Darlington New Nuclear Power Plant Project Joint Review Panel

___________________________________________________________________________________________________________________________________________________________________________________________

Projet de nouvelle centrale nuclaire de Darlington Commission dexamen conjoint

PMD 11-P1.221 File / dossier : 8.01.07 Date: 2011-02-21 Edocs:3682356

Written submission from Greenpeace

Mmoire de Greenpeace

In the Matter of

lgard de

Ontario Power Generation Inc. Environmental Assessment pursuant to the Canadian Environmental Assessment Act of a proposal by Ontario Power Generation for a Project that includes site preparation, construction, operation, decommissioning and abandonment of up to four new nuclear power reactors at its existing Darlington Nuclear Site located near Oshawa, Ontario, in the Municipality of Clarington and a Licence to Prepare a Site application for the Project under the Nuclear Safety and Control Act.

Ontario Power Generation Inc. Lvaluation environnementale, en vertu de la Loi canadienne sur lvaluation environnementale, du projet dOntario Power Generation qui inclut la prparation de lemplacement, la construction, lexploitation, le dclassement et labandon de jusqu quatre nouveaux racteurs nuclaires sur le site de la centrale nuclaire Darlington prs dOshawa (Ontario), dans la municipalit de Clarington, et une demande de permis de prparation de lemplacement, aux termes de la Loi sur la sret et la rglementation nuclaires.

Public Hearing March 21, 2011

Audience publique Le 21 mars 2011

Darlington New Nuclear Power Plant Project Comments on Ontario Power Generations Environmental Impact Statement Prepared by: Shawn-Patrick Stensil

February 2011

Summary
Modern environmental reviews require a proponent to answer the following question: does the proposed project create momentum towards a more sustainable society? If the answer is yes, the project may proceed. If the answer no, the project should either be significantly modified or outright rejected. Ontario Power Generations (OPG) proposal to build new reactors at Darlington fails this test and would in fact accelerate unsustainable development in Canada. OPGs request for approval should, therefore, be denied. OPGs Environmental Impact Statement (EIS) is only able to conclude building new reactors will make a positive contribution to sustainability by the omission and minimization of the projects significant costs and risks. In this submission, Greenpeace outlines why the approval of new reactors at Darlington would violate Canadas legal and policy commitments to sustainable development. The approval and construction of new reactors at Darlington would impose the threat of significant and irreversible harm, whether by the million year legacy of radioactive waste or the realistic risk of severe accidents or terrorist attacks over the next hundred years. Sustainability assessment calls for precaution when faced with the potential for causing significant and irreversible harm. This requires not simply minimizing such threats, but eliminating them. In the context of the current review, this requires an open and transparent review of the of non-nuclear energy alternatives to new reactors. A key flaw of OPGs sustainability assessment is its failure to evaluate the need for the project or assess alternatives without catastrophic risks. Worse, the intensive financial and infrastructural investment needed to proceed with new reactors at Darlington would essentially lock Ontario into nuclear reliance into the next century. New reactors would thus limit the flexibility of future generations to adapt to changing opportunities. Sustainability also requires project proponents to assume responsibility for the Projects full costs. OPGs sustainability assessment ignores the significant costs and risks created by its new reactors that will be transferred to the public, environment and future generations. It also avoids addressing the Polluter-Pays principle enshrined in Canadian law and policy. Taking these issues into consideration, Greenpeace concludes that OPGs sustainability assessment is misleading. Building new reactors at Darlington would not benefit Canadian society, but instead it will accelerate unsustainable development in Canada, block the development of clean energy alternatives and needlessly impose of risks and burdens onto future generations. For these reasons, the request to build new reactors at Darlington should be rejected.

Structure of Submission
This submission is divided into three sections and is supported by five attached documents. Section one provides high-level background and commentary on the need for, and the best practices associated with, a sustainability assessment. Section two discusses the principles of sustainability and how political and institutional interests have undermined them within the current review. Section three discusses the omissions, evasive assertions, and false assumptions that allow OPG to conclude that the Darlington project makes a positive net contribution to sustainable development. Section four summarizes the conclusions and Greenpeaces recommendations.

1. Sustainability Assessment Leaving a Positive Legacy

Sustainability assessment is a planning tool that can help decision-makers shift Canadian society away from the limits of environmentally corrosive growth. In contrast to environmental assessment, sustainability assessment goes beyond merely identifying and mitigating the negative impacts of an undertaking. Instead it seeks to ensure that all new or renewed projects, programs, plans and policies make a positive and enduring contribution to community well-being. 1,2 A projects benefits to society are not determined through a calculus of trade-offs between social, economic and ecological factors, but by its legacy. Simply put, sustainability assessment asks the following question: Does a projects economic, social and ecological effects create momentum towards a more sustainable society? Sustainability assessment, then, is created to discourage decisions that will result in the transfer of adverse effects or risks to future generations. 3 Such adverse effects would create a negative legacy in Canadian society and, as such, would be antithetical to sustainable development. This central aim of sustainability assessment (i.e., to impart long-lasting positive benefits to societies) is of particular importance in the present review of the Darlington project given the long-lived radioactive fuel and decommissioning wastes created by the project, as well as the impacts of realistic accident scenarios. The Panels evaluation of OPGs Environmental Impact Statement (EIS) is key to ensuring that the project contributes net social, economic, and ecological benefits to Canadians. This, in turn, is key to the credibility and effectiveness of the Darlington projects contributions to Canadas sustainable development goals. Given the scale and risk of irreversible harm posed by the project, Greenpeace urges the
1

Gibson, R. (2000) . Favouring the higher test: Contribution to sustainability as the central criterion for decisions under the Canadian Environmental Assessment Act. Journal of Environmental Law and Practice 10 (1): 39-54. 2 Pope. J., Annandale, D., & Morrison-Saunders, A. (2004), Conceptualising sustainability assessment. Environmental Impact Assessment Review 24: 595-616. 3 Gibson, R.B. (2006). Sustainability assessment: Basic components of a practical approach. Impact Assessment and Project Appraisal 24(3): 170-182.

Panel to seriously consider and rule on whether OPG has passed the contribution to sustainability test. As will be discussed, Canadian legislation and past precedents establish the Panels obligations to consider the question. To assist the Panel, Section 1.1 provides a summary review of sustainability planning and assessment best practices. These best practices outline how OPG should have incorporated sustainability concerns throughout decision-making in order to pass the contribution to sustainability test. Given the projects potential scale and long-term impacts , Greenpeace believes that these best practices must provide minimum criteria for assessing whether OPGs project passes the contribution to sustainability test. Section 2.0 describes some critical gaps in OPGs consideration of sustainability in planning the NND Project. Specifically, it highlights major shortcomings in OPGs methodology, consideration of the sustainability effects of the economic costs, radioactive wastes, and cumulative effects of the project; consideration of the sustainability effects of potential accidents and malfunctions; and consideration for the true need for the project, including alternatives to the project.

1.1 Sustainability assessment best practices: The minimum bar for Darlington
The significant and irreversible environmental effects associated with the construction of new reactors at Darlington requires a level of sustainability assessment that meets, at a minimum, the precedents set by Panel reviews of other large industrial projects under the Canadian Environmental Assessment Act. As will be discussed, Greenpeace believes that OPGs sustainability assessment has not met such a level or rigour. Previous reviews, notably of the Mackenzie Valley Pipeline project, Whites Point Quarry and Marine Terminal, and of the Voiseys Bay Mill and Mine project, provide Canadian precedents for sustainability assessment. These Panel sustainability evaluations demonstrate that thorough evaluations of the sustainability effects of projects can result in significant adjustments in those projects. Indeed, the aforementioned reviews resulted in decisions to reject the Whites Point Quarry and Marine Terminal projects. 4 In the Mackenzie Valley and Voiseys Bay examples, the sustainability-based evaluation underpinned the Panels requirements for significant adjustments in the projects in order to ensure net benefits to society. 5,6,7 In other words, these exemplary Panels embraced their responsibility to ensure that the projects would contribute net benefits to society. They did not hesitate to exercise considerable influence over the nature of such large industrial undertakings.

Fonesca, A., & Gibson, R. Application denied: BCs Kemess North and Nova Scotias Whites Point projects promised jobs and revenue, but the communities were looking for overall sustainability. Alternatives Journal 34 (4): 9-11. 5 Gibson, R. (2002) . EA in Canada: From wreck cove to Voiseys Bay: The evolution of federal environmental assessment in Canada. Impact Assessment and Project Appraisal 20 (3): 151-159. 6 Gibson, R.B. (2006). Sustainability-based assessment criteria and associated frameworks for evaluations and decisions: theory, practice and implications for the Mackenzie Gas Project review. http://www.ngps.nt.ca/registryDetail_e.asp?CategoryID=271] 7 Gibson, R.B. (2006). Sustainability assessment and conflict resolution: Reaching agreement to proceed to the Voiseys Bay nickel mine. Journal of Cleaner Production 14: 334-348.

Despite the considerable industrial and political support pushing for approval of these projects, the Panels rightly determined that, unmodified, the initial project designs would significantly accelerate Canadas current course toward unsustainable development. There is also considerable political and industry pressure on the current Panel to speedily approve the Darlington project without caveat. In light of the significant risks and irreversible long-lived effects of OPGs proposed undertaking, it is incumbent on this Panel to be prepared to exercise the same agency to modify and reject the current project. Based on the precedents set by the aforementioned Panels, the following sections, below, summarize the four essential steps in sustainability planning and assessment. Given the projects irreversible impacts, these four steps should be the minimum benchmark against which the Panel should evaluate whether OPG has passed the contribution to sustainability test. Request: In assessing and ruling on the adequacy of OPGs sustainability assessment, Greenpeace respectfully requests that the Panel use as a minimum the precedents of analysis established by past Panel Reviews.

Step One: Early Adoption of Sustainability Decision-Making Framework

The first step in evaluating the conclusions of OPGs sustainability assessment is to determine whether OPG adopted at the outset of planning a comprehensive sustainability framework to guide decision-making. In a previous submission to the Panel, Greenpeace provided a good example of sustainability decision criteria developed by Gibson et al. 8 (see Appendix A). These criteria are appropriate because they devote attention to a comprehensive set of social, economic, and ecological sustainability concerns in way that recognizes how these concerns are connected to each other. I To determine whether the conclusions of OPGs sustainability assessment is credible, the Panel must first ask: Where is the evidence that OPG adopted a comprehensive sustainability framework at the outset of planning in order to guide decision-making? Answering this question is essential to determining whether consideration for net contributions to social, economic, and ecological gains has shaped OPGs planning from the earliest stages. If OPG were to produce any evidence, the Panel must inquire into whether it is sufficient. To be sufficient, the Panel must be satisfied that any sustainability decision framework developed by OPG was (a) comprehensive, and (b) was adopted at the outset of planning. Greenpeace recommends that the Panel use Dr. Gibsons sustainability criteria in order to evaluate the comprehensiveness of any framework adopted by OPG. This is the criteria

Gibson, R.B., Hassan, S., Holtz, S., Tansey, J., & Whitelaw, G. (2005) . Sustainability Assessment: Criteria and Processes. London: Earthscan.

that the Ontario Power Authority, the planning authority responsible for recommending Ontario proceed with new reactors, used Dr. Gibsons sustainability criteria in its 2007 submission to the Ontario Energy Board. A sustainability decision framework must devote attention to: (a) cumulative effects, (b) links within and between social and ecological systems, present and future generations, and multiple scales, (c) efficiency in the use of natural resources, (d) the minimization of extractive damage and waste, (e) equity and justice in the distribution of costs and benefits within and between present and future generations, (f) the maintenance, protection, and enhancement of ecological systems, governance and administrative capacity, local economies, and basic individual needs, rights, and freedoms, (g) use of collaborative, participatory processes in decision-making, and (h) immediate and long-term implementation.

Step Two: Specifying Generic Sustainability Decision-Making Criteria

As mentioned, the first step in sustainability planning is to adopt at the outset of planning a comprehensive set of sustainability criteria to guide decision-making. Dr. Gibson criteria have been provided as a benchmark against which other frameworks can be evaluated. Gibsons criteria are generic in that they refer to broad sustainability goals that can be specified for any case and context. However, in addition, any sustainability framework applied in the Darlington Project should have been specified for the case and context. This second step specification involves gathering information about the particular stakeholder and sustainability concerns relevant to each sustainability criterion. It is best to begin with generic criteria and then sort the context-specific concerns appropriately under the generic criteria. The key test is to ensure that all of the fundamental sustainability requirements and all of the main case and context-specific concerns are recognized and incorporated in one consolidated framework. The end result would be a sustainability framework for decision-making that considers all contextual social, economic, and ecological issues and interests related to the case. The criteria could then be applied in decision-making to ensure that the project addresses the identified issues and interests in a way that ensures that the project contributes net benefits to community well-being.

Step Three: Consistent Application Through-out Decision-Making

With sustainability decision criteria established, the next step of sustainability assessment is to apply them consistently throughout decision-making. Judgment of the adequacy of OPGs sustainability assessment and, by extension, the net sustainability effects of the Darlington project, should, at a minimum, be based on whether

OPG applied a specified set of sustainability decision criteria to every component of the Darlington project. OPG must demonstrate how consideration for sustainability issues influenced its decisions with respect to all Project phases and components. This means that OPG must demonstrate how it compared the net sustainability contributions of core components of all Project phases. OPGs sustainability assessment should be judged on whether it has chosen options that deliver the greatest net benefits to community wellbeing. OPG should have begun with the intent to pursue the most promising options from among the available alternatives, including reactor refurbishment and all other energy options. Specifically, the sustainability criteria should have informed the following considerations with respect to Darlington project planning and decisionmaking: How interested stakeholders are to be engaged in the planning process and how different perspectives and scientific and traditional knowledge can be accommodated; What alternative operation, maintenance, decommissioning and abandonment options (e.g., reactor designs, condenser cooling methods, radioactive waste management options, storage of used fuel options, etc.) exist; What alternative operation, maintenance, decommissioning and abandonment options are most beneficial from a sustainability perspective; What potential effects (e.g., direct, indirect, cumulative, etc.) deserve detailed attention; What effects may be most significant in light of sustainability objectives; How any positive effects of the Project should be enhanced; How any adverse effects should be mitigated; Which trade-offs may be unavoidable and which ones might be most acceptable or least acceptable; and What monitoring and management requirements should be imposed on the project.

Step Four: Sustainability-Based Trade-Offs

In the fourth step essential to evaluating whether new reactors at Darlington will contribute net benefits to society, OPG must consider the trade-offs among the alternative means to carrying out the project, including the trade-offs among new build, refurbishment options, and other non-nuclear energy options. For example, some of the key trade-offs in the choice between new build nuclear versus refurbishment include, among others: Higher efficiency of new build versus less path dependency of refurbishment (i.e., with respect to nuclear generation as the major electricity source for Ontarians, and the transmission system); New and uncertain technology associated with new-build versus the better understood risks associated with the older mature technology of existing units.

Dr. Robert Gibson has developed a set of sustainability-based trade-off rules to guide planning, decision-making, and analysis when conflict arises between objectives, or when achieving one desired result compromises another 9 (see Appendix B). Briefly, the trade-off rules emphasize ensuring choices that allow for maximum progress towards sustainable development in ways that deliver benefits and gains in all sustainability criteria. Trade-offs should be considered as a last resort, only after alternatives have been evaluated in light of the sustainability criteria. The burden of proof or justification is placed on the proponent of a given undertaking. No trade-off that involves negative impacts on sustainability can be justified unless the alternative involves an even more significant negative impact. Similarly, no adverse impacts may be transferred from present to future generations unless the alternative involves acceptance of a more significant adverse effect. Trade-offs must be openly and explicitly justified based on sustainability principles and effective involvement of stakeholders. In summary, Greenpeace urges the Panel to seriously consider and rule on whether OPG has passed the contribution to sustainability test. Given the scale and risk of irreversible harm posed by the project, Greenpeace urges the Panel to ensure the methodology used to determine sustainability is both transparent and, at a minimum, on par with the precedents set by past review panels.

2. The Principles of Sustainability

Two principles underpin sustainable development: the Precautionary principle and the Polluter-Pays principle. As will be discussed, both of these principles are established in Canadian law and decision-making.

2.1 The Precautionary Principle

The precautionary principle is a central concept in environmental law and policy that requires that decision makers look before they leap when making decisions that could have adverse impacts on the environment or where the environmental impacts of the decision are not known. The principle has been adopted and applied by courts across Canada. The Supreme Court of Canada states: In order to achieve sustainable development, policies must be based on the precautionary principle. Environmental measures must anticipate, prevent and attack the causes of environmental degradation. Where there are threats of serious or irreversible damage, lack of scientific certainty should not be used as a reason for postponing measures to prevent environmental degradation.

Gibson, R.B., Hassan, S., Holtz, S., Tansey, J., & Whitelaw, G. (2005) . Sustainability Assessment: Criteria and Processess. London, Sterling: Earthscan.

Key to the application of the precautionary principle in the case of the current review is the consideration of alternatives to the project itself - including taking no action. While OPGs proposal to build new reactors comes with the realistic potential to create significant and irreversible harm, there are many known and tested energy options without such risks that could effectively address the claimed need for the project. The precautionary principle attempts not to mitigate potential harm but prevent it. As will be discussed, the lack of alternatives assessment is a central flaw of the OPGs sustainability assessment.

2.2 The Polluter Pays Principle

The Supreme Court of Canada unanimously upheld the polluter-pays principle. In a 2003 decision, the Court stated that the polluter-pays principle has become firmly entrenched in environmental law in Canada. The Court explained the principle as follows: To encourage sustainable development, that principle assigns polluters the responsibility for remedying contamination for which they are responsible and imposes on them the direct and immediate costs of pollution. At the same time, polluters are asked to pay more attention to the need to protect ecosystems in the course of their economic activities. 10 As will be discussed, OPGs Darlington project assumes the transfer responsibility for costs and risks onto Canadian society, the federal tax-payer and future generations. This is in violation of the Polluter-Pays principle. In assessing the projects contribution to sustainability, this violation of sustainability must be considered. OPGs sustainability assessment does not do this.

2.3 Politics Over Principles: Limitation of the Scope

Greenpeace believes that provincial and federal bureaucratic interests have acted to deliberately fast-track and limit the scope of the current review in order by-pass meaningful issues of public and environmental concern that would lead to the abandonment of the project. Greenpeace requests the Panel use its discretion under CEAA to return these issues to the scope of the review. Unlike other panel reviews, the scope of the current environmental review was not established by the appointed Panel members. Instead, the scope of the review was set by the Canadian Nuclear Safety Commission (CNSC), the government of Ontario and the Canadian Environmental Assessment Agency (CEAA) behind closed door. This process demonstrates a lack of transparency.

10 Stacey Ferrara and Jennifer Mesquita, Supreme Court of Canada Upholds a Ministers Application of the Polluter- Pays Principle, Gowling Lafleur Henderson LLP, October 2003.

Greenpeace believes the politics of electricity in Ontario and the fate of Atomic Energy of Canada Limited (AECL) federally drove the scoping of the current review. As a result, the principles of sustainable development have been undermined and not appropriately applied in OPGs Environmental Impact Statement (EIS). This is evidenced by the following in the scoping of the review: 1) deliberately limiting the scope of the environmental assessment; 2) characterizing the project description so as to exclude participation by the province; and 3) failing to operate the assessment process in an open and transparent manner. There are few developments that necessitate the use of precaution more than the construction of nuclear reactors given the resulting economic, social and environmental risks they create. Politics should not trump the principles of sustainability. Greenpeace reminds the Panel that its mandate is to uphold Canadas legal and policy commitments to sustainable development.

2.4 Darlington: Last Chance for AECL and CANDU Technology?

The undermining of sustainability principles in the current review, Greenpeace believes, has been driven by the desire to protect nuclear generations historic contribution against new clean energy options by sympathetic government agencies. Ontarios CANDU reactors have prematurely aged and after 2010 will begin to shut down for either expensive repairs or for permanent closure. With clean energy technologies taking much less time to deploy than new reactors, this is the biggest threat for both OPG and AECL maintaining their generation footprint in Ontario. As will be discussed in Section 3.3, quicker-to-deploy energy options, such as renewables, conservation and local generation are already cost-effective and their costs continue to decline. This put new nuclear at a disadvantage - if, that is, an objective public forum were used to assess nuclears cost-effectiveness. Aware of this, the nuclear industry , which has long-standing institutional support in both the federal and provincial agencies, has urged federal and provincial governments speed up approval processes and exempt nuclear stations from thorough reviews to accelerate their construction in order to stay relevant in the energy debate. Federally, for instance, Atomic Energy of Canada Limited (AECL) approached the CEA Agency in 2005 to ask how the environmental assessment process could be shortened. 11 Provincially, the former-heads of the Ontario Power Authority (OPA) and the Ontario Energy Board (OEB) asked former-CNSC president Linda Keen in 2006 to fast-track new reactor construction by grandfathering the safety standards for AECLs CANDU-6 reactor. President Keen refused this request. 12

11 12

Presentation, AECL Meeting with Louise Knox, Ontario Regional Director CEAA, November 8 2005 Linda J. Keen (President, Canadian Nuclear Safety Commission) to Dr. Jan Carr (Chief Exeuctive Officer, Ontario Power Authority), Licensing Process for New Nuclear Power Plants in Canada, January 27, 2006. Acquired through Access to Information.

10

Subsequent events show that the federal governments promotion of CANDU technology trumped its public interest of maintaining an independent nuclear safety regulator. The Harper government fired Keen in 2008 reportedly for her handling of the so-called radioisotope crisis. Keen, however, called the radioisotope crisis an excuse to fire her citing her previous imposition of modern standards to the CANDU-6. 13 Otherwise put, the independence of Canadas nuclear safety regulator has been compromised apparently to remove obstacles to the preservation of CANDU reactor technology in Canada. Greenpeace is deeply concerned that the independence of Canadas nuclear safety watchdog has been compromised by the firing of President Keen. Given current environmental review excludes discussion of safety issues based on the CNSC licensing framework, Greenpeace submits this should also be a concern for the Panel. Greenpeace believes similar institutional pressures to protect and promote the nuclear industry in Canada have limited the scope of the current environmental assessment. Specifically, the sustainability principles that have been accepted as Canadas policy and legal framework have been overlooked to assist in the approval of the current project.

2.5 Provincial Responsibilities: Is OPG or Ontario the Proponent?

Despite the fact that building new reactors at Darlington could cause significant and irreversible harm, OPG has not considered cost-effective non-nuclear alternatives in its environmental impact statement. This is in violation of the precautionary principle and undermines the adequacy of OPGs sustainability assessment. OPG argues that no such assessment is necessary based on the Environmental Impact Statement (EIS) guidelines, which state: As an assessment of provincial policy is not within the terms of reference of this joint review panel, the alternatives to the project need not include alternatives that are contrary to Ontarios formal plans or directives. In its submission to the Panel, Ontarios Ministry of Energy states its unequivocal support for OPGs exclusion of non-nuclear alternatives to the building new reactors at Darlington. 14 Greenpeace believes, however, such a position violates Canadas legal and policy commitment to sustainability. Nothing in the CEAA limits a federal panel from reviewing alternatives a provincial project. Key to any credible environmental assessment is a public review of the justification for a project and alternative means of addressing need for that project. No such review has taken place i for OPGs project, either provincially or federally. As noted, Greenpeace is deeply concerned that institutional interests for protecting the nuclear industrys place on Ontarios electricity system against more modern, sustainable and cost-effective sources of energies, has driven decisions to exempt or exclude nuclear from needs or alternatives assessments. In 2006, for example, the Ontario Government promulgated a regulation to exempt its nuclear-based electricity plan from a provincial environmental assessment, depriving the public of an opportunity to evaluate the nuclear option against other, less dangerous ,

13 14

Nucleonics Week, Volume 50 / Number 16 / April 23, 2009, p. 10 Ontario Ministry of Energy, PMD 11-P1.11, January 31, 2001, p. 12.

11

energy options. 15 To do so, Ontarios independent Environmental Commissioner noted, that the government violated its own environmental protection legislation and deprived the public of their rights to participate in decisions that could have great environmental significance for the people of Ontario. 16 The Ontario government subsequently made a political decision not to participate in this federal environmental review in order to avoid any discussion of alternative energy options. Internal documents obtained through the Freedom of Information Act by Greenpeace establish that the province has decided not to participate in the EA. The province explains the reason for refusing to participate as follows: Further to our joint discussions of October 25 with CEAA and the CNSC, and follow-up internal discussions between staff of the Ministries, it was agreed that the preferred option on Provincial participation in the EA process is for the Province not to participate through a formal agreement with the federal government and not to seek to participate in nominating a panel member. Instead the Province should limit its participation to making a submission to the federal EA panel. this option is preferred since it allows the Province to identify environmental issues important to the Province, but also limits the threat of legal challenge to the Province and the threat that the scope of the environmental assessment will be expanded to include provincial issues such as power system planning. [emphasis added] 17 Ontario has shown a pattern of actively avoiding any scrutiny or objective analysis of its energy policy and of ministerial directives being used limit the scope of the current review. As will be discussed, such a review is necessary given many of the assumptions used to establish the governments directives have proved erroneous. As noted, when a project may cause significant irreversible harm, the Precautionary principle requires other options be assessed. The intent is to eliminate the source of harm and not simply mitigate its impacts. No such transparent review of the need or new reactors at Darlington has taken place provincially or federally. Indeed, the provincial government has shown a pattern of acting to avoid any such assessment. Greenpeace would like to emphasize that nothing in the CEAA limits a federal Panel from reviewing alternatives a provincial project. To the contrary, given the potential consequences of OPGs proposal CEAAs sustainable development goals should motivate such an assessment.

2.6 Precaution calls for Questioning Ontarios Nuclear Directive

As noted, documents acquired by Greenpeace through the Freedom of Information Act indicate that the government of Ontario decided against participating in the current review

15 16

Martin Mittelstaedt. "Nuclear plan skips key green review", Globe and Mail, June 15, 2006. Press Release, Third Decision on Governments Electricity Plan Evades Environmental Bill of Rights, says Environment Commissioner, Environmental Commissioner of Ontario, June 19, 2006 17 Email correspondence between Cedric Jobe [Min of Energy] and Rick Jennings [Min of Energy] (December 11, 2006) RE: The Provincial Role in the Federal EA Process.

12

panel in order to limit the threat that the scope of the environmental assessment will be expanded to include provincial issues such as power system planning. Otherwise put, Ontarios Ministry of Energy has acted to avoid a transparent review of the justification and alternatives for building new reactors at Darlington. This is contrary to principles of sustainability and, if not addressed, undermines the credibility of the current review. Request: The Panel use its discretion under CEAA to ensure the justification, need and alternatives to build new reactors at Darlington is addressed in the current review. Greenpeace notes several significant omissions from the Ministrys of Energys submission to the Panel. These omissions raise questions about the need, justification and cost-effectiveness of the Darlington project. Given their relevance for the sustainability assessment, Greenpeace requests additional information from province, which is arguably the true proponent of the Darlington project. The Cost-effectiveness for New Reactors The Ministry of Energys submission to the Panel omits mention that its 2006 directive for OPG to begin planning for new reactors was based on advice from Ontario Power Authority in 2005. In its 2005 Supply Mix Advice, the OPA advised the government that new nuclear would be a cost-effective supply choice based on its conservative assumption that new reactors would cost $2,900 KWh or about $6 billion dollars for a 2000 MW station. 18 Under cross-examination at the Ontario Energy Board in 2008, the OPA admitted that new nuclear would be uneconomical if the cost topped $3,600 KWh. 19 In 2008, Ontarios Ministry of Energy began a procurement process for new reactors. In 2009, the Ministry of Energy suspended its procurement process saying bids received were significantly too high. While the government refused to release the bid prices, it was reported that AECLs bid topped $26 billion or about $10,000KWh for a 2,400 MW station. 20 This is significantly above the amount cites to justify the governments 2006 directive to proceed with new reactors. It is also well above the OPAs assessment of what would be cost-effective. Despite this significant increase in new reactor costs, Ontario has yet to publicly reevaluate the cost-effectiveness of new reactors. The OPA, however, has already undertaken such a review according to a Freedom of Information request by Greenpeace Canada. 21 Because OPA has refused to release this information, Greenpeaces request is currently being reviewed by Ontarios Information and Privacy Commissioner. It is noteworthy that Ontarios latest energy planning directive to the Ontario Power Authority states that new build nuclear will proceed where it can be achieved in a cost-

Canadian Energy Research Institute, Electricity Generation Technologies: Performance and Cost Characteristics prepared for the Ontario Power Authority, August 2005, p. 7. 19 In response to questioning at the Ontario Energy Board, the Ontario Power Authority admitted that at n overnight capital cost of $3,600 with an additional 2 cents in operating and maintenance costs and an 8 per cent discount rate, nuclear stations would be uneconomical. See: Ontario Power Authority, EB-2007-0707, Exhibit, Tab 43, Schedule 3. 20 Hamilton, Tyler. $26B cost killed nuclear bid, Toronto Star, July 24, 2009. 21 Greenpeace is attempting to acquire a May 4, 2010 document entitled OPA PRELIMINARY FINDINGS OF ECONOMICS OF NEWBUILD NUCLEAR.

18

13

effective manner. 22 Neither the Ontario government nor the OPA has indicated how its criteria for cost-effectiveness is based on the current estimates for new nuclear against other energy options. Request: The Panel should require the Ontario government to release its criteria and assessments for the cost-effectiveness of new nuclear during the current review. Electricity Demand The Ontario government issued its 2006 directive to build new reactors on the OPAs electricity demand forecast, which predicted the need for reactors in addition to the current fleet. Electricity demand, however, has not grown, but dropped since 2006. The governments submission attributes this decline to economic slowdown, but fails to acknowledge that the OPA simply over-estimated demand. The history of electricity planning in Ontario shows plans have always serious over-estimated demand growth and the need for additional supply. As a compliment to the current submission, Greenpeace is providing a copy of the report Review of the Ontario Load Forecast in the Integrated Power System Plan (IPSP) by Ralph Torrie and Doug Morrow. This report provides a summary of history of electricity planning in Ontario and how past electricity plans have always over-estimated the need for additional supply. It is relevant in the current context because justification for the governments 2006 directive was based on a similar over-estimated of demand growth. Whats more, Greenpeace believes the governments current Long Term Electricity Plan (LTEP) repeats yet again the same error. The LTEP assumes electricity demand will decline until 2020 as indicated by current publicly available forecasts. It then assumes electricity demand will again increase after 2020 justifying the need for additional supply. It should be noted Greenpeace has requested through Freedom of Information the long-term electricity forecast used by the OPA in developing the LTEP, but has been refused. Baseload Electricity Related to electricity demand forecasts used to justify the need for the Darlington reactors, is the projected need for baseload demand. Unmentioned in the governments submission to the Panel, is the ongoing issues of surplus baseload supply and the drop in electricity demand. Since 2006 the province has stated it will maintain nuclear capacity at 50% of supply to provide baseload power. This is coincidently in line with historic levels of installed nuclear capacity. Greenpeace has never been able to determine how the OPA and province calculate this need for baseload capacity. As noted, the purpose of this project has shifted with the decline in electricity demand since 2006. Instead addressing increased demand, Darlington reactors are now intended to replace the Pickering reactors when they close in 2020. As will be discussed, a portfolio of green energy technologies is more costeffective than new reactors.

22 Energy Minister Brad Duguid to Colin Andersen, CEO OPA, letter, February 17, 2011.

14

Whats more, given declining electricity demand, there may be no need to replace the Pickering reactors. However, the information required to accurately assess this has been withheld by the provinces electricity planning agency. Request: The Panel should request the Ministry of Energy explain its process for calculating the need for baseload electricity demand and the rationale for maintaining nuclear supply at historic levels. The Cap on Renewables, Local Generation and Conservation The Ministry of Energys submission to the Panel also omits mention that new renewable generation has been capped at approximately 10 15 % of supply in both the LTEP and its predecessor, the Integrated Power System Plan (IPSP). This cap is due to the commitment the government has made to keep nuclear at 50% of supply. As noted, the Ontario government has evaded any publicly assessment of the need or alternatives to new nuclear despite a significant increase in nuclear costs since the initial directives were issued in 2006. Greenpeace believes that this cap on renewable energy is antithetical to the principles of sustainability by denying future generations the flexibility and choice to reduce nuclear reliance. While the costs of nuclear technology has only ever risen, clean energy technologies have been showing significant cost reductions and efficiency gains. Request: The Panel deny approval of the Darlington project in order to maintain the flexibility for the expansion of cost-effective clean energy technologies. In summary, the aforementioned issues raise significant questions regarding the need, justification and cost-effeciveness of the Darlington project. The province of Ontario has repeatedly acted to avoid any scrutiny of these issues. Greenpeace is deeply concerned that opportunities to build a more sustainable electricity system in Ontario will be undermined if no such scrutiny takes place.

15

3. Darlington: Harming Sustainability

OPGs sustainability assessment concludes that building new reactors at Darlington would make a positive net contribution to sustainability. That is, the legacy of the Darlington project will not increase inertia towards sustainable development and not transfer negative impacts on future generations. The following sections argue this assessment is wrong. OPGs proposal to build new reactors at Darlington will accelerate momentum towards unsustainable development in Canada, including a serious imposition of risks and burdens on future generations. The following will highlight the omissions, evasive assertions, and false assumptions that allow OPG to conclude that the Darlington project makes a positive net contribution to sustainable development. Greenpeace concludes that if sustainability principles, such as a precaution and the polluter-pays, are applied to all aspects of OPGs proposal, it will be fail the net benefit to sustainability test. It must therefore be rejected. The following sections explain why the conclusion of OPGs sustainability assessment is misleading.

3.1 Lack of Methodology

After reviewing OPGs EIS, it is apparent that OPG planned the Darlington project without a sustainability framework. Then, after the bulk of the planning was complete, it undertook a sustainability assessment of the project. Otherwise put, OPG had no clear methodology to derive its conclusions. Its desired conclusions established methodology. OPGs sustainability assessment does not provide an adequate basis for the Panel to decide that OPG has passed the contribution to sustainability test. These reasons will be described in brief below: First, the framework OPG developed for the sustainability assessment is not grounded in an adequate understanding of core requirements for progress towards sustainable development. As previously described, Gibson et al. provide an in-depth explanation of these requirements in Sustainability Assessment: Criteria and Processes. Second, OPG does not describe how the evaluative framework addresses all of the environmental effects of the Project. Table 5.15-1 of the EIS provides a helpful summary of the likely environmental effects and residual adverse effects of the construction, operation and maintenance phases of the Project. The sustainability assessment must, at a minimum, address all of these effects. To sufficiently safeguard the well-being of Ontarians, the Panel must require OPG to rethink how key aspects of the Darlington project influence the extent to which the project contributes net benefits to society. Then, the Panel must impose appropriate adjustments. Subsequent sections, below, describe other critical gaps in OPGs sustainability analysis. Request: The Panel must acknowledge OPGs failure to adopt a comprehensive sustainability framework to evaluate the sustainability effects of the Darlington Project, and insist that the framework is consistently applied throughout planning and decision-making.

16

It is not adequate to finish planning and then assess the sustainability effects of the project. A comprehensive sustainability framework should have guided the project from the outset.

3.2 Economic Costs

A credible sustainability assessment must consider of the economic costs of building, operating and maintaining, and decommissioning and abandoning new reactors at Darlington. Greenpeace believes that because OPGs EIS omits or superficially addresses these issues, the positive conclusion of its sustainability assessment are at best misleading, and at worst, wrong. OPG has ignored how the economic costs of the Project affect the extent to which the Project contributes to sustainable development. OPGs conclusion that the Darlington Project makes a positive contribution to sustainability is inaccurate, based on its omission of these costs. Additionally, the proponent ignores the cumulative economic costs of the Darlington Project. Historic experience with nuclear projects in Canada and internationally provides empirical evidence that nuclear projects whether new build or radioactive waste management undergo significant cost escalation, which is often displaced from the project proponent to present rate and taxpayers as well as future generations. 23 This trend has already remerged in Ontario. OPG was directed by the Ontario government in 2006 to initiate the current environmental review after it was advised in 2005 by its electricity planning agency, the Ontario Power Authority (OPA), that the cost of new reactors would be more cost-effective than building other sources of green energy generation. They were mistaken. In 2005, for example, the OPA claimed that new reactors would cost $2,900/KWh or about $6 billion dollars for a 2000 MW station. In 2009, however, the Ontario government suspended its procurement of new reactors when the costs reportedly topped $26 billion or about $10,000/KWh for a 2400 MW station. 24 A credible assessment of the projects contribution or harm to sustainable development must consider these high and uncertain project costs in the context of other energy options. Request: The Panel should acknowledge OPGs failure to adequately consider the sustainability effects of the economic costs of the Darlington Project. This omission is critical in light of the uncertainties surrounding the current economic downturn in most of the world, as well as the uncertainties surrounding the costs of climate change. In particular, the Panel must ensure that OPG addresses: How the economic costs of the Darlington Project will affect the energy bills of Ontarians, i.e., who will pay for the Darlington Project and how long will it take to pay

23

Winfield, M., Jamison, A., Wong, R., & Czajkowski, P. (2006) . Nuclear Power in Canada: An Examination of Risks, Impacts and Sustainability. The Pembina Institute. 24 Hamilton, Tyler. $26B cost killed nuclear bid, Toronto Star, July 24, 2009.

17

for it; What mechanisms OPG has in place to ensure that any overrun costs do not result in higher energy bills for Ontarians; How OPG will secure the capital required for the long-term management (hundreds of years) of radioactive wastes. What mechanisms has OPG put in place to ensure that it can pay for the long-term management of radioactive wastes regardless of future circumstances (e.g., worldwide economic downturns, unforeseen climate change crises, unforeseen security crises, etc.)? Will provincial and federal interests to maintain the CANDU reactor industry lead to the construction of outdated Generation II reactor designs (Specifically the CANDU-6) that will impose higher levels terrorist and accident risk on Canadians for up to a hundred years?

3.3 No Need for New Reactors: Alternatives

OPGs sustainability assessment is built on an assumption that fundamentally contradicts sustainability assessment best practices. Specifically, OPG assumes that near-term jobrelated economic benefits are an acceptable trade-off for the production of long-lived radioactive wastes, which will transfer risks and costs to future generations (See section 3.4). An adequate sustainability assessment of the Darlington project would result in the avoidance of such trade-offs, except where the alternatives are worse. In the case of electricity generation in Ontario, many less environmentally invasive energy sources exist that would help create additional momentum towards a more sustainable economy. It is the position of Greenpeace that in the present context 25 new build reactors are the worst option available, especially given the range of renewable energy options in Ontario. Attached this submission a report entitled Ontarios Green Energy Plan 2.0: Choosing 21st Century Energy Options, which was produced by the Pembina Institute, Greenpeace and the Canadian Environmental Law Association. The report outlines how the assumptions used to justify the Ontarios 2006 electricity directives have been shown to be erroneous. Electricity demand is dropping instead of growing, green energy costs are dropping and new nuclear costs have increased dramatically. The report shows how at todays prices an investment in green energy options would be cheaper than purchasing new reactors. The report also outlines how Ontario initially established a policy to proceed with new reactors in 2006 based on the assumption that new reactors would be required beyond the existing fleet of reactors to address projected demand growth. At the time, Ontario it set a long-term target of 14,000 MW of nuclear supply. Since then, electricity demand is set to fall. New reactors, if constructed, would be intended to replace the Pickering nuclear station when it closes in 2020.

25

Ontario has a policy of phasing out coal electricity. With electricity demand falling, new reactors will be to replace aging nuclear supply. Reducing Ontarios reliance on nuclear supply will create a more resilient, flexible and diverse electricity system and economy.

18

As noted in Section 2.4, Ontarios Ministry of Energy, which issued the directive to OPG to seek approval for building new reactors at Darlington, has systematically avoided any public analysis of whether the project is in fact cost-effective or needed. The government did require the OPA, however, undertake a sustainability assessment as part of its submission to the Ontario Energy Board (OEB). Notably, the OPA used the sustainability framework by Dr. Robert Gibson in its filing to the OEB. The Green Energy Coalition, of which Greenpeace is a member, subsequently hired Dr. Gibson to evaluate the OPAs sustainability assessment. Dr. Gibson et al. concluded that the OPA conclusions were erroneous and that new nuclear in particular was the least desirable from a sustainability perspective given the alternatives. 26 Dr. Gibsons sustainability assessment of the OPAs 2007 IPSP has been attached to this submission. Request: The project should not be approved before there is a public and transparent assessment of the projects costs against other energy options. 3.4 Darlington: A Barrier to More Sustainable Alternatives As noted, the assumptions used to justify the governments 2006 directive to proceed with new reactors nuclear cost, demand growth and the viability of clean energy options have been shown to be wrong. It is essential, then, that the Panel consider other energy options in light of the irreversible effects that new reactors will have on future generations. Greenpeace also believes the panel must consider how the Darlington Project will affect the future of renewable energy supply in Ontario energy options that are just as reliable, cleaner with respect to life-cycle environmental impacts, and more cost effective for many reasons including the fact that renewable energy options do not require long-term radioactive waste management. The current electricity grid in Ontario has been constructed around nuclear as a major supplier for Ontarios baseload energy demand. If the Darlington Project is allowed to continue (i.e., based in part on the Panels evaluation of the Darlington Projects sustainability contributions) one major effect will be to further embed the electricity grid in its current physical pattern, which favours nuclear energy over renewable energy sources. Request: Greenpeace requests the Panel consider the impact approval the project will have on the ability and flexibility of future generations to develop more sustainable energy sources.

3.4 Radioactive Waste

Approving the construction of new reactors and continued radioactive waste production is fundamentally contrary to the purpose of sustainability assessment.

26 Gibson, R., Winfield, M., Markvart, T., Gaudreau, K., & Taylor, J. (2008) . An Analysis of the Ontario Power Authoritys Consideration of Environmental Sustainability in Electricity System Planning. Green Energy Coalition, Pembina Institute, Ontario Sustainable Energy Association.

19

Sustainability assessment is intended to reverse the direction of Canadian society away from unsustainable and destructive development that places increasing burdens on future generations. It is not about simply mitigating the most adverse environmental effects of a project, such as long-lived radioactive waste. 27 OPGs proposal to build new reactors will extend for decades the creation of long-lived radioactive wastes. This aspect of the project would be unacceptable if it were included in a sustainability assessment. OPG has avoided the inclusion of these wastes in its sustainability assessment by claiming that this issue will be considered in future environmental assessments, specifically, in the case of used fuel waste, by the Nuclear Waste Management Organization (NWMO). This claim, however, is erroneous. The fundamental assumption of the NWMOs risk assessment was that the volume of used nuclear fuel which needs to be managed was assumed to be limited to the projected inventory from the existing fleet of reactors. 28 The NWMOs Roundtable on Ethics also acknowledged that finding an ethical nuclear waste management approach is an intractable problem. While acknowledging we must find a method of managing existing wastes, it stated outright that producing additional wastes would be unethical: [g]iven the large stockpile of high level nuclear waste that already exists in Canada and that will be hazardous for thousands of years, some solution to managing wastes as safely and effectively as possible must be found. Even if no ethically optimal solution exists, it would be ethically justified to adopt the least unacceptable option available. By contrast, to justify new nuclear power plants or even replacing the ones now in place when they reach the end of their serviceable life, one would have to have an ethically sound waste management method, not just a least-bad one. 29 [Emphasis added] It should be noted that the Nuclear Fuel Waste Act required the NWMO to take into account ethical considerations when assessing the viability of differing waste management approaches. OPG has ignored such considerations in its EIS. The Advisory Committee of the NWMO made the following important statement in its report to government in 2005: The Advisory Council would be critical of an NWMO recommendation of any management approach that makes provision for more nuclear than the present generating plants are expected to create, unless it were linked to a clear statement about the need for broad public discussion of Canadian energy policy prior to a decision about future nuclear energy development. The potential role of nuclear energy in addressing Canadas future electricity requirements needs to be placed within a much larger policy framework that examines the costs, benefits and

See section Section 6.1.1 of OPG EIS. Assessing the Options The NWMO Assessment Team Report, June 2004, pg 14. Available at: http://www.nwmo.ca/Default.aspx?DN=1091,1090,199,20,1,Documents
28
29

27

See: Ethical and Social Framework (Version: June 24 2004) http://www.nwmo.ca/Default.aspx?DN=744,274,20,1,Documents

20

hazards of all available forms of electrical energy supply, and that framework needs to make provision for comprehensive, informed public participation. OPGs avoidance of this issue is unacceptable from a sustainability assessment perspective because, first of all, waste reduction is the paramount priority in any credible waste management strategy in this case, this can only be accomplished by ending the production of radioactive fuel waste by switching to more sustainable energy options as Canadas nuclear stations reach the end of their operational lives. Secondly, an environmental assessment is not a sustainability assessment. One of the key differences between them is that an environmental assessment is oriented towards assessing and mitigating biophysical impacts. A sustainability assessment moves beyond merely assessing and mitigating impacts by ensuring that projects contribute net benefits to society. In short, an environmental assessment is merely about investigating biophysical impacts and mitigation measures, while a sustainability assessment is about understanding how a project will contribute lasting positive benefits to community wellbeing. To elaborate on critical gaps in the EIS, OPG has still not assessed the sustainability effects of the risks associated with storage and removal of radioactive and other hazardous materials in all phases of the Project. As described in Chapter 12 of the EIS, the primary concern during decommissioning is the potential release of radioactivity into the environment, including exposure to humans. Chapter 12 describes the environmental effects of future decommissioning, including potential radiological effects on humans and the environment. Neither Chapter 12 nor OPGs response to Greenpeaces information requests IR, however, address the sustainability effects of these potential environmental impacts. The Panel must acknowledge that radioactive waste management is an integral component of all phases of nuclear energy generation. Thus, it should be central to the Panels evaluation of the sustainability contributions of the Darlington Project. Any neglect to consider the sustainability effects of radioactive waste management is unacceptable and irresponsible to Ontarians as this aspect of nuclear energy generation poses the most significant threats to immediate and long-term well-being. It must be noted that the radioactive waste repository promoted by the NWMO comes with numerous uncertainties that could create irreversible harm to the environment and future human societies. Please see the attached report, Rock Solid? A scientific review of geological disposal of high-level radioactive waste, for further information on this subject. If the Panel allows OPG to ignore the sustainability effects of radioactive waste management it would be allowing one of the most significant potential threats to the sustainability of the Darlington Project to go unchecked. Ontarians young and old need to know that the Panel is acting in their best interests in this regard. Request: OPG should not be permitted to proceed with the project until it has fully considered the risks, costs, ethics and uncertainties of continued radioactive waste production within its sustainability assessment. Request: Greenpeace requests the Panel acknowledge in its ruling that any approval to proceed with new reactors at Darlington would invalidate assumptions of the NWMOs

21

Adaptive Phased Management Approach. This would require the NWMO to stop its current search for a repository site and new consultations begun on a socially acceptable and technically possible waste management approach.

2.6 New Fuel Waste Polluter Pays

If properly assessed, OPGs proposal to build new reactors at Darlington would be rejected due to its transfer of the risks and burden of radioactive waste management onto future generations. Greenpeace requests the Panel assure the Polluter-Pays principle will be respected before any approval is made. This is particular importance in regard to new fuel waste. OPGs proposal to build new reactors at Darlington would result in the production of radioactive wastes that are toxic and longer-lived than most radioactive wastes. The costs of managing these wastes have not been assessed. If the project is approved, this will most likely result in a public subsidy to OPG. Marvin Resnikoff s, The Hazards of Generation III Reactor Fuel Wastes: Implications for Transportation and Long-Term Management of Canadas Used Nuclear Fuel is attached for your review In summary, Canadas operating reactors are Generation II CANDU designs that use natural uranium as fuel. All new reactors under consideration the AP-1000, the EPR and the Advanced CANDU - are Generation III designs using enriched uranium as fuel Generation III with a higher burn-up rate than existing CANDU fuel. This results in more toxic and long-lived radioactive wastes. The radioactive fuel waste produced by the Generation III reactor design proposed by OPG will be 2-158 times more radioactive than the waste of current CANDU fuel waste. While the waste produced by Ontarios current CANDUs will take one million years to decay to the levels of natural uranium, the waste from these prototype reactor designs will take up to 2.6 million years. Generation III waste would require changing the management plans for interim storage, transportation and long-term storage. The NWMO has not assessed these undoubtedly higher costs and has not consulted the public on such changes. The NWMOs Advisory Council admits these wastes types were not considered when the NWMO consulted Canadians on how to manage used radioactive fuel waste. Specifically, the Advisory Council stated that: a nuclear expansion scenario would likely entail fuel enrichment and new reactor technology, with spent fuel possessing new characteristics. These could affect the performance of the disposal technology and introduce a change in the outlook on reprocessing. Such technical aspects were not considered by NWMO in its study, which focused on existing facilities using natural uranium fuel. 30 Request: No approval for the Darlington project before OPG has assessed and

30

Nuclear Waste Management Organization Advisory Council Final Report, September 22, 2005.

22

internalized the costs of managing the waste produced by the Generation III reactors before approval. This information must be made available to public scrutiny.

23

3.0 Accidents and Malfunctions

The danger of accidents, malfunctions or terrorist attacks at a nuclear power plant remains one of the defining features of nuclear power generation because such events have the potential to create catastrophic and irreversible harm to Canadian society and the environment. While the CEAA requires accidents to be included in environmental assessments, the Canadian Nuclear Safety Commissions (CNSC) approach to implementing this requirement overlooks assessing a full range of realistic accidents in environmental assessments. Typically, the CNSC deems certain levels estimated risk acceptable to Canadians. These are based on accidents that would not result in radioactive material leaving the plant boundaries. Accident risks beyond this level of risk are deemed incredible and do not merit consideration within environmental assessments according to the CNSC even if those risks would result in catastrophic environmental damage. While the nuclear industry and its regulator claim publicly that catastrophic accidents are incredible and dont require consideration, the nuclear industry considers internally that such accidents are a realistic possibility. A special piece of legislation - the Nuclear Liability Act (NLA) - was created to indemnify the industry from paying the full costs of clean up and compensation in the event of an accident. This practice of ignoring accidents or threats deemed incredible when the nuclear vendors, suppliers and operators consider them realistic is contrary to precautionary principle. This transfer of residual risks for accident and malfunctions from nuclear operators and supplier to the Canadian public is in violation of the Polluter-Pays principle. In 2009, Greenpeace commissioned nuclear risk expert Dr. Gordon Thompson to assess the governments proposed revision to the Nuclear Liability Act against modern concepts of sustainability, precaution and the Polluter- Pays principle. His report, The Nuclear Liability and Compensation Act: Is it appropriate for the 21st century? is provided along with this submission. The report concludes that transfer of responsibility from the nuclear industry to the Canadian public acts as substantial subsidy to nuclear operators. It also puts other clean energy technologies at a disadvantage. Request: The Panel request that the federal government calculate the total subsidy provided to OPG for the operation of new reactors at Darlington from the Nuclear Liability Act. Request: The Panel not approve new reactors at Darlington until the federal government alignes its nuclear liability legislation with the Polluter-Pays principal.

3.1 Malevolent Events

Post September 11th, the CNSC adapted its approach to nuclear accidents for terrorist attack. While new reactors are reportedly being required to meet additional design

24

requirements to deter terrorist attacks, certain terrorist events are dismissed as incredible. Credible terrorist events are referred to as Design Basis Threats (DBT) while incredible terrorist events are called Beyond Design Basis Threats (BDBT). What these design requirements are and how they were determined is not public. The CNSC has not consulted Canadians on what types of terrorist events should be planned for. Greenpeace has attempted repeatedly to acquire information on how the CNSC has been developing these categories, but the requests are either denied outright or heavily censored. This lack of transparency regarding a safety issues that could cause irreversible harm is of significant concern in the current review. As mentioned, this environmental assessment excludes significant terrorist events based on the CNSCs non-transparent terrorism safety standards. This is unacceptable and contrary to principles of precaution. Terrorism events at a nuclear plant have the potential to cause irreversible harm to the environment and Canadian society. A precautionary approach requires an open and transparent process with the onus being placed on OPG to prove that the risks of its project are acceptable. If this is not required, the costs of potential terrorist events will clearly be born by Canadians. If approved, the Darlington Project will be the first nuclear station built in Canada post September 11th. Given these reactors could be operating into the next century, it is important that there is an open and full debate and understanding of the terrorism risks Darlington would impose on future generations Request: The project not be approved before a full examination of realistic terrorist attacks. Request: The project not be approved pending a public and transparent consultation on the level of terrorist resistance required in new reactors.

25

4.0 Conclusion and Recommendations

As the preceding discussion shows, OPGs proposal to build new reactors at Darlington would not benefit Canadian society. Instead it will accelerate momentum towards unsustainable development in Canada and needlessly impose of risks and burdens onto future generations. Based on its failure to meet the net benefit to sustainability test, Greenpeace requests the panel reject OPGs request for approval. Request: In assessing and ruling on the adequacy of OPGs sustainability assessment, Greenpeace requests the Panel use as a minimum the precedents of analysis established by past Panel Reviews. Request: The Panel use its discretion under CEAA to ensure the justification, need and alternatives to build new reactors at Darlington is addressed in the current review. Request: The Panel should require the Ontario government to release its criteria and assessments for the cost-effectiveness of new nuclear during the current review. Request: The Panel should request the Ministry of Energy explain its process for calculating the need for baseload electricity demand and the rationale for maintaining nuclear at historic levels. Request: The Panel should deny approval of the Darlington Project in order to maintain the flexibility for the expansion of cost-effective clean energy technologies. Request: The Panel must acknowledge OPGs failure to adopt a comprehensive sustainability framework to evaluate the sustainability effects of the Darlington Project, and require it to apply the framework consistently throughout planning and decision-making. It is not adequate to finish planning and then assess the sustainability effects of the project. A comprehensive sustainability framework should have guided the project from the outset. Request: The project should not be approved before there is a public and transparent assessment of the projects costs against other energy options. Request: The Panel should consider the impact approval the project will have on the ability and flexibility of future generations to develop more sustainable energy sources. Request: OPG should not be permitted to proceed with the project until it has fully considered the risks, costs, ethics and uncertainties of continued radioactive waste production within its sustainability assessment. Request: The Panel should acknowledge in its ruling that any approval to proceed with new reactors at Darlington would invalidate assumptions of the NWMOs Adaptive Phased Management Approach. This would require the NWMO to stop its current search for repository site and begin new consultations on a socially acceptable and technically possible waste management approach. Request: No approval for the Darlington project before OPG has assessed and internalized the costs of managing the waste produced by the Generation III reactors before approval. This information must be made open to public scrutiny.

26

Request: The Panel request that the federal government calculate the total subsidy provided to OPG for the operation of new reactors at Darlington from the Nuclear Liability Act. Request: The Panel not approve new reactors at Darlington until the federal government modernized and aligned its nuclear liability legislation with the Polluter-Pays principal. Request: The project not be approved before a full examination of realistic terrorist attacks. Request: The project not be approved pending a public and transparent consultation on the level of terrorist resistance required in new reactors.

27

Appendix A: Sustainability Decision Criteria

Core Decision-Making Criteria for Sustainability (Gibson et al., 2005, pg. 116 - 118)

Socio-ecological system integrity Build human-ecological relations to establish and maintain the long-term integrity of socio biophysical systems and protect the irreplaceable life support functions upon which human as well as ecological well-being depends. Livelihood sufficiency and opportunity Ensure that everyone and every community has enough for a decent life and that everyone has opportunities to seek improvements in ways that do not compromise future generations possibilities for sufficiency and opportunity. Intra-generational equity Ensure that sufficiency and effective choices for all are pursued in ways that reduce dangerous gaps in sufficiency and opportunity (and health, security, social recognition, political influence, etc.) between the rich and the poor. Intergenerational equity Favour present options and actions that are most likely to preserve or enhance the opportunities and capabilities of future generations to live sustainably. Resource maintenance and efficiency Provide a larger base for ensuring sustainable livelihoods for all while reducing threats to the long term integrity of socio-ecological systems by reducing extractive damage, avoiding waste and cutting overall material and energy use per unit of benefit. Socio-ecological civility and democratic governance Build the capacity, motivation and habitual inclination of individuals, communities and other collective decision-making bodies to apply sustainability requirements through more open and better informed deliberations, greater attention to fostering reciprocal awareness and collective responsibility, and more integrated use of administrative, market, customary and personal decision-making practices. Precaution and adaptation Respect uncertainty, avoid even poorly understood risks of serious or irreversible damage to the foundations for sustainability, plan to learn, design for surprise, and manage for adaptation. Immediate and long term integration Apply all principles of sustainability at once, seeking mutually supportive benefits and multiple gains.

28

Appendix B: Sustainability-Based Trade-Off Rules

Sustainability-Based Trade-Off Rules (Gibson et al., 2005, pg. 139-140)

Maximum net gains Any acceptable trade-off or set of trade-offs must deliver net progress towards meeting the requirements for sustainability; it must seek mutually reinforcing, cumulative and lasting contributions and must favour achievement of the most positive feasible overall result, while avoiding significant adverse effects. Burden of argument on trade-off proponent Trade-off compromises that involve acceptance of adverse effects in sustainability-related areas are undesirable unless proven (or reasonably established) otherwise; the burden of justification falls on the proponent of the trade-off. Avoidance of significant adverse effects No trade-off that involves a significant adverse effect on any sustainability requirement area (for example, any effect that might undermine the integrity of a viable socio-ecological system) can be justified unless the alternative is acceptance of an even more significant adverse effect. Generally, then, no compromise or trade-off is acceptable if it entails further decline or risk of decline in a major area of existing concern (for example, as set out in official international, national or other sustainability strategies or accords or as identified in open public processes at the local level), or if it endangers prospects for resolving problems properly identified as global, national and/or local priorities. Similarly, no trade-off is acceptable if it deepens problems in any requirement area (integrity, equity, etc.) where further decline in the existing situation may imperil the long term viability of the whole, even if compensations of other kinds, or in other places are offered (for example, if inequities are already deep, there may be no ecological rehabilitation or efficiency compensation for introduction of significantly greater inequities). No enhancement can be permitted as an acceptable trade-off against incomplete mitigation of significant adverse effects if stronger mitigation efforts are feasible. Protection of the future No displacement of a significant adverse effect from the present to the future can be justified unless the alternative is displacement of an even more significant negative effect from the present to the future. Explicit justification All trade-offs must be accompanied by an explicit justification based on openly identified, context specific priorities as well as the sustainability decision criteria and the general tradeoff rules. Justifications will be assisted by the presence of clarifying guides (sustainability policies, priority statements, plans based on analyses of existing stresses and desirable futures, guides to the evaluation of significance, etc.) that have been developed in processes as open and participative as those expected for sustainability assessments. Open process

29

Proposed compromises and trade-offs must be addressed and justified through processes that include open and effective involvement of all stakeholders. Relevant stakeholders include those representing sustainability-relevant positions (for example, community elders speaking for future generations) as well as those directly affected. While application of specialized expertise and technical tools can be very helpful, the decisions to be made are essentially and unavoidably value-laden and a public role is crucial.

30

Rock Solid?

A GeneWatch UK consultancy report

Rock Solid?
A scientific review of geological disposal of high-level radioactive waste

Written by Helen Wallace for Greenpeace International

September 2010

Contact details Greenpeace EU Unit Jan Haverkamp Email: jan.haverkamp@greenpeace.org

GeneWatch UK 60 Lightwood Road, Buxton, Derbyshire SK17 7BB Phone: 01298 24300 Email: mail@genewatch.org Website: www.genewatch.org
Registered in England and Wales Company Number 3556885

GeneWatch
UK
1

GeneWatch UK consultancy report September 2010

GeneWatch UK consultancy report September 2010

Acknowledgements
The author is grateful to Dr Rachel Western for drawing her attention to Nuclear Waste Advisory Associates' Issues Register and a number of the other publications cited; and to Dr. Rianne Teule and the advisory group from Greenpeace International for their helpful comments on a draft of this report.

Cover
The cover photograph by Eric Shmuttenmaer is licensed under a Creative Commons AttributionShare Alike 2.0 Generic license. The world's first nuclear reactor was rebuilt at this site in Red Gate Woods near Chicago in 1943 after initial operation at the University of Chicago.

GeneWatch UK consultancy report September 2010

Contents
Executive summary .............................................................................................................7 1. Introduction ..............................................................................................................9 2. Nuclear power and radioactive waste ....................................................................11 2.1. Harmful effects of radioactive wastes .................................................................13 3. The concept of deep geological disposal. .............................................................17 3.1. Safety assessment ...........................................................................................18 3.2. National programmes for geological disposal ................................................... 19 3.3. Potential for significant radiological releases? ...................................................21 4. Literature review of post-closure issues ..............................................................23 4.1. Corrosion of canisters, wastes, and repository structures .................................23 4.1.1. Corrosion of copper................................................................................23 4.1.2. Corrosion of copper by water ............................................................... 24 4.1.3. Role of microbes .................................................................................. 24 4.1.4. Steel corrosion and hydrogen gas generation .......................................25 4.1.5. Creep ....................................................................................................25 4.1.6. Summary of corrosion issues.................................................................26 4.2. Bentonite erosion and loss of buffer capacity...................................................26 4.2.1. Effects of heat and mineral changes on bentonite ................................ 26 4.2.2. Effects of saline water ............................................................................28 4.2.3. Effects of other minerals in clay .............................................................28 4.2.4. Chemical disturbance due to corrosion..................................................28 4.2.5. Effects of gas on the clay barrier............................................................29 4.2.6. The role of microbes: gas production and biomineralisation..................30 4.2.7. Summary of bentonite erosion and loss of buffer capacity ....................31 4.3. Solubility, sorption and transport of radionuclides ...........................................31 4.3.1. Geochemistry and buffer chemistry........................................................31 4.3.2. Colloids and complexation .....................................................................32 4.3.3. The role of microbes...............................................................................32 4.3.4. Release of radioactive gas .....................................................................33 4.3.5. Summary of solubility, sorption and transport of radionuclides ..............33 4.4. Bedrock properties and hydrogeology ...............................................................33 4.4.1. Groundwater flow in the bedrock and fractures .....................................33 4.4.2. Excavation damage................................................................................35 4.4.3. Gas flow ................................................................................................36 4.4.4. Summary of bedrock properties and hydrogeology ...............................36
GeneWatch UK consultancy report September 2010

4.5. Human intrusion and human error ...................................................................36 4.6. Ice ages and glaciation.....................................................................................36 4.7. Earthquakes .....................................................................................................38 4.8. Transport of radionuclides in the biosphere .....................................................38 5. Overarching unresolved issues ...........................................................................40 5.1. Safety assessment: the evidence base, the methodology and its limitations .......................................................................40 5.1.1. Unknowns, uncertainties and model validation ......................................40 5.1.2. Potential for bias in the assessment process .........................................42 5.2. Site selection, public opinion and radioactive waste inventories......................43 5.3. Costs ................................................................................................................46 6. Conclusion .............................................................................................................48

GeneWatch UK consultancy report September 2010

Index of Boxes and Figures

List of text boxes Box 1: Box 2: Box 3: Box 4: Box 5: Box 6: Box 7: Box 8: Categories of radioactive waste....................................................................11 Nuclear reprocessing ................................................................................. 12 Radioactivity ...............................................................................................13 Health effects of ionising radiation . .............................................................. 14 Radionuclides and deep geological disposal ................................................15 Existing difficulties with geological repository programmes ..........................20 The Swedish concept...................................................................................21 The French concept .....................................................................................21

List of figures Figure 1: Decay in radiotoxicity of spent nuclear fuel ..................................................16

GeneWatch UK consultancy report September 2010

Executive Summary
Worldwide, thirteen countries are actively pursuing long-term waste management programmes for high-level radioactive wastes resulting from nuclear electricity generation, but no country has yet completed an operational geological disposal facility for such wastes. The European Commission Joint Research Centres 2009 conclusion that the technology of geological disposal has developed well enough to proceed with stepwise implementation is based largely on a description of ongoing research projects and nuclear agency reports, and references only three papers published in scientific journals. Further, the Centres report falsely claims that it is mainly due to a lack of public acceptance that repository programmes in Germany and the UK have (temporarily) foundered, rather than because of safety issues. Similarly, the statement of the Organisation for Economic Co-operation and Developments (OECDs) Nuclear Energy Agency (NEA) that geological disposal is technically feasible and that a geological disposal system provides a unique level and duration of protection for high activity, long-lived radioactive waste is based on the collective views of its Radioactive Waste Management Committee, not on an analysis of the existing scientific evidence. Based on a literature review of papers in scientific journals, the present report provides an overview of the status of research and scientific evidence regarding the long-term underground disposal of highly radioactive wastes. This review identifies a number of phenomena that could compromise the containment barriers, potentially leading to significant releases of radioactivity: l Copper or steel canisters and overpacks containing spent nuclear fuel or high-level radioactive wastes could corrode more quickly than expected. l The effects of intense heat generated by radioactive decay, and of chemical and physical disturbance due to corrosion, gas generation and biomineralisation, could impair the ability of backfill material to trap some radionuclides. l Build-up of gas pressure in the repository, as a result of the corrosion of metals and/or the degradation of organic material, could damage the barriers and force fast routes for radionuclide escape through crystalline rock fractures or clay rock pores. l Poorly understood chemical effects, such as the formation of colloids, could speed up the transport of some of the more radiotoxic elements such as plutonium. l Unidentified fractures and faults, or poor understanding of how water and gas will flow through fractures and faults, could lead to the release of radionuclides in groundwater much faster than expected. l Excavation of the repository will damage adjacent zones of rock and could thereby create fast routes for radionuclide escape. l Future generations, seeking underground resources or storage facilities, might accidentally dig a shaft into the rock around the repository or a well into contaminated groundwater above it. l Future glaciations could cause faulting of the rock, rupture of containers and penetration of surface waters or permafrost to the repository depth, leading to failure of the barriers and faster dissolution of the waste. l Earthquakes could damage containers, backfill and the rock. Although computer models of such phenomena have undoubtedly become more sophisticated, fundamental difficulties remain in predicting the relevant complex, coupled

GeneWatch UK consultancy report September 2010

processes (including the effects of heat, mechanical deformation, microbes and coupled gas and water flow through fractured crystalline rocks or clay) over the long timescales necessary. In particular, more advanced understanding and modelling of chemical reactions is essential in order to evaluate the geochemical suitability of repository designs and sites. The suitability of copper, steel and bentonite as materials for canisters, overpacks and backfill also needs to be reassessed in the light of developing understanding of corrosion mechanisms and the effects of heat and radiation. Unless and until such difficulties can be resolved, a number of scenarios exist in which a significant release of radioactivity from a deep repository could occur, with serious implications for the health and safety of future generations. In this light, the existence in a number of countries of road maps for the implementation of deep disposal, and the rejection of other options, do not automatically mean that deep disposal of highly radioactive wastes is safe. At present, the following issues remain unresolved and have implications for policy development: l the high likelihood of interpretative bias in the safety assessment process because of the lack of validation of models, the role of commercial interests and the pressure to implement existing road maps despite important gaps in knowledge. Lack of (funding for) independent scrutiny of data and assumptions can strongly influence the safety case l lack of a clearly defined inventory of radioactive wastes, as a result of uncertainty about the quantities of additional waste that will be produced in new reactors, increasing radioactivity of waste due to the use of higher burn-up fuels, and ambiguous definitions of what is considered as waste l the question of whether site selection and characterisation processes can actually identify a large enough volume of rock with sufficiently favourable characteristics to contain the expected volume of wastes likely to be generated in a given country l tension between the economic benefits offered to host communities and long-term repository safety, leading to a danger that concerns about safety and impacts on future generations may be sidelined by the prospect of economic incentives, new infrastructure or jobs. There is additional tension between endorsement of deep disposal as a potentially least bad option for existing wastes, and nuclear industry claims that deep repositories provide a safe solution to waste disposal and so help to justify the construction of new reactors l potential for significant radiological releases through a variety of mechanisms, involving the release of radioactive gas and/or water due to the failure of the near-field or far-field barriers, or both l significant challenges in demonstrating the validity and predictive value of complex computer models over long timescales l risk of significant escalation in repository costs.

GeneWatch UK consultancy report September 2010

1. Introduction
This report examines the current state of scientific evidence regarding the geological disposal of spent nuclear fuel and other high-level and long-lived radioactive wastes. The European Atomic Energy Community (Euratom), which was founded in 1957 to promote the use of nuclear power in Europe, has been financing research in the area of geological disposal of highlevel radioactive waste for more than three decades and has provided considerable support to national research and development programmes.1 Worldwide, thirteen countries are actively pursuing long-term waste management programmes for high-level radioactive wastes resulting from nuclear electricity generation, but no country has yet completed an operational geological disposal facility for such wastes.2 The 2009 Euratom-funded Vision Document of the European Implementing Geological Disposal of Radioactive Waste Technology Platform (IGD-TP) states that a growing consensus exists that deep disposal is the most appropriate solution to disposing of spent nuclear fuel, high-level waste and other long-lived radioactive wastes, and that it is time to proceed to licensing the construction and operation of deep geological repositories for radioactive waste disposal.3 This conclusion is supported by the 2009 report of the European Commissions (ECs) Joint Research Centre (JRC), which states that our scientific understanding of the processes relevant for geological disposal has developed well enough to proceed with step-wise implementation.4 The IGD-TP Vision Document has been prepared by an Interim Executive Group with members from the nuclear waste management organisations SKB (Sweden), Posiva (Finland) and Andra (France) and the German Federal Ministry of Economics and Technology (BMWi). It adopts the vision that by 2025 the first geological disposal facilities for spent nuclear fuel, high-level waste and other longlived radioactive waste will be operating safely in Europe. The Director of Energy (Euratom) for the European Commissions Directorate-General for Research states in the Foreword: These will not only be the first such facilities in Europe but also the first in the world. I am convinced that through this initiative, safe and responsible practices for the long-term management of hazardous radioactive waste can be disseminated to other Member States and even 3rd countries, thereby ensuring the greatest possible protection of all citizens and the environment both now and in the future.5 The IGD-TP states that inherent in all the successful outcomes to date in European nuclear waste management programmes are judgements that safe geological disposal of spent nuclear fuel, highlevel waste, and other long-lived radioactive waste is achievable: In this context, the future RD&D [Research, Development and Demonstration] issues to be pursued, including their associated uncertainties, are not judged to bring the feasibility of disposal into question. This statement reflects the view expressed by the Radioactive Waste Management Committee (RWMC) of the OECDs Nuclear Energy Agency (NEA)6 that geological disposal is technically feasible and that a geological disposal system provides a unique level and duration of protection for high activity, long-lived radioactive waste. However, the OECD/NEA position is merely a collective statement, based on the views of the RWMC, not an analysis of the existing scientific evidence. Similarly, the IGD-TP report relies on a road map towards radioactive waste management developed by the European Nuclear Energy Forum7, and includes no references to papers in scientific journals. The ECs JRC report is largely a description of ongoing research projects; it cites only three papers published in academic journals (one of which dates from 1999) plus lists of background reports, largely published by the NEA and International Atomic Energy Agency (IAEA), and a few conference papers. The report makes no obvious links between these summaries of research activity and its conclusion that Europe is ready to proceed to implementation of deep geological disposal.8 In a rare example of a referenced claim, the JRCs statement that corrosion of steel (and the generation of hydrogen gas by this process) will not compromise the safety of a repository is based solely on an unpublished note of a panel discussion held in Brussels in 2007. Further, the report falsely claims that repository programmes in

GeneWatch UK consultancy report September 2010

Germany and the UK have (temporarily) foundered mainly for reasons of public acceptance, rather than because of safety issues. In contrast, the present report is based on a literature review of research on deep disposal published in peer-reviewed scientific journals. It provides an overview of the status of research and scientific evidence regarding the long-term underground storage of highly radioactive wastes, and asks whether this evidence supports the view that such wastes can be disposed of safely underground. It finds that significant scientific uncertainties remain and it accordingly questions whether strong conclusions in favour of deep disposal can be drawn until all the relevant issues have been addressed.

10

GeneWatch UK consultancy report September 2010

2. Nuclear power and radioactive waste

Nuclear reactors are used to generate electricity in 31 countries in the world.9 Currently, there are 438 operational nuclear power plants in the world, with a total net installed capacity of 372.038 GW(e).10 The IAEA lists 61 nuclear power reactors as currently under construction, mainly in China, Russia, South Korea and India.11 In Europe, new reactors are being built at Olkiluoto in Finland, Flamanville in France and Mochovce in the Slovak Republic. Globally, China is expected to be the fourth largest generator of nuclear power by 2025, behind the USA, France and Japan.12 Nuclear electricity generation creates large quantities of radioactive wastes, not only in nuclear power plants themselves, but at all stages of the nuclear chain, from uranium mining to decommissioning of nuclear facilities. The most highly radioactive wastes are those which are produced in the core of the reactor. The focus of this report is on spent nuclear fuel: this is nuclear fuel that has been involved in the nuclear chain reaction at the heart of the reactor (see Box 1). Some countries intend to dispose of spent nuclear fuel directly, but in other countries it is first reprocessed (Box 2). Reprocessing changes the characteristics of the wastes that will ultimately be sent to a repository. The amount of radioactive waste produced in a reactor depends on the reactor type. On the basis of data from 1992, the IAEA estimates that one years operation of a Light Water Reactor (LWR) producing 1GW of power typically results in spent fuel assemblies containing a total of 30 to 50 metric tonnes of heavy metal , with an initial activity of around 5 to 8.3 million TBq of radioactivity.13 According to the IAEA, current reprocessing procedures would separate about 15m3 of vitrified highlevel radioactive waste from this quantity of spent fuel. These figures are indicative only and have changed significantly with time. More modern reactors using higher burn-up fuel will produce smaller quantities of spent fuel but with higher levels of radioactivity per fuel rod. These changes can have significant implications for the safety case for a repository.14 Box 1: Categories of radioactive waste Naturally Occurring Radioactive Material (NORM) includes radioactive wastes created by mining and milling of naturally occurring uranium ores in order to produce fuel for nuclear reactors. Low-Level Waste (LLW) makes up the bulk of the volume of waste produced in the nuclear fuel chain. It consists of materials such as paper, rags, tools, clothing and filters, which may contain small amounts of mostly short-lived radioactivity. Intermediate-Level Waste (ILW) contains higher levels of radioactivity and normally requires shielding. It includes resins, chemical sludges, metal fuel cladding, and contaminated materials from the decommissioning of reactors or from nuclear reprocessing. Short-lived ILW is typically disposed of in shallow land burial, but long-lived ILW is destined for geological disposal. High-Level Waste (HLW) and Spent Nuclear Fuel both contain fission products (radioactive elements created when atoms are split in the nuclear chain reaction) and transuranic elements (see Box 5) generated in the reactor core. These are highly radioactive and generate heat due to radioactive decay. In countries where spent nuclear fuel is reprocessed, liquid high-level waste is separated from other radioactive waste streams (see Box 2) and is vitrified (turned into glass blocks) before disposal. Depending on the waste disposal concept the heat-generating spent fuel and high-level waste require a cooling period of up to several decades prior to ultimate disposal.

GeneWatch UK consultancy report September 2010

11

Box 2: Nuclear reprocessing Nuclear reprocessing involves treating spent nuclear fuel by means of a chemical process (usually by dissolving it in nitric acid15) after it has been removed from the reactor and stored for several years. The spent fuel is separated into plutonium, uranium, and high-level and intermediate-level wastes, and radioactive waste streams are also discharged as liquids into the sea or other water courses and as gases to air. Liquid high-level wastes are stored in tanks, which require constant cooling, and are later vitrified (turned into glass blocks). The volume of high-level waste contained in these glass blocks is smaller than the volume of the original spent nuclear fuel.16 However, reprocessing increases the total volume of radioactive material, and creates a large volume of long-lived intermediate-level wastes, which are usually also considered to require deep underground disposal.17 Four countries (France, India, Russia and the UK) currently have reprocessing plants which take spent nuclear fuel from non-military reactors on a commercial scale, while Japan and China have pilot plants and aim to reprocess commercially in the future.18 Reprocessing facilities were originally developed to extract plutonium from spent nuclear fuel in order to make nuclear weapons. The separated plutonium from commercial reprocessing is now mainly added to existing stockpiles, although small quantities are used in the production of mixed-oxide (MOX) nuclear fuel. Separated uranium was originally intended to be reused as nuclear fuel, but at present this rarely happens, probably as a result of its poor quality compared with fresh uranium (due to contamination with unwanted uranium isotopes). France, the UK, Russia and China are nuclear weapons states with a significant legacy of wastes from nuclear reprocessing for weapons production, as well as from their ongoing civil nuclear programmes. Reprocessing in these countries continues to generate large amounts of liquid high-level wastes.19 Japan has sent large quantities of spent fuel to the UK and France but is now planning to reprocess its own fuel. The USA reprocessed spent nuclear fuel in the past, although not on a commercial scale. It ceased the practice in 1997 due to concerns about the nuclear proliferation risks associated with separated plutonium, along with a combination of severe technical, economic and safety problems.20 In Europe, the Sellafield site in England and La Hague in France are the main reprocessing plants. Significant radioactive discharges to sea and air have been made from both sites over the past 60 years.21 The Strategy on Radioactive Substances adopted by the Oslo and Paris Convention (OSPAR) in 1998, which covers discharges to sea in the North-East Atlantic area, requires that by the year 2020 the discharges, emissions and losses of radioactive substances be reduced to levels where the additional concentrations in the marine environment above historic levels resulting from such discharges, emissions and losses are close to zero.22 With the exception of UK and France, OSPAR member states interpreted the Strategy to mean that reprocessing should cease and be replaced with storage of spent nuclear fuel (the nonreprocessing option).23 The UK and France have disputed the implications for reprocessing, but the UK has accepted that operational discharges from Sellafield from reprocessing should have fallen to zero by 2020.24 A number of other European countries have sent their spent nuclear fuel for reprocessing abroad. However, this practice has largely ceased due to concerns about costs, the harm to human health and the environment caused by the radioactive discharges, and the nuclear proliferation risk associated with separated plutonium.25 Vitrified high-level wastes and plutonium from past reprocessing are intended to be returned from the UK and France to the countries of origin. However, the return of intermediate-level wastes will be limited. In total, over 10,000 metric tonnes of spent nuclear fuel are being produced globally each year. The global inventory of spent nuclear fuel is expected to more than double to over 445,000 metric tonnes by 2020,26 with the highest percentage increases in developing countries. Yet, to date, no country has achieved an effective solution for the long-term management of spent nuclear fuel.27

12

GeneWatch UK consultancy report September 2010

According to 2004 statistics cited by the IGD-TP,28 the annual production in the EU Member States of radioactive waste and spent fuel considered suitable for deep geological disposal is 5,100m3 of longlived low- and intermediate-level waste (excluding that produced in Germany which is to be disposed of in the Konrad mine), 280m3 of high-level waste and 3,600 tonnes heavy metal of spent fuel. At the end of 2004, an estimated 220,000m3 of long-lived low- and intermediate level waste, 7,000m3 of high-level radioactive waste and 38,000 tonnes of heavy metal of spent fuel were stored in Europe. However, there are considerable uncertainties in these figures. In reprocessing countries such as the UK and France, spent nuclear fuel and reprocessed plutonium and uranium are not currently classified as nuclear waste, on the grounds that spent fuel is a recyclable material and that reprocessed uranium and plutonium might be used to make fresh fuel. This situation results in large volumes of radioactive material that may ultimately be buried in a repository not being included in the official inventories of radioactive waste in these countries.29 Plutonium (which is a nuclear weapons material) has no currently licensed disposal route and in practice most separated uranium is not reused. Spent nuclear fuel requires interim storage, to allow time for cooling after it is first removed from a reactor during refuelling. Wet storage involves keeping the spent fuel rods in racks under water in cooling ponds. Dry storage requires the use of casks designed to cool the waste by air convection and to protect it from fires and mechanical impacts. Interim storage in Europe is normally at the reactor sites or at centralised interim storage facilities; concerns about the safety of such facilities are beyond the scope of this report.30 In countries with reprocessing plants, spent nuclear fuel, liquid and vitrified high-level wastes and other radioactive wastes are stored at the reprocessing plant before and after reprocessing. Even if repositories are established by the projected dates in Sweden and Finland (the only countries which have so far selected sites), the amount of waste generated annually will account for much of the quantity scheduled to be transferred annually to the repository: hence it will take decades to reduce the amount of on-site radioactive waste significantly even once repositories are constructed.31 The focus of this report is on deep geological disposal of spent nuclear fuel and high-level nuclear wastes from reprocessing, i.e. heat-generating wastes. Long-lived intermediate-level wastes from reprocessing are also considered, but in less detail.

