ML20211H918
| ML20211H918 | |
| Person / Time | |
|---|---|
| Issue date: | 02/19/1987 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1575, NUDOCS 8702260324 | |
| Download: ML20211H918 (288) | |
Text
-
f ONGg_ NW M uni 1ED STATES
, g' '
1 NUCLEAR REGULATORY COMMISSION IN THE MATTER OF: DOCKET NO:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON WASTE MANAGEMENT i
l j
LOCATION: WASHINGTON, D. C. PAGES: 1 - 217 l
l DATE. THURSDAY, FE3RUARY 19, 1987 i
l l
^
m, < * * 'I
,P; C @ 'l fsgmz! 'I, L . -., ., r r .
j( t [a . .' b ' 'i~'
ACE-FEDERAL REPORTERS, INC.
l Official Rqcrters l
444 North Capitol Street Washington, D.C. 20001 8702260324 070219 (202)347-3700 PDR ACHS 7_;373 P. DR NAT ONWICE COVERAGE j
t
(} PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSIONERS' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS THURSDAY, FEBRUARY 19, 1987 The contents of this stenographic transcript of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards (ACRS), as reported herein, is an uncorrected record of the discussions recorded at the meeting held on the above date.
No member of the ACRS Staff and no participant at
() this meeting accepts any responsibility for errors or inaccuracies of statement or data contained in this transcript.
O
1 CR29879.0 BRT/sjg 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4 SUBCOMMITTEE ON WASTE MANAGEMENT
/
' 5
/
l Nuclear Regulatory Commission f
6 Room 1046 1717 H Street, N.W.
Washington, D. C.
7
, 8 Thursday, February 19, 1987 9
The subcommittee meeting convened at 8:30 a.m.
10
-11 ACRS MEMBERS PRESENT:
DR. DADE W. MOELLER, Presiding 13 DR. MAX W. CARBON 14 DR. J. CARSON MARK
^
15 DR. FORREST J. REMICK 16 DR. PAUL G. SHEWMON 17 18 ACRS CONSULTANTS PRESENT:
l
. 19 R. DILLON
~
W. KASTENBERG 20 K. KRAUSKOPF F. PARKER 21 M. STEINDLER 22 O. MERRIL, COGNIZANT ACRS Staff Member
, 23 24
- o 2e I
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800 336-6646
29879.0 2 BRT
- , g 1 PROCEEDINGS 2 MR. MOELLER: The meeting will now come t.o orde,r.
3 This is a meeting of the ACRS Subcommittee on Waste 4 Management.
5 I'm Dade Moeller, the chairman of the i
6 Subcommittee. We have in attendance with us today, the
) 7 following other members.of the ACRS: Forrest Remick, Max 8 Carbon, Paul Shewmon, and we anticipate having Carson Mark 2.
9 join us. We have a team of consultants with us, Martin 10 Steindler, Frank Parker, K. Krauskopf, and W. Kastenberg.
11 The purpose of the meeting is to -- oh, excuse
'12 me, R. Dillon. Thank you. He's so slender I didn't see t .
13 him.
(-- 14 The purpose of this meeting is to review several 15 pertinent nuclear waste management topics in accordance 16 with the proposed presentation schedule.
17 In the way of background on what we are to cover,
- 18 let me offer the following comments. The meeting will be a I
19 two-day meeting, today and tomorrow, and the principal
, 20 topics for discussion today are as follows.
?
i 21 First of all, the rulemaking for developing a 22 definition for high-level waste, and relative to this, DOE 23 has contracted with the Oak Ridge National Laboratory to 24 develop a proposed classification system for high-level and 25 other types of radioactive waste, and we have distributed i
i AcEJFEDERAL REPiORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6M6
. - _._._- __._ _ . _ , _ _ _ _ _ _ _ _ . - ~ _ _ . . . . _ . _ , . _ . _ . _ _ _ _ - . _ . _ . . _ _ . . . . _ - . _ _ . , _ _ . _ - . _ _ _ , .
29879.0- 3 BRT
() I this morning to each of you, a copy of the executive 2 summary of that report.so that you will be aware of it.
3 I would believe --
4 DR. REMICK: I don't think we've got it yet.
5 MR. MOELLER: Owen, have they not yet received 6 the executive suqmary of Oak Ridge? I think we'll pass 7 .that out right now. We'll want to take that into 8 consideration when we discuss the subject with the-NRC 9 Staff.
10 I know, in terms of classification for 11 high-level waste, certain ACRS members have raised 12 questions; for example, one of the questions raised is:
13 Where are control rods covered? Old, used control rods?
14 So we'll be discussing that as our first topic this morning.
15 The first topic will be draft allocations, and 16 here we are talking about looking for individual components 17 in a high-level waste repository as well as systems 18 consisting of several components, and how much weight or 19 how much faith can you put into a given system or a given 20 component within that system, in terms of meeting the overall 21 performance goals for a repository.
22 Another item that has come up within that is, if 23 you say I'm not taking credit for a given system or 24 component, should you not require that DOE censider any i 25 negative impact of that particular component or system. I ACE FEDERAL REPORTERS, }NC.
202-347 3700 Nationwide Coserage MX)-3346646
29879.0 4 BRT
/~T
( ,j 1 mean you can't just rule it out without at least looking --
2 looking at the potential negative impacts that it might 3 have.
4 The third item is assessing compliance with the 5 EPA high-level waste standards. This we have discussed 6 previously. In our discussions we realized that, certainly 7 from what we could see within the Waste Management 8 Subcommittee, that we did not have the talent that we 9 needed to really dig into that so we have Dr. Kastenberg ,
10 with us today to discuss that particular topic, and he has 11 also prepared written comments on the subject.
12 Then we are going to hear a briefing on the
- g- 13 description of the -- a briefing on the high-level waste
\ )) This, in essence -- perhaps I'm 14 licensing support system.
15 saying it in too simplistic a manner, but it is a computer 16 data bank in which the NRC hopes to include all the 17 references and background documents that are being used in 18 the evaluation of the high-level waste repository.
19 Then we'll wrap up the meeting today with a 20 discussion of the five-year plan for the high-level waste 21 program.
22 Tomorrow we'll begin with several reports from 23 the NRC Research Staff. The first one will be coverage of 24 the hydrology programs, domestic and foreign. The 25 Subcommittee, particularly Frank Parker, has been a prime j
Ace FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cmerage N W)-33MM6
4 '
$ 29879.0 5 BRT
( ): 1 mover in this area to be sure that the NRC Staff is working 2 with their counterparts in other countries to benefit, as 3 much as possible, from what is going on and what J
4 international cooperation is under way.
5 I believe that the NRC Staff is doing quite well
) 6 now in that area, so we'll hear a report on that. Then 7 we'll hear a discussion of the NRC waste package erosion 8 program, and Dr. Dillon is here to assist us with that.
4 f 9 And then, tomorrow afternoon we'll wrap it up with a 4
10 discussion of two of the primary documents that have 11 recently been issued with respect to the low-level waste j 12 program, and that is their standard review plan and their i
i 13 standard format and content document.
1 14 Then, lastly will be a discussion of a white i
! 15 paper which the NRC Staff has developed looking for a I
16 long-range plan for their low-level waste program. Today i 17 in essence is high-level waste and tomorrow afternoon is l 18 primarily low-level waste.
i 19 Each of the two days will conclude with an 20 executive session which is open to the public, and during 21 that time the Subcommittee will prepare -- will discuss
! 22 what we've heard during the day and begin to prepare
! 23 written comments on any topics that we believe require them.
24 Although it is subject to change and obviously subject to 25 what we conclude at the end of each day, my quess would be Act! FriotinAi. RrieonTiins, INC.
202 347 3700 Nationwide Cmcrage MX)-3% (M6
29879.0 6 l BRT 1
('y l
(,j 1 that we will try to prepare -- or I'm hopeful that we can l 2 prepare some comments on the definition of high-level waste; 3 some written comments. In terms of assessing compliance of 4 the EPA high-level waste standard with -- assessing 5 compliance, NRC and DOE repository, NRC licensed repository 6 with the EPA high-level waste standards, we have 7 Dr. Kastenberg's standards, so I'm not sure we'll prepare a 8 anything separate unless we have something to add to it.
9 We have another item which is not on the agenda, 10 which we must spend some time witnin terms of written 11 comments, and that is the draft paper that we have prepared 12 in which we are trying to compare the risk of a high-level i
13 waste repository to the risk of a nuclear power plant.
14 The ACRS Planning Subcommittee has been pushing 15 this Subcommittee for quite some time to prepare such a 16 report and in the minutes of their meeting of December 10, 1
17 1986, they pretty much mandated that we put this near the 18 top of our list of important topics and that we get on with 19 it. So we prepared a draft and we want to really take time 20 to discuss that and take your guidance on where we go from 21 here. The 5-year plan, unless the Staff request it, we'll 22 simply use our oral remarks and the transcript of a record 23 of our comments on that subject.
24 On the low-level waste side looking toward 25 tomorrow, I believe I'm correct in saying that the Staff
(~)h
\_
Aci Fiumnat. Riii'onTi:Rs, INC, 202.147 4 700 Nanonwide Cmcraye M00 Un#46
29879.0 7 BRT
() 1 specifically has requested written comments on their 2 standard review plan and on the standard format and content 3 document. Those are in published, quote unquote, " final 4 form." But, nonetheless, I believe they do want comments.
5 It may be that we will have -- that we will be unable to 6 prepare our comments by the end of this Subcommittee 7 mooting. If so, we'll simply let it spill over to another 8 one and cover it next time.
9 The other thing about those two documents is 10 that they are very thick and it may well be that what we'll 11 want to do here will be simply to assign different chapters 12 or sections to different consultants and handle it in that 13 way.
14 As a final comment before going on with the 15 other information about the meeting, I wanted to i n f o rm , or 16 be sure that all the Subcommittee members and consultants 17 knew that the ACRS last week submitted to Congress its 18 review of the NRC safety research program and in that we 19 had, I thought, a very good section on waste management, 20 comments on that. And that included, of course, what you 21 people have fed into the Subcommittee and, you know, the 22 results of our meetings over the past few months.
23 Owen Morrill is the cognizant Staff member for 24 today's meeting. The rules for participation in today's 25 meeting have been announced as part of the notice of this O
Acti 17tiDliRAl. Ill!!'ORTf!RS, INC, 202.ut Um NationwMe Cmeraye Hma %IMti
29879.0 8 BRT
.( ) 1 meeting that was published in the Federal Register on 2 February 4, 1987.
3 If there are members of the public that want to 4 have input into the Subcommittee meeting, please let Owen 5 know, or even,-I think as we run the meeting over the next 6 two days, if you simply raise your hand I -- and if the 7 opportunity is there, I'll recognize you and if you'll go 8 to the microphone, we'd certainly like to hear comments 9 that you have.
10 It's requested that every speaker first identify 11 himself or herself and speak with sufficient clarity and 12 volume so that everyone here can readily hear what is being 13 said.
14 We have received no written comments or requests 15 for time to make oral statements from members of the public 16 at this moment, but again, if you have comments, simply let 17 us know.
18 We will now ask if any of the Subcommittee 19 members or the consultants have comments that they want to 20 make prior to launching into the first topic? Does anyone 21 have anything?
22 Okay. Then we'll call upon Dr. Daniel Fehringer, 23 who will begin with a discussion of the rulemaking for 24 high-level waste definition. Dan, if you'll sit right here 25 at the table, it will be fine.
O ACE-FEDERAL. REvonTEns, INC.
202 347-)?00 Nationwide Cmerage RX).33MM6
29879.0 9 BRT
( ). 1 DR. REMICK: Dave, this is one question I think 2 Dr. Fehringer can answer. We just received.the draft 3 report that Owen gave us, " Closed classification system for 4 high-level and other radioactive waste." .Is that S' consistent on what we are going-to hear on the proposed
! 6 rulemaking? .Just reading as Dade was talking, this seems 7 like it's consistent but I'm not sure. If it is completely 8 consistent I won't continue to read the Oak Ridge paper.
9 DR. FEHRINGER: I haven't seen that particular 10 paper. You are referring to the Oak Ridge paper?
i l 11 DR. REMICK: Yes.
12 DR. FEHRINGER: I have seen earlier drafts of it 1
j s 13 and it's generally consistent with what I'm presenting 4 14 today.
i 15 MR. MOELLER: I would offer the comment, I have 16 read both of them as carefully as I could and, to me, the-i 17 main difference with the Oak Ridge approach is that they l
18 place heavy emphasis on the heat generating capacity of the l
a i 19 waste. That is one of their basic parameters upon which
- 20 they classify the waste.
i 21 Now, if you read the NRC document you can find 22 heat generation on page 36 or somewhere like that but it's 1
23 not as prominent.
24 DR. FEHRINGER: I have had passed out a set of
! 25 handouts that summarize what is in the advance notice that O
t 1 \
! /\CE-FEDERAL REPORTERS, INC.
! i 202-347 3700 Nationwide Coserage 800-336 4 46 i
_ . - , - m.__, . . _ - - . _ - , _
29879.0 10 BRT em is ) I we have sent downtown. Before I start on the handouts let 2 me give you the status of that advance notice. It has been 3 sent to the Office of the Secretary. I tried to determine 4 whether or not it had been signed out of the Secretary's 5 office and sent to the Federal Register for publication and 6 I was not able to determine that, but if it has not been 7 signed it will be in a day or two, and that would mean in 8 two or three weeks the Federal Register will have the type 9 set and have it published and distributed.
10 DR. REMICK: Why did you decide to go the 11 advance notice route rather than notice of proposed 12 rulemaking?
13 DR. FEHRINGER: We thought there were enouah p]s L
14 significant issues involved that we thought it would be 15 worth having two opportunities to public comment. In 16 particular, in the advance notice we wanted to try to get 17 some general agreement on the general concept of how we are 18 defining high-level waste, or if we cannot get that general 19 agreement we vant to know that before we put a lot of time 20 and effort into the supporting analyses. It was just to 21 provide that extra opportunity for public comment.
22 DR. REMICK: Thank you.
23 DR. FEHRINGER: On page 1 of my handouts I 24 summarize the purpose of this rulemaking, and that is to 25 revise the definition of high-level waste in part 60, to O
v ACE-FEDERAL REPORTERS, INC.
202-347 37(M) Nationwide Cmerage R(x)-33MM6
._ . _. - .- _ ~ ._ _ _ _ _ ._ _
4 29879.0 11 BRT
() 1 conform to the definition that is in the Nuclear Waste 2 Policy Act.
3 In the middle of the page, the definition from i 4 the Waste Policy Act is given. It has two parts: Clause A 5 refers to waste from reprocessing plants. It is the same j
6 as our current definition with one exception, and I have 7 underlined the relevant wording. If we would include as l 8 high-level waste only those wastes that have fission 9 products in sufficient concentrations.
I 10 Then clause B is very much different from our i
11 current definition. High-level waste would also include 12 other highly radioactive material that the Commission (g 13 determines requires permanent isolation.
\_) 14 Clause B is the part that is of most interest in i
4 15 this advance notice from the technical point of view. From 16 the policy point of view there's a significant question as 17 to whether or not we should define the "in sufficient 18 concentrations" wording, or whether we should continue to 19 classify waste more by source rather than with a numerical
! 20 definition, when they come from a reprocessing plant.
- 21 Just in contrast, at the bottom of the page is 22 the current part 60 definition. I think you are familiar l 23 with that. It refers to spent fuel or to liquid or solid 24 reprocessing wastes, from reprocessing spent fuel.
25 DR. SHEWMON: Sir, the first line in definition
! C:)
i ACEJFEDERAL REPORTERS, INC.
Nationwide Coverage 202 347 3700 800-336 4 646
. --m..-___,,-._._,.._.~,_-,,-._,-_.__._.,---,.__.-.___.... . . _ _ , ~ . - . . _ _ , . _ , , _ - . - -
29879.0 12 BRT
() 1 A, the " highly radioactive material," has never been 2 interpreted by the Staff to be the same as "in sufficient 3 concentrations"?
4 DR. FEHRINGER: I think it would be the same, 5 that the "in sufficient concentrations" fission products is 6 what makes a waste highly radioactive, and if the 7 concentrations were not sufficient, the waste would not be 8 highly radioactive, would not be high-level under Clause A.
9 DR. SHEWMON: So you are saying you may need the 10 second part, but I don't understand why you would if, 11 indeed, you said something about a definition of what was 12 highly radioactive instead of worrying only about fission 13 products.
O 14 DR. FEHRINGER: Clauses A and B refer to 15 different clauses.
16 DR. SHEWMON: I'm sorry, high radioactivity is A, 17 I'm only talking about A.
18 DR. FEHRINGER: Within A, I interpret the highly 19 radioactive and in sufficient concentrations as meaning the 20 same things. The insufficient concentrations is the 21 further interpretation of what highly radioactive means.
j 22 DR. REMICK: I could argue differently, though.
23 You could have something with a very short half-life, 24 highly radioactive but short half-life would meet that 25 definition, those words in isolation. Associated with O
Ace FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Cmerage 800-33MM6
29879.0- 13 BRT
() 1 reprocessing and so forth. In general, highly radioactive i 2 is more than the concentration, isotope --
3 DR. FEHRINGER: If it's measured in curies'in
-4 half-life, then it's correlated with radioactivity; if it's
! 5 measured in grams of volume of material then your point is 6 well taken; the concentration could be very low even with
- 7 high activity. Parallels the question of'transuranics and 8 induced activity in fission products.
9 It's a little odd that Congress focused so f 10 strongly on the fission products when that version was put i
11 in' Clause A, but we think it was meant to distinguish j- 12 wastes that look like transuranic wastes from wastes that i
13 really should be considered high-level wastes, and that's 14 the approach we are taking in the advance notice. If 15 there's not a sufficient level of radioactivity, then a f 16 waste would more properly be treated like a ' transuranic i
17 waste.
18 On page 2 I summarized what the current problem l
! 19 is with the existing definition of part 60. It is a 1 i i L
20 source-based definition and it does not correspond well to i 21 the hazard of a waste. A risk-based definition would
! 22 improve the match between wastes and disposal facilities. t i
23 The drawback, of course, is that a more precise 24 definition based on hazard or concentration would require
- 25 more characterization of wastes. Source-based definition l
i i
! ace-FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coserage Mn336-ua6 L
.~ - -. . - . . .. . _~ . -.. .. _
l i !
29879.0 14
.' BRT i
() 1: is simple, in the sense that you only need to know where i 2 waste came from in order to classify it.
i 3 A more precise definition of high-level waste 4 would identify the need-for waste generators to enter into 5 contracts for transfer of high-level waste to the '
6 Department of Energy; would allow the Department to plan i 7 for receipt of those wastes; and it would determine which l
8 of our regulatory requirements would apply to specific 9 types of waste.
j 10 In part 60, we have at least one requirement, 11 .the waste package requirement, that applies only to
! 12 high-level waste. Other parts of part 60 apply to any 1
l 13 waste that is disposed of in a depository. We need to put I '
14 the Department on notice regarding which waste will be 15 considered high-level and which not, so they can know which 16 requirements they must meet.
17 DR. REMICK: Excuse me --
l- 18 MR. PARKER: There's a third category, some of .
I i, 19 the defense establishments have high-level wastes not r
20 classified as high-level wastes. They also have DISs j
21 available now on what they propose to do and without l i
j 22 knowing what the definition is, it's almost impossible to I
t 23 determine what to do with those wastes.
I i 24 DR. FEHRINGER: That's a point I'll get to later.
i 25 Maybe I should mention it now. ;
)
! Ace I7EDERAt. REpoRTEns, INC.
- oL347 3700 Nationwide cmcrage un3%w6 i
_ . _-._____m_ _ _ _ . _ _ ._ - . _ _ _ . _ _ __ __ __ __
29879.0 15
.BRT $
i I ()
i 1 The Waste Policy Act has this definition, but
- 2 the Waste Policy'Act also says very explicitly that that i c i
j 3 definition is only for purposes of that act.
4 Our authority to license disposal of high-level 5 wastos derived from the Energy Reorganization Act and that l l 6 was not changed. Presumably we still have licensing i 7 authority for those wastos even if we classify them as !
j 8 something else under the Waste Policy Act. That's one of ;
I 9 the mossy aspects of this whole project. But the !
l 10 department does have opportunities to consider other i
i methods of disposal other than repository.
11 j 12 MR. PERRY: Doesn't the Reorganization Act, 11 though, separato the responsibility for the defense waste
([)
14 from the NRC, in terms of licensing the transuranic? ;
i 15 MR. MOELLER: Excuse me, that's Jack Perry. L 16 DR. FEllRINGER: I don't recall there's a
]
- 17 difference betwoon defenso wasto and disposal waste. t i
I 18 There's no NRC authority for licensing disposal of ;
i 19 transuranic wastos in the defonso sector and that's very I 20 clearly stated in one of the DOE authorization bills.
4 !
l 21 MR. PERRY: It's my understanding part of the 22 rationale before this dofinition was related to the defonso j 23 transuranics, but that's an undorstanding, not a fact that i I
24 I really have substantiated.
I 25 DR. FEllRINGER: It's still not clear what tho l I :
i I i Acit.171ti)iinAt. Ri:i>onnius, INC.
j 202-3471?m NationwIJe rmerage sn)m(M6 l
29879.0 16 BRT
() 1 point is, though. Wo recognize that wo do not have any 2 authority to license disposal of defense transuranics, but 3 we think there may be at least a few of those wastos that 4 should be classified as high-level, and that is one of the 5 issues that we are raising in this advance notico.
6 MR. PERRY: Some of the defenso waste streams do 7 contain low lovels of fission products --
8 DR. FE!! RINGER: Contain high levels of fission 9 products and those are the ones that possibly might be 10 candidatos for classification as high-level.
11 MR. STEINDLER: When you talk about candidato 12 for classification as high-lovel, what is the basis on 13 which you make that comment? Is that on the basis of the 14 anticipated disposal or economics?
15 DR. FEllRINGER: No. The characteristics of the 16 wanto. The concentrations of the different radionuclidos 17 present in the wasto.
18 MR. STEINDLER: So is there somo arbitrary 19 fooling among the Staff that, beyond a cortain 20 concentration, wo'll put a different label on that? You 21 know, what constitutos this fooling you are talking about.
22 DR. FEllP7NGER: Wo are not proposing to bo 23 arbitrary about it by any means, but wo do soo wantos that 24 look very much like the wantos that como from a 25 reprocosning plant. Whether wastos beqin to havo O
Acit.17tii)i Rai. Ittil>oR ri!Rs, INC.
20214717m mnonsude nwage m 3 vi u,4,,
29879.0 17 BRT
() I concentrations similar to reprocessing wastos then we think 2 they are candidates for classification as high-level.
3 DR. SilEWMON : With a half-life comparable to 4 that?
5 DR. FEllRINGER: Yos. Some of the transuranic 6 wastos have the long half-lives that are characteristic of 7 reprocessing waste and also have fission products, 8 contaminants, comparable to the figures product 9 concentrations in reprocessing wastos. Only small 10 fractions of DOE's wastos have that, but enough to be a 11 concern.
12 DR. REMICK: It would bo easy for me to conclude 13 that one should left the paragraph A definition alone and 14 have it source because, well it's kind of a tradition.
15 Maybo some of us fool wo know that what means. I'm not 16 sure that wo do. But I assumo that the down sido of that 17 is that there could be a lot of what might othorwise bo 18 classified as low-level waste which would havo to bo 19 handled as high-level wasto?
20 DR. FEllRINGER: That's exactly the issuo.
21 Particularly the wastos in the tanks at llanford.
22 Hany of thoso wanton -- most of them, havo 23 relatively low concontrations of nuclidos for novoral 24 reasonn. Ono is that they woro generated by a different 25 technology than Purox technology, and they woro much more O
Aci!.Ft:.i>iti<Ai. Illinoi< iiiits, INC.
l 202 m.non sanonede rowrage m .1 9,t u ,
l 29879.0 13 BRT 1 dilute to begin with. They have had the plutonium and 1
2 uranium separated, and now the cosium and strontium 3 separated and also there have boon wastos other than per ,
4 cycle reprocessing wastos put into those tanks, wastos from 5 other miscellaneous sourcos around the facility. So, some l
6 of those wastos are well below our class C limits that wo 7 refer to for shallow land burial. Those wastos are the 8 most obvious candidato for classification as something 9 other than high-level.
l 10 MR. STEINDLER: I guess I don't follow how 11 that's going to help it unloss you remove the sourco 12 identification.
l
! 13 DR. FE!! RINGER: That's what would be nocessary, 14 would be to classify them under the Wasto Policy Act on tho l
l 15 basis of the characteristics: The lack of sufficient l 16 concentrations of fission products.
l 17 MR. STEINDLER: At that stago of the como you 18 are prepared to have somebody simply go through and diluto l 19 all the wanton with aufficient inert matorial to got below 20 class C lovel and put it in shallow land burial?
21 DR. FEllRINGER: That's one of the problems 22 involved in numerical definition, you have to have somo 23 prohibition against dilution for that reason.
24 Oporationally --
25 HR. STEINDLER: I wann't suaqosting that's s
Acit.171!!)liRAl. RiiPORil!RS, INC.
I 202 147 17(o N.itiorimide Emcr6 ige !!(0 .116 (466
29879.0 19 BRT
() 1 necessarily a bad idea. I was just asking what the 2 standards will do about it.
3 DR. FEHRINGER: Yes. One of the questions we 4 face. I'm not sure this rulemaking is really the place to 5 address it. At least for our licensees, license conditions 6 on waste generators probably is the more effective way to 7 control any dilution that might occur. But it is a point 8 that we need to address.
9 MR. PERRY: Dan, under idea B, does the 10 determination of what requires permanent isolation -- is 11 that going to take into account the physical form of the 12 waste?
13 DR. FEllRINGER: Yes.
14 MR. PERRY: For instance, you did talk about the 15 llanford wastes which are water soluble and would be 16 amenable to rapid movement.
17 DR. FEllRINGER: Yes. I'll get to that in a 18 little bit more detail in a couple of more pages.
19 MR. MOELLER: Marty, I don't really know the 20 answer, you know, to your comment. I have thought about it 21 from time to timo.
22 Years ago, when there woro higher levels of 23 fallout in the United Statos and milk was contaminated, the 24 Public Iloalth Service at that time in their documents and 25 so forth always clearly stated that they woro against a O
l Acti lil!DiillAl, Illil'Ol[IllitS, INC.
20:44747m Nanonwide cmcrage sm.n6 ua6
m g,
29879.0 20 BRT h
( ). _1 milk-producing -- milk bottling company taking contaminated 2 ' milk and mixing it with uncontaminated to meet the standard.
3 -They always said you had to take the contaminated milk and 4 dump it or do something else with it.
5 MR. STEINDLER: I suppose that comes from the-6 very' low levels of the linear hypothesis, as the basis.
7 The thing I was looking for here is to see i 8 whether or not there's any risk-based thinking that goes h into these definitions. I don't yet see any. That may be
/10 too revolutionary at this point to ask for it. But in the
/ 1
~
11 last analysis isn't that really what we should be doing?
12 DR. FEHRINGER: On page 3, I make note that
-13 there's also the Low-Level Radioactive Waste Policy Act.
14 That act has been passed fairly recently. It does a couple 15 of things. First it defines low-level waste as waste that
. 16 is not high-level waste, or a couple of other categories.
17 Because of that when we define high-level, we will also be
! s 18 defining low-level. We will be drawing a line between the i
19 two classes of waste, in effect.
! 20 That act also establishes the responsibilities I
i 21 for disposal of radioactive wastes.
22 The last timo we discussed this advance notice, f
this act had not yet been passed and any definition of 23 24 high-level waste by the Commission would have had the l 25 effect of dividing up the responsibility between state and O
i /\CE FEDERAL REPORTERS, INC.
! 202-347-3700 Nationwide Coserage 800-336-6646
29879.0 21 BRT-(f 1 federal governments for disposal of wastes. That was a 2 very touchy issue and one of the reasons we found it 3 difficult to make progress on this project.
4 Congress has now set the class C limits as a 5 dividing line between state and federal government 6 responsibility for disposal. The Federal Government will 7 dispose of all wastes that have concentrations above class 8 C, whether they are high-level or whether they are a 9 special kind of low-level waste. So our definition of 10 high-level is irrelevant as far as the separation of 11 disposal responsibilities among state and federal 12 governments. And this rulemaking of course will not_ change c 13 that. That has been established in the act of Congress.
14 Page 4 summarizes our current waste classification system.
15 On the left --
t 16 MR. MOELLER: Excuse me a second, though. Back 17 on this, you still have a system approach. I'm following 18 you, any waste above class C is a federal responsibility.
19 Not only -- I mean for actual disposing and placing and so
-20 forth. But even the low-level wastes become either a 21 federal or a state responsibility, depending on how you 22 look at it.
23 For example, if it's an agreement state, the 24 agreement state licenses the low-level site and somebody 25 runs it, operates it, disposes of it. If it's a O
AcsFEDERAL REPORTERS. INC.
202-347-3700 Nationwide Coserage 800-336-6M6
29879.0 .
22 BRT p
(_) 1 nonagreement state, the NRC still licenses that low-level 2 facility.
3 DR. FEHRINGER: Right. But the state government 4 must make the arrangements to actually carry out the 5 disposal, either doing it themselves or by contract.
6 I'm not referring to the licensing here. Rather, 7 I'm referring to the responsibility for actually building a 8 facility and accepting wastes.
9 MR. MOELLER: Okay. I guess I wasn't straight 10 on that. You are saying to me that every low-level waste 11 facility in the United States, whether licensed by the 12 federal -- by NRC or by an agreement state, will be 13 operated by the state?
v 14 DR. FEHRINGER: Right.
15 MR. MOELLER: By a state?
16 DR. FEHRINGER: By a state or by a compact of 17 several states cooperating together.
18 MR. MOELLER: And the states, then, will 19 contract, perhaps, with some organization to actually 20 operate it for them, but the state holds the responsibility?
21 DR. FEHRINGER: Right. State holds title to the 22 land and ultimately takes over responsibility for long-term 23 care of the site.
24 DR. REMICK: But the licensee in the nonagreement 25 state could be a corporation but still the state would have
~
[V !
Ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
29879.0 23 BRT
() I the long-term responsibility? Licensee would not 2 necessarily be the state?
3 DR. FEHRINGER: I think that's correct. I'm not 4 familiar enough with that level of detail.
J 5 Page 4, I have a schematic representation on the 6 left of our current waste classification system. We have 7 classes A,.B and C well defined, at least the upper limits 8 of each class are defined. But above class C waste we do 9 have the source-based distinction. Wastes are low-level if 10 they come from sources other than a reprocessing plant; 11 they are considered high-level if they come from a 12 reprocessing plant, though they may have exactly the same 13 characteristics. The source is the only distinction 14 between them.
15 On the right-hand side of the page I just showed 16 the classification we would like to achieve. Draw a sharp 17 line which will distinguish high-level wastes from those 18 that are nonhigh-level. We think that will leave an above 19 class C class of wastes -- those that are not routinely 20 accepted for shallow land burial but still do not have the 21 characteristics that would make them high-level wastes.
22 DR. REMICK: When you say " desired" here, I 23 assume that this is the " proposed" in the advance notice?
24 The direction the NRC is heading?
25 DR. FEHRINGER: Right. Right.
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
29879.0 24
'BRT s
1 MR. STEINDLER: You.did mention that in the 2 current waste classification above class C, there, they are 3 source-related rather than concentrations or whatever. But 4 the implication is that's not true on the right side of 5 this page, and-that's not quite right, is it? Your 6 high-level definition is still source-related?
7 DR. FEHRINGER: For wastes other than 8 reprocessing wastes, we want to have a strictly risk-based 9 distinction. Draw a line. If they are above that they are 10 high-level. If they are below that they are low level.
11 MR. STEINDLER: But your discussion on
- 12 sufficient concentration, for example, at least left me 13 with the feeling that when all else fails, you go back to 14 the kind of things you find in the first cycle of solid 15 extraction as a standard or a guide or some measure of 16 appropriateness of what is sufficient.
17 DR. FEHRINGER: Yes. We list two options in the 18 advance notice. One is to do that because of legal 19 difficulties with existing laws. The second option is 20 apply the same standards, if they are below the line they 21 are low, above the line they are high. That causes some 22 awkwardness because of the the existing laws but 23 technically we think it makes some sense. Have the sharp 24 distinction and ignore the source of the waste; classify
! 25 the waste based upon the risk.
4 ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage MX)-336-6646
29879.0 25 BRT
() 1 MR. MOELLER: But what Marty is saying, and.I 2 agree with him if I'm interpreting it correctly -- in your 3 definition of high-level waste, the first definition, A, 4 you might have said: The highly radioactive material 5 resulting from the reprocessing of spent nuclear fuel 6 includes liquid waste produced directly in reprocessing and 7 any solid material derived from such liquid waste. And 8 then you would have said: As well as any liquid or solid 9 radioactive waste that contains fission products in 10 sufficient concentrations. Then you would have divorced it 11 from the source.
12 MR. STEINDLER: Well, almost.
13 I think the problem is that historically, the 14 Commission has been stuck with its original definition and 15 continues to adhere to that first cycle of solvent 16 extraction as the -- at least primordial standard against j
17 which everything seems to be judged.
18 Well, aside from the fact that that is a 19 floating crap game, quite frankly, it just doesn't seem to 20 fit, even ac a patch job, in the revisions that people are 21 currently looking at. I realize they are constrained 22 because they are unlikely to be able to change the Nuclear 23 Waste Policy Act. But it would be nice to avoid saying 24 because you are from across the tracks, you have to be 6 25 feet tall. That all 6 feet tall people have to use this O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 % *6
~29879.0 26
-BRT r '
1 elevator.
2 DR. SHEWMON: Or 6 feet underground.
3 MR. STEINDLER: Whatever.
4 MR. MOELLER: In part B, it has nothing about 5 the source. It's just "other highly radioactive material."
6 Then for part A it would have been better to have wiped out 7 any reference to the source?
8 MR. S'I EINDLER: Precisely.
9 MR. MOELLER: I agree totally.
10 MR. STEINDLER: That's my feeling.
11 DR. REMICK: Dade, the wording you proposed, as 12 I heard it, would not solve the problem that you would have 13 low-level waste from reprocessing that perhaps you don't 14 want to put in a high-level waste depository. You'd still 15 have that problem.
16 MR. MOELLER: You'd still have the problem.
j 17 Well, we are agreed that it should not be source-related 18 but should be -- depend upon the characteristics of the 19 waste and their associated risks.
i l 20 DR. FEHRINGER: I hope we made the point in the 21 text of the advance notice. Technically we think it should 22 .come out that way, too. We do have this problem with the 23 existing laws that we recognize, but our clear purpose, if l
24 we state it correctly, in the advance notice, is to 25 classify wastes based on their characteristics, rather than 1
l r
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646 L
29879.0 27 BRT r
(_)s I the source.
2 Page 5 is just another summary of the existing 3 waste classifications and I'll skip that without going into 4 any detail.
5 On page 6 I have summarized what we-think we can 6 accomplish in this rulemaking, and what the objectives are.
7 The first objective, we would like to place a definition of 8 high-level waste on a sound technical basis which relates 9 the classifications to the hazards of the wastes. That 10 would serve to notify waste generators of the need to enter 11 into contracts for transfer of wastes to DOE. People who 12 suddenly find they have high-level wastes need to make 13 arrangements to have them disposed of. And, as I said 14 earlier, will facilitate planning by DOE for receipt and 15 disposal of those wastes and will make clear which of our 16 regulations apply for disposal of different types of wastes.
17 The important point at the bottom of the page:
18 There will be no requirement in our regulations for
. 19 disposal of high-level wastes in a repository. The Waste 20 Policy Act comes close to requiring that because there is 21 no other disposal facility authorized. But if Congress T
22 would authorize some other means of disposal, our 23 regulations would permit that.
24 MR. MOELLER: Well, wait now. I saw that and 25 thought I understood it, but going back to your definition ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
_ .__ . _ _ . . . _ . . . - ~ _ _ . . _ _ . _ _ _ - . . ,_ __ _ _ _ _ _ _ . _ _ _ _ _
. . - - . .=. - .
29879.0 28 i- BRT 7- -
( l on page 1 of high-level rad waste, it must have the fission 2 products in sufficient concentrations and require permanent
- e 5
3 isolation.
4 Well, the only way you have at the moment to 5 require, to grant or to whatever the word is -- to give us 6 permanent isolation is a repository.
7 DR. FEHRINGER: Right.
8 MR. MOELLER: So then why don't high-level 9 wastes have to go in a repository?
f i 10 DR. FEHRINGER: If some other means of disposal i
11 would provide an equivalent means of isolation we would 12 find it acceptable at the moment.
I e i 13 MR. MOELLER: But you don't have it at the 14 moment.
15 DR. FEHRINGER: We don't know of any specific 16 project that's likely to provide that. The department went 17 through a number of alternatives several years ago, seabed 18 disposal, disposal in Arctic ice sheets and so on. Some of 19 those might provide an equal level, but they are not being 20 pursued.
21 One alternative that is of interest is the 22 current concept to dispose of some or all of the Hanford 23 waste tanks in place. There the department is proposing a 24 barrier on top of the tanks which they provide would 25 provide an equivalent level of isolation. They have a big l
(
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
29879.0 29 BRT
() 1 job ahead of them to prove that is so, but if they could 2 make that proof that would be an alternative we could find 3 permissible under our regulations.
4 DR. SHEWMON: Where they have gotten in' trouble 5 with some of those tanks in the past, or at least some, is 6 the barrier underneath, not the barrier above. What 7 happened to that?
8 DR. FEHRINGER: Their concept of a barrier would 9 prevent any moisture from traveling down to the tanks and i
10 effectively eliminate the need for any barrier below the 11 tanks.
12 DR. SHEWMON: They would solidify them and then
- ~ 13 if no water came in, nothing could get out; is that it?
14 DR. FEHRINGER: That's the concept. Right. The 15 barrier above the tank serves to trap any moisture that 16 falls on the site from precipitation and also serves as an 17 intrusion barrier in their conceptual design.
18 MR. PARKER: I might say that the Board of 19 Radioactive Waste Management, National Research Council, is 20 looking at just that problem right now. It will take about 21 two years before they are ready to report.
22 DR. REMICK: I have a question and I don't want 23 to nitpick words, but one, you say place the level 24 differences at a sound technical basis which relates waste 25 classification with waste hazards, where I see that, I mean i
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
l 29879.0 30 BRT
) I consequences, have you selected consequences rather than 2 risks? If I'm correct that hazard is equivalent to 3 consequence.
4 DR. FEHRINGER: You are right that I have used 5 the term a little loosely. On page 8 I'll get into that'in 6 a little more detail and discussion how we plan to treat 7 that.
8 DR. REMICK: All right.
9 DR. FEHRINGER: Page 7 shows an illustration of 10 the conceptual definition of waste -- high-level waste that 11 we are proposing in the advance notice. We are saying that 12 waste will be high-level only if it has two characteristics.
13 First, it is highly radioactive and second, it's in need of
- s 14 permanent isolation.
15 The vertical axis shows how we propose to define 16 highly radioactive. We will look at the concentrations of j 17 short-lived nuclides, which is where levels of 18 radioactivity come from. For the sake of illustration we 19 are saying that the class C limit of our current part 61 20 regulations makes a good illustration of where the line f
21 meet be drawn to separate highly radioactive from wastes 22 that are not highly radioactive.
! 23 I think that that pdrticular reference point 24 will probably remain in the proposed rule.
25 There is a good bit of subjectivity involved in l /\CE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
_ ., . - . _ _ _ _ _ , _ _ _ _ _ _ . _ _ _ _ . . . _ _ _ . _ _ . _ , , _ _ _ . _ _ _ . . . _ _ _ ~ _ . _ . _ _ . _ , . . - , , . _
29879.0 31 BRT
')
( 1 distinguishing' highly radioactive from non-highly 2 radioactive wastes. Class C limits do represent what I 3 think would generally be regarded as high levels of 4 radioactivity, so I think we know how to distinguish wastes 5 that are highly radioactive from those that are not.
6 MR. STEINDLER: How do those class C levels 7 compare to the intermediate-level vastes that are used in 8 Europe? Where they provide an intermediate definition or 9 definition of intermediate class between low and high-level 10 waste?
