ML20141K452
| ML20141K452 | |
| Person / Time | |
|---|---|
| Issue date: | 05/21/1997 |
| From: | NRC ADVISORY COMMITTEE ON NUCLEAR WASTE (ACNW) |
| To: | |
| References | |
| NACNUCLE-T-0114, NACNUCLE-T-114, NUDOCS 9705290154 | |
| Download: ML20141K452 (337) | |
Text
Official Transcript of Proceedings -
NUCLEAR REGULATORY COMMISSION fcNWT o//4
Title:
Advisory Committee on Nuclear Waste iTR08"(ACW). }
iRETului 0RIGINAl.'
tTO BJWHITE, 4 Docket Number: (not applicable) ACRS T- 2E26 D 'l f415-7130 .
[,THANKSi ~. ,,
- n. z.._ ;. c a Location: Rockville, Maryland Date: Wednesday, May 21,1997 b
)
\Vork Order No.: NRC-1119 Pages64-295 ACNWOFRCE COPY-RETAIN FOR M WMM 00MWilTEE BilllBllBlilll llllBllllHlllllllll n...
" ~ ~ ' " ~ ' '
NEAis R. GROSS AND CO., INC. " ~ '
Court Reporters and Transcribers b gg()()gg 1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005 j $$ h f Sk5 l0 .l]{k(
(202) 234-4433 DR DV N NA f CLE T-0114 PDR
k DIBCLAIMER PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGUIATORY COMMISSION'S ADVISORY COMMITTEE ON NUCLEJR WASTE MAY 21, 1997 The contents of this transcript of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Nuclear Waste on MAY 21, 1997, as reported herein, is a record of the discussions recorded at the meeting held on the above date.
This transcript has not been reviewed, corrected and edited and it may contain inaccuracies.
NEAL R. GROSS COURT REPORTERS ANDTRANSCRIBERS 1323 RilODE ISLAND AVENLT, NW (202)234-443I WASilINGTON, D C. 20005 (202)234 4433
64 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION
\_/
3 + 4 + + +
4 92ND MEETING I l
5 ADVISORY COMMITTEE ON NUCLEAR WASTE 6 (ACNW) 7 + + + + + l l
l 8 WEDNESDAY I 9 MAY 21, 1997 10 + + + + +
11 ItOCKVILLE, MARYLAND 2
12 + + + + +
13 The Advisory Committee met at the Nuclear i
k 14 Regulatory Commission, Two White Flint North, Room T2B3, 15 11545 Rockville Pike, at 8:30 a.m., Paul W. Pomeroy, s
16 Chairman, presiding.
l 17 i
18 COMMITTEE MEMBERS.
19 PAUL W. POMEROY Chairman 20 B. JOHN GARRICK Vice Chairman 21 WILLIAM J. HINZE Member 22 GEORGE M. HORNBERGER Member 23 24 25 N EAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE Isl AND AVE., N W (202) 234-443'J WASHINGTON, D C. 20005-3701 (202) 234-4433
65 1 ACNW STAFF PRESENT:
,, - ~ , , 2 John T. Larkins, Executive Director 1
i
. 3 Michele Kelton, Technical Secretary a
- 4 Richard K. Major 5 Howard J. Larson i 6 Lynn Deering j 7 Andrew C. Campbell 8 Richard P. Savio 9 Michael Markley 10 Carol A. Harris 4
i 11 Sam Duraiswamy i
i i 12 Theron Brown i
I
, 13 3
A)
)I
\v 14 ACNW CONSULTANT PRESENT:
i 15 Martin J. Steindler 1
l 16 ALSO PRESENT:
17 Steve Brocoum I
j 18 Abe Van Luik
'i i
19 Jack Bailey
$ 20 Carol Hanlon I
i 21 Mike Lee 1
22 Mike Bell 23 Norm Eisenberg 24 Margaret Federline r\
(j 25 Janet Kotra NEAL R. GROSS COURT REPORTERS AND TRANSCR flERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
66 1 A-G-E-N-D-A
, - - 2 AGENDA ITEM PAGE
(
v ;
3 Opening Remarks by Chairman Pomeroy 67 '
4 Views on the Defense-in-Depth Approach 5 S. Brocoum 69 6 J. Bailey 93 7 A. Van Luik 152 8 The Use of Expert Elicitation in the Review i 9 of a HLW Repository 10 S. Brocoum 215 11 M. Lee 231 12 Project Integrated Safety Assessment 1
l 13 C. Hanlon 269 1 rh I kv) 14 Adjournment l
1 1
15 16 17 18 19 20 21 22 23 24 O
t 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, O C. 20005-3701 (202) 2344433
67 1 P-R-O-C-E-E-D-I-N-G-S p_ 2 (8:37 a.m.)
(.
' '~' )
3 CHAIRMAN POMEROY: Can you please take your 4 seats and the meeting come to order?
5 This is the second day of the 92nd meeting of 6 the Advisory Committee on Nuclear Waste.
7 Today, the committee will first discuss the 8 use -- it will not do that. First, it will uiscuss the 9 defense-in-depth philosophy as it relates to radioactive 10 waste management.
11 Secondly, we'll discuss the use of expert 12 elicitation in the review of a high-level waste 13 repository.
/~'\
() 14 Third, discuss DOE's project integrated safety 15 assessment. Four, continue to prepare ACNW reports. And 1
1 16 finally, discuss committee activitie and future agenda, i l
17 Mr. Andrew Campbell, at the end of the table )
i I
18 on my right, is the designated federal official for the 1 l
1 19 initial portion of today's meeting. This meeting is being 20 conducted in accordance with the provisions of the Federal l
21 Advisory Committee Act. l l
22 We have received no written statements from 23 members of the public regarding today's session. Should 24 anyone wish to address the committee, please make your
(~'N
( ) 25 wishes known to one of the committee staff.
x_/
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C 20005-3701 (202) 234-4433 l
68 1 It is requested the each speaker use one of
_ 2 the microphones, identify himself or herself, and speak
3 with sufficient clarity and volume so that he or she can 4 readily heard.
5 We have a very busy schedule today. The first 6 item on our agenda, unless there are comments from my 7 colleagues, suitably quiet --
8 (Laughter.)
9 -- is the generic methodology. I'm sorry.
10 I've got to read the right page. Views on the defense-in-11 depth approach. And Dr. Garrick is our lead member for 12 this subject, and so, John, please lead off.
13 VICE CHAIRMAN GARRICK: Thanks, Paul.
f'^r k_) 14 The committee has a great deal of interest in 15 this subject. As you know, we have already had a 16 considerable amount of discussion on it at our last 17 meeting. We intend to hear from the DOE principally at 18 this meeting, and at our next meeting we also intend to 19 discuss this subject with the goal of perhaps following 20 that meetin: offering some advice to the Commission on 21 this subject.
22 The issue here is very much one of trying to 23 examine whether or not the traditional use of defense-in-24 depth philosophy should continue, or should be modified in
( w))
25 some way in the spirit of accommodating the new standard NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WAShsNGTON, D C. 20005-3701 (202) 234-4433
69 1 and the changes that are being considered in the 7, 2 regulations and the guidance to those regulations.
4 3 One of the key questions that this committee 4 is examining has to do with whether or not the 5 effectiveness of individual lines of defense can be 3 adequately quantified in reference to health and safety l 7 risk measures to justify some alternative approach. And 8 some of the issues that are involved here are the issue of 9 the independence of the individual lines of defense, 10 question of tradeoffs between natural barriers and 11 engineered barriers. There is the issue of the allowable 12 uncertainties.
13 And, in general, the question comes up as to l s 1
- 14 whether or not the role of the different barriers can be i
15 made sufficiently clear that they are -- in the context of l
16 the performance assessment, that some alternative approach 17 would be a more effective consideration for the regulation 18 of repositories and, in specific, reference to defense in ;
l 19 depth. l 20 So with that, I think we will begin. I think 21 that Steve Brocoum is going to start the discussion, and 22 I'll leave it to Steve to introduce the subsequent 23 speakers. Steve is from the Yucca Mountain Project and 24 the Department of Energy.
O)
(, 25 MR. BROCOUM: It was a pretty close call this i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHL, JE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
70 1 morning. I just rolled in here about two m .iutes ago.
73 2 And then when I went down to register, I needed to put a N] 3 phone number down. I needed to puC a person visiting and 4 a phone number. I thought of Lynn, and then, of course, I 5 had to look up the phone number.
6 So I want to make a couple of preliminary 7 comments here as we start. There will be three speakers 8 today talking about defense in depth. It will be myself, 9 who will kind of give an overview and an approach. There 10 will be Jack Bailey, who will talk in more detail, 11 particularly on the engineering part. He will go into 12 quite a bit of detail on some areas, some of the things 13 we're thinking about. Finally, there will be Abe Van O)
\-- 14 Luik, who is going to talk about TSPA in general and 15 related to defense in depth. I don't even see Abe. He's 16 here somewhere. I saw him. There he is, right there.
17 I have to be honest, though. In recent 18 months, at DOE defense in depth has not for us been a 19 front burner issue. Front burner issues for us is, what 20 is EPA going to do with their rule? What are we going to 21 do about an east-west drift? How is 960 going to go? We 22 were in rulemaking on that. Lots of other issues.
23 Defense in depth is not our front burner issue.
24 So to give a presentation, you need to put
<~s\
(
(_) 25 words on viewgraphs, and I've done that. This is kind of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
71 1 work in progress. I wouldn't like to call it a DOE
,- 2 position. We are starting to think through this.
3 The design -- as you know, we're doing the
, 4 viability assessment in September of '98. That design for 5 that -- it is a crucial period for that design as we are 6 deciding what to include and what not to include in that 7 design, and Jack is going to be talking about that for 8 ,some time. So I think I would like to characterize this i
9 as work in progress, and probably a year or two in the 10 future, if we get past VA, and if we go on to licensing, 11 we'll have thought a lot more about it. So I just want to 12 give that introduction.
13 We feel -- I can use a pointer, but I guess rN s 14 there isn't a pointer, 15 MS. DEERING: Right there.
16 MR. BROCOUM: We feel that the preclosure 17 repository system is amenable, more or less, to 18 traditional defense approaches, being as it's an 19 engineered feature. It's an operational feature. It has 20 a lifetime on the order of nuclear power plants. You 21 know, by prevention, mitigation, various kinds of multiple 22 barriers, use of conservatism, redundancy, diversity in 23 design. We'll go through that. Jack will do that later.
24 Using QA cont 2ols, emergency plans and procedures. Those A
! 25 kinds of things, we believe, can be applied to the s
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
72 1 preclosure.
I
, 2 For the postclosure, we feel you could, you U 3 know, make attempts to prevent releases, or, once they're 4 released, delay. You may be able to mitigate through 5 various features of transport after release. Of course, 6 you depend on multiple barriers. We're avoiding the word l l
7 " redundancy," because we don't think we have truly done, )
, 8 say, particularly con"ervatism, diversity, desigly.
9 We want our engineered barriers to compensate ,
I 10 for uncertainties in the natural barriers, and we would 11 like our natural barriers to compensate for the 12 uncertainties in the engineered barrier controls. !
l 13 With regard to postclosure performance, even' l j
()\
(_ 14 with the engineered barriers, you can decide what you're l i
15 going to design in. I mean, you can decide what features, 16 engineering features, you want to install. You can make a ,
I 17 decision. You can.do tradeoffs and decide whether you I
18 install them or not.
19 With regard to natural barriers, I mean, you ,
i 20 have what you have at the site. You can take credit for j 21 various barriers, or you can attempt to take credit, but ;
22 you have a site already. Whatever is there is what you l
23 have. And, of course, you have QA controls to make sure 24 you collect and analyze all of the information properly.
()
rx 25 A little background and historical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
73 1 perspective. In our 1986 environmental assessment, we 7- 2 made various conservative assumptions to compensate for m
3 uncertainties in the site and design information. We 1
4 assumed at that time a higher flux than we expected at 5 that time. There was an assumption that the flux just 6 went downward. It didn't get diverted. It could be 7 diverted by faults. It could be diverted to PTN.
8 We assumed a lower bound for the thickness of 9 zeolites. Notice we didn't take full credit for the 10 zeolites, and there was no credit taken at that time for 11 any effect of the heat on a system in terms of drying out j 12 the rocks around in the near field, or around the waste 13 packages, n.
- 14 In our 1988 site characterization plan, we had 15 objectives for the repository system. We wanted the 16 engineered barriers to limit the release of radionuclides 17 to natural barriers. We wanted to make sure -- and still 18 do -- that the engineered barriers did not adversely 19 impact natural barrier performance. And finally, we 20 wanted the natural barriers to perform to provide an 21 environment that was conducive to preserving the life of 22 the engineered barrier, and we want to provide conditions 23 where the transport of significant quantities of 24 radionuclides will take a long period of time.
O g j 25 There are, of course, issues that have been NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433 A -
74 1 raised over the years. The last, certainly since '91, we n 2 have -- DOE has provided strong support for a single risk-1 3 based quantitative criteria for postclosure performance.
4 We did that to the National Academy of Sciences. And, of 5 course, we've done it for the EPA. We've done it for the 6 NRC, and I think this group.
7 And in our current proposed revision, 10 CFR 8 960, which is on the street -- I think the public comment 9 period just closed on the 16th of May -- we rely on total J
10 system performance of natural and engineered barriers, l'1 each of which contain multiple barriers. We don't really 12 address defense in depth in 960 as we currently structured 13 it.
t h
(O 14 How are we thinking about it today? We're i i
15 using our previous design analyses and our TSPAs, and have i
16 provided a basis for our -- I put the word " evolving" 17 waste containment isolation strategy, which I'11 talk a 18 little bit about today. The waste containment and 19 isolation strategy establishes a framework for allocating 20 performance for the natural engineerea barriers and 21 provides the basis for describing the defense in depth for 22 the postclosure repository system.
23 Now, I want to caution that the waste 24 containment isolation strategy is an evolving -- and I rh (j 25 have that word on top of the viewgraph here -- and it's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
~ .
75 ;
l 1 being presented a little differently today than we have 7s 2 presented it in the past. And it's likely as we think i
- _ U 3 about it and talk about it, in turn, we will present it l
4 differently in the future.
l 5 We have four elements, as we're describing it
- 6 today -- limiting water contacting the waste packages, 7 having -- so that has both engineered and natural system 8 attributes, if you'd like; a robust waste package, which 9 is a near future; limited the mobilization of 1
i 10 radionuclides once the package fails; and finally, 11 delaying the transport, both through the engineered and 2
12 natural barriers and reducing the concentration. And 13 we'll talk a little about those, and I'll probably refer t%
-- 14 back to this diagram again. So these, again, as I just 15 said, are four key elements as we're describing it today.
16 Why is it evolving? There has been new or 17 improved site understanding over the last year, and I 18 think it started when ACNW was out in -- I think it was 19 last October at Yucca Mountain -- you know, the whole l
l 20 debate about the percolation flux. Currently, the numbers )
21 that people are debating range from one in 10 millimeters 4
1 22 a year, some people say two to 15. Al Beall had a model 23 between four and 10. I think the PA curves that Abe will i l
24 show later uses 6.2 millimeters.
f')\
\, 25 (Laughter.)
l N EAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
76 1 Which is --
. 2 CHAIRMAN POMERUY: That's good.
(N/
3 MR. BROCOUM: -- some sort of a mean out of 4 the four to 10 LBL model. Okay? So this is a very 5 important issue for us.
6 It is believed that, although the percolation 7 flux through the repository host may be somewhere in this 8 range, the amount that seeps into the repository itself is 9 likely to be less, just as the percolation flux through
- 10 the repository host rock is likely to be less than the 11 amount that infiltrates on the surface of the mountain.
12 So keep those in mind.
23 We believe that thermal effects will
- O
- \_) 14 redistribute the moisture, and it will slowly return to i
15 ambient conditions over several thousand years. Jack 1
16 Bailey will show you some curves. Unfortunately, they're 17 not for the current percolation flux, but they give you an la idea of what we're talking about. And during 19 transportation, radionuclide concentration will be reduced 20 by various mechanisms. You will diffuse, you will retard, 21 you will dilute, and so on.
22 However, it's difficult to make accurate 23 predictions because the natural system is heterogeneous, 24 even if you cannot fully quantify all of these parameters (O j 25 in your PA, that you have them present, and to some degree NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON, D C 20005 3701 (202) 234-4433
77 1 they provide defense in depth even though you're not 2 necessarily quantifying them.
p U 3 How we allocate performance to each system of 4 the repository depends on how effective that component 5 barrier and what the uncertainties around it are. The 6 improved understanding of the moisture conditions and the 7 better, again, definition of our spatial and temporal 8 uncertainties is used in sensitivity analysis, of which 9 Abe will show you several today on a total system 10 performance.
11 And for the barriers or the parameters that 12 are very important to how a site performs for those 13 particular barriers and parameters, especially those which O
V 14 have high uncertainties, that is where we want to focus 15 our efforts on providing defense in depth. So we want to 16 link defense in depth to overall system performance, and 17 we want to use TSPA to help -- and sensitivity studies to 18 help guide us to those barriers in performance that are 19 most important and those that have the most uncertainties.
20 That's kind of a summary of our philosophy.
21 How do we envision the repository working, and 22 what is the. defense in depth? We have natural barriers in 23 an arid and unsaturated zone which has limited 24 infiltration. The upper limit is the amount that
/
- i. 25 infiltrates through the surface of the mountain.
N NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
78 1 Presumably, some of it diverted, so the percolation flux 2 would have to be less, and what seeps through the
(,--)
'~ l 3 repository would have to be less. Some of it will be l
? 4 diverted, and that's the new term we're using. .
l 5 We will have engineered barriers. The drift 6 wall will be some sort of a capillary barrier. There are J I 7 come models that LBL has run that show up to 28 8 millimeters a year of percolation flux resulted in no 9 seepage. I'm looking at eight here. But at 280, so l
10 somewhere between 28 and 280, there war seepage.
1 11 The heat from the waste -- they reduced the I 1
12 available moisture. There is a lot more water now flowing 13 through the system. It may not be the dryout we thought !
