The safety case for geological disposal brings together the evidence and arguments that support the safety of a planned repository. It explains how the waste, the engineered barriers and the geological environment function together, how their behaviour may change over time, and how uncertainty affects the conclusions. Quantitative performance assessment, together with site observations, experiments, engineering evidence and independent review, forms the core of the safety argument.
WIA-ENE-026 v1.0 supports the safety case through records on waste classification, radionuclide inventory, containers, facilities, monitoring and traceability. The standard's phase document PHASE-1, Foundation & Classification, is a self-contained foundational document covering classification, the fundamentals of radioactive decay, radiation hazards, the regulatory basis and waste-tracking information systems. This chapter links that information base to geological disposal assessment and examines the evidence needed to support the safety of a specific site and design. Sections 3.1–3.5 cover disposal concepts, protection objectives, sites and designs; Sections 3.6–3.8 cover responsibility for evidence, records and research; and Sections 3.9–3.11 cover performance assessment, preparation and review, and complementary evidence; Section 3.12 collects related standards and guidance. Readers who wish to see the flow of the quantitative assessment first may begin with Section 3.9.
Geological disposal combines engineered and natural components to contain radioactive waste and to limit interaction between the waste and the environment accessible to people. IAEA Specific Safety Guide SSG-23, The Safety Case and Safety Assessment for the Disposal of Radioactive Waste (2012), provides guidance on structuring the overall safety argument for a disposal facility. Safety assessment considers the safety functions of each barrier, the interactions between barriers, and common causes that could degrade several barriers at once.
WIA-ENE-026 v1.0 lists the waste form, container, buffer, backfill and host rock as components of its disposal example. These components perform complementary functions. The extent to which each contributes to safety depends on the waste inventory, the repository design, groundwater conditions and the evolution of the site.
| Component | Contribution to safety | Main influencing factors |
|---|---|---|
| Waste form | Retains radionuclides within a solid matrix and influences the pattern of release once water has entered. | Matrix dissolution, chemical conditions, temperature, and the location of radionuclides within the waste. |
| Disposal container | Provides containment and delays contact between groundwater and the waste. | Material properties, fabrication quality, corrosion processes, mechanical loads, and interaction with surrounding materials. |
| Buffer | Conditions the environment around the container, restricts water movement, and can retard the migration of some radionuclides. | Density, saturation, swelling, groundwater chemistry, temperature, and radionuclide-specific migration behaviour. |
| Backfill and seals | Fill excavated openings and limit preferential migration along tunnels, shafts and other access routes. | Emplacement quality, material compatibility, interfaces, excavation damage, and long-term deformation. |
| Host rock | Physically separates the waste from the surface and influences groundwater flow and radionuclide migration. | Connected flow paths, rock properties, groundwater chemistry, geological evolution, and excavation effects. |
Where spent nuclear fuel is disposed of directly, the UO2 fuel matrix influences radionuclide release once groundwater reaches the fuel. The fuel cladding may make an additional contribution to safety, but crediting that function in the assessment requires a sound account of the cladding's condition and degradation. Release models for vitrified waste represent the glass matrix and its alteration behaviour. Container assessment considers welds and fabrication quality as well as the loads imposed during handling and disposal.
Bentonite swells as it hydrates and, under suitable conditions, can exhibit low hydraulic conductivity. Sorption is the binding of radionuclides to solid surfaces, including the surfaces of clay minerals. Sorption retards the migration of some radionuclides, and the degree of retention depends on the chemical species and the chemistry of the water. Backfill serves to limit preferential migration along the excavated openings created by repository construction.
Safety assessment addresses the interactions between barriers explicitly. Heat alters saturation and chemical reaction rates, and groundwater chemistry can affect corrosion, clay behaviour and radionuclide mobility. Excavation can create migration pathways that bypass rock with otherwise favourable properties. Degradation of one barrier can also change the conditions under which another operates. The safety argument therefore evaluates the system's evolution over time together with the dependencies between safety functions.
The biosphere is the environment into which released radionuclides enter and the setting in which potential exposures are assessed. Repository depth contributes to isolation, and radionuclide travel times depend on the actual migration pathways and processes. PHASE-1 of this standard describes the isolation period for high level waste as 100,000 years or more. The assessment sets the assessment timeframe on the basis of the radionuclide inventory, site evolution and the applicable criteria, and establishes how the functions credited to safety change over time.
A safety case is a structured body of evidence, analysis and reasoning assembled to support decisions at a given stage of repository development. It sets out what is currently understood and what uncertainties remain, and explains why the available evidence is sufficient for the decision in question. Its conclusions may support further investigation, design changes or progression to the next stage of development.
Examples of the elements that make up the arguments and evidence of a safety case include the following.
