Abstract
We have developed a “hybrid” scenario development method by combining bottom-up and top-down approaches and applied for the case of geological disposal of high-level waste. This approach provides a top-down perspective, by introducing a concept of safety functions for different periods and “storyboards”, which depict repository evolution with time on a range of spatial scales, and a bottom-up perspective, by identifying relationship between processes related to radionuclide migration and safety functions based on feature, event and process (FEP) management. Based on a trial study, we have specified work descriptions for each step of the hybrid scenario development methodology and confirmed that the storyboard provides a baseline and holistic overview for the FEP management and a common platform to involve close interaction with experts in various disciplines to understand the crossover phenomenological processes. We also confirmed that there is no conflict between the top-down approach and the bottom-up approach and the hybrid scenario development work frame fulfils the specified requirements for traceability, comprehensiveness, ease of understanding, integration of multidisciplinary knowledge and applicability to a staged approach to siting.
1. Introduction
Geological disposal systems for radioactive waste are designed to isolate the waste from human and its environment for the necessary times to ensure that no potential release from the wastes to the environment would constitute an unacceptable risk. Performance of geological repository, therefore, should be assessed to confirm that the geological disposal system has an ability to fulfil safety criteria for long term [1]. The performance assessments (PA) of the geological repository consist of scenario development, model development, consequence analysis and comparison of the results to safety criteria. Since the scenario development is a first stage in the PA, the scenarios define the sequences of events and processes to be considered in determining, as a function of time and space, the possible ways in which a geological disposal system will perform its fundamental roles of isolating the radioactive wastes from the human environment and limiting release of radionuclides. In the early work it was recognized that scenarios could be developed by combining the individual features, events and processes (FEPs) that characterize behaviour of geological disposal system. More specifically, as the first step of the scenario development, a comprehensive FEP list is developed on the basis of scientific understanding of the characteristics and evolution of a specific repository and its environment. This FEP list is checked for completeness against international FEP lists. Then, the FEPs are classified according to their likelihood of occurrence and extent of possible impact on system evolution. On the basis of this classification, a decision is made as to which FEPs to carry forward in the main part of the PA and which FEPs to exclude. Finally, scenarios are developed by combining the FEPs carried forward in systematic manner (e.g. OECD/NEA [1] and IAEA [2]).
Since comprehensive FEP list consists of more than 100 elements, a way of generating a small number of representative scenarios from a large number of FEPs in a systematic fashion is a complex task and implies that scenarios are synthesized in a “bottom-up” fashion. To develop scenarios based on this “bottom-up” approach (in other words, FEP-based approach), influence diagrams or interaction matrices have been used to show which FEPs influence (or are influenced by) other FEPs in particular scenarios. This process establishes a huge number of links between FEPs, and the procedure by which this complex network is reduced to a manageable level and specific links are selected for further analysis is unclear (in practice, it depends predominantly on expert judgment). Furthermore, FEPs as specified are limited in their capacity to form scenarios, since they cannot incorporate a description of how the FEP itself varies with time, which is system-specific.
More recently, a “top-down” approach for developing scenarios has become established [3–5]. This has been recognized in recent safety cases, which present scenarios as being developed based on system understanding and/or safety functions which is linked to the safety concept. This safety concept may include top level safety functions (isolation, containment and retardation) and more detailed safety functions that are specific to system components. In the top-down approach, the safety functions are used to describe the initial state and evolution of a system in relation to the safety concept. Furthermore, the FEP lists has been less visible and occasionally only used as a check of completeness of the key safety-relevant system components. It does, however, still exist in the wealth of phenomenological knowledge accumulated or FEP-based knowledge management to keep transparency for the derivation of the scenario. In fact, both bottom-up and top-down approaches have sometime been used in combination implicitly: one approach is primarily applied for establishing scenarios and the other serves as a complimentary tool.
Based on above observations, a methodology of scenario development should be explicitly combined the top-down approach and bottom-up approach to achieve a mutually complementary relationship. In this study, focusing on post-closure safety of a high-level waste (HLW), a “hybrid” approach to scenario development and analysis was developed, with the goal of establishing a practical procedure of scenario development. This provides a top-down perspective, by introducing a concept of safety functions for different periods and “storyboards”, which depict repository evolution with time on a range of spatial scales, and a bottom-up perspective, by identifying relationships between radionuclide migration processes and safety functions based on the FEP management. A traceable procedure then links a storyboard to a set of scenarios, associated with an assessment of their likelihood of occurrence. This has been developed and applied for the specific case of geological disposal of HLW.
