3591-3708,1972.
WIN75 Winograd, I.J. and W. Thordarson. Hydrogeologic and Hydrochemical
Framework South-Central Great Basin, Nevada-California, Withy Special
Reference to the Nevada Test Site. Professional Paper 712-C. Washington, D.C.:
U.S. Geological Survey, 1975. 206787.
WIT92 Wittwer, C.S., G.S. Bodvarsson, M.P. Chornack, A.L. Flint, L.E. Flint, B.D.
Lewis, R.W. Spengler, and C.A. Rautman. Design of a Three-Dimensional Site-
Scale Model for the Unsaturated Zone at Yucca Mountain, Nevada, High Level
Radioactive Waste Management, Proceedings of the Third International
Conference, Las Vegas, Nevada, April 12-16,1992. 1.263-271. La Grange Park,
Illinois: American Nuclear Society, Inc. 225904. 1992.
-------
YAN96 Yang, I.C., G.W. Rattray, and P. Yu, 1996. Interpretations of Chemical and
Isotopic Data From Boreholes in the Unsaturated Zone at Yucca Mountain,
Nevada, U.S. Geological Survey Water Resources Investigation Report 96-4058,
GS970108312271.001, U.S. Geological Survey, Denver, CO. 58 pp.
ZEL96 Zelinski, W.P., and R.W. Clayton. A 3-D Geologic Framework and Integrated
Site Model of Yucca Mountain: Version ISM 1.0 Civilian Radioactive Waste
Management System Management and Operating Contractor Document
BOOOOOOOO-01717-5700-00002, Rev. 1. Las Vegas, NV: TRW Environmental
Safety Systems Inc. 1996.
ZYV83 Zyvoloski, G.A. Finite Element Methods for Geothermal Reservoir Simulation."
International Journal for Numerical and Analytical Methods in Geomechanics, 7
(1), 75-86. New York, New York: John Wiley & Sons. 224068. 1983.
ZYV86 Zyvoloski, G.A. Incomplete Factorization for Finite Element Methods,
International Journal for Numerical Methods in Engineering, 23(6), 1101-1109.
New York, New York: John Wiley & Sons. 237569. 1986.
ZYV95 Zyvoloski, G., Z. Dash, and S. Kelkar. FEHM1.0, Finite Element Heat and Mass
Transfer Code, LA-12062-MS, Rev. 1, Los Alamos National Laboratory, Los
Alamos, New Mexico, 1995.
ZYV97 Zyvoloski, G.A., B.A. Robinson, Z.V. Dash, and L.L. Trease. User's Manual for
the FEHM Application — A Finite Element Heat- and Mass-Transfer Code, LA-
13306-M. Los Alamos, New Mexico: Los Alamos National Laboratory. 235999.
1997.
------- |