2.1. Harmful effects of radioactive wastes

Nuclear waste generates concerns because the radiation it emits (known as ionising radiation) can cause cancer and other serious illnesses in humans, and harm other living organisms (see Boxes 3 and 4). High-level nuclear waste is so radioactive that exposure to it is deadly: high doses of radiation cause skin burns, radiation sickness and death. Lower doses of radiation damage human cells in a way that increases the risk of diseases such as cancer; the higher the dose the greater the risk. If radioactive wastes leak from an underground repository they will expose people to low levels of radiation which can harm health and the environment; the safety assessment for a repository is required to take account of this. Box 3: Radioactivity The basic constituents of radioactive wastes are called radionuclides. These are atoms which are unstable and change to other more stable forms in a process known as radioactive decay, until a stable form is reached. The unit of radioactivity is the becquerel (Bq), defined as one decay per second. The half-life is a measure of how quickly a particular radionuclide decays: it is the time taken for the radioactivity to decay to half of its initial value. Different radionuclides have different half-lives, varying from fractions of a second to millions of years.32 After the decay of a radionuclide atom, the remaining nucleus can be either stable (i.e. nonradioactive) or unstable. If it is unstable, it will decay again: for some radionuclides long chains of decays result as one atom changes to another and then another, emitting radiation at each step. When a radionuclide decays it can emit alpha, beta or gamma radiation. Alpha radiation consists of two protons and two neutrons bonded together in a particle that is

GeneWatch UK consultancy report September 2010

13

identical to the nucleus of a helium atom. It can be emitted when a heavy radionuclide decays. Alpha particles are easily blocked (for example by a sheet of paper), but can be very dangerous if they are emitted inside the human body (for example, from a radionuclide breathed into the lungs, or ingested by eating or drinking contaminated food or water). Beta radiation consists of high-energy electrons (or positrons). It is more penetrating than alpha radiation and can penetrate living matter to some extent. However, it is less damaging, so the same amount of exposure does less damage than exposure to alpha radiation. Gamma radiation consists of electromagnetic radiation of very high energy. It is often produced at the same time as alpha or beta particles, or at the end of a long chain of decays. Gamma rays act like powerful X-rays which can pass through the human body, necessitating protection by thick shielding (for example lead or concrete). The harmfulness of radiation varies with the kind of radiation and its energy. Box 4: Health effects of ionising radiation The health effects of ionising radiation are not fully understood.33 The estimates of harm are based mainly on the ongoing study of survivors of the Hiroshima and Nagasaki bombings in 1945, supplemented by some more recent studies (e.g. of the effects of medical exposures to radiation and the Chernobyl accident). People can be exposed to radiation either externally, when radionuclides decaying outside the body expose it to ionising radiation, or internally if radionuclides are breathed in or swallowed (for example, by eating radioactively contaminated food). Some radionuclides bioaccumulate: i.e. they build up in the food chain, reaching higher concentrations in fish or seafood than in the surrounding water, and thereby posing a risk of increased radiation dose to anyone eating the contaminated food. Radiation can cause genetic damage to cells. Sometimes this damage can be repaired by mechanisms within the cell, but sometimes it can lead to the out-of-control growth of cancer cells. Damage to eggs or sperm can be passed on to future generations. Radiotoxicity is a measure of how harmful a radionuclide is to human health when inhaled or ingested: it depends on the type and energy of the radiation emitted and the radionuclides biochemical behaviour in the human body (for example, whether it is excreted quickly or builds up in bones or organs). The harm that is done depends on the dose of radiation received. But calculating this dose is not straightforward. The absorbed dose (measured in grays or Gy) is defined as the average amount of energy (in joules or J) that is deposited per unit mass (in kilograms or kg) of tissue from an exposure to radiation. The effective dose (measured in sieverts or Sv) is calculated by weighting for how harmful the type of radiation is thought to be (its estimated Relative Biological Effectiveness, RBE) and the relative sensitivity to radiation of the organs in the body which are expected to be exposed (e.g. lungs, liver, etc.). A sievert is the dose of a given type of radiation in grays that is expected to cause equivalent damage in humans to 1 Gy of X-rays or gamma radiation: but this is not known exactly. The International Commission on Radiological Protection (ICRP) is an advisory body which sets international standards on the calculation of doses and radiological protection.34 High-level radioactive wastes are so radioactive that the decay process generates significant amounts of heat. They contain a wide variety of radionuclides, each with different physical and chemical properties. Each radionuclide decays differently and has a different half-life. The physics of radioactive decay is well understood, but the inventory of radionuclides in the wastes is not well known. In addition, the chemistry of how wastes will behave in a repository is very complicated, because each element can take different forms and form a variety of compounds: some of these chemicals may dissolve easily and leak out of the repository in groundwater, while others may attach to the backfill or the surrounding rock and thus be contained more easily. Some can also

14

GeneWatch UK consultancy report September 2010

bioaccumulate in the food chain once they reach the living environment (known as the biosphere), and each one may have different health effects on humans exposed to it. A few of these radionuclides and their relevant properties are described in Box 5. Box 5: Radionuclides and deep geological disposal A chemical element is a pure chemical substance containing one type of atom. Each element has a different number of protons in its nucleus known as its atomic number. Isotopes are atoms of the same element but having different numbers of neutrons. Unstable isotopes are radioactive. Actinides. The actinides are a series of elements with atomic numbers from 90 to 103 (thorium to lawrencium, including uranium and plutonium). They are all radioactive and have a number of different radioactive isotopes. Only thorium and uranium occur in significant quantities in nature. Elements that are heavier than uranium are known as transuranic. Many actinide isotopes have long half-lives (tens of thousands of years) and are also highly radiotoxic. They exist in large quantities in spent nuclear fuel; successful containment of actinides is therefore very important in the safety case for a geological repository. Mobile radionuclides. Some radionuclides are expected to escape more easily from deep repositories in significant quantities because they are highly mobile in groundwater and have long half-lives, meaning that they are likely to reach the biosphere before they have decayed and so pose a risk to living organisms. These are mainly negatively charged (anionic) species (forms in which chemicals exist), which are not expected to be significantly retarded in the backfill or the rock. The main radionuclides of concern are iodine-129 (129I, half-life 15.7 million years), chlorine-36 (36Cl, half-life 300,000 years), selenium-79 (79Se, half-life 295,000 years) and technetium-99 (99Tc, half-life 212,000 years).35 These radionuclides are less radiotoxic than the actinides, but occur in large quantities in high-level radioactive wastes. Radioactive iodine that is ingested by humans tends to concentrate in the thyroid gland, where it can cause thyroid cancer and other problems. Technetium-99 bioaccumulates in the food chain, particularly in shellfish such as lobster.36 Selenium is an essential micronutrient for many organisms and selenium-79 can also bioaccumulate in the food chain. Carbon-14 (14C) has a half-life of 5,715 years and undergoes beta-decay into nitrogen-14. It is relevant to radioactive waste disposal because it is the main radionuclide that might escape from a repository as gas, in the form of carbon dioxide (CO2) or methane (CH4). In nuclear wastes, carbon-14 exists mainly in irradiated metals (especially steels). Smaller quantities in irradiated uranium can also impact on safety if the corrosion rate is high. There are particular problems with carbon-14 in the UK inventory of intermediate-level waste from nuclear reprocessing.37 Figure 1 shows how the radiotoxicity of spent fuel decays with time, on the basis of published calculations for spent fuel from an LWR with a burn-up of 33 GWd per tonne of heavy metal, initial enrichment 3.2% of uranium-235, and five years cooling.38 The radionuclide content and hence the decay curve will differ for higher burn-up spent fuels and those from different reactor types.

GeneWatch UK consultancy report September 2010

15

Figure 1: Decay in radiotoxicity of spent nuclear fuel Adapted from Bombini et al. (2009).39 1 total radiotoxicity of spent nuclear fuel; 2 plutonium and decay products; 3 minor actinides and decay products; 4 fission products; 5 uranium and decay products. 6 (for comparison) radiotoxicity of uranium ore. Units are Sieverts per million metric tonnes (Sv/Mt).

16

GeneWatch UK consultancy report September 2010

3. The concept of deep geological disposal

Research on nuclear waste disposal began in the1950s but a concerted attempt to solve the problem did not begin until the late 1970s. In 1976 the influential Flowers Report, published by the UK Royal Commission on Environmental Pollution, concluded that There should be no commitment to a large programme of nuclear fission power until it has been demonstrated beyond reasonable doubt that a method exists to ensure the safe containment of long-lived radioactive waste for the indefinite future.40 In April 1977, the Swedish Parliament passed the groundbreaking Nuclear Stipulation Act (Villkorslagen) that reinforced this standpoint by requiring the operators of nuclear power plants to have proven how and where a completely safe final storage facility could be constructed for spent nuclear fuel or reprocessed highlevel waste before operating permission was granted. In the USA, the Interagency Review Group on Nuclear Waste Management called for the development of geological repositories for high-level nuclear waste disposal in 1979.41 Since the adoption of these policies in the late 1970s, the focus of high-level nuclear waste disposal has been on burying wastes underground. Other options such as firing the waste into space in rockets, burying it under the Antarctic ice sheet or dumping at sea have been progressively ruled out as unfeasible and/or unsafe. As a result deep geological disposal has dominated research priorities for over 30 years.42 The option of deep geological disposal would involve excavating a repository in rock, hundreds of metres underground. The radioactive waste would then be put in containers which would in turn be placed in deposition holes in tunnels in the rock. Tunnels would be backfilled to keep the containers in place and to slow the release of radionuclides from the waste once the containers had corroded. The site is supposed to be chosen so that the flow of water through the waste and back to the surface would be slow enough for the radioactivity to decrease significantly before the living environment above the repository could become contaminated. The release of gas from corroding canisters and other structures, and radioactive gas from the waste itself, also needs to be considered, as does the risk of future earthquakes or glaciation affecting the repository. The geology of the chosen site and the engineered barriers around the waste are intended to be passively safe (i.e. not to require human intervention) after the closure of drifts and shafts. However, some designs would also allow retrieval of wastes should future generations decide to undertake this. The geological disposal concept involves multiple barriers in an attempt to ensure the long-term protection of the living environment. The key stages for implementation of geological disposal are: l establishment of the waste inventory l development of concepts and technologies l site selection and characterisation l design of the deep geological repository l safety demonstration based on scientific knowledge and demonstration of technology l licensing l construction and manufacturing l waste emplacement l backfilling and sealing l final closure. Siting a repository may take several decades and construction is expected to take another decade. Final closure is expected to be at least several decades more after the start of the operational phase. As well as the repository itself, encapsulation facilities would also be needed: here spent fuel or the vitrified waste from reprocessing would be placed in canisters or overpacks. Long-lived intermediatelevel waste is often encapsulated in concrete or bitumen and may be placed in steel barrels.

GeneWatch UK consultancy report September 2010

17

A transportation system would also be necessary to transport the highly radioactive wastes from the interim storage facilities to the encapsulation plant and on to the geological disposal facility.

3.1. Safety assessment

Before a proposed repository can be licensed for use, a safety assessment must be produced and approved by the relevant government regulators. The IAEA manages the Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste Management.43 It has published guidance documents on the siting of geological repositories44 and safety standards for their operation.45 The NEA has also developed guidelines for the post-closure safety case.46 The ICRP bases its recommendations on three fundamental principles justification of exposures, dose optimisation, and dose limits (which are not to be exceeded).47 The principle of justification requires that nuclear activities must be justified in the sense of doing more good than harm. The ICRP states that no practice involving exposures to radiation should be adopted unless it produces sufficient benefit to the exposed individual or to society to offset the detriment it causes. This principle also requires the collective dose (the average dose for a group exposed multiplied by the number of people in the group) to be weighed against the benefits of the practice (such as the generation of electricity). The collective dose gives an indication of the number of cancers and other adverse health effects that can be expected. Such cost-benefit analyses are inevitably subjective but, importantly, the requirement to consider the collective dose is intended to prevent the safety case being overly dependent on dilution and dispersion of radionuclides in the environment.48 There is no known safe dose of radiation, and a high collective dose can arise if a large number of people are collectively exposed to very low individual doses over time. The ICRP has set a dose limit of 1 milli-Sievert (mSv) per year for exposure situations where individuals received planned exposures from activities which are of no direct benefit to them (unplanned activities such as major nuclear accidents might sometimes exceed such limits).49 A lower dose constraint is set as an upper bound on the predicted dose that is allowed from planned future exposures such as that caused by waste leaking from an underground repository. For prolonged exposure to gradually accumulating long-lived radionuclides from a deep underground repository, the ICRP dose constraint is 0.1mSv per year (equating to a risk of death of one in a million per year).50 However, the ICRP is an advisory body and in practice different countries have taken different approaches to regulatory requirements. A 2007 review by the NEA found that dose constraints set by individual countries span a range of 0.10.3mSv per year, while risk constraints are set at one in 100,000 or one in a million per year.51 A major issue in the case of deep underground repositories is whether these limits can be met in practice, due to the difficulties of predicting the radiation exposures that will actually be experienced by future generations. Nevertheless the regulatory system is intended to reflect key principles, including the concept of inter-generational equity or the idea that there should be no undue burden on future generations as a result of producing nuclear electricity today.52 The dose constraint is always lower than the dose limit and represents a basic level of protection which sets an upper bound for a process known as dose optimisation. Optimisation is an iterative process that involves the identification of possible protection options and the selection and implementation of the best option under the prevailing circumstances. It is intended to ensure that doses are as low as reasonably achievable (ALARA), economic and social factors being taken into account.53 The most recent (2007) ICRP recommendations also expand the concept of radiological protection to protection of the environment, including the maintenance of biological diversity, the conservation of species and the health and status of natural habitats. However, the environmental impact of radionuclides has only recently begun to be considered and the main focus of safety assessments remains human exposures. Safety assessment requires the post-closure behaviour of the radioactive wastes in a repository to be predicted hundreds of thousands to millions of years into the future. The limitations of the computer models that are used to make these predictions and the difficulties of validating them i.e. of

18

GeneWatch UK consultancy report September 2010

confirming that they will give sufficiently reliable predictions over such long timescales are among the key issues for safety assessment.54,55 Computer models of the evolution of the engineered barriers and interaction between the multiple barriers have to be developed, to include all the complex thermal, mechanical, hydrogeological, chemical, and microbiological processes which will affect migration of the wastes as they are released from the containers.56,57 The release of radioactive water and gas through the rock also calls for complex computer models, taking account of chemical interactions with the rock, the transport of water and gas through cracks and fissures, and any potential fast routes for escape, such as via the backfilled tunnels and shafts of the repository or fractures and faults in the host rock.58,59,60 The effects of earthquakes (which can affect underground water flow or damage packaging, even when they are not major events), long-term climate change (which can alter sea level and underground hydrology) and the behaviour of future generations (who might for example dig a well above the repository at some point in the future) also need to be considered. Because many of the complex processes involved are poorly understood and many model assumptions impossible to verify, the question of whether computer predictions are sufficiently reliable to underpin a repository safety case is a matter of considerable debate. The chemical conditions inside the repository are very important because they will influence which chemical reactions can occur and at what rate. This in turn will affect the corrosion rates of the waste containers, the properties of the bentonite clay expected to be used as backfill, and how quickly the wastes dissolve and migrate through the backfill and rock. For example, corrosion of metals involves both oxidation and reduction.61 Relevant chemical properties in a repository will include how acidic or alkaline the groundwater becomes (its pH) during the lifetime of the repository and its redox (reduction-oxidation) potential. Solutions with a pH less than 7 are said to be acidic and solutions with a pH greater than 7 are said to be basic or alkaline. Reduction potential (Eh) is a measure of the tendency of a chemical species to acquire electrons and thereby be reduced. Both Eh and pH influence the type of chemical reactions that can occur. Eh-pH diagrams are commonly used in geology for assessment of the stability fields of different minerals and dissolved substances: they show under which conditions a mineral or chemical species is the most stable form.62 Understanding and predicting the rate of the complex chemical reactions which will occur underground is central to a robust repository safety case. However, many gaps in knowledge and uncertainties remain. In order to meet the safety requirements, predicted doses to a reference person living near the proposed repository are supposed to be calculated many generations into the future. The habits used as a basis for this calculation (e.g. consumption of foodstuffs and use of local resources) should be typical of the small number of individuals expected to be most highly exposed.63 There are obviously considerable uncertainties in defining these habits, as well as disagreements regarding the impacts of radiation on vulnerable groups such as children, babies and developing embryos.64 Because of the role that the geological surroundings are expected to play in containment of the wastes, site selection is a key part of the safety case.65 The inventory of wastes is also important because it determines the quantities of different radionuclides, the chemical reactions that will take place, the volume of rock likely to be needed and the amount of heat that will be generated by radioactive decay. Because high burn-up fuel contains increased amounts of long-lived hazardous radionuclides in the spent fuel, such as americium, curium and plutonium, for the same amount of energy produced, and generates significantly more heat, the proposed use of high burn-up fuel in new nuclear reactors could have significant implications for repository safety cases.66

3.2. National programmes for geological disposal

Repository programmes are at different stages in various countries, and involve several different approaches to containing highly radioactive wastes.67 To date, major problems with repository programmes have been encountered in several countries, for example the UK, Germany and the USA (Box 6).

GeneWatch UK consultancy report September 2010

19

Box 6: Existing difficulties with geological repository programmes United Kingdom: A planning inquiry into a proposal to build the first stage (known as a Rock Characterisation Facility) of a deep repository for long-lived intermediate-level wastes led to the rejection of the plans in 1997. The planning inspector concluded that the site near Sellafield was unsuitable for a repository for safety reasons.68,69,70 Various generic problems with deep disposal and the site selection process were also highlighted in the rejection of the plans. Although the proposal did not explicitly include high-level wastes, it was expected that the repository, if approved, would have been expanded to include these in the future. Following the advice of the House of Lords Science and Technology Committee71 (which reportedly advocated returning to the site72), the UK Government subsequently changed planning law so that in future the scientific evidence concerning safety at a site would not be cross-examined.73 It is now seeking volunteer communities for a repository close to the original Sellafield site, with the intention of starting a new programme to develop a repository which would accommodate high-level wastes and spent nuclear fuel as well as intermediate-level wastes. A recent change in government may mean that the planning process is revised again. Germany: In Germany, the deep disposal concept has been based on the use of rock salt as the host geological formation. From 1967 until 1978 the Asse II salt mine was used for disposal of low- and intermediate-level radioactive wastes, including some long-lived wastes. In January 2010, the German authorities decided that all the waste from Asse II needs to be retrieved and repackaged due to safety problems, including the leaking of saline water into the chambers.74 It has not yet been decided where the waste will be stored. The costs of this expensive operation will fall largely on German taxpayers. Repository shafts were constructed in 198590 in another salt dome site at Gorleben, selected for disposal of spent nuclear fuel as well as high-level waste from overseas reprocessing.75 However, in 2000 a moratorium was placed on activities at Gorleben as a result of continuing concerns about the suitability of rock salt for geological disposal. This moratorium was lifted in March 2010 to examine further whether Gorleben would be a suitable site for the final storage of spent nuclear fuel.76 The target date for commencing operation of a repository for spent nuclear fuel and high-level waste in Germany is still 2035, but this does not appear to be in any way realistic. United States: Yucca Mountain, Nevada was identified in 1987 as the sole US site to be investigated for a high-level waste repository.77 Plans at Yucca Mountain differed from those in other countries in that the waste was supposed to be placed above the water table, where it would not be in contact with the groundwater that flows through most rocks. However, a major concern was its siting in a geologically active area where there has been significant volcanic activity and faulting. The programme was halted in 2010 after the Obama administration announced that a new plan would be developed.78 Currently active programmes are mainly limited to two different approaches: the first developed by the Swedish Nuclear Fuel and Waste Management Company (SKB), and the second largely by the French nuclear waste management company ANDRA. The Swedish approach involves the disposal of spent nuclear fuel in copper canisters in crystalline rock (Box 7). The French approach involves the disposal of vitrified high-level waste in steel overpacks in clay rock formations (Box 8). Finland and Sweden plan to start operating deep geological repositories for direct disposal of spent nuclear fuel in 2020 and 2023 respectively, following the Swedish deep repository concept. Canada and South Korea intend to follow a similar approach, as does the UK for disposal of unreprocessed spent nuclear fuel from new nuclear reactors. France plans to start operating a deep geological repository for vitrified high-level waste from reprocessing in 2025. Belgium and Switzerland are also investigating a similar approach using clay host rocks.

20

GeneWatch UK consultancy report September 2010

Box 7: The Swedish concept79 In the Swedish concept for a deep geological repository, spent nuclear fuel will be placed in cast iron frames surrounded by 5cm thick copper canisters. The canisters will be deposited in granite bedrock at a depth of 500m and surrounded by highly compacted bentonite clay. Once the repository is closed, groundwater will come into contact with the canisters containing the wastes. The copper canisters are expected to corrode very slowly in the absence of oxygen: the target lifetime for containment of radioactive waste in the canisters is 100,000 years. The bentonite and surrounding crystalline bedrock are water-conducting, so the only absolute barrier to radionuclide migration will be the copper canisters, for as long as they remain intact. Once the canisters have corroded, radionuclides are expected to leak into the surrounding water. The bentonite clay is intended to act as a chemical buffer, slowing the movement of some radionuclides, particularly the highly radiotoxic actinides (see Box 5). It also gives the canisters mechanical support, as it swells in water. The bentonite clay and bedrock are expected to slow the movement of radionuclides to the biosphere. However, absolute containment until the waste has decayed is not expected and some of it will migrate to the surface in groundwater or as gas. Once radionuclides are close to the surface, the safety case relies on their dilution and dispersion in the aquifer above the repository and in the biosphere to limit the doses that humans receive through drinking or eating contaminated water or food. Box 8: The French concept80 The French concept for deep disposal differs from the Swedish one in two main respects. Firstly, the rock type will be clay, not crystalline; secondly, vitrified high-level wastes will be placed in steel overpacks rather than copper canisters. Steel is expected to corrode more rapidly than copper, so the safety performance of the repository will be more reliant on the surrounding bentonite and clay rock. Russia has been investigating the feasibility of salt, granite, clay and basalt as possible host rocks for geological repositories, but has no projected date for completion of a repository. China is investigating five potential repository sites, including a proposed underground research laboratory site in the Gobi desert, but is not expected to have an operational disposal facility until 2040 at the earliest.81 In Finland a disposal site has already been selected at Olkiluoto. In Sweden a site has been selected at Forsmark, on the east coast. In France the zone for disposal the village of Bure in Lorraine has been selected and the final site is to be specified by 2013. The Swedish, Finnish and French proposals are therefore the main focus of the remainder of this report.

3.3. Potential for significant radiological releases?

A number of low- and intermediate-level radioactive waste disposal sites have operated over the last 50 years. However, many of these supposedly final disposal sites have already caused unexpected environmental contamination, highlighting how difficult it is to predict what will happen to buried wastes, even over short timescales. Examples are the Dounreay nuclear waste shaft in Scotland, which exploded in 1977,82 the Centre de Stockage de la Manche storage site in France, where water supplies in the aquifer have become contaminated,83 and the Asse II salt mine in Germany84 (see Box 6). Moreover, the disposal of high-level wastes raises unprecedented challenges because of the very long half-lives and radiotoxicity of these wastes. Enthusiasts for deep geological disposal argue that there are examples (known as natural analogues) which demonstrate that geological formations are capable of isolating highly volatile and flammable substances such as oil and gas underground for hundreds of millions of years.85 Concentrated natural uranium deposits have been largely confined for millions of years at sites such

GeneWatch UK consultancy report September 2010

21

as Cigar Lake in Canada, and there is even an example of a natural underground nuclear reactor containing uranium and fission products in Oklo, Gabon.86 However, the emplacement of high-level waste in an underground repository would entail a major perturbation of the geological system, involving:87 (i) a large number of tunnels covering an area of several square kilometres (ii) the release of significant amounts of heat, initially of the order of tens of thousands of kilowatts per square kilometre (iii) intense radiation and significant quantifies of highly toxic radionuclides, each with its own complex chemistry. Nuclear Waste Advisory Associates, a UK-based consultancy, has listed over a hundred scientific and technical issues that remain to be resolved in relation to producing a robust safety case for the deep disposal of radioactive wastes.88 Significant releases of radioactivity from an underground repository could occur if the near-field or far-field barriers were breached in ways that allowed radioactive groundwater or gas to escape faster than expected. The current state of knowledge about these issues is considered in the literature review that follows.

22

GeneWatch UK consultancy report September 2010

4. Literature review of post-closure issues

4.1. Corrosion of canisters, wastes, and repository structures
Copper or steel canisters or overpacks will be used to contain the spent nuclear fuel or high-level waste when it is placed in the repository. As groundwater from the surrounding rock flows into the repository, these canisters or overpacks will begin to corrode and eventually their radioactive contents will be released into the groundwater. The focus of this report is on the Swedish and French designs in which the waste is below the water table and the backfill is expected to become saturated with water soon after the repository is closed. In the USA, the Yucca Mountain proposed site (now abandoned) was above the water table. However, rainwater was still expected to enter the repository and to cause corrosion.89,90 The Swedish safety case assumes that copper canisters 5cm thick will contain the wastes for 100,000 years, but there are serious question marks about the assumptions that have been made regarding the low corrosion rate of copper91 (discussed further below). Steel is expected to corrode much more rapidly than copper: with a typical design life of 1,000 years. Actual life may be significantly longer than design life and the predicted lifetime of the steel overpacks is of the order of 10,000 years.92 However, the safety case for the French approach remains much more dependent than its Swedish counterpart on the performance of the clay backfill and bedrock, due to the faster corrosion of steel. An inner material is also needed between the insides of the canisters and the spent fuel assemblies they contain, to prevent the gap filling with water and a criticality (nuclear chain reaction) from occurring. A cast iron insert will be used in the Swedish copper canisters; other materials (for example, glass or depleted uranium), each of which has different advantages and disadvantages, are being considered as possible alternative inserts in steel canisters in other countries.93 Corrosion of steel, and perhaps of copper, will release hydrogen gas into the repository. Corrosion of some wastes can also release carbon dioxide or methane, which may be radioactive (containing carbon-14). The build-up of gas pressure could be harmful, since a sudden release of pressure (or explosion) could damage the repository. Alternatively, slow release of gas could open up fractures in the backfill or rock, and speed up the release of some radionuclides from the repository.94 These issues are discussed in more detail below.

4.1.1. Corrosion of copper The Swedish concept for deep disposal uses copper canisters because the corrosion rates are expected to be extremely slow. The corrosion behaviour of copper canisters is expected to change with time as the conditions within the repository evolve from warm and oxidising initially to cool and anoxic in the long term.95 Copper corrodes in air due to the presence of oxygen, forming copper oxides. There will be air in the repository during the decades when it is operational (i.e. while the waste is being emplaced). However, after the repository is closed, safety cases assume that all the oxygen will be rapidly used up by the metabolism of oxygen-using microbes (aerobes) and other chemical reactions, so that the copper canisters can no longer corrode in this way. Nevertheless, there remains concern about the rate of corrosion of copper during the first 100 years or so, when oxygen and heat are both likely to be present. In Canada, coating copper canisters with a polymer is being considered as an option to provide protection during this early emplacement phase.96 After all the oxygen has been consumed, it is assumed that sulphide will be the primary corrosive agent for copper canisters in a repository, and corrosion will proceed with the formation of copper sulphide and hydrogen gas, although corrosion rates are predicted to be very slow.97,98,99

GeneWatch UK consultancy report September 2010

23

In general, it is presumed that the canisters will corrode in a uniform way, rather than through localised corrosion such as pitting.100,101 However, in reality, pitting may also occur and one of the scenarios that may result in early release of radioactivity is water flow into a defective canister.102 4.1.2. Corrosion of copper by water It has long been assumed that water alone does not corrode copper in an oxygen-free environment. If this assumption is wrong, the copper canisters used in the Swedish deep repository concept could corrode much more quickly than the current estimates suggest. The Swedish scientist Gunnar Hultquist first questioned this assumption in 1986, when he measured an increase in hydrogen concentration in the gas volume above copper in water.103 The results of his experiments are open to question due to problems with the probe used to measure hydrogen pressures. Nevertheless, an additional experiment published in 1989 supported the initial findings.104 These experiments have now been repeated: the researchers conclude that the results show that copper corrodes in water free of dissolved oxygen, forming a hydrogen-containing corrosion product.105 The hydrogen produced by corrosion in pure water is apparently found in the metal, the corrosion product and the water as well as in the gas phase.106 If this conclusion is correct, it has serious consequences for the repository safety case: calculations based largely on observations of corroded copper coins recovered from the Swedish Vasa warship, which sank in 1628, suggest that the copper canisters would need to be more than 1m thick, rather than 5cm, in order to last 100,000 years.107 One response to the experiments suggests that corrosion of copper by pure water alone cannot occur and that there is an alternative explanation for the measurements: namely that the hydrogen evolution observed was caused, not by the reaction of liquid water with copper but by the reaction of adsorbed water with the stainless steel walls of the vacuum container in which the experiment was conducted.108 Further, supporters of the Swedish safety case argue that the coins from the Vasa ship may have corroded because the water was polluted by sewage and contains hydrogen sulphide.109 However, the original authors disagree and have defended their findings.110,111 Other authors have suggested that the Eh-pH diagram for copper may differ from that currently assumed in the safety case, adding weight to the experimental evidence of Hultquist and colleagues that the corrosion of copper in water is not fully understood.112 In 2009, the Swedish National Council for Nuclear Waste (Krnavfallsrdet) held a seminar to discuss this dispute. The report of the seminar includes the views of experts on the findings to date and on additional research that should be conducted. 113 4.1.3. Role of microbes It has been known since the 1980s that microbes might survive in a deep geological repository and that the effects of microbial activity could have profound impacts on waste containment.114 In 1987, microbiology became a part of the Swedish scientific programme for deep disposal.115 Microbes could have a number of adverse effects on the safety of a nuclear waste repository, including causing corrosion of metal waste containers. There is now little doubt that life could survive in a repository in the form of microbes, despite the heat and radioactivity generated by the wastes. In experiments conducted in Canadas Underground Research Laboratory, culturable populations of microbes were found at all locations studied in the bentonite-based sealing materials.116 The microbes included heterotrophic aerobes, anaerobes and sulphate-reducing bacteria (SRB). (A heterotroph is an organism that cannot synthesise its own food and is dependent on complex organic substances for nutrition.) The microbes identified were more abundant at interface locations and absent only in those samples affected by heat and extreme drying (desiccation). Aerobic populations (those requiring oxygen) were significantly higher in the interface environments, especially at the rockbentonite interface. Conditions in the backfill region also appeared to be conducive to microbial activity, causing a reduction in oxygen, followed by a decline in aerobic populations and an increase in anaerobic populations (ie those not requiring oxygen, including SRB). The viable population was considerably larger than the culturable population, suggesting potential for future increased activity if conditions became more favourable.

24

GeneWatch UK consultancy report September 2010

Increased heat (and possibly some radiation) was found to increase nutrient availability in bentonitebased materials and to have a stimulating effect on microbial activity.117 Migration of micro-organisms through the bulk of the buffer appeared to be slow, but migration along the metallic holderbuffer interface was rapid, suggesting that cracks or interfaces may form preferred pathways for migration.118 The buffer used in the experiments was a mixture of sand and clay, rather than 100% bentonite. However, other experiments conducted in Sweden have found that the sulphate-reducing bacterium Desulfovibrio africanus is present in commercially available bentonite, and survives and is viable after exposure to high salt concentrations (which may occur in groundwater at depth) and temperatures of 100C for 20 hours.119 SRB are also a characteristic component of the Opalinus Clay formation, investigated as a potential repository host rock in Switzerland.120 SRB reduce sulphates to sulphides. Sulphides react with copper and could potentially corrode copper canisters in a repository. However, extrapolation from underground experiments suggests that the reaction rate is too low to significantly reduce the predicted lifetime of the canisters.121 In concepts where the repository is to be kept open for a long period of time, to allow for monitoring and possible retrieval of wastes, there may be added difficulties with microbes due to the presence in the ventilated caverns of a humid, oxygen-filled environment. This could provide many potential niches for microbial growth, which could then affect the integrity of the storage canisters.122

4.1.4. Steel corrosion and hydrogen gas generation In a deep repository, hydrogen will be produced by anaerobic corrosion of iron. In the French concept, iron will be present in the steel overpacks for the vitrified high-level waste. In the Swedish concept, the canisters will contain iron, which will be exposed only once the copper has been corroded or damaged. Hydrogen can also be produced by radiolysis (the dissociation of molecules by radiation) of the organic waste contained in some waste packages: for example, in the long-lived intermediate-level wastes generated by reprocessing in the UK and France. If copper corrodes in water alone, as some evidence has suggested (see Section 4.1.2), hydrogen may also be produced by this reaction.123 The corrosion rate of iron and steel may be significantly increased by the presence of gamma radiation. Experiments involving the measurement of hydrogen produced by corroding steel in artificial groundwaters suggest that the corrosion rate could be increased 10- to 20-fold in bentoniteequilibrated groundwater exposed to high levels of gamma radiation.124 The reasons for this are not fully understood. The pressure rise in a repository due to the formation of dissolved hydrogen, and the subsequent production of gas bubbles, might be sufficient to break or fracture the barriers (this is discussed further in Section 4.2.5). Hydrogen embrittlement of the corroding metal might also occur, with detrimental effects on the mechanical characteristics of the overpacks or canisters.125,126

4.1.5. Creep
Creep is the tendency of a solid material slowly to move or deform permanently under the influence of stresses. Creep in copper occurs readily at the high temperatures expected in a nuclear waste repository. In safety assessments calculations are necessary to show that the canisters will not rupture under the stresses to which they will be subjected. Calculations based on a creep model of the Swedish deep disposal canisters under the pressure and temperature conditions expected in the repository suggest that there will be very high stresses at the edges of the canisters, where creep rates will therefore be high, and that the cylindrical canisters will distort to an hour-glass shape in the repository due to elastic and creep deformation. However, the creep strain at the edges is still expected to be less than would be needed to rupture the canisters.127

GeneWatch UK consultancy report September 2010

25

4.1.6. Summary of corrosion issues The mechanisms for corrosion including the role of bacteria are not fully understood. This could result in both copper canisters and steel overpacks corroding more quickly than expected, allowing faster than predicted release of radionuclides into groundwater. A key issue is whether copper canisters corrode in water in the absence of oxygen: if so, their design life has been significantly overestimated. The intense radiation in the repository is also likely significantly to increase the corrosion rate of steel.

4.2. Bentonite erosion and loss of buffer capacity

The bentonite surrounding the canisters or overpacks is expected to provide physical support and to influence the chemistry of the repository, acting as a buffer and slowing the movement of some radionuclides particularly the highly radiotoxic actinides. However, a number of physical and chemical processes can affect bentonite in ways which could compromise safety. Groundwater transport through bentonite remains poorly understood, with diffusion probably taking place through the interlayers of clay particles.128 4.2.1. Effects of heat and mineral changes on bentonite The heat from the high-level waste in the repository will heat up the buffer/backfill and the surrounding rock of the different repository tunnels over a period of several decades as they are successively filled with the waste.129 The different components in a repository all have different expansion coefficients and the way they move and compress may lead to a significant change in the hydraulic properties of the interfaces between them. Heating could also cause significant pore pressure changes, particularly in clay rock, affecting the stress distribution, which could in turn damage the structure of the clay rock so that water flows through it more easily. Furthermore, the heat could induce convective flow of groundwater in the surrounding rock, along with significant vaporisation of groundwater, which may be ventilated in the pre-closure stage. This phenomenon complicates the prediction of how conditions in the repository will change with time, since the effects of water vapour as well as liquid water need to be considered. Experiments show that predicting the combined effect of heating and wetting on the bentonite requires coupled thermo-hydro-mechanical models.130,131 Complex interactions need to be included in these computer models, such as the effects of the wetting and swelling of the bentonite on water, gas and thermal flows and the effect of the changing thermal gradient on the transport of water vapour in the bentonite. Once the repository is sealed, there will be no escape of moisture and the excavated cavity will become re-saturated with groundwater, causing the bentonite to swell.132 The temperature will build up to a peak, which will be reached after some decades near the canister but may take hundreds of years in the far field. This is also the period when the hydraulic pressure will be rebuilt in the backfilled and sealed repository opening. Thermally induced flow or convection and coupled thermalmechanical processes will last much longer than the temperature pulse and could peak at about 10,000 years. The heat in a repository could have a significant negative impact on the properties of clay. The bentonite clay intended to be used in repositories consists mainly of montmorillonite, which is a member of the smectite group of minerals. Smectite is considered to be a good buffer material because it swells in contact with water slowing groundwater flow and also holding the waste canisters firmly in place and because it can retain radionuclides by sorption (a process in which they become incorporated in, or stick to, the clay particles). The swelling bentonite is expected to exert a swelling pressure on the canisters of up to 15MPa, generating considerable stresses133 (see Section 4.1.5). The high-level waste placed in a repository is expected to increase temperatures considerably until its radioactivity has decreased significantly. When smectite clay is exposed to high temperatures and

26

GeneWatch UK consultancy report September 2010

the geochemical conditions of a repository for a long time it could be transformed into other minerals with different physical and chemical properties. Smectite is converted to illite a clay-sized but non-expanding mineral in a reaction which becomes faster as the temperature increases. There are two possible chemical reactions, which depend on the chemical conditions. Experiments conducted in South Korea found that smectite transforms into randomly interstratified illitesmectite layers, and eventually into illite.134 The experiments showed that this reaction can affect the barrier properties of smectite clay which are required for a repository. When the temperature in the experiment increased (from 90C to 200C), the percentage of the expandable smectite layers in interstratified illitesmectite decreased and its sorption capacity was very significantly reduced (with a decrease in sorption distribution coefficient Kd of more than 90% for the caesium and nickel ions tested).135 The Swedish repository concept currently assumes that the negative effects of illite production can be limited by spacing the spent fuel canisters far enough apart to keep temperatures to less than 100C. However, the rate of conversion of montmorillonite to illite is in fact not yet known, and other mineralogical changes can also take place which are not yet well understood. A geological site where bentonites occur naturally at Kinkulle in Sweden suggests that a reduction of 5075% in the proportion of montmorillonite may have taken place over about 1,000 years, at temperatures estimated to have reached a maximum of 150C; this may imply much quicker changes in a repository than has been assumed, which could have serious implications for the safety case.136 A data synthesis reports that experiments at temperatures lower than 100C have also identified significant changes in the buffer: these include drops in swelling pressure of more than 50% in a Czech experiment, and a hundredfold increase in hydraulic conductivity (due to changes in particle structure) found in an experiment conducted in SKBs underground laboratory.137 The most obvious change observed in experiments is the dissolution of minerals such as calcite and feldspar, which are present in the clay, but dissolution of montmorillonite also occurs. This happens in all experiments in a large part of the buffer. Precipitation of iron and silicon then welds the stacks of clay particles together, giving rise to a permanent increase in hydraulic conductivity and a drop in sealing pressure meaning that water flows more easily through the heat-damaged clay, which loses its important sealing properties. A more detailed analysis of the Czech experiments suggests that the typical radius of the larger pores (macropores) in a mixture of 85% bentonite, 10% sand and 5% graphite increases threefold at 85C. The authors suggest that the observed microstructural changes in the experiment can be explained by mineralogical transformations, which have a serious impact on the geotechnical properties of the bentonite-based mixture.138 Altered mineralogy may also impact on other properties of the clay. For smectites, the risk of failure due to creep (the deformation of the clay under long-term strain, such as the weight of the canisters) is believed to be very small, with canisters expected to sink only a very small distance of the order of 10mm over a million years.139 The drop in sealing pressure in heat-damaged clay observed in the Czech experiments might affect this, although the author of this report is not aware of any in-depth studies of creep in heat-damaged clay. Because of the expected adverse impacts of heat on bentonite clay, SKB has developed thermal design criteria based on an upper temperature limit of 100C. The maximum expected bentonite temperature is a function of the thermal properties and geometry of the bentonite barrier, the thermal properties of the surrounding rock mass and the deposition geometry, i.e. the spacing between individual canisters and between tunnels. SKB has developed a computer model to predict maximum temperature based on these parameters.140 It has also conducted heating experiments in the sp Hard Rock Laboratory and used inverse modelling to try to calculate the thermal conductivity of the rock, on the basis of how the temperature changes.141 Uncertainties in the predictions appear to be dominated by spatial variability in the thermal properties of the rock. Some of the experimental measurements also appear to be influenced by groundwater movements. Allowing a margin of error, these studies can be used to determine the spacing of canisters that will be needed to meet the 100C temperature limit.142 The spacing of the canisters is one of the parameters with most impact on the size and cost of a repository (see Section 5.3). If the maximum temperature limit were lowered in order to meet

GeneWatch UK consultancy report September 2010

27

concerns about the effects of temperatures below 100C on bentonite this could significantly increase costs and make it more difficult to dispose of a given quantity of waste in the available volume of bedrock at a potential repository site. Heating can also release gases from clay host rocks. In a study performed in the Opalinus Clay in Switzerland, carbon dioxide and hydrogen sulphide were the most prominent gases released, both of which could interact directly with waste containers or wastes and/or change the chemical conditions in a repository.143 In the experiment, the clay rock in the test field was found not to be gas- tight.

4.2.2. Effects of saline water The salinity of groundwater can also affect the properties of bentonite. In experiments, mineral alteration of bentonite due to the accumulation of magnesium occurred in saline water at temperatures of 60C and 90C. The cation exchange capacity (CEC) decreased as the amount of magnesium increased presumably due to occupation of the internal surfaces and the distribution coefficient Kd for caesium in the altered bentonite was half that in the original, suggesting that the thermal alteration of bentonite in saline water affects the caesium sorption capacity.144 Caesium-137 was used in the experiment, as a chemical analogue for caesium-135, which has a much longer halflife and is considered to be one of the most important radionuclides in the safety assessment for a Japanese repository. The precipitated magnesium may also prevent the swelling of the bentonite. The CEC is a measure of the quantity of positively charged ions (cations) the clay can hold, so a reduced CEC combined with a higher hydraulic conductivity is likely to mean faster escape of some radionuclides. Preliminary experiments conducted in Spain show that swelling induced by the dissolution of salt crystals in clays may be significant in saline groundwaters.145

4.2.3. Effects of other minerals in clay Bentonite clay can also contain other minerals such as carbonates, quicklime, apatite and oxides. These minerals can be formed and dissolved in the matrix. For example, various sizes of carbonate nodule are found in bentonite mines, forming and dissolving as conditions change, and potentially creating large pores or gaps in the clay. Studies in Japan have shown that these minerals can be dissolved in acid, potentially opening spaces in the rock. However, this process may not be similar to what happens in nature.146

4.2.4. Chemical disturbance due to corrosion Bentonite is expected to be well-buffered, leading to stable pH conditions in the repository backfill.147 However, chemical disturbance due to corrosion could change the properties of the backfill. In the French repository concept, steel overpacks rather than copper canisters are expected to be used. The interactions between the corrosion products of steel, the surrounding groundwater and the bentonite are expected to create a chemical disturbance inside the engineered barrier system. Modelling of the system over 100,000 years predicts that the porosity of the bentonite will increase, due to changes in its mineralogy, and that both the Eh and pH will change significantly. However, the model suggests that there will be a feedback effect, involving the clogging of pores in the clay near each steel overpack, which will slow the initial high corrosion rate and its influence on the mineralogy.148 Iron frames used for spent fuel (which are to be contained inside the copper canisters in the Swedish design) will also create a chemical disturbance in the same way. The incorporation of corroded iron into clay can in theory act as a pump to accelerate corrosion. A UK model again shows slowing of corrosion after time due to clogging of pores in the clay, but the chemical reactions assumed to take place differ significantly from the model developed in France. The authors conclude that meaningful application of the model requires key missing data, such as solubilities and free energies at the mineralfluid interface. In addition, the chemical reactivity of hydrogen gas (which is assumed to be GeneWatch UK consultancy report September 2010

28

inert and diffuse away into the rock) may need to be taken into account.149 Experiments suggest that high concentrations of iron ions can be reached in bentonite without any mineralogical transformations but that CEC and swelling pressure may be reduced and hydraulic conductivity increased.150 As with magnesium accumulation (see Section 4.2.2), a reduced CEC combined with a higher hydraulic conductivity is likely to mean faster escape of some radionuclides. Large quantities of cement are also expected to be used in the repository for construction and sealing. The highly alkaline cement pore fluid may have adverse effects on bentonite, significantly reducing its swelling pressure and CEC.151 A number of minerals are expected to form as a result of cement-bentonite interactions152 and computer models of this process have been developed.153 The creation of highly alkaline fluids is expected to degrade the clay rock at the interface with the barriers in the French repository concept, and concrete engineered barriers may also be susceptible to attack by groundwater containing dissolved sulphates.154

4.2.5. Effects of gas on the clay barrier Corrosion of steel in the repository would lead to the generation of hydrogen gas in the backfilled tunnels, which could seriously affect repository safety if pressure build-up were to force fast routes through the bentonite or host rock or explosively damage their structure. Four principal mechanisms have been identified by which gases can pass through clay barriers:155

l two-phase (water plus gas) advective flow (i.e. due to bulk motion through the rock), under the
control of a combination of capillarity (the pull through the clay pores due to the attraction of molecules to the clay) and hydraulic gradient (difference in pressure)

l diffusion of gas through intervening fluid to neighbouring voids in the clay with lower gas
concentration

l deformation of the clay, creating larger pores to accommodate gas flow l fissuring and fracturing caused by gas breakthrough if the gas pressure becomes too high
(i.e. if it does not dissipate fast enough through the other mechanisms). Gas breakthrough is considered to be a serious potential problem as it could permanently damage the engineered barriers and surrounding rock. Diffusion is expected to be limited through the highly compacted clay. Experiments and computer modelling have therefore focused on demonstrating that gas could escape through advective flow before the gas pressure built up sufficiently to cause fissuring. However, advective flow could also have safety consequences because water would be pushed through the clay ahead of the gas. Advective gas flow could therefore speed up the transfer of radionuclides to the surface once groundwater had become contaminated (i.e. when the canisters or overpacks had corroded sufficiently to expose the wastes): this issue is considered further in Section 4.4.3. The mechanisms for gas breakthrough appear to be strongly dependent on the conditions of the tests.156 Gas breakthrough in bentonite is often abrupt, perhaps indicating channelling of gas or fracturing. In illite, the gas breakthrough pressure appears to be lower and less clearly defined: breakthrough seems to develop sequentially through many flow channels. Increased resistance to breakthrough can come from either an increase in saturation or an increase in clay density: the key consideration is the openness of potential flow channels. Above a certain degree of saturation (estimated at 93% in pure bentonite), breakthrough pressure rises sharply. Breakthrough is at least partly time-dependent and can occur at low pressures after a long period of time. Modelling has suggested that hydrogen gas breakthrough could occur in the conditions expected in a repository following the French concept (i.e. using steel overpacks or canisters) and that the gas problem is a key issue for the long-term performance of the clay barrier and hence for disposal safety.157 Hydrogen gas will form at the interface between the steel and bentonite and this could lead to over-pressurisation (gas build-up). Preferential pathways could form through fissuring of the rock, but could be closed again by the self-healing properties of clay, leading to a cyclical process in which the gas pressure again built up.158 However, a more recent model that includes feedback

GeneWatch UK consultancy report September 2010

29

mechanisms suggests that the degree and physical extent of gas pressure build-up may be much smaller than earlier models found.159 In this model, the gas pressure increases initially at the canister, but later decreases and eventually returns to a stabilised pressure that is slightly higher than the background pressure. In the Swedish repository concept, it has been assumed to date that corrosion of copper in the absence of oxygen will not occur and that the design life of the copper canisters is 100,000 years. If these assumptions are correct, hydrogen generation will be limited until the iron inside the copper canisters is exposed much later in the lifetime of the repository. If however corrosion of copper by water can occur in the absence of oxygen (see Section 4.1.2) the hydrogen generated by this reaction might also have significant implications for the safety case; however, the author of the present report is not aware of any studies that have attempted to quantify this. Additional steel may be introduced into a repository for other reasons, e.g. as structural support during excavation (necessary to keep structures open in the case of clay rocks160); or in the form of steel barrels containing long-lived intermediate-level wastes. Hydrogen generation from the corrosion of this steel also needs to be considered in the safety case.

4.2.6. The role of microbes: gas production and biomineralisation Microbes can corrode compacted bentonite, but current models suggest this would occur very slowly (at a rate of a few millimetres in 10,000 years).161 Nevertheless, other potential effects of microbes on the backfill, in particular their alteration of the mineral composition of bentonite or their generation of gases, may have significant implications for a repositorys safety case.162, 163 Microbes can both produce and consume gases164 and microbial gas production could cause a build-up of gas in a repository, potentially reducing the effectiveness of the clay-based barrier. The generation of carbon dioxide and other gases could also enhance radionuclide solubility and transport.165 Microbial processes could in addition affect adsorption/precipitation of radionuclides, chemical conditions and the creation of colloids (see Section 4.3.2). At the Severnyi repository of intermediate-level liquid radioactive wastes in Siberia, Russia, injection of waste has been shown to stimulate methane and hydrogen sulphide production by microbes, which existed at all depths investigated (up to 405m).166 Experiments in the Canadian Underground Rock Laboratory have shown that the production of gas (methane) in backfill is possible and could occur in the prevailing chemical conditions.167 The 1996 FEBEX experiment found that significant amounts of hydrogen, carbon and hydrocarbons were formed due to either thermal or microbial decomposition in the bentonite.168 More than 0.35m3 of carbon dioxide per 100kg of bentonite was formed, which may indicate that the gas could enhance the transport of radionuclides and microorganisms. The gas could also have a significant impact on the permeability of the buffer material by disrupting its mechanical structure, e.g. by increasing the size and frequency of the clay pores in the bentonite near the canisters, and hence its permeability. Microbial degradation of organic material within radioactive wastes, the main component being cellulose, can lead to generation of gases: this was a particular issue for the UK Nirex safety assessment programme due to the inclusion of large quantities of cellulose in the intermediate-level wastes destined for the proposed repository at Sellafield.169 Degradation products of cellulose can also enhance the solubility of radionuclides such as plutonium. Bacteria often play an important role in the production of minerals in a process known as biomineralisation.170 Depending on the conditions and micro-organisms available, biomineralisation can clog pores in rock by means of precipitation or enhance permeability due to decreased solid content.171 The conversion of montmorillonite to illite in the bentonite backfill (or surrounding clay rock) may have significant impacts on a repositorys safety case, as described above (Section 4.2.1). Findings that micro-organisms can dissolve smectite at room temperature (by reducing Fe(III)) have been described as a major challenge in the context of deep geological disposal, since they suggest that this process may happen much faster than predicted, even in the absence of significant heat.172

30

GeneWatch UK consultancy report September 2010

4.2.7. Summary of bentonite erosion and loss of buffer capacity The effects of intense heat on the bentonite backfill of a repository could seriously damage its ability to trap some radionuclides. Chemical and physical disturbance due to corrosion, gas generation and biomineralisation could also adversely affect the properties of the bentonite backfill.

4.3. Solubility, sorption and transport of radionuclides

4.3.1. Geochemistry and buffer chemistry The longer it takes for a given radionuclide to diffuse across the clay buffer, the lower the rate of release of that radionuclide from the near-field engineered barrier system will be, thanks to radioactive decay.173 Radionuclides released from the waste will precipitate when their concentrations in the pore water exceed their solubility in the water. This will limit the concentrations of many radionuclides in the buffer and thus their release rates to the surrounding rock. The chemical conditions of the buffer are expected to delay significantly the release of some radionuclides but not others. For example, in the French safety case, the mobile radionuclides chlorine-36, iodine-129 and selenium-79 are assumed to be non-sorbing in the clay rock, and are consequently expected to be the only three radionuclides that enter the biosphere during a millionyear timeframe (although it is possible that the solubility of selenium may be reduced by other complex chemical mechanisms).174 In contrast, the release of actinides, such as plutonium, is expected to be delayed by the chemical properties of the buffer. Thus the thickness of the buffer is expected to have a major impact on the release of relatively short-lived actinides, such as plutonium-241 (formed by the decay of curium245) and plutonium-239.175 However, the effectiveness of the buffer will depend on the chemical conditions (such as pH and Eh) and also on the physical and chemical form of the radionuclides. The speciation of radionuclides is the distribution of a radionuclide among different chemical species in a system. Species are defined by a wide variety of properties, such as charge, oxidation state, structure and degree of complexation.176 Safety can be significantly affected by issues such as whether radionuclides exist as particles (which may be more easily trapped in the bedrock or clay) or colloids (which may be much more mobile, see Section 4.3.2). There is particular concern that actinides, such as plutonium, might be transported long distances on colloids. Estimation of the transport of radionuclides from a repository requires careful prediction of the chemical and physical interactions of the radioactive waste with the bentonite buffers and surrounding rock over extremely long periods of time. The complex mechanisms involved include advective-diffusive transport of radionuclides in groundwater (including advection, diffusion, and osmotic and ion restriction effects) and geochemical reactions (complexation, exchange, precipitation, adsorption and desorption) under different temperatures and pressures. Preliminary safety assessments have assumed that the chemical retardation of radionuclides in the buffer can be calculated using a constant retardation factor, Kd. However, more sophisticated computer modelling of the interactions between the different chemical species and the buffer suggest that using the Kd approach does not provide a good approximation of contaminant transport and can result in significant errors.177 In particular, temperature has a great impact on the expected concentrations of contaminants in groundwater. Coupled thermo-hydro-mechanical-chemical processes will occur, which require complex modelling.178 The results show that there are still significant shortcomings to geochemical modelling and its applicability to real-world repository conditions. The geochemical suitability of a repository site is determined by the composition of the host rock and groundwater, which influence radionuclide solubility, chemical buffer capacity and radionuclide retention. However, selection of suitable conditions is generally not straightforward because of the multitude and complexity of the reactions involved.179 The chemical parameters used in reactive transport models are not known accurately due to the complex and heterogeneous conditions and there can be multiple alternative conceptual models, none of which explain the data.180

GeneWatch UK consultancy report September 2010

31

Hydrogen produced by the corrosion of canisters and overpacks could act as a reducing agent and change the chemical conditions in a repository. In particular, the reduction of aqueous or mineral sulphates and other oxidised species present in the site could change the original geochemical conditions. Experiments suggest that reaction rates are highly dependent upon temperature but can probably be ignored in nuclear waste performance assessment.181 However, further investigations are needed. 4.3.2. Colloids and complexation A colloidal system is a type of mixture in which one substance (the colloid) is dispersed evenly throughout another. Milk is an example of a colloidal system, consisting of globules of fat dispersed in a water-based liquid. Colloid particles have diameters ranging from 1nm to 1m and have a high surface area.182 Many radionuclides easily sorb onto colloids suspended in water and this can make them highly mobile and more easily transported through rock. Computer models that do not account for transport by colloids can therefore significantly underestimate the rate of transport of radionuclides in groundwater.183 Migration on colloids is of particular concern in the case of actinides, such as plutonium, which can be transported large distances in groundwater as colloids, and as a result could potentially be washed out of the bentonite buffer of a repository, rather than being retained there.184 There are still significant gaps in the understanding of the transport of actinides bound to minerals and colloids. However, experiments suggest that actinide speciation may be dominated by colloid forms.185 Humic matter is decayed organic matter, which is an important constituent of soil. Clay is an important source of humic colloids, which can have significant effects on radionuclide migration.186 The bentonite backfill of a repository could generate colloids, which could adsorb or incorporate radionuclides and transport them over long distances, or retain them by interaction with mineral surfaces or by agglomeration (gathering together as a mass).187,188,189 Bentonite colloids can diffuse within granite.190 Both solid particles and colloids could be detached from bentonite at the bentonite/granite interface in a repository and mobilised by the water flow. It has been shown that these colloids are very stable in low saline and alkaline waters, and could facilitate radionuclide transport in the fracture network of the excavation disturbed zone (EDZ) in the granite around a repository.191 Naturally occurring rare earth elements (REEs) can be used as chemical analogues for studying the behaviour of actinides. Preliminary studies at the Swedish Forsmark site suggest a strong association of REEs with colloids in the groundwater in the overlying aquifer but limited mixing and no evidence of transport from the bedrock groundwaters to the aquifer.192 The presence of oxidants can also enhance actinide transport significantly, due to the formation of complex species, which may increase solubility by orders of magnitude and potentially enhance mobility.193 Cementitious materials are commonly used to stabilise some radioactive wastes, such as long-lived intermediate-level wastes, which may be co-disposed with high-level wastes in some countries (such as the UK). In such wastes, cellulosic materials present in the wastes (tissues, cotton or paper) can exacerbate the above difficulties by forming organic compounds, which may then form complexes with actinides.194

4.3.3. The role of microbes Microbiological processes must be taken into account when modelling groundwater hydrogeochemistry: they are expected to be involved in many reactions which would not occur in a lifeless underground environment. Experiments in the sp Hard Rock Laboratory in Sweden confirmed the presence of SRB, which produce sulphide corrosive to copper, and autotrophic acetogens, which produce acetate from hydrogen and carbon dioxide. The analyses also showed that different rock fractures can have very different hydrogeochemical characteristics and

32

GeneWatch UK consultancy report September 2010

microbiological profiles.195 Measurements of the redox state (Eh) and pH at Forsmark and other sites in Sweden suggest that the waters are buffered by sulphate reduction, consistent with the presence of SRB.196 The presence of bacteria is important because microbes can affect the mobility of radionuclides in a number of ways.197,198 They can enhance radionuclide migration by sorption, or reduce it by immobilising radionuclides in biofilms. They can also influence the release of radionuclides by altering bulk water chemistry (especially pH and redox), by producing organic complexing ligands, or by direct accumulation onto or into cells. These complex biological effects on radionuclide transport are poorly understood but must be considered in a repository safety case.

4.3.4. Release of radioactive gas The principal source of gas in repository designs that use steel waste containers is expected to be hydrogen produced by the corrosion of the steel (see Section 4.1.4). The concerns relate to any damage to containment that might be caused by pressure build-up (see Section 4.2.5) and to the potential role of the gas in pushing radioactively contaminated water upwards out of the repository. However, carbon dioxide and methane are other gases that might be produced in a repository. These gases are likely to contain radioactive carbon-14 and may pose a radiological hazard in themselves as they leak from the repository. Carbon-14 has a high production rate in nuclear reactors and is released to the environment in discharges as well as through the disposal of radioactive waste.199 It has a long half-life (5,730 years) and high mobility in the environment. The majority of carbon-14 produced in reactors is either trapped in the spent nuclear fuel, structural materials or graphite moderator, or else produced in the reactor coolant. A large inventory of carbon-14 produced in nuclear power plants is captured in ion-exchange resins, which are not heat-generating and hence not classified as high-level wastes. However, carbon-14 is also contained in spent nuclear fuel, and in some countries long-lived non-heatgenerating wastes may also be co-disposed with high-level wastes in a deep repository. In the case of a repository for low-level radioactive waste, carbon-14 is the only radionuclide that is expected to contribute significantly to radiation exposure via the gas pathway, as all other gaseous radionuclides can be neglected due to their short half-lives, low inventories or low radiological relevance.200 For example, the Asse salt mine in Germany showed a continual release of carbon-14 during its operational phases as a result of mine ventilation, providing evidence of ongoing reactions in the waste.201 In many low-level waste facilities, the carbon-14 inventory is the limiting factor in meeting regulatory requirements. Carbon-14 can be released in groundwater or as carbon dioxide or methane gas. It is taken up by food crops and vegetables through photosynthesis and by root uptake from soils, and can then be ingested by humans. For the proposed US deep repository at Yucca Mountain, the collective dose due to carbon-14 was predicted to exceed the Environmental Protection Agencys limit, although the dose per person was low.202

4.3.5. Summary of solubility, sorption and transport of radionuclides Poorly understood chemical effects, such as the formation of colloids and the role of microbes, could speed up the transport of some of the more radiotoxic elements such as plutonium. Build-up of gas pressure in a repository could damage the barriers and force fast routes for radionuclide escape through crystalline rock fractures or clay rock pores. Radioactive carbon dioxide and methane could also be released.

4.4. Bedrock properties and hydrogeology

4.4.1. Groundwater flow in the bedrock and fractures Crystalline rocks contain fractures and faults, which are of critical importance in determining the flow of radionuclides out of a repository. In contrast, for clay rocks, used in the French concept, a key

GeneWatch UK consultancy report September 2010

33

assumption of the safety case is that transport would be by slow diffusion through the clay, rather than through cracks and fissures, which are assumed to be self-healing.203 Groundwater flow through crystalline rock takes place mainly through fractures, as the rock itself has very low permeability. However, flow through both fractures and porous rock needs to be considered in a safety assessment. This poses particular problems because of the very large degree of structural variation (known as heterogeneity) in the fracture systems, which means that the permeability of each piece of rock is different, and varies in different directions.204 The hydraulic conductivity can vary by one or two orders of magnitude at different points, leading to very different thermo-hydro-mechanical properties at different points in space. Despite the progress that has been made in scaling up the measured properties of fractured rock to try to predict overall flows, the problem is so complex that it has yet to be resolved. Producing accurate models of fractures in the rock through which radioactive water and gas can flow is difficult because it is hard to extrapolate from measurements on the surface of a block of rock in order to describe correctly the network of fractures inside it. This means that markedly different fracture densities, hidden in the rock, could be consistent with the same experimental data.205 An attempt to characterise a three-dimensional fracture network in a 1m3 block of granite has recently been published: however this is the first data-set of its kind.206 At the Forsmark repository site in Sweden, three major sets of deformation zones have been identified, plus a fourth subvertical zone. Two gently dipping brittle deformation zones seem to play an important role in determining the properties of the site, such as the distribution of stress, fracturing and transmissivity within fractures. Four main groundwater types are present.207 The complex groundwater evolution and patterns are a result of many factors, including present-day topography and proximity to the Baltic Sea; past changes in hydrogeology related to glaciation/deglaciation, land uplift and repeated marine/lake water regressions/transgressions; and organic or inorganic alteration of the groundwater composition caused by microbial processes or rock/water interactions. A major conclusion from site investigations is that changes from glacial rebound and hence hydrology seem to have a major influence on groundwater chemistry.208 Currently, the confidence concerning spatial variation is low due to relatively few observations having been made at depth, and there are significant uncertainties and discrepancies between models of the site as a result of the different assumptions made.209 Pore water in the rock has exchanged with water circulating in fracture networks over extremely long periods of time.210 Mixing models have been used in an attempt to understand how groundwater chemistry has developed at Forsmark and the alternative Swedish repository site Simpevarp/Laxemar through mixing of the main groundwater types along with water/rock interactions and biological reactions.211,212 However, the robustness of the model outputs is quite sensitive to the variables included. In addition, similar mixing proportions and mass transfers can be obtained using different reactions. Further, when chemical reactions produce an important compositional change the model may not correctly reproduce the mixing proportions.213 A coupled model of regional groundwater flow and solute transport, applied to the Simpevarp area, suggests that the main sensitivities are to the top surface flow boundary condition, the influence of variations in fracture transmissivity in different orientations (anisotropy), spatial heterogeneity in the regional deformation zones and the spacing between water-bearing fractures (in terms of its effect on diffusion through the rock matrix). Again, the best match to the observations may not be unique, introducing additional uncertainty.214 The large volume of accessible pores in fractured granitic rock may retard the migration of radionuclides through sorption onto the rock.215 However, the effects depend on the radionuclide, with experiments in Sweden suggesting that some actinides are retarded in the rock, while others (e.g. neptunium) may break through with hardly any retardation.216 Data from the Callovo-Oxfordian clay formation at Bure in the eastern Paris basin the proposed French repository zone suggests that groundwater residence times may be very long, although significant uncertainties remain due to the complexity of the hydrogeological system.217 The conductive layers in the clay are heterogeneous and there are several such layers identified as

34

GeneWatch UK consultancy report September 2010

porous zones.218 There is also an overpressure a pressure difference compared to the surrounding rock of 2060m within the Callovo-Oxfordian argillite, which has not yet been explained. The current preferred hypothesis is that this is due to chemical processes (chemical osmosis in which the rock behaves as a semi-permeable membrane).219,220 Osmotic flow of water has been shown to occur in samples of Opalinus Clay from Switzerland.221 4.4.2. Excavation damage Excavation causes significant stresses in rock and can change the apertures of fractures, which are important for determining the future groundwater flow through the wastes in a repository and the surrounding rock.222 Reduction of pore pressure will also occur during excavation as water is taken out of the system and gases that were under pressure in the water are released. These processes can influence fracture size and permeability, making it harder to predict water and gas flows after closure. After closure, it is not expected that the system will return to pre-excavation conditions, because of mechanical hysteresis (the effects of past stresses retained in the system). Excavation damage depends on local geological conditions and the excavation method: for example, it is greater in the case of drill and blast excavation than with mechanical excavation. The EDZ consists of a failed zone, in which blocks or slabs may detach completely from the surrounding rock; a damaged zone containing micro-cracks and fractures; and a larger disturbed zone where rock stress and water pressures may be altered. If high groundwater flow occurs in the EDZ, concerns include the possibilities that harmful chemical species may be transported from the surface to the engineered barriers, diffusion of radionuclides from the wastes into groundwater will be increased and fast routes for release of radionuclides could be created.223 When the stresses on the boundary of an underground excavation reach the rock mass strength, failure occurs. In good-quality hard rock, the failure process involves splitting and cracking, known as spalling. Calculations in Sweden suggest that the probability of spalling is low down to a depth of about 550m but that the probability increases below this.224 Explosive spalling (rock bursts) can occur in hard, brittle rock at these depths. At depth, it is likely that the excavations will induce stress concentrations above the rock mass strength. In addition, the heating from the spent nuclear fuel in a repository will increase stresses due to thermal expansion of the rock. These stresses could affect the stability of the rock mass pillars that surround the canisters and must be taken into account in the design.225 Repository construction will require the excavation of many underground openings. In the Swedish concept these range in size from the 1.8m diameter emplacement holes for the spent fuel, of which about 4,500 are needed, to an 8m wide x 15m high cavern required for the underground operations needed to transship spent fuel to different locations in the repository. The excavation-induced stresses form an EDZ in which hydromechanical and geochemical modifications induce significant changes in flow and transport properties.226 Strength degradation of the rock may occur over time due to micro-cracking or micro-fracturing.227 In clay rocks such as those in France, methods are being developed to limit the flow of groundwater through the EDZ by creating radial slots filled with bentonite to interrupt the flow.228 Clay-based seals may be key components in repository designs.229,230,231 In clay rocks studied in France and Belgium, an unpredicted hydraulic perturbation was found at a large distance (greater than 30m) from excavation in both clays. Herringbone fractures were observed ahead of the gallery excavation front and boreholes, and eye-shaped fracture patterns were also observed around boreholes.232 In the Opalinus Clay in Switzerland it has been found that the possibility of temperature-induced deformation of clay rocks due to the emplacement of high-level wastes cannot be neglected.233 At the surface, an uplift of up to 1m has been predicted. This is expected to occur smoothly and over a wide area, so is not considered likely to cause major damage to surface structures. However, below the surface significant damage could occur to tunnels and to tunnel linings, unless they are sufficiently strong or flexible.

GeneWatch UK consultancy report September 2010

35

4.4.3. Gas flow It is now recognised that the ability to understand and predict underground gas migration is crucial to the design and management of nuclear waste repositories. However, computer models of combined water and gas migration (known as two-phase flow) in an underground nuclear waste repository are still at an early stage of development.234,235,236,237,238 Considerable complexity exists due to the highly different porous media that may surround the gas-generating waste packages: concrete buffers or plugs, bentonite backfill, and damaged or fractured zones in different host rocks. Heat-damaged (cracked) bentonite backfill or clay rock and excavation-damaged or fractured rock may provide fast routes for gas escape. 4.4.4. Summary of bedrock properties and hydrogeology Unidentified fractures and faults, or poor understanding of how water and gas will flow through faults, could lead to the release of radionuclides in groundwater much faster than expected. Excavation of a repository could create fast routes for radionuclide escape through the part of the rock damaged by the excavation.

4.5. Human intrusion and human error

Other scenarios which should be considered include human intrusion.239 Spaces deep below ground may be subject to hydrocarbon or mineral extraction and increasingly used for geothermal energy production or for storage: for example, storage of gas for energy in salt caverns, or potentially storage of hydrogen or of CO2 as part of planned carbon capture and storage systems.240 This raises the possibility that future generations seeking to access such spaces may inadvertently drill into a repository and be exposed to potentially high levels of radiation. Repository sites are supposed to be chosen to minimise the risk of human intrusion by avoiding sites likely to be subject to the extraction of raw materials (minerals, coal, oil, gas) or drinking water or used for geothermal energy production.241 However, in practice it may be impossible to anticipate how future generations will wish to use underground space and resources. If human intrusion takes place in the form of underground drilling, radioactive wastes could be rapidly released. Solid material, which might be highly radioactive, could be rapidly ejected from a repository into a borehole during an exploratory drilling operation if the gas pressure in the repository exceeded the pressure of the column of drilling mud.242 Deliberate intrusion is also possible in that the contents of repositories could be attractive to some some of the wastes would be suitable for the manufacture of nuclear weapons and dirty bombs for thousands of years and the sites will also contain very substantial amounts of precious raw materials. Human error during the process of disposal is one of the hardest scenarios to identify and evaluate. Issues include the use of damaged canisters or overpacks and the disposal of poorly catalogued materials. If fresh, rather than irradiated, nuclear fuel were buried, it could undergo a nuclear chain reaction (criticality) while underground, potentially causing significant damage to the engineered barriers and the surrounding rock.243

4.6. Ice ages and glaciation

One of the greatest long-term threats to the integrity of deep repositories is likely to be the effects of future glaciation. Despite global warming, the next glaciation is expected to occur at 10,000 to 100,000 years in the future, and glaciation/deglaciation is likely to cause the most significant perturbation to a repository in this timeframe.244 There have been at least five ice ages in earths history; several factors are thought to be important in causing them, including changes in the earths orbit around the sun and variations in the suns output. The last glaciation ended more than 8,000 years ago but its effects on geology and groundwaters are still visible. Post-glacial rebound the slow upward movement of rocks which occurs after the weight of the ice has been removed is still occurring in regions that were under ice sheets, such as northern Europe and Canada. Repository sites in Europe and Canada are likely to be affected by future glaciations because future ice sheets are likely to extend over similar regions to those in the past.245 Direct erosion of glacial

36

GeneWatch UK consultancy report September 2010

troughs and fjords by ice in Europe was confined to Scandinavia, Scotland and the Alps, but more minor incisions and meltwater channels extended over a much larger area. Major fault displacements (scarps) occurred in northern Sweden and smaller ones in Finland and Canada. Sites near the margins of future ice sheets will be subject to repeated glacial advances and retreats and will thus undergo repeated rock stresses and hydrological changes. The greatest problems are likely in lowland regions exposed by the rapid retreat of thick ice fronts, where large lakes on or under thick warm-based ice are dammed by more distant cold-based ice. Groundwater in fractures dilated (opened up) by glacial unloading may reach over-pressures capable of hydraulically lifting blocks of bedrock or eroding more permeable rocks to depths of about 360m. When the load of grounded ice is lifted, deep accumulations of hydrocarbon gases may be capable of blowing craters or caves in bedrock. Rapid retreats of future ice sheets may therefore represent the horizon to practical safety assessments for nuclear waste repositories. Ice meltwater, which is alkaline, could significantly change the composition of the pore water around a repository and the chemistry of the bentonite buffer.246 An oxidising environment at depth would increase the solubility and mobility of many radionuclides and the corrosion of the canisters. Melting of glaciers is likely to be accompanied by oxidation and the formation of iron oxides, although studies in Sweden suggest that oxygenated waters do not readily penetrate beyond a depth of 100m, so a deep repository may be well-protected.247 However, modelling suggests that oxygen could reach long distances downwards during the lifetime of a repository and penetration will depend on hydraulic properties that may vary significantly with location, with, for example, oxidising conditions occurring after a relatively short time if a fracture connects the repository with a highly conductive fracture zone.248 The presumed depth of dissolved oxygen migration is greatly influenced by the assumptions that are made in the conceptual models.249 Research using samples of brines occurring in crystalline rocks in Canada, Finland and Sweden suggests that these waters have been concentrated from seawater, by freezing during glacial times.250 The researchers calculate that these brines were formed relatively recently (within a few hundred thousand years during the Pleistocene period, which began more than 2 million years ago and ended around 10,000 years ago) and that the consequent dynamic behaviour of the cryogenic fluids is in disaccord with the established consensus that the hydrological system in deep crystalline basement rocks is stagnant. They state that this finding should raise concern about the planning and construction of high-level nuclear waste repositories in such rocks. Measurements of minerals deposited in fractures at Forsmark show that different generations of fracture minerals are common, which implies that the fractures have been conductive several times and probably for long periods. Waters of quite different chemistry have been present at different times over the past million years as a result of repeated glaciation/deglaciation and transgression/regression of the Baltic Sea251. There are also many older fracture-filling events involving the migration of fluids many millions of years ago.252,253,254 A study of the release of uranium from the Palmotto natural uranium analogue site in Finland suggests that release occurred in two or three violent episodes in the last 300,000 years, probably due to repeated inflows of oxic glacial meltwater.255 At the UK Sellafield site, borehole measurements suggest that cold climate recharge occurred at depths of about 700m, probably during the Pleistocene glacial periods between 2 million and 10,000 years ago.256 Modelling of the effects of a future glaciation on a hypothetical repository in the Canadian Shield, based on the last glacial cycle between about 60,000 and 11,000 years ago, suggests that under extreme conditions permafrost is able to develop down to the assumed 500m repository depth or lower. During ice sheet advance, there is a rapid rise in hydraulic head (pressure due to the ice sheet), high groundwater velocities (two to three orders of magnitude higher than under non-glacial conditions) and deeper recharge from surface water. During ice sheet retreat, the gradients reverse. In the fracture zones, the upward hydraulic gradient lasts for about 100 years, whereas in the rock matrix at depth it can last for tens of thousands of years.257 The effects of temperature, salinity, stressdependent permeability, permafrost and large-scale isostasy (i.e. the effects of the weight of the ice on the rock) were omitted from the coupled hydro-mechanical computer model of water flow in the rock underneath the ice, although the formation of permafrost may have significant effects on groundwater flow and chemistry.258 The modelled head distribution is 3D, reflecting the 3D nature of

GeneWatch UK consultancy report September 2010

37

the geometry of the fracture zones in the rock and the subglacial drainage channels.259 The details of the fracture network modelled significantly affect the response. A model of the effects of glacial cycles on the Bure potential repository site in France has identified a memory effect of the last glaciation at depth, where as much as 80% of the last glacial maximum head disturbance may remain.260 The long-term effects of glaciation on repository safety could be very serious, potentially involving a large release of radionuclides due to glacial flushing from a damaged repository zone. Future glaciations could cause faulting of the rock, rupture of containers and penetration of surface and/or saline waters to the repository depth, flushing out radionuclides as the ice melts. Future glaciations therefore place a serious limit on the predictability of containment of the buried wastes.

4.7. Earthquakes
Inactive faults may be reactivated during the lifetime of a repository and earthquakes could severely damage the containment system, including the canisters, backfill and the rock. Networks of monitoring stations in north-west Europe have identified the positions of seismic events since the 1970s. In Britain, there are two regional-scale clusters of seismicity, one occupying the length of onshore western Britain and the southern North Sea, the other in the northern North Sea. However, this seismicity data only represents a few decades of observations and it could be argued that this length of historical record is not very relevant to earthquake hazard assessment over periods of tens of thousands of years. Seismic reflection data indicate that the fault density is as great in the areas of the UK that have been seismically quiet in the historical period as it is anywhere else; given the difficulty in declaring a fault extinct, such faults could become seismic hazards during the lifetime of a repository.261 The Prvie Fault system in northern Sweden contains faults that have been reactivated since the Precambrian and which are strong candidates for future movement under suitable stress conditions.262 In Finland and Sweden, changes in the mass of glacial ice sheets associated with periodic advances and retreats of ice are associated with very strong earthquakes. It is difficult to predict the extent to which faults may be reactivated by glaciations.

4.8. Transport of radionuclides in the biosphere

Once radionuclides reach the biosphere, they may expose humans to radiation in a variety of ways. As part of the safety assessment of a proposed repository, computer models are used to calculate expected doses to humans via pathways such as ingestion of radionuclides in drinking water and food, inhalation of radionuclides, and external radiation from radionuclides in soils.263 Computer models of the behaviour of relatively well-known radionuclides in scenarios such as a nuclear accident can give reasonable predictions. For example, a comparison of nine computer models of ecological transfer and thyroid doses resulting from the release of iodine-131 following the Chernobyl nuclear accident found agreement within a factor of ten with dose measurements.264 However, different radionuclides move in different ways in the near-surface environment, including in soils, lakes and streams.265 There may be multiple migration mechanisms involved, including transport by air, water, particulate matter and biota, which further complicate dose estimates.266 Estimates of the effects of radionuclide exposure on health may also be revised in future as scientific understanding improves.267 The speciation of radionuclides is of great importance for biological uptake, accumulation and biomagnification.268 Radionuclide transfer from soils to food crops can vary considerably with the radionuclides, plant species, soil types and times of deposition, and there is considerable uncertainty regarding these transfer factors.269 Many data gaps also remain in factors governing the transfer of radionuclides in animal feedstuffs to domestic farm animals, which will contaminate the human food chain via meat and milk.270 Repositories located near to the coast are expected to discharge some radionuclides into the marine environment and here too there are uncertainties regarding the bioaccumulation of radionuclides in different species of fish and shellfish, and particularly in the rates

38

GeneWatch UK consultancy report September 2010

of sorption and re-release (desorption) of radionuclides into and from seabed sediments over long timescales.271 Until recently, it was assumed that plants did not play an important role in the transport of actinides such as plutonium in the biosphere. However, studies of the US plutonium-contaminated site at Savannah River have shown that a large proportion of the buried plutonium has unexpectedly migrated upward. Simulations indicate that because plants create a large water flux, small concentrations taken up in plants over long periods may result in a measurable concentration of plutonium on the ground surface.272 This finding will not be relevant to repository safety if actinides are contained by sorption in the bentonite backfill deep in the repository. However, the concentration of plutonium by plants could be an issue of concern if it is transported to an aquifer faster than expected, perhaps in the form of colloids (see Section 4.3.2). Plutonium has also been detected in groundwater in the prevailing flow direction in a borehole close to the vault at the Maiiagala shallow radioactive waste repository in Lithuania. Investigation of possible colloid-mediated transport is planned. The presence of tritium and carbon-14 in groundwater at the Maiiagala site (which operated from 1963 to 1989) also suggests possible uptake of these radionuclides in plants, with measurements confirming the transfer of tritium to plants.273 Other mechanisms of radionuclide transport and accumulation, as well as impacts, may be missed because the current approach to radiological protection is based on simplification of systems, rather than acknowledging and addressing complexity.274 A more ecosystem-focused approach would recognise multiple feedbacks (such as the ways that organisms can affect environmental concentrations of radionuclides, as well as vice versa), the limitations of extrapolations and the potential importance of indirect and ecosystem effects over long timescales.275 One example of an area that is only just beginning to be studied is the accumulation of radionuclides in invertebrates, including beetles, ants, spiders and millipedes, which are a major dietary component of many animals and therefore one potential route into the human food chain.276 Climate change including both global warming and future glaciation will change ecosystems significantly, including drastic changes from aquatic to terrestrial systems and vice versa as sea levels rise or fall at a particular location. This prospect poses additional challenges for radiological protection.277 Currently, different climate states are considered in safety assessments, but not the transitions between them: this means that some scenarios that might result in higher releases such as the accumulation and then release of radionuclides below an ice shield during a glaciation event are not included in the models.278 Processes at the biosphere/geosphere transition zone (such as groundwater recharge rates) are also neglected, although they may be essential for modelling radionuclide mobility during climate transition phases. A typical scenario for future exposures presumes the existence of a group of people living above a repository and deriving all its water from a well in the aquifer above the waste. The water is used for drinking by humans and animals, exposing people directly via the water and via meat, milk and eggs from the livestock; and also for irrigation, exposing people via soil contamination, plant uptake, and ultimate ingestion of soil and plants, as well as via external exposure and inhalation of suspended soil.279 There are significant social uncertainties regarding future human behaviour, as well as uncertainties in the physical, chemical and biological behaviour of each radionuclide. Further, because radionuclides are assumed to be diluted in the well, the above scenario may not always be the highest exposure route for future generations, compared with, for example, consumption of fish or shellfish in which radionuclides have bioaccumulated.280

GeneWatch UK consultancy report September 2010

39

5. Overarching unresolved issues

5.1. Safety assessment: the evidence base, the methodology and its limitations
The literature review set out above suggests that significant releases of radioactivity from a deep underground repository could occur in a number of ways:

l Copper or steel canisters and overpacks containing spent nuclear fuel or high-level
radioactive wastes could corrode more quickly than expected.

l The effects of intense heat generated by radioactive decay, and of chemical and physical
disturbance due to corrosion, gas generation and biomineralisation, could impair the ability of backfill material to trap some radionuclides.

l Build-up of gas pressure in the repository, as a result of the corrosion of metals and/or the
degradation of organic material, could damage the barriers and force fast routes for radionuclide escape through crystalline rock fractures or clay rock pores.

l Poorly understood chemical effects, such as the formation of colloids, could speed up the
transport of some of the more radiotoxic elements such as plutonium.

l Unidentified fractures and faults, or poor understanding of how water and gas will flow through
fractures and faults, could lead to the release of radionuclides in groundwater much faster than expected.

l Excavation of the repository will damage adjacent zones of rock and could thereby create fast
routes for radionuclide escape.

l Future generations, seeking underground resources or storage facilities, might accidentally

dig a shaft into the rock around the repository or a well into contaminated groundwater above it.

l Future glaciations could cause faulting of the rock, rupture of containers and penetration of
surface waters or permafrost to the repository depth, leading to failure of the barriers and faster dissolution of the waste.

l Earthquakes could damage containers, backfill and the rock.

Although computer models of some of these processes have undoubtedly become more sophisticated, fundamental difficulties remain in predicting the relevant chemical and geochemical reactions and complex coupled processes (including the effects of heat, mechanical deformation, microbes and coupled gas and water flow through fractured crystalline rocks or clay) over the long timescales necessary. To date, preliminary safety assessments have been produced for the selected sites in Forsmark, Sweden281 and Olkiluoto, Finland282 and the selected region of Bure, France.283 All these assessments have been produced by the nuclear waste management organisations SKB (Sweden), Posiva (Finland) and Andra (France) themselves. Safety assessments have also been produced in the past for the Yucca Mountain site (now abandoned) and for the failed plan to bury long-lived radioactive wastes near Sellafield in the UK.284 According to the Finnish nuclear waste disposal company, Posiva, the following issues are still pending final resolution and will be addressed in future updates:285

l the initial state of the site (e.g., in situ stresses, the fracture network, hydrogeochemical
conditions at repository depth)

l the impact of the EDZ and thermal spalling on the hydraulic evolution l the evolution of buffer/backfill saturation (e.g. time to reach full swelling pressure) and its
consequences for the performance of the engineered barrier system

40

GeneWatch UK consultancy report September 2010

l flow paths to and from the repository (e.g. groundwater flows at canister scale, release paths
for radionuclides)

l reaction rates and (experimental) evidence of the sequence of hydrogeochemical reactions in

the near field, especially with respect to the reactions leading to the production of sulphide

l the impact on the engineered barrier system of cementitious materials and other stray
materials used in repository construction

l the long-term performance of repository closing and sealing materials and the consequences
for safety

l the impact of external conditions related to glaciations e.g. taliks (unfrozen layers of ground
in regions of permafrost) and glacial meltwater intrusion on the long-term performance of the engineered barrier system. However, there remain fundamental difficulties in resolving these issues, as discussed below. 5.1.1. Unknowns, uncertainties and model validation A landmark paper published in 1994 argued that verification and validation of numerical models of natural systems is impossible.286 This is because natural systems are never closed and because model results are always non-unique. Models can be confirmed by the demonstration of agreement between observation and prediction, but confirmation is inherently partial. Computer models can only be evaluated in relative terms, and their predictive value is always open to question. The objective of site investigations for a nuclear waste repository is not primarily to produce a geoscientifically true model, but rather to provide a basis for good decisions. Results are subject to uncertainty not only due to inherent variability but also due to lack of knowledge (epistemic) uncertainties, including systematic bias, which can have a large influence on the results. It is usually assumed that the underlying physics or chemistry of the problem being modelled is fairly well understood and that there is no fundamental misunderstanding of the problem prior to investigation. However, there are situations where an investigation may provide surprising information, calling for a revised conceptual model of the problem.287 Historic examples include collapses in fish stocks, the effects of CFCs on the ozone layer, and the harm to health caused by X-rays and asbestos.288 One problem is that many different models may be consistent with the available data.289 Therefore even perfectly calibrated models (e.g. those that fit the data at a particular site) may have limited or no predictive value (i.e. they may not adequately represent the necessary processes as conditions change with time).290 Similarly, models that work well in the laboratory may not apply to real-world conditions. For example, the advection-diffusion equation is used to predict the transport of solutes in soils. However, it neglects the possibility of preferential fast transport routes, particularly on colloids, and therefore failed to predict the unexpected pollution of steams and groundwaters with pesticides and other contaminants.291 Another problem is the difficulty in finding a parameter set that adequately represents a given location, because places are unique in their characteristics and boundary conditions and their uniqueness is inevitably to some extent unknowable.292 This means that a model that has been refined to be fit for purpose at one location will not necessarily work at another, or in different future circumstances, if the parameters used to define the new site or circumstances are inadequate to represent important processes. Theoretically, it should be possible to take a pragmatic approach which would allow researchers to consider all the possible models that might fit the data and, by hypothesis testing using experimental data, rule out those models that breach safety requirements.293 However, most safety assessment programmes remain wedded to the idea that there is a single best fit model, rather than focusing on exploring possible alternative models of the site, some of which may show the proposed repository to breach safety requirements. Further, it is by no means clear that sufficient data can be collected, or sufficiently safe sites exist, to rule out scenarios which involve significant radiological releases.

