|Note: These comments relate to the Draft Regulatory Assessment Principles for Controlled Facilities (DRAP) that can be obtained from ARPANSA PO Box 655 Miranda NSW 1490.
In commenting on the draft documents provided, it should be understood that we are referring to the operation of a 20 Mw production/research reactor, a radioisotope production plant and to the production and storage of radioactive wastes at the Lucas Heights site in Sydney.
The Foreword to the DRAP notes that they will be used by, amongst other entities, "the general public at large who may wish to satisfy themselves that ARPANSA is applying 'international best practice' to the regulatory assessment of controlled facilities". Whilst this may be an underlying aim held by many of the staff of ARPANSA, its performance so far has not travelled very far down the road to satisfying the public. Its granting of a licence to site a new reactor at Lucas Heights, the dismissal of all public comments opposing the application and its token acceptance of the material supplied by ANSTO, was a miserable start.
The public awaits the results of the applications to operate the obsolete plant that exists at Lucas Heights but expects little.
The Foreword is guilty of self-praise in comparing the DRAP ('a long and distinguished pedigree') with those used by the Nuclear Safety Bureau and, presumably ARL and the Safety Review Committee. It should be remembered that this was the regime described by Professor McKinnon as being 'unduly fragmented'. It is to be hoped that ARPANSA will not become yet another toothless tiger, merely assisting the nuclear industry along its way.
An area of disappointment is that ARPANSA, being a new regulatory body, a new broom, did not start at the beginning but at the end. Its first task turned out to be to accept an application and then grant a licence for a reactor site. This was dancing to the tune being played by ANSTO that had a timetable to hold. Instead of getting its rules, standards, principles firmly in place it decided to fill them in later. This is not good enough. Which brings me to:
An example of these unresolved tasks is, where does ARPANSA get its standards from? It was my understanding that the assortment of standards across Australia were to be considered by ARPANSA, brought up the date and consolidated into a national code. Part of this process would be the removal of the NH &MRC guidelines on radiation protection and maximum dose levels for industry workers and for the public. Not so it appears. Page 17 of the DRAP states that the recommendations of the NH & MRC as published in Recommendations for Limiting Exposure to Ionising Radiation (Radiation Health Series No. 39), 'form the basis of radiation Protection in Australia'.
Question: How do the NH & MRC's dose limits compare with those set down in the ARPANSA Regulations??
Question: Will the NH & MRC be removed from the equation and if so, when is this expected?
Question : At what stage will Australia have mandatory levels of radiation exposure?
The existing recommendations also allow the 'averaging' of levels of radiation exposure to individuals as well as allowing higher levels than those recommended 'in special circumstances'. So at the whim of (whom) an operator or a regulator, a victim of an accidental exposure can slide easily from a dangerously high dose to an innocuous one if it has been averaged over 12 months or even 5 years.
It is noted that there is also reference in the Bibliography on page 35, to the NH & MRC publication Intervention in Emergency Situations Involving Radiation Exposure, 1990. This document is described as being under revision, a fact that will be well received, as its exposure levels to the public at which intervention is suggested are outrageously high.
Question: Can you define Special Circumstances?
Question: Do you really believe that the public should accept that the NH & MRC dose ranges for primary intervention (sheltering) levels for the public at between 5 and 50 mSv whole body dose? Or between 50 and 500 mSv when evacuation 'would be considered'? It should be emphasised that these are recommendations only. Which leads on to:
Off Site Emergency Response
It is noted that ARPANSA does not have any responsibility for implementing or executing any off site emergency response measures in the event of an accident. The whole responsibility to protect the health and safety of the public from the harmful effects of radiation falls, not on ARPANSA but on the NSW State Emergency Services agencies.
They, in turn, rely on advice from ANSTO which also has no responsibility for health or safety outside its site boundaries.
Principle 122 a) refers to the 'operating organisation preparing and periodically updating an emergency plan which:.is prepared in consultation with public authorities that act on the advice of the operating organisation and ARPANSA'. This is the first time that I have heard of this plan, unless that is you refer to the range of DISPLANS that range from local up to State level. The DISPLANs are no more than management plans offering no detail.