2. Development of the methodology for scenario development
2.1. Requirements of hybrid approach
Some requirements to develop a new approach of the scenario development are identified based on an OECD/NEA workshop on scenario development in 2000 [6]:
2.1.1. To demonstrate or try to ensure completeness or sufficiency
Achieving sufficient completeness is a key issue in scenario development. Although the hybrid approach has top-down perspectives, we explicitly define a bottom-up work based on the FEP management to try to ensure completeness or sufficiency.
2.1.2. To deal with safety function
In this hybrid approach, we introduce safety functions and storyboard (see section 2.2) to provide straightforward directions to decide which FEPs to include in the PA. They form a way of structuring understanding to focus on FEPs relevant to system safety, avoiding blind investigation of processes that do not contribute to barrier performance.
2.1.3. To deal with changes with time
Since safety functions change with time, depending on the long-term evolution of the disposal system and its geological environment, scenarios should reflect this temporal progression. Furthermore, the Nuclear Safety Committee of Japan (NSC) has developed a scenario classification for low-level radioactive waste disposal that includes consideration of long-term changes of the geological environment resulting from uplift/erosion, sea-level change and climate change in the base scenario [7]. We thus intend that compatible evaluation of relevant time frames is included in our scenario methodology.
2.1.4. To provide traceability and transparency
Traceability and transparency require formal justification of what has been included/excluded in each scenario, description of how each key process is treated and recording all information. In this hybrid approach, we take a procedure for record keeping based on FEP lists and catalogues (which includes description of meaning of FEPs, interactions with other FEPs, accumulated scientific knowledge, uncertainties, etc). Furthermore, the FEP lists and catalogues are linked to holistic understandings associated with temporal system evolution sequences through the safety function in the hybrid approach. This is intended to establish a mutually complementary relationship between the top-down approach and the bottom-up approach.
2.1.5. To guide decisions concerning future work
In addition to focusing on improvement of assessment methodology, iterative scenario development may contribute to overall programme guidance. Since Japanese programme on the geological disposal is developed in a stepwise fashion, at an early generic stage, scenario development aids understanding of the main features of a disposal system, identifying key processes, determining preliminary scenarios and guiding effort to reduce uncertainties and synthesize multidisciplinary knowledge. At later stages, specific geological settings, tailored designs and more detailed site evolutions will be examined and scenario development then supports comparison of different options and may be particularly important in communicating resulting decisions to regulators and other stakeholders. The hybrid approach is intended to be applicable to a staged approach to siting, facilitating integration of system understanding in a form that is suitable for building (or reviewing) the arguments that form the basis of comprehensive safety case.
2.2. Work frame for hybrid scenario development
2.2.1. Work frame
Based on the requirements and policies of this study descried in section 2.1, a “hybrid” scenario development method that combines aspects of bottom-up and top-down approaches has been developed. The overall procedure for constructing the scenarios (hereinafter referred to as the “work frame”) is illustrated in . This consists of a number of “Tasks” (T-1 to T-5), forming a work frame that is linked to an associated FEP management process. In general, there are several categories of scenario which should be considered in the PA; in our case we define base scenario, less probable variant scenario and extremely low probability scenario. This work frame is intended to be applicable for all scenarios, since the events and processes which are considered “uncertainties” in the base scenario are identified at each step in the work frame. Scenarios without the base scenario are then derived from these uncertainties.
Figure 1. Work frame of hybrid scenario development.
2.2.2. Work description for the specific tasks
To start with scenario development, the geological environment is described, based on literature information or the output of site investigations, and an appropriate disposal system is chosen based on assessment of legal requirements, regulatory guidelines, the characteristics of the site and the properties of the waste under consideration. Based on these boundary conditions, the scenario development is pursued. The developed work frame is elaborated in the following steps.
• T-1
The safety of the geological disposal of HLW in Japan is assured by the selection of a geological environment, design and installation of a well understood engineered barrier system (EBS) and confirmation of safety by a rigorous PA process. Essentially, the safety concept is constructed by these three elements when applied in a consistent manner. The geological environment and the disposal system are associated with a safety concept, which summarizes the conceptual basis of arguments that the repository is safe. The safety concept, from a scenario development perspective, must be built on well-understood features and processes that ensure safety, even in the light of uncertainties and possible detrimental phenomena. A detailed description of the safety concept involves identifying the functions of the disposal system that are key contributors to safety, during both the operational phase and after closure. In our methodology, the concept of safety function is introduced to establish the safety concept and designed which barriers or functions contribute to the safety in the specific time in this task. An advantage of building the safety concept using safety functions is that they can explain and justify the functions provided by each barrier and identify the time periods which they are expected to perform their various functions and also the alternative or additional safety functions.
