Geologic Disposal of High-Level Radioactive Waste—Status, Key Issues, and

Trends
Jens Birkholzer, James Houseworth, Chin-Fu Tsang
Earth Sciences Division
Lawrence Berkeley National Laboratory

Abstract:
The permanent disposal of high-level radioactive waste is one of the major technical hurdles that
must be addressed if electrical power production by nuclear energy is to remain viable. The main
challenge is that the waste must be effectively isolated from interactions with the biosphere for
hundreds of thousands of years. A number of permanent disposal options have been proposed
and reviewed by various countries and scientific organizations during the past few decades, and
there appears to be a consensus today that mined geologic disposal is the most practical and
effective method. Several variations on mined geologic disposal are being intensively studied by
waste-producing countries. These investigations address a wide range of scientific questions,
such as the behavior of geological and engineered barriers over time, or the use of quantitative
modeling and qualitative observational evidence to demonstrate the safety of disposal. The
present paper provides an overview of current approaches, scientific issues, and safety
assessments related to mined geologic repositories for high-level radioactive waste.

1
Outline:
1. Introduction

2. Background
2.1 Radioactive Waste Inventory and Nuclear Fuel Cycles
2.2 Alternative Geologic Disposal Options
2.3 Governance and Management

3. Mined Geologic Disposal

3.1 Engineered Barriers
3.2 Natural Barriers
3.3 Retrievability

4. Physico-Chemical Processes Affecting a Mined Geologic Repository and Their Studies in

Underground Research Laboratories
4.1 Four Stages in the Development of a Geologic Repository
4.2 Underground Research Laboratories

5. Recent Trends in Assessing the Long-Term Safety of Mined Geologic Disposal

5.1 Safety Standards
5.2 The Safety Case
5.3 Quantitative Safety Assessments
5.4 Long-Term Predictability

6. Conclusions

2
Key Words for Indexing:
Barriers, Isolation, Repository, Safety

Glossary:
Engineered barriers are the engineered part of the disposal system, primarily consisting of the
waste form, waste package, backfill, ground support, excavated openings of the repository, and
any seals and plugs used to close such openings.

Natural barriers are formed by the rock formations around the repository and along the
radionuclide transport pathways that affect the movement of fluid, colloids, dissolved
constituents, heat, and transmission of mechanical stress.

Biosphere is the part of the earth and atmosphere where living organisms exist.

High-level radioactive waste is spent fuel or spent fuel reprocessing waste that has high levels
of radioactivity and generates significant quantities of heat.

Spent nuclear fuel is fuel that has been irradiated in a nuclear reactor to the point that the build-
up of fission products and transuranics interferes with the fission reaction. As a result, it is
removed from the reactor and replaced with fresh fuel to maintain the reactor’s energy output.

Once-through fuel cycle is where spent fuel is disposed of directly as high-level waste.

Reprocessing is the separation of spent fuel into components that can be returned as fuel to the
nuclear reactor and components that are waste.

Acronyms:
CFM: colloid formation and migration
EDZ: excavation damaged zone
ESF: Exploratory Studies Facility
FEBEX: Full-scale Engineered Barriers Experiment
FEPs: features, events, and processes
HADES: High Activity Disposal Experimental Site
THM: thermal-hydro-mechanical
URL: underground research laboratory

3
1 Introduction

The future of nuclear power is confronted with several challenges, such as societal concerns
about catastrophic accidents (particularly after the disaster at Fukishima Daiichi) or economic
realities related to the high cost of nuclear installations. But perhaps the issue that has over the
last decades triggered the greatest and most consistent public concern is the high-level
radioactive waste generated by nuclear power plants and other nuclear installations (1). The
overarching complication of radioactive waste management is the long-lived toxicity of the
waste, which requires isolating it from the biosphere for many hundreds of thousands of years.
An international consensus has emerged that such isolation can best be provided by disposal of
the waste in geologic repositories, a strategy that today is pursued by most countries in
possession of nuclear waste. It is now widely accepted among experts from national waste-
management organizations and related international bodies—such as the International Atomic
Energy Agency (IAEA) or the Nuclear Energy Agency (NEA) —that the burial of high-level
radioactive waste in mined geologic repositories is technically feasible, and that it can provide
adequate protection to humans and the environment.

This assessment is not necessarily shared, however, by all stakeholders, and issues related to
public distrust, inconsistent policies, or political changes have challenged or disrupted disposal
programs in several nations (2)—such as Canada, the United Kingdom, Germany, and, most
recently, the United States—and no high-level waste disposal facilities are yet in operation
anywhere in the world. Some other nuclear nations, on the other hand, are moving forward as
scheduled with their site selection and site approval process. According to current plans, Finland,
Sweden, and France are expecting the start of geologic repository operations in the 2020–2025
time frame (2).

While the disposal of nuclear waste continues to face technical, social and political difficulties,
existing and emerging nuclear nations do not have the luxury of simply ignoring the issue.
Substantial amounts of high-level waste already exist, and more are being produced.
Transmuting waste in reactors or accelerators could theoretically reduce waste volumes, but
never completely, and the technology for advanced transmutation still needs to overcome

4
considerable challenges. Surface storage of high-level waste can provide a temporary relief, but
has its own environmental (and other) risks, and is certainly no long-term option. At this time,
there is no realistic alternative to long-term waste isolation via geologic disposal.

This article reviews some of the key questions related to disposal of high-level waste in mined
geologic repositories, from a technical and scientific perspective. It starts with an overview of the
nuclear waste disposal problem, including current waste inventory, waste types, alternative
disposal options, and management/institutional issues, followed by a basic description of the
waste isolation capability of mined geologic repositories and their multiple barrier functions. The
article then gives a technical discussion of the physical and chemical processes affecting the
long-term performance of a mined geologic repository, and describes how these processes are
being studied in underground research laboratories (URLs). The final section presents issues
associated with demonstrating safety via qualitative and quantitative assessment of disposal
system performance and the necessity of appropriately accounting for uncertainty.

2 Background

High-level radioactive waste is mainly associated with the generation of nuclear power and, to a
lesser extent, the production of nuclear weapons. With regards to power generation, the high-
level waste is either comprised of spent nuclear fuel or constitutes the radioactive material
remaining after reprocessing of spent fuel (http://www.nrc.gov/waste/high-level-waste.html).
High-level waste is characterized by high levels of radioactivity and the generation of significant
quantities of heat due to decay. Low- and intermediate-level radioactive waste is considered to
be less hazardous and is typically disposed of separately (http://www.oecd-nea.org/brief/brief-
06.html) (3). A separate category used in the United States, transuranic waste, is waste that
contains long-lived transuranic radionuclides but does not generate significant quantities of heat.
There is an operating permanent disposal facility for transuranic waste, the Waste Isolation Pilot
Plant, near Carlsbad, New Mexico. This mined geologic disposal facility is constructed in a
bedded salt formation at a depth of about 650 m (4, 5).

5
2.1 Radioactive Waste Inventory and Nuclear Fuel Cycles

The current inventory of high-level waste may be summarized in terms of the radioactivity
measured in curies (Ci), where 1 Ci is 3.7 × 1010 becquerels or decays per second. The total
amount of commercial spent (or used) fuel worldwide is about 7.6 × 1011 Ci, while the amount of
reprocessed waste is about 2.0 × 109 Ci (6). From these rough estimates, it is clear that most of
the radioactivity (>99%) is in the spent fuel. The quantities of spent fuel are not evenly divided
among the waste-producing countries. About 60% of the total inventory of spent fuel is found in
just three countries, 29% in the United States, 20% in Canada, and 11% in Russia, and 91% of
the existing spent fuel is from ten countries. The worldwide inventory of spent fuel is about
180,000 metric tons of heavy metal (as of 2005) and additional waste is being generated at a rate
of about 10,000 metric tons of heavy metal per year (6).

The radioactivity of spent fuel is highest at the point it leaves the reactor or the nuclear facility.
There are many short-lived radionuclides that decay rapidly, initially causing the radioactivity to
decrease fast. Successively longer-lived radionuclides dominate the total radioactivity as the
waste ages outside the reactor. Figure 2.1 shows that the radioactivity level of the spent fuel is
mainly caused by fission products for the first few hundred years, but is then dominated by
actinides at longer times (7). A similar pattern is found for thermal energy release in terms of
heat generation being dominated at early times by fission product decay and at later times by
actinides (8).

Different fuel cycles result in different treatments for spent fuel. For the “once-through” fuel
cycle (i.e., without reprocessing of waste), nuclear power facilities store the spent fuel on site in
spent fuel pools immediately after removal and often transfer them to dry cask storage after
about 5 years of cool-down (http://www.nrc.gov/reading-rm/doc-collections/fact-sheets/dry-
cask-storage.html). The waste is expected to remain in such interim storage for several decades,
covering the period of highest toxicity and heat generation, before placement into a permanent
disposal facility is envisioned. While disposal will always be needed at some point in the future,
the concept of extended storage for durations of 60 to more than 100 years has been suggested as
an alternative to developing permanent disposal systems in the near term. Additional R&D
would be required to investigate the feasibility and safety of extended storage beyond 60 years of
disposal system lifetime (9).

6
Alternatives to the once-through cycle involve some type of reprocessing of spent nuclear fuel
(10). The only alternative currently used commercially is the modified-open cycle using mixed-
oxide fuel. More advanced fuel cycles have been proposed, but need to overcome technical and
economical challenges, and are decades away from commercial implementation. In theory, these
fuel cycles could significantly reduce the volume of high-level waste and could eliminate certain
actinides in the waste, but the need for long-term waste disposal would still remain (10).

2.2 Alternative Geologic Disposal Options

A wide variety of long-term disposal options have been considered over the past decades,
including methods that are not related to geologic disposal, e.g., shooting the waste into space
(11), but most of these have been dismissed based on safety concerns, cost, or other fundamental
problems. The only options considered potentially viable today are all based on geologic
disposal. Of these geologic disposal options, recent evaluations (or re-evaluations) by various
countries and international organizations continue to support mined geologic disposal as the most
practical and effective method for permanent disposal of high-level waste (10, 11, 12, 13). Basic
design and safety considerations for mined geologic repositories are described in Section 3.

The closest alternative to mined geologic disposal is to place the waste into boreholes of up to
5 km depth. Recent expert reviews found deep borehole disposal to be technically promising (10,
11), but also recognized that the concept has not been investigated as thoroughly as mined
geologic disposal, and requires further research and development. Other geologic disposal
options that have been discussed in the past, such as placement of waste on or within the ice
sheet in Antarctica (14), subseabed disposal, and burial beneath islands (10, 11, 12, 15). These
latter alternatives have not been pursued, mainly because of difficulties concerning international
policies and treaties, opposition from environmental organizations, and public acceptance issues.