But one wonders why any of this is necessary. We have listened to ANSTO's claims of an earthquake with a return period of 10,000 years (with a ground movement at a speed yet to be negotiated) - of a not credible worst case accident (sudden loss of coolant) that could not happen, releasing only one millionth of the core radioactivity - or that, 'With a frequency of occurrence of the Reference Accident one in a million per year, the maximum risk from this accident of an individual developing a fatal cancer is calculated to be less than one in six thousand million (1 in 6,000,000,000) per year and the maximum risk of any harmful health effect is less than one in four thousand million.(1 in 4,000,000,000) per year'.
Who would believe such hyperbole? So far only the Commonwealth Government, Environment Australia, and ARPANSA. Is it any wonder that the State Emergency Service Agencies are relaxed? So relaxed in fact that, at a recent meeting of the Local Emergency Management Committee when the [new] Chairperson asked those around the table if they were familiar with the NH & MRC document on Intervention in Emergency Situations Involving Radiation Exposure 1990 , the only person who knew what she was referring to was the representative from ANSTO!
These are the people who are assuring the community that they are fully informed and would know exactly what to do in such an emergency. After 53 years this is appalling. As we are all aware, ANSTO has no responsibility for what might happen outside its fence apart from supplying advice as required. Add to this the note in your section 5.59 that ARPANSA does not have any off-site emergency responsibility either, then the decision of the Government, Environment Australia and ARPANSA to collectively agree that Lucas Heights is the best - and only - site in Australia for a new reactor becomes increasingly more absurd.
The only public awareness of any off-site emergency measures, said to be important in 5.59, is to stay inside and wait for further instructions. Whilst this may appear to fulfil your assessment principles (under the term of your Act to cover your back) without actually getting out and examining the readiness of the service agencies it is really meaningless. If principles 123 (h) and (i) are carried out you may find that there is a lot of work to be done.
The final page of the NH & MRC document says that the success of all countermeasures depends on a clear understanding of their purpose. And, more importantly, public authorities should develop programmes to educate and prepare communities living in a potential near-field situation. The stress being on pre-event preparation. This has never been attempted (not surprising considering that the service agencies are not aware of the document) and representations to the relevant State authority have proved fruitless. In fact the only information ever supplied to the general public in the area was a poorly worded leaflet put out in 1996 and then re-issued two years later. Both laid stress on the unlikelihood of any accident and ANSTO's safety record - which negated any usefulness in the rest of the information.
The Regulatory Framework
ARPANSA has existed now for two years. Its top tier is said to be the ARPANS Act and the accompanying Regulations. Can you truthfully agree that these represent the most up to date and comprehensive documents and that they are in line with International Best Practice (IBP)? How often have they been examined in the light of your experiences so far and been found to be less than perfect? What is the process for improving them? Do any such moves come from the Government or would they bubble up, slowly, from the coalface?
I ask this because, at the recent public meeting held in Sutherland, the question was raised of the US system of a judicial public inquiry where arguments can be tested before an independent commissioner. Apparently this is mandatory with all nuclear installations. If I recall correctly you replied that your legislation does not call for this "but perhaps it should". On the other hand there is nothing in the legislation that rules out ARPANSA calling for an independent inquiry. ANSTO would object and the Government would be furious but, as our local Federal member always stresses, ARPANSA is Australia's independent nuclear regulator.
It should also be pointed out that all large projects that are planned in NSW are subject to Commissions of Inquiry. How can ARPANSA really believe that it works to IBP when its processes are far lower than those that are mandatory at State level? It seems that if a project has been arbitrarily decided on by the Commonwealth Government then the subsequent processes work to the lowest common denominator.
Background, Objective and Scope
1.10 Change the last sentence to 'Where clarification is required, ARPANSA must be consulted.'
1.12 Refers to economic analyses being the responsibility of the operating organisation and is not addressed in the principles. However the operator is expected and allowed to, use the obnoxious ALARA principle. In its present form ALARA allows the operator to decide whether to 'take into account economic and social factors' in areas of safety and emissions to the public. Until this is amended by the removal of the final clause it gives the operator power over the regulator rather than the opposite.