• T-2
Since scenarios define the long-term evolution of the geological environment, behaviour of the disposal system and the sequences of events and processes to be considered in the PA, integrating system understanding is a key challenge for scenario development. In order to capture the essence of specific scenarios with a focus on post-closure safety, T-2 provides a synthesis of understanding, which forms the technical basis for subsequent scenario development tasks. The goal is to illustrate evolution of the key safety-relevant components of the entire repository system on the basis of the judgment of experts with enough experiences in this field to have an overview of integrated system performance. The evolution of individual barriers and processes related to radionuclide migration are described by using illustrative figures and keywords in the form of “storyboard”. The term “storyboard” is commonly used for a panel or series of panels on which a set of sketches is arranged depicting consecutively the important changes of scene and action in a series of shots as for a film, television show or commercial. Here, storyboard refers to a time sequence of images indicating a potential evolution scheme for a repository system. At early stage of the repository project, T-2 provides a synthesis of understanding, which forms the technical basis for subsequent scenario development tasks. As a repository programme progressed, the storyboard is elaborated for license application.
• T-3
Based on the storyboard, the behaviour of individual system components can be examined in a holistic manner and key processes relevant to safety functions are identified. There are two components within this task. Firstly, safety functions outline the understanding of system features that have an essential role in ensuring the performance of the repository. Secondly, relationships between processes and safety functions are identified by considering the evolving role of different engineering and natural barriers in constraining radionuclide release and migration. While the storyboard helps give holistic understanding with respect to the specific scenario context, it is difficult to describe detail processes and their relationships due to space limitation. Much effort in the past has focused on the challenge of development of analytical tools to comprehensively identify and characterize interactions between safety-relevant processes. For example, in Sweden, SKB has applied a thermal, hydrological, mechanical and chemical (THMC) matrix for assessing the complex interactions during system evolution [8]. In Switzerland, Nagra introduced the concept of a “Super FEP” to reduce the number of FEPs and simplify identification of the relationships between processes by using an interaction matrix of Super FEPs [5]. This study uses a form of process influence diagram (PID) based on the relationships between the processes which are compiled in the FEP catalogue (e.g. JAEA’s FEP database [9]). To maintain a sequentiality of evolving disposal system, an outcome of the subsystem description in certain time is used as an initial condition for analysis of the next segmentation in time. This enables to evaluate detrimental effect that might have impact to safety functions across the segmentations.
• T-4
The following task integrates sets of evolution of safety functions, and processes relevant to the radionuclide migration, which are evaluated in terms of their likelihood of occurrence and the possible extent to which processes may have positive or negative impacts to the safety function with time. Uncertainties associated with the safety functions are evaluated through sensitivity analysis or uncertainty analysis. Here, the storyboard in T-2 can provide the scenario context and a holistic viewpoint to guide T-3 and T-4, which gives rise to directions for identifying key processes. On the other hand, T-3 and T-4 can provide further information about important processes and features relevant to system safety as feedback to expand the system understanding captured in the storyboard. This is intended to try to ensure completeness or sufficiency of developed scenario. Therefore T-2, T-3 and T-4 are carried out and refined in an iterative manner and those tasks are essential elements in the hybrid approach to overcome weakness of one approach with advantage of the other approach. Finally T-4 results in a set of scenarios that summarize possible behaviour of safety functions, processes relevant to radionuclide migration and their uncertainties at different time periods.
• T-5
Finally, the scenarios are assigned to specific classes, based on regulatory guidelines.
In parallel, a comprehensive and project-specific FEP catalogue is established, which provides a technical synthesis of the available scientific understanding of the characteristics and evolution of the repository and its environment (left column of ). In this hybrid approach, the FEP catalogue has a role to provide scientific basis for establishing scenario and to check that there are no significant oversights from T-1 to T-4. On the other hand, key processes that might have serious impact to safety functions are identified through system understanding or uncertainty analysis at that point (right column of ). This helps to identify priorities for which FEP should be focused as future R&D. In more practical terms, it will be important to decide, particularly at an early stage of site selection, how to handle lack of information on a particular site condition, or uncertainty contained in the information. In this context, the hybrid approach would be applicable to a staged approach and useful in making it clear what issues remain to be addressed and what type of technology development is required based on integration of multidisciplinary knowledge. All above information should be kept records in the FEP catalogue. In this task, existing FEP lists, such as the NEA International FEP list [10], provide a comprehensive generic starting point, which is tailored to the design and geological setting under consideration. This tailoring process focuses on FEPs relevant to radionuclide release and migration and associated uncertainties, which is fully documented. The process of FEP management as mentioned above has been described for HLW in the Second Progress Report of the Japan Nuclear Cycle Development Institute (hereinafter referred to as the “H12 report”) [11]. Therefore, in the following, we focus on the procedure of progressing from T-1 to T-5, as shown in .