2.3 Governance and Management

Institutional processes are important in the development of high-level waste disposal systems.
Given the natural tendency for the societal cost and impact of any waste to fall outside the
purview of normal market economic forces, governance is needed to bring these costs and
impacts into balance. Governance for radioactive waste is generally implemented through a

7
governmental regulatory process, which has been followed by all countries faced with high-level
radioactive waste disposal issues (16). Some differences among nuclear nations exist and are
mainly in the details of how the regulations have been written, in terms of what levels of dose
and risk rates, as well as the time frame of the regulations (Section 5.1).

The implementing organizations tasked with management of nuclear waste disposal around the
world are more diverse than regulatory bodies. Generally speaking, the implementing
organizations fall into one of four categories: government agencies, private corporations,
government-owned corporations, and public-private partnerships (16). Implementing
organizations are faced with an array of difficult issues: (a) complex technical issues that carry
large degrees of uncertainty and significant short and long-term risks; (b) a high degree of
political and public controversy that makes for difficult relationships with stakeholders; and (c)
difficulty in maintaining stable financing and resource support, political legitimacy, and
structural integrity over the time required to develop a permanent disposal system (likely to be
close to 100 years) (17). To address these issues, management organizations are advised to (a)
implement a decentralized authority structure; (b) openly acknowledge uncertainties; (c)
proactively interact with affected localities, institutions, and the public; and (d) ensure that
funding is segregated from general government revenues (2, 17). Part of the management
challenge is to retain public confidence during the inevitably long process of site selection, site
characterization, and repository development. A major step toward achieving such confidence
may be through consensus-based approaches, which have been successful in Finland and Sweden
(10).

3 Mined Geologic Disposal

In a mined geologic disposal system, waste is placed in mined cavities or tunnels approximately
250 to 1000 m below the surface (http://world-nuclear.org/info/inf04ap2.html). A fundamental
premise for the disposal system is that it is intended to operate independently without relying on
monitoring or institutional controls (although the system may be operated with such controls for
a limited time after closing the repository) (18).

8
The long-term safety of a mined geologic repository is based on the containment and long-term
isolation provided by a multibarrier system consisting of engineered and natural barriers (and
their sub-barriers). The basic idea is that any one barrier may be insufficient to ensure the
necessary degree of containment and isolation and that other barriers need to be able to
compensate and provide additional insurance (18). The disposal system is built to maximize the
redundancy of barriers to ensure that the performance of the disposal system is acceptable even if
some components of the repository system fail to perform as expected.

Engineered barriers are designed to completely contain the waste initially, during the period of
highest toxicity and greatest public concern. Since no engineered barrier can guarantee complete
containment over the extraordinarily long time during which some components of the waste
remain hazardous, limited releases of radionuclides from the engineered containment system are
to be expected at some point far in the future. It is at this point that the natural barriers
surrounding the geologic repository combine with the engineered barriers to attain the desired
performance. The main functions of the natural barriers are to shield the repository from changes
in the geologic environment, promote performance of the engineered barriers, delay radionuclide
transport, and allow for dilution of radionuclide concentrations to limit exposure and ensure
safety (18). As an example, Figure 3.1 illustrates the main barriers in the Swedish program for
high-level radioactive waste disposal.

3.1 Engineered Barriers

The engineered barriers of a mined geologic repository are comprised of a number of

components that may be identified in terms of their safety function. The main components are
the waste form, the waste canister, and backfill material, as well as tunnel seals and plugs.

The main safety function of the waste form is to provide structural stability and resistance to
waste dissolution, slowing the release of radionuclides into the aqueous phase (19). Spent fuel is
in a solid form and is typically encased in zircaloy or stainless steel cladding (13, 19). Fuel
cycles using reprocessing produce liquid wastes that need to be stabilized in a solid form for
disposal, often using materials such as borosilicate glass, ceramic, and glass ceramic (20).
Numerous other waste forms also exist, although they represent smaller quantities of waste.

9
The safety function of the waste package is to contain the waste, prevent its contact with
groundwater, and delay radionuclide transport for as long a period as possible. Two safety
strategies for the waste package have been proposed. In one strategy, the container material is
not corrosion resistant (e.g., carbon steel) but has sufficient thickness to substantially delay
releases before corrosion penetrates the package. (The corrosion products can also play a role in
retarding radionuclide transport following failure of the package.) The second strategy is to use a
corrosion-resistant material such as copper or nickel-based alloys. In some nations, such as
Canada, both are implemented by using a carbon-steel inner container and a copper outer shell
(19).

Another engineered-barrier component included in many waste disposal programs is a backfill

(or buffer) material that fills the space in the repository openings between the waste package and
the rock. The backfill provides mechanical support to the repository excavation and limits
rockfall damage to the waste packages, blocks preferential flow and transport pathways along the
excavation, and retards radionuclide movement within the repository tunnels. In most geologic
environments, the backfill material is a bentonite or a bentonite-sand mixture. In rock salt, the
backfill may be made of crushed salt. Tunnel and shaft seals and plugs provide similar safety
functions in the repository excavations between waste emplacement locations; they are often
made of the same materials.

3.2 Natural Barriers

The geologic environment in which the repository is located comprises the natural barrier. Key
characteristics for a successful natural barrier are (13):

 Long-term (millions of years) geological stability, in terms of major earth movements

and deformation, faulting, seismicity and heat flow;
 Low groundwater flow at repository depths, which can be shown to have been stable for
periods of at least tens of thousands of years;
 Stable geochemical or hydrochemical conditions at depth, mainly described by a reducing
environment and a composition controlled by equilibrium between water and rock
forming minerals;
10
  Good engineering properties, which readily allow construction and operation of a
repository.

The rock in which waste emplacement occurs is referred to as the “host” rock. A number of host
rock types have been investigated for repository development in various nations. These generally
fall within the broad categories of crystalline rock, clay-based rock, salt, and volcanic rock.
Natural barriers ensure isolation of the waste and provide a stable predictable environment.
Isolation is needed to prevent human contact but also to shield the repository from fluctuations in
climate and related effects on groundwater flow/composition and erosion/deposition of
sediments. Isolation also concerns the flow rate and velocity of groundwater movement through
the geological environment. Since transport with flowing groundwater is the main mechanism
for radionuclide migration to the biosphere, the magnitude of flow (a) impacts the rate of
radionuclide release from the repository and (b) the rate of arrival of radionuclides at some point
in the environment that leads to human contact.

Some host rock types limit water flow as a result of very low rock permeability, which is
typically true for many clay-based rocks and salt. In other cases, such as some crystalline and
volcanic rocks, permeability may be higher because of natural rock fractures, but the hydraulic
driving forces may be limited, again resulting in low flow rates. The mineralogy of the rock and
the aqueous geochemical conditions also play an important role through chemical interactions
between the rock, water, and radionuclides. These interactions can result in radionuclide
partitioning between the mobile aqueous phase and the immobile rock surface, and cause a
substantial delay in the rate of radionuclide migration, even for systems with significant flow.

3.3 Retrievability

An important design aspect of mined geologic disposal is the question of retrievability of the
waste during the first decades or centuries after emplacement. There are two main reasons for
wanting the option of retrieving the waste: (a) to reverse storage if performance of the repository
is found to be unacceptable, and (b) to allow for recycling of the waste if advanced technologies
in the future make this possible. To some extent, retrievability is at odds with the reason for
disposal, which is to isolate the waste. Retrievability for mined geologic disposal varies

11
according to the host rock type. Hard rock such as granite tends to allow for easier retrievability
than soft rock like salt (21). Retrievability is also affected by the engineering design of a mined
geologic repository—for example, the type of tunnel support and the use of backfill.

4 Physico-Chemical Processes Affecting a Mined Geologic Repository and

their Studies in Underground Research Laboratories

At its core, the development of a geologic repository for the disposal of radioactive waste is a
scientific and engineering problem. Public acceptance and regulatory requirements rely on an
adequate scientific understanding of the physico-chemical processes within and around a waste
repository, and a detailed knowledge of the geological characteristics of the repository site. As
mentioned earlier, the design of a waste repository is expected to be such that it allows
emplacement of waste into the repository with minimum risk, and that it provides waste isolation
for hundreds of thousands of (and even up to a million) years into the future. These are unique
expectations, far beyond what are expected of human-constructed engineering structures to date.
Locating a repository in deep geologic formations takes advantage of the in situ conditions
having geologic time scale. For example, at depths of 500 m, the age of the groundwater could
be hundreds of thousands of years, clear evidence that groundwater flow at the site is very slow.
The approach adopted by waste management organizations internationally is a systematic and
careful investigation of the physical and chemical processes that may occur within and around a
repository—from its initial stage of construction, to the time of waste emplacement, all the way
to the time period long after the repository closure. These processes are evaluated based on field
studies, laboratory research, and mathematical modeling, in order to assess their impact on the
isolation performance of the repository far into the future.

Efforts along these lines in the countries planning for radioactive waste repositories have been
substantial over the last 30 years, and often involve major underground research laboratories
(URLs). URLs are systems of underground rooms at the anticipated repository levels (200–500
m in depth) that have been mined through tunnels or shafts. These URLs allow direct
observations of conditions and processes at these depths and provide opportunities to conduct
full-scale tests to evaluate the physical and chemical responses to repository construction, heat

12
release, and chemical migration of emplaced waste, as well as to long-term natural events, such
as seismicity and glaciation on the land surface.

The safety of a repository depends on various factors: (a) in the early stages, those factors
affecting the stability of the waste form and the durability of the waste packages; (b) at later
stages (when barriers may have been breached), those factors affecting the transport of
radionuclides by flowing groundwater, from the engineered barrier into the host rock and
eventually to the biosphere. The latter factors, which are mostly of a hydrogeologic nature, are
the main focus in this section. We first present relevant physico-chemical processes that may
occur during the lifetime of a nuclear waste repository and explain how these can affect the
hydrogeologic characteristics of the host rock. We then introduce some major currently active
URLs to convey the level of scientific efforts being applied in a number of countries.