11 DR. FEHRINGER: I can't answer that right 12 offhand, but I can give you another reference point on the
(. 13 14 class C concentration limits in part 61.
approximately equivalent to the concentrations that They are
- 15 actually exist in the reprocessing waste at Savannah River, 16 the cesium and strontium levels in those wastes are about 17 at the class C limits. That's one more reason why I think 18 that those are high levels of radioactivity.
19 MR. STEINDLER: I don't understand the standard 20 that you are using. Just because Savannah River turns out 21 particular material, that therefore constitutes a 22 high-level of activity. I realize --
23 DR. FEHRINGER: Most people would generally 24 agree the Savannah River reprocessing wastes are highly -
25 radioactive. It is a subjective call, but I don't think
-p V
i ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-66*6
29879.0 32 BRT-I) .1 many people would disagree with that.
2 MR. STEINDLER: The question is a frame of 3 reference. Certainly I would not care to have my 4 bedsprings made out of Savannah River waste. But that 5 doesn't necessarily mean that from the standpoint of.both
, 6 disposal and the economics of disposal and the extent to 7 which you provide confinement for an undefined period of 8 time, that they would constitute a particularly high-level 9 waste. Their burnups are low. By now the. cooling times in 10 some of those tanks is, you know, at least one half-life 11 down.
12 On the basis of arbitrary numerology, I have e 13 some problems that say simply because it was generated at 14 Savannah River in the South it's a high-level waste.
15 DR. FEHRINGER: No, that's not the point. The 16 point is that we need a reference point for what is and is 17 not highly radioactive.
l 18 Kroft & Coker tried to use heat generation rates 19 to arrive at that and to some extent they also tried to use
- 20 external radiation levels. I'm not sure whether that 21 remained in the version that you were given.
22 Both of those may be useful. I also think that 23 we may find that the potential for risk to a human intruder 24 from various disposal means may provide some insights into 25 what is and is not highly radioactive.
O ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage MX)-336-6646
P
'29879.0 33 BRT (m) 1 MR. STEINDLER: Now you are gradually drifting, 2 I think, unfortunately at rather glacial speeds, to the 3 risk-related issues. You address them in Rogers and 4 company, reasonably well, whether you folks were fussing 5 around with the definitions of class A, B and C.
6 It's a little startling that the same 7 organization that did, I thought -- what I thought was a 8 fairly decent job in the low-level waste definitions, 9 totally comes apart as soon as you get past level C.
10 I grant you, and that's not necessarily -- it 11 sounds worse than it is -- I grant you that you have a 12 problem with the Nuclear Waste Policy Act which is
, 13 anchoring you on one side, and you are going to be unable
- 14 to tear yourself loose from that. But the rationale that j 15 says it's commonly accepted that a waste is high-level is 16 not really in keeping with your past approaches.
17 DR. FEHRINGER: Let me reiterate once more. We 18 are trying to-get the concept established that a waste is 19 high-level only if it has two characteristics: first, it's 20 highly radioactive, and second, it's in need of permanent 21 isolation. We need to define what we mean by both of those 22 terms. Right at the moment I'm trying to sort out what is 23 and is not highly radioactive. In a minute I'll discuss 24 what we mean, what we think the term " requires permanent
] 25 isolation" means.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 800-336-6M6
- , - - - - .~ - -,--.-.-.----- - ,--- --- . -. . . . , --
29879.0 34 BRT
,m
(,) 1 The highly radioactive one I find difficult to 2 put on a sound basis because there's no good reference 3 point. It's a subjective call, whether something is or is 4 not highly radioactive.
5 As I mentioned just a minute ago, that Kroft &
6 Coker have tried to work with heat generation rates and 7 external radiation levels, and I find that a little bit 8 lacking, in a way, because they are focusing primarily on 9 the management of a waste before disposal. I would prefer 10 to look at the risk of a waste after it is disposed of.
11 That's where I'm finding a good reference point hard to 12 arrive at.
13 The need for permanent isolation, though, is a b,s 14 characteristic that I think we do have a good handle on.
15 What we need to do is look at the types of disposal 16 facilities that might be used that are not as good as a 17 repository, and determine what wastes those can accept 18 safely, wastes that cannot be safely accepted in those 19 facilities, then, will be determined to need permanent 20 isolation and must go to a repository or equivalent. I 21 think that split between a repository and the next least 22 good facility -- awkward wording -- but the less secure 23 facility is the line that divides permanent isolation from 24 nonpermanent isolation.
25 MR. STEINDLER: So your comment concerning O
ace FEDERAL REPonTEns. INC.
l m .347 37oo sation iae coseraye son.336.y u
29879.0 35 BRT
( l pe rmanent is a geologic comment rather than a risk or 4 2 half-life or waste-related comment; is that right? If I 3 have material that has an 8-day half-life, permanent ;
e 4 isolation for that may be, for example, 50 half-lives.
5 DR. FEHRINGER: We are saying in our view l
6 permanent isolation corresponds to long-term hazards that i 7 require the long-term stability that a repository provides.
8 MR. STEINDLER: But that's independent -- that's 9 what I'm saying. That's independent of what is in the 10 wasto.
3 11 If I have a highly radioactive waste that i
12 contains an 8-day half-life material, you, I think, would i 13 require some kind of permanent isolation which, i t 14, unfortunately I think would drive you into a repository.
15 DR. FEHRINGER: No. We are saying that
- 16 pe rma ne nt isolation, the need for permanent isolation 17 occurs only when you have long half-lived materials.
f l 18 MR. STEINDLER: What's long? '
i l 19 DR. FEHRINGER: 100 years or so. We are trying 20 to make a split based on the short-lived versus the i
21 long-lived nuclide tables in the class C definitions and a i
22 100 year half-life is approximately the split between those i 23 two tables.
. 24 Notably, cesium and strontium, with 30-year
- 25 half-lives are considered short half-lives in those tables.
1 I /\CE-FEDERAL REPORTERS, INC.
! 202-347-3700 Nationwide Coverage SU)-336-(M6
29879.0 36 BRT
-( ) 1 And that's the point where I think you find a definite 2 distinction between shallow land burial and repository.
3 Shallow land burial provides isolation for a few centuries 4 but not for millenia, and that corresponds quiteJwell with-5 the implicit half-life distinction between short'and 6 long-lived in part 61.
7 MR. PARKER: In light'of Monty's remarks, have-8 you looked at the classification -- the Swiss system? They 9 have gone totally to a risk-based classification system.
10 There are three kinds of repositories, three kinds of waste, 11 A, B, and C, and it's strictly risk-based. I think it 12 would be well worth your while to look at their methodology.
13 DR. FEHRINGER: Do they have numerical Ot- 14 definitions?
15 DR. SHEWMON: Where is that published, Frank? I 16 asked for a copy a month ago and it hasn't been forthcoming.
17 It presumably means the Staff isn't finding it easy to lay 18 their hands on it.
19 MR. PARKER: The summary volumes -- there's one 20 in English --
21 DR. SHEWMON: Do you know whether or not the NRC 22 library should or could have a copy of that?
23 MR. PARKER: They could if they are easily 24 available.
25 DR. SHEWMON: From who?
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 800-336-fM6
- . ~,
29879.0 37 BRT
() 1 MR. PARKER: From NAGRA, N-A-G-R-A, it's the 2 acronym for Nationale Genossenshaft Fur die Lagerung --
3 something of that sort.
4 DR. SHEWMON: Thank you.
5 MR. PARKER: I can get the cite reference.
6 DR. SHEWMON: Fine. I would like to see a copy 7 in the library down there.
8 DR. FEHRINGER: Getting back to page 7, on the 9 horizontal axis we also want to draw a line that will 10 distinguish wastes that will require permanent isolation 11 from those that do not. Again, for the sake of 12 illustration, we are using the class C limits of part 61 to
't s 13 make that distinction. The table of long-lived nuclides in 14 part 61 will serve that purpose. That's what I've referred 4 15 to as table 1 on this graph.
16 This effectively divides all wastes into four 17 categories. The lower left-hand corne,r, we have the 18 existing low-level wastes, what we might now call ordinary
, 19 low-level wastes, which will be classes A, B and C wastes.
20 These are neither highly radioactive nor in need of i 21 permanent isolation. The lower right-hand quadrant would 22 have a type of above class C low-level waste that needs
{
23 permanent isolation but is not highly radioactive. An 24 example would be DOE's transuranic waste that contain 25 long-lived nuclides, but do not have very high levels of 3 /~T
- (/
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
l 29879.0 38 BRT T*
1 radioactive materials.
2 In the upper left-hand quadrant we'd have a 3 second type of above class C low-level waste. These are 4 . wastes that are highly radioactive but not long-lived.
5 Example would be the cesium and strontium capsules at the 6 Hanford site and also the example presented earlier where 7 an 8-day half-lived sample might be extremely radioactive 8 but not long-lived. Only those on the right hand would be 9 high-level; those both highly radioactive and in need of 10 isolation. Most reprocessing wastes we think will fall 11 into that quadrant unless something has been done to them 12 like what has happened at Hanford where the short-lived 13 radioactive material has been removed from those wastes.
14 On page 7A I have tried to sketch in where I T
15 think different wastes will fall in this type of a 16 classification scheme.
17 In the lower left-hand quadrant there are the i
18 decontaminated salt streams from Savannah River and West 19 Valley. These salts would contain only very low residual l
20 levels of radioactive materials. In the upper right hand l 21 quadrant we have spent fuel and commercial reprocessing l
l 22 waste, if we should ever have commercial reprocessing.
i 23 Also the Savannah River and West Valley classes 24 will also certainly be in the upper 'right-hand quadrant.
25 The Hanford wastes are shown by the circle more or less in f
i l
! /\CE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
. - . _ _ , _ _ , . _ _ _ _ _ _ _ _ _ _ _ _ - , , _ . . _ . . _ . . ~ - - , _ _ _ _ , _ _ - -. _ . . ,.
4
~
29879.0 39 BRT p)
(_ I the middle of the figure. Many of those will be 2 nonhigh-level wastes, based on the best information I have 3 available. Some will be long-lived wastes in need of 4 pe rmanent isolation because of the transuranic content.
5 Some would be classified as ordinary low-level wastes:
6 Both the long-lived and short-lived nuclides appear to be 7 below the class C limits of part 61.
I 8 Also, DOE's existing transuranic wastes will 9 almost all be in the lower right-hand quadrant, but I have 10 deliberately drawn the line slightly above the horizontal 11 divider to indicate that a very small fraction of those 12 wastes do have high concentrations of fission products and 13 under this class any indication scheme would be classified l 14 as high-level waste.
15 DR. REMICK: How about -- I can't think of the 16 name of the -- Three Mile Island -- the ones that used a 17 primary system -- would they still fall on high-level waste i
18 according to this activity? Or low level?
19 DR. FEHRINGER: I believe it was primarily 20 short-lived nuclides. If they had long concentrations 21 above the class C limits, then they would be high-level, 22 but I don't think those demineralizer beds had that 23 characteristic.
24 MR. PERRY: Several shipments have been made of 25 those materials to Hanford already.
, [~)
%J i
l ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 800-336-6M6
29879.0 40 BRT
/^\
(_) 1 DR. REMICK: Yes. I know. Currently it's 2 low-level waste. I was wondering under this definition 3 would they be low-level or high-level? I was always a 4 little squeamish whether those should be called low level.
5 DR. FEHRINGER: That's the example of a type of 6 material that could be classified fairly unambiguously with 7 this type of a definition.
8 MR. STEINDLER: I'm sorry, classified as what?
9 DR. FEHRINGER: As either high-level or low 10 level.
11 MR. STEINDLER: That's unambiguous?
12 DR. PEHRINGER: It would be with this type of 13 classification system.
14 MR. STEINDLER: Which would they fall in, do you 15 think?
16 DR. FEHRINGER: I can't say offhand, because I 17 don't know what the concentrations of long-lived nuclides 18 were in those demineralizer beds --
19 MR. STEINDLER: Not very high. But short-lived 20 -- according to your definition, short-lived nuclides 21 extremely high.
22 DR. FEHRINGER: That's my understanding also of 23 the characteristics. They would then be in the upper 24 left-hand quadrant. They would be a special time of 25 low-level waste, high activity but not long-lived.
.O ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6646
29879.0. 41 BRT k 1 MR. PERRY: Dan, do you think it might be 2 possible under the kind of four quadrant system that you 3 are laying out here to subdivide, say, the Hanford waste in 4 such a way that they hit the three nonrepository 5 requirements sections? That, in effect, is what's 6 happening with the stroatium and cesium capsules.
7 DR. FEHRINGER: Yes. That has occurred to some 8 extent already. By sampling each of the individual tanks, 9 I think they can determine wnich ones would be in the 10 high-level category and classified that way. It is my 11 understanding based on the information I have available 12 that only half a dozen or so of the tanks are likely to 13 have sufficient concentrations to be classified as 14 high-level.
15 I don't think it is very practical to further 16 process those wastes but I don't know that much about 17 chemical engineering, whether it might be possible or not.
18 Just by the process of sampling tanks, though, they can 19 divide up the wastes and classify them appropriately.
20 That's one of the things we told them in their -- in our 21' comments on their EIS for managing those wastes, we thought 22 they ought to sample the tanks, get a better idea of what 23 they have to work with, and process individual wastes 24 appropriately according to the concentrations that they 25 find.
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M6
29879.0 42 BRT 7m
(_) 1 MR. PARKER: I have another question on your 2 example. You talk someplace in your report about 3 significant quantities. That doesn't show here; it's 4 strictly concentrations. I think it's fairly important 5 because if you are concerned about nonproliferation of 6 nuclear materials, then that certainly becomes a very 7 important aspect. I was wondering how you address that in 8 this figure?
9 DR. FEHRINGER: The total inventory of 10 radioactive material does not enter into this figure and it 11 is one of the things that we have asked for comment on. We 12 think there is a need to consider the total inventory in 13 two ways.
14 First of all, very small quantities of waste 15 should not be considered high-level regardless of the 16 concentration. If someone has just a trivial amount of 17 activity that is being used in a laboratory, for example, 18 it is not properly considered as high-level, regardless of 19 concentrations, we don't think.
20 The second thing that may be more important is 21 consideration of the total inventory of radioactive 22 material at a disposal site, independent of concentrations.
23 Concentration is a fairly good surrogate for the hazard or 24 the risk of a waste material, but to some extent the total 25 amount of activity also is important, particularly when O
v ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-33MM6
29879.0 43 BRT i
(m_) 1 looking at population risk, and taat is something we want 2 to try to work into the definition, if we can.
3 It may be too awkward to work with, but we'd 4 like to at least consider it.
5 MR. PARKER: Shouldn't you have at least as a 6 minimum definition what IAEA considers safeguards levels.
7 Anything above that certainly ought to be in the high-level 8 and therefore has to go to a repository on very secure 9 storage -- disposal?
10 DR. FEHRINGER: I hadn't thought of getting into 11 safeguards considerations.
12 MR. PARKER: With spent fuel you have to worry gg 13 about it.
\, '/
14 DR. FEHRINGER: Spent fuel presumably will have 15 the short-lived concentrations necessary to be classified 16 as high-level. The transuranic wastes, if we over had them 17 from the commercial sector, would presumably go to a 18 facility that is much like a repository, as DOE is now 19 planning to use WIPP for its own transuranic wastes.
20 I think I would prefer to keep the safeguards 21 considerations separate from the classification as 22 high-level or nonhigh-level. If safeguards considerations 23 are important they can be dealt with separately.
24 MR. PARKER: The reason I bring it up is because 25 of the discussion of whether or not to go to repositories O
ACE FEDERAL REPORTERS, INC.
202-347 37(O Nationwide Coserage 8(Xb336 M46
29879.0 44
- BRT -
() I within a certain time frame hinges heavily upon the 2 safeguards problem. There's a lot of material in the 3 literature on just that topic.
4 MR. MARK: Frank, what do you mean by "the 5 safeguards problem"? What is that imaginary thing you are 6 talking about?
7 MR.-PARKER: There's a lot of sentiment now not 8 to go to depository, material, indefinite storage. The 9 English, for example, are proposing. material above ground 10 or --
11 DR. REMICK: Is this spent fuel, Frank?
12 MR. PARKER: Spent fuel and for them 13 particularly.
14 MR. MARK: What are we worrying about? What is 15 somebody worrying about?
16 MR. PARKER: The amount of fissionable material 17 that is readily.available.
18 MR. MARK: Oh, you mean somebody is going to dig 19 it up and make bombs out of that dirt?
20 MR. PARKER: They don't have to dig it up.
21 That's the point. It's above ground, easily accessible 22 because it is in interim storage.
23 MR. MARK: There is some plutonium, like 1 24 percent or so --
25 MR. PARKER: Right.
O ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336 (M6
29879.0 45 BRT
) 1 MR. MARK: And they are going to make bombs out 2 of it?
3 MR. TARKER: They can easily do chemical 4 separation of it.
5 MR. MARK: It isn't so easy and you darn well 6 know it. It's possible.
7 MR. PARKER: Well, I think Barney will bear me 8 out there's a relatively famous paper that shows how easy 9 it is to do it.
10 MR. MARK: I'm aware it is possible. I don't 11 buy the word it's easy. It takes an awful lot of handling, 12 an awful lot of doing. And for very little.
. 13 MR. PARKER: That's not what they said. They 14 said that any country that had a brewery, a winery or a 15 cheese plant -- dairy, within --
16 MR. MARK: They can run a reactor of their own 17 and get it out directly.
18 MR. STEINDLER: Reactors are dangerous.
19 (Laughter.)
- 20 MR. MARK
- I think this is overdone by a 21 fantastic amount.
22 DR. SHEWMON: They do have to be able to move 23 hundred-ton casks or something and be left alone for the --
24 MR. MARK: Months.
25 DR. SHEWMON: Months required to sort of break ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 80)-33MM6
s j
29879.0 46 /
BRT ,, ",",
,l' O 1 tui ev. 9et it ee retea ero the wette. rue it thro #94 2 .their milk factory or whatever it is, cheese factory.
3 MR. MARK: In.the same time they couldb uild a 4
4 reactor and take it out at home comfortably.
5 DR. SHEWMON: Not quite, but if there's any, Kind 6 of authorities there, presumably they would have reacte'd to 7 it by then. After all, they run through the BBC studios at 8 the drop of a hat. They ought to be able to go find a ,
9 cheese factory. ,
10 MR. MARK: I do believe we ought not6t'o' i n f l a m e ,
11 this concern. It's a very far-out concern. That's all I 12 meant to say. p 13 In the things I read, I see no referende to O 14 cobalt 60. That has a what, five-year half-life?
15 DR. FEHRINGER: Right. ,
16 MR. MARK: Is that why it's not mentioned?
17 DR. FEHRINGER: That's correct. In the.. existing 18 part 61 regulations, there is no upper limit on 'the .,
~,
19 concentrations of cobalt 60 that would be considered class ;
20 C: Theoretical specific activity of cobalt 60 would still
( 21 be below the class C category for that nuclide, so it's not 22 a candidate for classifying waste as high-level or not i
23 high-level. i 24 MR. MARK: Okay. I vaguely suppoced that might i 25 be the case, but I know it is one of the main ingredients I
O l
! ACE FEDERAL REPORTERS INC.
j 202-347 3700 Nationwide Coserage 800 336 6646
- p
. ~ . _ . . , - , _ , _ - , .
ls 29879.0 47 l BRT
. ?,.
/" .
Vg ,Tj, ,, 1- of wastes that might go into the stream.
I 2 The other question: Surely it would not be
. 3 wrong, or there might even be a point to looking at this 4 waste from the point of view of comparative beta and gamma s 5 activity. The beta activity is only through ingestion or 6 direct exposure. The gamma is -- just if it gets into the 7 , atmosphere or into the environment it spreads. And in a
.8 number of cases I wondered about that split and whether it "L 9 wasn't a significant split that should be called to 10 attention or should be taken into account? After 300 years,
-c 11 the waste is mostly beta. After 10 years a good part is 12 gamma. And so one measure of concern might be modulated by
? ~. 13 things like that. It's a question only.
L
/ 14 DR. FEHRINGER: To some extent I guess I think I
- 15 understand what you are s'aying. I think you mean, say
- 16 alpha as being the long-lived nuclides, though, rather than 17 beta.
18 MR. MARK: Or the alpha, it's only by ingestion 19 that you have to worry at all. And by beta it's almost the i 20 same.
21 DR. FEHRINGER: Yes.
. i 22 MR. MARK: By gamma it's not the same. And in 23 the early years, gamma is a prominent candidate from the
cesium, I believe. In the later years it's gone. And I 24 25 think the longer-lived things, like iodine 129 and 1 J 4
/\CE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 800-336-6M6
- _ , __ _ ._.._.__ - - - _ . _ _ _ - . - . . _ . _ . _ . _ . . _ _ - . . _ . - . . . _ . . . , _ _ . _ ~ _ . . - -
29879.'O 48 BRT
!(). 1 plutonium and so forth, are either alpha or beta.
2 DR. FEHRINGER: Yes. Those affect the hazard 3 that a waste represents. I think we can adequately handle 4 that through the analyses we'll do of the hazard of a waste 5 after it's disposed of. We look at the pathways by which 6 the materials might return to man's environment and come 7 into contact with man. Those types of considerations come 8 factored in.at that point.
9 Iodine 129, of course, has its own unique 10 characteristics, biological as well as radiological. And 11 it is a concern over the long term. The transuranics 12 also --
13 MR. MARK: There is a transition, somewhere, g'u) '
14 whereby the presence of the stuff in the environment is 15 threatening or the presence is only threatening if you eat i
16 it. Or breathe it. And that point is not at all clear 17 from things I read.
18 DR. REMICK: Dan, I was under the assumption 19 that spent fuel was automatically considered high-level 20 waste, but you said something, maybe I misunderstood, that 21 made me question it. You said something about, you assume 22 spent fuel would have enough activity to be classified as 23 high-level waste. Am I correct that high-level wastes in 24 the advance notice would be automatically high-level waste?
25 Or am I incorrect?
.O V
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-M>46
29879.0 49 BRT oO
() 1 DR. FEHRINGER: The Waste Policy Act 2 distinguishes spent fuel from high-level waste, so for the 3 purpose of the Waste Policy Act, spent fuel is addressed 4 separately.
5 In part 60 we have included spent fuel as a type 6 of high-level waste.
7 DR. REMICK: Right.
8 DR. FEHRINGER: Any spent fuel from a commercial 9 power plant obviously will have adequate concentrations.
10 If one had a small amount of -- or an amount of low burn-up 11 spent fuel from a research reactor or something like that, 12 I guess there's a possibility it might not be high-level 13 waste. That is the only reason for the qualified statement
%.J 14 earlier.
15 DR. REMICK: So it would not automatically be
, 16 high-level waste under that condition? In other words, you i
i 17 are still saying most of it would be high-level waste.
18 Some of it might not be? Is that what you are saying?
19 I thought when I read the advance notice I 20 thought automatically any spent fuel would be high-level i
21 waste, but you seem to be wavering.
22 DR. FEHRINGER: As the advance notice is worded, 23 that is correct. Spent fuel is still a category of
[ 24 high-level waste regardless of concentrations and I should i 25 not have even tried to open up the research file type of --
l O
l
/\CEJFEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6M6 i
29879.0 50 BRT l) 1 DR. REMICK: I'm just trying to understand.
2 Going back to Frank's-comment, and I agree very 3 much of where Carson is coming from, but the question of 4 quantity, would you include quantity in a definition of 5 high-level waste? Or would you include quantity on a 6 restriction on the repository or the low-level waste burial 7 ground as a means of concern over the quantity being a 8 strategic amount or something like that? I'm not quite i
9 sure how you'd include that in a definition. It seems-to l
10 me that's kind of a restriction on the burial location.
11 Maybe I'm wrong, maybe I'm not seeing the whole problem.
12 DR. FEHRINGER: That's been my difficulty fm 13 looking at inventory also, it doesn't fit well into a
(-) 14 definition.
15 I hadn't thought of inventories in safeguards 16 terms before. So that's new to me. But looking at total 17 inventory from the public health risk point of view has 18 some merit. I just don't know exactly how to factor that 19 into a definition. Concentration seems to be the better i 20 representation of risk or surrogate for risk. But I think 21 we do need some kind of a lower cutoff where a trivial 22 total quantity of material does not need to be classified 23 as high-level. Perhaps we could -- I think we can figure a 24 way to do that.
I 25 MR. STEINDLER: One other nitpicking point, l
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
'29879.0 51 BRT
.( p) 1 perhaps, you keep talking about spent fuel. But that 2 normally,.I think, in the parlance of the folks in the 3 waste business, has included fuel that was prematurely 4 discharged for one reason or another, which in fact is not 5 spent. .
6 Are you going to get folks to tell you that you 7 have thrown subassembly out of a reactor because the silly 8 thing failed and you really don't want to fuss with it and 9 somebody says: Well, that's not really repository material 10 because it's not spent. It's only been in the reactor 20 11 days before it cracked open. What is your response? Or is 12 this a trivial issue?
g 13 DR. FEHRINGER: In a way it may not be uJ 14 particularly important. This definition --
15 MR. STEINDLER: The fact is that is precisely 16 the kind of junk that got buried in West Valley.
17 DR. FEHRINGER: This definition is being 18 proposed for part 60 which is a regulation for a repository.
I 19 The definition does not directly reach out to other aspects 20 of the agency's licensing, so there would be no particular 21 requirement, based on this definition, that a waste be 22 disposed of in any particular way.
t 23 MR. STEINDLER: People are surely going to look i
24 to this definition that,you folks have put together as the t
25 definition for the whole regulatory --
l (
l
/\CEJFEDERAL REPORTERS, INC.
I 202 347 3700 Nationwide Coserage m)-336-6646
. - - ~ . _ - - - , , . , _ . - _ . _ _ _ _ . _ _ _ _ , _ , . _ _ . . . _ , - _ . . . _ . _ _ _ . , , _ . _ . - . .
c
.29879.0 52 BRT
()' 1 .DR. FEHRINGER: I realize that. Legally there 2 would not'be any implication beyond part 60 but in a 3 practical sense there will be.
4 I believe we use the words " irradiated reactor 5 fuel" rather than " spent fuel" in our definition.
6 MR. STEINDLER: If you did then you are home 7 free. But I didn't see it.
8 DR. FEHRINGER: I think so. We used 9 " irradiated reactor fuel."
10 MR. KASTENBERG: Could I ask you a question?
11 You keep using the term " risk." What is your measure of 12 risk in this case?
13 DR. FEHRINGER: I'm glad you asked. Let me turn O 14 to page 8. We discussed a little bit earlier the 15 ' difficulties with defining the term " highly radioactive."
16 That is a term that we will have to wrestle with. We may 17 end up being a little subjective with it. But when we want 18 to determine what is or is not in need of permanent 19 isolation, we propose to use the approach that is listed on 20 page 8.
21 First of all we want to establish the acceptance l
22 criteria for an appropriate match between a waste and a 23 disposal facility. This is where the term " risk" enters in.
24 Included in this definition of risk will be 25 individual dose rates, those taken from part 61. We may 1
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-3364M6
d 29879.0 53 BRT- .
- n
() 1 need to also include some measure of population impacts.
2 That we'll just have to look at and see whether that is 3 going to be'more important than the individual risk.
4 In part 61, in developing class A, B and C 5 limits, individual risks were the measure that was used.
6 The criteria were dose rates no greater than 25 millirem 7 .per year to a person at the site boundary and no more than 8 500 millirem to an accidental human intruder. That was 9 primarily an external dose rate -- external dose. It was 10 just 500 millirem, not an annual dose rate criteria.
11 MR. KASTENBERG: What's the difficulty with 12 population dose or population impact?
w ,
13 DR. FEHRINGER: I don't know that there's a
(%.)
14 difficulty with it. I'm just not sure whether it's 15 necessary to add an extra population risk measure or 16 whether the individual risk criteria for part 61 will be 17 adequate.
i 18 MR. KASTENBERG: Have you thought about any 19 other societal impacts? Economic impact? Or the fact that 20 you may impact an aquifer or something, t at has economic 21 value or other value?
22 DR. FEHRINGER: I'm not proposing to include 4
j 23 those at this time. Economics were a greater consideration 24 in the part 61 classifications where we find classes A, B 25 and C.
O ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage Rx)-336-6M6
29879.0 54 BRT
/"'%
() 1 In this case I'm not sure that economics is a 2 proper basis for classifying wastes as high-level or 3 nonhigh-level. I would prefer to stick to the public 4 health issues. In the same way I would prefer not to look 5 at the environmental impacts as being controlling. I think 6 those can be dealt with separately, things like potential 7 contamination of aquifers.
8 So'I would prefer to limit the criteria strictly 9 to ones of public health, if possible, to individual risks, 10 and, if necessary, add in population -- radiological risk.
11 once we have established the acceptance criteria, 12 then we would describe the features of what I have been 13 calling the intermediate disposal f acilities. These are 14 the facilities that are almost but not quite as secure as a 15 deep geologic depository.
16 The third step would be to characterize the 17 wastes in terms of the physical and chemical form of the 18 waste, concentrations of radionuclides and the volumes, any 19 other characteristics that are relevant; fourth, develop or 20 adapt an existing assessment methodology which would be i
21 used to predict how well these intermediate disposal 22 facilities would perform with the particular types of 23 wastes placed in them; and, fifth, we would evaluate the 24 performance of these facilities. The purpose of the 25 evaluation would be to determine the maximum concentrations ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 2 0-336-6646
4
- 29879.0 55 BRT
) 1 of wastes that can be disposed of in these intermediate 2 facilities; wastes that cannot be safely disposed of in 3 these facilities would then be judged to require permanent 4 isolation and we would say that then they need to go to a 5 repository or equivalent. If they also were-highly 6 radioactive,'then they would be classified as high-level 7 waste.
8 The point here is that we do not'think it 9 sufficient-to examine the waste alone. We think the match 10 of wastes with disposal facilities is what really 11 determines the risk to public health, and we must look at 12 .the combination in order to classify wastes.
13 MR. MOELLER: Frank?
14 MR. PARKER: You might be interested this is 15 exactly what they did in Project Gasneer, procedure they 16 followed. Interestingly enough they also came up with the 17 conclusion that for the present they are not going to have 18 low-level waste. They are only going to have intermediate 19 -- repositories, the repository will only be nor low -- for 20 intermediate waste and high-level waste. They reserve the 21 right to have a low-level repository in the future, but at 22 the present they aren't going to go to that.
23 DR. PEHRINGER: Does that mean they are going to 24 dispose of low-level waste by shallow land burial then, or 25 store it?
O ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-u46
29879,0 56 BRT
() 1 MR. P'RKER: They are going to put it at 2 intermediate de- or greater --
3 '
- s. CARuoN: That's the Swiss way?
4 MR. PARKER: The Swiss.
5 DR. REMICK: Dan, you said you were considering 6 a dose of 25 millirem but I was not clear if you said per 7 year or just 25 millirem.
8 DR. FEHRINGER: 25 millirem per year for the 9 person living adjacent to the site. And a single dose of 10 500 millirem to a potential human intruder.
11 DR. REMICK: How did you arrive at the 25 12 millirem?
13 DR. FEHRINGER: Those were the numbers that were 14 used in the part 61 classifications. The.25 millirem per 15 year corresponds to the environmental standards for the 16 uranium fuel cycle and the 500 millirem corresponds to the 17 annual dose limit in part 20, for nonoccupationally exposed 18 people.
19 DR. REMICK: It's fairly consistent with the 20 safety goal for reactors. I didn't know if there was any 21 consideration of that or not.
22 DR. FEHRINGER: I think that work was done 23 before safety goals were promulgated, so I don't think 24 there was a correlation.
25 MR. MOELLER: Frank?
O
~
ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmcrage 800-33MM6
29879.0 57 BRT 1 MR. PARKER: 25 for the whole cycle?
2 DR. FEHRINGER: Yes.
3 MR. PARKER: But for this one facility you are 4 .also going to use the 25, is that right? So you are 5 assuming no interaction?
6 DR. FEHRINGER: Yes. Right.
7 MR. PARKER: Is that reasonable, considering i
8 that it's liable to be collected?
9 DR. FEHRINGER: Well I'm not sure that an 10 intermediate type of disposal facility would be. This is 11 the type of facility that's almost as good as a repository.
j 12 I'm not sure it would be collected with any other nuclear J
13 facility. Shallow land burial of course is different.
O 14 MR. . PARKER: The Swiss are doing that. The i
15 intermediate storage and the low-level waste, they are both i
l 16 at reactor sites.
17 MR. MOELLER: We are nearing the end of the
! 18 allotted time, but let's go ahead and finish up on the l 19 coverage of the handout.
20 DR. FEHRINGER: Okay. On page 9 I just want to 21 reiterate an earlier point that I made. From a technical 22 point of view we would prefer to have a numerical 23 definition of high-level waste apply to all wastes 24 regardless of source. We do have the Energy Reorganization 25 Act still in effect and not altered by the Waste Policy Act.
i :
! Acu. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33MM6 i
-,,---,,---...,.-,__,._..._.__,_,,,..__,.c_, ._..,_.,_.._,.,.._,__c._._____,___---- _ , . _ _ . _ _ , . , _ _ , . . _ _ _ , . _ , _ _ . _ _ , , , , - . , , _ _ . . .
29879.0 58 BRT A
(_) 1 That is what gives us our licensing authority for 2 high-level wastes and it is the view of our legal staff 3 that that is a source-based piece of legislation wh_ich we 4 cannot alter by any definition that we might develop. That 5 has led to including in the advance notice an offshoot 6 which would continue to look at reprocessing wastes based 7 on the source.
8 Technically we don't think it's the best way to 9 go, but legally and administratively it does have some 10 merit and we've asked for public comment on which option we 11 should take.
12 MR. MOELLER: On page 9 I didn't understand the 13 part, the last two lines. In other words, you exempt waste 14 from research and development activities. What was the 15 point there? What does that gain for you?
16 DR. FEHRINGER: I believe that wording allowed 17 DOE to use high-level wastes for testing such as they are 18 planning to carry out at the waste isolation pilot plant, 19 without having NRC licensing of that testing.
20 MR. MOELLER: Okay. I didn't understand.
21 DR. FEHRINGER: I don't know of any other 22 implication for that.
23 Page 10, I summarize some other issues that are 24 discussed in the advance notice. We do not plan to have 25 any impact of significance on DOE defense-only facilities.
O ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-33HM6
29879.0 59 BRT
) 1 There is a possibility of a little bit of a spillover 2 effect from our definition on the transuranic wastes that 3 DOE has. In a strict legal sense, our definition would not 4 affect those wastes, because they are disposed of in a i
5 defense-only facility, but public opinion being what it is, 6 we know that there will be a comparison between what DOE 7 does and what our definition says.
8 There certainly is no intent, though, to make a 9 licensing gap for the NRC licensing of the waste isolation 10 pilot plant or of any other defense facility.
i
! 11 Many of the tank wastes at Hanford would likely
}
12 be low-level wastes under this plan and again there will be 13 a comparison in the public's mind between those wastes and i 14 ours under this classification, and we are aware.that this j 15 will generate a lot of controversy. There may be some 16 minor effects on the nuclear waste fund. This is primarily 17 an administrative problem that the Department of Energy 1
18 might have. The Waste Policy Act required people to have 19 entered into contracts for payments into the fund by a j 20 particular day. If we now classify other things as i
! 21 high-level, that is impossible for them to have done. But 22 it's a point that the department and Congress can work out i 23 without too much difficulty.
I 24 We do not think there will be any significant 25 effects on the Commission's waste confidence rulemaking.
I i
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Cmerage 800 336-6M6
.~. . . . - - . . . _ . - . . _ . , . . _ _ _ . . , . _ . . _ . _ _ _ _ _ _ _ . . _ _ . , , _ , _ _ . _ . , . _ _ _ _ . . . , . . , _ _ ,, _ . _ _ . . _ _ , _ _ _ , _ _ , _ . . _ _ _ _ _ _ _
29879.0 60 BRT (m_) 1 If there are any, we think it will be a beneficial effect.
2 This will help to make sure that there is a home for all 3 wastes. And there will be no change in the status of the 4 naturally occurring and accelerator-produced radioactive 5 materials. The NRC doesn't have licensing for those 6 materials and our lawyers tell us we have no basis fo,r 7 classifying those materials as being high-level or 8 nonhigh-level.
9 We will consider those as we carry out our 10 analyses and likely we will say, if we had authority, this 11 is the way they would be classified, but legally we cannot 12 give them names.
-)
'~
13 The last page of the handout is a page from the 14 advance notice and it gives the conceptual definition of 15 high-level waste as it is included in the advance notice.
16 The first part of the definition is the same as the 17 existing definition, in part 60. Then we say: solids 18 ~ would be classified as high-level only if they contained 19 radioactive materials which exceed the concentrations in 20 table 1 and table 2. It is important that a waste must 21 exceed the concentrations of both tables to be classified 22 as high-level.
23 The two tables are the tables from the class C 24 limits of part 61. They are there for the sake of 25 illustration and we will either provide a technical support O
ACE FEDERAL REPORTERS, INC.
202-347-3X0 Nationwide Coserage MO-33M646
29879.0 61 BRT r~s
() I for those tables or develop alternative tables when we 2 develop a proposed rule for this project. That's the end 3 of my presentation.
4 MR. MOELLER: Well, I don't understand, then, 5 this last item. You have strontium and cesium in there, 6 which you told us you classify as short-lived materials.
7 DR. FEHRINGER: Rig h t'.
8 MR. MOELLER: And yet if they exceed this 9 concentration you are going to call it high-level waste.
10 DR. FEHRINGER: We need to go back to page 7 to 1 11 illustrate this. I think a picture is definitely better 12 than words in this case.
f- 13 MR. MOELLER: 7 or 7-A? Well, 7 --
b) 14 DR. FEHRINGER: Either one.
15 MR. MOELLER: Okay.
16 DR. PEHRINGER: Table 2 is the table of 17 short-lived nuclides for part 61. If a waste has
. 18 concentrations exceeding the values in that table then it 19 would be considered highly radioactive. On the horizontal i
20 axis we have the long-lived nuclides listed and those are 21 displayed in table 1 and, to the right of the table 1 22 values is where waste becomes in need of permanent 23 isolation. It is only in the upper right-hand quadrant 24 where a waste is high-level and that's only when a waste 25 exceeds both table 2 and table 1.
O v
ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage MXL3M (M6
29879.0 62 BRT
() 1 MR. MOELLER: Okay. So you are saying -- all 2 right -- that it has to exceed something in both tables.
3 DR. FEHRINGER: Right.
4 MR. MOELLER: All right.
5 In terms of the Subcommittee, you'll notice on 6 page -- I believe it's 24 of the advance rulemaking they 7 are giving the questions that the public is being asked to 8 answer. We might take a moment and just go down through 9 those questions.
10 MR. MARK: In general connection with, I think, 11 stuff on page 7, tables 1 and 2, I felt -- had a feeling 12 that there was something lacking in the way of stipulation 13 of the time at which these judgments were to be made.
14 I mean if I've got 40-year-old stuff, the cesium 15 is different than if it's four-year-old, and other things 16 in that same period. I have' the general question, then, 17 whether it isn't useful, perhaps even necessary, to say 18 something about the time at which these readings are to be used?
19 20 DR. FEHRINGER: Perhaps we should have said that 21 more explicitly in the advance notice. Because this 22 definition goes into part 60, I had thought it was implicit 23 that the definition applies at the time when wastes are 24 received at a part 60 facility for disposal.
25 Since there is a tendency for people to look O
Ace FnnenAi. ReconTens. INC.
202 347 37(N) Nationwide Cmcrage MO 3%6M6
29879.0 63 BRT
() 1 into other areas of the Commission's licensing, I think we 2 should have made it more specific. It was our intent.
3 MR. MARK: That would answer my concern.
4 DR. FEHRINGER: Okay.
5 MR. MOELLER: Then you have no objection to a 6 waste generator holding the waste for 10 years, say, to get 7 down below or into another category?
8 DR. FEHRINGER: That's correct. We want to look 9 at the nature of the waste when it is received for disposal.
10 MR. MARK: I believe if that were made clearer --
11 it would help.