IAEA SF-1, Fundamental Safety Principles (2006), states in Principle 7, "Protection of present and future generations", that people and the environment, present and future, must be protected against radiation risks. This principle sets an overarching protection objective. A repository assessment gives it concrete form by specifying the assessment timeframe, the scenarios and the protection criteria to be applied.
The public effective dose limit of 1 mSv/year applies, within the relevant planned exposure framework, to exposures arising from regulated activities, and excludes medical exposure and exposure to natural background radiation. IAEA SSR-5, Disposal of Radioactive Waste (2011), treats a disposal facility as a single source and requires that it be designed so that the calculated dose to a representative person in the future, arising from natural processes that could affect the facility, does not exceed a dose constraint of 0.3 mSv/year or a risk constraint of the order of 10−5 per year (SSR-5, para. 2.15(b)). These constraints support prospective assessment and optimization within the overall system of radiological protection.
The two values differ in scope. The public dose limit applies to the sum of relevant exposures from regulated activities, whereas the dose constraint for disposal is used, within that limit, as a criterion for assessing and optimizing a single facility. The safety case specifies the national criteria that apply, the exposure situations included in the assessment, and how disruptive events and inadvertent human intrusion are treated. Comparisons with numerical criteria are evaluated together with the adequacy of the inventory, scenarios and model assumptions.
SSR-5 provides graded criteria for inadvertent human intrusion after closure. If the annual dose to people living around the site is expected to be less than 1 mSv, efforts to reduce the probability of intrusion or to limit its consequences are not considered warranted; if it is in the range 1–20 mSv, reasonable efforts are made at the facility development stage to reduce the probability or consequences of intrusion through design optimization; and if it could exceed 20 mSv, alternative disposal options or the separation of the radionuclides giving rise to the higher dose are considered. These criteria are treated separately from the dose constraint applied to natural evolution.
WIA-ENE-026 v1.0, in §5.1.2 of the standard, gives 1 mSv/year for public exposure and 0.25 mSv/year as a site-boundary criterion for nuclear power plants. This site-boundary value is interpreted as a criterion applicable to nuclear power plants. A geological disposal assessment defines the criteria to be applied to operational and post-closure scenarios and specifies the exposure pathways and decision rules alongside them.
ICRP Publication 122 (2013) addresses radiological protection specifically for the geological disposal of long-lived solid radioactive waste. When this system of protection is applied in calculations, appropriate radionuclide-specific dose coefficients are selected, and the exposure pathways and population groups on which those coefficients are based are stated.
Operational assessment covers activities such as the handling and emplacement of waste packages, ventilation and worker protection. Equipment reliability and maintenance records can support this assessment. Post-closure assessment examines how the closed repository evolves, including containment, release and migration. This requires evidence describing the long-term behaviour of the waste, the barriers and the geological environment.
Where an equipment reliability and maintenance data framework developed for another industry is used as a reference for repository operational equipment records, the difference in scope between its original application and repository equipment is stated. Repository siting and post-closure assessment draw on geological, hydrogeological and geochemical evidence. Assessment of ventilation, drainage and handling equipment focuses on operational reliability, whereas assessment of the closed disposal system focuses on long-term containment and radionuclide migration.
Passive safety is the concept of ensuring long-term protection through the functions of the disposal system itself. Requirement 5 of SSR-5 requires that safety be ensured by passive means to the fullest extent possible through the siting, design, construction, operation and closure of the disposal facility, and that the need for actions after closure be minimized. Long-term protective functions do not depend on continuing active intervention. The purposes and regulatory basis of records, institutional controls and post-closure monitoring are considered separately from the physical safety functions credited to the safety of the repository.
Site evaluation begins with a conceptual understanding of the geological environment. Relevant evidence includes rock structure, groundwater flow and chemistry, mechanical conditions, thermal properties, and the processes that could change the site. Measurements are interpreted with regard to the spatial extent they represent. A hydraulic test in a single borehole shows the conditions investigated by that test; applying the result more widely requires evidence on the rock domain surrounding the repository.
Characterization links observations to safety issues. Identifying connected fractures, for example, helps in evaluating potential flow paths. Groundwater chemistry provides information for judging corrosion and radionuclide speciation, and rock-mechanics observations support evaluation of excavation damage and the stability of excavated openings. Measurement records include location, depth, test conditions, method, detection limit and uncertainty. Where the observations can be explained by more than one model, alternative interpretations are also considered.
Suitability criteria distinguish factors that would prevent a site from meeting the conditions necessary for safety from characteristics that can be compared among potentially suitable sites. The criteria applied require a clear technical or regulatory basis. Site comparison records state the document and edition that establish each criterion and the procedure by which it is applied in decision-making.
Multi-criteria decision analysis can make comparisons more transparent by setting out the criteria used, the scoring rules and the weights. The weighted overall result is used to compare potentially suitable sites. Meeting the applicable safety requirements is a precondition and is evaluated separately from preferences concerning accessibility or cost.