4.1 Four Stages in the Development of a Geologic Repository

The development of a geological repository for radioactive waste can be divided into four
periods (22). The first period of repository development may be defined as the repository
construction stage, which extends from tunnel excavation to a few days after tunnel lining and
ground support installation (if necessary for tunnel stability). The second period, called the open
tunnel stage, lasts until emplacement of waste and backfill materials. Then there is the third
period, named here the exploitation stage, which marks the start of the strong perturbation due to
the decay heat generated by the radioactive waste. This stage lasts until a few hundred or
thousand years after repository closure, when the waste heat has greatly decreased and its effects
become less significant. Finally, the fourth period is the long-term postclosure stage, which is
the main period of concern for long-term performance and safety assessment. It is typically
during this period that the integrity of the engineered barriers becomes questionable and the
potential release of radionuclides from waste canisters must be considered. Key issues are then
the migration of radionuclides through the backfill materials and subsequent transport in the host
rock. The isolation capacity of the geologic repository for the long term and on the large scale
depends on (a) the initial hydrogeological characteristics of the host rock, (b) the potential
perturbation of these host-rock characteristics as a result of near-field processes and effects

13
occurring during the previous three stages, and (c) other changes in host-rock characteristics due
to natural phenomena and events, such as seismic activity or changes in climate.

The key processes during these four stages of geological repository development are discussed in
the following subsections, in the context of the different rock types considered suitable for
hosting a geologic repository. Crystalline rocks or hard rocks, such as granite, have been selected
in Finland, Sweden and China; indurated (hardened) clays are being investigated in France and
Switzerland; and plastic clays are being studied in Belgium. Furthermore, in Germany, Japan,
and the United States, salt is being considered as one of several host rock alternatives.

In a hydro-mechanical sense, crystalline rock and plastic clay or salt form the end members of
various material classes. While crystalline rock is nearly perfectly brittle material, plastic clay
and salt can be described as ductile or visco-elastic-plastic material, the latter having the
important property that any fractures formed can seal in time through rock creep and
deformation. However, near the walls of ventilated underground openings, brittle behavior is
also observed in plastic clays due to suction conditions. Indurated hard clays and sedimentary
rock are transitional materials, which are brittle at low water content, but ductile at high water
content. These differences in geomechanical behavior are important to repository safety, because
they influence the extent to which new transport pathways for radionuclides may be created near
repository tunnels as a result of structural damage from tunnel construction, ventilation, and
waste emplacement.

Key Processes During the Construction Stage

The construction stage represents a major perturbation of the rock formation near emplacement
tunnels, with possible creation of new openings (fractures) in the subsurface rock and significant
changes in the hydro-mechanical conditions on the new rock walls. The stress field in the rock is
redistributed around the tunnels, and the tunnel wall surface is moves inward until restrained by
tunnel lining and support (in the case of soft rock), or until it stabilizes by its own strength (in the
case of hard rock).

Basically there are two sources of rock damage that may occur near the new rock walls. First,
there is the potential for damage caused by the excavation method itself, and second, there are

14
mechanical changes caused by stress redistribution around the newly excavated opening. In
addition, there are the effects of back pressure on rock deformation by emplacement of tunnel
lining and support (23). For hard and brittle crystalline rock, the excavation activity could by
itself induce significant damage, depending on the excavation method used. For example, if drill
and blast methods are used, an excavation damaged zone (EDZ) could extend from 0.1 m to as
much as 1.5 m into the rock, increasing its permeability to flow by two or three orders of
magnitude in this region. In contrast, direct excavation damage is not so significant in other rock
types, especially when a tunnel-boring machine is used.

The stress redistribution around a newly excavated tunnel induces a multiprocess coupled
response in the rock (24, 25, 26). The resulting rock movements cause a volumetric deformation
of the pore space that, when coupled with the low rock hydraulic conductivity, gives rise to
strong pore-water-pressure variations that will then dissipate with time. These pore-pressure
changes have a significant influence on the effective stress state in the rock and may contribute
to rock failure processes. Creating the underground opening also influences the hydrological
properties of the clay rock. The region adjacent to the tunnel rock may change from water-
saturated to unsaturated conditions (and suction can develop), which can change the behavior of
clay rocks from plastic to brittle (27).

Key Processes During the Open Tunnel Stage

This stage may be defined as the period between the completion of excavation and lining
installation and the emplacement of waste and buffer. It may last from a few months to a few
decades. During this period, the rock wall is in contact with the atmosphere in the (typically
ventilated) tunnel. The atmosphere, with its generally lower humidity, imposes a new hydric
condition at the tunnel wall or the lining, with its own permeability and air entry pressure, thus
changing the local effective stress. Importantly, this condition is not a simple equilibrium
between the relative humidity in the air and in the host rock, since the hydric transfer processes
depend on several other factors such as air velocity and skin behavior.

In the case of crystalline rock, while tunnel ventilation and temperature changes probably do not
have a strong effect on mechanical properties, air entering the rock results in an oxidizing, two-
phase flow condition in the EDZ. This oxidizing condition may cause chemical and biological

15
activities and a possible clogging effect, whereas the two-phase flow implies a reduction of the
effective water-permeability values.

In clays (and possibly also in salt), where mechanical properties are sensitive to moisture
content, the effects of dynamic changes in humidity and temperature due to tunnel ventilation are
larger than in crystalline rock. These changes cause rock desaturation, which, in turn, gives rise
to capillary forces and hence an increase in clay cohesion and strengthening. At the same time
desaturation may increase tensile stress (as a result of shrinkage) and the potential for bond
failure, and may change the degree of property anisotropy. Furthermore, increases in humidity in
the tunnel air can cause clay swelling and rock softening in the rock next to the wall. The former
may increase compressive stresses, leading to additional damage to the softened rock. Cyclic
seasonal changes in humidity and temperature over a number of years have generated
discontinuities in the vicinity of the underground research tunnel at Tournemire in France (28,
29, 30). On the other hand, as time goes by these discontinuities may close and seal due to clay
deformation and moisture-induced swelling, which can lead to a reduction in permeability to that
of the undisturbed formation. A number of studies have been conducted to investigate swelling
behavior and permeability evolution in clays (31, 32, 33, 34).

Key Processes During the Exploitation Stage

At this stage, waste emplacement represents the start of a period of thermal perturbation, backfill
will be in place, and the repository will be closed. Decay heat from radioactive waste diffuses
away through the backfill and the near-field rock to the far field. Groundwater will flow from the
far-field host rock towards the emplacement tunnels, leading to resaturation of the backfill and
the EDZ. This is a slow process governed by low rock permeability, and will not be uniform over
the repository domain. Resaturation is also affected by heat release from the emplaced waste,
which will tend to dry the backfill and rock close to the waste canisters, and cause the vapor to
flow outward and condense in the cooler region away from the heat source. These transient
changes in water content, associated with the interplay between resaturation and temperature
gradient, may have a significant dynamic impact on rock properties, especially in clays and rock
salt, where the rock strength is a sensitive function of water content (22). Heating will also
generate transient pore-pressure buildup resulting from differential thermal expansion, leading to

16
changes in effective stress state. Studies addressing heating effects using laboratory
measurements and field tests include those by Delage et al. (35), Zhang et al. (36), and Gens et
al. (37).

Key Processes During the Post-Closure Stage

A few hundred to a few thousand years after the closure of a nuclear waste repository, the heat
generation from the waste will be much reduced. In repositories that use bentonite-based
components as backfill material, a return of pore-water pressure to the original hydrostatic level
will result in swelling of the bentonite. Major processes during this stage include self-sealing of
fractures in clay and salt rock and bentonite materials, long-term chemical and biological
degradation of backfill and rock materials, and the degradation of the tunnel support system.
There is also concern that chemical reactions (especially those involving canister and rock
support materials) may produce gases, the pressures from which could build up in the canister
deposition holes and create stress-induced damage. The effects of gas production and pressure
buildup on radionuclide transport have recently received considerable attention (38, 39, 40, 41,
42).

It is during this stage, perhaps after tens to hundreds of thousands of years, that the integrity of
the waste form and the canisters may be breached and radionuclides might dissolve into the
groundwater present near the canisters. Radionuclides need to first migrate through the backfill,
where chemical retardation and sorption onto clay materials can significantly retard the transport.
The retardation mechanisms depend significantly on the characteristics of the backfill that has
undergone some physico-chemical changes due to heating from the radioactive waste and
subsequent cooling. Then, as the released radionuclides migrate into the near field of the
repository, the properties of the EDZ play an important role, in particular raising the question of
whether connected flow paths have been generated. Radionuclides that emerge from the EDZ
represent the source term for migration into the far field rock. Flow and transport in the far field
involve a number of processes, which, for example, include flow channeling, dispersion, matrix
diffusion and sorption in a host rock in which fracture pathways are present. Some of these
processes retard solute transport, while others cause preferential flow faster than the average

17
flow rates. A thorough understanding of these processes is essential for estimating the far-field
migration of radionuclides, which in turn determines possible radiation dose or health risks.

All these factors have to be investigated in the context of possible long-term changes in the
natural environment of the repository system, such as from glaciation and climatic change. For
example, in Canada and Scandinavia, one would anticipate glaciation, which has a cycle of about
20,000 years and produces an ice sheet of 1–4 km thick weighing on the land surface. Other
events to be considered over long time periods may have low probability but high impact. These
are generally disruptive events such as seismic or volcanic activity, which have been included,
for example, in the long-term safety assessment for the Yucca Mountain site in Nevada, United
States (43, 44). Regarding climatic change and seismic events, there is some thought that perhaps
their impact at repository depth may be limited, but nevertheless these possibilities must be
evaluated.

4.2 Underground Research Laboratories

In this subsection, we shall present currently active URLs, which are being utilized in many
countries to study various processes of importance described in the last subsection. The need for
URLs was identified in the early years of research into geologic disposal. A first example is the
Stripa Project (45), which was an international collaborative program of research in an
abandoned iron mine in Sweden carried out over the years 1977 to 1992. The results of the
project significantly advanced the state of science of fracture hydrology and rock
thermomechanics, and demonstrated the value of concerted underground research at, or close to,
the anticipated repository level. The success of the Stripa Project led to later development of
other URLs specifically constructed for waste repository research. Among these were some
major URLs no longer active today, such as the Exploratory Studies Facility (ESF) at Yucca
Mountain (46). The ESF was distinct from other URLs in that the volcanic tuff unit proposed as
repository host rock resides in the vadose zone above the groundwater table. Since the main
transport mechanism for radionuclides from a repository is through water flow, a partially water-
saturated site has certain advantages (47), but also has its own set of processes, which were
systematically evaluated in the URL (43).