12 DR. FEHRINGER: Right.
13 MR. MOELLER: Who looks at the risk of the waste 14 generator holding the waste for this extra period of time 15 as compared to the risk of getting it in the disposal 16 facility?
17 DR. FEHRINGER: The Commission can look at that 18 if the waste generator is a licensee, the same way we look 19 at the rest of its operations.
20 If it's the Department of Energy then it's their 21 own internal decision.
22 MR. MOELLER: All right. Looking at page 24, at 23 these questions, I believe we should take some time and go 24 over them. The first one is that two options are presented 25 for defining the reprocessing wastes and they want to know (v~)
Acti-FliDl! rat. RitmR11tRs, INC.
202 347 3700 Nationwide Cmcrage m 0-3M 6646
- - . - - . - . - - . - ~ . _ . - - _ - - - - - - - - . . _ _ _ _ _ _ - - - . . ~ _ , _ . - . - , _ . , , , . - - - . - . . , - - . . . , _
t 29879.0 64 BRT fs
( ,) I which of the two approaches is preferable.
2 The first proposes to define the sufficiency of 3 the product concentrations in a manner analogous to its 4 treatment of highly radioactive and requiring permanent 5 isolation. The second interprets it as encompassing all 6 those wastes which have heretofore been considered 7 high-level wastes under appendix F to Part 50.
8 Do the Subcommittee members have any preferences?
9 DR. REMICK: I personally have preferences for 10 the first option.
11 MR. MOELLER: I voted for option 1 myself.
12 MR. PARKER: I --
g- 13 MR. MOELLER: Just don't hold us to this, but
'\_)g 14 let's go down the questions.
15 DR. REMICK: I think it's unfortunate there you 16 said "by examining the hazards posed by wastes" rather than 17 by risk. Because as you said later on it's not only the 18 waste but where you are going to put it that determines the 19 risk. That's more consistent.
20 MR. MOELLER: The second one, I'm not going to 21 bother reading them, but it is the current class C 1
22 concentration limits. Would an alternative set of 23 concentration limits be preferable? And how should they be 24 derived? Do you have any off-the-cuff comments on this?
1 25 MR. PARKER: If you are going to use risk-based t
Ace Frionnai. IliiroRTiins, INC.
202 347-37m Nationwide Onerage m 34 ud6
._- _. -- .. .-~.-. . _. - . ~
< 29879.0 65 BRT
() 1 methodology, then how can you set it without having --
2 MR. MOELLER: Made the analysis. Okay.
3 DR. FE11 RINGER: Let me point out that this 4
4 question is directed at the highly radioactive axis where 5 we think there's less of a risk basis for defining that 1
6 term. We are asking if there is, asking for guidance, 7 really.
I 8 MR. PARKER: I guess my reaction is if you are 9 going to use a risk basis you probably ought to do~the 10 whole works under the risk-based methodology.
11 MR. MOELLER: Well, on second thought, rather 12 than trying to go down through all of these now, which may i
13 not be too productive, I think we might do it in our i 14 executive session and sort of reach some consensus 15 positions then.
16 What, Dan, do you want from us? What can we do 17 that will be most helpful? Do you want some comments on 18 these nine questions or what?
19 DR. FEllRINO",R : That would be very helpful. The i
20 next step in this project is to develop a proposed rule and 21 any comments that you have along with comments from the f.
l 22 public will provide us guidance in that.
23 In particular, with the advance notice, we are i
24 trying to determine whether there is or is not agreement i 25 that this whole conceptual approach of having two 4
O ace FEDERAL REPORTERS, INC.
l 202 347 3700 Nationwide Coserage 14X)-3364M6
29879.0 66 BRT.
() 1 characteristics to define high-level waste is'a correct one.
2 -If you could comment on that, I think that would be very 3 useful.
4 MR. MOELLER: Say that again, the last statement?
5 DR. FEHRINGER: Our conceptual approach is to 6 require two characteristics before waste is high-level.
7 MR. MOELLER: Yes. Okay.
8 DR. FEHRINGER: That basic concept is one that 9 we are looking for guidance on.
10 DR. REMICK: Dade, on that, I realize we are all 11 knowing in field, considerable wisdom, and certainly we 12 thought of everything that the public might think of -- but 13 I don't know. Since this is advance notice it seems to me 14 it would be beneficial to know what the public has to 15 comment. That might flavor our comments on this. There 16 are things we just might not think about.
17 MR. MOELLER: Generally the Committee does wait 18 for the public comments and even for the Staff's analysis 19 of the public comments before we offer ours. I suppose 20 that's the cautious approach.
21 DR. FEHRINGER: The timing may be just about 22 right to have some comments received by the time of your 23 next meeting. If so I'd be glad to tell you what the 24 nature of these comments is.
25 MR. MOELLER: All right. We'll talk about it, O
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cmerage MX)-336-6646
'29879.0 67 BRT !
() I then, in our executive session and decide-what we want to 2 do. I think, though, you've heard in general, of course, 3 comments as you have gone through the presentation, so--
4 unless there are other comments or questions on this topic, 5 I think we'll thank Dan and take a break.
6 DR. REMICK: One comment I would make, Dan.
7 This is not to be critical, but when I started reading the 8 advance notice of proposed rulemaking I thought I 9 understood it. The more I read I became confused. That's 10 not necessarily a criticism because I read it only once and 11 sometimes you have to reread these things, but I would have 12 found some of the things -- your presentations today were 13 extremely helpful for me, tying these things together. If 14 the Commission sent the advance notice back for any comment 15 or changes before it goes out, you might consider some of 16 these simplified things that you provided for us. It would 17 be very helpful to the public, I think, in understanding 18 what this is all about.
19 DR. FEHRINGER: It's a valid criticism that we 20 did not do the best job of writing on this. I'm not sure 21 why it came out the way it did, but other people have had 22 difficulty understanding it. I'm going to be making as 23 many presentations as I can reasonably carry out like this 24 one this morning. In a few minutes I go across the street 1
25 to give the same presentation to the Atomic Industrial C:)
l
/\CE. FEDERAL REPORTERS, INC.
! 202-347-37(x) Nationwide Coserage 8(X)-33MA46
-_. ._.-._.._.-.__-.-..._-._._-__..m._, . . _ . , _ - , . . _ . _ _ . ,_.. _ ...-._, ..
29879.0 68 BRT
(~h Forum and I hope that that will, at least, provide the
(_j 1 2 mechanism to explain what we should have said more clearly 3 in the advance notice.
4 MR. MOELLER: Then I think every member of the 5 Subcommittee and consultants should, again, hindsight is 6 always good -- but should have had SECY 86-328, which talks 7 about each of the -- the reasoning behind each of the 8 segments -- sections, and so forth.
9 MR. MARK: Without let's say assuming the 10 availability of SECY 86-3 -- 8 -- whatever it was.
11 MR. MOELLER: 328.
12 MR. MARK: The thing should be presented, gg 13 written, in a way that didn't require that one have either (l that or a number of other things within reach. It should 14 15 be understandable and it could be improved, as Forrest 16 suggested, by carrying that thought in mind. I'm talking 17 about somebody that doesn't have SECY this or that and 18 table this or that. But it's still possible to make it 19 comprehensible and more effort might be put on that thought.
20 DR. REMICK: We'd like to see it as clear as 21 ACRS letters.
22 (Laughter.)
23 MR. MARK: Oh, much clearer than any ACRS letter.
24 Good heavens.
25 MR. PARKER: That figure I think is really very b
LJ ACE FEDERAL REPORTERS, INC.
202-347-37(O Nationwide Coserage MO-33MM6
(
29879.0 69 BRT l () I helpful. 7-A.
2 MR. MOELLER: And I think you have a tough row 3 to hoe. Because you are coming up with, you hope, a --
4 ultimately a risk-based classification system which may 5 show that the Hanford wastes are not high-level wastes and 6 so you are trying to anticipate, you know, the problems.
7 Let's take 15 minutes.
8 (Recess.) ,
- 9 MR. MOELLER
- The meeting will resume, and let
)i 10 me welcome Pauline Brooks and Seth Coplan, from the j 11 Division of Waste Management, and they will be discussing L 12 with us the subject of performance allocation. Welcome.
l 13 MS. BROOKS: I would like, first of all, to give 7
14 you an overview of what we would like to talk about today.
l 15 That's on page 26 of your handout.
16 First, why performance allocation is needed, I f
17 primarily as a tool for site characterization planning and-o 18 then what performance allocation is. Deal with some 19 definitions and some approaches that have been agreed upon 20 between NRC and DOE to doing performance allocation. And
)
21 walk through an example of how it might be done.
22 Finally, we'll discuss DOE's planned approach to 23 performance allocation in terms of, first of all, Appendix >
24 A of the draft plan amendment, which I believe was sent 25 down to you, and an upcoming meeting, March 3 and 4,
,i
(
l I
! ACE. FEDERAL. REronTens, INC.
} 202-347-3700 Nationwide cmerage 8oo-33 u u 6
,,,_-,,--,c.,,.-_-__-n--
29879.0 70 BRT
( ,j 1 generic meeting, DOE headquarters, will brief us on how 2 they are doing performance allocation and the status of 3 their work on it.
4 MR. MOELLER: Could you help me? I know you are 5 tying it now, into the DOE draf t mission plan. Will you 6 later talk quite a bit more about that?
7 MS. BROOKS: Yes, I intend to.
8 MR. MCELLER: Then I'll wait.
9 MS. BROOKS: On page 3 I need to tell that you 10 there is a major omission. The last bullet was 11 inadvertently left off in the final typing, and when I get 12 to that point I will tell you what that wording should be.
13 Performance allocation is intended to be a 14 technical management tool to help DOE design its site 15 testing program. The part 60 performance objectives were 16 laid out to give DOE some flexibility in choosing ways of 17 meeting the performance objectives at different sites, and 18 performance allocation maintains its flexibility taking 19 advantage of site features to tailor the design and the 20 program to a specific site.
21 It also provides structure for a site test 22 program, which will, in the end, make it easier for us to 23 review the test program. The structure will come about in 24 terms of how much credit DOE wants to take for components.
25 Now, an integrated systems view, we will see later, is O
Acti FliDl!RAl. IlliPORIl!RS, INC.
202 347 3700 Nanonwide Coserage M rk116 (446
29879.0 71 BRT
() 1 maintained in doing a performance allocation, so this means 2 that when DOE says that it is taking certain credit for --
3 credit for certain components, it will also be saying what 4 components it will be holding in reserve or redundant 5 status, and having identified all the components and their 6 functions it will be able to say, well, we are not taking 7 credit for this at all but we know what they are and what 8 their functions will be, and we are in a position to study 9 what possible offects it has.
10 That means, then, that wo are assured that thoro 11 will be enough studied to understand the system and to 12 avoid doloterious offects of those components, if they are 13 not, presumably, taking credit for.
14 And the missing bullet is that " helps determine 15 specifically what investigations are noodod." And 16 sub-bullets under that are " kind of tests; how many tests; 17 what paramotors are i;nportant; and what lovel of cortainty 18 is required." ,
19 Do you want me to repeat the whole thing?
20 MR. KASTENBERG: Just the last.
21 MS. 13 ROOKS: What lovel of cortainty is 22 required.
23 Moving to the chart on pago 4, tackle somo of 24 the definitions. First of all, performance allocation
- 25 itself. What do wo mean when wo use that term? We are O
Acti Flintinai. Riivoariins, INC.
202.u?mm Nanonwide Cmerage Moo. Ut> W(i
29879.0 72 BRT
() I thinking of a systematic iterative process of assigning 2 tentative performance goals and the associated desired 3 confidence level for the high-level waste disposal system, 4 subsystems and components.
5 The performance goals are defined -- or we have 6 agrood with DOE that this is how we will look at 7 performance goals: numerical values performance, 8 tentatively assigned to performanco measures of the systems, 9 subsystems and components.
10 MR. PARKER: Why do you say " tentatively"?
11 MS. BROOKS: We'll soo a little later why we say 12 tentatively a little bit later, but I'm referring to the 13 fact that you have to got started early in performing your 14 site characterization plans and at that point there's a 15 limited amount of information.
16 MR. KASTENBERG: Those numerical values are 17 reliability goals? Will you get into that?
18 MS. BROOKS: In terms of the performance goals 19 they would bo. They would actually be what are you going 20 to mean? What property are you going to measure with 21 respect to a given component and what value would you liko 22 that to bo?
23 Again, when wo put the whole picture in I think 24 you'll see.
25 MR. KASTENilERG I guess the reason I raised O
Acit.17ti ntinai. RiiPonitins, INC, 202M Um Nationwide Gnaage mim mui
. 1 29879.0 73 BRT
() 1 question is we were presented some material based on some 2 meetings you had in April of '85 and later on, and I 3 detected a negative feeling'towards this approach in the 4 material that we were given; and now I sense a very 5 positive approach towards it. I have not been at past 6 Subcommittee meetings but I just wondered what has been 7 happening.
8 DR. SHEWMON: You wouldn't have heard very much 9 about it.
10 MR. KASTENBERG: Okay.
11 MS. BROOKS: I believe there was some discussion 12 of performance allocation back in October of '85 with the 13 ACRS Subcommittee.
O 14 To give a little history of what I'm talking 15 about with agreements between DOE and NRC on how they are 16 going to do this, if you look at the meeting summary for 17 the April 17th meeting with DOE, you will find there was a 18 great deal of rescissions to the wholo concept of 19 performance allocation. In terms of DOE being af raid that 20 it would be stuck with meeting goals that just weren't --
21 were unmeotablo, and then eventually be held to it. And 22 DOC feeling that confidence levels, for examplo, should not
! 23 be addressed in the performance allocation.
1
! 24 Then, in September of the same year, '85, tho --
1 25 a number of agroomonts were struck. We managed to l
Ace.FEDERAt. REPORTERS, INC.
202 347 37m Nationwide Cmerage Mm.316-(M6
r
?-
29879.0 74 BRT
() 1 communicate at that meeting, and what I'm talking di>out 2 today is based on the agreements of the September meeting'.
3 MR. MARK: In what I read about that, I believe 1
4 I read the things that were made available, amongst'the 5 preferments allocation -- there could be som'e difference 6 about the metal of the sheathing of the wasto, it might be
~. .-, ~
i 7 1000 years -- you say okay, count 1000 years for that. Or-s ,< ,
8 it might of course be 300 years, but we are going to count. i 9 on the Bentonite to make up the next 300 years, or -
10 something like that. I saw no reference at all to ]
11 geochemistry of the surrounding area.
12 There was water transport rates, and so forth. ,
13 Is that one of the items that would come in here? This 14 stuff is going to adsorb such a fraction or not? ..It wasn't 15 mentioned in the things. At least I didn't see it clearly. !
16 MS. BROOKS: It probably would not have been 17 mentioned simply becauso, so far, most of the materials ,
^
18 discussing performance allocation have selected examplos 19 and have not presented an "across the board picture" of all 20 the factors that would be considered.
- 21 HR. MARK: But if someone were able to point out 22 my repository plan is right in the middle of a largo bed of 23 argillaceous clay or somothing, you would be prepared to 24 nay: Well, that's good for some measure of something. ,
25 MR. MARK: They would probably propose that some O
Acit.Rmi!RAl. Riironinns, INC. >
- o .347 37 saiion iae emera,e mit 3m <un j
]
- c-Ud/, m i 29879.0 75 l- BRT i . ..
() 4 1 credit be given to that aspect of the system and it would j 2 be up.to us -- and that you would be prepared to recognize i
3 if you believe that it was well taken?
4 MS. BROOKS Yes. We would have to review it 1
j 5 for that purpose.
i 6 MR. MOELLER: Frank?
j 7 MR. PARKER: I guesn I was confused in your l -
8 overview. You say this is a tool for site characterization ,
j 9 and planning. I assume that this meant that you were going -
l 't j
(. '
10 to look at the site characterization -- this would help
!- 11 look at the site characterization plan, which I would not 12 have thought would deal with all the man-made barriers.
13 And yet the discussion involves the man-made barriers. And i 14 the second question is, if you are going to-deal with the 1
h[ 15 man-made barriers, isn't that already written in 10 CFR 60 i
4 16 on some of those -- aren't those allocations already given 1 17 there?
f 18 MS. BROOKS: Back to the first part of your i l
} 19 question, in site characterization planning they are going i
j 20 to be laying out tests for the site itself, knowing that l 21 tentatively what the design of the engineered part of the 1
! 22 system might be. And, also, wanting to find out what are t
23 the characteristics of the site that would bound the ,
24 required properties of the engineered system. So that --
l 1
j 25 MR. PARKER: I guess I would have thought that l
l C:)
i 1
ace FEDERAL. REPORTERS, INC. !
l 202 347 3700 Nationwk!c Cmerage 800 33MM6
q 29879.0 76 BRT
() 1 the site would have to be reasonably well characterized 2 before they could determine exactly what they would need 3 for the man-made barriers.
4 MS. BROOKS: Before they could determine exactly; 5 yes. But in terms -- the word " tentative," that we talked 6 about before. This is why you would start out with some 7 tentative figures to box things in to get started.
8 DR. SHEWMON: Frank, one of the other aspects of 9 that is, even given 10 CFR60, there's a question whether 10 you are going to take any question for the Bentonite, any 11 question for the stainless steel can that this stuff is 12 there in, or whether you are going to only rely, in the 13 case of salt on 4 inches of steel which is there as yet a J
-)
14 third layer.
15 So the estimate coming out of the whole package 16 is defined by that, but how it is distributed is a tactical 17 problem, or strategic one.
18 MS. BROOKS: Now, on the performance goals --
19 MR. STEINDLER: I'm sorry. Ask one more 20 question? Is the performance target limited to 10 CFR 60?
21 Or are you also addressing the issue of what is written in 22 the EPA rules?
23 MS. BROOKS: Effectively we are also addressing 24 that, and effectively the revised version of part 60 rules 25 incorporate the EPA standard.
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 77 BRT
_ j.3
(_) 1 MR. STEINDLER: Okay. Because otherwise the 2 issue becomes somewhat academic; you have a part 10 to the 3 5th, you have a part in 1000 travel time -- a few of these 4 things get to be fairly routine.
5 MS. BROOKS: Right. The achievement of the 6 performance goal. First of all, performance goals are 7 meant to guide the test program so what you want to do is 8 evaluate whether your testing has shown that you are
~
9 meeting the performance goals. And you would take the test 10 data and perform performance assessments that are 11 sufficiently detailed; answer the question: Are you close 12 to meeting that goal? And they are primarily to guide the 13 testing and design.
.g-)g
\.
14 Now, the other important part of a performance 15 allocation is a statement of the desired confidence level.
16 A definition of " desired confidence level" is that it is 17' the probability that the actual value of an uncertain 18 parameter or performance measure is at least as favorable 19 as the performance goal.
20 The confidence level is intended to express how 21 well DOE thinks it needs to meet a tentative performance 22 goal. It should be stated in quantitative terms, if 23 possible, and always it should be stated on the most 24 defensible basis. We recognize that the confidence level 25 need not be statistically meaningful and may be set by Oa ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 78 BRT g
( l expert judgment. Certainly, the initial performance 2 allocation, that's the most likely way it will be 3 determined. And -- yes?
4 MR. MARK: In what I have read, there were two 5 different definitions of " confidence. level," one offered by 6 the DOE and one of fered by the NRC. You quoted one. Which 7 was that? And is there now agreement that that is the one 8 that may be used?
9 MS. BROOKS: The agreement was that the NRC's 10 version would be used.
11 MR. MARK: And that's what you read?
- 12 MS. BROOKS: What I read you is actually a fs 13 variant on that.
U 14 In the meeting we said the probability that the 15' true value of an uncertain parameter -- oh boy -- an 16 uncertain performance measure, if I remember correctly --
17 is at least as favorable as a specified value. And I have 18 in this case for clarity made some changes.
19 MR. MARK: That's the definition which NRC is 20 willing to accept and DOE is also?
21 MS. BROOKS: Yes.
22 MR. MARK: Fine.
23 DR. REMICK: Does this solve the controversy 24 that was going on on reasonably assurance and reasonable 25 expertation, if I have my words correct? Or am I in the ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
29879.0 79 BRT
-w wrong --
' (_) I 2 MS. BROOKS: As far as I know all it does is 3 solve the controversy as to what a confidence level is.
4 You will find sometimes reading the DOE literature that 5 they don't necessarily use confidence level, desired 6 confidence level, or they may use indication of confidence; 7 but we are meaning the same things when we use those terms.
8 MR. KASTENBERG: How will you use this in 9 evaluating what DOE does? If they come in with a parameter 10 and they tell you: Well, we feel the confidence is low and 11 you both agree on what " low" is, how do you use that? Do 12 you go back and tell them to do more testing or more
, 13 experiments?
O 14 MR. COPLAN: The reason this definition got 15 changed a little bit from what we had before is that it is 16 focusing on desired confidence level. In other words, they 17 are going to be saying, or we are asking them to say: What 18 level of confidence do you think you will need in this goal 19 when you have completed site characterization, in order to 20 be able to proof up your repository, so we have departed 21 somewhat from the basic statistical concept right off the 4
22 bat. The idea there is that they be saying up front: How 23 good -- well, implicitly, how good a test program they 24 think they are going to need in order to actually make the 25 determination in the long run.
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M6
29879.0 80 BRT 1 MR. MOELLER: I don't understand the first 2' bullet, where it says, I guess, that the desired confidence 3 level expresses how well DOE thinks it needs to meet a 4 tentative performance goal. That sounds to me like the DOE 5 can say: Well, I've decided not to meet that one or I 6 think I'll-meet it halfway or something. I don't 7 understand the bullet.
8 To me --
9 MS. BROOKS: It's certainly not meant to express 10 the way it comes across to you.
11 MR. STEINDLER: I think you have it right. But 12 the sum of all the parts eventually have to get to the 13 desired performance.
14 MR. MOELLER: Well, if that's the point, I'm 15 okay.
16 MR. STEINDLER: The DOE is allowed to select.
17 MR. MOELLER: But I thought it meant that it 18 expresses the degree of assurance that DOE would give us, 19 that they are meeting this particular goal. Not just how 20 well they meet it, but the degree of assurance with which 21 they are meeting it.
22 MR. STEINDLER: You are saying the same thing, 23 aren't you?
24 MR. MOELLER: Is that the same thing?
25 MR. DILLON: Just about. That's really what you O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
29879.0 81 BRT
() 1 are talking about. If 300 years is as good as 1000 years, 2 why, you have a'certain flexibility, perhaps.
3 MR. MOELLER: I could see a totally different-4 situation, Marty, with what you are talking about. Of 5 course the total system must meet the goals, and I think 6 NRC has said that DOE should have flexibility in deciding 7 how much weight to put on each barrier. But that's not 8 what I read here.
9 MS. BROOKS: Again, I think that as we go 10 through some examples it may help on that.
11 MR. MOELLER: Okay. I'll wait on that.
12 MS. BROOKS: If it doesn't I'll tackle it again.
13 I think the final bullet which says that the 14 desired ~ confidence level may be stated as high, medium or 15 low, provided the terms are defined, is what gives -- what 16 we might expect to see in the very first initial 17 performance allocation.
18 MR. PERRY: Excuse me, I think you have passed s
19 over that third bullet there about -- that the confidence 20 level need not be statistically meaningful?
21 MS. BROOKS: The desired confidence level that 22 they start out stating would not necessarily be the same l
23 thing you'd expect to see if you had done all the work, 24 gathered all the data and established what the confidence 25 level is.
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage MG336-6M6
29879.0 82 BRT A
t, )
1 MR. COPLAN: "Better risk" might be a better 2 choice of words than " meaningful."
3 MR. PERRY: It occurs to me that statistical 4 accuracy of corrosion data may well be plus or minus 50 5 percent. But then I got into real discomfort with what
~6 level of confidence you have in the data; 50 percent of 7 what? l 8 MR. COPLAN: I think this comes back to'the 9 point that --
10 . MR. KASTENBERG: That's not what you mean here.
11 MR. COPLAN: -- that I think Dade was getting at 12 a few' moments ago. We ought to round out that decision now, 13 I think, rather than put it off.
14 What we are figuring is that in some instances, 15 DOE may have some flexibility as to how much reliance they 16 can put on it. By putting more dollars, more effort into 17 it they can get more out of it.
18 In other instances, I think, they are going to 4
19 be in a situation where the state of the art or just Mother i 20 Nature is going to put them in a position that they may not
, 21 be able to do better than, you know, a 50 percent level of 22 confidence or a 60 percent level of confidence.
23 There, as Dr. Steindler pointed out, the main 24 concern is how do things come together in toto to make the ,
25 system work out at a high enough degree of confidence to be A
V l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 80 4 336-6646
. . . . = ..
29879.0 83 BRT (q,j 1 able to get through the licensing procedure? And what we 2 are really asking in the performance allocation scheme is 3 for DOE to recognize that, to lay out that. For this 4 particular subsystem we are not expecting to be able to 5 ever get to a high-level of confidence, for example, but 6 these other systems in consequence are going to require 7 fairly high-level of confidence. This is what we intend to i 8 do there.
1 9 MS. BROOKS: Now, on the sixth chart, we have 10 agreed with DOE as to many of the aspects of how a 11 performance allocation will be done.
12 First of all, we recognize a performance ew 13 allocation should be developed as early as possible, since
.Q 14 it is designed to help guide the test program planning.
15 We have agreed that the performance allocation 16 should specify the particular barriers to be relied on and 17 which ones will be held in reserve or redundant status and, 18 since they will be using a top-down approach they will have 19 identified the other barriers that they might not be taking 20 credit for, if they will be identified for further study as 21 to whether their existence would degrade the performance of 22 other components that they might be taking credit for or 23 holding in reserve.
24 MR. MOELLER: Now, if I were DOE -- I was trying
- 25 to think what approach I would take on this, and do I have ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
g. - , , - ,-yg s,- - - , - - - - - - , - - - , - , - , -
29879.0 84 BRT -
_O_ 1 the opportunity to change my level of performance for each 2 barrier if I gather data and want to relax that one or 3 tighten it up? I can do that?
4 MS. BROOKS: That's the beauty of the thing. If 5 you find that you have to relax one then you tighten up 6 another.
7 MR. MOELLER: Fine.
8 MR. KASTENBERG: Then maybe what you need to add 9 to this is that you have some internal consistency, so that 10 when you start changing allocations you still have the same 11~ top level and the same confidence level for that top level.
12 Because when you do these, they tend not to be simple sums.
- g. 13 You know, when you propagate uncertainty you don't just add
\-)
14 them all up.
- 15 MS. BROOKS
- I would think that would be the 16 thing they would be working for and we would be reviewing 17 it for.
18 MR. KASTENBERG: Somewhere in this you need some
- 19 what I would call a consistency measure.
i 20 MR. COPLAN: Pauline is going to describe this 21 again on the next page, I think, but there are a variety of
- 22 ways that conservatism can be worked into the system in the 23 beginning and that ties, really, to how they might adjust 24 the goals as they go through site characterization, reduce 25 uncertainty.
i l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
29879.0 85 BRT
/^N
.( ) 1 MR. PARKER: You say assigning these goals and 2 being able to revise it is the beauty of these things but 3 we always see when you change it -- say you don't have a 4 parameter -- that you are not going to be able to show as 5 much confidence as you originally thought, that's going to 6 be perceived almost immediately by everybody outside as 7 relaxing the standard.
8 MS. BROOKS: Not if you are going to offset it, 9 I think, offset it by tightening up on another component.
10 MR. STEINDLER: I think you are right.
11 MR. PARKER: That's not the way it happens in 12 the states. Maybe that's the way it happens in Washington 13 but it sure doesn't happen in the states.
14 MR. KASTENBERG: In the provinces.
15 MR. STEINDLER: I don't see they have a whole 16 lot of choice.
. 17 Let me ask you one. question: You say this is an 18 approach agreed to by DOE. When did you folks arrive at j
19 this agreement?
l 20 MS. BROOKS: Primarily in the September '85 21 meeting.
l i 22 MR. STEINDLER: So DOE has looked at this for 23 about a year and a half and has been able to do its 24 planning and so forth?
25 MS. BROOKS: Yes.
l l
- /\CEJFEDERAL REPORTERS, INC. l
~
202-347-3700 Nationwide Coverage 800-336-6646 e t
29879.0. 86 BRT
( l MR. STEINDLER: All right.
2 MR. MOELLER: And when you said these things 3 should be specified early, when will you start receiving 4 the barriers -- you know, the listings for each barrier?
5 MS. BROOKS: It will be definitely a part of 6 chapter 8 of the site characterization plan for each site.
7 MR. MOELLER: Okay.
8 DR. SHEWMON: I'm uneasy about what you call 9 your reserves here. In a sense it is related to what Frank 10 asked. I don't know exactly where it is, but there's this 11 thing about all of a sudden we decide that the waste form 12 is 10 orders of magnitude more_ reactive. You say never 13 mind, we haven't ever taken any credit for this other can 14 here and so we are still safe.
15 These reserves that you have or whatever your 16 word is --
17 MS. BROOKS: Reserve or redundant status; yes.
18 DR. SHEWMON: --
is that comething that won't 19 have to be looked at at all? It seems to me politically 20 you are in a better position if you say we have got two 21 orders of magnitude safety right up front. Then if you 22 have to fall back an order of magnitude you are okay, 23 instead of like all of a sudden like a rabbit out of a hat 24 you say: Gee whiz, that didn't make any difference anyway.
25 MS. BROOKS: DOE has a number of options in l
l 1
i
/\CEJFEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6 l
29879.0 87 BRT
{n
_) 1 looking at the three categories. The item for which you i 2 are taking credit, the item that you are holding in reserve, 3 as I said, or redundant status, which means you might take 4 credit for several, more than one item.
5 DR. SHEWMON: I hope these all get listed fairly 6 early.
7 MS. BROOKS: They should be. And then the ones 8 that you are not taking credit for.
9 Now, the level and type testing that you do 10 would depend on the component and how critical you think it 11 is to the system.
12 Continuing on, page 7, performance allocation 13 will initially be established by a technical management 14 judgment. It is understood that available information will 15 be incomplete. And the performance allocation will include 16 conservatism as needed to account for uncertainties, and 17 one way of doing this is to have the goals for reserve i
18 components. Another is to set the goals higher than t
19 actually needed. And there is an interplay there, to see 20 the combination.
i 21 We believe that performance allocation preserves 22 DOE's flexibility to determine which components will be 23 relied on and for how much performance. And we also have 24 the biannual updates of the site characterization plan so e
! 25 that the performance allocation can be and should be O
ACE FEDERAL REPORTERS, INC.
j 202-347-3700 Nationwide Coverage 800-336-6M6
I29879.0 88 BRT
() 1 revised periodically to reflect the site characterization 2 and test results.
3 MR. MOELLER: In this, is there a formula that 4 is generally accepted so that I would know how much 5 conservatism I have'to incorporate in order to allow for a 6 given uncertainty? Is that a standard, universally 7 acceptable approach? Or is there a standard approach for 8 that?
9 MS. BROOKS: ' There are standard approaches but 10 there's no real formula.
11 MR. MOELLER: How are you going to decide, then, 12 how much conservatism is needed for a given -- to offset a 13 given uncertainty?
4 O. 14 MS. BROOKS: Were we planning on discussing that 15 later today?
16 MR. COPLAN: No. I think the answer to that.is 17 we don't know the answer to that yet. It's the gut 18 question involved in licensing a repository, really.
19 MR. MOELLER: And again, then, that brings into 20
- question the point that several ACRS members have been
> 21 asking about, and that is: Do you have DOE's computer 22 codes for assessing the performance of individual 23 components and systems in totality, and how do you decide 24 when those computer codes or programs are acceptable to you?
25 Do you have your own and compare yours to theirs? Who is O
i I /(CE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 89 BRT.
() I the judge? Who makes the judgment on which computer 2 programs are acceptable to NRC for assessing the 3 performance or estimating the performance of a repository?
4 MS. BROOKS: Well, there is ongoing work on our 5 part and on DOE's part regarding verification and 6 validation and the benchmarking of codes. In addition to 7 all the international work that is going on, hydracoin and 8 Intracoin and so forth, there are groups that have gotten 9 together to establish and are working on the basic guidance 10 for doing this sort of thing, so the most important thing 11 is to make sure that those codes are properly verified, 12 validated and benchmarked to the extent possible.
13 MR. MOELLER: And do you have a written O 14 agreement with DOE at the moment that tells how this is 15 going to be done?
. 16 MS. BROOKS: No, we don't.
17 MR. MOELLER: Somebody is working on that? Or 18 you have a target date when it will be done?
19 MS. BROOKS: There was a meeting, I think --
20 John, did you attend a meeting in which this sort of thing l 21 was discussed? Okay. At any rate there are no agreements 22 at this point.
23 MR. STEINDLER: How do you intend to satisfy the l 24 problem that a licensing board will have in the event that 25 both you and the applicant are asked to define and explain I
(
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage MG336-6646
29879.0 90 BRT 1 the methodologies that are involved here?
2 MS. BROOKS: Actually I think I missed about the 3 first 5 words you said.
4 MR. STEINDLER: I'm assuming that this whole 5 process may well end up in front of.a licensing board, 6 which is, in effect, a public domain issue. So you are 7 going to have to somehow or another explain to the public 8 at large how you've gone about satisfying yourself that the 9 criteria that DOE has used are criteria that you would 10 accept. How are you going to do that?
11 MR. COPLAN: I think there's a --
12 MR. STEINDLER: The whole question of commuter 13 codes is likely to be subjected to a great deal of
- 14 mistrust. Certainly by the licensing board. For good 15 reason.
16 MR. PARKER: For good reason.
17 MR. STEINDLER: Yes, for good reason. You are 18 right.
19 MR. COPLAN: If I'm understanding the drift of 20 the discussion here, I think there are a couple of points 21 where maybe we haven't been as clear as we need to be.
22 First off, this performance allocation process 23 is a tool. It's not something part 60 would require, and 24 ultimately when we do appear before a licensing board what
- 25 is going to be at issue is whether or not the Department of t
(
l l
ACEJFEDERAL REPORTERS, INC. !
I 202 347-3700 Nationwide Coserage 800 336-6M6
l l
~29879.0 91 BRT C
i) 1 Energy meets the criteria, performance objectives that are 2 in part 60.
3 We see the performance allocation process as 4 being a way of getting a handle on the various test 5 programs for the site, for the design components, way in l 6 advance of the licensing proceeding, and being able to make 7 appropriate adjustments during site characterization, 8 depending on the results that come in. In other words, 9 there is going to be something of an iterative process 10 going on here.
11 As far as the computer codes go, I guess I would 12 have thought of that as something of a separate issue, but 13 we see that as kind of a second leg of three legs that go iO ' 14 to this basic question of uncertainty. The three legs that 15 we see there are: first, an issue regarding the 16 completeness of scenarios: How well has DOE done at 17 identifying what is going to happen to a repository?
18 The second leg being, all right, assuming that 19 they have identified what is going to -- what can happen to 20 it, how well are they doing at modeling what the 21 consequences of those things are? And, final leg, is one 22 that has to do with data and parameter uncertainties.
23 Focusing on that second leg, namely, how well are they 24 doing at modeling, again the direct answer to the question 25 is: We don't know yet how we would qualify codes. One of
/\CE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
, . . - - ~ _ . - - . _ . . , - , . - - , ,-
3 --..-,.-y.,,._----.--.--,.--,---y, - - -
29879.0 92 BRT
(~ 1 the things that we expect to be working on over the next t
2 several years is developing criteria that can be used to 3 judge the adequacy of the validation program, and at that 4 point it will become virtually a moot point whether it is 5 their codes, our codes, but one way or another there would 6 be some standard for you to judge the adequacy of the codes.
7 MS. BROOKS: On the eighth chart we tackle a 8 hypothetical, highly idealized example of how performance 9 allocation might work, and I want to say ahead of time that 10 this is strictly not to be construed as saying: This is 11 how it should be, these particular things should be 12 included in the performance allocation.
13 There are the four objectives of part 60 that 14 need.to be met. This is where we are going to start, with 15 the regulatory requirements. One on containment time, the 16 one on release rate, the one on groundwater travel time, 17 and the EPA standard.
18 MR. PERRY: Excuse me. On the groundwater 19 travel time, isn't that on the undisturbed groundwater 20 travel time?
21 MS. BROOKS: The basic objective is on the 22 preclosure groundwater trial time. The reason for the word 23 "postclosure" here, if you look on the second column above 24 targets and these retardation factors, we are going to have 25 to in some way consider the hydrologic regime, O
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6M6
29879.0 93
-BRT g 1 post-emplacement and that was just a way of noting that, 2 but basically, yes, the objective is pre-emplacement.
3 MR. PERRY: Actually it's not pre-emplacement; 4 it's undisturbed.
> 5 MS. BROOKS: Undisturbed; right.
6 MR. PARKER: I guess I'm confused. Why do you 7 have " unsaturated zone" in here? Do you think it's 8 specifically a Yucca problem?
9 MS. BROOKS: That's the other thing we need to 10 warn you about. No, we just figured that was an easy one 11 to use in the example. We are not necessarily thinking of 12 Yuca Mountain. It's not meant to be a real site in any 1
13 sense.
O 14 MR. PARKER: Where does the 5000 years come from, 15 or is it 1000 years?
16 MS. BROOKS: That's just a proposed --
l 17 MR. PARKER: But isn't the number in the i
18 regulations already 1000 years?
19 MS. BROOKS: This is something -- DOE can select 20 a higher figure if they choose to for this initial 21 performance allocation, and we are just recognizing that, i
22 Also, the second column shows, first the barrier 23 that we are concerned with and then the particular 24 component that DOE might say they are taking credit for, 25 primarily; relying on.
l i
()
i ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
29879.0 94 BRT
() 1 So that for containment time the barrier.would 2 be the waste package and the component of it might be a 3 canister. Then the performance target they may choose 4 would be the 1000 year containment time with a confidence 5 level of 8 percent.
6 Again, with the fully engineered barrier system 7 which would have to provide for the 10 to the minus fifth 8 per year release rate, they may choose to rely on the waste 9 form which you have, with a'certain percent confidence 10 level.
11 The unsaturated zone, they would be identified 12 as a barrier and they might, for the component, choose a 13 particular unit in the unsaturated zone and say the travel 14 time through that should be a certain length of time. Then, 15 for the EPA standard they've got to meet our performance 16 objectives and make sure that the release to the accessible 17 environment is within limits. So they might want to reach 18 all the above targets, considering what is happening with 19 the flow, and then certain retardation factors.
20 MR. MOELLER: Then X, Y and 2 are just three 21 nuclides?
22 MS. BROOKS: Just three nuclides, right.
23 That's sort of the way it would work or could 24 work at the top level.
25 MR. STEINDLER: I have a little trouble seeing ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
~
29879.0 95 BRT
( l the allocation aspect of performance allocation in this 2 performance target table.
3 MS. BROOKS: It might better have been 4 performance goal. But for the canister containment time, f
'5 the goal should be -- we assume that they took 1000 years 6 containment time.
, 7 MR. STEINDLER: So next to it there is a column 8 that says " regulatory requirement." Containment time of 9 1000 years is certainly one of them.
10 MS. BROOKS: 300 to --
11 MR. STEINDLER: So there the performance 12 allocation is direct?
13 MS. BROOKS: We went on the high side of it.
14 MR. STEINDLER: It's 300 to 1000, depending on i
15 something which is not very well defined.
16 MR. MOELLER: They cannot, though, go below the i
17 300 years, because that is a specified barrier?
18 MR. COPLAN: That's correct.
19 MR. STEINDLER: There's no combination of trade 20 office that will get them to 1000 years. In fact there 21 doesn't seem to be any reason why they should allocate 1000 22 years. It doesn't do anything for them further down the 23 line except a little additional --
24 MS. BROOKS: It gives them a little leeway down i
25 there, a little overlap?
i O
i
/\CE FEDERAL REPORTERS, INC.
(
202-347-3700 Nationwide Coverage 800-336-6M6
29879.0 96 BRT A
T ) 1 MR. STEINDLER: You mean in case 8 percent is 2 too high? Is that what you are saying? That implies 3 there's some absolute --
4 MS. BROOKS: In case some other component of the 5 waste package, for example, doesn't help it meet the 6 containment time.
