Radioactive Waste Management eBook — Chapter 3: Geological Disposal Safety Case

IAEA SSG-23 safety case per multi-barrier concept (waste form + container + buffer + host rock + biosphere), site selection per ISO 14224 reliability + multi-criteria analysis, performance assessment per probabilistic methods.

This chapter elaborates on the Geological Disposal Safety Case portion of Radioactive Waste Management (WIA-ENE-026). The radioactive waste management canonical envelope shape, API surface, and protocol exchanges are designed to interoperate across implementations published under the WIA Standards family. Each Phase file in `spec/` documents one architectural layer (envelopes, REST surface, federation, ecosystem) and the present chapter focuses on the geological disposal safety case subset of that layering.

Implementations of geological disposal safety case need to reconcile the per-standard expectations of IAEA GSG-1 (Classification of Radioactive Waste), IAEA SSR-5 (Disposal of Radioactive Waste), IAEA SSG-23 (Safety Case for Geological Disposal), IAEA Joint Convention on Spent Fuel and Radioactive Waste Safety, IAEA SSR-6 (Transport of Radioactive Material). The Phase 1 envelope class for Radioactive Waste Management carries the per-host identifier and per-tenant identifier so that audit records can be reconstructed across hosts that federate through the Phase 3 protocol layer.

Operators stepping into geological disposal safety case typically follow this trajectory: stand up the reference simulator at `simulator/index.html`, walk the canonical envelope per the CLI helper at `cli/radioactive-waste.sh envelope`, then exercise the Phase 2 endpoints with curl/Postman against a host running the reference container at `wia/radioactive-waste-host:1.0.0`. Once those steps complete, integrators wire the real backend one endpoint at a time per the Phase 1 §A.* envelope catalogue and the Phase 2 §B.* endpoint catalogue.

A common operational concern in the geological disposal safety case space is observability. Every Phase 1 envelope SHOULD emit a structured log line at the host's audit transport with: timestamp per RFC 3339, host identifier, tenant identifier, envelope class, envelope identifier, operation outcome, and a W3C Trace Context `traceparent` propagated end-to-end. Phase 2 surfaces this trace identifier as the `X-WIA-Trace-Id` response header. Phase 3 protocol exchanges propagate the trace identifier inside the exchange envelope so that a federation crossing remains correlatable end-to-end.

The geological disposal safety case portion of the standard composes with: WIA-OMNI-API for credential storage, WIA-AIR-SHIELD for runtime trust list, WIA-INTENT for workload intent declaration, and WIA Secure Enclave for sealed-data envelopes where personal data is processed. The composition lets one host running multiple WIA family standards reuse one identity, one signing-key chain, and one audit transport rather than maintaining N parallel per-standard implementations.

Normative references touched in this chapter

Implementation worksheet

Cross-standard composition recap

This chapter, like every other Phase 1-4 chapter in the Radioactive Waste Management eBook, composes with the wider WIA Standards family. Implementations that adopt the canonical envelope reuse the cross-standard audit transport (W3C Trace Context plus OpenTelemetry semantic conventions), the cross-standard identity (WIA-OMNI-API), and the cross-standard runtime trust list (WIA-AIR-SHIELD) without per-standard re-implementation. The federation handshake described in Phase 3 §3 follows the same challenge-response pattern as the rest of the WIA Standards family so that one operator can federate across multiple standards using one signing-key chain and one audit transport.

3.1 The IAEA SSG-23 Multi-Barrier Concept

Geological disposal places high-level radioactive waste at depths of several hundred meters or more within a stable host rock, securing isolation periods on the scale of tens of thousands to one million years. The IAEA Specific Safety Guide SSG-23 ("The Safety Case and Safety Assessment for the Disposal of Radioactive Waste", 2012) adopts the multi-barrier concept as the central organizing idea: five isolation barriers in series, designed so that the failure of any single barrier still leaves the remaining barriers performing their isolation function.