(~'
(_,) i 14 we'd have several years ago, but there will probably be a I
15 reduction in moisture. And Jack particularly will talk l
, 1 16 about various ways we're thinking of diverting the i
17 seepage. We're thinking of drip shields and back fill and 1
l 18 ceramic coatings, and various features of those sorts. '
19 With regard to the waste package, we are f 20 planning -- currently, we are planning to use a corrosion 21 resistant inner barrier, and the corrosion allowance cut 1
22 of barrier, which will prolong the life of the waste 23 package. We are looking into these two layers producing i 1
24 galvanic protection of the inner layer. We are looking
/~'Tj 25 into ceramic coating on the outside of the package, which
(
'u/
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N.W. l (202) 234 4433 WASHINGTON, D C 20005-3701 (202) 234 4433 I
79 l
1 may even extend the life longer, in terms of preventing i
-s 2 against corrosion. This adds some diversity of design due
(,)
1 3 to different material. I 4 We're looking at using backfill. We've had a 5 lot of debate over the years on backfill. As you probably I 6 know, the PA people love backfill, and engineers hate it, I 7 because it is a very difficult thing to emplace and work 8 around. It may offer mechanical protection for the i 9 diversion system and from rock falls, for example. And I 10 depending on how it's done, it could limit advective flow 11 on the waste package surface, making all of the processes j 12 go that much slower. I l
13 CHAIRMAN POMEROY: Steve, I know you don't I~h j
\s l 14 want to put any numbers on the waste package numbers --
15 waste package time at this point, but can you give us just 16 an order of magnitude? Are you thinking in terms of 17 thousands of years now?
18 MR. BROCOUM: I think we're thinking in terms 19 of thousands of years. And the thermal pulse, at least at 20 the low flux rate, was on the order of 3,000 years or so.
21 I'm not sure what the thermal pulse -- what the estimates 22 are on a thermal pulse today with the higher flux. But we 23 wanted to get the waste package more or less to 24 accommodate the heat -- the thermal load early on.
gs
( ,) 25 Is there any number, Abe, or - -
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
80 1 MR. VAN LUIK I think it's less than a 2 thousand years.
9, 3 MR. BROCOUM: Now it's less than a thousand 4 years. But anyway, we're looking for the maximum lifetime 5 we can get. The issue we've always had is demonstrating 6 it in a hearing. ;
7 CHAIRMAN POMEROY: Yes. I i
8 MR. BROCOUM: That whole reasonable assurance 9 argument.
10 And, of course, the waste packages will not i
11 fail all at once. They will fail over time, and that 12 failure will be represented by some kind of a distribution i 13 function.
O i
k~ 14 For some radionuclides, the solubilities will 15 limit mobilization. You know, in neptunium, we've been 4
16 using solubilities that c.re several orders of magnitude 17 lower than is thought to be correct.
I 18 Cladding reduces the waste form area that is '
i 19 exposed to solubilities. Now there is a big debate 20 whether we could take credit for cladding or not. And, in !
21 fact, if you take credit for cladding, it improves your i
22 performance by about one order of magnitude. )
l 23 If we cannot take credit, because we can't l l
l 24 convince ourselves or the NRC, it is still there. That's j
(~)\
( 25 an important point. So it's a qualitative -- it may be a NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W. ,
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
i 1
81 1 1
l 1 quantitative, if you can take credit, or it may be a l l
d
,. 2 qualitative if you can't. But the fact is it's still
[ ) :
V 3 there.
j 4 You know, a long waste package lifetime 1
5 through the thermal pulse limits the alteration of the ,
i 6 performance inside of it. And this is an important issue. ,
1 7 This gets back -- we don't want to do anything in the 8 engineered barrier that would impact the natural barriers. !
l 9 And since now we're thinking -- the designers are thinking 10 of some kind of a concrete lining, this becomes important. )
11 You don't want to change the alkalinity of the water j 12 beyond a certain point. And there are other issues here 13 every time we think there is some kind of an engineered g
(_/ 14 feature.
15 With regard to the -- once it gets out of the 16 waste package and it gets into the engineered and natural 17 barriers, there is a potential for putting some sort of 18 sorptive mater.tal beneath the waste package in inverts, 19 and the engineers are considering that. I think it's on 20 Jack Bailey's list.
21 There is, of course, the backfill, which I 22 already mentioned. We know that matrix diffusion, both in l
23 the unsaturated and the saturated zone, will occur.
24 Again, even if you can't fully quantify it, you know it's
,o (v) 25 there. For some radionuclides you'll have sorption.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
82 1 Again, we may not be able to fully understand it, but we
,,s 2 know we have sorption. We have the Calico Hills with lots
()
3 of zeolites.
4 When a concentration is reached for the water 5 table, of course, then you will have dilution. You will 6 have mixing. You will have dispersion. And how much of 7 this occurs, of course, there will be a lot of debate.
8 Where your standard requires you to take the measurement 9 as you meet a standard will be very important.
10 And as you withdraw water, as a farmer or a 11 subsistence farmer or a person, from a well there is more 12 mixing, a lot of debate whether one can take credit for 13 this with EPA and with NRC. But the fact is it will (3
\/ 14 occur. So even if you cannot take credit, there will be l
15 some conservatism introduced, you know, because there will I 16 be mixing.
17 With regard to disruptive processes and events 18 -- well, obviously, early on we considered the 19 probabilities of significant disruptive events in a site 20 screening process. The current approach is to analyze 21 these features, events, processes, on the basis of 22 likelihood and their basis of consequences. TSPA will be 23 used to evaluate consequences for a limited number of 24 these events that have more or less a high likelihood.
- s q,) 25 The one that comes to mind, of course, is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
83 1 volcanism, which I'll touch on later when I talk about 1
,3 2 expert elicitation.
(V \
3 So, in summary, we have robust waste packages,
\
4 much more robust than we had in the SCP days. We're going 1
l 5 to have enhanced engineered barriers. We have a whole 6 menu of things we are considering.
l 7 And the amazing thing to me on this is that a ,
1 a couple of weeks ago the engineers came in and gave us an )
i 9 internal briefing on the kinds of things that they were l 10 considering. And even the most complex features of total 11 cost was not greater than $1.8 billion on top of current 12 costs. They said that was an estimate that could be 100 13 percent too low and 50 percent too high.
in 14 So even if you add all of the features the 15 engineers would like to add, you're only talking 1.8, with 16 an error of 900 to 3.2 billion, at our current 17 understanding. Again, that is all very preliminary, but I 18 just wanted to put it on the table.
19 The natural barriers provide defense in depth.
20 Some of it is quantitative, but some we know is there, but 21 we may not be able to quantify it. But it's there. You 22 know, it doesn't go away just because you don't consider 23 it in your analysis.
24 And again, our goal is to use conservative
()
f3 25 assumptions for those elements with high uncertainty and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
I 84 l 1 high impact on performance during our performance j 73 2 assessments. That is kind of our goal. So we want to tie b 3 the defense in depth of our overall system performance.
4 We don't want to get in a position where we have subsystem 5 requirements that are not connected, or, as the famous 6 word is, there is no nexus between subsystem requirements 7 and the overall repository performance.
8 I think Abe is going to -- you're going to 9 talk on this a little bit, aren't you, because I didn't 10 see it in your viewgraphs. But basically, there is an 1
11 issue here that when you impose a subsystem requirement, !
12 you are imposing a conceptual model. And that becomes the 13 driving force in some cases and detracts from the overall rm
>i
\/ 14 system performance.
15 So that is my introduction, and much of this 16 will be repeated in greater detail by Mr. Bailey as he 17 goes through some examples.
18 VICE CHAIRMAN GARRICK: Steve, a general 19 question.
20 MR. BROCOUM: Yes.
21 VICE CHAIRMAN GARRICK: You referred to the 22 waste containment and isolation strategy as evolving.
23 MR. BROCOUM: Yes.
24 VICE CHAIRMAN GARRICK: My sense is that the
/'N
( ),
25 strategy isn't evolving. It seems to me that has always NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
85 1 been your strategy are these four steps. But what is 2 evolving is the way in which you are implementing that 7_
l l
~
3 strategy.
4 MR. BROCOUM: That's correct. Much of the 5 elements today are very similar to the elements we've had 6 in the past. That is correct.
7 However, we've had a difficult time, because 8 it's a difficult issue, in getting the scientists, 9 engineers, and the regulatory people to agree on a 10 comprehensive strategy that we could publish. And so we 11 published a summary. We didn't even publish it. I think 12 we released it, but we never published a summary, and 13 we're updating that summary. But we've had a very n
(_) 14 difficult time getting a scientific and engineering buy-in !
l 15 on a more comprehensive -- we call it a technical basis of 16 that strategy.
17 So we're working it, but it has -- and the 18 fact that you worked it has been good. It causes a lot of 19 people to face up to a lot of issues and to debate and to 20 argue and, you know, to think about. But we haven't 21 gotten closure on that yet.
22 VICE CHAIRMAN GARRICK: Yes. I have lots of 23 other questions, but I want to defer them to the other 24 speakers. But in the meantime, let me ask my colleagues
,a 25 -- Marty, do you --
()
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C 20005-3701 (202) 234 4433
l 86 l
MR. STEINDLER:
1 Yes. I've got one, if you p._
2 don't mind. You indicated several times that for some of
('-) 3 these barriers quantification is either not possible or l
l 4 hasn't been done. What do you do with the either l
1 5 qualitative or hand waving discussion and about the fact !
6 that a particular barrier exists, if you're either 7 unwilling or unable to quantify it and, therefore, have 8 some clear indication of its benefits? l 9 MR. BROCOUM: Well, I think, you know, we are 10 writing a comprehensive -- I'm talking natural barriers 11 first. We are writing a comprehensive site description 12 right now, which will be issued next year. That's kind of
, 13 a summary of all of the knowledge in a natural system that
/%.
(_) 14 we have. We'll pulling it all together. We will 15 obviously attempt to define all of the barriers, and we're 16 not going to ignore them.
17 But, for example, if the poinc of measurement 18 is 20 kilometers away, you've got an area in a saturated 19 zone that goes 20 kilometers. We are not going to have a 20 hole every hundred feet to define that geology. We will 21 have a few holes and some seismic and geophysical lines.
22 So there will always be a degree of uncertainty, for l 23 example, on the distribution of rock types, fractures, and 24 other natural features.
I)
U
. With regard to the engineered barriers, we l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C- 20005 3701 (202) 234-4433 l
87 1 have the ability to design them. However, there can be j_ 2 processes that occur that are so slow that we haven't seen
(\~') 3 them in the timeframe of our studies, but may occur over 4 the millennia. We refer to that as the unknown unknowns.
5 Who knows? I mean, there will be things that are that 6 way.
7 I want to make one more point here, because 8 I'll tell you what our internal goals are in doing our 9 TSPAs right now. When we do our TSPAs, the PA people have 10 told us, "Well, you know, first, to really feel 11 comfortable, we would like to be one order of magnitude of 12 whatever standard you're shooting for." So if you're 13 shooting for 15 millirems, the PA people would like to be
, n
% 14 one and a half.
I 15 And then we have this whole issue of defense l
16 in depth and how you define defense in depth for a natural l
l 17 system. So we've added another order of magnitude just to 18 kind of take into defense in depth. So if we're shooting l
l 19 for a 15 millirem standard, internally our goal is .15 20 millirems. That's what we're trying to do.
21 Again, because of all of the other l
l 22 uncertainties, both in natural and potentially engineered 23 systems., because we're talking again about, you know, 24 compressing the timeframes. So --
( 25 MR. STEINDLER: So you're not really talking NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON, D C. 20005 3701 (202) 234 4433
88 1 about lack of quantification. You're just talking about 73 2 lack of precision.
l )
~
3 MR. BROCOUM: You can put it that way. We 4 will quantify it to the extent we can.
i G VICE CHAIRMAN GARRICK: George? Bill?
6 MEMBER HINZE: Steve, in terms of the 7 environmental barriers, the geological barriers, you have 8 used the term dilution, dispersion, retardation, all 9 dealing with the saturated zone. What do your sensitivity 10 studies -- do you have any information on what your 11 sensitivity studies tell you in terms of what are the 12 critical unknowns regarding the saturated zone? And what i
13 is being done to bring that knowledge together before you o
- i )
! 's/ 14 have to do even a preliminary design for VA?
15 MR. BROCOUM: You know, the saturated zone was l 16 not very important in the SCP program, because at that l
17 time we thought we had a release-based standard. And we l
18 still don't have a standard. But assuming we get a dose-l 19 based standard, then the saturated zone becomes very 20 important. So we, of course, are doing the C well
(
21 complex.
22 We are adding some tests in addition to those 23 we had already planned to that. We are adding three 24 boreholes, one to the north, one in the middle of the n
( ,) 25 western part of the block, one to the south. And we are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
89 1 considering -- this is not so much for saturated zone, but
,f - 2 we are considering an east-east drift. So we are adding
) 3 more boreholes for the saturated zone, I guess would be 4 the best answer I can give you right now. I don't know if 5 there is anybody else here that can comment on that.
6 There was also consideration of another C well 7 complex. I'm told right now that is not really -- there I
8 is discussion about that, and they will be able to do 9 another test further to the south. And we may have to.
10 But right now, I don't think that's in our plan.
11 MEMBER HINZE: So are these plans to achieve 12 these before you get to the design as part of the VA?
13 MR. BROCOUM: No. This -- it is now hay of O.
I
)
(./ 14 '97.
15 MEMBER HINZE: Amen, brother.
16 (Laughter.)
17 MR. BROCOUM: No. No.
18 CHAIRMAN POMEROY. That's an answer.
19 MR. BROCOUM: The VA, remember, is a point in )
20 time. It's a status of where we are. We will incorporate l 1
21 all of the information we can in it. But in terms of I 1
22 planning a new test or new boreholes, depending on -- and 23 maybe some information can be. It depends on how fast the 24 boreholes are drilled. I don't know the schedule of the
("N 4
()
25 boreholes. I think the first one is WT-24 is the one up l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE., N W, (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l l
90 1 north wnere the water table starts to come up.
7- 2 MEMBER HINZE: Thank you, Steve.
3 MEMBER HORNBERGER: So is the one to the north 4 just below the high gradient area, in the high gradient 5 area?
6 MR. BROCOUM: It is kind of in it. It is kind 7 of in -- their other borehole that was drilled in that 8 area was drilled in the early '80s. It's the first modern 9 borehole up there. I don't have a map.
10 MEMBER HORNBERGER: And do you have any idea 11 how deep? No?
12 MR. BROCOUM: I don't know how deep it is.
13 The second one is in the middle of a block called SD-6, O
(_) 14 and there's a third one underway. So there will be three 15 new boreholes over the next year or so. I suspect they 16 will all be drilled over the next year. How much 17 information we have is another issue.
18 VICE CHAIRMAN GARRICK: Paul?
19 CHAIRMAN POMEROY: I just have -- I have 20 several questions, but I'm not going to ask them of Steve.
21 But I do have one that is just for planning purposes. Do 22 you have some feeling, recognizing the evolving nature of 23 the waste containment and isolation strategy, when that 24 might occur? Because it's on our priority list as well.
/~N
( ,) 25 MR. BROCOUM: There's a level 2 milestone --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
91 1 I'm looking at Jean. When is that level 2 milestone?
,, 2 September.
t
') 3 CHAIRMAN POMEROY: September of this year?
4 MR. BROCOUM: September of this year. Let's 5 see where we are then. But I can tell you, it has been a 6 very difficult thing. So I'll be on the hook for that.
7 Level 2 is a milestone for the director, so I will be on 8 the hook for that one.
9 VICE CHAIRMAN GARRICK: Just a final unfair 10 question, but, Steve, you're the one to ask it.
11 (Laughter.)
12 At least to direct it to. I'm a great 13 believer in the what if game. And if it turns out -- and n\ 14
's / you know that sooner rather than later -- and I don't know 15 how to characterize this in a numerical sense. But 16 suppose it turns out that you've got enough evidence now 17 that clearly indicates that to meet any kind of reasonable 18 standard there is a 50 percent or greater dependence on 19 engineered barriers.
20 Has there been any kind of discussion with the 21 regulators or with the folks that are going to have to 22 deal with this as to whether or not there is a threshold j 23 or there is a dependence at which it would constitute a 24 real problem? And wouldn't it be better to address that x
(;j) v 25 sooner rather than later?
NEAL R. GROSS 1
COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433 l
r
l 92 )
1 MR. BROCOUM: I see a lot of regulators in the 2 room here.
7-i )
'#~
3 (Laughter.)
4 We have discussed that internally, and we got 5 asked that question by the NRC in various forms, usually )
l 6 at various, you know, Appendix 7 type or technical 7 exchanges. I think the concept of separating out the site 8 -- and now I'm speaking as myself here -- ceparating the j 9 site and saying the site is good or bad, independent of a l l
10 system, is kind of a red herring.
11 I think in 960 we made a mistake back in the 12 early '80s when we tried to define the good attributes of 13 the site, not knowing what kind of a system you're going
(^
i l
(-)T 14 to put in. So I truly believe that it is the whole system l l
15 and how the system operates as a whole tnat really counts.
16 And so as we tried to indicate on these 17 viewgraphs, we're trying to create a system, an 18 environment, so that the engineering system can work as 19 well as possible, and an environment so that the natural 20 system itself will not be negatively impacted and will 21 work as well as possible. So I don't usually think of it 22 as 49 percent engineering, 51 percent natural barrier. I 23 think that's a red herring and detracts from the real 24 issues at hand -- is how is the overall system likely to
/";
( ) 25 perform,
! NEAL R. GROS'i COURT REPORTERS AND TRNJSCRIBERS l
1323 RHODE ISLAND AVF ,, N.W.
l (202) 234-4433 WASHINGTON, D C. 200353701 (202) 234-4433
93 1 VICE CHAIRMAN GARRICK: Well, we'll probably 7s 2 address this more later. But it's the red herring, the l i
.(') 3 equivalent to the devil is in the detail. I 4 (Laughter.) ,
I 5 Okay. Thank you very much.