Added to this, when ARPANSA says that an alteration or improvement must be initiated but adds 'as soon as is practical' it passes on all power to the operator. (Refer to the Safety Review Committee and the saga of the high level liquid waste from technetium production. No teeth - no action!)
1.13 Refers to Australia being a contracting party to the International Convention on Nuclear Safety. Appendix 6 notes that it was ratified in 1996. Does it have to have enabling legislation for it to be a legal requirement in Australia? Is any legislation in place?
It is noted that the Joint Convention on the Safety of Spent Fuel Management and on the Safety of radioactive Waste Management, whilst signed in 1988, has not been ratified. Will this drag on for decades as with the conventions on compensation after a nuclear accident?
2.1 Suggests that regulatory bodies should strive to find indicators that will show up poor safety performance before accidents occur. Have a look at the report of the Canadians who came out to Lucas Heights in 1990 to appraise ANSTO's operation. Its notes on safety culture are interesting. Since then we have had the spent fuel cocooned in rainwater in 'dry storage' - the incidents involving spent fuel where we wait for the report as to its root causes - the irradiation of staff in the radioisotope production plant. Where were the SOP's? What were the supervisors doing for all those years? Was management too busy to notice the problems? Does ARPANSA accept ANSTO's claims of an excellent safety culture without investigating the actuality?
2.2 It is noted that ARPANSA encourages open communication with the operating organizations. Does this refer to the senior management level only? Or have you spread the word throughout all levels at ANSTO where a staffer might be reluctant to supply embarrassing information to ARPANSA for fear of reprisal? I mention this because I recall a former ANSTO scientist who made a submission to the Research Reactor Review (RRR) and who later attended consultation meetings with the community and ANSTO, asking for a policy on 'whistle blowers'. The matter did not go any further. There are also current employees at Lucas Heights who pass on information about accidents but always anonymously.
Question : Does ARPANSA have a whistle blower policy? Does ANSTO?
Defence in Depth
2.3 May I draw your attention to a couple of instances where defence in depth has proved less than adequate. First the reprocessing plant at Dounreay in Scotland. At the end of a chequered life one of the final straws that led to its closure was that of the electricity supply to the plant. Outside contractors were called in to carry out some work. They used a mechanical mobile digger that sliced through the mains supply cable. It could happen to anyone you might say, but, the defence in depth planners had put the reserve emergency cable in the same trench and it too got the chop!
In July 2000 there was an incident at Kingsford Smith airport in which the radar system failed as did back-up system no.1 and back-up system no.2. This shows that even "defence in depth" safety plans, often mentioned by ANSTO and parroted by ARPANSA, are sometimes ineffective when most needed. The scientific community, even more than the rest of us, knows that Murphy's Law is the most consistent law in our tiny part of the universe, but it can't be factored into a computer model.
The radar incident led to the NSW minister concerned appearing on TV looking suitably serious. Predictably he said that there would be a full inquiry to ensure that it could not happen again. A fortnight later the same thing happened.
This is not to say that defence is not an admirable idea but it shows that even with it firmly in place it does not always work. Yet another reason not to site a nuclear reactor and the hazards associated with its production facilities near a residential district.
Another accident that had all the regulatory checks and double checks in place took place in Holland last year. The accident (should we describe it as an incident?) at the fireworks factory in the Netherlands should have sent ARPANSA some kind of message. Two days after the explosion there was an interview on the ABC Parliamentary and News Network, PNN, with a local official. He said that "The factory was licensed, the storage facility inspected, the licence has recently been updated and, to help the process, the army bomb unit had been called in to endorse its safety. It was safe.........(pause)........but the accident happened." The town looked like a war zone after a bombing raid.
It has not been established whether arson/sabotage was involved but the fact that a factory with the capability of causing such damage was even considered in such an area, not to say licensed, beggars the imagination. Such is the confidence of regulators who have no responsibility for accidents at a plant they have licensed.
2.4 This section extols the virtues of defence in depth. May I quote from a recent letter to Dr Loy following a reply from ANSTO about the use of the stainless steel tanks for storing the highly radioactive liquid waste that results from the production of Technetium. "The liquid waste from current molybdenum production continues to be transferred to stainless steel tanks before it is solidified."