2.3. Trail run for specific tasks
Since the scenario development is closely coupled to geological environment and system design, particularly in Japan where siting is based on a volunteering approach and design will be tailored to individual sites, the safety assessment, system design and site investigation are proceeded in an iterative manner. However, in order to confirm an applicability of the hybrid approach to the practical scenario development work and to elaborate work description in each step in the hybrid approach, a trail run was implemented assuming the geological environment and the system design of the H12 report.
2.3.1. Development of the safety concept: T-1
Based on the Japanese HLW disposal concept and safety assessment of the H12 report, here, the safety concept is depicted to identify the functions of the disposal system that are key contributions to safety during after closure. illustrates a set of safety functions and their applicability for different periods of the safety assessment. For example, physical containment by the overpack is a key safety function during initial 100 years. This is to ensure a long enough period of time for decay to the most of shorter lived radionuclides that emit a significant amount of radioactivity and heat. This safety function is supported by the low permeability of the buffer to limit infiltration of O2 in the groundwater and the chemical buffering to maintain the reducing condition. After overpack failure, the glass waste form immobilizes radionuclides and restricts their release into the surrounding porewater. Besides, the buffer material contributes to containment of the radionuclides within the EBS through diffusion, retardation, low elemental solubility in porewater and colloid filtration. The characteristics of the buffer, which limits the influx of groundwater and maintains chemistry in mildly alkaline conditions, can be considered supportive this key safety function to provide appropriate hydraulic and chemical environment in order to restrict the radionuclide migration. These functions must be guaranteed for relevant time periods by siting suitable environment in order to build a credible safety case. In particular, sufficient geological stability for the geological repository system and less possibility of human intrusion in the future should be focused as the suitable environment.
Figure 2. Illustration of expected safety functions for different periods.
Unlike defence in depth of existing nuclear facility which relies on multiple and independent barriers, safety of geological disposal can be ensured by mutually dependent multi-barriers or multi-safety functions. Since some safety functions are closely linked to each other as mentioned above, identifying the interaction between these functions is vital to establish the reliability of the overall disposal system safety case. shows relationships between the safety functions during different time periods for the Japanese HLW disposal concept (the upper figure shows the thermal phase and the lower figure shows the period of approximately 105 years after overpack failure). The arrow shows a dependant relationship between the supporting and the receiving functions. Furthermore, establishing such interactions leads to a better understanding of total system safety and the propagation of processes which may have significant impact to the safety function. Possible events and processes that could perturb individual safety functions are identified and analysed in later tasks.
Figure 3. Relationship among the safety functions for different periods (upper figure: thermal phase, lower figure: period approximately 105 years after overpack failure).
2.3.2. Description of evolution of the geological environment and disposal system: T-2
In this task, we have developed “storyboards” as a convenient and user-friendly way to synthesize the complex multidisciplinary knowledge involved. The storyboard illustrates system components on various physical scales and in a time sequence that highlights repository-relevant evolution milestones. shows a format of storyboard placing in array of time scale (row) and system components (column). Initially, evolutions of the geological environment and expected natural phenomena are described considering characteristics of the specific site as the fundamental information to the scenario development.
Figure 4. Format of storyboard.
Each box in the storyboard highlights expected system conditions during a specific time period identifying the safety-relevant phenomena that define ongoing evolution of barrier system components. Working over storyboards has been seen to facilitate communication between geologists, design engineers and safety assessors and play an essential role in developing comprehensive system understanding from a top-down perspective. Such understanding is essential for development of improved PA, in which long-term evolution of the geological environment is treated more realistically. For example, phenomena such as climate change and sea level change must be included in description of expected evolution, along with their hydrological or geochemical influences on safety functions. This perspective on treatment of natural phenomena is compatible to a recent requirement of the NSC [6]. and 6 show contents of the storyboard of the EBS scale during thermal phase and period of approximately 105 years respectively after overpack failure based on the safety assessment of the H12 report. The illustration diagram consists of Thermal (T), Hydrological (H), Mechanical (M) and Chemical (C) conditions in specific time period. At this stage, to capture the essence of specific scenarios and to integrate technical basis is important. In this context, key processes in the light of the THMC are roughly described in the storyboard. For example, during thermal phase, heat from the vitrified waste can be listed one of the most dynamic processes and resaturation of the bentonite and chemical condition in the porewater are thought as key processes to maintain physical containment of the overpack. The period of approximately 105 years after overpack failure, detrimental processes to the safety functions of the buffer, such as the effect of high pH groundwater from concrete liner and the effect of corrosion products of the overpack and nuclides migration from vitrified waste through the buffer material can be considered as crucial processes. Through the trial run, it is recognized that T-2 thus provides the required transparency and traceability in preparation for detail process description, in a manner that is easy to understand, while integrating interdisciplinary knowledge with a focus on relevant assessment time frames: these are selected on the basis of the periods over which key safety functions are expected to operate. In addition, the highly focused, visual description of the storyboard forms an ideal interface for both developing interdisciplinary consensus and communicating this to other stakeholders. Additionally, a key issue in Japan involves assessing natural phenomena at times beyond present condition of plate tectonics, which requires expert judgment of geologists who understand the context in which this assessment will be used. The uncertainties associated with natural phenomena should be dealt with as variant scenarios.