7 MR. STEINDLER: That's buried in the 8 percent; 8 isn't it?
9 MS. BROOKS: In case the canister doesn't do it 10 or they look and find the canister plus whatever else is in 11 the waste package together won't do it, they've still got a 12 chance to play with those numbers. But they can't play-13 with them except by looking at the whole picture, and where 7S V 14 they change one they need to -- they would have to 15 compensate by changing others.
16 MR. PERRY: Earlier presentations on this topic 17 would suggest that the containment time of the barrier, the 18 first item going across, would be a result of allocation of 19 performance of individual subsystems, if you will, or 20 subcomponents of the waste package, such as backfill, any
, 21 added material, such as Bentonite around the canister, the l 22 canister wall, an overpack over the spent fuel and so forth.
l 23 Where is -- do you have an example showing how i 24 that subfractionation would go? This does not appear --
25 this table does not appear to represent the kind of ;
O ,
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
29879.0 97 BRT
('
( l performance allocation description that we've seen 2 previously. This merely says the waste canister package 3 will meet 1000 years, which is from part 60, at a certain 4 confidence level.
5 MR. COPLAN: This is quite simplified, really.
6 MR. PERRY: Obviously.
7 MR. COPLAN: We are not suggesting the type of 8 allocation figure you are describing is not something the 9 DOE couldn't do.
10 MR. PERRY: It's more than that. Second --
i 11 MR. COPLAN: In fact, we would like to see how 12 they plan to break out the components. In doing that, that 13 helps us to get a better understanding and put us in a 14 better position to be able to review the test programs that 15 they were describing in the site characterization plans 16 that are intended to show that they can get thus and so 17 much performance out of an overpack, and so much out of a 18 canister and so on.
19 John, is there anything you would care to add to 20 that?
, 21 John Vogelwede is from our waste section.
22 MR. VOGELWEDE: Jack is right. There's been 23 some discussion at the DOE projects about further divisions 24 in performance allocation. A good example is the use of 25 zine alloy cladding as a barrier on the SALP program. It
- /\CEJFEDERAL REPORTERS, INC.
'02-347-3700 Nationwide Coverage 800-336-6646
29879.0 98
-BRT' r
1 appears that they will not be relying on this as a barrier.
2 It would be one of those things that are held in reserve.
3 At the TUP program it looks like they will be relying on it 4 so they'll get 800 years, zircalloy an additional 400 years, 5 getting over a 1000 year limit on containment time.
6 DR. SHEWMON: What does it cost to call up this 7 out of reserve, or whatever it is.
8 MR. VOGELWEDE: That's a difficult question, 9 Dr. Shewmon.
10 DR. SHEWMON: Especially if you are using it on 11 one site but not at another site.
12 MR. VOGELWEDE: We won't have to look, for the 13 moment, at brine-zircalloy combinations, so that's ,
14 something to be said for the allocation plan, but from 15 DOE's standpoint to hold something in reserve and planning 16 on using it up front is going to require some corrosion 17 tests with zinc alloy, whereas simply indicating that 18 component of the barrier system is available but not 19 bringing it up front means that our experimental programs 20 have to be made above and beyond showing that it's not 21 detrimental.
22 DR. SHEWMON: Okay.
23 MR. PERRY: John, in his last sentence, picked 24 up one negative aspect of that, showing it's not 25 detrimental.
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
i 29879.0 99 BRT 73 Of course, the point that you raise is that if
() 1 2 they try to call it up, you know, halfway through the 3 hearing process, that means that some -- maybe two to five 4 years of corrosion studies would be necessary to show that 5 the zircalloy will, in fact, provide the 4- to 500 years or 6 however it is they need to get by at a sufficient 7 probability or confidence.
8 MR. COPLAN: That's correct. That's why one 9 would hope that, as they lay out the allocation, they say 10 we are going to hold the zircalloy in reserve but they are 11 also taking into account what kind of time frame they would 12 be involved in, in actually taking credit for it if they 13 did have to call upon it, and start appropriate testing fS V 14 during site characterization to allow them that flexibility.
15 MR. PERRY: Out of discretion I won't comment 16 further on that since we have some SALP people here.
17 DR. REMICK: In the reactor licensing area, one 18 would describe the same thing differently. You say one 19 doesn't take credit for it. You are not holding it in 20 reserve. There's a difference in those words but I think 21 it's the same thing. It's something that's there, but in 22 your analysis you are not taking credit for it.
23 It has a little different flavor to it, I think, 24 than holding something in reserve.
25 MR. COPLAN: This is a little bit different. It i
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
29879.0 100 BRT
,~
1 goes to the point that was raised earlier about what is, I 4
2 think, largely a perception problem, adjusting goals during 3 site characterization, where if, for example, as DOE is j
4 doing their site characterization they find that a 5 particular barrier is not going to do as well as they had 6 hoped at the beginning and they have to adjusts that goal 7 downward and adjust something else upward, a way of dealing 8 with what might be perceived as playing fast and loose with 9 everything is to start outright from the beginning and say:
10 This is my goal for barrier X. I am holding barrier Y in 11 reserve. I need to show that barrier Y -- if I'm going to 12 have to take credit for barrier Y, what I am going to have
- 13 to do to be able to take that credit is go through such and 14 such test program.
i 15 DR. REMICK: You sharpen your pencil, sure.
j 16 MR. COPLAN: But you are describing, to a degree, 17 in advance how you plan to sharpen your pencil if you have 18 to.
19 DR. REMICK: Oh, I see. So you would set 20 performance objectives for those items in reserve also if 21 you need them, I see, ahead of time.
22 MR. COPLAN: That's correct. That's right.
23 MR. MOELLER: On this hypothetical example it 24 seems to me there are two items you are addressing, or two 25 parameters or two performance factors. One is the time at l
l ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 800-336-6646
29879.0 101 BRT
, 1 which the canister ultimately begins to leak. That's the 2 1000 years. And the 5000 years also ties into that.
3 Because if you could keep it from leaking for 5000 years 4 you wouldn't get the movement.
l 1 5 Then the second thing is, once it leaks, how 6 fast does it leak? And that's the second and fourth items, 7 or how fast does what it leaks move? So there are the two 8 items.
9 DR. SHEWMON: You're going so well, would you l 10 explain what the last three lines mean, too? Is that i 11 something I'm supposed to understand or is X63 Y127 Z249 12 just so you put some numbers there?
,e 13 MS. BROOKS: Hypothetical nuclides.
14 DR. SHEWMON: Oh, they are nuclides.
15 MR. MOELLER: They are nuclides, and the 16 retardation factors are the ones assigned to those three 17 for some --
18 DR. SHEWMON: Those factors I'll understand 19 after I read what Carl gave me for a reference. Pardon me, 20 Jack.
21 MR. PERRY: The point you were approaching which 22 I think really is more critical is -- are going to be the 23 definitions of what constitutes small fractions -- I'm 24 sorry, I forget the terms right now -- small fraction 25 release that's allowed during the first 300 to 1000 years.
(
m_-
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800 33MM6
29879.0 .
102 BRT
-O 1 That's the really critical item. And what constitutes 2 release? Mere penetration of the canister? Or does it 3 actually have to get outside the waste package, which then 4 puts a requirement that you determine in advance what an 5 undesigned backfill system, how it is going to perform 6 under an unknown recharge rate from the water that may be 7 coming into the system?
8 MR. PARKER: So you won't get off too quick, may 9 I ask two questions? One is since the site 10 characterization plan is, I guess, pretty well complete, 11 almost ready to be issued, how are they treating that? I 12 mean, you are talking about this perspective, but those 13 plans are practically done now.
O 14 MS. BROOKS: As I understand it, what we will 15 find out at this meeting in March with DOE headquarters is 16 how they have gone about using performance allocation to link 17 the site characterization plans and how they are going to 18 use it in the site characterization plans and how far they 19 have gotten in doing that.
20 I understand the first site characterization l 21 plan is still expected -- what, in July, is it?
22 MR. COPLAN: I think that's what we are hearing 23 at this point.
24 What the March 3 and 4 meeting is briefing, 25 really, is that DOE is going to describe to us how they l
l 1
/\CEJPEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 80433MM6
29879.0 103 BRT s
(q_j 1 factored performance allocation into the site 2 characterization plans.
3 From discussions that I have had with them just 4 to set up the agenda for the meeting, my understanding is 5 that this has had kind of a pervasive influence on the way 6 they have written the site characterization plans. That 7 they have used this as kind of the fundamental logic for 8 developing the plans, deciding what it is that they need to 9 do, how various parameters influence one another, what 10 implications that has for the test plans and what degree of 11 testing is going to be needed to determine various 12 parameters.
- g- 13 MR. PARKER
- Since it's practically done, maybe b my next comment is sort of superfluous, but it seems to me 14
! 15 I would have turned it all around since the primary purpose 16 of all of this is to set up the test plans and the site
- 17 characterization plan, and to set up proper testing you 18 need to know what the objective is. So I would have 19 phrased it in those terms rather than saying this is the 20 performance that we have to achieve, and then you wouldn't 21 be stuck with this perception problem that you are going to 22 have by setting up performance goals that you are going to 23 have to change.
24 MR. COPLAN: That is the intent of this, really, 25 to establish objectives for those programs, and a way of O
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
'29879.0 104 BRT
.(qj 1 our being able to review them in'some sort of objective 2 terms rather than just saying: Well, we understand that 3 there now exists test X,.Y or Z, why aren't you using that 4 one, too?
5 MS. BROOKS: Well, that oversimplified' example 6 engendered a lot of discussion.
7 (Laughter.)
8 We wanted to take up an example of how 9 specifying the goal and the associated confidence level 10 would help in selecting test programs, and.on chart 9, laid 11 out what the problem is.
12 We want to establish through performance s 13 allocation what the performance goal and desired confidence
, 14 levels are and we, again, oversimplifying these things, are 15 assuming that the goal in this case is having a '
16 conductivity of less than 10 to the minus 8, and the 17 associated or cesired confidence level is 90 percent.
18 That's a 10 percent chance of not meeting that goal. So 19 that's our starting point.
20 Now we have soma .u lt ,cograms available, and 21 our problem is to determine which test program is suitable.
22 We know that test program A is good to plus or minus two 23 orders of magnitude and test program B is good to one order 24 of magnitude.
25 DOE has a choice there. Their concerns would be O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 -105
'BRT
() I with the cost and time consumption. How long will the test 2 take and that sort of thing? And our concern is primarily 3 what test is adequate for the job?
4 The other thing that you know from early work on 5 the site is what -- you may have a preliminary estimate of 6 the conductivity, and so you want to look at that 7 preliminary estimate for that parameter and see if the 8 combination of the information you've got will help you 9 determine which test program to use, and there were three 10 cases, in terms of what the site information is.
11 At the moment, case 1 would be that the 12 conductivity is 10 to the minus 11. If you turn to the 13 next chart you will find sort of a graphic representation-
~
14 of how you might go about choosing the test program.
15 The first case, you've got the field reading of 16 10 to the minus 11, and your goal always, in all the cases, 17 is 10 to the minus 8. Now, on test program A, presumably 18 that bar is drawn at two orders of magnitude and we are 19 concerned with the upper side. We are talking about 20 confidence levels toward the goal. And in the first case 21 both tests A and B are, with their error bars lying below 22 the desired goal.
23 In that case, obviously the deciding pointed is 24 going to be some other factor. As far as we are concerned 25 they are both adequate. As far as DOE is concerned, they
/\CE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80 4 336-6646 g
,. - _ - . . _ . _ . . - . . . _ _ _ _ . . . , - , , _ , ~ . _ , - _ . _ . - . _ , _ _ - - - - - . _ _ _ _ _ _ - , _ - _ - - - - . - - . . - . .
29879.0 106 BRT
) 1 need to make a choice based on practical considerations.
2 In the second case, though, they don't have that kind of 3 choice, because the error bar for test program A is now 4 getting into the area of the. goal 10 to the minus 8, 5 whereas test B is not. And so they would be constrained to r 6 use test program B in that particular situation.
7 Finally, if the situation were as in case 3, 8 they would have to go find another program.
9 DR. REMICK: That's the thing I don't understand, 10 because that's an estimated value, 3 times 10 to the minus 11 9. That might be the upper error bound when you actually 12 do some testing, although you could rule it out.
-, 13 MS. BROOKS: It's a preliminary figure. Yes.
14 MR. KASTENBERG: That's what's confusing. You 15 are using the word " actual."
16 MS. BROOKS: " Actual" probably was not the best 17 choice. As I was looking at it, it's preliminary, it's
- 18 really preliminary.
19 MR. KASTENBERG: It's an estimate, and the way 20 you have drawn it is the best estimate, and that's where I 21 think the confusion is. But I'm just curious about 22 something, too. On page 9 you talk about a confidence 23 level of 90, if that were the target. Then on 9-A you have f 24 80. If it's bounded from below the 90 and the 80 are not 25 the same.
ACE FEDERAL REPORTERS, INC.
202-347-370) Nationwide Coserage MX)-336-6M6
29879.0 107 BRT (3
y_j 1 MS. BROOKS: Essentially what I was trying to do 2' is translate that 90 percent confidence level, one-sided .
3 confidence interval, in terms of the 150 percent chance of 4 not meeting the goal on that side, and when I-drew the 5 error bars I had both sides, and so I needed to get the 10 6 percent accounted for on the right side of the bar.
7 MR. KASTENBERG: Okay.
8 MR. PERRY: I don't see where. In case 3'they 9 would have to reject test A or B, because you say what you 10 do is you go back and you change your performance 11 allocation to adapt to the situation that you have.
12 MS. BROOKS: That is the other option that they
- ', 13 have.
14 MR. MOELLER: I agree on case 3. It seems to me.
15 case 3, the parameter or the goal is not met, and you relax
, 16 this and tighten something else.
, 17 MS. BROOKS: You would use neither test but you 18 go on to change your test allocation.
19 MR. MOELLER: It seems' funny to say if you don't 20 get the answer you want find new methods of testing.
f 21 MS. BROOKS: But we are talking about a 22 situation, I guess, where you are looking at a problem in 23 the beginning and you are trying to determine whether you I
24 can do it and, if you can do it, what is the best way to go 25 at it. You've got to get some initial boxing in of your O
l l
ACE. FEDERAL REPORT ERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646 l.
. _ _ _ _ _ _ . . . _ _ , _ _ _ _ . _ , _ _ . - . _ . _ _ _ _- _ __ _ .__ . _ .-. ~ _ _ _ _ _ . _ _ _
29879.0 108 BRT q
- 1) I figures to get started on it.
2 MR. MOELLER: But case 3 tells me that the 3 conductivity for'this, you know, medium or whatever it is, 4 does notJmeet the goal.
5 MR. COPLAN: At least the odds are that's what 6 they are going to find.
7 MR. KASTENBERG: But that assumes what you call 8 "actually," and is " preliminary," is somewhere.near the 9 median and the distribution is a straight-line, flat
- 10 distribution. It may be that neither one is the case. You 11 may have some other distribution which has peaked near'your '
- 12 initially measured value, and test program A or B may be
( - 13 perfectly adequate.
14 MS. BROOKS: We are oversimplifying again.
.15 MR. KASTENBERG: I think you understand the 4
16 problem but this example is much -- you are led to the 17 wrong conclusion by the example.
i 18 MR. COPLAN: It is too simplified. We had
! 19 another -- we tried to lay it out in a different way where 20 we were using distributions and whatnot. The whole thing 21 got so complicated it was really hard to talk through it,
[ 22 so --
23 MR. KASTENBERG: I saw that.
24 MR. COPLAN: -- yesterday we opted to do it this l
l 25 way and I think maybe we went too far the other way.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-(M6
29879.0 109 BRT l) 1 MR. PERRY: I find it interesting, too. Before 2 on the allocation, generally you were talking about 3 performance of engineered barriers or man-made or 4 man-designed things. Now you are talking about performance 5 allocation on physical parameters of the host rock, in this 6 case, conductivity. I just find that an interesting thing 7 worthy of note.
8 MS. BROOKS: One of the sticking points with DOE 9 was exactly that point. How can you allocate.any 10 performance to nature?
11 MR. PERRY: That's right.
12 MS. BROOKS: Essentially what you are doing is 13 trying to find out what is out there and giving an estimate O 14 of it and saying: Okay, that's it. What does that mean we 15 have to live with, in designing the system?
16 MR. KASTENBERG: You are allocating to the --
17 your basis of your knowledge.about nature, not to what 18 nature is doing but what you think you know what nature is 19 doing.
20 MR. COPLAN: Also, as soon as you start a design, f
21 before you've got good information, really sound 22 information about nature, about the environment in which i 23 your design is functioning, you really are implicitly 24 allocating some level of performance to nature on the basis 25 of what you have chosen to be your design. And part of I
l
/\CEJFEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336-6646
4 1
29879.0 -110 l
.BRT l
. () I what we were hoping.to get out of performance allocation in 2 the SCPs was a relatively clear statement of what level of ;
3- performance, or not so much performance as what the design 4 was implicitly demanding from the site, in order to be able 5 to function.
1 6 MR. PARKER: I have the same problem but in 7 somewhat different te rms . The way I understand it, at 8 least, you are going to have a 100 percent allocation. So 9 any time you find that you can't meet that, then you are 10 going to have to change everything else around. It seems 11 to me I would set it up with a 1 percent allocation, so you 12 are overdoing it, expect the allocation to be 100 times 13 more than it needs to be, and therefore you are always 14 going to be on the safe side if you come even close to what This way it looks like you have a failure every 15 you want.
16 time you have to change it. I think that's a terrible 17 perception.
18 MR. COPLAN: Again, we haven ' t tried to tell DOE ;
i 19 do it by just meeting things or overspecifying or whatever.
20 I think that's something that they've got to decide, really, 21 how they want to do that. What we are interested in is 22 what is the strategy that underlies the site i
23 characterization program.
i 24 MS. BROOKS: On page 10 I just summarized, 25 really, something about DOE -- what we have said, mostly, l
(~)T ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-3346M6
29879.0 111 BRT A
T ,) 1 about DOE's approach to performance allocation. As Seth 2 said, we are essentially interested in how they are going 3 to lay it out. The decision is theirs. And the 4 flexibility to deal with it is theirs.
5 The last chart, really, is taken from the draft 6 amendment and represents the way DOE lays out its -- what 7 they call the issue resolution strategy. The heart of that 8 strategy is performance allocation, if you notice between 9 the two horizontal bars on here.
10 MR. PERRY: I had leaped ahead and looked at 11 this and note that you don't have a loop on here. You 12 don't have the failure -- or they do not.
13 MS. BROOKS: They do not have the loop. That is 14 what I would add to it.
15 MR. PERRY: That's the point that was made 16 earlier and Dr. Parker just made. It's going to be a 17 terrible perception on the part of the public to come 18 around and every six months be changing this and have to 19 come around for two years of additional work when they get 20 stuck. I think it's a problem.
21 MR. MOELLER: Other questions on this?
22 MS. BROOKS: Essentially in the performance 23 allocation books. As DOE has laid it out, they are going 24 to -- in number 3, which is licensing strategy, they are l
25 going to state what site features, engineering features, i
i ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
29879.0 112 BRT
( ). 1 models and analyses are going to be used to prepare proper 2 data for the license. And then the output of that step 3- will be a statement of the components to be relied on, held 4 in reserve, under status, or zero credit.
5 Then, on step 4 they will be identifying 6 particular performance measures and setting the performance 7 goals and -- in here they call it-indications of confidence, 8 what we have been calling desired confidence level.
9 On to 5, they identify specific parameters that 10 have to be measured or have to be determined, rather, in 11 order to get a figure for the performance measure.
12 One of the things, again I'm using a natural 13 example -- let's say that the licensing strategy for a site 14 is to rely on the unsaturated zone as a barrier. Then the 15 component to be relied on would be identified, perhaps, as 16 a particular unit in the unsat zone, and what you would 17 want to measure is the groundwater flow through that unit.
18 And a parameter that you might need there would be the 19 effective velocity, and you need to set a value for that, a 20 target value for that, and indicate what level of 21 confidence you need.
i 22 From that you would develop your testing 23 strategy, knowing what parameters you need to measure in 24 order to get to the effective porosity. And, finally, you r
25 would provide all supporting -- presumably DOE would f
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336 6646 i
m I
i 29879.0 113 BRT I) 1 provide all the supporting information needed to achieve 2 the confidence in that figure, that they set for themselves.
3 Are there any other questions?
4 MR. MOELLER: I suppose if I were DOE and were 5 setting these performance measures or goals for each of the 6 barriers within the system, I would initially -- I'm just 7 thinking out loud, but I guess I would initially set them 8 relatively high, meaning such that the combination would 9 not meet the total, you know, NRC requirements or EPA's 10 goals, on the assumption that as I get in and measure each 11 one of them a number of them will be well below the level I 12 set, and therefore the final product will meet it or the 13 final system will meet the standards or licensing 14 requirements. Otherwise, if'I set them -- if I chose 15 numbers where my limits for all of the barriers were such 16 that when I multiplied them out I met the goal, then I'm 17 sure as the devil going to exceed several of them when I 18 get down and measure them.
19 MR. COPLAN: If I'm understanding you right --
20 MR. MOELLER: I'm trying to develop in my own 21 mind a procedure for setting these individual system and 22 component goals, and trying to think how I would approach 23 it.
24 MR. COPLAN: What I thought I understood you to 25 be saying, though, is you would pick the goals in a way O
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 8ak336-6M6
29879.0 114 BRT 1 that they wouldn't necessarily add up to meeting the 2 regulatory criteria.
3 MR. MOELLER: They would far exceed the 4 regulations.
5 MR. COPLAN: Far exceed in terms of a failing to 6 meet?
7 MR. MOELLER: Failing to meet.
8 MR. COPLAN: We would have a problem with that.
9 MR. MOELLER: Then I would come at you and say:
10 Aha, now when I measure these individually and show you 11 what I actually have versus my goal I'm going to be well 12 below the goal, and hopefully I won't exceed a single one
, ~3 13 of my goals and I won't be embarrassed.
~'
14 MR. PERRY: Hopefully, Dade, the Staff -- NRC 15 Staff, if the DOE came in and said: Well, this site is not 16 going to meet it but we are going to go ahead and test it
{
17 anyway -- would then tell them not to test the site at all.
18 DR. REMICK: I think you might set it, so you 19 just barely met the overall goal and show most of these 20 improve with time so you are really more conservative.
21 MR. PARKER: Wouldn't you want to do it the 22 other way, as I suggested? Meet it by two or three orders 23 of magnitude, so therefore when you fail to meet certain of 24 the individual ones that you are still okay?
25 MR. PERRY: But you still have to reallocate for s-l ACE. FEDERAL REPORTERS, INC.
202 347 37tW) Nationwide Coserage N O-33MAM
h 29879.0 115 BRT p)
( I those intermediate levels.
2 MR. PARKER: Oh yes. Sure.
3 MR. PERRY: And go ahead on that basis, which is 4 essentially no different, then, from just going ahead and 5 measuring all the data that you think is necessary in the 6 first place without going through trying to guess where you 7 are going to end up four years from now.
8 MR. MOELLER: Bill?
9 MR. KASTENBERG: I have two comments and a 10 question. The two comments are: One, there's quite a bit 11 of literature with respect to reactors on allocation of 12 goals. I know Brookhaven did a study for the -- an 2 13 NRC-funded study -- on a method for allocating the safety 75 V 14 goals that have been promulgated. You are probably 15 familiar with that.
16 I know GE and Rockwell have used it. I guess 17 the Brookhaven work was in a retrospective mode, having the 18 reactor built already and having data. And the GE and 19 Rockwell work, as I recall, was in a prospective mode 20 similar to yours where they are were looking at advanced 21 reactors and asking questio7s: How would I allocate the 22 different systems and so on?
23 I think they are weak on the uncertainty part 24 but you may want to take a look at the literature and see 25 what has been done to help you formulate your approach.
.fD V
ACE FEDERAL. REPORTERS, INC.
202-347-3700 Nationwide Coserage lux)-336-m46
.. - _ - - - - -. . ~. .
29879.0 116 BRT-
.() 1 The second comment is one I made earlier, but'I 2 just want to reiterate that when it comes to aggregating i 3 uncertainties, that you have to be careful. I think the 4 example of drawing those straight lines reinforces my 5 comment. Some may well have well-defined' distributions; 6 some you may have no knowledge. I think you have to be 1
7 careful.
8 My question is, though, if initially DOE comes 9 in and goes the route of just high, medium and low, how 10 will you deal with that? If there are numbers, you know, i 11 there are some techniques that you can use and people have 12 used them in the literature to aggregate the uncertainties 13 and play with the reallocation. But how will you deal and O-14 how will you define what you mean by "high," " medium" and I
15 " low," and how will you deal with them?
16 MS. BROOKS: One of the things we agreed with i 17 DOE is they would define what they mean by "high," " medium" 18 and " low" when they use the terms, so presumably that would 19 be in some -- in terms of numbers or small range of numbers.
20 MR. KASTENBERG: Then how will you make the i 21 adjustments?
i
( 22 MS. BROOKS: I'm sorry, what adjustments?
i 23 MR. KASTENBERG: Well, suppose you need to l
l 2+ reallocate because they come in with certain things of 25 admittedly low certainty or high uncertainty?
O ACE. FEDERAL REroRTEas, INC.
I 202 347 3700 Nationwide Coserage mig-336446 I
l
29879.0 117 BRT
'( l MS. BROOKS: We would point out the defects in 2 _the performance allocation. It's up to them to make the 3 adjustment.
4 MR. PERRY: What would happen in a case like 5 your example, A and B, and you propose certain bounding or 6 scattered data and the tests after they are run come out 7 well beyond that? That is, instead of an 80 percent 8 variation -- or whatever example you want to use -- your 9 data come out wrong? Then you are going to have to l
10 reallocate anyway; is that correct? Or go to another test?
11 I mean if the tests when actually run show the 12 data fail to meet your expectation?
- 13 MS. BROOKS: Again I think they would have 14 options, and options with respect to other tests with 15 respect to other ways of allocating it, basically.
16 MR. PERRY: So then what has performance 17 allocation gained you?
18 MS. BROOKS: First of all, it gains a way of
, 19 tracking what has happened, what they are planning to do.
j 20 As I said initially, it gives a structure to the logic of 21 the test program, basic structure. And that structure, i'
22 when it exists, then gives everybody involved a ch3nce to 3 23 see what is happening when one factor or another is changed j 24 in the system.
i 25 MR. KASTENBERG: I guess that's good for them,
! C) .
1 ACE. FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cmerage 800 3 t6MA6
29879.0 118 BRT (O I and I appreciate that in their planning, but how do you as
()
2 the regulator then decide what they have done is adequate.
3 I haven't heard that. Maybe you haven't formed that in 4 your own minds yet?
5 MS. BROOKS: There are two things in terms of 6 that. There are certain general questions you ask in 7 reviewing a performance allocation; and, secondly, they are, 8 as I understand it -- this afternoon you'll be talking 9 about our implementing the five-year plan where we will be 10 talking about developing staff review capabilities for both 11 the SCPs and license application.
12 MR. COPLAN: I hadn't expected to get into that
>s 13 aspect this afternoon.
s,]
14 MS. BROOKS: Oh, really?
15 MR. COPLAN: But I think the way we would expect 16 to be using performance allocation is, first off, from the 17 standpoint of looking at the overall allocation of 16 pe r f o rma nce , to make sure that they weren't doing what Dade 19 suggested earlier, namely, proposing to characterize a 20 state that they don't appear to have confidence that it can 21 really meet the criteria; and then having assured ourselves 22 that they have, proposing to characterize a site, that they 23 believe they can allocate performance in which way that 24 will meet the criteria eventually, look at the test 25 programs that they are proposing to use to determine the O
N)
Acn-Enonna RuPonTnns INC.
202 347 3700 Nationwide Coserage 800-336-6646
f 29879.0 119 BRT 4
() 1 various parameter. If they are saying, for example, that 2 eventually we want to have high confidence that the 3 conductivity is less than some amount, and just looking at 4 what their proposing, would indicate that there is no way 5 that they are ever going to have high confidence in the 6 conductivity based on that sprinkling of tests, I think we 1
7 would come back and say we don't think that the program 8 that you are planning is going to get you there. What we 9 think you may need is some additional wells drilled and we l 10 think that this would help. But, in any case we think you i 11 need to go back and figure out better how you are going to i
- 12 determine that particular parameter. You are not going to 13 get what you want.
1 I
14 MS. BROOKSt Seth has given an example I think i
- 15 that expresses it. One of the things -- there are several
! 16 key things we would look for, and obviously the first one 17 is: Will it meet the performance objectives set by the
?
18 rule. And I think it was brought up earlier about the need 19 for consistency and coherence in the performance allocation i
1 20 and, as Seth was saying, we've got to make physical sense; I 21 and those are things key things I think we would be looking 22 for, and there are numerous other questions that fall out i
i 23 of that.
i 24 The other thing, DOE has agreed to provide a 2
25 rationale for the performance allocation and that, of ,
C)
ACE FEDERAL, REronTEns, INC.
202-347-370) Nationwide Coserage Nuk33MM6
29879.0 120 BRT
()
I course, is what we would be looking at very closely.
2 MR. KASTENBERG: I guess you have an obligation 3 to also provide a rationale to them when you say --
4 MS. BROOKS: Exactly.
5 MR. KASTENBERG: -- well, we think this is not 6 sufficient. You are going to have to tell them why you 7 think that. That's way was getting at, you are going to 8 have to develop your own rationale for why you think they 9 are right or wrong.
10 MR. COPLAN: That's correct. And we have not in 11 all instances prepared ourselves to do that yet. That's 12 what we are going to be working on. That's what Pauline 13 was alluding to.
14 MR. KASTENBERG: The five-year plan.
15 MR. COPLAN: I will be touching this afternoon 16 on how we intend to be developing technical review plans, 17 but I wasn't going to get to the level of detail that 18 Pauline was implying.
19 MR. MOELLER: Your comment, or Dr. Kastenberg's 20 comment of using the nuclear power plant approach for any 21 lessons to be learned, or benefits, it seems to me that's a 22 very good suggestion. Like with the power plants you had 23 the safety goals, you know, qualitative and so forth, and 24 then you went to a probability of core melt and so forth.
25 And then in the process of examining all of this a number O
Ace FnonRAi. RneonTEns, INC.
202 347 37(X) NationwiJe Coserage 800-336 4 46
. _ . ~ _ . _ _ . _ _ . . , . . _ . , . _ - , _ . _ _ _ _ _ _ . - - _ ,
29879.0 121
- BRT
() 1 of people finally came up and said: Well, we need also a 2 goal for containment failure.
3 Well, the same could be true here as we go 4 through the sequence and learn more about the barriers.
5 One of the parameters -- again, I don't know that this 6 makes too much sense -- but it might be you'll have to set 7 a limit on heat generation capability, you know, of 8 whatever is placed in the repository. So I think that's a
} 9 very good thought. :
i j 10 When did you say we would first start seeing --
1 11 I guess when Dee comes in we'll see the individual i
12 components and systems and the goals for each?
13 MR. COPLAN: We are not sure how far they have i
j 14 broken things down in the site characterization. We'll 15 know that, I think, a lot better by March 4th. But I 16 certainly my impression of the discussions that I have had l 17 just in formulating the agenda is that they have carried 1
i 18 things pretty far.
} 19 DR. SdEWMON: There's also a phrase called l 20 " defense in depth" which sounds a little bit better than 21 " pulling something else out of the hat."
22 (Laughter.)
! 23 It may mean the same thing.
l 24 MR. MOELLER: Jack?
25 MR. PERRY: Along the lines of availability, I 1
f ACE. FEDERAL. REPORTERS, INC.
202 347-3700 Nationwide tmerage 80)-336-M>46
29879.0 122
-BRT 1 understand DOE's most recent schedule is May for Hanford 2 and NNWSI, and next March for SALP. That's present.
3 schedules.
4 MR. MOELLER: What do you want from us on this?
5 MR. COPLAN: I think that was the main thing.
6 MR. MOELLER: Fine. Well, thank you very much, 7 for an interesting review, and I'm sure we've all learned a 8 lot. We'll now, then, take one hour'for lunch.
9 (Whereupon, at 12:03 p.m., the meeting was -
10 recessed, to reconvene at 1:00 p.m., this same day.')
11 12 13 14 15 16 17 l 18 l 19 f
! 20 t
21 22 23
! 24 1
4 25 O
4 ace FEDERAL REPORTERS, INC.
202 347-37(X) Nationwide Coserage 8(X).336-6M6
. _ - . . , _ , - - . . _ . _ _ _ , _ - . , - _ . - . _ . . , . . _ . , _ , . . . . _ _ . . __. _ _ . _ - _ _ . . - _ _ _ , _ . _ _ - ~ . - - _
29879.0 123 BRT
() 1 AFTERNOON SESSION (1:00 p.m.)
, 2 MR. MOELLER: The meeting will come to order.
1 3 The:next item , as we have mentioned on our meeting agenda, 4 is to discuss assessing the compliance with the EPA -- of 5 the repository with the EPA high-level waste standard, and
~
6 we have with us Bob Cranwell as well as William Kastenberg 7 and Seth Coplan. The main time, or the initial 8 presentation will be that of Dr. Kastenberg and I say 1
9 presentation just -- just informal discussion. You might l 10 -- well, you have -- each person has been given your i
- 11 handouts so if you'll lead off?
i 12 MR. KASTENBERG: To begin with, I was given a 4
13 set of questions to address, I believe four major questions.
O 14 But before getting into the questions and the way I 15 responded it is helpful to go back and look at the EPA 4
16 containment requirements, which are summarized in the NUREG
. 17 that I was asked to review, NUREG CR 4510, assessing 18 compliance with the EPA high-level waste standard and j 19 overview.
J 20 Basically the thrust of the document is to l 21 comply with the containment requirement which I have just 22 sketched out on this first sheet, and when you look at the 23 requirement, it is basically a probabilistic requirement.
24 It has a couple of parts to it which I think are worthwhile i
I 25 repeating: One, that the disposal system shall be designed 1
l ace. FEDERAL REPORTERS, INC.
! 202 347 3700 Nationwide Cmcrage 800-336 4 46
. ~ - - . . , _ . . _ _ _ . . , .. _ . - _ .m, _ ,, _ , - . _ _
29879.0 124 BRT
() I to provide " reasonable expectation," and we'll come back to 2 that " reasonable expectation" later; that the cumulative 3 release of high-level waste to the accessible environment 4 for 10,000 years from all.significant processes and events 5 shall be, one, less than one~ chance in 10 of exceeding the 6 quantities calculated in. table 1, less than one chance in 7 1000 of exceeding those quantities by a factor of 10, and 8 -- I left out the word " events" -- events less than one 9 chance in 10,000 shall not be considered.
10 MR. MOELLER: Let me ask on the first two 11 bullets, are those two bullets compatible in that you have 12 less than one chance in 10 of exceeding the values; then if 13 you skip to one chance in 1000, shouldn't it be 100 times 14 the value?
- 15 MR. KASTENBERG
- Not necessarily. If you wanted l 16 to do it linearly you could and, in fact, if you look down 17 two sheets you will see -- and pardon, this is the artist
! 18 at work, you will see that you can sketch the EPA limit as l 19 a series of steps. If you want to take the linear approach 20 that you just suggested you have a series of linear steps 21 there.
I 22 MR. MOELLERt Sure. It's not linear. Okay.
23 MR. MARK: But the notion of one chance in 10 to 24 the 5th of not happening and ignore it, that is in 25 violation of some of the other things we have been saying.
l
(:)
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmcrage HlX)-3364M
29879.0 125 BRT
) 1 MR. KASTENBERG: That may be. I'm just quoting 2 what is in the EPA document.
3 MR. PARKER: 10 to the 8th isn't it?
4 MR. KASTENBERG: There's a statement here that 5 says initiating events less than one chance in 10,000.
6 MR. MOELLER: 10 to the 4th.
7 MR. PARKER: In 10,000 years.
8 MR. MARK: Oh, it's 10 to the 9th per year.
9 okay.
10 MR. KASTENBERG: Then just to review on the 11 second page there are some excerpts from table 1.
12 Basically table 1 has a list of radionuclides and then for fs 13 each radionuclide you are given a release limit per 1000 14 megawatt-tons -- metric tons of heavy metal, or in curies.
15 I just chose four of them as an example because it shows 16 the range. The numbers range from 100 to 10,000. Then the 17 other radionuclides fall in between and then there's a 18 little formula one can use. Obviously in a repository site 19 you'd have a mixture of radionuclides and they'd give you a 20 way of summing when you have such a mixture.
21 MR. STEINDLER: What do you mean by "or in 22 curies." These numbers are in curies, aren't they?
23 MR. KASTENBERG: This is a ratio of release 24 limit per either curies or, if you wanted to do it, by 25 metric tons of heavy metal. So apparently they are giving O
ACE.Fl!DI!RAl. RiiPoRTliRs, INC.
202 347 37(N) Nationwide Cmcrape lux)-336416
29879.0 126 BRT (3
(,j 1 it to you in either one of two ways. You can do it on a 2 per weight or per curie basis, as I understand the document.
3 MR. STEINDLER: I don't think so.
4 MR. PARKER: It's curies per thousand metric 5 tons.
6 MR. STEINDLER: Heavy metal waste equivalent.
7 MR. KASTENBERG: Then I interpreted the table 8 wrong. It says " release limit per thousand metric tons of 9 heavy metal or other unit of waste, curies."
10 MR. STEINDLER: I think the release limit is --
11 as I say in the case of Ameridium, it's 100 curies per 12 quantity of waste equivalent to 1000 metric tons of heavy 13 metal discharged from the factor.
(n) 14 MR. KASTENBERG: You have to get them to fix 15 their table.
16 MR. MARK: If there's a difference of opinion 17 between Kastenberg and Steindler on how to read this lousy 18 report, that should be called attention to.
19 DR. SilEWMON: We'll strike " lousy" report.
20 MR. MOELLER: So noted. Then we agree those are 21 in curies per that quantity of waste.
22 MR. KASTENBERG You would interpret that as 23 curies per 1000 metric tons.
24 MR. DILLON: And that's in 10,000 years?
25 MR. KASTENBERG: This in release limits for O
Acn I71!DliRAI Ill!PORTl!RS, INC.
202 m noo Nationwide cmcran mo tw,46
29879.0 127 BRT
() I containment requirements, cumulative releases to the 2 accessible environment for 10,000 years after disposal.
3 So that is background. That is the compliance 4 criteria.
5 Now, the questions I was asked to address: Does 6 this report represent a satisfactory first step in 7 addressing this matter? In particular, is the methodology 8 as outlined reasonable and appropriate? Basically the 9 methodology, to review, contains four steps. The first 10 step shown on page 4 is scenario development and screening; 11 the second step is consequence assessment; the third step 12 is sensitivity and uncertainty analysis; and the fourth 13 step is regulatory compliance assessment.
14 When you consider both the Staff and in this 15 case Sandia National Laboratory, their contractor's 16 experience, with probabilistic risk assessment and related 17 issues and recognize that this compliance criteria is 18 specified in probabilistic terms, then I would say this is 19 a reasonable first step. If I were asked to do it with my 20 background I would approach it pretty much the same way and 21 I would be hard pressed at there point to say: No, they 22 are of f base. Let's do it a different way. So in that 23 sense I believe this is a good first step.
24 Is the methodology reasonable and appropriate?
25 I believe it is. Again, however, as I mentioned, I have Acti FliDI!RAL RiironTtins, INC.
202-W Mm Nanonakle Cmcrage Mm 33MM6
29879.0 128 BRT 1 some reservations on how it would be implemented.
2 To go on to the next questions just to get you 3 ready, when I look at those four steps through a 4 performance assessment I see a number of key issues. There 5 may be others but these are some that I see, and one that 6 the Staff recognizes, in terms of scenario development and 7 screening, are determining the probabilities. That is a 8 key issue and we'll come back to that later; determining 9 the magnitude of the uncertainty, that is a key issue in 10 bullet number 3.
11 One that they don't mention that I feel is going 12 to be important, and just in an informal discussion with
~
s 13 Seth before, I think the Staff is aware of this although gJ 14 it's not mentioned in the report, somewhere along the line 15 we are going to have to communicate the results of any of 16 these PRAs, and in particular with respect to regulatory 17 compliance. And I think some detail has to be paid, in the 18 end, of how you communicate the results of these analyses.
