Mr. Ian Fairlie, Northumberland Environmental Protection

MR. FAIRLIE: Thank you, Chairman. First of all, I would like to say that I am not a doctor, I am a PhD student, and I think that calling me "doctor" is a use of an unconservative assumption. But thank you anyway.

In view of the short time that we have got this afternoon, I am going to concentrate my talk on two aspects of the papers that are in front of you; in other words, collective doses and reprocessing as an option. And perhaps tomorrow, when I have got more time, I will come on to the use of international comparisons, if I may.

So will move straight into collective doses.

I think it is fair to say to members of the panel that the use of collective doses has a checkered career; there is certainly no equanimity of view about whether to calculate and whether to use collective doses in the international radiation community.

It has got a bit of a checkered career in the sense that back in the 1960s, for example, a number of authorities were considering the use of collective dose limits; and in fact the very first country in the world to introduce a collective dose limit was Canada, the AECB back in 1965 introduced a collective dose limit of 100 person Sv per nuclear power station per year. That doesn't exist anymore.

But there is a limit in Sweden of 5-person Sv per thousand megawatts of installed capacity, installed nuclear capacity.

Perhaps I should tell you what "collective dose" is. It is an attempt, however inexact, to put some figures on collective detriment from radionuclide discharges. You will be aware, obviously, from the work done by ... the excellent work done by AECL, that the thing which is used as a comparison is average doses to average members of the critical group.

Now, I would argue that in addition to that parameter, collective doses should also be connected ... or calculated I should say, and I notice that in the guidelines put out by the panel a few years ago, that you asked AECL to calculate collective doses, and I have not been able to see any mention of collective doses certainly in the post-closure assessment and the case study and in the EIS. There is some mention in the pre-closure, but we are talking about post-closure right now.

In very simple layman's terms, what "collective dose" is, is the product of the population exposed and the average to that population, to an individual member of that population.

Nevertheless, I would go beyond that and say that when it comes to calculating collective dose, conventionally there are three methods of doing this; you can calculate it to local populations, to a regional population, or to a global; in other words, the whole world population.

Now, in the added representations given by AECL on May 17th this year, AECL put forward calculations for a regional collective dose. However, a number of the major nuclides under consideration here, which are Carbon-14 and Iodine 1-129 and CL-36, are globally distributed; and that means that, whether we like it or not, we have to calculate global doses.

Now, there are a number of serious question marks about whether we should do this and the reality of the results. I have set out in my paper from paragraphs 23 to 27 a number of misgivings about global collective doses. It is true that if you are measuring the effects of very, very small amounts of radiation, or small amounts of radioactivity, calculated over large numbers of people for a period of time in the future, you can put a question mark against what this actually signifies.

Despite that, Chairman, and members of the panel, a number of authoritative, international bodies have come to the conclusion that we have to make these calculations anyway; in particular UNSCEAR and the IAEA, and I have referenced these materials at the back.

I have attempted to do the same.

What basically, in very simple terms, is done, is one constructs global transport models, say for iodine -- and it has been done for iodine and carbon -- looking at the weather patterns throughout the world, the uptake patterns throughout the biosphere into human beings, the dose conversion factors in human beings, and you can make what we call a global dose conversion factor. So that if you have a model for injecting, say, 1 terabecquerel of Carbon-14 into the atmosphere anywhere in the world, you can calculate what the collective dose would be throughout the word. In other words, this would be a dose that would be untruncated in terms of time to a population of 10 billion people, ten 10th (sic). This has been done by the United Kingdom's National Radiological Protection Board; they have their global transport models.

And what I have done is multiplied the inventories of the various important radionuclides by this global dose conversion factor; very simple, one figure multiplied by another, nothing complicated about it. But I would stress that the models that I have used are internationally accepted and that the calculation of these global doses should really be done by the AECL.

If I had been able to, from the figures in the EIS, I would have liked to have calculated the global dose rates; in other words, not just the amount which we estimate has come out after 10,000 years or after 100,000 years, but actually measured the annual rates. But I'm afraid I've not been able to do that with the figures that are available in the EIS and in the case study.

Perhaps the panel could ask AECL to produce those figures at a later time.