Because weights express preferences, rankings may change if preferences change or if site property values, taking uncertainty into account, differ. To show that a comparison is sound, the evidence used in the judgement is presented, and the stability of the ranking is examined by varying the weights and site parameters within plausible ranges.
Stakeholder engagement helps identify concerns and uses of the environment that the assessment may need to address. The scientific basis of safety claims, the reasons for siting decisions and the applicable legal engagement requirements are each presented separately. Stakeholder forums provide a venue for discussing these matters, while the basis of the assessment is established by citing the relevant technical and regulatory documents.
A reference design provides a consistent starting point for investigating materials, geometry, thermal behaviour and migration. Design dimensions are meaningful when interpreted in relation to the conditions and safety functions of the design concerned. For each configuration assessed, a consistent set of dimensions is defined in line with the design basis.
The KBS-3 type canister used for the direct disposal of spent nuclear fuel consists of a copper shell about 5 cm thick and an inner insert that provides structural integrity. The insert supports the fuel and provides mechanical strength. The geometry and structural function of the insert are evaluated separately from the thickness of the copper shell.
The KBS-3 example in §6.3 of the standard gives a copper thickness of 50 mm and a bentonite buffer thickness of 350 mm. The 350 mm value belongs to the KBS-3 reference design example; its relationship to the 500 mm that appears in the facility record example in §4.2 of the standard is set out in the Standard Reading Notes, item 6, in Section 1.3. The design basis records the dimensions of the geometry actually applied, together with their sources.
Reflecting the safety-function durations given in §6.3 of the standard in a site-specific assessment requires supporting evidence. Container lifetime, buffer performance and geological isolation are evaluated against the conditions and processes expected at the repository concerned.
§6.3 of the standard gives a depth of 430 m for Onkalo and a planned operating period of 2025–2120, while §9.2 records its status at the time the standard was written as "under construction". The former is a planned schedule; the latter is the status at a particular point in time (Standard Reading Notes, item 5, in Section 1.3). When project status is cited, the reference date and the actual stage are confirmed from official sources of the implementer (Posiva) or the regulator (STUK), and the reference date is recorded with it. The depth and geometry of individual facility areas are confirmed from detailed design records.
International comparisons become clearer when they focus on a specific function or assessment method. Comparing container behaviour requires checking that materials and groundwater conditions are comparable; comparing migration in the host rock requires describing the flow paths and geochemical characteristics. When evidence from one site is applied to another, the inventory, design and environmental conditions that govern the results are examined.
| Case | Disposal concept to examine | Differences to check |
|---|---|---|
| Onkalo, Finland | Interaction of KBS-3 type containers and buffer with the host rock in the crystalline bedrock at Olkiluoto (predominantly migmatitic gneiss). | Site-specific groundwater chemistry, fractures and design information, and the distinction between scheduled items and milestones actually achieved in the operational stage. |
| Forsmark, Sweden | The roles of the engineered barriers and the crystalline host rock in the KBS-3 type disposal concept planned for the Forsmark site. | Site-specific geological conditions, detailed dimensions and model inputs, even where the overall disposal concepts are similar. |
| Cigéo, France | Diffusion, excavation effects and engineered barrier functions in a disposal concept using the Callovo-Oxfordian claystone of the eastern Paris Basin as host rock. | Differences in migration pathways compared with crystalline rock systems, and differences in how safety functions are shared between the engineered barriers and the host rock. |
Technical comparison of disposal concepts and reporting on project progress serve different purposes. When describing project status, the reference date is stated and planned milestones are distinguished from those actually achieved. When comparing calculation results, the radionuclide inventory, model boundaries, scenarios, exposure pathways and the basis of the reported statistics are likewise aligned so that differences in results can be interpreted.
Crystalline rock, argillaceous formations and rock salt are examples of geological environments studied for disposal. The Waste Isolation Pilot Plant (WIPP) in the United States is an example of disposal in salt. Located 26 miles (about 42 km) southeast of Carlsbad, New Mexico, it permanently isolates defense-generated transuranic (TRU) waste in rooms excavated about 655 m (2,150 feet) underground in an ancient salt formation. Disposal began in 1999. Suitability is judged by considering together the properties and evolution of the formation, the engineered design and the waste inventory.
| Geological environment | Potential contribution | Matters to confirm from site evidence |
|---|---|---|
| Crystalline rock, such as granite and gneiss | Mechanically competent rock can support underground construction. A low-permeability rock matrix can provide additional retention through pores and mineral surfaces accessible to radionuclides. | Fracture connectivity, groundwater flow paths, and the representativeness of measured fracture properties, matrix diffusion and sorption. |
| Argillaceous formations, including claystone | Low hydraulic conductivity can produce diffusion-dominated migration conditions. Clay minerals can retain some chemical species. | Effects of excavation damage, gas, heat and changes in pore water chemistry on migration and mechanical behaviour. |
| Rock salt | Intact salt can have low permeability, and creep can contribute to the closure of excavated openings. | Effects of brine, dissolution, heat load, deformation and interfaces on the evolution of the repository. |
Travel times depend on path length, groundwater flow, fracture connectivity, diffusion, sorption, solubility and the chemical form of the radionuclides. Connectivity can matter more than the number of fractures, and connected preferential pathways may exist even where the average hydraulic conductivity is low. Because sorption varies with the radionuclide and the chemical conditions, retention is described in a way that reflects the characteristics of each process.