Question: This system of storage was deemed by the SRC to be an accident waiting to happen in 1988. It noted that an accident could have off-site consequences. Its contents are slowly being solidified but the same tanks are still being used. How many levels of D-in D are in place for this condemned system?
3.3 States that the prime responsibility for the safety of facilities rests with the operating organisation. One wonders whether this is entirely correct in law. ARPANSA has a great deal of responsibility in licensing the site, close to a residential area; for passing the design specification; for granting a construction licence. Should the unthinkable happen and a new reactor did cause harm to the local population then we all know that we would have to take ANSTO to court under common law and prove negligence. Perhaps this could be expanded to include Government Agencies that approved the technical side of the project. Who knows?
Principle 4 (a) calls for the operator to demonstrate that it has financial capability. You are aware that ANTSO was unable to repair the surveillance system at Little Forest Burial Ground for 18 months due to lack of funds.
Principle 5 (b) refers to environmental assessments by environmental protection agencies. You are aware that the NSW EPA has no jurisdiction over any of the activities at Lucas Heights as it is Commonwealth property. For this reason it has refused to carry out any radiation monitoring since the law governing the site was changed.
Question: To which EPA do you refer?
Principle 5 (c) says that the operator must demonstrate compliance with all relevant legislation and any obligations of Australia under international treaties.
Question: Does this include those that the Commonwealth has signed but not ratified and for which there is no legislation?
Principle 7 refers to staffing levels and training.
Question: Does ANSTO have sufficient trained staff to run HIFAR and a new reactor at the same time?
Question: What is the situation in Australian universities for the training of nuclear engineers and operators?
Conservative Proven Design and Engineering Practice
3.4 is vague and therefore too difficult to understand. The question of what can be described as 'conservative' has now become a factional issue. ARPANSA and ANSTO claim conservatism in their approach to design but, when faced with a raising of the seismic bar they both claimed that the new figure was too conservative and demanded a recount. You can't have it both ways.
3.5 "The greater the innovation in the design of systems, structures and components, the greater the need for demonstration of performance and reliability." Who could fault such a noble statement? The approved builder of a replacement reactor has only one demonstration unit, that in Egypt.
Question: What truly independent assessor is going or has gone to Egypt to talk to the operators, the engineers, the research scientists and the accountants? (This comment also covers principal 16 )
Principle 14 calls for the operator to have recognised quality practices accreditation that is applied to the facility.
Question: What was the accreditation that applied to the storage of the liquid wastes from radioisotope production and for those wet spent fuel rods that were standing in 'rainwater' for several years?
Feedback of Operational Experience
3.7 says that 'decades of nuclear facility operation have led to improvements in safety culture and the implementation of defence in depth.'
Question : Has Dr Loy access to the Canadian report on safety culture at ANSTO dated 1990? It is hoped that safety culture has improved over the last ten years.
Principle 16, dot point 2 refers to personnel performance assessment and counselling and retraining.
Questions: What happened to the person whose job it was to inspect the spent fuel dry storage? And to the two people who were irradiated in the radioisotope production plant? More to the point, what happened to their supervisors and to the managers of their sections?
4.1 Includes a reference to 'a licence or an authorisation'. Is this a typing error? ANSTO has operated for years under an authorisation supplied by itself. It was because of this that ARPANSA was formed and part of its function is to supply or deny licenses.
The reference to ALARA, and in principle 18, calls for a repeat of my usual complaint. If you are to be honest you should add, each time ALARA is quoted, the conditional clause, 'taking into consideration economic and social factors' and define it. Better still, ARPANSA should become a world leader and remove the offensive clause altogether.
4.5 Highlights the difference between a PSAR and an SAR. It presents a problem in that the PSAR is the theoretical assessment as perceived by the manufacturer. It is presented to ARPANSA which, when its peer review is complete, will give a licence to construct. I recall that, at the recent public meeting, ARPANSA said that it would consider the reference accident as a basis for approval to construct. Then, when the reactor was built and the money spent, it would consider a worst-case accident when the licence for operation was applied for. Surely this is like having a good look at the open stable door the week after the horse has well and truly shot through.