6 I guess now we're going to hear from Jack i l
1 7 Bailey, DOE. i l
1 8 MR. BAILEY: Good morning, gentlemen. I am )
i l
9 Jack Bailey. I am, however -- I knew the title slide i 10 would be good for something. I am, however, the Deputy 11 Manager for Engineering & Integration Operations with the 1? M&O contractor. The agenda lists me as being with DOE, l 13 but I am, in fact, with the M&O.
rN l
! ) '
N/ 14 I'm going to talk to you today, as Steve said, 15 about defense in depth for the repository engineered 16 barrier system. I'm going to talk about the various 17 options and the various aspects and the ways that we can 18 work with engineering, in concert with the site, to affect 19 the overall performance of the system, as Steve described 20 it. You will see some slides that look pretty familiar.
21 In preclosure, as Steve said, it is an 22 operational period that is dictated pretty much, in our 23 opinion, and I believe by the NRC, by the standard rules 24 of reactor regulation, if you will, looking for prevention n
() ,
25 of the occurrence of events, looking at mitigation of the l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
1 (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234 4433 l
1 94 '
i 1 events when the what ifs happen, when the likelihood 1 l
gs 2 occurs, multiple barriers. You want to have some i
\
k-) 3 consrrvatism, you want the redundancy, you want the dual 4 systems, you want diversity if, in fact, you have an event 5 that requires diverse means to deal with common cause.
6 You use a QA program to ensure you get in 7 place what it is that you design for and you get the 8 performance that you expected. You have very descriptive 9 and very precise radiation dose standards, which you can l l
10 model. You use emergency plans and procedures, and )
)
11 included in that is the human element. It is both the 1
l 12 evacuation of the site, but I also include in that the ;
l 13 ability of an operator to deal with an upset. The l g)
\/ 14 operator can go and have some human intervention and solve 15 the problem.
16 And, of course, there are limiting conditions 17 for operations, limiting conditions for operations there 18 co that you can begin from inside the box. The analysis 19 says you start here, so you have to make sure you start 20 there. We have to do that for the preclosure.
21 I always like to talk about that because the 22 engineers have to design the preclosure, and you have to 23 stick with the preclosure. And we have to create a 24 facility that works for 50 or 100 or 150 years, whatever 25 is finally determined to be correct. We have to make a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
l 95 1 facility that works, protects the health and safety of the 7_ 2 public in that timeframe, and yet still does all of the
( )
3 things that it has to do for the postclosure period. :
4 And sometimes those are competing. Sometimes 1
5 things you do in preclosure to ensure control, handling, G radiological release, are a little tougher on the 7 postclosure.
8 But when you go to the postclosure, it isn't a l 9 whole lot different. We want to prevent things. We want 10 to prevent the waste package from failing. In mitigation, i
11 if it does fail, we want to keep it in the package as long 12 as we can. And then once it gets out, we want to hold it.
13 We want to condition it so it will work better in l/m)
\_/ 14 accordance with the na; ural environment.
I 15 And when we deal with prevention, we want to 16 make sure that we set up an engineering system that works 17 well with the conditions that we're going to face. And 18 I'm going to talk about that a little bit more.
19 In terms of multiple barriers, clearly, there 20 is conservatism. Clad is a good example that Dr. Brocoum 21 talked about. Can we demonstrate, through a series of 22 tests, precise numerical values for what the clad is 23 worth? Well, we're doing those tests to try and figure 24 that out.
n
(~_ 1I 25 But we want to protect the clad. Even if we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
96 1 can't take as much credit as we want for it, or perhaps p_ 2 not even get it into the system, we're not sure about i !
3 that. We want to probably protect the clad so that it's 4 there. We don't want to give up a barrier if we can 5 possibly protect it. And those are the kinds of decisions 6 that we have to go through in this defense-in-depth 7 approach. Let's not give up something for something else 8 unless the result is overwhelming.
9 Redundancy is hard to do, because of the long 10 timeframas and the processes that occur, and having more 11 and more barriers, but we'll go through that and talk.
12 There is some ability to have redundancy in effect, or 13 perhaps it is redundancy in the processes that are t
\_ 14 affecting you -- prevention or mitigation.
15 And diversity -- again, it's a little hard, 16 but try and think through some other way to do it. Rather 17 than just making the corrosion barrier bigger and bigger, 18 maybe there is something else that we can do as a means.
i 19 And as Steve described, there is a natural 20 system and there is an engineered system, and we want to 21 play those two together. And there are uncertainties, and l
22 I'm going to talk about that as well. And we want to hav.e 23 those uncertainties so that if we have uncertainties here, 24 then perhaps there is something in the natural system that
,- s
) 25 will cover it. If there's something in the natural (J
u NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
97 1 system, perhaps the' engineering system will cover it, and 7- 2 we'll talk about that. They work tcgether. It's a top-
! \
3 to-bottom system.
4 Yes?
5 VICE CHAIRMAN GARRICK: Jack, not to interfere 6 here, but --
7 MR. BAILEY: Oh, no. Please.
8 VICE CHAIRMAN GARRICK: -- I think that the 9 committee is very aware of what it is you want to do.
10 MR. BAILEY: Yes.
11 VICE CHAIRMAN GARRICK: And that --
12 MR. BAILEY- I'll get that.
13 VICE CHAIRMAN GARRICK: -- all of these steps V 14 are involved, but we're hoping that we can spend most of 15 our time talking about how you're going to do this.
16 MR. BAILEY: Yes. I'm almost there.
17 I want to talk about the QA program, because 18 what we do during preclosure, we have a QA program to try 19 and make sure what goes into the postclosure will perform 20 the way that we want it to. We again have dose standards.
21 We have a goal to shoot for. And as Steve said, we want 22 to do it with margin -- our conservatism argument. And 23 again, we're going to have limited conditions for 24 operations during preclosure to make sure that the 7m (x_/ ) 25 postclosure is set up for operation.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASH:NGTON, D C. 20005 3701 (202) 234-4433
98 1 The goals -- the engineered barriers need to i i
e~s 2 work in concert. This is our design goals. We want them l
( ) l
\
'~' / 1 3 to work in concert with the features. They can adversely 4 impact the natural barriers. We don't want to put l 5 something down there that will ruin the mountain.
l l
6 And we want multiple barriers. We want to ,
l 7 delay failure of the waste package. We want to delay a release of radionuclides from the waste package, as I ;
9 said. And we want to mitigate the effects of that 10 radionuclide release once it occurs. .
1 l
11 I have to meet the preclosure requirements, as i 12 I said. We have to do all of these things, because that's j 13 where we have to start. Then, what we want to do is we r~s l
- 14 want a design that provides an acceptable performance for 15 the expected case. We want to figure out what is the most 16 likely thing that is going to happen there when we put a 17 design in place that gives us gcod performance with regard 18 to what we believe is the most likely case.
19 Once we have that design, then we want to go 20 back and use multiple barriers and look at our whole menu 21 of items to improve our confidence in the engineering 22 system performance, considering those uncertainties in the 23 natural processes, those things that we're not quite sure 24 of and the way they happen naturally, and those (j 25 uncertainties in response to the design features Because NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
99 1 this is such a long-term process, as Steve said, funny
,. S 2 things may happen and we have to do the what if game --
\
% ,/
l 3 what happens if -- and consider it.
4 That's a step-wise process of evaluating the 5 uncertainties, first in the input parameters and then in 6 the response of the engineered features to those input 7 parameters.
8 And I say it again, expected case -- let's do 9 good. Evaluate the options for design features. Use the 10 performance assessment to analyze our performance 11 contributions, and then evaluate the PA, run the cases, 12 run lots of cases and find out where our uncertainties are 13 -- use the sensitivity systematically, design features '
kl -
14 with regard to data uncertainties, and, on the next page, 15 with regard to the design features and options. Let's 16 just walk through them all.
17 And then, when we're done, select the 18 appropriate design features to pick up what we really want 19 to accomplish with regard to an expected case, and then 20 dealing with the uncertainties.
21 Now that's what we've begun to do. We are not 22 done. I won't be able to show you a design today that 23 says, "Here is where we're starting, and here is what 24 we're adding," but I wi31 show you the things we're
('))
(
25 considering in trying to get there. Abe is going to go NEAL R. GROSS COURT REPO9TERS AND TRANSCRIBERS 1323 RHODE 8SLAND AVE , N W (202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
100 l 1 through a case that we have tentatively selected to work,
,, 2 and then show some additional features. But I'm going to e
N ); i 3 talk about the whole menu for a while here.
4 First, however, I want to talk'about the 5 inputs. As an engineer, I want to know what the heck it i 6 is that I'm facing, because it may affect the designs that l l
7 we have to have. We use a series of models. This is your I 8 science -- unsaturated zone, thermal hydrology, thermal I 9 chemistry, unsaturated zone, saturated zone. Those are 10 all of the things that transpire in a mountain and either 11 provide inputs to or receive the inputs from the i 12 engineering. l 13 This goes into the PA and all of the good es
(
(_)
14 coupled processes, and all of the nice time-dependent 15 interactions, all of that stuff occurs. That is very 16 interesting. Being a graphical kind of guy, what does 17 that really mean? What is it that I have to design for?
18 And what happens when I start getting into some of the 19 odder cases?
20 So we sat down -- and I have some pieces here.
21 These are what you're going to see that's preliminary.
22 They were generally done from work done at the ACD 23 timeframe, or slightly thereafter. So they are not 24 completely representative of either the latest flux
() 25 conditions or necessarily the design, but which I have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433 ,
101 1 some calculations of so I could show some numbers.
,_ 2 I have a couple on flux, since flux is of i
! /
3 great interest. And you'll see what we looked at is you 4 have a flux somewhere here in the middle, five or six 5 millimeters per year. Remember that we're getting five or 6 six inches per year at the surface, five or six inches of 7 rain per year at. the surface. On an average basis, we 8 expect to get around five. You may have some wet years 9 where you get 10 times that amount. You have those kinds 10 of years. You have some other periods which may be twice 11 as high. And you have your pluvial periods where it may 12 be much, much higher.
13 So you look at an average flux, and then you
/m, V) 14 look at what happens because of a high rainfall. So you 15 might be seeing numbers like that on an occasional basis, 16 which is what this is intended to show.
17 You also have the mobilized water driven by 18 the heat, the initial heating. We're finding that the 19 package really draws a lot of water. We don't think it is 20 going to get in the drift, but it is something that we 21 have to take a look at because there is a great deal of 22 mobilized water.
23 Now, that says from a design condition I have 24 to deal with a steady flow, and then I have some higher p
25 fluxes. And again, this is in 50/100 millimeters per year (a)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
102 1 on an average basis. That's a general corrosion humidity-
-i
, 2 type concern.
3 Now, the obvious problem is that you have the 4 potential for some episodic flows. What if you have some 5 -- the water runs into the cracks, runs and makes it to 6 the drifts? Well, in this case, let's pretend that we 7 have except that once a year you may get a single B fracture receiving an annual flux from a 400 square meter 9 area, and that number is obviously being negotiated.
10 We're talking about that, but it has some basis. Some I 11 people will defend it, some won't, but we're talking --
12 but a 400 square meter area in a week.
13 So we pretend a whole year's worth of rain
(~'\
(_) 14 falls in a week. It hits a fracture, and two and a half i 15 percent of the waste packages receive all of the flow from l 16 those fractures. Boom.. That boils down to a small garden i
17 hose running for about a week. That's what that means.
18 That is a different kind of a design. That says a drip 19 shield to keep water off of a package is pretty useful.
20 The other suggests that I need te design 21 against relative humidity, and I probably n, ome kind 22 of a corrosion barrier or a ceramic barrier to keep things 23 off for the long term. This ones comes and goes. That 24 one doesn't. This is how I get into the different cases.
,m
( ,) 25 The base case is probably five or six or 10 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
103 l
. 1 millimeters a year. But then I've got to deal with i
,s 8
2 excursions, minor excursions, and then I have to deal with j 4 l
^'#
3 episodes of flooding so to speak. So I try and think l
4 through it in those kinds of manners. l l
9 And I have a couple of other charts --
6 temperature-time. This was done on the aerial load that )
1 i
7 we're looking at. It was done on the ACD layout, which '
8 was a certain spacing of drift packages. It was done j 9 without backfill, but it was done at the very low flux.
10 This is done, I believe, at one millimeter per 11 year. And you can see that you get a significant -- you 12 get a significant drift wall temperature from your hot 13 packages, which is the 10-year old 21 PWR packages. The
- 14 waste package is currently 21 PWR packages. And you can 15 see that for the Hanford defense high-level waste package, 16 for a glass package, not nearly as warm.
17 It's a pretty good effect. It pushes the 18 water away, gets you through the thermal pulse. It tends 19 to keep the water away. It's good for you. This one --
20 not quite as good. Can we use this to shield those? Part 21 of our strategy. That's what we have to think through 22 when we look at graphs like this.
23 Humidity-time -- pretty much the inverse of 24 the thermal. You have a period where you can drive the (v ) 25 humidity way down. Here you can see the Hanford defense NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
104 1 high-level waste. You don't do nearly so well. You don't s 2 get it out of the -- if you believe the 80 to 85 percent
\
NJ ,
3 is where general corrosion occurs, you don't get it out of 1 i
4 there for very long, if at all. If you believe the 50 or i 5 60, because you have ionic salts on it, then you don't get 6 it out of .it at all.
7 So those are . things that we have to look at of 8 how do we protect these kinds of packages with these kinds 9 of packages, although there is a much lower radiological 1 10 content in that.
11 And again, this is for the point load. The 12 packages together may help us, which is one of the things l
13 we're doing our math on this year.
(q 4
\' 14 pH and ionic concentration -- nominally, a pH 15 slightly basic. Normal cement -- maybe we can use a low 16 pH cement to keep it from going as basic. We're looking 17 at different kinds of materials, and you'll see that.
18 We're running sensitivity studies to see if we get a 19 different result. Does it help us?
20 If we go over here and you look at 21 microbiologically induced corrosion, or pitting, then you 22 may end up with a very acidic condition inside of here.
23 Very different problem. Very different problem. And 24 we're looking at materials with regard to MIC resistivity, I
- n. i
(/ 25 and the ceramics, of course, become pretty interesting.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASH NGTON, D C. 20005-3701 (202) 234-4433
105 1 We're trying to resolve pits, although the galvanic action 2 may help us as well.
73
( I x/
3 And down here, the ionic concentration -- one 4 of those things of putting it in a mountain. The
- 5 mobilized water -- well, it gets hotter, more soluble, 6 gets a higher residence time inside the rock. We chose as 7 our standard J-13, which is a well out by the site with 8 the ionic concentration we believe is equilibrated since 9 the water has been there a long time.
10 We chose that as our standard and concluded
~
11 that as you drive this water, as you mobilize this water, 12 it gets hot, it has a chance to pick up nuclides, it 13 doesn't get a chance necessarily to fully cool. off. And k_el 14 you may end up with some fairly high ionic concentrations. l 15 So we may have to give some consideration to a different I
16 type chemistry as it comes back in -- an important piece 17 for us to think about back into the pulse.
18 And down here, water that has -- condensate 19 that hasn't equilibrated, water that comes up and falls 20 immediately, for example. Clearly, that isn't a problem 21 during the preclosure because we'll have enough 22 ventilation and enough heat and such to keep it away. But 23 that may be something that we have to take care of 24 immediately after postclosure.
, ,a 23 And then we look at mechanical load. Over
()
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
106 1 time, this is going to collapse. You have a period here.
. 2 Somewhere out here -- and we just took a date for our rock
( )
3 fall, there is no -- I'm not trying to predict when it's
\
4 going to fail particularly with this chart. But we looked 5 at an eight meter drop. That was the calculation we had 6 for the starter tunnel. That was just the math tha.t we 7 had at hand.
8 And you can see a six-ton spherical rock 9 impacting the package is a very high load. And so we 10 certainly want to keep very high loads like that off of 11 the packages for the long term, at least until the tunnel 12 starts to slowly collapse. We don't want the rock fall to 13 impact the package and perhaps damage the package or r~x 4 L
(_./ 14 damage the different coatings or materials of the package.
15 This may alone drive backfill as a mechanical 16 protection, not even trying to gather the other effects 17 that it has. We may want to do it just from the 18 standpoint of protecting the package so we don't break 19 them from heavy loads. And it turns out that the robust 20 package, the in place package, will easily handle this 21 from a preclosure point of view.
22 Now, the TSPA, in the past, has a lot of 23 capability of modeling lots and lots and lots of features, s
24 but we typically have focused on the subsurface layout f~S t
v) 25 size facing thermal load, support and ventilation system, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D C. 200 5 3701 (202) 2344433
1 107 l
I what effects it has. The engineered barrier system has ,
1
, 2 been mostly invert materials for the chemistry in the
\)
3 system, the packing and the backfill materials, and the 4 use of flow diversion. Can we keep the flow off the 5 package?
1 6 And the waste package has tied itself mostly I 7 to the size and the thermal load, and the materials and i 8 the fabrication techniques, so we can try and maintain the 9 package and have confidence that the package will get l l
10 there.
11 The charts that you're going to see today from l 12 Abe pretty much cover that stuff again. But we went l
13 through and said, "What are all of the things we can do?
,r y C) 14 What affects performance? What should we be taking credit 15 for? What can we engineer into the system? Since we have 16 to be there anyway, what can we do?" A lot of stuff. A j 17 lot of stuff.
I
! 18 I've gone through and tried to characterize it 19 roughly in accordance with the way Steve described the l
{
20 waste isolation strategy with his four pieces, limiting 21 the moisture of the package or the waste package i
r i
22 environment, the robust waste package, limiting the 23 mobilization, and working in conjunction with the natural 24 feature for mitigation once it is released from reducing
/~'N 25 concentration.
!\_-)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE :5 LAND AVE., N.W.