18
A list of major URLs with ongoing R&D is given in Table 4.1. These URLs are located at depths
from 240 to 500 m. Some have been initiated in the 1980s, but still maintain very active research
programs. Most are generic URLs (i.e., located at a site with representative rock, but not
designated to host a repository) while others, such as the Onkalo URL in Finland (48, 49) or the
Bure URL in France (50, 51) are at a potential repository location. The latter site-specific type
serves not only a research purpose, but is often expected to have a prototyping and performance
confirmation function, meaning that full-scale mock-ups of an eventual repository tunnel are
built underground for early testing of engineered and natural barrier components. The formations
being studied range from crystalline rock like granite to very soft plastic clay rock like Boom
Clay at Mol, Belgium. Among URLs of the same rock type, there can be significant differences,
such as the age of the formation (ranging from 30 million to 1.5 billion years), the salinity of
groundwater, the fracture density, and the occurrence of major fracture and fault zones. In
general, the objectives of URLs may be summarized as follows:

 To compare the data on rock properties and in situ processes obtained from geophysical
surveys and surface-based boreholes with those from direct measurements in the URL at
depth. To verify that investigations from the ground surface and boreholes can provide
sufficient data concerning key safety-related properties of the rock at repository level.

 To develop and verify the methods and technologies needed for site characterization
at depth.

 To study deep rock responses to excavation and to heating from the waste, and to study
radionuclide migration processes in the host rock. To develop and verify numerical
models for description of groundwater flow, mechanical deformation, heat transfer,
radionuclide migration, and evolution of chemical conditions during operation of a
repository and after its closure.

 To test and demonstrate the technology for carrying out repository construction, waste
emplacement, backfill, and repository closure procedures, to ensure that all the
components function at a high level for maintaining repository short-term and long-term
safety.

19
In addition, URLs provide a way for international collaboration with fruitful cross fertilization of
ideas and methodologies (52). URLs are also critically needed by training personnel for
conducting the various tasks in repository development, including repository design,
construction, field tests, and modeling for safety assessment. Below, we shall give a brief
description of two selected URLs to illustrate the scope and effort of URL research.

The first is the Grimsel URL, situated in crystalline rock in the Swiss Alps at a depth of about
400 m in the granite and granodiorite of the Aar Massif formed some 300 million years ago. The
tunnel system at the site, over a kilometer in length (Figure 4.1), has been used for a large
number of field experiments since the beginning of its operation in 1984. Some of the
experiments are indicated in the figure. In the current phase, the research program at Grimsel
covers about 10 field tests and involves 17 research organizations from 10 countries as well as
the European Union (53).

Of particular interest among the field tests conducted at Grimsel is the FEBEX experiment, an in
situ full-scale heater experiment evaluating the thermal-hydro-mechanical (THM) behavior of
the geologic repository (54). In this test, two heaters, used to represent the heat-releasing
radioactive waste, were placed horizontally into a tunnel and surrounded by a swelling clay
buffer constructed of compacted bentonite blocks. The two heaters were turned on in 1997 with
the temperature maintained at 100°C at the bentonite-heater interface. In 2002, one of the heaters
was removed, while the other has continued its operation to date, making it the longest
experiment of this type. All through the years of operation, hundreds of sensors have provided
detailed physical and chemical data on the bentonite and rock. Furthermore, samples of
bentonite, materials, and used instruments from the area of the first heater, which was removed
in 2002, were studied and analyzed (55). The results form a very valuable data set regarding the
transient behavior of buffer and rock materials under natural conditions for over a decade. The
current plans are to continue the heating and monitoring of the second heater until 2014, and then
it will also be excavated in 2015.

Another series of experiments at Grimsel, conducted as part of the colloid formation and
migration project (CFM), seek to better understand transport mechanisms in a fracture shear zone
that could potentially form a fast flow path in the host rock (56). Radionuclides or other

20
surrogate species were released into the shear zone, and then their arrival was measured at some
distance away from the release point to measure the “breakthrough” as a function of time. Such
tests were performed for different types of radionuclides, for sorbing and nonsorbing tracers, for
transport with and without colloids present. Current CFM experiments directly evaluate the
performance of backfill material with respect to swelling, erosion, and colloid generation by
emplacing a bentonite plug into a borehole completed in the fracture shear zone.

The second URL example is the HADES (High Activity Disposal Experimental Site) excavated
at 223 m depth in Boom Clay, a tertiary clay formation in Mol, Belgium. Since its construction
in 1980, many experimental investigations have been conducted at the site. Figure 4.2 (top)
shows the construction history of the HADES facility (57). The main tunnel is about 200 m in
length, with an internal diameter averaging about 4 m, from which experiments were conducted
in boreholes and side galleries. An extensive summary is provided by Bernier et al. (26).

One current focus of work in HADES is the Praclay Gallery constructed in 2007, in which a Seal
Test and a Heater Test were initiated in 2010 and 2011, respectively. The layout of the gallery is
shown in Figure 4.2 (bottom). The Heater Test involves heating a 30 m long gallery section for
10 years with many monitoring sensors, for the purpose of investigating the THM behavior of
plastic clay under the most hydromechanically challenging conditions occurring around a
repository (58, 59). To this aim, the heated section of the gallery was fully saturated before
starting the heating, and a hydraulic seal was installed to separate the heated from the unheated
sections of the gallery. This installation makes up the Seal Test, which was conducted in 2010,
and enabled testing of hydraulic seal functionality under heated repository conditions.

As can be seen from Table 4.1 and the above discussions, a considerable amount of scientific
work has been conducted over the last 30 years or more to study various physico-chemical
processes potentially occurring within and around a radioactive waste repository. Constructing a
URL and performing various experiments in a URL are major investments. As a result,
significant advances in geosciences and engineering have been made. These advances not only
benefit repository construction and its safety assessment, but also represent major progress in the
scientific fields of hydrogeology, hydrogeochemistry, geophysics, rock mechanics, and geology.
The problem of geologic repository development, with the extraordinary requirement of being

21
able to make convincing predictions many years into the future, is extremely challenging. There
are no alternatives to conducting systematic and careful scientific and engineering research,
which includes not only research in URLs, but also studies of natural analogues (60, 61). This is
indeed happening in many of the countries with radioactive waste management programs today.

5 Recent Trends in Assessing the Long-Term Safety of Mined Geologic

Disposal

This section describes the current practice and recent trends in assessing the future safety of a
geologic disposal site, in order to provide a basis for regulatory (and societal) approval. As
discussed in the previous section, any safety assessment needs to start with an adequate technical
understanding of how the main waste-isolation barriers behave. Based on a solid technical
foundation, a safety case can then be made, in which various lines of evidence are presented to
quantify and substantiate that the repository will be safe over a long time period, and that the
predicted performance meets regulatory standards.

5.1 Safety Standards for Repositories

While management of high-level radioactive waste is a national responsibility, several

multinational organizations have played an important role in establishing a consensus on the
principles of waste disposal and harmonizing safety standards and best practices for safety
assessments among different countries. The most important international bodies are the
International Atomic Energy Agency (IAEA) and the Nuclear Energy Agency (NEA). In part
because of the consensus building of the IAEA and NEA, there is general agreement
internationally on the fundamental principles and ethical duties of radioactive waste
management, which require, for example, that future generations shall not shoulder an undue
burden from today’s waste and shall not be exposed to greater levels of radiation than are
acceptable today (62, 63).

Differences remain between national waste-management policies regarding the specific safety
standards that should be used to judge the suitability of a geologic repository, as well as the
methods for demonstrating compliance with the standards (2, 63). Regulations typically define as

22
primary safety standards a certain level of radiation protection for a “representative” person
living in the vicinity of the repository under defined conditions (such as the consumption of a
defined amount of groundwater from the area). Radiation protection is either defined as
maximum radiation dose or is given in terms of health risk limit, or both. The main health risk
associated with nuclear waste disposal is the potential of long-term low-level radiation exposure,
which increases the risk of cancer or genetic mutation (65, 66). The health-risk limit is usually
defined as the maximum likelihood of the representative person suffering health effects during
his or her lifetime, measured in terms of probability per year. Dose and health risk are connected
via “dose-to-risk” conversion factors. There is ongoing discussion on the uncertainty of “dose-to-
risk” factors, caused by the difficulty of assessing health effects resulting from chronic radiation
exposure (21, 67), and it has also been pointed out that health impacts might change in the not-
too-distant future because of the expected progress in preventing and curing cancer (68).

Today, most international programs have converged on comparable dose constraints and risk
limits as safety standards, to be demonstrated for time periods ranging from 10,000 years to a
million years. Some countries have different safety and performance standards for the first
several thousand years of repository lifetime compared to the remaining time period. Dose
constraints for the earlier time periods vary between 0.1 and 0.3 millisievert per year, while risk
limits vary between 10-5 and 10-6 per year. In comparison, the global annual average effective
dose from typical natural background radiation is about ten times higher, at approximately 2.4
millisievert (66), and the risk of 10-6 per year is about twice the risk of being struck by lightning
(21). This immediately invites the question (not discussed in this paper) as to how the public
perceives different types of risk and how public acceptance might be improved by better risk
education (69, 70, 71). In the context of providing multiple lines of evidence for the safety case
(see next section), the value of additional safety indicators—complementary to the regulated
safety standards—has received increasing attention (72). Examples of useful additional
indicators are quantities describing the performance of sub-systems, such as the longevity of an
engineered barrier or the magnitude of groundwater flow through the EDZ.

23
5.2 The Safety Case

Organizations responsible for implementing waste-management solutions generally present a

safety case to the regulators or other decision makers at specific points in the process of
repository development (72, 73). What is meant by safety case is a synthesis of evidence,
analyses, and arguments to quantify and substantiate that a repository will be safe over a long
time period. A central element of the safety case is a quantitative assessment of the future
performance of the repository system, via comprehensive modeling studies, followed by a
comparison of the results with the regulatory standards (see Section 5.3). The guiding principles
behind a safety case have evolved over the last decade, as it was realized that building
confidence in long-term safety assessments requires more than just the results of quantitative
performance predictions. The safety case, as it is understood and accepted today, needs to also
present independent lines of (sometimes qualitative) evidence and reasoning to support the
quantitative assessment (and the assumptions made therein), and to provide decision makers with
a basis for judging whether sufficient confidence exists in the performance evaluation (73).
Independent lines of evidence may take the form of a combination of alternative predictive
models, may involve broadbrush “insight” models based on first principles, or may utilize
independent results from natural analogues (21). For example, principles of thermodynamics
may guarantee the stability of copper (a canister material in some designs) in deep anaerobic
groundwater. Or the existence of ancient natural uranium deposits may provide evidence that
radionuclides can be stable in a geologic environment over millennia (72).