The whole idea of basing a construction licence on a PSAR prepared by the vendor and based on its own national standards, is wrong. Regardless of 'standard IBP' practices the vendor must be made aware of the seismic parameters of the design and how a 'worst case' accident - sudden loss of coolant, a power excursion or an act of sabotage - (possibly taking those described by Tony Wood or by Daniel Hirsch) should be factored into the design. Anything less would be a dereliction of duty by ARPANSA.
Question: When will the seismic parameters be finally decided? Who will make this decision? Will the decision be made public?
Question: When will the appointment of the peer reviewers be notified to the public?
Question: Will the peer review be made available to the public at the same time that it is asked to comment on the licence application? To have it appear half way through - or even later - the period of public comment, would be inappropriate, time-wasting and unacceptable.
Questions: Has ARPANSA enough staff qualified to consider the approval of a nuclear reactor? If not, will it look elsewhere for such experts? Would any such appointees include current or former employees of ANSTO? Will the background of such appointees be publicly announced?
Principle 22 says that the SAR will include the numerical values for the operational limits and conditions, and the proximity to the relevant safety limits, for normal and abnormal operation.
Question: As these will be provided by ANSTO, who sets out the relevant safety limits? Are they negotiated, decided by ANSTO or determined by ARPANSA?
Principle 25 is a very badly worded section. Did a lawyer put it together? That it includes 'taking into account the very low probability of such events' and 'that such accidents do not dominate the total risk (in terms of consequences and frequency of occurrence)' totally wipe out any confidence that we might have had at the start of your document.
If this is the way your staff are to be guided then it should be pointed out to them that if a theoretical occurrence has a return period of 10,000 years - and the data goes back only a few decades - it can return next Monday. It could also happen the following Saturday if averages are all we are looking at. Please rewrite this principle in at least two and better still, three sentences. Remove all references to low probability and frequency of occurrence.
Design-Basis and Beyond-Design-Basis Accidents
4.8 Again reference to 'but because of their lower frequency of occurrence than design basis accidents, may not result in intolerable risk'. This would immediately convey to a newcomer to the staff that there is nothing to worry about and make an assessment accordingly. The use of the words 'may not' is far too vague. It may not - it may. If the assessors are confident enough to accept all risks on behalf of the community, without reference to them, then they should be bold and say 'will not'. Then, should the unthinkable occur, we would know who to take to court.
4.12 Question : Will the PSA on a new reactor be considered at levels 1,2 and 3? Or will it be less than this as was carried out on HIFAR?
Principle 27 should have the addition of c) Worst case accidents. How can you simply ignore them?
Principle 35 brings to mind the allegations that the INVAP reactor built for Egypt has had design and safety problems. This has been refuted, I believe by ANSTO in a late submission to the Senate Inquiry. As ANSTO has a vested interest in the matter what has ARPANSA done to check it?
Principle 36 e) 'takes into account human factors'. Exactly how is this achieved? Most accidents in industry occur through human error or by deliberate human intervention.
Question : Who is responsible for these reviews, ANSTO or ARPANSA? It brings to mind the recent situation in the aviation industry where the regulator, CASA, was unaware that routine checks had not been carried out on passenger jets for up to six months. Last week CASA introduced a system of fines should it occur again (another case of the bolted horse). Will ARPANSA be a better regulator/overseer than CASA?
Defence In Depth
This has been dealt with in an earlier section. However, if the culture is built around the feeling that a) an accident cannot happen and b) but if it does no great damage will result from it, the result may be apathy and the good old Aussie motto 'She'l be right!' will be the guiding light. There is also the stress resulting from management decisions such as happened eleven or twelve years ago. The days of permanent employment are long gone and events in the US involving disgruntled workers show what can result. Ergonomic chairs would have no effect in such circumstances.
Regarding communication systems, I have observed telephone failures during emergency exercises held at ANSTO, including that to used by the Communications Manager. Maybe they will work when really needed.
Question: Where does the figure of 30 minutes come from in Principle 49?
I am uncertain as to the meaning of the word 'redundant' when used in principle 52. Could you please define it for me?
Principle 54 (d) says that the design of a facility takes into account the feasibility of emergency response. I am yet to be convinced that the SES really do know how to handle the kind of emergency that ANSTO says can't happen and for which ARPANSA perceives as not credible.