(212) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
108 1 And we tried to look -- and this is just the 3 2 engineer in me. Well, it has a lot of red dots. Do we
- )
~~
3 have a lot? Do we have a lot of yellow? Do we have a lot 4 of green? Do we have a lot of blue? First, do I have a l
5 lot of features to choose from? Is my menu complete? And !
l 6 you're going to see this again on a matrix on the next -
l 7 couple of pages, just another way of sorting it so that I l
8 can walk through it, and we can make those kinds of l
9 decisions.
10 But here are the pieces of the strategy that I l 11 have to work with, and I have a different set of criteria l
12 on the following page. And you'll see I have lots of drip 13 shields in here, and I'd better come to the second piece Y-) 14 for the design options.
15 This is most everything we've thought of, and 16 in some cases we have different ways of accomplishing it.
17 The drip shield is a good example. I've put my little 18 quonset hut out of the material -- free standing, so there 19 is no strain. I've put one that I've placed directly on 20 the waste package, which obviously has some corrosion 21 issues to deal with of having those two things in contact.
22 I have my bat wing, as we call it, suspended 23 in the backfill where I put backfill in. I lay the drip 24 shield over the package and I cover it with more backfill
(_) 25 to protect it mechanically. Or you could even, by the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
109 1 white, coat the package in ceramic, which serves as a drip
, s 2 shield as well.
', ~)s 3 Clearly, some of these have advantages. This 4 is the most desirable, because I can emplace it later, and S I've got 100 years to choose material to make it just 6 right and get it in the right place. On the other hand, 7 as Steve pointed out, it is difficult to put in backfill i 8 above these packages and then place a bat wing, if you l 9 will, or a corrugated one, or build one on top of it, and 1
10 then cover it again. '
11 So there are some difficulties here. Some of 12 them -- that one poses operational problems. That one is 13 pretty easy. That one poses structural problems. But t
s/ 14 they all accomplish the same purpose. And as I said, we 15 also have the ability to coat the package. Or we can coat l
16 the package and put a drip shield on, which Abe is on the 17 case of. So I have several things that I can do there.
18 To walk around real quickly, we have cladding 19 credit. We're looking at the clad. We want to protect 20 the clad, and not intentionally cause the clad to fail.
21 This year we are running tests, we have started tests,
,22 associated with the actual degradation of the waste form 23 itself. Oddly enough, the pellets were ceramic. They are 24 pretty resistant to going away, too.
p
(_) 25 So we're doing testing to see what the NEAL R. GROSS i COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
110 1 dissolution of the fuel itself looks like, and see if that
,, 2 is something that we can reasonably add into the model.
! I 3 The cladding credit itself -- when you come to 4 the waste package, we have a corrosion allowance material.
5 We have 10 centimeters of carbon steel. We have a G corrosion resistant material, a couple centimeters of 7 nickel-based steel, corrosion-resistant type material, the 8 idea being to let the high corrosion period when it's hot 9 and moist, as you saw from the other charts, let that 10 degrade the outer material.
11 And then, when it gets a little cooler and go 12 to the corrosion-resistant material, which because of the 13 kinetics and the type of material, tends to resist very
/]
(_) 14 well. So you have a sacrificial that gets to the one that 15 you want to protect. And we're pleased to report it 16 appears that because of the differences in the material, 17 we actually are able to generate a galvanic potential 18 between the two.
19 And we can cause the outer one to 20 preferentially corrode so that if, in fact, we get a pit 21 or an opening into or through the corrosion allowance 22 material down to the corrosion-resistant material, the 23 galvanic action will, in fact, cause the corrosion 24 allowance material to corrode preferentially, and so
) 25 deferring the corrosion-resistant material. And if we can NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
111 1 push that out far enough, your kinetics get real good and 2 you can make that package very robust.
r .
'~'
/
3 We're looking at a ceramic coating. I have an 4 inside or an outside here. I don't think the inside is as 5 practical as the outside, but you could coat the package 6 in ceramic. If you can find a way to make it bond, then 7 you've eliminated a great deal of the problems that you i
8 have with regard to the initial corrosion. l l
9 Clearly, a bonding process thbt will last that 10 period of time is very interesting. The mechanical 11 protection of that is very interesting, the thermal 12 expansion questions, and we are working on all of those 13 issues as we speak.
El 14 We have large and small and large. We have 15 large packages, small packages, small and large packages, 16 and, in drift, a vertical borehole or a horizontal 17 borehole. Why are they on there, you say? Well, the
- 18 other thing this has to be -- and the reason it's here --
l 19 is that we do have to do 60.21 alternative studies. We 20 have to look at all kinds of repositories, all kinds of l
21 options, and make our deci sions.
I 22 So we're going to look at all of these kinds 23 of things and take them probably not to equivalent levels 24 of detail, because some we're going to submit for a
,-)
(
(
25 license, some we're going to reject, but we have to have a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
112 1 case that says, "Where do we look at them? What was their
, . , 2 advantage and their disadvantage? And how do they
( )
3 integrate into the whole system?"
4 So what you have here is a draft list, because 5 I know of one that is missing from here as I reviewed it 6 again, a draft list of all of those things we hope to have 7 been through and have reports on by the time of -- by the 8 VA. Not to be reported in the VA, but by the end of '98, 9 once we have concluded where we think we are in VA, we 10 will have tried to have worked through all of these to 11 move forward into the licensing process, and, if 12 necessary, make an adjustment to the design.
13 What we have to do for the VA is we have to 14 look at the expected conditions. We have to decide, what 15 does the science tell us right now? What do we know?
16 What do we believe our uncertainties to be? And we're 17 going to go through a case where we select certain of 18 these. We're going to run that expected case. We're 19 going to run our sensitivities. We're going to choose 20 additional features. And that's what we're going to go 21 forward with to the VA.
22 We may get new science. We may have results i 23 from our testing programs on the different materials. And 24 it may cause us to make changes. But as Steve said, the
/')
t 25 VA is a point in time. We're going to walk up to a point NEAL R. GROSS COURT REPORTERS AND TRANSCR'BERS r 1323 RHODE ISLAND AVE., N W.
l (202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
113 1 and say, "Here is a design, and here is what we know. And 2 here are other pieces that we have in our toolbox that we 73 L] 3 can work with.'
i I
4 I'm sorry for the digression. But when you 1
5 get here, clearly that isn't going to be shown in the VA. !
l 6 As we continue to walk around, you have a 7 pedestal. It's important we design the pedestal so as not 8 to exacarbate corrosion of the packages -- create stress 9 points. We have the .nvert. You'll notice I have 10 additives, non-restricted and restricted. We're looking 11 at additives that we can place, not so much in the invert, l 12 but things that we can place on top of the invert. Build 13 it, put it in the burlap bag -- as I like to call it. It
(\
i )
N/ 14 would probably have to be something nonorganic. But 15 basically, place the material in there.
16 There are some materials that show some 17 propensity for absorption of long-lived radionuclides.
18 Our problems are concentration. The stuff that we want to 19 catch may not come available for 10- or 20- or 30,000 20 years, and with some moisture in here it may all be gone 21 before we can use it. So we're looking at those kinds of 22 issues. But we're looking to see, is there something we i 23 can throw down here?
24 And we looked at both the non-restricted and.
,m
( ,) 25 the restricted. The reason I put that in is some are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 Rh0DE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 1
i
114 1 mineral in nature. They can be put down there pretty g3 2 readily. Some are more metallic in nature and provide
() 3 some health issues. And so those I really can't look at 4 prior to VA, so we are going to do those studies to see if 5 there are some other types that we can put in. But I 6 won't necessarily have that done in that timeframe. But 7 we're certainly not ruling it out.
8 As I said, I went through the various -- as I 9 said, I have three different types of drip shields that 10 we're looking at. The backfill -- I have an upper 11 backfill, which is a rock fall protection, mechanical. I 12 don't believe it does teo much for me with regard to p
13 performance, other than protection of the package.
( I i 14 It can be used to limit flow and humidity. We 15 put up Richards barrier. I don't like trade names. It 16 really should be a multi-level barrier, multi-material 17 barrier, something that tries to draw water a;ay from the 18 package and keep the water away. It might be a single 19 layer, if we find we gather enough performance, it's 20 easier to put in, we have a higher confidence that we can 21 put it in.
22 And we might add something to it, similar to 23 using the invert, to try and condition the water as it 24 leaves -- prepare t mountain. And that is a wag. We're
./~'s
(_,) 25 thinking that through. We think there is something we can NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
& O2) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
115 1 do there, but we're not sure what it is.
7 2 When you come to thermal design, here is where
!' ') 3 the preclosure becomes kind of interesting. Clearly, the 4 aerial mass load can become important. How hot is it?
5 How do you want to mobilize the water? Do you want it 6 hot? Do you want it cold?
7 The waste package spacing in a point or a a line? The drawings I showed you -- significant 9 differences between hot packages and cold packages. Can !
10 we move them closer together in a line and shield each l
l 11 other? Can we shield them so that one protects another 12 with heat? Because we believe the heat helps us in the 13 long run of keeping the water away. l
.~x l
(_)) 14 With that, we can do some spent nuclear fuel 15 assembly blending. And we may have to do that in the 16 facility. We're not sure yet. We are evaluating that.
17 Meet the criticality limit, which is my second one. To 18 meet the 18Kw limit. We do place a limit on the total 19 heat in the package so as not to cause adverse reaction 20 and problems to the rock and the concrete itself. So we 21 have those kinds of limits. We may have to blend fuel.
22 We may not be able to just take things and throw it 23 together. We may have to have a little lag storage -- put 24 stuff together to get it into the right piece.
r~N iv) 25 On the other hand, we could potentially, with NEAL R. GROSS COURT REPORTERS AND TRANSCR'BERS 1323 RHODF ISLAND AVE., N W (202) 234-4433 WASHING TON, D C. 20005-3701 (202) 234-4433
l 116 l
1 that kind of a storage, load up packages to a higher heat
.s 2 level than hot / cold, hot / cold. We might be able to move
( }
^#
3 those heats around a bit, which we're also looking at, and l
4 that's to control thermal variability, j 5 We're looking at waste packaging sequence. ,
1 6 Once we get it down, how do we load it into the drifts? l 1
7 We certainly don't want to be stuck necesscrily putting a 1
8 lot of hots together and a lot of colds together, so we're 9 looking at keeping four drifts open and being able to load 10 whichever drift is most important for the thermal. And we 11 can do that on the subsurface or the surface. We can hold 12 it on top. We can hold it on the bottom.
13 MEMBER HINZE: Jack, in terms of the materials (3
(_ l 14 that are planned to be put into the repository, are all of 15 them such that they can go inside the canisters such as 16 you are picturing there?
17 MR. BAILEY: Yes. That is our intent at this 18 point in time. Our waste streams consist of commercial 19 nuclear fuel, commercial spent nuclear fuel, which is the 20 bulk of what we get. One plan is longer than the other, 21 so we have a longer package.
22 The defense high-level waste -- we're 23 basically using the same package and we're packing either 24 four or five of the glass logs, which are actually (n/
v 25 canisterized into a stainless steel package, and loading NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
t 1
117 l
1 those into the package. And the defense -- the DOE spent l
7_s 2 nuclear fuel, which we are now looking at, we believe that
!, \
3 we are also going to receive that inside of a stainless ,
l 4 steel package. And we will probably co-load it with the l l
5 glass logs. That is our belief at this point in time. l 6 And we did that mostly for criticality reasons. i 7 VICE CHAIRMAN GARRICK: Is retrievability a 8 factor in these design considerations? l 9 MR. BAILEY: Yes, it is. We are required to I
10 maintain retrievability for the regulation, and our 11 approach -- and I don't have a drawing of that very well, 12 but I'll talk about it for a minute -- the way that we 13 have retrievability set up, or the way we have delivery of
?m
( )
x._ / 14 the package set up, is that you build a drift, you lay in 15 the inverts, you put up the pre-cast segments -- and I l I
16 haven't gotten to that, but that's wh.at this is meant to 17 be. Right now, we're looking at pre-cast put-up segments.
18 There is rail lines right here, and those rail i 19 lines are intended to use a gantry crane. We'll set the 20 pedestal, and then we will bring in a waste package and 21 set it down on the pedestals, unhook the slings, and bring 22 the gantry crane back out and allow the package to sit 23 there.
24 So that at the time of emplacement you will
,< y f.,
%- ) 25 have the drift, the invert, the pedestal. We could I
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASH:NGTON, D.C. 20005-3701 (202) 234-4433
118 1 conceivably add in invert material at thie point in time
,m 2 prior to putting it in, provided we -- I'll go slower.
t 1
\v l 3 The additives -- we could conceivably add additives here 4 below the pedestal, away from the rails, before we 5 emplace, which would probibly be easier because you can 6 locate it better without the package in place. And then 7 you would put the package in place. And that is pretty 8 much all that would go in there until closure.
9 You might -- if we want to do a ceramic 10 coating, you might coat the package before you put it in.
11 Some of the ceramics -- they have the capability of being i
12 put there in place. And so you could actually dress one 13 up if you wanted to. You could coat it, take it down, set c
14 it, and dress it up in place if you had to. So the i
15 ceramic might go in place. Other than that, all of this )
16 would be done at closure, once retrievability is no longer 17 required. l l
l 18 The place I haven't caen is the drift liner, 19 and you'll notice I can use normal or low pH concrete, as -
l 20 I discussed with my other chart. Pre-cast concrete -- and 21 we're looking at three kinds. We're looking at pre-cast, 22 which is the segmented piece that you look at with an 23 invert, and then several segments. We're looking at a 24 cast-in-place, which would be a poured, if you will, dw Q) 25 behind the machine, or you actually come back and do it NEAL R. GROSS COURT REPORTERS AND TRAMSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
119 l
1 again. We are also looking at steel sets. ;
1
,3 2 We are also looking at -- if the PA guys tell
! )
'~~
3 us that concrete is just unacceptable, we have nc 4 foreclosed the steel option. We have the ability to put i
5 in a completely steel, if necessary, and that is being 6 designed and running through the math now. l 1
7 And for the VA, we have the capability of '
B using any of these three. We're, in fact, carrying all l 9 three designs, and he is going to evaluate all three i
10 designs so that we can make a decision in that regard.
l 11 You'll notice here the crown joint. The PA 12 guys like to -- I like this one, because the way they like 13 to put up cast-in-place concrete, it turns out, is you i
i 14 like to put it up, as you might expect, get those last two 15 pieces, spread them, put in a crown joint, and then let 16 them come back and relax into position and bear on 17 themselves. And they like to put that at the top -- you 18 know, it's symmetrical, and put it right on the overhead.
19 Unfortunately, you make yourself a nice little 20 drip area right within the shadow of the waste package, so 21 we're looking at putting that over here. Certainly 22 capt le . Small item. The PA guys like it. They think --
23 they look at me askance, because how long is that going to 24 be good for? The stuff is going to fall Well, we may A
) 25 decide to cause this thing to fall so that this piece NEAL R. GROSS COURT REFORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
120 I falls.
,_ 2 We may design it that way so that that piece t ;
\ /
3 of concrete lays on top of the backfill for a long time 4 and helps divert in that regard a: well. We may go to 5 that level. So we'll probably put the crown joint over on 6 the side, so that at least during that preclosure period, 7 and while the ground support is intact, you don't have 1 8 that naturally built joint right across the top. ,
I 9 Now, clearly, just to be fair, it is a three-10 toot segment, so there is a joint this way as well. But l l
13 if -- you know, there is no crown joint in the 12 longitudinal, but there is one in the axial. So we're 13 going to have to look at that and see if there is a way to C.) 14 seal or a way to divert, but that is a whole lot more 15 likely because you've got a place to take the water. You 16 can drip it into grooves and run it down the side and 17 perhaps by something.
18 And up here on the top, the surface control, 19 we can do land withdrawal. We can control the area. Now, 20 administrative controls for 10,000 years, I don't think I 21 can go to. It would be nice if I could do that one, but I 22 don't think I can. But we can look at land withdrawal.
23 We also can look at drainage structures on the 24 surface. I think we have to do that, if for no other (w.s ) 25 reason than to answer the what ifs. Why don't you rip the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
121 1 top of the mountain off and pave it? I think we have to
,,_ 2 go through and find out what that does for us and if we
('") 3 can do it and how.
4 And the one that I missed, unfortunately, was 5 the altered near-field rock -- a couple of interesting 6 pieces there. One, the tunnel boring machine tends to 7 create some stress effects, as you might expect, in the 8 rock There is some argument as to whether or not that l 9 causes water to drain around the drift or causes water to i l
10 go to the drift, and we've got to figure that one out.
11 But something is going to happen there. Clearly, you have 12 a stress affected zone.
13 The other thing we are finding is that in this I
(]
(_) 14 heat pulse, we're finding to a certain extent that as the 15 moisture drives all of these minerals, these dissolved 16 solids up, they go up and then they retreat as the 17 temperature changes. And they seem to be leaving some 18 behind. And, in fact, you may end up with kind of a 19 natural umbrella being built over this.
20 when you drive this thing all the way to the 21 top of its pulse and it sits there and percolates for a 22 while before it starts back down, you end up leaving a lot 23 of deposits. And there is some belief that those deposits 24 may form somewhat of a partial shield, if not a whole 7
25 shield. And so we're going to spend a little bit of
()
i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISi.AND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
122 1 effort looking at that and see if we can control it.
2 And if that turns out to have a big effect, 7_
( )
3 then how we do our thermal loading and how we put things 4 in is pretty important. Unfortunately, that's one of 5 those things that is very hard to quantify and prove. But 6 it's one of those layering arguments, as I call them, that 7 says there is something there that may be helping us, and 8 we ought to talk about it, and we ought to show it to you.
9 There is one missing -- test, test, test.
10 There is one missing -- filler. I didn't talk about 11 filling the waste package with material. We can do that 12 with steel shot. We can do it with other materials. It 13 makes the packages a little harder to move. That's why it n
(_)
14 isn't on there. My engineers really don't want to pick up ,
i 15 a package that heavy, but it will go back on there. )
16 And it has -- it hurts us thermally. Oddly l
17 enough, it doesn't conduct the heat as well. On the other 18 hand, it has some good effects for criticality, which I 19 haven't talked about other than up in here. And if you l 20 notice, I didn't talk about it in the beginning either, 21 but I will in a moment.