Figure 5. Storyboard of the EBS for the period of thermal phase.
T-2 thus provides the required transparency and traceability, in a manner that is easy to understand, while integrating interdisciplinary knowledge with a focus on relevant assessment time frames: these are selected on the basis of the periods over which key safety functions are expected to operate.
2.3.3. Description of detailed processes and their relationship to safety functions: T-3
T-3 provides more detailed analysis of the key processes that may significantly impact identified safety function. Because experts are required to focus on detailed characterization of specific local processes, it is often difficult to ensure that all larger scale influences and interactions between processes are taken into account. While the storyboard helps solve this problem by providing the scenario context and the boundary conditions in an easily understood form, it is too much trouble to describe detail relationships and interaction influences between THMC processes in the storyboard due to space limitation. T-3, therefore, is focused on interaction influences between THMC processes and their propagation to the safety functions for the assessment time periods identified in the storyboard and must carefully assess the time-dependency of both processes and the parameters that characterize them. This task has used a form of PID based on the relationships between the processes.
and 8 illustrate the relationship between processes during the thermal period and the period of approximately 105 years after overpack failure. Engineered barrier system components examined here are the vitrified waste, steel overpack, bentonite buffer, concrete liner (ordinary portland cement) and excavation disturbed zone (EDZ). Initially the possible relationships between processes are listed within the form of PID as described in section 2.2.2, and then some links are identified as negligible processes in terms of PA through detail analysis (T-4). and 8 also include the relationships between the safety functions and processes. The rationale of introducing the safety functions into the PID is to identify not only key processes, but also perturbation processes that may seriously influence the safety functions. This is useful to guide derivation of variant scenarios, while keeping tight rein on potentially huge increases of the number of assessment cases if relevance to performance is not kept in mind.
Such detailed description may require scoping calculations to allow different processes to be put in perspective and guide assessment of their influence on safety functions. For example, describes processes and their relationships during the thermal phase based on a number of scoping calculations which were implemented in the H12 report and the further R&D. From the specification of model vitrified waste and assumed 50 years cooling prior to disposal, the temperature within the buffer will reach a maximum (less than 100°C) within about 10 to 30 years after emplacement of the EBS [12]. Groundwater initially contains oxygen trapped within the EDZ during the operational phase and the unsaturated buffer contains trapped air. O2 will be consumed by the reactions with buffer minerals and corrosion of the hot overpack, leading to development of anaerobic conditions.
Groundwater will interact with the concrete liner. After hydration, the solid phases within concrete include portlandite, Calcium-Silicate-Hydrate gel, ettringite, etc [13]. Initially, soluble components (especially NaOH/KOH) are leached from the concrete. Due to reaction with component minerals, buffer porewater pH will be reduced (possibly to around pH ∼11.5) [13]. This pH range favours uniform corrosion of the overpack, hence reducing the risk of localized corrosion. However, if evaporation of porewater occurs near the overpack, significant quantities of salts may be precipitated [14]. Although porewater subsequently dissolving salts may have a composition favouring localized corrosion, such salt composition will rapidly move by diffusion to the outside of buffer with re-saturation. The risk of localized corrosion is thus dependent on system-specific details of evolution of the EBS with a defined geological environment. This type of analysis is a key component of T-4.
In the past, some “problem areas” (e.g. gas, colloids, microbes and organics) may have been excluded from the base scenario due to lack of quantitative understanding or scoping calculation. In contrast, it should be emphasized that the aim in T-4 is to develop as realistic and complete a description of system evolution as possible. For example, gas generation could have a negative impact to some safety functions. Such process is described in the PID based on the system understanding. However, hydrogen production by gamma radiolysis of water can be shown to be negligible and production by anaerobic corrosion of the overpack is problematic only if it generates gas too quickly for it to be removed by molecular diffusion in solution. Past concerns here can be attributed to clearly very conservative gas generation assumptions, associated with a corrosion rate based on not the system understanding, but the design minimum life time of the overpack. Therefore, more realistic calculations or experimental studies can support that the gas effect is not significant process.