The newly formed ANSTO Emergency Management Arrangements Public Information Working Party (AEMAPIWP) will be meeting on 2 March and I understand that ARPANSA will be represented. I hope that you will ask the pertinent questions.
Inspection, Testing and Maintenance
Principle 65 'Inspection, testing and maintenance frequencies of items important to take account of the categorisation by safety significance of the items.'
Once again I return to the case of the 'dry' spent fuel rods sitting in rainwater. If ANSTO could not control this over a period of ten or more years does ARPANSA have the number of inspectors needed to check on all the items that need routine inspection? I am reminded of the recent situation at Sellafield when it was discovered that QA inspectors had been falsely recording the perfection of their MOX pellets. When it was made public a whole batch was returned from Japan. Heads rolled. The paperwork was filled in but the records were false.
Question: When modifications are called for by ARPANSA how soon would it act if ANSTO were recalcitrant? I refer to the modification to the water tower politely asked for by ARPANSA.
Principle 70 'The form, locations and quantities of nuclear material are specified, monitored and recorded.'
Question: Did ARPANSA ever get the report from ANSTO as to the root causes of the 'dry' spent fuel standing in rainwater? Just before your time but what action have you taken?
This section ignores the comments on page 12, section 3.4.2 of ARPANSA's SER for the site approval for a new reactor. As usual it has caught the bull by the tail rather than the horns. 'Contingency plans describing alternative strategies for disposal of radioactive waste and spent nuclear fuel will also be required with the application for a licence to construct the replacement reactor. A licence to operate the reactor would not be issued by ARPANSA without there being clear and definite means available for the ultimate disposal of radioactive waste and spent nuclear fuel'
Surely this is back to front. The means of ultimate disposal should be a requirement before construction, not after. I can imagine, when the plant is built and the $300 million - or more - has been spent, the Minister of the day is told by ARPANSA that it won't be licensed to operate because 'there is no clear and definite means available for the ultimate disposal of radioactive waste and spent nuclear fuel'.
To this end, Principle 73 is too weak when compared with the SER. It should be re-worded to emphasise its importance to the assessors.
This is really a chilling section. It shows no evidence of regard for the local community. It talks of collective effective doses to the most exposed population - of contamination of the environment not being so extensive as to restrict the long-term use of land around the site - of the collective effective dose being a relevant representation of societal risk - a level of dose that is considered to represent an acceptable level of risk in the circumstances
Question: What about short and medium term use? What about any loss of property value? These are not covered by commercial or government insurance? What of the societal effect to the local community? Not ARPANSA's responsibility!
Question: What can be seen as relevant or acceptable risk if the community is not asked its opinion? In the light of no commercial insurance? Unwillingness of the government to provide insurance cover?
Question: Why do the authorities always ignore the social disruption that would prevail, even if only a reference level accident occurred?
The DRAP have been produced with slight regard to the first objective of ARPANSA which is to protect the health and safety from the effects of radiation. They have been produced by engineers without prior reference to the community. The CEO is obliged under the Act to take note of comments from the community but the single example of this - the site licence application - gives the community no confidence that it has or will be, listened to.
Since ARPANSA was formed two years ago it has not been allowed time to start at the beginning. Rather than getting its guidelines, principles, radiation dose levels and emission levels into proper order, it has been pressed by the Government and ANSTO for urgent licences. As it is also an organisation that works for a Government Department its independence as a regulator is perceived to be under threat and its priorities organised for it.
There is uncertainty as to how or whether ARPANSA is setting the rules and regulations for radiation protection for the whole of Australia or it is still relying on the recommendations of the NH & MRC which were highly controversial. Do ARPANSA's regulations refer only to Commonwealth operations or for those outside that sphere? Will there be two or more sets of standards at the end of the day? One set merely recommendations, the other mandatory but providing room for manoeuvre by an operator?
There is little evidence in the DRAP that the human society side of its assessment process is considered. Was there anyone on the team that put this paper together with humanities knowledge and skills? I do not include Public Relations managers in this field.
There is a lot of work to be done before the RAP is finally released.
Convenor Nuclear Study Group
Sutherland Shire Environment Centre