Argillaceous formations can exhibit anisotropy, so measurements and models take account of the direction of migration and deformation. In rock salt, impurity layers and the distribution of brine can affect migration, corrosion conditions and long-term deformation. Examining these features shows how general rock properties manifest at an individual site.
Excavation changes the system being assessed. Even where the undisturbed rock has favourable properties, the excavation-damaged zone, stress disturbance and interfaces around backfill or seals can affect migration. The assessment therefore considers the original geological environment together with the changes introduced by construction.
The Konrad repository in Germany makes use of an iron ore mine where mining ceased in 1976 and is planned for the disposal of low- and intermediate-level radioactive waste with negligible heat generation at a depth of about 850 m. When comparing facilities intended for different categories of waste, the waste categories and thermal output are stated along with the actual geological environment. When experience gained with one waste category is applied to another, differences in heat load, release processes and migration conditions are examined.
A safety case clearly assigns responsibility for the evidence supporting each claim. The technical tasks and responsible parties involved in preparing and reviewing that evidence are set out below. Legal responsibilities follow the relevant national framework.
Issues raised in review are resolved technically on the basis of evidence, calculations, design changes or a sound justification of the conclusions, and the outcome is recorded. Each issue is tracked together with the action taken and its basis. Progress of the scientific evaluation and the status of administrative decisions are recorded separately in accordance with applicable law and approved procedures.
WIA-ENE-026 v1.0 captures information usable in assessment within its waste package and facility data structures. The value of this information lies in making it possible to trace waste records through to assessment inputs. Site models, the arguments justifying the scenarios, and the supporting evidence connect these inputs to conclusions about repository safety.
| Defined record or field | Use in assessment | Interpretation required |
|---|---|---|
| packageId and wasteClass | Identify the waste package and build the inventory to be assessed. | Acceptability is judged by reviewing the classification code against the repository's acceptance criteria. |
| radiologicalData.isotopes and radiologicalData.totalActivity.referenceDate | Establish radionuclide quantities and the time basis for inventory calculations. | Inventory consistency is ensured by checking and reconciling the measurement and calculation sources, decay data, uncertainty and any missing radionuclides. |
| radiologicalData.thermalOutput | Supports thermal analysis and emplacement planning. | A consistent time basis is applied and the change in thermal output over time is taken into account. |
| containerInfo, physicalProperties and processing | Describe the packaging, waste form and conditioning history. | Transport or storage packaging is distinguished from the disposal container, and the supporting materials evidence is stated. |
| traceability.burnup, traceability.dischargeDate, traceability.coolingTime and traceability.chainOfCustody | Used to reconstruct the inventory and maintain continuity of waste package records. | Calculations, hand-overs and acceptances, and recorded inventory changes affecting the contents assessed are explained. |
| facilityId, location.geology and engineeredBarriers | Link the facility description to geological and barrier models. | Summary entries are supported by investigation results, design documents and materials characterization. |
| monitoring and monitoringSystems | Provide observations relevant to operating conditions and model evaluation. | Measurements are interpreted taking into account measurement location, calibration, uncertainty and observation period. |
The following example shows part of a waste package record. It illustrates how activity, thermal output and uncertainty information are carried into the assessment using the fields defined in §4.1 of WIA-ENE-026 v1.0. The radionuclide list contains a single example entry.
{
"packageId": "RW-2025-KR-00001234",
"timestamp": "2025-12-25T10:30:00Z",
"wasteClass": "HLW-1",
"radiologicalData": {
"totalActivity": {
"value": 1.5e15,
"unit": "Bq",
"referenceDate": "2025-12-25"
},
"thermalOutput": {
"value": 2.5,
"unit": "kW"
},
"isotopes": [
{
"isotope": "Cs-137",
"activity": {
"value": 8.5e13,
"unit": "Bq"
},
"halfLife": {
"value": 30.17,
"unit": "years"
}
}
],
"uncertainties": {
"activityUncertainty": 5,
"doseRateUncertainty": 10,
"unit": "percent"
}
}
}packageId links the waste package to its associated records. timestamp indicates when the record was made, and totalActivity.referenceDate indicates the reference time for the total activity. To align the inventory to a common time basis that accounts for radioactive decay and daughter ingrowth, a measurement or calculation reference date is also defined for the activity of each radionuclide.