Barrier 1: Waste form - for direct disposal of spent nuclear fuel, the UO_2 pellets and zircaloy cladding; for reprocessing residues, the borosilicate vitrified waste form or the Synroc synthetic mineral phase.
Barrier 2: Disposal container (canister, overpack) - multiple shells of corrosion-resistant metals such as copper, steel, or titanium. The Finnish KBS-3 reference design uses a 5 cm copper outer shell over a 5 cm cast-iron inner shell.
Barrier 3: Buffer - compacted blocks of bentonite (montmorillonite clay). When groundwater enters, the buffer swells to seal the host-rock to container gap. The clay's sorption capacity delays nuclide migration substantially.
Barrier 4: Host rock - granite (Swedish SKB, Finnish Posiva, Korean KAERI), claystone (French Cigeo, Swiss Nagra), or rock salt (German Konrad, U.S. WIPP). The host rock delays nuclide migration on a timescale of tens of thousands of years, depending on fracture density, hydraulic conductivity, and redox conditions.
Barrier 5: Biosphere isolation - the host-rock thickness between the disposal depth (500 to 1,000 m) and the biosphere (surface and water resources). The barrier serves as the primary safety margin against human-intrusion scenarios and natural-erosion scenarios.

3.2 Composition of the Safety Case

The "Safety Case" defined by IAEA SSG-23 is a documentation regime that quantitatively demonstrates the long-term safety of a disposal facility. It comprises five core components: (1) System Description, (2) Safety Assessment, (3) analysis of model and parameter uncertainty, (4) Supporting Arguments, and (5) Synthesis. The quantitative target derived from IAEA SF-1 ("Fundamental Safety Principles", Principle 8 - "The protection of present and future generations") is an individual effective dose at or below 1 mSv per year, or a potential dose rate at or below 0.3 mSv per year for systems with a natural host rock of substantial thickness (SSR-5 §3.32 normal scenario).

3.2.1 Probabilistic Performance Assessment (PA)

The PA of a geological disposal facility uses Monte Carlo sampling to combine host-rock hydrogeology parameters (hydraulic conductivity, fracture density, redox potential, groundwater chemistry) with canister corrosion parameters (uniform corrosion rate, localized corrosion distribution, weld defect frequency) to produce a probability density function (PDF) of nuclide release rates. The reference simulator panel 2 "shielding calculator" demonstrates the first-order PA simplification (single nuclide, one-dimensional diffusion). The PA reference codes are GoldSim (Sandia National Laboratories, USA), MELODIE (ANDRA, France), and ECOLEGO (SKB, Sweden). KAERI has acquired the GoldSim license and operates a Korean-site model in the KAERI-PA-2020 code.

3.3 ISO 14224 Reliability Analysis and Site Selection

ISO 14224 "Petroleum, petrochemical and natural gas industries - Collection and exchange of reliability and maintenance data for equipment" was originally a reliability standard for petroleum and petrochemical facilities, but it is borrowed by disposal facilities as the standard vocabulary for active equipment reliability databases (ventilation, drainage, shielding doors). IAEA SSG-23 §6.3 recommends using ISO 14224 conformant reliability data as safety-assessment parameters.

3.3.1 Multi-Criteria Decision Analysis (MCDA)

Site comparison decisions for geological disposal facilities are made using the multi-criteria decision analysis (MCDA) recommended by IAEA SSG-14 "Geological Disposal Facilities for Radioactive Waste". Eight to twelve criteria are typically aggregated with weighted summation: geological stability, hydrogeology, earthquake frequency, climate variability, accessibility, population distribution, ecological impact, economic cost, social acceptance, mineral-resource potential, cultural-heritage distribution, and international-convention compliance. The Korean Special Act on the Management of High-Level Radioactive Waste enabling regulations will codify eight "exclusion criteria" and fourteen "suitability criteria" for the basic-survey and detailed-survey stages of site selection.

