In October 2019 there was an IAEA Integrated Regulatory Review Service (IRRS) visit to the UK. Its report can be found <here>.
The IAEA state that: “The Integrated Regulatory Review Service helps host States strengthen and enhance the effectiveness of their regulatory infrastructure for nuclear, radiation, radioactive waste and transport safety.
IRRS teams evaluate a State’s regulatory infrastructure for safety against IAEA safety standards. The teams compile their findings in reports that provide recommendations and suggestions for improvement, and note good practices that can be adapted for use elsewhere to strengthen safety. Mission reports describe the effectiveness of the regulatory oversight of nuclear, radiation, radioactive waste and transport safety and highlight how it can be further strengthened”. <here>
Prior to the visit the UK authorities conducted a self-assessment and presented a preliminary action plan and supporting documents. The IRRS team, which consisted of 18 senior regulatory experts from 14 IAEA Member States, 2 IAEA staff members and 1 IAEA administrative assistant, and 3 observers, reviewed these and a number of other documents before their visit and then spent two very busy weeks in the UK. This included interviews with 16 regulatory bodies and governmental departments.
Of particular interest to me are the references to emergency planning.
The mission commented that the “emergency planning zones established under REPPIR 2019 are not fully in alignment with the requirements of GSR part 7”. They recommend that the “Government should review the UK EP&R framework to explain how the requirements of GSR Part 7 are met in terms of planning zones and distances, and if any gap exists develop appropriate regulatory requirements”.
We must remember that GSR part 7 is IAEA advice and its section 2 states that it is “established in addition to and not in place of other applicable requirements, such as those of relevant binding conventions and national laws and regulations”. It goes on to say that where there is conflict between the GSR-7 and other requirements “the government or the regulatory body, as appropriate, shall determine which requirements are to be enforced”. I would expect that the ONR would have to champion UK regulation over IAEA advice.
We know that the UK “planning zones” do not match those of the IAEA. The UK zones have developed over many years and have, in the past, suited the UK emergency planning framework. REPPIR-19 was an opportunity to undertake a review of planning zones but it was an opportunity missed. The current system of a DEPZ with a torturous definition and an arbitrary outline planning zone does nobody any favours.
GSR-7 defines a precautionary action zone (PAZ) where arrangements are made to implement urgent protective actions and other responses before any significant release in order avoid or to minimize SEVERE DETERMINISTIC effects. This is severe accident territory and a release profile consistent with older designs of contained reactors for which a containment failure after several days of heating up was conceivable. So the PAZ as described in GSR-7 does not seem to make a great deal of sense in the modern world.
The next IAEA zone is the urgent protective action planning zone (UPZ). This is an area where arrangements have been made to initiate urgent protective actions and other response actions, if possible before any significant release of radioactive material occurs, on the basis of conditions at the facility, and after a release occurs, on the basis of monitoring and assessment of the radiological situation off the site, in order to reduce the risk of stochastic effects. This is broadly similar to the plans at many British sites where some protective actions are initiated on declaration and then thought is given to extending their scope and range if conditions merit it. It is important to realise that, in the UK, the default protective action areas are contained within the DEPZ but not defined by it.
The IAEA have an extended planning distance (EPD), beyond the urgent protective action planning zone, for which arrangements are made to implement further protective actions if monitoring and assessment on the day show that they may reduce stochastic effects if implemented within a day to a week or up to a few weeks following a significant radioactive release. UK outline planning and the gap between the automatic protective action zone and the DEPZ, sort of covers this zone.
Finally the IAEA define an ingestion and commodities planning distance (ICPD) beyond the extended planning distance where plans are in place to protect the food chain and water supply. That this zone is missing in the UK regulation does not mean that the relevant protective actions are not given the attention they deserve. The control of potentially contaminated food and drink is covered in REPPIR-19 (it is part of the operator’s consequence report and mentioned throughout guidance).
The “zones” are a bit arbitrary; are a planning tool and are best reserved for describing the national concept of operations to be applied to a fleet of reactor sites rather than to a particular site. Excellent emergency plans could be written without any use of the terms DEPZ and OPZ. What really matters is that the emergency plan is capable of initiating sensible default protective actions without delay and then rapidly considering the situation and responding to the particular characteristic of the emergency as those characteristics emerge.
I’d prefer to see a process in which the protective actions comes first and the zones second. Sensible plan compontents include:
- On-site. UK plans tend to be quiet about what happens to the people (possibly several hundred) on the site. I’ve heard reservations about evacuating the site despite the fact that it is probably the only sensible thing to do because it will alarm sheltering residents. Cooping employees up in “mustering stations” i.e. the works canteen does not seem viable beyond a few hours and provides them with little protection.