The isotopes array is the starting point for radionuclide-specific calculations, and thermalOutput supports analysis of thermal behaviour. In this example the total activity represents the whole waste package, while the radionuclide array is a partial record listing only Cs-137. The assessment ensures consistency by reconciling the full radionuclide inventory with the total activity of the waste package.
The uncertainty magnitudes in the example are 5% for activity and 10% for dose rate. To propagate uncertainty quantitatively, it must be explained whether these values represent a standard uncertainty, a coverage interval or some other bound. Probability distributions and correlations also require a sound basis. On the basis of these definitions, the way the uncertainty entries are used in Monte Carlo calculations is determined.
The standard places host_rock in the engineeredBarriers array of the facility data structure, alongside waste_form, container, buffer and backfill. In terms of physical role, the host rock is a natural barrier. This distinction is the basis for assigning responsibilities for engineering design control and for geological characterization.
The distinction between the spent nuclear fuel codes (HLW-1 in the main text of the standard and HLW-SF in PHASE-1) follows the Standard Reading Notes, item 3, in Section 1.3. Assessment records preserve the classification notation and, where a project-specific mapping is applied, state its basis.
For the half-life of Cs-137, the adopted value and its source are recorded and applied consistently, as explained in the Standard Reading Notes, item 1, in Section 1.3. Each assessment states the basis for the nuclear data adopted and uses the same values in every calculation that depends on those data. Before calculation, the reference dates of the inventory data are harmonized and units are stated. In particular, when a decay constant is combined with elapsed time, the time units of the two values are checked and made consistent.
A project evidence index can link waste package and facility records to reports, model versions, calculations and review decisions. This form of document control connects the waste management records of this standard to the wider body of evidence supporting the safety case.
WIA-ENE-026 v1.0 defines GET /packages/{packageId}, GET /facilities/{facilityId} and GET /facilities/{facilityId}/inventory for retrieving waste package, facility and inventory information. The assessment fixes the records and reference time to be used in the input dataset and confirms that the container information represents the disposal design being assessed. The traceability.chainOfCustody field records the chain of custody (hand-over and acceptance records), while model applicability and safety conclusions are supported by the relevant assessment evidence.
Repository research is a process of progressively refining understanding of the disposal system. Early work explores candidate materials and geological environments; later work tests key assumptions for the selected design and site. Research is iterative, with results feeding back into further investigation and testing. A revised groundwater model may change the priorities for materials testing, and materials that behave differently from expectations may require design changes.
Materials research may investigate waste form alteration, container corrosion, clay swelling and the interaction of radionuclides with mineral surfaces. Results are presented together with the conditions under which they were obtained. Temperature, water chemistry, mineral composition and specimen preparation all affect the applicability of results to a repository.
The sorption distribution coefficient Kd expresses the relationship between sorbed and dissolved concentrations under the assumptions of a particular model. Treating it as a constant can be useful within a limited range. Extrapolation takes account of speciation, competing solutes and mineral surfaces. The applicability of the coefficient depends on the conditions represented by the measurements and the model.
Coupled thermal–hydraulic–mechanical–chemical studies investigate interactions that are difficult to capture through individual measurements. Heating can redistribute moisture, swelling can change stress and permeability, and chemical alteration can change migration properties. Such evidence helps decide which processes can be simplified and which should be represented explicitly in models.
Underground investigations help evaluate the rock response to excavation, groundwater conditions, and the behaviour of engineered materials at larger scales. Interpretation of the results takes account of disturbance caused by excavation, the duration of observation, and differences in conditions between the research location and the disposal site.
When evidence from one location is applied to another, the conditions that govern the process concerned are compared. The applicability of corrosion observations depends on the materials and the chemical environment; the applicability of tracer observations depends on the flow paths investigated and the migration mechanisms. Rock composition, fracture networks, stress, groundwater chemistry and experimental scale form the basis for such comparisons.
Numerical verification confirms that a calculation correctly implements its equations. Key checks include satisfaction of conservation laws, consistency of units, agreement with analytical solutions for simplified problems, and convergence as time steps and grids are refined. Model validation examines, on the basis of observations and experiments, whether a model adequately represents the processes and conditions to which it is applied. Its conclusions state the range of conditions and behaviour actually evaluated.
Code-to-code comparison helps in reviewing numerical implementation. A meaningful benchmark specifies common inputs, boundary conditions, the quantities compared and the basis for the tolerances set. Agreement of results under those conditions is interpreted together with independent evidence on whether the common assumptions adequately describe the site.
Research priorities are set according to how significantly uncertainties affect safety. If several plausible fracture models predict different release pathways, further characterization may be more useful than fine-tuning parameters that have little influence on the conclusions. Long-term extrapolation is an argument supported by multiple lines of evidence, and the limits of its applicability are stated.