I would like to show you now the figures that I have come up with. I have come up with two tables. The first table is the worst case scenario; in other words, looking at the inventory as presented in the EIS study and calculating the global population dose which would result if all of that were inadvertently discharged up to the atmosphere. This is unlikely, obviously, but nevertheless it still is the worst case scenario. It is something which could occur, for example, if there was a criticality excursion; again unlikely.

But, nevertheless, I think it's important that we understand the nature of the scale of the hazards that we are dealing with, and that is why I calculated this.

And I will repeat what I'm doing here; is that I'm looking at the ... this is the nuclide, this is the total inventory -- I'm sorry these figures are rather small -- this is the inventory in terabecquerels. For some reason the AECL don't seem to use terabecquerels, they always use kilograms or moles, I don't know why, but they are easily converted.

And this is the dose conversion factor, the global dose conversion factor which has been applied after using the figures from NRPB, and this is the result. For Carbon-14, for example, we wind up with a collective dose of about 10 to the 6th person Sv, providing a much smaller figure of about 35,000 person Sv.

Now, this is an important conclusion in my mind, because what it shows you is that the key nuclide of concern, as far as I'm concerned anyway, is Carbon-14 rather than Iodine-129. It's true that in terms of individual doses to members of the critical group that Iodine-129 is the major contributor, but when you look at collective dose you realize that suddenly that Carbon-14 is much, much more important.

The second table that I have produced is looking at the actual discharges as predicted by the AECL after 10,000 years, and you can see there are big reductions here.

Sorry. Not 10,000 years, after ten hundred thousand years.

These are the radionuclides again, the fraction, this is the inventory, fraction release, dose conversion, again collective dose.

I would like to apologize right now, Mr. Chairman, that it was pointed out to me by a member of AECL that I had picked up the wrong figure for Carbon-14, it should be actually three times 10 to the minus 10, which means instead of 4, we're dealing with .00000013 -- in other words, you can ignore it -- and the collective dose after 10 to the 5th years in fact does come from Iodine-129.

Now, obviously this assumes all of the assumptions taken by the AECL in its case study, the post-closure case study are correct. You can see there is a major reduction involved.

I would like to move on now to the second part of my talk, which is on reprocessing.

It is quite noticeable that, if you look at the EIS documents and the case study, that there are a number of places where the words "and reprocessed high-level waste" have been added after the words "spent fuel," and in my personal view I think they have been added well after the report has been written, and they are a political inclusion.

Given the wealth of detail and the large amount of science which is included in the documentation, the paucity of information about reprocessing and the lack of discussion stands out in my mind, and I hope to remedy that in the next few minutes that I've got left, Chairman.

By the way, I don't know what the green light means. Does that mean I've got a few minutes?

THE CHAIRMAN: The green light means you begin to talk, at thirteen minutes the amber light comes on, at fifteen you have used up your time.

MR. FAIRLIE: And after sixteen I get cut off at the mike, do I?

THE CHAIRMAN: Yes, the trap door opens and you drop into Hades --

MR. FAIRLIE: Okay, then.

Very briefly, reprocessing is a chemical process whereby plutonium and uranium are separated from spent fuel. It is a rather nasty and heavily chemical process where spent fuel rods are dissolved in boiling nitric acid, and by various chemical process, and the plutonium and uranium are extracted and separately stored. The wastes which are left over, which are the fission products and minor actinides are, in fact, right now being put into a vitreous matrix and put into canisters which are then stored. The EIS in the various documents, when it talks about vitrified high-level wastes, are really talking about stainless steel canisters in which the vitreous waste is contained, not about uranium or the plutonium which is going to be stored separately.

Reprocessing, in many people's view, suffers from a number of serious disadvantages, both from an economic point of view, from an environmental point of view, from a proliferation point of view.

But I want to talk about wastes, because that is what we are here to talk about this afternoon basically.

The key point is that if you look at -- this slide is in kinescope and technicolour here -- spent fuel over here on my left, there are basically two things you can do with it; you can send it through to reprocessing here, or you can keep it as spent fuel. And you can see what happens when you reprocess; you create a large number of waste streams, all of these waste streams are created. And indeed in Britain almost ... well, more than two-thirds of its intermediate-level waste is created solely by reprocessing, and about half of it is low-level waste.