19 flow do you communicate the magnitude of the uncertainty so 20 that the Commissioners or whoever makes the final decision 21 understands what they have.
22 The issue that I think was mentioned this 23 morning very briefly, this question of what is reasonable 24 expectation that these will be mot, that's not addressed in 25 this report. Somewhere along the line the Staff will have O
ace FEDERAL. REponTuns INC.
202 347-37W Nationwide Cmcrage 800 3MM
29879.0 129 BRT gs
() I to address that.
2 Then, last but not least, pertaining to scenario 3 development and screening, there is an example in here and 4 it goes through and shows how you can screen out various 5 scenarios and, unless I misunderstood, there is no formal 6 method for doing this. Basically the way it stands now it 7 sounds like you are just relying on the judgment of the 8 people that are doing the assessment. If they were to 9 leave the Staff and someone else were to pick it up later, 10 there would be no way of duplicate -- going through the 11 same steps they did to arrive at the event. There doesn't 12 seem to be a formalism about that.
ys A 13 MR. CARBON: There doesn't seem to have been by 14 the Staff?
15 MR. KASTENBERG: The Staff would have to answer.
16 DR. REMICK: Last I heard it was a controversy 17 whether it was expectation or assurance.
18 MR. COPLAN: I starred an answer on that and I 19 noticed that Dan Fehringer has come back. He has been 20 involved in rulemaking to adopt the standard and 21 incorporate it in part 60. But, essentially, yes, there 22 has been some controversy over whether to adhere to the 23 language " reasonable assurance" --
in part 60, which is a 24 term that the Commission has a lot of experience with -- or 25 whether to adopt what is essentially a new term in the
(
ace.FliDERAL llEPORTiiRS, INC.
202 347 3700 Nationwide Cmerage N n3364M6
29879.0 130 BRT (f 1 Commission's experience but carries with it, we believe, ,
2 the same meaning as " reasonable assurance," as it is N.-
i 3 intended to be applied in part 60. In the proposed rule, 4 that went out we stuck with the language " reasonable 5 assurance." o 6 Dan, would you want to add anything to that?
7 DR. PEHRINGER: In the other rulemaking that we 8 have under way to adopt theEPAstandards,we'arthworking 9 in some text that tries to describe more clearly what we 10 mean by " reasonable assurance" and explain why we think it 11 means the same thing EPA meant with their term " reasonable 12 expectation."
n 13 It is a little bit different than the past ,
\ )
14 application of reasonable assurance in reactor licensing: ,,
15 We have different considerations, longer periods of time to ' .
16 be concerned about, and so on. And we have some text in ,
17 the Federal Register notice that we are working on through 18 that other rulemaking. Basically it's the same type of 19 information that I presented to you at your last meeting ,
20 where we talked about that rulemaking. Assurance means 21 different things, depending on the type of decision to be 22 made, the types of information at hand to base the decision, l
j 23 on, and so forth.
24 MR. MARKt Dade?
25 HR. MOELLER: Yes?
O Acti.FrintinAi. Rt!PonTiins, INC, 202 347 3700 Nationwide rmerage m n 316 (M 6
J 29579.0 131 BRT- ,
(j 'l MR. MARK: On the same point, " reasonable 2 assurance, reasonable expectation," and in this context, we 3 are talking about the possibility of there being a volcano jf ^'4 or there being-a new ice age or a new flooding from the sea.
5 You can approach that like saying: Well, there's an ice 6- age every 10,000 years and therefore the chance of an ice
'7 age each year is 10 to the minus 40, like next year, 10 to
, + ~
8 the minus 40. That's hallowed by the linear hypothesis for 9 radiation exposure, which is a lot of --
10 (Laughter.)
. 11 -- nonsense.
. r
! 12 Now, you could also approach it by saying, well, 13 there's.no chance for 5000 years. There's a threshold.
l
! ! 14 And then you could start a linear line from there with a
- 15 mean time of 10,000 years.
- 16 The arithmetic that you might do on those
- 17 different assumptions is, indeed, going to come out t
[ ,
18 differently. It should.
i 19 The same for volcanoes. There is no chance that i
i 20 next year at Los Alamos there will be a reactivation of the I 21 Via Grande volcano. But there is a chance in a million I. 22 years that something might happen there -- another, it will i
I, 23 happen somewhere else, of course.
! 24 So if you are going to use the linear hypothesis t i j 25 which makes the arithmetic dandy and makes sense Lo
. s
"' ace-FEDERAL. REl'ORTERS, INC.
202 347 37(M) NationwiJe Cmerage H(ub3364M6 i
29879.0 132 -
BRT f I- 1 unavailable, that's one thing. And the other is, if you 2 are going to specify your reasonable expectation on the 3 basis of: Well, no new ice ages for 5000 years, therefore 4 expectation zero; and after that who knows, you can draw it 5 yourself? Have you given thought to that? And, if you 6 have, or whether you have or not, something should be 7 specified in the way in which you approach this, as to how 8 you approach to do it.
9 MR. COPLAN: I guess the answer to the question 10 is: Yes, we have given thought to it.
11 MR. MARK: I was prepared for that, of course.
12 MR. COPLAN: We don't know an answer. And, you know, we are sensitive to the fact that that and questions J
-) 13 14 like it are, I think, going to be among the more d.*ficult 15 questions that are going to have to be addressed during a 16 licensing proceeding and, consequently, during the period l 17 of time between now and the time that proceeding starts.
, 18 One of the things that we have done, in an effort to start 19 to get a handle on some of those questions, is that we had 20 the contract with Sandia, I guess I should say as part of.
21 the contract we have with Sandia. They are preparing a 22 report actually in several stages, the first one of which 23 is approaching conclusion,-in which they are taking stock 24 of what methods exist for predicting consequences of some 25 of these events, estimating probabilities of some of these O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 133 BRT
() 1 various events, and pointing out to us where there is hope 2 to be able to improve the situation over the next several 3 years and, in other instances, what methods are available 4 to us now.
5 Would you want to elaborate on any of that?
6 MR. MARK: Well, a certain amount of the answer 7 will depend on the way in which you approach those things.
4 8 Now, to have a meteor impact is perhaps pure stochastic 9 process. Once in every X years a meteor hits you on the 10 head. That you can afford to draw as a linear curve. But 11 for ice ages; no.
12 MR. MOELLER: I guess you are saying, too, you
~ 13 could almost wait until you saw the ice age coming to draw 14 your curve, because you know it can't happen tomorrow or 15 the next year.
16 MR. MARK: Precisely, but it's in the 10,000 17 year estimates, and the question is, how is it in there?
18 Is it-in there as a no-threshold linear curve? It 19 shouldn't be.
20 DR. SHEWMON; Your point is not if there's an 21 ice age and it throws 1000 feet of ice across this
- 22 repository we need be concerned, but the question of how 23 they are handling things of this sort? Is that correct?
24 MR. MARK: That is it exactly.
25 MR. COPLAN: That's the way I was interpreting m
l (~ )
l l
ACE FEDERAL REPORTERS, INC.
i 202-347-3700 Nationwide Coserage 800-336-6 4 6
29879.0 134 BRT 1 the question.
2 MR. MOELLER: Of course the whole idea of 3 reasonable assurance or reasonable expectation was to make, 4 I believe, licensing simpler.
5 MR. STEINDLER: I don't think that's quite right.
6 I think make it possible.
7 MR. MOELLER: Make it possible is a better word.
8 MR. COPLAN: The term'is one that is intended to 9 cut through the idea that there is going to be a fair 10 amount of uncertainty, to put it mildly, around some of the 11 these kinds of events and to, well, recognize that 12 uncertainty, that it is going to exist and that kind of 13 assurance that one might expect to have in licensing a
- ~,]
14 reactor is not-necessarily going to be had for a repository,
- 15 and that incorporating that concept into part 60 we are 16 saying we don't think that kind of assurance is necessary 17 for a repository.
18 MR. MARK: Well, I think -- particular items 19 have crossed my mind. Sea flooding to reproduce lake, l 20 whatever it was, that covered all at Minnesota, for 21 instance, needn't be in in a linear fashion. It shouldn't i
22 be in. Ice age shouldn't be in in a linear fashion.
i 23 Volcanoes shouldn't be in in a linear fashion, because you
. 24 can anticipate volcanoes for several thousand years.
25 Meteorites, put them in linear, if you want, or leave them i
i l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
29879.0 135 BRT f 1 out if you want. It won't matter.
2 How you handle those things surely has an 3 influence on your final grinding out of probability.
4 DR. FEHRINGER: Dan Fehringer again. I would 5 like to respond to that in one way.
6 Where events do not have a constant probability 7 of occurrence per year it is possible to treat them in a 8 fairly simple way nevertheless by-defining.several 9 subcategories of that type of events.
j 10 For example, one scenario is an ice age occurs 11 within the first 100 years after depository closure. The 12 probability is zero.
-3 13 Another event is an ice age occurs between the 1
' Q 14 year 100 and the year 1000. Possibly a moderately higher 15 probability; then between the year 1000 and 5000, the 16 likelihood is somewhat greater; and the. year 5000 to 10,000.
17 By dividing up the descriptions of these types of events 18 one can take into account the changing likelihood and still 19 not make the analysis overly complicated.
20 MR. MARK: It isn't a change in likelihood so 21 much as a change in uncertainty. What you are saying would 22 not be offensive to me.
23 DR. FEHRINGER: Right. Right. Okay.
24 MR. MOELLER: Where are we?
l 25 MR. KASTENBERG: Question number 2 --
l-
/\CE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
29879.0 136 BRT-gs 1 MR. MARK: Excuse me, Bill.
t._)
2 MR. KASTENBERG: No, that's all right. That's 3 good.
4 Question number 2 is, have the proper questions 5 been addressed? Are there significant omissions in the 6 report?
7 I guess my feeling is that I think the Staff has 8 raised the proper questions but I'm not convinced that all 9 of them have been addressed. In particular I refer to 10 probabilities and uncertainty because I think that's really 11 -- as Carson mentioned and others, that's really a key.
l 12 When you are dealing with probabilities of the 13 kind we are talking about that are rare events that are 14 expected to occur or not occur over 10,000 years, basically 15 we do not have actuarial data on a lot of these, so you 16 either have to report to probabilistic or deterministic 17 models, and more than likely they will rely on expert 18 opinion.
19 Just to give you a flavor, I looked at a i
20 companion report done at Sandia. There's a table in the
- 21 companion report which you should have on the next page, t
22 and it shows you for a lot of the physical phenomena, the 23 natural phenomena that are important -- first, how the 24 probabilities might be obtained and then whether they feel i 25 they can obtain them with some certainty or accurately or
[
l l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
29879.0 137 BRT
) 1 only with limited confidence.
2 At first blush when you look at this it's a 3 little bit of everything. The probabilities in terms of 4 confidence spread all over the place, and in terms of how 5 these probabilities would be obtained, they are going to be 6 obtained by basically every which way the Staff and the DOE
- 7 are going to be able to obtain them.. But what is 8 particularly interesting is that in many of the instances 9 it is going to be expert opinion. It's going to be 1
10 subjective estimates of probability.
11 This is an area that I think the Staff has
. 12 raised the issue. But when you read-the report -- at least 13 this represents their thinking -- they are not --
it looks 14 like they are not ready to state how they are going to do 15 this. In other words, how are they going to use expert l 16 opinion? And how are some of these subjective estimates to 17 be used to establish what the probabilities are and the 18 uncertainties and so on. I think this is an area where the 19 Staff has to really do some homework.
20 So that's kind of important.
21 The second point or part of that I have on the 22 next page, marked page 6, is on uncertainty itself, and the 23 Staff, again, raises the right issues. The three issues
- 24 are, with respect to scenario uncertainty, the issue of 25 completeness; with respect to model uncertainty the fact
(
l
/\CE FEDERAL REPORTERS, INC.
1 202-347-3700 Nationwide Coserage 800-336-6646
- _ , - - . , _ . ___ ._ . . ~ . _ . . . . _ . . , - _ . . . _ - _ . .. . _ , _ . -, ,. - . - _ _ . ,
u 29879.0 - 138 t BRT-1 that there is a lack of knowledge of physical processes, ,
2 and so we don't really know what some of the models may;be,
! 3 and then parameter uncertainty, given the models, what is 4 the data that goes in there? f 5 I made a rather facetious comment in~my answer 6 to question number 2. I hope the Staff doesn't take this-7 too much to heart because they are not alone, that they 8 have a very nice description in there on parameter 9 uncertainty, and all the various methods that you can use.
10 But they never really say which one they like, and I said, 11 well, this is a nice introduction for a-journal article or 12 MS or PhD thesis for somebody. But somewhere along the 13 line you know you have to come down and say these are the
.O' 14 methods we are going to use and so on.
15 The issue of model uncertainty, I think, is
- 16 going to be a real tough one, and I think up until recently
! 17 people really didn't have an appreciation for the effect of 18 model uncertainty.
19 You can do all of the sensitivity analyses in
- 20 the world and if you have a phenomena that changes on you 21 and you don't have the right model it ain't worth a hill of i
l 22 beans. The example I always give the students is a 23 transition from laminar to turbulent flow in fluid dynamics.
24 You can do all the sensitivity analyses you want, but if 25 you don't model the transition as something entirely ;
-(:)
e ACE FEDERAL REPORTERS, INC.
i 202 347-3700 Nationwide Coserage 800-336-6646
29879.0 139 BRT i
O q_j 1 different, it doesn't mean anything. So I think you 2 recognize that but you don't really say how you are going 3 to tackle it, what you are going to do about it. So that's 4 a concern that I have.
5 Then, last but not'least, the two questions that 6 I was asked: Could the approach suggested by applied to a 7 real-world situation? And, does the report present any 8 basis for assurance that an actual proposed site can be 9 shown to meet the standard?
. 10 I guess with respect'to the first one I tend to 11 be conservative myself, and I would be reluctant at this 12 point to say: Yes, that you could use it for a real-world 13 situation. I just don't think we have the data and the 14 models to do that. I think that there needs to be some 15 homework done, as I mentioned, in terms of model 16 uncertainty, data uncertainty, how expert opinion would be 17 used to generate data and so on. In fact, one of the 18 things -- well, we'll get to that.
19 Then the last question, I said: Does the report 20 present any basis for assurance that an actual proposed 21 site can be shown to meet the standard? And corresponding 22 NRC requirements? Again, I feel it's a little premature to 23 come out yes on it.
24 MR. STEINDLER: Are you saying you can 25 distinguish amongst sites?
O ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33MI>t6
29879.0 140 BRT
( l MR. KASTENBERG: At this point?
2 MR. STEINDLER: Yes.
3 MR. KASTENBERG: Again, not based on the data.
4 You might be able to do some screening, prioritize, but I 5 don't know that you would be'able to say that, yes, these 6 meet the standard and these do not.
7 MR. STEINDLER: But the response from two 8 separate sites is not going to be sufficiently close you 9 can't tell them apart?
10 MR. KASTENBERG: Right.
11 Now, I want to comment on the last thing I 12 jotted down. I was just a little troubled by it and, again, 13 it's my own conservative nature. There's a nice example in 14 the report and it shows a little bit of your thinking of 15 how you would apply this. I have this problem, when in a 16- report like this you start to use numbers -- okay? I have 17 just seen too many times where you as a staff or others 18 write a seminal report and you pull some numbers out as 19 examples and little by little they become part of the 20 folklore.
21 Then five years from now people start 22 referencing your example and that's the number they use, 23 TID-14,844. You go back -- it was an example of how you 24 would do a site analysis -- and all of a sudden those 25 release fractions show up, everybody uses them. In O
ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
_ . . _ _ _ _ _ . _ = . . _ _ _ _ . . - . . . . - - _ . _ . . _ _ . . . _ , _ , - , _ . , _ -
29879.0 141 BRT
() 1 WASH-1400 they used certain release fraction, evacuation 2 radii and so on, and you asked Norm Rasmussen why you did 3 it and he said, well, they seemed like reasonable numbers 4 at the time, and all of a sudden you have a licensing issue 5 today and in the computer code are the same numbers that he 6 used because he felt they were reasonable numbers to use.
7 So that is just my own conservative nature, that you have a 8 nice example, ,you have used some numbers and you state up 9 front these are only examples, but nonetheless I worry that 10 they will creep in.
11 What I would like to see, and I_have a little 12 table here, when you look at your numbers they seem to fall 13 into three categories. They are either numbers on the 14 probabilities on the order of one to one'in 100; numbers on 1
15 the order of one to 10 to the minus 4, and 10 to the minus 16 7 or 10 to the minus 8. If I were doing my example I would 17 use that these events are the kind that might happen; will 18 happen; these are likely to happen, over 10,000 years, and 19 these are very likely not to happen over 10,000 years; and
( 20 I would construct my examples along those lines and work 21 like that. Again, that's just a personal feeling that I 22 have, in a seminal work starting to put numbers out that 23 people start to believe in, after all.
24 MR. MARK: I'm very much with you, Bill.
25 Because you can so easily multiply something into 10 to the l
O I
j /\CE FEDERAL REPORTERS, INC.
! 202-347-3700 Nationwide Coserage 800-336-6646
29879.0 142 BRT
/
(m),
1 minus 4, but you can't multiply it into "likely to happen."
2 MR. KASTENBERG: That's why I like that, at 3 least at this stage of the game.
4 Then I also felt, while I was just preparing for 5 today to think about and summarizing what are the needs --
6 my own view of what some of the needs are, and again this 7 is not meant to be all-inclusive but things that came to me, 8 is that I think the Staff and the DOE and others have to 9 agree on the methods they are going to use for expert 10 opinion. I can't see that the DOE is going to go their way 11 and you are going to go your way and in the end you are 12 going to end up in a spitting contest, your experts versus s 13 theirs. I think you have to establish some ground rules U 14 right at the beginning.
15 Similarly with respect to the data base, even 16 though you don't have a complete data base, even though it 17 is uncertain and so on, I have a better feeling after what 18 I heard this morning that you are working with DOE, but you 19 need to establish at least an agreed-upon data base to 20 begin to do some calculations that, even though you may not 21 like it and there are uncertainties, that you both agree 22 that's where you want to start.
23 I think on this formalism for the screening, you 24 know, you have an example. You show how you did it for 25 that example but I would like to see you develop a
/m ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6646
e
-29879.0 143 BRT
) formalism for how you are going to do the screening so that
, 1 2 if you guys get promoted to other jobs an other. people come 1
i 3 in they know how to go about doing it and others on the 4 outside can repeat what you did to arrive at your results.
5 I think this quantification of uncertainty, we talked about 6 that a little bit this morning. It applies to this as well.
7 How do you propagate and aggregate uncertainty?
8 It's very, very important especially this morning you are 9 talking about reallocating goals -- reallocating criteria, 10 and-each has its own uncertainty, and you can fall into a 11 lot of traps, as we saw in the example this morning.
- 12 So I think you have to work on that a little bit.
13 Then, last but not least, I feel that the public 14 will be involved in this, others will be involved, and you
- 15 should start thinking at an early stage of how you are i
16 going to represent and communicate the risk. Are you going 17 to be giving point estimates, perhaps? You might give 18 confidence levels. I think right from the beginning you 19 should think about how this risk is going to be 20 communicated and represented to the Commission, to the 21 public, to the Congress, whoever. It is not an easy thing, 22 and I would like to see it happen right off the top.
23 MR. MOELLER: Thank you much. Do you have 24 comments? Why don't you go ahead, then.
25 MR. CRANWELL: Okay. First I would like to l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 144 BRT
(- I thank Dr. Kastenberg for his review of this report.
(_) I 2 think this type of review is really critical in this early 3 stages of this type of analysis.
4 I wrote some comments down here. Also I hope 5 the reference, not by you but by somebody, as it being a 6 lousy report, is not taken in a literal sense. Because I 7 would take objection to that. I don't think it's a lousy 8 report at all. I hope you didn't mean it that way.
9 MR. MARK: Not entirely. -
10 MR. CRANWELL: First of all we were asked to 11 keep the report fairly short when we wrote it. This work 12 actually represents about a decade of development work, 73 13 commuter code work and so forth. But we were asked to kind V 14 of keep it very concise and very short in this particular 15 report, concentrating on showing in a concise way how you 16 might go about demonstrating compliance with the EPA 17 standard as it stands now. So we didn't put a lot of 18 detail in there.
19 We do have accompanying documents and reports 20 that go into great detail on our sensitivity and 21 uncertainty analysis techniques, what we recommend, the 22 advantages and disadvantages of the others. Why we don't 23 recommend those.
24 We didn't put a lot of that in here. Maybe we 25 should have, but again we were asked to keep it rather ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
29879.0 145 BRT p
(_) 1 short.
2 In terms of the numbers that we quoted for the 3 probabilities, again I would agree with you that 4 qualitative description is nice but that doesn't work when 5 the EPA suggests that you generate a CCDF to show 6 compliance with a standard. To generate a CCDP you have to 7 put numbers, probabilistic values for those scenarios or 8 for those events in process to generate that CCDF in the 9 way that is recommended in here. So, to demonstrate that 10 generation of that, we had to put numbers down. We 11 couldn't use qualitative descriptions of those. That was 12 the reason.
rm s 13 But I agree with what you are saying. I don't N] 14 care how much you caveat that, there's a tendency later on 15 to use those numbers, but that's why we did that in here.
16 Okay?
17 Let's see.
18 MR. MARK: You are speaking in effect of two 19 reports, one which covers the field and the other which l
l 20 stipulates the way to act. Your numbers don't have to be
\
l 21 in the first. They do have to be in the second.
22 MR. CRANWELL: That's right. I would agree.
23 MR. MARK: And that first one should possibly be 24 free of such assumptions and use terms of the sort that l 25 Bill was referring to. Where the second one says, well, I
(s
~
l l
l l /\CE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
29879.0 146~ .
BRT
( l for this we'll take 10 to the minus 6 to 10 to the minus 8..
2 That would, a'little bit, free the concern Bill raised and 3 you raised. If the first is for the general statement.of 4 the approach and the second is'for the implementation of 5 that, the details could be in the second and recognized as 6 hypothetical and tentative and could be changed. Whereas 7 the intention would be in the first. That, presumably, 8 does not change. Then that first should be as scrutable as 9 it can be made. The second could be as detailed as is felt.
10 necessary.
11 MR. CRANWELL: I think the earlier drafts had 12 qualitative -- likely -- reasonable -- I don't know. They 13 had some qualitative description to them.
O 14 In terms of the scenario development or 15 selection, there is a methodology or formalism available 16 and it is, I guess -- do you have a copy? It is reference --
17 it is NUREG 1667, where we have developed a methodology for 18 selecting and screening scenarios -- by Cranwell, Gasowski,
) 19 and Mortis, 1982-A, in the references -- where we show a i 20 systematic procedure for developing scen'arios which 21 includes the selection, the screening, the combining of
- 22 scenarios or events in process as to form scenarios and so l
23 forth.
i 24 So we feel like there is a methodology available i
i 25 and would even encourage you to look at it and see what you i
I /\CEJFEDERAL REPORTERS, INC.
( 202-347-3700 Nationwide Coserage 800-336-6M6 i
, - - - - - - . - - - . , , , , . , - . ..-------,..n,--,, -,.r, - ,--..---- - , , , ,--, - -- , - - - - - - -..~.
v . - -
'29879.0 147 BRT gw
(_) I think of that. I could send you a copy of the report, if 2 you would like.
3 MR. STEINDLER: Do you expect another-4 methodology to come up with a different set of answers?
5 MR. CRANWELL: Probably. But --
6 MR. STEINDLER: If you do, just to carry it on 7 through to tell you where I'm going, if you do, then I 8 would guess that both the applicant and the NRC would be 9 asked to defend their unique methodologies in front of the 10 Yakima Indians, who are going to say to you that isn't the 11 right methodology, ours is, and we get a different answer.
12 How do you resolve an issue like that other than j -
13 by rulemaking early on, which is a legal nicety to take 14 care of technical' uncertainties?
15 MR. COPLAN: Maybe there are some differences in 16 understanding of the word " methodology." But I guess I 17 would have given a different answer than Bob.
18 MR. STEINDLER: Okay.
19 MR. COPLAN: I think you'd get, with different 20 methodologies, the same scenarios, assuming that you were 21 using the same criteria to screen with and you were using 22 the same inputs in terms of events, transport mechanisms.
23 I think from there on it becomes pretty much --
24 MR. STEINDLER: Then you say the methodologies 25 don't make a tremendous difference. But I do get two ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-33M486
4 29879.0 148 4
BRT 1
() 1 different answers.
I 2 MR. CRANWELL: I would agree. I was thinking 3 maybe a totally different methodology where the-screening, 4 criteria for screening scenarios -- 1 5 MR. STEINDLER: I was, too. What I was.looking 6 for: Is it reasonable where some legitimate activity
- 7. generates the conclusion that your selection process is 8 knots the only way to go? Mine is at least as good as 9 yours? And now you are in the process of having to defend
! 10 a brand-new set of scenarios that you either hadn't thought c 11 about or rejected earlier? And you are going to be asked 12 to justify in some fashion or another that yours is a valid es 13 method of looking at it or in fact is the only valid method
' Q 14 of looking at it?
i 15 MR. MOELLER: But they also stated, or don't we 16 agree that if they have a scenario that totally bounds ,
17 yours --
18 MR. STEINDLER: Yes. -
l
! 19 MR. MOELLER: Then they are okay. It's only if I
20 you come up with one that is outside of their bounds.
21 DR. SHEWMON: That's going to be very hard to do f
- 22 with -- I don't want to say the complete lack of data but 23 the uncertainties that are going to be here. At least it I 24 make the reactor business look nice and certain because at i
I 25 least there we have a certain amount of operating i
i ACEJFEDERAL REPORTERS, INC.
j 202-347-3700 Nationwide Coverage 800-336-6646 l _ .
. _ , _ _ _ _ _ _ , . , _ . , , . _ _ , , _ , . , ~ - - . - , _ _ . _ _ . . - _ _ . _ _ . . _ -,
b 29879.0 149 BRT
{%
() 1 experience over a few thousand reactor / years.
2 MR. MOELLER: In the standard review plan which 3 we are going to look at tomorrow, of course you know this, 4 they have a routine release section and an accidental 5 release section. And the routine release section gives, 6 you know, eight, 10, 12 scenarios. Then the accident 7 doesn't give anything. Well, it says: Refer back to the 8 routine release scenarios.
9 I wondered why, in the standard review plan, you 10 didn't give us some specific accident scenarios?
11 MR. COPLAN: I think that you must be' thinking 12 of low-level -- you threw me there for a minute.
13 MR. MOELLER: Okay. You are right. That's 14 low-level waste. Okay.
15 I mixed them up for the moment.
16 MR. CRANWELL: I think because of the -- and I'm 1
17 not trying to build up this methodology we have for 18 scenarios, but I think it would be difficult to come up, 19 because it is very comprehensive and we have tried to look, 20 whether we developed that scenario selection and screen 21 criteria technique, we have tried to look at what little 22 had been done before that and incorporate that into this 23 methodology also.
24 So it seems to be comprehensive enough. I 25 really think you would find it difficult or someone would O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
29879.0 150 BRT 1 find it difficult to come up with a different -- and in 2 terms of that or in terms of wheth'er this thing is 1
3 applicable or not, I might point out the whole methodology
. 4 in here is being applied to the WIPP site right now and the i 5 scenario selection and screening methodology is part of 6- that being applied to WIPP. It's been applied to WIPP in 7 the past and, in a report by Bingham & Barr in 1982, it was 8 applied to NNWSI in Hunter and others, in 1981 and '82, and 9 is currently being applied to the Nevada test site by Bob 10 Gasowski, who is in the audience, to select scenarios.for 11 those particular sites.
12 So it is being scrutinized and applied to 13 specific sites right now.
14 MR. STEINDLER: Has that same methodology been 15 applied to, I don't know, aquifer selection or some process 16 where some currently industrialized process, where you have 17 got actual experience to see whether or not it fits?
18 MR. CRANWELL: I might say some of the 19 components of the methodology have. Not the overall 20 methodology but certain components of it; yes. They have i 21 been applied in that area and in the petroleum industry 22 also.
23 MR. MOELLER: Will NRC ultimately agree with DOE i
24 on a set series of scenarios? You'll use the standard l 25 methods to, or systematic methods to develop them and O
l ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
{
29879.0 151 BRT 4
. ()
1 you'll finally, then, agree on five or 10 that you will 2 apply?
i 3 MR. COPLAN: I would hope that we'd reach that i 4 point. Short of that, at least agree on a methodology.
5 MR. MARK: I can concur in what Dade is 6 suggesting and you meant, you would hope to.
7 You were talking'to actual people, however, in l 8 DOE. DOE is not that amorphous thing of which you might 9 expect agreement or not. From the people you talked to do j 10 you expect to?
11 MR. COPLAN: I think we will. I think we will.
12 The reason that I was. thinking that it is more
-13 likely that we could reach agreement in terms of 14 methodology than specific sets of scenarios, at least in a 15 near time, is that I think it is more likely that we could
! 16 reach agreement on criteria for screening scenarios than 17 actual values of probabilities for scenarios or potential 18 consequences of scenarios.
19 MR. MARK: I think that's all you can say. I
! 20 think you and I could agree on something or other and then 21 we'd get different answers.
)
4 22 MR. COPLAN: Yes.
23 MR. CRANWELL: I'm aware of a few situations i 24 where DOE is currently using this scenario methodology in-i 25 their analysis. So it is being applied there also.
- C:)
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 152 BRT
<x-()_
1 I think those were the main comments that I 2 wanted to speak to you. Again I really appreciate your 3 review of this document because I think this is important.
4 MR. KASTENBERG: I wanted to just come back to 5 the uncertainty. You mentioned a companion document where 6 you discussed various methods for treating uncertainty, 7 which ones and which not -- I would like to get a copy of 8 that also.
9 But I'm just curious, in this brief document you 10 mentioned four or five methods that can be used for 11 parameter uncertainty.
12 MR. CRANWELL: Right.
~
,3
, 13 MR. KASTENBERG: You don't say anything about 14 model uncertainty. I know that's a tough one. In fact we 15 have a PhD student working on a thesis in that area right 16 now.
17 Do you have anything written on that?
18 MR. CRANWELL: Every computer code model that we 19 developed for the NRC for this type of work goes through --
20 and we publish documents on it -- goes through a 21 verification process, a comparison with field data. I will 22 not call it a validation because we have some strong 23 feelings on ever being able to fully validate some of these 24 computer codes in what I think is validation, so we call 25 them verification and fuel comparisons or benchmarking, if G
L)
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
. - =. - - - . - - , . , . . . . . .
29879.0 153 ,
BRT C'1 1 you want, with other codes -- comparison with field data.
(,j 2 And they also go through a software quality assurance 3 program that requires all these kinds of analyses. If 4 that's an issue that -- and it is avoided a lot because 5 it's an area of uncertainty you can't quantify, you just 6 can't quantify. So it is avoided a lot because of the fact 7 that the EPA wants -- it's a risk-based standard and I 8 don't know how you would fold that, since you can't 9 quantify it, how you fold it into a CCDF. You can't, 10 really. And it's something we are working with the NRC on 11 now and maybe publishing some kind of a document that says:
12 Okay, DOE, once you have gone this far toward " validating a 13 code," that's good enough then.
14 But it is really a big issue and it's something 15 we are working on, but that's as far as I can tell you 16 right now.
17 MR. MOELLER: Paul?
18 DR. SHEWMON: I'm trying to get a handle --
19 pardon me if I use this, I know a little bit about it.
20 There is questions of kind of phenomena, and his 21 turbulent / laminar flow criteria is one. The other type of
- 22 thing is, given a complex phenomena that you understand 23 roughly, how do you set up a set of constituent equations 24 or whatever, that will approximate it?
25 Do you feel that in most areas you understand ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage M)0-336-6646
29879.0 154 BRT
.(/~S) I the first and your problems are of the latter sort? Or 2 what? And let's leave out whether the software does what 3 you want it to do.
4 MR. CRANWELL: There are still problems in that 5 area. I think the laminar flow -- there's a big concern 6 about whether that's realistic or not. There's a concern 7 with Darcy's law is appropriate for modeling groundwater 8 flow at a lot of these sites. But again hopefully a lot of 9 these questions can be resolved in calibration exercises, 10 where you take the model embodied in a computer code, you 11 take site data and you see whether you can match with that 12 code what's going on -- match with that code what you are 13 observing going on at the site. But I still hate to say 14 that that blesses the code in terms of: All right, let's 15 go ahead and do simulations for 10,000 years.
16 DR. SHEWMON: You know as well as I do, one may 17 fit the data by taking six of your variables and evaluate 18 them or maybe by deciding that you do or don't understand 19 the physics of what you should be using for the equations.
20 MR. CRANWELL: That's right. And that's where a 21 person doing that kind of fitting, you have to watch over
- 22 that very carefully because you can play games with a code.
23 A person who knows a code can play all kinds of games with
- 24 it and force it to fit what is going on at a site. So you
- 25 have to be very careful with that. I agree.
I i
l ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-334 6646 i
- - ~. . _ _ . -. - . . - - . -- .
= - ~ --
29879.0 155 BRT
)- 1 MR. KASTENBERG: I want to ask you a question on 2 another subject.
i' 3_ MR. CRANWELL: Okay.
} 4 MR. KASTENBERG: One concern we have in doing i
i 5 reactor PRAs is is-that you have an instantaneous picture 6 of the plant and then you base all of your initiating 7 events and sequences on that picture of the plant. Then if 8 a year from now you have a severe-transient, still leaving 9 the plant intact, you have some degradation, maybe in the I
10 vessel because you had overcooling or whatever; and we have 4
11 really not factored that kind of thing in.
12 When Carson raised this question before, the s
fx 13 fact occurred to me that over a 10,000-year period you are
'.Q 14 apt to have certain geological or climatic events that may i
- 15 keep your repository intact, operating the way it should, 16 and yet it's in a different condition. Now if you have j 17 your earthquake or whatever, you have a whole new set ~- is -
18 that factored into your screening and thinking? I didn't
)
I 19 catch that.
20 MR. CRANWELL: That is not pointed out in here.
l 21 We tried to account for that by hypothesizing what I would 22 call different conceptual modelings of what the site looks 23 like. The way we look at our analyses is we have a -
i 24 conceptual model, a picture of what the site is like when 25 you place waste there. Then you perturbate that to account I
/\CE FEDERAL REPORTERS, INC.
1 202-347-3700 Nationwide Coserage 804336-6M6
- _ - , ~ . _ _ . . . . . , , _ - , . _ _ , . _ . _ , _ . __ . . _ _ _ _ _ . _ , _ _ _ . _ . _ _ - . _ . . , . _ . . - . - - , - - _ -
29879.0 156 BRT p)
(, I for these scenarios. Okay? You take perturbations of that, 2 impose these scenarios on that by taking perturbations of 3 that conceptualization of what you think the site is like, 4 but to account for the uncertainty in that, we vary those, 5 the conceptualization of the site.
6 Maybe the layering is this rather than what we 7 think it'is. Maybe the groundwater, the water budget is 8 something else rather than what we think it is and we do 9 those uncertainty analyses to see what the impact is on 10 what we get out the end. I don't know if that's.what you 11 are talking about --
12 MR. KASTENBERG: Yes. And the sort of idea that 13 for the first 5000 years what you had assumed, the way you 14 have characterized the site may be correct, but somewhere 15 in the 5000th year you had a big earthquake and 1000 years 16 later-you had some severe floodings and so on and now for i
17 the last 4000 years of the life of the site it is a i
18 different site than it was. That's what I mean.
l I
19 MR. CRANWELL: We account for that. We take i, 20 perturbations of this thing at different points in time.
21 MR. KASTENBERG: Okty. That's the point.
22 DR. SHEWMON: Let me bring up a variance, mildly 23 relevant.
24 Looking through an issue of Science several 25 years ago, there was a short word by a geologist, somebody i
()
l f /\CEJFEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 l
i
~ . - _ . _. . .
29879.0 157 BRT
(-%, .
g) 1 who had been spelunking scientifically down in Kentucky, 2 let's say. They had been underground in these limestone 3 caves one day. It started to flood. The waters rose, 4 torrential rain on the surface, they got the heck out of 5 there.
6' When they could come back in a couple of weeks 7 they found that all that water flowing through, they had 8 really chewed the hell out of things and their basic 4
9 conclusion was when people average over 100 years they try 10 to take models that will give a uniform erosion rate of 11 this. What they had found was that in a week things had 12 receded a heck of a lot when you have this torrential rain.
13 You come back, how often do they have a O 14 torrential rain in Nevada? Maybe once a century, I don't i.
15 know.
4 16 MR. CRANWELL: Again, we try to account for 17 those changes. Maybe not in a few weeks, but similar to 18 what Dan Fehringer said earlier, we do what we call a 19 multiple steady state analysis where we break the states of 20 nature up into these little epochs of time and change the I
21 system as we go through it. We don't treat it as steady l
22 state over 10,000 years, but as a multiple steady state
. 23 system to try to see if it gives us those effects.
24 MR. MOELLER: Other questions or comments?
, 25 MR. COPLAN: Just a couple. Bob, at the 1
I l
i /\CE FEDERAL REPORTERS, lNC.
202-347-3700 Nationwide Coserage 800-336 4 46 i __
t
/
29879.0- '158 ,
BRT '~*
x-() I beginning of his comments, tried to put the report into .
2 some context and I would like to elaborate on that since I .
3 was responsible for asking for it. At the time we asked
- 4 Bob to prepare the report, it wasn't too long after the EPA.
5 standard had become final and an awful lot of people were 6 having a very hard time just grasping what the standard was 7 about. It is a probabilistic standard, talks in terms of 8 CCDFs and all kinds of things that many people that are 9 involved in repository programs f rom DOE's end, our end, -
10 affected states and tribes, didn't have the vaguest idea '
A:
4 11 what was being talked about.
12 MR. MOELLER: You might include the ACRS in that.
- - 13 (Laughter.)
( 14 MR. COPLAN: We asked Bob to put together a 15 simple 10-page example, 10 pages, that would give people an i
i 16 opportunity to get some idea of what these concepts were i
17 and that's really what this report represents. At this i
l 18 point I'm happy that he resisted the 10-page idea and we I
19 ended up with several times that, but that was really the 20 intent of the report.
21 The second point that I wanted to make is that 22 we, I think, are pretty much in agreement with 23 Dr. Kastenberg as to the potential soft spots in being able 24 to license a repository.
25 One of the activities that, on starting, that1we i
ACE FEDERAL REeoRTERs, INC.
- 202-347-3700 Nationwide Coverage 800-336-6M6
.;[ ,
s 29879.0 159 BRT l 1 touched on very briefly in the last Subcommittee meeting is 2 an effort to try to develop and reach agreement with DOE 3 and affected states and tribes, on methodologies for
,4 dealing with the uncertainty that surrounds the 5 implementation of the standard. Specifically, we are 6 hoping to be able to reach some sort of resolution on at 7 least a methodology, for identifying scenarios, and, you 8 know, I don't know at this point -- I can't say how far we 9 could really go in terms of, perhaps, going to a rulemaking
, 10 on such a methodology but it is something that we could 11 consider. With respect to model validation, again, we see 12 that as one of the fundamental problems that's going to be 13 involved in licensing. And there, what we would hope to be 14 able to accomplish is to establish some sort of criteria 15 that could be used to determine the goodness of a j 16 validation program and finally, as was pointed out during
- 17 this discussion, there are methodologies available for
- 18 doing data and parameter uncertainty analysis and we would
[- 19 like to, maybe, be able to focus on one or several that 20 could be considered to be good ones to use as part of the 21 licensing process.
l 22 MR. MOELLER: Yes. Carson?
! 23 MR. MARK: Are we about through with this 24 session?
25 MR. MOELLER: Yes.
ACE FEDERAL REPORTERS, INC.
i 202-347-3700 Nationwide Coserage 800-336-M46
29879.0 160 BRT r~m
() 1 MR. MARK: A're our witnesses also to be 2 discharged?
3 MR. MOELLER: I think they are nearing the end.
4 MR. MARK: I would like to make my own comment 5 that I believe we are very grateful for the people who have 6 come to talk these things over and to Professor Kastenberg.