Performance assessment examines how a repository may evolve and how that evolution affects containment, radionuclide release and potential exposure. The six linked modules below are an example illustrating the logical flow of a typical assessment. Scenario development and uncertainty analysis are carried out across all six modules.
| Module | Function | Information passed to the next step |
|---|---|---|
| 1. Inventory and evolution through radioactive decay | Defines the initial radionuclide inventory and reference date and accounts for radioactive decay and daughter ingrowth. Tracks heat generation where needed. | Time-dependent inventory and thermal conditions for the waste being assessed. |
| 2. Release from the container and waste form | Models container integrity, water ingress and release from the waste form according to the selected scenario. | Release rates or boundary concentrations of radionuclides entering the surrounding materials. |
| 3. Buffer, backfill and seals | Models migration and retention within the engineered materials, taking account of associated changes in saturation, chemical conditions and structure. | Fluxes or concentrations at the interfaces with the host rock and other migration pathways. |
| 4. Migration in the host rock | Models groundwater transport, diffusion, dispersion, sorption and radioactive decay along geological pathways supported by evidence. | Time-dependent releases to the surrounding environment. |
| 5. Biosphere transfer and exposure | Links releases to environmental concentrations and specified exposure pathways, applying consistent water and mass balances. | Intakes or external exposure quantities for the representative person assessed. |
| 6. Dose assessment and interpretation | Applies appropriate dosimetric models and coefficients, sums the relevant contributions, and compares the results with the criteria applicable to the scenario. | Dose over time, individual contributions, uncertainty information, and implications for the safety argument. |
The inventory to be assessed is defined to match the actual waste and the assessment timeframe. Short-lived radionuclides can be important for early heat generation and operational protection, whereas long-lived radionuclides and their daughters can influence later releases. Selection of the radionuclides to be assessed considers the origin of the waste, radionuclide quantities, decay chains, mobility and potential contribution to exposure over the assessment timeframe.
For example, I-129, Tc-99 and Np-237 may be candidates for consideration in long-term assessment, depending on their presence and behaviour in the waste and the disposal system. Cs-137 and Sr-90 can contribute substantially to initial activity and heat generation. Because ingrowth of daughters from the decay of a parent radionuclide can increase the inventory of individual radionuclides, the members of the relevant decay chains are calculated in linked form.
The source term describes the quantity of radionuclides released from the waste to its surroundings and how it changes over time. Subsequent calculations track migration through the container, buffer, backfill and host rock and link it to environmental concentrations and exposure pathways. The relative importance of advection, diffusion, dispersion, sorption, chemical reactions and radioactive decay varies with the radionuclide and the site.
When modules are linked, the physical quantities passed between them are clearly distinguished. Concentration, release rate, intake, activity and dose are different quantities. Converting between them requires, depending on the conversion, water flow rates, geometry, exposure assumptions or dosimetric information. Consistent treatment of radioactive decay, sorption and retention ensures that each process is handled as intended at the boundaries between models.
A normal-evolution scenario represents the expected overall evolution of the repository under stated assumptions. Alternative scenarios examine departures from the expected path that affect the results, such as a different pattern of groundwater evolution or the loss of a safety function credited in the assessment. Inadvertent human intrusion is subject to separate scenario-definition rules and interpretation. The assessment explains how assumptions about the consequences and probability of intrusion enter into the judgement.
Deterministic calculations help explain individual mechanisms and selected combinations of assumptions. Probabilistic calculations, including Monte Carlo sampling, examine how specified input distributions and correlations affect the results. Both approaches rest on sound scenarios and conceptual models.
Monte Carlo results are conditional on the chosen model, input distributions and dependencies. Distributions and correlations require a physical and empirical basis. Sampling groundwater conditions and sorption parameters independently, ignoring their chemical dependence, can produce combinations that are hardly compatible with each other. The number of samples is set by examining the convergence of the quantities of interest. Model completeness and the adequacy of the input distributions are evaluated separately.
For example, the interval between the 5th and 95th percentiles contains the central 90% of the simulated distribution. This is an uncertainty interval conditional on the chosen model and inputs. Interpreting a statistical confidence interval requires defining the quantity being estimated and stating the sampling and estimation procedure. When describing percentiles obtained from simulation, it is therefore made clear which distribution they summarize and what uncertainty that distribution represents.
The results distinguish parameter uncertainty from alternative conceptual models and scenarios. They also state whether a statistic represents the dose at a particular time or the peak dose obtained separately in each simulation. A percentile of the per-simulation peak values and the maximum of a time series of time-specific percentiles summarize different quantities.
Shielding calculations evaluate how materials modify a radiation field. Photon attenuation, for example, depends on energy, shield composition and geometry. Repository migration calculations track the movement of radioactive material through water, pores, fractures and environmental compartments.