GeneWatch UK consultancy report September 2010

41

5.1.2. Potential for bias in the assessment process Scientific bias has been well studied in the medical research literature, where several types of interpretative bias (bias in the analysis of data, rather than in the measurements themselves) have been identified:294

l confirmation bias evaluating evidence that supports the scientists preconceptions

differently from evidence that challenges these convictions

l rescue bias discounting data by finding selective faults in the experiment in order to
rescue the original hypothesis

l mechanism bias being less sceptical when underlying science furnishes credibility for the
data, meaning that the interpretation of results is in line with prior expectations

l time will tell bias the phenomenon whereby different scientists need different amounts of
confirmatory evidence, because deciding when evidence is sufficient to make a decision is inevitably subjective

l orientation bias the possibility that the hypothesis itself introduces prejudices and errors
and becomes a determinate of experimental outcomes. In the field of deep disposal, the likelihood of interpretative bias is high and the potential safety implications considerable, because the wastes involved remain highly dangerous for thousands to millions of years and there is no mechanism to validate computer model predictions over the long timescales involved. In systems whose properties are spatially and temporally heterogeneous (variable) at different scales the concept of the observer as an impartial, totally unbiased bystander becomes meaningless.295 Models of environmental systems, including radioactive waste disposal, involve numerous subjective choices about system structure, boundary conditions, feasible values for parameters, characterisation of input data, scenarios for future predictions and how the performance of the model should be evaluated.296 Environmental models are therefore mathematically ill-posed or ill-conditioned, meaning that the information content available to define a modelling problem does not allow a single mathematical solution.297 Failure to recognise this can easily lead to overconfidence in a particular computer model or the assumptions that underpin it. It is clear from historical and contemporary examples drawn from many fields a recent example being the credit crunch of 2007 that highly expert regulators and private risk modellers sometimes exhibit herd behaviour and may fail to anticipate rare and unexpected events. Such dangers are greatest when the discussion of the issues and computer model-building are highly complex and are comprehended only by a highly expert group, because they are then less 298 likely to be open to public scrutiny or challenge by outsiders. Numerous articles in the medical literature have also found that bias is strongly influenced by commercial interests.299,300,301,302,303,304 This suggests that the selection of a particular computer model and set of parameters may be not only subjective, but also easily biased towards giving the preferred outcomes. Availability of alternative expertise and funding can therefore strongly influence whether there are sufficient critical perspectives to identify problems with the safety case for a radioactive waste repository. For example, at the UK Nirex planning inquiry, the objecting groups had a total budget one hundredth that of Nirex but nevertheless succeeded in demonstrating significant problems with the safety case by involving sufficient alternative expertise. Nirex produced seven expert witnesses to discuss the technical, geoscience and engineering issues: all except one were directly employed by Nirex.305 For the objectors, Cumbria County Council produced five experts (one employed by it), focusing on site selection, hydrogeology and overall risk. Greenpeace presented five experts (one employed by it) to discuss site selection, geology, hydrogeology, flow modelling, geochemistry, and comparable investigations worldwide. Friends of the Earth fielded nine experts (two employed by it) to tackle government policy, geology, site 3-D structure, hydrogeology, fluid flow in fractures, engineering and geochemistry. In one example of the expert evidence presented, alternative groundwater flow modelling by researchers at the University of Glasgow, funded by the Greenpeace Environmental

42

GeneWatch UK consultancy report September 2010

Trust, suggested that much faster groundwater return flow times than those calculated by Nirex were more consistent with its borehole measurements,306 implying that Nirexs risk calculations might be two orders of magnitude in error.307 Similarly, much of the work highlighting concerns about the potential for copper to corrode in water (see Section 4.1.2) has been unfunded. The official research programme did not identify this problem. These examples suggest that other problems may remain unidentified due to lack of sufficient independent scrutiny. Bias can be exacerbated by claims that deep disposal must be workable because road maps towards its implementation exist in a number of countries, significant amounts of research have been done, and other alternatives have been discarded as technically or economically unfeasible or unsafe.308,309 In Finland, a Posiva researcher speaking anonymously to the Finnish Broadcasting Company in May 2010 expressed concerns about pressure on scientists to meet the schedule for approval of the Olkiluoto repository despite doubts about the reliability of the copper canisters, bentonite backfill and tunnel backfill. The researcher reportedly said: The results of research are decided beforehand. Then we find data that gives the desired result. If there is information that does not back up the result, it is ignored..310 It seems likely therefore that there could exist other serious problems with deep repository proposals, which have not been identified due to lack of resources and funds for independent scrutiny of data and assumptions. In each country with a deep disposal programme, regulators are responsible for reviewing safety cases and ultimately for licensing facilities.311 Although this can include some independent research and development to support decision-making, regulators are in practice largely dependent on the data collection, analysis and computer modelling produced by the nuclear waste disposal companies. The majority of the funding for research, development and demonstration (RD&D) in waste management comes from the nuclear industry and follows the research agenda set by the industrys implementing organisations.312 Reliance on industry-funded research, although consistent with the principle that the polluter pays, is likely to introduce interpretative bias in repository safety assessments. On the other hand, significant sums of public money invested via Euratom are not being used to fund independent scrutiny. It is of particular concern that Europes IGD-TP states that it is open only to stakeholders endorsing the vision and willing to contribute positively and constructively to the objectives and goals of the platform in other words, critics of deep disposal are supposed to be excluded from the research programme and hence from Euratom funding.313 Greenpeace has now joined the IGD-TP but only on condition that it is not required to subscribe to the vision. Members of the Executive Group consist of organisations either responsible for implementing a waste management programme or formally responsible for the RD&D programme needed for implementation. It is difficult to see how adequate independent scrutiny of data and assumptions can take place in such circumstances.

5.2. Site selection, public opinion and radioactive waste inventories

Sweden has involved local communities in the decision-making process and given them a veto at each stage of the site selection process for a deep repository. Following the example set by Sweden and the past failures of site selection processes in many countries, there has been a shift in most countries since the 1980s away from finding the best geological site for disposal towards finding a site that is considered good enough and where repository construction is considered politically as well as technically achievable. The site selection process then takes more account of other factors, particularly acceptability to the local population and proximity to existing nuclear facilities.314 The UK and Canada have been particularly active in attempting to be seen to follow the Swedish approach by introducing new public participation and consultation programmes for nuclear waste disposal decisions.315

GeneWatch UK consultancy report September 2010

43

The Canadian nuclear waste management programme settled on deep disposal as a solution by the mid-1970s, initially projecting site selection by the mid-1980s, construction of a repository by the late 1980s, and an operating repository by 2000. Intense public opposition thwarted this programme, which failed to gain popular acceptance at a public inquiry, but a new process of site selection is currently under way.316 It remains to be seen whether technical problems and scientific uncertainties are truly open to public scrutiny, or whether official claims that deep disposal has been found to be safe from a technical perspective will remain unchallenged. In the UK, proximity to existing nuclear facilities previously led to a focus of investigations on the suitability of Sellafield as a site for the planned Nirex intermediate-level waste repository, which was rejected in 1997. The new approach therefore differs little on one level in that the same area has now been shortlisted again; however, this time people living near the final site are expected to be offered compensation. On the other hand, planning law has been changed so that the scientific evidence cannot be cross-examined (see Box 6). Site selection processes based on volunteerism typically now involve some form of financial compensation for the local population and perhaps other benefits, such as employment and new roads or other infrastructure. For example, in Slovenia two communities close to the countrys only nuclear power plant competed for the financial compensation available for hosting a repository for low- and intermediate-level waste.317 However, a voluntarist approach to site selection for a deep geological repository presumes that a number of sites that are both geologically suitable and publicly acceptable exist, and that safety will not be compromised by offering financial incentives to poor or marginalised communities. In practice, offering financial compensation risks undermining the requirement to optimise radiological protection (i.e. to use the best available techniques to minimise radiation exposures in the future). Further, as the European Commissions JRC acknowledges, a suitable site might simply not exist in a given country seeking to implement the deep disposal option.318 A study of the siting of a low-level waste repository in South Korea identified four factors that influenced local acceptance: perceived economic benefit, risk perception (which has strong negative effects), trust and perceived competition for the facility.319 In Sweden, one study has suggested that people in the two municipalities shortlisted for a spent nuclear fuel repository have less precautionary attitudes to risk than the general population.320 There is thus a danger that concerns about repository safety and impacts on future generations may be sidelined in communities which volunteer to host a repository, especially if they are economically dependent on the compensation, infrastructure or jobs offered to them. A recent survey of public attitudes in Finland provides strong evidence that residents in the municipality of Eurajoki, where the Olkiluoto disposal site is to be situated, perceive a threat to the safety, health and well-being of future generations from the planned repository.321 The site was shortlisted when the use of purely geological criteria had been abandoned, and was inserted onto a list of 101 potential sites as an exception, based on the short transport distance that would be required for the wastes already stored at the nearby nuclear reactors. The survey, based on 606 responses which qualified for analysis, found that 63% expected a positive effect on employment and economic development in the area, but that the facility was widely expected to have a harmful impact on rural non-farm livelihoods (fishing, hunting, forest product gathering etc.), the state of the environment near the facility and the image of the area (to outsiders). Nearly 60% of respondents agreed with the statement Nuclear waste poses a continuous threat to the lives of future generations and only 23% disagreed with it. A majority felt that the repository posed a threat to the safety of future generations (55%), the health of future generations (55%) and the well-being of future generations (52%). The information provided by the survey is important because this is the first municipality in the world where the views of local residents have been able to be studied following a site selection decision for a deep underground repository for spent nuclear fuel. The survey suggests that, far from being convinced about the long-term safety of the proposed facility, residents have reservations and a high level of concern about future generations. However, these reservations have not led to the community vetoing the site, thanks to a package of economic benefits negotiated between the municipality and the nuclear industry in 19992000.322

44

GeneWatch UK consultancy report September 2010

This tension between long-term safety and short-term economic incentives has also been seen in the UK, where compensation is now being offered to communities that volunteer to host a repository. Selection of the Sellafield site in 1991 for a repository for long-lived intermediate-level waste involved a process which gave zero weighting to safety criteria on the grounds that all short-listed sites could meet regulatory requirements. However, the chosen site met none of the geological criteria or guidelines that had ever been developed to identify appropriate sites.323 The rejection of planning permission for a Rock Characterisation Facility (the first phase of the planned repository) at this site in 1997 was the third time the then UK disposal company Nirex had its plans rejected.324 A key issue at the inquiry was the site selection process and Nirexs failure to optimise radiological protection using best available techniques. The site was chosen for non-scientific reasons, in a decision-making process which concealed its true geological problems, leading geologists who gave evidence against the plans to conclude that the planning inspectors comprehensive dismissal of the site would make it hard to return to it.325,326 Over a decade later, a new planning system has now been adopted to facilitate the construction of new nuclear reactors and the associated nuclear waste repositories.327 New geological criteria which do not exclude the Sellafield area have been developed, despite continued concerns that the area is geologically unsuitable.328,329 Volunteer communities are being sought for a repository, to receive high-level as well as intermediate-level wastes from past and current reprocessing, plus spent nuclear fuel from new nuclear reactors, and three communities near Sellafield have expressed an interest.330,331 The new planning system thus appears designed to allow construction of a repository to proceed at or near the previously rejected site. It remains to be seen whether this process is successful at building public confidence. In several countries, including Canada, the UK, Sweden and Finland, the difficulties of implementing deep disposal have been exacerbated by government decisions to build new nuclear reactors, threatening to create new wastes before a solution has been agreed or implemented for existing wastes. Rather than reiterating the conclusion of the 1976 Flowers report that new reactors should not be constructed in the absence of a safe means of containing the wastes, the UK Government has adopted an active programme of new reactor construction, claiming that there is now a consensus that deep geological disposal is safe. In a sign of tension, the UK Committee on Radioactive Waste Management has clearly stated that its conclusions and recommendations regarding deep disposal are intended to apply only to committed wastes, not to wastes generated by new nuclear power stations.332 Issues include the ethical concerns associated with producing new wastes before a solution has been demonstrated, and the increased difficulty of finding a suitable volume of rock. In Finland, where a new reactor is currently under construction, the nuclear waste company Posiva has rejected the proposal that it dispose of waste from the new reactor, which is owned and managed by a different company.333 The construction of new reactors will increase not only the volume of wastes to be disposed of but also the average level of radioactivity per rod of spent nuclear fuel, since next-generation reactors are likely to use higher burn-up fuel. This may have implications for repository safety cases.334 Thus, in addition to the tension between the economic benefits being offered to host communities and long-term repository safety, there is a tension between endorsement of deep disposal as a potentially least bad option for existing wastes, if the scientific and technical difficulties can be resolved at some point in the future, and nuclear industry claims that deep repositories provide a safe solution which will allow the sustainable expansion of the industry. Yet there is little public support for the idea that the problem of high-level nuclear waste has been dealt with in the sense that it can now be got rid of safely. According to a 2008 Eurobarometer survey, in Greece, Sweden, France, Germany and Finland around 80% of respondents totally or tend to agree that there is no safe way of getting rid of high-level radioactive waste.335 Of EU residents as a whole, 41% totally agreed that there is no safe way of getting rid of such waste, while under a third (31%) tended to agree. Only 14% disagreed and a similar percentage did not know or had no opinion on the issue. The idea that there is no safe way of getting rid of high-level waste had slightly more support in Finland in 2008 than in 2005, while Cypriot, Lithuanian, Hungarian, Latvian and Dutch respondents seemed to have become more convinced by the opposite statement, i.e. that there is a safe way of getting rid of it.

GeneWatch UK consultancy report September 2010

45

The 2008 Eurobarometer survey also found that public opinion was rather divided across the EU over deep underground disposal of high-level waste. In Finland, Sweden and Hungary this idea received more support than anywhere else in the 27-member EU. Majorities in Luxembourg and Belgium did not agree that deep underground disposal represents the most appropriate solution and the largest single response in France, Poland, Italy and Latvia was also disagreement. In some countries very high proportions of citizens answered that they did not know whether deep underground disposal is the best solution. The respondents were asked which things would worry them the most if a disposal site for radioactive waste was built in the area where they live. The main issues of concern were the possible effects on the environment and health (51%) and the risk of radioactive leaks (30%). Of all those surveyed, eight out of ten responded that one of these two issues would worry them the most. EU residents would clearly want to be directly consulted and would like to participate in the decision-making process, should this hypothetical situation take place well over half of respondents (56%) indicated that they would want to be personally involved. Just over one in five (22%) indicated that they would prefer local non-governmental organisations to participate in the decision-making process, while 15% felt that they would rather let the responsible authorities decide. The IGDTP Vision Document states that it is essential to develop dialogue with the general public so as to share the extensive scientific and engineering work underpinning the conclusions made by the OECD NEA that geological disposal is technically feasible and provides a unique level and duration of protection. This issue is being dealt with in the Forum on Stakeholder Confidence in the Radioactive Waste Management Committee of the NEA336. This suggests that, rather than genuinely seeking to address scientific and technical concerns, the nuclear industry and advocates of new nuclear reactors, such as Euratom and the NEA, are actively engaged in a public relations exercise focused on the claim that no major issues remain to be resolved. One example of such advocacy is a web-based communication system funded by the Japan Nuclear Safety Organisation to seek social consensus on high-level waste disposal.337

5.3. Costs
The global market for nuclear decommissioning and clean-up is estimated at 300 billion (360 billion) over the next 30 years.338 The costs of deep geological disposal are significant: for example, South Korea has estimated the cost of its proposed spent fuel repository as 2.6 billion euros.339 The cost of the copper canisters is one of the key components of the cost of a nuclear waste repository built according to the Swedish concept. In South Korea, 14,210 canisters will be needed to dispose of spent fuel consisting of 36,000 tonnes of uranium from the two existing reactor types (11,375 pressurised water reactor (PWR) canisters and 2,835 CANDU canisters). As part of a costestimation exercise in Korea, the cost of a CANDU canister consisting of a 5cm copper outer shell with a cast iron insert was calculated at 171,415 and the cost of a PWR copper canister produced using the cheapest method at 156,776 (2006 prices). In these calculations, the material cost was about 43% of a canisters total manufacturing cost (the rest being mainly labour costs), and the manufacturing cost of the canisters represented about 32% of the total disposal costs.340 South Korea accordingly plans to use a thinner (1cm rather than 5cm) copper canister to reduce costs;341 however, this will impact on containment and hence on safety. There is uncertainty regarding the future costs of both the main materials needed to implement the Swedish deep disposal concept: copper powder for the canisters and bentonite for the backfill.342 The repository layout will also influence costs due to the cost of constructing and backfilling the tunnels and the costs of the bentonite needed for the disposal holes.343 For example, placing several spent fuel canisters in long horizontal disposal drifts is cheaper than excavating individual vertical disposal holes accessed via tunnels. However, this option is more sensitive to the site geology because a single large fracture zone in a long disposal drift could destroy the whole drift.344 The design of a repository in fractured rock may need to be optimised to minimise the number of locations where water-conducting fractures are intersected.345 Some of the concerns highlighted in the literature review above could be mitigated by changes to the repository design. However, major changes would have significant impacts on projected costs.

46

GeneWatch UK consultancy report September 2010

Examples of proposed changes identified in the literature search include:

l thicker copper canisters l wider spacing between canisters (to reduce the adverse impacts of high temperatures on
bentonite, or to seek to avoid water-conducting fracture zones)

l purer bentonite (to limit mineral changes with heat) l increased excavation depth (to limit gas bubbles through higher pressure, increase
groundwater flow return times, give greater protection from glaciation and reduce microbial activity). All of the above would increase costs significantly. Increasing depth would also increase the risk of rock bursts (localised earthquakes) due to the high pressures at depth.

GeneWatch UK consultancy report September 2010

47

6. Conclusion

A scientific consensus on deep disposal?

The European Commission Joint Research Centres report bases its claim that there is a scientific consensus on the deep disposal of high-level radioactive wastes on the existence of road maps towards implementing this option in Finland and Sweden.346 The Implementing Geological Disposal of Radioactive Waste Technology Platform states that the recommendation of the OECD Nuclear Energy Agencys Radioactive Waste Management Committee is based on work over several decades by the international scientific and technical community in which alternatives such as launching nuclear waste into space, ocean dumping, disposal under continental glaciers, sub-seabed disposal and long-term supervised storage were carefully studied and discarded.347 However, the existence of road maps and the rejection of other options do not automatically mean that deep disposal of highly radioactive wastes is safe. On the contrary, the present reports review of papers published in peer-reviewed scientific journals has identified a number of scenarios in which a significant release of radioactivity could occur, with serious implications for the health and safety of future generations. The following phenomena could compromise containment in a deep repository:

l Copper or steel canisters and overpacks containing spent nuclear fuel or high-level
radioactive wastes could corrode more quickly than expected.

l The effects of intense heat generated by radioactive decay, and of chemical and physical
disturbance due to corrosion, gas generation and biomineralisation, could impair the ability of backfill material to trap some radionuclides.

l Build-up of gas pressure in the repository, as a result of the corrosion of metals and/or the
degradation of organic material, could damage the barriers and force fast routes for radionuclide escape through crystalline rock fractures or clay rock pores.

l Poorly understood chemical effects, such as the formation of colloids, could speed up the
transport of some of the more radiotoxic elements such as plutonium.

l Unidentified fractures and faults, or poor understanding of how water and gas will flow through
fractures and faults, could lead to the release of radionuclides in groundwater much faster than expected.

l Excavation of the repository will damage adjacent zones of rock and could thereby create fast
routes for radionuclide escape.

l Future generations, seeking underground resources or storage facilities, might accidentally

dig a shaft into the rock around the repository or a well into contaminated groundwater above it.

l Future glaciations could cause faulting of the rock, rupture of containers and penetration of
surface waters or permafrost to the repository depth, leading to failure of the barriers and faster dissolution of the waste.

l Earthquakes could damage containers, backfill and the rock.

48

GeneWatch UK consultancy report September 2010

References

1 2 3 4

European Commission. 2009. Implementing Geological Disposal of Radioactive Waste Technology Platform: Vision Document. October 2009. http://www.igdtp.eu/Documents/VisionDoc_Final_Oct24.pdf Rogers, K.A. 2009. Fire in the hole: A review of national spent nuclear fuel disposal policy. Progress in Nuclear Energy 51: 281-289. European Commission. 2009. Implementing Geological Disposal of Radioactive Waste Technology Platform: Vision Document. October 2009. http://www.igdtp.eu/Documents/VisionDoc_Final_Oct24.pdf Falck, W.E., Nilsson, K. F. 2009. Geological disposal of radioactive waste. European Commission Joint Research Centre. JRC Reference Report EUR 23925 EN. http://ie.jrc.ec.europa.eu/publications/scientific_publications/2009/LRJRC_Reference_Report_IE_Geological%20Disposal.pdf European Commission. 2009. Implementing Geological Disposal of Radioactive Waste Technology Platform: Vision Document. October 2009. http://www.igdtp.eu/Documents/VisionDoc_Final_Oct24.pdf OECD/NEA, 2008. Moving forward with geological disposal of radioactive waste: An NEA RWMC collective statement. NEA/RWM(2008)5/REV2. http://www.nea.fr/html/rwm/docs/2008/rwm2008-5rev2.pdf European Nuclear Energy Forum. 2008. Developing a roadmap to comprehensive long term radioactive waste management in the EU. Memo from Working Group Risk, 23rd January 2008, Brussels. http://ec.europa.eu/energy/nuclear/forum/bratislava_prague/working_groups/risks/radio_waste_en.pdf Falck, W.E., Nilsson, K. F. 2009. Geological disposal of radioactive waste. European Commission Joint Research Centre. JRC Reference Report EUR 23925 EN. http://ie.jrc.ec.europa.eu/publications/scientific_publications/2009/LRJRC_Reference_Report_IE_Geological%20Disposal.pdf International Atomic Energy Agency (IAEA). 2010. Power Reactor Information System. Accessed 07. June, 2010. Available at: http://nucleus.iaea.org/sso/NUCLEUS.html?exturl=http://www.iaea.or.at/programmes/a2/ International Atomic Energy Agency (IAEA). 2010. Power Reactor Information System. Accessed 07.June, 2010. Available at: http://www.iaea.org/programmes/a2/index.html International Atomic Energy Agency (IAEA). 2010. Power Reactor Information System. Accessed 07.June, 2010. Available at: http://www.iaea.org/programmes/a2/index.html US Energy Information Administration. 2009. Nuclear consumption forecast. http://www.eia.doe.gov/oiaf/ieo/pdf/ieoecgtab_h5.pdf. Accessed: 13 June 2010. IAEA. 2008. Estimation of global inventories of radioactive waste and other radioactive materials. IAEATECDOC-1591. http://www-pub.iaea.org/MTCD/publications/PDF/te_1591_web.pdf Resnikoff, M., Travers, J., Alexandrova, E. 2010. The hazards of generation III reactor fuel wastes. Greenpeace Canada. May 2010. Chandler J., Hertel, N. 2009. Choosing a reprocessing technology requires focusing on what we value. Progress in Nuclear Energy 51: 701-708. Hgselius, P. 2009. Spent nuclear fuel policies in historical perspective: An international comparison. Energy Policy 37: 254-263. Beken T.V., Dorn, N., Daele, S.V. 2010. Security risks in nuclear waste management: Exceptionalism, opaqueness and vulnerability. Journal of Environmental Management 91: 940-948. Hgselius, P. 2009. Spent nuclear fuel policies in historical perspective: An international comparison. Energy Policy 37: 254-263. IAEA. 2008. Estimation of global inventories of radioactive waste and other radioactive materials. IAEATECDOC-1591. http://www-pub.iaea.org/MTCD/publications/PDF/te_1591_web.pdf

5 6

10 11 12 13 14 15 16 17 18 19

GeneWatch UK consultancy report September 2010

49

20 21 22 23

Hgselius, P. 2009. Spent nuclear fuel policies in historical perspective: An international comparison. Energy Policy 37: 254-263. Gleizon, P., McDonald, P. 2010. Modelling radioactivity in the Irish Sea: From discharge to dose. Journal of Environmental Radioactivity 101: 403-413. OSPAR Commission. Radioactive Substances. Available on: http://www.ospar.org/content/content.asp?menu=00220306000000_000000_000000 OSPAR Decision 2000/1 on Substantial Reductions and Elimination of Discharges, Emissions and Losses of Radioactive Substances, with Special Emphasis on Nuclear Reprocessing. Copenhagen, 2000. DECC (2009) UK strategy for radioactive discharges. July 2009. http://www.decc.gov.uk/assets/decc/What%20we%20do/UK%20energy%20supply/Energy%20mix/Nucl ear/radioactivity/1_20090722135916_e_@@_dischargesstrategy.pdf Rogers KA 2009. Fire in the hole: A review of national spent nuclear fuel disposal policy. Progress in Nuclear Energy 51: 281-289. IAEA. 2007. Operation and Maintenance of Spent Fuel Storage and Transportation Casks/Containers. pp.1 IAEA-TECDOC-1532. Available at http://wwwpub.iaea.org/MTCD/publications/PDF/te_1532_web.pdf Rogers KA 2009. Fire in the hole: A review of national spent nuclear fuel disposal policy. Progress in Nuclear Energy 51: 281-289. European Commission. 2009. Implementing Geological Disposal of Radioactive Waste Technology Platform: Vision Document. October 2009. http://www.igdtp.eu/Documents/VisionDoc_Final_Oct24.pdf DEFRA/NDA. 2007. Radioactive materials not reported in the UK radioactive waste inventory. Department of Environment, Food and Rural Affairs (DEFRA) and the Nuclear Decommissioning Authority (NDA). March 2008. Lee, C.-M. 2010. Assessment of radiological safety of Wolsung site at site boundary considering crack impact. Progress in Nuclear Energy 52: 374-378. Rogers KA 2009. Fire in the hole: A review of national spent nuclear fuel disposal policy. Progress in Nuclear Energy 51: 281-289. US Environmental Protection Agency. Commonly encountered radionuclides. http://www.epa.gov/radiation/radionuclides/index.html Report of the Committee Examining Radiation Risks of Internal Emitters (CERRIE) 2004. www.cerrie.org www.icrp.org Altman, S. 2008. Geochemical research: A key building block for nuclear waste disposal safety cases. Journal of Contaminant Hydrology 102 (3-4): 174-179. Gleizon, P., McDonald, P. 2010. Modelling radioactivity in the Irish Sea: From discharge to dose. Journal of Environmental Radioactivity 101: 403-413. UK Nirex Ltd. 2006. C-14: How we are addressing the issues. February 2006. No. 498808. Bomboni, E., Cerullo, N., Lomonaco, G. 2009. Assessment of LWR-HTR-GCFR Integrated Cycle. Science and Technology of Nuclear Installations 2009: Article ID 193594, 14 pages. doi:10.1155/2009/193594. Available on: http://www.hindawi.com/journals/stni/2009/193594.cta.html Bomboni, E., Cerullo, N., Lomonaco, G. 2009. Assessment of LWR-HTR-GCFR Integrated Cycle. Science and Technology of Nuclear Installations 2009: Article ID 193594, 14 pages. doi:10.1155/2009/193594. Available on: http://www.hindawi.com/journals/stni/2009/193594.cta.html Royal Commission on Environmental Pollution Sixth Report (1976); Chairman Sir Brian (now Lord) Flowers: Nuclear Power and the Environment. Solomon, B.D., Andrn, M. , Strandberg, U. 2009. Thirty years of social science research on high-level nuclear waste. Conference on Managing Radioactive Waste: Problems and Challenges in a Globalized World. University of Gothenburg, Sweden, December 15-17, 2009. http://www.cefos.gu.se/digitalAssets/1291/1291675_Solomon__paper_.pdf Solomon, B.D., Andrn, M. , Strandberg, U. 2009. Thirty years of social science research on high-level nuclear waste. Conference on Managing Radioactive Waste: Problems and Challenges in a Globalized

24

25 26

27 28 29

30 31 32 33 34 35 36 37 38

39

40 41

42

50

GeneWatch UK consultancy report September 2010

World. University of Gothenburg, Sweden, December 15-17, 2009. http://www.cefos.gu.se/digitalAssets/1291/1291675_Solomon__paper_.pdf 43 44 45 46 47 48 49 50 See IAEA http://www-ns.iaea.org/conventions/waste-jointconvention.htm IAEA. 2004. Siting of Geological Disposal Facilities Safety Series No. 111-G-4.1. http://wwwpub.iaea.org/MTCD/publications/PDF/Pub952e_web.pdf AEA Safety Standards. 2006. Geological Disposal of Radioactive Waste. Safety Requirements, No. WS-R-4, 2006. http://www-pub.iaea.org/MTCD/publications/PDF/Pub1231_web.pdf OECD/NEA, 2004. Post-closure Safety Case for Geological Repositories. OECD, NEA, Paris www.nea.fr/ html/rwm/reports/2004/nea3679-closure.pdf Wrixon, AD. 2008. New ICRP recommendations. Journal of Radiological Protection 28: 161-168. Lindell, B. 2000. [Editorial]. On collective dose. Journal of Radiological Protection 20(1):1. Wrixon, AD. 2008. New ICRP recommendations. Journal of Radiological Protection 28: 161-168. Larsson, C-M. 2009. Waste disposal and the recommendations of the International Commission on Radiological Protection Challenges for radioecology and environmental radiation protection. Journal of Environmental Radioactivity 100: 1053-1057. OECD-Nuclear Energy Agency. 2007. The long-term regulatory criteria for radioactive waste disposal: Towards a common understanding of the objectives, challenges and practical issues. NEA/RWMC/RF(200)1/ROV. OECD-NEA, Paris. http://www.nea.fr/rwm/reports/2007/nea6182regulating.pdf Falck, W.E., Nilsson, K. F. 2009. Geological disposal of radioactive waste. European Commission Joint Research Centre. JRC Reference Report EUR 23925 EN. http://ie.jrc.ec.europa.eu/publications/scientific_publications/2009/LRJRC_Reference_Report_IE_Geological%20Disposal.pdf Pentreath, R.J. 2009. Radioecology, radiobiology, and radiological protection: frameworks and fractures. Journal of Environmental Radioactivity 100: 1019-1026. Oreskes, N., Schrader-Frechette, K., Belitz. K. 1994. Verification, validation and confirmation of numerical models in the Earth sciences. Science 263: 641-646. Beven, K. 2002. Towards a coherent philosophy for modelling the environment. Proceedings of the Royal Society of London 458: 2465-2484. Tsang, C. -F., Stephansson, O., Hudson, J.A. 2000. A discussion of thermo-hydro-mechanical (THM) processes associated with nuclear waste repositories. International Journal of Rock Mechanics and Mining Sciences 37(1-2):397-402. Mulligan, C.N., Yong, R.N., Fukue, M. 2009. Some effects of microbial activity on the evolution of claybased buffer properties in underground repositories. Applied Clay Science 42: 331-335. Montarnal, Ph., Mugler, C., Descostes, M., Dimier, A., Jacquot, E. Presentation and use of a reactive transport code in porous media. Physics and Chemistry of the Earth 32: 507-517. Metz, V., Kienzler, B., Schssler, W. 2003. Geochemical evaluation of different groundwater-host rock systems for radioactive waste disposal. Journal of Contaminant Hydrology 61: 265-279. Tsang, C. -F., Stephansson, O., Hudson, J.A. 2000. A discussion of thermo-hydro-mechanical (THM) processes associated with nuclear waste repositories. International Journal of Rock Mechanics and Mining Sciences 37(1-2):397-402. Protopopoff, E., Marcus, P. 2005. Potential-pH diagrams for hydroxyl and hydrogen adsorbed on a copper surface. Electrochimica Acta 51: 408-417. Takeno, N. 2005. Atlas of Eh-pH diagrams. Geological Survey of Japan Open File Report No. 419. http://www.gsj.jp/GDB/openfile/files/no0419/openfile419e.pdf Pentreath, R.J. 2009. Radioecology, radiobiology, and radiological protection: frameworks and fractures. Journal of Environmental Radioactivity 100: 1019-1026. Report of the Committee Examining Radiation Risks of Internal Emitters (CERRIE) 2004. www.cerrie.org Fyfe, W.S.. 1999. Nuclear waste isolation: an urgent international responsibility. Engineering Geology 52: 159-161.

51

52

53 54 55 56

57 58 59 60

61 62 63 64 65

GeneWatch UK consultancy report September 2010

51

66 67 68 69 70 71

Resnikoff, M., Travers, J., Alexandrova, E. 2010. The hazards of generation III reactor fuel wastes. Greenpeace Canada. May 2010. Rogers, K.A. 2009. Fire in the hole: A review of national spent nuclear fuel disposal policy. Progress in Nuclear Energy 51: 281-289. Haszeldine, S., Smythe, D. 1997. Why was Sellafield rejected as a disposal site for radioactive waste? Geoscientist 7(7): 18-20. McDonald, C.S., Jarvis, C., Knipe, C.V. 1996. RCF planning appeal by UK Nirex Ltd. Report No. APP/HO900/A/94/247019, DOE. McDonald, C. 2007. Letter: Flaws in search for nuclear waste site. The Guardian. 28th June 2007. http://www.guardian.co.uk/world/2007/jun/28/nuclear.uk House of Lords Science and Technology Committee. 1999. Management of Nuclear Waste. 24 March 1999. 3rd Report of Session 1998-99 HL 41 ISBN 0 10 404199 4 http://www.publications.parliament.uk/pa/ld199899/ldselect/ldsctech/41/4109.htm#n51 Irving, A. 1999. Gosforth will fight dump. Whitehaven News. 8th April 1999. HM Government. 2009. Infrastructure planning: How will it work? Can I have my say? http://www.communities.gov.uk/documents/planningandbuilding/pdf/infrastructureplanningwork.pdf Asse II. Website on: http://www.endlager-asse.de/cln_094/EN/1_Home/home_node.html European Commission. 2009. Implementing Geological Disposal of Radioactive Waste Technology Platform: Vision Document. October 2009. http://www.igdtp.eu/Documents/VisionDoc_Final_Oct24.pdf Bundesministerium fr Umwelt, Naturschutz und Reaktorsicherheit (BUMF). 2010. Bundesumweltminister Norbert Rttgen Wir mssen uns der Verantwortung fr die Entsorgung radioaktiver Abflle endlich stellen, Pressemitteilung Nr. 037/10, Berlin, 15.03.2010. Available at http://www.bmu.de/pressemitteilungen/aktuelle_pressemitteilungen/pm/45767.php Rogers, K.A. 2009. Fire in the hole: A review of national spent nuclear fuel disposal policy. Progress in Nuclear Energy 51: 281-289. Pickard, W.F. 2010. Finessing the fuel: Revisiting the challenge of radioactive waste disposal. Energy Policy 38: 709-714. Strm, A., Andersson, J., Skagius, K., Winberg, A. 2008. Site descriptive modelling during characterization for a geological repository for nuclear waste in Sweden. Applied Geochemistry 23: 1747-1760. Altmann, S. 2008. Geochemical research: A key building block for nuclear waste disposal safety cases. Journal of Contaminant Hydrology 102 (3-4): 174-179. Rogers, K.A. 2009. Fire in the hole: A review of national spent nuclear fuel disposal policy. Progress in Nuclear Energy 51: 281-289. http://www.dounreay.com/decommissioning/shaft-and-silo ACRO. 2009. Gestion des dchets radioactifs: les leons du Centre de Stockage de la Manche. Centre sans mmoire, centre sans avenir? Greenpeace France 25th June 2009. http://www.greenpeace.org/raw/content/france/presse/dossiers-documents/rapport-gestion-desdechets-radioactifs.pdf Asse II. Website on: http://www.endlager-asse.de/cln_094/EN/1_Home/home_node.html Rempe, N.T. 2007. Permanent underground repositories for radioactive waste. Progress in Nuclear Energy 49(5): 365-374. Rempe, N.T. 2007. Permanent underground repositories for radioactive waste. Progress in Nuclear Energy 49(5): 365-374. Tsang, C. -F., Stephansson, O., Hudson, J.A. 2000. A discussion of thermo-hydro-mechanical (THM) processes associated with nuclear waste repositories. International Journal of Rock Mechanics and Mining Sciences 37(1-2):397-402. Nuclear Waste Advisory Associates. 2010. NWAA Issues Register. March 2010. www.nuclearwasteadvisory.co.uk King, F., Kolar, M., Kessler, J.H., Apted, M. 2008. Yucca Mountain engineered barrier system corrosion model (EBSCOM). Journal of Nuclear Materials 379: 59-67.

72 73 74 75 76

77 78 79

80 81 82 83

84 85 86 87

88 89

52

GeneWatch UK consultancy report September 2010

90 91

Qin, Z., Shoesmith, D.W. 2008. Failure model and Monte Carlo simulations for titanium (grade-7) drip shields under Yucca Mountain repository conditions. Journal of Nuclear Materials 379: 169-173. Hultquist, G., Szaklos, P., Graham, M.J., Belonoshko, A.B., Sproule, G.I., Grsjo, L., Dorogokupets, P., Danilov, B., AAstrup, T., Wikmark, G., Chuah, G.-K., Eriksson, J.,-C., Rosengren, A. 2009. Water corrodes copper. Catalysis Letters 132: 311-316. Johnson, L., King, F. 2008. The effect of the evolution of environmental conditions on the corrosion evolutionary path in a repository for spent fuel and high-level waste in Opalinus Clay. Journal of Nuclear Materials 379: 9-15. Puig, F., Dies, J., de Pablo, J., Martnez-Espara, A. 2008. Spent fuel canister for geological repository: Inner material requirements and candidates evaluation. Journal of Nuclear Materials 376: 181-191. Graham, J., Halayko, K.,G., Hume, H., Kirkham, T., Gray, M., Oscarson, D. 2002. A capillarity-advective model for gas break-through in clays. Engineering Geology 64: 273-286. King, F., Kolar, M., Maak, P. 2008. Reactive-transport model for the prediction of the uniform corrosion behaviour of copper used fuel containers. Journal of Nuclear Materials 379: 133-141. Mortley, A., Bonin, H.W., Bui, V.T. 2008. Radiation effects on polymers for coatings on copper canisters used for the containment of radioactive materials. Journal of Nuclear Materials 376: 192-200. Bennett, D.G., Gens, R. 2008. Overview of European concepts for high-level waste and spent fuel disposal with special reference waste container corrosion. Journal of Nuclear Materials 379: 1-8. Taniguchi, N., Kawasaki, M. 2008. Influence of sulphide concentration on the corrosion behaviour of pure copper in synthetic seawater. Journal of Nuclear Materials 379: 154-161. Hwang, Y. 2009. Copper canister lifetime limited by a sulphide intrusion in a deep geologic repository. Progress in Nuclear Energy 51: 695-700. King, F., Kolar, M., Maak, P. 2008. Reactive-transport model for the prediction of the uniform corrosion behaviour of copper used fuel containers. Journal of Nuclear Materials 379: 133-141. Rosborg, B., Werme, L. 2008. The Swedish nuclear waste program and the long-term corrosion behaviour of copper. Journal of Nuclear Materials 379: 142-153. Posiva Oy. 2007. Expected evolution of a spent nuclear fuel repository at Olkiluoto. Revised October 2007. http://www.posiva.fi/files/346/Posiva2006-05_revised_081107web.pdf Hultquist, G. 1986. Hydrogen evolution in corrosion of copper in pure water. Corrosion Science 26: 173177. Hultquist, G., Chuah, G.K., Tan, K.L. 1989. Comments on hydrogen evolution from the corrosion of pure copper. Corrosion Science 29: 1371-1377. Szaklos, P., Hultquist, G., Wikmark, G. 2007. Corrosion of copper by water. Electrochemical and SolidState Letters 10(11): C63-C67. Hultquist, G., Szaklos, P., Graham, M.J., Sproule, G.I., Wikmark, G. 2008. Detection of hydrogen in corrosion of copper in pure water. Presented at the International Corrosion Congress 2008.Paper No. 3384. NACE International, Houston, USA. Hultquist, G., Szaklos, P., Graham, M.J., Belonoshko, A.B., Sproule, G.I., Grsjo, L., Dorogokupets, P., Danilov, B., AAstrup, T., Wikmark, G., Chuah, G.-K., Eriksson, J.,-C., Rosengren, A. 2009. Water corrodes copper. Catalysis Letters 132: 311-316. Johansson, L.-G. 2008. Comment on Corrosion of Copper by Water. Electrochemical and Solid-State Letters 11(4): S1-S1. Werme, L.O., Korzhavyi, C. 2010. Comment on Hultquist et al. Water corrodes copper. Catalysis Letters 135:165-166. Szaklos, P., Hultquist, G., Wikmark, G. 2008. Response to the Comment on Corrosion of Copper by Water. Electrochemical and Solid-State Letters 11(4): S2-S2. Hultquist, G., Szakalos, P., Graham, M.J., Belonoshko, A.B., Rosengren, A. 2010. Reply to Lars O. Werme et. al.: Comments on Water Corrodes Copper. Catalysis Letters 135: 167-168. Protopopoff, E., Marcus, P. 2005. Potential-pH diagrams for hydroxyl and hydrogen adsorbed on a copper surface. Electrochimica Acta 51: 408-417. Swedish National Council for Nuclear Waste (Krnavfallsrdet). 2009. Mechanisms of copper corrosion

92

93 94 95 96 97 98 99 100 101 102 103 104 105 106

107

108 109 110 111 112 113

GeneWatch UK consultancy report September 2010

53

in aqueous environments. A report from the Swedish Council for Nuclear Wastes Scientific Workshop on 16th November 2009. Report 2009: 4e. http://www.karnavfallsradet.se/sites/default/files/dokument/287588_Engelsk_Rapport_2009_4_W.pdf 114 115 116 Merroun, M.L., Selenska-Pobell, S. 2008. Bacterial interactions with uranium: An environmental perspective. Journal of Contaminant Hydrology 102 (3-4): 285-295. Pedersen, K. 1999. Subterranean microorganisms and radioactive waste disposal in Sweden. Engineering Geology 52: 163-176. Stroes-Gascoyne, S. 2010. Microbial occurrence in bentonite-based buffer, backfill and sealing materials from large-scale experiments at AECLs Underground Research Laboratory. Applied Clay Science 47: 36-42. Stroes-Gascoyne, S. 2010. Microbial occurrence in bentonite-based buffer, backfill and sealing materials from large-scale experiments at AECLs Underground Research Laboratory. Applied Clay Science 47: 36-42. Stroes-Gascoyne, S. 2010. Microbial occurrence in bentonite-based buffer, backfill and sealing materials from large-scale experiments at AECLs Underground Research Laboratory. Applied Clay Science 47: 36-42. Masurat, P., Eriksson, S., Pedersen, K. 2010. Evidence of indigenous sulphate-reducing bacteria in commercial Wyoming bentonite. Applied Clay Science 47: 51-57. Mauclaire, L., McKenzie, J.A., Schwyn, B., Bossart, P. 2007. Detection and cultivation of indigenous microorganisms in Mesozoic claystone core samples from the Opalinus Clay Formation (Mont Terri Rock Laboratory). Physics and Chemistry of the Earth 32: 232-240. Masurat, P., Eriksson S., Pedersen K. 2010. Microbial sulphide production in compacted Wyoming bentonite MX-80 under in situ conditions relevant to a repository for high-level radioactive waste. Applied Clay Science 47: 58-64. West, J.M., McKinley, I.G., Neall, F.B., Rochelle, C.A., Bateman, K., Kawamura, H. 2006. Microbiological effects of the Cavern Extended Storage (CES) repository for radioactive waste A quantitative evaluation. Journal of Geochemical Exploration 90: 114-122. Szaklos, P., Hultquist, G., Wikmark, G. 2007. Corrosion of copper by water. Electrochemical and SolidState Letters 10(11): C63-C67. Smart, N.R., Rance, A.P., Werme, L.O. 2008. The effect of radiation on the anaerobic corrosion of steel. Journal of Nuclear Materials 379: 97-104. Bonin, B., Colin, M., Dutfoy, A. 2000. Pressure building during the early stages of gas production in a radioactive waste repository. Journal of Nuclear Materials 281(1): 1-14. Fron, D., Crusset, D., Gras, J-M. 2008. Corrosion issues in nuclear waste disposal. Journal of Nuclear Materials 379: 16-23. Lai-Zhe Jin, L.-Z., Sandstrm, R. 2009. Non-stationary creep simulation with a modified ArmstrongFrederick relation applied to copper canisters. Computational Materials Science 46(2):339-346. Bourg, I.C., Bourg, A.C.M., Sposito, G. 2003. Modeling diffusion and adsorption in compacted bentonite: a critical review. Journal of Contaminant Hydrology 61: 293-302. Tsang, C. -F., Stephansson, O., Hudson, J.A. 2000. A discussion of thermo-hydro-mechanical (THM) processes associated with nuclear waste repositories. International Journal of Rock Mechanics and Mining Sciences 37(1-2):397-402. Tsang, C.-F., Jing, L., Stephansson, O., Kautsky, F. 2005. The DECOVALEX III project: A summary of activities and lessons learned. International Journal of Rock Mechanics and Mining Sciences 42: 593610. Chen, Y., Zhou, C., Jing, L. 2009. Modeling coupled THM processes of geological porous media with multiphase flow: Theory and validation against laboratory and field scale experiments. Computers and Geotechnics 36: 1308-1329. Tsang, C. -F., Stephansson, O., Hudson, J.A. 2000. A discussion of thermo-hydro-mechanical (THM) processes associated with nuclear waste repositories. International Journal of Rock Mechanics and Mining Sciences 37(1-2):397-402. Lai-Zhe Jin, L.-Z., Sandstrm, R. 2009. Non-stationary creep simulation with a modified ArmstrongFrederick relation applied to copper canisters. Computational Materials Science 46(2):339-346.

117

118

119 120

121

122

123 124 125 126 127 128 129

130

131

132

133

54

GeneWatch UK consultancy report September 2010

134 135 136

Lee, J.O., Kang, I.M., Cho, W.J. 2010. Smectite alteration and its influence on the barrier properties of smectite clay for a repository. Applied Clay Science 47: 99-104. Lee, J.O., Kang, I.M., Cho, W.J. 2010. Smectite alteration and its influence on the barrier properties of smectite clay for a repository. Applied Clay Science 47: 99-104. Pusch R., Kasbohm, J., Thao, H.T.M. 2010. Chemical stability of montmorillonite buffer clay under repository-like conditions A synthesis of relevant experimental data. Applied Clay Science 47: 113119. Pusch R., Kasbohm, J., Thao, H.T.M. 2010. Chemical stability of montmorillonite buffer clay under repository-like conditions A synthesis of relevant experimental data. Applied Clay Science 47: 113119. Prikryl, R., Weishauptov, Z. 2010. Hierarchical porosity of bentonite-based buffer and its modification due to increased temperature and hydration. Applied Clay Science 47: 163-170. Pusch R., Zhang, L., Adey, R., Kasbohm, J. 2010. Rheology of an artificial smectitic clay. Applied Clay Science 47: 120-126. Sundberg, J., Back, P-E, Christiansson, R., Hkmark, H., Lndell, M., Wrafter, J. 2009. Modelling of thermal rock mass properties at the potential sites of a Swedish nuclear waste repository. International Journal of Rock Mechanics and Mining Sciences 46: 1042-1054. Sundberg, J., Hellstrm, G. 2009. Inverse modelling of thermal conductivity from temperature measurements at the Prototype Repository, sp HRL International Journal of Rock Mechanics and Mining Sciences 46: 1029-1041. Yang, S.-Y., Yeh, H-D. 2009. Modeling transient heat transfer in nuclear waste repositories, Journal of Hazardous Materials 169 (1-3), 108-112. Jockwer, N., Wieczorek, K., Fernndez, A.M. 2007. Measurements of gas generation, water content and change in the water distribution in a heater experiment in the underground laboratory Mont Terri. Physics and Chemistry of the Earth 32: 530-537. Suzuki, S., Sazarashi, M., Akimoto, T., Haginuma, M., Suzuki, K. 2008. A study of the mineralogical alteration of bentonite in saline water. Applied Clay Science 41: 190-198. Mokni, N., Olivella, S., Alonso, E.E. 2010. Swelling in clayey soils induced by the presence of salt crystals. Applied Clay Science 47: 105-112. Fukue, M., Fujimori, Y., Sato, Y., Nakagawa, T., Mulligan, C.N. 2010. Evidence of the production and dissolution of carbonate phases in bentonite formations. Applied Clay Science 47: 133-138 Wersin, P. 2003. Geochemical modeling of bentonite porewater in high-level waste repositories. Journal of Contaminant Hydrology 61: 405-422. Marty, N.C.M., Fritz, B., Clment, A., Michau, N. 2010. Modelling the long term alteration of the engineered bentonite barrier in an underground radioactive waste repository. Applied Clay Science 47: 82-90. Savage, D., Watson, C., Benbow, S., Wilson, J. 2010. Modelling iron-bentonite interactions. Applied Clay Science 47: 91-98. Carlson, L., Karnland, O., Oversby, V.M., Rance, A.P., Smart, N.R., Snellman, M., Vhnen, M., Werme, L.O. 2007. Experimental studies of the interactions between anaerobically corroding iron and bentonite. Physics and Chemistry of the Earth 32: 334-345. Karnland, O., Olsson, S., Nilsson, U., Sellin, P. 2007. Experimentally determined swelling pressures and geochemical interactions of compacted Wyoming bentonite with highly alkaline solutions. Physics and Chemistry of the Earth 32: 275-286. Savage, D., Walker, C., Arthur, R., Rochelle, C., Oda, C., Takase, H. 2007. Alteration of bentonite by hyperalkaline fluids: A review of the role of secondary minerals. Physics and Chemistry of the Earth 32: 287-297. Yamaguchi, T., Sakamoto, Y., Akai, M., Takazawa, M., Iida, Y., Tanaka, T., Nakayama, S. 2007. Experimental and modeling study on long-term alteration of compacted bentonite with alkaline groundwater. Physics and Chemistry of the Earth 32: 298-310. Trotignon, L., Devallois, V., Peycelon, H., Tiffreau, C., Bourbon, X. 2007. Predicting the long term durability of concrete engineered barriers in a geological repository for radioactive waste. Physics and Chemistry of the Earth 32: 259-274.

137

138 139 140

141

142 143

144 145 146 147 148

149 150

151

152

153

154

GeneWatch UK consultancy report September 2010

55

155 156 157 158 159 160

Graham, J., Halayko, K.,G., Hume, H., Kirkham, T., Gray, M., Oscarson, D. 2002. A capillarity-advective model for gas break-through in clays. Engineering Geology 64: 272-286. Graham, J., Halayko, K.,G., Hume, H., Kirkham, T., Gray, M., Oscarson, D. 2002. A capillarity-advective model for gas break-through in clays. Engineering Geology 64: 272-286. Gall, C. 2000. Gas breakthrough pressure in compacted Fo-Ca clay and interfacial gas overpressure in waste disposal context. Applied Clay Science 17: 85-97. Ortiz, L., Volckaert, G., Mallants, D. 2002. Gas generation and migration in Boom Clay, a potential host rock formation for nuclear waste storage. Engineering Geology 64: 287-296. Xu, T., Senger, R., Finsterle, S. 2008. Corrosion-induced gas generation in a nuclear waste repository: Reactive geochemistry and multiphase flow effects. Applied Geochemistry 23(12): 3423-3433. Falck, W.E., Nilsson, K. F. 2009. Geological disposal of radioactive waste. European Commission Joint Research Centre. JRC Reference Report EUR 23925 EN. http://ie.jrc.ec.europa.eu/publications/scientific_publications/2009/LRJRC_Reference_Report_IE_Geological%20Disposal.pdf Nakano, M., Kawamura, K. 2010. Estimating the corrosion of compacted bentonite by a conceptual model based on microbial growth dynamics. Applied Clay Science 47: 43-50. Fukue, M., Fujimori, Y., Sato, Y., Nakagawa, T., Mulligan, C.N. 2010. Evidence of the production and dissolution of carbonate phases in bentonite formations. Applied Clay Science 47: 133-138. Mulligan, C.N., Yong, R.N., Fukue, M. 2009. Some effects of microbial activity on the evolution of claybased buffer properties in underground repositories. Applied Clay Science 42: 331-335. Merroun, M.L., Selenska-Pobell, S. 2008. Bacterial interactions with uranium: An environmental perspective. Journal of Contaminant Hydrology 102 (3-4): 285-295. Mulligan, C.N., Yong, R.N., Fukue, M. 2009. Some effects of microbial activity on the evolution of claybased buffer properties in underground repositories. Applied Clay Science 42: 331-335. Nazina, T.N., Kosareva, I.M., Petrunyaka, V.V., Savushkina, M.K., Kudriavtsev, E.G., Lebedev, V.A., Ahunov, V.D., Revenko, Y.A., Khafizov, R.R., Osipov, G.A., Belyaev, S.S., Ivanov, M.V. 2004. Microbiology of formation waters from the deep repository of liquid radioactive wastes Severnyi. FEMS Microbiology Ecology 49: 97-107. Stroes-Gascoyne, S. 2010. Microbial occurrence in bentonite-based buffer, backfill and sealing materials from large-scale experiments at AECLs Underground Research Laboratory. Applied Clay Science 47: 36-42. Mulligan, C.N., Yong, R.N., Fukue, M. 2009. Some effects of microbial activity on the evolution of claybased buffer properties in underground repositories. Applied Clay Science 42: 331-335. Mulligan, C.N., Yong, R.N., Fukue, M. 2009. Some effects of microbial activity on the evolution of claybased buffer properties in underground repositories. Applied Clay Science 42: 331-335. Fukue, M., Fujimori, Y., Sato, Y., Nakagawa, T., Mulligan, C.N. 2010. Evidence of the production and dissolution of carbonate phases in bentonite formations. Applied Clay Science 47: 133-138. Mulligan, C.N., Yong, R.N., Fukue, M. 2009. Some effects of microbial activity on the evolution of claybased buffer properties in underground repositories. Applied Clay Science 42: 331-335. Mulligan, C.N., Yong, R.N., Fukue, M. 2009. Some effects of microbial activity on the evolution of claybased buffer properties in underground repositories. Applied Clay Science 42: 331-335. Malekifarsani, A., Skachek, M.A. 2009. Effect of precipitation, sorption and stable of isotope on maximum release rates of radionuclides from engineered barrier system (EBS) in deep repository. Journal of Environmental Radioactivity 100: 807-814. Altman, S. 2008. Geochemical research: A key building block for nuclear waste disposal safety cases. Journal of Contaminant Hydrology 102 (3-4): 174-179. Malekifarsani, A., Skachik, M.A. 2009. Calculation of maximum release rates in alternative design changes in the thickness of the buffer for the engineered barrier system (EBS) in deep repository by using Amber code. Progress in Nuclear Energy 51: 355-360. Salbu, B., Skipperud, L. 2009. Speciation of radionuclides in the environment. Journal of Environmental Radioactivity 100: 281-282. Darban, A.K., Yong, R.N., Ravaj, S. 2010. Coupled chemical speciation-solute transport model for

161 162 163 164 165 166

167

168 169 170 171 172 173

174 175

176 177

56

GeneWatch UK consultancy report September 2010

prediction of solute transport in clay buffers. Applied Clay Science 47: 127-132 178 179 180 181 Montarnal, Ph., Mugler, C., Descostes, M., Dimier, A., Jacquot, E. Presentation and use of a reactive transport code in porous media. Physics and Chemistry of the Earth 32: 507-517. Metz, V., Kienzler, B., Schssler, W. 2003. Geochemical evaluation of different groundwater-host rock systems for radioactive waste disposal. Journal of Contaminant Hydrology 61: 265-279. Aggarwal, M., Ndiaye, M.C.A., Carrayrou, J. 2007. Parameters estimation for reactive transport: A way to test the validity of a reactive model. Physics and Chemistry of the Earth 32: 518-529. Truche, L.,Berger, G. Destrigneville, C., Pages, A., Guillaume, D., Giffaut, E., Jacquot, E. 2009. Experimental reduction of aqueous sulphate by hydrogen under hydrothermal conditions: Implication for the nuclear waste storage. Geochimica et Cosmochimica Acta 73(16): 4824-4835. Alonso, U., Missana, T., Patelli, A., Rigato, V. 2007. Bentonite colloid diffusion through the host rock of a deep geological repository. Physics and Chemistry of the Earth 32: 469-476. Kersting, A.B., Eferd, D.W., Finnegan, D.L., Rokop, D.J., Smith, D.K., Thompson, J.L. 1999. Migration of plutonium in groundwater at the Nevada test site. Nature 397: 56-59. Geckeis, H., Rabung, T. 2008. Actinide geochemistry: From molecular level to the real system. Journal of Contaminant Hydrology 102 (3-4): 187-195. Kunze, P., Seher, H., Hauser, W., Panak, P.J. 2008. The influence of colloid formation in a granite groundwater bentonite porewater mixing zone on radionuclide speciation. Journal of Contaminant Hydrology 102 (3-4): 263-272. Wold, S., Eriksen, T. 2007. Diffusion of humic colloids in compacted bentonite. Physics and Chemistry of the Earth 32: 477-484. Filby, A., Plascke, M., Geckeis, H., Fanghnel, Th. 2008. Interaction of latex colloids with mineral surfaces and Grimsel granodiorite. Journal of Contaminant Hydrology 102 (3-4): 273-284. Heberling, F., Brendebach, B., Bosbach, D. 2008. Neptunium(V) adsorption to calcite. Journal of Contaminant Hydrology 102 (3-4): 246-252. Finck, N., Stumpf, T., Walther, C., Bosbach, D. 2008. TRLFS characterization of Eu(III)-doped synthetic organo-hectorite. Journal of Contaminant Hydrology 102 (3-4): 253-262. Alonso, U., Missana, T., Patelli, A., Rigato, V. 2007. Bentonite colloid diffusion through the host rock of a deep geological repository. Physics and Chemistry of the Earth 32: 469-476. Missana, T., Alonso, U., Turrero, J.M. 2003. Generation and stability of bentonite colloids at the bentonite/grantite interface of a deep geological radioactive waste repository. Journal of Contaminant Hydrology 61: 17-31. Rnnback, P., strm, M., Gustafsson, J.-P. 2008. Comparison of the behaviour of rare earth elements in surface waters, overburden groundwaters and bedrock groundwaters in two granitoidic settings, Eastern Sweden. Applied Geochemistry 23: 1862-1880. Geckeis, H., Rabung, T. 2008. Actinide geochemistry: From molecular level to the real system. Journal of Contaminant Hydrology 102 (3-4): 187-195. Gaona, X., Montoya, V., Cols, E., Griv, M., Duro, L. 2008. Review of the complexation of tetravalent actinides by ISA and gluconate under alkaline to hyperalkaline conditions. Journal of Contaminant Hydrology 102 (3-4): 217-227. Hallbeck, L., Pedersen, K. 2008. Characterization of microbial processes in deep aquifers of the Fennoscandian Shield. Applied Geochemistry 23: 1796-1819. Auqu, L., Gimeno, M.J., Gmez, J., Nilsson, A.-C. 2008. Potentiometrically measured Eh in groundwaters from the Scandanavian Shield. Applied Geochemistry 23: 1820-1833. Merroun, M.L., Selenska-Pobell, S. 2008. Bacterial interactions with uranium: An environmental perspective. Journal of Contaminant Hydrology 102 (3-4): 285-295. Pedersen, K. 1999. Subterranean microorganisms and radioactive waste disposal in Sweden. Engineering Geology 52: 163-176. Yim, M.-S., Caron, F. 2006. Life cycle and management of carbon-14 from nuclear power generation. Progress in Nuclear Energy 48: 2-36. Bracke, G., Mller, W. 2008. Contribution to a more realistic approach in assessing the release of C-14

182 183 184 185

186 187 188 189 190 191

192

193 194

195 196 197 198 199 200

GeneWatch UK consultancy report September 2010

57

from low-level radioactive waste repositories. Journal of Contaminant Hydrology 102 (3-4): 210-216. 201 202 203 204 Bracke, G., Mller, W. 2008. Contribution to a more realistic approach in assessing the release of C-14 from low-level radioactive waste repositories. Journal of Contaminant Hydrology 102 (3-4): 210-216. Yim, M.-S. 2006. Life cycle and management of carbon-14 from nuclear power generation. Progress in Nuclear Energy 48: 2-36. Grambow, B. 2008. Mobile fission and activation products in nuclear waste disposal. Journal of Contaminant Hydrology 102 (3-4): 180-186. Tsang, C. -F., Stephansson, O., Hudson, J.A. 2000. A discussion of thermo-hydro-mechanical (THM) processes associated with nuclear waste repositories. International Journal of Rock Mechanics and Mining Sciences 37(1-2):397-402. Tsang, C.-F., Jing, L., Stephansson, O., Kautsky, F. 2005. The DECOVALEX III project: A summary of activities and lessons learned. International Journal of Rock Mechanics and Mining Sciences 42: 593610. Dowd, P.A., Martin, J.A., Xu, C., Fowell, R.J., Mardia, K.V. 2009. A three-dimensional fracture network data set for a block of granite. International Journal of Rock Mechanics and Mining Sciences 46: 811818. Strm, A., Andersson, J., Skagius, K., Winberg, A. 2008. Site descriptive modelling during characterization for a geological repository for nuclear waste in Sweden. Applied Geochemistry 23: 1747-1760. Follin, S., Stephens, M.B., Laaksoharju, M., Nilsson, A.-C., Smellie, J.A.T., Tullborg, E.-L. 2008. Modelling the evolution of hydrochemical conditions in the Fennoscandian Shield during Holocene time using multidisciplinary information. Applied Geochemistry 23: 2004-2020. Laaksoharju, M., Smellie, J., Tullborg, E.-L., Gimeno, M., Molinero, J., Gurban, I., Hallbeck, L. 2008. Hydrogeochemical evaluation and modeling performed within the Swedish site investigation programme. Applied Geochemistry 23: 1761-1795. Waber, H.N., Smellie, J.A.T. 2008. Characterisation of pore water in crystalline rocks. Applied Geochemistry 23: 1834-1861. Laaksoharju, M., Gascoyne, M., Gurban, I. 2008. Understanding groundwater chemistry using mixing models. Applied Geochemistry 23: 1921-1940. Molinero, J., Raposo, J.R., Galndez, J.M., Arcos, D., Guimer, J. 2008. Coupled hydrogeological and reactive transport modeling of the Simpevarp area (Sweden). Applied Geochemistry 23: 1957-1981. Gmez, J.B., Auqu, L.F., Gimeno, M.J. 2008. Sensitivity and uncertainty analysis of mixing and mass balance calculations with standard and PCA-based geochemical codes. Applied Geochemistry 23: 1941-1956. Hunter, F.M.I., Hartley, L.J., Hoch, A., Jackson, C.P., McCarthy, R., Marsic, N., Gylling, B. 2008. Calibration of regional palaeohydrogeology and sensitivity analysis using hydrochemistry data in site investigations. Applied Geochemistry 23: 1982-2003. Yoshida, H., Metcalfe, R., Seida, Y., Takahashi, H., Kikuchi, T. 2008. Retardation capacity of altered granitic rock distributed along fractured and faulted zones in the orogenic belt of Japan. Engineering Geology 106: 116-122. Kienzler, B., Vejmelka, P., Romer, J., Fanghanel, E., Jansson, M., Eriksen, T.E., Wikberg, P. 2003. Swedish-German actinide migration experiment at SP hard rock laboratory. Journal of Contaminant Hydrology 61: 219-233. Lavastre, V., La Salle, C.L.G., Michelot, J.-L., Giannesini, S., Benedetti, L., Lancelot, J., Lavielle, B., Massault, M., Thomas, B., Gilabert, E., Bourls, D., Clauer, N., Agrinier, P. 2010. Establishing constraints on groundwater ages with 36Cl, 14C, 3H, and noble gases: A case study in the eastern Paris basin, France. Applied Geochemistry 25: 123-142. Delay, J., Rebours, H., Vinsot, A., Robin, P. 2007. Scientific investigation in deep weels for nuclear waste disposal studies at the Meuse/Haute Marne underground research laboratory, Northeastern France. Physics and Chemistry of the Earth 32: 42-57. Distinguin, M., Lavanchy, J.-M. 2007. Determination of hydraulic properties of the Callovo-Oxfordian argillite at the bure site: Synthesis of the results obtained in deep boreholes using several in situ investigation techniques. Physics and Chemistry of the Earth 32: 379-392.

205

206

207

208

209

210 211 212 213

214

215

216

217

218

219

58

GeneWatch UK consultancy report September 2010

220

Delay, J., Distinguin, M., Dewonck, S. 2007. Characterization of a clay-rich rock through development and installation of specific hydrogeological and diffusion test equipment in deep boreholes. Physics and Chemistry of the Earth 32: 393-407. Horseman, S.T., Harrington, J.F., Noy, D.J. 2007. Swelling and osmotic flow in a potential host rock. Physics and Chemistry of the Earth 32: 408-420. Tsang, C. -F., Stephansson, O., Hudson, J.A. 2000. A discussion of thermo-hydro-mechanical (THM) processes associated with nuclear waste repositories. International Journal of Rock Mechanics and Mining Sciences 37(1-2):397-402. Bckblom, G., Martin, C.D. 1999. Recent experiments in hard rocks to study the excavation response: implications for the performance of a nuclear waste geological repository. Tunnelling and Underground Space Technology 14: 377-394. Martin, C.D., Christiansson, R. 2009. Estimating the potential for spalling around a deep nuclear waste repository in crystalline rock. International Journal of Rock Mechanics and Mining Sciences 46: 219228. Andersson, J.C., Martin, C.D. 2009. The sp Pillar Stability Experiment: Part I Experiment design. International Journal of Rock Mechanics and Mining Sciences 46: 865-878. Andersson, J.C., Martin, C.D., Stille, H. 2009. The sp Pillar Stability Experiment: Part II Rock mass response to coupled excavation-induced and thermal-induced stresses. International Journal of Rock Mechanics and Mining Sciences 46: 879-895. Lin, Q.X., Liu, Y.M., Tham, L.G., Tang, C.A., Lee, P.K.K., Wang, J. 2009. Time-dependent strength degradation of granite. International Journal of Rock Mechanics and Mining Sciences 46: 1103-1114. Delay, J., Vinsot, A., Krieguer, J.-M., Rebours, H., Armand, G. 2007. Making use of the underground scientific experimental programme at the Meuse/Haute-Marne underground research laboratory, North Eastern France. Physics and Chemistry of the Earth 32: 2-18. Sugita, Y., Fujita, T., Takahashi, Y., Kawakami, S., Umeki, H. Yui, M., Uragami, M., Kitayama, K. 2007. The Japanese approach to developing clay-based repository concepts An example of design studies for the assessment of sealing strategies. Physics and Chemistry of the Earth 32: 32-41. Martino, J.B., Dixon, D.A., Kozak, E.T., Gascoyne, M., Vignal, B., Sugita, Y., Fujita, T., Masumoto, K. 2007. The tunnel sealing experiment: An international study of full-scale seals. Physics and Chemistry of the Earth 32: 93-107. Van Geet, M., Volckaert, G., Bastiaens, W., Maes, N., Weetjens, E., Sillen, X., Vallejan, B., Gens, A. 2007. Efficiency of a borehole seal by means of pre-compacted bentonite blocks. Physics and Chemistry of the Earth 32: 123-134. Wileveau, Y., Bernier, F. 2008. Similarities in the hydromechanical response of Callovo-Oxfordian clay and Boom clay during gallery excavation. Physics and Chemistry of the Earth 33: S343-S349. Klubertanz, G., Folly, M., Hufschmied, P., Frank, E. 2008. Impact of the thermal load on the farfield and galleries of a HLW-repository. Physics and Chemistry of the Earth 33: S457-S461. Smai, F. 2009. A model of multiphase flow and transport in porous media applied to gas migration in underground nuclear waste repository. Comptes Rendus Mathematique 347(9-10), 527-532. Bourgeat, A., Jurak, M. 2010. A two level scaling-up method for multiphase flow in porous media: numerical validation and comparison with other methods. Computational Geoscience 14: 1-14. Aquino, J., Francisco, A.S., Pereira, F., Souto, H.P.A. 2008. An overview of Eulerian-Lagrangian schemes applied to radionuclide transport in unsaturated porous media. Progress in Nuclear Energy 50: 774-787. Javeri, V. 2008.Three dimensional analysis of combined gas, heat and nuclide transport in a repository in clay rock including coupled thermo-hydro-geomechanical processes. Physics and Chemistry of the Earth 33: S252-S259. Alkan, H., Muller, W. 2008. Approaches for modeling gas flow in clay formations as repository systems. Physics and Chemistry of the Earth 33: S260-S268. Lee Y.-M., Hwang, Y. 2009. A GoldSim model for the safety assessment of an HLW respository. Progress in Nuclear Energy 51: 746-759. Evans, D., Stephenson, M., Shaw, R. 2008. The present and future use of land below ground. Land Use Policy 26S: S302-S316.

221 222

223

224

225 226

227 228

229

230

231

232 233 234 235 236

237

238 239 240

GeneWatch UK consultancy report September 2010

59

241

Falck, W.E., Nilsson, K. F. 2009. Geological disposal of radioactive waste. European Commission Joint Research Centre. JRC Reference Report EUR 23925 EN. http://ie.jrc.ec.europa.eu/publications/scientific_publications/2009/LRJRC_Reference_Report_IE_Geological%20Disposal.pdf Knowles, M.K., Hansen, F.D., Thompson, T.W., Schatz, J.F., Gross, M. 2000. Review and perspectives on spallings release models in the 1996 performance assessment for the Waste Isolation Pilot Plant. Reliability Engineering and System Safety 69: 331-341. Calic, D., Ravnik, M. 2010. Criticality calculations of spent fuel in deep geological respository. Nuclear Engineering and Design 240: 668-671. Chan, T., Christiansson, R., Boulton, G.S., Ericsson, L.O., Hartikainen, J., Jensen, M.R., Ivars, D.M., Stanchell, F.W., Vistrand, P., Wallroth, T. 2005. DECOVALEX III BMT3/BENCHPAR WP4: The thermohydro-mechanical responses to a glacial cycle and their potential implications for deep geological disposal of nuclear fuel waste in a fractured crystalline rock mass. International Journal of Rock Mechanics and Mining Sciences 42: 805-827. Talbot, C.J. 1999. Ice ages and nuclear waste isolation. Engineering Geology 52: 177-192. Arcos, D., Grandia, F., Domenech, C., Fernndez, A. M., Villar, M.V., Muurinen, A., Carlsson, T., Sellin, P. Hernan, P. 2008. Long-term geochemical evolution of the near field repository: Insights from reactive transport modelling and experimental evidences. Journal of Contaminant Hydrology 102 (3-4): 196-209. Dideriksen, K., Christiansen, B.C., Frandsen, C., Balic-Zunic, T., Mrup, S., Stipp, S.L.S. 2010. Paleoredox boundaries in fractured granite. Geochimica et Cosmochimica Acta 74: 2866-2880. Sidborn M., Neretnieks, I. 2008. Long-term oxygen depletion from infiltrating groundwaters: Model development and application to intra-glaciation and glaciation conditions. Journal of Contaminant Hydrology 100: 72-89. MacQuarrie, K.T.B., Mayer, K.U., Jin, B., Spiessl, S.M. 2010. The importance of conceptual models in the reactive transport simulation of oxygen ingress in sparsely fractured crystalline rock. Journal of Contaminant Hydrology 112: 64-76. Starinsky, A., Katz, A. 2003. The formation of natural cryogenic brines. Geochimica et Cosmochimica Acta 67(8): 1475-1484. Tullborg, E.-L., Drake, H., Sandstrm, B. 2008. Palaeohydrogeology: A methodology based on fracture mineral studies. Applied Geochemistry 23: 1881-1897. Sandstrm, B., Tullborg, E.-L. 2009. Episodic fluid migration in the Fennoscandian Shield recorded by stable isotopes, rare earth elements and fluid inclusions in fracture minerals at Forsmark, Sweden. Chemical Geology 266: 135-151. Drake, H., Tullborg, E.-L., Page, L. 2009. Distinguished multiple events of fracture mineralization related to far-field orogenic effects in Paleoproterozoic crystalline rocks, Simpevarp area, SE Sweden. Lithos 110: 37-49. Sandstrm, B., Tullborg, E.-L., Larson, S.A., Page, L. 2009. Brittle tectonothermal evolution in the Forsmark area, central Fennoscandian Shield, recorded by paragenesis, orientation and 40Ar/39Ar geochronology of fracture minerals. Tectonophysics 478: 158-174. Rasilainen, K., Suksi, J., Ruskeeniemi, T., Pitkanen, P., Poteri, A. 2003. Release of uranium from rock matrix a record of glacial meltwater intrusions? Journal of Contaminant Hydrology 61: 235-246. Bath A., Richards, H., Metcalfe, R., McCartney, R., Degnan, P., Littleboy, A. 2006. Geochemical indicators of deep groundwater movements at Sellafield, UK. Journal of Geochemical Exploration 90: 24-44. Chan, T., Christiansson, R., Boulton, G.S., Ericsson, L.O., Hartikainen, J., Jensen, M.R., Ivars, D.M., Stanchell, F.W., Vistrand, P., Wallroth, T. 2005. DECOVALEX III BMT3/BENCHPAR WP4: The thermohydro-mechanical responses to a glacial cycle and their potential implications for deep geological disposal of nuclear fuel waste in a fractured crystalline rock mass. International Journal of Rock Mechanics and Mining Sciences 42: 805-827. Stotler, R.L., Frape, S.K., Ruskeeniemi, T., Ahonen, L., Onstott, T.C., Hobbs, M.Y. 2009. Hydrogeochemistry of groundwaters in and below the base of thick permafrost at Lupin, Nunavut, Canada. Journal of Hydrology 373: 80-95. Chan, T., Stanchell, F.W. 2005. Subsurface hydro-mechanical (HM) impacts of glaciation: Sensitivity to

242

243 244

245 246

247 248

249

250 251 252

253

254

255 256

257

258

259

60

GeneWatch UK consultancy report September 2010

transient analysis, HM coupling, fracture zone connectivity and model dimensionality. International Journal of Rock Mechanics and Mining Sciences 42: 828-849. 260 Jost, A., Violette, S., Goncalves, J., Ledoux, E., Guyomard, Y., Guillocheau, F., Kageyama, M., Ramstein, G., Suc, J.-P. 2007. Long-term hydrodynamic response induced by past climatic and geomorphic forcing: The case of the Paris basin, France. Physics and Chemistry of the Earth 32: 368378. Stewart, S. 2002. Exploring the continental shelf for low geological risk nuclear waste respository sites using petroleum industry databases: a UK case study. Engineering Geology 67: 139-168. Juhlin, C., Dehghannejad, M., Lund, B., Malehmir, A., Pratt, G. 2010. Reflection seismic imaging of the end-glacial Prvie Fault system, northern Sweden. Journal of Applied Geophysics 70: 307-316. Brennwald, M.S., van Dorp, F. 2009. Radiological risk assessment and biosphere modelling for radioactive waste disposal in Switzerland. Journal of Environmental Radioactivity 100: 1058-1061. Zvonova, I., Krajewski, P., Berkovsky, V., Amman, M., Duffa, C., Filistovic, V., Homma, T., Kanyar, B., Nedveckaite, T., Simon, S.L., Vlasov, O., Webbe-Wood, D. 2010. Validation of 131I ecological transfer models and thyroid dose assessments using Chernobyl fallout data from the Plavsk district, Russia. Journal of Environmental Radioactivity 101: 8-15. Rnnback, P., strm, M. 2007. Hydrochemical patterns of a small lake and a stream in an uplifting area proposed as a repository site for spent nuclear fuel Sweden. Journal of Hydrology 344: 223-235. Monte, L. 2010. Modelling multiple dispersion of radionuclides through the environment. Journal of Environmental Radioactivity 101: 134-139. Report of the Committee Examining Radiation Risks of Internal Emitters (CERRIE) 2004. www.cerrie.org Salbu, B., Skipperud, L. 2009. Speciation of radionuclides in the environment. Journal of Environmental Radioactivity 100: 281-282. Choi, Y.-H., Lim, K.-M., Jun, I., Park, D-W, Keum, D.-K., Lee, C.W. 2009. Root uptake of radionuclides following their acute soil depositions during the growth of selected food crops. Journal of Environmental Radioactivity 100: 746-751. Howard, B.J., Beresford, N.A., Barnett, C.L., Fesenko, S. 2009. Quantifying the transfer of radionuclides to food products from domestic farm animals. Journal of Environmental Radioactivity 100: 767-773. Gleizon, P., McDonald, P. 2010. Modelling radioactivity in the Irish Sea: From discharge to dose. Journal of Environmental Radioactivity 101: 403-413. Kaplan, D.J., Demirkanli, D.I., Molz, F.J., Beals, D.M., Cadieux Jr., J.R., Halverson, J.E. 2010. Upward movement of plutonium to surface sediments during an 11-year field study. Journal of Environmental Radioactivity 101: 338-344. Gudelis, A., Gvozdaite, R., Kubareviciene, R., Lukoevicius, S., utas, A. 2010. On radiocarbon and plutonium leakage to groundwater in the vicinity of a shallow-land radioactive repository. Journal of Environmental Radioactivity 101: 443-445. Brchignac, F., Doi, M. 2009. Challenging the current strategy of radiological protection of the environment: arguments for an ecosystem approach. Journal of Environmental Radioactivity 100: 11251134. Brchignac, F., Doi, M. 2009. Challenging the current strategy of radiological protection of the environment: arguments for an ecosystem approach. Journal of Environmental Radioactivity 100: 11251134. Mietelski, J.W., Maksimova, S., Szwalko, P., Wnuk, K., Zagrodzki, P. Blazej, S., Gaca, P., Tomankiewicz, E., Orlov, O. 2010. Plutonium, 137Cs and 90Sr in selected invertebrates from some areas around Chernobyl nuclear power plant. Journal of Environmental Radioactivity 101: 488-493. Larsson, C-M. 2009. Waste disposal and the recommendations of the International Commission on Radiological Protection Challenges for radioecology and environmental radiation protection. Journal of Environmental Radioactivity 100: 1053-1057. Kirchner, G. 2010. Use of reference biospheres for proving the long-term safety of radioactive waste repositories. Journal of Environmental Radioactivity 101: 435-437

261 262 263 264

265 266 267 268 269

270

271 272

273

274

275

276

277

278

GeneWatch UK consultancy report September 2010

61

279

Albrecht, A., Miquel, S. 2010. Extension of sensitivity and uncertainty analysis for long term dose assessment of high level nuclear waste disposal sites to uncertainties in the human behaviour. Journal of Environmental Radioactivity 101: 55-67. Kirchner, G. 2010. Use of reference biospheres for proving the long-term safety of radioactive waste repositories. Journal of Environmental Radioactivity 101: 435-437. SKB. 2006. Long-term safety for KBS-3 repositories at Forsmark and Laxemar a first evaluation. Main report of the SR-Can project. November 2006. http://www.skb.se/upload/publications/pdf/TR-0609webb.pdf Posiva Oy. 2007. Expected evolution of a spent nuclear fuel repository at Olkiluoto. Revised October 2007. http://www.posiva.fi/files/346/Posiva2006-05_revised_081107web.pdf Andra. 2005. Dossier 2005 Argile. Safety evaluation of a geological repository. http://www.andra.fr/download/andra-international-en/document/editions/270va.pdf Baker, A.J., Chambers, A.V., Jackson, C.P., Porter, J.D., Sinclair, J.E., Sumner, P.J., Thorne, M.C., Watson, S.P. 1997. Nirex 97: An Assessment of the Post-closure Performance of a Deep Waste Repository at Sellafield. Volume 3: The Groundwater Pathway. http://www.nda.gov.uk/documents/biblio/detail.cfm?fuseaction=search.view_doc&doc_id=2388 Posiva Oy. 2007. Expected evolution of a spent nuclear fuel repository at Olkiluoto. Revised October 2007. http://www.posiva.fi/files/346/Posiva2006-05_revised_081107web.pdf Oreskes, N., Schrader-Frechette, K., Belitz. K. 1994. Verification, validation and confirmation of numerical models in the Earth sciences. Science 263: 641-646. Back, P.-E., Christiansson, R. 2009. Value of information analysis for site investigation programs accounting for variability, uncertainty and scale effects with the sp HRL prototype repository as an example. International Journal of Rock Mechanics and Mining Sciences 46: 896-904. European Environment Agency. 2001. Late lessons from early warnings: the precautionary principle 1896-2000. Environmental Issue Report No. 22. Luxembourg. Beven, K. 2002. Towards a coherent philosophy for modelling the environment. Proceedings of the Royal Society of London 458: 2465-2484. Carter, J.N., Ballester, P.J., Tavassoli, Z., King, P.R. 2006. Our calibrated model has no predictive value: An example from the petroleum industry. Reliability Engineering & System Safety, 91(10-11), 13731381. Beven, K. 2002. Towards a coherent philosophy for modelling the environment. Proceedings of the Royal Society of London 458: 2465-2484. Beven, K. 2002. Towards a coherent philosophy for modelling the environment. Proceedings of the Royal Society of London 458: 2465-2484. Beven, K. 2002. Towards a coherent philosophy for modelling the environment. Proceedings of the Royal Society of London 458: 2465-2484. Kaptchuk, T.J. 2003. Effect of interpretative bias on research evidence. British Medical Journal 326: 1453-1455. Baveye, P. 2003. The emergence of a new kind of relativism in environmental modelling: a commentary. Proceedings of the Royal Society of London. A. 460: 2141-2146. Beven, K. 2003. Reply to: The emergence of a new kind of relativism in environmental modelling: a commentary by Philippe Baveye. Proceedings of the Royal Society of London A. 460: 2147-2151. Beven, K.J. 2006. A Manifesto for the Equifinality Thesis. Journal of Hydrology 320 (1-2): 8-36. Beken,T.V., Dorn, N., Daele, S.V. 2010. Security risks in nuclear waste management: Exceptionalism, opaqueness and vulnerability. Journal of Environmental Management 91: 940-948. Stelfox, H., Chua G., ORourke, K., Detsky, A. 1998 Conflict of interest in the debate over calciumchannel antagonists. New England Journal of Medicine 2: 101-106. Bhandari, M. Busse, J.W., Jackowski, D.,et al., 2004. Association between industry funding and statistically significant pro-industry findings in medical and surgical randomized trials. Canadian Medical Association Journal 170(4):477-80. Friedman, L., Richter, E. 2004. Relationship between conflicts of interest and research results. Journal of General Internal Medicine 19: 5156.

280 281

282 283 284

285 286 287

288 289 290

291 292 293 294 295 296 297 298 299 300

301

62

GeneWatch UK consultancy report September 2010

302 303 304 305 306 307 308

Lexchin, J., Bero, L.A., Djulbegovic, B., Clark, O. 2003. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. British Medical Journal 326: 1167-1170. Hartmann, M., Knoth, H., Schulz, D., Knoth, S. 2003. Industry sponsored studies in oncology vs. studies sponsored by nonprofit organisations. British Journal of Cancer 89: 1405-1408. Katan, M.B. 2007. Does Industry Sponsorship Undermine the Integrity of Nutrition Research? PLoS Medicine 4(1): e6 doi:10.1371/journal.pmed.0040006 Haszeldine, S., Smythe, D. 1997. Why was Sellafield rejected as a disposal site for radioactive waste? Geoscientist 7(7): 18-20. McKeown, C., Haszeldine, R.S., Couples, G.D. 1999. Mathematical modelling of groundwater flow at Sellafield, UK. Engineering Geology 52: 231-250. Haszeldine, S., Smythe, D. 1997. Why was Sellafield rejected as a disposal site for radioactive waste? Geoscientist 7(7): 18-20. Falck, W.E., Nilsson, K. F. 2009. Geological disposal of radioactive waste. European Commission Joint Research Centre. JRC Reference Report EUR 23925 EN. http://ie.jrc.ec.europa.eu/publications/scientific_publications/2009/LRJRC_Reference_Report_IE_Geological%20Disposal.pdf European Commission. 2009. Implementing Geological Disposal of Radioactive Waste Technology Platform: Vision Document. October 2009. http://www.igdtp.eu/Documents/VisionDoc_Final_Oct24.pdf Finnish Broadcasting Company YLE. 2010. Ydinjtteen loppusijoittamisen lupa-aikataulu liian kire. 26 May 2010. http://yle.fi/uutiset/kotimaa/2010/05/ydinjatteen_loppusijoittamisen_lupaaikataulu_liian_kirea_1708124.html Vuorinen, A. 2008. Regulators role in development of Finnish nuclear waste disposal program. Progress in Nuclear Energy 50: 674-679. CARD Project. 2008. A Co-ordination Action on Research, Development and Demonstration Priorities and Strategies for Geological Disposal. Final Report May 2008. European Commission. 2009. Implementing Geological Disposal of Radioactive Waste Technology Platform: Vision Document. October 2009. Solomon, B.D., Andrn, M. , Strandberg, U. 2009. Thirty years of social science research on high-level nuclear waste. Conference on Managing Radioactive Waste: Problems and Challenges in a Globalized World. University of Gothenburg, Sweden, December 15-17, 2009. http://www.cefos.gu.se/digitalAssets/1291/1291675_Solomon__paper_.pdf Durant, D. 2009. Responsible action and nuclear waste disposal. Technology in Society 31: 150-157. Durant, D. 2009. Responsible action and nuclear waste disposal. Technology in Society 31: 150-157. Slovenia: Agreement on a site for LILW repository reached. ENS News, Issue 27. Winter 2010. http://www.euronuclear.org/e-news/e-news-27/slovenia.htm Falck, W.E., Nilsson, K. F. 2009. Geological disposal of radioactive waste. European Commission Joint Research Centre. JRC Reference Report EUR 23925 EN. http://ie.jrc.ec.europa.eu/publications/scientific_publications/2009/LRJRC_Reference_Report_IE_Geological%20Disposal.pdf Chung, J.B., Kim, H.-K. 2009. Competition, economic benefits, and risk perception in siting a potentially hazardous facility. Landscape and Urban Planning 91: 8-16. Sjberg, L. 2009. Precautionary attitudes and the acceptance of a local nuclear waste repository. Safety Science 47: 542-546. Kojo, M., Kari, M., Litmanen, T. 2010. The socio-economic and communication challenges of spent nuclear fuel management in Finland. The post site selection phase of the repository project in Eurajoki. Progress in Nuclear Energy 52: 168-176. Kojo, M., Kari, M., Litmanen, T. 2010. The socio-economic and communication challenges of spent nuclear fuel management in Finland. The post site selection phase of the repository project in Eurajoki. Progress in Nuclear Energy 52: 168-176. Mather, J. 1997. The history of research into radioactive waste disposal in the United Kingdom and the selection of a site for detailed investigation. Environmental Policy and Practice 6: 167-177. Kelling, G., Knill, J. 1997. The Nirex story: a geological perspective. Geoscientist 7(7): 10-13.