7 MR. MOELLER: Well, they have been very helpful 8 to us, particularly in an area where we do not have the 9 people who could really dig into it on our own. So it has 10 been an education.
11 Okay. That finishes that up. I had my 12 appreciation on behalf of the subcommittee in Carson Mark's 13 remarks, and I think looking at the schedule, perhaps we
~
14 could take 10 minutes and that will help us get back on 15 schedule.
16 (Recess.)
17 MR. MOELLER: The meeting will resume. The next 18 item is a discussion of the high-level waste licensing 19 support system and this will be, I gather, a slide 20 presentation as well as stepping across the hall and seeing 21 an actual demonstration of the system. Our speaker is 22 Mr. A. Bender, from NMSS. Welcome.
23 MR. BENDER: Thank you. The name is Avie Bender, 24 I'm a project manager with the Division of Waste Management,
.s 25 NMSS.
l l
l ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646
4 29879.0 161 BRT 1 My discussion is going to deal not so much about 2 uncertainty or a forthcoming avalanche -- maybe that of 3 paper -- but it is something that I think is of significant
. 4 importance to the licensing process, specifically'with 5 trying to streamline that process.
6 I think it has become pretty evident, based on 7 NRC's experience, that one of the major reasons for delays 8 in licensing-proceedings has been the amount of time 9 required to send and receive information. It is.a very 10 lengthy process where one has to look for documents at the 11 last hour.
12 We tried to convey this information to the 13 Department of Energy starting in-about 1984 and at that 14 time we had some trouble convincing them there was an 15 actual need to develop such a system. It's pretty clear 16 there's no way for us to review this license in a 17 three-year period if there is no record system in place.
18 In order to convince the DOE of that we initiated back in 19 '85 a pilot project to develop a prototype of the system to 20 let them understand the kind of capabilities we were 21 looking to get from this system.
22 We had a series of meetings with DOE, the states 23 and the tribes, in which we tried to define some of their
=
24 requirements. We initiated a pilot system of our own, 25 demonstrated it over a year and a half and DOE was finally l
4 ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-334M46
.- - m- ..
29879.0 162' BRT I convinced in January of 1986 to develop such a system. We 2 signed an agreement to develop an interagency-coordinating 3 committee to serve as a forum for public participation and 4 allow the states and tribes to raise their concerns as to' 5 what they would like to see in such a system.
6 This week in fact, just a couple of days ago, 7 the DOE has released a request for proposal to develop the 8 licensing support system. The NRC has a prototype of such 4
9 a system but it is going to be up to the Department of 10 Energy to develop the formal full production system.-
11 (Slide.)
12 There are many issues that are involved in its es 13 development and many of those can only be answered by the 14 forthcoming negotiated rulemaking process.
! 15 MR. MOELLER: If a private contractor then 16 develops the system, they will then turn it over to you?
17 or will you utilize the benefits of their work?
18 MR. BENDER: As it now stands the DOE will be 19 developing a system and making that system available to the 20 NRC, the states and the tribes. In fact we will be 21 responsible for inputting information into that system.
22 MR. STEINDLER: How about individuals?
23 MR. BENDER: The parties to the process, 24 individuals, states and tribes and DOE.
25 MR. STEINDLER: Just states and tribes?
O 1
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8@3364M6
.~. . - . . _ _ . . _ - _ . _ _ - -
- - _ _ . , _ _ . _ . . . - . . .-__-..._.---_..--.-_._.-_-m.-_--.
29879.0 163 BRT
()
~
1 MR. BENDER: But that still has to be defined in 2 the negotiated rulemaking.
3 MR. STEINDLER: Before'you leave that, 4 apparently you don't believe you are going to be subjected 5 to criticism if you make a system of that kind' essentially 6 not available to the public at large?
7 MR. BENDER: That's one of the issues that has 8 to be addressed. DOE is very'well aware of that and 9 through the waste fund I think they will have the capacity 10 to make that system available. It is the very intent of 11 why we are doing this.
12 DR. SHEWMON: Why should they have it available r~s 13 to them? Why is it unavailable to them?
L) 14 DR. REMICK: They might not have money available 15 for terminals and printers, as individual parties.
16 MR. BENDER: As you know, the states and tribes 17 are well-funded by the DOE and they have developed systems 18 of their own.
19 DR. REMICK: But the parties might be other than 20 Indian tribes and states. They could be individuals, I i 21 assume.
22 MR. BENDER: At that time terminals could be set 23 up at local public document rooms to allow access.
24 DR. REMICK: That was the answer I got, that DOE 25 was looking at making this available in some public i,
(
i
/\CE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 164.
BRT 1 document room so general members of the public could get it.
2 MR. BENDER: There are many issues involved in 3 this. Not so much of the technical issues of how do you 4 create this information management system, but some of the 5 issues, questions have come up, and I'm sure these are some 6 of the questions in your mind as well: What information is 7 going to get into this system? What is the time frame?
8 Shall we be putting in references going back to the 1940s?
9 or where do we draw the line?
10 What are the procedures for getting this 11 information into the system? And, if, in fact, this is 12 intended to be a system that will be used to minimize the
(~g 13 need to rely on discovery and it will be the sole source of
("' 14 information for the licensing proceedings, what sanctions 15 can be created for those parties that do not provide 16 information into the system and then show up at the last 17 minute and say: Well, here is a report that you guys are 18 not aware of that shows such and such; we would like to 19 have all of this information available into the system and 20 very accessible up front.
21 One of the key points here is that the major 22 issues have to be raised today, not in the 1990s. And by 23 making such a system available as soon as possible, much of 24 this information is accessible to the public and therefore
- 25 questions are going to be raised. That's an added l
/\CE FEDERAL REPORTERS, INC.
{
202-347-3700 Nationwide Coserage 8 m 336 & t6
, , ~ - -.-n,~
,--w,-,,,n, , - . . - - --,-,,-,,,,m.-,-w.,-mn wr--r - , - ,,--,-rn~ . _ , - , ,,~~,-.,,-m ,-v-,-~ , , - - , - , , - . , , ,
29879.0 165 BRT 1 advantage.
2 MR.' MARK: I don't understand the bottom bullet.
3 Is there some difficulty about looking at electronically 4 transmitted paper? Different from handwritten?
5 MR. BENDER: One of the options of building a 6 data base is to have individuals submit information in 7 electronic format. There are various ways of doing that 8 and you have to have some types of procedures as to how you 9 are to prepare documents and how you are to send them,
- 10 receive them and put them into the system. These are 11 questions that are going to be raised in negotiated 12 rulemaking and at that point we will have some answers.
-3 13 MR. STEINDLER: Did you say this was going to be 14 the sole source of information in the licensing process?
f 15 MR. BENDER: It is intended this will be a 16 system that will be used by all parties to the process and 17 will be the sole source of the information. That's correct.
18 MR. STEINDLER: That means -- do I understand 19 correctly that that implies that there will be no other
! 20 source of information admitted to the licensing process?
21 MR. BENDER: There will be other sources of l
i 22 information, obviously, but it's the intent of this system 23 to make all of that information available under one single 24 system and at the time that the license application is
( 25 submitted, that in fact becomes the single source for l
(3 s/
I ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336-6M6
, . . . _ - -.-- _ . - - - _ . , _ , _ . _ . , , , _ . . . _ _ _ . . - ~ _ , - _ _ - . _ _ _ . - - . _ . . _ _
29879.0 166 BRT
.,' b kJ 1- information.
2 MR. STEINDLER: You are not helping me much. If 3 you cut off your data base say in 1940 and you ignore all 4 literature of any sort beyond that -- earlier than that, 5 are you telling me that I cannot go to a license hearing 6 and bring in a 1935 document?
7 MR. BENDER: I'm not saying that. That's one of 8 the issues I raised before. It's not clear'at this point 9 what is the point in time that we cut off on the data as to 10 what goes on the system and what doesn't. It is 11 conceivable that some information will be needed, but not 12 everything will be put into the syste6 as far as being able pg 13 to search on it. Okay?
%,) The system in itself couid be made up of a 14 15 number of components and one component could be information 16 that's already accessible in the public domain, and since 17 it is accessible, there's no need to put r'iat into the 18 system for searching. But some of those questions that you 19 are raising now I cannot give you a clear answer on, 20 because these are questions that have been raised by other 21 parties to the process and they can only be addressed in 22 the form of a negotiated rulemaking.
23 MR. MOELLER: I understand what Marty is saying.
24 I don't know even where it cuts off, not only in time but 25 in subject matter.
4 O
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage m)-336-6M6
'29879.0 167 BRT r~m 1 I could claim a basic chemistry thing could help
-(v) 2 me. I~could claim that the Bible should be in there 3 because it helps me philosophically or morally to deal with 4 the issues.
5 Max?
6 MR. CARBON: I was just agreeing.
7 MR. MOELLER: I don't know where it stops.
8 MR. BENDER: I don't know either at this point, 9 but what has to be clear is that once we get through this 10 negotiated rulemaking process we will have some compromise.
11 MR. MARK: How will you arrange for electronic 12 transmission of those favorite chapters of the Bible?
13 DR .' SHEWMON: You could probably find it all on 14 a CD. I don't know.
15 It is my understanding, or was earlier, that if 16 somebody submits a document that they want to be part of 17 the hearing, it must be submitted not only on paper but
- 18 also in electronic format?
19 MR. BENDER: That's one of the possibilities, 20 yes.
21 DR. SHEWMON: It seems to me that's a reasonable 22 way to get most things into the data base. What other 1
23 things the philosophers want --
1 24 MR. BENDER: I'll be discussing that.
25 MR. CARBON: Not only the time frame and so on, 4
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
-, . . . . _ . . - - . ~ . . . _ _ _ . . _ _ . - . . _ _ . _ - , _ . _ , . _ . , . _ . _ . . _ , - . _ . _ . . _ _ , . _ _ _ _ . . . . _ . . . , _
29879.0 168 BRT 1 where do you draw the limits, but is it really at all 2 practical to talk about limiting it that way? Can you 3 truly, do you think, limit it such that somebody comes up 4 with something else, you are not going to let him bring it 5 in?
6 DR. SHEWMON: He can bring it in; just bring a 7 floppy disk.
8 MR. CARBON: He says if it's not in there it's 9 not there.
10 MR. BUNTING: My name is Joe Bunting. I may be 11 able to shed some light on some of the questions you are 12 asking.
N 13 When we advocated this idea to the Commission we i
14 agreed that not every mom and pop going to participate in 15 the proceeding would be required to possess a computer and 16 the kinds of equipment necessary to get this into the 17 system. But that we would convert what you now typically 18 know as the local public document rooms, to have a system 19 like this at which mom and pop could come in, access the 20 system, there would be people there to tell them how to use 21 it and if they wanted to insert favorite chapters from the 22 Bible, the system would be there to take that information 23 and put it in the system and convert it to the proper 24 format. That is what we had envisioned.
25 The notion is that we are looking for how to p
- 1 l
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 800-33MM6
29879.0 169 BRT
(,).)
( 1 save time in the proceedings and we want to be able to 2 eliminate someone asking 0-1 and 0-2 and not having it in 3 the system for everyone to see.
4 Bring in a document that no one knows everything 5 about, then everybody gets weeks to look at it and examine 6 it. The world stops while that happens.
7 We don't have the perfect answers to all the 8 questions raised. He are really turning over a new rock.
9 There's a lot of unknowns in this thing. We are dealing 10 with them the best we can. Yes, sir?
11 DR. SHEWMON: Did you say that at least one of 12 the precedents for this was that the judge in the U
-) 13 Johns-Mansville case said that --
14 MR. BUNTING: That's correct. Johns-Mansville 15 litigation case was conducted electronically in San 16 Francisco. My understanding is he ruled if you wanted to 17 participate in the proceeding, not only did you need money 18 for a lawyer, but you needed to have money to do it 19 electronically. If you couldn't afford that, you couldn't i
l 20 afford to participate in the proceeding.
21 MR. MOELLER: Perhaps one statement, or one l- 22 approach you could take which would help some of us -- some 23 of the questions we've just asked is to say that you want a 24 system that is as complete as possible or is as reasonably 25 achievable, and not claim 100 percent. You'd still save a l
l e'%
Y-]
l l
l ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8m33MM6
29879.0 170 BRT
()
n l lot of time.
2 MR. BENDER: There will obviously have to be a 3 level of uncertainty, but that has to be understood by the 4 parties up front and agreed to.
5 MR. BUNTING: One of the limitations we are not 6 going to put into the system -- pub 1'ish textbooks, for 7 instance. They are normally available in the public domain.
8 We are talking about documents that would be unique to the 9 proceeding.
10 So if someone found a 1942 document, for 11 instance, that's not in the system, we'll put it in the 12 system. Whatever you think is old information that the
- 13 parties think are pertinent to the proceedings, we'll put CD 14 it in there.
15 DR. SHEWMON: The Gideon Society will give us a 16 certain number of documents.
17 DR. REMICK: I can see the advantage to this.
18 Basically DOE and NRC say everything that we have we are 19 putting in there. Then if there's a discovery request, 20 basically you say: It's in there. They can't ask -- can't 21 delay the process by asking for documents outside there.
22 The problem is, if you have somebody that's 23 trying to be clever and withhold information, you do need 24 something there, if they want to introduce that as evidence 25 later on, there has to be some kind of high threshold on O
\_/
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage W A336 4 46
'29879.0- 171 BRT r
()s I why they did not make that available to everybody else in 2 that LSS.
3 MR. BUNTING: That's correct. In the negot'iated 4 rulemaking we are looking for some innovative ways to have 5 sanctions just to take care of that kind of gambit. We 6 don't have the answer, but that's our goal.
7 MR. PARKER: Someplace in the documentation we 8 were given it says you expect 16 million documents. Can
! 9 you tell us where you arrived at that and what are you 10 going to do with it? How are we going to access it?
11 MR. BENDER: I have heard that number. The l 12 alternatives to not having a system are pretty frightening.
13 Some of the other challenges facing:us is there are systems
()
4 14 in place, various DOE facilities, such as the DOE Salt 15 Project, and they have a computerized system, data 16 management in place, the DOE Las Vegas is now preparing to 17 develop a system of their own. Each one of these sites has
- 18 their own enormous record management problems that they-19 have to deal with.
! 20 DOE headquarters at this point has no record
- 21 management system at all. They are planning to develop 22 that and, as I mentioned before, states and tribes who are i
. 23 being funded by the DOE have developed their own systems as 24 well and the NRC is now proceeding with its transitional 25 licensing support system. This is a little bit troublesome, O
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage Mn336-6M6
-- ~- -- - - - . - - .-
1 29879.0 172 BRT 1
() I because each one-of these entities has their own procedures 2 and their reliability as far as being able to recall 3 documents from each one of these systems varies at this 4 point.
4 5 It is not clear to me about the integrity of the 6- data base in each one of these systems, with the exception j
7 of the NRC's of course.
i 8 So what we are trying to do is to move into some
.. 9 kind of single, comprehensive system which must be j
, 10 trustworthy and has to be the single source of information I
, 11 for the licensing proceedings. And DOE, in their requests i 12 for proposal, I'm sure is going to be dealing with this
!. 13 subject. The main concern here is merely the proliferation 1 14 of paper. What I'm going to show you here is only the j 15 estimates for the NRC', but as far as the high-level waste 16 program is concerned, the number of documents are growing
- 17 exponentially.
1 18 Slide).
19 Given that our budget is only 4 percent that of 20 the DOE you can multiply that by a factor of 25 or so.
- 21 Another 'way to look at this is you can translate 22 that into something like this. Okay? That's what it looks 23 like.
! 24 Can you imagine, you get into a situation when
- 25 someone says to you we have to very quickly find out, now, ACE FEDERAL REPORTERS, INC.
202 4 47 3700 Nationwide Cmerage 8(Xb336&t6
r 29879.0 173 BRT
() 1 all the documents related to performance allocation, or the 2 very subject matter of uncertainty that you were dealing 3 with today and very quickly come up with a chronology of-4 events of individuals who were responsible for decisions 5 and here are the files. Go to it and find it. Okay?
6 This was just a small pictureaof.our document i 7 control center. And it is getting worse and worse.
4 8 The system that we have created, which I will 9 describe to you, allows you to literally, quickly within 10 seconds, go into these documents and pull out information 11 and during the break if you are interested I can give you a
! 12 demo of something like that in the other room.
i 13 (Slide.)
14 As of today, as I mentioned before we had a i 15 pilot project we developed to convince the DOE to develop.
Our pilot has been very successful and, in 16 the formal LSS.
- 17 fact, we are beginning to recognize now that it is 18 important that the NRC begin the process of converting its 19 own documents into electronic format, so that at the time 20 when the DOE's system is operational and is in place, we l 21 can then provide the DOE with our information so that they
?
22 can load it into their system and make it accessible to all 23 the parties and all the other individuals involved.
- 24 As of today we have about 1000 high-level waste 1 25 documents, averaging about seven pages apiece, which are t
i ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cmcrage mig-33M646
__ ~ _ - - _ _ _ _ _ . . . . _ , _ _ _ . _ . . - . - - - . _ . . . _ _ _ . . _ - _ , __
29879.0 174 BRT
<x
() 1 accessible full text. It's an operational system.
2 We also have about 900 Congressional questions 3 and answers, also available full text search, as well as 10 4 CFR 60. You can get access to this data base even from IBM 5 PCs or word processing equipment, locally within the 6 Willste Building. But you can also get access to this 7 information from any remote terminal, so if you are out on 8 the West Coast and you want to access the information that 9 you saw on the slide before, all those files, you can do 10 that through the telephone. And that's what I'll be 11 showing you later.
12 MR. PARKER: Did I understand you to say that 13 the average length document of those thousands is seven 14 pages?
15 MR. BENDER: Each individual document is about 16 seven pages.
17 MR. PARKER: Dave, you have been sending us the 18 wrong stuff.
19 MR. BENDER: I should add some of those
- 20 documents have attachments such as vugraphs. Maybe it's 21 wishful thinking.
22 In addition to the full text system which we 23 demonstrated a year ago, we recognize that we do have 24 another problem and that problem is what do we do with the 25 images. Many of these memos come with all the attachments, O
ace FEDERAL REPORTERS, INC.
I 202147 37(N) Nationwide Coserage 8(X)-33M646
. _ _ .~ . _ . . _ _ _ . .
29879.0 175 BRT A
() 1 photographs and handwritten notes and so forth. How can 2 one access that on computer? Well, it is very difficult.
3 The next phase of our pilot is now seeking to 4 address that problem by being able to capture images of 5 documents. Some of the members of the ACRS, I think, have 6 seen this system. It is very difficult to talk about that, 7 much easier to describe it and show it to you, but I do 8 have some slides and some pictures which I will pass around.
9 So the next phase now is to go beyond the full 10 text search capability and get access to the original 11 images of the documents. I'll give you a schematic of the 12 system.
13 (Slide.)
14- The original prototype system which is now 15 operational is this portion right here. What you are 16 seeing is, the information resides on a mainframe in Tampa, 17 Plorida. In the Willste building we have terminals known 18 as 3278s, word processing equipment, IBM 5520s, and a laser 19 printer. All of this equipment and hardware is tied to a 20 cluster controller. This box, then, communicates with the 21 mainframe in Tampa, Florida.
22 Another way of accessing this mainframe, is you 23 can do it from a remote PC.
24 So this operational system here is an 25 accomplishment in itself, in that it does give you very O
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Cmerage 800-336-6646
29879.0 176 BRT Ik
%/
1 quick access to the information. But only to the text, not 2 to the images or photographs or maps.
3 This second system here which we have started 4 just recently is a microcomputer-based system which allows 5 you to do exactly the same thing as far as full text search, 6 but at the same time also display the image of the document 7 on the screen. Which is something, actually, pretty unique 8 in the country, in fact. It's a single work station. It 9 is a prototype of what we would like to see in the near 10 future. The experience we are getting in operating this 11 system is being shared with the DOE and we are hoping that 12 they will make full use of that experience. In fact we 13 have in preparation a memorandu,m of understanding between 14 Davis-Rusche to pretty much formalize that DOE should be 15 working very closely with us as they go ahead and develop i
i 16 the system, and recognize some of the capabilities that we i
l 17 have developed over the past few months.
- 18 (Slide.)
i 19 Just a little bit about information retrieval.
- 20 How many of you in the audience have used computer 21 equipment to access data bases to retrieve information?
l l 22 Has anyone? Okay. Usually when you access a data base, i
i 23 with the exception of LEXIS and some of the other full text 24 systems, you do a search of an abstract or a surrogate of i 25 the document and you normally search an author or subject ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-346M6
29879.0 177 BRT t'%
( ) 1 or addressee, maybe some kind of a file code. That --
v 2 basically all the information you then get is just that top 3 portion. Okay? There's some problems with that which I 4 will describe to you.
5 The system that we are developing and advocating 6 is a system that not only allows you to do a search of the 7 surrogate of the document but also, at the same time, 8 allows you access to the full text. This is an example of, '
9 let's say, an internal memo. So again that slide that you 10 saw with all those files gives you immediate access to the 11 content of the information within the document as opposed 12 to the surrogate of the document.
13 The reason that we have placed emphasis on full O 14 text search as an important feature of any information 15 retrieval system is because I ran a little experiment with 16 our own system. You recall the little header that you saw 17 before, the surrogate document. I queried the system and 18 asked, just searching on header now, I asked the system:
19 Give me all correspondence dealing with volcanic activity 20 at Yucca Mountain, and within 3 seconds I got a read of 21 zero. No documents in the system.
22 Many of the existing systems within the NRC can 23 only allow you to s.earch on a surrogate of the document so, 24 in effect you possibly lose access to some important 25 documents.
O ACE FEDERAL. REPORTERS, INC.
202 347-37(X) Nationwide Coserage N N)-33MM6
29879.0 178 BRT (qj 1 I then conducted a full text search with the 2 same query and I got 42 hits. Meaning there were 42 3 documents in which the word " volcanic" and " Yucca 4 Mountain" appeared.
5 After a view of that document it may be that in 6 fact only two or three of the documents are actually 7 relevant, but otherwise if you are strictly searching on a 8 surrogate you would have lost that document. Multiply this 9 by the 16 million potential documents and you begin to see 10 the real critical need to have a very effective information 11 management system which combines the features of full text 12 and surrogate searches.
13 I ran similar tests for correspondence dealing
'~#
14 with some problems associated with trench-14, something 15 that our staff is working on. By simply searching the 16 header again -- which, by the way was indexed for us by a 17 contractor specializing in this field -- zero documents.
18 With the full text search I got three documents.
19 It was only when I knew very specifically what I 20 was looking for that the result was the same. In other 21 words, when you know the date of the memo and the name of 22 an individual, then you are very likely to get it and in 23 fact you could go to a hard copy and pull that out because 24 you know specifically what you are looking for.
25 MR. STEINDLER: You said the system was O
V ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 20-33MM6
29879.0 ' 179
< BRT
() 1 assembled for you by a contractor who was expert in the 2 field. Expert in what field?- Document retrieval?
3 MR. BENDER: The individuals who wrote the 4 header for us, or the abstract, are individuals that --
5 that was the Aerospace Corporation individuals who are 6 working in the field of information management. I would 7 say the aptitude of the individuals who created the header 8 was several years of college and some graduate studies as 9 well. And they reviewed the documents and put that 10 information in there.
11 I should add that, for example, in the DOS data 12 bases a lot of this abstracting is done by part-time
- 13 college students at night.
14 MR. STEINDLER: There's your problem. Get 15 somebody who understands something about waste management, 16 you might not have had such a bad record on the other 17 recall.
18 MR. BENDER: So based on my little experiment
. 19 here, I then went into the literature.
20 MR. MOELLER: Excuse me. The headers -- are the i
21 headers written by these college students?
22 MR. BENDER: Yes, right.
23 DR. MOELLER: They are. It's not just using the 24 title of the article?
25 DR. SHEWMON: This is a key word arrangement?
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6646
29879.0 180 BRT
() 1 MR. BENDER: Somebody looks at it and says I 2 have to fill in these field, key words, author, subject, 3 and maybe one brief statement about what the document is 4 about. An abstract.
5 So when you are searching the document years 6 later, you have to rely -- there has to be some congruity 7 between what that individual determined to be important and 8 what you may feel is important in the document. Some of 9 the times that might not be the same.
10 So taking that somewhat uncontrolled experiment ]
11 I conducted I went into the literature just to see what 12 other people are saying about this. This is an individual, 13 Carol Tenopir, she worked out of the University of Hawaii O' 14 and specializes in the field of full text search. She 15 conducted a search of her own on surrogates, abstracts and 16 the full text. And what she found was this: When you are 17 searching only in the full text, if you search in full text, 18 she got these -- an average of these many documents. Out i 19 of those 17 only 3.5 were actually relevant. Okay?
i 20 By searching only in an abstract it went down to I 21 2.4. By searching only on controlled vocabulary, key words, 22 it was 3.1. Bibliographic with the abstract produced a 23 little bit higher recall. But a larger percentage of 24 relevant documents.
l 25 The best combination appears to be able to 1
( /\CE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide CmeraFe 800-3364M6
-_m..
4 29879.0 181
, BRT search in both the full text and the surrogate of the 4
(f I 2 document.
i 3- comparing these numbers it appears there really 4 isn't that much of a difference, you know, 4.5 to 3.5.
L -
5 Again you have to multiply that by the number.of documents
- i. 6 we are dealing with and ask yourself the question. Can you 7 put up with this kind of a potential earlier?
l 8 If everyone agrees to it, yes, then so be it.
I 9 DR. SHEWMON: I don't understand the difference 10 between the first two up there. Are you saying that words 11 which were germane to the search came in the surrogate but 12 not in the text?
]
13 MR. BENDER: Between these two?
14 DR. SHEWMON: Yes.
i 15 MR. BENDER: What that means in the first one
- 16 you have the option of searching both the header and the i-l 17 text at the same time. In the first search you are
- l. 18 searching -- you are saying to the system: I don't want to j 19 look only to surrogate. Search this and also search this 20 at the same time. Okay?
21 DR. SHEWMON: And in the other case you only 22 searched the text.
I
! 23 MR. BENDER: This portion.
I 24 DR. SHEWMON: And it was better if you had the l 25 header and text than just the text; which means there's I
l ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Cmcrage mm.33MM6
29879.0 182 BRT o)
(
1 information in the header that's not in the article.
2 MR. BENDER: Correct.
3 DR. SHEWMON: I find that hard to believe.
4 DR. REMICK: Is it the author in combination 5 with the word? Is it word strings? Why is it better?
6 MR. BENDER: The reason it was better in this 7 particular case was because there was, I believe, the use 8 of a source --
9 MR. CARBON: Excuse me, use of what?
10 MR. BENDER: The source. Something like a 11 synonym file. If the subject of this memo deals with 12 airplanes, but for whatever reason, nowhere 1. there 13 mention of a Boeing 707, the person who does the 14 abstracting picks up on that and includes that information 15 in the abstract. Okay? So now you have the added 16 insurance.
17 So by searching on both you got the hit on the 18 document.
19 DR. REMICK: How could he put Boeing 707 if it 20 wasn't in the text.
21 MR. BENDER: The other way around.
22 I think you know what I'm saying.
23 So the point here is that there are a number of 24 ways to develop an information data base and I think people 25 should be aware of some of the limitations of existing
()
v ACE FEDERAL. REvonTens, INC.
202 347-37(x) Nationwide coserage w ri-3 w ( M 6
29879.0 183 BRT
() 1 methods of putting information into systems.
2 (Slide.)
3 What I would like to do now is talk a little bit 4 about how we actually build a data base.
5 MR. MOELLER: Even the articles that you put 6 into the data base where you essentially photograph a page, 7 you can still search out individual words?
8 MR. BENDER: Okay. On that particular system 9 which I think some of the other individuals have not seen, 10 I will describe exactly what happens at this point.
11 If you can visualize the stacks of information 12 in the files that you saw before, information comes into 13 our docket control center which we like to term the eye of
( 14 the needle, flow of all the in f o rma tion . That information 15 is scanned by the imaging system. The image is verified.
16 The image is then captured and is sent to a' searchable 17 image handling system at the same time, and it gets to be a 18 little confusing here, but if this information came from 19 internal to the Division of Waste Management, that means it 20 is already on our system, and we have it in electronic 21 format, so that the ASCII text is already available for the 22 searchable full text system. In those instances when the 23 information comes in from the outside, the image has to be 24 run through an optical character reader to recognize the i
25 text and then convert that into a searchable ASCII format.
ACE FEDERAL. REPORTERS, INC.
202-347 3700 Nationwide Cmerage 800-3 h(M6 l
29879.0 184 BRT f%
,) 1 The only way to conduct a full convertible text is for the 2 computer to understand the ASCII text, so there's a picture 3 between a picture on a computer and a picture which is the 4 ASCII text, between digitized information and pure ASCII.
5 In the meeting that we had the picture that you 6 saw on the terminal was an image of the document. It was 7 one step before that image was interpreted into an ASCII 8 format.
9 In esserce, and it is a microcomputer based 10 system, the second phase of our pilot, we have dual systems 11 which are transparent to the user. So we have the 12 searchable ASCII text and we have the image. This allows 13 you to conduct a full text search, document comes up on the U
14 screen, you scroll through it and you say, well, this memo 15 is very interesting but I notice that it says something 16 here about an attachment, some maps. Where are those maps?
17 Well, the maps and the original image of the 18 memo with the signatures are located on this image system 19 and the way these two systems are linked is by a common 20 access number which is assigned at the time this 21 information gets into the system. So you do your full text 22 search or the search of the header as I showed you before, 23 and then you may be satisfied with the information you 24 receive here or, if you need the image, you go into the 25 image system which is just a number of key strokes away and Act FimnnAt Riti>oirriins, INC.
202-347 3700 Nationwide cmerage soo 3364 a 6
29879.0 185 BRT O 1 voo o#11 #9 ea t orieteet 1 eoe-2 These other boxes that I have here deal with 3 some new technologies and I'll give you an example of that.
4 This is really the most difficult aspect, to get 5 the information into ASCII, and we are striving to get most 6 of that from our word processing equipment, and that was 7 the emphasis we placed earlier about electronic mail and 8 receiving that in that format, because once we receive it 9 in an electronic format, we don't have to go through an 10 optical character reader where there is some room for error.
11 Once this data base is created it presents a 12 number of options. One option is to maintain it in a 13 mainframe system and access it from terminals across the O 14 country. The other option is to take this information and 15 put it on something called CD-ROM, and I'll explain what 16 that is, or an optical disk, take that optical disk and 17 send it to a different location in the country. Again, 18 visualizing the information that you saw before, the file
! 19 folders.
i 20 You can either ship this to the site or you can l 21 send this. Okay? You can store about a million pages of 22 information on one of these optical disks. Okay? You can
{
, 23 store the text and the images. So there is a revolution 24 really going on now in the record management industry, and f 25 in order to access the information on such an optical disk, l0 ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage Hub 33MM6 l
i
29879.0 186 BRT-
-( ) I what you would need is a computer like an IBM AT, 2 specialized high resolution monitor, and a device -- a 3 controller and a drive to insert this into and then read 4 the information.
5 When we are dealing with millions of documents 6 this appears to hold great promise. Imagine to be able to 7 get this massive amount of information, create,maybe one or 8 two of these diskettes, these optical disks and send them 9 to the sites? And right there in the local public document 10 room, have access to the whole data base. Okay?
11 So these are some of the new technologies which 12 we are now testing and I think they hold some great promise.
13 MR. PARKER: Aren't they doing that already with 14 some legal systems, and actually updating and throwing away l 15 the old ones because they are so cheap once you get the 16 first one done?
j 17 MR. BENDER: Theoretically, once you capture the 18 information on optical disk you can get rid of your 19 original files.
20 MR. PARKER: On a current system like this, you 21 can send those out periodically and then when you add on 22 new information just throw away the old one. Those are 23 only a couple of dollars. It's making the first one is the 24 expensive one.
25 MR. BENDER: To make the first one is about O
ace FEDERAL RevonTEns INC.
202 347 3700 Nationwide Cmcrape 800-33MA46
29879.0 187
-BRT
() 1 $5000 and to make each additional one is about $10 or $15.
2 MR. STEINDLER: How much money is NRC spending i~
3 on this exercise?
4 4 MR. BENDER: How much money? The initial effort 5 to develop this first imaging system is approximately 6 S300,000. There was $100,000 to purchase the equipment and I 7 the rest was for contractor support to take that equipment l
8 and to integrate it.
9 MR. PARKER: Are you using scanners to get some 10 information into the data base?
11 MR. BENDER: Really the best way to see this is i
12 to come over to our shop and I'll show it to'you; 13 I do have some pictures which I'll pass around.'
j 14 So, some of the questions, again.
15 (Slide.)
16 How does one go ahead and build this data base?
i j 17 As far as the NRC is concerned there are three options.
18 You can take the documents and rekey them; you can capture 19 them through an optical character reader and do your 4
20 characterization and get the ASCII text that I mentioned 21 before; but what appears to be the best way is to capture 22 them directly from the source. The reason for that being 23 over 60 percent of the documents that we create in the 24 Division of Waste Management are produced in our 5520 25 system. And the kind of documents that I'm talking about O
ACE FEDERAL REvonTEns. INC.
202 347-37(W) Nationwide Coserage M(x)-33MM6
29879.0 188 BRT (O,
1 have to do with technical position memos, trip reports, 2 site visits, Commission papers and Congressional Os and As.
3 So the majority of documents can already be 4 captured in electronic format. It has the great advantage 5 in that it reduces the labor intensiveness of building a 6 data base.
7 As far as this "other" that's concerned, we have 8 put, for example, in the FFRDC, a requirement for a new 9 major support contractor to provide us information in 10 electronic format. And it is conceivable that in the near 11 future DOE may require something like that as well to 12 minimize the need to rely on optical character readers.
s 13 What I would like to do -- there was no way for 14 me to bring the new system out, but I would like to -- I 15 have some slides and I would like to pass some of these 16 pictures around. I don't know if you can see this, but 17 we'll pass it around.
18 The unique feature about the new system that we 19 are developing is that it allows you to conduct this full i
20 text search and then bring up the original image of the 21 document. Okay? I'm show the slide, but at the same time i
22 I would like to pass that around.
23 MR. MOBLLER: If a document has a typographical 24 error, which some of the older'ones I presume would, will 25 that throw the search off?
l 'I
/\CE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8f&336-6646
29879.0 189 ,
BRT
("% MR. BENDER: If there's a typo in &.he electronic t,,_j 1 2 version of the document and in the original when it went 1 3 into the system, at that point it will -- there will be a 4 typo in.the system.
5 MR. MOELLER: I meant, let's say I'm searching 6 for the word repository and that word is misspelled with a 7 typographical error. Will it pick that up and call it out 8 for me?
9 MR. BENDER: One of the features of the full 10 text search system is before you conduct any search it's 11 always wise to use a wild card approach. So, for example, 12 if you are doing a search on vulcanism but you are not 13 really sure how that word has been used, what you would do 14 is type in vole dollar sign, and in a matter of seconds 15 you'd have a listing -- it's the stem of the word, all the 16 words that are like that. And when you do that you may in 17 fact pick up some typos and may pick up some other words
! 18 that you didn't think existed within the documents, and it 19 gives you a chance to search on the group as well.
I 20 MR. MARK: I'm impressed with the magnificent, l 21 almost incredible capability of this system you have.
Marty asked about the cost, S300,000 for just the machinery.
i 22
( 23 Are you saving any money at all for people to read the 24 miserable documents and not understand them?
25 MR. BENDER: The major cost savings is in the l
l ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6M6
29879.0 190 BRT
() 1 future. Actually -- right now. For example just 2 illustrate. We have a need to constantly respond to 3 Congressional questions *. Many times we've already answered
^
4 the same question. So an individual has to go to the files 5 and look for a similar answer that we may want to give at 6 this point.
7 MR. MARK: I would hope he could have made up 4 8 his mind just from what he understood.
9 MR. BENDER: Using this system obviously it's 10 much faster. I don't know how well you can see this, but 11 the imaging system at this point is a menu driven system.
12 It is still being refined. It allows the user to sit at 13 the station and do a search of the data base.
L 14 MR. MOELLER: Can you put a book under the front 15 of that?
16 MR. BENDER: How is that?
17 This is the imaging part of the system so a 18 document comes in through the window and the operator has 19 an opportunity to oither scan the documents and enter them 20 into the system or do a search. Once the document is in 21 the system, it can then be readily retrieved.
22 (Slide.)
23 What we are trying to show here is that the 24 combination -- combination of having both the full text 25 search capability and the access to the original image. So, O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800 336-6646
(
. _ . _ _ _ , _ _ _ . . _ _ ,_ _ . _ . _ _ _ . _ , _ _ _ . ~ _ _ _ . , _
29879.0 191 BRT
() 1 in this case a search was done on the word " Death Valley."
2 Okay? So you very quickly have access to the ASCII text 3 that comes up and can scroll through the document, find all 4 the occurrences of " Death Valley," okay, and see whether in 5 fact this is the document you were looking for.
6 If you are then interested in getting the image 7 to the document, you can then press one of the keys and the 8 original image of the document will then come up as well.
9 On the terminal.
10 (Slide.)
11 So we can go from that, then picture on the 12 terminal, you can then go to the original image of the 13 document. This picture was taken directly from the 14 computer terminal. It is very crisp and clear and this 15 information can be printed on a laser printer and the 16 output is as good as the quality of the document that went 17 into the system.
18 The other advantage of course is being able to 19 capture things such as maps. This happens to be a map of 20 the Yucca Mountain site. Again, the imaging quality is 21 very good. This picture doesn't really do it justice, but 22 you would have to see the system up close yourself.
23 So, very quickly, documents can be brought into 24 the system, scanned, captured. The labor intensive part of 25 it is the conversion to the ASCII -- to the searchable O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
{
i
~
29879.0 192 BRT
(,j - 1 portion of the document.
2 So, in summary, as far as the system is 3 concerned, you can do full text search on the cover memo of 4 the document, no matter how many pages that is, and then if 5 the document has 'a number of attachments, the'whole package 6 in its entirety is stored on an optical disk similar to i
7 what I showed you before and all of that information can be 8 readily retrieved.
9 MR. PARKER: Can't you make it a part of every 10 contract now that everybody has to give you a diskette in 11 ASCII format?
12 MR. BENDER: Yes.
13 MR. PARKER: Are you doing that already?
14 MR. BENDER: That's what we will be doing with 15 the FFRDC.
16 DR. SHEWMON: Will you settle for ASCII or 17 Display-rite III?
18 MR. BENDER: The ideal way to get the 19 information would be to get it both in hard copy and in the 20 back a diskette that has the information in ASCII.
21 DR. SHEWMON: That's not an answer to my 22 question. The FFRDC proposal went out that things had to 23 be submitted in Display-rite III files, not ASCII files.
24 My question was would you settle for ASCII 25 instead of Display-rite III?
O V
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage MXL336-M>46 m
l 29879.0 193 BRT
( ). 1 MR. BENDER: Whether it's Display-rite III or 2 ASCII honestly doesn't really matter. Because if it's in 3 Display-rite III there's a feature that allows you to very t
4 quickly save the document as an ASCII file, or if you 5 already get it in ASCII, you save yourself a step.
6 So either way would be fine. But we would want ,
7 to avoid getting information from a number of different 8 word processing systems.
9 Display-rite III was mentioned because it is the 10 word processing system that is being used throughout-the 11 NRC.
12 MR. STEINDLER: This system is not going'to be f3 13 archiving or storing into its base contractor documents U 14 that are normally published in public domain, is it?
15 MR. BENDER: At this point the way our docket 16 control center is set up, there is no way that we can 17 arbitrarily say, let's just forget about this pile of 18 information and only concern our self with this pile.
19 Until such time that we implement strict procedures, we are 20 now saving all the information. At this point, unless I'm 21 told otherwise, all the information having to do with the t
22 high-level waste program is going into this system.
23 MR. STEINDLER: So all the NUREG documents and f
24 NUREG CRs that you are going to get out of contractors are 25 going to end up in there, even though I can buy them from i
I l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 194
- j. BRT
() I whatever the outfit is?
1- 2 MR. BENDER: One exception may be the NUREG i
3 documents, but I'm not really sure whether in fact that 4 would get into the system.