Typical safety cases will start with a description of the overall safety strategy for the repository,
followed by an explanation of the assessment basis, a demonstration of evidence, analyses, and
arguments, and finally a synthesis of results (see Figure 5.1). Guidance documents call for
transparency in the presentation, traceability with respect to all key assumptions, flexibility in
planning and implementation, and openness regarding remaining questions (72, 73). Chapman
and McCombie (21) compiled key principles of safety cases that help reduce uncertainty and
increase confidence, as follows:

24
 Use robust designs and analyses
 Aim for simplicity
 Apply good science
 Use a structured approach, including iterative assessments
 Use multiple lines of reasoning, a range of models, and natural analogues
 Document the elicitation of expert judgment (74)
 Perform quality assured analyses and have these peer reviewed

Most of these principles are straightforward, but the first two items deserve further discussion.
Robustness refers to the repository system itself but also to the analyses used for assessing
safety. A robust repository system starts with a stable geological environment in which key
characteristics are unlikely to change significantly over long time frames. Robustness extends to
the engineered barrier components that need to be well understood, tested, and resistant to
degradation under a broader range of conditions than expected in the geological environment.
Robustness is also related to the degree of redundancy achieved by having multiple barriers
operating in concert to isolate the waste and prevent/delay radionuclide transport release to the
biosphere. It is important that such complementary safety function, and the level of redundancy
as a function of time, is demonstrated by conducting quantitative safety assessments for specific
scenarios in which barriers are assessed individually (see following section). Robustness and
simplicity are closely associated. Aiming for a repository site with simple geology, physics,
chemistry, and design enhances transparency of and confidence in the safety case, and will
furthermore allow for less complexity (and more robustness) in the safety analyses.

5.3 Quantitative Safety Assessment

What is meant with quantitative safety assessment is a “quantification of the overall level of
performance, analysis of the associated uncertainties, and comparison with the relevant design
and safety standards” (18). Quantitative safety assessments are an important contribution to the
development of the long-term safety case for deep geologic disposal of radioactive wastes. In
addition to contributing to the safety case for a repository, quantitative safety assessments also
aid in identifying key processes relevant to safety and focusing research, provide an

25
understanding of the value of individual barriers, and allow comparison between alternative
waste-management options (18). Note that we use in this paper the term “safety assessment” for
other labels sometimes used in the literature (e.g., “performance assessment”) (75, 76).

A quantitative safety assessment generally needs to provide answers to the following questions:
(a) What are the performance-relevant features, events, and processes (FEPs) that need to be
considered? (b) How likely are these? (c) What are their consequences? (d) How reliable (or how
uncertain) are the answers to the first three questions? A typical workflow starts with a
systematic selection and analysis of FEPs that can influence performance of the specific site,
usually starting from the review of comprehensive generic FEPs databases (77). The next step is
to develop the conceptual and mathematical models for predicting individual components of the
repository system. After the component models (or submodels) have been tested to establish
confidence and parameter ranges (including establishing their uncertainties), they need to be
linked with each other in a system-level model of the entire repository. The system-level model
can then be applied to determine the behavior of the whole system under consideration. Because
different sources of uncertainty need to be accounted for—scenario uncertainty, conceptual
model uncertainty, and parameter uncertainty—this final step generally involves some
quantification of uncertainty, usually conducted by repeating simulations over a sufficient
number of alternative scenarios, conceptual models, or parameters.

Review of the recent literature suggests consistency between international programs on basic
principles and methodologies of safety assessments (73, 78), but also reveals a few areas where
open questions remain and discussion is ongoing. Below, we briefly touch on two of these areas:
(a) how the fidelity of the system-level model and their underlying submodels can be ensured,
and (b) how uncertainty and sensitivity should be treated.

The overall fidelity of a safety assessment model starts with the fidelity of the individual
submodels that support the overall system level model. These are typically detailed process-
driven models, which need to be thoroughly tested and validated against experimental data. As
mentioned in Section 4.2, URLs offer valuable data for model testing, because experimental
studies of the expected conditions in emplacement tunnels and the surrounding rock can be

26
conducted at or near full spatial scale. Time, however, is another issue. No experimental study
can even marginally approach the long time periods that quantitative safety assessments need to
cover. Confidence in geologic modeling over long time scales can be gained from natural
analogues (60). An example of a natural analogue is the Peña Blanca uranium ore deposit in
northern Mexico, which contributed to a better understanding of the long-term transport
processes and of the long-term stability of minerals in arid regions similar to the Yucca
Mountain site (61). Almost all national disposal programs are therefore involved in natural
analogue studies and use data—often qualitative, sometimes quantitative—from analogues to
support model predictions.

For use in system-level safety assessment models, the results from individual component models
often need to be simplified, in a process sometimes referred to as model abstraction. It is
important in the abstraction process that the simplifications and assumptions made are realistic
and appropriate, and that the uncertainties of the primary models are represented by their simpler
surrogates (43). To ensure appropriate abstraction, the simplified models need to be carefully
compared to the primary models, and the remaining differences need to be evaluated in the
context of the overall system-level impact. Another concern for the fidelity of a system-level
model is that output from one submodel is used as input for another, which can lead to
consistency problems (79). The trend in recent years has been to reduce the number of
submodels and the number of abstractions in a system level model. For example, advances in
computer power may now allow simulating the near-field and far-field processes in one detailed
model rather than a sequence of submodels, or may allow use of certain primary models in a
system-level uncertainty analysis without simplification or abstraction. The distinction between a
system-level model and a detailed process model is becoming increasingly blurry and will
continue to do so in the future (79).

With regards to addressing uncertainty in long-term performance predictions, national disposal

programs all agree with international guidance that uncertainty is an “unavoidable aspect of
planning and regulating deep disposal programs” (21), but they differ considerably in how
uncertainty is to be assessed in a safety assessment (2). Probabilistic safety assessments are
evaluations in which the uncertainty of scenarios or parameters is accounted for explicitly by

27
assigning probability values/distributions and conducting performance predictions over the
whole uncertainty space, thus assessing “global” uncertainty. Deterministic uncertainty
assessments, on the other hand, explore a subset of selected uncertainties by performing
“targeted” simulations for variations from an expected case, thereby assessing “local”
uncertainty (80). Regulators in the United States have required a probabilistic safety assessment
for the Yucca Mountain site (43). In contrast, the Finnish, French, and German programs
evaluate deterministically how a proposed geologic repository would perform for a small set of
defined scenarios. A combination of probabilistic (for probable scenarios) and deterministic (for
less-probable scenarios) uncertainty evaluation is called for in Sweden.

Most experts would probably agree that probabilistic assessments are more useful in theory
because the entire uncertainty space is queried, but would also acknowledge that these methods
are not necessarily simple nor are they straightforward. For example, because probability values
or probability distributions functions are not easy to define, there may be a tendency to make
conservative choices for parameters that are not well constrained, e.g., by selecting an upper-
bound value or by choosing a very wide uncertainty distribution. Such intended conservatism
can, in rare circumstances, lead to a reduction of calculated overall risk, a counter-intuitive result
often termed “risk dilution” (79). A more likely outcome of conservatism in parameter choices is
that the performance prediction does not represent a realistic assessment of risk (81). Since the
results of individual submodels are propagated with the system-level model, the calculated
overall risk can be overly conservative even if reasonable conservatism is employed in each
submodel. To evaluate the importance of explicit and implicit conservatism, probabilistic safety
assessments should generally be supported by (and compared to) deterministic calculations (82).

Another important question is whether sensitivity cases considered in a safety assessment should
be bundled together and rolled up into one total dose value (lumping) or whether they should be
treated separately (splitting) (80). In principle, lumping of scenarios or events ensures that all
possible future conditions with their respective likelihood are incorporated into one (time-
dependent) uncertainty estimate (43), but it also means that high-probability and low-impact
scenarios are attributed to the same level of risk as low-probability and high-impact scenarios.
Individual cases with poor performance might then either dominate over, or might be diluted

28
within, a multitude of more favorable cases. Splitting of scenarios or events, on the other hand,
has its disadvantages, because it is difficult to quantify the relative importance of various discrete
cases and its aggregated overall performance. Nevertheless, several nations (e.g., Sweden) and
institutions advocate the splitting of scenarios, in particular when less likely scenarios with
natural disruptive events are considered (84, 85).

In reviewing the above considerations, it appears that a robust quantitative safety assessment for
geologic disposal should consist of a set of separate, purpose-driven, and complementary
performance predictions with probabilistic and deterministic calculations, as well as with
aggregated total assessments and separate scenario/subsystem analysis. The latter would include,
but should not be limited to, analysis of scenario classes. Separate performance calculations
should also be conducted for individual barriers (e.g., “What if the engineered barrier would
fail?”) or other relevant sensitivity cases. System-level models need to be supported by
demonstrations that the underlying primary models are valid. Because of the technical
complexity of disposal systems, quantitative safety assessments can be so complicated that their
logic, methodology, and outcome may be hard to comprehend even for experts, let alone the
public (82). Repository programs may have to produce extremely complex safety assessments,
but they should also develop simplified yet realistic versions for the general technical and public
audiences (85, 86).

5.4 Long-Term Predictability

A topic of ongoing discussion in safety assessments is how to handle the long time scale over
which the radioactive waste may remain hazardous (87, 88). Many factors defining the evolution
of a repository involve complex processes that are quite challenging to predict over years or
decades. Even the most stable materials and geologic environments, over long enough time
scales, are subject to perturbing events and long-term changes, which make quantitative
predictions more and more uncertain. Eventually, uncertainties can become so overwhelming
that a quantifiable performance analysis may not be meaningful (e.g., 89, 90, 91). With the
exception of France, however, most national programs today require safety assessments for time
periods between 100,000 and one million years. The question arises then what level of protection

29
should reasonably be required for such distant times and how compliance should be
demonstrated.

The rationale for selecting time periods that encompass hundreds of thousands of years is based
on the ethical principle that safety needs to be assessed for as long as the waste presents a hazard.
This argument often involves comparison of the radiological activity of the waste with that of
natural systems (see Figure 2.1), suggesting that typical waste returns to activity values below
natural levels somewhere between tens of thousands to hundreds of thousand years. Another
rationale is that the compliance period should extend beyond the time of predicted peak dose. In
most disposal concepts, the maximum dose impacting the biosphere is expected after several
hundred thousand years or even later, due to the longevity of the engineered barriers and the
retention capacity of the host rock.

Advocates for shorter compliance periods have cautioned that predictions of several thousand
years into the future are neither meaningful nor reliable because of the increasing intrinsic
uncertainties of longer-term projections. Some have argued that 10,000 years is about the
duration of the human civilization as we know it, and that the protection of individuals beyond
that time is “not a reasonable expenditure of today’s resources” (91). Nevertheless, it appears
that the international community is converging towards acceptance of a one- million-year safety
assessment period, but, in an attempt to balance ethical and technical considerations, is
advocating for “time-graded containment objectives” (21, 87). In this concept, different safety
standards are defined for different time periods, with hard protection criteria used in the short
term giving way to softer criteria in the very long term (based on 21).