309 310

311 312 313 314

315 316 317 318

319 320 321

322

323 324

GeneWatch UK consultancy report September 2010

63

325 326 327 328 329 330 331 332

Haszeldine, S., Smythe, D. 1997. Why was Sellafield rejected as a disposal site for radioactive waste? Geoscientist 7(7), 18-20. McKeown, C., Haszeldine, R.S., Couples, G.D. 1999. Mathematical modelling of groundwater flow at Sellafield, UK. Engineering Geology 52: 231-250. Greenhalgh, C., Azapagic, A. 2009. Review of drivers and barriers for nuclear power in the UK. Environmental Science & Policy 12: 1052-1067. Macalister, T. 2007. Sellafield not fit for nuclear waste disposal. The Guardian. 2nd November 2007. http://www.guardian.co.uk/business/2007/nov/02/nuclearindustry.greenpolitics McDonald, C. 2007. Letter: Flaws in search for nuclear waste site. The Guardian. 28th June 2007. http://www.guardian.co.uk/world/2007/jun/28/nuclear.uk Allerdale might host N-waste dump. Whitehaven News. 10th December 2008. http://westcumbriamrws.org.uk/ CoRWM. 2007. Re-iteration of CoRWMs Position on Nuclear New Build. September 2007 http://www.corwm.org.uk/pdf/2162%202%20%20CoRWM%20position%20on%20new%20build%20reiterated.pdf

333

Kojo, M., Kari, M., Litmanen, T. 2010. The socio-economic and communication challenges of spent nuclear fuel management in Finland. The post site selection phase of the repository project in Eurajoki. Progress in Nuclear Energy 52: 168-176. Resnikoff, M., Travers, J., Alexandrova, E. 2010. The hazards of generation III reactor fuel wastes. Greenpeace Canada. May 2010. EC. 2008. Attitudes towards radioactive waste. Special Eurobarometer 297. June 2008. European Commission. http://ec.europa.eu/public_opinion/archives/ebs/ebs_297_en.pdf www.nea.fr/html/rwm/fsc.html Kugo, A., Yoshikawa, H., Wakabayashi, Y. Shimoda, H., Ito, K., Uda, A. 2008. Study on risk communication by using Web system for the social consensus toward HLW final disposal. Progress in Nuclear Energy 50: 700-708. Beken T.V., Dorn, N., Daele, S.V. 2010. Security risks in nuclear waste management: Exceptionalism, opaqueness and vulnerability. Journal of Environmental Management 91: 940-948. Kim, S.K., Lee, M.S., Choi, H.J., Kwak, T.-W. 2009. Progress of a cost optimization for an HLW repository in Korea. Progress in Nuclear Energy 51: 401-408. Kim, S.K., Chun, K.S., Choi, H.J., Choi, J.W., Kwak, T.-W. 2007. Cost estimation of the canisters for an HLW repository in Korea. Progress in Nuclear Energy 49: 555-566. Kim, S.K., Lee, M.S., Choi, H.J., Kwak, T.-W. 2009. Progress of a cost optimization for an HLW repository in Korea. Progress in Nuclear Energy 51: 401-408. Kim, S.K., Lee, M.S., Choi, H.J., Choi, J.W., Revankar, S.T. 2009. Availability of a probabilistic cost estimation for the price effect of Cu powder and bentonite on a HLW disposal cost in Korea. Progress in Nuclear Energy 51: 649-657. Kim, S.K., Lee, M.S., Choi, H.J., Kwak, T.-W. 2009. Progress of a cost optimization for an HLW repository in Korea. Progress in Nuclear Energy 51: 401-408. Kim, S.K., Lee, Y., Choi, J.W., Hahn, P.S., Kwak, T.-W. 2007. A comparison of the HLW underground repository cost for the vertical and horizontal emplacement options in Korea. Progress in Nuclear Energy 49: 79-92. Suyama, Y., Toida, M., Yanagizawa, K. 2009. Study of an optimization approach for a disposal tunnel layout, taking into account the geological environment with spatially heterogeneous characteristics. Nuclear Engineering and Design 239: 1693-1698. Falck, W.E., Nilsson, K. F. 2009. Geological disposal of radioactive waste. European Commission Joint Research Centre. JRC Reference Report EUR 23925 EN. http://ie.jrc.ec.europa.eu/publications/scientific_publications/2009/LRJRC_Reference_Report_IE_Geological%20Disposal.pdf European Commission. 2009. Implementing Geological Disposal of Radioactive Waste Technology Platform: Vision Document. October 2009. Http://www.igdtp.eu/Documents/VisionDoc_Final_Oct24.pdf

334 335 336 337

338 339 340 341 342

343 344

345

346

347

64

GeneWatch UK consultancy report September 2010

GeneWatch
UK

The Nuclear Liability and Compensation Act:

Is it appropriate for the 21st century?

November 2009

BY GORDON R. THOMPSON

Institute for Resource and Security Studies

www.greenpeace.ca

THE NUCLEAR LIABILITY AND COMPENSATION ACT: Is it Appropriate for the 21st Century?:
BY GORDON R. THOMPSON Institute for Resource and Security Studies Prepared under the sponsorship of Greenpeace Canada Copyright November 2009 GREENPEACE CANADA 33 Cecil St. Toronto, Ontario M5T 1N1 www.greenpeace.ca 1 800 320-7183

AbStRAct A bill has been introduced into Canadas parliament to create the Nuclear Liability and Compensation Act (NLCA), which would replace the existing Nuclear Liability Act. Both Acts pertain to civil liability for damage arising from an incident at a nuclear installation. This report assesses the proposed NLCA. In performing that task, the report focuses on the application of the NLCA to nuclear power plants (NPPs) in Canada. The report further focuses on potential incidents at NPPs where the incident involves an unplanned release of radioactive material to the environment or within a plant. Criteria used here to assess the NLCA are based upon three concepts that would play major roles in a properly-constructed 21st century policy framework for nuclear power. The concepts are: sustainability; the precautionary principle; and full accounting of costs (which includes the polluter-pays principle). This reports assessment of the NLCA identifies significant deficiencies. Options for improving the NLCA are set forth here. About the InStItute foR ReSouRce And SecuRIty StudIeS The Institute for Resource and Security Studies (IRSS) is an independent, nonprofit, Massachusetts corporation, founded in 1984. Its objective is to promote sustainable use of natural resources and global human security. In pursuit of that mission, IRSS conducts technical and policy analysis, public education, and field programs. IRSS projects always reflect a concern for practical solutions to resource and security problems. About the AuthoR Gordon R. Thompson is the executive director of IRSS and a research professor at Clark University, Worcester, Massachusetts. He studied and practiced engineering in Australia, and received a doctorate in applied mathematics from Oxford University in 1973, for analyses of plasma undergoing thermonuclear fusion. Dr. Thompson has been based in the USA since 1979. His professional interests encompass a range of technical and policy issues related to sustainability and global human security. He has conducted numerous studies on the environmental and security impacts of nuclear facilities, and on options for reducing those impacts. For example, Dr. Thompson prepared a report in 2000 for the Standing Committee on Energy, Environment, and Natural Resources of the Canadian Senate, discussing the accident risk posed by the Pickering A nuclear generating station. AcknowledgementS This report was prepared by IRSS under the sponsorship of Greenpeace Canada. Shawn-Patrick Stensil assisted the author by obtaining information that was used during preparation of the report. The author, Gordon R. Thompson, is solely responsible for the content of the report.

Greenpeace is one of the worlds most effective and best-known environmental organizations, with almost three million supporters worldwide in thirty countries. Greenpeace is an independent global campaigning organization that acts to change attitudes and behaviour, in order to protect and conserve our environment and promote a peaceful future. Greenpeace has had a long-standing interest in nuclear issues, and has worked to promote a shift away from nuclear power and fossil fuels towards sustainable energy systems based on conservation and renewable technologies.
iNstitutE foR REsouRCE AND sECuRity stuDiEs

27 Ellsworth Avenue, Cambridge, Massachusetts 02139, USA Phone: 617-491-5177 Fax: 617-491-6904 Email: info@irss-usa.org

Cover photo taken by thespeak, Creative Commons License 2006 http://www.flickr.com/photos/thespeak/150544424/sizes/o/

Executive Summary
A bill has been introduced into Canadas parliament to create the Nuclear Liability and Compensation Act (NLCA), which would replace the existing Nuclear Liability Act. Both Acts pertain to civil liability for damage arising from an incident at a nuclear installation. Under the new Act, the liability of the operator of the affected installation would be limited to $650 million, compared to the limit of $75 million under the present Act. This report assesses the proposed NLCA, focusing on its application to nuclear power plants (NPPs), and on incidents involving an unplanned release of radioactive material to the environment or within a plant. Damage from such an incident could substantially exceed other types of damage covered by the NLCA. The NLCA is assessed here using three concepts that would play major roles in a properly-constructed 21st century policy framework for nuclear power. The concepts are: sustainability; the precautionary principle; and full accounting of costs (which includes the polluter-pays principle). Criteria based on these concepts are used here to assess the NLCA. The criteria are expressed as questions, as follows: Criterion #1: full accounting information disclosure Would the NLCA require Canadas nuclear industry and government to publish a full spectrum of information relevant to nuclear risk and nuclear insurance, including: (a) an assessment of the risk of operating each NPP and other nuclear installation; and (b) the insurance coverage of offsite and onsite damage that is provided for each NPP and other nuclear installation, premiums paid for this coverage, and reinsurance arrangements? Criterion #2: full accounting damage coverage Would the NLCA ensure that commercial insurance covers the full range of potential offsite and onsite damage arising from incidents at each NPP and other nuclear installation? Criterion #3: full accounting accounting for risk costs as costs of production Would the NLCA ensure that the risk costs of operating each NPP and other nuclear installation are accounted for as costs of production? Criterion #4: the precautionary principle Would the NLCA be compatible with coherent and consistent application of the precautionary principle in Canada? Criterion #5: sustainability Would the NLCA be compatible with a transition of Canadas economy toward sustainability? mAjoR fIndIngS In regard to Criterion #1, the NLCA would not require Canadas nuclear industry and government to publish a full spectrum of information relevant to nuclear risk and nuclear insurance. Thus, the present situation would continue, in which the information published about nuclear risk is often incomplete or misleading, and there is a dearth of information about the operation of the nuclear insurance sector. This situation is incompatible with democracy, given the public-policy significance of nuclear risk and nuclear insurance. In regard to Criterion #2, the NLCA would not ensure that commercial insurance covers the full range of potential damage arising from incidents at NPPs. The liability of commercial insurers would be limited to $650 million, and a substantial portion of that exposure would be reinsured by the Canadian government. Monetized damage to third parties could far exceed $650 million. Studies by the Canadian nuclear industry show that offsite health costs from some potential accidents at existing NPPs could exceed $50 billion. The industry says that such accidents have low probability. Yet, offsite health costs could exceed $1 billion for accidents that the industry concedes are credible because their probability is comparatively high. Moreover, there are reasons to doubt that accident probabilities are as low as the nuclear industry claims. Nuclear insurers assume much higher probabilities. Also, the potential for malevolent acts must be considered. Overall, the probability of an event causing offsite damage exceeding $50 billion is substantially higher than is shown by industry studies.

In the USA, a two-tier system of nuclear insurance provides about $11 billion of coverage for damage to third parties. The Price-Anderson Act governs this arrangement. In Germany, there is no limit on an NPP operators liability for damage to third parties. Each NPP operator in Germany must provide security of 2.5 billion Euro toward its liability. An unplanned release of radioactive material could cause economic damage to the operator of an NPP even if most of the released material is contained within the plant buildings. Studies by the nuclear industry show that accidents with a comparatively high probability could cause damage to the operator in excess of $650 million. For example, an industry study shows that one type of accident at the Darlington nuclear station with an estimated probability of 3 per 1,000 reactor-years (i.e., 48 percent over 40 years, at this 4-unit station) would cause damage between $790 million and $2,500 million to the operator. It appears that no insurance coverage of such an accident exists now in Canada or would exist under the NLCA. Although the damage would directly affect the operator, its costs would ultimately be borne by society at large. Insurance coverage of damage to the operator is provided in the USA. In regard to Criterion #3, the NLCA would not ensure that the risk costs of operating each NPP are accounted for as costs of production. This report contains quantitative analysis to address that issue, and the findings are presented in Table ES-1. That table shows that risk costs would far exceed insurance premiums. There is no other mechanism to incorporate risk costs as production costs. Thus, most of the risk costs would be borne by the public at large, representing a large, implicit subsidy of nuclear power. For example, Table ES-1 shows that operation of an existing NPP in Canada creates offsite risk costs of 2.7 to 5.4 cent per kWh of electricity produced, plus additional onsite risk costs of 2.7 to 5.6 cent per kWh. Under the NLCA, the only evident accounting for these risk costs as costs of production would be through payment of insurance premiums amounting to 0.02 cent per kWh, to cover offsite damage to third parties up to a $650 million limit. Thus, the public now provides, and would provide under the NLCA, an implicit subsidy to NPP operators of 5.4 to 11.0 cent per kWh. In 2007, the total, implicit subsidy across all Canadian NPPs was $4.8 billion to 9.7 billion for the year. (Canadian nuclear generation was 88.2 billion kWh in 2007.) Such a large subsidy is a significant violation of the polluter-pays principle.

Table ES-1
Risk Costs of Nuclear Generation in Canada: summary of this Reports findings
Note: This table, with notes, appears in the body of the report as Table 9-8. Magnitude of Risk Costs and Insurance Premiums for an Existing CANDu Plant 2.7 to 5.4 0.02 2.7 to 5.6 No explicit premium is evident for a New Generation iii Plant 1.5 to 15.4 As for existing CANDU plant? Smaller amount than for existing CANDU plant No explicit premium is evident

CAtEGoRy of iMPACts fRoM uNPLANNED RELEAsEs of RADioACtiVE MAtERiAL

Category of Risk Costs and the insurance Premiums that are Paid to Provide Coverage of these Costs Risk Costs (2009 Can cent per kWh) Insurance Premiums Under the NLCA (2009 Can cent per kWh) Risk Costs (2009 Can cent per kWh) Insurance Premiums Under the NLCA (2009 Can cent per kWh)

offsite Impacts

onsite Impacts

The analysis underlying Table ES-1 involves various uncertainties and assumptions. Other analysts might make different assumptions, yielding different findings. However, the assumptions made here are reasonable, and the findings are robust. Risk costs exceed insurance premiums by a factor of 270 to 550. Although limited information is publicly available regarding the premiums set by nuclear insurers, accessible data show that nuclear insurers make assumptions about nuclear risk that are consistent with the assumptions made here.

In regard to Criterion #4, the NLCA would not be compatible with coherent and consistent application of the precautionary principle in Canada. Underlying the NLCA is an assumption that consideration of the risk of unplanned radioactive releases from NPPs should, in the practical context of liability and insurance, be confined to comparatively small releases. That assumption is implicit in the choice of $650 million as a liability limit, and is confirmed by official sources. Yet, that assumption is contradicted by studies conducted by the Canadian nuclear industry, analogous studies conducted in other countries by industry and government, experience with the 1986 Chernobyl accident and the 1979 Three Mile Island accident, and a well-recognized potential for malevolent acts at NPPs. Those sources of information show that damage far in excess of $650 million is a realistic possibility. The NLCA ignores that possibility, and would, therefore, be incompatible with the precautionary principle. Moreover, the NLCAs liability limit has no basis in either systematic technical analysis or public debate, both of which are central to application of the precautionary principle. A relevant technical analysis the Severe Accident Study was initiated by the Canadian government in the late 1980s but never completed. In regard to Criterion #5, the NLCA would not be compatible with a transition of Canadas economy toward sustainability. There are at least three major aspects of this incompatibility. First, the NLCA would violate the precautionary principle, thereby tending to suppress policies and actions flowing from that principle. As a result, opportunities to reduce the risk of NPP operation would be lost, and Canada would bear an unnecessarily high risk of a radioactive release with devastating impacts on the environment, economy, and society. Second, the NLCA would create a large, implicit public subsidy for nuclear power, thereby distorting decisions about public and private investment in systems for production and use of energy. In that way, the NLCA would suppress innovation in sustainable energy systems, and would inhibit market processes for cost-effective deployment of such systems. Third, the NLCA would continue the present climate of incomplete and often misleading information about nuclear risks and nuclear insurance. That climate is antithetical to the open, science-based, and participatory exchange of information that is necessary for Canadas transition to sustainability. The polluter-pays principle was upheld unanimously by the Supreme Court of Canada in a 2003 decision. The precautionary principle has important roles in the Canadian Environmental Assessment Act of 1992 and the Federal Sustainable Development Act of 2008. Yet, the NLCA would violate both the polluter-pays principle and the precautionary principle. Thus, the NLCA would be incompatible with existing Canadian law, and may be incompatible with Canadas international commitments. Also, the NLCA would create subsidies for nuclear generation of electricity. One subsidy is implicit in the $650 million liability limit. Another subsidy would be the Canadian governments reinsurance of risk that the nuclear insurers refuse to cover. Both subsidies may be incompatible with Canadas existing laws and international commitments.

optIonS foR ImpRovIng the nlcA The preceding paragraphs identify significant deficiencies in the proposed NCLA. Options that could address those deficiencies, to varying extents, include: option #1: Disclosure of nuclear insurance data The NLCA would require the timely publication of data on nuclear insurance coverage, premiums, reinsurance, and compensation. option #2: full assessment of NNP risk Voting by Parliament on the NLCA would be preceded by publication of a comprehensive assessment of the risk posed by NPP operation in Canada. As part of this option, the NLCA could provide for its own review at specified intervals, preceded each time by an updated NPP risk assessment. option #3: full commercial insurance The NLCA would prohibit nuclear reinsurance by the Canadian government. option #4: increased liability limit The NLCA would have a third-party liability limit much higher than $650 million, or no limit. option #5: Coverage of onsite damage The NLCA would require NPP operators to have commercial insurance coverage for onsite damage, including loss of electricity production. option #6: Economic channeling The NLCA would require economic channeling of liability, as under the Price-Anderson Act. option #7: operator-pooled, second-tier coverage The NLCA would require all Canadian NPP operators to participate in pooled, second-tier coverage via retroactive premiums, as under the Price-Anderson Act. A variant of this option would involve all US NPP licensees and all Canadian NPP operators participating in a common pool that provides second-tier coverage.

mAjoR RecommendAtIonS The following recommendations are offered: R1. The NLCA should not be enacted in its present form. R2. The Canadian government should prepare a comprehensive assessment of the risk posed by NPP operation in Canada, the opportunities for reducing that risk, and the accompanying risk costs and riskreduction costs. R3. The Canadian government should prepare a legal analysis of the compatibility of the NLCA with Canadas existing laws and international commitments. R4. The Canadian government should prepare a systematic study of options for improving the NLCA, to include the options outlined in this report. R5. Parliament should sponsor a public event at which representatives of a wide range of stakeholders would discusses the analyses called for in recommendations R2 through R4, and would present their own analyses. R6. A revised NLCA should be prepared, informed by the public event called for in recommendation R5.

Table of Contents
1. Introduction ...............................................................................................................................................................................................................................7 2. An Outline of the NLCA and Nuclear Insurance in Canada...............................................................................................................9 3. Issues Relevant to Assessing the NLCA: Sustainability, Precaution, & Full Accounting ........................................ 12 3.1 Imperatives and Principles of Sustainability ...................................................................................................................... 12 3.2 Prudence, Uncertainty, & the Precautionary Principle .............................................................................................. 14 3.3 Full Accounting of Costs .................................................................................................................................................................. 16 3.4 Criteria for Assessing the NLCA ................................................................................................................................................ 17 4. Nuclear Liability from Global & North American Perspectives..................................................................................................... 18 5. Assessing the Risk Posed by Nuclear Power Plants.......................................................................................................................... 22 6. Options for Reducing the Risk Posed by Nuclear Power Plants .............................................................................................. 25 7. Risk Posed by Nuclear Power Plants in Canada ................................................................................................................................... 26 7.1 Scope of this Discussion ................................................................................................................................................................. 26 7.2 Types of Nuclear Power Plant that Could Operate in Canada .......................................................................... 26 7.3 Potential Radioactive Releases and their Offsite Impacts ..................................................................................... 26 7.4 Potential Onsite Impacts of Fuel-Damage Events ....................................................................................................... 27 8. The Canadian Governments Consideration of Risk-Related Costs ...................................................................................... 28 9. This Reports Estimation of Risk-Related Costs ..................................................................................................................................... 29 9.1 Scope of this Discussion ................................................................................................................................................................. 29 9.2 Costs of Measures Intended to Reduce Risk ................................................................................................................. 29 9.3 Risk Costs of Offsite Impacts of Radioactive Releases .......................................................................................... 30 9.4 Risk Costs of Onsite Impacts of Fuel-Damage Events ........................................................................................... 32 9.5 An Overview of Risk Costs ............................................................................................................................................................. 32 10. Regulatory Issues and Secrecy......................................................................................................................................................................... 33 11. Deficiencies in the NLCA, and Options for Improvement ............................................................................................................ 36 12. Conclusions and Recommendations ........................................................................................................................................................... 39 13. Bibliography ...................................................................................................................................................................................................................... 40

Tables, Figures, and Appendix (See next pages)

List of Tables and Figures (The Tables and Figures appear after Section 13.)
tAbleS
Table 1-1: Classification of Potential Accidents and Malfunctions at a Nuclear Power Plant Table 3-1: The Twelve Principles of Green Engineering, According to Anastas and Zimmerman Table 5-1: Some Potential Modes and Instruments of Attack on a Nuclear Power Plant Table 5-2: The Shaped Charge as a Potential Instrument of Attack Table 5-3: Safety Goals for a New Nuclear Power Plant, as Specified in CNSC Regulatory Document RD-337 Table 6-1: Selected Options to Reduce the Risk of a Spent-Fuel-Pool Fire at a Nuclear Power Plant that Employs High-Density Pool Storage Table 7-1: Radioactive Releases and Offsite Impacts for the 1986 Chernobyl Accident and Some Potential Accidents at Nuclear Power Plants: Selected Data Table 7-2: Licensee Estimates of Frequency and Population Dose for Potential Accidental Atmospheric Releases of Radioactive Material from Some CANDU Stations in Ontario Table 7-3: Ontario Hydro Estimate of the Risk of Onsite Economic Impacts from Fuel-Damage Events at the Darlington Nuclear Power Plants (Existing CANDU Plants) Table 9-1: Estimation of Cost to Transfer Spent Fuel from a PWR Spent-Fuel Pool to Dry Storage After 5 Years of Storage in the Pool Table 9-2: Risk Indicators for Potential Accidental Atmospheric Releases of Radioactive Material from the Bruce A Station Table 9-3: Risk Costs of Offsite Impacts of Accidental or Malevolent Releases of Radioactive Material from Nuclear Power Plants in Canada: Selected Cases Table 9-4: Range of Risk Costs of Offsite Impacts of Accidental and Malevolent Releases of Radioactive Material from Nuclear Power Plants in Canada: Existing or New Plants Table 9-5: Selected Range of Risk Costs of Offsite Impacts of Accidental and Malevolent Releases of Radioactive Material from Nuclear Power Plants in Canada: Existing or New Plants, Excluding Releases from Spent-Fuel Pools Table 9-6: Policy-Applicable Risk Costs of Offsite Impacts of Accidental or Malevolent Releases of Radioactive Material from Nuclear Power Plants in Canada Table 9-7: Risk Costs of Onsite Impacts of Fuel-Damage Events at Existing CANDU Plants, Using an Ontario Hydro Estimate of the Risk of Economic Impacts at the Darlington Plants Table 9-8: Risk Costs of Nuclear Generation in Canada: Summary of this Reports Findings Table 10-1: Selected Approaches to Protecting Canadian Critical Infrastructure From Attack by Sub-National Groups, and Some Strengths and Weaknesses of these Approaches

fIguReS
Figure 5-1: Core Damage Frequency for Accidents at a Surry PWR Nuclear Power Plant, as Estimated in the NRC Study NUREG-1150 Figure 5-2: Core Damage Frequency for Accidents at a Peach Bottom BWR Nuclear Power Plant, as Estimated in the NRC Study NUREG-1150 Figure 5-3: Conditional Probability of Containment Failure Following a Core-Damage Accident at a Surry PWR or Peach Bottom BWR Nuclear Power Plant, as Estimated in the NRC Study NUREG-1150
6

Appendix (This Appendix appears after the Tables and Figures.)

Designing Nuclear Power Plants to Pose a Comparatively Low Level of Risk

1. Introduction
A bill has been introduced into Canadas parliament to create the Nuclear Liability and Compensation Act (NLCA).1 The new Act would replace the existing Nuclear Liability Act. Both Acts pertain to civil liability for damage arising from an incident at a nuclear installation. Under the new Act, the liability of the operator of the affected installation would be limited to $650 million, compared to the limit of $75 million under the present Act. This report assesses the proposed NLCA. In performing that task, the report focuses on the application of the NLCA to nuclear power plants (NPPs) in Canada. 2 The report further focuses on potential incidents at NPPs where the incident involves an unplanned release of radioactive material to the environment or within the plant. Those foci reflect the fact that an unplanned release of radioactive material from or within an NPP could cause damage substantially exceeding the damage arising from other types of incident covered by the NLCA.3 In assessing the NLCA, this report indirectly assesses the existing Nuclear Liability Act. Applying the assessment criteria used here, the existing Act is equivalent to or weaker than the NLCA. Thus, deficiencies identified in the proposed NLCA are present to an equal or greater extent in the present Act. cRIteRIA foR ASSeSSIng the nlcA The major purpose of the NLCA would be to facilitate the future operation of NPPs in Canada. Those NPPs could be existing or new plants.4 Historically, national governments adopted legislation on nuclear liability because the private insurance sector was unwilling to extend coverage to NPPs. Governments took that action to facilitate the growth of what was then a new industry. Now, nuclear power is a mature technology that has been in use for half a century. If this technology is to remain in use over the coming decades, it should do so within a policy framework that is appropriate for the 21st century. The NLCA should be assessed in the context of such a framework. Detailed articulation of a 21st century policy framework for nuclear power is a task beyond the scope of this report. It is clear, however, that three concepts would play major roles in such a framework. The concepts are: sustainability; the precautionary principle; and full accounting of costs (which includes the polluter-pays principle). Section 3 of this report outlines each concept, and sets forth criteria, pertinent to each concept, that are used here to assess the NLCA. typeS of AnAlySIS uSed heRe This reports assessment of the NLCA according to sustainability and precautionary-principle criteria uses qualitative analysis. In applying full-accounting criteria, however, this report employs both qualitative and quantitative analysis. The quantitative analysis estimates the risk of operating an NPP, assigns that risk a monetary value, and determines the extent to which the monetized risk would receive insurance coverage under the NLCA. Monetized risk that is not covered by insurance can be designated as uninsured risk cost or as externality cost. It should be noted that the risk of operating an NPP has some attributes that cannot be quantified. Moreover, quantitative estimation of risk involves uncertainty and assumptions. Also, assigning a monetary value to a unit of quantitatively-estimated risk involves assumptions. In the quantitative risk analysis that is presented here, various assumptions and estimates are used. Some of these assumptions and estimates originate with the Canadian nuclear industry, and some are by the author. The findings should be viewed as illustrative in both cases. Other analysts might employ different assumptions, and their numerical findings would differ accordingly. Ultimately, assessment of the NLCA must be qualitative.

7 1 2 Becklumb and Dufresne, 2009; Henault, 2009. The current bill is designated as bill C-20. It is functionally identical to bills C-63 and C-5 that were introduced during the 39th parliament but not enacted. Throughout this report, the term nuclear power plant means a fission reactor and its associated equipment, including equipment to produce electricity. Future NPPs of the Generation IV type, if any operate in Canada, might also produce hydrogen, potable water, and/ or process heat. At three sites in Ontario (Pickering, Bruce, and Darlington), NPPs are clustered in groups of four to make up nuclear generating stations. In those instances, the individual NPPs are often described as units. NPPs would be less dominant in terms of potential damage if a uranium enrichment plant, nuclear fuel reprocessing plant, or high-level radioactive waste repository became operational in Canada. There is discussion about building new NPPs in Canada. At present, no order has been placed for a new NPP.

3 4

defInIng And cAtegoRIzIng RISk The term risk is used here to refer to the potential for, and impacts of, unplanned releases of radioactive material to the environment or within an NPP.5 Such releases could occur as a result of accidents or malevolent acts, as discussed below. The releases would be unplanned in the sense that they would not be expected to occur during routine operation. Nevertheless, the potential for occurrence of unplanned releases is acknowledged by the nuclear industry and its regulators, in Canada and elsewhere. A more precise term for the type of risk considered here would be residual risk. In that term, the word residual characterizes the risk that remains after the nuclear industry has complied with the requirements imposed by its regulators. In Canada, the primary regulator of the safety and security of NPPs is the Canadian Nuclear Safety Commission (CNSC). This report examines risk with two assumptions. First, construction and operation of NPPs in Canada consistently reflect good-faith efforts by the nuclear industry to comply with CNSC regulations. Second, CNSC personnel consistently make good-faith efforts to enforce CNSC regulations. An unplanned release of radioactive material could occur as a result of an accident or a malevolent act. The Canadian Environmental Assessment Agency (CEAA) acknowledges the need to consider both types of incident in an environmental impact statement (EIS) for the construction of a new NPP. CEAA describes such incidents as accidents and malfunctions, which are defined as a category of events that includes accidents of a traditional type (events attributable to human error, natural phenomena, etc.) together with intentional, malevolent acts. Consideration of malevolent acts in an EIS for a commercial nuclear installation is a comparatively new development in the field of environmental assessment. CEAA has published EIS guidelines that contain a classification of accidents and malfunctions at an NPP. 6 The CEAA classification has been refined by this author, as shown in Table 1-1. This report focuses on accidents and malfunctions in the first two rows of Table 1-1. Those rows pertain to an unplanned release of radioactive material from the reactor core or from spent fuel. Such events dominate the potential for damage that is relevant to the NLCA. The damage could occur within an NPP or in the surrounding environment. In addressing damage in the surrounding environment, this report focuses on potential unplanned releases to the atmosphere. The released material would be carried downwind in a radioactive plume, and some of that material would be deposited from the plume onto vegetation, buildings, the ground surface, etc. SecRecy The nuclear industry, and the bodies that regulate its safety and security, are prone to secretive behavior. Such behavior reflects a variety of motives. The nuclear industry has a legitimate need to protect trade secrets, but has also been known to hide embarrassing information that should be disclosed. Regulators have often chosen to not assess, or not disclose, the upper end of the range of risk associated with nuclear installations, reflecting a paternalistic view of the publics ability to use such information. In recent years, as the potential for intentional, malevolent acts has become a salient issue, industry and regulators have become more secretive. This trend reflects the fact that each existing and currently-proposed NPP in Canada and elsewhere has a limited ability to ride out an attack by a well-prepared, sub-national group.7 The nuclear industry and regulators have two, broad motives for secrecy regarding the risk of malevolent acts at NPPs. First, they do not wish to encourage an attack. Second, they are reluctant to admit their failure, over several decades, to ensure that resistance to attack is a priority in NPP design. The first motive matches the public interest, but the second does not. While NPPs continue to operate, it is difficult to disentangle the two motives. Thus, excessive secrecy persists. As a result, public discussion of nuclear risk and nuclear insurance does not properly address the threat of attack. This report strikes a balance in discussing that threat. The report describes, in a general way, the risk of malevolent acts at NPPs, but does so without disclosing information that would assist potential attackers.

5 6 7

The term risk is often used to refer to the arithmetic product of: (i) a quantitative indicator of adverse impact; and (ii) the quantitative probability that the impact will occur. In this report, the term is used in a more general sense, to encompass a range of qualitative and quantitative information about the potential for an adverse outcome. CEAA at al, 2008. That document was prepared by CEAA and the Canadian Nuclear Safety Commission, in consultation with Fisheries and Oceans Canada, Transport Canada, and the Canadian Transportation Agency. A sub-national group of attackers is a group that can operate without support or direction from a nation-state.

StRuctuRe of thIS RepoRt The remainder of this report has twelve sections. Section 2 describes the NLCA and the nuclear insurance sector in Canada. Section 3 develops criteria for assessing the NLCA, drawing upon three high-level issues related to nuclear power: sustainability; the precautionary approach; and full accounting of costs. Section 4 provides global and North American perspectives on nuclear liability and insurance. Then, Section 5 discusses the general problem of assessing the risk posed by NPPs, and Section 6 discusses options for reducing that risk. Section 7 examines the risk posed by NPPs in Canada. Risk-related costs are then discussed from two perspectives. Section 8 outlines the Canadian governments consideration of risk-related costs, while Section 9 provides this reports estimation of those costs. Section 10 discusses regulatory issues, secrecy, and their relevance to the NLCA. Drawing upon all the preceding sections, Section 11 summarizes the major deficiencies in the NLCA that are identified in this report, and discusses options for correcting those deficiencies. Conclusions and recommendations are set forth in Section 12, and a bibliography is provided in Section 13. All documents cited in this report, whether in the footnotes to the narrative or in the reports tables and figures, are listed in the bibliography. Tables and figures, numbered according to the relevant section of the report, appear after Section 13. An appendix appears at the end of the report. This Appendix, which is especially relevant to Section 6, shows how NPPs could be designed to pose a comparatively low level of risk.

2.An Outline of the NLCA and Nuclear Insurance in Canada

The proposed Nuclear Liability and Compensation Act would replace the existing Nuclear Liability Act. Major provisions of the NLCA are outlined here, drawing upon a summary by the Parliamentary Information and Research Service and another summary by an advisor to Natural Resources Canada.8 Both the existing Act and the proposed NLCA make the operator of a nuclear installation exclusively liable for damage arising from an incident at the installation. The liability is absolute, which means that a victim does not need to prove negligence in order to make a claim. The operators exclusive liability is through legal channeling, which differs from the economic channeling that occurs in the USA under the PriceAnderson Act. The difference between these modes of channeling is discussed in Section 4, below. A major difference between the NLCA and the present Act is the NLCAs higher limit on liability. Under the NLCA, the operators liability would be limited to $650 million, compared to the limit of $75 million under the present Act. The $650 million limit could be increased by regulation, and must be reviewed every five years by the responsible minister. The rationale for a $650 million limit has been described by an advisor to Natural Resources Canada as follows:9 The $650 million limit reflects a balance of considerations. First, it addresses foreseeable, rather than catastrophic Chernobyl-type accidents. Second, it reflects insurance capacity that can be available at reasonable costs. It puts Canada on par with liability limits in many other countries. And, finally, it is responsive to recommendations of the Standing Senate Committee on Energy, Environment and Natural Resources which, in 2001, recommended that the operator liability limit be increased from the current 75 million dollars to an amount in line with the Paris and Vienna Conventions of over 600 million dollars. There are various other differences between the NLCA and the present Act. For example, the NLCA is more explicit in stating that Parliament can appropriate funds for additional victim compensation beyond the $650 million limit, should that be necessary. Also, the NLCA would extend the claims limitation period to 30 years for bodily injury and death, while continuing the present 10-year claims limitation period for property damage. The NLCA would expand the definition of compensable damage. Under the NLCA, compensable damage would include bodily injury, damage to property, and psychological trauma resulting from those effects. Also included would be economic losses arising from bodily injury, property damage, or psychological trauma, together with costs arising from loss of use of property. However, costs arising from an affected nuclear installations failure to provide electricity would not be compensable. Costs of measures to repair or mitigate environmental damage would be compensable if the measures are ordered by a government agency.

8 9

Becklumb and Dufresne, 2009; Henault, 2009. Henault, 2009.

ApplIcAtIon of the nlcA to mAlevolent ActS The NLCA would not apply to an incident resulting from an act of war, hostilities, a civil war, or an insurrection. It would, however, apply to an incident arising from terrorist activity, as defined in subsection 83.01(1) of the Criminal Code. That definition encompasses a wide range of incidents, with the following specific exception:10 [Terrorist activity] does not include an act or omission that is committed during an armed conflict and that, at the time and in the place of its commission, is in accordance with customary international law or conventional international law applicable to the conflict, or the activities undertaken by military forces of a state in the exercise of their official duties, to the extent that those activities are governed by other rules of international law. ApplIcAtIon of the nlcA to onSIte dAmAge According to the Parliamentary Information and Research Service, the NLCA would not apply to damage to a nuclear installation that experiences an incident, if the operator is responsible for that damage. It is not clear, however, that the NLCA would apply to this type of damage if the operator were not responsible. Such a situation could arise, for example, if the incident were caused by a natural event (e.g., an earthquake) or a malevolent act.11 It appears that there is no specific provision in Canadian law or insurance practice to address liability or compensation for onsite damage at an NPP that experiences an unplanned release of radioactive material. More specifically, this author has not identified such a provision. Apparently, Canadian operators of NPPs self-insure for onsite damage. This practice was established during a period of several decades when all the NPP operators were provincially-owned corporations. More recently, a private entity Bruce Power has become an NPP operator.12 Bruce Powers insurance arrangements for onsite damage are unclear.

nucleAR InSuRAnce In cAnAdA Three pools of insurance companies provide nuclear insurance in Canada. Each pool is approved under the existing Nuclear Liability Act, and has held that position since 1976. The pools are: Nuclear Insurance Association of Canada (NIAC); American Nuclear Insurers (ANI), a US pool; and Nuclear Risk Insurers (NRI), a UK pool. NIAC, the Canadian pool, consists of domestic insurance companies and Canadian subsidiaries or branches of foreign insurance companies. In Canada, NIAC is the primary insurer, and ANI and NRI are co-insurers. A substantial fraction of NIACs present coverage of NPP operators is re-insured with the Canadian government. In 2008, the government was providing $585 million of third-party liability coverage for Canadas 18 operating NPPs, which amounts to an average of $32 million per NPP, or 43 percent of the $75 million liability limit.13 Re-insurance by the Canadian government would continue under the NLCA, in order to provide coverage for risks that insurers are unwilling or unable to cover.14 For this purpose, the government would continue to operate the Nuclear Liability Reinsurance Account. Any deficit in that account would be paid from the Consolidated Revenue Fund. In other words, the Canadian government would provide nuclear reinsurance from general revenue.

10

10 Definition accessed on 19 October 2009 at: <http://www.canlii.org/en/ca/laws/stat/rsc-1985-c-c-46/latest/rsc-1985-c-c-46.html>. 11 In such a situation, it might be argued that the operator had failed to ensure that the installation could ride out a foreseeable natural event or malevolent act, and was therefore responsible. 12 Bruce Power operates, but does not own, the Bruce A and Bruce B nuclear generating stations in Ontario, and has proposed the construction of new NPPs. 13 CNSC, 2008d, page 73. 14 Henault, 2009.

The Canadian nuclear insurers have made it clear that they would not provide coverage under the NLCA without reinsurance by the Canadian government. At an October 2009 conference in Toronto, their representatives stated:15 Insuring all of the heads of damage under the NLCA for the full Cdn. $650 million limit will be challenging. We understand that countries that are signatories to the Conventions face similar challenges. Insurers have expressed concerns with insuring a number of the new heads of damage under the NLCA. Of particular concern is environment impairment. However, there are also concerns with respect to economic loss, psychological trauma and terrorism. Fortunately, in Canada, the Federal Government already reinsures a portion of the exposure under the current NLA and has expressed a willingness to continue to reinsure a portion of the exposure under the NLCA. Without such reinsurance, it would be impossible to insure the full exposure under the NLCA.

opeRAtIonAl dAtA About nucleAR InSuRAnce: pRemIumS, etc. There is a dearth of publicly-available operational data for the nuclear-insurance sector in Canada. No information about coverage of onsite damage at NPPs could be obtained for this report. Sparse information was obtained about premiums paid for third-party liability coverage, as discussed below. An attempt to obtain information about nuclear insurance was made in 2008 in the context of a proceeding (EB-2007-0707) before the Ontario Energy Board. Relevant interrogatories were submitted to the Ontario Power Authority (OPA) by a group of intervenors.16 OPA provided no information in response to those interrogatories. A research paper sponsored by the Canadian government states that a third-party liability premium of $125,000 per RY was paid in 1995 for the Darlington nuclear generating station.17 That premium can be adjusted upward to 2009 $ by a factor of 1.36, using the CPI inflator published by Bank of Canada.18 A further upward adjustment can be made by a factor of 650/75 to account for an increase in the liability limit from $75 million to $650 million. Those adjustments yield an estimated premium of $1.47 million (2009 $) per reactor-year (RY) for third-party liability coverage for a Canadian NPP under the proposed NLCA. That estimate is used in this report, in the absence of official data. In view of the extrapolation from 1995 data, we assume that terrorist-risk coverage is not included in this estimated premium. After the attacks of September 2001 on New York and Washington, the NIAC insurers were unwilling to provide full coverage for third-party liability associated with terrorist incidents. Since 2003, the Canadian government has provided the bulk of that coverage through reinsurance. In 2006, for example, the Canadian government provided 80 percent of the terrorist-risk coverage for Canadas 18 operating NPPs for an aggregate premium of $280,000.19 That amount is equivalent to a premium of $280,000/(18 x 0.8) = $19,400 per RY (2009 $). The latter premium can be adjusted upward by a factor of 650/75 to account for an increase in the liability limit, resulting in an estimated premium of $0.17 million per RY (2009 $) for terrorism-risk third-party liability coverage under the proposed NLCA. The two preceding paragraphs yield an estimated total premium of $1.64 million per RY (2009 $) for coverage of an NPP under the NLCA.20 Assuming a plant capacity of 0.88 GWe and a capacity factor of 0.9, that premium translates to an amount of 0.024 cent per kWh of nuclear generation (2009 Can cent).

15 Murphy et al, 2009. 16 The group consisted of the Green Energy Coalition, the Ontario Sustainable Energy Association, and the Pembina Institute. The relevant interrogatories were numbered 91 through 99. 17 Heyes and Heyes, 2000, page 93. 18 It could be argued that the premium should not be adjusted for inflation, because the $75 million liability limit has remained unchanged. That argument neglects profit and administrative costs, and the insurers expectations regarding claims below the liability limit. Without an inflation adjustment for the premium, this reports estimate of uninsured risk costs would rise slightly. 19 Lunn, 2007. 20 Given perfect forecasting and no profit or administrative costs, an insurance premium of $1.64 million per RY for coverage of $650 million would imply an event probability of 2.5 per 1,000 RY.

11

3. Issues Relevant to Assessing the NLCA: Sustainability, Precaution, & Full Accounting
3.1 Imperatives and Principles of Sustainability
During recent decades, citizens and governments have increasingly recognized the need to organize human affairs within the context of a finite Earth. One manifestation of that need is human-induced, adverse change in the climate.21 Other signs of stressed ecosystems are also evident. The Millennium Ecosystem Assessment determined that 15 out of the 24 ecosystem services that it examined are being degraded or used unsustainably, including fresh water, capture fisheries, air and water purification, and the regulation of regional and local climate, natural hazards, and pests.22 By abusing ecosystems in this manner, we deplete renewable resources that are essential to human life. Nonrenewable resources are also being depleted. For example, a growing body of analysis predicts a peak in world oil production within the next few decades.23 In our well-populated, competitive world, limits to the availability of resources and ecosystem services have implications for peace and security. For example, analysts are considering the potential for climate change to promote, through its adverse impacts, social disorder and violence.24 It is increasingly evident that nations must cooperate to protect and share the Earths resources. International agreements such as the Framework Convention on Climate Change reflect that imperative. National policies on a range of issues including energy, agriculture, forestry, transport, minerals, and urban planning must be consistent with global needs. Engineers must develop new design approaches, as shown in Table 3-1. Policy choices made now will determine the opportunities available to future generations. The future implications of current policy choices have been examined by analysts convened by the Stockholm Environment Institute (SEI).25 These analysts identified six possible worldwide scenarios for human civilization over the coming century and beyond. In some scenarios, the world faces chronic, unresolved problems and conflicts. In others, the world descends into barbarism. The most attractive scenario, with the greatest opportunities for future generations, is one that the SEI analysts described as a New Sustainability Paradigm. The concept of sustainability was brought to wide public attention by the World Commission on Environment and Development in 1987. WCED discussed the concept in terms of sustainable development, to emphasize that sustainability is compatible with improvement in the conditions of life for poorer societies. Since 1987, the concept of sustainability has been widely endorsed by governments and other entities. Yet, there has been comparatively little progress in making the concept operational at the level of specific policies and plans. In an effort to address that problem, the Organization for Economic Cooperation and Development (OECD) initiated a three-year project in 1998, seeking to identify sustainability principles and indicators that can be used in policy making. One product of the effort was a report by the OECD Nuclear Energy Agency (NEA), published in 2000, that discussed commercial nuclear power in the context of sustainable development.26 the neA vIew of SuStAInAbIlIty In discussing the concept of sustainability, the NEA report took as its starting point the WCED definition of sustainable development as development that meets the needs of the present without compromising the ability of future generations to meet their own needs. The NEA report elaborated on that definition by suggesting that sustainability involves the passing on to future generations of a stock of capital assets, which could be human-made, natural, or human and social. Human-made assets include buildings, machinery, and infrastructure. Natural assets include the environment, and the renewable and non-renewable resources that it can supply. Human and social assets include education, health, scientific and technical knowledge, cultures, institutions, and social networks.
12

21 22 23 24 25 26

IPCC, 2007. MEA, 2005, page 1. GAO, 2007. Campbell et al, 2007. Raskin et al, 2002. NEA, 2000.

According to the NEA, strong sustainability involves the preservation of an asset in its present form. That approach is relevant, for example, to ecosystems that are essential and irreplaceable. Earths atmosphere fits that category. An alternative approach is weak sustainability, whereby the loss of one asset (e.g., an area of forested land) is offset by creation of another asset (e.g., development of a city on the formerly forested land). The weak-sustainability approach requires tradeoffs, which create the potential for conflicts within and between generations. The strong-sustainability approach is conceptually simpler, but is rarely encountered in its pure form. For example, human-induced emissions of CO2 to the atmosphere cannot be eliminated instantly, but must be reduced over time. With the best of intentions, we cannot pass on to the next several generations an atmosphere containing CO2 at the present concentration. The NEA report contained a general discussion of nuclear power from the perspective of sustainability. That discussion addressed many of the relevant issues, including emissions of CO2 and other greenhouse gases. The NEA report did not, however, provide an analytic framework that could be used to assess the sustainability of a proposed program of nuclear power, or to compare the sustainability of that program and the sustainability of other strategies to meet energy needs. The NEA report discussed the potential for an NPP to experience a large, unplanned release of radioactive material to the environment. Such a release would substantially degrade human-made, natural, human, and social assets in the affected locations. For example, contaminated land and buildings would be abandoned, and exposed populations would experience higher rates of cancers. Thus, the release could have significant, adverse effects on sustainability. According to the NEA, the probability of a large, unplanned release is low. Thus, a conceptual and analytic framework is needed to assess the sustainability implications of potential events with large, adverse consequences and low probability. That issue is not unique to nuclear power. For example, capture and sequestration of CO2 is an energy option that could allow use of fossil fuels without adverse effects on Earths climate. If a proposed project involves CO2 sequestration on a large scale, an assessment of the sustainability of the project should examine the probability and consequences of a large, unexpected release of sequestered CO2 to the atmosphere. Analysts in the nuclear industry and its regulatory bodies address the potential for high-consequence, lowprobability events by defining an indicator called risk. They typically define that indicator as the arithmetic product of a numerical indicator of consequences and a numerical indicator of probability.27 They frequently argue that equal levels of risk should be equally acceptable to citizens. That argument has been made so often that it has become dogma. Yet, the argument is not a scientific statement. It is, instead, a statement representing a particular set of values and interests. The NEA report recognized that citizens may be more concerned about the potential for a highconsequence, low-probability event than about the potential for a low-consequence event with the same nominal level of risk. That concern can reflect a legitimate set of values and interests, skepticism about estimates of low probability, doubt that the complexity of consequences can be represented by simple indicators, and recognition that new phenomena can come into play when thresholds of consequence are exceeded.28 The NEA report recommended that concerns of this type be heard, respected, and addressed by governments. Acceptance of that recommendation would require the abandonment of previous dogma about the acceptability of risk. Public-engagement processes and research could then be used to develop a new paradigm of risk and its acceptability. cAnAdAS commItment to SuStAInAbIlIty Canada is officially committed to the principles of sustainability through the Federal Sustainable Development Act, which states at paragraph 5: 29 The Government of Canada accepts the basic principle that sustainable development is based on an ecologically efficient use of natural, social and economic resources and acknowledges the need to integrate environmental, economic and social factors in the making of all decisions by government.
13

27 In this report, the term risk is used in a more general sense, to encompass a range of qualitative and quantitative information about the potential for an adverse outcome. 28 There is evidence that the consequences of the 1986 Chernobyl reactor accident exceeded thresholds that brought new social and political phenomena into play. Many observers conclude that the accident undermined the legitimacy of the USSR government, contributing significantly to the breakup of the USSR. See, for example: Parsons, 2006. 29 Accessed on 22 October 2009 at: <http://laws.justice.gc.ca/en/showtdm/cs/F-8.6>.

3.2 Prudence, Uncertainty, & the Precautionary Principle

A prudent citizen or public official will always give careful consideration to potential adverse outcomes of a proposed action. If the proposed action is the construction and operation of an NPP, two potential adverse outcomes will be particularly salient. One potential outcome, as mentioned above, would be a large, unplanned release of radioactive material from the plant to the environment. The second potential outcome would be the diversion of fissile or radioactive material from the plant by a national government or a sub-national group, for use in nuclear or radiological weapons. The organizations that construct, operate and regulate nuclear power plants will implement measures intended to prevent these outcomes. Nevertheless, a prudent observer will consider the possibility that the preventive measures will fail. In view of the severe, adverse impacts that would be associated with these outcomes, it would be imprudent to ignore the possibility of their occurrence.30 ASSeSSIng the potentIAl foR An unplAnned ReleASe The potential for a large, unplanned release of radioactive material is typically regarded by the nuclear industry and its regulators as a safety issue. An analytic art, known as probabilistic risk assessment (PRA), has been developed to estimate the probabilities and consequences of potential releases. The first PRA for an NPP was known as the Reactor Safety Study, and was published by the US Nuclear Regulatory Commission (NRC) in 1975.31 A PRA for an NPP considers a range of scenarios (event sequences) that involve damage to the reactor core. The initiating events are categorized as internal events (human error, equipment failure, etc.) or external events (earthquakes, fires, strong winds, etc.). The core-damage scenarios that arise from these events are termed accidents. PRAs typically do not consider initiating events that involve intentional, malevolent acts, although PRA techniques can be adapted to estimate the outcomes of such acts. The NRC adapted PRA techniques in developing its 1994 rule requiring protection of each NPP against attack using a vehicle bomb.32 A modern NPP has safety features reactor shut-down systems, core cooling systems, etc. with independent, redundant and diverse components. A core-damage accident at such a plant would, in many potential cases, involve a combination of independent failures that coincide, thereby overcoming the plants safety features.33 By contrast, during an intentional attack on an NPP, the plants safety features would be challenged by a common factor the attackers intellectual and practical capabilities. Attackers with the motivation and resources to mount a significant attack would be likely to plan the attack with the specific intention of overcoming the plants safety features and causing a large radioactive release. Attacks on buildings in New York and Washington in September 2001 demonstrated that an attack on a civilian facility by a skilled, highly-motivated and well-resourced sub-national group is a credible event. Many observers agree that an NPP is a potential target of a future attack of this kind. Faced with this threat, risk analysts, regulatory bodies and plant designers must modify their approach. The traditional safety paradigm is insufficient. To understand the threat, risk analysts must think like skilled attackers. Regulatory bodies must capture those insights in appropriate rules and guidance documents. Resistance to attack must become an explicit objective in the design of an NPP. A large, unplanned release of radioactive material from a nuclear power plant would be a comparatively rare event. Any estimate of the probability of such an event will be highly uncertain.34 Many PRAs for NPPs have been performed, and the results are useful for various purposes. However, PRA estimates of the probabilities of accidental releases should not be regarded as definitive, scientific findings. Such estimates rely on numerous assumptions and judgments. There is no certainty that all of the relevant factors are captured by a PRA. Findings of very low probability cannot be validated by direct experience.35 At present, there is no statistical basis for a quantitative estimate of the probability of a large release caused by an intentional, malevolent act. It does not follow, as some have suggested, that malevolent acts should be ignored in risk analysis for NPPs. In a policy or planning context, risk analysts could use judgment to assign minimum probabilities to postulated acts. That judgment could be combined with technical assessments of the vulnerability of plants to the postulated acts. Those assessments would rely, in part, on PRA techniques.
30 31 32 33 See Table 1-1 for an additional perspective. NRC, 1975. NRC, 1994. In some core-damage accidents, a common cause such as a powerful earthquake would simultaneously overcome a number of safety features. 34 Hirsch et al, 1989. 35 There have been two core-damage accidents involving unplanned releases of radioactive material from commercial nuclear power plants. Those accidents occurred at Three Mile Island in 1979 (involving a small release) and Chernobyl in 1986 (involving a large release).

14

ASSeSSIng the potentIAl foR A dIveRSIon of mAteRIAl A second potential adverse outcome of operating a nuclear power plant is, as mentioned above, a diversion of fissile or radioactive material from the plant, for use in nuclear or radiological weapons. That possibility is typically regarded by the nuclear industry and its regulators as a safeguards issue. Canada and many other countries have safeguards agreements with the International Atomic Energy Agency (IAEA). The purposes of these agreements include the prevention of diversion of fresh or spent fuel from NPPs. If such diversion is successfully prevented, the fissile and radioactive material in the fuel will remain protected. In the context of plant design, it is important to note that spent fuel could be diverted from some types of NPP with comparative ease. Heavy-water reactors, such as the Canadian CANDU design, are in this category because they employ on-line refueling. At those plants, prevention of diversion must rely heavily on administrative measures. By contrast, light-water reactors (LWRs) undergo batch refueling at intervals of one or more years. Thus, diversion of spent fuel from an LWR plant is comparatively difficult. Note that most (80 percent) of existing NPPs worldwide are LWRs. Diversion of spent fuel from a Canadian NPP for malevolent use is a comparatively unlikely event, given the present social and political environment in Canada. The same cannot be said for every location in the world where a Canadian-supplied CANDU plant is operating.36 A diversion event at such a plant could potentially be significant in terms of Canadian liability. Thus, the nexus between the NLCA and the risk of diversion should be examined. This report does not undertake that task. the pRecAutIonARy pRIncIple (pRecAutIonARy AppRoAch) The preceding discussion addresses two potential adverse outcomes of operating a nuclear power plant an unplanned release of radioactive material, and a diversion of fissile or radioactive material. The probability of either outcome is highly uncertain. Yet, either outcome would be significant from the perspective of the sustainability of nuclear power. In a policy or planning context, citizens and policy makers must grapple with this conjunction of uncertainty and significance. The precautionary principle offers guidance in such situations. This principle has been much discussed, and is incorporated in laws and regulations in Canada and elsewhere. It is discussed in Canadian government guidelines for the preparation of an EIS for a new NPP. Those guidelines tend to use the terms precautionary principle and precautionary approach interchangeably, and they introduce the concept as follows: 37 One of the purposes of environmental assessment is to ensure that projects are considered in a careful and precautionary manner before authorities take action in connection with them, in order to ensure that such projects do not cause significant adverse environmental effects. The Precautionary Principle informs the decision-maker to take a cautionary approach, or to err on the side of caution, especially where there is a large degree of uncertainty or high risk. In the Canadian Environmental Assessment Act of 1992, the concept of precaution appears twice in the Purposes section.38 First, at 4 (1) (a), the Act states that one of its purposes is to ensure that projects are considered in a careful and precautionary manner before federal authorities take action with them, in order to ensure that such projects do not cause significant adverse environmental effects. Then, at 4 (2), the Act states that federal government entities shall, in administering the Act, exercise their powers in a manner that protects the environment and human health and applies the precautionary principle. The Act further states, at 4 (1) (b), that one of its purposes is to encourage responsible authorities to take actions that promote sustainable development and thereby achieve or maintain a healthy environment and a healthy economy. Thus, the Act seeks to promote principles of sustainability and of precaution. That general commitment has been applied to specific cases by panels convened under the Act.39
15

36 37 38 39

For additional information and analysis, see: Thompson, 2008b. CEAA et al, 2008, Section 2.5. Justice Department, 2007. Gibson, 2000.

In April 2007, the Canadian government issued the Cabinet Directive on Streamlining Regulation.40 That directive sets forth six objectives for regulation by the federal government. The third of those objectives states that the government will: Make decisions based on evidence and the best available knowledge and science in Canada and worldwide, while recognizing that the application of precaution may be necessary when there is an absence of full scientific certainty and a risk of serious or irreversible harm. One application of that objective would be to anthropogenic climate change. In that instance, the harm would be serious and irreversible if no action were taken to reduce emissions of greenhouse gases. Yet, there might not be full scientific certainty about the extent to which emissions should be reduced. The above-stated objective would call for early action, without waiting for full scientific certainty. In the context of this report, serious and irreversible harm could arise from the taking of an action. The action would be the operation of an NPP, if the design of the plant created a significant potential for an unplanned release of radioactive material, or for diversion of fissile or radioactive material. In this instance, the above-stated objective would favor the blocking of the plants operation, even though the potential for harm could not be characterized with full scientific certainty as to consequences and probability. Canadas Federal Sustainable Development Act of 2008 gives the precautionary principle a prominent role in planning for sustainability. Paragraph 9 of the Act calls for the development and periodic updating of a Federal Sustainable Development Strategy based on the precautionary principle.41

3.3 Full Accounting of Costs

The concept of sustainability introduces new categories of cost into analyses of the costs and benefits of potential actions. In considering the costs and benefits of a large energy project, such as an NPP, it is no longer sufficient to consider a narrow range of costs, as was typically done in the mid-20th century. It is now necessary to consider costs over the full life cycle of a project, together with costs arising from the environmental and social impacts of the project. An essay placed on the Canadian Nuclear Associations website describes the importance of a full accounting of costs. It calls for a life-cycle perspective and the internalizing of costs that were previously neglected. It states:42 Nuclear energy should argue vigorously for a level playing field, where the costs of external impacts are internalized in the costs of energy sources, and where sustainable development considerations are given adequate weight. Maintaining a level playing field requires that a full spectrum of information about the costs of any large energy project is publicly available. Democracy and the functioning of energy markets both require the publication of all relevant costs, with minor exceptions to protect privacy. Part of full accounting is to ensure that costs are properly attributed. An important mechanism for such attribution is the polluter-pays principle. This principle is defined by Environment Canada as follows:43 Polluter Pays Principle (Principe du pollueur-payeur) A principle under which users and producers of pollutants and wastes should bear the responsibility for their actions. Companies or people that pollute should pay the costs they impose on society.

16

40 41 42 43

Government of Canada, 2007. Accessed on 22 October 2009 at: <http://laws.justice.gc.ca/en/showtdm/cs/F-8.6>. Morrison, 2000. Accessed on 17 October 2009 at: <http://www.ec.gc.ca/cppic/en/glossary.cfm?view=details&id=178>.

The Supreme Court of Canada unanimously upheld the polluter-pays principle in a 2003 decision. The decision supported the Quebec Minister of Environment against Imperial Oil. In its decision, the Court stated that the polluter-pays principle has become firmly entrenched in environmental law in Canada. The Court explained the principle as follows:44 To encourage sustainable development, that principle assigns polluters the responsibility for remedying contamination for which they are responsible and imposes on them the direct and immediate costs of pollution. At the same time, polluters are asked to pay more attention to the need to protect ecosystems in the course of their economic activities.

3.4 Criteria for Assessing the NLCA

Sections 3.1 through 3.3, above, describe three concepts that would play major roles in a properlyconstructed 21st century policy framework for nuclear power. The concepts are: sustainability; the precautionary principle; and full accounting of costs. From these concepts, one can articulate criteria for assessing the NLCA. Five criteria are set out here, expressed as questions. The first question, about full accounting, has a simple, factual answer. Questions two and three, also about full accounting, can in part be answered through quantitative analysis. Questions four and five, about the precautionary principle and sustainability, require qualitative answers. In this report, the NLCA is assessed according to the extent to which it would respond to the content of each question. The questions follow. Criterion #1: full accounting information disclosure Would the NLCA require Canadas nuclear industry and government to publish a full spectrum of information relevant to nuclear risk and nuclear insurance, including: (a) an assessment of the risk of operating each NPP and other nuclear installation; and (b) the insurance coverage of offsite and onsite damage that is provided for each NPP and other nuclear installation, premiums paid for this coverage, and reinsurance arrangements? Criterion #2: full accounting damage coverage Would the NLCA ensure that commercial insurance covers the full range of potential offsite and onsite damage arising from incidents at each NPP and other nuclear installation? Criterion #3: full accounting accounting for risk costs as costs of production Would the NLCA ensure that the risk costs of operating each NPP and other nuclear installation are accounted for as costs of production? Criterion #4: the precautionary principle Would the NLCA be compatible with coherent and consistent application of the precautionary principle in Canada? Criterion #5: sustainability Would the NLCA be compatible with a transition of Canadas economy toward sustainability?

17

44 Ferrara and Mesquita, 2003.

4.Nuclear Liability from Global & North American Perspectives

In assessing the NLCA, it is important to view the relevant issues from both global and North American perspectives. The global perspective is important because Canadas economy is integrated with the global economy, and Canadas nuclear industry is part of that integration. The North American perspective is important for two reasons. First, the potential for cross-border effects from nuclear incidents irrevocably links the nuclear liability regimes of the USA and Canada. Second, the nuclear liability regime in the USA offers lessons that should be considered in updating Canadas regime. In taking a global perspective, the first step is to examine the prospects of nuclear power over the coming decades. ScenARIoS foR futuRe uSe of nucleAR poweR woRldwIde Nuclear power is in a transitional phase. Annual, worldwide capacity additions peaked in 1985 and have been modest since 1990.45 If construction of nuclear power plants does not resume, total capacity will decline as plants are retired. Observers view this situation in widely differing ways. Some call for a nuclear power renaissance in which nuclear generating capacity rises substantially. Others prefer or expect a scenario in which nuclear capacity declines, leading to eventual disappearance of the industry. The most ambitious visions of the nuclear renaissance are exemplified by a technology roadmap issued under the auspices of the US Department of Energy in 2002.46 The roadmap proposed the development and use of a range of Generation IV nuclear fission reactors that would push against engineering limits in a variety of respects. Some reactor types would produce hydrogen as well as electricity, thereby providing fuel for use in vehicles and other applications. Reactors would be deployed in such large numbers that uranium reserves would become depleted during the latter part of the 21st century. To prepare for that eventuality, large-scale reprocessing would begin during the next few decades, and breeder reactors would be deployed beginning in about 2030. A less extreme but still highly ambitious vision of the nuclear renaissance is contained in a study published under the auspices of Massachusetts Institute of Technology in 2003.47 The authors saw no need for reprocessing or breeder reactors during at least the next 50 years. They offered an illustrative scenario for expansion of nuclear capacity using Generation III reactors whose designs would involve a comparatively small evolutionary step from the designs of present reactors. In the scenario, annual worldwide production of nuclear-generated electricity would rise by a factor of 4 to 6 between 2000 and 2050. Many observers doubt the merits of nuclear power, and seek or expect a decline in its use.48 Some argue that nuclear power can and should be phased out, even during an effort to dramatically reduce greenhouse gas emissions from electricity generation.49 Others argue that scenarios for expansion of nuclear capacity are implausible, and that the commercial nuclear industry is in evident decline.50 the economIcS of nucleAR poweR Current trends suggest that the nuclear industry will decline over the coming years, rather than experience a renaissance. The major trend that suggests this outcome is a rapid, steep rise in the projected costs of generating electricity from new NPPs. Alternative options for meeting electricity needs and reducing greenhouse gas emissions are less costly.51 Nevertheless, a number of national governments are committed, at least nominally, to the construction of new NPPs in their countries. A continued commitment to this goal would oblige governments to provide direct or indirect subsidies to new NPPs. If governments do provide such subsidies, they will continue a longstanding practice. A recent study has concluded:52 There are numerous ways by which governments have organized or tolerated subsidies to nuclear power. They range from direct or guaranteed government loans to publicly funded research and development (R&D). Direct ownership of subsidized nuclear fuel chain facilities, government funded nuclear decommissioning and waste management, generous limited liability for accidents and the transfer of capital costs to ratepayers via stranded cost rules or special rate-basing allowances are all common in many countries.
18 45 46 47 48 49 50 51 52 IAEA, 2006. NERAC/GIF, 2002. Ansolabehere et al, 2003. Romm, 2008. Makhijani, 2007; Greenpeace International, 2007. Schneider et al, 2009. Cooper, 2009. Schneider et al, 2009, page 7.

Given this history, any assessment of the NLCA should pay close attention to its possible role in subsidizing nuclear power in Canada. If the NLCA would provide a subsidy, its proponents should acknowledge and justify that subsidy. the nucleAR lIAbIlIty RegIme In the uSA Much of law and practice around the world regarding nuclear liability and compensation can be traced back to the development of commercial nuclear power in the USA in the 1950s. Initially, the private sector was only willing to become involved in nuclear R&D if the US Atomic Energy Commission would include hold harmless clauses in the applicable contracts, thus channeling liability to the State. Beginning in 1954, however, the US government shifted third-party liability to the emerging nuclear industry. That move acted as a brake on private nuclear investment, because the insurance sector could not provide insurance on a normal commercial basis. In response, the Price-Anderson Act was passed in 1957, channeling liability to the operator while limiting that liability.53 The Price-Anderson Act introduced economic channeling, which differs from the legal channeling that prevails in Canada. The difference has been described as follows:54 In regimes of legal channelling, the operator is the only party which victims may legally hold responsible for a nuclear accident, i.e. no civil lawsuits may be initiated against any other party (suppliers, designers, constructors, etc.) and on any other civil basis than the channelling legal basis. Ordinary tort law is, in other words, set aside. This is so, even when the operator did not even remotely contribute to the nuclear accident. Moreover, unless a contract expressly states otherwise, the operator does not have a right of recourse against these other parties and fully bears the financial liability burden of a nuclear accident vis--vis third parties. In systems of economic channelling, victims can initiate civil lawsuits both against the operator and against any of the other parties involved (suppliers, designers, etc.) in line with ordinary tort law. However, the operator, whose insurance needs to cover the other parties third party liability as well (i.e. an omnibus coverage or umbrella insurance), ultimately needs to indemnify these parties. The result is similar to legal channelling in that the operator bears the financial liability burden of the nuclear accident vis--vis third parties. However, economic channelling leaves the legal reality unscathed and does not set ordinary tort law aside, whereas legal channelling distorts the underlying legal construction and sidesteps ordinary tort law. The functioning of economic channelling is best illustrated by the Three Mile Island accident of 1979, where all defendants the operator, the designer and the constructor of the plant were represented by a single law firm. In the late 1950s, US firms began exporting nuclear technology and materials to the emerging nuclear industry in Europe. They became concerned about their potential liability for damage arising from a nuclear incident in Europe. That concern was resolved by successfully promoting legal channeling as the basis of the 1960 Convention on Third Party Liability in the Field of Nuclear Energy (Paris Convention), which determined nuclear liability regimes in Europe. US nuclear suppliers also successfully promoted legal channeling as the basis of the 1963 Vienna Convention on Civil Liability for Nuclear Damage (Vienna Convention). Western European nuclear suppliers subsequently promoted legal channeling as the basis for nuclear liability regimes in Eastern Europe, to facilitate their exports to that region. This history has been summarized as follows:55 Put simplistically, legal channelling was introduced in the legal systems of Western Europe by the Paris Convention under US pressure and in the legal systems of Eastern Europe by the Vienna Convention under Western-European pressure.

19

53 Ameye, 2009. 54 Ameye, 2009. 55 Ameye, 2009.

In the USA, economic channeling continues as the basis of the Price-Anderson Act. That Act was most recently amended in 2005. It holds the operator liable for third-party damage up to $300 million. For a single-unit NPP site, the average annual premium for insurance of that amount is currently about $400,000.56 If damage exceeds $300 million, a second tier of insurance is provided via a retroactive premium levied on each licensed NPP in the USA. The retroactive premium is adjusted for inflation. In 2005, its maximum value was $15.8 million per year per NPP, and the maximum cumulative value was $100.6 million per NPP.57 At present, 104 NPPs are licensed in the USA. Thus, the second tier of insurance amounted to $10.5 billion in 2005.58 As a separate matter, the US Nuclear Regulatory Commission (NRC) requires each licensee of an NPP in the USA to obtain insurance coverage for onsite damage. The minimum required coverage is currently $1.06 billion per NPP site, and actual coverage is $2.75 billion. This coverage does not include loss of electricity production. The coverage is provided by Nuclear Electric Insurance Limited (NEIL), a mutual insurance system created by a group of NPP operators.59 NEIL also provides separate coverage of loss of electricity production. The covered amount is up to $4.5 million per week, with a maximum claim of $490 million.60 lIAbIlIty AcRoSS the cAnAdA-uSA boRdeR An incident at a Canadian NPP could cause significant damage in the USA, and vice versa. The NLCA would allow the Canadian government to enter into an agreement with the US government, whereby parties in each country have access to the liability and compensation regime in the other country.61 At present, the law regarding cross-border claims is unclear. An injured party in the USA could make a claim in a US court against the operator of a Canadian NPP that experienced an incident, or against a supplier to that operator. If the US court accepted the claim, it could seek enforcement of the claim in Canada. Enforcement would pose a challenge to the relevant Canadian court, a challenge that has been described as follows:62 A Canadian court may be unwilling to enforce a very large US judgment for damages where doing so would frustrate the purpose of the Nuclear Liability Act by draining the available pool of funds for claimants, if US plaintiffs would be substantially or fully compensated and Canadians would be left with little to no compensation. However, the political acceptability of such a situation of under-compensation in the US would be so difficult that, as a practical matter, the Canadian government would be forced to make up the difference. If Canada can afford to provide some $4 billion to protect the domestic automobile manufacturers, it seems unlikely that it could limit payment to victims of a nuclear incident, on either side of the border, to a mere $75 million. It has been suggested that Canada could resolve this problem by becoming a party to the 1997 Convention on Supplementary Compensation for Nuclear Damage (CSC). The USA ratified the Convention in 2008, but it has not yet entered into force. If Canada ratified the Convention and it entered into force, then a US party injured by an incident at a Canadian NPP could not, apparently, sue in a US court.63

20 56 WNA, 2009. Given perfect forecasting and no profit or administrative costs, an insurance premium of $400,000 per RY for coverage of $300 million would imply an event probability of 1.3 per 1,000 RY. 57 These amounts include a 5 percent allowance for legal costs. 58 Faure and Borre, 2008. 59 Faure and Borre, 2008. 60 NEIL PowerPoint presentation at the Nuclear Inter Jura Congress, Toronto, 5-9 October 2009. 61 Henault, 2009. 62 Roman et al, 2009. 63 Roman et al, 2009.

nucleAR lIAbIlIty RegImeS outSIde noRth AmeRIcA Outside North America, there is a patchwork of nuclear liability regimes at the national level. A coherent, international structure is gradually emerging that links these national regimes and establishes a common approach. At the center of that structure are the Paris Convention and the Vienna Convention, as mentioned above. These conventions were developed under the aegis of NEA and IAEA, respectively. Their territorial scope is linked by the 1988 Joint Protocol. They have been amended in various respects. The CSC is a related, stand-alone convention, developed under the aegis of IAEA. 64 In 2004, parties to the Paris Convention signed a set of amending Protocols, upgrading the Conventions level of compensation. The 2004 Protocols expand the scope of damage that is compensable, extend the time horizon of compensable personal injury or death from 10 years to 30 years, and raise the liability limit. The new liability limit for each incident is 1,500 million Euro, supported by three tiers of compensation. The first tier is 700 million Euro of operator liability, nominally met by commercial insurance.65 The second tier is 500 million Euro of public funds provided by the government of the country in which the nuclear incident occurs. The third tier is 300 million Euro of public funds provided by the governments of other parties to the Paris Convention.66 The 2004 Protocols are not yet in force. A major reason is that NPP operators have had difficulty obtaining insurance coverage for a liability of 700 million Euro.67 This difficulty came to public attention in the UK in September 2009. Press reports disclosed that nuclear insurers in the UK were refusing to cover the 700 million Euro of operator liability required by the 2004 Protocols. The insurers especially objected to the scope of damage and the time horizon specified by the Protocols. It appears that this problem will be resolved by the UK government covering a substantial portion of the operator liability, for both existing and new NPPs.68 Germany is a party to the Paris Convention, but has no limit on the operators liability. Each NPP operator in Germany must provide security of 2,500 million Euro, of which 256 million Euro is provided by insurance. The remaining security is provided collectively by all of the NPP operators in Germany. In this collective arrangement, the directly-affected operator bears the primary burden of liability, and the others provide backup. 69 As an indication of nuclear-insurance premiums in Europe, consider the case of EDF, the French NPP operator. EDF has current liability of 91 million Euro, covering 31 million Euro of this liability through insurance and the remaining 60 million Euro through its own reserves. To provide 31 million Euro of insurance coverage for 58 NPPs in 2008, EDF paid an annual premium of 6.4 million Euro. 70 Thus, the premium was 110,000 Euro per RY. Given perfect forecasting and no profit or administrative costs, an insurance premium of 110,000 Euro per RY for coverage of 31 million Euro would imply an event probability of 3.5 per 1,000 RY.

21 64 Faure and Fiore, 2008; WNA, 2009. 65 Under the 2004 Protocols, a countrys government could meet the operators first-tier liability of 700 million Euro, or some portion of that liability, from public funds. 66 Faure and Fiore, 2008; WNA, 2009. 67 Faure and Borre, 2008. 68 Pagnamenta, 2009. 69 Faure and Borre, 2008. 70 Faure and Fiore, 2008.

5. Assessing the Risk Posed by Nuclear Power Plants

There is a large body of technical literature addressing the risk posed by NPPs. Much of this literature assesses the potential for, and consequences of, an atmospheric release of radioactive material following accidental damage to nuclear fuel. The fuel could be in the reactor core, the spent-fuel pool, or elsewhere in the plant. Such literature typically falls under the rubric of probabilistic risk assessment, as described in Section 3.2, above. In the PRA field, the events that initiate an accidental release are categorized as internal events (human error, equipment failure, etc.) or external events (earthquakes, fires, strong winds, etc.). PRAs typically do not consider initiating events that involve intentional, malevolent acts, although PRA techniques can be adapted to estimate the outcomes of such acts. PRAs for NPPs are conducted at Levels 1, 2 and 3, in increasing order of completeness, as discussed below. A thorough, full-scope PRA would be conducted at Level 3, and would consider internal and external initiating events. The findings of such a PRA would be expressed in terms of the magnitudes and probabilities of a set of adverse environmental impacts, and the uncertainty and variability of those indicators. The adverse impacts would include: (i) early human fatalities or morbidities (illnesses) that arise during the first several weeks after the release; (ii) latent fatalities or morbidities (e.g., cancers) that arise years after the release; (iii) short- or long-term abandonment of land, buildings, etc.; (iv) short- or long-term interruption of agriculture, water supplies, etc.; and (v) social and economic impacts of the above-listed consequences. The magnitudes and probabilities of such adverse impacts would be estimated in three steps. First, a Level 1 PRA analysis would be performed. In that analysis, a set of event sequences (accident scenarios) leading to fuel damage would be identified, and the probability (frequency) of each member of the set would be estimated. The sum of those probabilities across the set would be the total estimated fueldamage probability.71 Second, a Level 2 PRA analysis would be performed. In that analysis, the potential for release of radioactive material to the atmosphere would be examined across the set of fuel-damage sequences. The findings would be expressed in terms of a group of release categories characterized by magnitude, probability, timing, isotopic composition, and other characteristics. Third, a Level 3 PRA analysis would be performed, to yield the impact findings described above. In that analysis, the atmospheric dispersion, deposition and subsequent movement of the released radioactive material would be modeled for each of the release groups determined by the Level 2 analysis. The dispersion modeling would account for meteorological variation over the course of a year. Then, the adverse environmental impacts of the released material would be estimated, accounting for the materials distribution in the biosphere. If done thoroughly, this 3-step estimation process accounts for uncertainty and variability at each stage of the process. A thorough, full-scope, Level 3 PRA is expensive and time-consuming. It yields estimated impacts expressed as statistical distributions of magnitude and probability, not as single numbers. Even after such a thorough effort, there are substantial, irreducible uncertainties in the findings.72 empIRIcAl vAlIdAtIon of pRA fIndIngS Direct empirical evidence for the validity of PRA findings is limited. Worldwide operating experience of commercial nuclear power plants through 2008 is about 13,400 RY, and Canadian experience is about 580 RY.73 Two events involving substantial damage to a reactor core have occurred worldwide while that experience was accruing. At Three Mile Island Unit 2 in 1979, the reactor core was severely damaged but there was a comparatively small radioactive release to the environment. At Chernobyl Unit 4 in 1986, a substantial fraction of the core inventory of radioactive material was released to the atmosphere. This limited experience allows one to estimate the probability of a core-damage accident as 1.5 per 10,000 RY, and the probability of a large atmospheric release as 0.7 per 10,000 RY.74
71 The term core-damage frequency (CDF) is often encountered. This term refers to the annual probability of severe damage to nuclear fuel in a reactor core. 72 Hirsch et al, 1989. 73 Extrapolated from Table 1 of: IAEA, 2006. 74 2/13,400 = 1.5 per 10,000; 1/13,400 = 0.7 per 10,000.

22

nuReg-1150 The high point of PRA practice worldwide was reached in 1990 with publication by the NRC of its NUREG-1150 study, which examined five different nuclear power plants using a common methodology.75 The study was well funded, involved many experts, was conducted in an open and transparent manner, was done at Level 3, considered internal and external initiating events, explicitly propagated uncertainty through its chain of analysis, was subjected to peer review, and left behind a large body of published documentation. Each of those features is necessary if the findings of a PRA are to be credible. There are deficiencies in the NUREG-1150 findings, which can be corrected by fresh analysis and the use of new information. The process of correction is possible because the NUREG-1150 study was conducted openly and left a documentary record. PRA practice in the USA has degenerated since the NUREG-1150 study. Now, PRAs are conducted by the nuclear industry, and the only published documentation is a summary statement of findings. NRC formerly sponsored independent reviews of industry PRAs, but no longer does so. Thus, PRA findings have lacked credibility for at least a decade. In other countries, including Canada, PRA practice has experienced similar degeneration.76 Figures 5-1 through 5-3 show some findings from the NUREG-1150 study that are relevant to this report. The findings are for a pressurized-water-reactor (PWR) plant at the Surry site, and a boiling-water-reactor (BWR) plant at the Peach Bottom site. These plants typify many of the Generation II plants in the present worldwide fleet of nuclear power plants. Using the Livermore seismic estimates, the NUREG-1150 findings for these two plants are roughly comparable with the experience-derived probability estimates mentioned above a coredamage probability of 1.5 per 10,000 RY, and a large-release probability of 0.7 per 10,000 RY. the potentIAl foR mAlevolent ActS At nucleAR poweR plAntS CNSC has established criteria for the design of new NPPs. These criteria, expressed in the document RD-337, include resistance to attack as a design objective.77 To date, CNSC has not specified the threats that will be considered in applying the design criteria. CEAA guidelines for an EIS for a new NPP require the consideration of accidents and malfunctions that include malevolent acts.78 This author has commented upon those guidelines.79 A consultant to CNSC has examined potential modes and instruments of attack on an NPP, and has recommended an approach to incorporating these threats in the design criteria for new plants.80 Among the instruments of attack considered by the consultant were a large commercial aircraft, an explosiveladen smaller aircraft, and an explosive-laden land vehicle. Table 5-1 describes some potential modes and instruments of attack on an NPP, and also describes the defenses that are now provided at US plants. There is no defense against a range of credible attacks. Defenses at Canadian plants are no more robust than at US plants. Among the instruments of attack mentioned in Table 5-1 is a large commercial aircraft. In September 2001, aircraft of this type caused major damage to the World Trade Center and the Pentagon. However, such an aircraft would not be optimal as an instrument of attack on a nuclear power plant. Large commercial aircraft are comparatively soft objects containing a few hard structures such as turbine shafts. They can be difficult to guide precisely at low speed and altitude. A well-informed group of attackers would probably prefer to use a smaller, general-aviation aircraft laden with explosive material, perhaps in a tandem configuration in which the first stage is a shaped charge. Table 5-2 provides some information about shaped charges and their capabilities. There is no statistical basis for a quantitative estimate of the probability that am NPP will be attacked. However, if a given attack scenario is postulated, one can apply PRA techniques to estimate the conditional probabilities of various outcomes. NRC took that approach in developing its vehicle-bomb rule of 1994.81
75 NRC, 1990. 76 In Canada, it appears that PRAs are no longer available for independent review. To illustrate, Greenpeace Canada requested a copy of the PRA for the Pickering B units. CNSC has refused to order Ontario Power Generation (OPG) to provide this PRA. In so doing, CNSC has accepted OPG's argument that the PRA should be available only to OPG personnel on a need to know basis. See: CNSC, 2008b. This approach, although it may be well-intentioned, will inevitably create an entrenched culture of secrecy that will suppress a clear-headed understanding of risks. A more sophisticated approach could allow independent review of the PRA without disclosing information that would assist malevolent actors. 77 CNSC, 2008a. 78 CEAA et al, 2008. 79 Thompson, 2008a. 80 Asmis and Khosla, 2007. 81 NRC, 1994.

23

RAdIoActIve ReleASeS fRom StoRed Spent fuel At NPPs in the USA and elsewhere, large amounts of spent fuel are stored under water in pools adjacent to reactors. All US pools currently employ high-density racks, to maximize the amount of spent fuel that can be stored in each pool. This practice has been adopted because it is the cheapest mode of storage of spent fuel. Unfortunately, the high-density configuration would suppress convective cooling of fuel assemblies if water were lost from a pool. Several reputable studies have agreed that loss of water from a pool would, across a range of water-loss scenarios, lead to spontaneous ignition of the zirconium alloy cladding of the most recently discharged fuel assemblies. The resulting fire would spread to adjacent fuel assemblies and propagate across the pool. Extinguishing the fire, once it had been initiated, would be difficult or impossible. Spraying water on the fire would feed an exothermic reaction between steam and zirconium. The fire would release a large amount of radioactive material to the atmosphere, including tens of percent of the pools inventory of cesium-137. Large areas of land downwind of the plant would be rendered unusable for decades. Loss of water could arise in various ways as a result of an accident or an intentional, malevolent act.82 Fortunately, measures are available for dramatically reducing the risk of a fire in a spent-fuel pool, as discussed in Section 6, below. As discussed in Section 7, below, three designs of NPP are being considered for construction in Ontario. Two of these plant designs the AREVA and Westinghouse designs are PWR plants. It appears that both vendors envision the equipping of each plants spent-fuel pool with a set of high-density racks.83 That practice would bring with it the potential for a large atmospheric release of radioactive material (especially Cesium-137) from the pool. This author is not aware of any study on the potential for an accidental release of radioactive material from spent fuel stored at an NPP employing a CANDU reactor. Absent such a study, the potential remains unknown. cnSc poSItIon on npp RISk CNSC has articulated safety goals for a new NPP, as shown in Table 5-3. The safety goals were first set forth in a 2007 draft document. Revised safety goals were then adopted by CNSC in 2008. As shown in Table 5-3, the safety goals that were ultimately adopted by CNSC represent a significant retreat from the draft goals set forth in 2007. A logical explanation for that retreat is a CNSC determination that compliance with the 2007 goals could not be demonstrated for new plants of the types being considered for construction in Canada. Thus, CNSCs current position is that the probability of a large release of radioactive material to the environment from a new NPP in Canada is less than 1 per 1 million RY. Apparently, that probability does not account for malevolent acts. CNSC has not specified whether the stated probability is a mean value or some other expression of the probability density function. In this report, it is assumed that CNSCs stated probability is a mean value. It would also be reasonable to assume an uncertainty factor of 10.84 Given those assumptions, CNSCs current position would be that the 95th percentile probability of a large release from a new plant would be 1 per 100,000 RY. The 95th percentile value can be regarded as a high confidence estimate. The discussion in the preceding paragraph refers to new NPPs. In the context of life extension of an existing plant, CNSC requires a licensee to conduct an Integrated Safety Review, whose purposes include determination of reasonable and practical modifications that should be made to systems, structures, and components, and to management arrangements, to enhance the safety of the facility to a level approaching that of modern nuclear power plants [emphasis added], and to allow for long term operation.85 CNSC does not articulate numerical safety goals for existing plants.