5 MR. MARK: You have spoken of many hundreds of 6 thousands of pieces of paper. Do you have a search routine 7 which will tell you which 10 you ought to read?
8 MR. BENDER: We are getting into artificial 9 intelligence, and we seem to have a lot of that. So maybe 10 we can address that -- but anyway -- where do we go from 11 here? A number of activities that are ongoing. Our main 12 objective now is to continue to develop this traditional
46
29879.0 201 BRT
() 1 MR. MOELLER: So we are not missing these; they 2 don't exist at the moment?
3 MR. COPLAN: That's correct. And as I get into 4 this farther, I think you'll find that the focus of the 5 activity that I am going to be describing is somewhat 6 narrower than that; it is focusing primarily on the 7 technical issues and capability to conduct technical 8 reviews.
9 MR. PARKER: Could I ask you another question 10 along the same lines that Dave did? Can we take this as a 11 precursor to what the high-level one might be like? The 12 standard review plan for low-level waste?
13 MR. COPLAN: We are not really sure yet how much
( 14 detail we'd be getting into in a review plan. As I get 15 farther into this, I'll describe some of the thoughts we 16 .have as to how we are going to proceed along these lines.
17 MR. PERRY: You might want to compare the SCP 18 review plan, or standard guide -- content format for the SCP
[
19 -- that might give them an idea of what the applications l 20 are going to look like.
1 21 To give a relative idea, it's going to be a very i
22 big document compared to the low-level one. l I
23 MR. COPLAN: Oh, yes. The SCP alone, we are 24 told, is probably going to be on the order of several !
25 thousand pages.
O ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
l l
29879.0 202
.BRT N ,
4 1 MR. PARKER: The review plan?
2 MR. COPLAN: The site characterization plan that 3 DOE is preparing. The license application, who knows?
4 In' implementing this five-year plan, there are 5 several items that stand out and those are really hit on in 6 the next couple of charts that John had presented to you 7 before, one of which is this concept of identifying and 8 prioritizing open items, developing mechanisms to focus the 9 development of guidance and NRC/ DOE interactions on closure 10 of these open items; and the development of a Staff review 11 capability.
12 The activity that we have under way is one that 13 is focused on resolving what we have come to call O 14 " compliance demonstration issues." I'm just referring to 15 it as " resolution of compliance demonstration issues."
16 Its purpose is to implement the five-year plan 17 for generic open items. It has as its objectives to 18 identify key licensing open items; to identify strategies 19 and specific mechanisms to resolve them; to identify what 20 must be done to assure an independent integrated Staff 21 review capability; and to establish milestones and 22 schedules for open item resolution and development of 23 review capability.
24 The approach that we are taking in this is, 25 first off, to identify what we have come to call O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
___n
29879.0 203 BRT
(<s) 1 " compliance demonstration issues." These would be an 2 embodiment of the findings that would have to be made under 3 Part 650. One naturally starts to focus on the performance 4 objectives and the siting and design criteria in that 5 context.
6 For each such compliance demonstration issue, we 7 then proceeded to identify key generic open items that 8 would specify the findings and technical conclusions needed 9 to be made or that are implicit in the compliance 10 demonstration issue, and identify what guidance is needed 11 and what methodologies need to be developed in order to 12 support the findings and technical conclusions.
13 The next step is, having gone that far, to 14 identify methods for resolving issues where that might be 15 possible and the working assumption is that at the start 16 that they can all be resolved. Which ones we would end up 17 focusing on would be a question of resources and priorities, 18 but the idea is to identify the methods for resolution, 19 evaluate open items and identify areas where guidance 20 should be developed for the DOE. To the extent practicable, 21 resolve these open items prior to the license application 22 submittal using such mechanisms as generic technical 23 positions, topical reports, which the DOE might submit that 24 we could review; meetings with the DOE and to put things to 25 bed once and for all, where practicable, to go to a ace FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coserage 800-336-6646 l
.- __ - _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - _ - _ _ _ _ _ _ _ - _ _ _ _ _ _m
29879.0 204 BRT
'( ) 1 rulemaking.
2 The next step in this process is one of 3 identifying mechanisms that are needed to develop and 4 support the Staff review capabi~1ity. What I mean by that 5 is that over the next several years, one of the things that 6 we are going to have to do is assure that we have in place 7 a capability to condu'ct an independent integrated review of 8 a DOE license application and what are the mechanisms, what 9 kind of approaches should we employ in order to see to it l 10 that we do have that kind of a capability in place when we
, 11 need it. So -- and then to develop the milestones, L
l 12 schedules. identify specific steps for open item resolution, 13 development of review capability, prioritize the open items 14 and, finally, establish schedules.
15 In going through those steps we were guided by 16 certain principles that you have seen before in the package 17 that John presented last July, but essentially this would 18* be a proactive as opposed to a reactive approach; that.the 19 program would be focused on the key licensing decisions 20 that need to be made; that the process be open and 21 documented and that it provide for early involvement with 22 DOE, the affected states and Indian tribes, and it include 23 the development of a formal mechanism for its 24 implementation.
25 That is another activity that we are just O
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
29879.0 205 BRT (9_) 1 getting started on. I'm not going to talk about that today.
2 We can come back and talk about that when we've made a 3 little bit more progress on it.
4 With the next chart that states the compliance 5 demonstration issues and really starting the example. The 6 compliance demonstration issues that we identified were, 7 firs't of all, the eight -- eight of the compliance 8 demonstration issues derived directly f rom the performance 9 objectives of Part 60. I'm not going to walk through those.
10 There were five others that cut across the first 11 stage and either address a common element or assure an 12 integrated program, these being uncertainty, quality y 13 assurance, development of format and content guide for the d 14 license application, development of review plans, and a 15 plan to integrate the activities that go on under all these 16 other work plans.
17 I guess I just threw another term here that I 18 hadn't used before. What we are doing for each of these
. 19 compliance demonstration issues is develop what we call a
( 20 " work plan." It's simply a set of milestones and schedules 21 that are intended to accomplish the development of the 22 review capability, get open items resolved. That means 23 developing a generic technical position and would include I 24 milestones and schedules to develop that position.
25 The next chart just focuses in on one of the i
t
! ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-3364446 k
29879.0 206 BRT
() I compliance demonstration issues, namely the waste package 2 containment. Just by way_of example, the open items that 3 we identify are -- to get some detail, so I'll just hit the 4 high spots of it. But there were two broad areas, the 5 first of which is performance analysis of the waste package 6 designs and the second broad area is the waste package 7 testing. The way we approached that, in way of performance 8 analysis of waste package design, the way the regulation 9 reads is that a waste package must contain the waste for a 10 period of 300 to 1000 years, the thought being that the 11 Department of Energy could present and defend sometime 12 within that range as being the appropriate time for their g 13 design.
(_) 14 That means that somewhere along the line Staff 15 is going to be in the position of having to review, and the 16 Commission is going to be in the position of having to make 17 findings on whether or not the time that the Department of 18 Energy is proposing is appropriate.
19 So, one of the activities that we figure would 20 be needed in order to resolve that particular item is to 21 develop criteria for an acceptable waste package 22 containment time, something that will help us to know 23 whether or not what they are providing is adequate.
24 A number of other items there, I wouldn't 25 propose t> walk through them in detail, but each one of O
C/
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8%3364M6
29879.0 207 BRT 1 them had in mind to try to finger an srea where there is a 2 potential stumbling block to licensing if we don't start to 3 get our ducks lined up now.
4 MR. STEINDLER: Am I correct in assuming that 5 under the criteria for acceptable waste package containment 6 time, issues only arise if the time elected by DOE is less 7 than 1000 years?
8 MR. COPLAN: I would think you are right in that I
9 assumption. I can't think of a reason why not, but I would 10 be inclined to hedge it.
11 MR. PERRY: Seth, would you raise a question if, 12 in fact, their projection was accurate? I mean;-- say they 13 say it's 1000 years. Say I don't believe them.
14 MR. COPLAN: That's the second part of that.
15 That is all the rest of it, the other items that are ,
16 identified there, are all focused on -- all right, now we 17 have agreed on what the time is. How do we determine t
18 whether or not the package is really measuring up to that?
19 Okay. So, having identified those as the open 20 items for waste package, the next step would be to identify 21 mechanisms that could be used to get those items resolved.
22 Generic technical positions at this stage of our planning 23 would be a primary mechanism for providing guidance and 24 resolution of the open items. These would be developed 25 through a standardized -- it should be set of milestones.
l O
ACE FcDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
29879.0 208 BRT
() 1 The standardized milestones I will come back to in a moment, 2 but they are intended to incorporate the idea of getting 3 states, tribes, DOE involved early on in this process of 4 reaching agreement on resolutions for the issues. But 5 other mechanisms would also enter into play. These, of 6 course, would include topical reports which I mentioned 7 earlier; letters to DOE, meetings, and rulemakings is kind 8 of the ultimate way of closing out an issue.
9 Selection of specific mechanism for each open 10 item will depend on priorities, resources and 11 appropriateness to the particular open item.
, 12 The review capability would be developed through 13 a standardized set of activities that we worked out while 14 going through this process that I'll touch on in a moment 15 and that -- let's go back to GTPs and the kinds of things 16 that the milestones were intended to assure are dealt with 17 while the positions are developed. They involve DOE,
~
18 states and tribes early, and throughout the entire process.
19 The milestones were intended to adequately ventilate open 20 items to the technical community, peer review groups; they 21 assure NRC's request for comments are responded to and 22 documented; and they provide DOE, states and tribes with 23 documented responses to comments on draft technical 24 positions. So the milestones really include the basic 25 elements of what we have been doing now.
- C
- )
i ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
_. ~ ._ _ __ _ _
4 29879.0 209 BRT
-( l Yes?
2 DR. SHEWMON: Are these milestones to start 3 during the time you are reviewing these things after the 4 package is in or before?
5 MR. COPLAN: The generic technical positions are 6 positions of the kind that the Committee has actually 7 reviewed several of them at this time for us: Groundwater 8 travel time, waste package reliability position, and what 9 we do there is we are developing what represents a Staff 10 position and approach.
11 DR. SHEWMON: But when you say " involve the i
! 12 states," my impression is the State of Texas tells you to 13 get lost, rather than wanting to get involved, doesn't.it?
14 If it's that civil.
15 MR. COPLAN: What we mean by " involving them" is 16 making a position or tentative position available for their 17 comment, reaction,. at a number of stages through our i
18 process of development.
19 We have put in place a system so that if we
[ 20 don't get a response, say after the opportunity to comment
, 21 has been noticed in the Federal Register, we send out a 22 tickler letter, and we will be responding in a documented 23 way to the comments that we do get and try to continue a 24 dialogue on the basis of comments that we get.
25 DR. SHEWMON: Okay. Thank you.
ACE FEDERAL REPORTERS, INC, 202-347-3700 "ationwide Coverage 800-336-6646
-29879.0 210 BRT
() 1- MR. COPLAN: With regard to the development of 2 Staff review capability, the activities that we've 3 identified there that would be scheduled for each of the 4 open items that are identified under compliance 5 demonstration issues would be, first to establish review 6 criteria or standards based on the finding or technical 7 conclusion that's specified; secondly, to develop a review 8 strategy in coordination with the task group that is being 9 set up within the division to be called the "modeling 10 strategy task group." The modeling strategy task group 11 would include people from each of the disciplines involved 12 in repository review and it is going to be chaired by 13 Pauline Brooks who spoke to you earlier this morning.
O- 14 Their. job will be to review each technical 15 strategy that is developed for conducting a review against 16 these standards or criteria, and assure that there is 17 integration across the strategies and that there is j 18 consistency in them with the modeling strategy document 19 that we talked to you about, I guess not quite a year ago.
20 The first job of the task group is going to be to revise 21 that, taking into account your comments among others. Then, 22 having come up with a set of these strategies -- I guess 23 maybe I ought to say one other thing about them to flesh 24 them out a little bit better.
25 For the SCP, we are going to do this -- not just l
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
29879.0 211 BRT ,
t o) ,
1 to prepare for a license application. We are intending to 2 do this prior to the SCP review and at least for the SCP 3 review, what we have in mind is something that would be 4 fairly concise, say simply the identification of whatever 5 technical conclusions need to be made that are implicit in 6 the particular open item that was identified; what 7 standards or criteria would be used,to focus the review and 8 then a -- in a few paragraphs, say page and a half, two 9 pageo -- lay out what the technical approach is going to be 10 for conducting the review against those criteria or 11 standards.
12 So we are talking about a fairly brief document
,e m 13 that is intended, as much as anything else, to, one, get
!/
14 people thinking in advance about how they are going to know 15 what they see is good enough; how they are going to conduct 16 a review to make that determination and, two, to assure 17 that within the different technical disciplines there --
18 that the right people are expecting to be talking to each 19 other and getting the right information from each other and 20 providing the right information to each other.
21 Once that is in place, the next step in 22 developing this review capability would be to determine 23 whether the existing contractual efforts that we have, 24 Staff activities and research efforts, support the needs 25 that are called'for in the strategy and, if not, to develop
(~'i V
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6M6
- .j
29879.0 212 BRT
(,) 1 further plans that would acquire that capability.
2 The current status of this effort is that we've 3 got kind of a first good cut at a complete set of work plans, 4 in other words a work plan for each of the compliance 5 demonstration issues and they are not undergoing a 6 management review within the division.
7 The management review is intended, among other 8 things, to point out whatever holes we might still have in 9 terms of open items that have been identified; areas where 10 we have overlooked connections between plans, for example, 11 there may be places where a single activity within the area 12 of hydrology might serve to lay to rest several open items
,- s 13 under several of the compliance demonstration issues so e l v that they would appear in several of the work plans at this 14 15 point. We have overlooked that.
16 Then, once we've gotten to that point, the next 17 step is to prioritize the items, determine milestones, 18 schedules, dates and resource allocation. That is going to 19 be a very key and actually very difficult step.
20 We put the plans together to this point. We 21 haven't really made any effort to screen anything out. We 22 haven't culled anything. We santed to try to get the full 23 set of open items and there is an awful lot there. This is 24 probably more there than we would be able to handle at, say, 25 even the level of trying to develop generic technical r'
(' _)s ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide CoseraFe 800-336-6 4 6
29879.0 213 ,
BRT
() 1 position for each and every open item. There's too much 2 there for that. So we are going to have to do some very 3 careful sorting and planning as to how we are going to 4 tackle all of that.
- 5 The next thing is that we have a couple of other 6 efforts that are getting under way in the division that are 7 also going to impact on whatever we come out with, one of 8 which is an effort to identify site-specific concerns so 9 that we'd be working from the site level to identify things.
I 10 We are going to need to be sure that the two activities, 11 the generic activity and the site-specific activity are 12 going to meet.
13 Secondly, there is another activity that we are 14 calling a " requirements analysis," that is under way, that 15 is looking at the entire Part 60, not just the technical 16 part. Starting from the level of what the specific 17 requirements are in the reg, trying to identify what 4
18 criteria may need to be or areas, really, where criteria f
19 may need to be developed and what connections there are 20 between the various findings.
21 This is an activity that's getting started now.
22 It will probably move along fairly rapidly and we'll come i
23 back and talk with you about that one as we've made more 24 progress on it.
25 Finally, once we get all the plans in place, we f
(:)
/\CEJFEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
l i
29879.0- 214 BRT
() 1 are going to want to develop an automated means of 2 monitoring and updating compliance.
3 MR. MOELLER: On the identification of these 4 open items and resolution and assigning priorities and so 5 forth to them, do you plan to use the licensing support 6 system? Will you draw on that computer system?
7 MR. COPLAN: I expect that we will. We'll be 8 feeding information into it. Once the system is fully 9 operational it will make available to us information that 10 the Department of Energy is going to be relying on and 11 other potential parties to a proceeding.
12 MR. MOELLER: In the material we reviewed for 13 the meeting, one was this NRC's role under the Nuclear l 14 Waste Policy Act which I find very helpful, where you cite 15 the sections and what you are doing.
16 There's one item in there that I had not really 17 noted too much previously, and that was provision number 46, 18 and it says, "Section 223(d) requires the U.S. to provide
,. 19 -- to be prepared to cooperate and provide technical l
20 assistance to nonnuclear weapon states in the field of 21 spent fuel storage and disposal."
s 22 What are you doing on that?
23 MR. COPLAN: I'm going to refer that one to Joe s
24 Bunting.
i 25 MR. MOELLER: Does this mean you won't help a O
/\CEJFEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6 l
\ --
29879.0 215 BRT
() I nuclear weapon state, I guess like France or U.K.? John?
2 Go ahead, Joe.
3 MR. BUNTING: I'm going to try and pinch hit.
4 I'm not the expert in this area, but I believe in the 5 Federal Register, right after the Act was passed, the 6 Office of Research has lead on this area and I believe that 7 notices were published in the Federal Register to indicate 8 the kinds of services that the Commission had available to 9 provide in this area. To my knowledge, that's what we have 10 done so far.
11 MR. MOELLER: It says here, "a Federal Register 12 notice was published following coordination with DOE, ACDA, 13 and the State Department on March 30, 1983, and then there p%)
14 have been annual updates, 4/6 of '84 and 4/5 of '85.
s 15 MR. BUNTING: That's correct.
16 MR. MOELLER: And 4/3 of '86. I guess if we are 17 interested we should look it up.
18 But does it mean you won't help nuclear weapon
["'c i
19 states? Or you are not required to?
20 MR. BUNTING: That's how I read it.
21 MR. MOELLER: Do we have other questions on the 22 five-year plan? John?
23 MR. LINEHAN: There is one thing I would like to d, 24 add regarding the DOE five-year slip and the effect on our 25 five-year plan. The plan itself is very independent of DOE's
(
,s ACE FEDERAL REPORTERS, INC.
( ,
202-347-3700 Nationwide Coserage 800-3364M6
[..'
'p. -
I 29879.0 216
~A 5BRT
/
-- A
(_) 1 schedules. What it does is lay out what we need to do 2 prior to the receipt of the license application and if you 3 - look at the activities that we lay out there, development 4 of guidance, development of independent Staff review 5 capabilities, development of methodologies for resolving 6 issues, I believe about 2/3 of our resources are spent on
. 7 activities that would be NRC initiatives. The 1/3 would 8 rely on activities that would be DOE initiatives, such as 9 review of DOE site characterization plans, things of that 10 nature. So it is very independent.
11 That was purposely done because of the schedule 12 slippages that had occurred in the past and we realized to 13 get our own program in order we had to get something that 14 was independent of the slips that may have occurred, so the 15 . bottom line is we really don't see any significant impact.
16 MR. MOELLER: Are there other questions or 17 comments on this?
18 Well, thank you. I know it has been a long day 19 and we appreciate your coming down and meeting with us and 20 presenting this and other topics.
21 I think with that, then, we'll wrap up the 22 formal Subcommittee meeting and, following this formal 23 coscion we, of course, will go into executive session, 24 still open to the public, in which we will address two 25 items. I would like for us to try to see what we can do l
l l
/\CE FEDERAL REPORTERS, lNC.
202-347-3700 Nationwide Coverage 800-336-6646
29879.0 217 BRT l) I with the nine questions which we are supposed to address in 2 terms of the definition of high-level waste and we will -
3 distribute those nine questions for you to look at, and we 4 will also be looking at the draft document on the 5 comparison of the risk of a nuclear power plant and 6 high-level waste repository.
7 I note here, Owen has reminded me that. John 8 Linehan'wants to speak for a few minutes on the mission 9 plan.
10 MR. LINEHAN: Time permitting it was just 11 responding to a request you had made to talk in general 12 about things that may be going on.
13 ' ' MR. MOELLER: All right. Go ahead and do that-
. O 14 and then we'll go into the executive session.
15 (Discussion off the record.)
16 MR. MOELLER: The suggestion is made that we'll i 17 handle this, then, in the execntive session, so we'll do 18 that. With that then we'll recess the formal meeting andf 19 we'll reconvene tomorrow morning at 8:30. -
20 (Whereupon, at 4:15 p.m., the hearing was 21 adjourned, to reconvene at 8:30 a.m., February 20 1987.)
22 23 i 24 25 +
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
. -,,, _ - . . . - _ . . _ , . . _ . , _ _ . - _ - . . - _ , . - - - - _ . _ _ _ . _ - - - . _ . - - _ . _ . . _ - . _ ~ _ _ _ _ _ - - . - .- . - _ - . . . _ , , - , . - . , - - - - ,
I CERTIFICATE OF OFFICIAL REPORTER O
This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the matter of:
NAME OF PROCEEDING: ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON WASTE MANAGEMENT DOCKET NO.:
PLACE: WASHINGTON, D. C.
DATE: THURSDAY,' FEBRUARY 19, 1987 were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission.
(sigt) w (TYPED)
JOEL EITNER Official Reporter ACE-FEDERAL REPORTERS, INC.
Reporter's Affiliation l
- O
9 O
ADVANCE NOTICE OF PROPOSED RULEMAKING (ANPR)
DEFINITION OF "HIGH-LEVEL RADIOACTIVE WASTE" O
Daniel J. Fehringer February 19, 1987 O
1 I
O
\J ADVANCE NOTICE OF PROPOSED RULEMAKING (ANPR)
DEFINITION CF "HIGH-LEVEL RADI0 ACTIVE WASTE" PURPOSE REVISE THE DEFINITION OF "HIGH-LEVEL RADI0 ACTIVE WASTE" (HLW) IN 10 CFR PART 60 TO CONFORM TO THE DEFINITION OF HLW IN THE NUCLEAR WASTE POLICY ACT OF 1982.
NUCLEAR WASTE POLICY ACT OF 1982 "HIGH-LEVEL RADI0 ACTIVE WASTE" MEANS:
(A) THE HIGHLY RADI0 ACTIVE MATERIAL RESULTING FROM THE REPROCESSING 0F SPENT NUCLEAR FUEL, INCLUDING LIQUID WASTE PRODUCED DIRECTLY IN REPROCESSING AND ANY SOLID MATERIAL DERIVED FROM SUCH LIQUID WASTE THAT CONTAINS FISSION PRODUCTS IN SUFFICIENT CONCENTRATIONS; AND (B) OTHER HIGHLY RADI0 ACTIVE MATERIAL THAT THE COMMISSION, CONSISTENT WITH EXISTING LAW, DETERMINES BY RULE REQUIRES PERMANENT ISOLATION.
CURRENT PART 60 DEFINITION "HIGH-LEVEL RADI0 ACTIVE WASTE" OR "HLW" MEANS: (1) IRRADIATED REACTOR FUEL, (2) LIQUID WASTES RESULTING FROM THE OPERATION OF THE FIRST CYCLE S0LVENT EXTRACTION SYSTEM, OR EQUIVALENT, AND THE CONCENTRATED WASTES FROM SUBSEQUENT CYCLES, OR EQUIVALENT, IN A FACILITY FOR REPROCESSING IRRADIATED REACTOR FUEL, AND (3) SOLIDS INTO WHICH SUCH LIQUID WASTES HAVE BEEN CONVERTED.
O
l k
I '
WHAT IS THE PROBLEh, i 1. THE-CURRENT HLW DEFINITION IS BASED ON THE SOURCE RATHER THAN THE HAZARD OF A WASTE. A RISK-BASED DEFINITION WOULD IMPROVE THE j MATCH BETWEEN WASTES AND DISPOSAL FACILITIES.
- 2. A NORE PRECISE DEFINITION OF HLW IS NEEDED T0:
--IDENTIFY THE NEED FOR WASTE GENERATORS TO ENTER INTO CONTRACTS FOR TRANSFER OF HLW TO 00E. -
() --ALLOW DOE TO PLAN FOR RECEIPT OF WASTES.
~
--DETERMINE WHICH OF THE NRC'S REGULATORY REQUIREMENTS APPLY TO SPECIFIC TYPES OF WASTES.
I 1
[
I I
l lO t
i l
r r.... . + - ,9 .+-wr,, y9..,m.,_.-,.,9_ _%ye. ,,r.,,- , . , . _m_ ..m,__
a
-3
~- / ,
LOW-LEVEL RADI0 ACTIVE WASTE POLICY AMENDMENTS ACT 4
"' LOW-LEVEL RADI0 ACTIVE WASTE' MEANS'RADI0 ACTIVE NATERIAL THAT IS NOT HIGH-LEVEL RADI0 ACTIVE WASTE, SPENT NUCLEAR FUEL, OR [ MILL TAILINGS]."
BECAUSE OF THIS WORDING, A DEFINITION CF HIGH-LEVEL HASTE ALSO SERVES
.T0 CLASSIFY LOW-LEVEL WASTES.
THE ACT ESTABLISHES RESPONSIBILITIES FOR RADI0 ACTIVE WASTE DISPOSAL:
i
()
\_/
FEDERAL GOVERNMENT FOR "AB0VE CLASS C" STATES FOR CLASS C AND BELOW.
NO CHANGES IN RESPONSIBILITIES WOULD RESULT FROM A RULEMAKING TO DEFINE
^
HIGH-LEVEL WASTE.
4 i
a O
r
- f.- v' -
,.e-. ,-r ,-c- - b _ , ffm . qw gm -- ,--,.--m%.- --- .-------%-r- -. , --en-,--..-,,-.,#, w,--v---, - - . . _ , . .---- - - . , -, , -- - + - -.---e,------e.---
p.___.... . . _ . . . . _ _ . . . . . . _ . . . - . - . _ . . _ _ . . _ - . _ . - . _ . . . _ . _ - _ _ - . _ _ . . _ _ . . . . . . _ , . - . . . . _ . _ _ , . . _ .- ,
J
~ ~
O O O 4
- i 4
i -
1 . ,
I LOW-LEVEL IF 1 HLW IF FROM l HIGH-LEVEL !
j NON-REPROCESSING l REPROCESSING l
j "ABOVF CLASS C"-
l j i
- l l L j t CLASS C CLASS C ;
I INCREASING
, CONCENTRATION ,
! CLASS B CLASS B i
CLASS A CLASS A ;
l l
I
! CURRENT WASTE CLASSIFICATIONS. DESIRED WASTE CLASSIFICATIONS.
1 1
(WASTES OTHER THAN SPENT FUEL) (NOTE: NO REVISIONS TO CLASSES A, B, & C)'
l t
i
\
4
^
i l - - _. - - - ,. - - , . , _ _ _
5 (3
t/
WASTE I SPECIFIC l CURRENTLY l DISPOSAL CLASSES l WASTE TYPES l DEFINED l RESPONS,IBILITY l I I I I I I SPENT FUEL ( l i I PART 60 i HIGH-LEVEL l REPROCESSING l l FEDERAL GOV'T I WASTES I l
.._________9_____________------t--------------t l l UNDEFINED I
- ___"AB0VE CLASS C"__9
_____________9.________________
I l I LOW-LEVEL l CLASS C l l l 1 I l CLASS B l PART 61 l STATE GOV'TS I I I l CLASS A i l O
L) o
J f
O WHAT IS OUR APPROACH FOR RESOLVING THE PROBLEM?
THROUGH A RULEMAKING, DEVELOP A NUMERICAL DEFINITION OF "HIGH-LEVEL KASTE."
OBJECTIVES OF RULEMAKING --
- 1. PLACE THE DEFINITION OF HLW ON A SOUND TECHNICAL BASIS WHICH RELATES WASTE CLASSIFICATIONS TO WASTE HAZARDS.
- 2. NOTIFY WASTE GENERATORS OF NEED TO ENTER CONTRACTS FOR TRANSFER OF
~
WASTES TO D0E.
- 3. FACILITATE PLANNING BY DOE FOR RECEIPT AND DISPOSAL OF WASTES.
- 4. MAKE CLEAR WHICH OF NRC'S REGULATIONS APPLY FOR DISPOSAL OF DIFFERENT TYPES OF WASTES.
0 REQUIREMENT (IN NRC REGULATIONS) FOR DISPOSAL OF HLW IN A GEOLOGIC N_0 REPOSITORY. IF CONGRESS AUTHORIZES, OTHER DISPOSAL METHODS COULD BE USED.
O
()
l l l l Concentrations l l of Short-Lived l l Radionuclides l l l Low-Level Waste l High-Level Waste l -Above Class C 1 -Exceeds concentrations l " Highly Radioactive" l of Tables 1 & 2 l -Example: Cs & l -Both " Highly Radioactive" l Sr Capsules l and " Requires Permanent l l Isolation" Table 2 from l l 10 CFR 61 l l l l l l l l l l l l l Low-Level Waste l Low-Level Waste l -Classes A, B, &C 1 -Above Class C l of 10 CFR 61 l " Requires Permanent Isolation" l -Neither " Highly l -Example: TRU weste l Radioactive" nor l l " Requires Permanent l l Isolation" l l l Table 1 from Concentrations 10 CFR 61 of Long-Lived Radionuclides CONCEPTUAL DEFINITION OF HLW INCLUDED IN ANPR.
7
'NJ V V l l
,r' l Cs & Sr Capsules l Spent Fuel &
Concentrations l l Commercial i Reprocessing l Wastes i of Short-Lived l l '
Radionuclides l l l Low-Level Waste i High-Level Waste /
I I ._ -
, . - -x l l , .
l l ' SRP & WV JGlass Table 2 from 10 CFR 61 I
l l /l D2/
i
.' 'N
]
l l
Hadford/ Tanks \
( l/ TRU Wastes \
l \ l i I l Low-Level Waste 't. Low-Level Waste l 1 l 1:
I SRP & WV Salts ll l l l h l l ) l l v I Table 1 from Concentrations 10 CFR 61 of Long-Lived Radionuclides EXAMPLES OF WASTE CLASSIFICATIONS WITH CONCEPTUAL DEFINITION OF ANPR 7a
q 8 O
APPROACH FOR IDENTIFYING CONCENTRATION LEVELS TO DEFINE HLW.
- 1. ESTABLISH ACCEPTANCE CRITERIA (INDIVIDUAL DOSE, POPULATION IMPACTS,ETC.).
- 2. DESCRIBE FEATURES OF " INTERMEDIATE" DISPOSAL FACILITIES.
- 3. CHARACTERIZE WASTES (FORMS, CONCENTRATIONS, VOLUMES, ETC.).
- 4. DEVELOP (OR ADAPT EXISTING) ASSESSMENT METHODOLOGY.
- 5. EVALUATE DISPOSAL FACILITY PERFORMANCE. WASTES WHICH CANNOT SAFELY BE DISPOSED OF IN " INTERMEDIATE" FACILITIES WILL BE .
MAJOR POINT -- IT IS NOT SUFFICIENT TC EXAMINE THE WASTE ALONE.
WASTE CLASSIFICATION IS INTENDED TO PROVIDE A MATCH BETWEEN WASTE HAZARDS AND DISPOSAL FACILITY CAPABILITIES. THEREFORE, A WA5TE CLASSIFICATION SYSTEM MUST CONSIDER THE RISK OF THE WASTE AFTER DISPOSAL IN VARIOUS TYPES OF FACILITIES.
I
{
!O l'
-e b.
ENERGY REORGANIZATION ACT OF 1974 PROVIDES NRC LICENSING AND REGULATORY AUTHORITY FOR:
l (3) FACILITIES USED PRIMARILY FOR THE RECEIPT AND STORAGE OF HIGH-LEVEL RADI0 ACTIVE WASTES RESULTING FROM ACTIVITIES LICENSED UNDER [THE ATOMIC ENERGY] ACT.
(4) RETRIEVABLE SURFACE STORAGE FACILITIES AND OTHER FACILITIES AUTHORIZED FOR THE EXPRESS PURPOSE OF SUBSEQUENT LORG-TERM STORAGE OF HIGH-LEVEL RADI0 ACTIVE WASTE GENERATED BY THE [D0E],
WHICH ARE NOT USED FOR, OR ARE PART OF, RESEARCH AND DEVELOPMENT ACTIVITIES.
O
10
,-k OTHER ISSUES
- 1. EFFECTS ON DOE DEFENSE-0NL) FACILITIES. NO CONGRESSIONAL INTENT TO IMPOSE NRC JURISDICTION FOR WIPP. NEVERTHELESS, SOME " REMOTELY-HANDLED" TRU WASTES INTENDED FOR WIPP WOULD APPEAR TO BE hlGH-LEVEL WASTES USING TPE PROPOSED CLASSIFICATION. MANY " TANK" WASTES AT HANFORD WOULD LIKELY BE LLW UNDER THE PROPOSED CLASSIFICATION, BUT NRC LICENSING JURISDICTION MIGHT STILL REMAIN DUE TO THE WORDIhG 0F THE ENERGY REORGANIZATION ACT. .
O 2. EFFECTS ON NUCLEAR WASTE FUND. MINOR DIFFICULTIES RELATED TO TIMING OF CONTRACTS AND PAYMENTS IhTO THE WASTE FUND MAY REQUIRE REMEDIAL LEGISLATION.
- 3. EFFECTS ON WASTE CONFIDENCE RULEMAKING. SP0ULD BE FAVORABLE BY HELPING TO ASSURE THAT PLANNING IS INITIATED FOR DISPOSAL OF ALL WASTES, REGARDLESS OF SOURCE.
- 4. NATURALLY-0CCURRING AND ACCELERATOR-PRODUCED MATERIALS. NO CHANGE.
O
s O
n: ...
, 2./ ; 3 a..
., . . <. . . :. .,, n, s u. . -. " ..gn.-,ws.
. . . (.). , o:.
.-..c..
J r.2.-...
...'r.,
..;,, sa) . -- . . < . . . ' -.
.-2 ^' - . 2
.-> u . . .s -r . . .
.i-
".-.#. .ej.'=.9 c 2.'..f ,c-I "... , c--". *.2 . '. .r " .=."a wa .=.2 '...-.
s.1.-=... a. c.. . m.. c r. . . . . .
s ,.,,, ..,.. , . 2..-g ,.a.+.'.- ,,...2, ': - e-r -% 4- '.- v c l e r. . , i. '. a #. . - i 'i . .,> ',s'-
r 2. ,,, ,, . 2. 2. . < i.y :
i r r. a ". ' u . =. - .-.2 . .- . r f U t:- 1, ( .', ) s o l "i s i r . .- #. . 'i 69 e. .,. 1 4 ,, ~ i .
- e. .....3.", * , .A p,J. . 2g}'Ma.a p c " { veg .= *. j . .. a. ,
- v. .g3 - . o.*aEi f a a* g o e. . e p
.. 3 ,- .,. 3 . g
- 3 ic s.-s. a. .s a,. r. . .
- a. .2 ~. ; ,.. .. _. ., .. , r ... ,... . .,. . .,. 3 9 n -c.. g. . c . , - l i v =. d r _> " i a,, -
.. m. a ., s ...
- 3. ... . ,, s .
1 a a, d, s ", .o r. +.- 1 4 v e do
- r. ,. '. *. .2 , ', r, - . * . a. . a . '. .- s e x . v. " i m. , _". a. v a l u a. s o f T,2.b i a, e .m. - - . . i. s. s. -
.- i. ~- . . .- . 2. . . . . ' s . s
. ~
- a. v. .- a. s. ". i. . 3+r=.
~
. v'iuas e'e. T c~s'l a. 2.
.- . 1 ia, .
.u.0 .
1c...s C o n.n a. n..t , . i c n t
. ,2 .
, - s s : ..< ,,iv.jia. (c,1/, )
- i . . -
e .,. a. :... .. ..3. ,. 4.-.3..
w ... .
_I
.s C-1* 3 Ni-63 700 c_
- 4. _.. 3 i r, m c +. .
s "e .'.
e . : 4 ,, ,. . . . , ..
O-
. c. ... ,1 .. 3.
. .a1
,m 7u m0 til _:-3 i,. u,_... u-. .. _ m.
- e. 3
+ : 2 ,.
cs ia,/ *..e,- u
, " , . ... _ .- 4 . . c .- . . 0.2
.2
.. . d :.
, . _ . a. .;
i t. , a. 0.0:
, ., ,, a . o. .,. .r .. .. ; -, ,,
inmJ, .3 .
o ,,, r i. .n n t a
P.u _9 ,i .s.au-^02
- ,,, en --
t-... o,,u J.-
% tes for Tables 1 ar.d 2.
'i s - y- a. s a. . , , a ,~ n .'. . ' . - s
' '. '. a m.1 . x
- U r a. of r1 .'..1'
, a.-> 3
- 'e.
2 y= : } a. { s
- o h o. g-4 9. ,- . -.n . ...3 . l. a. .
i - a.
. ....s.e .
. - n. i .:. .3..a h
.m -. :
, wj ;.. *44 ., .,
.-4 ,
, . .'r . 2.. . j ,s , .2
. . .. 4 J. 3 4
.. a .. . .2. .2 . . .
= . .. -,.*., e. n.,a .., 3*.. a.
, a. .m
.. .~.3 J a-._. .
- E
. ...m. a.
g
.. c. g . ,., 3. 4. e3 s e a. 5. .s= e.
.4 .
$' 2 ..* .JJl ,.. 7 # .a.
. ,**9,
- . *. ..
- d.
f* *P 1 -*.. . * ,,[.
'.g,,'p P,9 . ..
O 19
..o .
l
/
i
/
m M $
8 e
C4 W
H-b LL.
I
~
e s aw'"
J
- tJ
'J '
4 l
Ia
-4
+
W b.
C; c.
r-.
Y D d.'
% .- i W'.