22 Questions on the chart? I can always put it 23 back up.
24 VICE CHAIRMAN GARRICK: That's quite a chart.
/~%
(v ) 25 (Laughter.)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIB'ERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
123 1 MR. BAILEY: Yes, sir. And 1 don't think I s 2 have everything on it yet. I don't think I have all of
( )
'~'
3 the explanations down, and boy, do I hit on a lot of 4 people that I want better answers. But it does a very 5 good job of focusing us on what we're trying to do, which 6 is why we created the chart and why we're trying to deal 7 with the next chart, which is how us rational, logical 8 engineers think.
9 CHAIRMAN POMEROY: Before you do that, can I 10 just a quick question?
11 MR. BAILEY: Yes, sir.
12 CHAIRMAN POMEROY: I recognize that we're not 13 supposed to think about retrievability after closure. But
(
rm E/ ) 14 does anybody, in the course of engineering, look at the 15 question of whether we're making it extremely difficult 16 .for ourselves versus easier?
17 MR. BAILEY: Oh, yes.
18 CHAIRMAN POMEROY: Is that one of the things 19 that people look at?
20 MR. BAILEY: Yes.
21 CHAIRMAN POMEROY: Or if we, for some 22 catastrophic reason, had to go back and ask for --
23 MR. BAILEY: That's after postclosure?
24 CHAIRMAN POMEROY: Yes. After closure, not --
r~
(,N) 25 MR. BAILEY: After closure. After NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
124 1 postclosure. That would be a good one, fm 2 (Laughter.) l b 3 When is that? )
4 (Laughter.)
5 No, after closure, our intent is that to get 6 the license to close, we believe we have to be able to 7 have an awfully solid case that says this is the right j 8 thing to do. And in order to do that, if we need things I 9 like -- I have my black and white version of it which you 10 probably don't want. If you want to do things like drip 11 shields like this, or you want to put backfills in, or you !
i 12 want to put restricted materials in, those things become l 13 very difficult.
O V 14 If we start putting a restricted material in l
15 here, for example, boy, did retrieval just get really l 16 hard. Up until i hen -- well, let me back up for a second.
17 Clearly, we have to be able to deal with a package that 18 suffers a rock fall. I mean, we have 120 miles of tunnel 19 open for 50 or 100 years, and we design for it, and we i 20 analyze for it -- look at my regulator over there -- and l
21 we'll show that we have protection for it.
22 We also have to be able to recover that. And 23 so we are designing -- and they are kind of funky looking 24 things, but they are basically extravagant backhoes that O
25 can cause you to dig one of these out. They can put up N EAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 13;3 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005 3701 (2u2) 234-4433
125 1 shoring. It looks like a big tow truck that can pick the n ? package up. So there are ways to deal with what we call
( )
3 waste package movement and problems in the preclocure, 4 where if the walls collapse we can go down and get it.
5 Now, that could clearly be extended into the 6 postclosure arena.
7 CHAIRMAN POMEROY: If you needed to.
8 MR. BAILEY: If we had to, but it would make 9 it very, very difficult. The regulation suggest.s that you 10 should be able to do retrieval during the same timeframe 11 as emplacement, which is 25, 35 years. If we load all of 12 this kind of stuff in, I don't think we could do it in 25 13 or 35 years.
p b 14 On the other hand, it is after closure, and so 15 we would expect at the time, I believe -- probably a 16 policy question for Dr. Brocoum -- but I believe we would 17 expect that at the time of closure that the retrieval 18 requirement would be lifted on us. But is a factor that, 19 if we put a restricted material in here, clearly sending 20 people down to muck this thing out. isn't going to work if 21 you have some kind of a restricted material here. So that 22 becomes very important.
l 23 But everything else is still there. I mean, 24 you still have an air space up above. We can't fill this
(~%
i{j 25 to the top. We still have ventilation. We could l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISLAND AVE., N W.
i (202) 234 4433 WASHlNGTON. D C. 20005-3701 (202) 234-4433
1 1
126 1 conceivably -- the ventilation is in the middle, so we l
2 could conceivably still have the ventilation and keep the 7_ >
r ,
3 v( Itilation viable if we had to, if you wanted to dig them 4 out.
5 So those are factors that could enter into the 6 decision, but --
7 CHAIRMAN POMEROY: And somehow they're in your 8 mind as -- 1 9 MR. BAILEY: Yes. Yes.
l' 10 CHAIRMAN POMEROY: -- it goes along.
11 MR. BAILEY: We look at this very much, and I 12 really like the design because I only foreclose - I hate 13 to use that word. There's only a few things that I really O)s_ 14 set. I set the package and its materials.
15 CHAIRMAN POMEROY: Right.
16 MR. BAILEY: And that's while I have testing 17 going. And clearly, we're at the beginning of our long-18 term testing. But before we actually start manufacturing l 19 packages, we have the opportunity to have probably lo l 20 years worth of accelerated testing done. So we probably 21 have a pretty good idea of materials.
22 We should be able to make significant 23 decisions on how to protect the cladding or not protect 24 the cladding. The operations of this have the ability to
/s
( 25 affect the cladding. It's sensitive to heat -- pretty (f
NEAL R. GROSS COURT REPORTERS AND TRANSC-RiBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D.C. 20005 3701 (202) 234-4433
127 1 high temperatures, Lut we have to be careful how we do 7_s 2 this. So the selection of the ceramic becomes real I ') 3 important to make sure that we don't overheat the clad.
l 4 The placement of the backfill in time is 5 important, so that we don't overheat the clad and cause 6 the clad to fail. Because when Abe talks to you, he is going to talk to you about the assumptions on clad 8 failure. And so it's important to maintain the integrity 9 of the clad if we possibly can.
10 We set the pedestal, and we set the materials 11 insi de the dr.i f t . And beyond that, the rest of it is I 12 pretty much aiditive. And we can make choices, we can 13 select new materials, they can get smarter people than me l'h
(-) 14 on the job that can think up new and oddball things to put 15 on there, and we have some time to work with that.
16 Clearly, our job is to demonstrate with 17 reasonable acsurance that we can accomplish this at the 18 time of LA. In reality, you can continue to do things, 19 because if you have a base case that looks like it works, 20 clearly you may be able to improve on it, or you may find 21 that you have flaws with it as you continue the testing.
22 CHAIRMAN POMEROY: Great. Thank you, Jack.
23 MR. BAILEY- I go back to my design features 24 evaluation matrix. I'm not going to try and go through
./
25 the Xs and the Ys with you, but I wanted you to see the (u %)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 HHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTCN. D C. 20005-3701 (202) 234-4433
l 128 1 thought process. Down one side I have all of my features,
-S 2 and they're the same features that are on the other chart
'~
3 with cne exception. I made a mistake. I didn't get the 4 multi-level barrier in there. They got typed separately 5 and I missed it in the review. So I apologize for that.
6 But if you take a look, I had some postclosure 7 goals that I talked to you about. What am I trying to do a in engineering? I'm trying to delay the breach --
9 prevention. I'm trying to prolong the time from waste 10 package breach to waste release, and I keep the stuff in 11 the package. And finally, once it is released from the 12 package, what can I do mitigating its release and 13 conditioning itself to deal with the rest of the i )
k 14 environment?
15 Those are my goals. So I went through and I 16 ranked everything that I did with primary and secondary.
17 What is this thing really doing for me? And if you stay 18 off of this side of the chart and look, you will find that 19 we have done a heck of a lot of designing for prevention.
20 We aze trying to keep that package intact. We worked at 21 it really hard.
22 By the way, we get some good pieces on 23 prolonging and mitigating. Mitigation is not quite as 24 filled as I would like, and I've challenged my people to n
h 25 i
sit down and think about how do we get a few more things NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
129 1 in that column. And there's another page to this, which I l
g 2 will flop up in a minute, but it's the same thought L) '
3 process we went through.
l 4 This gives me a visual of what is there. It 5 gives me a visual of, hnnn, look at all of the things that 6 I can do for prevention -- almost a redundancy argument.
7 If I put another column on here and look at the factors j 8 that affect it -- and that's what we':ce going to do --
9 then I can start looking if this factor affects this one, 10 this factor affects this one, and maybe I can get the 11 thoughts of diversity into it. So we have to think 12 through that. We can look at, what do we get good credit 13 for? What do we not take too much credit for?
,a s I
'w/ 14 Wnen you come over to the postclosure 15 environments -- well, let me stop for a second, because 16 there are a couple of other postclosure goals. They are i
17 subs -- one of them is a sub of delay the breach, and that 18 is a microbictogically induced corrosion. It probably 19 deserves .tt s own column , so that we can see how to deal 20 with that. We have to make sure that we've covered that.
22 , I didn't -- that's an emerging issue, and it was hard to 22 get on the chart so I didn't put it.
23 The other is -- the postclosure goals --
24 clearly, we have to deal with criticality, and criticality
,rx
(,! 25 folds itself to a certain extent under delaying, if I keep NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C 20005-3701 (202) 234-4433
130 1 the package intact. I don't have it. Prolonging the time
,s 2 of release. Well, that's desirable because once it gets N) 3 into the EBS you have to look at the ex-package 4 criticality. So we have to look at in-package, we have to 5 look at ex-package.
6 And in actual fact, I don't have it with me, 7 but we've made up a similar matrix dealing solely with l
8 criticality and the basic steps that occur with 9 criticality. And we're looking at the design features and l 10 the options to see what really gives us some work on 11 interfering with the capability of criticality.
1 12 And so we're going to walk through that kind l l
13 of a process -- the same tn_ng. What is the expected I
(')
(/ 14 case? What are the things we can do for the expected 15 case? Wnat are the sensitivities? What are the things i
16 that are really problematic for us? Making sure not to 1
1 17 exacerbate the expected case while we look at the odd 18 cases or the more uncertain cases. I 19 And, of course, it ic real important not to 20 get overly conservative, because you may drive a design 21 with too much conservatism. The strength of PA, in my i 22 mind, is to look at what is likely to occur. Don't overdo
, 23 it. That's why when they brought me the charts on the 24 flooding they had the thing four times higher Wait a
,m
( ,) 25 minute. You know, 1 don't want to put fins on this thing NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 AHODE ISLAND AVE., N W.
l
, (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
l 131 1
1 unnecessarily. Let's figure out what is a reasonable i 1
-3 2 value. Let's find what we think is going to happen. !
. )
\
3 Let's not force ourselves through conservatism into doing 1
l 4 too much.
5 The uncertainties review is where we'll take 6 the likely case and then deal with the uncertainties.
l 7 Don't make the extravagant case the design case. Which is ..
8 a little different than you look at in the Part 10 world.
9 Take the absolute worst condition of this and i 10 this and this and you've got it covered. Instead of that, 11 we want to look at the expected case, and then we want to 12 treat our uncertainties. And we want to make sure we fill 13 in enough of these columns, so that when the thing we
,q f
C/ 14 don't expect to have happen -- we have some capability to 15 deal with it.
16 On the postclosure environments, I looked at 17 flux, humidity, chemistry, rock fall, and drift collapse.
18 In reality, I should have broken flux into two and looked 19 at it at the mist, and looked at it as the episodic. And 20 I will do that. I have not yet for this chart, because 21 there is different design pieces and which one is better 22 against which.
23 I also don't have on here radiologically 24 induced corrosion, which is a subject which I am not as n
25 familiar with as I'd like, and I missed the expert (v)
NEAL R. GROSS COURi REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON. D C 20005-3701 (202) 234-4433
132 1 elicitation yesterday so I have to catch up on it. But 7_ 2 there is probably another environment that we have to deal 1'-)
~
3 with here that I don't have listed. l 4 And what we looked at is, what are we really 1
5 protected against? And what did I find? Hey, we've been 6 worrying about water a whole lot. Humidity -- yeah, well, 7 it kind of comes with water. T"a chemistry -- not as much l 8 as I'd really like. We really need to think about that 9 chemistry aspect. And the rock fall and drift collapse, 10 we haven't looked very much at mechanical loads.
11 People have not really been too worried about 12 crushing the package or dinging the package, or what l 13 happens to the package in the long haul. And maybe n
f ')
(_' 14 everybody has in the back of their mind that it's going to 15 be backfilled and just generally collapse onto itself as 16 it degrades. But we have to go through that and get out 17 of the back of the mind, get it up front, and put it into 18 a thoughtful process that says, how have we covered this?
19 Then, the second page is more of the first 20 with the rest of the pieces. And we just walk through it 21 all And we'll be continuing to evaluate this, doing 22 these evaluations, providing these reports that says here 23 is what we select and why, all with the able help of PA 24 who does our evaluations for us.
A
( ) 25 You've heard this. Design to the expected NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
133 1 case. Determine the key sensitivities. Do the 7~ 2 sensitivities to the PA. What really drives the PA?
Let's make sure we have the base case really covered. The i
4 value uncertainties in your data -- what is going on in l 5 the mountain? How do our inputs arrive? Look at the l 6 uncertainties of those. Make sure that we can deal with i
7 them. Evaluate the uncertainties in the rv.sponse of the !
8 design.
9 Take a look at all of them for conservatism, 10 which may be orders of magnitude, as Steve has suggested.
11 It may be -- I didn't wear my belt, but I am wearing 12 suspenders -- your belt and suspenders of having extra 13 things that maybe you can't take full credit for but p
s- l 14 they're there. I have to go back to my chart. I left one 15 off. In fact, I'll go back right now for things that --
16 how long can you take cred1+ for it? Because I did miss 17 one as I walked around it. I always do.
10 Affect on the zeolites -- the thermal load has 19 the capability of affecting the zeolites. If we make it 20 too hot, we can degrade their capability and how much of 21 them is there. Another one of our goals is we don't want 22 to burn off the zeolites. It's a barrier. It's there. I 23 don't know -- maybe Abe can tell you how much water 24 actually makes it through the zeolitic beds, but why take
/
() ,
25 it away? It's available. It's a barrier. How much NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
134 1 credit we get for it -- probably not a lot.
fs 2 But if we're wrong in the what if game, Dr.
! \
'~
3 Garrick, what if the water goes that way instead of that 4 way, well, I don't want to take away a barrier that I 5 might need. And that was on conservatism, probably a good 6 example, I hope.
7 Redundancy, diversity, and we need to make a a few more columns. And then we have all of that stuff. We 1
9 look at the total TSPA, the total total system performance i
i 10 assessment, and we look at all of the design features, and j I
11 we look at it in the entire system and make our 12 selections. It isn't my choice. It's the system's I 1
13 choice. I provide the menu. )
14 And I hope I escaped the preclosure soon 15 enough.
16 VICE CHAIRMAN GARRICK: Thanks, Jack. I do 17 like your superchart.
18 MR. BAILEY: Okay. Well, it puts things in 19 perspective.
20 VICE CHAIRMAN GARRICK: Right.
21 MR. BAILEY: It gives you an idea of what it 22 is that you're really looking at. And I would like to add 23 one thing to that. There is almost nothing on that chart 24 that hasn't been here for three to five years.
,m 25 VICE CHAIRMAN GARRICK: Right.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W-(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
1 ll 135 i
1 MR. BAILEY: We have talked about it. We have i
,, 2 looked at it. All the chart does is the chart says, "Here i
- i
~
3 is all of the things we have to work with." And the 1
1 4 diffe.;ent -- the other charts tell you, "Here is how we 5 c a n u s e '*. hem . " So we're trying to bring that together.
\
6 sVICE CHAIRMAN GARRICK: Well, before our j 7 break, let's go around the room a little bit and get some 8 questions. ,
9 Marty? j l
10 MR. STEINDLER: Yes. Your large chart strikes j 11 me as an open-ended invitation to break the Treasury's l
12 back.
33 (Laughter.)
V 14 MR. BAILEY: That's why there is no -- well, 15 there is a little bit of green on it. But that doesn't 16 mean they're expensive.
17 MR. STEINDLER: You clearly have to have, I 18 assume, some kind of method of prioritizing, number 1.
19 Number 2, I didn't hear any words about criteria for l
20 performance of each of these barriers. I also didn't hear 21 any words about independence. I heard something about i
22 diversity, and somehow that seems to be a magic term, 23 aside from the fact that I'm not sure what it means in 24 engineering terms.
A t., j 25 Where are you in some kind of a systematic 1
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE , N W.
- (202) 234-4433 WASHINGTON. O C. 20003-3701 (202) 234-4433 l
136 1 approach to, well, cost-benefit or setting criteria for
,-. 2 each of the various barriers that you're even thinking i, \,
3 about? What is it you're trying to get done here in terms 4 of more hard numbers?
I 5 MR. BAILEY: A fair question. What we're l
6 doing is we're actually walking through each feature 7 through the PA, as best we can model it, and they've got j 8 to do so.e tweaking. Unfortunately, they can't model it 9 all today. We're walking through each teature to see what 10 each features buys us independent of all of the others.
11 What does this feature do or that feature?
12 Now, clearly, there has to be some kind of a 13 base case. Clearly, you're going to have some kind of a l rh
.. 14 waste package and you're going to have some kind of a --
15 there are certain pieces that have to be there, but for 16 the additional features, for walking through them one at a 17 time to see what kind of a performance we get.
l 18 What we're learning from that exercise is that 19 most of these things worked coupled. An individual piece 20 isn't really very helpful. It gives you something, and 21 then the next piece will give you something, and the next 22 piece will give you something. But if you put three of l
l 23 them together, you get seven times something. And so 24 we're walking through each individual one to learn about
('%
( ) 25 the piece -- what it does and what its sensitivities are, l NEAL R. GROSS i COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D C. 20U05-3701 (202) 234-4433
137 1 so that we can learn that piece. And then we're trying to
,. 2 put them together into groups.
(' ^ ~ ')
3 MR. STEINDLER: I'm not clear as to whether or 4 not the analysis of what a particular piece can do for you l l
5 is based on state of the art, currently gold-plated, the 6 best you can buy, or whether it has some bounds on it.