In Canada, you are blessed, you don't reprocess your fuel. You don't even have intermediate-level waste as a result. What I would strongly suggest to you is, if Canada were unwise enough to go down the reprocessing path, it would be creating a large number of problems for itself in the waste sphere, and I certainly hope it doesn't do so.

However -- and I've only got a couple of sentences -- I understand from the terms of reference of the panel, that really you are restricted to looking at vitreous wastes as a waste form, how it compares with spent fuel; in other words, you are looking at it as a purely waste or technical matter.

I have looked into this, and it is my conclusion that vitrified high-level waste compares very poorly with spent fuel as a waste form. You see, what happens is that most people think about vitrified high-level waste as being a monolithic glass block, and nothing could be further from the truth.

When vitrified high-level waste is poured into the stainless steel canisters, it cools, and on cooling, because of the different thermal coefficients of expansion between stainless steel and glass, it shatters, micro-shatters in fact. And indeed I was at a conference in Las Vegas about a month ago where I actually saw photographs of cross-sections of canisters -- these photographs were taken by the Department of Energy at Spiny River where they are currently carrying out studies into vitrified high-level waste -- and if you look at these photographs you see incredible shattering around the perimeter of the glass inside the canisters, and in the middle large voids. The net result is an increase in the surface area of the glass by factors of 10 to 100 or so.

Indeed, a study done in the United Kingdom, which had participation of a number of persons from AECL Canada, showed that the dissolution which compared vitrified high-level waste on the one hand with spent fuel on the other, showed that the dissolution rates of all nuclides, not just the immediate volatile nuclides but all of the nuclides, was greater from vitrified high-level waste than from spent fuel. The conclusion was that spent fuel was a much better waste form.

And my recommendation to the panel is that a question be put to AECL as to why reprocessing has been injected into the EIS almost arbitrarily without any justification or rationale. I think the matter should be gone into because of the importance of it in much more detail than it has been.

Thank you, Chairman.

THE CHAIRMAN: Thank you, Mr. Fairlie.

Questions from the panel on either of the two subjects which Mr. Fairlie has addressed with admirable brevity? Denis Brown.

DR. D. BROWN: Mr. Fairlie, I noticed your reference to global collective doses, and the work of Titley calculating these for the U.K. and LPB. I looked at the references and I see that in publication. I didn't know that --

MR. FAIRLIE: In fact, it has been published since --

DR. D. BROWN: It would be interesting if a copy could go to the Secretariat so that we all saw it.

MR. FAIRLIE: Yes.

DR. D. BROWN: But my question to you is, when fuel comes out of the pond storage and goes into storage on site, on ground storage, surely you are going to get more releases to the atmosphere than you are once it has been canned and placed underground and is not being released to the atmosphere but to groundwater systems from which it may well get to the ocean and be taken up in the ocean without ever entering the atmosphere.

MR. FAIRLIE: If I may repeat your question. You are saying that when the spent fuel rods come out of the reactor and go into wet storage --

DR. D. BROWN: When they come out of the cooling pond and go into above-ground storage, then there is, as I see it, as I understand the process, nothing but the sheathing around the reactor fuel rods to prevent the fission products from diffusing out to atmosphere, and so they will diffuse out a lot more quickly during above-ground storage than they will canned in containers and placed underground where their only route to the atmosphere would be through surface water supplies, and where a good proportion of them, anyway, would be carried down to the sea, which would act as a pretty infinite reservoir, make it unlikely they would get back to the atmosphere before they decayed.

MR. FAIRLIE: I haven't seen any figures which suggest the emanation of volatile nuclides, say, from dry storage canisters. In any event, it depends what kind of canister you're talking about. You could have ventilated ones or, as in Germany, unventilated ones.

Off the top of my head, I would have said that the nuclide emissions of volatile gases from the spent fuel rods would be very limited indeed, at least for the first hundred years or so. However, that's off the top of my head; I would have to go back and look at the books to figure it out.

I haven't seen anybody talk of this. I thought we were looking at what would happen when you put the stuff underground rather than dry storage above ground.

THE CHAIRMAN: Pieter Van Vliet.