24

82 Alvarez et al, 2003; National Research Council, 2006; Thompson, 2007. 83 Thompson, 2008a. 84 In this report, the term uncertainty factor is used to designate the ratio of the 95th percentile value to the mean value of a probability density function. 85 CNSC, 2008c, page 4.

6. Options for Reducing the Risk Posed by Nuclear Power Plants

There are numerous opportunities for reducing the risk posed by an NPP. Those opportunities are, in principle, substantially greater for a new NPP than for an existing plant. The design of the new plant could benefit from new technical knowledge. The safety and security criteria that the plant must meet could be more stringent than the criteria to which existing plants were designed. In practice, however, the new plants being considered for construction in Canada would not pose a substantially lower risk than do the existing plants, for reasons discussed below. tRendS In conStRuctIon, SAfety And SecuRIty of nucleAR poweR plAntS Nuclear power is in a transitional phase, moving toward an uncertain future. Annual, worldwide capacity additions peaked in 1985 and have been modest since 1990.86 If construction of NPPs does not resume, total capacity will decline as plants are retired. During the nuclear industrys start-up phase (1956-1970), capacity additions worldwide averaged about 1 GWe per year. In the decade 1971-1980, worldwide capacity increased at an average rate of about 12 GWe per year, increasing to an average rate of 20 GWe per year during the period 1981-1990. During the period 1991-present, the rate of capacity addition has been much lower, averaging about 4 GWe per year, with an even lower rate since 2000.87 This construction history has left the world with a fleet of existing NPPs that are mostly in the Generation II category. The basic designs of these plants were laid down more than three decades ago. At that time, risk goals were less demanding than the goals now articulated by CNSC.88 There was, for example, less concern by industry and regulators about the potential for malevolent acts. During the 1970s and 1980s, plant vendors and other stakeholders identified innovative design options that could have reduced the risk of NPP operation to a level substantially below the level posed by the Generation II designs. Some of those design options such as ASEA-Atoms PIUS design are discussed in the Appendix to this report. The nuclear industry did not adopt innovative designs such as the PIUS design. Instead, the industry chose to pursue Generation III designs that represent a comparatively small evolutionary step from the Generation II designs now in use. That decision has yielded a level of risk, for new NPPs, that is not substantially lower than the risk posed by existing plants.89 Regulators, including CNSC, could have sought to steer the industry toward innovative, lower-risk designs, by adopting highly stringent criteria for safety and security. Instead, regulators have accommodated the nuclear industry, by adopting criteria that are achievable by Generation III designs. The process of accommodation is clearly evident in CNSCs relaxation of its safety goals for new NPPs, as discussed in Section 5, above. RISk ReductIon At exIStIng nucleAR poweR plAntS At an existing NPP, efforts to reduce risk could be made in areas including: (i) physical modification of the plant (capital additions); (ii) new procedures for operation and maintenance; (iii) personnel enhancement (training, etc.); (iv) enhanced site security (guards, gates, etc.); (v) enhanced capability for onsite damage control (firefighting, etc.); and (vi) enhanced capability for offsite emergency response. Table 6-1 illustrates the potential for reducing risk at an existing plant. The table shows options for reducing the risk of a fire in a spent-fuel pool equipped with high-density racks. As explained in Section 5, above, such a fire could occur if water were lost from the pool. High-density pool storage of spent fuel is standard practice at existing NPPs in the USA, and would be a likely practice at new PWR or BWR plants in Canada, if any are built.
25

86 87 88 89

IAEA, 2006. Keystone Center, 2007, pp 25-26. Okrent, 1981. Thompson, 2008a.

7. Risk Posed by Nuclear Power Plants in Canada

7.1 Scope of this Discussion
Sections 7.2 through 7.4, below, outline the risk posed by operation of existing or new NPPs in Canada. A comprehensive assessment of that risk would be a major undertaking. Although some relevant studies have been done, there are major gaps in knowledge. For the purpose of estimating risk costs, this report draws from available sources to present a broad-brush picture of risk. Where possible, risk estimates developed by the nuclear industry are used here.

7.2 Types of Nuclear Power Plant that Could Operate in Canada

All of the existing NPPs in Canada are CANDU plants. At present, 18 of these plants are operational. All are in the Generation II category. All but two are in Ontario. The Ontario plants are unusual because they share safety and support systems (e.g., a vacuum building) in 4-unit or 8-unit blocks. Most NPPs in the world do not share systems to this extent. new plAntS The Ontario government is contemplating the construction of new NPPs at the Darlington site. Three types of NPP are candidates, as follows: (i) the US EPR, a PWR plant offered by AREVA; (ii) the AP1000, a PWR plant offered by Westinghouse; and (iii) the ACR-1000, a CANDU plant offered by Atomic Energy of Canada Limited (AECL). Each of these plant types is in the Generation III category, as mentioned above.90 There is no operational experience for any of them, and limited construction experience. Two EPR plants are being built, in Finland and France. In April 2009, concrete pouring began for the first of four AP1000 plants ordered for a site in China. No order has been placed for an ACR-1000 plant. As a Canadian product, the ACR-1000 is a likely candidate for construction in Ontario. However, the technical competence of AECL the vendor of the ACR-1000 plant is under question because of AECLs experience with the MAPLE reactors. AECL built these two reactors at the Chalk River laboratories to produce medical isotopes. In May 2008 the MAPLE reactors were scrapped without ever becoming operational. AECL had concluded that the reactors were unfit to operate, and that their deficiencies could not be rectified within any reasonable budget and timeframe.91

7.3 Potential Radioactive Releases and their Offsite Impacts

PRAs, despite their limitations, are important sources of information about the potential for, and consequences of, releases of radioactive material from NPPs. For the new plants being considered for use in Ontario, PRAs are not available. For the existing plants in Canada, there is a body of PRA-related literature, with limitations as discussed below. Unfortunately, Canada lacks a fully developed PRA culture. PRAs performed in Canada for CANDU reactors find very low probabilities for large releases. Based on those findings, the PRAs do not estimate the radiological impacts of large releases. Yet, the low probabilities are not credible.92 The practice of ignoring large releases deprives citizens and policy makers of needed information. For example, in a recent analysis of the risk of continued operation of the Pickering B station, the largest release considered included 71 TBq of Cesium-137.93 That is a comparatively small release, and is categorized as such in Table 5-3.

26

90 Some plant designs are said to be in a Generation III+ category. That designation has no technical meaning, because it presumes a generally-accepted classification scheme that does not exist. 91 Thompson, 2008a, Section 5. 92 Thompson, 2000; IRSS, 1992. 93 SENES, 2007, Table B.5.3-1.

The high point of PRA practice in Canada was reached by Ontario Hydro in its preparation of the Darlington Probabilistic Safety Evaluation (DPSE).94 DPSE was conducted for internal initiating events only. It was conducted to Level 3, except that the impacts of the largest releases in Ex-Plant Release Category 0 (EPRC 0) were not evaluated. It was not subjected to an official, independent review. Thus, DPSE did not rise to the quality of NRCs NUREG-1150 study, which is discussed in Section 5, above. A focused review of DPSE was conducted by a team led by this author.95 Several deficiencies were revealed. For example, DPSE had failed to identify an event sequence involving failure of service water supply that would be familiar to analysts conducting PRAs for PWR plants. In light of that and other deficiencies in DPSE, our team concluded that a reasonable estimate of the probability of a large, accidental radioactive release to the atmosphere from the Darlington plant would be 1 per 10,000 RY. Our value is comparable to the probability derived from occurrence of the TMI and Chernobyl accidents. (As mentioned above, those events suggest a core-damage probability of 1.5 per 10,000 RY, and a large-release probability of 0.7 per 10,000 RY.) Interestingly, our value is also comparable to the 95th percentile (high-confidence) value of DPSEs estimate of the probability of release category EPRC 0, adjusted to account for external initiating events. The adjusted, 95th percentile probability of EPRC 0 is 1.2 per 10,000 RY.96 PRAs are not available for the new NPPs that might be built at the Darlington site in Ontario. Lacking a PRA, one can take two approaches to estimating the probability of a large, accidental atmospheric release from a new plant. First, one can assume a 10-fold reduction in release probability for a new (Generation III) plant, by comparison with an existing (Generation II) plant. That assumption yields a release probability, for a new plant, of 1 per 100,000 RY. Second, one can rely on the CNSC safety goals, which are discussed in Table 5-3. Those goals now state that the probability (assumed here to be the mean probability) of a large release from a new NPP is below 1 per 1 million RY. As discussed in Section 5, above, one can reasonably assume that CNSCs position is that the 95th percentile (high-confidence) probability of a large release is 1 per 100,000 RY. Table 7-1 assembles a set of selected data about radioactive releases and their offsite impacts. Those data support this reports estimation, in Section 9, below, of the risk costs of offsite impacts. The metric used in Table 7-1 for offsite impacts is lifetime population dose. In this report, population dose is assumed to scale linearly with the amount of Cesium-137 released to the atmosphere. That assumption is supported by the finding that most of the offsite population dose from the 1986 Chernobyl accident was from Cesium-137.97 Table 7-2 shows estimates of population dose and frequency for potential accidental atmospheric releases of radioactive material from three nuclear generating stations in Ontario. These estimates are by the licensees (i.e., operators). Each station has four CANDU units. Each release shown is from a single unit (i.e., NPP).

7.4 Potential Onsite Impacts of Fuel-Damage Events

A range of fuel-damage events could occur at an NPP, and the fraction of the material released from the fuel that reached the environment would vary according to the characteristics of each event. Ontario Hydro, in its DPSE study, estimated the risk of onsite economic impacts from fuel-damage events at the existing CANDU nuclear power plants at the Darlington site. That estimate considered only accidents initiated by internal events. Table 7-3 shows Ontario Hydros findings, adjusted to 2009 Can $. An interesting observation from Table 7-3 is that almost 90 percent of the risk of onsite economic impacts arises from fuel damage categories FDC 6 through FDC 9. The lowest mean probability for one of these categories is 1 per 500 RY. Thus, the overall risk is dominated by events with a comparatively high probability. To illustrate the risk of onsite impacts, consider fuel damage category FDC 7. Table 7-3 shows that this event has an estimated probability of 3 per 1,000 RY (i.e., 48 percent over 40 years, at the 4-unit Darlington station) and would cause damage between $790 million and $2,500 million to the operator.
27

94 Ontario Hydro, 1987. 95 IRSS, 1992. 96 DPSE (Ontario Hydro, 1987) states in its Table 5-6 that the probability of EPRC 0 is 4.4 per 1 million RY. Applying an uncertainty factor of 14 (see Table 5-5 of DPSE), and a multiplier of 2 to account for external initiating events, one finds a 95th percentile value for EPRC 0 of 1.2 per 10,000 RY. 97 DOE, 1987.

8. The Canadian Governments Consideration of Risk-Related Costs

The proposed NLCA limits the operators third-party liability to $650 million. The Canadian governments choice of a $650 million limit reflects a balance of considerations, which have been set forth by an advisor to Natural Resources Canada.98 (See a relevant quote in Section 2, above.) Part of the governments rationale is that the $650 million limit addresses foreseeable, rather than catastrophic Chernobyl-type accidents.99 This statement alleges that a Chernobyl-scale release of radioactive material is not foreseeable. Yet, that very event occurred in 1986. Also, the 1979 Three Mile Island accident could, with comparatively minor variation, have yielded an atmospheric release comparable to that at Chernobyl. Moreover, government and industry studies performed in the USA (see Section 5, above) and Canada (see Section 7, above) show that large, accidental releases of radioactive material are entirely foreseeable. The potential for such releases is further exacerbated by the well-recognized possibility of malevolent acts at NPPs. One could reasonably expect that the Canadian governments choice of a $650 million limit would be based upon detailed analysis, informed by practical experience, PRAs, and assessments of the risk of malevolent acts. No such analysis is evident. A related report was prepared for CNSC, but it considered only design-basis accidents.100 The releases considered in that report included only noble gases and iodine isotopes. The largest Iodine-131 release considered was 24 TBq.101 Yet, CNSCs safety goals for new NPPs categorize a small release as a release that exceeds 1,000 TBq of Iodine-131. (See Table 5-3.) economIc ImpAct of ceSIum-137 depoSItIon In toRonto A study sponsored by Defence Research and Development Canada provides useful information about the economic impact of an unplanned release of radioactive material. The study estimated the economic impact of an open-air explosion of a radiological dispersal device (i.e., a dirty bomb) at the CN Tower in Toronto.102 The assumed release consisted of 37 TBq of Cesium-137. The estimated economic impact varied considerably, according to the cleanup standard that was assumed in the analysis. That standard was expressed in terms of the radiation dose rate that would remain after completion of the cleanup. For a cleanup standard of 5 mSv per year, the estimated economic impact would be $28 billion, whereas for a cleanup standard of 0.15 mSv per year the impact would be $250 billion. The magnitudes of those impacts are interesting, considering that the assumed release (37 TBq of Cesium-137) is a tiny fraction of the release that could occur from an NPP. CNSCs safety goals for new NPPs categorize a large release as a release that exceeds 100 TBq of Cesium. (See Table 5-3.) Each NPP at the Darlington site in Ontario where there are four CANDU plants contains about 65,000 TBq of Cesium-137 in its reactor core.103 ImplIcIt SubSIdy vIA the nlA The Canadian government gave financial support to a study, by two UK analysts, of the implicit subsidy that arises from the provisions of the Nuclear Liability Act.104 The study sought to take an empirical approach to estimating the implicit subsidy. It generated a range of findings for that indicator, expressed as cent per kWh of nuclear generation. The approach taken was interesting. Unfortunately, however, the study had severe flaws as follows: (i) the study assumed a formula, for the probability density function of monetized offsite damages, that lacked a credible rationale; (ii) the assumed formula can, depending on the parameters chosen, yield a probability density function with a very long right-hand tail that is inconsistent with the phenomena that can cause offsite damages; (iii) the study did not disclose the parameters used for its formula; (iv) the analysts misunderstood expert findings regarding worst-case damages, believing (incorrectly) that these findings referred to the probability that damages would be equal to or greater than a particular $ value; and (v) equation (3) of the study shows a lower bound of integration that is too high by six orders of magnitude.
28

98 Henault, 2009. 99 Henault, 2009. 100 ISR, 2003. 101 ISR, 2003, Tables 7 and 35. 102 Cousins and Reichmuth, 2007. 103 ISR, 2003, Table 36. Also see Table 7-1 of this report, which shows a Darlington NPP reactor core inventory of 67,000 TBq of Cesium-137. 104 Heyes and Heyes, 2000.

SeveRe AccIdent Study An engineer who apparently has long experience in the Canadian nuclear industry, John W. Beare, has commented on the Canadian practice of not examining the consequences of the largest potential releases from NPPs, stating:105 If the Commission [CNSC] is concerned about the cost-benefit aspects of its safety requirements it could start by completing the Severe Accident Study research project started about 1988 but never completed. The conclusion of the preliminary study is that, in the event of a catastrophic accident, a release of radioactive material proportionately as large as that from Chornobyl could not be ruled out. In the case of a water-cooled reactor like CANDU such a release could be in the form of a relatively cool aerosol and not be dispersed as much as at Chornobyl. The radiation doses close to the reactor could be higher than at Chornobyl. It appears that Beare is referring to a Severe Accident Study conducted under the aegis of the Atomic Energy Control Board, whose functions were taken over by CNSC in 2000. Beare says that the study was initiated in 1988, while NRC was working on its NUREG-1150 study. The NUREG-1150 study was completed in 1990, establishing the high point of PRA practice for NPPs.106 Canadas analogous study was, apparently, abandoned.

9. This Reports Estimation of Risk-Related Costs

9.1 Scope of this Discussion
Drawing upon analysis in preceding sections of this report, Sections 9.2 through 9.5 address the cost implications of the risk of operating NPPs in Canada. Section 9.2 discusses the costs of measures intended to reduce risk. Section 9.3 provides quantitative estimates of the risk costs of offsite impacts of unplanned radioactive releases. The releases could arise from accidents or malevolent acts. Section 9.4 provides quantitative estimates of the risk costs of onsite impacts of fuel-damage events. An overview of risk costs is provided in Section 9.5. The estimates presented here reflect assumptions and sources that are identified in the narrative of the report, and in the notes to tables and figures. A reader could repeat the analysis with different assumptions. Here, the term risk costs refers to monetized impacts of potential radioactive releases, on a per annum basis. The monetization of impacts involves assumptions, as discussed below.

9.2 Costs of Measures Intended to Reduce Risk

As discussed in Section 6, above, options are available to reduce the risk of operating NPPs. Each option has a cost, which can be compared to its benefit. The NLCA would be a mechanism for risk mitigation, by providing compensation for damage. Risk reduction and risk mitigation are closely related, and should be considered within the same framework. Sustainability, the precautionary principle, and full accounting of costs are the three pillars of that framework, as discussed in Section 3, above. The NLCA would impose direct costs, in the form of premiums. If the premiums are insufficient to cover the risk costs, the NLCA would also impose indirect costs, in the form of an implicit public subsidy for nuclear power. That subsidy could crowd out investment in risk-reduction options. Some of those options, and their costs, are sketched here. Experience in the USA during the 1970s and 1980s provides important information about the cost implications of efforts to reduce the risk from operating NPPs. During that period there was growing awareness of safety issues, leading to actions that involved cost increments. The growth of awareness was significantly, but not exclusively, attributable to the occurrence of the TMI accident in 1979.
29

105 Beare, 2005, paragraph 192. 106 NRC, 1990.

Charles Komanoff, in a book published in 1981, examined the escalating trends of costs associated with nuclear generation in the USA.107 He showed that efforts to reduce risk were a major driver of cost escalation, and he predicted that this effect would continue during the 1980s. A subsequent compilation of data showed that his prediction was correct.108 Construction/capital costs in the 1970s averaged 1.95 cent per kWh (1990 $), but rose to an average of 3.51 cent per kWh (1990 $) in the 1980s. Annual capital additions grew from an average of 0.35 cent per kWh (1990 $) in the 1970s to 0.89 cent per kWh (1990 $) in the 1980s.109 Efforts to reduce risk were a major driver of those trends. Analysts examining the potential for a nuclear power renaissance are well aware of the history of cost escalation.110 Plant vendors and other advocates of the renaissance recognize that substantial cost escalation will prevent their ambitions from being realized. They hope to curb this escalation through measures such as standardizing of designs and streamlining of regulation. It is not clear, however, that they fully appreciate the potential for an unplanned release, at any NPP in the world, to override those measures.111 Such an event, whether caused by an accident or a malevolent act, would increase public pressure for adoption of riskreducing measures at plants in Canada and elsewhere. That pressure could become especially powerful if the public became aware that the nuclear industry had rejected innovative plant designs such as the PIUS design in favor of Generation III designs that pose a higher risk. exAmpleS of the coSt of RISk ReductIon Table 9-1 provides an example of a capital-addition cost that would lead to risk reduction at a new or existing PWR. This example would be relevant to Canada if a PWR plant were built at the Darlington site or elsewhere. The cost would arise from the re-equipping of the plants spent-fuel pool with low-density racks, an option shown in the first row of Table 6-1.112 That measure would require the transfer of spent fuel to dry storage after 5 years of storage in the pool. Another example of a cost of risk reduction is the additional expenditures in Canada since 2001 to enhance security measures at nuclear installations. Capital costs for these measures have totaled about $300 million, and ongoing costs are about $60 million per year.113 Licensees are bearing the majority of these costs. It can be presumed that most of the expenditure has been at NPPs.

9.3 Risk Costs of Offsite Impacts of Radioactive Releases

The potential for unplanned radioactive releases from NPPs in Canada is discussed in Section 7.3, above, with a focus on atmospheric releases. That discussion also addresses the offsite radiological impacts of releases, with a focus on lifetime population dose (collective dose commitment). Table 7-1 summarizes information that is relevant to an assessment of the risk costs of offsite impacts. Note that Table 7-1 considers the potential for releases from reactor cores and spent-fuel pools. monetIzIng RAdIologIcAl ImpActS One of the steps in assessing the risk costs of radiological impacts is to assign a monetary value to radiation dose. Here, the relevant dose is the lifetime population dose arising from a release. That dose is statistically linked to the incidence of radiation-caused morbidity and mortality in the exposed population. The potential for a population dose, as a result of an unplanned release from an NPP, would be a foreseeable outcome of a decision to operate the plant with a particular level of risk. As discussed in Section 6, above, options are available to reduce the risk. Implementation of those options would involve expenditures. Thus, the monetary value to be assigned to population dose should reflect the tradeoff between the cost of receiving a dose and the cost of avoiding that dose. That tradeoff is made routinely in the operation of NPPs, in the context of small, routine releases of radioactive material. The tradeoff is formalized through the concept of keeping radiation exposure as low as reasonably achievable (ALARA).

30 107 108 109 110 111 Komanoff, 1981. Komanoff and Roelofs, 1992. Komanoff and Roelofs, 1992, pp 17-20. Hultman et al, 2007. The 1986 Chernobyl accident had a less visible effect on cost trends than did the 1979 TMI accident. Two factors may explain that outcome. First, the Chernobyl accident occurred in a closed, non-Western society. Second, annual capacity additions were already beginning to decline in 1986. 112 The cost estimate in Table 9-1 assumes that the pool would be re-equipped with low-density racks prior to the 11th year of plant operation. 113 Frappier, 2007.

CNSC has provided regulatory guidance for implementing the ALARA concept.114 The CNSC guidance document notes that implementation of ALARA requires the assigning of a monetary value to population dose, and refers the reader to an IAEA report that discusses specific monetary values.115 The IAEA report is titled, Optimization of Radiation Protection in the Control of Occupational Exposure.116 Table III-2 of the IAEA report shows that owners of US nuclear power plants were, in the early 1990s, assigning an average value of US $1 million to each person-Sv of occupational exposure. The same value (in Can $) was used at the Gentilly plant in Canada. Analysis published in 1992, by a team led by this author, noted that NRC was then recommending a value of $1,000 per person-rem ($100,000 per person-Sv) for population dose in the ALARA context. NRC also used a dose value of $1,000 per person-rem at that time to determine if a risk-reducing plant modification was economically justified. A value of $1,000 per person-rem, if updated to a 1992 value to account for inflation and new scientific information about the health effects of radiation, would amount to $1 million per person-Sv.117 In this report, a potential population dose is assigned a value of $1 million per person-Sv in 1992 Can $. That value is adjusted upward to 2009 Can $ by a factor of 1.36, a CPI inflator provided by Bank of Canada. eStImAtIng RISk coStS As discussed in Section 7, above, Table 7-2 shows licensee estimates of frequency and population dose for potential accidental atmospheric releases from some CANDU stations. The population dose can be monetized as described above. Table 9-2 shows the results of that monetization for the Bruce A station. Similar tables could be prepared for the Bruce B and Pickering B stations, using data from Table 7-2. Table 9-2 shows, for example, that release EPRC 2 from the Bruce A station would cause offsite health damage of $49.1 billion. The licensee estimates the probability of an EPRC 2 release at 3.6 per 10 million RY. Release EPRC 7 has a higher estimated probability 2.6 per 100,000 RY and would cause offsite health damage of $1.05 billion. Note that these probability estimates are mean values, and the 95th percentile value could be an order of magnitude higher. Also, these probability estimates do not account for external initiating events and malevolent acts. As shown in Table 5-3, CNSC specifies a large-release probability of no more than 1 per 1 million RY as a safety goal for new NPPs. That probability value is often said by regulators and the nuclear industry to be a threshold, above which potential accidents are credible. Using that criterion, release EPRC 7 at Bruce A is clearly credible, and release EPRC 2 would be credible if external initiating events, uncertainty, and malevolent acts were considered.118 Table 7-1 assembles a set of selected data about radioactive releases and their offsite impacts. Those data are used to develop the release cases shown in Table 9-3. Application of a population-dose value of $1.36 million per person-Sv to those cases yields the risk costs shown in Table 9-3. In that table, the CANDU plants at the Darlington site represent the existing CANDU plants across Canada. Table 9-4 extends Table 9-3 by showing the risk costs for specific values of: (i) the plants capacity factor; and (ii) the probability of a release caused by malevolent action. In addition, Table 9-4 combines the risk costs for accidental and malevolent releases. The capacity factors shown in Table 9-4 and elsewhere in this report are illustrative, and do not imply that NPPs in Canada will achieve any particular capacity factor. The malevolent release probabilities (MRPs) shown in Table 9-4 are not the product of statistical analysis. Instead, they provide a range of quantitative probabilities that serves as a proxy for a qualitative assessment of the potential for malevolent action. Table 9-4 shows some very large risk costs for releases from spent-fuel pools at new plants. In the worst instance, the risk costs for pool releases would be 99 cent per kWh.119 That is a remarkable finding, but is a true reflection of the risk posed by storage of a large amount of spent nuclear fuel in a high-density pool.

31 CNSC, 2004. CNSC, 2004, page 3. IAEA, 2002. IRSS, 1992, Volume 2, page 22. As another example, consider release EPRC 5A at the Pickering B station. Table 7-2 shows an estimated probability of 7.1 per 10 million RY for this release. Monetization of the health impact of this release at $1.36 million per person-Sv yields damage of $1.2 billion. 119 In this instance, pool release cases 1B-H (high accidental-release probability) and 2B (high malevolent-release probability) are combined, and the plants capacity factor is 0.8. 114 115 116 117 118

As discussed in Section 6, above, options are available for dramatically reducing the risk of a release from a spent-fuel pool, especially by reverting to the use of low-density racks and transferring fuel to dry storage. The cost of implementing that option would be comparatively modest, as shown in Table 9-1. This report assumes implementation of that option. Table 9-5 simplifies the findings in Table 9-4, by excluding releases from spent-fuel pools and by assuming that the capacity factor of each plant would be 0.9. Further simplification is provided in Table 9-6, which sets forth policy-applicable risk costs of offsite impacts of radioactive releases. The term policy-applicable means that these estimates are appropriate for policy purposes, such as assessing the NLCA. Table 9-6 considers releases caused by accidents and by malevolent acts, and shows risk costs for existing plants and new plants.

9.4 Risk Costs of Onsite Impacts of Fuel-Damage Events

The potential for onsite impacts of fuel-damage events at existing CANDU plants in Canada is discussed in Section 7.4, above. Table 7-3 shows an estimate by Ontario Hydro of the risk of such onsite impacts. The risk is expressed in 2009 Can $ per RY of plant operation. Table 9-7 converts Ontario Hydros findings to risk costs of onsite impacts of fuel-damage events. Ontario Hydro considered only accidents initiated by internal events. It is reasonable to double Ontario Hydros estimate of risk to account for external initiating events and malevolent acts. It is also reasonable to use the Darlington estimate for all existing CANDU plants in Canada, because all but two of these plants are in multiunit blocks. Note that the dominant component of the risk costs shown in Table 9-7 is the cost of replacement power due to forced outage of NPPs. (See the notes to Table 7-3.) Analysts considering the economics of nuclear generation in Canada should be wary of double counting the costs of forced outages.

9.5 An Overview of Risk Costs

The estimated risk costs in Tables 9-6 and 9-7 are combined in Table 9-8, which summarizes this reports overall findings regarding the risk costs of nuclear generation in Canada. Also shown in Table 9-8 are the estimated insurance premiums that would be paid under the NLCA to provide coverage of the risk costs. Table 9-8 shows that operation of an existing NPP in Canada creates offsite risk costs of 2.7 to 5.4 cent per kWh of electricity produced, plus additional onsite risk costs of 2.7 to 5.6 cent per kWh. Under the NLCA, the only evident accounting for these risk costs as costs of production would be through payment of insurance premiums amounting to 0.02 cent per kWh, to cover offsite damage to third parties up to a $650 million limit. Thus, the public now provides, and would provide under the NLCA, an implicit subsidy to NPP operators of 5.4 to 11.0 cent per kWh. In 2007, Canadian nuclear generation was 88.2 billion kWh.120 Thus, the total, implicit subsidy across all Canadian NPPs during 2007 was $4.8 billion to 9.7 billion. conSIStency of theSe fIndIngS wIth nucleAR InSuReRS ASSumptIonS The analysis underlying Table 9-8 involves various uncertainties and assumptions. Other analysts might make different assumptions, yielding different findings. However, the assumptions made here are reasonable, and are consistent with assumptions made by nuclear insurers. Limited information is publicly available regarding the premiums set by nuclear insurers. Nevertheless, enough information has been disclosed to show how nuclear insurers view the risk posed by operating NPPs. Underlying Table 9-8 are assumptions including: (i) the probability of an accidental release from an existing NPP in Canada is 1 per 10,000 RY; and (ii) the probability of a malevolent release is between 1 per 1 million RY and 1 per 10,000 RY. (See Table 9-6.) These probability values can be compared with the event probabilities that are implicit in nuclear premiums. To make this comparison, the indicator used here is implied probability of event (IPOE), defined as the insurance premium (in $ per RY) divided by the covered amount (in $). IPOE is a simplified event probability (per RY) that assumes perfect forecasting and no profit or administrative costs.

32

120 Statistics Canada, 2009.

In Canada, a premium of $125,000 per RY was paid in 1995 for $75 million of coverage. (See Section 2, above.) That is equivalent to an IPOE of 1.7 per 1,000 RY. In the USA, a premium of $400,000 per RY is paid for $300 million of coverage. (See Section 4, above.) That is equivalent to an IPOE of 1.3 per 1,000 RY. In France, a premium of 110,000 Euro per RY was paid in 2008 for 31 million Euro of coverage. (See Section 4, above.) That is equivalent to an IPOE of 3.5 per 1,000 RY. Clearly, commercial nuclear insurers do not accept nuclear industry claims that the probability of a radioactive release is very small. Instead, the insurers take a conservative position that is consistent with the empirical record, including the Chernobyl and Three Mile Island events. The probability assumptions underlying Table 9-8 are less conservative than the insurers assumptions. It should also be noted that commercial insurers refuse to provide full coverage of $650 million of operator liability in Canada (see Section 2, above) and 700 million Euro of liability in Europe (see Section 4, above).

10. Regulatory Issues and Secrecy

In all industrial sectors except the nuclear sector, the public is protected against adverse, unplanned impacts by two major mechanisms. First, corporations are fully liable for economic damage arising from adverse impacts, which creates a powerful incentive to adopt measures that enhance safety and security. Insurance companies reinforce the incentive by setting premiums, deductibles, and other conditions that account for risk. Second, government regulators impose requirements related to safety and security. The NLCA would, by limiting an NPP operators liability, hinder the functioning of the first mechanism. The primary mechanism for protecting the public would be government regulation, in the form of CNSC requirements related to safety and security. Thus, an assessment of the NLCA must account for the quality of CNSC regulation. That subject is addressed here, with attention to the role of secrecy. At least four major questions arise regarding the quality of CNSC regulation. First, are CNSCs criteria for the design and siting of new NPPs adequate? Second, given CNSCs present reliance on a probabilistic paradigm of safety, is the PRA culture in Canada adequate to support that paradigm? Third, does CNSC have the necessary independence and authority to perform its functions? Fourth, what is the role of secrecy? deSIgn And SItIng cRIteRIA CNSCs design and siting criteria for new NPPs reflect a paradigm in which potential accidents and malfunctions at an NPP are categorized as design-basis events that would not lead to a substantial radioactive release if the plant functioned as designed, and beyond-design-basis events that could lead to such a release.121 Events in the latter category are addressed probabilistically, using the concept of risk.122 Although purportedly scientific, the risk concept as currently applied is actually a form of dogma. The probabilistic paradigm reflects fundamental weaknesses in design. Generation II nuclear power plants, and the proposed Generation III plants, are unable to ride out a variety of credible events outside their design basis. An alternative paradigm is available, as illustrated by the proposed design and siting criteria set forth in the Appendix to this report. Within the traditional paradigm, CNSC has progressively weakened the plant design criteria that are set forth in CNSC document RD-337. Two instances of weakening have been documented by this author.123 First, during preparation of the Draft version of RD-337, CNSC withdrew a requirement to prioritize inherent safety. That action was taken to allow the continued operation of existing CANDU plants, and to promote the sale of new CANDU 6 plants by AECL. Second, during the transition from Draft to Final versions of RD-337, CNSC weakened its quantitative safety goals. That action made it easier for plant vendors to claim that their designs comply with the safety goals. cAnAdAS pRA cultuRe As discussed in Sections 5 and 7, above, Canada never achieved a fully developed PRA culture.124 Moreover, Canadas PRA culture has degenerated during the past two decades. Now, PRAs are performed in secret and find extremely low probabilities for large releases. Based on those probability findings, the PRAs do not estimate the environmental impacts of large releases. Yet, the low probabilities are not credible. The practice of ignoring large releases deprives citizens and policy makers of needed information.
121 For additional information, see: Thompson, 2008a. 122 In CNSC's risk paradigm, the probabilities of beyond-design-basis threats (events involving malevolent acts) are addressed qualitatively, but not always explicitly. Often, a threat is ignored, which implicitly assigns a probability of zero to that threat. 123 Thompson, 2008a, Section 4.2. 124 For additional information, see: Thompson, 2008a.

33

Canadas isolation from best practice in the PRA field leaves CNSC unprepared to implement its probabilistic safety paradigm. For example, CNSC currently lacks the ability to credibly determine if a new plant complies with the quantitative safety goals set forth in Table 5-3. cnScS Independence And AuthoRIty A credible regulator must be able to demonstrate, on a sustained basis, its independence from political pressure, and its ability to exert authority.125 In the case of CNSC, a related challenge is to discard CNSCs traditional regulatory approach, which has ad hoc and incestuous qualities, and to adopt a more modern and professional approach. One piece of evidence that illustrates the deficiencies in the traditional CNSC approach is the progressive weakening of the document RD-337 as it moved from pre-Draft to Draft to Final versions. Supporting evidence includes the vagueness of the life extension requirements in CNSC document RD-360.126 Contrasting evidence, indicating the existence of professionalism within the CNSC, includes the CNSC Staffs insistence that Ontario Power Generation conducts detailed studies of safety issues related to life extension of the Pickering B units.127 Tension between the CNSCs traditional regulatory approach and a more professional, politically-independent approach is illustrated by events in late 2007 and early 2008 related to AECLs operation of the NRU reactor at Chalk River.128 That reactor had been producing a substantial fraction of the radioisotopes used for medical tests and procedures worldwide. Its continued operation was particularly important in light of AECLs failure to make the MAPLE reactors operational. In November 2007, CNSC ordered AECL to cease operation of the NRU reactor, pending the upgrading of safety systems. CNSC had been dissatisfied with AECLs progress in making the upgrade. In December 2007, the Canadian parliament voted to override CNSCs order, and the NRU reactor was re-started. Continuing conflict between the CNSC President and the Canadian government led to the Presidents dismissal in January 2008. An independent evaluation of these events yielded findings including the following:129 Based on a review of these events, and related internal and external communications of both organizations, a fundamental observation of the Talisman Team is that the CNSC regulatory program and the AECL regulatory compliance program are expert based and not process based. The regulatory effectiveness of both organizations can be significantly improved by developing and implementing formal processes, to be used for establishing and complying with regulatory requirements. The Canadian governments dismissal of the CNSC president, Linda Keen, created concern among nuclear regulators in other countries. At the 2008 Nuclear Safety Convention review meeting, Keens dismissal was discussed as follows:130 Her case was cited as illustrating insufficient regulatory independence, and the final statement from the 2008 meeting underlined the need to maintain strong and independent regulators. Keen has alleged that the NRU situation was an excuse to fire her. She cited as a larger reason her earlier refusal to grandfather safety standards for AECLs CANDU 6, a type of NPP designed in the 1970s, insisting instead that the CANDU 6 be reviewed against 2007 standards. 131 That refusal reduced AECLs prospects for marketing the CANDU 6 in Canada and elsewhere. AECL is a politically influential Crown corporation. SecRecy And npp RISk The nuclear power industry and its regulators, in Canada and elsewhere, are prone to secretive behavior. For some purposes, such as protection of trade secrets or the privacy of personnel, secrecy is widely accepted and is comparatively harmless. In the context of safety and security, however, secrecy has significant, negative impacts. Experience shows that an entrenched culture of secrecy is not compatible with a clear-headed, sciencebased understanding of the risk of NPP operation. Entrenched secrecy perpetuates dogma, stifles dissent, creates opportunities for corruption, and can create a false sense of security. In illustration, the culture of secrecy in the former USSR was a major factor contributing to the occurrence of the 1986 Chernobyl reactor accident.132
125 126 127 128 129 130 131 132 CNSC, 2007b. CNSC, 2008c. Schaubel, 2008. CBC News, 2008. Talisman, 2008, page i. MacKenzie and MacLachlan, 2009, page 9. MacKenzie and MacLachlan, 2009, page 10. Thompson, 1998.

34

Some countries have a tradition of governmental secrecy. The USSR was a prominent example, and suffered accordingly. Other countries, including Canada and the USA, have recognized the benefits of an open society. In the latter group of countries, a considerable amount of information about NPPs and their risks has been publicly available until recent years. Much of this information has been accessible through the regulatory agencies. Since the attacks on New York and Washington in September 2001 that situation has changed, as discussed below. Prior to September 2001, there were differences between the USA and Canada regarding the availability of information about accidents and malfunctions at NPPs. One difference related to the creation of information, rather than its dissemination. In the USA, numerous PRAs were conducted at Level 3, and there was considerable scientific and public debate about the risks of large releases of radioactive material. In Canada, by contrast, the environmental impacts of large releases were not assessed. Also, the Canadian nuclear industry and its regulator were comparatively reluctant to disclose information that they created.133 As a result of these factors, Canada has seen less scientific and public debate about the risks of large releases than has occurred in the USA. Since September 2001, NRC has become notably more secretive. NRC has justified that position by pointing to the risk of attack on NPPs and other nuclear facilities. Yet, NRC has not required its licensees to significantly strengthen the existing plants. Nor has NRC required that robustness against attack be a major design objective for new plants. One could reasonably suspect that a major motive for NRCs secrecy is protection of the status quo. NRC opposes requests by state and local governments and citizen groups that malevolent acts be considered in EISs.134 An oppoRtunIty to Reduce SecRecy In cAnAdA CNSC has departed from current US practice by including resistance to malevolent acts in its proposed design criteria for new NPPs. Similarly, CEAA guidelines depart from US practice by requiring consideration of malevolent acts in an EIS for a new NPP. CNSC, CEAA, and the other relevant agencies deserve commendation for this new approach. It remains to be seen, however, if CNSC will accommodate the consideration of malevolent acts without resorting to excessive secrecy. CNSC could greatly reduce the need for secrecy by requiring that new NPPs be highly robust against attack. If a plant is robust against credible attacks, secrecy about its design and operation serves no purpose. For example, the proposed design and siting criteria set forth in the Appendix would provide a high degree of robustness. If new NPPs in Canada were designed to meet such criteria, secrecy about their design and operation would not be necessary. the wIdeR context of SecRecy A declared motive for the secrecy that now surrounds the nuclear industry is fear that a nuclear installation could be attacked by a sub-national group. Yet, secrecy has substantial adverse effects on a society. Thus, a rush to secrecy hands a victory to hostile, sub-national groups. A more mature, multi-faceted response is needed. Table 10-1 illustrates the factors that should be considered in developing an integrated strategy to protect NPPs and other elements of critical infrastructure. One of the protective approaches described in Table 10-1 is to build infrastructure facilities that are highly robust and inherently safe. The incremental costs of that approach are not necessarily onerous. In illustration, Table 9-1 shows the estimated incremental cost of transferring spent fuel from a PWR spent-fuel pool to dry storage after 5 years of storage in the pool. That action would allow the pool to be re-equipped with low-density storage racks. If the pool were re-equipped in that manner, attackers could no longer achieve combustion of the spent fuel simply by removing water from the pool. The incremental cost, beginning in the 11th year of plant operation, would be 0.04 cent per kWh.

35 133 For example, the performance of an independent, focused review of the Darlington Probabilistic Safety Evaluation was initially hindered by Ontario Hydro, which delayed its provision of a full set of DPSE documents for more than 6 months. When those documents were eventually provided, they revealed significant deficiencies in DPSE. See: IRSS, 1992, Volume 2, Annex I, pp a-c. 134 Since 2002, NRC has been in litigation with a group of intervenors, led by San Luis Obispo Mothers for Peace, who have requested that an EIS be prepared for a new spent-fuel-storage installation at the Diablo Canyon site, to address the environmental impacts of malevolent acts. The intervenors are supported by a ruling from the 9th Circuit of the US Court of Appeals. NRC refuses to implement that ruling at sites beyond the reach of the 9th Circuit, and the intervenors allege that NRC has not properly implemented the ruling in the Diablo Canyon instance. The interests of the intervenors extend beyond the Diablo Canyon spentfuel-storage installation. They seek a legal precedent with wider application.

11. Deficiencies in the NLCA, and Options for Improvement

Five criteria for assessing the NLCA are set forth in Section 3.4, above. Information and analysis provided in this report allow the NLCA to be assessed against those criteria. Deficiencies in the NLCA are evident, as discussed in the following paragraphs. nlcA defIcIencIeS RelAted to full AccountIng In regard to Criterion #1 (Full accounting information disclosure), the NLCA would not require Canadas nuclear industry and government to publish a full spectrum of information relevant to nuclear risk and nuclear insurance. Thus, the present situation would continue, as described in various sections of this report. Information published in Canada about the risk posed by operation of NPPs is often incomplete or misleading, and there is a dearth of information about the operation of the nuclear insurance sector. This situation is incompatible with democracy, given the public-policy significance of nuclear risk and nuclear insurance. In regard to Criterion #2 (Full accounting damage coverage), the NLCA would not ensure that commercial insurance covers the full range of potential damage arising from incidents at NPPs. The liability of commercial insurers would be limited to $650 million, and a substantial portion of that exposure would be reinsured by the Canadian government. Monetized damage to third parties could far exceed $650 million. Studies by the Canadian nuclear industry show that offsite health costs from some potential accidents at existing NPPs could exceed $50 billion. Table 9-2 shows estimated offsite health costs from potential accidental releases at the Bruce A station, derived from an industry study. The EPRC 2 release would impose health costs of $49.1 billion. Application of the same methodology to the Pickering B station shows that an EPRC 1 release at that station would impose health costs of $52.0 billion.135 The nuclear industry says that accidents with such large consequences have low probability. Yet, offsite health costs could exceed $1 billion for accidents that the industry concedes are credible because their probability is comparatively high. For example, Table 9-2 shows that an EPRC 7 release from the Bruce A station would impose health costs of $1.05 billion, and the probability of the release is 2.6 per 100,000 RY.136 That probability is well above the credibility threshold of 1 per 1 million RY that is often cited by industry and regulators. Moreover, there are reasons to doubt that accident probabilities are as low as the nuclear industry claims. Nuclear insurers assume much higher probabilities. (See Section 9.5, above.) Also, the potential for malevolent acts must be considered. Overall, the probability of an event causing offsite damage exceeding $50 billion is substantially higher than is shown by industry studies. (See Sections 5 and 7, above.) In the USA, a two-tier system of nuclear insurance provides about $11 billion of coverage for damage to third parties. In Germany, there is no limit on an NPP operators liability for damage to third parties. Each NPP operator in Germany must provide security of 2.5 billion Euro toward its liability. (See Section 4, above.) An unplanned release of radioactive material could cause economic damage to the operator of an NPP even if most of the released material is contained within the plant buildings. Studies by the nuclear industry show that accidents with a comparatively high probability could cause damage to the operator in excess of $650 million. Table 7-3 shows findings from a study by Ontario Hydro regarding the onsite costs of potential accidents at the Darlington nuclear station. For example, an accident in the FDC 7 category has an estimated probability of 3 per 1,000 RY (i.e., 48 percent over 40 years, at this 4-unit station) and would cause damage between $790 million and $2,500 million to the operator. It appears that no insurance coverage of such an accident exists now in Canada or would exist under the NLCA. Although the damage would directly affect the operator, its costs would ultimately be borne by society at large. Insurance coverage of damage to the operator is provided in the USA. (See Section 4, above.)
36

135 Table 7-2 shows that an EPRC 1 release at Pickering B would cause a population dose of 38,200 person-Sv. Monetization of that dose at $1.36 million per person-Sv yields health costs of $52.0 billion. 136 If an accidents probability is 2.6 per 100,000 RY, then its cumulative probability over 40 years for a population of 18 NPPs (Canadas operational population of NPPs at present) is 2 percent.

In regard to Criterion #3 (Full accounting accounting for risk costs as costs of production), the NLCA would not ensure that the risk costs of operating each NPP are accounted for as costs of production. This report contains quantitative analysis to address that issue, and the summary findings are presented in Table 9-8. That table shows that risk costs would far exceed insurance premiums. There is no other mechanism to incorporate risk costs as production costs. Thus, most of the risk costs would be borne by the public at large, representing a large, implicit subsidy of nuclear power. For example, Table 9-8 shows that operation of an existing NPP in Canada creates offsite risk costs of 2.7 to 5.4 cent per kWh of electricity produced, plus additional onsite risk costs of 2.7 to 5.6 cent per kWh. Under the NLCA, the only evident accounting for these risk costs as costs of production would be through payment of insurance premiums amounting to 0.02 cent per kWh, to cover offsite damage to third parties up to a $650 million limit. Thus, the public now provides, and would provide under the NLCA, an implicit subsidy to NPP operators of 5.4 to 11.0 cent per kWh. In 2007, the total, implicit subsidy across all Canadian NPPs was $4.8 billion to 9.7 billion for the year. (See Section 9.5, above.) Such a large subsidy is a significant violation of the polluter-pays principle. (See Section 3, above.) The analysis underlying Table 9-8 involves various uncertainties and assumptions. Other analysts might make different assumptions, yielding different findings. However, the assumptions made here are reasonable, and the findings are robust. Risk costs exceed insurance premiums by a factor of 270 to 550. Although limited information is publicly available regarding the premiums set by nuclear insurers, accessible data show that nuclear insurers make assumptions about nuclear risk that are consistent with the assumptions made here. (See Section 9.5, above.) nlcA defIcIencIeS RelAted to the pRecAutIonARy pRIncIple In regard to Criterion #4 (The precautionary principle), the NLCA would not be compatible with coherent and consistent application of the precautionary principle in Canada. Underlying the NLCA is an assumption that consideration of the risk of unplanned radioactive releases from NPPs should, in the practical context of liability and insurance, be confined to comparatively small releases. That assumption is implicit in the choice of $650 million as a liability limit, and is confirmed by official sources. (See Sections 2 and 8, above.) Yet, that assumption is contradicted by studies conducted by the Canadian nuclear industry, analogous studies conducted in other countries by industry and government, experience with the 1986 Chernobyl accident and the 1979 Three Mile Island accident, and a wellrecognized potential for malevolent acts at NPPs. Those sources of information show that damage far in excess of $650 million is a realistic possibility. (See Sections 5 and 7, above.) The NLCA ignores that possibility, and would, therefore, be incompatible with the precautionary principle. (See Section 3, above.) Moreover, the NLCAs liability limit has no basis in either systematic technical analysis or public debate, both of which are central to application of the precautionary principle. A relevant technical analysis the Severe Accident Study was initiated by the Canadian government in the late 1980s but never completed. (See Section 8, above.) nlcA defIcIencIeS RelAted to SuStAInAbIlIty In regard to Criterion #5 (Sustainability), the NLCA would not be compatible with a transition of Canadas economy toward sustainability. There are at least three major aspects of this incompatibility. First, the NLCA would violate the precautionary principle, thereby tending to suppress policies and actions flowing from that principle. As a result, opportunities to reduce the risk of NPP operation would be lost, and Canada would bear an unnecessarily high risk of a radioactive release with devastating impacts on the environment, economy, and society. (See Section 6, above, for a discussion of options for risk reduction.) Second, the NLCA would create a large, implicit public subsidy for nuclear power, thereby distorting decisions about public and private investment in systems for production and use of energy. (See the discussion above regarding Criterion #3.) In that way, the NLCA would suppress innovation in sustainable energy systems, and would inhibit market processes for cost-effective deployment of such systems. Third, the NLCA would continue the present climate of incomplete and often misleading information about nuclear risks and nuclear insurance. (See Sections 2, 5, 7, 8 and 10, above.) That climate is antithetical to the open, science-based, and participatory exchange of information that is necessary for Canadas transition to sustainability.

37

IncompAtIbIlIty of the nlcA wIth exIStIng cAnAdIAn lAw The polluter-pays principle was upheld unanimously by the Supreme Court of Canada in a 2003 decision. The precautionary principle has important roles in the Canadian Environmental Assessment Act of 1992 and the Federal Sustainable Development Act of 2008. (See Section 3, above.) Yet, the NLCA would violate both the polluter-pays principle and the precautionary principle. Thus, the NLCA would be incompatible with existing Canadian law, and may be incompatible with Canadas international commitments. Also, the NLCA would create subsidies for nuclear generation of electricity. One subsidy is implicit in the $650 million liability limit. Another subsidy would be the Canadian governments reinsurance of risk that the nuclear insurers refuse to cover. (See Section 2, above.) Both subsidies may be incompatible with Canadas existing laws and international commitments. oPtioNs foR iMPRoViNG thE NLCA The preceding paragraphs identify significant deficiencies in the proposed NCLA. Options that could address those deficiencies, to varying extents, include: option #1: Disclosure of nuclear insurance data The NLCA would require the timely publication of data on nuclear insurance coverage, premiums, reinsurance, and compensation. This option would have negligible direct cost. It would help to provide the open, science-based, and participatory exchange of information that is necessary for Canadas transition to sustainability. option #2: full assessment of NNP risk Voting by Parliament on the NLCA would be preceded by publication of a comprehensive assessment of the risk posed by NPP operation in Canada. The Canadian government could prepare that assessment, building upon the Severe Accident Study that the government initiated in the late 1980s but never completed. (See Section 8, above.) As part of Option #2, the NLCA could provide for its own review at specified intervals, preceded each time by an updated NPP risk assessment. option #3: full commercial insurance The NLCA would prohibit nuclear reinsurance by the Canadian government. This option would help to ensure that nuclear insurance premiums reflect the full risk of NPP operation. option #4: increased liability limit The NLCA would have a third-party liability limit much higher than $650 million, or no limit. Other countries have demonstrated this option. (See Section 4, above.) In the USA, a two-tier system of nuclear insurance provides about $11 billion of coverage for damage to third parties. In Germany, there is no limit on an NPP operators liability for damage to third parties, and each NPP operator must provide security of 2.5 billion Euro toward its liability. option #5: Coverage of onsite damage The NLCA would require NPP operators to have commercial insurance coverage for onsite damage, including loss of electricity production. In the USA, the NRC requires insurance coverage for onsite damage other than loss of electricity production, and additional insurance is available to cover loss of electricity production. (See Section 4, above.) option #6: Economic channeling The NLCA would require economic channeling of liability, as under the Price-Anderson Act in the USA. Economic channeling would not set ordinary tort law aside, which the NLCAs currently-proposed legal channeling would do. Under economic channeling, suppliers of nuclear technology, services, and materials would have stronger incentives to promote high levels of safety and security at NPPs. (See Section 4, above.) option #7: operator-pooled, second-tier coverage
38

The NLCA would require all Canadian NPP operators to participate in pooled, second-tier coverage via retroactive premiums, as under the Price-Anderson Act. In the USA, this approach provides about $11 billion of total coverage for damage to third parties. (See Section 4, above.) A variant of Option #7 would involve all US NPP licensees and all Canadian NPP operators participating in a common pool that provides second-tier coverage. Implementing that variant would require a treaty between Canada and the USA.

12. Conclusions and Recommendations

Major conclusions of this report include:
C1. Future operation of NPPs in Canada should be consistent with a policy framework that is appropriate for the 21st century. C2. The NLCA should be assessed against a framework as described in conclusion C1. Five criteria are developed in this report to support such an assessment. C3. The NLCA would not meet any of the five criteria described in conclusion C2. The NLCAs failure to meet these criteria would arise from a variety of significant deficiencies. C4. The NLCA would be incompatible with existing Canadian law, and may be incompatible with Canadas international commitments. C5. Options are available to improve the NLCA. Some options are outlined in this report.

Based on the preceding conclusions and the body of this report, the following actions regarding the NLCA are recommended:
R1. The NLCA should not be enacted in its present form. R2. The Canadian government should prepare a comprehensive assessment of the risk posed by NPP operation in Canada, the opportunities for reducing that risk, and the accompanying risk costs and risk-reduction costs. R3. The Canadian government should prepare a legal analysis of the compatibility of the NLCA with Canadas existing laws and international commitments. R4. The Canadian government should prepare a systematic study of options for improving the NLCA, to include the options outlined in this report. R5. Parliament should sponsor a public event at which representatives of a wide range of stakeholders would discusses the analyses called for in recommendations R2 through R4, and would present their own analyses. R6. A revised NLCA should be prepared, informed by the public event called for in recommendation R5.

39

13. Bibliography
(Alvarez et al, 2003)
Robert Alvarez and seven other authors, Reducing the Hazards from Stored Spent Power-Reactor Fuel in the United States, Science and Global Security, Volume 11, 2003, pp 1-51.

(CEAA et al, 2008)

Canadian Environmental Assessment Agency and other federal agencies, Guidelines for the Preparation of the Environmental Impact Statement for Ontario Power Generations Darlington B New Nuclear Power Plant Project, Draft (Ottawa: CEAA, September 2008).

(CNSC, 2008a)
Canadian Nuclear Safety Commission, Design of New Nuclear Power Plants, RD-337 (Ottawa: CNSC, November 2008).

(Ameye, 2009)
Evelyne Ameye, Channelling of Nuclear Third Party Liability towards the Operator: is it Sustainable in a Developing Nuclear World or is there a Need for Liability of Nuclear Architects-Engineers?, paper for the Nuclear Inter Jura Congress, Toronto, 5-9 October 2009.

(CNSC, 2008b)
Canadian Nuclear Safety Commission, Written submissions regarding the request by Greenpeace Canada on the release of the Pickering B Probabilistic Risk Assessment Document, CMD 08-H4.29, 11 April 2008.

(Anastas and Zimmerman, 2003)

Paul T. Anastas and Julie B. Zimmerman, Design Through the 12 Principles of Green Engineering, Environmental Science and Technology, 1 March 2003, pp 95-101.

(CNSC, 2008c)
Canadian Nuclear Safety Commission, Life Extension of Nuclear Power Plants, RD-360 (Ottawa: CNSC, February 2008).

(Ansolabehere et al, 2003)

Stephen Ansolabehere and nine other authors, The Future of Nuclear Power: An Interdisciplinary MIT Study (Cambridge, Massachusetts: Massachusetts Institute of Technology, 2003).

(CNSC, 2008d)
Canadian Nuclear Safety Commission, 2007-08 Annual Report (Ottawa: CNSC, 2008).

(Asmis and Khosla, 2007)

G. J. Kurt Asmis and Jagjit Khosla, Report to the CNSC, Guidance on Meeting Regulatory Expectations for the Engineering Safety Aspects of Protection from Malevolent Events, RSP-0218 (Ottawa: Asmis Consulting, 19 March 2007).

(CNSC, 2007a)
Canadian Nuclear Safety Commission, Design of New Nuclear Power Plants, RD-337, Draft (Ottawa: CNSC, October 2007).

(Beare, 2005)
John W. Beare, Review of ACR-LBD-001, Licensing Basis Document for New Nuclear Power Plants in Canada, Draft dated 2004 December (Ottawa: CNSC, file No. 34R240-2, 31 March 2005).

(CNSC, 2007b)
Canadian Nuclear Safety Commission, Regulatory Independence: Law, Practice and Perception A Report to the Canadian Nuclear Safety Commission (Ottawa: CNSC, August 2007).

(Becklumb and Dufresne, 2009)

Penny Becklumb and Robert Dufresne, Bill C-20: An Act Respecting Civil Liability and Compensation for Damage in Case of a Nuclear Incident (Ottawa: Parliamentary Information and Research Service, 21 April 2009).

(CNSC, 2004)
Canadian Nuclear Safety Commission, Regulatory Guide, Keeping Radiation Exposures and Doses As Low as Reasonably Achievable (ALARA), G-129, Revision 1 (Ottawa: CNSC, October 2004).

(Bruce Power, 2005)

Bruce Power, Environmental Assessment Study Report, Volume 1: Main Report, Environmental Assessment of Bruce A Refurbishment for Life Extension and Continued Operations Project (Mississauga: Bruce Power, December 2005).

(Cooper, 2009)
Mark Cooper, The Economics of Nuclear Reactors: Renaissance or Relapse? (Vermont: Institute for Energy and the Environment, Vermont Law School, June 2009).

(Campbell et al, 2007)

Kurt M. Campbell et al, The Age of Consequences: The Foreign Policy and National Security Implications of Global Climate Change (Washington, DC: Center for Strategic and International Studies, Center for a New American Security, November 2007).

(Cousins and Reichmuth, 2007)

Tom Cousins and Barbara Reichmuth, Preliminary Analysis of the Economic Impact of Selected RDD Events in Canada, presentation at the CRTI Summer Symposium 2007, Gatineau, Quebec, 11-14 June 2007. CRTI is the CBRNE Research and Technology Initiative, a program of Defence Research and Development Canada. The conference proceedings (available from CRTI) list the presentation as CRTI 05-0043RD, titled Economic Impact of Radiological Terrorist Events.

40

(CBC News, 2008)

CBC News, Nuclear safety watchdog head fired for lack of leadership: minister, 16 January 2008, accessed at <http://www.cbc.ca/canada/story/2008/01/16/keenfiring.html> on 21 May 2008.

(DOE, 1987)
US Department of Energy, Health and Environmental Consequences of the Chernobyl Nuclear Power Plant Accident, DOE/ER-0332 (Washington, DC: DOE, June 1987).

(Henault, 2009)
Jacques Henault, Overview of Canadian Nuclear Civil Liability Legislation, paper for the Nuclear Inter Jura Congress, Toronto, 5-9 October 2009.

(Faure and Borre, 2008)

Michael G. Faure and Tom Vanden Borre, Compensating Nuclear Damage: A Comparative Economic Analysis of the U.S. and International Liability Schemes, William & Mary Environmental Law & Policy Review, Volume 33, 2008, pp 219-287.

(Heyes and Heyes, 2000)

Anthony Heyes and Catherine Heyes, An empirical analysis of the Nuclear Liability Act (1970) in Canada, Resource and Energy Economics, Volume 22, 2000, pp 91-101.

(Hirsch et al, 1989)

H. Hirsch and three other authors, IAEA Safety Targets and Probabilistic Risk Assessment (Hannover, Germany: Gesellschaft fur Okologische Forschung und Beratung, August 1989).

(Faure and Fiore, 2008)

Michael G. Faure and Karine Fiore, The civil liability of European nuclear operators: which coverage for the new 2004 Protocols? Evidence from France, International Environmental Agreements, Volume 8, 2008, pp 227-248.

(Hultman et al, 2007)

Nathan E. Hultman and two other authors, What History Can Teach Us about the Future Costs of US Nuclear Power, Environmental Science & Technology, 1 April 2007, pp 2088-2093, plus Supporting Information.

(Ferrara and Mesquita, 2003)

Stacey Ferrara and Jennifer Mesquita, Supreme Court of Canada Upholds a Ministers Application of the PolluterPays Principle, Gowling Lafleur Henderson LLP, October 2003.

(IAEA, 2006)
International Atomic Energy Agency, Nuclear Power and Sustainable Development (Vienna: IAEA, April 2006).

(Frappier, 2007)
Gerry Frappier, Enhanced Security in the Nuclear Industry, Frontline Security, Spring 2007, accessed from the CNSC website, <http://www.nuclearsafety.gc.ca/eng/ newsroom/articles/2007-04-18_nuclear_safety.cfm>, on 28 May 2008.

(IAEA, 2002)
International Atomic Energy Agency, Optimization of Radiation Protection in the Control of Occupational Exposure, IAEA Safety Reports Series No. 21 (Vienna: IAEA, 2002).

(GAO, 2007)
US Government Accountability Office, Crude Oil: Uncertainty about Future Oil Supply Makes It Important to Develop a Strategy for Addressing a Peak and Decline in Oil Production, GAO-07-283 (Washington, DC: GAO, February 2007).

(IPCC, 2007)
Intergovernmental Panel on Climate Change, Fourth Assessment Report, Climate Change 2007: Synthesis Report (Geneva: IPCC, 2007).

(Gibson, 2000)
Robert B. Gibson, Favouring the Higher Test: Contribution to Sustainability as the Central Criterion for Reviews and Decisions under the Canadian Environmental Assessment Act, Journal of Environmental Law and Practice, Volume 10, 2000, pp 39-56.

(IRSS, 1992)
Institute for Resource and Security Studies, Risk Implications of Potential New Nuclear Plants in Ontario (Toronto: Coalition of Environmental Groups for a Sustainable Energy Future, November 1992).

(Government of Canada, 2007)

Government of Canada, Cabinet Directive on Streamlining Regulation, came into effect 1 April 2007, accessed at <http://www.regulation.gc.ca> on 17 December 2007.

(ISR, 2003)
International Safety Research, Review of the Coverage Limit in the Canadian Nuclear Liability Act, Task 5, Final Report, presented to CNSC, 17 September 2003.

(Justice Department, 2007)

Canada Department of Justice, Canadian Environmental Assessment Act (1992, c. 37), current to 1 December 2007, accessed at <http://laws.justice.gc.ca> on 5 January 2008.

(Greenpeace International, 2007)

Greenpeace International, Nuclear Power is Not the Answer to Climate Change (Amsterdam: Greenpeace International, 2007).

(Hannerz, 1983)
K. Hannerz, Towards Intrinsically Safe Light Water Reactors (Oak Ridge, Tennessee: Institute for Energy Analysis, February 1983).

(Keystone Center, 2007)

Keystone Center, Nuclear Power Joint Fact-Finding (Keystone, Colorado: Keystone Center, June 2007).
41

(Komanoff and Roelofs, 1992)

Charles Komanoff and Cora Roelofs, Fiscal Fission: The Economic Failure of Nuclear Power (Washington, DC: Greenpeace USA, December 1992).

(NRC, 1994)
US Nuclear Regulatory Commission, 10 CFR Part 73, RIN 3150-AE81, Protection Against Malevolent Use of Vehicles at Nuclear Power Plants, Federal Register, Volume 59, Number 146, 1 August 1994, pp 3888938900.

(Komanoff, 1981)
Charles Komanoff, Power Plant Cost Escalation: Nuclear and Coal Capital Costs, Regulation, and Economics (New York: Van Nostrand Reinhold Co., 1981).

(NRC, 1990)
US Nuclear Regulatory Commission, Severe Accident Risks: An Assessment for Five US Nuclear Power Plants, NUREG-1150 (Washington, DC: Nuclear Regulatory Commission, December 1990).

(Lunn, 2007)
Gary Lunn, Minister of Natural Resources Canada, letter of 8 June 2007 to Shawn-Patrick Stensil, Greenpeace Canada (GPC), responding to Petition No. 193 submitted by GPC to the Auditor General of Canada on 23 January 2007.

(NRC, 1987)
US Nuclear Regulatory Commission, Report on the Accident at the Chernobyl Nuclear Power Station, NUREG-1250 (Washington, DC: Nuclear Regulatory Commission, January 1987).

(MacKenzie and MacLachlan, 2009)

Rennie MacKenzie and Ann MacLachlan, Ex-CNSC president looks at options after losing challenge to dismissal, Nucleonics Week, Volume 50, Number 16, 23 April 2009.

(NRC, 1975)
US Nuclear Regulatory Commission, Reactor Safety Study, WASH-1400 (NUREG-75/014) (Washington, DC: Nuclear Regulatory Commission, October 1975).

(Makhijani, 2007)
Arjun Makhijani, Carbon-Free and Nuclear-Free: A Roadmap for US Energy Policy (Muskegon, Michigan: RDR Books, 2007).

(Okrent, 1981)
David Okrent, Nuclear Reactor Safety: On the History of the Regulatory Process (Madison, Wisconsin: University of Wisconsin Press, 1981).

(MEA, 2005)
Millennium Ecosystem Assessment, Ecosystems and Human Well-Being: Synthesis (Washington, DC: Island Press, 2005).

(Ontario Hydro, 1987) Ontario Hydro, Darlington NGS Probabilistic Safety Evaluation: Summary Report (Toronto: Ontario Hydro, December 1987). (OPG, 2008) Ontario Power Generation, Pickering B Risk Assessment Summary Report (Toronto: Ontario Power Generation, November 2008). (OPG, 1999) Ontario Power Generation, Bruce NGS B Risk Assessment Summary Report (Toronto: Ontario Power Generation, November 1999). (Overbye et al, 2002) Thomas J. Overbye and three other authors, National Energy Supergrid Workshop Report (Urbana-Champaign, Illinois: University of Illinois, November 2002). (Pagnamenta, 2009) Robin Pagnamenta, UK taxpayer may be forced to take on nuclear risk after insurers refuse to offer cover, TimesOnline, 9 September 2009, accessed at <http://business.timesonline.co.uk/tol/business/ industry_sectors/natural_resources/article6826650. ece>.

(Morrison, 2000)
Robert Morrison, Nuclear Energy and Sustainable Development, 15 November 2000, accessed at: <http:// www.cna.ca>.

(Murphy et al, 2009)

Dermot Murphy and four other authors, The Role of Insurance in Canadas Nuclear Civil Liability Regime: Possible Solutions for Insuring the Expanded Scope of Damage under Canadas New Regime, paper for the Nuclear Inter Jura Congress, Toronto, 5-9 October 2009.

(National Research Council, 2006)

National Research Council Committee on the Safety and Security of Commercial Spent Nuclear Fuel Storage (a committee of the Councils Board on Radioactive Waste Management), Safety and Security of Commercial Spent Nuclear Fuel Storage: Public Report (Washington, DC: National Academies Press, 2006).

(NEA, 2000)
Nuclear Energy Agency, Nuclear Energy in a Sustainable Development Perspective (Paris: OECD, 2000).
42

(NERAC/GIF, 2002)
Nuclear Energy Research Advisory Committee (US Department of Energy) and Generation IV International Forum, A Technology Roadmap for Generation IV Nuclear Energy Systems, GIF-002-00 (Washington, DC: DOE, December 2002).

(Parsons, 2006) Robert Parsons, Chornobyl 20 Years After: A Nuclear Nightmare Becomes a Political Disaster, Radio Free Europe / Radio Liberty, 20 April 2006, accessed at <http://www.rferl.org> on 10 January 2008. (Raskin et al, 2002) Paul Raskin et al, Great Transition: The Promise and Lure of the Times Ahead (Boston, Massachusetts: Stockholm Environment Institute, 2002). (Roman et al, 2009) Andrew Roman and two other authors, Canada and International Nuclear Liability, paper for the Nuclear Inter Jura Congress, Toronto, 5-9 October 2009. (Romm, 2008) Joe Romm, The Self-Limiting Future of Nuclear Power (Washington, DC: Center for American Progress Action Fund, June 2008). (Schaubel, 2008) T. E. Schaubel, CNSC Staff, letter to D. Patrick McNeil, Ontario Power Generation, CNSC Staff Review of Pickering NGS-B Integrated Safety Review Safety Analysis Safety Factors Report, 7 April 2008. (Schneider et al, 2009) Mycle Schneider and three other authors, The World Nuclear Industry Status Report 2009 (Paris: August 2009). This report was commissioned by the German Federal Ministry of Environment, Nature Conservation, and Reactor Safety. (SENES, 2007) SENES Consultants Limited, Credible Malfunction and Accident Scenarios Technical Support Document (Final), Refurbishment and Continued Operation of Pickering B Nuclear Generating Station Environmental Assessment (Richmond Hill, Ontario: SENES Consultants, December 2007). (Statistics Canada, 2009) Statistics Canada, Electric Power Generation, Transmission and Distribution, 2007 (Ottawa: Statistics Canada, April 2009). (Talisman, 2008) Talisman International LLC, Atomic Energy of Canada Limited, National Research Universal Reactor Safety System Upgrades and the Canadian Nuclear Safety Commissions Licensing and Oversight Process: A Lessons Learned Report, June 2008.

(Thompson, 2008a) Gordon R. Thompson, Scope of the EIS for New Nuclear Power Plants at the Darlington Site in Ontario: Accidents, Malfunctions and The Precautionary Approach (Cambridge, Massachusetts: Institute for Resource and Security Studies, November 2008). (Thompson, 2008b) Gordon R. Thompson, Risks of Operating CANDU 6 Nuclear Power Plants: Gentilly Unit 2 Refurbishment and its Global Implications (Cambridge, Massachusetts: Institute for Resource and Security Studies, October 2008). (Thompson, 2007) Gordon R. Thompson, Risk-Related Impacts from Continued Operation of the Indian Point Nuclear Power Plants (Cambridge, Massachusetts: Institute for Resource and Security Studies, 28 November 2007). (Thompson, 2000) Gordon Thompson, A Review of the Accident Risk Posed by the Pickering A Nuclear Generating Station (Cambridge, Massachusetts: Institute for Resource and Security Studies, August 2000). (Thompson, 1998) Gordon Thompson, Science, democracy and safety: why public accountability matters, in: F. Barker (editor), Management of Radioactive Wastes: Issues for local authorities (London: Thomas Telford Ltd., 1998). (Watson et al, 1972) M. B. Watson and four other authors, Underground Nuclear Power Plant Siting (Pasadena, California: Environmental Quality Laboratory, California Institute of Technology, September 1972). (WNA, 2009) World Nuclear Association, Civil Liability for Nuclear Damage, August 2009, accessed at <http://www.world-nuclear.org/info/inf67.html>.

43

Table 1-1
classification of potential Accidents and malfunctions at a nuclear power plant

tyPE of ACCiDENt oR MALfuNCtioN MoDE of iMPACt of ACCiDENt oR MALfuNCtioN Accidents initiated by internal Events Accidents initiated by External Events X Releases and Diversions initiated by intentional, Malevolent Acts X

unplanned release of radioactive material from the reactor core unplanned release of radioactive material from spent fuel, during storage or transfer to/from storage unplanned release of radioactive or hazardous chemical material from another part of the plant diversion of fissile or radioactive material for illicit use

Not applicable

Not applicable

noteS: (a) The symbol X indicates that there is a potential for accidents and malfunctions in the designated category. (b) For further background, see: Thompson, 2008a.

44

Table 3-1
the twelve Principles of Green Engineering, According to Anastas and Zimmerman
Designers need to strive to ensure that all material and energy inputs and outputs are as inherently non-hazardous as possible. It is better to prevent waste than to treat or clean up waste after it is formed. Separation and purification operations should be designed to minimize energy consumption and materials use. Products, processes, and systems should be designed to maximize mass, energy, space, and time efficiency. Products, processes, and systems should be output pulled rather than input pushed through the use of energy and materials. Embedded entropy and complexity must be viewed as an investment when making design choices on recycle, reuse, or beneficial disposition. Targeted durability, not immortality, should be a design goal. Design for unnecessary capacity or capability (e.g., one size fits all solutions) should be considered a design flaw. Material diversity in multi-component products should be minimized to promote disassembly and value retention. Design of products, processes, and systems must include integration and interconnectivity with available energy and materials flows. Products, processes, and systems should be designed for performance in a commercial afterlife. Material and energy inputs should be renewable rather than depleting.

principle 1 principle 2 principle 3 principle 4 principle 5 principle 6 principle 7 principle 8 principle 9 principle 10 principle 11 principle 12

SOURCE: Anastas and Zimmerman, 2003.

45

Table 5-1
Some potential modes and Instruments of Attack on a nuclear power plant

AttACk MoDE/iNstRuMENt

Characteristics
Could involve heavy weapons and sophisticated tactics Successful attack would require substantial planning and resources Readily obtainable Highly destructive if detonated at target Readily obtainable Highly destructive at point of impact More difficult to obtain than pre-9/11 Can destroy larger, softer targets Readily obtainable Can destroy smaller, harder targets Difficult to obtain Assured destruction if detonated at target

Present Defenses at us Plants

Alarms, fences and lightlyarmed guards, with offsite backup Vehicle barriers at entry points to Protected Area None if missile launched from offsite None

commando-style attack

land-vehicle bomb Small guided missile (anti-tank, etc.) commercial aircraft

explosive-laden smaller aircraft

None

10-kilotonne nuclear weapon

None

noteS: (a) This table is adapted from: Thompson, 2007, Table 7-4. Further citations are provided in that table and its supporting narrative. (b) Defenses at Canadian plants are no more robust than at US plants. See: Frappier, 2007.

46

Table 5-2
the Shaped charge as a potential Instrument of Attack

CAtEGoRy of iNfoRMAtioN

selected information in Category

Shaped charges have many civilian and military applications, and have been used for decades

general information

Applications include human-carried demolition charges or warheads for anti-tank missiles Construction and use does not require assistance from a government or access to classified information The German MISTEL, designed to be carried in the nose of an un-manned bomber aircraft, is the largest known shaped charge Japan used a smaller version of this device, the SAKURA bomb, for kamikaze attacks against US warships Developed by a US government laboratory for mounting in the nose of a cruise missile Described in detail in an unclassified, published report (citation is voluntarily withheld here)

use in world war II

A large, contemporary device

Purpose is to penetrate large thicknesses of rock or concrete as the first stage of a "tandem" warhead Configuration is a cylinder with a diameter of 71 cm and a length of 72 cm When tested in November 2002, created a hole of 25 cm diameter in tuff rock to a depth of 5.9 m Device has a mass of 410 kg; would be within the payload capacity of many general-aviation aircraft A Beechcraft King Air 90 general-aviation aircraft will carry a payload of up to 990 kg at a speed of up to 460 km/hr A used King Air 90 can be purchased in the US for $0.4-1.0 million

A potential delivery vehicle

SOURCE: Thompson, 2007, Table 7-6. Further citations are provided in that table and its supporting narrative.

47

Table 5-3
Safety goals for a new nuclear power plant, as Specified in cnSc Regulatory document Rd-337
(ATTENTION: The table shows the safety goals in the Draft version of RD-337; the notes show how the goals were weakened in the Final version.)

sAfEty GoALs tyPE of outCoME

sum of frequencies of all event sequences that can lead to this outcome
should be less than shall not exceed 1 per 100,000 plant-years 1 per 1 million plant-years

Small Release to the environment (more than 1 pbq of Iodine-131) large Release to the environment (more than 0.1 pbq of cesium-137) core damage (significant core degradation)

1 per 1 million plant-years 1 per 10 million plant-years

1 per 1 million plant-years

1 per 100,000 plant-years

noteS: (a) The table as shown describes the safety goals set forth by the CNSC in October 2007 in the document: Design of New Nuclear Power Plants, RD-337, Draft. See: CNSC, 2007a, page 5. (b) In November 2008, the CNSC Staff published a Final version of RD-337. See: CNSC, 2008a. Section 4.2.2 of the Final version sets forth safety goals that exhibit the following changes from the table above. First, the numerical goals in the should be less than category are abandoned. Second, the numerical goals in the shall not exceed category are retained, but with different language. RD-337 now states that the sum of frequencies of all event sequences that can lead to a specified outcome is less than a numerical value. Both changes represent a significant weakening of the safety goals.

48

Table 6-1
Selected options to Reduce the Risk of a Spent-fuel-pool fire at a nuclear power plant that employs high-density pool Storage

oPtioN

Passive or Active?

Does option Address fire scenarios Arising from: Malevolent Acts? other Events?

CoMMENts

Re-equip pool with low-density, open-frame racks

Passive

Yes

Yes

Would substantially reduce pool inventory of radioactive material Would prevent auto-ignition of fuel in almost all cases Spray system must be highly robust

Install emergency water sprays above pool

Active

Yes

Yes

Spraying water on overheated fuel could feed Zr-steam reaction Could delay or prevent auto-ignition in some cases

mix hotter (younger) and colder (older) fuel in pool

Passive

Yes

Yes

Would be ineffective if debris or residual water block air flow Could promote fire propagation to older fuel

minimize movement of spent-fuel cask over pool

Active

(Most cases)

No

Yes

Could conflict with adoption of lowdensity, open-frame racks

deploy air-defense system (e.g., Sentinel and phalanx) at plant

Active

Yes

No

Implementation would require presence of military personnel at plant

develop enhanced onsite capability for damage control

Active

Yes

Yes

Would require new equipment, staff and training Personnel must function in extreme environments

SouRce: Thompson, 2007, Table 9-1. Further citations are provided in that tables supporting narrative.

49

Table 7-1
Radioactive Releases and offsite Impacts for the 1986 chernobyl Accident and Some potential Accidents at nuclear power plants: Selected data
fraction of Cesium-137 inventory Released to Atmosphere
Reactor core: 40% (estimated from known release) Spent fuel: 0% (known)

ACCiDENt CAsE

inventory of Cesium-137 Available for Release

Probability of Release to Atmosphere

Lifetime Population Dose (million person-sv)

chernobyl unit 4 in 1986 (capacity = 1.0 gwe)

Reactor core: 223 PBq Spent-fuel pool: ?

Reactor core: 1.0 (known) Spent-fuel pool: 0 (known)

1.2 (estimated from known release)

generation II plant (generic)

Varies by plant type & mode of spent-fuel storage

Reactor core: 1/13,400 = 7.5E-05 per RY Spent fuel: ? Reactor core: 1.0E-04 per RY (estimated) Spent-fuel pool: 0 (assumed) Reactor core: 7.4E-05 per RY (estimated) Spent-fuel pool: 2.0E-06 per RY (est.) Reactor core: 1.0E-06 per RY (CNSC goal) Spent fuel: 1.0E-06 per RY (CNSC goal)

Varies by plant type and scenario

Varies by plant type, scenario and site

darlington cAndu plant (capacity = 0.88 gwe)

Reactor core: 67 PBq Spent-fuel pool: ?

Reactor core: 50% (est.) from each of two reactors Spent fuel: 0% (assumed) Reactor core: 23% (est.) Spent fuel: 50% (est.)

2.7 (estimated)

Indian point unit 2 pwR plant (capacity = 1.08 gwe) Generation III plant at Darlington site
(amounts in this row are normalized to a plant capacity of 1.0 GWe)

Reactor core: 420 PBq Spent-fuel pool: 2,500 PBq

Reactor core: 390 PBq Spent-fuel pool: 2,300 PBq

Reactor core: 50% (est.) Spent fuel: 50% (est.)

Reactor core: 7.9 Spent fuel: 46

(both amounts are extrapolated from the Darlington CANDU case)

NOTES: (a) Actual releases would include isotopes in addition to Cesium-137. (b) RY = reactor-year (c) Population dose is also known as collective dose commitment. Lifetime dose is typically calculated for a 50-year period. (d) Data in the first row are from: DOE, 1987; NRC, 1987; IRSS, 1992, Volume 2, Annex III. Lifetime population dose is from: DOE, 1987, Table 5.16. (e) In the second row, the probability of a substantial reactor-core release is determined by the occurrence of one such event (at Chernobyl Unit 4) during the 13,400 RY of worldwide commercial reactor operation accrued through 2008. (f) Data in the third row are from: IRSS, 1992, Volume 2. The probability of a reactor-core release is an IRSS estimate for internal + external initiating events, excluding malevolent acts. The estimated lifetime population dose is a weighted average over the set of the most frequent weather conditions at Darlington, where that set accounts for 20% of the frequency of all weather conditions at the site. Dose was calculated by the MACCS code up to a distance of 1,000 km, assuming no relocation of populations. The estimated release of Cesium-137 is 67 PBq for two reactors, which is equivalent to a release of 67/(2 x 0.88) = 38 PBq per GWe. (g) Data in the fourth row are from Entergy and the author, in: Thompson, 2007. Entergy estimates a core-damage probability, accounting for internal + external initiating events + uncertainty, excluding malevolent acts, of 1.4E-04 per RY. Entergys estimate of the conditional probability of an Early High release is adjusted here to account for containment bypass during High/Dry core-damage sequences (see: Thompson, 2007, Table 5-3). The estimated probability of a release from the spent-fuel pool is taken from the NRC study NUREG-1353 (see: Thompson, 2007, Table 6-2). (h) In the fifth row, inventories of Cesium-137 are adjusted from the Indian Point Unit 2 inventories in proportion to plant capacity. Release probabilities are set to the CNSC safety goal for a Large Release (see Table 5-3). The estimated release of Cesium-137 is 0.5 x 390 = 195 PBq for the reactor core and 0.5 x 2,300 = 1,150 PBq for the spent-fuel pool. Lifetime population dose is extrapolated from the Darlington CANDU case by assuming that dose is proportional to the release of Cesium-137, yielding an estimated dose of (195/67) x 2.7 = 7.9 million person-Sv for the reactor release and (1,150/67) x 2.7 = 46 million person-Sv for the spent-fuel release.

50

Table 7-2
licensee estimates of frequency and population dose for potential Accidental Atmospheric Releases of Radioactive material from Some cAndu Stations in ontario

Ex-PLANt RELEAsE CAtEGoRy

Bruce A
freq. (per Ry) 1.5E-07 3.6E-07 6.4E-08 8.6E-08 2.2E-09 (EPRC 5) 9.1E-07 2.6E-05 7.0E-10 2.8E-05 3.0E-05 Pop. Dose (p-sv) 32,700 36,100 10,500 8,700 22,700 (EPRC 5) 2,750 770 40 240 7

Bruce B
freq. (per Ry) 2.8E-09 9.1E-08 2.7E-08 6.9E-08 2.0E-10 (EPRC 5) 1.8E-07 3.5E-06 4.7E-10 5.9E-06 5.8E-05 Pop. Dose (p-sv) 32,900 35,400 10,400 8,730 22,700 (EPRC 5) 2,800 760 40 240 7

Pickering B
freq. (per Ry) 1.1E-10 1.0E-11 1.0E-11 2.4E-10 7.1E-07 2.1E-08 1.0E-11 1.0E-11 1.3E-06 1.0E-06 ? Pop. Dose (p-sv) 38,200 16,000 19,800 7,920 900 230 3,600 22,000 11 40 ?

epRc 1 epRc 2 epRc 3 epRc 4 epRc 5A epRc 5b epRc 6 epRc 7 epRc 8 epRc 9 epRc 10

NOTES: (a) Bruce A estimates are from: Bruce Power, 2005, Table 3.5.11-5. Quantities shown are mean estimates. Population dose is to 200 km radius. It appears that frequency estimates cover internal initiating events. (b) Bruce B estimates are from: OPG, 1999, Tables 2-11 and 2-12. Quantities shown are mean estimates. Population dose is to 200 km radius. Frequency estimates cover internal initiating events and loss of offsite power. (c) Pickering B estimates are from: OPG, 2008, Table 5; and SENES, 2007, Table 5.3-3. Quantities shown are mean estimates. Population dose is to 100 km radius. Radiation risk coefficient is 0.05 delayed fatalities per person-Sv. Frequency estimates cover internal initiating events, loss of offsite power, and some screenhouse failures. (d) EPRC definitions differ across the studies cited. (e) RY = reactor-year.