Figure 8 shows relationships between the process and the safety function focusing on EBS scale after overpack failure based on the storyboard (see Figure 6). The buffer material has a central role to ensure that no significant release from the wastes to the geosphere. In this context, the safety functions directly involved nuclide migration, which are diffusion, retardation, solubility and low permeability, and associated detrimental processes such as the effect of the high pH groundwater from the concrete liner [e.g. 15] and the effect of corrosion product of overpack to the buffer material [e.g. 16], which were not paid careful attention in the H12 report, may be focused in this time period. It should be therefore analysed qualitatively or quantitatively the extent to which these processes will contribute the alternation of the buffer and their performance will decline due to the alteration. However, through this trail run, it was recognized that the further R&D is needed to understand the mechanism of processes for the assessment cases. A sensitivity analysis or an uncertainty analysis associated with PA parameters related to the safety function of the buffer can provide complimentary information to develop the specific scenario. This type of studies is considered in T-4 and carried out in an iterative manner to identify priorities for which uncertainties should be focused as future R&D.
Figure 6. Storyboard of the EBS for the period of approximately 105 years after overpack failure.
Figure 7. Process influence diagram of the EBS for the period of thermal phase (N: negligible process, hexagonal box: safety function).
Figure 8. Process influence diagram for the EBS for period of approximately 105 years after overpack failure (N: negligible process, hexagonal box: safety function).
2.3.4. Uncertainty analysis of safety functions and development of radionuclide release scenarios: T-4
The final step in assembling the time sequences of the processes describing evolving safety functions includes explicit assessment of associated uncertainties. Scoping calculations can have a role here, particularly as a tool to show that specific issues can be eliminated from further consideration. and 8 show possible relationships of processes with negligible significance, which are indicated by the circle “N”. For example, the effect of precipitation of salts at the inside of the buffer may affect the corrosion of the overpack. However, as described in T-3, when the buffer is saturated by infiltration of the groundwater, they are redissolved and migrate towards the outside of the buffer. Nevertheless, if there is a possibility that they could have an effect on important safety functions, such processes will be dealt as uncertainties and analysed in T-4. Technical knowledge of associated uncertainties that includes effect to the safety function and possibility of the occurrence will be comprehensively considered in the scenario classification (T-5).
Finally, T-4 results in a set of scenarios that summarize possible path of system evolution, behaviour of safety functions and processes relevant to radionuclide migration at different time periods. outlines evolution of expected safety functions and their evolution, processes relevant to radionuclide migration and associated uncertainties for the thermal period and the period of approximately 105 years after overpack failure. The evolution of safety function is summarized based on the relationship between safety functions in and description of detail processes in and . The descriptions of processes relevant to radionuclide migration in provide fundamental information about required model development and associated data sets for the assessment cases. For example, glass dissolution (radionuclide release from glass), diffusion and retardation in the buffer, and elemental solubility in , which are key components to radionuclide migration model in the buffer, are linked to the system understanding described from the T-2 to the T-4 through the safety functions. This structured approach is intended to facilitate the transparency and traceability of the process from the descriptions of system evolution to the derivation of assessment cases.
Table 1. Evolution of safety functions, processes relevant to radionuclide migration and their uncertainties.
2.3.5. Scenario classification: T-5
In Japan, regulatory guidelines under development are expected to specify different performance goals or assessment approaches for different classes of scenario. The final task, T-5, thus involves definition of assessments cases for the various scenario classes that result from T-4. Since supporting scoping calculations are conducted within T-3 and T-4, the scenarios output in this task include not only description of system understanding, but also the models and data which can be used for subsequent PA consequence analysis.