Loss of containment, waste form dissolution, diffusion and sorption are handled by release and mass transport models. Radioactive decay can affect both shielding and migration assessments, while the other governing processes and the required inputs vary from one assessment to another.
WIA-ENE-026 v1.0 defines POST /dose/calculate for radiation dose calculation and POST /dose/decay for radioactive decay calculation. Long-term disposal assessment links radioactive decay and dose assessment with scenario-specific release, geological migration and environmental exposure models.
The following calculation is an example for checking a single migration relationship. All conditions are hypothetical and set for illustration, and the result is to be interpreted as distinct from any actual repository, laboratory measurement or result of a particular organization. It assumes idealized one-dimensional diffusion in an unbounded homogeneous medium. The diffusion coefficient is held constant, and the effects of advection, sorption and radioactive decay are excluded.
In this model the mean-square displacement is twice the product of the diffusion coefficient and the elapsed time. As a worked example, assuming a diffusion coefficient of 1.0 × 10−12 m2/s, the time for the root-mean-square displacement to reach 0.50 m is obtained as follows.
D = 1.0e-12 m²/s
L = 0.50 m
L² = 2 D t
t = L² / (2 D)
t = 1.25e11 s
t ≈ 4,000 yearsThis example result indicates the statistical extent of spreading of material in the idealized model stated. Predicting a minimum travel time, a barrier lifetime or breakthrough through a finite buffer each requires a model suited to that purpose. Repository calculations also incorporate the source history, actual boundaries, appropriate migration coefficients and other relevant processes.
This calculation helps check units and numerical implementation. Assessing annual dose and interpreting its regulatory significance requires linking release, migration, biosphere and dosimetric calculations and securing evidence to support each input.
This single-radionuclide example shows how inventory calculation links to release, migration and concentration calculations. It uses the C-14 half-life of 5,730 years defined in §3.3 of WIA-ENE-026 v1.0. All other numerical inputs are illustrative assumptions, and the results are example values. The start of release is taken as the time origin, with an initial activity of 1.0 × 106 Bq, a constant release coefficient of 0.001/year and a travel time of 100 years. At the point of arrival, complete mixing into a water flow of 1,000 m3/year is assumed.
In this example model the inventory decreases through radioactive decay and release, and only radioactive decay is applied to material in transit. Travel time and water flow rate are held constant. Sorption, dispersion, daughter ingrowth and differences in chemical form are excluded from the calculation. These assumptions simplify an educational example, and their applicability to a real disposal system is evaluated separately.
lambda = ln(2) / T_half (T_half = 5730 years)
A(t) = A0 * exp(-(lambda + k) * t)
J_release(t) = k * A(t)
J_arrival(t + tau) = J_release(t) * exp(-lambda * tau)
C(t + tau) = J_arrival(t + tau) / QHere A0 is the initial activity and A(t) the remaining activity, both in Bq. The decay constant lambda and the release coefficient k are in 1/year. The variable t is the time elapsed since the start of release, and tau is the travel time. J_release and J_arrival are the activity transfer rates at the point of release and the point of arrival, respectively, in Bq/year. Q is the water flow rate, and C is the mean activity concentration in that water.
| Quantity | Example result | Interpretation |
|---|---|---|
| Remaining activity 100 years after the start of release | About 8.94 × 105 Bq | The remaining inventory reflects both radioactive decay and release. |
| Release rate at 100 years | About 8.94 × 102 Bq/year | The remaining activity at that time multiplied by the release coefficient. |
| Arrival rate at 200 years | About 8.83 × 102 Bq/year | Reflects the additional radioactive decay undergone by material released at 100 years during its 100 years in transit. |
| Mean concentration at 200 years | About 0.883 Bq/m3 | The arrival rate divided by the water flow rate assumed in the example. |
Radioactive decay within the source inventory and radioactive decay in transit apply to different time intervals. The quantities are also distinguished by their units: Bq denotes activity, Bq/year an activity transfer rate, and Bq/m3 an activity concentration. The 100-year travel time is a value assumed in this example model, and the diffusion example in Section 3.9.4 uses separate assumptions.
Linking concentration to ingestion dose requires the water intake and an applicable ingestion dose coefficient. For constant concentration and intake, the relationship is as follows.
E_rate = C * U * gIf U is the annual water intake in m3/year and g is the ingestion dose coefficient in Sv/Bq, then E_rate is in Sv/year. The dose coefficient is selected with a stated basis appropriate to the radionuclide, exposure pathway and age group. To compare the calculated dose with the applicable safety criteria, these additional inputs are recorded and the exposure scenario is defined.
Preparing a safety case means organizing accumulated technical work into an argument that others can review. It is structured so that a reviewer can work back from the conclusions through the scenarios, models and parameter selections to the supporting observations. A practical sequence for preparation is as follows.