51

Table 7-3
ontario hydro estimate of the Risk of onsite economic Impacts from fuel-damage events at the darlington nuclear power plants (existing cAndu plants)

fuEL DAMAGE CAtEGoRy

Est. Mean Probability (uncertainty factor) 3.8E-06 per RY (UF = 6) 2.0E-06 per RY (UF = 6) 8.0E-05 per RY (UF = 6) 4.7E-04 per RY (UF = 4) 3.0E-05 per RY (UF = 10) 1.0E-04 per RY (UF = 10) 2.0E-03 per RY (UF = 10) 3.0E-03 per RY (UF = 5) 2.0E-03 per RY (UF = 10) 2.3E-02 per RY (UF = 3) totAL Risk

Est. onsite Economic impacts (million 2009 Can $) ? 6,400 to 11,500 5,800 to 10,200 3,400 to 5,900 3,400 to 6,200 2,700 to 5,200 1,900 to 3,700 790 to 2,500 120 to 600 390 to 700

Risk of onsite Economic impacts (million 2009 Can $ per Ry) using Mean Estimate of fDC Probability ? 0.013 to 0.023 0.46 to 0.82 1.60 to 2.80 0.10 to 0.19 0.27 to 0.52 3.80 to 7.40 2.40 to 7.50 0.24 to 1.20 8.97 to 16.10 17.9 to 36.6 using 95th Percentile Estimate of fDC Probability ? 0.077 to 0.14 2.80 to 4.90 6.40 to 11.10 1.02 to 1.90 2.70 to 5.20 38.0 to 74.0 11.90 to 37.5 2.40 to 12.0 26.9 to 48.3 92.2 to 195.0

fdc 0 fdc 1 fdc 2 fdc 3 fdc 4 fdc 5 fdc 6 fdc 7 fdc 8 fdc 9

NOTES: (a) Estimates are from the Darlington Probabilistic Safety Evaluation (DPSE). See: Ontario Hydro, 1987, Tables 5-2, 5-8 and 5-9. For additional data from the full version of DPSE, see: IRSS, 1992, Volume 2, Annex IV. (b) DPSE provided cost estimates in 1985 Can $. These are adjusted here to 1991 Can $ by a multiplier of 1.25 (see: IRSS, 1992, Volume 2, Annex IV), and from 1991 Can $ to 2009 Can $ by a multiplier of 1.36 (CPI inflator from Bank of Canada). The combined multiplier is 1.70. (c) DPSE did not estimate the risk of onsite economic impacts for FDC 0. (d) These estimates are limited to fuel damage in a reactor core or a fueling machine, caused by accidents initiated by internal events. (e) Replacement power is the dominant component of the estimated onsite economic impacts. The other component considered by DPSE is the cost of decontamination and repair. (f) The range of estimated onsite economic impacts is from a best estimate (lower bound) to a probable maximum (upper bound). (g) The Darlington station has four CANDU units (plants) that share many safety and support systems (e.g., fueling duct and vacuum building), which means that a fuel-damage event at one unit could readily lead to adverse impacts on the other units. DPSE determined that accidents in categories FDC 1 through FDC 9 would lead to forced outage of all four units. For example, given the occurrence of an FDC 1 accident, the estimated duration of the forced outage would be 45-72 months for all four units, and an additional 65-126 months for the unit that suffered fuel damage. (h) The uncertainty factor (UF) in the second column is DPSEs estimate of the ratio of the 95th percentile value to the mean value.

52

Table 9-1
estimation of cost to transfer Spent fuel from a pwR Spent-fuel pool to dry Storage After 5 years of Storage in the pool

EstiMAtioN stEP Average period of use of a fuel assembly in the reactor core period of storage of a spent-fuel assembly in the spent-fuel pool, prior to transfer to dry storage point in plant history when transfer of spent fuel to dry storage begins

EstiMAtE 5 years

5 years

11th year of plant operation

Average annual transfer of spent fuel from pool to dry storage capital cost of transferring spent fuel from pool to dry storage (given a dry-storage cost of $200 per kgu, and a mass of 450 kgu per fuel assembly) capital cost of transferring spent fuel from pool to dry storage (given a plant capacity of 1.08 gwe, and a capacity factor of 0.9)

36 fuel assemblies

$3.2 million per year 0.04 cent per kWh of nuclear generation

NOTES: (a) This calculation employs data that apply to the Indian Point 2 nuclear power plant in New York state. Comparable data would apply to a new PWR plant in Canada. (b) Data in this table are from Tables 2-1 and 9-2 of: Thompson, 2007. (c) The capital cost begins in the 11th year of plant operation, and continues while the plant operates. (d) The cost can be regarded as being in 2009 Can $.

53

Table 9-2
Risk Indicators for potential Accidental Atmospheric Releases of Radioactive material from the bruce A Station

Ex-PLANt RELEAsE CAtEGoRy EPRC 1 EPRC 2 EPRC 3 EPRC 4 EPRC 5A EPRC 5B EPRC 6 EPRC 7 EPRC 8 EPRC 9 EPRC 10

frequency (per Ry)

1.5E-07 3.6E-07 6.4E-08 8.6E-08 2.2E-09 (EPRC 5) 9.1E-07 2.6E-05 7.0E-10 2.8E-05 3.0E-05

Population Dose (person-sv)

32,700 36,100 10,500 8,700 22,700 (EPRC 5) 2,750 770 40 240 7

Monetized health impact ($ million in 2009 Can $)

44,500 49,100 14,300 11,800 30,900 (EPRC 5) 3,740 1,050 54 330 10

NOTES: (a) Frequencies are mean estimates by the licensee, as shown in Table 7-2. (b) Population doses are mean estimates by the licensee, as shown in Table 7-2. (c) Population dose is converted here to monetized health impact using a coefficient of $1.36 million per person-Sv (2009 Can $).

54

Table 9-3
Risk costs of offsite Impacts of Accidental or malevolent Releases of Radioactive material from nuclear power plants in canada: Selected cases

RELEAsE CAsE

Lifetime Population Dose (million person-sv) Reactor release: 1.5

Probability of Release

Risk Costs of Release (million 2009 Can $ per Ry) Reactor release: 210

Risk Costs of Release (2009 Can cent per kWh) Reactor release: 2.4/C

CAsE 1A: Accidental release at existing cAndu reactor CAsE 1B-L: Accidental release at new gen III reactor or spent-fuel pool (lower probability)

1.0E-04 per RY

Reactor release: 7.9 Spent-fuel release: 46

Reactor core: 1.0E-06 per RY (CNSC goal) Spent fuel: 1.0E-06 per RY (CNSC goal) Reactor core: 1.0E-05 per RY (10 x CNSC goal) Spent fuel: 1.0E05 per RY (10 x CNSC goal) MRP per RY

Reactor release: 11 Spent-fuel release: 63

Reactor release: 0.13/C Spent-fuel release: 0.72/C

CAsE 1B-h: Accidental release at new gen III reactor or spent-fuel pool (higher probability)

Reactor release: 7.9 Spent-fuel release: 46

Reactor release: 110 Spent-fuel release: 630

Reactor release: 1.3/C Spent-fuel release: 7.2/C

CAsE 2A: malevolent release at existing cAndu reactor CAsE 2B: malevolent release at new gen III reactor or spent-fuel pool

Reactor release: 1.5

Reactor release: MRP x 2.1E+06

Reactor release: (MRP x 2.4E+04)/C

Reactor release: 7.9 Spent-fuel release: 46

Reactor core: MRP per RY Spent fuel: MRP per RY

Reactor release: MRP x 1.1E+07 Spent-fuel release: 6.3E+07

Reactor release: (MRP x 1.3E+05)/C Spent-fuel release: (MRP x 7.2E+05)/C

NOTES: (a) This table is developed from the data shown in Table 7-1. (b) Population dose (in person-Sv) and risk costs (in $ per RY) are shown here for a 1 GWe-capacity plant, and can be scaled linearly to other capacities. (c) C = average annual capacity factor of a plant. (d) Malevolent release probability (MRP) = probability (per RY) that a malevolent act will yield a large atmospheric release. (e) In this table, lifetime population dose is assigned a monetary value of 1992 Can $1 million per person-SV. That value is converted to 2009 Can $ using a CPI inflator of 1.36, from Bank of Canada. (f) In the first and fourth rows, the release contains 38 PBq of Cesium-137. Population dose is scaled linearly from the Darlington CANDU case. (g) In the second, third and fifth rows, the reactor release contains 195 PBq of Cesium-137, and the spent-fuel release contains 1,150 PBq of Cesium-137. Population dose is scaled linearly from the Darlington CANDU case. 55

Table 9-4
Range of Risk costs of offsite Impacts of Accidental and malevolent Releases of Radioactive material from nuclear power plants in canada: existing or new plants

Risk Costs of RELEAsEs (2009 CAN CENt PER kWh) Av. Capacity factor (C) = 0.8
Malevolent Release Prob. (MRP) = 1.0E-04 per Ry Malevolent Release Prob. (MRP) = 1.0E-06 per Ry

CAsEs

Av. Capacity factor (C) = 0.9

Malevolent Release Prob. (MRP) = 1.0E-04 per Ry Malevolent Release Prob. (MRP) = 1.0E-06 per Ry

cASeS 1A & 2A: Release at existing cAndu plant

Case 1A: 3.0 Case 2A: 3.0 totAL: 6.0 Case 1B-L (reactor): 0.16

Case 1A: 3.0 Case 2A: 0.03 totAL: 3.0 Case 1B-L (reactor): 0.16 Case 1B-L (pool): 0.9 Case 2B (reactor): 0.16 Case 2B (pool): 0.9 totAL: 2.1 Case 1B-H (reactor): 1.6 Case 1B-H (pool): 9.0 Case 2B (reactor): 0.16 Case 2B (pool): 0.9 totAL: 11.7

Case 1A: 2.7 Case 2A: 2.7 totAL: 5.4 Case 1B-L (reactor): 0.14 Case 1B-L (pool): 0.8 Case 2B (reactor): 14.0 Case 2B (pool): 80.0 totAL: 95.0 Case 1B-H (reactor): 1.4 Case 1B-H (pool): 8.0 Case 2B (reactor): 14.0 Case 2B (pool): 80.0 totAL: 103.0

Case 1A: 2.7 Case 2A: 0.03 totAL: 2.7 Case 1B-L (reactor): 0.14 Case 1B-L (pool): 0.8 Case 2B (reactor): 0.14 Case 2B (pool): 0.8 totAL: 1.9 Case 1B-H (reactor): 1.4 Case 1B-H (pool): 8.0 Case 2B (reactor): 0.14 Case 2B (pool): 0.8 totAL: 10.3

cASeS 1b-l & 2b: Release at new gen III plant (lower accident prob.)

Case 1B-L (pool): 0.9 Case 2B (reactor): 16.0 Case 2B (pool): 90.0 totAL: 107.0 Case 1B-H (reactor): 1.6

cASeS 1b-h & 2b: Release at new gen III plant (higher accident prob.)

Case 1B-H (pool): 9.0 Case 2B (reactor): 16.0 Case 2B (pool): 90.0 totAL: 117.0

NOTE: Amounts in this table are calculated from the formulae shown in Table 9-3.

56

Table 9-5
Selected Range of Risk costs of offsite Impacts of Accidental and malevolent Releases of Radioactive material from nuclear power plants in canada: existing or new plants, excluding Releases from Spent-fuel pools

CAsEs
(Excluding Releases from spent-fuel Pools)

Risk Costs for Accidental and Malevolent Releases, Assuming Av. Capacity factor of 0.9 (2009 Can cent per kWh) Malevolent Release Prob. (MRP) = 1.0E-04 per Ry 5.4 14.1 15.4 Malevolent Release Prob. (MRP) = 1.0E-06 per Ry 2.7 0.28 1.5

cases 1A & 2A: Release at existing cAndu plant cases 1b-l & 2b: Release at new gen III plant (lower accident prob.) cases 1b-h & 2b: Release at new gen III plant (higher accident prob.)
NOTES: (a) Amounts in this table are from Table 9-4.

(b) There are two rationales for excluding releases from spent-fuel pools when assessing risk costs. First, the potential for such releases could be greatly reduced by adopting alternative modes of storage of spent fuel. Second, the inventory of Cesium-137 in a pool would grow over time, reaching its maximum value after several decades of reactor operation.

57

Table 9-6
policy-Applicable Risk costs of offsite Impacts of Accidental or malevolent Releases of Radioactive material from nuclear power plants in canada

CAsE

Risk Costs for Accidental or Malevolent Releases (2009 Can cent per kWh) high Probability of Malevolent Release 5.4 15.4 Low Probability of Malevolent Release 2.7 1.5

existing cAndu plant new generation III plant

NOTES: (a) Amounts in this table are from Table 9-5. (b) Releases from spent-fuel pools are excluded here. (c) The average capacity factor of the plant is assumed to be 0.9. (d) High probability of malevolent release = 1 per 10,000 RY; low probability = 1 per 1 million RY. (e) Here, the probability of an accidental release from an existing CANDU reactor is 1 per 10,000 RY, and the probability of an accidental release from a new Generation III reactor is 1 per 100,000 RY.

58

Table 9-7
Risk costs of onsite Impacts of fuel-damage events at existing cAndu plants, using an ontario hydro estimate of the Risk of economic Impacts at the darlington plants

Value of indicator
iNDiCAtoR using Mean Estimate of Probabilities of fuel Damage Categories 17.9 to 36.6 (million 2009 Can $ per RY) 0.26 to 0.53 (2009 Can cent per kWh) 0.5 to 1.1 (2009 Can cent per kWh) using 95th Percentile Estimate of Probabilities of fuel Damage Categories 92.2 to 195.0 (million 2009 Can $ per RY) 1.33 to 2.81 (2009 Can cent per kWh) 2.7 to 5.6 (2009 Can cent per kWh)

Risk of onsite economic impacts Risk costs of onsite economic impacts (oh estimate for internal initiating events only) Risk costs of onsite economic impacts (internal initiating events + external events + malevolent acts)
NOTES: (a) This table is developed from data shown in Table 7-3.

(b) Ontario Hydro considered the occurrence of accidents involving Fuel Damage Categories FDC 1 through FDC 9, but not the most severe Category (FDC 0). (c) Ontario Hydro considered fuel damage in a reactor core or a fueling machine, caused by accidents initiated by internal events. (d) Values in the first row are from Table 7-3. Values in the second row are calculated from the first row. (e) Values in the third row are adjusted upward from values in the second row by a factor of 2, to account for accidents initiated by external events, and for malevolent acts. (f) Each Darlington plant has a capacity of 0.88 GWe. A capacity factor of 0.9 is assumed here.

59

Table 9-8
Risk Costs of Nuclear Generation in Canada: summary of this Reports findings

CAtEGoRy of iMPACts fRoM uNPLANNED RELEAsEs of RADioACtiVE MAtERiAL

Category of Risk Costs and the insurance Premiums that are Paid to Provide Coverage of these Costs Risk Costs (2009 Can cent per kWh)

Magnitude of Risk Costs and Insurance Premiums for an Existing CANDu Plant 2.7 to 5.4 for a New Generation iii Plant 1.5 to 15.4 As for existing CANDU plant? Smaller amount than for existing CANDU plant No explicit premium is evident

offsite Impacts Insurance Premiums Under the NLCA (2009 Can cent per kWh) Risk Costs (2009 Can cent per kWh) onsite Impacts Insurance Premiums Under the NLCA (2009 Can cent per kWh) No explicit premium is evident 0.02

2.7 to 5.6

NOTES: (a) Risk costs in the first row are from Table 9-6. (b) Risk costs in the third row are from Table 9-7, using the 95th percentile probability estimate, and considering internal events + external events + malevolent acts. (c) Insurance premiums are as discussed in Section 2 of this report.

60

Table 10-1
Selected Approaches to protecting canadian critical Infrastructure from Attack by Sub-national groups, and Some Strengths and weaknesses of these Approaches

APPRoACh

stRENGths

WEAkNEssEs Could promote growth of sub-national groups hostile to Canada, and build sympathy for these groups in foreign populations Could be costly in terms of lives, money, etc. Implementation could be slow and/or incomplete Requires ongoing international cooperation Could destroy civil liberties, leading to political, social and economic decline Could suppress a true understanding of risk Could contribute to political, social and economic decline Requires ongoing expenditure & vigilance May require military involvement

offensive military operations internationally

Could deter or prevent governments from supporting sub-national groups hostile to Canada

International police cooperation within a legal framework

Could identify and intercept potential attackers

Surveillance and control of the domestic population

Could identify and intercept potential attackers

Secrecy about design and operation of infrastructure facilities

Could prevent attackers from identifying points of vulnerability

Active defense of infrastructure facilities (by use of guards, guns, gates, etc.)

Could stop attackers before they reach the target Could allow target to survive attack without damage, thereby enhancing protective deterrence Could substitute for other protective approaches, avoiding their costs and adverse impacts Could reduce risks from accidents & natural hazards

Robust and inherently-safe design of infrastructure facilities

Could involve higher capital costs

61

Figure 5-1
Core Damage frequency for Accidents at a surry PWR Nuclear Power Plant, as Estimated in the NRC study NuREG-1150

NOTES: (a) This figure is adapted from Figure 8.7 of: NRC, 1990. (b) The bars range from the 5th percentile (lower bound) to the 95th percentile (upper bound) of the estimated core damage frequency (CDF). CDF values shown are per reactor-year (RY). (c) Two estimates are shown for the CDF from earthquakes (seismic effects). One estimate derives from seismic predictions done at Lawrence Livermore National Laboratory (Livermore), the other from predictions done at the Electric Power Research Institute (EPRI). (d) CDFs are not estimated for external initiating events other than earthquakes and fires. (e) Malevolent acts are not considered.

62

Figure 5-2
Core Damage frequency for Accidents at a Peach Bottom BWR Nuclear Power Plant, as Estimated in the NRC study NuREG-1150

NOTES: (a) This figure is adapted from Figure 8.8 of: NRC, 1990. (b) The bars range from the 5th percentile (lower bound) to the 95th percentile (upper bound) of the estimated core damage frequency (CDF). CDF values shown are per reactor-year (RY). (c) Two estimates are shown for the CDF from earthquakes (seismic effects). One estimate derives from seismic predictions done at Lawrence Livermore National Laboratory (Livermore), the other from predictions done at the Electric Power Research Institute (EPRI). (d) CDFs are not estimated for external initiating events other than earthquakes and fires. (e) Malevolent acts are not considered.

63

Figure 5-3
Conditional Probability of Containment failure following a Core-Damage Accident at a surry PWR or Peach Bottom BWR Nuclear Power Plant, as Estimated in the NRC study NuREG-1150

NotE: This figure is adapted from Figure 9.5 of: NRC, 1990. 64

APPENDIX
Designing Nuclear Power Plants to Pose a Comparatively Low Level of Risk137
The most reliable option for reducing the risk posed by a nuclear power plant would be to design the plant according to highly stringent criteria of safety and security. During the 1970s and 1980s, some plant vendors and other stakeholders sought to develop designs that could meet such criteria. One design approach was to provide a highly robust containment which might be an underground cavity to separate nuclear fuel from the environment. Another approach was to incorporate principles of inherent or intrinsic safety into the design. The two approaches could be complementary. undeRgRound SItIng In the 1970s, there were several studies on constructing NPPs underground. Those studies are exemplified by a report published in 1972 under the auspices of the California Institute of Technology (Caltech).138 The report identified a number of advantages of underground siting. Those advantages included highly-effective confinement of radioactive material in the event of a core-damage accident, isolation from falling objects such as aircraft, and protection against malevolent acts. Based on experience with underground testing of nuclear weapons, the report concluded that an appropriately designed plant would provide essentially complete containment of the radioactive material liberated from a reactor core during a core-damage event. The Caltech report described a preliminary design study for underground construction of an LWR plant with a capacity of 1,000 MWe. The minimum depth of the underground cavities containing the plant components would be 150 to 200 feet. The estimated cost penalty for underground siting would be less than 10 percent of the total plant cost. In an appendix, the Caltech report described four underground nuclear reactors that had been constructed and operated in Europe. Three of those reactors supplied steam to turbo-generators, above or below ground. The largest of those reactors and its above-ground turbo-generator made up the Chooz plant in France, which had a capacity of 270 MWe. In describing the European reactors, the report noted:139 The motivation for undergrounding the plant appears to be insurance of containment of accidentally released radioactivity and also physical protection from damage due to hostile military action. Since the 1970s, underground siting of NPPs has been considered by various groups. For example, in 2002 a workshop was held under the auspices of the University of Illinois to discuss a proposed US-wide supergrid. That grid would transmit electricity via superconducting DC cables and liquid hydrogen, which would provide cooling to the DC cables and be distributed as fuel. Much of the energy fed to the grid would be supplied by nuclear power plants, which could be constructed underground. Motives for placing those plants underground would include reduced vulnerability to attack by nature, man or weather and real and perceived reduced public exposure to real or hypothetical accidents.140 the pIuS ReActoR In the 1980s the reactor vendor ASEA-Atom developed a preliminary design for an intrinsically safe commercial reactor known as the Process Inherent Ultimate Safety (PIUS) reactor. An ASEA-Atom official described the companys motives for developing the reactor as follows:141 The basic designs of todays light water reactors evolved during the 1950s when there was much less emphasis on safety. Those basic designs held certain risks, and the control of those risks led to an increasing proliferation of add-on systems and equipment ending up in the present complex plant designs, the safety of which is nevertheless being questioned. Rather than to continue into this blind alley, it is now time to design a truly forgiving light water reactor

65

137 138 139 140 141

A lengthier version of this discussion is provided in: Thompson, 2008a. Watson et al, 1972. Watson et al, 1972, Appendix I. Overbye et al, 2002. Hannerz, 1983, pp 1-2.

in which ultimate safety is embodied in the primary heat extraction process itself rather than achieved by add-on systems that have to be activated in emergencies. With such a design, system safety would be completely independent of operator actions and immune to malicious human intervention. The central goal of the PIUS design was to preserve fuel integrity under all conceivable conditions. That goal translated to a design specification of complete protection against core melting or overheating in case of: any credible equipment failures; natural events, such as earthquakes and tornadoes; reasonably credible operator mistakes; and combinations of the above; and against: inside sabotage by plant personnel, completely knowledgeable of reactor design (this can be considered an envelope covering all possible mistakes); terrorist attacks in collaboration with insiders; military attack (e.g., by aircraft with off-the-shelf non-nuclear weapons); and abandonment of the plant by the operating personnel.142 To meet those requirements, ASEA-Atom designed a light-water reactor the PIUS reactor with novel features. The reactor pressure vessel would contain sufficient water to cool the core for at least one week after reactor shut-down. Most of that water would contain dissolved boron, so that its entry into the core would inherently shut down the reactor. The borated water would not enter the core during normal operation, but would enter through inherent mechanisms during off-normal conditions. The reactor pressure vessel would be made of pre-stressed concrete with a thickness of 25 feet. That vessel could withstand an attack using 1,000-pound bombs. About two-thirds of the vessel would be below ground. ASEA-Atom estimated that the construction cost of a four-unit PIUS station with a total capacity of 2,000 MWe would be about the same as the cost of a station equipped with two 1,000 MWe conventional light-water reactors. The PIUS station could be constructed more rapidly, which would offset its slightly lower thermal efficiency. Thus, the total generating cost would be about the same for the two stations. ASEA-Atom estimated (in 1983) that the first commercial PIUS plant could enter service in the early 1990s, if a market existed.143 To date, no PIUS plant has been ordered. deSIgn cRIteRIA foR ReducIng RISk Table App-1 sets forth criteria for designing and siting a nuclear power plant that would pose a risk substantially lower than is posed by the Generation II plants that are now in use worldwide, and by the Generation III plants whose construction in Ontario is being considered. These criteria are similar to ASEA-Atoms design specification for the PIUS plant. Thus, there is evidence that the criteria set forth in Table App-1 are achievable. If ASEA-Atoms cost projections were accurate, there would be no overall cost premium for complying with such criteria. An initial review of the three types of Generation III plants whose construction in Ontario is being considered shows that none of the three designs could meet the criteria in Table App-1.144

Table App-1
criteria for design and Siting of a new nuclear power plant that poses a Residual Risk Substantially lower than is posed by generation II or III plants

66

APPLiCAtioN of CRitERiA
142 Hannerz, 1983, page 3. 143 Hannerz, 1983, pp 73-76. 144 Thompson, 2008a, Section 5.

Criteria

safety performance of the plant during reactor operation (design-basis criteria)

no significant damage of the reactor core or adjacent stored spent fuel in the event of: Loss of all electrical power (AC & DC), compressed air, other power sources, and normal heat sinks for an extended period (e.g., 1 week); Abandonment of the plant by operating personnel for an extended period (e.g., 1 week); Takeover of the plant by hostile, knowledgeable persons who are equipped with specified explosive devices, for a specified period (e.g., 8 hours); Military attack by specified means (e.g., 1,000-pound air-dropped bombs); An extreme, specified earthquake; Conceivable erroneous operator actions that could be accomplished in a specified period (e.g., 8 hours); or Any combination of the above.

safety performance of the plant during reactor refueling (design-basis criteria)

A specified maximum release of radioactive material to the accessible environment in the event of: Loss of reactor coolant at a specified time after reactor shut-down, with replacement of the coolant by fluid (e.g., air, steam, or unborated water) creating the chemical and nuclear reactivity that would maximize the release of radioactive material, at a time when the plant's containment is most compromised; and Any combination of the events specified above, in the context of reactor operation.

site specification (radiological-impact criteria)

In the event of the maximum release of radioactive material specified above, in the context of reactor refueling, radiological impacts would not exceed specified values regarding: Individual dose; Population dose; and Land areas in various usage categories that would be contaminated above specified levels.

NotEs: (a) The criteria in the first two rows of this table would apply to spent fuel stored adjacent to the reactor core. Separate criteria would apply to an independent facility for storing spent fuel, whether onsite or offsite. (b) For a more detailed discussion, see: Thompson, 2008a.

67

www.greenpeace.ca
Greenpeace Canada 33 Cecil St. toronto, Ontario, M5T 1N1 454, avenue Laurier Est, 3e tage, Montral (Qubec), H2J 1E7 1726 Commercial Drive, Vancouver, British Columbia, V5N 4A3 6238 - 104 Street NW, Edmonton, Alberta, T6H 2K9
1 800 320-7183

The Hazards of Generation III Reactor Fuel Wastes

Implications for Transportation and Long-Term Management of Canadas Used Nuclear Fuel

MARVIN RESNIKOFF, PH.D JACKIE TRAVERS, AND EKATERINA ALEXANDROVA

Radioactive Waste Management Associates

www.greenpeace.ca

MAY 2010

The Hazards of Generation III Reactor Fuel Wastes:

Implications for Transportation and Long-Term Management of Canadas Used Nuclear Fuel
BY Marvin Resnikoff, Ph.D ,Jackie Travers, and Ekaterina Alexandrova, Radioactive Waste Management Associates Prepared under the sponsorship of Greenpeace Canada Copyright May 2010 GREENPEACE CANADA 33 Cecil St. Toronto, Ontario M5T 1N1 www.greenpeace.ca 1 800 320-7183

Greenpeace is one of the worlds most effective and best-known environmental organizations, with almost three million supporters worldwide in thirty countries. Greenpeace is an independent global campaigning organization that acts to change attitudes and behaviour, in order to protect and conserve our environment and promote a peaceful future. Greenpeace has had a long-standing interest in nuclear issues, and has worked to promote a shift away from nuclear power and fossil fuels towards sustainable energy systems based on conservation and renewable technologies.

Radioactive Waste Management Associates

526 W. 26th Street, #517 New York, NY 10001 About Radioactive Waste Management Associates (RWMA): Founded in 1989, RWMA is an expert team of scientists and engineers that assists governments and communities to evaluate and address the risks of radioactive management. It has provided expert analysis for numerous environmental and regulator reviews on projects involving radioactive sources, including for the State of Nevada. This report was prepared by RWMA under the sponsorship of Greenpeace Canada. RWMA is solely responsible for the content of this report.

Executive Summary
Canadas present generation of nuclear plants was built with no prior plan as to how to manage the radioactive wastes it would produce. Canada is arguably on the cusp of repeating this mistake. In 2006, the Ontario government directed Ontario Power Generation (OPG) to begin the approvals process to build as many as four additional reactors at the Darlington Nuclear Station east of Toronto. The three new reactor designs under consideration differ significantly from all CANDU reactors to date in that they use enriched uranium as a fuel source. All currently operating CANDU reactors in Canada use natural uranium as fuel source. This shift from natural to enriched reactor fuel creates additional radiological hazards for reactor operation and for radioactive waste management. There are few developments that necessitate the use of precaution more than the construction of nuclear reactors, given the resulting economic, social, and environmental risks. Typically, environmental assessments are the tool through which possible environmental consequences of projects are evaluated. Environmental assessments of the proposed new reactors in Canada, however, are failing to adequately examine a key impactradioactive waste. This report examines the likely impacts of continued radioactive waste production from a new generation of reactors of so-called Generation III designs. The Canadian public has been deprived of an objective forum for reviewing the potential impacts of continued radioactive waste production. Federal government agencies have excluded the long-term management of radioactive waste from the environmental reviews on new nuclear stations in Canada, citing the Nuclear Waste Management Organizations (NWMO) legislated responsibility for managing used nuclear fuel waste. The NWMO, however, has not consulted Canadians on the wastes that would be created by reactors of the proposed new designs, nor demonstrated that these new wastes can be managed safely in the long term. To inform the discussion on radioactive waste management in Canada, this report identifies the unassessed risks and costs that would be created by the use of enriched uranium in Generation III reactors in Canada.

New Reactor DesignsNew Waste Hazards

Unlike all of the currently operating CANDU reactors in Canada, which use natural uranium, all of the reactors being considered for construction in Canada use enriched uranium fuel, a factor which was not covered by the NWMO in its risk assessments or public consultations. New reactor fuel waste will be between 2 and 158 times more radioactive than the waste produced by existing CANDU reactors. While the use of enriched uranium will lower the volume of radioactive waste, it will increase the amount of long-lived hazardous radionuclides contained in it, such as americium, curium and plutonium, when compared to the waste of the natural uranium fuel of CANDU reactors, for the same amount of energy produced. The radioactive waste produced by CANDU reactors that have already been operational will take almost one million years to decay to levels of radioactivity equivalent to that of natural uranium. Waste from the proposed light-water designs for new reactors will take approximately 2.6 million years (see Section 5.4.1). The higher burn-up (fission eventssee Section 4.1) of fuel in Generation III reactor designs also poses new challenges to the safe operation of the reactor, due to the increase in wear on materials and components. Higher fuel burn-up will result in increased environmental and human health consequences in the event of an accident that leads to the release of radioactive substances. All of the reactors being considered for construction in Canada are untested prototypes. There is, then, no empirical evidence to support the claims to safety.

IMPACTS ON THE NWMOS PROPOSED FUEL WASTE MANAGEMENT APPROACH

After three years of public consultations, in 2005 the NWMO recommended a radioactive waste management strategy called Adaptive Phase Management (APM) to the Canadian federal government. APM proposed that the used nuclear fuel wastes be moved from storage at reactor sites to centralized storage and finally to deep-rock disposal. The federal government subsequently accepted the proposal and directed the NWMO to begin implementation. The public consultations and risk assessments, however, used to develop the NWMOs waste management approach only considered how to reduce risks from the finite waste stockpiles produced during their expected life-spans by Canadas existing CANDU reactors. The continuation of production of radioactive waste and the introduction of new forms of this waste call into question many of the assumptions of the NWMOs plan. Implementing the NWMOs Adaptive Phase Management approach while continuing to produce radioactive waste well into the future increases the risks to Canadian society instead of reducing the risks. The NWMOs proposed consolidation of radioactive waste at a centralized site arguably only reduces risks if the wastes are limited.. The increased hazards attached to the next generation of fuel waste will require more robust interim storage facilities and longer storage at reactor sites, heightening risks from terrorism. The increased radioactivity of new reactor fuel wastes will boost the consequences of an accident or sabotage during transportation to any future centralized storage site. The higher radioactive and thermal output from new fuel wastes will complicate the engineering challenges in designing any future deep geological repository. New reactor waste generates significantly more decay heat, thereby influencing water migration, package corrosion, and geological deformation of any repository. Generation III fuel waste will also require more underground acreage per ton than current Generation II CANDU waste. A geological repository that would contain both used enriched uranium fuel and current used CANDU fuel would require a volume of water comparable to over 2.3 Lake Ontarios to dilute the radioactive contents to concentrations of safe drinking water. This is over 6 times more than the water volume that would be required to dilute a geological repository containing only currently produced CANDU used fuel.

Repositories and Increasing Financial Costs, Hazards

The estimated cost of building Generation III reactors has increased significantly over the past decade. After receiving competitive bids from Areva, Westinghouse, and Atomic Energy of Canada Limited, the Ontario government suspended its procurement of new reactors, saying the bids were billions too high. In addition to the increasing front-end cost of constructing Generation III reactor designs, increases in the back-end uncertainties and costs of fuel management will further erode the economic desirability of new reactor construction. With no technically proven and socially acceptable repository operating anywhere in the world, cost estimates of such are highly uncertain. The estimated costs under proposals for deep geological repositories in Canada and the United States have risen dramatically and can be expected to continue increasing. The move toward enriched reactor fuels implies wholesale changes to the designs of storage and long-term waste management facilities and to transportation modes, which will increase costs. The increased hazards of new fuel waste will multiply fuel management costs and uncertainties, but the magnitude of these increases has not been assessed. The NWMO wrongly claims continued waste production will reduce nuclear fuel management waste costs per unit of production, by assuming additional wastes will be the same in nature and hazards as Generation II CANDU wastes. This report concludes that the magnitude of these increases in costs and hazards has not been adequately assessed, and that if these factors are left unassessed and underestimated, proceeding with new nuclear reactor construction would be granting a de facto subsidy to the industry, unfairly paid for by tax-payers, rate-payers and future generations.

Table of Contents
1. Introduction...................................................................................................................................................................................................................5 2. NWMO Final Study.................................................................................................................................................................................................9
2.1 Concept 1: Deep Geological Disposal in the Canadian Shield...................................................................................................... 9 2.2 Concept 2: Storage at Nuclear Reactor Sites......................................................................................................................................10 2.3 Concept 3: Centralizing Storage, Above or Below Ground..........................................................................................................10 2.4 NWMOs Assessment and Consultation Process. .............................................................................................................................11

3. NWMO Recommendation: Adaptive Phase Management.................................................................................. 12

3.1 Phase 1: Preparing for Central Used Fuel Management................................................................................................................12 3.2 Phase 2: Central Storage and Technology Demonstration...........................................................................................................13 3.3 Phase 3: Long-Term Containment, Isolation, and Monitoring.....................................................................................................14 3.4 Cost of Implementation......................................................................................................................................................................................14 3.5 Associated Risks and Uncertainties............................................................................................................................................................15

4. Impacts of Expanded Nuclear Waste Production.......................................................................................................... 17

4.1 Proposals for Generation III Reactors........................................................................................................................................................17 4.1.1 Atomic Energy of Canada Limited ACR 1000..........................................................................................................................19 4.1.2 Westinghouses AP1000......................................................................................................................................................................20 4.1.3 Arevas EPR...................................................................................................................................................................................................20

5. New Reactor Designs: Increased Waste Hazards......................................................................................................... 22

5.1 Operation. ...................................................................................................................................................................................................................23 5.2 Interim-Storage at Reactor Sites. .................................................................................................................................................................24 5.3 Transportation. .........................................................................................................................................................................................................26 5.3.1 Number of Shipments.............................................................................................................................................................................26 5.3.2 Transportation Accidents and Sabotage Attacks...................................................................................................................27 5.4 Long-Term Disposal in Geologic Repository. ........................................................................................................................................29 5.4.1 High Burn-upHigh Hazard...............................................................................................................................................................29 5.4.2 Environmental Impacts...........................................................................................................................................................................30

6. Cost...................................................................................................................................................................................................................................... 34
6.1 Canadian High-Level Waste Repository...................................................................................................................................................34 6.2 High-Level Waste Repository at Yucca Mountain, Nevada. .........................................................................................................35 6.3 Implications of High Burn-up Fuel................................................................................................................................................................35

7. Conclusion................................................................................................................................................................................................................... 36 References.......................................................................................................................................................................................................................... 38

List of Tables
Table 1. Nuclear Fuel Wastes in Canada as of 30 June 2009.................................................................. 6 . Table 2. Existing CANDU Nuclear Power Reactors in Canada.................................................................. 7 Table 3. Summary of Existing CANDU (Generation II) Reactors and Generation III+ Reactor Types Proposed for the Darlington New Build............................................................................. 21 . Table 4. Summary of the Fuel Types for Proposed Generation III Reactors, Normalized to the Average Net Electricity Generation of a CANDU Reactor........................................... 22 Table 5. Scenarios for the Waste Repository Proposed by NWMO........................................................ 29 . Table 6. Liters of Water Needed to Dilute the Full Repository.................................................................. 30

1. Introduction
Canadas present generation of nuclear plants was built with no prior plan as to how to manage the radioactive wastes they would produce. The Nuclear Fuel Waste Act (NFWA) came into force on 15 November 2002, more than forty years after Canadas nuclear power reactors started producing radioactive fuel waste. The NFWA specified that the Nuclear Waste Management Organization (NWMO), which was constituted from Canadas radioactive waste producersOntario Power Generation (OPG), New Brunswick Power (NBP), HydroQubec (HQ) and Atomic Energy of Canada Limited (AECL)would recommend within three years (i.e., by November 2005) one of three possible approaches for the management of Canadas existing stockpiles of nuclear fuel waste. These approaches included disposing of the fuel in a deep geological repository in the Canadian Shied, storing the fuel at nuclear sites, and storing the fuel in a centralized storage site either above or below ground.1 After a three-year period of consulting the Canadian public and studying how to best reduce the risks posed by existing stockpiles of radioactive waste, the NWMO recommended the Adaptive Phase Management (APM) plan for storing and managing the existing high-level radioactive fuel waste in Canada. The APM approach essentially re-proposed Atomic Energy of Canada Limiteds (AECL) long-time plan to dispose of radioactive fuel waste in an underground repository, with a step-wise implementation process and a timeline of up to 300 years. It has been observed that the APM approach of the NWMO is the culmination of a series of public policy failures in efforts to site a geological repository, along with the declining support for nuclear projects.2 Support for the disposal of Canadas high-level radioactive waste in a deep geological repository was first declared in a joint statement of the federal and Ontario governments in 1974, following Indias explosion of an atomic bomb that year. The public uproar caused by the revelation that diversion of spent nuclear fuel produced by Canadian nuclear technology had been used to construct an atomic bomb initiated revisions to Canadas nuclear proliferation policy and highlighted the risks of fuel waste.3 Previous to the joint statement by the two governments, AECL and the two provincial electricity utilities with radioactive waste at the timeOntario Hydro and Hydro-Qubechad endorsed the plan known as monitorable and retrievable storage (MRS), because of the flexibility it provided and because permanent disposal had not been proven a viable alternative. With increased public understanding of how used nuclear fuel could contribute to atomic arms proliferation, MRS became politically contentious and this motivated the turn toward deep geological disposal, despite the latters technical uncertainties. This initial support for MRS, however, has been omitted from the communications of both AECL and the NWMO, arguably to diffuse questions on the technical aspects of deep geological repositories.4 In 1977, a panel which was commissioned by the federal government and chaired by F.K. Hare further solidified the federal policy for disposing of radioactive waste by recommending that the federal government develop policies to establish a deep geological repository in the Canadian Shield. The Hare Report further recommended that two sites be selected by 1983 and that a final site be operational by 1990.5 In response, the federal and Ontario governments issued a joint policy statement in 1978, instructing AECL and Ontario Hydro to develop storage and disposal technologies. Site selection was required by 1983, with a repository to be operational by 2000.6 In 1978, however, the Ontario Royal Commission on Electric Power Planning, which did not rely only on evidence from industry, concluded that the scientific basis for the deep geological disposal was inadequate to ensure the protection of future generations. The Commission recommended a moratorium on reactor construction if proof to the contrary was not provided by 1985.
5

1 2 3 4 5 6

Canada Department of Justice. Nuclear Fuel Waste Act: 12(2). Durant, D. Radwaste in Canada: A Political Economy of Uncertainty. Journal of Risk Research, 12:7, pp. 897-919. (2009.) Bratt, D. The Politics of CANDU Exports. Toronto: University of Toronto Press. (2006). Cited in Durant, D., 2009, p. 901. Aikin, A.M., J.M. Harrison and F.K. Hare. Green Paper on Nuclear Waste Management. EMR Report EP 77-6. Ottawa: Department of Energy, Mines and Resources (1977). Durant, D. Radwaste in Canada: A Political Economy of Uncertainty. Journal of Risk Research, 12:7: 902 (2009).

In reaction to the mounting community opposition to AECLs efforts to site a repository and to the public skepticism of reactor safety following the Three Mile Island accident, the federal and Ontario governments directed AECL and Ontario Hydro to win approval of the deep geological repository concept before siting could begin.7 It took AECL fifteen years to finally fully respond to the governments directive. In 1994, AECL released an environmental impact statement (EIS) on its deep geological repository concept.8 A panel known as the Seaborn Panel, after its chairman Blair Seaborn, studied the proposal for years. Regarding deep geological disposal, the Seaborn Panel concluded that From a social perspective, safety of the AECL concept has not been adequately demonstrated for a conceptual stage of development.9 The non-industry evidence and testimony during the hearings demonstrated to the Panel that public trust and social acceptance are undermined when supportive government bureaucracies and waste producers control the technical knowledge and determine the approaches to waste management. In its final report, the Panel recommended that the nuclear waste be managed by an agency at arms length from industry and from the government.10 The Chrtien government, however, ignored this recommendation of the Panel and under the NFWA gave control to an agency comprised of nuclear waste producers (Ontario Power Generation [OPG], Hydro-Qubec, New Brunswick Power, and AECL)the NWMO. As of 2005, the NWMOs plants had produced over 42,000 tonnes of used nuclear fuel, which is stored at Canadas five power reactor sites, in Ontario, Quebec and in New Brunswick, as well as at AECL laboratories at Chalk River, Ontario, and Whitshell, Manitoba (Table 1).11

Table 1. Nuclear Fuel Wastes in Canada as of June 30, 200912

LOCAtION Bruce, Ontario Darlington, Ontario Douglas Point, Ontario (at the Bruce site) Gentilly, Quebec Pickering, Ontario Point Lepreau, New Brunswick Whiteshell, Manitoba Chalk River, Ontario Total Number of Fuel Bundles: Waste Owner OPG OPG AECL AECL & Hydro-Qubec OPG NB Nuclear Power AECL AECL Total Number of Used Fuel Bundles 901,595 356,504 22,256 115,658 598,687 121,758 2,268 4,886 2,123,612

There are five power reactor sites in Canada, with 22 reactors in various states of operation, repair, or permanent shut-down (Table 2). In addition to these power reactors, there are three experimental reactors: Gentilly-1 at Bcancour, Quebec, Douglas Point at the Bruce site in Ontario, and NPD (Nuclear Power Demonstration) facility at Rolphton, Ontario. These retired stations are under the management of AECL.

Energy, Mines and Resources Minister (ERM) and Ontario Energy Minister (OEM). Canada/Ontario Joint Statement on the Radioactive Waste Management Programme. Ottawa: Ministry of Supply and Services (5 June 1978). 8 Atomic Energy of Canada Limited (AECL). Environmental Impact Statement on the Concept for Disposal of Canadas Nuclear Fuel Waste. Pinawa: AECL: AECL-10711, COG-93-1. (1994.) 9 Canadian Environmental Assessment Agency (CEAA). Nuclear Fuel Management and Disposal Concept (Seaborn Report). EN-106-30/11998E. Ottawa: Public Works and Government Services Canada (1998): Ch. 5.2.2.1. 10 Ibid. 11 Nuclear Waste Management Organization (NWMO). Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p.350. 12 Garamszeghy, M. Nuclear Fuel Waste Projections in Canada2009 Update. NWMO TR-2009-30. Nuclear Waste Management Organization, Toronto, Ontario (2009): Table 1.

Table 2. Existing CANDU Nuclear Power Reactors in Canada13

LOCAtION Nuclear Power Company Licensee Year InService Net Electrical Power Produced (MW(e)) Current Status

Bruce NuclearStation, Ontario Bruce A - 1 Bruce A - 2 Bruce A - 3 Bruce A - 4 Bruce B - 5 Bruce B - 6 Bruce B - 7 Bruce B - 8 Darlington, Ontario Darlington 1 Darlington 2 Darlington 3 Darlington 4 Bcancour, Qubec Gentilly 2 Pickering, Ontario Pickering A - 1 Pickering A - 2 Pickering A - 3 Pickering A - 4 Pickering B - 5 Pickering B - 6 Pickering B - 7 Pickering B - 8 OPG OPG OPG OPG OPG OPG OPG OPG 1971 1971 1972 1973 1983 1984 1985 1986 515 515 515 515 516 516 516 516 Operating Permanently shut down in 2005 Permanently shut down in 2005 Operating Operating Operating Operating Operating Hydro-Qubec 1983 635 Operating OPG OPG OPG OPG 1992 1990 1993 1993 881 881 881 881 Operating Operating Operating Operating Bruce Power Bruce Power Bruce Power Bruce Power Bruce Power Bruce Power Bruce Power Bruce Power 1977 1977 1978 1979 1984 1985 1986 1987 750 750 750 750 795 822 822 795 Undergoing refurbishment Undergoing refurbishment Operating Operating Operating Operating Operating Operating

Point Lepreau, New Brunswick NB Power Point Lepreau Nuclear

1983

635

Undergoing refurbishment

All of the reactors in operation in Canada are CANDU (CANadian Deuterium Uranium) heavy-water reactors. Unlike internationally predominant light-water reactor designs, which use enriched uranium as fuel, Canadas CANDU reactors use natural uranium as fuel. This use of natural uranium has been promoted as an advantage of the CANDU design because it avoids dependence on foreign uranium enrichment technology. As will be discussed, the natural uranium fuel affects the toxicity, longevity and volume of waste produced by CANDU reactors. As required by the NFWA, in 2002 the NWMO undertook a study of and a public consultation process on the three waste management approachesstorage at reactors sites, centralized storage, and deep geological disposal. The NFWA required the NWMO to compare the benefits, risks and costs of each approach, as well as ethical, social and economic considerations associated with that approach.14
13 Ibid.: p. 11. 14 Canada Department of Justice. Nuclear Fuel Waste Act: 12(4).

Notably, a fundamental assumption of the NWMOs risk assessments and consultations was that the volume of used nuclear fuel which needs to be managed was assumed to be limited to the projected inventory from the existing fleet of reactors.15 Otherwise put, nuclear waste production would end after the current fleet of reactors reaches the end of their operational lives.

Indeed, the Advisory Committee of the NWMO made the following statement in its report to the government in 2005:
The Advisory Council would be critical of an NWMO recommendation of any management approach that makes provision for more nuclear fuel waste than the present generating plants are expected to create, unless it were linked to a clear statement about the need for broad public discussion of Canadian energy policy prior to a decision about future nuclear energy development. The potential role of nuclear energy in addressing Canadas future electricity requirements needs to be placed within a much larger policy framework that examines the costs, benefits and hazards of all available forms of electrical energy supply, and that framework needs to make provision for comprehensive, informed public participation.

The Advisory Council also highlighted that new reactors would use enriched fuels, creating hazards that had not been assessed during the NWMOs study and consultations between 2002 and 2005:
a nuclear expansion scenario would likely entail fuel enrichment and new reactor technology, with spent fuel possessing new characteristics. These could affect the performance of the disposal technology and introduce a change in the outlook on reprocessing. Such technical aspects were not considered by NWMO in its study, which focused on existing facilities using natural uranium fuel.16

Specifically, the Advisory Committee stated in its final report to the federal government:
we emphasize, as did many other participants in the process, that any significant change in the amount or type of used fuel to be managed (whether due to phase out or expansion of the nuclear program) should trigger a review of the work undertaken by the NWMO to date.17 A month after the NWMO submitted its recommendation for APM to the federal government in 2005, the Ontario Power Authority (OPA), a new provincial planning agency, recommended that Ontario plan for and expedite the approvals to build new nuclear reactors.18 On 12 June 2006, the Ontario government accepted OPAs proposal and directed OPG to begin the planning process to build new nuclear stations at Darlington.19 A year later, the federal government accepted the NWMOs recommendations and directed the agency to proceed with implementing APM.20 Aside from the risks posed by the continued waste production, the introduction of nuclear fuel that is much more radioactive into the mix of used fuel that must be transported and managed in the long term must be assessed. NWMO acknowledges that the radiological hazards of Generation III fuel assemblies are much greater than those of CANDU fuel bundles, and believes that additional studies, safety assessments, facility designs and technology demonstrations would be required for each waste form.21 Since the NWMOs proposal for a deep geological repository only takes into consideration Canadas existing reactor fleet and a projected number of used CANDU fuel bundles based on current operations, the human and environmental health risks, as well as repository construction costs associated with these new types of fuel, have not been adequately estimated. Furthermore, the risk assessments on which NWMO consulted the public did not include the health risks that would result from the new type of used nuclear fuel. In this report we examine the long-term hazards posed by this new generation of reactors and their used fuels, as well as their impact on transportation and financial costs.
15 16 17 18 19 20 21 NWMO. Assessing the Options. NWMO Assessment Team Report (June 2004): p. 14. Advisory Council to the NWMO. Nuclear Waste Management Organization Advisory Council Final Report. (22 September 2005). Ibid. Ontario Power Authority (OPG). Supply Mix Advice Report. (December 2005.) Duncan, D. (Minister of Energy). Letter to James Hankison (Chief Executive Officer), 16 June 2006. Natural Resources Canada. Canadas Nuclear Future: Clean, Safe, Responsible. Press release, 14 June 2007. Russell, S. Preliminary Assessment of Potential Technical Implications of Reactor Refurbishment and New Reactor Build on Adaptive Phase Management. Nuclear Waste Management Organization, Toronto, Ontario (December 2008).

2. NWMO Final Study

As required by the NFWA, the NWMO initiated its study of nuclear fuel waste management in 2002. The NFWA specified that NWMO study and compare the following approaches: 1. deep geological disposal in the Canadian Shield, based on the concept described by Atomic Energy of Canada Limited; 2. storage at nuclear reactor sites; and 3. centralized storage, either above or below ground.22 The NFWA also mandated the NWMO to consider the benefits, risks and costs of each approach, as well as the ethical, social and economic considerations associated with that approach.23 The following sections discuss the three technical concepts mandated by the NFWA.

2.1 Concept 1: Deep Geological Disposal in the Canadian Shield

Concept 1 is AECLs proposal for disposing of radioactive waste in the Canadian Shield, an idea which has existed in some form since the 1970s. The NFWA stipulates that the NWMO examine the concept, as described in AECLs submission to the Seaborn Panel.24 AECLs proposal for deep geological disposal involves storing all of Canadas used nuclear fuel in a centralized repository at a depth of 500 to 1,000 meters in the Canadian Shield. The Canadian Shield, also called the Laurentian Plateau, is a massive area of exposed Precambrian crystalline igneous and high-grade metamorphic rocks. Geologic shields, in general, are flat, very tectonically stable areas. Over a period of approximately 30 years, used nuclear fuel would be transported via road, rail or water shipments from existing interim storage facilities to the repository, where it would be repackaged in long-term corrosion-resistant containers. These containers would be moved to rooms excavated deep in the rock of the Canadian Shield, and kept in these rooms, in boreholes drilled in the floors of these rooms, or in horizontal underground tunnels, depending on site-specific conditions. The underground excavations would be backfilled and sealed and the isolation of the repository from humans and the surface environment would be provided via natural and engineered barriers. The NWMO predicted that the construction of such a facility could be completed by 2035 at the earliest. It was estimated that it would take approximately 15 years to site the deep geological repository location, another 15 years to design and construct the repository, 30 years to receive, repackage, and dispose of all used nuclear fuel, 70 years to monitor the repository, 12 years to decommission the site, and approximately 13 years to completely close the site. In 2004, it was estimated that the total cost of implementing Concept 1 would be $6.2 billion (in 2004 dollars).25 Notably, the Seaborn Panel was not convinced of the safety of AECLs deep geological disposal method, stating that From a social perspective, safety of the AECL concept has not been adequately demonstrated for a conceptual stage of development.26

22 23 24 25

Canada Department of Justice. Nuclear Fuel Waste Act: 12(2). See Nuclear Fuel Waste Act: 12(4). See Nuclear Fuel Waste Act, 12(2). NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): pp. 121125. 26 CEAA. Nuclear Fuel Management and Disposal Concept (Seaborn Report). EN-106-30/1-1998E. Ottawa: Public Works and Government Services Canada (1998): Ch. 5.2.2.1.

2.2 Concept 2: Storage at Nuclear Reactor Sites

All of Canadas radioactive fuel waste is currently stored at reactor sites. Concept 2 would make these interim waste storage facilities permanent for the foreseeable future. But this method is not currently sustainable, due to the short life expectancy of the containers in use at the present time to store used nuclear fuel. When used nuclear fuel is removed from a CANDU reactor, it is first placed in wet storage, for approximately seven to ten years. Wet storage provides the greatest amount of shielding for when the fuel is the hottest, or most radioactive. Once the radioactivity of the used nuclear fuel has significantly decreased, the fuel is then transferred to dry concrete and steel storage containers and stored at a licensed facility at the reactor site. The dry storage containers currently in use have a design life of approximately 50 years, although the expected life of the containers is estimated to be at least 100 years. Concept 2 would require either the expansion of all pre-existing dry storage facilities at the reactor sites or the construction of new, long-term storage facilities at the reactor sites or other suitable locations. Regardless of which option is deemed more feasible, the used nuclear fuel currently stored at the reactor sites will have to be transferred from existing storage containers to new ones, as the lives of the dry storage containers currently in use are shortly going to expire. If Concept 2 is approved, it is proposed that new dry storage containers would be designed to last between 100 and 300 years. Therefore, each storage facility will need to be completely refurbished or replaced approximately every 300 years. NWMO estimates that it would take approximately five years to site and gain approval of new storage facilities, an additional five years to design and construct the new facilities, approximately 35 to 40 years to transfer existing used nuclear fuel to new long-term storage containers and facilities, and up to 300 years for each monitoring and refurbishing cycle. NWMO estimates that Concept 2 would cost approximately $2.3 billion to $4.4 billion (2004 dollars) for one complete 300-year cycle.27

2.3 Concept 3: Centralizing Storage, Above or Below Ground

Concept 3 involves the creation of a single, new, long-term used nuclear fuel storage facility at a centralized location. Conceptual designs have been developed for facilities both above and below ground and storage options include casks and vaults in storage buildings, surface modular vaults, casks and vaults in shallow trenches, and casks in rock caverns. As discussed in Concept 2, the design limits of the new storage containers would mean an expected life of 100 to 300 years and the facility would have to be refurbished or replaced approximately every 300 years. If Concept 3 were accepted, the used nuclear fuel stored at the Canadian reactor sites on an interim basis would have to be transported to the new central facility and repackaged. NWMO estimates that it would take approximately 10 years to site an appropriate above or below ground location for a storage facility, an additional 10 years to design and construct the facility, approximately 40 years to receive, repackage, and store all used nuclear fuel, and approximately 100 to 300 years of monitoring before the facility is refurbished or decommissioned. It would take approximately 30 years to decommission the facility and for each facility decommissioning repeat event. NWMO estimates that the total cost of implementing Concept 3 would range between $3.1 billion to $3.8 billion (2004 dollars) for one 300-year cycle.28

10

27 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): pp. 126128. 28 Ibid., pp. 129132.

2.4 NWMOs Assessment and Consultation Process

As mandated by the NFWA, between 2002 and 2005 the NWMO consulted Canadians on the three waste management options. To carry out its consultation process, the NWMO hired firms to organize focus groups, workshops, dialogues with targeted groups, and polls. In 2003, the NWMO first consulted and reported on what questions Canadians wanted answered.29 Then it asked and reported on how Canadians understood the three proposed waste management approaches.30 This process sought to maximize the areas of consensus among Canadians and to characterize them as Canadian wants, such as adaptability, phased decision-making, robust governance, and citizen participation.31 Darrin Durant of York University has suggested a general rule for NWMOs method of determining Canadian wants: Eligibility for the class of Canadian wants is a function of whether the preference or value or goal is not fundamentally incompatible with what had been the operating mantra of AECL and Ontario Hydro since 1981: that something needed to be done about nuclear waste, that whatever was to be done should commence soon, and that a chosen course was not to be confused with decisions about the waste generating industry.32 With a view to the aforementioned Canadian wants, the NWMO considered each of the three waste storage options. Deep geological disposal (Concept 1) was viewed as providing little flexibility for future generations to be able to influence the management of used nuclear fuel or to make fundamental changes to the plan without incurring considerable additional costs. The requirement that the repository must be located in the Canadian Shield, as opposed to other suitable repository locations, narrowed the focus for siting, which could result in a less fair distribution of the costs and risks to specific geographic and economic regions of Canada. In addition, if a used nuclear fuel container breached, it might be difficult for a future generation to detect the breach in a timely way and take corrective action.33 Storage at the site (Concept 2) was considered to place too much responsibility on future generations to actively manage the used nuclear fuel produced by people today. Since there are multiple sites to be managed, the potential costs and risks passed on to future generations could be higher than the costs and risks associated with just one centralized facility. In addition, the prolonged existence of used nuclear fuel storage facilities at the reactor sites was not originally proposed to and understood by the communities and businesses that have chosen to locate themselves in the vicinity of the reactor sites.34 Centralized storage (Concept 3) was also considered to place too much responsibility on future generations to actively manage the used nuclear fuel produced by people today. In addition, the suitable area selected for above- or below-ground long-term storage of used nuclear fuel may not be in an area where the regional community benefits from the production of nuclear power energy.

11

29 30 31 32 33

NWMO. Asking the right questions? The Future Management of Canadas Used Nuclear Fuel. Discussion Document 1. Toronto (2003). NWMO. Understanding the Choices: The future management of Canadas Used Nuclear Fuel. Discussion Document 2. Toronto (2004). NWMO. Assessing the Options: Future Management of Used Nuclear Fuel in Canada. NWMO Assessment Team Report (June 2004): p. 9. Durant, D. Radwaste in Canada: A Political Economy of Uncertainty. Journal of Risk Research. Vol. 12, nos. 78 (2009): p. 911. NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 169. 34 Ibid: p. 170.

3. NWMO Recommendation:

Adaptive Phase Management

In November 2005, the NWMO proposed what it called a fourth option of nuclear waste management to the Canadian federal government. The NWMO called this fourth option Adaptive Phase Management (APM), a three-step process of moving used fuel waste from storage at the reactor sites to a centralized location, where a deep geological repository would eventually be built. The site of the deep geological repository would be voluntary and the host community would participate in the implementation of each step. Critics have argued that Adaptive Phase Management is simply a repackaging of AECLs deep geological disposal proposition (Concept 1).35 The federal governments response to the NWMO proposal indicates that the NWMO proposal may be used to promote the expansion of nuclear power. The federal government, in accepting the Adaptive Phase Management plan and directing the NWMO to begin implementation, called the initiative vital to the future of nuclear energy in Canada.36 The following sections describe, in greater detail, the features of each phase of the Adaptive Phase Management process.

3.1 Phase 1: Preparing for Central Used Fuel Management

The temporary storage of used nuclear fuel at CANDU reactor sites would be maintained while a central site with rock formations suitable for shallow underground storage, an underground characterization facility, and a deep geological repository would be selected. This site could be located in a suitable rock formation such as the crystalline rock of the Canadian Shield in central Canada or the Ordovician sedimentary rock basins in western Canada. In 1977, an independent expert group researched suitable rock types in Canada for a deep geological repository for used nuclear fuel. It was found that salt, crystalline rock, sedimentary rock, and volcanic tuff would be suitable for a deep geological repository but that efforts should be focused on crystalline rock. Several studies conducted since the 1980s have found that sedimentary rock formations are relatively simple, homogenous, and thick and have favorable geotechnical properties. Sedimentary rock formations have low hydraulic conductivity, which means the flow of groundwater in the formations is very slow.37 Based on current CANDU reactor operations, the land requirement for the surface buildings and associated facilities (such as waste repackaging or construction facilities) would be approximately 2 kilometers (km) x 3 km, or about 600 hectares (ha). The footprint of the shallow underground storage facility would be about 515 meters x 450 meters, or about 23 hectares, and would be approximately 50 meters below the surface. The footprint of the deep geological repository would be about 1.8 km2, or about 452 acres.38 The depth of the repository, however, would depend on the total number of fuel bundles or assemblies disposed of there, the heat output of this fuel, and site-specific factors such as the thermal conductivity of the rock mass. Once a proper location is sited, the licensing process, triggering the environmental assessment process, will then be initiated. Site characterization, safety analyses and environmental assessments for the shallow underground storage facility, the underground characterization facility, the deep geological repository, and transportation routes to these sites from the interim storage locations will be carried out. Once a license to prepare the site is issued, transportation canisters will be designed and certified.

12

35 Edwards, G. Following the Path Backward: A Critique of the Draft Study Report of the Nuclear Waste Management Organization Entitled Choosing the Way Forward. Canadian Coalition for Nuclear Responsibility ( September 2005). 36 Natural Resources Canada. Canadas Nuclear Future: Clean, Safe, Responsible. Press release, 14 June 2007. 37 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 118. 38 Russell, S. Preliminary Assessment of Potential Technical Implications of Reactor Refurbishment and New Reactor Build on Adaptive Phase Management. Nuclear Waste Management Organization, Toronto, Ontario (December 2008).

Subsequently, a license to construct the underground characterization facility will be obtained and the decision as to whether or not to proceed with construction of a shallow underground storage facility and to transport used fuel to the central site for storage will be made. If the decision is made to construct the shallow underground storage facility, construction and operating licenses will be obtained and the shallow underground storage facility and the underground characterization facility will be built. NWMO estimates that it would take approximately 20 years to site a central facility and an additional 10 years to design and construct the underground characterization facility and the optional shallow underground storage caverns, if the decision is made to do so. Thus, it would take approximately 30 years for Phase 1 of NWMOs Adaptive Phase Management process to be completed.39 However, NWMO estimates that the deep geological repository will be in service in 2035, only 25 years from today.40

3.2 Phase 2: Central Storage and Technology Demonstration

If the decision is made to construct the shallow underground storage facility, then the transportation of used nuclear fuel from the reactor sites to the repository will begin. At this phase of the Adaptive Phase Management process, it will have to be decided whether the transportation of used nuclear fuel will be by road, rail, or water, depending on the location of the central facility in relation to the reactor sites. Based on current CANDU operations, it is estimated that the number of shipments of used nuclear fuel from the interim-based reactor storage sites to the repository would be approximately 53 road shipments per month for 30 years, five rail shipments and 36 road shipments per month for 30 years, or 2 water shipments and 36 road shipments per month for 30 years.41 Subsequently, an emergency transportation response plan will be developed and the appropriate security provisions will be made. If the decision to construct a shallow underground storage facility is not made, used fuel will remain at the interim-based reactor storage sites until the deep geological repository is constructed and available for use. Continuous research and testing at the underground characterization facility will be conducted in order to determine the suitability of the site for a deep geological repository. At this point, citizens would be involved in assessing the site location, the technology used at the site, and the timing for placement of used nuclear fuel in the deep repository. Assuming citizens responded positively to the Adaptive Phase Management process, the time frame for when the deep repository should operate would then be determined. Final design and safety analyses required for obtaining an operating license for the deep geological repository and associated surface handling facilities will then be carried out and completed. It is possible that transportation containers, and possibly facilities in which these containers would be produced, could be needed during this phase of the Adaptive Phase Management process. If so, transportation containers and the facilities needed to produce the containers will be constructed at this time. NWMO estimates that it will take approximately 30 years for Phase 2 of the Adaptive Phase Management process to be completed.42

13

39 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): pp. 133144. 40 Hung, M. Financial Implications of Used Fuel Volume Variation in Long Term Management2008 Update. Nuclear Waste Management Organization, Toronto, Ontario APM-REP-03780-0001 (December 2008). 41 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 123. 42 Ibid., pp. 133144.

3.3 Phase 3: Long-Term Containment, Isolation, and Monitoring

If the decision is made to store used nuclear waste in the shallow underground facility, this waste will now be retrieved and repackaged into long-term storage containers. If used fuel is still being stored at interim-based reactor storage sites, waste will now be transported to the central repository location for repackaging into long-term storage containers. These containers will have a corrosion barrier and a claimed design life of approximately 100,000 years, based on the designs of the Finnish and Swedish radioactive waste management programs repository containers. A secondary, inner steel container will also be included, so that containers can withstand environmental effects such as glaciation or tectonic movement. All the fuel containers will be placed in the deep geological repository for final containment within a network of horizontal access tunnels, excavated room, or boreholes drilled into the floors of the excavated rooms in stable rock approximately 500 to 1,000 meters below the surface. Based on current CANDU operations, it is assumed it will take approximately 30 years to place all used fuel containers in the deep repository.43 One-hundred-percent bentonite clay will be used to seal off all access to the used nuclear fuel containers and to fill any void spaces in the repository. This is intended to limit the movement of groundwater and dissolved materials in the repository. It is assumed that both the Canadian Shield and Ordovician sedimentary rock basins have long-term stability, good rock strength, and low groundwater flow. 44 The deep repository will be continuously monitored over an estimated period of 240 years so that the performance of the repository system can be assessed. Access to the repository will also be maintained, in case the fuel would need to be retrieved. Closure of the facility is estimated to occur in approximately 300 years and post-closure monitoring would take place from the surface. The decision to decommission the underground characterization facility and any remaining longterm experiments or demonstrations of technology will be left in the hands of a future generation. Future generations will also be left with the decisions of exactly when to close the repository, when to decommission the surface-handling facilities, and the nature of any post-closure monitoring of the system. NWMO estimates that decommissioning and closure activities will take approximately 25 years or more and that post-closure monitoring would be indefinite.

3.4 Cost of Implementation

NWMO estimates that it will cost approximately $14.9 billion (2008, constant dollars) to carry out all three phases of the Adaptive Phase Management process, calculating that 3.6 million CANDU fuel bundles will be produced by 2034, the end of all of Canadas existing CANDU nuclear power reactors lifetimes. Based on this projection, the deep geological repository would require 160 placement rooms, holding 70 used nuclear fuel containers per room, assuming 324 used fuel bundles are stored in each container. Each additional year of operation with the current fleet of CANDU nuclear reactors requires four additional placement rooms in the deep geological repository and one more year of repository operation. Each additional year of operation of Canadas current fleet of nuclear power reactors will increase the design size of the deep geological repository by approximately 2.5 percent. With waste production limited to the current fleet of reactors, NWMO assumes that it will take 30 years to transport, repackage, and dispose of the projected 3.6 million used fuel bundles produced by all existing CANDU reactors by the year 2034.45 The deep geological repository will be capable of receiving 120,000 used CANDU nuclear fuel bundles per year46. The Nuclear Fuel Waste Act (NFWA) specifically assigned the responsibility for financing the long-term management of Canadas used nuclear fuel to the companies that are the owners of the used nuclear fuel wastes: OPG, Hydro-Qubec, NBP Nuclear, and Atomic Energy Canada Limited (AECL). Under the NFWA, each of these used nuclear fuel owners is required to establish a trust fund and make annual payments to the fund in order to implement the NWMOs Adaptive Phase Management plan.
43 Ibid. 44 Ibid. 45 Russell, S. Preliminary Assessment of Potential Technical Implications of Reactor Refurbishment and New Reactor Build on Adaptive Phase Management. Nuclear Waste Management Organization, Toronto, Ontario (December 2008). 46 Hung, M. Financial Implications of Used Fuel Volume Variation in Long Term Management2008 Update. APM-REP-03780-0001. Nuclear Waste Management Organization, Toronto, Ontario (December 2008).

14

The NWMO developed a funding formula to determine the cost-sharing responsibilities of each used nuclear waste owner for the long-term management of Canadas used nuclear fuel. This formula incorporates the highest cost option for implementing the Adaptive Phase Management plan, provides for reasonably foreseeable and unforeseeable events, and collects funds over the assumed economic life of the nuclear reactors producing the used nuclear fuel bundles. Under this formula, each waste owner will pay based on the quantity of waste produced and their subsequent usage of the repository. The timing and number of used nuclear fuel bundles will also be accounted for in the sharing of repository costs. As of 30 June 2006, it was estimated that the transfer of OPG-owned used nuclear fuel from interim-based storage facilities at the reactor sites to the repository will begin in 2035. It is anticipated that the transfer of Hydro-Qubec, NBP Nuclear, and AECL waste will not begin until 2050. As of 30 June 2006, OPG is to account for 90.8 percent of all repository costs, whereas Hydro-Qubec, NBP Nuclear, and AECL are to account for 3.9, 4.2, and 1.2 percent, respectively, of all repository costs. Cost-sharing percentages will be reviewed and updated periodically on a minimum of a five-year cycle. Annual trust fund contributions are broken down into committed and future costs. Committed costs are the costs for used nuclear fuel bundles that were known to be produced as of 30 June 2006. Future costs are the costs for used nuclear fuel bundles generated from 1 July 2006 onward. All trust fund contributions for committed used nuclear fuel bundles will be fully submitted by used nuclear waste owners by 2035 so that all funds necessary to meet the complete construction costs of the deep geological repository will be available by the time the construction begins. The current average cost for a deep geological repository for 3.6 million CANDU used nuclear fuel bundles is $4,140 per fuel bundle. This cost is based on the size and the total number of fuel bundles estimated to be received by the repository; it does not incorporate the thermal heat output of each fuel. Trust fund contributions for future used nuclear fuel bundles will be deposited in the year following the generation of the bundles. For example, a waste owners contribution to its trust fund in 2009 will cover the cost of disposal for used nuclear fuel bundles generated between 1 July 2007, and 30 June 2008. At present, the cost of disposing a projected 3.6 million CANDU used nuclear fuel bundles in the deep geological repository is estimated to be $14.9 billion (2008, constant dollars).

3.5 Associated Risks and Uncertainties

NWMOs Adaptive Phase Management approach is a reformulation of the long-standing proposal for a deep geological repository. The debate about the safety of such a disposal method has continued throughout the globe for years, but severe risks and uncertainties pertaining to the issue remain. The design, safety, and effectiveness of deep geological repositories have been studied worldwide for several decades through the use of underground research laboratories (URLs). Although research has made progress in understanding deep geologic disposal methods, it is still uncertain that such facilities can provide ultimate safety in the long term. General uncertainties on whether a satisfactory design is possible will continue to exist for a long time because the kinetics of processes taking place in the geological medium within a repository are very slow and would take decades to better understand.47 Many of the scientific predictions stated in various reports use models to overcome the challenge of long timescales; however, the typical uncertainty associated with a long-term model should prevent any claims of total validity. 48 In 1991, the French Parliament passed its Waste Act, mandating that a 15-year research program for deep geological disposal of radioactive waste be carried out. In 1998, the French Parliament selected a clay formation in the northeast as an underground research laboratory site.49 In 2006, after 15 years of inconclusive research, the failure to find a solution led the French Parliament to authorize the continuation of research on disposal and on long-term storage of spent nuclear fuel.50 As of the writing of this report, France has been unable to name an acceptable deep geological repository site.51
15
47 International Atomic Energy Agency (IAEA). The use of scientific and technical results from underground research laboratory investigations for the geological disposal of radioactive waste. IAEA-TECDOC-1243 (September 2001). 48 Shrader-Frechette, K. Risk and Uncertainty in Nuclear Waste Management. NWMO (November 2003). 49 National Research Council. Disposition of High-Level Waste and Spent Nuclear Fuel: The Continuing Societal and Technical Challenges. Committee on Disposition of High-Level Radioactive Waste through Geological Isolation, Board on Radioactive Waste Management (2001): p. 55. 50 Dorfman, P. High Burn-up Radioactive Spent Fuel. Nuclear Consultation Group, Helsinki, Finland (8 May 2009). 51 Alexander, W.R., and L.E. McKinley. Deep Geological Disposal of Radioactive Waste. Elsevier Ltd., Amsterdam, The Netherlands. (2007): p. 184.

In addition to scientific uncertainty, public opposition has led to a moratorium on the underground disposal of used nuclear fuel in The Netherlands and in Spain. In Sweden all used nuclear fuel is stored above ground in a central interim storage facility,52 but there appears to be movement toward siting a waste repository near two of the countrys reactor sites.53 In The Netherlands, extensive studies on deep geological repositories in salt domes conducted over several decades were shut down due to growing public opposition to the deep geological disposal of used nuclear fuel. Now, the Dutch government is focusing on retrievable, longterm storage of radioactive wastes.54 In Spain, all field work pertaining to the deep geological disposal of radioactive waste has been stopped due to opposition from the public. Subsequently, Spain has not been able to select a host rock formation or set a target date for any potential deep geological repository.55 Similar circumstances exist in the United States as well. For over 20 years, the US has been unable to designate an acceptable site for the deep geological disposal of used nuclear fuel and high-level radioactive waste. Yucca Mountain was originally designated such a site by the US Congress in 1987, and after much delay due to strong public and political opposition, a license application was finally submitted in 2008. Approximately 23 years and US$38 billion later, the current Obama Administration declared that Yucca Mountain is not a workable option for a nuclear waste repository56 and that the US needs to be looking at other alternatives.57 US Energy Secretary Steven Chu submitted a motion to the Nuclear Regulatory Commission to withdraw the Yucca Mountain proposal with prejudice. Currently, the US has no long-term management plan for its used nuclear fuel and high-level radioactive wastes other than storing such wastes at reactor sites. In addition to the predisposed uncertainty, deep geological repositories have been shown to undergo severe accidents and miscalculations. For example, a German geological repository which was converted from a temporary storage facility in 1965 started leaking radioactive brine in 1988 and the accident was not discovered until June 2008.58 Due to the high number of tunnel constructions and decades of use, this repository is now losing its stability. The capping mass in the pit is currently shifting 15 centimeters (cm) a year and this shifting may soon lead to an uncontrollable increase in water inflow into the dry waste repository.59 A second German deep geological repository has deteriorated to such an extent that its risk of collapse is almost certain. Over 4 million cubic meters of salt-concrete have been pumped into the repository in an attempt to temporarily stabilize it.60 Past nuclear waste management experience in the US also provides little assurance about safety: all US nuclear wastes stored away from reactors, to date, have eventually leakedas at Oak Ridge, Tennessee; Maxey Flats, Kentucky; and Hanford, Washington.61 The federal and Ontario governments have endorsed deep geological disposal as the solution for Canadas used nuclear fuel waste since the mid-1970s. The technical studies used to support the safety of the method were developed by Crown corporationsAECL and Ontario Hydroas directed by their respective government shareholders. However, the public opposition, in part motivated by mistrust of the industrys case concerning the safety, prevented implementation of the deep geological disposal concept. Despite the controversies, it has been observed that the NWMOs study did not question the technical scenario produced by AECL and Ontario Hydro. NWMO, instead, established the social acceptability of deep geological disposal as its primary task.62 Yet, it failed to address the root criticism behind the publics opposition, which is the overall mistrust of the safety case developed by vested interests such as AECL and Ontario Hydro. The objectivity of the safety case for deep geological disposal is undermined by OPG and AECLs historic control over financing and directing research. This reality led the Seaborn Panel to conclude that any technical case must also address the issues of social safety. A Seaborn Panel member has observed that the federal government, in response to the panels recommendations, minimized the technical problems with deep geological disposal.63

16

52 Eureka County Yucca Mountain Information Office. The Nuclear Waste Dilemma: An International Perspective. Nuclear waste update newsletter. Eureka County, NV (Fall 2001). 53 World Nuclear Association. Nuclear Power in Sweden. <http://www.world-nuclear.org/info/inf42.html> (March 2010). 54 Alexander, W.R., and L.E. McKinley. Deep Geological Disposal of Radioactive Waste. Elsevier Ltd.: Amsterdam, The Netherlands (2007): p. 184. 55 Ibid., p. 185. 56 Mascaro, L. Yucca Mountains Death Just a Few Steps Away. Las Vegas Sun. Las Vegas, NV (2 February 2010). 57 Mascaro, L. Obama Administration: Were Done With Yucca. Las Vegas Sun. Las Vegas, NV (29 January 2010). 58 Deutsche Welle. Problems at Germanys Asse II Nuclear Waste Repository. 20 August 2008. 59 Institut fur Gebirgsmechanik GmbH. Mountain-mechanical State Analysis of the Shaft Arrangement System of Asse II. Short Report. Leipzig, Germany (11 September 2007). 60 Federal Ministry for the Environment, Nature Conservation and Nuclear Safety. Safety Requirements Governing the Final Disposal of HeatGenerating Radioactive Waste. Bonn, Germany (July 2009). 61 Shrader-Frechette, K. Risk and Uncertainty in Nuclear Waste Management. NWMO (November 2003). 62 Durant, D. Managing Expertise: Performers, Principals, and Problems in Canadian Nuclear Waste Management. Science and Public Policy (April 2006). 63 Wilson, L. M. Nuclear Waste: Exploring the Ethical Dilemmas. Toronto: United Church Publishing House (2000).

4. Impacts of Expanded Nuclear Waste Production

The NWMOs APM, which recommended that Canadas used fuel waste be centralized at one site and eventually disposed of in a deep geological repository, was derived based on two key assumptions: first, that waste production would be limited to the current generation of Canadas existing reactors; and second, that waste would be produced by natural uraniumfueled CANDU reactors. A change in these assumed conditions would justify re-examining conclusions. It would also call into question the legitimacy of public consultations based on these assumptions. The NWMOs independent advisory board emphasized this fact in September 2005, stating: The NWMO Study Report provides a framework for proceeding to address existing and expected used nuclear fuel from the current fleet of reactors. However we emphasize, as did many of the participants in the engagement process, that it does not provide a green light for expansions of nuclear power production beyond the lifespan of the current fleet. As we said in Section 1, any significant change in the amount or type of used fuel to be managed should trigger a review of the work undertaken by the NWMO to date.64 Following the submission of the NWMOs recommendation to the federal government in November 2005, the Ontario government directed OPG to begin the planning for building additional reactors at the Darlington site in June 2006. The NWMO recommended centralizing used nuclear fuel at one site in order to reduce the overall risks posed to Canadian society, saying, for example, it limits exposure to hazards.65 The NWMO also stated that, with centralization, transportation risks are limited in time duration.66 Simply put, centralized used nuclear fuel contained at one site poses fewer risks than having waste at multiple sites. The justification for waste centralization in NWMOs Adaptive Phased Management is lost, however, if production of used nuclear fuel waste is extended. Arguably, centralizing fuel storage in such a situation would increase the overall hazards to society by increasing production sites and initiating the open-ended transportation risks. The NWMOs Adaptive Phase Management assumes that 3.6 million used CANDU fuel bundles will be produced by Canadas existing CANDU reactor fleet by the year 2035, at which time a centralized repository will have been constructed and begun accepting used nuclear fuel waste shipments. All the estimated hazards to society are strictly based on this definitively projected inventory. Extending production of used nuclear fuel in Canada, especially used nuclear fuel that is much more hazardous than that of CANDU reactor wastes, will greatly increase the hazards imposed on Canadian society as there can only be an inevitable increase in the number of production sites and used fuel shipments sent to the repository.

4.1 Proposals for Generation III Reactors

Most nuclear power reactors operating in the world today are categorized as Generation II designs, which were built up until the late 1990s. All of Canadas reactors were ordered between 1964 and 1974 and fall into the Generation II category. All reactors being proposed for construction at the Darlington site in Canada are categorized as Generation III designs, which are promoted as having incremental improvements in safety systems and costs compared to Generation II designs. Canadas Generation II CANDU reactor designs, however, are distinguished from most other reactor designs by the use of natural uranium fuel and of heavy water as a neutron moderator. Internationally, the majority of operating reactors use enriched fuel and light, or natural, water as a moderator.
17

64 Crombie, D. (Advisory Council Chair). Letter to Ken Nash (NWMO Board of Directors Chair). Advisory Council to the NWMO. (September 2005.) 65 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 169. 66 Ibid.: p. 175.

Additions to worldwide nuclear capacity peaked in 1985. The principal cause of the decline in reactor construction was the higher-than-estimated construction costs of Generation II reactor designs. To address this constraint to growth, nuclear vendors have sought to significantly reduce the construction and operation costs in Generation III reactor designs. (Nuclear vendors often refer to certain reactor designs as Generation III+. With few Generation III reactors in operation worldwide and no generally-accepted classification system, this report will refer to all next-generation reactors as Generation III.) In 2007, OPG identified nine reactor designs that it would consider at Darlington, including both AECLs untested Generation III Advanced CANDU Reactor (ACR-1000) and the Enhanced Generation II CANDU-6; five pressurized-water reactors (PWRs); and two boiling-water reactors (BWRs).67 In March 2008, the OPGs shareholder, the Ontario government, narrowed the designs under consideration, eliminating the Enhanced CANDU-6, three pressurized-water reactors (APWR, OPR 1000, and APR 1400), and one boiling-water reactor (ABWR).68 GE-Hitachi subsequently withdrew its Economic Simplified Boiling Water Reactor (ESBWR) design from procurement process.69 This left three Generation III reactor designs under consideration: AECLs ACR-1000, Arevas USEPR, and Westinghouses AP-1000. None of these designs is currently operating anywhere in the world. Unlike all power reactors currently operating in Canada, all of the designs use some form of enriched fuel. The use of enriched fuel reactor designs will be a significant change to Canadas nuclear fuel industry. From a historical perspective, the CANDUs use of natural uranium fuel was promoted as an attribute. With Canadas abundant supplies of uranium, a reactor design independent of expensive foreign uranium enrichment facilities was promoted as a means of building Canadas energy independence. Natural uranium is 99.28-percent uranium-238 (U-238) and 0.72-percent uranium-235 (U-235). Enriched uranium is a type of uranium in which the percent composition of uranium-235 has been increased through the process of isotope separation. In this process, natural uranium is concentrated in U-235 isotopes by separating U-238 from U-235 and removing the U-238 isotopes. Since U-235 is the only uranium isotope that is fissile with thermal neutrons (in small amounts), enriched uranium having a higher-percent composition of U-235 will sustain longer nuclear fission reactions than natural uranium. To sustain a nuclear reaction in a reactor, neutrons must be slowed down to a speed at which they are more likely to split atoms. This, in turn, releases more neutrons that will split atoms, and so on and so forth. The slowing of neutrons requires the use of a neutron moderator which absorbs some of the neutrons kinetic energy, slowing them down to a sustainable speed. Water is an excellent neutron moderator and thus used in most nuclear reactors. However, water is also an excellent neutron absorber and will absorb such a great amount of neutrons that there will be too few left to react with the small amount of U-235 in natural uranium-fuel reactors. Thus, CANDU reactors use heavy water as a moderator, since heavy water already contains an extra neutron and this reduces the number of neutrons the water can absorb. Using heavy water, however, adds a significant cost to CANDU operation. Alternatively, enriched uraniumfueled reactors can use light, or natural, water as a neutron moderator. Enriched uranium fuel has a higher burn-up than natural uranium fuel. Burn-up is the measure of fission events that occur in a fuel source, expressed in units of megawatt-days per metric ton of uranium (MWD/t-U). This means that a given amount of fuel with a higher burn-up will undergo a greater number of fission reactions than the same amount of fuel with a lower burn-up, and therefore generate a greater amount of electricity.

18

67 OPG. Project Description for the Site Preparation, Construction and Operation of the Darlington B Nuclear Generating Station, Environmental Assessment. Toronto: OPG (12 April 2007). 68 Infrastructure Ontario. Request for Proposals: Nuclear Procurement Project. RFP No. OIPC 08-00-1027 (7 March 2008). 69 Hamilton, T. GE-Hitachi Wont Bid for Reactor. Toronto Star (9 April 2008).