3.4 Korean Geological Disposal R&D

Korea has conducted geological disposal R&D since 1997, led by the Korea Atomic Energy Research Institute (KAERI). The principal facilities and programmes are:

KURT (KAERI Underground Research Tunnel, KAERI campus in Daejeon, 2006-): Korea's first underground R&D facility, a 90 m deep horizontal tunnel in granite host rock. KURT hosts the K-DEMO module for coupled thermo-hydro-mechanical-chemical (THMC) testing of bentonite buffer, fracture-tracing experiments, and in-situ nuclide sorption tests. International cooperation links it with Finnish Onkalo, Swedish Aspo, and Swiss Grimsel underground facilities.
K-AKRWMS (Advanced Korean Reference for Waste Management System): the Korean reference design for geological disposal, developed by KAERI from 2007 to 2017. Assumes approximately 36,000 spent fuel bundles plus the equivalent of approximately 100,000 reprocessing-canister units, at a depth of 500 m, with a 70 cm bentonite buffer thickness and a disposal canister of 5 cm copper plus 5 cm cast iron.
High-level siting procedure (2024-): Under the 2024 Special Act, KORAD has been designated as the implementer of the siting procedure. The 37-year roadmap allocates 13 years for the siting committee and basic survey, 7 years for detailed survey, 7 years for interim storage construction, and 10 years for permanent disposal facility construction.

3.5 Candidate Host Rock Comparison

Three host rock types are internationally recognized as standard for geological disposal: granite, claystone, and rock salt.

Granite: A crystalline rock where fractures can be traced. Examples include Swedish SKB Forsmark (currently under construction), Finnish Posiva Onkalo (operational from 2023), and the Korean KAERI KURT. Hydraulic conductivity is concentrated in fractures, which gives the tracing model a clear advantage, but fracture-distribution uncertainty is the major uncertainty source in PA.
Claystone: Very low hydraulic conductivity and high sorption capacity. Examples include French ANDRA Cigeo (Bure, target operational 2031) and the Swiss Nagra Mont Terri R&D facility. The host rock itself acts as a nuclide-migration delay barrier, but the excavation-damaged zone (EDZ) is the key variable in PA.
Rock salt: Plasticity provides self-sealing of fractures. Examples include German Konrad (low- and intermediate-level) and U.S. WIPP (military transuranic waste, operational since 1999). The long-term plasticity behavior is the key PA assessment area.

3.6 Korean Geological Disposal Infrastructure Mapping

Regulator: NSSC issues siting and operating licenses and reviews the Safety Case. KINS verifies the safety-assessment codes and reviews the site-suitability technical evidence.
Implementer: KORAD - the Korea Radioactive Waste Agency - is the statutory implementer of the high-level siting procedure under the 2024 Special Act and operates the secretariat of the siting committee.
R&D institutions: KAERI in Daejeon operates KURT, the K-AKRWMS design, and the GoldSim PA code. KIGAM (Korea Institute of Geoscience and Mineral Resources, Daejeon) leads geological surveys and geophysical exploration. KICT (Korea Institute of Civil Engineering and Building Technology) leads EDZ assessment.
Statute: The Special Act on the Management of High-Level Radioactive Waste (January 2024) provides the statutory basis for siting, the siting committee, and interim storage construction. The enabling regulations are under development in 2025.
Policy lead: The Ministry of Trade, Industry and Energy (MOTIE) leads the High-Level Radioactive Waste Management Basic Plan (2021-2060). The Ministry of Science and ICT (MSIT) leads the disposal R&D budget (KAERI R&D programmes).
International cooperation: The OECD/NEA Working Party on Long-Term Safety Assessment of Geological Repositories (WP-LTSA), the IAEA International Project on Demonstrating the Operational and Long-Term Safety of Geological Disposal Facilities (GEOSAF Part III), and the bilateral cooperation between Finnish Posiva Onkalo and Korean KAERI (since 2018).
Social acceptance: The OECD/NEA Forum on Stakeholder Confidence (FSC) guidelines apply. Korea will operate community-information meetings, environmental-impact assessment, and local-council consent as legally mandated stages in the siting process.
Workforce development: The Korean Radioactive Waste Society (KRWS) is operated jointly by KAERI, KINS, and KORAD. UNIST, KAIST, Seoul National University, and Hanyang University all operate disposal-R&D graduate programmes in nuclear engineering.

3.7 Safety Case Envelope Example

The following example envelope describes a Safety Case submission for a candidate granite site under Korean siting procedure.