MR. Van VLIET: You mentioned the collective dose and distribution of radionuclides on a global basis. If you calculate the collective dose on a global basis, either from the theoretical maximum or from what the actual emission is, it presupposes two things; and that is that, number one, that there is a global distribution of the material under consideration; and secondly, that there is an ingestion or inhalation, or whatever, of the particular radionuclides for each person in the world. What kind of mechanisms do you see that might make this even plausible?

MR. FAIRLIE: The global distribution?

DR. D. BROWN: Yes.

MR. FAIRLIE: Oh, well, it's well accepted. For example, Carbon-14 would travel as C-14 methane or as CL or carbon dioxide or Carbon-1 oxide, there are no shortage of transport gases as far as that's concerned.

The models have been fairly well developed, there has been about three or four models, I seem to remember, and they have compartments for carbon dioxide or carbon monoxide, and in the atmosphere, in land, biosphere, in the ocean, deep ocean and deep ocean sediments, you can get models of varying and increasing complexity with up to 100 compartments, and the key thing is that it doesn't really matter all that much, there is not much variation between the models.

The main reason is because of the very long-lived nature of these nuclides, and Carbon-14, five thousand, seven hundred years, it doesn't really matter; you could have five models, ten models -- I'm sorry -- five compartments, ten compartments, or fifteen compartments, the half-life of it is so long, the mixing is so thorough, that it doesn't really matter, and they all come down to about the same figure; like 110 person Sv per terabecquerel --

DR. D. BROWN: So eventually someone anywhere would end up --

MR. FAIRLIE: Yes. As long as this was ... this would be a discharge to atmosphere, not a discharge to sea. There would be slightly lower figures if it was to sea.

MR. Van VLIET: Thank you.

THE CHAIRMAN: Dougal McCreath.

DR. McCREATH: Thank you, Mr. Chairman.

Thank you, Mr. Fairlie. I need some help with collective dose. You make the comment in your very comprehensive review, for which I congratulate you, when you are talking very small average doses --

MR. FAIRLIE: Correct.

DR. McCREATH: -- and you say, "Of course, such small doses are often trivial as far as the individual is concerned." And the next sentence says, "However, added over the world's population, these small doses become far from trivial."

And you will have to help me with the logic, because I don't understand that. If the dose to every individual of course is trivial, I don't understand why ten trivials make something not trivial.

MR. FAIRLIE: Well, it depends where your standpoint is, or your viewpoint. The viewpoint of that of an individual, thinking about himself or herself, the risk is very small indeed. But from a public policy-maker's point of view, we're talking about thousands of deaths in the future. In other words, the public policy-maker cannot afford the luxury perhaps of just taking an individualistic point of view, they also have to take on board the public health considerations as well; in other words, looking at the detriment to populations rather than to an individual.

DR. McCREATH: I guess I still don't understand. I will work at it.

Just turn to your table 1 and table 2 so that I understand that --

MR. FAIRLIE: Could I add to that?

DR. McCREATH: Please, yes.

MR. FAIRLIE: And that is that if, for example, the average dose to an individual from discharge, say, of a micro Sv, that if you were to put that around the whole world -- you've got a population of 10 billion -- you multiply that micro Sv by 10 billion and you come up with large numbers of person Sv.

DR. McCREATH: I understand the arithmetic. I don't understand the logic.

MR. FAIRLIE: I see.

DR. McCREATH: That is my problem. I continue to not understand how, if it is of course trivial --

MR. FAIRLIE: Trivial to the individual. Sorry.

DR. McCREATH: I understand. -- for each individual in the population, I just don't understand why it becomes less trivial in terms of mortality because you have got a lot of people to each of whom it's trivial. I'm really having difficulty wrestling with that, and I will continue to wrestle with it --

MR. FAIRLIE: Can I get one more try --

DR. McCREATH: Please, yes. If --

---Reporter Appeals:

MR. FAIRLIE: They are random throughout the population; we don't know who is going to get them, who is going to come up with either cancer or with genetic changes, we just don't know.

DR. McCREATH: Indeed.

MR. FAIRLIE: We do know, from our theories of radiation's effects, that effects will occur if you expose twenty people to 1 Sv each, we know that on average one person will die, or will have serious genetic changes in the future generation; we know that.

DR. McCREATH: At 1 Sv I understand that.

MR. FAIRLIE: Yes. So what we're doing here is we are calculating the doses to a large population and we know, from the figures we got, that a certain number will die.