)
i 1
7 Clearly, for example, if you want to put ceramic coating a on a chunk of metal, you can make it the thickness of the 9 average car paint, and you know how long that lasts. l 10 MR. BAILEY: Yes. )
11 MR. STEINDLER: Or you can do a heck of a lot 12 better job on it.
13 MR. BAILEY: Yes.
/7
'(s') 14 MR. STEINDLER: What frame of reference are 15 you using to do this evaluation?
16 MR. BAILEY: Well, we have a 10,000-year 17 standard, so we try and do -- and I think Steve alluded to 18 it a little bit. We try and look at, how do we do it with 19 regard t) the 10,000-year standard? We try and look at 20 what does our peak dose look like over a lifetime. How 21 many orders of magnitude can we drive this down? See what 22 kind of a performance it gives us. And we try and pick 23 some periods that make sense to us. The therma pulse is 24 one that makes sense to us, to get past the thermal pulse.
c
() 25 So we're looking at it in those kinds of ways.
NEAL R. GROSS COURT REPORTERS AND TRANSCR BERS 1323 RHODE ISLAND AVE., N W (202) 234 4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
138; 1 MR. STEINDLER: Let me stop you.
,eq 2 MR. BAILEY: I'm still not addressing your A
3 question?
4 MR. STEINDLER: No. ,
5 MR. BAILEY: Okay. I 6 MR. STEINDLER: The 10,000-year standard -- l 7 your comment implies that you want this waste package --
8 the engineered barrier aystem to last 10,000 years.
9 MR. BAILEY: No, no, no. I try and look at 10 results at the end of the 10,000 years, and then I go back 11 and look and see, when does it fail? What is the 12 probability density of when which ones fail? And how does 13 it drive it? And can I shift that out?
/^' ,
t /
! -- 14 What we're finding is that if we can shift a 15 couple thousand years -- I know that is hard for an j 16 engineer to say. If you can shift it, the corrosion 17 kinetics get really interesting, and this corrosion-18 resistant material, boy, it will last a long time. So if 19 there is a way to push this thing out a little ways, you l
j 20 get tremendously better performance. And those are the i
l 21 kinds of sensitivities that we're learning as we go 22 through individual and coupled evaluations. And that's 23 what we're trying to shoot for. That really wasn't as 24 clear to us before.
,r3
() ,
25 With regard to cost, we're trying to cost them NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
139 1 as best we can. Some of it is developing technology. We
,w 2 make the phone calls. We talk to the vendors. We find
() 3 out what is current now. And then we work at what kind of 4 material we believe will give us the performance that we 5 want to dial in to the PA.
6 When Abe talks to you -- they're just going to 7 be dying for you, Abe -- when Abe talks to you, you're 8 going to see he has different lengths of waste package 9 lives, or of the ceramic coating life. Is it a 3,000-year l 10 ceramic or 10,000-year ceramic? Tough numbers, I 11 understand.
12 But you look at those kinds of numbers, and 13 you look at what kind of performance you get, and ther we !
14 come back and he says, "3,000 will do us." What kind of a 15 package do you need for 3,000? What kind of ceramics?
! 16 What kind of materials? What is the cost of that? How 17 can you do it? What do you have to do?
18 And so it's an iterative process of, I think I 19 can do something like this, he evaluates it, and we have 1
20 to come back and find out what he expects from us. So l
21 it's a back and forth process to do all of those things.
22 And as Dr. Brocoum said, we have tried to do 23 some costing on very basics. I did not cost the thickness 24 of a package per se to the millimeter. We talked to
( ,) 25 vendors about what it takes to do a coating of a package NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N.W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
140 1 to some nominal depth, which we believe, based on the 2 literature, is probably enough in talking to the s >
3 scientists. And so that was how we came up with the rough 4 cost.
5 When he does that better, we'll take our 6 science, we'll find out what kind of numbers we need, 7 decide our margins, and then we'11 do the real math. But 8 it turns out for the case of ceramics that it ain't a lot
)
9 different when you get down to doing that. And if it is 10 extremely sensitive to it, then we have to do it. And no, !
j 11 I didn't put cost on my chart. It will be on his chart, 12 because it has to be there. It has to be a tradeoff.
13 VICE CHAIRMAN GARRICK: George. I
,q il 14 MEMBER HORNBERGER: I have two questions, ;
15 Jack. One is more a philosophical. I'11 start with that 16 one. To what extent -- well, let me preface it. Suppose 17 we actually do, at some time in the future, get to 18 implement some des 1gn. To what extent do you build into 19 your thinking a " design as you go" kind of concept?
20 That is, this ic, all fine, it's a paper study.
21 The question is, once you get to the actual drifts and see 22 the conditions of the rock, to what extent do you envision 23 your hands being tied that the spacing will be exactly 30 i 24 meters and the backfill will be exactly this?
,m What flexibility do you let in your thinking (v) 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 2344433
141 1 for changes during construction?
,f- 2 MR. BAILEY: Good question. Having come out I
'~#
3 of the reactor world and the Navy world for a short 4 period, I'm a pretty structured guy. It wau quie- a piece 5 of excitement getting miners to do exactly what you say.
6 Hmm, I think we'll turn a little here; I think we'll go 7 down a little here; I don't like that; let's put some 8 other things over here.
9 And so naval calibration screw is getting hit 10 here for the fact that you can't do things perfectly. And 11 the miners do a very good job of keeping track of that.
12 Clearly we want to characterize the block. We want to 13 drive the drift so that we have a good capability of being
- y i i
\_) 14 able to put them where we want them to go.
15 We don't think that we're going to have too 16 much trouble in putting them where we want to go. We can 17 design that and we can do the analysis for it. And we 18 have the capability to do the as built. Backfill -- they 19 have told me very clearly that, you know, don't ever show 20 it smooth on top.
21 We're never going to take a rake back and make 22 that thing smooth, it's going to be lumped in there. And l
23 that's one of the reaecns -- the batwing is clearly my 24 favorite drift shoot. It's suspended, it's up there, it's
( 25 never going to get moved by anything short of an extreme i
A I
! NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
I 142 1 seismic event. !
i 1
rg 2 And you know, it has very little load forces ;
N]-
3 on it. There's no corrosion. It's a wonderful design.
4 But boy, trying to put backfill in to make it smooth l 5 enough and rounded and lay something like that in? Very, l
6 very difficult when they sat down and actually showed me l
7 how you do this kind of work. 1 8 And so we haven't given up on that because I'm 9 the boss. But it's had to move to a back burner because 10 it's harder to do. So we do exactly what you say. We 11 look at what kind of variations can we have, what kind of 1
12 variabilities can we have. You know, good question of -- '
13 what if we have to go in and get some packages?
7.s (k ) 14 What if we heat up a drift? The intent is to 15 close the drift off after we fill it up during preclosure.
16 So you have a drift here. It's cooked for five years, 17 It's nice and hot. And we dec].de we have to empty that
' 18 drift out. It has a structural problem or something's 19 wrong.
20 We have the capability and the design to 21 ventilate that drift, cool it down so that we can get 22 people in, take it out. We just put a cycle in a drift.
23 That has to be accounted for in the overall analysis. And 24 it affects us. The altered field rock that I talked about
()
(~~
25 where perhaps we can create that umbrella, well we just l
NEAL R. GROSS COURT ilEPORTERS AND TRANSCRIBERS l 1323 PHODE ISLAND AVE , N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4 33 l
143 1 did something funny to that one because it was driving and eS 2 now we may have it emptied for a year or two years.
! 1
</
3 And then we're going to go drive it again. So 4 we have to look at those kinds of things. You're exactly 5 correct. And that's why I looked at LCO's for the 6 technical specification -- limiting conditions for 7 operations for technical specifications.
8 We have to be able to deal with those off 9 normal bases. We have off sets. We have to be able to 10 look at how we put the packages in. We have to run the 11 math of what the heat plumes are going to look like. We 12 have to be careful doing those things.
13 In actual fact, we try and separate the drifts l
[, T
'l 14 for the most part so that that's just the line that we're 15 interested in and not the area. So a lot of the aspects 16 of the odd pieces that happen in mining probably shouldn't 17 affect us badly.
18 MEMBER HORNBERGER: The other question I have l 19 is, you've mentioned during your talk that you envision i
20 having roughly ten years to do testing. Do you envision 21 doing testing that approaches the system or are we talking !
22 about testing of individual components?
i l
23 MR. BAILEY: My comment was -- that particular !
24 comment was associated with the materials testing that we
,\
n (l,,./ 25 have ongoing at Livermore which is a very large set of l l
NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
144 l 1 specimens which are suspended above water and various
,.s 2 human environments and temperatures. We're trying to find
('~') 3 out -- do an accelerated corrosion process. l 4 My comment was really centered on the fact 5 that before we buy our first waste package, if we're on !
6 schedule, it's about ten years from now. And so before we 7 actually buy that first waste package, we'll have eight to 8 ten years worth of testing that says we're pretty 9 confident in our material. l 10 Now will we ever turn those off? I don't I 11 know. That's a budget question. That's a program i 12 direction. Clearly we want to get as much data as we can.
13 I don't think that the idea is to study this thing for 15 j
,a
' r i
! (_/ 14 years and 20 years, write an LA, and then walk away from 15 .. t and just -- it's going to work and close our eyes.
l I
16 I think you have to keep studying and you have 17 to keep gathering data. In fact, we have a performance 18 confirmation program required by Part 60. So my comment 19 was generated really at the ability to do procurements and 20 having a fairly good idea. And I think it might even --
21 it probably came before, but it was really generated at 22 what do I put in the drift and what have I made decisions 23 on early on.
24 And one of those things we want to make a good (n) 2E decision on is the waste package. They're expensive, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLANI AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1 l
145 i 1 they're big. You have to handle fuel. And we're going to
, , , 2 weld them up. It's an expensive -- and it is clearly the i )
3 center of this feature. So we want to try and make sure l 4 and get that right. l l
5 So if we can -- we'll be changing that I 6 material up until the day that we order it, if need be.
7 And if there's too much flexibility in it, we won't order l l
8 them. If we can't be certain of the answer, we won't 9 order them. But we are doing those kinds of tests. And 10 clearly you learn more in five years than you do in two.
11 And you learn more -- there's a diminishing 12 return. l l
13 VICE CHAIRMAN GARRICK: Bill.
p)
- k. 14 MEMBER HINZE: At the risk of suggesting that 15 you might add something to your super illustration, let me 16 ask you -- l 17 MR. BAILEY: Throw me a pen.
18 MEMBER HINZE: Let me ask you about a couple 19 of things that I didn't hear --
20 MR. BAILEY: Sure.
21 MEMBER HINZE: -- you chat about. And they 22 relate to the layout of the repository.
23 MR. BAILEY: Yes.
24 MEMBER HINZE: The footprint of the repository O.
( ,) 25 and setback distances, how do they enter into your design NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
146 1 considerations?
7s 2 MR. BAILEY: The footprint is actually taken
! )
3 into account in the aerial mass load. We look at where we t
4 yrenreasonablyput it, the quality of the rock, and how we 5 waaC to load the packages in. And that's what really i
6- determines ove need for real estate.
7 M.4BER HINZ~.; Do you have any criteria other 8 than ther al ioading in terms of the footprint?
9 . .AILEY: Oh, yes; it has to be accessible, 10 there's certain geological pieces. I probably can't get l 11 into them as well as some of the other people. But yes, l
12 there are other reasons.
13 MEMBER HINZE: Do you have the information !
r~s i s' ) l
'w /' 14 upon which to do these designs at this time? l l
15 MR. BAILEY: We are proceeding with those 16 designs at this point in time. We are continuing to learn ;
1 17 from the ESF. And in answer to George's question, we are 18 in ESF and we are learning a lot about mining and what 19 kinds of problems we have, and we're examining the data 20 from the ESF to make sure that we have that right.
21 Clearly more data -- when we get more data, 22 we'll take it into account and use it.
23 MEMBER HINZE: Certainly in terms of the 24 isclation of the waste, setback distances could be 1
g ~s
( ,) 25 critical.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C 20005 3701 (202) 234-4433
147 1 MR. BAILEY: Yes, yes.
7-4 2 MEMBER HINZE: What -- I don't see that on
\w] there.
3 Where's that coming into effect?
4 MR. BAILEY: That is, of course, an important i 5 parameter. I didn't put it ic here because I focused on 6 corrosion more so than -- it should be in the mechanical 7 load section. Most likely that's where it ought to be. I l 8 hadn't put it there.
9 We do have some numbers that we've chosen.
10 I'm sorry, I don't remember them. Fifteen or twenty 11 meters, I believe.
12 MEMBER HINZE: I don't care what the numbers 13 are. I'm concerned about how they're being chosen. j f3 l 1
\' 14 MR. BAILEY: I can't really go into very much 15 detail on that. We have looked at that and we have 16 thought about that. But I don't know the specifics to it.
17 MEMBER HINZE: Okay, thanks.
18 MR. BAILEY: I'll get an answer back to you.
19 VICE CHAIRMAN GARRICK: Paul.
20 CHAIRMAN POMEROY: Let me just try one quick l 21 on and perhaps it's very close to Marty Steindler's 22 question. It seems to be at first blush here when you 23 look at even the various coupling phenomena that may take 24 place, that every one of these will win you something in a t
,) 25 -- in terms of you'll get a positive effect --
- NEAL R. GROSS l CoVRT REPORTERS AND TRANSCRIBERS
( 1323 RHODE ISI.AND AVE., N W.
l (202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
148 1 MR. BAILEY: Yes.
,c3 7 CHAIRMAN POMEROY: -- in the performance
- d
\s' 3 assessment evaluation that you're going to do.
4 And therefore, it would seem like you must 5 have some criteria for saying where is the cut off. If we 6 -- should we use everything that gives us any additional 7 assessment? You comment, I think, was that you were a 8 minimalist or something to that effect in terms of you 9 want to use the minimum number of things that achieve the 10 desired goal. 1 11 Is it a performance goal we're working towards i
12 then or is it a performance cost? It's a multiplicity of I 13 goals. ;
l I V 14 MR. BAILEY: Yes, it is. 1 l
l 15 CHAIRMAN POMEROY: I'm concerned about what 16 the criteria are for what the cut off is from a scientific 17 basis -- or an engineering basis. I'm sorry.
18 MR. BAILEY: That's -- it's a fair question 19 and I'd like to have a better answer for you. I think of 20 it in terms of the matrix that I haven't finished. I like 21 to look at what it is I'm trying to accomplish. What are 22 my goals? And am I accomplishing my goals? And it needs 23 to be broken up more.
24 It needs to be broken up into redundancy or
,m.
b) 25 margin or good thought as opposed to hard numbers. What NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
O 149 1 is my certainty .in those numbers? Where are -- how do I 2 do this for my base case? And then looking at the (7
v -)/
3 uncertainties. What are the uncertainties I have to deal J l
4 with? l 5 It isn't clear I can deal with every 6 uncertainty. It just isn't clear that you can handle l 7 everything completely. So you have to make some decisions 8 in that regard. And those are hard decisions I think to 9 make at this point in time. l 10 CHAIRMAN POMEROY: Right.
11 MR. BAILEY: Because we have preliminary data l l
12 and it's hard to set that criteria. So I look for my s 13 columns filled in. I need a lot more columns. I look at p)
\
L- 14 my inputs and I try and do it in a rigorous manner. I've l 1
15 got to meet a goal. I've got to meet a radiological goal.
I 16 But then how do I deal with the rest of it?
17 Can I really show that I have -- defense-in-18 depth was the title of the talk. Can I really show that I 19 have lots of ways to deal with my primary problems? And 20 then do I have lots of ways to deal with the environments 21 that it's going to cae and those uncertainties? And we 22 have to go through and make that chart.
23 And then we have to look at the total 24 performance and make sure there's no exacerbation -- what r's
(_) 25 the trade offs are. I can't give you a criteria per se.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
150 j 1 Clearly you can throw everything on the -- you can throw
[ ,._ 2 everything on. It isn't clear to me that everything has (s)
' ~ '
l 3 to go on, that you won't get that much better a piece of 1
4 performance.
5 There may be some exacerbation. I mean, the 6 clad -- I think you have to protect the clad. That's my 7 personal opinion. I think we have to protect the clad 8 very much so. I think that's really important. Because 9 we g. : it in pretty good shape. And backfill's 10 problematic for clad. Ceramics are problematic for clad.
11 But they do a lot too.
12 CHAIRMAN POMEROY: It seems to me your stance 13 is pretty good. I guess I'm concerned about, you know, in
- 14 when we may all have gone onto other things by the time 15 some of this gets in its final form. If somebody else 16 comes along that, say, isn't a minimalist, they might say 17 let's put everything in and push it down how ever many 18 orders of magnitude everything will give you.
19 Somewhere along the line maybe we ought to 20 think about whether or not you can establish some --
21 MR. BAILEY: Yes.
22 CHAIRMAN POMEROY: -- radiological criteria 23 or, as Steve was saying, a couple of orders of magnitude 24 below what you're striving for or something like that.
l
! ()
(~j 25 Otherwise, it does seem as though Marty's comment does l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005 3701 (202) 234 4433
151 1 into play that it does -- it can cost you a lot of money.
fy 2 MR. BAILEY: Yes, I believe you're correct.
N] '
3 We have to have a criteria. We can't just say well, this 4 looks like the right thing to do. I think Steve is the 5 policy maker here.
6 CHAIRMAN POMEROY: Okay.
7 MR. BAILEY: I'll tell you the engineering 8 piece, Steve.
9 MR. BROCOUM: To some degree, we're working in 10 a vacuum because we don't have a standard. And you know, i
11 whether a standard is implemented at say 30 kilometers or 12 five kilometers, whether it's 15 millirem or 500 millirem 1
1 13 makes a -- you know, we don't have a standard. l
[~'N
)
i N -)
14 So I think to some degree we're in a vacuum. ;
15 And we have to -- internally we've come up with some cases 16 we're working to to help bound the problem and help us 17 understand as EPA's going on to rule making. But at this 18 time, absent a standard and absent an updated 60, we're 19 kind of a little bit in a vacuum.
20 So I think we're looking at a lot of things, 21 so we can go in any direction the standards and the 22 regulations move.