 Initial Time Period of 500 to 1000 Years

Total containment is proposed for this initial time period, which is the period of highest
toxicity and greatest public concern. This time period may overlap with a period of
monitoring and retrievability.
 Second Time Period of Up To 100,000 Years
During this time period, a dose constraint on the order of 0.1 to 0.3 millisievert per year
would be prescribed, similar to current radiological protection principles.

30
 Third Time Period of Up To 1,000,000 Years
During this last time period, the proposed objective is that the eventual redistribution of the
residual activity in the repository is equivalent to the range of natural radiation hazards. Thus
it would be recognized and accepted that beyond the natural “cross-over time” (i.e., the time
when even the longest-lived waste eventually returns to values below the natural toxicity
level) there is no logical or ethical reason for trying to set radiation protection standards that
are more strict than the range of natural radioactivity exposure experienced today.

There is no full closure yet on some of the questions being raised above (e.g., Blandford et al.,
2011). However, it is clear that a safety case with a compliance period of up to one million years
will have to be made in most international programs. What is not settled today is whether
disposal safety over very long time periods should be demonstrated by quantitative (and possibly
probabilistic) assessment of dose or health risk, or rather via qualitative or semi-quantitative
reasoning. Proponents of quantitative methods maintain that these have value in a safety
assessment as long as the large uncertainties involved with the calculation results are
acknowledged. The argument is that quantitative calculations of dose and risk over such long
times should not be viewed as a realistic prediction of the future, but rather as an illustration of
potential impacts to provide evidence that supports other lines of reasoning and qualitative
analysis (87, 88). Others maintain that the results from such calculations are so questionable that
they should not be used to support decision making (90, 91, 92). Instead they propose rigorous
quantitative safety assessment for the human time scale (thousands or ten-thousands of years),
but prefer qualitative analysis for the geologic time scale.

6 Conclusions

Because of its long-lived toxicity, high-level radioactive waste requires disposal systems that can
provide long-term isolation over very long times. Evaluations (and re-evaluations) by various
countries and international organizations have identified mined geologic disposal as the most
practical and effective way to ensure safe permanent disposal. Carefully selected and well
designed geologic disposal systems with multiple, passive, engineered and natural barriers are
expected to provide an acceptable level of safety for current and future generations.

31
There has been a considerable level of scientific effort on mined geologic disposal over the last
30 years in geological formations ranging from granitic rock to salt and clay. These research
activities, which include the development and use of several large-scale URLs, have resulted in a
better understanding of the physical, mechanical, and chemical processes expected within and
near geologic repositories. URLs also provide unique opportunities for international
collaboration that stimulates the development of new ideas and methodologies.

The long time frame associated with high-level waste requires that safety assessments for
geologic disposal need to be carried out very far into the future. This requirement is a major
challenge and can only be addressed in an integrated approach composed of several key
elements: (a) the systematic and careful research of relevant processes through laboratory
research, numerical modeling and field tests in URLs; (b) the detailed study of natural analogues
to evaluate the long term behavior of natural systems related to fluid flow and radionuclide
transport; and (c) the synthesis of quantitative safety analyses with a convincing science base and
multiple independent lines of evidence in a transparent, traceable, and robust safety case.

Uncertainties are inevitable, however, and they need to be explicitly accounted for. Different
methods exist to conduct uncertainty evaluations and more research is needed to determine
which uncertainty quantification approaches are best suited. Of equal importance is that results
from safety assessments and uncertainty analyses need to be formulated in a clear way
understandable to the public and stakeholders. Open communication is critical to achieve and
maintain public support and acceptance over the long development period required for selecting,
investigating, licensing, constructing, and operating a repository system.

Summary Points

1. Radioactive waste requires complete containment on the order of a few thousand years,
because of the high-radiation intensity of the waste during this initial period. Substantial
isolation over much longer time periods, up to a million years, is needed to limit exposure to
the waste because of the potential hazard of chronic low-radiation-dose exposures.

32
2. Mined geologic disposal is considered the most practical and effective way to safely isolate
the waste over long time periods. It relies on multiple, passive, engineered and natural
barriers to provide an acceptable level of safety for current and future generations.

3. Many technical aspects of the waste disposal problem have been studied systematically for a
variety of geologic environments, and this effort has resulted in significant advances in
associated aspects of engineering and geosciences.

4. Regulatory and public acceptance requires the development of a safety case for a given
disposal system. The safety case ideally contains multiple lines of evidence, including
quantitative performance predictions supported by laboratory and field data, plus qualitative
information from natural analogues or other geological observations.

Future Issues

1. Having to deal with the radioactive waste remains one of the main challenges to the prospect
of global nuclear energy utilization.

2. While many experts believe that the disposal of waste in a geologic repository is technically
feasible and safe, public perception is mostly negative. Several countries and organizations
have come to realize that public acceptance of waste disposal sites can be increased by active
stakeholder participation in the decision process. Also, countries like Finland, Sweden, and
France are moving forward with their license application process and are expected to have
operating geologic repositories in the 2020 to 2025 time frame. A successful repository
operation in one of these countries would possibly increase confidence in other nuclear
nations.

3. Despite decades of R&D and significant advances regarding the technical complexities of
disposal, there is still need for continued research because of the unique requirements and
inevitable uncertainties of long-term predictability into the far distant future.

33
4. Questions remain as to which safety standards are reasonable in light of technical and ethical
considerations and whether adherence to these standards over very long time periods should
be (or can be) demonstrated by quantitative analyses or rather qualitative reasoning.

Acknowledgement

This paper has been reviewed by Drs. Dan Hawkes and Hui-Hai Liu at Lawrence Berkeley
National Laboratory. Their constructive comments are greatly appreciated. This work was
funded by and conducted for the Department of Energy, Office of Nuclear Energy, under DOE
Contract No. DE-AC02-05CH11231.

34
Literature Cited

1. Ramana MV. 2009. Nuclear Power: Economic, Safety, Health, and Environmental Issues of
Near-Term Technologies, Annual Review of Environment and Resources, 34: 127-152.
2. NWTRB. 2011. Experience Gained From Programs to Manage High-Level Radioactive
Waste and Spent Nuclear Fuel in the United States and Other Countries - A Report to
Congress and the Secretary of Energy. April 2011. (Arlington VA: U S Nuclear Waste
Technical Review Board).
3. International Atomic Energy Agency (IAEA). 2009. Classification of Radioactive Waste,
General Safety Guide No. GSG-1, IAEA, Vienna, 2009. http://www-
pub.iaea.org/MTCD/publications/PDF/Pub1419_web.pdf
4. Andrews A. 2006. CRS Report for Congress, Radioactive Waste Streams: Waste
Classification for Disposal, Order Code RL32163,
http://www.fas.org/sgp/crs/misc/RL32163.pdf

5. DOE. 1997. Waste Isolation Pilot Plant Disposal Phase Final Supplemental Environmental
Impact Statement; Chapter 4, Department of Energy, Carlsbad Area Office, Carlsbad, New
Mexico. http://www.wipp.energy.gov/library/seis/V1-4o7.pdf

6. IAEA. 2008. Estimation of Global Inventories of Radioactive Waste and Other Radioactive
Materials. IAEA-TECDOC-1591, ISBN 978–92–0–105608–5,ISSN 1011–4289, IAEA,
Vienna, 2008. http://www-pub.iaea.org/MTCD/publications/PDF/te_1591_web.pdf
7. Hedin A. 1997. Spent nuclear fuel—How dangerous is it? Svensk Kärnbränslehantering AB,
Technical Report TR-97-13.
8. Massachusetts Institute of Technology (MIT). 2003. The Future of Nuclear Power.
Massachusetts Institute of Technology. ISBN 0-615-12420-8.
http://web.mit.edu/nuclearpower/pdf/nuclearpower-full.pdf.
9. NWTRB. 2010. Evaluation of the Technical Basis for Extended Dry Storage and
Transportation of Used Nuclear Fuel. http://www.nwtrb.gov/reports/reports.html.
10. Blue Ribbon Commission (BRC). 2011. Blue Ribbon Commission on America’s Nuclear
Future—Draft Report to the Secretary of Energy. Draft Report. July 29, 2011. Available at
www.brc.gov.

35
11. NAS. 2001. Disposition of High-Level Waste and Spent Nuclear Fuel: The Continuing
Societal and Technical Challenges, National Academy Press, ISBN: 0-309-56764-5,
http://www.nap.edu/catalog/10119.html.
12. Nuclear Waste Management Organization (NWMO). 2005. Choosing a Way Forward: The
Future Management of Canada’s Used Nuclear Fuel (Final Study).
www.nwmo.ca/studyreport/?action=downloadfile&id=341
13. IAEA. 2003. Scientific and Technical Basis for the Geological Disposal of Radioactive
Wastes. Technical reports series, ISSN 0074–1914 ; no. 413, STI/DOC/010/413, ISBN 92–0–
100103–7, IAEA, Vienna, http://www-
pub.iaea.org/MTCD/Publications/PDF/TRS413_web.pdf.
14. Hebel LC, Christensen EL, Donath FA, Falconer WE, Lidofsky LJ, Moniz EJ, Moss TH,
Pigford RL, Pigford TH, Rochlin GI, Silsbee RH, Wrenn ME, Frauenfelder H, Cairns TL,
Panofsky WKH, Simmons MG. 1978. Report to the American Physical Society by the Study
Group on Nuclear Fuel Cycles and Waste Management, Reviews of Modern Physics, 50(1),
Part II.
15. North DW. 1997. Unresolved problems of radioactive waste: Motivation for a new paradigm.
Physics Today, 1997, American Institute of Physics.
16. NWTRB. 2009. Survey of National Programs for Managing High-Level Radioactive Waste
and Spent Nuclear Fuel. A Report to Congress and the Secretary of Energy. (Arlington VA:
U.S. Nuclear Waste Technical Review Board).
http://www.nwtrb.gov/reports/nwtrb%20sept%2009.pdf
17. Cook BJ, Emel JL, Kasperson RE. 1990. Organizing and managing radioactive waste
disposal as an experiment. Journal of Policy Analysis and Management 9 (3): 339–366. John
Wiley & Sons, Inc.
18. IAEA. 2006. Geological Disposal of Radioactive Waste— Safety Requirements. IAEA Safety
Standards Series No. WS-R-4. (Vienna: International Atomic Energy Agency). Jointly
Sponsored by IAEA and OECD/NEA.
19. NEA. 2003. Engineered Barrier Systems and the Safety of Deep Geological Repositories,
ISBN 92-64-18498-8, OECD 2003. http://www.oecd-nea.org/rwm/reports/2003/nea3615-
ebs.pdf.