DR. McCREATH: Well, then surely what you are saying is, to those individuals this dose was not trivial.

MR. FAIRLIE: But we don't know who the individuals are.

DR. McCREATH: I understand that. But if you say these tiny doses are trivial --

MR. FAIRLIE: To the individual.

DR. McCREATH: But then you have to say to some individuals it's not trivial. Right?

Let's leave that, because I don't want to take up time. I think I understand what you're saying.

Let me turn, if I may, to tables 1 and 2, just so I understand those clearly. In table 1, as I understand it, the hypothesis is that if the entire inventory of nuclides is released --

MR. FAIRLIE: Worst case scenario.

DR. McCREATH: Yes, I understand. -- if the entire inventory is released, divided amongst the world's population of 10 to the tenth persons, you end up with these collective doses, and if you then apply a risk factor of 5 percent, et cetera, et cetera, you end up saying this would give about 50,000 excess fatal cancers throughout the world.

Presumably the reason that we are storing used fuel is to avoid complete release, worst case scenarios. And let's assume for a moment that there is some success in doing that, which is your table 2.

MR. FAIRLIE: Correct.

DR. McCREATH: So if I understand table 2 then, we are saying if you take the proponent's calculations at 100,000 years -- that is not the 10,000 regulatory, but the 100,000 years which is approximate peak from their calculations -- if you take the worst case of their calculations at least, and go through the same calculation, if you were to apply the same risk factors to those figures, would I be correct in understanding that you would end up with a total world-wide excess of fatal cancers of between one and two persons?

MR. FAIRLIE: Correct.

DR. McCREATH: Thank you.

MR. FAIRLIE: That's making all the ... with all the assumptions built into the --

DR. McCREATH: I just have one other question, if I may, to help me with the collective dose.

I understand you make the statement that this is often useful and helpful and seem to be thus in comparing what you call radiological practices --

MR. FAIRLIE: Correct.

DR. McCREATH: -- current radiological practices.

MR. FAIRLIE: Correct, yes.

DR. McCREATH: It's not clear to me, if one takes the worst case scenario, what radiological practices this approach would help us compare. I mean, what are we comparing?

MR. FAIRLIE: No. Well, the various studies on collective dose, in particular Lindell's study, point out three uses of collective dose; and the first one is to get a handle on the total risk involved. That's one. Put that to one side.

A second use is in comparing similar radiological practices. For example, the use of a titanium container versus a copper container. In other words, you do your calculations and you calculate your collective doses using the one scenario or the other.

DR. McCREATH: So that would be the table 2 kind of approach?

MR. FAIRLIE: Sort of thing, yes.

DR. McCREATH: Okay.

MR. FAIRLIE: In other words, you would calculate your collective doses using the titanium container and figure out what you get at the end of the day, and do the same thing using your copper container, and compare the two results.

Now, that's fairly robust, because you're using ... it's a comparison, you're using exactly the same assumptions throughout, and you can rely on that comparison.

DR. McCREATH: But it would not change the table 1 type of approach, is what I guess I'm trying to --

MR. FAIRLIE: It would be table 2 ... I guess --

DR. McCREATH: Okay. Thank you. I understand.

Thank you, Mr. Chairman.

THE CHAIRMAN: Thank you.

DR. D. BROWN: You have commented on the collective dose. It is known that 10 MilliSieverts to ten people gives the same risk as 1 MilliSievert to a hundred people, and I agree totally. But when you equate collective dose to collective detriment for doses way below 1 MicroSievert, this becomes much more controversial and has certainly not been established.

I fully appreciate and use collective dose as a valuable tool for comparing two different installations, but I think it gets very dangerous to equate collective dose automatically with collective detriment.

MR. FAIRLIE: May I respond to that?

THE CHAIRMAN: Yes.

MR. FAIRLIE: As I mentioned in my preamble, the concept of collective dose has enjoyed periods of being in fashion and out of fashion. At the present moment it is coming back into fashion again, and the reason why is because of the adoption by ICRP -- sorry, acronyms flourish in this field; I take it that you know International Commission on Radiological Protection -- of their latest recommendations on radiation protection, assumes conservatively that a linear dose response relationship with no threshold. That has a number of concomitant conclusions, a number of things follow on from that, and that means that there is a detriment all the way down to zero. No matter how small it is, there is a calculable detriment. Even at tiniest doses imaginable, there is some sort of detriment.