The documents produced in this process may include a system description, engineering design reports, a safety assessment report, supporting evidence reports and an integrated synthesis report. These documents use a common inventory and design basis. A change in buffer material, for example, can affect engineering specifications, migration parameters and the safety argument at the same time.
To make calculations reproducible, the editions of input data, nuclear data, units, reference dates, scenarios, boundary conditions, model and code versions, and post-processing methods are recorded. For probabilistic calculations, the input distributions, correlations and sampling settings needed for reproduction are also preserved. Linking these records through waste package identifiers connects the calculation basis with the waste package chain of custody while preserving the distinct purpose of each record.
Uncertainty is evaluated not only by its numerical magnitude but also by its effect on decisions. Uncertainty in fracture connectivity can change migration pathways, and uncertainty in release rates can change the magnitude or timing of the peak dose. Parameter sensitivity analysis helps set priorities for further investigation. Evaluation of alternative conceptual models examines different interpretations of system behaviour.
The safety case is updated in response to significant changes in the development stage, applicable requirements or evidence. New construction observations, a revised inventory or unexpected monitoring trends can prompt re-examination of earlier assumptions, whether or not the calculation software has changed. If container materials, buffer design or the waste to be accepted change, the affected safety functions and calculations are reviewed. Update records state the evidence or design that has changed, the effect on the conclusions, and the schedule of the decision concerned.
Monitoring results become evidence supporting a safety function when their relationship to that function is explained. Temperature measurements can help evaluate thermal models, whereas evaluating future corrosion requires additional evidence on the relevant materials, chemical conditions and processes. The safety case explains how observations are used to test assumptions and what action is taken when observations depart from the expected relationships.
The overall judgement states the conclusions together with their scope of application. It identifies processes for which evidence is limited, conclusions that are sensitive to alternative models, and questions to be addressed by further investigation. This allows the basis for decisions to be traced from the overall safety argument back to the underlying observations and assumptions.
Natural analogues provide evidence on processes operating over timescales or spatial scales that are difficult to reproduce directly in engineering experiments. Relevant topics include mineral alteration, uranium mobility, diffusion, retention, and changes in solubility under different redox conditions. The more clearly the processes under study and the limits of the comparison are explained, the more useful the analogue is to the safety argument.
The natural nuclear fission reactors at Oklo in Gabon are a well-known case for studying the retention and redistribution of nuclear reaction products remaining in geological materials. The isotopic composition of uranium and the isotopic distribution patterns of stable fission products are evidence of past fission and of subsequent material behaviour. Interpretation combines mineralogical and isotopic evidence, radioactive decay relationships and reconstruction of geological history. When reconstructing the behaviour of Cs-137 and Sr-90, the isotopic signatures remaining after radioactive decay are examined together with mineralogical and geochemical evidence, and initial production is distinguished from subsequent radioactive decay and migration.
The Cigar Lake uranium ore body in Canada is a case for examining the retention and migration of uranium in the ore body and its surroundings. Even while the deposit has persisted, mineral replacement, redistribution and later groundwater movement can have produced a complex history. When this evidence is applied to radionuclides in waste, differences in the chemical forms, sorption and migration processes that govern the behaviour of each radionuclide are taken into account.
Archaeological metal artefacts are classed as anthropogenic analogues that provide complementary observational evidence on corrosion. Their use involves comparing their manufacturing history, burial conditions and remaining material with the planned disposal container. Because surviving artefacts form a record that conveys limited information about artefacts that may already have degraded, the representativeness of the preserved examples is also considered.
An argument based on an analogue begins with a specific question, such as whether a retention mechanism persists over a defined range of chemical conditions. It then sets out the features observed, the interpretation that explains them, and the similarities and differences between the analogue and the repository. Temperature, groundwater chemistry, material composition, flow conditions and geological history can all limit the extent to which the evidence can be applied to a repository.
A present-day distribution may be explained by several combinations of initial conditions and subsequent processes. Interpretation therefore also considers alternative histories consistent with the observations. Analogue evidence supports the plausibility of specific processes and long-term models within the conditions examined.
Analogue observations can support process models, prompt re-examination of assumptions, or bound the range of possible behaviour. Their value lies in complementing laboratory studies, site characterization and quantitative assessment with evidence obtained under different conditions and at different scales. The scope of safety claims is matched to the processes and environmental conditions that the analogue represents.
The final safety argument identifies where multiple lines of evidence agree and where further evidence is needed, and explains how the remaining uncertainties affect decisions. Numerical results and analogue observations contribute to demonstrating safety when the relationships between evidence, interpretation and safety claims are clearly explained.
The following publications address waste classification, disposal safety and radiological protection from complementary perspectives. When they are used in a safety case, the concepts or requirements applied are identified and their relationship to the relevant regulatory framework is stated.
WIA Standards — WIA-ENE-026 — Chapter 3. Hongik Ingan (弘益人間) — Benefit All Humanity.