{
  "version": "1.0",
  "host_id": "korad.kr.siting-secretariat",
  "tenant_id": "kaeri.pa-team",
  "envelope_class": "WIA-ENE-026.SafetyCase",
  "envelope_id": "ulid:01HKABCD12345678...",
  "site_candidate": "candidate-granite-A",
  "host_rock_type": "granite",
  "depth_m": 500,
  "buffer_thickness_cm": 70,
  "canister_design": "copper-5cm-castiron-5cm",
  "pa_simulator_code": "KAERI-PA-2020",
  "peak_individual_dose_msv_per_year": 0.087,
  "regulatory_target_msv_per_year": 0.3,
  "safety_case_status": "draft-for-public-comment",
  "audit": {
    "timestamp": "2026-05-15T11:33:45Z",
    "traceparent": "00-9d8c7b6a5f4e3d2c1b0a9d8c7b6a5f4e-2b3c4d5e6f7a8b9c-01"
  }
}

3.8 Performance Assessment Worked Example

A representative Performance Assessment (PA) of the Korean K-AKRWMS reference design proceeds as follows. The source term for one disposal canister carries the time-evolved inventories of 18 key nuclides (I-129, Tc-99, U-235, U-238, Np-237, Pu-239, Pu-240, Pu-241, Am-241, Am-243, Cm-244, Cs-137, Sr-90, C-14, Mo-100, Zr-93, Sn-126, Sm-141). The canister corrosion model applies a uniform-corrosion rate distribution (mean 1 nm/y under reducing conditions for copper, 10 nm/y for cast iron) together with a localized pitting frequency drawn from KAERI K-DEMO laboratory data. Container breach is conventionally assumed to occur between 100,000 and 1,000,000 years after disposal, depending on the parameter draw.

The buffer model (70 cm bentonite) applies a sorption distribution coefficient (K_d) of approximately 5 m^3/kg for U, 10 m^3/kg for Np, and 50 m^3/kg for Pu under reducing conditions, derived from KURT in-situ sorption tests. The fractured-granite host rock applies a fracture transmissivity distribution calibrated against KURT borehole hydraulic tests, with mean fracture spacing of 4 m and mean fracture aperture of 50 micrometers. Nuclide migration through the host rock fracture network proceeds by advection plus matrix-diffusion-and-sorption.

The Monte Carlo simulation of 10,000 parameter realizations through the coupled six-module model produces a peak individual effective dose distribution. For the K-AKRWMS design under normal-evolution scenarios, the 95th-percentile peak individual dose is approximately 0.08 mSv per year, well below the IAEA-recommended 0.3 mSv per year threshold for natural-host-rock disposal systems (SSR-5 §3.32). The PA also evaluates human-intrusion scenarios (inadvertent drilling, unintentional excavation) with stochastic timing parameters, with peak doses in the 0.5 to 2 mSv per year range that, while exceeding the normal-scenario target, fall within the IAEA "constrained optimization" envelope for low-probability scenarios.

3.9 Comparison with International Reference Sites

The K-AKRWMS design is benchmarked against three international reference sites for credibility. (1) Finnish Posiva Onkalo (Olkiluoto granite, 450 m depth, operational since 2023) - the closest analog, with the same KBS-3 canister design (copper outer + cast-iron inner) and bentonite buffer. (2) Swedish SKB Forsmark (granite, 470 m depth, under construction with target operational date 2030) - identical canister and buffer design; differences in fracture network density. (3) French ANDRA Cigeo (claystone Callovo-Oxfordian, 500 m depth, target operational 2031) - different host rock (claystone instead of granite), different canister (steel only), but similar overall safety case structure.

The OECD/NEA Working Party on Long-Term Safety Assessment of Geological Repositories (WP-LTSA) recommends cross-comparison of PA results across these sites as a credibility check. KAERI participated in the 2022 WP-LTSA Code Verification Project, with KAERI-PA-2020 results agreeing within plus-minus 10 percent of GoldSim reference outputs for the standard benchmark cases. This benchmark agreement strengthens the credibility of future Korean Safety Case submissions to NSSC.