And it's for that reason that a number of publications have come out in recent years about collective dose. Because if you assume that there is a detriment down at very, very, very small doses, then the idea of collective dose in invigorated and you have to start looking at collective dose.

THE CHAIRMAN: Any questions from the floor? Microphone number ... oh, sorry, AECL did you have any questions, or SRG? Just a moment, please --

DR. FRIND: Emil Frind, SRG. I just have a very brief comment regarding your remarks using models ... item number 15 on page 5. You quote several references that criticize the use of complex models.

I suggest that you have taken that a little bit out of context, because the subject of model validation is currently the subject of scientific debate and there is an equal number of very credible papers out there that support the use of models, and I think one of those authors that support the use of models is indeed McCombie, who you mention in the same paragraph, and I think he supports the use of models provided you know what you're doing.

MR. FAIRLIE: May I reply?

THE CHAIRMAN: Are you finished?

DR. FRIND: Yes.

MR. FAIRLIE: In my statement I make it quite clear that McCombie said ... was careful; he thought that the use of complex hydrogeological models was okay for heuristic purposes, but was less happy about it for predictive purposes. And that is a direct quote from his paper.

And I would agree that certainly we have learned ... the science of geology has taken huge leaps and bounds in the last few years as a result of the present kinds of studies that we are doing, Mr. Chairman, but nevertheless the use of complex geological models still raises some question marks for detailed predictions.

DR. FRIND: It may be a direct quote, but there is also other papers, other authors that argue the opposite. That is what I wanted to point out.

THE CHAIRMAN: AECL, I believe you now wish to make one comment, and then I will turn to the other microphones.

DR. JOHNSON: Lawrence Johnson from AECL. On your page 4, point number 13, you commented about unconservative assumptions related to the solubility of UO2.

I should point out that I believe that's a misunderstanding of what we had presented, and is probably based on the view that we used the median value simulation, which is only used as an example. In fact, the average value is several orders of magnitude higher and is not dissimilar, really, from the values used in other international assessments, and that point is clarified in the report that was presented on an International Comparison of Disposal Concepts written subsequent to the reference that you gave.

So I believe that is --

MR. FAIRLIE: Which report is that?

DR. JOHNSON: It is a report we presented to the panel on the International Comparison of Disposal Concepts. One of the authors of it is a co-author of the study from NAGRA. So I believe it is erroneously stated here that we used a very much lower solubility. In fact, our average value is not dissimilar from the --

MR. FAIRLIE: What was the figure that you used?

DR. JOHNSON: The average is about 3 times 10 to the minus 7 moles per litre; which is about the same as some of the other studies.

MR. FAIRLIE: I see. Chairman, could I add --

THE CHAIRMAN: I'm sorry?

MR. FAIRLIE: Well, there is two reasons, really. This is a rather technical point --

THE CHAIRMAN: I think probably it would be wise if you compared notes on the two --

MR. FAIRLIE: Can we do that afterwards, and then --

THE CHAIRMAN: You certainly could do that. And then if it must come back, fine. Because I think it is important that we give the others the opportunity to put questions which may be of a broader nature.

Microphone number 3.

MRS. de QUEHEN: Mr. Fairlie, don't you think we tend to lose sight of --

THE CHAIRMAN: Would you give your name, please?

MRS. de QUEHEN: Ella de Quehen.

MR. FAIRLIE: Hello, Ella.

MRS. de QUEHEN: Don't you think we tend to lose sight of one objective? When we are looking at global effects and collective dose, we tend to think in terms of one facility. However, if this concept becomes acceptable to the nuclear community, then there could be a proliferation of such technology throughout the world, and hence collective dose would jump enormously. Don't you think that is a responsibility we have here? We are not really just thinking in terms of one facility, but of what happens if such a concept becomes acceptable.

MR. FAIRLIE: Well, it's difficult.

THE CHAIRMAN: That causes me a little bit of problem. I know exactly what you are saying, and I think in a global sense you are quite correct. But our mandate is to look at the handling, the safe management long term of nuclear waste from the Canadian reactors here in Canada, and I think it might lead us into rather distant fields if we pursued that whole item, so I would hope that Dr. Fairlie could reply fairly briefly.