36
20. Stefanovsky SV, Yudintsev SV, Gieré R, Lumpkin GR. 2004. Nuclear waste forms. In: Gieré
R. & Stille P. (eds) 2004. Energy, Waste, and the Environment: A Geochemical Perspective,
236: 37–63. Geological Society, London, Special Publications.
21. Chapman N, McCombie C. 2003. Principles and Standards for the Disposal of Long-Lived
Radionuclide Waste, Elsevier, Waste Management Series, Volume 3.

22. Tsang CF, Barnichon JD, Birkholzer J, Li XL, Liu HH, Sillen X. 2012. Coupled thermo-
hydro-mechanical processes in the near field of a high-level radioactive waste repository in
clay formations. International Journal of Rock Mechanics and Mining Sciences 49: 31–44.

23. Tsang C-F, Bernier F, Davies C. 2005. Geohydromechanical processes in the EDZ in
crystalline rock, rock salt, and indurated and plastic clays, in the context of radioactive waste
disposal. International Journal of Rock Mechanics and Mining Science 42: 109–125.

24. Vietor T, Armand G, Nyonoya S, Schuster K, Wieczorek K. 2010. Excavation induced

damage evolution during a mine-by experiment in opalinus clay. Clays in natural and
engineered barriers for radioactive waste confinement. 4th International Meeting. March,
2010. Nantes, France.

25. Wileveau Y, Bernier F. 2008. Similarities in the Hydro-Mechanical Response of Callovo-

Oxfordian Clay and Boom Clay during Gallery Excavation. Physics and Chemistry of the
Earth 33: S343–S349.

26. Bernier F, Li XL, Bastiaens W. 2007. Twenty-five years’ geotechnical observation and
testing in the Tertiary Boom Clay formation. Géotechnique 57(2): 229-37.

27. Delage P, Le TT, Tang AM, Cui YJ, Li XL. 2007. Suction effects in deep Boom Clay block
samples. Géotechnique 57(1): 239–244.

28. Rejeb A, Cabrera J. 2006. Time-dependent evolution of the excavation damaged zone in the
argillaceous Tournemire site (France). Proceedings “GeoProc 2006,” Invited Lecture for the
2nd International Conference on Coupled THMC Processes in Geosystems and Engineering,
Nanjing, China, May 22-25, 2006, pp. 65–74.

29. Massmann J, Uehara S, Rejeb A, Millard A. 2009. Investigation of desaturation in an old

tunnel and new galleries at an argillaceous site. Environmental Geology 57(6), 1337–1346.

37
30. Rutenberg M, Czaikowski O, Lux K-H. 2009. Time-dependent deformation of claystone rock
mass—Numerical simulation and field measurement data. Proceedings of European
Commission TIMODAZ-THERESA International Conference, Luxembourg, September–
October 2009.

31. SELFRAC 2007. Fractures and self-healing within the excavation disturbed zone in clays.
Final Technical Report (Project FIKW-CT2001-1001-00182). European Commission EUR
22585, 2007.

32. Lanyon GW, Marschall P, Trick T, de La Vaissière R, Shao H, Leung H. 2009.

Hydromechanical evolution and self-sealing of damage zones around a microtunnel in a
claystone formation of the Swiss Jura Mountains. 43rd US Rock Mechanics Symposium and
4th U.S.-Canada Rock Mechanics Symposium, Asheville, NC, June 28th–July 1, 2009;
ARMA 09-333.

33. Zhang C-L, Rothfuchs T. 2008. Damage and sealing of clay rocks detected by measurements
of gas permeability. Physics and Chemistry of the Earth 33: 363–373.

34. Bock H, Dehandschutter B, Martin CD, Mazurek M, deHaller A, Skoczylas F and Davy C.
2010. Self-sealing of fractures in argillaceous formations— Evidence, mechanisms and
implications for performance assessment (An NEA Clay Club Project). Clays in Natural and
Engineered Barriers for Radioactive Waste Confinement. 4th International Meeting. March
2010. Nantes, France.

35. Delage P, Sultan N, Cui YJ, Li XL.2009. Permeability changes of Boom Clay with
temperature. Proceedings of European Commission TIMODAZ-THERESA International
Conference, Luxembourg, September 29–October 2009.

36. Zhang C-L, Wieczorek K, Rothfuchs T, Armand G, Lebon P. 2009. Responses of the
Opalinus Clay to heating during the HE-D experiment at Mont Terri. Proceedings of
European Commission TIMODAZ-THERESA International Conference, Luxembourg,
September–October 2009.

37. Gens A, Vaunat J, Garitte B, Wileveau Y. 2007. In situ behaviour of a stiff layered clay
subject to thermal loading: Observations and interpretation. Géotechnique 57(2): 207–228.

38
38. Marschall P, Horseman S, Simmi T. 2005. Characterisation of gas transport properties of the
Opalinus Clay, a potential host rock formation for radioactive waste disposal. Oil & Gas
Science and Technology 60(1): 121–139, Rev. IFP.

39. Shao H, Schuster K, Delay J, Wang W, Kolditz O. 2009. Analysis of dilatancy-controlled gas
flow along clay bedding. Proceedings of European Commission TIMODAZ-THERESA
International Conference, Luxembourg, September–October 2009.

40. Yu L, Weetjens E. 2009. Summary of Gas Generation and Migration: Current State-of-the-
Art. Mol, Belgium: SCK•CEN, 2009 —29 pp. (External Report of the Belgian Nuclear
Research Centre; ER-106; European Commission FP7 FORGE. Fate of Repository Gases).-
ISSN 1782-2335, 2009.

41. Hoxha D, Do D-P, Wendling J, Poutrel A. 2010. Role of HM coupling mechanisms on the
evolution of rock masses around nuclear waste disposals in the context of gas generation.
Clays in Natural and Engineered Barriers for Radioactive Waste Confinement, 4th
International Meeting. Nantes, France, March 2010.

42. Duveau G, Davy CA, Skoczylas F, Shao J-F, Dubois T. 2010. Creation of gas pathways
through water-saturated argillite: experimental and numerical approaches. Clays in Natural
and Engineered Barriers for Radioactive Waste Confinement. 4th International Meeting,
Nantes, France, March 2010.

43. SNL. 2008, Total System Performance Assessment Model/Analysis for the License
Application Application, Volume I, II, and III, ANL-WIS WIS-PA PA-000005 REV 00
(Sandia National Laboratory).
44. SNL. 2008. Probabilistic Volcanic Hazard Analysis Update (PVHA-U) for Yucca Mountain,
Nevada, TDR-MGR-PO-000001 Rev 01(Sandia National Laboratory).

45. Witherspoon PA. 2000. The Stripa Project. International Journal of Rock Mechanics and
Mining Science 37(1): 385–396.

46. Bodvarsson GS, Boyle W, Patterson R, Williams D. 1999. Overview of Scientific

Investigations at Yucca Mountain—the Potential Repository for High-Level Nuclear Waste.
Journal of Contaminant Hydrology 38(1-3): 3–24.

39
47. Birkholzer JT, Li G, Tsang C-F, Tsang Y. 1999. Modeling Studies and Analysis of Seepage
into Drifts at Yucca Mountain. Journal of Contaminant Hydrology 38(1-3): 349-384.

48. Aalto P, Aaltonen I, Ahokas H, Andersson J, Hakala M, Hellä P, Hudson JA., Johansson E,
Kemppainen K, Koskinen L, Laaksoharju M, Lahti M, Lindgren S, Mustonen A, Pedersen K,
Pitkänen P, Poteri A, Snellman M, Ylä-Mella M. 2009. Programme for Repository Host Rock
Characterisation in ONKALO (ReRoc). Posiva Oy Workin Report 2009-31.

49. Posiva. 2011. Olkiluoto Site Description. Posiva Oy, Report.

50. Armand G, Wileveau Y, Morel J, Cruchaudet M, Rebours H. 2007. Excavated damaged zone
(EDZ) in the Meuse Haute-Marne underground research laboratory. Proceedings of the 11th
Congress of the ISRM – Lisbon, Portugal, 9-13 July 2007 – pp. 33–36.

51. Delay J, Lebon P, Rebours H. 2010. Meuse/Haute-Marne centre: next steps towards a deep
disposal facility. International Journal of Rock Mechanics and Geotechnical Engineering
2(10): 52-70.

52. Tsang C-F, Stephansson O, Jing L, Kautsky F. 2009. DECOVALEX Project: from 1992–
2007. Environmental Geology 57: 1221–1237.

53. Vomvoris S, Blechschmidt I, Vietor T, Johnson L. 2011. NAGRA'S activities in the URLs in
Switzerland: Update and outlook, Proceedings of the 13th Intern. HLRWM Conf. April 10-
14, 2011, Albuquerque, New Mexico, USA.

54. Gaus I, Thurner E, Vahanen M, Martín Martín P-L, Mayor JC, Garcia-Sineriz J-L, Gens A.
2009. FEBEX in situ test - Showing the value of very long term (>10 years) experiments.
ICEM'09 / DECOM'09. The 12th International Conference on Environmental Remediation
and Radioactive Waste Management, 11-15 October 2009, Liverpool. Final Program, 12.

55. Villar MV (ed.) 2006. FEBEX Project final report: Post-mortem bentonite analysis.
Publicación técnica 05-1/2006. Empresa Nacional de Residuos Radiactivos, Madrid.

56. Schaefer T, Seher H, Hauser W, Walther C, Geckeis H, Degueldre C, Yamada M, Suzuki M,

Missana T, Alonso U, Trick T, Blechschmidt I. 2009. The CFM (Colloid Formation and
Migration) Project at the Grimsel Test Site (GTS, Switzerland): First Field Experiments.
Abstract NO: PB5-6.- Migration 2009, 12th International Conference on the Chemistry and

40
 Migration Behaviour of Actinides and Fission Products in the Geosphere, Kennewick,
Washington, September 20 - 25, 2009, 220.

57. Van Marcke P, Bastiaens W. 2010. Construction of the Praclay Experimental Gallery at the
Hades URF. Clays in Natural and Engineered Barriers for Radioactive Waste Confinement:
4th International Meeting. March, 2010. Nantes, France.

58. Li X-L, Bastiaens W, Weetjens E, Sillen X. 2009. The THM boundary conditions control in
the design of the large scale in situ PRACLAY heater test. Proceedings of TIMODAZ/
THERESA Coupled THMC Conference, Luxembourg, September 29–October 1, 2009.

59. Weetjens E, Sillen X. The source term and the geometry of the PRACLAY heater test.
Internal report of EURIDICE, Belgium, 2004.