For example, after generating a net power of approximately 689 MW of electricity, a single Generation II CANDU reactor will have produced approximately 86,400 kilograms (kg) of used fuel per year. After generating the same net power, a single Generation III ACR-1000 reactor that uses enriched uranium with a higher fuel burn-up will have only produced approximately 33,000 kg of used fuel per year, a little more than one-third the used fuel production of a CANDU reactor. Furthermore, a single AP1000 reactor will only produce approximately 15,170 kg of used fuel per year (approximately one-fifth of a CANDUs used fuel production), and a single EPR reactor will only generate approximately 12,490 kg of used fuel per year (approximately one-seventh of a CANDUs used fuel production).70 The nuclear fuel requirements for these Generation III reactors are much smaller than that of a Generation II CANDU reactor because these new reactor types use enriched uranium as opposed to the natural uranium which is currently used by all CANDU reactors. Enriched uranium contains a greater amount of uranium-235 (U-235), the only fissile isotope of uranium in appreciable amounts. One approach to improving the economics of Generation III reactor designs has been to extract more energy from the nuclear fuel in order to maximize electricity generation. The Generation III CANDU design (the ACR-1000) seeks to improve its economics by abandoning a principal design characteristic of the CANDU, the use of natural uranium, and switching to slightly enriched uranium (SEU). Light-water designs, which already use enriched fuel, seek to increase fuel burn-up and energy extraction by increasing fuel enrichment and by a longer period of fuel use during operation. These moves to improve the economics of Generation III reactor designs increase reactor operational risks and the hazards and challenges of fuel waste storage. These hazards are described in much greater detail in Section 5.0 of this report. The following section describes the fuel cycles of the reactor designs currently under consideration at the Darlington site in Canada.

4.1.1 AtOMIC ENERGy OF CANADA LIMItED ACR-1000

The Advanced CANDU Reactor (ACR-1000) is AECLs proposed Generation III reactor design. The ACR-1000 is a light-water-cooled nuclear power reactor, with a 60-year design life, that uses heavy-water for neutron-moderating. This reactor type is similar in design to the original CANDU reactor but utilizes lightly enriched uranium fuel having up to a 2.5-percent enrichment of U-235. Since the uranium fuel used in this reactor is slightly enriched, the reactor incorporates a light-water coolant system with a separate heavy-water neutron moderator. The ACR-1000 reactor uses CanFlex ACR fuel bundles, a variation of the CanFlex LVRF bundle. These fuel bundles are 10.3 cm in diameter, 0.5 meters in length and contain a total of 43 fuel elements, amounting to 16.2 kg of uranium. Similarly to the CANDU fuel bundles, the uranium in these fuel bundles is in the form of high-density uranium dioxide fuel pellets sealed inside zirconium alloy tubes. Also like in the CANDU reactor, CanFlex ACR fuel bundles can be removed and replaced in the ACR1000 reactor without reactor shutdown. The average burn-up of used nuclear fuel in an ACR-1000 reactor is 20,000 MWD/t-U, as opposed to the average 8,300 MWD/t-U in a CANDU reactor. The ACR-1000 is designed to produce a net power of 1,085 MW of electricity. The total expected number of used fuel bundles produced over the lifetime of a single ACR-1000 reactor is 192,600.71

19

70 Garamszeghy, M. Nuclear Fuel Waste Projections in Canada2009 Update. NWMO TR-2009-30. Nuclear Waste Management Organization, Toronto, Ontario (2009). 71 Garamszeghy, M. Nuclear Fuel Waste Projections in Canada2009 Update. TR-2009-30. Nuclear Waste Management Organization, Toronto, Ontario NWMO (2009).

4.1.2 WEStINGHOuSES AP1000

The AP1000 reactor is designed by Westinghouse Electric Company and is the first Generation III reactor to receive final design approval from the Nuclear Regulatory Commission.72 This reactor is a light-water reactor, specifically a pressurized-water reactor (PWR). In this reactor, water coolant is pumped, under high pressure, to the nuclear reactor core, becoming superheated. The heated water then transfers thermal energy to a steam turbine that generates electricity. Light water is used as both the reactor coolant and the neutron moderator. The nuclear fuel used in the AP1000 reactor is enriched uranium fuel. The initial core fuel load is between 2.4-percent and 4.5-percent enriched, and each subsequent fuel reload has an average enrichment of approximately 4.8 percent. The uranium fuel of this reactor is be contained in fuel bundles, but in a conventional PWR fuel assembly. PWR fuel assemblies are much larger than CANDU fuel bundles and are rectangular in shape. The AP1000 PWR fuel assemblies have a square diameter of 21.4 cm, a length of 4.8 meters, and contain 538.3 kg of enriched uranium. The enriched uranium fuel contained within the AP1000 PWR fuel assemblies is in the form of high-density uranium dioxide fuel pellets, which are sealed in Zirlo (a recently developed alloy of zirconium) fuel rods. The PWR fuel assemblies specific to the AP1000 reactor will have a 17 x 17 fuel rod matrix structure, consisting of a total 264 fuel rods, 24 control rods, and 1 instrumentation tube. In order to refuel the AP1000, the reactor must be shut down every 12 to 24 months and a portion of the core, approximately one third, must be replaced.73 The AP1000 reactor is planned to produce a net power of 1,090 MW of electricity. The average burn-up of the AP1000 used nuclear fuel assemblies is 60,000 MWD/t-U. The total number of fuel assemblies expected to be produced over the lifetime of a single AP1000 reactor is 2,700.

4.1.3 AREvAS EPR

The European Pressurized Reactor, generically known internationally as an evolutionary power reactor (EPR), is a Generation III pressurized water reactor (PWR). This reactor operates in the same fashion described above for the AP1000 reactor. The nuclear fuel used in the EPR is up to 5-percent enriched uranium. The structures of the uranium fuel pellets and the PWR fuel assembly for this reactor are almost identical to those described above for the AP1000 reactor, except that the EPR fuel assemblies incorporate M5 cladding material instead of Zirlo and contain 527.5 kg of uranium instead of 538.3 kg. Similarly to the AP1000 reactor, the EPR requires that the reactor be shut down every 12 to 24 months for refueling.74 It is estimated that the EPR will produce a net power of 1,600 MW of electricity. The total burn-up of an EPR PWR fuel assembly is 62,000 MWD/t-U. The total number of fuel assemblies expected to be produced over the lifetime of a single EPR reactor is 3,300. Three Generation III EPR reactors will be constructed at the Darlington site if this reactor design is selected by OPG.75 Table 3 displays a summary of the new nuclear reactor types proposed to be introduced to Canada.

20
72 Westinghouse. AP1000 Public Safety and Licensing. (13 September 2004). 73 Garamszeghy, M. Nuclear Fuel Waste Projections in Canada2009 Update. NWMO TR-2009-30. Nuclear Waste Management Organization, Toronto, Ontario (2009). Westinghouse. 2004. AP1000 Public Safety and Licensing. (13 September 2004.) <http://www.westinghousenuclear.com/ AP1000/public_safety_licensing.shtm> Accessed March 2010. 74 Garamszeghy, M. Nuclear Fuel Waste Projections in Canada2009 Update. NWMO TR-2009-30. Nuclear Waste Management Organization, Toronto, Ontario (2009). 75 OPG. Environmental Impact Statement for the New Nuclear at Darlington Project. Toronto, Ontario (30 September 2009): pp. 245.

Table 3. Summary of Existing CANDU (Generation II) Reactor and Generation III Reactor Types Proposed for the Darlington New Build

CANDU

ACR-1000 Heavy water moderated, light watercooled 1,085

AP1000

EPR

Reactor Type

Pressurized heavywater reactor

Pressurized lightwater reactor

Pressurized lightwater reactor

Net Power (MW(e))

689

1,090

1,600

Fuel Type

CANDU fuel bundle (4 variations)

CanFlex ACR fuel bundle

17 x 17 PWR fuel assembly Refueling shutdown every 12 to 24 months; portion of core replaced 2.44.5% avg. for initial core; 4.8% avg. for reloads 60,000

17 x 17 PWR fuel assembly Refueling shutdown every 12 to 24 months; portion of core replaced

Fueling Method

Refuel without shutdown; entire bundle replaced

Refuel without shutdown; entire bundle replaced

Fuel Enrichment Fuel Burn-up (MWD/t-U) Fuel Assembly Mass (kg U)

0.72%

2.5%

Up to 5.0%

8,300

20,000

62,000

19.2 (average)

16.2

538.3

527.5

21

5. New Reactor Designs: Increased Waste Hazards

The proposed shift to Generation III reactor designs which use enriched reactor fuel with an increased fuel burn-up will lead to the production of used nuclear fuel that is much more hazardous than the fuel generated by Generation II reactors. In its environmental assessment of the construction of new reactors at Darlington, OPG did not analyze the hazards associated with the transportation and long-term management of Generation III used nuclear fuel at a centralized repository, as it assumes that any off-site facility will have been subject to its own approval process.76 Since the NWMOs APM for a centralized deep geological repository is limited to the projected used nuclear fuel inventory of Generation II reactors in Canada, the addition of Generation III type reactor wastes to the repository falls outside the scope of the NWMOs APM. If OPGs new reactor builds continue as planned at the Darlington site, then Generation III reactors, which have not yet been tested or operated anywhere in the world, will be allowed to operate in Canada without any analysis of the hazards and consequences of their used nuclear fuel waste. The addition of the Generation III used nuclear fuel with its higher burn-up will impose increased environmental and human health risks at every stage of the APM process, from the operation of the nuclear reactors to the long-term monitoring of the used nuclear fuel disposed of in the deep geological repository. Generation III reactor used nuclear fuel will not only affect the cost of construction and the size of the deep geological repository, but also the interim-storage and the transportation of the used nuclear fuel. Since nuclear fuel with higher burn-up contains a greater inventory of radionuclides, the health hazards imposed on the workers handling it, as well as on members of the public who reside along its shipping routes or within the vicinity of the deep geological repository, will be greatly affected also. In order to compare the hazards of higher burn-up Generation III reactor fuel to those of fuel generated by CANDU reactors, the electricity generationexpressed as megawatts electric: MW(e)of the ACR-1000, AP1000, and EPR Generation III reactors were normalized to average electricity generation of current CANDU (Generation II) reactors in Canada. In this way, it is fair to compare the hazards of the Generation III reactors to the hazards of the existing reactors. Below, Table 4 displays the various parameters of the proposed new reactors, in terms of the average net electricity generation (689 MW(e)) of Canadas CANDU reactors.

Table 4. Summary of the Fuel Types for Proposed Generation III Reactors, Normalized to the Average Net Electricity Generation of a CANDU Reactor*
CANDU Fuel Burn-up (MWD/t-U)** Annual Used Fuel Production (t-U/year per reactor) Annual Used Fuel Production (# of fuel assemblies per year per reactor) Lifetime Used Fuel Production (t-U per reactor) Lifetime Used Fuel Production (# of assemblies per reactor) 8,300 86.4 4,500 3,456 180,000 ACR 1000 20,000 33.0 2,038.4 1,981 122,305 AP1000 60,000 15.2 28.4 920 1,707 EPR 62,000 12.5 23.7 749 1,421

22

*689 megawatts electric. **MWD/t-U: megawatt-days per ton of uranium.

76 OPG. Environmental Impact Statement for the New Nuclear at Darlington Project. Toronto: Ontario. (30 September 2009): pp. 2-16.

As seen in the table above, CANDU reactors require a much greater amount of uranium to generate the same net power in electricity as Generation III reactors. However, since Generation III reactors use enriched uranium fuel with a higher burn-up than that of natural uranium fuel, the inventories of longlived, hazardous radionuclides will be much greater. The following sections describe the increased hazards and consequences of the enriched, higher burnup nuclear fuel during the operation of Generation III reactors, the interim storage of used fuel at reactor sites, the transportation of used Generation III reactor fuel to a centralized deep geological repository, and long-term disposal of Canadas used nuclear fuel.

5.1 Operation
Using fuel with higher burn-up will result in the addition of hazards during operation of the nuclear power reactors. First of all, the increased lifetime of the fuel will contribute to the risk of reactor accidents. The fuel cladding in a reactor using fuel with high burn-up will become thinner over time, thus increasing the risk of radioactive release.77 Likewise, deformation of the fuel rods due to the higher burn-up may cause the control rods to stick and hinder the flow of cooling water. Such potential risk of cooling failure would require more advanced emergency cooling.78 Not only does the chance of a reactor accident become more likely with high burn-up fuel, but if a severe accident occurs, such as in an EPR reactor using fuel with 65,000 MWd/t-U burn-up, it would cause double as many cancer cases as an accident in the largest present-day reactors.79 Cladding failure, due to burn-up-dependent effects, may also occur during transfer and handling operations, imposing a much greater threat on nuclear workers.80 Currently, there is very limited data to show that the cladding of used nuclear fuel with burn-ups greater than 45,000 MWD/t-U will remain undamaged during the licensing period.81 The design of fuel cladding capable of staying in the reactor for long periods of time does not consider long-term consequences.82 The radionuclides of greatest concern during the operation of a nuclear reactor are those that are considered fission products, such as cesium-137, iodine-129 and -131, and strontium-90. These radionuclides are produced as a result of fission reactions that occur when a large nucleus, like that of U-235, is bombarded with neutrons and splits into two smaller nuclei, or fission products. Isotopes of cesium and iodine were the primary radionuclides released during the Chernobyl accident in 1986. Fission products are considered such hazards because many of them, particularly cesium-137, have the ability to volatilize and be released into the environment in gaseous form. As discussed in Section 3, fission reactions are what sustain the operation of a nuclear power reactor. The greater the number of fission reactions that occur in a reactor, the greater the amount of electricity produced. If two reactors, regardless of their design, are to generate the same amount of electricity, then the number of fission reactions that occur within those reactors will be the same. This is to say that the number of fission reactions that occur within a CANDU reactor generating an annual net average of 689 MW(e) will be the same as in an ACR-1000, AP1000, or EPR Generation III reactor generating the same annual net amount of electricity. The only difference is that the Generation III reactors will require a much lower quantity of fuel to generate this same amount of electricity. However, Canada plans to extract more energy from the nuclear fuel used in proposed Generation III reactors in order to maximize electricity generation. As opposed to existing CANDU reactors, which generate an average net power of 689 MW(e) per reactor, ACR-1000, AP1000 and EPR Generation III reactors will generate net powers of 1,085, 1,090, and 1,600 MW(e), respectively, per reactor.83 This can only be accomplished by the generation of more fission products in Generation III reactors than in currently existing CANDU reactors. Therefore, if a nuclear accident or malfunction were to occur at a Generation III reactor running at maximum power, the severity of the consequences imposed on the workers operating that reactor will be greatly increased.
23
77 IAEA. Selection of Away-from-Reactor Facilities for Spent Fuel Storage: A Guidebook. TECHDOC-1558. (2007.) 78 Schmitz F. & Papin J.High Burn-up Effects on Fuel Behaviour under Accident Conditions. Journal of Nuclear Materials. Vol. 270 (1999): pp.55-64. 79 Large and Associates. Assessments of the Radiological Consequences of Releases From Existing and Proposed ERP/PWR Nuclear Power Plants in France. (2007.) 80 Fairlie, I. Estimated Radionuclide Releases and Collective Doses from the Rokkasho Reprocessing Facility. Greenpeace Japan (2008). 81 United States Nuclear Regulatory Commission. Standard Review Plan for Spent Fuel Dry Storage Facilities, Final Report. NUREG-1567. (2000.) 82 IAEA. Optimization Strategies for Cask Design and Container Loading in Long Term Spent Fuel Storage. TECDOC-1523 (2006). 83 Garamszeghy, M. Nuclear Fuel Waste Projections in Canada2009 Update. Nuclear Waste Management Organization, Toronto, Ontario. NWMO TR-2009-30 (2009).

5.2 Interim-Storage at Reactor Sites

The higher radioactivity of enriched fuel would dramatically affect the engineering requirements for interim fuel waste storage facilities. To mitigate increased transportation risks, interim storage for used fuel may need to be extended. According to the NWMOs APM, all CANDU used nuclear fuel will be stored at licensed storage facilities at the reactor sites for approximately 30 years before being transported to the centralized deep geological repository. Obviously, the used nuclear fuel currently in storage at the reactor sites will have been cooled much longer than 30 years. Once used nuclear fuel is removed from a CANDU reactor, it is stored in wet storage for approximately seven to ten years.84 Used nuclear fuel is the most radioactive immediately after it is removed from the reactor, and water provides the greatest shielding from radioactivity and decay heat released from the used nuclear fuel. After 10 years, the radioactivity of CANDU used nuclear fuel is assumed to decrease to approximately 0.1 percent of its initial radioactivity.85 Once the radioactivity and decay heat of the used nuclear fuel have significantly decreased, the used fuel is transferred to dry storage containers, where it will remain until a centralized deep geological repository is constructed and begins operating. According to the NWMOs Final Study, the radioactivity of CANDU used nuclear fuel will decrease to approximately 0.03 percent of its initial radioactivity after 30 years, and to 0.01 percent of its initial radioactivity after 100 years.86 The radionuclide inventory of used nuclear fuel can primarily be broken down in terms of fission products and actinides. As discussed earlier, fission products are those radionuclides that are created as a result of a fission reaction. Actinides are radionuclides with atomic numbers of 90 to 103 and include uranium, plutonium, americium, curium and neptunium. These radionuclides are very longlived, with half-lives of up to 15,600,000 years (curium-247). NWMO compares the hazards of CANDU fuel to the hazards of uranium over time,87 but the hazards of plutonium, curium, americium, and neptunium are much greater. The number of fission reactions that occur in a nuclear reactor dictates the amount of energy produced by that reactor. Since this analysis assumes that CANDU and Generation III reactors will produce the same amount of net energy, the amount of fission reactions that occur in each reactor for the same energy output will be the same for all reactor types. Therefore, the focus of this analysis is on the amount of actinides produced by each of the Generation III reactor types in comparison to a CANDU reactor that generates the same amount of electricity. Workers at nuclear reactors are responsible for directly handling this fuel at two different stages, first when the fuel is transferred from wet storage to dry storage containers and second when the fuel is transferred from dry storage containers to transportation casks. Thus, the interactions between workers and the used nuclear fuel most likely occur first when the fuel has been cooled for approximately 10 years and then again at 30 years. Generation II CANDU used nuclear fuel bundles have a total actinide radioactivity of 1.05E+13 Bq (becquerels; E = 1018) after being cooled for 10 years, and a total actinide radioactivity of 4.18E+12 Bq after being cooled for 30 years.88 Any exposures to unshielded CANDU fuel that has been cooled for less than 50 years will give a radiation dose rate of 1,150 mSv (millisieverts) per hour at a distance of 0.3 meters. This is enough to provide a fatal dose of 5 Sv (sieverts) of radiation to a nuclear worker after only four hours of exposure89.

24

84 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 22. 85 Ibid.: p. 341. 86 Ibid. 87 Ibid. 88 Tait, J.C., I.C. Gauld, and G.B. Wilkin. Derivation of Initial Radionuclide Inventories for the Safety Assessment of the Disposal of Used CANDU Fuel. Report No. AECL-9881. Atomic Energy of Canada Limited, Whiteshell Nuclear Research Establishment, Pinawa, Manitoba (August 1989). 89 Nuclear Waste Management Organization (NWMO). Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 344.

Although similar in design to a CANDU reactor, a Generation III Advanced CANDU Reactor (ACR1000) uses enriched fuel with a burn-up of 20,000 MWD/t-U. Normalized to the electrical power generation of a single CANDU reactor, the actinide radioactivity of a used fuel bundle is 2.40E+13 Bq after it is cooled for 10 years and 1.02E+13 Bq after it is cooled for 30 years.90 Thus, the radioactivity of an ACR-1000 fuel bundle is approximately 2.3 times greater than that of a CANDU fuel bundle after being cooled for 10 years and 2.4 times greater than that of a CANDU fuel bundle after being cooled for 30 years. A nuclear worker handling a used ACR-1000 fuel bundle could thus receive a fatal dose of 5 Sv of radiation in as little as two hours instead of four hours. Assuming the AP1000 reactor operated at the same power level as a CANDU reactor, the total actinide radioactivity of an AP1000 fuel assembly would be 3.42E+15 Bq when it is first removed from the reactor, 1.55E+15 Bq after it is cooled for 10 years, and 6.60E+14 Bq after it is cooled for 30 years.91 The radioactivity of an AP1000 PWR fuel assembly is approximately 148 times that of a CANDU fuel bundle after being cooled for 10 years and 158 times that of a CANDU fuel bundle after being cooled for 30 years. A nuclear worker handling a used AP1000 PWR reactor fuel assembly could thus receive a fatal dose of 5 Sv of radiation in as little as 1.5 minutes instead of four hours. Similarly, if an EPR reactor operated at the same power level as a CANDU reactor, the total actinide radioactivity of a PWR fuel assembly would be 2.33E+15 Bq when it is first removed from the reactor, 1.06E+15 Bq after it is cooled for 10 years, and 4.5E+14 Bq after it is cooled for 30 years.92 The radioactivity of an EPR PWR fuel assembly is approximately 101 times that of a CANDU fuel bundle after being cooled for 10 years and 108 times that of a CANDU fuel bundle after being cooled for 30 years. A nuclear worker handling a EPR PWR fuel assembly could thus receive a fatal dose of 5 Sv of radiation in as little as 2.25 minutes instead of four hours. In summary, the radioactivity of long-lived actinides in used nuclear fuel from the new Generation III reactors will be between 2 and 158 times greater than that of CANDU used fuel at the approximate time of disposal at the repository. In addition, it would take approximately 50, 100,000, or 20,000 years, for an ACR-1000, AP1000, or EPR fuel assembly, respectively, to cool to the radioactivity of a CANDU used nuclear fuel bundle that has been cooled for 30 years. Currently, the life expectancy of all short-term storage containers used in Canada is 50 years, with a maximum life expectancy of 100 years.93 OPGs environmental assessment for the New Nuclear Project at Darlington accounts for the need for different interim storage containers depending on the type of Generation III reactor that is selected for the Darlington Site.94 However, as demonstrated in this section, the radioactivity of long-lived actinides in of Generation III used fuel at 10 and 30 years cooled is much greater than that of CANDU used fuel at those same intervals. Therefore, the used Generation III fuel stored on-site, above ground, will be much more radioactive than used CANDU fuel currently stored on-site. The addition to interim nuclear fuel waste storage sites of used fuel that is high burn-up and more radioactive will greatly increase the potential consequences from a sabotage attack or other malevolent act against the site. The consequences of malevolent acts against Generation III used nuclear fuel storage containers were not specifically analyzed by OPG in its Environmental Impact Statement for the New Nuclear Project at Darlington,95 nor were they considered by the NWMO in its Adaptive Phase Management final study. If the New Nuclear Project at Darlington continues as proposed, the consequences that could be suffered by members of the Canadian public residing in the vicinity of the site, as a result of an act of sabotage against a Generation III used fuel storage container, will not have been assessed before the construction of the Generation III reactors begins there.

25
90 RISKIND, Version 2.0. Argonne National Laboratory. S.Y. Chen and Bruce M. Biwer, contacts (July 2002). The radionuclide inventories of all Generation III reactor used nuclear fuel were derived using the spent nuclear fuel database of the RISKIND computer program, which was developed by Oak Ridge National Laboratory for the US Department of Energy and the Office of Civilian Radioactive Waste Management. 91 Ibid. 92 Ibid. 93 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 22. 94 OPG. Environmental Impact Statement for the New Nuclear at Darlington Project. Toronto, Ontario. (30 September 2009): pp. 262. 95 Ibid.: Ch. 2, pp. 5-1 to 5-2.

5.3 Transportation
The extended production of radioactive fuel waste will cause an increase in transportation risks beyond those envisioned in the NWMO study. Transporting to the repository fuel that is much hotter than CANDU fuel will greatly increase the consequences of a potential accident or sabotage attack for members of the public who reside along used nuclear fuel transportation routes.

5.3.1 NuMbER OF SHIPMENtS

NWMOs Final Study for the Adaptive Phase Management plan assumes 3.6 million CANDU used fuel bundles will require transport to the centralized deep geological repository. The NWMO estimated that it would take approximately 636 road shipments per year, or 60 rail shipments and 432 road shipments per year, or 24 water shipments and 432 road shipments per year, for a total of 30 years, to transport this volume of fuel. This would result in maximums of 19,080 road shipments, or 1,800 rail shipments plus 12,960 road shipments, or 720 water shipments plus 12,960 road shipments of CANDU used nuclear fuel bundles over 30 years.96 Two different shipping containers will be used to transport CANDU used nuclear fuel bundles from interim storage at the reactor sites to the centralized repository. Irradiated fuel transportation casks for baskets or modules (IFTC/BM) may be used to transport used fuel by road, rail, and water. Shipments by road will consist of one IFTC/BM per shipment, by rail 10 IFTC/BM per shipment (2 casks per railcar) and by water 32 IFTC/BM per shipment. The maximum allowable number of used CANDU fuel bundles per IFTC/BM is 192.97 Dry storage container transportation packages (DSCTPs) may be used to transport used fuel by rail and water only. These containers are much larger than IFTC/BM and are capable of holding up to 384 used CANDU fuel bundles. Each DSCTP by water shipment will eventually be broken down and transferred to two IFTC/BM and transported the remainder of the way to the centralized repository via road shipment.98 The shipping containers that will be used to transport Generation III used nuclear fuel assemblies are different in design from IFTC/BM and DSCTPs. In the United States, transportation casks such as the GA-4 and GA-9 truck casks99 and the Holtec HI-STAR rail casks are used to transport PWR and BWR used nuclear fuel assemblies. The GA-4 truck cask is designed to transport four PWR fuel assemblies per cask and the GA-9 truck transports nine BWR fuel assemblies per cask. The Holtec HI-STAR casks can hold approximately 24 PWR fuel assembles per cask and approximately 68 BWR fuel assemblies per cask. However, it is possible for rail casks to transport up to 26, 28 or more PWR fuel assemblies.100 The introduction of Generation III nuclear reactors is projected to extend the production of radioactive waste in Canada to the year 2085 or beyond. Moreover, the introduction of these reactors, which use enriched uranium fuel with higher burn-up, will increase the amount and in addition change the type of used nuclear fuel requiring permanent disposal in Canada. Depending on the type of Generation III reactors selected to be built at the Darlington site, as many as 770,400 CanFlex ACR used fuel bundles or 10,800 PWR used fuel assemblies will be generated and require permanent disposal, if the quantity of energy extracted from the Generation III reactors at Darlington is maximized. If the Generation III reactors operate so that their electricity production is equivalent to that of current CANDU reactors, the amount of used fuel generated by the year 2085 would then be reduced to 489,200 CanFlex ACR used fuel bundles and 6,830 PWR used fuel assemblies. Comparatively, four CANDU reactors operating for 60 years would generate 1,080,000 used CANDU fuel bundles.

26
96 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 25. 97 Cogema, 2003. Logistics of Transportation of Used Fuel to a Centralised Facility. Report of study carried out for Ontario Power Generation, New Brunswick Power, Hydro-Qubec and Atomic Energy of Canada Limited. Report No. 500276-B-009 (24 September 2003): p. 11. 98 Ibid. 99 Civilian Radioactive Waste Management System and TRW Environmental Safety Systems Inc. GA-4/9 Legal Weight Truck Cask Systems Review and Status. Paper presented to the Nuclear Waste Technical Review Board, Arlington, VA, by D.J. Nolan (14 June 1995). <http:// www.nwtrb.gov/meetings/1995/june/nolan.pdf>. 100 Holtec. Topical Safety Analysis Report for the HI-STAR 100 Cask System. Holtec Report No. HI-941184 (1998).

Since less used nuclear fuel will be generated by Generation III reactors compared to that by existing CANDU reactors, and therefore require less transport to the centralized repository, it would then seem advantageous to implement Generation III reactors. This is not the case, however, as the hazards associated with these new enriched uranium fuel assemblies are much greater than those of CANDU natural uranium fuel bundles and thus the consequences of a transportation accident or sabotage event involving Generation III used fuel assemblies will be much greater. It would take approximately 5,625 road shipments to transport 1,080,000 CANDU used fuel bundles from the Darlington site in Ontario to the centralized repository, whereas it would only require 2,550 road shipments (less than half of the CANDU shipments) to transport 489,200 used CanFlex ACR fuel bundles and 1,707 road shipments (less than one-third of the CANDU shipments) to transport 6,830 PWR used fuel assemblies. Risk is measured as the expected frequency of an event multiplied by the severity of the consequences of that event. Although it appears that the risk of transporting used nuclear fuel would decrease as the number of shipments of used fuel decreases (i.e., Generation III reactors instead of CANDU reactors are implemented at the Darlington site), this is not necessarily the case, as the radionuclide inventory of Generation III used fuel assemblies is much greater than that of CANDU used fuel bundles (see Section 5.2). In the event of a transportation accident or sabotage attack, the risk of transporting a large number of CANDU used fuel shipments with small radionuclide inventories (high frequency x low consequences) is essentially the same as that of transporting a lesser number of Generation III used fuel shipments with much greater radionuclide inventories (low frequency x high consequences). Therefore, members of the public who reside along the transportation routes to the centralized repository will not benefit from the shipment of Generation III reactor used nuclear fuel instead of CANDU used nuclear fuel. Depending on the severity of a transportation accident or sabotage event involving Generation III used nuclear fuel, the consequences could be seven times greater than those of an accident involving CANDU used nuclear fuel. Section 5.3.2 discusses the consequences of transportation accidents and sabotage attacks in much greater detail. Although the risk of transporting a large amount of CANDU fuel or a small amount of Generation III used nuclear fuel is essentially the same, the addition of any number of fuel bundles or fuel assemblies that exceeds the total of 3.6 million fuel bundles projected by NWMO in its Adaptive Phase Management final study will inherently increase the overall risk to society. The public was not consulted on these increased risks. In addition, OPG made no mention of the number of shipments, nor the types of shipping containers, required to transport Generation III used nuclear fuel from the Darlington site to a centralized repository, leaving the transportation and long-term management of Generation III used nuclear fuel in the hands of the NWMO. As discussed throughout this report, the NWMOs final study neglects the incorporation of any Generation III reactor used fuel waste, but it recognizes that additional studies, safety assessments, facility designs and technology demonstrations would be required for each waste form.101 No such assessments have been performed by OPG or NWMO.

5.3.2 tRANSPORtAtION ACCIDENtS AND SAbOtAGE AttACkS

With the shipment of used nuclear fuel from interim storage facilities to a centralized repository comes the risk of transportation accidents and sabotage attacks along the used nuclear fuel transportation routes. By rule of the NFWA, the NWMO at this time does not have to select an exact location for the deep geological repository proposed in its Adaptive Phase Management plan. However, NWMO has proposed that the repository be located within a region that benefits from the activity of Canadas nuclear power reactors, as they should have the greatest responsibility in the management of the waste streams generated by these reactors. Currently, there are only four provinces within Canada that benefit from Canadas nuclear power reactors: Ontario, New Brunswick, Quebec, and Saskatchewan.102

27

101 Russell, S. Preliminary Assessment of Potential Technical Implications of Reactor Refurbishment and New Reactor Build on Adaptive Phase Management. Nuclear Waste Management Organization, Toronto, Ontario (December 2008). 102 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 40.

Regardless of the exact chosen location of the proposed deep geological repository, it is inevitable that Generation III used nuclear fuel generated at the Darlington site will have to be transported by road and/or rail through Ontario, Canadas most largely populated province, putting the greatest amount of people at risk in the event of a transportation accident or sabotage event en route to the deep geological repository. In the event of a such an accident or attack, the walls of a shipping container could be penetrated or destroyed and the internal fuel bundles or assemblies breached, resulting in the release of radioactive material from the container. The release of specific radionuclides such as Cesium-137 (Cs-137) and other gamma emitters is of greatest concern, as these radionuclide impose the greatest long-term health risks to society. Cs-137 is a long-lived fission product (half-life = 30.03 years) contained within used nuclear fuel. Cs-137 decays via beta decay into barium-137m (Ba-137m), which is a primary gamma ray emitter. Gamma rays are strong forms of ionizing radiation and do not have a definite range in matter; the intensity of gamma rays can only be weakened by increasing the thickness of the material absorbing the rays. Therefore, gamma rays can cause a great deal of damage to living human tissue.103 The consequences that could result from a transportation accident or sabotage event involving a shipment of used nuclear fuel from a Generation III reactor are much greater than those involving a shipment of used fuel from a CANDU reactor. This is because the Generation III shipments contains up to 2.5 times the amount of Cs-137 (in Bq) of a CANDU road shipment and 7.6 times the Cs-137 of a CANDU rail shipment. The health risks imposed upon members of the public who reside along used nuclear fuel transportation routes would therefore be greatly increased if Generation III fuel is to be generated at the Darlington site in Ontario. Such health risks include increases in expected cancer fatalities, cardiovascular disease, dermal disease, and birth defects in children born to parents exposed to ionizing radiation. OPG has not evaluated the risks of accidents or sabotage events involving transportation of any amount of Generation III reactor used nuclear fuel from new-build reactors at the Darlington site to a centralized deep geological repository in its environmental impact statement.104 Instead, OPG analyzed transportation accidents involving transfer of low-level radioactive waste (LLW) and intermediate-level radioactive waste (ILW), which do not include used nuclear fuel, to off-site disposal facilities. This lack of analysis of a potential transportation accident or sabotage attack on a Generation III used nuclear fuel shipping container can be attributed to OPGs understanding that any off-site nuclear waste repository will have been subject to its own approval process.105 This assumes NWMO would consider such impacts at a later date, after decisions to proceed with new reactors had been approved. As noted, the NWMOs risks assessments and consultations used to develop its Adaptive Phase Management did not account for the transportation of Generation III used nuclear fuel. The risks imposed on Canadian society and the environment by the transportation of Generation III fuel waste materials have not been evaluated. Generation III reactors might remain in operation until the year 2085 and beyond, placing Canadians residing along the transportation routes to a centralized deep geological repository in the immediate range of these unassessed consequences for as long as 90 or more years.

28

103 Cember, H. Introduction to Health Physics. 3rd ed. McGraw-Hill (1996): p. 134. 104 OPG. Environmental Impact Statement for the New Nuclear at Darlington Project. Toronto, Ontario (30 September 2009): Technical Support Documents 13, 27, 28. 105 OPG. Environmental Impact Statement for the New Nuclear at Darlington Project. Toronto, Ontario. (30 September 2009): pp. 216.

5.4 Long-Term Disposal in Geologic Repository

5.4.1 HIGH buRN-uPHIGH HAzARD
Fuel with high burn-up uses enriched fissile uranium and is left in the reactor for a longer period of time, making the spent fuel significantly more radioactive than the conventional spent fuel from CANDU reactors. But exactly how much more hazardous is it? One measure of comparison is to look at the number of years that it would take for spent nuclear fuel disposed of in the repository to cool down through radioactive decay and become equally radioactive to natural uranium ore (2. 47E+04 Bq/g [becquerels per gram]).106 According to the NWMO, CANDU used nuclear fuel will decay to the radioactivity of natural uranium in less than one million years.107 On the other hand, Generation III used nuclear fuel will take up to 2.6 million years to decay to the same level.108 Since this method does not appropriately address the radiological health hazards associated with this waste, another measure of comparison is to consider the amount of water needed to dilute the nuclear fuel deposited in the proposed repository to concentrations that would comply with the Nuclear Regulatory Commissions effluent water standards. The radioactivity of individual actinides remaining in CANDU, ACR-1000, AP1000, and EPR reactor fuel after it has cooled for one million years were computed and normalized to the output of electricity produced per bundle/assembly of that reactor. Then, the total projected amount of bundles and assemblies that will be placed in the repository according to NWMOs plan were considered. The NWMOs final study proposed that about 3.6 millions bundles of spent CANDU fuel will be deposited into the repository. If up to four new Generation III reactors are built at the Darlington site in Ontario, the total spent fuel deposited into the repository by 2085 will include the 3.6 million used CANDU fuel bundles projected by the NWMO, with the addition of up to 489,200 CanFlex ACR used fuel bundles or 6,830 PWR used fuel assemblies, assuming the new reactors generate the same average net electricity as do current CANDU reactors. The total potential used fuel generated from the new Generation III reactor builds at Darlington that will require disposal at the centralized repository is broken down into the following four scenarios. Scenario 1: Generation III reactors are not constructed at the Darlington site and the total amount of used fuel to be disposed of remains at 3.6 million used CANDU fuel bundles, as projected by the NWMO in its final study. Scenario 2: 3.6 million used CANDU fuel bundles, as well as 489,200 CanFlex ACR used fuel bundles generated by four ACR1000 reactors at the Darlington site require disposal in the deep geological repository. Scenario 3: 3.6 million used CANDU fuel bundles, as well as 4,263 PWR used fuel assemblies generated by three EPR reactors at the Darlington site require disposal at the deep geological repository. Scenario 4: 6,830 PWR used fuel assemblies generated by four AP1000 Generation III reactors at the Darlington site require disposal at the deep geological repository. (See Table 5.)

Table 5. Scenarios for the Waste Repository Proposed by NWMO

Scenario 1 Used CANDU bundles CanFlex ACR bundles PWR assemblies 3,600,000 0 0 Scenario 2 3,600,000 489,200 0 Scenario 3 3,600,000 0 4,263 Scenario 4 3,600,000 0 6,830
29

106 Agency for Toxic Substances and Disease Registry. Public Health Statement for Uranium. United States Department of Health and Human Services, Public Health Service, Atlanta, GA (September 1999; updated August 2008). <http://www.atsdr.cdc.gov/toxprofiles/ phs150.html>. Accessed March 2010. 107 NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005): p. 341. 108 It would take AP1000 and EPR used nuclear fuel approximately 2.6 million years to decay to the same radioactivity of natural uranium. It would take ACR-1000 used nuclear fuel approximately 1.74 million years to decay to the same level.

Based on these four scenarios, the volumes of water that would be needed to dilute the contents of the repository in each situation are listed in Table 6, and are also compared to the total water in Lake Ontario.

Table 6. Liters of Water Needed to Dilute the Full Repository

Liters of Water (E = 1018) Scenario 1 Scenario 2 Scenario 3 Scenario 4 6.02E+14 8.08E+14 1.94E+15 3.78E+15 Number of Lake Ontarios 0.37 0.50 1.18 2.30

If one takes into account NWMOs most recent deep geological repository expansion assumptions,109 it would take approximately 8 Lake Ontarios to dilute the radioactivity of the full repository.

5.4.2 ENvIRONMENtAL IMPACtS

The NWMO final study considered only the hazards posed by the disposal of spent CANDU nuclear fuel in a deep geological repository, based on the currently existing CANDU reactor fleet in Canada. It did not consider the impact and challenges Generation III reactor fuel waste will have on the proposed repository. A significant new risk to repository operation will be the much higher thermal output of high burn-up waste. Hotter fuel will generate more decay heat, which will enable greater thermoconvective forces affecting the water migration, waste package corrosion, and even geological deformation of the repository. The following discussions highlight the typical hazards and uncertainties associated with a repository, showing that the NWMO final study has not provided an adequate safety assessment for the disposal of high-burn-up fuel.

Thermal Convection
Convective forces constitute a major safety consideration for geological nuclear waste repositories, due to their effects on the local hydrology. Decay heat produced by the waste packages in the repository is transferred to the host rock and to the surrounding environment by thermal radiation. As a result of this heat transfer and the subsequent temperature differences between warmer and cooler sections, large-scale convective cells (fractures) appear along the lengths of the drifts (emplacement tunnels).110 111 In the early stages of repository emplacement, decay heat may warm the near-drift conditions to aboveboiling temperatures. As a result, water in the pores will vaporize and raise the pressure, which in turn will drive the vapor away from the heated area into distant rock formations or back into the drift. Essentially, warm vapor-rich gases will move from heated sections to cool sections and deposit the moisture on cooler rock surfaces through condensation. Vapor that is pushed out into the distant rock formations condenses and enhances the liquid flux, while the vapor that is pushed back into the drift raises the relative humidity within the repository cavity.112 Eddies caused by vertical air movement establish an axial transport of heat and moisture along the drift and cause changes in the surrounding saturation fluxes. It must be recognized that at the start the vapor inflow is from evaporation of the initial pore water; however, after the first few hundred years most of the evaporation comes from percolating water.113 This signals that the volume of the evaporating water can be replenished. The changes in saturation fluxes raise several concerns about the safety of a deep geological repository.
109 Garamszeghy, M. Nuclear Fuel Waste Projections in Canada2009 Update. NWMO TR-2009-30. Nuclear Waste Management Organization. Toronto, Ontario (2009). 110 Zharikov, A.V., et al. Permeability of Rock Samples from the Kola and KTB Superdeep Boreholes at High P-T Parameters as Related to the Problem of Underground Disposal of Radioactive. Petrophysical Properties of Crystalline Rocks. Geological Society of London (2005). 111 Webb, S.W., and A. Reed. In-Drift Natural Convection and Condensation. Yucca Mountain Project Report. MDL-EBS-MD-000001 REV 00. Las Vegas: Bechtel SAIC Company (2004). 112 Halecky, N., et al. Natural Convection in Tunnels at Tucca Mountain and Impact on Drift Seepage. Lawrence Berkley National Laboratory (September 2009). 113 Bahrami, D., et al. Water, Vapor, and Salt Dynamics in a Hot Repository. Materials Research Society (2007).

30

First, radionuclide migration occurs mainly through groundwater transport. The groundwater flow consists of a forced component caused by regional flow and the thermoconvective component caused by the described heat generation.114 The resulting convective cells would provide an easy mechanism for the transport of potentially contaminated water from the areas undergoing radioactive decay to areas where no waste is emplaced.115 Although some radionuclides will be less soluble in higher temperatures produced by decay heat, other radionuclides, including uranium, neptunium, and americium, will have higher solubilities with increasing temperatures.116 Additionally, during the evaporation of water from the drift wall into air space, the dissolved ions from the source will precipitate and accumulate as solids in the rock pores and fractures. Salts deposited early on, during the first two thousand years, will be transported toward the drift roof and the surface due to the vertical liquid flow.117 The combination of water transport by convective eddies and the continuous salt accumulation in pores and fractures can distribute radionuclides from the source to the surrounding environment. As stated above, vapor may also be pushed back into the drift, raising the relative humidity inside the repository. Such a phenomenon would have a considerable effect on the acceleration of waste package corrosion. In general, the thermally mobilized water can dominate seepage into emplacement drifts for thousands of years, even if the practical thermal power density is low. Any changes, temporary or permanent, in flow pathways resulting from thermally coupled processes can result in diverting condensate flow toward waste packages.118 In addition to heightened moisture levels inside the repository, the episodic water flow into the drift may cause the dripping of concentrated salts from the roof into the waste packages.119

Failures of Waste Containers

Radioactive waste packages consist of waste form and the waste container. The container must meet specifications consistent with waste form and the conditions surrounding the waste burial site. In general, waste that is to be stored for long periods of time (>50 years) is encapsulated in stainless steel or concrete containers.120 It is highly likely that water will eventually reach contact with the waste. Stainless steel containers corrode over time at a speed largely determined by the surrounding environment. Container corrosion occurs through general or through local processes. Research shows that general corrosion removes about 2 mm of the container thickness in 100,000 years. Local corrosion, however, is more significant and includes stress corrosion cracking, crevice corrosion, and pitting corrosion.121 The container design, especially its welded sections, has to be able to withstand such local stresses, which leaves human error a major contributor to the risks associated with package failures. More than 60 percent of hazardous-materials accidents are the result of human error.122 Furthermore, certain environmental conditions may exacerbate the local corrosion processes. The presence of salts, contaminants, or chemicals, as well as inadequate moisture levels and ambient temperatures may affect the container. For example, galvanic corrosion can be accelerated by the presence of Magnox, aluminum, or graphite in close proximity to other metals.123 The deposition of salts caused by the convective cells around the waste repository represents a situation which would exacerbate the waste package corrosion. 124

114 Zharikov, A.V., et al. Permeability of Rock Samples from the Kola and KTB Superdeep Boreholes at High P-T Parameters as Related to the Problem of Underground Disposal of Radioactive. Petrophysical Properties of Crystalline Rocks. Geological Society of London (2005). 115 Halecky, N., et al. Natural Convection in Tunnels at Tucca Mountain and Impact on Drift Seepage. Lawrence Berkley National Laboratory (September 2009). 116 Hardin, E. L., and D.A. Chesnut. Synthesis Report on Thermally Driven Coupled Processes. Lawrence Livermore National Laboratory (15 October 1997). 117 Bahrami, D., et al. Water, Vapor, and Salt Dynamics in a Hot Repository. Materials Research Society (2007). 118 Hardin, E. L., and D.A. Chesnut. Synthesis Report on Thermally Driven Coupled Processes. Lawrence Livermore National Laboratory (15 October 1997). 119 Bahrami, D., et al. Water, Vapor, and Salt Dynamics in a Hot Repository. Materials Research Society (2007). 120 International Atomic Energy Agency (IAEA). Requirements and Methods for Low and Intermediate Level Waste Package Acceptability. IAEA-TECDOC-864 (February 1996). 121 Environment Agency. The Longevity of Intermediate-Level Radioactive Waste Packages for Geological Disposal: A Review. Issue 1. United Kingdom (August 2008). 122 Shrader-Frechette, K. Burying Uncertainty: Risk and the Case against Permanent Geological Disposal of Nuclear Waste. Berkeley: University of California Press (1993). 123 Environment Agency. Gas Generation and Migration from a Deep Geological Repository for Radioactive Waste: A Review of Nirex/NDAs Work. Issue 1. United Kingdom (November 2008). 124 Environment Agency. The Longevity of Intermediate-Level Radioactive Waste Packages for Geological Disposal: A Review. Issue 1. United Kingdom (August 2008).

31

In addition to the outside environment, the internal environment of the container also has influence on the container corrosion, especially because it is more caustic, wetter, and may have more reactive agents. For example, it has been postulated that graphite present in the actual waste can lead to galvanic coupling and enhance corrosion of stainless steel containers. Cement containers can also be affected by the internal environment of the waste package. Certain metals present in the waste can change the cement formulation.125 In addition the constituents of cement can dehydrate at elevated temperature and transform to more crystalline minerals. Chemical reactions involving these recrystallized phases can affect water chemistry and relative humidity of the nearfield environment.126 Although some research attempts to show the long-term effectiveness of modern waste containers, such as Nirexs study on hydrated cement containers, presently research cannot definitively demonstrate success or failure of waste packaging. One of the issues is the fact that such research is done using natural analogues, which makes it impossible to state predictions with full confidence.127 In addition, many of the waste package corrosion tests have been made based on static repository environments. However, three-dimensional models that consider changes in barometric pressure and thermal differences between waste packages suggest that a more dynamic system exists within a repository and indicate that the predictions based on static models are inadequate.128 Accidents associated with leakage of waste containers are also not excluded. Posiva, the Finnish organization mandated to manage Finlands nuclear waste, conducted an environmental impact statement on a nuclear fuel repository there. It concluded that although the waste packaging is designed to withstand age and corrosion for the most part, the possibility of individual containers leaking and releasing contaminants into the environment cannot be excluded. Canister leakage can occur due to an originally damaged container or to major geological changes over thousands of years.129

Geological Stability
While a repository would always be placed in a geologically stable area, it is impossible to predict any long-term geological activity, including tectonic plate shifts, earthquakes, and ice ages. 130 Furthermore, the structure of the repository will be affected by the decaying wastes within it. Fractures in the host rock may open and close in response to rising and falling temperatures.131 The actual heat from radioactive decay of spent fuel will have the potential to expand the volume of the repository and raise the ground surface.132 Finally, gas generated by the corroding metals and decaying wastes can produce fractures in the surrounding host rock.

Gas Generation
The generation of gas in the repository is a realistic and likely possibility, which presents a number of safety concerns. Several pathways contribute to gas generation, including: the corrosion of metals in the waste and/or packaging, radiolysis of water and organic molecules in the packaging, and microbial degradation of organic wastes. Certain conditions expedite the gas-generating processes. For example, the availability of water assists in metal corrosion and the availability of oxygen and moisture from backfill accelerate the degradation of organic matter, thus contributing to gas generation.

32

125 Ibid. 126 Hardin, E. L., and D.A. Chesnut. Synthesis Report on Thermally Driven Coupled Processes. Lawrence Livermore National Laboratory (15 October 1997). 127 Environment Agency. The Longevity of Intermediate-Level Radioactive Waste Packages for Geological Disposal: A Review. Issue 1. United Kingdom (August 2008). 128 Bahrami, D., et al. Water, Vapor, and Salt Dynamics in a Hot Repository. Materials Research Society (2007). 129 Posiva. Posivas Expansion of the Repository for Spent Nuclear Fuel, Environmental Impact Assessment Report. Finland (2006.) 130 Ibid. 131 Hardin, E. L., and D.A. Chesnut. Synthesis Report on Thermally Driven Coupled Processes. Lawrence Livermore National Laboratory (15 October 1997). 132 Posiva. Posivas Expansion of the Repository for Spent Nuclear Fuel, Environmental Impact Assessment Report. Finland (2006.)

Once gas is formed within the repository it can create a hazard by causing structural damage due to over-pressurization of the chamber, by altering the local hydrology of underground formations, by producing a flammability risk, or by releasing gaseous radionuclides to the surface. Over-pressurization of the chamber due to gas build-up is the most influential pathway by which gas can affect water flow. The potentially immense pressure can cause fractures in the rock formations, creating a pathway for water to enter the chamber and for gas to escape it. Gas can also have an impact on the hydrology of the site. If a gas cushion is accumulated within the repository it can force out the water. Likewise, groundwater can also be trapped by streams of gas bubbles and be induced due to instabilities in gas pathways. Finally, certain colloids and chemical species can attach to gas-water interfaces and migrate throughout the groundwater formation. In order to prevent the collection of gas inside the chamber, grouts can be designed to release this gas. Although this alleviates the challenge of structural damage and impact on hydrology, some of the released gas may be flammable and will include gaseous radionuclides. Gases that escape to the surface are likely to be dominated by hydrogen and methane, making them a potential flammability hazard. Carbon-14 is a radionuclide that will be released in the most volume. It will escape to the surface mainly in the form of methane because Carbon-14 in the form of carbon dioxide is likely to react with the container or dissolve in groundwater. Other radionuclides that can be released as gases are H-3, Rn-222, Kr-85, and I-129.133 Many uncertainties remain regarding gas migration, such as migration of gas in cementitious materials, migration of gas through the geosphere, chemical and physical interactions between repository materials and gases, and appropriate gas solubilities in saline groundwater. 134

33

133 Environment Agency. Gas Generation and Migration from a Deep Geological Repository for Radioactive Waste: A Review of Nirex/ 134 Ibid.

NDAs Work. Issue 1. United Kingdom (November 2008).

6. Cost
It follows that cost estimates for such a fuel management approach as described above must be considered uncertain and speculative. No deep geological repository has ever been constructed worldwide and the projected costs of repositories in both the United States and Canada have escalated dramatically. The potential need to store fuel wastes having higher burn-up will create additional uncertainties and costs. These increased backend fuel management costs may have a substantial impact on the economics of new reactor construction. If left unassessed and unaccounted for, new-build reactor construction could be enabled by an implicit subsidy and expose Canadians to significant future liabilities. In the section below, we discuss the escalation of cost for a high-level waste repository. This material is taken from submitted testimony to the Ontario Energy Board.135 In the following section, we discuss the novel nature of high burn-up fuel and how it will add to the escalating cost.

6.1 Canadian High-Level Waste Repository

The NWMOs proposed nuclear fuel waste repository has been designed to store fuel bundles from Generation II CANDU reactors. The estimated cost of this repository has increased dramatically. In 1979, AECLs Whiteshell Nuclear Research Establishment estimated the costs of building the surface facilities and vault for a high-level waste repository designed to hold 334,000 metric tons of used fuel as $237 million for the surface facilities and $233 million for the vault, totaling $470 million. These costs include construction, engineering, administration, overheads, taxes and a contingency of 20%. The annual operating costs were estimated to be $99185 million for the surface facilities and $1925 million for the vault, for a total range of $118210 million. The transportation costs were estimated to be $43172 million, with an annual operating cost of $12180 million.136 The total cost was estimated to be between $643 million and $1.032 billion dollars. Adjusted for inflation, that is between $1,776,753,149 and $2,851,647,355, in 2006 dollars, or $5,320/MTU (metric tons of uranium) and $8,538/MTU.137 In 2004, the OPG, Hydro-Quebec, New Brunswick Power and AECL commissioned C Tech Development Corporation to perform a study to develop designs for a deep geologic repository. That estimate, which includes the cost of siting, constructing and operating a high-level waste repository was $11.750 billion, in 2002 dollars.138 The repository is expected to hold 3.6 million fuel bundles and operate for 30 years. To determine the cost of transportation, the four power companies commissioned Cogema Logistics. It calculated that the cost of transportation would be $954 million, in 2002 dollars.139 Since each fuel bundle contains 19 kg of uranium, the repository would hold about 68,400 metric tons of used fuel.140 That is, the proposed NWMO repository is 1/5 the size of the original Whiteshell concept. Adjusting for inflation we obtain $13,977,822,158, in 2006 dollars, or $204,354/MTU.141 Comparing the two repository concepts, normalized in terms of MTU, we see there is a 390687% increase over the original estimate. In addition to calculating the total megawatt-hours (MWh) of OPGs nuclear division, the total MWh for Canadas entire power reactor fleet, including New Brunswick Power (NBP) and Hydro-Qubec (HQ), was calculated. The calculations for NBP and HQ used their generating capabilities and their average lifetime capacity factor.142 143

34

135 Resnikoff, M. Application for Approval of the Integrated Power System Plan and Procurement Processes by the Ontario Power Authority, on Behalf of the Provincial Council of Women of Ontario. Radioactive Waste Management Associates before the Ontario Energy Board. EB-2007-0707. (1 August 2008.) 136 Johansen, K., et al. Second Interim Assessment of the Canadian Concept for Nuclear Fuel Waste Disposal Volume 3: Pre-Closure Assessment. (1985). 137 Bank of Canada. Inflation Calculator Other Rates and Statistics. http://www.bankofcanada.ca/en/rates/inflation_calc.htm. Accessed June 2008. 138 C Tech Radioactive Materials Management. Conceptual Design for Deep Geologic Repository for Used Nuclear Fuel. Prepared for Ontario Power Generation, New Brunswick Power, Hydro-Qubec and Atomic Energy of Canada Limited (December 2002). 139 Ontario Power Generation, Hydro-Qubec, New Brunswick Power, and Atomic Energy of Canada. Costs of Alternative Approaches for the LongTerm Management of Canadas Nuclear Fuel WasteDeep Geologic Disposal Approach. (March 2004). 140 Forsberg, Charles W. Description of the Canadian Particulate-Fill Waste-Package (WP) System for Spent-Nuclear Fuel (SNF) and its Applicability to Light Waster Reactor SNF WPS with Depleted Uranium Dioxide Fill. Oak Ridge National Laboratory (October 1997). 141 Bank of Canada. Inflation Calculator Other Rates and Statistics. http://www.bankofcanada.ca/en/rates/inflation_calc.htm. Accessed June 2008. 142 Juzhou, Zhu, Jianqiang, Shan, McQuade, Dennis. Operational Characteristics and Management of the Qinshan Phase III CANDU Nuclear Power Plant. Xian Jiaotong University, and Atomic Energy of Canada Ltd. 143 Then they were multiplied by 365 days in a year and 24 hours. The total megawatt-hours for NBP, HQ and OPG are 3,312,230,205. The total cost of the repository was then divided by the total MWh to the cost per MWh. We found the historic cost to be between .03 and .05 cents per MWh. For the present cost we used a value from the Ontario Power Authority, which estimated the cost to be 1.68/MWh. See: Ontario Power Authority. Economic Analysis of Gas-Fired and Nuclear Generation Resources. EB-2007-0707. Exhibit D, Tab 3, Schedule 1 (2007).

6.2 High-Level Waste Repository at Yucca Mountain, Nevada

The detailed design and engineering of the Yucca Mountain repository proposition is further advanced than the Canadian repository proposal and yet the project was recently cancelled by the Obama administration. The observed escalation of costs over time is potentially informative for future trends in repository development.

CHARACtERIzAtION AND CONStRuCtION

In 1984, the Department of Energy (DOE) estimated that the characterization144 costs for the Yucca Mountain Repository would be approximately US$1.5 billion and the cost of construction would be $1.1 billion.145 These estimates as adjusted for inflation became $3,000,479,902 (characterization) and $2,200,351,928 (construction), in 2006 dollars, or $5,200,831,830 in total.146 The original opening date for the proposed repository, 1998, has been pushed back to the year 2020, if the repository opens at all. In 2006, DOE estimated that Yucca Mountain would be completed in 2017 at a cost of $23 billion, which includes the cost of characterization and construction.147 This is a 342% increase over the original estimate of $5,200,831,830. Since DOE intends to ship waste by rail to Yucca Mountain, and there is no rail line within 125 miles of the proposed repository, DOE intends to build a new 319-mile rail line from Caliente, NV. The projected construction cost for this proposed rail line is $3.155 billion (in 2007 dollars).148 Adjusted back to 2006 dollars, the cost is $2,994,210,876.149 This cost for a rail line must be added to the cost of the repository. The repository is expected to hold 70,000 metric tons of spent fuel. The total historical cost of the project was estimated to be $7.717 billion, or $110,233/MTU, in 2006 dollars. The current projections, including the rail line, imply the project will cost $26.116 billion, or $373,090/MTU, in 2006 dollars. This is a 238% increase over the original estimate. In sum, the costs have continued to rise over time, and this rise must be taken into account in determining the cost uncertainty for siting and constructing a waste repository.

6.3 Implications of High Burn-up Fuel

The proposed use of Generation III reactor designs in Canada implies wholesale changes to the storage, transportation and fuel waste disposal systems. Costs and uncertainties can be expected to increase. The storage and transportation casks will be different from those for CANDU fuel waste. The road and rail trailers will be much heavier. The disposal system and underground configuration in the repository will also be different. While NWMO argues that production of greater quantities of Generation II CANDU used fuel would mean lower cost per CANDU bundle for disposal, this is not true for a PWR or BWR fuel assembly, since so many design changes will occur. New and different casks and carriages will have to be designed and built. For the repository, new disposal containers, underground carriages, and disposal systems will have to be developed. Generally, used fuel with high burn-up will require more underground acreage per ton than CANDU fuel waste. The additional tonnage of high burn-up used fuel will not lower the disposal cost per ton of additional fuel waste but will add to it.

144 Characterization means exploration of the geological formation to determine suitability for waste disposal. 145 Department of Energy. Environmental Assessment: Yucca Mountain Site, Nevada Research and Development Area. DOE/RW--0073Vo1.2. United States Government (May 1986). 146 Freidman, S. Morgan. The Inflation Calculator. <http://www.westegg.com/inflation/>. Accessed June 2008. Data before 1975 are the Consumer Price Index statistics from Historical Statistics of the United States (USGPO, 1975). Data since 1975 are from the annual Statistical Abstracts of the United States. 147 United States Government Accountability Office. Yucca Mountain Project: Information on Estimated Costs to Respond to Employee E-mails That Raised Questions about Quality Assurance. <http://www.gao.gov/new.items/d07297r.pdf>. Accessed June 2008. 148 Tetreault, S. Railroad Cost Estimates for Yucca Top $3 Billion. Las Vegas Review Journal (24 July 2007). 149 Freidman, S. Morgan. The Inflation Calculator. <http://www.westegg.com/inflation/>. Accessed June 2008.

35

7. Conclusion
In conclusion, OPG is proposing to build as many as four Generation III reactors, designs which have not yet been tested or operated anywhere in the world, at the Darlington site in Ontario, Canada. Generation III nuclear reactors use enriched uranium fuel having high burn-up, much different from the natural uranium currently used in Canadas existing CANDU reactor fleet. Canada has struggled for forty years to develop a long-term plan for managing the used nuclear fuel produced by Generation II reactors. The NWMO developed a proposal that Canada use its Adaptive Phased Management approach to move waste to a deep geological repository, based on calculations that assumed a finite amount of Generation II reactor fuel waste. The addition of new, more-hazardous types of used nuclear fuel will impose much greater health and environmental hazards on the Canadian public. These hazards have not been assessed by OPG or the NWMO. The Canadian public has been deprived of an objective forum for reviewing the potential impacts of continued radioactive waste production. OPG and the NWMO have not consulted Canadians on the wastes created by proposed new reactor designs or demonstrated that these new wastes can be managed safely in the long term. The hazards of Generation III nuclear reactor operation and radioactive waste management associated with using enriched uranium fuel far exceed the hazards of using natural uranium fuel. The higher burn-up of the fuel of Generation III enriched uranium reactors has the ability to increase the wear of materials and reactor components, thus posing operational dangers. The higher radioactivity of the fuel will also increase the consequences of an accident or act of sabotage during transportation to the centralized location. Even though the enriched uranium fuel waste would be smaller in volume, it will include radionuclides that are much more long-lived and will remain hazardous for a significantly longer time frame. The higher decay heat of Generation III fuel will increase the engineering challenge of designing a repository. Not only will more underground acreage per ton be needed than for CANDU fuel waste, but the higher decay heat will influence the water migration, package corrosion, and geological deformation of a repository. In addition to these known amplified hazards, there are also uncertainties associated with the Generation III reactor designs because these reactors have never been tested. Comparing the new Generation III reactor fuel to Generation II CANDU reactor fuel we find the following differences in hazards: Generation III reactor fuel waste will be 2158 times more radioactive than the CANDU reactor fuel waste. Radioactive waste produced by all operating CANDU reactors will take less than one millions years to decay to levels equivalent to natural uranium. Alternatively, waste from proposed new reactors will take about 2.6 million years to decay to the same levels. A geologic repository containing used fuel from the current CANDU reactors and prospective new Generation III reactors at Darlington would require a volume of water greater than two Lake Ontarios to dilute the radioactive contents to the concentrations in safe drinking water. This is 6 times more than the water volume needed to dilute a repository holding just CANDU fuel waste. If all proposed new Generation III reactor builds are implemented in Canada, the waste volume requiring disposal at a centralized deep geological disposal site would need the equivalent of as many as 8 Lake Ontarios to dilute the radioactive contents to the concentrations in safe drinking water.

36

In addition to environmental and health impacts, Generation III reactors also present higher costs, which have not been adequately assessed. As is, the estimated cost for a deep geological repository in Canada or the US has risen dramatically and is expected to continue increasing. Having to deal with a new type of radioactive waste from Generation III reactors would require making changes to the storage, transportation, and long-term management facilities, increasing the cost even more. The NWMO is wrong in its claims that the continued waste production will reduce the nuclear fuel waste management cost per unit of production, because it assumes that the additional wastes will be the same in nature and hazard as CANDU wastes. Without a proper assessment of the increasing costs, new-build reactors will be unfairly aided through a de facto subsidy paid for by tax-payers, ratepayers, and future generations.

37

References
Advisory Council to the NWMO. Nuclear Waste Management Organization Advisory Council Final Report. (22 September 2005) Available at: <http://nwmo.ca/advisory>. Agency for Toxic Substances and Disease Registry (ATSDR). Public Health Statement for Uranium. United States Department of Health and Human Services, Public Health Service, Atlanta, GA (September 1999; updated August 2008). <http://www.atsdr.cdc.gov/toxprofiles/ phs150.html>. Accessed March 2010. Aikin, A.M., J.M. Harrison and F.K. Hare. Green Paper on Nuclear Waste Management. EMR Report EP 77-6. Ottawa: Department of Energy, Mines and Resources (1977). Alexander, W.R., and L.E. McKinley. Deep Geological Disposal of Radioactive Waste. Elsevier Ltd.: Amsterdam, The Netherlands (2007). Atomic Energy of Canada Limited (AECL). Environmental Impact Statement on the Concept for Disposal of Canadas Nuclear Fuel Waste. Pinawa: AECL-10711, COG-93-1 (1994). AECL. CANDU Reactors: ACR-1000 (Advanced CANDU Reactor). Accessible from website: http://www. aecl.ca/Reactors/ACR-1000.htm. (2010.) Bahrami, D., et al. Water, Vapor, and Salt Dynamics in a Hot Repository. Materials Research Society (2007). Bank of Canada. Inflation Calculator Other Rates and Statistics. <http://www.bankofcanada.ca/en/rates/ inflation_calc.html>. Accessed June 2008. Bratt, D. The Politics of CANDU Exports. Toronto: University of Toronto Press (2006). Canada Department of Justice. Nuclear Fuel Waste Act (2002). Canadian Environmental Assessment Agency (CEAA). Nuclear Fuel Management and Disposal Concept (Seaborn Report). EN-106-30/1-1998E. Ottawa: Public Works and Government Services Canada (1998). Cember H. Introduction to Health Physics. 3rd ed. McGraw-Hill (1996). Civilian Radioactive Waste Management System and TRW Environmental Safety Systems Inc., 1995. GA-4/9 Legal Weight Truck Cask Systems Review and Status. Paper presented to the Nuclear Waste Technical Review Board, Arlington, VA, by D.J. Nolan (14 June 1995). Accessible at: <http://www.nwtrb.gov/meetings/1995/ june/nolan.pdf>.
38

Crombie, David (Advisory Council Chair). Letter to Ken Nash (NWMO Board of Directors Chair). Advisory Council to the NWMO. (September 2005.) C Tech Radioactive Materials Management. Conceptual Design for Deep Geologic Repository for Used Nuclear Fuel. Prepared for Ontario Power Generation, New Brunswick Power, Hydro-Qubec and Atomic Energy of Canada Limited (December 2002). Department of Energy. Environmental Assessment: Yucca Mountain Site, Nevada Research and Development Area. DOE/ RW--0073-Vo1.2 (May 1986). Available at: <https:// www.etde.org/etdeweb/servlets/purl/138186-nrG6Bs/ webviewable/>. Deutsche Welle. Problems at Germanys Asse II Nuclear Waste Repository. 20 August 2008. <http://www.dw3d.de/ dw/article/0,2144,3571028,00.html>. Dorfman, P. High Burn-up Radioactive Spent Fuel. Nuclear Consultation Group, Helsinki, Finland (8 May 2009). Duncan, D. (Minister of Energy). Letter to James Hankison (Chief Executive Officer), 16 June 2006. Durant, D. Managing Expertise: Performers, Principals, and Problems in Canadian Nuclear Waste Management. Science and Public Policy (April 2006). Durant, D. Radwaste in Canada: a Political Economy of Uncertainty. Journal of Risk Research. Vol. 12, nos. 78 (2009). Edwards, G. Following the Path Backward: A Critique of the Draft Study Report of the Nuclear Waste Management Organization Entitled Choosing the Way Forward. Canadian Coalition for Nuclear Responsibility (September 2005). Available at: <http://www.ccnr.org/ follow_path_back.pdf>. Energy, Mines and Resources Minister (ERM) and Ontario Energy Minister (OEM). Canada/Ontario Joint Statement on the Radioactive Waste Management Programme. Ottawa: Ministry of Supply and Services (5 June 1978). Environmental Agency. The Longevity of Intermediate-Level Radioactive Waste Packages for Geological Disposal: A Review. Issue 1. United Kingdom (August 2008). Environmental Agency. Gas Generation and Migration from a Deep Geological Repository for Radioactive Waste: A Review of Nirex/NDAs Work. Issue 1. United Kingdom (November 2008). Eurkea County Yucca Mountain Information Office. The Nuclear Waste Dilemma: An International Perspective. Nuclear waste update newsletter.

Cogema, 2003. Logistics of Transportation of Used Fuel to a Centralised Facility. Report of study carried out for Ontario Power Generation, New Brunswick Power, Hydro-Qubec and Atomic Energy of Canada Limited. Report No. 500276-B-009 (24 September 2003).

Eureka County, NV (Fall 2001). Fairlie, I. Estimated Radionuclide Releases and Collective Doses from the Rokkasho Reprocessing Facility. Greenpeace Japan (2008). Federal Ministry for the Environment, Nature Conservation and Nuclear Safety. Safety Requirements Governing the Final Disposal of Heat-Generating Radioactive Waste. Bonn, Germany (July 2009). Forsberg, C. W. Description of the Canadian ParticulateFill Waste-Package (WP) System for Spent-Nuclear Fuel (SNF) and its Applicability to Light Waster Reactor SNF WPS with Depleted Uranium Dioxide Fill. Oak Ridge National Laboratory (October 1997). Freidman, S. M. The Inflation Calculator. <http://www. westegg.com/inflation/>. Accessed June 2008. Garamszeghy, M. Nuclear Fuel Waste Projections in Canada2009 Update. NWMO TR-2009-30. Nuclear Waste Management Organization, Toronto, Ontario (2009). Halecky, N. et al. Natural Convection in Tunnels at Tucca Mountain and Impact on Drift Seepage. Lawrence Berkley National Laboratory (September 2009). Hamilton, T. GE-Hitachi Wont Bid for Reactor. Toronto Star (9 April 2008). Hardin, E. L., and Chesnut, D. A. Synthesis Report on Thermally Driven Coupled Processes. Lawrence Livermore National Laboratory (15 October 1997). Holtec. Topical Safety Analysis Report for the HI-STAR 100 Cask System. Holtec Report No. HI-941184 (1998). Hung, M. Financial Implications of Used Fuel Volume Variation in Long Term Management2008 Update. APM-REP-03780-0001. Nuclear Waste Management Organization, Toronto, Ontario (December 2008). Infrastructure Ontario. Request for Proposals: Nuclear Procurement Project. RFP No. OIPC 08-00-1027 (7 March 2008). Institut fur Gebirgsmechanik GmbH (Institute for Rock Mechanics GmbH). Mountain-mechanical State Analysis of the Shaft Arrangement System of Asse II. Short Report. Leipzig, Germany (11 September 2007). International Atomic Energy Agency (IAEA). Requirements and Methods for Low and Intermediate Level Waste Package Acceptability. IAEATECDOC-864 (Feb 1996) International Atomic Energy Agency (IAEA). The use of scientific and technical results from underground research laboratory investigations for the geological disposal of radioactive waste. IAEA-TECDOC-1243. (September 2001.)

International Atomic Energy Agency (IAEA). Optimization Strategies for Cask Design and Container Loading in Long Term Spent Fuel Storage. TECDOC-1523. (2006.) International Atomic Energy Agency (IAEA). Selection of Away-From-Reactor Facilities for Spent Fuel Storage: a Guidebook. TECHDOC-1558. (2007.) Johansen, K., Dunford, W.E., Donnelly, K.J., Gee, J.H., Green, B.J., Nathwani, J.S., Quinn, A.M., Rogers, B.G., Stevenson, M.A., and Tamm, J.A. Second Interim Assessment of the Canadian Concept for Nuclear Fuel Waste Disposal Volume 3: Pre Closure Assessment. (1985.) Juzhou, Zhu., Jianqiang, Shan., McQuade, Dennis. Operational Characteristics and Management of the Qinshan Phase III CANDU Nuclear Power Plant. Xian Jiaotong University and Atomic Energy of Canada Ltd. Large and Associates. Assessments of the Radiological Consequences of Releases From Existing and Proposed ERP/PWR Nuclear Power Plants in France. (2007.) Mascaro, L. Obama Administration: Were Done with Yucca. Las Vegas Sun. Las Vegas, NV (29 January 2010). Mascaro, L. Yucca Mountains Death Just a Few Steps Away. Las Vegas Sun. Las Vegas, NV (2 February 2010). National Research Council. Disposition of High-Level Waste and Spent Nuclear Fuel: The Continuing Societal and Technical Challenges. Committee on Disposition of High-Level Radioactive Waste through Geological Isolation, Board on Radioactive Waste Management. (2001.) Natural Resources Canada. Canadas Nuclear Future: Clean, Safe, Responsible. Press release, 14 June 2007. Available at: <http://www.nrcan.gc.ca/media/>. Nuclear Waste Management Organization (NWMO). Asking the Right Questions? The Future Management of Canadas Used Nuclear Fuel. Discussion Document 1. Toronto (2003). NWMO. Understanding the Choices: The Future Management of Canadas Used Nuclear Fuel. Discussion Document 2. Toronto (2004) NWMO. Assessing the Options. NWMO Assessment Team Report (June 2004). Available at: <http://www. nwmo.ca/Default.aspx?DN=1091,1090,199,20,1,Docu ments>. NWMO. Choosing a Way Forward: The Future Management of Canadas Used Nuclear Fuel (Final Study). Toronto, Ontario (November 2005). Ontario Power Authority. Economic Analysis of Gas-Fired and Nuclear Generation Resources. EB-2007-0707. Exhibit D, Tab 3, Schedule 1 (2007).

39

Ontario Power Authority. Supply Mix Advice Report. (December 2005.) Ontario Power Generation, Hydro-Qubec, New Brunswick Power and Atomic Energy of Canada. Costs of Alternative Approaches for the Long-Term Management of Canadas Nuclear Fuel WasteDeep Geologic Disposal Approach. (March 2004.) Ontario Power Generation (OPG). Project Description for the Site Preparation, Construction and Operation of the Darlington B Nuclear Generating Station, Environmental Assessment. Toronto: OPG (12 April 2007). OPG. Environmental Impact Statement for the New Nuclear at Darlington Project. Toronto: OPG (30 September 2009). Posiva. Posivas Expansion of the Repository for Spent Nuclear Fuel, Environmental Impact Assessment Report. Finland (2006). Resnikoff, M. Application for Approval of the Integrated Power System Plan and Procurement Processes by the Ontario Power Authority, on Behalf of the Provincial Council of Women of Ontario. Radioactive Waste Management Associates before the Ontario Energy Board. EB-2007-0707. (1 August 2008.) RISKIND, Version 2.0. Argonne National Laboratory. S.Y. Chen and Bruce M. Biwer, contacts (July 2002). Russell, S. Preliminary Assessment of Potential Technical Implications of Reactor Refurbishment and New Reactor Build on Adaptive Phase Management. Nuclear Waste Management Organization, Toronto, Ontario (December 2008). Schmitz F. & Papin J. High Burn-up Effects on Fuel Behaviour under Accident Conditions. Journal of Nuclear Materials. Vol. 270 (1999). Shrader-Frechette, K. Burying Uncertainty: Risk and the Case against Permanent Geological Disposal of Nuclear Waste. Berkeley: University of California Press (1993). Shrader-Frechette, K. Risk and Uncertainty in Nuclear Waste Management. NWMO (November 2003). Tait, J.C., Gauld, I.C., and G.B. Wilkin. Derivation of Initial Radionuclide Inventories for the Safety Assessment of the Disposal of Used CANDU Fuel. Report No. AECl-9881. Atomic Energy of Canada Limited, Whiteshell Nuclear Research Establishment, Pinawa, Manitoba (August 1989). Tetreault, S. Railroad Cost Estimates for Yucca Top $3 Billion. Las Vegas Review Journal (24 July 2007).
40

Webb, S.W., and A. Reed. In-Drift Natural Convection and Condensation. Yucca Mountain Project Report. MDL-EBS-MD-000001 REV 00. Las Vegas: Bechtel SAIC Company (2004). Westinghouse. AP1000 Public Safety and Licensing. <http://www.westinghousenuclear.com/AP1000/public_ safety_licensing.shtm>. Accessed 13 September 2004). Wilson, Lois M. Nuclear Waste: Exploring the Ethical Dilemmas. Toronto: United Church Publishing House (2000). World Nuclear Association. Nuclear Power in Sweden. <http://www.world-nuclear.org/info/inf42.html> (March 2010). Zharikov A.V., et al. Permeability of Rock Samples from the Kola and KTB Superdeep Boreholes at High P-T Parameters as Related to the Problem of Underground Disposal of Radioactive. Petrophysical Properties of Crystalline Rocks. Geological Society of London (2005).

United States Government Accountability Office. Yucca Mountain Project: Information on Estimated Costs to Respond to Employee E-mails That Raised Questions about Quality Assurance. <http://www.gao.gov/new. items/d07297r.pdf>. Accessed June 2008. United States Nuclear Regulatory Commission. Standard Review Plan for Spent Fuel Dry Storage Facilities, Final Report. NUREG-1567. (2000.)

David I. Poch
14 August 2008 Ms. Kirsten Walli, Board Secretary Ontario Energy Board By e-mail and RESS

Barrister

tel. (613) 264-0055 fax (613) 264-2878

Dear Ms. Walli: Re: EB-2007-0707 Exhibit L-8-2, ICF report on load forecast issues Further to my letter of August 1st, the principal author of the ICF report, Mr. Ralph Torrie, has experienced a severe confluence of deadlines and travel commitments (notably work urgently needed by the Ontario Ministry of Environment for upcoming cabinet deliberations on climate change policy). This has delayed the availability of the attached report until today. Unfortunately, an effort to rely upon other ICF staff to meet the earlier filing date proved to be unworkable given the extensive background to this case. As is apparent from the document, the work was data intensive and relied to a considerable degree on knowledge of the Ontario electricity planning history, economic context and the details of OPAs filings and responses to interrogatories. GEC-Pembina-OSEA has previously expressed its desire to see the hearing proceed expeditiously. Eight of the nine reports the groups commissioned were filed on August 1st. Unfortunately, the GEC-Pembina-OSEA intervention team had no advance notice of the difficulty facing ICF and accordingly, we had not applied for permission to file late. Given the repeated delays we have experienced in obtaining this report over the last two weeks, we were not in a position to reliably predict when the report would be available to enable us to seek an alternative deadline from the Board. We apologize for the delay and ask that the document be received by the Board on the basis that we will accept interrogatories on it until August 22nd and we will respond by the original response deadline of September 2nd. In this manner we hope that any inconvenience to the Board and parties will be minimized and no prejudice or delay in the hearing progress will result. Please note that I am out of the country during the week of August 11th. Sincerely,

<original signed by>

David Poch Cc: all parties

1649 Old Brooke Road, Maberly, Ontario K0H 2B0

e-mail:dpoch@eelaw.ca

Green Energy Coalition

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2

BEFORE THE ONTARIO ENERGY BOARD IN THE MATTER OF sections 25.30 and 25.31 of the Electricity Act, 1998; AND IN THE MATTER OF an application by the Ontario Power Authority for review and approval of the Integrated Power System Plan and proposed procurement processes.

Review of the Ontario Load Forecast in the Integrated Power System Plan (IPSP)
by Ralph Torrie & Doug Morrow ICF International
Filed August 14, 2008

(David Suzuki Foundation, Eneract, Greenpeace Canada, Sierra Club of Pembina Institute Ontario Sustainable Energy Association
Canada, World Wildlife Fund Canada)

prepared for: Green Energy Coalition

1986 Panda symbol WWF-World Wide Fund For Nature (also known as World Wildlife Fund) WWF is a WWF Registered Trademark

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 1 of 49

Review of the Ontario Load Forecast in the Integrated Power System Plan (IPSP)

Prepared by: Ralph Torrie and Doug Morrow ICF Consulting 277 Wellington St. West, St. 808 Toronto, ON M5V 3E4 Canada

August 14, 2008

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 2 of 49

Some Historical Context

Long range forecasting of electricity demand has been the central reason that previous attempts at electric power investment planning in Ontario have failed. The great power system expansion plans put forward by Ontario Hydro in the mid-1970s and then again in the late 1980s never materialized because they were based on forecasts that turned out to be so far off the mark that none of the dozens of power plants proposed in those earlier efforts was ever built. In fact, the forecasts that underpinned those planning efforts proved to be so wrong so quickly after they were done that the reviews and associated public hearings associated with those earlier planning efforts (the Royal Commission on Electric Power Planning, the Environmental Assessment of the Demand Supply Plan) did not even run their course before the proposals were withdrawn or shelved for the indefinite future.1 In the mid-1970s, Ontario Hydro put forward a long range investment plan for Ontarios electricity system that was based on a forecast of electricity that saw the demand doubling roughly every ten years, with the capacity of the system passing through the 70,000 MW mark around 1997, and then nearly doubling again by 2008. The generation supply plan (LRF48A) included the addition of 120,000 MW of new capacity by the summer of 2008, provided by nuclear and coal in about a 2:1 ratio. The nuclear units increased in size from 850 MW to 1200 MW to 2000 MW over the course of the supply plan; the coal units were standardized at 750 MW each. By 2008, the plan included 56 nuclear reactors, post-Darlington, and 50 coal-fired units, none of which were ever built. In the 1977 forecast, the system peak in 1997 was projected to be 57,000 MW, with low and high variations of 36,000 MW and 67,000 MW, respectively. The plan asserted that it is unlikely the actual load will be outside the range of these scenarios2 the actual peak demand in 1997 was 22,200 MW. The 1977 forecast was so egregiously wrong that thirty years later the system peak has not reached the mark that the 1977 forecast predicted it would pass through 25 years ago, in 1983. In the late 1980s Ontario Hydro came forward with another power system plan -- Providing the Balance of Power.3 By this time, the traditional top down macroeconomic correlation methods for forecasting electricity demand (that had proven so inadequate in the 1970s) were being supplemented at Ontario Hydro with a set of end use models that allowed the forecast to be developed with more detail than had been the case in the past. Notwithstanding the end use
1

Given that it has been wrong-headed load forecasting that has been the central weakness in previous attempts at long range electric power investment planning in Ontario, it is ironic that reviews and assessments of those plans often take the load forecast as a given, even to the point of ruling it out of bounds as a suitable topic for inclusion in the review of the investment plan. 2 Ontario Hydro, Planning of the Ontario Hydro East System, Report No. 573 SP, November 15, 1977.Part 2, page 3-1 ff. 3 Ontario Hydro, Providing the Balance of Power: Ontario Hydros Plan to Serve Customers Electricity Needs, Ontario Hydro, Toronto, 1989.

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 3 of 49

disaggregation and the fact that the projected electricity growth rate was much lower than had been the case with the earlier generation of forecasts, the forecast that underpinned the Balance of Power Plan completely misread the dynamics of the commodity market. It had peak demand reaching 34,900 MW by 2005 and 39,800 MW by 2014 and asserted that there was only a 10% probability the actual load would fall below a lower bound forecast that reached 28,700 MW by 2004 and 33,500 MW by 2014. Actual demand peaked at 25,000 MW in 2004, 10,000 MW below the median forecast in Providing the Balance of Power and 3,700 MW below even the lower boundary (less than 10% probability) of the forecast bandwidth. These and most other historical forecasts of electricity demand that have been done to support long range electric power planning in Ontario over the past 30 years were not too high because they underestimated the amount of conservation and electricity efficiency improvement that would take place over the forecast period. They were too high because they failed to capture (or even attempt to capture) the underlying dynamics of the electricity commodity market. While there have no doubt been and continue to be improvements in the efficiency of electricity use, until now at least such improvements comprise a relatively small contribution to the reason why the aggregate demand for electricity has turned out to be so much lower than predicted. For example, in the mid-1970s when the total consumption of electricity was in the range of 85 TW.hours, Ontario Hydros load forecast projected it would grow to more than 325 TW.hours by 1997. Electricity demand in 1997 was 140 TW.hours. Only a fraction of the 185 TW.hour shortfall from the forecast could possibly be attributed to conservation and efficiency most of it was due to the evolving structure of electricity demand which rendered moot the assumptions in the 1977 load forecast. In 1988, when the Providing the Balance of Power supply plan was put forward by Ontario Hydro, annual electricity consumption in the province was 140 TW.hours and forecast to grow by some 85 TW.hours by 2005, to 225 TW.hours. By 2005 the actual demand was 155 TW.hours. As it turned out, the misreading of the future demand for electricity represented by the load forecast that underpinned the rationale for the Demand Supply Plan was a much larger problem with the Demand Supply Plan than the aggressiveness of the demand and conservation component of that plan. While the environmentalists and Ontario Hydro were debating before the Environment Assessment Board whether it might be possible to get 10 or 20 or even 35 TW.hours of electricity conservation out of the forecast load by 2010, the real issue would turn out to be that the demand itself would be lower than the forecast by more than twice the most aggressive estimate of DSM potential. As had been the case in the earlier round of planning, the large gap between the forecast and what actually transpired was well above even the most aggressive analyses of the realizable potential for conservation and efficiency. As had also been the case in the earlier round of power system planning, the forecast was high primarily because it failed to capture the market dynamics of electricity demand, dynamics in which role of kilowatt-hours in providing

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 4 of 49

underlying energy service demands was undergoing major changes for reasons that had very little to do with the price of fuels and electricity or the technological efficiency of electricity using devices. The demand for kilowatt-hours of electricity is derived from and part of more fundamental demands for amenities, including for example energy end use services such as lighting; information processing and telecommunications; convenient and quick access to recreation, culture and the performing arts; and interior working and living spaces that are warm in winter and cool in the summer. These amenities are provided with an array of techniques and technologies, and electricity varies from one energy end use vector to the next and over time with regard to its importance and relative contribution to the overall cost of providing the amenity. Electricitys competition over the past thirty years has not been primarily other energy commodities, or even the greater efficiency of electricity use itself; rather, electricitys competition has come from an underlying improvement in the electricity productivity of the Ontario economy that is much deeper and much broader than the mere technological efficiency improvement of electricity use.