MR. FAIRLIE: Yes, that is probably true. I would prefer to say something that is in my report; and that is that I understand that roughly speaking about a quarter of all the fuel which has been created, which is intended to be put inside the repository ... Well, one way of looking at this is to try and figure out a way of cutting down the amount of fuel that we are putting into repository, just talking about this one. There has not been much examination of that. And I'm talking about, for example, different fuel inventory regimes in the sense of fuel burnups and different reactor operating machines way of trying to cut down the source terms that we are talking about going into the repository, there hasn't been much discussion about that, and I would like to have seen that.

THE CHAIRMAN: Microphone number 2.

DR. RUBIN: Thank you, Mr. Chairman. It is Norman Rubin.

Dr. Fairlie, Ian, you mentioned that AECB no longer applies it population dose limits to nuclear generation stations. That --

MR. FAIRLIE: That is what I understand.

DR. RUBIN: -- is contrary to my understanding, and I wonder if somebody in the room knows. I believe those were part of the siting guide, or I believe that is what the document is called, which contains single mode and dual mode failure criteria and population dose limits, and my understanding is that those are still in effect and still form the basis of licensing. Although they are not generally exceeded, and therefore they are still academic, we get closer as the population increases around the nuclear generating stations.

Anyway, perhaps somebody can come up later, or the AECB perhaps can undertake to find the answer.

In your discussion with Dr. McCreath, I believe he said that the fraction release in table 2 was the worst case release, and you agreed with that. Is that AECL's worst case release in table 2?

MR. FAIRLIE: No, no, no, no, no. Table 1 was the worst case scenario.

DR. RUBIN: Right. But I thought that phrase came up again in table 2. That's an average outcome, is that right?

DR. McCREATH: May I clarify?

THE CHAIRMAN: Dougal McCreath.

DR. McCREATH: Well, no, I understand fully that table 1 is a complete release of the entire nuclide inventory.

DR. RUBIN: I understand that too.

DR. McCREATH: Table 2 represents, from AECL's calculations, the upper end of their 100,000 year spectrum.

MR. FAIRLIE: That's right.

DR. McCREATH: Therefore, in that sense, the worst case of their calculations --

DR. RUBIN: The 95th percentile worst case, perhaps we can say, to clarify.

DR. McCREATH: That's correct.

DR. RUBIN: Also let me just mention briefly, Dr. McCreath, before you joined the panel I spoke at some length about population dose, and in my two written submissions to the panel I have gone into that in some length, and on page 13 of my March 1st '96 submission I quoted from the AECB's Advisory Committee on Nuclear Safety, its document Principles and Guidelines for Radioactive Waste Disposal Facilities, where all of section 4 of that document, in fact, was entitled "Collective Risk and the ALARA Principle," and specifically how that should be applied, and I think it's fair to say, as Ian said, several of the authors endorsed but said caution is applicable, and one of the reasons caution is applicable is the kind of zero infinity problem that you expressed puzzlement over, how a number could be close enough to zero that an individual might not be willing to cross the street to avoid that risk, and yet society might well want to avoid imposing the risk on all of its members.

THE CHAIRMAN: Microphone number 1.

MR. WILSON: Dr. Fairlie, my name is Ian also. Wilson is my second name, though. I am with the Canadian Nuclear Association.

Wouldn't you agree that the application of the linear hypothesis to small doses, of a collection of small doses, goes with the caution that its an upper bound in terms of estimates? In other words, the range of estimate ranges from zero to the number calculated by the linear hypothesis; it really isn't a predictive so much as it is a cautionary estimate?

MR. FAIRLIE: Yes. I didn't hear what you said, but I will answer to what I thought I heard you say.

Yes, the linear no threshold hypothesis is a conservative assumption, no doubt about it, and I think that it's wise to take that precautionary approach for the time being.

I don't see much argument about this at all, because almost all of the international authorities are quite agreed on the adoption of this approach.

MR. WILSON: Would you also agree that, based on your calculations of collective dose from your table 2 illustration, that the total collective dose calculated could also be said to be trivial?