60. Simmons AM, Stuckless JS. 2010. Analogues to Features and Processes of a High-Level
Radioactive Waste Repository, USGS Professional Paper 1779, U.S. Geological Survey,
Reston, Virginia.
61. Levy S, Goldstein S, Dobson PF, Goodell P, Ku T-L, Abdel-Fattah A, Saulnier G., Fayek M.,
De La Garza R. 2011. Peña Blanca Natural Analogue Project: Summary of activities.
International High-Level Radioactive Waste Management Conference, Albuquerque, NM,
April, 2011.
62. International Atomic Energy Agency (IAEA). 1995. Principles of Radioactive Waste
Management. Safety Series No. 111-F (Vienna: International Atomic Energy Agency).
63. International Atomic Energy Agency (IAEA). 2011. Radioactive Waste Management
Objectives. IAEA Nuclear Energy Series NW-O. (Vienna: International Atomic Energy
Agency).
64. NEA. 2009. Towards Transparent, Proportionate, and Deliverable Regulation for Geologic
Disposal. NEA/RWM/RF2009(1). (Paris: Organization for Economic Cooperation and
Development).
65. National Research Council. 2006. Health Risks from Exposure to Low Levels of Ionizing
Radiation: BEIR VII – Phase 2, Committee to Assess Health Risks from Exposure to Low
Levels of Ionizing Radiation. http://www.nap.edu/catalog.php?record_id=11340.

41
66. UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation) 2000.
Sources and Effects of Ionizing Radiation, 2000 Report to UN General Assembly, Vol. 1:
Sources, and Vol. 2: Effects, with Scientific Annexes, New York.
67. West JM, Shaw RP, Pearce JM. 2011. Environmental Issues in the Geological Disposal of
Carbon Dioxide and Radioactive Waste, in Toth, F.L. (ed.), Geological Disposal of Carbon
Dioxide and Radioactive Waste, Advances in Global Change Research, 44: 81–102.
68. Moeller DW. 2009. Lauriston S. Taylor Lecture: Yucca Mountain Radiation Standards,
Dose/Risk Assessments, Thinking Outside the Box, Evaluations and Recommendations,
Health Physics, 97(5), 376–391.
69. De Groot JIM, Steg L. 2011. Psychological perspectives on the geological disposal of
radioactive waste. In Toth FL (ed.), Geological Disposal of Carbon Dioxide and Radioactive
Waste, Advances in Global Change Research 44: 339–363.
70. Fischhoff B, Slovic P, Lichtenstein S, Read S, Combs B. 1978. How safe is enough? A
psychometric study of attitudes towards technological risks and benefits. Policy Sci. 9: 127–
152.
71. Peters E, Burraston B, Mertz CK. 2004. An Emotion-Based Model of Risk Perception and
Stigma Susceptibility: Cognitive Appraisals of Emotion, Affective Reactivity, Worldviews,
and Risk Perceptions in the Generation of Technological Stigma. Risk Analysis, 24: 1349–
1367.
72. NEA. 2004. Post-Closure Safety Cases for Deep Geological Disposal of Geological
Repositories. NEA-3679. ISBN 92-64-02075-6. (Paris: Organization for Economic
Cooperation and Development).
73. NEA. 2008. Safety Cases for Deep Geological Disposal of Radioactive Waste: Where Do We
Stand? NEA-6319. ISBN 978-92-64-99050-0. (Paris: Organization for Economic
Cooperation and Development).
74. O’Hagan A, Caitlin EB, Daneshkhah A, Eiser, JR, Garthwaite P, Jenkinson D, Oakley J,
Rakow T. 2006. Uncertain Judgments: Eliciting Experts’ Probabilities, John Wiley, London.
75. International Atomic Energy Agency (IAEA). 1993. Radioactive Waste Management
Glossary. (Vienna: International Atomic Energy Agency). International Atomic Energy
Agency (IAEA). 1995. Principles of Radioactive Waste Management. Safety Series No. 111-
F (Vienna: International Atomic Energy Agency).

42
76. Rechard RP. 1999. Historical relationship between performance assessment for radioactive
waste disposal and other types of risk assessment. Risk Analysis, 19(5): 763–807.
77. NEA. 2000. Features, Events, and Processes (FEPs) for Geologic Disposal of Radioactive
Waste: An International Database. NEA, Technology & Engineering. (Paris: Organization
for Economic Cooperation and Development).
78. Nuclear Energy Agency (NEA). 1997. Lessons Learned from Ten Performance Assessment
Studies. (Paris: Organization for Economic Cooperation and Development).
79. Maul PR, Metcalfe R, Pearce J, Savage D, West JM. 2007. Performance assessments for the
geological storage of carbon dioxide: Learning from the radioactive waste experience.
International Journal of Greenhouse Gas Control. 1, 444–455.
80. Roehlig K-J, Plischke E. 2009. PAMINA, Performance Assessment Methodologies in
Application to Guide the Development of the Safety Case - Review of Existing Fly
Probabilistic Assessments: The Regulator’s Perspective on the PSA Approach, Project
Report, European Commission 6th Framework Program, September 2009.
81. IRT (International Review Team). 2002. An International Peer Review of the Yucca
Mountain Project TSPA-SR. A Joint Report by OECD/NEA and IAEA.
82. NWTRB. 2011. Technical Advancements and Issues Associated with the Permanent
Disposal of High-Activity Wastes - Lessons Learned from Yucca Mountain and Other
Programs—A Report to Congress and the Secretary of Energy. June 2011. (Arlington VA:
U.S. Nuclear Waste Technical Review Board).
http://www.nwtrb.gov/reports/technical%20lessons.pdf
83. ICRP (International Commission on Radiological Protection). 2000. Radiation protection
Recommendations as Applied to the Disposal of Long-Lived Solid Radioactive Waste. ICRP
Publication 81, Annals of ICRP 28/4.
84. SKB. 2006. Long-term Safety for KBS-3 Repositories at Forsmark and Laxemar—A First
Evaluation: Main Report of the SR-Can Project. SKB Technical Report TR 06 09, October
2006. http://www.skb.se/Templates/Standard____24109.aspx
85. SNL. 2008. Total System Performance Assessment Model/Analysis for the License
Application Application, Volume III, Appendix L “Simplified TSPA”, ANL-WIS WIS-PA
PA-000005 REV 00 (Sandia National Laboratory).

43
86. EPRI (Electrical Power Research Institute). 2009. EPRI Yucca Mountain Total System
Performance Assessment Code (IMARC) Version 10 – Model Description and Analyses,
Final Report 1018712, June 2009.
87. NEA. 2004. The Handling of Timescales in Assessing Post-closure Safety. NEA-4435. ISBN
92-64-02161-2. (Paris: Organization for Economic Cooperation and Development).
88. NEA. 2009. Considering Timescales in the Post-closure Safety of Geological Disposal of
Radioactive Waste. NEA No. 6424. (Paris: Organization for Economic Cooperation and
Development)
89. Ewing RC, Tierney MS, Konikow LF, Rechard RP. 1999. Performance Assessments of
Nuclear Waste Repositories: A Dialogue on Their Value and Limitations, Risk Analysis,
19(5): 933–958.
90. MacFarlane AM, Ewing RC, eds. 2006. Uncertainty Underground—Yucca Mountain and the
Nation’s High-Level Nuclear Waste. MIT Press. Cambridge MA. 431 pp.
91. Blandford ED, Budnitz, RJ, Ewing, RC. 2011. What Does 1 Million Years Mean to a
Regulator, Nuclear News, November 2011.
92. Long JCS, Ewing RC. 2004. Yucca Mountain: Earth science issues at a geologic repository
for high-level radioactive waste. Annual Review of Earth and Planetary Sciences. 32:363–
401.

44
Web Sites Cited

Nuclear Energy Agency (NEA): http://www.oecd-nea.org/brief/brief-06.html

SKB: http://www.skb.se/Templates/Standard____24109.aspx

U.S. Nuclear Regulatory Commission: http://www.nrc.gov/reading-rm/doc-collections/fact-

sheets/dry-cask-storage.html, http://www.nrc.gov/waste/high-level-waste.html

World Nuclear Association: http://world-nuclear.org/info/inf04ap2.html

45
Tables

Table 4.1: Major Currently Active URLs

URL Rock Formation Depth Year Initiated

Grimsel, Switzerland Crystalline Rock ~ 400 m 1984
HADES, Belgium Boom Clayc ~ 240 m 1984
Äspö, Sweden Crystalline Rock ~ 460 m 1995
c
Mont Terri, Switzerland Opalinus Clay 250–320 m 1996
c
Tournemire, France Argillite 250 m 1996a
Bure, France Argillitec 500 m 2000 b (completed
2009)
Onkalo, Finland Crystalline Rock 400–500 m 2004 b
Mizunami, Japan Crystalline Rock 500 m 2005 b
Horonobe, Japan Sedimentary Rock 300 m (500 m 2006 b
planned)
a
Year when two tunnels for repository research were drilled off a century-old railway tunnel.
b
Year URL construction started; the construction is ongoing, but field testing has already been initiated in the
completed part of tunnel.
c
Opalinus Clay and Argillite are indurated clays, representing rock with strength in between crystalline rock and
plastic clay such as Boom Clay.

46
Figures

Figure 2.1: Radioactivity of spent fuel relative to the natural radioactivity of

mined uranium ore [from Hedin (7)].

47
 Figure 3.1: Barriers for radioactive waste disposal.
Source: SKB (http://www.skb.se/Templates/Standard____24109.aspx)

48
Figure 4.1: Layout of the Grimsel URL tunnel network [from Vomvoris et al. (53)]. The list of field tests
in Phase V1 (2003–2013) are indicated in the lower left part of the figure.

49
Figure 4.2: Layout of HADES URL at Mol, Belgium (top). Schematic of the Praclay In-Situ Experiment
constituted of the Gallery and Crossing Test, the Seal Test, and the Heater Test (bottom). [Van Marcke
and Bastiaens (57)]

50
Figure 5.1: Typical elements of a safety case [NEA, (72)].

51
DISCLAIMER

This document was prepared as an account of work sponsored by the United States
Government. While this document is believed to contain correct information, neither the
United States Government nor any agency thereof, nor The Regents of the University of
California, nor any of their employees, makes any warranty, express or implied, or
assumes any legal responsibility for the accuracy, completeness, or usefulness of any
information, apparatus, product, or process disclosed, or represents that its use would not
infringe privately owned rights. Reference herein to any specific commercial product,
process, or service by its trade name, trademark, manufacturer, or otherwise, does not
necessarily constitute or imply its endorsement, recommendation, or favoring by the
United States Government or any agency thereof, or The Regents of the University of
California. The views and opinions of authors expressed herein do not necessarily state or
reflect those of the United States Government or any agency thereof or The Regents of
the University of California.

Ernest Orlando Lawrence Berkeley National Laboratory is an equal opportunity

employer.