3. Discussion of test application
As the result of trial test cases to examine practicality of scenario development using the hybrid scenario development work frame, the following observations were recorded focusing on differences of bottom-up approach and top-down approach:
| • | The generic bottom-up approach comprises the elements of FEP collection, classification and screening, followed by scenario construction. In this approach, FEP processing is always major theme. Although a number of methods for handling FEPs have been developed (e.g. event tree, logic diagrams, fault tree etc.), it is clear that these methods were much less formalized. On the other hand, in the hybrid scenario approach, the overview of system evolution is firstly depicted, and then key FEPs that might have significant impact to safety functions are indentified. Due to these processes, we confirmed that understandability and traceability that are basic requirements of the scenario development was improved through the attempt of scenario development. Furthermore, by providing a focus on safety relevance, the number of processes considered could be reduced and key processes were evaluated at only the needed level of detail. | ||||
| • | The derived scenario should be included the present (initial state of the system) and the future evolution of the system in whether it is the top-down approach or the bottom-up approach, but it is acknowledged that in some safety assessments system description covers only a representative time [e.g. 5,11]. On the other hand, the hybrid approach segments the evolution of the repository into some “representative situations” from the phenomenological view point, such as thermal, hydraulic, mechanical and chemical processes, and storyboards can succinctly illustrate a set of temporal evolution sequences which provide a holistic overview for subsequent identification of relevant scenarios. As a result, we confirmed that this approach enables to evaluate whether the disposal components with regard to the safety functions fulfil their expected performance in the segmented period. This process is quite consistent with fundamental idea of the safety concept of the geological disposal: the safety functions have their own time characteristics and post-closure safety shall be provided by performing multiple safety functions simultaneously or time-differently. This advantage of hybrid approach is particularly important in Japan, where there is a need to determine long-term evolution of the potentially complex geological environments, considering factors such as climate change, sea level change and uplift/erosion, which may be significant in volunteer sites. | ||||
| • | In general, the top-down approach consists of identifying the safety functions and focusing on what processes could jeopardize one or more safety functions. In this approach, to ensure completeness of potential processes with regard to the safety functions is major concern. This is because only top-down approach itself will be insufficient to handle scientific knowledge that is essential elements to develop scenarios with confidence. In the hybrid approach, we confirmed that the storyboards provided an effective interface for linking the FEP based phenomenological knowledge and safety functions, which can form the basis of audits to demonstrate completeness. | ||||
| • | Throughout the overall timeframe of scenario development thermal, hydraulic, mechanical and chemical processes have their own time and space characteristics, which determine the successive state of the disposal system. Storyboard provided a time and space frame to analyse the evolution of the disposal system, a common platform to involve close interaction with experts in various disciplines to understand the crossover phenomenological processes and to support consensus building: not only for experts in diverse disciplines, but also for non-technical stakeholders. Descriptions with regard to propagation of processes in the storyboard, however, should be sufficient to describe relevant phenomena in a manageable level of detail. We recognized that deciding an appropriate level detail is a sensitive matter in this hybrid approach. Too much superfluous detail not only complicates subsequent consequence analysis, but also risks losing the transparency and ease of communication that were objectives of this study. One possible solution is that the storyboards are drafted by highly experienced generalists but, thereafter, details are refined and technical quality assured by iterative review by appropriate experts. | ||||
| • | By systematically recording the processes of scenario development using conceptual models, key words and an easily interpreted PID, the transparency and traceability of scenario development is enhanced. A critical issue is ensuring that all expert judgments and justifications for decisions associated with each scenario are traceable and reasonable. | ||||
The scenario development work frame in Figure 1 shows a straightforward work flow. However, in practice, it is seen that each task involves several iterations and it is clear that the entire exercise is worth repeating to maintain an overview of the status of the safety concept as project progresses and R&D expands the supporting knowledge base.
4. Conclusions
We have developed a hybrid approach that are combined the top-down approach and the bottom-up approach. In the hybrid approach, a top-down perspective gives an overview of system evolutions and crucial safety functions to provide straightforward directions to decide which FEPs to include in the PA. One of the key components of this work frame is the construction of storyboards. The storyboard that describes system evolutions based on segmentation in space and time provides a baseline and holistic overview for the FEP management and a common platform to involve close interaction with experts in various disciplines to understand the crossover phenomenological processes.
On the other hand, a bottom-up perspective gives fundamental scientific knowledge for establishing scenarios, which is used to keep transparency for the derivation of the scenario. Through a trial of the methodology, we confirm that there is no conflict between the top-down approach and the bottom-up approach, and the hybrid scenario development work frame fulfils the specified requirements for traceability, comprehensiveness, ease of understanding, integration of multidisciplinary knowledge and applicability to a staged approach to siting.
Scenario development is a complex activity that does not involve a standard methodology. Furthermore, scenario development is no simple sequential activities then, undergoes iterations in the frame of safety case; it needs to be tailored to the requirements of specific national programmes. In Japan, it will play a key role in ongoing communication between implementer and the regulator and other stakeholders in order to develop regulatory guidelines and encourage volunteer host communities to come forward. In the next stage of development, the work frame will be applied to a wider range of repository concepts and siting environments to check that it has the flexibility required to respond to the challenges of the volunteering siting programme.
Acknowledgements
The authors acknowledge Dr. I. McKinley of McKinley Consulting for useful comments on the scenario methodology and some polishing of the English. Thanks to anonymous reviewers for their valuable comments to improve the original manuscript.