MR. FAIRLIE: In table 2, this is what the calculations are after using the case study, the post-closure case study and all the assumptions inherent therein, and that's what the result comes out as. I haven't made any comments at all about the wiseness or otherwise of all the assumptions in the post-closure case study. I'm saying that, based on AECL's figures, this was what the result would be.

A SPEAKER: (Inaudible)...(no microphone)...

MR. FAIRLIE: No, I didn't say that. I said that's what the results would be.

MR. WILSON: No, but those which would result in the prediction of between zero and two deaths in the period of 100,000 years could hardly be said to be significant. Surely you would agree that that would be considered by most reasonable people to be very trivial indeed.

MR. FAIRLIE: Well, I --

THE CHAIRMAN: I think we are just getting into judgment there, and I doubt that we will handle it.

May I also point out that we are already ten minutes over the time when we should have been at presentation and questions and answers. If there are additional questions, I would urge that they be very brief, because I can only give another couple of minutes given the other presentations for this afternoon.

Microphone number two.

MS. LAWSON: Pat Lawson, speaking to Ian Fairlie. In this matter of collective dose, do you arrive at your conclusions through mathematical modelling?

MR. FAIRLIE: Yes.

MS. LAWSON: As with the panel member, I too have struggled with this subject in attempting to calculate the dose for the population of Port Hope, and I would just like to add two other factors here that seem to me very important; one of them in the matter of collective dose is the whole time factor. And is that calculated into the mathematical modelling?

MR. FAIRLIE: Yes, it is.

MS. LAWSON: The other one that I don't see how it could be calculated into the modelling is the multi-factorial characteristics of human health.

THE CHAIRMAN: A final question from microphone number 2.

DR. RUBIN: Yes. I was interested in your conclusion that at very low doses, which is clearly what we are talking about in the table 2 scenario, that the linear hypothesis of ICRP is a conservative assumption.

First of all, I believe you and I are both old enough to remember when other models were used by ICRP which had thresholds, and since then they have adopted a model as being more appropriate which doesn't have a threshold. Could you confirm that first, and I will do a quick follow-up --

MR. FAIRLIE: As I can remember way back then, yes.

DR. RUBIN: Yes. And isn't it scientifically true to say that we can no more prove conclusively that the linear hypothesis number is correct, or that it isn't zero, we can no more prove that than we can prove that the effect isn't, say, twice as high at those tiny levels?

MR. FAIRLIE: Yes. That's a good way of putting it.

DR. RUBIN: Thank you.

MR. FAIRLIE: But the evidence that we have for dose response relations has been limited to fairly high doses, Chair, and when you extrapolate down, you enter a foggy area, and the ICRP has adopted a conservative hypothesis that it's a linear relationship down to zero, and --

DR. RUBIN: I guess "conservative" means it overstates, and we don't know that either. It's better than assuming there is no effect, that's for sure, but it's not a worst case, as Ian Wilson suggested and as you seem to agree.

MR. FAIRLIE: Well, no, I wouldn't say it's a worst case. I would tend to use the word "conservative" in the way that it's normally meant to be used, Chairman.

THE CHAIRMAN: Thank you very much --

MR. FAIRLIE: Can I finish with one statement, Chair?

THE CHAIRMAN: A very brief one--

MR. FAIRLIE: Very brief. The AECL has come in for a lot of stick in this study, and in my case, my opinion, many of the criticisms are valid. Nevertheless, there are a number of individual scientists who work for AECL whose work is very good indeed, and I want to put that on public record. I won't point out who they are, but their work is very good.

THE CHAIRMAN: Thank you, Dr. Fairlie, for the presentation and your full answers to the questions.

Our next presentation is by Mr. Phil Richardson on behalf of Northwatch. This one, because Mr. Richardson is not in town, in fact not in the country, will be coming to us by telephone. I trust that Mr. Richardson has been alerted to the ground rules, a fifteen-minute maximum presentation. When it comes to any questions afterwards, would you just remember, please, there must be a very brief pause, a split-second pause between putting the question and getting an answer because of the delays on transatlantic transmission.

Mr. Richardson, wherever you are, you are on.

MR. RICHARDSON: Thank you. Can I assume that I can be heard?

THE CHAIRMAN: If you can increase the volume, it would help.

MR. RICHARDSON: Okay. I shall now begin.