23 CHAIRMAN POMEROY: That's good. Thank you, 24 Steve.
/~%
( k I'd also like to say thank you, Jack. You
(_,/ 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C 20005 3701 (202) 234-4433 l
152 I i
1 always make this subject an exciting one. I personally 1
2 appreciate it.
.r-]
\_/
l 1
I 3 MR. BAILEY: Thank you, sir, I l
4 VICE CHAIRMAN GARRICK: All right, I think 5 it's time to take a break. We'll take a ten minute break. l 6 (Whereupon, the foregoing matter went off the i 1
7 record at 10:26 a.m. and went back on the l 8 record at 10:38 a.m.)
l 9 CHAIRMAN POMEROY: Let's reconvene, please.
10 VICE CHAIRMAN GARRICK: Okay, just to show my l
11 colleagues that I'm not obsessed with performance 1 12 assessment, we put it last and we're putting the major 13 squeeze on it.
( )
k/ 14 So Abe, I know you're flexible and we are 15 pushing to terminate this discussion as close to 12:00 as 16 we possibly can.
17 MR. VAN LUIK: I suppose my microphone is on, 18 but terminating me by 12:00 sounds like a release.
19 (Laughter.)
20 VICE CHAIRMAN GARRICK: Why are you so happy?
21 MR. VAN LUIK: One of the things -- is this 22 microphone working?
23 Okay, one of the things about being part of an 24 integrated program is it gives you the opportunity to g
() 25 deflect questions to the people who are feeding you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE. N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1 l
153 1 information. For example, design and site questions we l 1
,m 2 defer to others because we're not working off the same
)
l
~'
3 song sheet.
l 1
4 On the other hand, it makes my life a little )
5 more difficult because it used to be that if we didn't l 6 like what they gave us, we would just assume something 7 different. i 1
8 (Laughter.)
9 And now, if I misspeak, I'm sure that Jack and 10 any representatives from the site program will jump up and 11 say no. So life is easier in some ways and more difficult 12 in others.
13 I was asked to talk about three things for
)
L/ 14 this talk: to give some overview of what we've learned 15 from our past work; to give us a status on what we're 16 doing for TSPA-VA, what can we expect there; and then to 17 talk about how we in PA are working into this whole milieu 18 of defense-in-depth and design.
19 The print on my viewgraphs is slightly larger 20 than it is on your handouts, which is good, but it's 21 probably still not readable. What I've done is divided 22 this history discussiori up to the environmental assessment 23 SCP which is the '86-88 time frame, a before and after 24 type snapshot.
(j -
25 In my opinion -- and I know that there's other l
NEAL R. GROSS COUFC REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASH;NGTON. D C. 20005-3701 (202) 234-4433
154 1 work that I could have reported. But in my opinion, one l
,_s 2 of the keystone total system performance casessments was i \
v 3 done by the National Academy of Sciences and the Waste 4 Isolation Systems Panel "eport, otherwise known as the 5 Pigford Report.
6 They did simplified modeling and gave us the 7 initial warnings concerning the possibility of relatively 8 high doses when you're discharging radionuclides into 9 groundwater in a closed basin. This was a slap at the 10 EPA's idea that a cumulative release performance measure 11 is appropriate for all cases and all times.
12 And he also showed -- while he was slapping 13 people around, he also si.ovred that it was possible to meet fh
's-14 the 10 CFR 60 subsystems requirements and still fail the i 15 EPA cumulative release standard, showing that there was 16 not a good nexus between the subsystem and the total 17 system requirements.
18 This was a very interesting report. It was, 19 of course, funded by DOE. And it managed to slap DOE and 20 it managed to slap EPA and slap NRC. And this is the kind 21 of contractors that we love. They are independent.
22 (Laughter.)
23 The Nuclear Regulatory Commission sponsored 24 some system assessments by Sandia to evaluate the draft n
25 EPA standard. They did a number of different sites. They (w_-)
! NEAL R. GROSS
! COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISLAND AVE., N W.
l (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
155 1 did a simplified model of a repository at Yucca Mountain 2 in the saturated zone.
73 (v) 3 By the way, he also calculated for a saturated 4 zone. But then, at the last minute, when he found out we 1
5 had bait and switched him to the unsaturated zone, he l 6 added a few thousand years groundwater travel time and 7 said just move the left side of the picture up three l l
8 points and it's still valid.
9 But they did a similar thing for the saturated !
10 zone. And here again, we must give them credit. They i
11 pointed out that even in the saturated zone, when you i
12 start getting towards the surface of the saturated zone, 13 you have potentially oxidizing conditions and underscore j
/~ ;
/
(-)
1 14 the importance of understanding the flcw fields in the 15 saturated zone and the transport parameters particularly 16 under oxidizing condition, are you correct in assuming the 17 solubilities that you are assuming. l l
18 The Environmental Protection Agency did a I 19 total system performance assessment in 1985. Again, 20 simplified modeling. But for the first time actually 21 including low probability disruptive events. Suggests
]
22 that a Yucca Mountain repository capable of meeting the j 23 draft EPA standards with a warning, however, that if 24 unsaturated zone fluxes were substantially higher than r"s 25 assumed, consequences would be higher.
()
NEAL R. GROSS 1
COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
I I
156 ,
)
1 And they updated these for the 1993 !
l 7~3 2 background information document supporting the WIPP
(' )
3 standard doing only actinides basically for transuranic l l
4 waste. We did some calculations in 1984 and 1985 PL in I 5 Sandia. simplified modeling again making exemptions l i
6 thought to be conservative at that time. !
i 7 I think that's an important thing to point i 8 out. And suggesting again Yucca Mountain repository 9 capable of meeting the standards. What do all of these 10 things have in co:nmon? And I tried to underscore it each 11 time. They did simplified modeling, each of which said if 1 12 certain aspects are not as we thought they were, all bets l 13 are off and our conclusions don't hold.
(N/ )
14 I think that's important to point out in j 15 fairness to those simplified -- they were very well aware 16 that they were simplified. What's changed since the SCP?
17 Well, one thing Jack just pointed out is we have gone a 18 long way from small containers vertically in place and 19 boreholes made of relatively thin outer barrier of 20 stainless steel.
21 At that time, the maximum capacity was three 22 PWR assemblies, four BWR assemblies, or orie high level 23 waste canister. Now we're looking at large multi-barrier 24 container of nickel "ased alloy within carbon steel (s
( ,) 25 horizontally placet n drifts --22 BWR's, 44 BWR's, or NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
ll 157 1 five high level waste glass containers with one DOE spent s 2 fuel canister co-located within the same package.
(
i ,/
)
3 So a very different concept for how we view 4 the engineered barrier.
5 How about the natural system? What's changed G there? At the time of the SCP and the EA, we only had 7 very short term of observations. We thought the net 8 infiltration was estimated to be less than a few 9 millimeters per year. I think up to four was the maximum 10 in Monizer and Wilson.
11 And many people thought that it might even be 12 negative. We had no real estimates available of net i
13 infiltration under weather clinates. We thought that the
/ \
k- 14 ef fects were to be minimal bac:4 on the presence of the 15 Sierras, very large scale argument. We based our 16 solubilities at that time largely on literature values 17 and did retardatitu estimates in the modeling for 18 preliminary laboratory work and literature.
19 Where have we come since then? We're into the 20 second generation of flow and transport models. We're 21 incorporating and calibrating to data from the repository 22 level that have su msted a net flux under current 23 conditions between two to 15. Steve said one to ten.
I 24 Later you'll see ourselves using one to ten. j
/~N s.,,) 25 One to ten is what is currently in the model NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D.C. 20006-3701 '202) 234-4433
158 1 coming from LBL. Two to 15 is a little expansion on that.
7 s, 2 And basically I put the little symbols there to say these
(] 3 are uncertain numbers. Steve made fun of our 6.2. You 4 know, that's -- when you got a calculator, that's what 5 happens.
6 (Laughter.)
7 We have an improved basis for estimating 8 solubilities. We have a site specific data base for 9 representative radionuclide sorption parameters. And the 10 flux range will increase to perhaps triple the current 11 values under episodic weather climate conditions.
12 Those are kind of the new bounds that we're 13 working in.
("%
'- 14 VICE CHAIRMAN GARRICK: Abe, just a simple, i
15 quick question here. Why wouldn't you just put in a 16 distribution?
17 MR. VAN LUIK: Oh, we do. We do put in a i
! 18 distribution. The reason the 6.2 came up was for the l
19 sensitivity studies that we're doing. To vary everything i
20 at the same time kind of muddles the issue.
l 21 VICE CHAIRMAN GARRICK: So is the 6,2 indeed a i
22 central tendency parameter from a distribution?
23 MR. VAN LUIK: Yes, it is.
24 VICE CHAIRMAN GARRICK: Okay.
g
() 25 MR. VAN LUIK: Yes, it is; yes. And we will NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005 3701 (202) 234 4433
159 I get into that in my third part of my talk. But basically,
.s
, 2 again, major changes in the way that we see the natural
( )
v -
3 system behaving.
l I
4 How about performance assessment itself? If 5 we're looking at the EA SCP time, we use simple models.
6 The unsaturated zone flux was the most important parameter l 7 determining releases for 10,000 years. And conservative ]
8 estimates at that time again, those -- we thought they j 9 were conservative then -- indicated releases would be I 10 minimal.
11 The engineered system failed and disappeared j l
12 after a specific time usually not exceeding a thousand i 13 years, sometimes 200, 300 years. The warning was sounded 1
,/
l U 14 that at higher flux it would mean higher releases again 15 based on that kind of modeling.
1 16 And doses were not really addressed in the 17 project until the PNL study that was done for headquarters 18 in 1988 which predicted very low doses again because of 19 low flux and optimistic neptunium solubility assumptions.
20 Optimistic is -- actually it was the wrong neptunium 21 solubility.
22 They were using the value from the literature 23 we had used before for reducing conditions and had not 24 upped it five orders of magnitude for up to oxidizing ry
( ) 25 conditions.
%d NEAL R. GROSS COURT REPORTERS ANJ TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
! 160 I I
I 1 Where are we now? Well, we have process-level
,, 2 models. That's very different from the simplified
/ \
l 3 assumption models for the engineered system and the 1
1 4 natural system. These are nearing completion for TSPA-VA.
5 And the next part of my talk is to tell you how they're 6 nearing completion; what are we doing to make them 7 different. !
8 We are incorporating that process-level 9 modeling into TSPA through abstraction, testing, and 1
10 creating thereby a more thorough and defensible TSPA 11 capability. All pathway doses are to be considered at the 12 time of peak dose and for various hypothetical locations
)
1 13 and potential dose recipients. I
\J 14 This is the vacuum that Dr. Brocoum was 15 talking about. If we had a definite regulatory 16 environment to work in, we would only do one location for i 17 whatever that location was. I just put this bottom 18 paragraph in there because I want you to know that, from 19 my perspective, the preliminary sensitive studies that 20 we're doing do not call into question the general 21 expectation that TSPA-VA will likely suggest the Yucca 22 Mountain repository is a viable option.
l 23 Boy, that's a lot of words.
24 (Laughter.)
A
! ) 25 But I just wanted you to know. I'm still l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l 1 (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433 J
161 1 positive, okay?
l q 2 (Laughter.)
i ;
w ./
3 The worst punishment I've ever had to endure 4 is to have to read my own viewgraphs.
5 How about TSPA philosophy? Well, this is a 6 very important point. When we were in the SCP stage, we 7 were thinking very simplistic models. And we assigned 8 subsystem and even component performance goalu, and we 9 stated right in the SCP that we were doing this until a l
10 complete TSPA capability was available.
11 We feel that this was very similar to the 1 l
12 NRC's effort to include subsystcm performance requirements 1
1 13 in Part 60 and that it suffered the same drawbacks. How I\ s') 14 did we do this? Well, we used informal expert judgements.
15 We got people together basi: ally in a rubber room and said j 16 okay, we want this thing to work this way.
17 We know that the influence is here. What do 18 you think are good partial estimates of partial 19 performance measures? And there is table after table 20 after table of EPPM's in the SCP. We informally thus, 21 through expert judgement, drove system requirements -- and 22 we did have a standard in place at that time -- down to l 73 subsystem and even component levels 24 One point that Steve made this morning that I s
., ) 25 like to make over and over again is that these tables NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
162 1 assumed and were locked into a simplified, and it turns
,g- 2 out, unsubstantiated conceptual model. Now since that
( )
w/
3 time, we've done site characterization and design process-4 level modeling and TSPA analyses that have been matured by 5 that process-level modelings.
v And now we're using, as Jack pointed out just 7 a while ago, the TSPA as a tool to directly support the 8 evaluation of data uncertainty, design assumptions, and 9 alternative conceptual model significance. And this is a 10 sizeable issue in and of itself.
11 What we have found from moving from this phase 12 to this phase is that the EPPM and subsystem component 13 performance goals are valid for a specific conceptual
', / ')
i
\/ 14 model only. And that since that particular conceptual 15 model that they were based on, it is no longer our 16 conceptual model. They are obsolete.
17 That approach just doesn't work. We find that 18 lower-level surrogates for the primary public health and 19 safety concern are either derived directly from that 20 primary concern -- in other words, they either have a very 21 strong nexus through being based on the same exact 22 conceptual model and derived from it, and therefore they 23 really add no assurance.
24 All they do is take that point and say that
(*h (m) 25 means this point and this point if the above is true. Or ,
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
163 1 they will be more conservative. If you say well, we 7, 2 recognize that this subsystem requirement is based on the i
1.."/
3 same conceptual model, however there is uncertainty in 4 that conceptual model; therefore, we'll knock it down an l
5 order of magnitude, it will be more conservative.
6 But does it really add to public health or 7 safety, or has it just become more conservative? And 8 actually, it could become the limiting case for your 9 licensing argument since it will be more conservative than 10 the regulatory mandated public health and safety concern.
11 So to us, it either adds no value or it adds 12 an artificial conservatism into the whole process which is 13 why we are now looking, rather than assigning temporary t >
C/ 14 interim performance goals for each barrier in the 15 engineered system, we are looking -- and I will give you 16 an example in the third part of the talk.
17 We are looking at what does this do to total 18 system performance holding other things constant.
19 Having gone through this history, let's move 20 now into the status of TSPA and our plans. Most of you 21 have a copy of the TSPA-VA plan of last September. And we 22 want you to know that, by and large, it is being 23 implemented. So when you read in there that we are doing 24 certain things this year, we are doing them.
Q/ 25 Nine out of the ten workshops on abstraction NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4 4 3 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
164 1 of process models for input to TSPA have been completed.
,.- 2 And I believe the NRC staff or center representatives of 3 the NRC have observed each of these workshops. We have 4 done sensitivity analyses and we are working on developing 5 the TSPA-VA model.
6 We are conducting a series of expert 7 elicitations. We have a peer review initiated. And in 8 July in San Antonio, Texas, where it should be nice and 9 cool, we have a technical exchange scheduled and a meeting 10 with the ACNW.
11 What's our schedule? You love these diagrams 12 I know. They have so much technical content. We are in 13 the planning and conducting stage. We will then construct g
U 14 and document the abstractions, freeze the process models 15 somewhere in this time frame, freeze the date in pretty 16 much that same time frame, complete the reference case, i
l 17 and then complete the draft TSPA-VA documents by the 12th ;
1 18 of June of 1998.
19 So as you can probably figure out, this is our i
20 major activity for this year. But as you'll find out 21 later, there are also other activities. What's the logic?
22 Well, we have process models, as Jack pointed out a while 23 ago, that feed into TSPA. Those process models have 24 uncertainties.
,a
( ) 25 For some of the key process models, we have an v
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 2344433 l
165 1 expert elicitation either ongoing. UZ flow is pretty well r~s 2 finished. We're looking at adding SZ flow and transport, i
\
3 thermal hydrology, waste form dissolution and waste 4 package degradation. This one is going on even as we 5 speak. Yesterday and today there's a meeting of this 6 group.
7 And these ale leading to the input to TSPA-VA l
8 on the process model level. We also have a peer review in )
9 progress. They have just finished the TSPA orientation 10 phase where they basically came up to speed on what TSPA's 11 are all about. They are starting, beginning this -- later )
1 12 in June, the process model phase where they're going to i 13 look at these models and pass judgement on them.
k_ 14 They are looking at the scenarios that are 1
15 going to be included in TSPA-VA. They're looking very l 16 much at the abstraction process which is all these 17 workshops that we've been holding. And the workshops 18 kicked off six month of calculational effort for each 19 workshop, and they're looking at the total feedback 20 picture.
21 And then at the end, they're going to look at
'2 the TSPA-VA and give us comments on the preparation, i
23 analysis, and documentation differences that they would 24 suggest when we're -- you know, after we're done with the (3) 25 TSPA-VA for the TSPA-LA.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
166 l
1 So this peer review panel, their last gasp of
~s 2 work will be to learn from the TSPA-VA what we should do
\
~
3 different for the TSPA-LA. We do the analyses, we 4 document them, and we issue the VA next year.
5 CHAIRMAN POMEROY: Are those external peer 6 review panels?
7 MR. VAN LUIK: This is an external peer review 8 panel. And I'll give you the names in just a second.
9 CHAIRMAN POMEROY: Excellent. Thank you.
10 MR. VAN LUIK: The schedule of the process 11 model abstraction testing workshop and summary -- these 12 are all done. This one is the only one left. And 13 unfortunately, I didn't catch that until this morning.
(D
\/ 14 But just last week we changed the date on this one.
15 It's going to be June 2nd and 3rd rather than 16 3rd, 4th, and 5th, with spill over to the 4th if it needs 17 to, but we don't think so. And the reason we changed it 18 is because it's all my fault. I have to be at an offsite 19 the 4th and 5th, so they moved it over to accommodate me.
20 Because I said for this one I definitely want to be there 21 since I have opinions on this subject.
22 No knowledge, just opinions.
23 (Laughter.)
24 So what good are these workshops that we've g
( ,) 25 been doing? Well, the first thing is identify -- and the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
167 1 way the workshops are structured, we have design people,
,e' 2 we have site people, and we have PA people. So together QJ' 3 we identify and agree on the key performance-related 4 issues associated with each process model.