References
- OECD/NEA . 1992 . Safety Assessment of Radioactive Waste Repositories: Systematic Approaches to Scenario Development , Paris, France : OECD .
- IAEA . 2004 . Safety Assessment Methodologies for Nuclear Surface Disposal Facilities (ISAM), Vol. 1, Review and enhancement of safety assessment approaches and tools , Vienna : IAEA .
- SKB . 2006 . SR-Can Project: Long-term Safety for KBS-3 Repositories at Forsmark and Laxemar – A First Evaluation, TR-06-09 SKB, Stockholm, Sweden
- ANDRA . 2005 . Dossier 2005:Argile, Tome, Safety Evaluation of a Geological Repository ANDRA, Châtenay-Malabry Cedex, France
- Nagra . 2002 . Project Opalinus Clay:Safety Report, Demonstration of disposal feasibility for spent fuel, vitrified high-level waste and long-lived intermediate-level waste (Entsorgungsnachweis), Technical Report 02-05 , Nagra, Wettingen, Switzerland .
- OECD/NEA . 2001 . Scenario Development Methods and Practice, An Evaluation Based on the NEA Workshop on Scenario Development, OECD/NEA Madrid, Spain
- Nuclear Safety Commission of Japan . 2010 . Policy of the Safety Assessment of Sub-surface Disposal after the Period for Active Control , Tokyo, Japan : Nuclear Safety Commission of Japan .
- SKB . 1999 . Deep repository for spent nuclear fuel, SR97-Post-closure safety, Main Report Volume I, TR-99-06 , SKB, Stockholm, Sweden .
- Kanzaki , Y. , Takeda , S. and Kimura , H. 2009 . FEP Database for Scenario Construction on Groundwater Migration in Safety Assessment or a Geological Disposal of High-level Radioactive Waste , JAEA-Data/Code 2009-011, Japan Atomic Energy Agency (JAEA) , Ibaraki, Japan . [in Japanese].
- OECD/NEA . 2000 . Features, Evens and Processes (FEPs) for Geologic Disposal of Radioactive Waste: An International FEP Database , Paris, France : OECD .
- Japan Nuclear Cycle Development Institute (JNC) . 2000 . H12: Project to Establish the Technical Basis for HLW Disposal in Japan , JNC, Ibaraki, Japan : Supporting Report 3, Safety Assessment of the Geological System, JNC TN1410 2000-004 .
- Japan Nuclear Cycle Development Institute (JNC) . 2000 . H12: Project to Establish the Technical Basis for HLW Disposal in Japan , JNC, Ibaraki, Japan : Supporting Report 2, Repository design and Engineering Technology, JNC TN1410 2000-003 .
- Japan Atomic Energy Agency, and The Federation of Electric Power Companies of Japan . 2007 . Second Progress Report on Research and Development for TRU Waste Disposal in Japan , JAEA and FEPC, Tokyo, Japan : JAEA-Review 2007–10, FEPC TRU-TR2-2007-01 .
- Japan Atomic Energy Agency (JAEA) . 2010 . Development of assessment on chemical effect in the disposal system , Tokyo, Japan : Commissioned Report from Ministry of Economy, Trade and Industry . [in Japanese]
- Kosakowski , G. , Blum , P. , Kulik , D. , Pfingsten , W. , Shao , H. and Shingh , A. 2009 . Evolution of a generic Clay/Cement interface: First reactive transport calculations utilizing a Gibbs energy minimization based approach for geochemical calculations . J. Environ. Sci. Sustainable Soc , 3 : 41 – 49 .
- Ishidera , T. , Xia , X. , Idemitsu , K. and Kikuchi , Y. 2008 . Corrosion products from carbon formed in compacted bentonite under reducing conditions . J. Nuc. Sci. Tec , 45 : 763 – 772 .
- Nuclear Waste Management Organization of Japan (NUMO) . 2004 . Evaluating Site Suitability for a HLW Repository, NUMO-TR-04-04 NUMO, Tokyo, Japan
- Japan Nuclear Cycle Development Institute (JNC) . H17: Development and management of the technical knowledge base for the geological disposal of HLW , Repository Engineering Technology, JNC TN1400 2005-022, JNC, Ibaraki, Japan, 2005. [in Japanese] .
- Japan Atomic Energy Agency (JAEA), Database on R&D of Safety Assessment in the CoolRep/Kernel, JAEA Knowledge Management System (2009) Available at http://kms1.jaea.go.jp/CoolRep/index.php/kernels/2009-06-26-09-47-26. (http:\\kms1.jaea.go.jp\CoolRep\index.php\kernels&\2009-06-26-09-47-26.)