The Current Forecast An Overview

By the time the Ontario Power Authority was formed, there had been a lengthy hiatus in long range electric power planning in Ontario. The last long range forecast had been done in the early 1990s at Ontario Hydro. The power industry in Ontario had been restructured, with no single player having responsibility for long range planning. The Independent Electricity System Operator (IESO) maintained a forecast, but it was focused on the short term and was generated using top down correlations with labour force and household formation, was not intended for and did not have the internal structure to support long range planning or DSM potential analysis. The OPA has recently expressed interest in taking an integrated approach to the modeling and analysis of DSM and the future demand for electricity, an approach that would at least provide the possibility of the type of scenario and portfolio risk analysis that is needed to better prepare for a range of plausible long term directions of Ontarios electric power system.4 While the OPA has indicated it expects such an approach can be put in place in less than a year, they are just getting underway in this direction. Meanwhile, the IPSP is based on the traditional approach -- a single line forecast against which a separate analysis of DSM potential is conducted. The load forecast underpinning the IPSP is depicted in Figure 1, and a close examination reveals a significant departure from recent trends. For one thing, the forecast predicts that the
4

Ontario Power Authority, REQUEST FOR PROPOSALS FOR CONSULTING SERVICES TO DEVELOP A 20-YEAR ENERGY AND DEMAND FORECAST AND A CONSISTENT CONSERVATION POTENTIAL ASSESSMENT OR TO DEVELOP A MODEL TO PRODUCE SUCH FORECASTS AND ASSESSMENT, Issued June 30, 2008, Toronto, Canada.

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 5 of 49

rate of demand growth, which has been falling for decades, will turn around and begin increasing, with electricity demand growth actually accelerating over the forecast period. It is also evident that the forecast has already diverged significantly from actual demand trends. The last year for which historical data was available when the forecast was produced was 2005, when the demand hit an all time high of 157 TW.hours. In the IPSP forecast, demand growth rates stop falling and turn around, with forecast demand in 2007 returning to 157 TW.hours in 2007 and climbing slowly to 159 TW.hours in 2010 before beginning to accelerate to 195 TW.hours in 2027. But by the end of June 2008, electricity consumption in Ontario was down 2.75% compared to the first six months of 2007. Total annual electricity use in Ontario in 2008 is on track to be lower than it has been since 1999, and this year will be the third year in a row electricity consumption has been lower than the all-time high reached in 2005 of 157 TW.hours. It is likely that 2008 electricity use will be in the range of 148 TW.hours, nearly 10 TW.hours below the IPSP load forecast issued in 2007, just one year earlier.

Figure 1
Historical and Forecast Grid Electricity Consumption in Ontario
200 180 160 140 120 100 80 60 40 20 0 1987 1992 1997 2002 2007 2012 2017 2022 2027
TW.hours

Historical OPA Forecast

The extent to which the forecast diverges from past trends can be seen more clearly by comparing the twenty-year average growth rates for the historical years with the twenty year

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 6 of 49

averages for the forecast years. The average growth rate of electricity demand in Ontario for the past ten years has been about 0.5% per year5 and in the twenty years prior to 2007 the average annual growth rate was 0.7% per year. This twenty year average growth rate has been declining steadily for over 25 years, as shown in Figure 2. (In this figure, the growth rate shown for each year is the average annual growth rate for the twenty year period leading up to that year; for example, the growth rate shown for 1996 is 2.0%, the average annual growth rate of the 1976-1996 period.) In the Integrated Power System Plan, this longstanding historical trend is assumed to be reversed, and electricity demand rates begin to grow and even accelerate over the 2007-2027 forecast.

Figure 2
Electricity Demand Growth in Ontario, Historical and Forecast 20 Year Average Annual Growth Rates
6.0% 5.0% 4.0% 3.0% 2.0% 1.0% 0.0% 1980

1985

1990

1995

2000

2005

2010

2015

2020

2025

2030

Historical Data

Forecast Period

In the OPA forecast, electricity demand growth averages 1.1% per year over the next twenty years, as compared with 0.7% over the 1987-2007 period and with what will likely be about 0.5% over the 1988-2008 period. Between 1987 and 2007, electricity consumption in Ontario grew by 26 TW.hours; in the OPA forecast it grows by 38 TW.hours over the 2007-2027 period.

The OPA load forecast notes that the annual growth rate from 1995 to 2005 (a high demand year) was 1.3%, but it has been declining every year since and is on track to drop below 0.5% by the end of this year.

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 7 of 49

In this sense, the load forecast underpinning the IPSP follows in the tradition of the previous load forecasts we have referred to above, which have acknowledged the historical decline in electricity demand growth rates but then go on to assume (or to adopt assumptions that lead to the conclusion) that the decline will stop more-or-less immediately in the forecast year and even reverse and begin climbing again over the forecast period.6 Perhaps this time will be different, and the decline in the long range average growth rate of electricity demand that has been going on in Ontario for over 40 years since it peaked in 1965 will turn around and begin to grow again in the next year or two as the IPSP forecast predicts. To explore this possibility further we need to look more closely at the sector and end-use makeup of the IPSP forecast.

The IPSP Load Forecast A Closer Look

The IPSP forecast is a somewhat cobbled together effort. The economic growth rates and associated end use breakdowns are based largely on the outdated NRCan 2006 Energy Outlook. Additional detail has been filled in by MKJ Associates and Marbek who were commissioned by the OPA to take a national study of conservation and demand management potential for all fuels for all of Canada7 (which had also relied on the NRCan 2006 Outlook as a reference point) and adapt it for the case of Ontario electricity.8 When ICF did an preliminary assessment of the DSM potential for the OPA in 2005,9 the only long range forecast available was based on the IESO ten-year forecast of the day, as extended to 2025 by Navigant10. That forecast used the historical correlation between electricity demand, housing starts and labour force growth to project electricity demand into the future and it projected a long range growth of electricity demand of 0.9%, reaching 186 TW.hours in 2025, very close to the 189 TW.hours in 2025 in the IPSP forecast. In fact, the OPA forecast that

As the electricity demand growth in Ontario hovers around zero it is important to realize that a zero growth rate for electricity demand does not represent a floor below which demand growth rates cannot fall for a sustained period. Indeed, the ICF sectoral calibration concluded that residential electricity demand in Ontario will likely trend downward over the forecast period, even as population, households and incomes continue to grow. As a consideration of the derived nature of electricity demand suggests, there are clearly conditions under which there can be negative growth in the contribution of kilowatt-hours to meeting underlying human amenities and energy end use services, even as the supply of those amenities and services continues to grow. 7 Marbek Resource Consultants Ltd. and MK Jaccard and Associates (MKJA). Demand Side Management Potential in Canada: Energy Efficiency Study. Submitted to: Canada Gas Association, May 2006. 8 MK Jaccard and Associates (MKJA). Modelling and Scenario Documentation. Prepared for: Ontario Power Authority Power System Planning (Ex. D, Tab 4, Attach. 6), September 2006. 9 ICF Consulting. Electricity Demand in Ontario Assessing the Conservation and Demand Management (CDM) Potential. Prepared for: Ontario Power Authority (OPA), November 2005. 10 Navigant Consulting Ltd., Avoided Cost Analysis for the Evaluation of CDM Measures Presented to Hydro One Networks Inc., June 14, 2005. The Navigant forecast is an extrapolation the Median Growth Scenario in the IESO 10-Year Outlook: Ontario Demand Forecast from January 2005 to December 2014 (March 31, 2004).

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 8 of 49

underpins the rationale for the IPSP is essentially the same as the forecast we analyzed in 2005, albeit with a higher long range growth rate (1.1% vs. 0.9%). To estimate DSM potential, however, it is necessary to have a forecast that contains a high level of sector and end-use disaggregation. To estimate how much electricity could be saved by a particular technology in a particular sector/end use application (for example, office building lighting) one first needs to know or have an assumption about how much electricity will be used by office building lighting, and what the saturations and market shares of current technologies are for providing office building lighting. These issues were addressed in more detail in the report produced at the time, but the point is that in order to do a DSM potential assessment it is first necessary to have a disaggregated breakdown of electricity, at least for the end uses being targeted by the proposed DSM programs. That is what ICF had to do in 2005 in order to estimate the DSM potential relative to the forecast, and that is why MKJ and Marbek also had to develop a disaggregated forecast before they could proceed with their DSM potential estimates. Both the ICF and the MKJ/Marbek analyses were done against the backdrop of the NRCan Outlook (which we know is in good agreement with the forecasts for the Ontario economy on which the OPA forecast is based). The MKJ/Marbek end use forecast reflects the OPA forecast of 1.1% long term growth in electricity demand. The ICF analysis calibrated to the 0.9% growth rate in the IESO/Navigant forecast, but only by making assumptions about activity growth rates and natural rates of efficiency and productivity improvement that biased the demand upward so as to hit the IESO/Navigant forecast. The ICF study concluded that electricity demand growth in Ontario would stay well below 0.9% per year unless there was a reversal of historical trends, for example the assumption of unmoderated commercial building floor area growth while at the same time assuming no improvement in building electricity intensities over the entire forecast period. The study concluded: While the IESO forecast growth rate of 0.9 percent per year is low by the standards of the 1970s and 1980s, it is significantly higher than the 0.5 percent year average of the period since 1990. The forecast demand could prove low if the electricity-intensive industries grow faster than the rest of the economy, or if electricitys market share of space and water heating increases, or if the rate of new commercial building construction accelerates relative to the trend of the past fifteen years, but these circumstances represent reversals of recent trends. If recent trends continue, then the risk of the forecast being too high is more likely.11

ICF Consulting. Electricity Demand in Ontario Assessing the Conservation and Demand Management (CDM) Potential. Prepared for: Ontario Power Authority (OPA), November 2005.

11

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 9 of 49

To further understand the reasons why the OPA forecast is projecting a reversal of the longstanding historical trend, we looked at each of the sectors separately in some detail; we present here some of the key observations from that analysis.

Residential Sector
Structure of Electricity Demand
Electricity use in the residential sector depends on a number of factors, including population growth rates, the number of households, the proportion of households that use electricity for space and water heating and the energy efficiency of electricity-specific end uses12, such as lighting. Since electricity is a subset of energy, it is instructive to examine to relative contribution of electricity to the total energy needs of the residential sector. As shown in Figure 3, the contribution of electricity to total energy use remained relatively flat over the 1990-2005 period, ranging from approximately 33% in 1990 to 30% in 2005. The most significant development to take place over the historical period was the increasing share of natural gas at the expense of both electricity and heating oil. In 2005, fully 61% of total residential energy use was derived from natural gas, up from 52% only 15 years earlier.

12

This refers to end uses whose only energy source is electricity.

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 10 of 49

Figure 3
Share of Total Energy Use in Residential Sector by Energy Source, 1990-2005 70% 60% 50% 40% 30% 20% 10% Wood 0%
19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05

Natural Gas Electricity Heating Oil Other

Ye ar

As in the commercial and industrial sectors, the largest end use for energy consumption in the residential sector is space heating. As shown in Figure 4, energy demand for space heating in 2005 totaled 89,400 GW.h, or approximately 58% of total energy use in the sector. Space heating thus commands a larger share of total energy demand than all other end uses combined. When interpreting the energy demand of different end uses, it is important to keep in mind that energy demand for lighting and space cooling can only be met by electricity. As a result, these end uses can be thought of as captive markets for electricity. By contrast, electricity must compete with other energy sources (principally natural gas) for use in space heating, water heating and, to a far lesser extent, in ranges and clothes dryers. Any transition away from the use of electricity in space heating will have disproportionately large effects on total electricity demand.

10

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 11 of 49

Figure 4
Share of Energy Use by End Use, 1990-2005 180,000 160,000 140,000 120,000
GWh
Space Cooling Lighting Appliances Water Heating Space Heating

100,000 80,000 60,000 40,000 20,000 0

19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05
Year

The relative contribution of residential electricity use to total electricity use in Ontario has remained roughly constant over the past half-century. As shown in Figure 5, the relative share of residential sector electricity use shifted by only 10% over the 1958-2005 period, ranging from a low of 27% in the early 1970s to a high of 37% in 1990. The residential sector is the only sector not to have undergone a significant proportional change since the 1950s.

11

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 12 of 49

Figure 5
Total Electricity Final Demand in Ontario by Sector, 19582005
60% 50% 40% 30% 20% 10% 0%
19 58 19 61 19 64 19 67 19 70 19 73 19 76 19 79 19 82 19 85 19 88 19 91 19 94 19 97 20 00 20 03

Residential Commercial Industrial

Year

Historical trends
As shown in Figure 6, final electricity demand in the residential sector increased dramatically over the past half-century, growing from approximately 8,000 GW.h in 1958 to 46,000 GW.h in 2005. While the broad historical pattern can therefore be characterized as one of increasing demand, Figure 6 also illustrates how growth in electricity use experienced a marked slowdown beginning around 1990. Notable year-on-year decreases in total residential electricity demand occurred in the early 1990s, 1996 and, most recently, in the early 2000s, where electricity demand declined from an all time high of approximately 50,000 GW.h in 2003 to 46,000 in 2005.

12

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 13 of 49

Figure 6
Electricity Final Demand in Ontario in the Residential Sector, 1958-2005
60,000

50,000

40,000

GWh

30,000

20,000

10,000

0
98 94 90 86 82 78 72 74 66 68 58 60 62 64 70 76 80 84 88 92 96 00 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 02 04

Year

The observed leveling of electricity demand growth in the Ontario residential sector is noteworthy in that many of the aforementioned drivers of residential electricity demand including household, floor space and population growthgrew much more quickly than electricity use over the same period. This trend, which is explored in greater detail in Figure 7, is a central feature of residential sector electricity use since 1990.

13

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 14 of 49

Figure 7
Relative Growth of Residential Electricity Use and Key Drivers
1.80 1.60 1.40 1.20 1990=1 1.00 0.80 0.60 0.40 0.20 0.00 Total Floor Space (million m2) Population Households (thousands)

Ontario GDP (1997 millions)

Actual Electricity Demand

Figure 7 shows the relative growth in actual electricity use and a number of key demand drivers over the historical period (normalized to 1990). The graph illustrates that electricity use did not keep pace with growth in households, total floor space or population. This reflects the growing electrical productivity of the Ontario residential sector. Despite near flat growth in electricity use from 1990 to 2005 (electricity use increased by less than 1% from 1990-2005, with average annual growth of 0.05%), the number of households in Ontario increased by 29% over the same period, growing from 3.6 million in 1990 to 4.7 million in 2005. A similar pattern is found with population (which increased by 22%) and total floor space (which increased by 32%). While GDP would not be expected to be correlated with residential sector electricity use, it is included in Figure 7 as a point of reference to show the decoupling of electricity demand and economic growth in the province (a trend that is also borne out in the commercial and industrial sectors). Expressing these variables as ratios shows that electricity use per household and per capita declined substantially during the historical period (see Figure 8). In 1990, the average Ontario household used 12.5 MW.h of electricity. By 2005, that figure had dropped to 9.8 MW.h, a decrease of 22%. In terms of a per capita comparison, the average Ontario resident used 4.4 MW.h of electricity in 1990 and 3.6 MW.h in 2005, a decrease of 18%. Figure 8 also shows how the number of Ontario residents per household remained essentially flat during the 1990-2005 14

19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05
Year

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 15 of 49

period, during which time the relative composition of Ontarios housing stock (i.e., single detached, single attached, apartments and mobile homes) did not change significantly.13

Figure 8
Relative Growth of Key Demand Ratios in the Residential Sector
1.2 Electricity Use Per Capita (MWh per Ontario resident)

0.8 Electricity Use Per Household (MWh per Ontario household) 1990=1

0.6

0.4

0.2

People per houeshold

0
19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05

Year

While total residential electricity demand remained essentially flat between 1990 and 2005 (45,274 GW.h in 1990 vs. 45,630 in 2005), the end use composition of electricity demand underwent a number of profound changes. The first change was that seven end uses experienced negative growth in terms of absolute electricity demand. From 1990-2005, end use demand declined for freezers (-56%), water heating (-45%), refrigerators (-44%), dishwashers (-39%), clothes washer (-23%), clothes dryer (-9%) and space heating (-8%). Given that the stock in the Ontario economy of each of these end uses increased (often dramatically, such as the case with dishwashers) during the period, the decreases point to rapid and systematic energy efficiency improvements that occurred across these product categories beginning in the mid 1990s.
According to the OEE, the percentage breakdown of Ontarios housing stock in 1990 was as follows: single detached (56%), single attached (13%), apartments (30%), and mobile homes (1%). In 2005, the breakdown was as follows: single detached (57%), single attached (14%), apartments (29%), and mobile homes (1%).
13

15

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 16 of 49

Figure 9
Percentage Breakdown of Residential Electricity Use by End Use, 1990 vs. 2005 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1990 2005
Space Heating Space Cooling (Central) Lighting Other Appliances Refrigerator Clothes Dryer Range Water Heating Freezer Space Cooling (Room) Dishwasher Clothes Washer

Expressed in terms of the change in their relative contribution to total residential sector electricity demand, four of these seven end uses underwent significant declines, with the others remaining relatively constant (i.e. <1%). In 1990, refrigerators and water heating accounted for 17% and 12%, respectively, of residential sector electricity use. By 2005, these figures had dropped to 9% and 6%. Less dramatic declines occurred for freezers (7% of total sector demand in 1990 vs. 3% of total sector demand in 2005) and space heating (20% of total sector demand in 1990 vs. 18% of total sector demand in 2005).

16

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 17 of 49

Figure 10
Average Annual Growth Rate in Electricity Use by End Use, 1990-2005
10% 8% 6% 4% 2% 0% -2% -4% -6% -8% 1
Freezer Water Heating Refrigerator Dishwasher Clothes Washer Clothes Dryer Space Heating Range Lighting Space Cooling (Room) Other Appliances Space Cooling (Central)

17

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 18 of 49

Figure 11
Electricity Use by End Use, 1990 vs. 2005 10,000 9,000 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0
W as D he is r hw Co as ol he in r g (R oo m ) Fr ee W at ze er r H ea tin g R an C lo ge th es D ry R er ef O rig th er er at Ap or pl ia Sp nc ac es e Co Li gh ol ti n in g g (C en Sp tra ac l) e He at in g

GWh

1990 2005

C
14

lo th es

The second major change was that absolute electricity use increased significantly for five end uses, including ranges (4%), lighting (7%), room space cooling (60%), other appliances14 (84%) and central space cooling, which increased from 2,000 GW.h in 1990 to 7,300 in 2005, a gain of 263%. While the relative contribution to total residential sector electricity use remained constant for lighting, ranges and room space cooling, it increased markedly for other appliances (8% in 1990 compared to 14% in 2005) and central space cooling (4% in 1990 compared to 16% in 2005). The increase in end use demand for other appliances and central space cooling is clearly a function of shifting consumer preferences for these devices. The number of other appliances in Ontario grew from approximately 38 million in 1990 (roughly 10 per household) to 74 million by 2005 (roughly 16 per household). At the same time, electricity use per other appliance dropped from approximately 95 KWh in 1990 to 88 in 2005 (see Figure 12).15
The OEE defines other appliances as small appliances such as microwaves, televisions, cable boxes, video cassette recorders, stereo systems and computers. 15 The sudden decline in electricity use per other appliance from 2004 to 2005 could be a result of an error in the OEE Comprehensive Energy Use database. The decline (from roughly 100 KWh in 2004 to 89 KWh in 2005) is a function of both a significant decrease in the amount of total electricity use from other appliances reported over 2004-2005 and a sizeable increase over 2004-2005 in the number of other appliances in the economy.

Sp ac e

18

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 19 of 49

Figure 12

Electricity Use per Other Appliance, 1990-2005

102 100 98 96 KWh 94 92 90 88 86 84 82
02 03 04 20 90 91 92 93 94 95 96 97 98 99 00 01 20 20 19 19 19 19 19 19 19 19 19 19 20 20 20 05

Year
Looking at space cooling, in 1990, there were 997,000 central space cooling systems in Ontario, or 0.3 per household. By 2005, there were 2.4 million systems in Ontario, or 0.5 per household. Figure 13, which compares the saturation rate for these two end uses, shows that growth in central space cooling systems per household appears to be tapering off, which could put downward pressure on residential sector electricity growth going forward.

19

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 20 of 49

Figure 13
Saturation Rates of Other Appliances and Central Space Cooling Systems 0.6 0.5 0.4 0.3 0.2 0.1 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Other Appliances (Hundreds) per Household

Central Space Cooling Systems per Household

Year

Analysis of the IPSP Forecast

Figure 14 shows actual electricity use in the residential sector from 1990 to 2005 and projected use from 20002025 under the IPSP forecast. The IPSP reference case anticipates an increase in residential electricity demand from 41,900 GW.h in 2000 to 53,830 GW.h in 2025, an increase of 11,930 GW.h or 28%.16 Analyzing the 5 year overlap (2000-2005) between actual electricity demand data from the OEE and forecasted electricity demand from the IPSP shows that the latter is tracking relatively closely to historical trends. The IPSP forecast for 2000 was 41,900 GW.h, while actual demand was 42,730 GW.h. A gap of approximately 2,000 GW.h between actual and forecasted data emerged in 2002, but it narrowed to a negligible amount by 2005. Indeed, the difference between actual electricity use in 2005 and the IPSP forecast was only 307 GW.h.

Since data in the IPSP reference case are provided in six, five-year increments (e.g. 2000,2005,2010,2015,2020 and 2025), data points for the intervening years were interpolated. The same method is followed in the commercial and industrial sectors.

16

20

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 21 of 49

Figure 14
Actual and Projected Electricity Demand in the Residential Sector 60,000 50,000 40,000 GWh 30,000 20,000 10,000 0
20 90 93 96 99 02 05 11 08 14 17 20 19 19 19 19 20 20 20 20 20 20 20 23

Actual Electricity Demand

IPSP Forecast

Year

Looking at electricity demand per household (Figure 15), the IPSP anticipates a modest decline over the forecast period, reaching 8.5 MW.h per household by 2025. Demand per household in the IPSP forecast for the overlap period of 2000-2005 is slightly lower than actual demand per household not because of differentials in electricity demand but rather because the number of households in the IPSP forecast is larger than the number of households in the OEE database.17 Despite this discrepancy, the IPSP forecast anticipates a slowing down and ultimately a reversal of the pronounced trend towards declining energy use per household observed from 1990-2005. From 1990-2005, electricity use per household declined on average by 1.61% per year. Demand
The IPSP states that its economic growth projections (including forecasts for households) are based on growth rates found in the NRCan Energy Outlook 2006 (see D-4-1, Attachment 6, pg. 20). However, investigation revealed that the IPSP household forecast does not appear to correspond to data in the source document. For example, the IPSP forecasts 4.96 million households in 2005, while the NRCan Energy Outlook 2006 forecasts 4.73 million. See Canadas Energy Outlook: The Reference Case, 2006, NRCan, Analysis and Modelling Division, pg. 183. The origin of the household data in the IPSP is therefore unclear.
17

21

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 22 of 49

per household under the IPSP forecast is expected to decline on average by 0.24% per year. Moreover, demand per household under the IPSP forecast is expected to begin increasing in 2020, albeit very marginally.

Figure 15
Actual and Projected Electricity Demand per Household

14 12 10 8 6 4 2 0
19 93 20 05 20 08 20 02 20 14 20 11 20 17 19 90 19 96 19 99

MWh

Actual demand per household (MWh)

IPSP Forecast

Year

In terms of the end use distribution of residential electricity use in the OPA forecast, it identifies thirteen end use categories, which generally correspond with the twelve end use categories used by the OEE.18 This relatively seamless overlay facilitates between historical end use demand and projected end use demand under the IPSP forecast. As mentioned above, the IPSP projects that residential sector electricity use will increase by approximately 11,900 GW.h from 2000-2025. Over 99% of this increase is expected to come
The 12 overlapping end use categories are: central space cooling, clothes dryer, clothes washer, dishwasher, freezer, lighting, other/minor appliances, range, refrigerator, room space cooling, space heating and water heating. The 13th standalone category identified in the IPSP but not in the OEE database is for electric furnace fans.
18

20 20

20 23

22

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 23 of 49

from just two end uses: lighting and other appliances. Figure 16compares actual and projected electricity use from lighting, while Figure 17 looks at demand from other appliances.

Figure 16
Actual and Projected Electricity Demand for Lighting

12,000 10,000 8,000

GWh

Actual Lighting Demand

6,000 4,000 2,000 0

90 93 99 02 05 08 11 14 17 19 19 19 20 20 20 20 20 19 20

Forecasted Lighting Demand

Year

Insofar as lighting is concerned, the IPSP assumes a slight increase in the penetration of compact fluorescent lighting. This assumption has been overtaken by the federal policy to phase out incandescent lighting by 2012. The IPSP reference case does not address the effects of this policy shift because it does not take into account any conservation and demand management (CDM) programs implemented after 2000. While the government plan does not make it illegal to use incandescent bulbs, the IPSP reference forecast is inconsistent with expected technological trends in that it uses a constant figure of 0.055 MW.h per incandescent bulb across the entire forecast period. Since this figure is higher

20

20

20

23

96

23

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 24 of 49

than the new standard for incandescent bulbs that will take effect in 2012, the IPSP reference case almost certainly overstates electricity demand for lighting. The specific impact on electricity demand of a wholesale transition from incandescent bulbs to more energy efficient alternatives such as compact fluorescents is far from trivial. Shifting 50% of the number of forecasted incandescent bulbs in 2025 to compact fluorescent bulbs would trim a full 3,500 GW.h from the IPSP reference case forecast for residential sector electricity demand in 2025 (6.5% of total forecasted 2025 demand for residential sector). This conforms with the success of similar government programs in other jurisdictions and may understate the case. According to internal ICF models, compact fluorescent bulbs could overtake incandescent bulbs in the Canada-wide residential lighting market as early as 2014. In the IPSP reference case, incandescent bulbs command a 94% share of the total 2025 light bulb stock with compact fluorescents capturing 5.8%.19

19

The remaining 0.2% is captured by light emitting diodes.

24

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 25 of 49

Figure 17
Actual and Projected Electricity Demand for Other Appliances 16,000 14,000 12,000 10,000
GWh

Actual Other Appliances Demand

8,000 6,000 4,000 2,000 0

90 93 96 99 02 05 08 11 14 17 20 19 19 19 19 20 20 20 20 20 20 20 20 23

Forecasted Other Appliances Demand

Year

Figure 17 illustrates the steep growth trajectory anticipated in the IPSP reference case for electricity use from other appliances. Other appliances are by far the largest contributor to the increase in residential sector electricity demand forecast in the IPSP. Of the 11,900 GW.h increase forecast over 2000 to 2025, fully 8,470 (71%) comes from other appliances. Thus the forecast for the residential sector as a whole depends greatly on anticipated demand for these appliances. While the basis for residential electricity growth in the OPA forecast rests almost totally on the growth of electricity consumption of other appliances, the supporting analysis in the MKJ/Marbek calibration is particularly weak. Per household electricity use in this category grows from 1,500 kW.hours to 2,400 kW.hours over the forecast period, as shown in Figure 18, but the basis for this projection is unclear. We know that the stock of miscellaneous electricity using devices will grow during this period, but how much the related electricity will grow, if at

25

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 26 of 49

all, depends on how much they are used and per device use will tend to decline as the number of devices per household increases, even as household size itself continues to decline. To the extent standby losses are an important contributor to this end use category, it is also worth noting that there is an intense international research and technological development effort underway to address this issue, an effort that will likely have a significant impact on the time scale of this forecast.

Figure 18
Actual and Projected Electricity Demand from Other Appliances per Household 2.5

2 MWh per household

1.5

Actual electricity demand from other appliances per household Forecasted electricity demand from other appliances per household

0.5

14

17

11

20 20

96

99

90

93

02

05

08

20

20

19

19

20

19

19

20

Another issue with respect to the MKJ/Marbek calibration of the forecast has to do with the importance of the end use calibration of historical electricity use both in providing a sound basis for projecting future electricity use and as a reliable basis for DSM potential analysis. In and end use forecast such as the OPA forecast, in which different end uses are forecast at different rates, if the initial allocation of electricity use to end use categories is wrong, then the forecast will be growing electricity demand at the wrong rate. For example, if the initial calibration assigns too little electricity to an end use that has a relatively slow growth rate, and assigns it instead to an end use that has a relatively high growth rate, then the forecast will be high.

20

Year

20

20

23

26

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 27 of 49

There is a compounding problem with end use misallocation, and that is the effect it has on any subsequent estimate of DSM potential. For example, if too electricity use is allocated to an end use that has a relatively high DSM potential, and instead is allocated to an end use with relatively low DSM potential, then when the DSM potential analysis is done, it will tend to underestimate the overall potential for DSM to moderate the forecast. In the MKJ/Marbek residential electricity forecast, an example of this type of compounding error occurs with respect to electricity use in furnace fans. The MKJ/Marbek calibration assumes only about a million Ontario households have furnace fans, with total base year electricity use of 1,200 GW.hours, averaging about 1,200 kW.hours per fan. However, there are some 3,000,000 houses in Ontario with furnace fans, three times more than the MKJ/Marbek estimate, so a figure of 3,600 GW.hours seems a more reasonable base year number for this end use. (With the increasing use of these fans for air circulation and as part of central cooling systems, the 1,200 kW.hour average figure may also bee too low, further exacerbating the calibration error for this end use.) Because the base year is calibrated to actual electricity use, the missing furnace fan electricity use (the difference between 3,600 GW.hours or more and the MKJ/Marbek estimate of 1,200 GW.hours) must be misallocated to other end uses. The total number of furnace fans in Ontario grows at the rate of new houses that install forced air heating systems; if the base year electricity use that should have been assigned to furnace fans has been misallocated to end uses that grow faster than this (e.g. other appliances) the resulting forecast will be high. When the error is as large as it appears to be in the case of the residential furnace fans, the effect on the overall forecast can be significant. The furnace fan case also illustrates the problem that end use misallocation can have on subsequent analysis of DSM potential. The new ECM motors reduce furnace fan electricity use by 75%;20 although evidently not captured in the MKJ/Marbek DSM technology database, the residential furnace fan has a very high potential for electricity improvement. Because it is also one of the largest residential electricity end uses (probably the largest in homes that do not have electric space or water heating), the large under-allocation of furnace fan electricity use in the OPA forecast means that the DSM potential for this end use will also be underestimated. In a case like this, where the end use is large and has above average efficiency improvement potential, the resulting underestimation of DSM potential can be significant.

Gusdorf, J., Hayden, S., Entchev, E., Swinton, M., Simpson, C., and Castelian, B. Final Report on the Effects of ECM Furnace Motors on Electricity and Gas Use: Results from the CCHT Research Facility and Projections. Canadian Centre for Housing Technology, NRCC-38500, 2003.

20

27

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 28 of 49

Commercial Sector
Structure of Electricity Demand
Electricity demand in the commercial sector is driven by a number of different factors. The variable with the tightest historical correlation to actual electricity demand has been commercial sector floor space, however other important determinants of electricity use include commercial sector GDP and the proportion of commercial institutions using electricity for space heating. Like energy use in the residential sector, energy use in the commercial sector is dominated by natural gas and electricity (see Figure 19). In 2005, 49% of total energy use came from natural gas, while 42% came from electricity.

Figure 19
Share of Total Energy Use in Ontario Commercial Sector by Energy Source, 1990-2005 Natural Gas

60% 50% 40% 30% 20% 10% 0%

90 96 92 94 02 98 19 19 19 19 19 00 20 20 20 04

Electricity

Light Fuel Oil and Kerosene Heavy Fuel Oil Steam

Other

Year

Segmenting energy by end use (Figure 20), we see that space heating accounts for the largest proportion of energy use, representing 52% of total energy demand in 2005. Figure 20 provides 28

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 29 of 49

little evidence of any expected changes in the composition of end use demand going forward, although the increased share of space cooling over 2004-2005 could be an indication of growing future demand for this end use.

Figure 20
Share of Commercial Sector Energy Use by End Use, 1990-2005 140,000 120,000 100,000 GWh 80,000 60,000 40,000 20,000 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Street Lighting Space Cooling Lighting Auxiliary Motors Auxiliary Equipment Water Heating Space Heating

Year

The largest user of electricity in the commercial sector at the sub-sector level is offices21, which, in 2005, accounted for nearly 40% of total electricity consumption in the sector (see Figure 21).

Offices includes activities related to finance and insurance; real estate and rental and leasing; professional, scientific and technical services; and public administration.

21

29

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 30 of 49

Figure 21
2005 Commercial Sector Electricity Use by Sub-Sector
Accommodation and Other Services Food Services 2% 7% Wholesale Trade 6%

Arts, Entertainment and Recreation 3%

Retail Trade 15%

Health Care and Social Assistance 8%

Transportation and Warehousing 4%

Educational Services 14%

Information and Cultural Industries 2%

Offices 39%

Historical trends
Electricity demand in the commercial sector increased from approximately 5,000 GW.h in 1958 to 53,000 GW.h in 2005 (see Figure 22) . In 2005, the sector accounted for 37% of total residential electricity use (compared to 31% for residential and 32% for industrial). This represents an increase in share from 1990 when the commercial sector accounted for 30% (compared to 35% for both residential and industrial). The commercial sector is thus the only sector whose electricity demand consistently grew during the 1990-2005 period, both in absolute terms and in terms of relative contribution to overall electricity demand in the province.

30

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 31 of 49

Figure 22
Electricity Final Demand in Ontario in the Commercial Sector, 1958-2005
60,000

50,000

40,000

GWh

30,000

20,000

10,000

0
98 96 86 76 74 64 58 60 62 66 68 70 72 78 80 82 84 88 90 92 94 00 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20 02 04

Year

Figure 23 examines the historical relationship among the main drivers of commercial sector electricity demand, including total commercial floor space, Ontario GDP, commercial sector GDP, and actual electricity use. This analysis shows that electricity demand has traditionally tracked very closely to commercial sector floor space. From 1990-2005, commercial sector electricity demand grew on average by 1.72% per year, while floor space grew at 1.75% per year. The graph also shows how electricity demand diverged from economic output (measured in terms of Ontario GDP and commercial sector GDP) during the historical period.

31

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 32 of 49

Figure 23
Relative Growth of Commercial Electricity Use and Key Drivers
1.80 1.60 1.40 1.20 1990=1 1.00 0.80 0.60 0.40 0.20 0.00
90 91 92 93 94 95 96 97 01 98 99 00 02 03 04 19 19 19 19 19 19 19 19 19 20 20 19 20 20 20 20 05
Actual Electricity Demand Commercial Sector GDP (1997 millions) Ontario GDP (1997 millions) Total Floor Space (million m2)

Year

32

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 33 of 49

Figure 24
Relative Growth of Key Ratios in the Commercial Sector
1.4 Commercial Sector GDP (1997 millions) per million m2 of commercial sector floor space Electricity Demand (GWh) per million m2 of commercial sector floor space Commercial Sector GDP (1997 millions) per GWh of electricity demand
20 02 19 92 19 94 19 96 19 98 20 04 20 00

1.2

1990=1

0.8 0.6

0.4

0.2

0
19 90

Year

Figure 24 shows that the electrical productivity of the commercial sector (measured in terms of commercial sector GDP produced per unit of electricity) increased significantly during the 19902005 period. In 1990, 1 GW.h of electricity use generated $4.8 million in commercial sector GDP (in constant 1997 dollars). In 2005, the same amount of electricity produced $6.1 million in commercial sector GDP (in constant 1997 dollars), a 27% increase. Counterevidence to this trend occurred during the bear market of 2001-2002, where electrical productivity declined sharply. Since that time, however, the decoupling of electricity demand and economic growth has continued. Figure 24 also illustrates how electricity demand per unit of commercial sector floor space remained essentially flat from 1990-2005, with the exception of a notable dip during the economic slump of the early 2000s. Segmenting total electricity use in the commercial sector over the 1990-2005 period by end use yields a number of insights into how consumer demands for electricity are changing (Figure 25). The most notable trend has been the growth in electricity demand for space cooling, which grew by 260% during the period, increasing from 4,900 GW.h in 1990 to over 16,000 GW.h by 2005.

33

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 34 of 49

An equally compelling trend has been the steady growth in electricity use for auxiliary equipment22, which increased by 86% from 1990-2005 (average annual growth of over 4%). The most dramatic decreases over the period occurred for water heating (-8% average annual growth), space heating (-7% average annual growth) and lighting (-2% average annual growth), while growth was flat for auxiliary motors (-1% average annual growth) and street lighting (0.2% average annual growth). In the case of water heating and space heating, the decreases were largely a function of electricitys eroding market share at the expense of natural gas (natural gas increased its share of the water heating market from 77% in 1990 to 83% in 2005 and the space heating market from 77% in 1990 to 82% in 2005).

Figure 25

Commercial Sector Electricity Use by End Use

20,000 18,000 16,000 14,000 12,000 GWh 10,000 8,000 6,000 4,000 2,000 0
90 92 94 96 98 00 19 19 19 19 19 20 02 20 20 04

Space Heating Water Heating Auxillary Equipment Auxillary Motors Lighting Space Cooling Street Lighting

Year

Auxillary equipment includes stand-alone equipment powered directly from an electrical outlet, including computers, photocopiers, refrigerators and desktop lamps.

22

34

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 35 of 49

When interpreting trends in end use electricity demand in this sector, it is important to understand that in most (but not all) cases, data for 2005 diverged sharply from the historical pattern that had been established up until that point. Nowhere is this more evident than for water heating. Looking at 1990 to 2005 data, electricity demand for water heating dropped 71%, from 760 GW.h to 225 GWH, for an average annual decline of approximately 8%. However, based on 1990 to 2004 data, electricity use for water heating increased by 19% and grew, on average, by over 1% per year. Similar interpretation issues are at play when assessing trends in end use demand for lighting, which, depending on the termination year, either decreased by 2% (on average) per year, or increased by 0.3% (on average) per year.

Analysis of the IPSP Forecast

Under the IPSP reference case forecast, commercial sector electricity demand is expected to grow from 47,980 GW.h in 2000 to 64,430 GW.h in 2025, an increase of 16,450 GW.h or 34%. Figure 26 shows actual electricity use in the commercial sector from 1990 to 2005 and projected use from 20002025 in the IPSP forecast. The average annual growth rate during the 1990-2005 period was 1.72%, while the average annual growth rate across the IPSP forecast is 1.19%.

Figure 26
Actual and Projected Electricity Demand in the Commercial Sector 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0
19 90 19 99 19 93 20 02 19 96 20 11 20 05 20 08 20 20 20 14 20 23 20 17

GWh

Actual Electricity Demand

IPSP Forecast

Year

35

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 36 of 49

As demonstrated in Figure 27, the IPSP forecast anticipates a steady decline in the electrical intensity of the commercial sector, measured as electricity use per million m2 of commercial floor space.

Figure 27
Actual and Projected Intensity (GWh per million m2 of floor space) in the Commercial Sector 250

200

150
GWh

Actual intensity

100

50

Forecasted intensity

0
1990 1993 1996 1999 2002 2005 2008 2011 2014 2017 2020 2023 Year

Industrial Sector
Structure of Electricity Demand
The structure of electricity demand in the industrial sector is distinct from that in the residential and commercial sectors. The industrial sector can be further divided into various sub-sectors, and electricity demand in each of these sub-sectors is driven by a unique set of drivers. In general, electricity use in the industrial sector can be said to track physical output measures, such as Mt of pulp and paper or Mt of mineral ores.

36

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 37 of 49

The largest single contributor to energy use in the industrial sector is natural gas (see Figure 28). Electricity commands a lower proportional share of the energy market in the industrial sector than in either the residential or commercial sectors.

Figure 28
Share of Total Energy use in the Industrial Sector by Energy Source, 1990-2005
45% 40% 35% 30% 25% 20% 15% 10% 5% 0%
90 92 94 96 98 00 02 19 19 19 19 19 20 20 20 04

Natural Gas Electricity Diesel Fuel Oil, Light Fuel Oil and Kerosene Heavy Fuel Oil Still Gas and Petroleum Coke LPG and Gas Plant NGL Coal Coke and Coke Oven Gas Wood Waste and Pulping Liquor Other

Year

Unlike the share of natural gas in the residential and commercial sectors, the share of natural gas use in the industrial sector declined over the historical period. In 1997, natural gas accounted for nearly 40% of energy use in the sector, while its share by 2005 had dropped to 34%. This decrease was offset by an increase in the share of still gas and petroleum coke and, to a lesser extent, wood waste and pulping liquor. In terms of energy consumption by individual industries, Figure 29 shows that the composition of energy demand did not change substantively from 1990-2005, with the Other Manufacturing, Iron and Steel and Pulp and Paper industries accounting for over 60% of total energy use.

37

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 38 of 49

Figure 29
Share of Energy Demand by Industry 300,000 250,000 200,000 GWh 150,000 100,000 50,000 0
90 92 94 96 98 19 19 19 19 19 00 02 20 20 20 04

Other Manufacturing Iron and Steel Pulp and Paper Petroleum Refining Cement Chemicals Mining Construction Smelting and Refining Forestry Year

Historical trends
Electricity demand in the industrial sector increased from approximately 15,500 GW.h in 1958 to 43,000 GW.h in 2005. In 2005, the sector accounted for slightly over 30% of total electricity use in Ontario, the smallest share of all three sectors. The share of industrial sector electricity use in Ontarios total electricity demand has declined unabatedly since the late 1950s. Reflecting the economic challenges facing Ontarios heavy industry sector, electricity use in the industrial sector in 1990 was approximately 3,000 GW.h higher than in 2005. The industrial sector is the only sector whose electricity demand declined substantially during the 1990-2005 period.

38

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 39 of 49

Figure 30
Electricity Final Demand in Ontario in the Industrial Sector, 1958-2005
50,000 45,000 40,000 35,000 30,000

GWh

25,000 20,000 15,000 10,000 5,000 0

86

92

00

84

88

90

94

96

98

58

60

62

64

66

68

70

72

74

76

78

80

82

19

19

20

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

19

Industrial electricity use and some its key drivers are illustrated in Figure 31. Industrial sector GDP began to diverge from total GDP during the economic slump of the early 2000s and is expected to contribute an increasingly small share of total GDP going forward (as commercial sector activity takes its place as the main economic growth driver in the province).

19

Year

20

20

02

04

39

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 40 of 49

Figure 31
Actual and Projected Growth of Industrial Electricity Use and Key Drivers
1.8 1.6 1.4 1.2 1990=1 1 0.8 0.6 0.4 0.2 0
GWh of electricity demand per Industrial Sector GDP (1997 millions) Electricity Demand (GWh) Industrial Sector GDP (1997 millions) Ontario GDP (1997 millions)

90

94

92

96

98

00

19

19

19

19

19

20

20

Year

As in the commercial sector, electricity demand in the industrial sector did not track economic output (whether measured as total GDP or industrial sector GDP) during the historical period, with the divergence between the two sets of activities beginning in 1990. This relationship speaks to the growing electrical productivity of the Ontario economy identified earlier. This relationship is also borne out by the downward trajectory of electricity demand per unit of industrial sector GDP shown in Figure 31.

20

04

02

40

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 41 of 49

Figure 32
Electricity Use by Industry
20,000 18,000 16,000 14,000 12,000 GWh 10,000 8,000 6,000 4,000 2,000 0
19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05
Iron and Steel Other Manufacturing Mining Pulp and Paper Smelting and Refining Petroleum Refining Cement Chemicals

Year

Focusing on electricity use at the sub-industry level, Figure 32 shows the historical trajectory of electricity demand in the eight industries catalogued by the OEE.23 Measured in absolute terms, electricity demand did not change significantly over the period in any industry, with the exception of other manufacturing24, where demand plummeted from 15,900 GW.h in 2001 to 13,900 GW.h in 2004. Measured in percentage terms, however, electricity demand in most industries underwent dramatic changes (see Figure 33). Electricity use in the chemicals industry declined by nearly 35% between 1990-2005 while, at the other end of the spectrum, electricity demand in the smelting and refining industry grew by 88%.

Since the construction and forestry industries did not use electricity during the historical period, they are excluded from this analysis. 24 The Office of Energy Efficiency uses other manufacturing as a residual category.

23

41

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 42 of 49

Figure 33
Percentage Change in Electricity Demand by Industry, 1990-2005

Chemicals Mining Other Manufacturing Petroleum Refining Cement Pulp and Paper Iron and Steel Smelting and Refining -60% -40% -20% 0% 20% 40% 60% 80% 100%

Comparing changes in electricity demand with changes in industry-specific GDP (see Table 1) shows that two industries cement and other manufacturingexperienced negative growth in electricity demand and positive growth in economic output, reflecting improved electrical productivity (or reduced electrical intensity) over the period. The divergence is particularly striking in other manufacturing. Far from declining electrical intensity, two industries (pulp and paper and smelting and refining) faced rising electricity demand and declining economic output over the period, implying declining economic utility for each GW.h of electricity.

42

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 43 of 49

Table 1
Growth in Electricity Use, 1990-2005 0.00% 4.31% 88.54% -8.96% -3.13% -34.28% 38.60% -12.58% 0.00% -28.56% Growth in GDP (millions 1997$), 1990-2005 18.35% -6.11% -59.91% -31.96% 5.59% -7.08% 7.12% 73.42% 15.96% -16.36%

Construction Pulp and Paper Smelting and Refining Petroleum Refining Cement Chemicals Iron and Steel Other Manufacturing Forestry Mining

Analysis of the IPSP Forecast

Under the IPSP reference case forecast, industrial sector electricity demand is expected to grow from 43,650 GW.h in 2000 to 58,800 GW.h in 2025, an increase of 16,000 GW.h or 35%. The IPSP thus anticipates more growth in electricity use in the industrial sector (both in absolute and percentage terms) than in either the residential or commercial sectors. This forecast is interesting in light of the fact that industrial electricity demand experienced negative growth over the 19902005 period (the only sector in which negative growth took place). The average annual growth rate from 1990-2005 was -0.42%, while the average annual growth rate of the IPSP forecast is 1.2%.

43

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 44 of 49

Figure 34
Actual and Projected Electricity Demand in the Industrial Sector 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0
90 96 99 93 05 14 17 02 08 19 19 19 19 11 20 20 20 20 20 20 20 20 20 23

GWh

Actual Electricity Demand

IPSP Forecast

Year

The overwhelming majority (87%) of the forecasted growth comes from the other manufacturing sector.25 Figure 35 shows how the IPSP forecast failed to anticipate the pronounced decline in other manufacturing electricity demand that begin in 2000. Since this is base year of the IPSP forecast, the IPSP forecast assumes that growth (however marginal) that had occurred up until that point would continue. This explains the jump that occurs in 2005 between actual electricity demand and the IPSP forecast.

Unlike the OEE, which simply defines other manufacturing as a residual sector, the IPSP states that other manufacturing is comprised of the following sub-sectors: electronics and other; transportation equipment; furniture, printing and machinery; wood products; leather, textiles and clothing; rubber and plastics products; and food, tobacco and beverage.

25

44

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 45 of 49

Figure 35
Actual and Projected Electricity Demand in the Other Manufacturing Sub-Sector 35,000 30,000 25,000 20,000 15,000 10,000 5,000 0
11 17 20 90 93 08 99 02 05 96 14 20 19 19 19 19 20 20 20 20 20 20 20 23

GWh

Actual Electricity Demand

IPSP Forecast

Year

Figure 36 tracks actual GDP growth in the other manufacturing sub-sector and projected growth in the IPSP forecast.

45

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 46 of 49

Figure 36
Actual and Projected GDP Growth in the Other Manufacturing SubSector 160,000 140,000 120,000
GDP ($1997 millions)

100,000 80,000 60,000 40,000 20,000 0

19 90 19 93 19 96 19 99 20 02 20 05 20 08 20 11 20 14 20 17 20 20 20 23

Actual GDP

IPSP Forecast

Year

By assuming a relatively high growth rate for electricity demand and a relatively low growth rate for GDP, the IPSP forecast diverges sharply from the historical pattern over the 1990-2005 period when it comes to the productivity of the other manufacturing sub-sector, measured in terms of GDP per unit of electricity demand (see Figure 37). The increasing electricity productivity of the other manufacturing sector is one of the most visible and important trends to emerge in industrial sector electricity use over the 1990-2005 period. Other manufacturing productivity nearly doubled over the historical period, growing from 2.7 in 1990 to 5.4 in 2005. During the 5 year overlap period (2000-2005) for which we have actual data and forecasted numbers from the IPSP, we can see that the IPSP dramatically underestimated the productivity of this sub-sector. While the historical trend line points toward continued productivity gains in this sub-sector, the IPSP reference case anticipates a flattening out of productivity in other manufacturing followed by declining productivity beginning in 2020.

46

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 47 of 49

Figure 37
Actual and Projected Productivity in the Other Manufacturing SubSector 7
GDP ($1997 millions) per GWh of Electricity Use

6 5 4 3 2 1 0
1990 1993 1996 1999 2002 2005 2008 2011 2014 2017 2020 2023 Year

Actual productivity

IPSP Forecast

47

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 48 of 49

Conclusion
Previous long term power sector expansion plans for Ontario were rendered useless (and their associated public reviews irrelevant) largely because they were based on long range forecasts that incorrectly assumed (or adopted assumptions that led to the conclusion that there would be) a reversal of the longstanding downward trend in the electricity demand growth rate. The OPA forecast and the associated MKJ/Marbek repeat this pattern, supporting a forecast of Ontario electricity demand in which growth rates stop falling, turn around, and then accelerate through the forecast period. The absolute amount of growth in the OPA forecast (some 40 TW.hours over a twenty year period) is lower than the absolute growth in the previous ill-fated long range forecasts of electricity demand in Ontario, and unlike in previous planning rounds, is now comparable in size to the technical potential for DSM. This makes it all the more important to develop the capacity (models, analysts, databases) to be able to take an integrated approach to the future demand for electricity, in which the forecast and the DSM potential estimates are no longer separate and sequential exercises. The end use calibration of the OPA forecast does not provide a convincing case that history will not repeat itself in this current round of long range electric power planning. An examination of the MKJ/Marbek end use calibration of the end use forecast indicates a number of instances where the demand growth derives from assumed or unsubstantiated departures from historical trends with respect to the growth of the activity drivers, with respect to the relationship between the activity drivers and electricity demand, and with respect to the relative growth rates of end uses with variable natural conservation potential. The electricity growth in the residential and commercial sectors is highly concentrated in a couple of end uses -- almost all residential electricity growth is in the other appliance category and nearly 90% of commercial sector electricity growth is for lighting, but the underlying justification for this lopsided distribution of growth is not convincing. In the industrial sector the forecast growth rests on the assumed departure from recent trends toward greater electricity productivity, and instead assumes deterioration in electricity productivity. Even accepting the forecast level of electricity demand, the end use calibration of the electricity demand underlying the MKJ/Marbek in at least some instances has under-allocated electricity to important end uses with high DSM potential (e.g. residential furnace fans) at the expense of end uses where the end use and related DSM potential is poorly understood and assumed to be below average (e.g. residential other appliances). This results in an end use calibration that sets up a DSM potential analysis that will return too low a result as the end use with the high DSM potential is underrepresented in the forecast.

48

Filed: August 14, 2008 EB-2007-0707 Exhibit L Tab 8 Schedule 2 Page 49 of 49

An overestimate in the forecast will also result in an overestimate in the DSM potential (one cannot apply DSM to a kilowatt. hour of demand that never materializes) but the impact on the aggregate DSM potential is dominated by opportunities to save electricity (and peak) from the existing demand, and so the impact of a lower forecast on the DSM potential estimate is dampened considerably. If the long range decline in electricity demand growth rates in Ontario continues through this forecast period, very little if any of the 40 TW.hours of growth in this forecast would come about. The potential for DSM would also be reduced, but only to the portion of the DSM potential that derives from load growth. If the forecast growth in Ontarios electricity demand were not to materialize, the corresponding technical potential for DSM would decline by roughly 25% as compared to the potential estimated relative to the OPA forecast.

49

AUGUST 2010

Its time for Ontario to upgrade its green energy plans.

Ontarios Green Energy Plan 2.0

Choosing 21st Century Energy Options

Executive Summary

Should the province replace its aging nuclear plants with new nuclear stations or with a portfolio of green energy options?
As Ontarios aging nuclear reactors are retired over the next 15 years, the province must replace its out-dated 20th century electricity system with a new one that will power the province into the 21st century.
Until now, the province has pursued two goals building new nuclear reactors while expanding its supplies from renewable energy sources. As long as electricity demand was growing, and renewable sources were marginal, both goals seemed compatible. But thats no longer the case. The situation in Ontario has drastically changed. The provinces electricity demand has been falling for the last four years and will likely continue to decline. At the same time, the success of the Green Energy Act has led the province to procure more wind, solar and bio-energy than expected over the next 17 years. With electricity demand falling, coal being phased out, and the natural gas capacity that will be used to balance the system in the medium-term largely already in place, Ontario faces a stark choice: Should the province replace its aging nuclear plants with new nuclear stations or with a portfolio of green energy options? New nuclear reactors will limit future green energy investments in Ontario by constraining transmission access and overall grid flexibility. By replacing aging nuclear with modern green energy, Ontario would realize the following benefits: Greater savings: Building a new nuclear plant will cost ratepayers anywhere from 12 to 48 per cent more than delivering that same amount of power using a mix of renewable and more efficient options.

HOME SOLAR PANEL

Long-term green jobs: A green energy portfolio would create an additional 27,000 new jobs over a 10 year period. Proven: All of the new nuclear reactor designs being considered by the Ontario government are untested prototypes. Green energy technologies have been proven to work in other jurisdictions to replace significant sources of old electricity production, and they are increasingly technologically and cost efficient. Flexible options: Nuclear power plants are designed to run at full output all the time and are not responsive to changes in demand. Renewable sources, meanwhile, can be incrementally developed faster and in smaller increments, which more appropriately matches gradual changes in supply and demand. Protection: Ontarians are still paying for the cost overruns for reactors built decades ago. Ontarios progressive Green Energy Act, however, protects electricity consumers by requiring green power developers to pay for cost overruns. This report makes the case for the green portfolio as the more cost-effective option and the one which best prepares this province to prosper in the 21st century. Investing in renewable energy systems is the most effective way to achieve the government's economic and environmental goals while providing a sustainable legacy for future Ontarians.

The New Context: Green Power Exceeding Expectations

Just a few years ago we were told it was impossible for green power to replace aging nuclear stations, such as Pickering. Today, however, its clear we can.
In 2005, the provinces energy planners told the Ontario government that new or re-built nuclear reactors were more cost-effective while green power and conservation were either too expensive or unreliable. In retrospect, it is clear the government received bad advice. The estimated cost of building new nuclear plants has almost tripled and reactor refurbishment projects have gone substantially over budget and suffered significant delays. Ontarios ground breaking Green Energy Act has positioned the province as North Americas leader in renewable power development. In 2009, Ontarios 1,000 MW of wind power produced 2.3 terawatt hours4 of electricity equivalent to the power used in over 400,000 houses every year, while the output from Ontarios coal plants was down to 8.9 terawatt hours. In the last six months, the province has contracted for an additional 4,800 MW of new renewable energy generation to be built within the next five years under the Green Energy Act, which would generate roughly 11.4 terawatt hours annually. At the same time, Ontario Power Generation (OPG) is proceeding with plans to convert some of its coal burning units to produce 2 terawatt hours annually from biomass. In total, Ontario has already procured more green energy in 2010 than it expected to over the next 17 years. (For a summary of all green energy contracted so far and a comparison with the governments 2007 electricity plan see the Appendix.)

Despite this new landscape, the provincial government has not While the provinces nuclear energy projects have failed to formally changed its plan to resuscitate the provinces aging meet cost targets, the growth of the cleantech sector has nuclear supply. Staying the nuclear course will expose Ontarians to exceeded expectations, both increasing costs while delaying internationally and in Ontario. and diverting investment in The thriving green energy green energy. industry has confounded its critics and surpassed even the In 2006, the Ontario government gave the Ontario Power Authority But it doesnt have to be most optimistic projections. (OPA) minimum targets for renewable development over the next that way. The retirement of 20 years, while giving it a maximum for nuclear development.13 In 2008, the United Nations the antiquated Pickering But in the OPAs 2007 energy plan to the Ontario Energy Board, reported that for the first time, nuclear station provides the minimum renewable targets appeared as maximums, and the more money was invested development of green power was cut short in order to ensure there an opportunity to upgrade globally in wind, solar and was enough room on the electricity grid for nuclear energy. Ontarios green energy other forms of renewable supply. Instead of replacing The reality is that once we have maximized the contribution of electricity generation than the Pickering station conservation and efficiency, nuclear and renewables are direct into new nuclear, coal and competitors for space on the grid (See the graph Nuclear Blocks with costly replacement natural gas combined.1 In Wind Expansion in the Appendix). The governments policy of reactors, a green energy spite of the global recession, replacing aging nuclear reactors with new ones is an expensive portfolio would be a logical barrier to the expansion of Ontarios green energy industry. As 2009 was an even better step in the evolution of the aging nuclear stations are retired, cost-effective green energy year.2 The International Energy provinces electricity system options should be allowed to replace them. Agency recently reported that and economy. By doing so, solar power is likely to be Ontario would be protecting cost competitive with current the health of its present and future residents. The 20,000 electricity rates by 2020 if leading governments continue to tonnes of highly radioactive waste produced by Pickering are support that maturation of solar technology in the interim years.3 still currently stored at site. While these wastes cannot be eliminated, the province can avoid producing more of it.

Leveling the playing field

"We may not need any [new coal or new nuclear], ever I think baseload capacity is going to become an anachronism."
Jon Wellinghoff, Chair, Federal Energy Regulatory Commission, April 21, 2009.5

The 3,000 MW produced by Pickering reactors represents 15 per cent of Ontarios electricity. A diverse supply of renewable sources, combined with advanced smart grid technology and demand-side management, can replace the traditional model of base and peak load. In Ontario, this can be done at a lower cost than replacing Pickering with more nuclear power.

Room to Breathe: Ontarios Falling Electricity Demand

Thanks in part to the success of the governments conservation programs, electricity demand is now expected to decrease over the next decade. Instead of building additional electricity supply, we can now focus on using modern green energy options to replace retiring nuclear stations.
In 2005, the provinces electricity planners warned that Ontarians could face blackouts due to a combination of rapidly increasing electricity demand and the shutdown of the provinces aging reactors. These dire projections led Ontario to commit to purchasing new nuclear reactors. Today, these predictions of imminent blackouts appear to have been unfounded. Instead of increasing, electricity consumption has actually been dropping since 2005, even before the global recession. Demand for electricity in Ontario is projected to continue to decrease, even as the economy recovers. The North American Electric Reliability Corporation (NERC) predicts electricity demand in Ontario will fall an average 0.7 per cent a year between 2009 and 2018. This 9.5 per cent drop in annual consumption, almost the equivalent to the output of three Pickering reactors, eliminates the predicted need for new reactors.6 After decades of energy planners assuming that demand will always rise, they are now recognizing that we can meet our energy needs with less total energy use, while still growing our economy. The demand for electricity is dropping in Ontario as a consequence of changes in the industrial structure, successful conservation programs, and the replacement of old capital with new, more efficient equipment. The Independent Electricity System Operator (IESO) says this change in its forecast of future demand is due to changes in the economy (both the high Canadian dollar and the recent recession); having conservation programs in place and planned, and increases in embedded generation (on-site power generation like rooftop solar).7 In other words, while additional nuclear generation was said to be needed to meet growing demand just a few years ago, electricity consumption can now be expected to modestly decline, eliminating the need for some of the current nuclear supply.

Demand for Electricity in in Ontario is dropping, even as population and GDP and GDP Are Increasing Demand for electricity Ontario is Dropping Even as Population are increasing
180 170

Demand for electricity in Ontario is dropping, even as population and GDP are increasing 160
150 140 130 120 110 100 90 80 70 60
1975
1983 1976 1984 1977

180

160 150 140 130 120 110 100 90 80 70 60

1975 1976

Total Energy Consumption in Ontario (TWh)

Total Energy Consumption in Ontario (TWh)

170

1978

1979

1988 1980

1989 1981

1982

1992 1983

1993 1984

1985

1986

1997 1987

1998 1988

1989

2001 1990

2002 1991

1992

1993

2006 1994

2007 1995

1996

1997

2011 1998

2012 1999

2000

2015 2001

2016 2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

1977

1978

1979

1980

1981

1982

1985

1986

1987

1990

1991

1994

1995

1996

1999

2000

2003

2004

2005

2008

2009

2010

2013

2014

2017

Actual Electricity Consumption (1975 - 2009)

Actual Electricity Consumption (1975 - 2009)

Ontario Power Authority Reference Forecast (prediction made in 2005)

Ontario Power Authority Reference Forecast (prediction made in 2005) (December 2009 prediction)

Independent Electricity System Operator Independent Electricity System Operator (December 2009 prediction)

2018

2018

Nuclear: Over Budget (Again)

The nuclear industry has a history of low-balling its cost estimates. This makes reactor projects appear economical.
The up-front costs of building nuclear plants are high. It is also complicated work that often takes longer than anticipated. Historically, Ontario has had bad luck when constructing nuclear stations. Not a single reactor in Ontarios history has ever been built on time or on budget. Projects to refurbish old reactors have also been late and significantly over budget. Not long ago, Ontarians were told that nuclear reactors would be inexpensive to construct. In 2005, the Ontario Power Authority (OPA) assumed a new nuclear plant would cost about $2,900/ KWh or about $6 billion for a 2,000 MW station. After the government suspended its procurement of new reactors in 2009, it was reported that the cost to build AECLs untested Advanced CANDU reactor was over $10,000/kW or $26 billion for a 2400 MW station.8 At this price, building two new reactors would consume the provinces entire 20-year nuclear budget, which included the reconstruction of 12 reactors, in addition to the new ones. The OPA has admitted that building new reactors at these levels would not be cost effective.9 Maintaining aging reactors is also expensive. In August 2005, Ontario abandoned restarting two of the Pickering A reactors because of high costs. In response, Bruce Power CEO Duncan Hawthorne said his companys ability to come in on time and on budget refurbishing two of the Bruce A reactors would be a test case for future nuclear projects in Ontario, saying If we cant do this, dont talk nuclear again in this province.10 Bruce Power failed to deliver. By 2008, the restart was more than a year behind schedule and Ontario ratepayers are on the hook to pay $237.5 million in cost overruns.11 In addition to the overruns at Bruce A, the refurbishment of the Point Lepreau nuclear station in New Brunswick is significantly delayed and overruns have cost the federal taxpayer hundreds of millions of dollars. Ontario Power Generations decision not to refurbish the Pickering B reactors is an admission that it is not economical to keep CANDU reactors running. It is currently examining the viability of refurbishing the Darlington nuclear station, but is unsure on how much it will cost. Estimates range between $6 and $10 billion.12 Given the troubling history of nuclear power in this province, Ontario is fortunate that the Green Energy Act provides an excellent foundation for its energy future. However, unless the government changes its policy of replacing nuclear with nuclear, Ontario will have to stop procuring additional green energy under the Green Energy Act.

Comparing Nuclear Cost Estimates

$12 000

Comparing Nuclear Cost Estimates

The nuclear industry has a history of underestimating the cost of nuclear plants. The bar to the left shows the actual cost for the Darlington nuclear station, the last nuclear station completed in Canada in 1992. The next estimates are early reactor vendor estimates for next generation reactor designs. The red bar in the centre is the conservative estimate used by the Ontario Power Authority in 2005 in advising the government to purchase new nuclear plants. The bars to the right show estimates provided by non-industry financial analysts. The bar on the far right is AECLs reported bid for building its Advanced CANDU in Ontario in 2009. All estimates are adjusted to 2010 dollars.

$10 000

Cost Per kW Installed (adjusted to 2010 dollars)

$8 000

$6 000

$4 000

$2 000

$Actual cost Westinghouse's of Darlington AP-1000 (last reactor built (2005) in Ontario) ACR-1000 (2005) The OPA's "conservative" cost estimate (CANDU-6, 2005) Standard & Poor (May 2007) Moody's Investment Service (October 2007) Moody's Investment Service (May 2008) Reported Bid Price for AECL's Advanced CANDU (2009)

Sources: Darlington construction costs, Ontario Hydro; Ontario Power Authority 2007; Standard & Poor and Moodys Investment Service; the Toronto Star.

Ontario Green Energy 2.0: Proven, Doable and Diverse

Its time for Ontario to upgrade its green energy plans.
The province has already made commendable progress in building a green economy. Ontario is on track to phase out its coal stations by 2014 and replace them with a mix of conservation, green energy, and cleaner gas generation. Much of this progress, however, will come to a halt if the government stays the nuclear course. The Green Energy Plan 2.0, outlined below, presents an affordable and forward-thinking option. It is less risky than buying a new nuclear station. The 3,000 MW of capacity in the six reactors at the Pickering plant currently provide about 15 per cent of Ontarios overall electricity when they are operating well. Instead of relying on new untested nuclear plants, Ontario could replace the contribution from these aging reactors to the provinces supply with a portfolio of proven hydro, wind, solar, biomass, Combined Heat and Power (CHP), conservation and efficiency options. Under the current Ontario electricity plan, retiring nuclear stations are to be replaced by new or rebuilt nuclear reactors to continue to provide up to 14,000 MW of supply, while green energy will be capped at about 5,300 MW. However, by increasing targets for renewables, green power could replace nuclear just as it has been allowed to replace coal. This diversity of supply will only benefit the province, which will no longer have to rely on this troubled and costly source. Industry has already demonstrated its confidence in the economics and potential of green energy in Ontario. A 2009 survey by the OPA found over 15,000 MW of renewable energy projects already in the planning or development stage,14 more than double what was predicted in its 2007 electricity plan. By the end 2009, the OPA had received applications for 8,000 MW worth of green energy projects.15 In early 2010, more than 500 new green energy projects had been approved across Ontario. Many of these projects will be built in communities by farmers, municipalities, businesses, and public institutions such as schools and hospitals.

A Green Energy Portfolio Can Replace the Pickering Nuclear Station

25

This chart illustrates how a portfolio of green energy sources can easily make up for the electricity produced by the deteriorating Pickering B nuclear Station.16 Ontario can already keep the lights on with less nuclear generation. The provinces electricity system is capable of handling large periods of time when many of its nuclear power stations are offline for repairs. Of the 11,300 MW of nuclear supply in Ontario, in 2009 (a relatively reliable year for nuclear power) the average nuclear output was just over 9,400 MW, while the entire system operated for 45 days at a nuclear output lower than 8,300 MW the equivalent of operating without the entire Pickering station. Ontarios green energy legislation provides many of the right conditions for conservation and renewable energy to thrive. But ifNuclear energy is ever to reach its full potential, the green government must revise its 2006 commitment to maintaining Additional Efficiency/Conservation nuclear at 50 Heat and Power per cent of supply. Otherwise, the government Combined will cause clean energy to remain a marginal source of power Biomass/Biogas/Landfill in Ontario, despite the innovative Green Energy Act.
Solar Off Shore Wind On Shore Wind
Nuclear

20

Contribution to Ontarios Annual Electricity Demands (TWh/yr)

15

10

Hydro

Additional Efficiency/Conservation

Green Portfolio (FiT)

Nuclear Additional Efficiency/Conservation Combined Heat and Power Biomass/Biogas/Landfill Solar

Combined Heat and Power Biomass/Biogas/Landfill Solar Off Shore Wind On Shore Wind Hydro

Picketing A & B

Green Energy Plan 2.0: The Affordable Choice

No one could blame Ontarians for asking this simple question: Why risk billions of dollars on untested reactors, when proven green energy can keep the lights on at lower risks and lower costs?
The revelation in 2009 that two new reactors would cost $26 billion equivalent to the total cost of the governments entire long-term nuclear spending plan proves that Ontario cannot afford to stay on the nuclear path. By allowing modern green power to replace the 1960s-era Pickering nuclear station, this green portfolio shows how the Green Energy Act could be put to work. Charting such a course is cheaper and will diversify investment in power generation. Better still, it reduces the risks of relying on an expensive form of energy generation to supply our needs. The graph below shows how building a new nuclear plant will cost rate payers 12 to 48 per cent more than delivering that same amount of power proposed in our Green Energy Plan 2.0. While more affordable, this plan also provides better protection to Ontario ratepayers. Ontarios progressive Feed-In Tariff ensures that ratepayers only pay for electricity generated, ensuring any cost overruns or unforeseen liability expenses are borne by the developer not Ontarians. This is not the case for nuclear power. Today, Ontarians continue to pay for cost overruns from reactors built decades ago. Better still, the plan protects both the provincial ratepayer and the federal taxpayer. In suspending its procurement of new reactors in 2009 because AECLs winning bid was billions of dollars too high, the Ontario government asked the federal government to lower the price; that is, it asked the federal taxpayer to subsidize its nuclear plans. Our Green Energy Plan 2.0 shows Ontario can use Green Energy Act to meet Ontarios electricity needs without a federal bailout.

"Nuclear generation has a fixed design where construction costs are rising rapidly, while other renewable technologies are still experiencing significant advancements in terms of energy conversion efficiency and cost reductions."
Moodys Investment Service, 2008.

Green Energy Plan 2.0 is conservative and assumes todays Feed-In Tariff prices. Over time, Feed-In Tariffs are intended to decline while projects that are already approved will remain fixed for 20 years. Either way, our Green Energy Plan 2.0 is already the more affordable choice at todays prices.

A Green Energy Plan Is Cheaper Than New Reactors

20 18 16

20,0

Portfolio Costs of replacing Pickering

The bar to the left shows how, using current Feed-In Tariff rates, a green energy portfolio made up of renewable energy sources will cost Ontario ratepayers less than generation from a new nuclear power station.17 The bars to the left provide recently reported cost estimates for new nuclear stations.

15,1
14

13,5
12 10 8

$/MWh Generation

6 4 2 0 Green Portfolio (FiT prices) Moodys Investment Service 2008 AECLs Reported Bid Price for the Advanced CANDU 2009

Green Jobs Plan 2.0: More Green Jobs

A Green Energy Plan 2.0 would allow Ontarios green workforce to continue growing and diversifying the provinces economy.
Aside from saving provincial ratepayers money on electricity bills, the figure below illustrates the sectors in which our suggested upgrade to the provinces green energy plans would create an additional 27,000 new jobs over a 10 year period.18 Ontario is already seeing progress being made on creating a green collar workforce. In January 2010, the province signed a $7 billion deal with the Korea-based Samsung Group. Samsung committed to building four manufacturing plants that will produce renewable technology such as wind turbines and developing 2,500 megawatts of wind and solar farms in Ontario. This investment is expected to generate more than 15,000 jobs.19 Part of the allure of Ontario for Samsung were the provinces regulations and policies such as the Green Energy Act that reward investment in renewables. The provinces domestic content requirements, for example, require at least 25 per cent of wind project costs and 50 per cent of large solar project costs to come from Ontario goods and labour. Along with guarantees in prices for energy generated from renewable sources, companies will have the confidence to invest in Ontario, hire workers, and produce and sell green energy. The growth in the green jobs sector can and should continue. A recent study by Blue Green Alliance, a coalition of environmental and labour groups, estimated that 90,000 jobs could be created with green energy over the next decade 20 by replacing aging nuclear stations with green energy as they retire. One of the major benefits of the Green Energy Act is that it allows renewable energy producers across the province to connect to the grid not just those working in a nuclear facility. Aboriginal communities, homeowners, farmers, schools, factories, co-ops, as well as large-scale commercial generators will be able to boost local economies and create jobs by selling green energy to the provinces electricity grid. In the green energy future, everybody wins. Unlike jobs in the nuclear industry, an upgraded green energy plan will bring more diverse jobs to all corners of Ontario. The province can expect to see jobs in wide-ranging sectors such as manufacturing, industrial efficiency, clean generation, home retrofitting, and offshore developments.

Distribution of Jobs Created by Replacing Nuclear Power with a Green Portfolio

800 400

1 300
1 300

800 400

5 400
5 400

Portfolio Jobs
Over the next 10 years, 27,000 new green jobs could be created to implement a portfolio equivalent to the power generated by Pickering A & B.
Hydro

6 400
6 400

Hydro On-Shore Wind On-Shore Off-Shore Wind Wind Off-Shore Wind

Solar

Biomass/Biogas/Landfill

Solar

Combined Heat and Power

Biomass/Biogas/Landfill Combined Heat and Power

Additional Efficiency/Conservation Additional Efficiency/Conservation

8 900 3 900
3 900

8 900

The Green Future Starts Now

Following the coal phase-out, the Ontario government must revise its policy on replacing aging nuclear facilities with new ones to continue developing a modern green energy economy.
Building a 21st century energy system means that Ontario must learn from its 20th century mistakes with nuclear power. Clean energy sources must be given room to grow in order to realize their potential. The Ontario governments role is to provide direction and guidance to encourage the provinces transition to a green energy future. In 2008, then-Minister of Energy and Infrastructure George Smitherman stopped the Ontario Energy Boards review of the Ontario Power Authoritys 2007 long-term electricity plan and instructed it to review and enhance its long-term targets for renewables, conservation, and decentralized energy within six months.21 At the time, the minister insisted nuclear would still remain at 50 per cent of supply, inadvertently limiting significant enhancements to green targets.22 Since that time, it has become clear that green energy can play a more significant role in Ontarios energy plan. Nuclear costs are increasing at a time when demand forecasts are decreasing, making it an unsuitable fit for Ontarios needs. The Green Energy Act could if permitted encourage renewable power to thrive by stimulating more investment in clean technology. Despite this, the current Energy and Infrastructure Minister Brad Duguid has stated that green energy development will be limited to targets set several years ago (about 10 per cent of supply)23 targets that were set when nuclear costs were believed to be significantly lower. The government needs to provide clear direction to our electricity planning agency to avoid investing in risky nuclear energy again. A 21st century energy system cannot depend on 20th century thinking. The growth of green energy in Ontario has been driven by the governments commendable and successful coal phase-out. It is replacing dirty coal power with green energy, conservation, and cleaner gas generation. There is no reason this momentum cannot continue, allowing green energy to take the next step in Ontario. Right now, Ontario has a once-in-a-lifetime opportunity to replace the six aging reactors at the Pickering nuclear station with safer and more sustainable options. By making wise decisions today, Ontario will usher in a new era of prosperity for tomorrow.

Recommendations
1. Direct the Ontario Power Authority to replace the Pickering reactors by increasing its mid-term baseline targets (between the years 2015 and 2020) for renewables, conservation, and Combined Heat and Power. 2. Forgo or delay24 buying new reactors. 3. Follow through on commitments to establish a Feed-In Tariff for Combined Heat and Power generation in order to enable the development of diversified baseload generation. 4. Instruct the Ontario Power Authority that aging nuclear facilities can be replaced by cost effective green energy options.

Highlights of Ontarios Green Energy Plan 2.0

Adopting a portfolio of renewable energy sources has numerous benefits:
Doable All the energy options in the portfolio are proven to work and can easily meet and surpass the green targets established in 2006. Diverse Instead of risking billions of dollars on an untested reactor, this green portfolio would provide power diversity from proven sources: onshore and offshore wind; local, residential, and industrial power stations; and efficiency programs. Disperse Combined Heat and Power (CHP) stations could provide efficient baseload power to hospitals, schools, and industrial facilities across Ontario instead of being centralized in a distant location. Conservative The OPA already intends to surpass its original targets for wind power for 2014. The additional wind capacity proposed here is less the OPAs own deployment estimates for 2014.25 Cost effective Feed-In Tariff rates are scheduled to be reviewed and likely decline over time for new projects, while projects that are already approved will remain fixed for 20 years. Meanwhile, nuclear power costs have continued to escalate.

Endnotes
1 Macalister, Terry, (2009). Renewables took bulk of global energy investment in 2008, says UN. The Guardian, accessed online: < www.guardian.co.uk/environment/2009/jun/03/renewables-energy-spending>, April 20, 2010. 2 Roney, J. Matthew, (2010). Wind Power Soared Past 150,000 Megawatts in 2009. Earth Policy Institute, accessed online at: <www.earthpolicy.org/ index.php?/indicators/C49/wind_power_2010>, March 30, 2010. 3 International Energy Agency. IEA sees great potential for solar, providing up to a quarter of world electricity by 2050. Media Release, accessed online at: <http://iea.org/press/pressdetail.asp?PRESS_REL_ID=301>, May 11, 2010. 4 Independent Electricity System Operator. Wind Power in Ontario Generates a New Record in 2009, Media Release, accessed online at: <http:// www.ieso.ca/imoweb/media/md_newsitem.asp?newsID=5019>, January 8, 2010. 5 Energy Regulatory Chief Says New Coal, Nuclear Plants May Be Unnecessary. New York Times, accessed online at: <http://www.nytimes. com/gwire/2009/04/22/22greenwire-no-need-to-build-new-us-coal-ornuclear-plants-10630.html>, April 22, 2009. 6 Hamilton, Tyler. Hydro use decreasing, Toronto Star, October 30, 2009. 7 Personal communication from Alexandra Campbell of the IESO with Keith Stewart of WWF-Canada, January 13, 2010. 8 Hamilton, Tyler. $26B cost killed nuclear bid, Toronto Star, July 14, 2009. 9 In response to questioning at the Ontario Energy Board, the Ontario Power Authority admitted that at an overnight capital cost of $3,600 with an additional 2 cents in operating and maintenance costs and an 8 per cent discount rate, nuclear stations would be uneconomical. See: Ontario Power Authority, EB-2007-0707, Exhibit, Tab 43, Schedule 3. 10 Erwin, Steve. Ontarios Nuclear Future at Risk If New Project Goes Over Budget: Bruce CEO, The Canadian Press, October 20, 2005. 11 Leslie, Keith & Romina Maurino. Ont. Electricity consumers on hook for Bruce Power cost overruns: critics say, The Canadian Press. April 17, 2008. 12 Minutes, 2011/2012 Regulated Facilities Payment Amounts Stakeholder Information Session 2. Ontario Power Generation, April 1, 2010. 13 Ontario Energy Board. Report of the Board on the Review of, and Filing Guidelines Applicable to, the Ontario Power Authoritys Integrated System Plan and Processes, pp. 56, accessed online at: <http://www.oeb.gov. on.ca/documents/cases/EB-2006-0207/IPSP_report_final_20061227. pdf>, May 14, 2010. 14 Survey results available online at: <http://www.powerauthorigy.on.ca/fit/ Page.asp?PageID=924&Content ID=10106>. 15 Ontarios Feed-In Tariff Program Backgrounder. Ontario Power Authority, accessed online at: <http://www.powerauthority.on.ca/Page. asp?PageID=122&ContentID=7136>, May 14, 2010. 16 Generation data were developed based on what would supply the equivalent of 3,000 MW of nuclear power generation (85% capacity factor assumed), while continuing to meet peak supply and reliability demands. Land-based and offshore wind power, often considered one of the more challenging power sources to integrate large quantities of, would account for 6,400 MW of supply (including what is already built and contracted). This is significantly less than the 10,000 MW scenario that was modeled in the Ontario Wind Integration Study performed by General Electric for the Ontario Power Authority, the Independent Electricity System Operator and Canadian Wind Energy Association in 2006. This scenario suggested that 10,000 MW of wind power could be integrated into the Ontario system by 2020 by adding less load following capacity than is already built in the province. 17 Renewable power generation costs based on 2010 Feed-In Tariff rates, additional conservation costs are estimated by the Ontario Power Authority to be 2.7 c/kWh, Combined Heat and Power system costs estimated to be 8.5 c/kWh. Nuclear plants have relatively low operating costs and the bulk of the cost of delivered power is related to their capital costs. The 20 cent/ kWh figure for new nuclear reactors that have a capital cost of $10,800/kW is based on an extrapolation of the Ontario Power Authority's cost estimate ($2,900/kW capital costs = 8.6 cents/kWh, of which only 0.3 cents is Operating and Maintenance costs) and Moody's cost estimate ($7,500/kW capital costs = 15.1 cents/kWh, of which only 1.5 cents/kWh is variable costs). These costs would be higher if the reactors don't achieve the high capacity factors (~90 per cent) that are predicted for them, as the fixed capital costs must be spread over fewer delivered kWh. 18 Job creation figures were developed based on the model results of the analysis in the 2009 study by Robert Polin and Heidi Garrett-Peltier of the Political Economy Research Institute (PERI) at the University of Massachusetts-Amherst Building the Green Economy available online at <www.bluegreencanada.ca>. The jobs figures include both direct and indirect employment recognizing the level of investment required to meet the green energy portfolio, the labour intensity of this spending and the level of local content of the spending. 19 Denette, Nathan. Hamilton: Samsung deal keeps jobs from going south, The Canadian Press, accessed online at: <http://www.greenenergyact.ca/Page.asp?PageID=924&ContentID=1398>, May 14, 2010. 20 Pollin, Dr. Robert . Building the Green Economy. Commissioned by WWF-Canada, Blue Green Alliance and the Green Energy Act Alliance, May 2009. 21 Smitherman, George (Ontario Minister of Energy and Infrastructure). Amendment to the Supply Mix Directive, Issued September 17, 2008. Available online at: <http://www.powerauthority.on.ca/Storage/83/7831_ Ministry_ Directive_PSP_Sept_18_08.pdf>. 22 Maurino, Romina. Ontario eyes bigger focus on renewable energy to meet power needs, Canadian Press, September, 18 2008. 23 Spears, John. 184 power producers are given green light, Toronto Star, April 9, 2010. 24 According to the Ontario Electricity Act, the Ontario Power Authority is tasked with revising the Integrated Power System Plan (IPSP) every three years. A decision on whether new nuclear stations are economical and/or needed could be delayed until completion of the next iteration of the IPSP in three years. Such a decision could still allow the various environmental and regulatory approval processes to continue, keeping the nuclear option open for a future government. 25 The OPAs 2007 IPSP application estimated that there would only be 3005 MW of wind generation in 2014 with wind development stopping completely in 2019 (Source: OPA, EB-2007, Exhibit D, Tab 9, schedule 1, p. 8). The OPA now estimates that installed wind capacity will reach 5,300 MW in 2014. (http://www.powerauthority.on.ca/Storage/111/15898_Integrating_Renewable_Energy_Supply_Resulting_ from_Uptake_of_the_FIT_and_microFIT_Programs.pdf) This portfolio proposes just under 2000 MW of wind generation in 2014.

Appendix: Green Energy Exceeding Expectations

Nuclear Blocks Wind Expansion
The long-term electricity plan the OPA submitted to the Ontario Energy Board in 2007 capped the long term development of renewable energy in Ontario at 5,312 MW of wind solar and biomass. According to the plan, wind, solar and biomass development would stop in 2020 just after additional reactors came online. Wind development would be capped at 4685 MW, solar at 88 MW and biomass at 539 MW. The graph to the right shows how wind was planned to flatline after new reactors went online in 2018 and 2019.

Source: OPA 2007

The table to the right shows how Ontario has already procured more renewable energy electricity than was anticipated in the OPAs 2007 electricity plan. By 2010, Ontario had already procured 1255 MW of solar power compared to the long-term target of 88 MW planned in 2007.

Energy source
Solar Wind Biomass Total Annual Production (Solar, Wind and Biomass)

New renewables in IPSP (2007)

88 MW 4685 MW 539 MW 12.4 TWh

Announced new renewables March 2010

1255 4692 1178 14.7 TWh

The table below provides a breakdown of green energy sources that have been procured by the government. Ontario's Growing Renewable Energy Supply
Existing wind power (IESO 2009 output data) Biogas / Biomass Renewables from March 10, 2010 announcement Wind (onshore) Water Solar Biogas Biomass Renewables from April 8, 2010 announcement Solar Water Wind (onshore) Wind (offshore) Wind (onshore) Samsung Targets Solar
(based on proposal to convert Atikokan and 2 units at Nanticoke)

Capacity (MW)
1,162 6.5 1 0.9 103 31 19 652 192 1,229 300 2000 500 1121

Output (MWh)
2,300,000 29,609 2,453 4,652 108,274 141,211 86,549 685,382 992,333 3,014,491 919,800 4,905,600 525,600 2,000,000 15,715,953

OPG Biomass

Total New Renewable Generation by 2015

A 21st century energy system cannot depend on 20th century thinking. Following the coal phase-out, the Ontario government must revise its policy on replacing aging nuclear facilities with new ones to continue developing a modern green energy economy.

renewableisdoable.com

PREPARED BY Tim Weis, P.Eng., of the Pembina Institute, Shawn-Patrick Stensil of Greenpeace and Dr. Keith Stewart. August 2010