5a
% _w
.g d
%* s no E, W -
9
OVERVIEW e
O WHY PERFORMANCE ALLOCATION IS NEEDED A TOOL FOR SITE CHARACTERIZATION PLANNING O WHAT IS PERFORMANCE ALLOCATION DEFINITION APPROACH EXAMPLE ,
O DOE'S PLANNED APPROACH TO PERFORMANCE ALLOCATION DOE'S DRAFT MISSION PLAN AMENDMENT PLANNED MARCH 1987 GENERIC MEETING ACRS BRIEFING 2 87/02/18 O O O
WHY PERFORMANCE ALLOCATION IS NEEDED O TECHNICAL - MANAGEMENT TOOL TO HELP DOE DESIGN ITS SITE l TESTING PROGRAM PART 60 PERFORMANCE OBJECTIVES GIVE DOE FLEXIBILITY IN CHOOSING WAYS TO MEET OBJECTIVES AT DIFFERENT SITES PERFORMANCE ALLOCATION MAINTAINS FLEXIBILITY FOR TAKING ADVANTAGE OF SITE FEATURES TO TAILOR DESIGN AND PROGRAM TO A SPECIFIC SITE PROVIDES STRUCTURE FOR A SITE TEST PROGRAM HOW MUCH CREDIT DOE WANTS TO TAKE FOR COMPONENTS RETAINS INTEGRATED SYSTEMS VIEW ASSURES ENOUGH STUDY TO UNDERSTAND SYSTEM AND AVOID DELETERIOUS EFFECTS l
ACRS BRIFFING 3 87/02/18 O O O
WHAT IS PERFORMANCE ALLOCATION DEFINITIONS PERFORMANCE ALLOCATION:
SYSTEMATIC, ITERATIVE PROCESS OF ASSIGNING TENTATIVE PERFORMANCE GOALS AND ASSOCIATED DESIRED CONFIDENCE LEVELS FOR THE HLW DISPOSAL SYSTEMS, SUBSYSTEMS AND COMPONENTS PERFORMANCE GOALS:
NUMERICAL VALUES OF PERFORMANCE TENTATIVELY ASSIGNED TO PERFORMANCE MEASURES OF HLW DISPOSAL SYSTEMS, SUBSYSTEMS AND CGMPOhENTS ACHIEVEMENT EVALUATED USING PERFORMANCE ASSESSMENTS, PRIMARILY TO GUIDE TESTING',AND DESIGN ACRS BRIEFING 4 87/02/18 9 9 O
DEFINITIONS (CONTINUED) ,
DESIRED CONFIDENCE LEVEL THE ' PROBABILITY THAT THE ACTUAL VALUE OF AN UNCERTAIN PARAMETER OR PERFORMANCE MEASURE IS AT LEAST AS FAVORABLE AS THE PERFORMANCE GOAL O EXPRESSES HOW WELL DOE THINKS IT NEEDS TO MEET A TENTATIVE PERFORMANCE GOAL O SHOULD BE STATED IN QUANTITATIVE TERMS, IF POSSIBLE AND ON THE MOST DEFENSIBLE BASIS O NEED NOT BE A STATISTICALLY MEANINGFUL CONFIDENCE LEVEL O MAY BE SET BY EXPERT JUDGMENT O MAY BE STATED AS "HIGH," " MEDIUM" OR " LOW" PROVIDED THE TERMS ARE DEFINED ACRS BRIEFING 5 87/02/18 O O O
1 APPROACH AS AGREED TO BY DOE O A PERFORMANCE ALLOCATION SHOULD BE DEVELOPED AS EARLY AS POSSIBLE O THIS PERFORMANCE ALLOCATION SHOULD SPECIFY:
PARTICULAR BARRIERS RELIED UPON LEVEL OF PERFORMANCE FOR EACH BARRIER DESIRED LEVEL OF CONFIDENCE FOR LEVEL OF PERFORMANCE ACRS BRIEFING 6 87/02/18 O O 9
APPROACH (CONTINUED)
O ESTABLISHES INITIAL ALLOCATION BY TECHNICAL-MANAGEMENT JUDGMENT O INCLUDES CONSERVATISM NEEDED TO ACCOUNT FOR UNCERTAINTIES GOALS FOR RESERVE COMPONENTS GOALS SET HIGHER THAN NEEDED 0 PRESERVES DOE FLEXIBILITY TO DETERMINE WHICH COMPONENTS WILL BE RELIED ON, AND FOR HOW MUCH PERFORMANCE O THE PERFORMANCE ALLOCATION SHOULD'BE REVISED PERIODICALLY TO REFLECT SC TEST RESULTS. THIS WILL BE DONE IN BIANNUAL SCP UPDATES ACRS BRIEFING 7 87/02/18 O 9 O
HYPOTHETl CAL EXAMPLE BARRIER / PERFORMANCE CONFIDENCE COMPONENT TARGET LEVEL CONTAINMENT WASTE PACKAGE / 1000 YEARS A%
TIME CANISTER RELEASE RATE EBS/ WASTE FORM 10 -5 / YEAR 8%
GWTT, POST CLOSURE UNSATURATED ZONE 5000 YEARS C%
EPA STANDARD ABOVE TARGETS X-63 15 D%
AND THESE Y-127 25 RETARDATION FACTORS Z-249 5 ACRS BRIEFING 8 87/02/18 O S O
o o o
~
i i
}
i IMPORTANCE FOR TEST PROGRAM EXAMPLES OF SELECTING TESTS I
l
~0 GOAL -- CONDUCTIVITY LESS THAN 10 ESTABLISH THROUGH THE PERFORMANCE ALLOCATION
- CONFIDENCE LEVEL -- 90% PROCESS WHAT THE PERFORMANCE GOAL AND DESIRED CONFIDENCE j LEVEL ARE i TWO TEST PROGRAMS LOOK AT AVAI LABLE TEST-l PROGRAMS TO DETERMINE
! TEST PROGRAM A -- +2 ORDERS OF MAGNITUDE WHICH TEST PROGRAM IS l TEST PROGRAM B -- Il ORDER OF MAGNITUDE SUITABLE
]
l j IF PRELIMINARY ESTIMATE OF CONDUCTIVITY IS LOOK AT PRELIMINARY i
_yy ESTIMATE OF PARAMETER CASE 1 10 -10, EST PROGRAM A l CASE 2 USE TEST PROGRAM B CASE 3 3X 3 x 10_g 10 ,USE NEITHER
)j ,
i
't l ACRS BRIEFING 9 87/02/18
O O
' ~
CONCEPTUAL L USTRA TI ON i
1 i
i CASE 1 CAs e' 2 CASE 3 i
2 A 6
. -s -r Ge<i to.r - - - - - - - - - -
ca l ts ---.- - - - - - - - - - - - Goal,so --- - - - ---- ------
h a! .q
.. A a sxio Ah!Jy8 ,. ,
)
, A s
)
! kbl 10
m ,
i l
I i BhRs = 80% Car,Rdence l~nterazir (Er<or Bands.) '
)
(l e catwde
, to(; chuce either bmd o-F being) 4
-I ,
9a -
I f
I .
7, . , .
i O .
O ,
O !
4 1 P
(
l -
+
t DOE'S PLANNED APPROACH TO PERFORMANCE ALLOCATION
- l I i O DOE'S DRAFT MISSION PLAN AMENDMENT f
O PLANNED MARCH 1987 GENERIC MEETING i t
I I
f i
i
, p
,s
_e'-
e 6
ea '
- i 1 .
l ACRS BRIEFING 10 87/02/18
- i' <
.i ,
.. _[ .-
I __-._ _
-..,..-_,I. ,. -
- fl = . -- -
IL -
1s 1
g DEVELOP SVSTEM IDENTIFY REGULATORY I
- DESCRIPTION REOulREMENTS O.. . $h l
. EB 't b
t' l '-
9 DEFINE ISSUES s.
- a U.
s if 1I g
A ) L 3
SET LICENSING STRATEGY 1 f 8
4 IDENTIFY INFORMATION NEEDS:
g PERFORMANCE MEASURES. IDENTIFY PARAMETERS. SET A { SET PERFORMANCE " GOALS" AND PARAMETER " GOALS." AND SET " INDICATIONS OF SET " INDICATIONS OF wa CONFIDENCE" CONFIDENCE" a.
W 3r e
DEVELOP TESTING STRATEGY.
IDENTIFY TESTS VARIASLES.
' AND PARAMETERS TO SE MEASURED g4 ' 1 f 1f s' y g
CONDUCT INVESTIG ATIONS ;
II _ _ _
6 ANALYZE RESULTS lf ESTABLISH TH AT INFORM ATION NEEDS ARE SATISFIED if
'O USE INFORM ATION TO R ESOLVE ISSUES 1r 11 s
DOCUMENT RESOLUTION
- () Figure A 2. Issue resolution strategy.
Souhce : 1) rah
- Missicn Plu A mud meZ, u s Do E, Juua *U 19 E '
to a
_ . ]
l %.) {
1 1
i l DiVlS1ON OF WASTE MANAGEMENT l
l HIGH LEVEL WASTE TRANSITIONAL LICENSING l
l' SUPPORT SYSTEM (TLSS)
I, STATUS BRIEFING TO
- THE ACRS FEBRUARY 19, 1987 i . .
I i
i I
h 4
l l
i f JOSEPH BUNTING, BRANCH CHIEF PHILIP M. ALTOMARE, SECTION LEADER AVI BENDER, SENIOR PROJECT MANAGER i '
l I l /~gn avego, 1 87/02/18
f Q .ym p.
a U u 4
LICENSING SUPPORT SYSTEM (LSS)
HISTORICAL PERSPECTIVE 4
9 NRC CANNOT ACillEVE A 3 YEAR LICENSE REVIEW REQUIREMENT WITHOUT AN INFORMATION/ RECORD MANAGEMENT SYSlEM 9 NRC I)OL . SIATL AND 1RIBES MEI TO DISCUSS NEED FOR INFORMAT1ON MANAGEMENT SYSTEMS 1988e-1986 e NRt INillAfl0 A PILOT PROJECT IN 198', TO DEMONSTRATL CONCEPT TO DOE e D< >l ALRLiD TO DEVELOP LSS (1/86) AND AN INTERAGENCY COORDINATING COMMITTEE WAS ESTABLISHED IORUM FOR PUBLIC PARTICIPATION DLFINE LSS ISSUES e Dof. 10 RELEASE RFP FEBRUARY 1987 0 NIGOTIAILD RULEMAKlNG ADDRESS AND RESOLVE LSS ISSUES 1INAL DEF iNI TION ON NATURE AND SCOPE OF THE LSS
- l tiLW/87/02/18 2 87/02/18 8
~ ~ ~ ~
Q? 'O C' "
SAMPLE LSS ISSUES TO BE ADDRESSED BY NEGOTIATED RULEMAKING (51 FED. REG. 45342)
O CATEGORIES OF INFORMATION RELEVANT TO HLW LICENSING DECISION O TIMEFRAME O RULES FOR PRIVILEGED INFORMATION O LSS ADMINISTRATION O WilEN/WHAI SHOULD BE ENTERED?
O PROCEDURES t
O SANCTIONS FOR ENSURING DOCUMENTS GET INTO LSS O FORMAT O SHOULD ALL DOCUMENTS BE ENTERED IN FULL TEXT?
e llOW TO HANDLE ELECTRONIC SUBMISSION OF DOCUMENTS?
I t
l l
l HLW TLSS 3 87/02/18
~ ~ '
c, , , , . . , ,
MAJOR TECHNICAL CHALLENGES O STREAMLINING THE DATABASE CREATION PROCESS O PROVIDING AN " INTERIM" SYSTEM TO STORE / RETRIEVE DOCUMENTS NRC TLSS DOE SALT DIFFRENT RECORD MANAGEMENT PROCEDURES DOE BWIP
STATE / TRIBAL SYTEMS i
1 O KEEPING UP WITH THE EXPONEN11AL PROLIFERATION OF PAPER i
4 i
- O OTHER CHALLENGES . . . . . TO BE ADDRESSED BY NEGOTIATED RULEMAKING i
l l
l HLW/87/02/18 5 87/02/18 l
l
o Os i WASTE MANAGEMENT DOCUMENT ACQUISITIONS i 1981 TO 1999 LEGEND 480- -
CU MULATIVE l
c <. a ACQUISITIONS 8
350- -
m g300-
- =
250- -
- 6. . .
5 g ,.o. .
z 100- -
7 l7 . . . . . ..
81828384858687885990919293949596979899 YEAR
q.
6.i n e,e e. . 4.5thchii ub4Fdth CS.O cataAttM '- .
a '
'?
t l'
, NRC TLSS AS OF TODAY ,
i i l
I e FULL TEXT SEARCH / RETRIEVAL - OPERATIONAL 1000 HLW DOCUMENTS 900 CONGRESSIONAL Q'S AND A'S 10CFR 60 ACCESS TO DATABASE e --
LOCAL PC'S REMOTE (WITH MODEM)
"RLAL ilME CAPTURE" 9 IMAGL CAPTURE AND RETRIEVAL - START MARCH 1987 4 PROIOIYPE OF IMAGE / TEXT OPTICAL DISC SYSTEM IS BEING DEMONSTRATED HLW/87/02/18 6 87/02/18
!I 1
I 1B'1 INFOR'IAT10!1 NETWORK REr10TE ACCESS TO -
ONI.If1E FULL TEXT SEARCH DATABASE
.M PC'S 3278' S IBM 5520'S IASER PRINTER j m -l FM -
n
/ 3270 ACCESS WORK STATIONS
- CONT 901.LD f ' ' --' r- '
RDIOTE PC s 4
oo oo _s "L*= ,$ L R h ,$"L R '
< /
s:""
/ }
Vu s
],
/ / '
sx1mn f=3 if
/
l =-- I -O-- 1 -
l 7m " ,
ItOST WORK STATION 200 MB SCANNER LASER OPTICAL HARD DISC PRINTER DISC STORAGE OPTICAL DISC l'1 AGE AilD FULL TEXT SEARCH AllD RETRIEVAL SYSTEM
l l
l O
= =-
sq mrr Portion of original image of a letter m2s W 76 "., N4' 4*"
~ "' ,
I m.t. 7,.xi.""*"IE" a.
.. n ,
3.tJ.rt: 4.el d INJ1BC fl P .f B F.t 8'.I '.' WPMP E".'.7.,'l".E'".'"e!'4ll'7y' 2.
ei.a' "'J'::l.::..l .
!."4,,7, y . :i L.a. ".::.:".a?4."..yl"#' l'0" f
.-.,.....i,.,
O SEARCH llAS MADE /
Ofi THE WORDS I
" Death Valley" l
VIEW FROM A SPLIT SCREEil HIGH RESOLUTI0f1 MO.'ilTOR THE UPPER PORTIO 4 0F T!iE SCREEil SHOWS All ORIGI::AL DOCUME!iT WHILE TiiE LOWER PORTIO.'l SHOWS THE ASCII TEXT. THE ORIGI:;AL IMAGE ".AY BE SEEit COMPLETELY OR MAY BE PRI:lTED. (in this examole the two documents are not related) e n-- - . . . -, , _ _-, . , . - . , _ - . , -, _ _ _ , , . - - , _ , _ . . , _ , - , _ _ _ _ .
~
j l7 0 SAMPLE SEARCH USING FULL TEXT AND HEADER INFORMATION SEARCH STATEMENT RECALL WlTH HEADER RECALL WlTH FULL TEXT ALL CORRESPONDENCE DEALING WlIH VOLCAN.C ACTIVITY AT YUCCA MT. 0 42 ALL CORRESPONDENCE DEALING WITH CURRENT EVALUATION OF TRENCH-14 0 3 MEMO DATED 10/01/85 BY NNWSI ON-SITE REP. I 1 CORRESPONDENCE DEAL 1NG WITH GEOCHEMISTRY ISSUES AT YUCCA 27 50 SEARCHES CONDUCTED WlTHOUT BENEFIT OR SYNONYM OF THESAURUS F1LE SEARCH OF S00 DOCUMENTS FROM NNWSI DATABASE HLW/87/02/18 7 87/02/18
( .- . _ . . . _ _ _ _ _ -
~
4
- MEAN OF RELEVANT AND TOTAL DOCUMENTS RETRIEVED BY SEVERAL SEARCH METHODS (SOURCE
- CAROL TENOPIR ONLINE REVIEW VOL 9. NO.2 1985)
AVERAGE AVERAGE l TOTAL RELEVANT '
RECALL DOCUMENTS i
FULL TEXT WITH BIBLIOGRAPHIC 21.2 4.5 !
ONLY FULL TEXT 17.8 3.5 l ABSTRACT 2.4 1.0 l CONTROLLED VOCABULARY 3.1 1.2 1
l BIBLIOGRAPHIC WITH ABSTRACT 5.3 2.0
! AND CONTROLLED VOCABULARY I
l l
HLW TLSS 7 87/02/18
'TF W W 1
i l
I l
HOW DO YOU BUILD A DOCUMENT i DATABASE?
l 1
i O REKEY DOCUMENTS COMING INTO DCC i
O USE OPTICAL CHARACTER READER i O CAPTURE DIRECTLY FROM SOURCE I
i I
Op Figure 1 Text and Image Capture Process O O 4
i
- OPTICAL SCAN IMAGE CHARACTER IMAGE
- VERIFICATION
- RECOGNITION i I ASCll TEXT REFORMAT WORD * - TO SOFTWARE PROCESSING
- EDITING /QA/
INDEX SPECIFICATION ASCil TEXT i r g
- CD-ROM CD-ROM :
SEARCHABLE SEARCHABLE A S C ll T E X T IMAGE SYSTEM SYSTEM OPTICAL DISC MAINFRAME ; JUKEBOX
- -O R- - -
TEXT IMAGES I
i 12" OPTICAL DISC
i O) O O ,
I e
- NON-NRC GENERATD NRC GDERATD I
LETTERS TECHN ITItl5 REPORTS MEMOR OTNR ,
h l AM N 5
'cEEi#m"MP@meci
\ 55m m entin G N fl N ! E N l S EttN TER Df' ABASE 3 ROLE NRC DOCKET EW CENTER
s i
DJ O 1
l o
WHERE DO WE GO FROM HERE?
I e NRC TO DEVELOP A TRANSITIONAL LICENSING SUPPORT SYSTEM TO MANAGE ITS OWN RECORDS UNTIL DOE SYSTEM iS AVAILABLE e DOE RELEASE RFP FOR DEVELOPING THE LSS ON FEBRUARY 18, 1987 8 INTERAGENCY COORDINATING COMMITTEE MEETING MARCH 25, 1987 l
e PROPOSED MOU BETWEEN DAVIS-RUSCHE RE: ROLE OF LSS 4 NEGOTIATED RULEMAKING ACTIVITIES PROCEEDING ON SCHEDULE l-t HLW/87/02/18 9 87/02/18 1
.i
./ v O T Y
- 1^ e
, d I E o 4 %
'h 4 h Sh N-e
)b N * %
B i %
i s$sk 5
~
t w
e
\ U k "q 3 t I \ "4 h.m o o
q c s o W
e E
g o
g i o
s N
t 9
(
x y 1
O i '(
% fN i V
y U
i d.i1 i T j
{ i z \*
i t 4 i 4 i
- w. 3 15 1e 4 . c i
k
- R
- g j s ; e S li e s s h
s
,t 3
w 43 4 N $ $ k$ $ +t k e x N
-fo I y s D v t T - g W 4
N 4 e t 0 .
% M {
h d R
7 D
y +-
O c
'$ 2s 4
o
- L4
\
\ N t \20 o D l
Q I N
R S .
e i <e
- \
N D N T
4 9 I' s 2 .s b *,\
3 tQ 3 . y 4'
4 Q
s 5 y 4 o
d S % N s (
4 h 4 N
@ k n, n
- . u I
kk $ . ?
v Q
tt t k
= g s %[*M s u
)k '
l k l d y g R
( <
bds e
ge 0
- 1 u s Oe
_._J
' M*
I t-lO i
Oo "a.s' s
~ e O.
+c .g>
4 t>-
h 9 4
6 +
4 NM
-o o
h ..
- h. f,. 1
. - a 1 3,
t a- -
g l
PN -
- c d
N %O
. ie r -
i
. d %s
.. - iTo
~
(8V '* A g @
i -
- '* 2. 'e. 'e
- C ooo 'o.r vi y 4
"' ##/
- U h Af7'7'7*J/
S f.
O ramemoce n.rsessmea r e Sc wiveo dever o m eur +vo Sce*WAldf 00AJSEQ UCMC6 kSzcsCA M Eevstrwiry +vo L/ocert rsairy Agat.y.ii.r Neruosroey r dou pimact As.s exse ewr O
, vey zu ves bereemwic; Aeoannicines
=
,berewm, air; L% een rear y i
, 0 AA M urji C A- r /0 Al Of OLTA Ah 2' 'T Re6 ut .e ron.y Cem oeia ser
,,, "Sensonnec.c ExpeerA rio4- "
,b No Fox m a t. Mgruoo.c pga Scwf 4
m.=d_J. ~ .-.a - a L -
6 e.
O berce ws; peoa a a m es _
- Ro 4a d'; 4 / N'oc/w/s beJe~ -, sic /f as G,oe < t goik a ~
O Sec her E-/
4
(
O m
~.
t- C $ *C C
(\ 3 u
- h 8 M E u k 5 C m m
O U 3 >
- m D. C H u tc e 9 >_. u k la2 4 C m; h et, GO u u * > 0 a O 9 G G eC m m U m -
U k
E 80 m o
o o
m U m o E C C G D C O C E O 2 .C O m M E
- u u u
- e G C O m a U m O
= * 'w w = = O e
= N z = u en e > c Probabilities can be deter =ined
~
with near certainty .
j fairly accurately h b only with li=ited cenfidence l I [
Probabilities Obtained I Edf's derived fro: local data 1
I u
- Edf's derived fre: regicnal data
-- = :
j 2
~ _
Edf's derived fro: global data j
- 5 g Pdf's or cdf's derived fro: [
l -
theore tical bases :
u - ~~-
O
.c 1.i=ited observation .
C I E=pirical Bayesian =ethods '
u Bayesian methods
.E. -
u g
. Expert opinions :
=
u; -
G Delphi methods :
u -
u Relative probability 3
estimation :
i en Figure E-1. Nomal bases for probabilities assigned to geologic events, r processes, and features of concern to various areas of interest OW; considered in this report.
E-5 8tt.
9 O
. sca n n a ,o unen reary ISSt/E .
$0 M PLC TrA.)Ex.s
- A10de/ Uomrsisry zssu e .- L ad< ef Icuaw kdye e{
phy1,ha / processe s.
O .n-n une a rmory
[g g u g ,* f0.eskm OM.h f S $" A DA L eer h a fies.
rwees cx/vsm pa <~ . s -
2,scosses n n.a + 4"4a e .f .
Aa./y dd n io & 6 /o b Diffeve M 20 hsourios or . % aio/ nona.
Uowersidry &,.
X 4 4 th g-RAJ T.S SA0048sLITY /I /o,
- co yn SL )/Lt. N4s*t*W l' O ~ 0. 0 /
l l
l Alf L/b{Ely @ r jo 'l I
H+ns ar VEky L / WELy <Jo 7 /0 - I fo /o
~
" " ~ ~
\O AJ E6'b1 b) MS Goos Mtlk0d3 he c Mp e+s[ opine &
l b) /I
hr ~ & A aJ t.
Qua,,Apah4 (Pvv
- 4 /'M"J'4) d)
D 6/ Umech,;,,p.
@ Eep a.1e., Aek C n ~ a,,,ea n'4 (Pac .
o
t.
=
~
h O
February 1, 1987
- e+
To: Owen S. Merrill, ACRS Staff ~
From: William E. Kastenberg jgf(/, -
Subject:
Review of " Assessing Compliance with the EPA High Level Waste Standard: An Overview".
follow Following are my comments regarding the subject report. They the questions raised in Al Newsom's letter of December 17, 1986. This report is an edited version of my preliminary review dated January 16, 1987 plus some additional comments.
- 1. Does this report represent a satisfactory first step in addressing this matter? In particular, 1s.the methodology as outlined reasonable and appropriate ?
The methodology presented in the report is composed of four general steps:
a) Scenaro development and screening b) Consequence assessment c) Sensitivity and uncertainty analysis d) Regulatory-compliance assessment Steps a) and b) are based on both the NRC Staff's and their
' contractor's (in this case Sandia National Laboratory) experience with probabilistic risk assessment and related issues. Given the probabilistic nature of the EPA standard, the methodology as outlined appears to be a satisfactory first step. The methodology is reasonable because at present, it may very well be the only way to begin an assessment of this kind.
As discussed below, I have some reservations with the implementation of the method, particularly as it pertains to part c) and d).
- 2. Have the proper questions been addressed? Are there significant ommissions in the report?
In general, most of the significant questions have been raised, but they have not been addressed.
involve The main omaissions sensitivity and uncertainty analysis and their use in assessing regulatory compliance. There is a very nice discussion on sensitivity and uncertainty analysis which reviews and critiques the various methods. It would make a very nice introduction to a paper in a journal. But that is it. There is n'o why, guidence, nor any discussion as to which methods will be used, and how. The example in Section 3 of the report does not (m discuss this at all. Moreover, there is no discussion about how (j uncertainty will be factored into regulatory compliance except to quote from EPA :
It states that because there will be substantial uncertainties in a
projecting disposal-system performance...the Standard requires reasonable expectation...that compliance will be achieved."
The concept of ' reasonable expectation' needs to be addressed by
, _ __-_- ATIA CHMMr& ---- - --- ~
i 3
b S
-) the NRC Staff. 'e
- 3. Could the approach suggested here be applied in a world" situation ? 'real ',
The major problem confronting the application to real world problems is the lack of data for scenario development and screening. This aspect of the methodology requires probabilities for a number of processes (some natural such as i
seismic and climatic and scue human such as future exploration for resources) which are expected to occur over the 10,000 year period required by the Standard. Since many of these processes involve rare events, and the data cannot be collected over such time scales, the probabilities will be determined by a) probability models b) deterministic models c) expert opinion.
The major processes of interest fall into the following categories: hydrology climatology, volcanology, tectonism, seismicity, resource exploration, mining engineering, thermomechanical effects, and geochemistry. Almost all of the i natural processes involve probabilities that require expert l
Opinion for determination of data. (One of the companion reports
[ .
goes into this in more detail).
l
/
k >) Before the methodology can be applied to real world problems an " acceptable data" base should be established, or at least one that is agreeded upon as an initial data base. The report does
, not address how expert opinion will be used to generate the probabilities and I am not aware of any universally accepted method of doing so.
The draft report (NUREG/CR-3964)
- Techniques for Determining Probabilities of Events and Processes Affecting the Performance of Geologic Repositories"- March 1986, is especially usefull in this regard. The authors are able to group the uncertainties associated with geologic phenomena into three groups: 1) those "with high confidence", 2) those that can be determined " fairly accurately" and, 3) those that can be determined "only with limited confidence ( i . e. with inherently large uncertainty)".
I have enclosed a copy of Figure E.1 which gives some indication of both the degree of uncertainty using the nomenclature above and the method by which such probabilities are determined. The latter represents a variety of methods which are both subjective and objective and the probabilities are seen to span the whole range of uncertainty. The report is a good
! starting point for establishing a set of ground rules for qstablishing a data base and can also be used to begin to define how such data can be obtained and used.
- 4. Does the report present any basis for assurance that an s)T actual proposed site can be shown to meet the EPA standards and corresponding NRC requirements with the needed confidence?
Based upon the comments I have made above, it is too premature to
.__ _ _ _ _ _ . . _ _ _ _ _ _ ._-_- m ________ _ __ _ _ _ _ _ _
i I -
O conclude that a basis exists for assurance that a proposed site _
can be shown to acet the standards. Because of the uncertainties and their lack of quantification the needed confidence has met '
been adequately established. A combination of this methodology, engineering judgement and other mechanistic considerations should be used in the near term to evaluate such facilities.
Additional Comments
~
The report uses a number of values for the probabilities of the various process and they generally fall into three categories: 1.0-0.1, 10E-4, and 10E-6-10E-8. Although many of these are stated to be examples only, I am afraid they may become part of the folklore and people will begin to take them for gospel after a while. We have seen this happen too many times in the PRA and Source Term business (dating back to TID 14844). It seems to me that in the formative stages of dealing with the EPA Standard and especially during the screening process, all we can say is that the processes: will happen, are likely to happen, or are very likely not to happen, over the 10,000 year period of consideration. I would feel much more comfortable with a qualitative approach at this stage of the game.
l fm ., Although the examples do give a good illustration of how the
(_) methodology will be used they raise a number of other fundamental questions. One involves scenario development. There seems to be quite a bit of judgement as to how and when some scenarios are combined and/or deleated. There does not appear to be a framework or algorithem for making choices and if not documented clearly, the choices will not lend themselves to scrutability. In this regard the authors point out that some events are not
' mutually exclusive. Moreover, during a 10,000 year period a number of different events can occur, each one leaving the site in a different and perhaps degraded condition. Linear summation may not be appropriate under such conditions.
In summary, the methodology is appropriate at this point in time but the data base (with respect to probabilities) neede considerable work. The Staff must address a number of issues concerning uncertainty, expert opinion and the screening process.
There are a number of other details which need review concerning the treatment of the trees in the example, but these are best left for a meeting with the Staff and its contractors.
t
Se W$
we e
- C & E
- O & **
- we m "
M e w
- u k e E w m
O we U
we
& C b & b tt e T b u k b la2 .c E k 4 W &
U
.Y t
l k -
C O
M os U
a en
.O O
9 E
se ee .Oe u =e C C e 3 C C C S C e .C C =* b* E ** M U U s a G C *C on a u .e C e, we w - e e o e ac @ z = u in e > u Probabilities car. be dete =.ined with near certainty j
[
fairly accurately E 0 '!
=
only with li=ited cenfiden:e I =
Prebabilities Obtained I Edf's derived fro: local data e
Edf's derived frc: re;;ienal data _-
(;u a Edf's derived fro: g1:bal da:a 5
~
w _
, o
> Pdf's or cdf's derived frc= ;
~
theoretical bases -
o
.c Li ited observatica .
C
- i k I i
.c E=pirical Bayesian me thods -
)
- Bayesian methods :
~
E -2 -
c Expert opinions E w -
t Delphi methods :
=
u -
g Relative probability -
.c estimation :
2 -
an Ffgure E-1. Normal bases for probabilities assi 5ned to geologic events, processes, and features of concern to various areas of interest Q considered in this report.
. U E-5
O D . .
O -
l l
I IIPLB13ffATION OF HLW FIVE YEAR 11Af1 ACP.S IiRIEFifiG '
FEBRUARY.19,1987 -
SET 11 M. COPLAN b
I I
, _ , . - - . -j., ,
j F 0 -
O -
^y
" ~ 2 .
j 'f..f t q . .
J of$
i si r
(NEfWIEW $ ,/
f
~
REVIEW FIVE YFAR PLNI OBJECTIVES NO APPROACH DISCUSS STATUS OF IMPLEMENTATION OF!PL#1 FOCUSIflC Of! Ml ONGOING PLN4NING ACTIVITY -- RESOLUTION OF COPFLIANCE i DEf0NSTRATI0tlISSUES ,
PURPOSE NO OBJECTIVES APPROACH EXAMPLE STATUS ,
4 i , ,
I h
0 1
$ p-
e 9
e O
s b
n y
- g. .
m.
cr g
u.
d
bb b. O -
e FIVE YEAR PLAN - HIGi LEWLWASTE REPOSITORY PROGRAM GOALS #fD OBJECTIVES AGGRESSIVE PROGRAM FOCUSED Otl ACTIVITIES NECESSARY TO PROVIDE SUFFICIENT LICENSING GUIDANCE TO DOE AND SUFFICIEt(T If(TERACTION WITil DE, STATES, INDIn TRIBES,AND OTilER AGENCIES IN ORDER TO IDENTIFY LICENSING CPEN ITEMS AND BEGIN TE PFOCESS OF RESOLVING TifEN. .
AGGRESSIVE PROGRAM THAT STRIVES TO -ASSURE A FORMAL RESOLifTION OF LICENSITE OPEtl ITEMS PRIOR TO THE LICENSIM HEARIflG, TO Til,E EXTENT PRACTICABLE.
DEVELOP AN INDEPENDENT TECFWICAL CAPABILIYY TO PEVIEW DOE'S LICENSE APPLICATION WITHIll A 3 l4 YEAR TIE FRAME. , -
- IDEtlTIFY AND ELIMINATE, TO THE EXTEllT P05NIBLE, IMPEDIENTS TO NETIllG NRC'S STATUTORY TIE FRAE FOR LICENSE PROCEED'A AND IDEt(TIFY AND IMPLEMENT EFFICIEllCIES IN THE LICEllSIllG PROCESS.
P t
4 a
3-f
~
! O Q O -
t DEVELOP SYSTEMATIC APPROACH FOR DEVELOFMNT OF GUID'ANCE AND FOCUSING Pf0GRAfi ON EARLY IffflTIFICATION AND CLOSURE OF OPEN ITEMS OPEN ITEM IDalTIFICATION AND PRIORITIZATI0ll.
DEVELOP NCHANISH TO' FOCUS DEVELOPMENT OF GUllWJCE AND NRC/ DOE INTERACTIONS ON FORMAL CLOSURE OF OPEf! ITUIS.
{
d 6
4 I
y.._ _ . . _ _ _ . - _ _ _
t DEVELOREfT OF INDEPEEENT TE0HICAL EVIEW CAPABILITY ESTABLISH REVIEW CRITERIA NID PEVIEW APPROACH FOR EACH COWLIANCE IU10fL"TRATION ISSUE l -
FOR VARIOUS' PHASES OF THE PROGRAM
- SCP SCP UPIMTES LICENSE APPLICATION MODELING STPATEGY LOCIA1EllT i
6 i
i l
e +
k 1
9
. 5
-. W4M' 9
O 9
E s
p E
e tb
.22 b-O. ~
2":
- 5 -
g.-
8
!6 5
a l
l h
l
. \
4 RJRf0SE AND OBJECTIVES FURPOSE: TO IffLETErlT THE FIYE YEAR PLAN FOR GENERIC OPEN ITEMS.
OBJECTIVES:
IDENTIFY KEY LICENSIP(G OPEN ITEMS IDENTIFY STRATEGIES AND SPECIFIC MECHANISMS TO RESOLVE TilEM IDENTIFY MiAT MUST BE DONE TO ASSURE AN INDEPENDENT, INTEGRATED STAFF REVIEW CAPABILITY '
ESTABLISHMILESTONES/hCHEDULES'FOROPENITEMRESOLUTION AND DEVELOPMENT OF REVIDI CAP /BILITY 4 9 l
8 l
l 7
i i
i l, APPROA01' IDENTIFY KEY CG1PLINJCE DEMONSTRATION ISSUES (El@0DIKffT OF THE FINDINGS)
PERF0PNf1CE OBJECTIVES OF 10 CFR PART 60 INCLUDIllG TEOfflCAL SITING AND DESIGl CRITERIA i
) IDEifTIFY KEY GENERIC OPEN ITEMS FOR EAQi cr$1PLIANCE DEMONSTRATIOff ISSUE SPECIFY TE FINDINGS AND T5ClflICAL CONCLUSIONS TO BE MADE IDENTIFY t. flat GUIDAIJCE IS NEEDED AND hHAT METil0DOLOGIES j NEED TO BE DEVELOPED TO SUPPORT THE FINDINGS AND TECifflCAL-CONCLUSICf!S e
e 9 *I * ,
e
r] .
k APPROACH i
i IDElfflFY METHODS FOR RESULUTI0ff EV/10 ATE OPEN ITEMS AND IDENTIFY AREAS MERE GUIDANCE SHOULD EE DEVELOPED FOR THE DOE TO THE EXTEffT PRACTICABLE, RESOLVE OPEf1 ITEMS PRIOR TO LA SUBMITTAL USING SUCH KEY MECH # list 1S AS:
GU'S TOPICAL REPORTS ftETIl1GS' RULFFAKING IDEt(TIFY ECHANISNS NEEDED TO DEVELOP AND SUPP0kT STAFF REVIEW C/PADILIT/
DEVELOP MILEST0flES/ SCHEDULES IDENTIFY SPECIFIC STEPS FOR OPEN ITEM RESOLUTIOli #1D DEVELOPMENT OF REVIEW CAPABILITY ,
PRIORITIZE OPEN ITEMS ESTABLISH SCHEDULES 1 9
5 Q .
O .
1 4
\
GUIDELINES FOR THE FIVE YEAR PLAMilNG APPROACH i
PROACTIVE AS OPPOSED TO REACTIVE 4 FOCUS THE PROGir'1 ON THE XEY LICENSING DECISI0llS TilAT FUST BE FADE WITH RESPECT TO 10 CFR PART 60 PERFORMANCE OBJECTIVES AfD SITIfJG NO DESIGN CRITERIA. ,
OPEll #S DOCLPENTED PROCESS FOR DEVELOPfENT OF GUIDANCE AND EARLY IDEllTIFI-CATION, PRIORITIZATION AND RESOLLITI0li 0F OPEll ITEMS, SCP/SCA PROCESS IS PPINCIPAL FORM
! PROVIDE FOR EARLY AND FULL INVOLVEMENT WIT 11 DOE, STATES no INDIAN TRIBES i-INCLIFE DEVELOPMEllT OF A FORMAL MECHANISM FOR IFFLUElTfATION i
10 t
% 6. O -
f -
C&PLTANCE DEMDIETPATILW ISSUES i
i EIGfi C0ffLIANCE DEMONSTRATIrt! ISSUES DERIVE DIRECTLY
' FROM TIE PERFORMANCE OBJECTIVES
, OF 10 CFR PART 60, I -
PRE-CLOSURE RADIOLOGICAL PROTECTION, i -
RETRIEVABILITY WASTE PACKAGE CONTAlltiEffT ENGIllEERED BARRIER SYSTEM RELEASE P/.TE GROUNDWATER TRAVEL tit 1E INDIVIDUAL PROTECTION (EPA STntva) .
l -
GROUNDWATER PROTECTION (EPA STnuPD)'
i -
EPA CONTAIPPENT STANDAFD i
! FIVE CUT ACROSS THE FIRST EIGfr AllD EITHER ADDRESS A COP 90N ELEMEffT OR ASSURE All INTEGRATED PROGRAM l
UflCERTAlflTY j -
QUALITY ASSURM'CE 1
FORMAT AND CONTENT GUIDE FOR THE LICENSE APPLICATION.
REVIEW PLANS INTEGPATION I
t 11
O O O -
i i
EXMPLE -- WASTE PACKAGE C0flTAlff9fT OPEN ITEMS 1
o PERFORMANCE NMLYSIS OF WASTE PACKAGE DESIGNS I -
CRITERIA FOR ACCEPT /ILE W.STE PACKAGE CONTAliffffT TIE DETEPNiilATI0fl 0F INPlK PARAETERS AND DATA FXTPAPOLATION OF SHORT-TERM TEST DATA l
ENVIROMNT/1 PARNETERS #fD DATA WASTE PACKAGE RELIABILITY NMLYSIS l
ACCEPTANCE CRITERIA FOR PERFORMANCE ASSESSMiiT CODES i -
APPLICABLE ASW/ ANSI CODE ffWIPEENTS i
0 WASTE PACKAGE TESTING -
CRITERIA FOR WASTE PACKAGE TEST PROGRAPS ENVIP0ffENTAL CONDITIONS FOR TESTING PRIOR TO ACQUISITION OF SC DATA APPLICABLE ASE/ ANSI CODE REQUIREENTS i
IMPLEEffTATION OF PEER REVIEW P#!EL REcutEl#MTIONS l .
i l
't i 12
', J l
, EXNfLE-WASTE PACKAGE CONTAIWENT I .
l GEERIC TEOflICAL POSITIONS (GTP'S) WOULD BE A PRIIMRY ECHANISM l FOR PROVIDING GUIDANCE AND RESOLUTICU 0F T E OPEN ITEMS. TESE WOULD BE DEVELOPED THROUGH A STN & &DIZFD TEST OF MILESTONES.
l OTHER ECHANISMS It'CLUDE RULEMAKINGS, TOPICAL REPORTS, LETTERS i TO DOE, AllD l'CETINGS. SELECTION OF TE SPECIFIC fEOfANISM FOR EAOI OPEli ITBi WILL DEPEND ON PRIORITIES, RESOURCES, AND i APPROPRIATENESS TO TE PARTICULAR OPEN ITEM i *
!' FEVIDI CAPABILITY NOULD BE DEVELOPED Tl!RGJGil A STNEARDIZED SET OF ACTIVITIES IDENTIFIED AS PART 0F TNF OVERALL PLANNING EFFORT.
i
, . +
e l .
l
'h
j ()
i GlP F11LESTQES
? -
i MILESTONES TO PREPARE GTPS PROVIDE ECHANISMS TO:
1
- INVOLVE D T , STATES WlD TRIBES EARLY ON
} AND THROUGHOUT TIE EilTIRE PROCESS l
i -
ADEQUATELY VEffTIlf,TE OPEN ITEMS TO THE TEOf11 CAL l CdfM4ITY AND PEER REVIEW GROUPS l -
ASStfE NRC'S PEQUEST FOR CortENTS ARE RESPONDED TO AND DOClMNTED .
l . .
i - PROVIDE TE, STATES AND TRIBES WITil DOClMNTED j RESPONSES TO COPHEllTS ON DRAFT TECif4ICAL POSITIONS l
i i
) ]Il l<
O O O '
) LGIVITIES TO DBt10P STKF lEVIDi CA!YJ31LIW KEY STANDARDIZED ACTIVITIES ESTABLISH GENERIC REVIEW CRITERIA / STANDARDS BASED Gfi THE FINDING OR TECHNIC 0AL CONCLUSION THAT IS SPECIFIED DEVELOP A GENERIC EEViFW STRATEGY IN C00RDINATIOil WITH Tile fl0DELifE STRATEGY TASK GROUP
- DETERMIfE METIER EXISTIfM COlGRACTS,'GTAFF ACTIVITIES AND RESEAPClf EFFORT 3 SUDPORT T E NEEDS C/1 LED TOR Ill THE GENERIC REVIEW S11MTEGY
- IF NOT, DEVELOP PLANS TO ACQUIRE THE F.'EEDS i
i i
i I
i l
l
{ f>
4 O O. -
i
! M AK PLANS STATUS 1
CLTTfffT STATUS:
TE FIRST GOOD C11T OF A COPPLETE SET OF M'RK PU11S IS UNDERGOING MANAGEPS(T
., REVIEW
- FtfTURE PLANS:
i FUFTriER REFINE AND INTEGRATE Tif WORK l PLANS BASED ON THIS PP/!EW l -
PRI0ldTIZE DIE OPEN ITEMS AND DETErdillE
- 11ILE3 TONE SCHEDULE DATES AND PES 000CE ALLOCKt 10fi ,
J FURTHER REFIf1E BASED Or! EEEll!TS 'CF OTER
! PLNiill1G EFFORTS OriCE MILESTONES SCHEDJLE IMTES NO FUNFCES
, ARE ALLOCATED, IEELOP N1 #fror%TED MANS OF MONITORIlJG NJD UFIATING TE WODK PLANS 9
9 I
i j 16
,