- An automatic protective action plan where shelter/exclusion and stable iodine are pre-planned in detail and initiated without discussion on declaration over an area likely to require them in a reasonably foreseeable emergency. (This could be a keyhole shape informed by the wind direction on the day).
- A deliberative protective action plan that looks at how the protective actions of shelter and stable iodine could be extended further downwind if required and under what circumstances. This plan should detail the monitoring required to support decision making, the decision making process and how the protective actions will be achieved in a timely manner.
- An agricultural precautionary protective action plan, where thought is given to how far downwind food interventions might be needed as an automatic action and as a deliberated action, what these might be and how they might be achieved. Informing farmers of the implications of this would be part of the public information cycle.
- An evacuation plan looking at the circumstances under which authorities might want to evacuate areas close to the site (including the potentially hundreds of people on the site) and how it could be done.
- A communication plan considering how people in the area will be informed of the plans and their parts in them, before any event and how they will be alerted and advised on the day.
The US concepts of “plume exposure zone” and “ingestion pathway emergency planning zone” are rather more logical than the IAEA ones.
Neither the GSR-7 or REPPIR-19 planning zones definitions are ideal. Since REPPIR-19 has recently been introduced and the planning zones all reviewed there is likely to be little appetite in the UK to make any changes so it will be interesting to see how the ONR cope with this recommendation.
Another observation made by the mission was that “ONR does not have previously agreed format for plant data and information transfer during an emergency” coupled with the suggestion that “ONR should consider establishing pre-defined communication with the operating organizations in terms of plant data and other information during emergencies”.
The big questions here are “what plant data would be useful to ONR?” and “What would they do with it if they had it?”
If ONR were going to analyse plant data in real time and use it to generate advice to the local responders and the national government they would have to greatly extend their expertise in reactor accident management. This would only be a good idea if (a) there is something worth measuring i.e. there are parameters such as temperature, pressure, radiation levels, flow rates that can give the responders better knowledge of what is happening and what is likely to happen next (b) that data is measured and displayed somewhere (c) the ONR know what it means and will definitely be there to interpret it and (d) we don’t really trust the operator to correctly analyse and report the situation.
If ONR just need the data to be better informed spectators then I’d rather not bother.
I remember talking around this subject several times in relation to the rather primitive Magnox reactors. The conclusion was that there were very few parameters that were useful and could be measured and transmitted after a major cooling circuit failure and ignition of a fuel channel fire or two and unless they had happened there wasn’t really a problem. We always thought it would be different with PWRs which have far more instruments and loss of cooling accident sequences with periods where temperatures and pressures could be rising and threatening containment integrity.
Another observation was that “The RCIS provides ONR with adequate infrastructure to respond in emergencies and its staff has been increased significantly in recent years. However, ONR does not have an overarching emergency response plan that defines its response objectives, the organizational response structure and functions, how the response actions are coordinated within the RCIS and its external stakeholders, etc. There are RCIS procedures for each position; however, these procedures are not linked together with an overarching document. The new ONR management system, under development, does not currently include a sub-process of ONR EP&R capability maintenance”.
It is a bit surprising that ONR has such a large structure and has recently extended it without actually articulating its objectives. I wonder if everyone has the same view about what it is for.
The mission goes on to observe that “the ONR does not have an overarching emergency response and preparedness plan to coordinate the response functions and maintain response capability within the RCIS. The action plan identified the ONR does not have a formal training and qualification programme for its staff responding to an emergency” and suggests that “The ONR should consider integrating its response arrangements into a response and preparedness plan and formalize training and qualification of emergency response staff”. This could be summarised as “if you are going to do something, understand why you are doing it, work out how you are going to do it and make sure your people know how to do it on the day”. On the face of it, this is sensible advice.
Having been on both sides of this type of exercise I recognise that only a small fraction of the worth of the exercise is held in the final report. Being on the receiving side and trying to justify your plans and planning process against a polite but sustained challenge from a team of experts who are used to looking at things differently forces you to think deep in a way that the day job seldom does. You learn a lot.
Similarly being on the away team you read reports and think you’ve found gaps but, in discussion, you become to realise that different is not wrong and often where you see gaps you’ve missed the filling in a different component of the plan. They do some things, maybe a lot of things differently to you and many of them they do better than you. Everybody learns, everybody wins.
Keith Pearce, January 2021