5 We key these issues to the waste isolation and 6 containment strategy hypotheses. We key them to the NRC's 7 KTI's. We prioritize these issues using -- and you will l 8 catch me in a total contradiction here -- using surrogate 9 performance-related criteria as a temporary expedient.
10 Now why are we doing that? Our point is that 11 we recognize that these are temporary expedients. These 12 are not -- you know, these are things -- okay, when we're 13 working within this model and looking at this process,
/,_N
( l 52 14 let's say if it moves, what we think the natural condition 15 is by an order of magnitude.
16 That will define whether it's important or 17 not. It's that simple. We develop approaches to address 18 the key issues within the context of the VA in terms of 19 sensitivity analyses and abstraction plans. And these are 20 the activities that go on for at least six months, if not 21 nine months on some cases, after the workshops.
22 So the workshop again is to agree on what the 23 problems are and get us to agree on the approach to t
24 addressing these problems. And then we go to work and
(_y) 25 address them. And in some cases, this is unplanned work NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 123 RHODE ISMND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
168 1 and has to be run through a change control process to make l
,s 2 sure that it gets done.
~
3 And no work gets done unless resources are 4 devoted to it.
5 The expert elicitations feed to TSPA in this j i
6 way. For UZ flow, they have confirmed the magnitude of l 1
7 infiltration flux. And they have given us some bounds in j 8 the uncertainty and infiltration and percolation fluxes. )
1 9 The report for this should be to us within -- we haven't i 10 seen it yet officially.
11 Should be to us within the next few days. The 12 waste package degradation one is ongoing even as we speak.
13 They are looking at a range of likely galvanic protection l 7-L- 14 percentage. In other words, they're going to give us --
15 we want to take galvanic protection credits. Therefore, 16 the TSPA-VA giving us a range over which they think that's 17 legitimate.
18 And by the time for the LA, the testing will 19 be done to give us an experimental basis on top of that.
20 And then they're giving us ranges of degradation models 21 for corrosion resistant materials and ceramic coatings.
22 For saturated zone flow and transport, we want to look at 23 range of reasonable dispersivities and range of mixing 24 percentages.
,Q)
( 25 There are other aspects to all of these NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2h4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
169 1 things. We think the other aspecte -- you know, they will
,- 2 become conversant with the other aspects. But these are i i
~#
3 the main things that we want out of these bodies as far as 4 giving us some help for the TSPA-VA. That doesn't mean 5 it's the final answer for all time.
6 The peer review panel we have -- Chris Whipple 7 Is the chairperson. Bob Budnitz, Paul Witherspoon, Rod 8 Ewing, Joe Payer, and Dade Moeller. The kick off meeting 9 was held in February, and we expect in June of 1997 to 10 receive their interim review comments based on their first 11 six months of activity. ,
12 And like I said, the first six months of 13 activity was basically an orientation. Whether they n
L- 14 learned from reading the old TSPA's, the review comments 15 that we expect back from them is probably the same thing 16 that you've been telling us for a long time, you need to 17 write these things so that other people can understand 18 them.
19 And for the second six months, they will be 20 looking intensely at the process-level modeling that's 21 being done in support of TSPA.
22 Moving right along to the third part of my 23 talk, where our connection is -- our nexus is with 24 defense-in-depth. It's an important thing to say that, (p) 25 even though we have just focused all about TSPA-VA, there v
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
170 1 are other ongoing performance assessment activities.
,_ 2 It's a large organization and it works hard.
( )
~'
3 We do system level and supporting models updates 4 continually. We update them and test them as new site 5 design information becomes available. Even as we are 6 looking at this process of formally using abstraction to 7 create the TSPA total system model, the minute that Bo 8 Bodvarssen, for example, comes out with a new rendition of 9 his infiltration and flux model for the unsaturated zone, 10 we update our model just as a matter of course.
11 We use these models which, you know, will be 12 developed into the TSPA-VA through the process I just 13 described. But we use these models to support the ongoing fw I )
im/ 14 process of evaluating the adequacy of construction and 15 testing controls. This is the DIE process that you may or 16 may not be familiar with.
17 But anytime there is the design or an activity 18 in the ESF that introduces materials or changes the shape 19 or anything else, we support that by doing an evaluation 20 of whether or not this has implications for total system 21 '
performance in the post-closure period.
22 And then we also use the models for scoping l
l 23 sensitivity analyses in support of design and testing. So 24 this is a continuing level of effort within the PA program gy 25 (v) that is not really focused on the TSPA-VA activity, but l
NEAL R. GROSS COURT REPORTERS AND TRANSCRlBERS j 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
171 1 it's continually updated by the results of that activity.
2 So what have they done for the program lately 1
\/ 3 just in the last few weens in fact? You know from the 4 presentations that we made last year and this year that we 5 continually update the TSPA 1995 case. What we have done 6 very recently is updated it with current site model
- / information to allow comparative analyses.
8 We have taken Flint's very latest, this years 9 infiltration map. We have taken Bodvarasen's 1997 10 percolation flux estimate which gives an average of six 11 millimeters per year. Steve made fun of our 6.2, but you 12 see that we astutely erased the 6.2 because we knew he was 13 c;ing to do that.
i }
's t
14 But it's actually 6.2 and it's off of a range 15 from four to ten. We assumed 30% of the flux was in 1
16 fractures. This is what Bo's model is now giving us. We l l
17 looked at fracture seeps inCo ten percent of the drifts.
18 That's an assumption that we are making because the drift 19 scale modeling is not quite as mature as the other i
20 modeling that we're looking at, so we're doing that l 21 ourselves. I 22 But we have niche tests and other things in 23 place in our planning for this coming fiscal year to 24 actually experimentally look at this. We have revised the l
(,) 25 thermal hydrology based on Livermore work. We have --
NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N W (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
172 1 based on a conference word between Livermore and Los 7 s f
2 Alamos, we have reduced our solubility of neptunium by a
\_/
3 factor of 100.
4 And we have assumed a saturated zone flux of
- 5 .3 meters per year. I think I showed you last time that 6 this was a very important variable. This came up in the 7 discussion with Dr. Brocoum, and that's one of the reasons 8 that we're also doing an expert elicitation on saturated ;
1 9 zone flow and tran. port is because it turns out that that l 10 is an important parameter.
11 CHAIRMAN POMEROY: Abe, I don't want to spend j 12 a lot of time on this, but just why did you reduce the l 13 neptunium solubility by a factor of 100?
/,_N
\'
]
14 MR. VAN LUIK: As you probably know, we have a 15 devoted critic on the TRB that has been telling us for 16 years that we're using the wrong solubility, that it's 17 three orders of magnitude too high.
18 CHAIRMAN POMEROY: Yes.
19 MR. VAN LUIK: Probably because of prompting 20 by that individual on the TRB, Livermore and Los Alamos 21 put their heads together and reinterpreted the 22 experimental basis on which the original estimates were 23 made and said, you know, there is a point there and they 24 decided that it was probably two orders of magnitude too r~x
() 25 high.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, O C. 20005-3701 (202) 234-4433 l?
173 1 They have actually documented that. And I 73 2 think they're making a very good case for that. So given t
t)
3 that the site program has given us that, it is our 4 obligation then to show that in the TSPA. And the TSPA-VA 5 will include a discussion of what that value is and why it 6 was picked the way it was. l 1
7 But it's consistent with the experimental work l 8 that was done assuming a different conceptual way of l 9 looking at it with less conservatism in it. So the i i
i 10 engineering alternatives, as I already said, we're looking 11 at carbon steel alloy 625 waste packages. We're looking 12 at point loading in this particular reference, sensitivity l 13 study. :
,e
_ 14 We're looking at galvanic protection.
15 Seventy-five percent of the carbon steel must be corroded 16 before you start pitting Alloy 625. And as a part of the 17 sensitivity study, we assumed that ten percent of the 18 packages had ineffective galvanic protection.
19 I guess that's -- you know, just trying to be 20 conservative. No cladding credit, no backfill, no ceramic 21 coat, and no drip shield. So that is our updated TSPA 22 1995 case. And then we're going to start adding things to 23 that to do the sensitivity study.
24 VICE CHAIRMAN GARRICK: Why do you do that?
(v) 25 MR. VAN LUIK: Why do we do that?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
174 ;
1 VICE CHAIRMAN GARRICK: Yeah, why do you c 2 profess to be doing risk-based performance assessment on i i l
3 the one hand, and then on the other hand *otally 4 obliterate it with unreasonable assumptions? l i
5 (Laughter.) ;
6 CHAIRMAN POMEROY: Subtle question. ,
I 7 (Laughter.) j 8 VICE CHAIRMAN GARRICK: We can go on if you 9 want.
10 (Laughter.) '
11 MR. VAN LUIK: Okay, here's the reason we do 12 that and the reason I'm showing this example. When we do i
13 a performance assessment that looks at compliance i g -
k._) 14 arguments, then we want to be totally probablistic and put 15 all the uncertainties that we know and have into the 16 calculation.
17 When we're looking at if I add this to my 18 design, is it important or not, then in order to simplify 19 these calculations and be able to do two or three of them 20 within a couple of hours, we basically take either a mean 21 valne or another, you know, estimated approach so that we 22 don't vary everything at the same time and muddy the 23 waters, j 24 When y au see these considerations in the TSPA- l
,o I
( ) 25 VA, they will be totally clouded with the uncertainty that
%d NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433 l
175 1 deserve to be clouded with. And you will see a CCDF that 2 includes all of that uncertainty as, you know, you would
(,, i
(/
3 want to see it.
4 But these are strictly scoping sensitivity 5 analyses to just look at the engineering alternatives that 6 are being presented to us in a quick way. We will do a 7 more definite job once they start to hone in on one or two ;
l 8 alternatives. ;
I l
9 VICE CHAIRMAN GARRICK: Yeah, and the only 10 thing I would recommend is that you consult your i l
11 contemporaries'on the WIPP project which have learned the )
i 12 lesson that once you start presenting in the regulatory 1 13 arena, conservative models under the pretense of a risk-p
'U) 14 based approach, that they're very -- it's very difficult 15 u disengage from the positions and conclusions that you l l
16 arrive with those very conservative models.
17 And you know, it just seems to me that this is l 18 something that you really have to be very careful about.
19 MR. VAN LUIK: I think you're exactly correct.
l 20 And when we did the dry run for Dr. Brocoum, in fact he 21 hammered on this and said make sure that you explain that i
22 these are just scoping sensitivity analyses and that they 23 have -- they make no statement as to whether or not we 24 meet standards or anything else.
<3 i ) 25 We're sensitive to the issue. At the same G'
H EAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
176 1 time, we're caught between a rock and a hard place.
gg 2 Because when the engineers come to us with, you know, U
3 these options, they say how quickly can you run them and 4 tell them, you know, give us guidance for the next set of 5 thought that's being done.
6 These are not final decision making things.
7 These are inputs to the engineering decision making 8 process.
9 So I think you're correct. And --
10 MR. STEINDLER: Can I ask one other question 11 on the same line?
12 MR. VAN LUIK: Of course.
13 MR. STEINDLER: Do you have any idea as to
- ' ') 14 whether or not the omission of cladding credit backfill, 15 ceramic coating, drip shield, wouldn't that in fact )
1 16 overshadow totally any benefits that you might get out of 17 the sensitivity studies?
l 18 In other words, the issue of sensitivity of a l
1 19 particular variable change may be totally lost if in fact 20 you had included more realistic approaches. You know what i
21 sort of -- what's the quality of the answer that you get l 22 by in fact being what you call conservative. l l
l 23 MR. VAN LUIK: Yeah, I think this will become 24 a little bit clearer when I show my next viewgraph im q,) 25 because, in fact, we wanted to construct that base case NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
WASHINGTON, D C. 20005-3701 (202) 234-4433 (202) 234-4433 )
l
)
177
! 1 and then add each of those components in one at a time to j ,s 2 see what the effects would be.
I %-) 3 And so I think it's not like we're presenting 4 the base case as the case that we want to present. We are 5 saying we have to start here and then we add this and see 6 what it does to the base case, add this -- add them in 7 unison. And in fact, they ran 17 different cases, but I'm 8 only showing four here.
9 What kind of results did we get? Well, let me 10 run a couple of caveats first. These are scoping l
11 sensitivity analyses. They are not pictures of what we 12 think the mountain will perform as with these barriers.
13 We're looking at 100,000 year or up to 100,000 year dose I
( \
'- 14 history drinking water based on ICRP-30, not based on 15 ICRP-2.
16 And so you can multiply these by about ten to 17 get the total all pathway dose. And we're looking at a 18 five kilometer boundary. Again, conservatism is implied 19 by that. Because we, you know, believe that the actual 20 people that are going to receive this dose in the future, 21 if there ever is a dose, will be considerably farther than l l
22 five kilometers from the site.
23 So case one is the one that you just 24 criticized as being unrealistic because it did not include
/~'
( )s 25 cladding, cera nic, or any of these other things. And NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
i 178 1 this, as you can see, here is the 10,000 year line. And
)
l 7s 2 this is the issue that we brought up last time when 1 !
( )
<J s 3 showed the sensitivity of these results for a 10,000 year .
4 standard to sensitivity to the saturated zone flux.
5 We used .3 for these calculations. If it had 1
1 6 been one, this line would be moved to the left and we 7 would have a higher dose before 10,000 years. But really, 8 when your dose peak comes up this fact across the 10,000 l
9 year line, then it would not behoove us and it would 10 certainly not help the design process for us to cut off l 11 the calculation at 10,000 years.
12 We would learn absolutely nothing. So let's l
13 see what happens when we add in the 90% of the fuel gets q
t k '/ 14 cladding protection. And this is using a cladding model 15 that we feel is realistic, but not particularly 16 conservative. We used to say that 99% of the fuel -- you 17 know, that one percent had premature pin failure.
18 But now we recognize that basically, because 19 the cladding model is not as conservative as we'd like it
[
l 20 to be but we, you know, have yet to do the work to bring l
l l
21 it to a more defensible state, we just assumed that ten l
22 percent of the cladding would fail at the start.
23 And you see that this reduces the doses an 24 o-der of magnitude. If you multiply these doses by ten to
(~N q) 25 get the whole -- you know, all pathway doses, it doesn't NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N W.
(202) 234-4433 WASHINGTON, D 1 20005-3701 (202) 234-4433
179 1 matter because you're still looking at an order of
,7- 2 magnitude.
! 4
'~
3 If we add in the ceramic coating and we assume 4 that it's uniformly distributive failure from five to '
5 15,000 years, this is based on expert judgement. We had 6 one knowledgeable expert who didn't expect failure until l
7 10,000 and said that it would probably all fail by 20,000, i
i 8 We back that up with his agreement to five to 1
9 15,000. Then we have this kind of effect. And we also l 10 ran the case of this without the cladding and, you know, 11 all back and forth. But this is just to illustrate the 1
12 process we're going through. 4 i
13 And you say yes, there's another order of
/~m I l
\_/} 14 magnitude at least from that barrier. What happens if we
! 15 add in the drip shield? And the drip shield we're talking 16 about here is the one that just completely overlays the 17 package in contact with the ceramic.
l l 18 And what happens if that drip shield lasts for 19 the total time, 100,000 years? Well, you have another 20 almost two order of magnitude reduction in the dose.
l 21 However, it's unrealistic to have it last that long. So 22 what we are doing now is modifying the model so we can 23 have a time varying distribution for that failure.
24 And when we in fact allow that drip shield to (n) 25 fail across the board at 20,000 years, then this goes up NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
180 1 this way and meets this curve and comes out this way. So 7s 2 you know, that's not rocket science. But these are the I
\-~s) s kinds of calculations that we're doing.
4 Now what happens when we take this same exact 5 calculation and move it into Amargosa Valley where the 6 farmers live? And you can see that things are reduced by 7 at least an order of magnitude. And all of the 8 relationships still stay the same because the saturated 9 zone modeling -- basically, once it's into the saturated 10 zone system, there is a constant degree of dispersion and 11 diffusion -- not diffusion, dilution.
12 So everything comes down to very small 13 numbers. And the question then becomes, if this becomes (3 >
k/ 14 the point at which you are going to regulate, where do you 15 cut off this design? I think that question was asked 16 earlier. And of course, that's a question not only of 17 policy makers, but also of how defensible are each of 18 these components.
19 But this is just an illustration, just a 1
1 20 scoping sensitivity analysis of the type of things that l l l l
21 we' re doing in coordir.ation with, not in support of, but l 22 we're doing it in direct coordination with the design I l
23 folks. This ia what we're doing for them showing them j 24 that yes, there is indeed some defense-in-depth here at 30 (m,) 25 kilometers.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
181 l l
l 1 Obviously even case one, you know, shows some 1
l
,g 2 positive aspects of -- for sure you'11 meet a 10,000 year
~) 3 standard like WIPP has. If you apply that same standard i 4 to all time, you have a pretty good chance of showing that )
5 this is a safe site. And you can add some of these !
i 6 barriers, if not all of them, to provide you more layers 1
i 7 of assurance that you will meet a health and safety based l l
l 8 standard.
l l
9 So I don't know how many ways that I can l l
I 10 caveat this saying don't run out and publish these because 11 these are internal calculations used to help sharpen the i
12 process. !
13 What else are we doing? Well, even as we I
,m !
- )
V 14 speak, the same team that did these two viewgraphs last I
15 week for this presentation is also looking at alternative 16 ceramic coatings, alternative degrees of the galvanic 17 protection, alternative corrosion resistant materials.
i 18 In fact, there's a joke going around now which 19 is that we can create a million dollar waste package by 1
20 assuming C-22 as the corrosion resistant material and l
l 21 spraying it with titanium oxide as the ceramic. And that l 22 would be the Cadillac of all waste packages. But that's, 23 I think, the extreme bound to which we're carrying these 24 analyses.
O.
(,i 25 We're looking at alternative cladding NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
182 1 degradation models, alternative thermal loading designs,
,e~x, 2 alternative drip shields, alternative backfills, and
(/ I 3 looking at what the effects are of sorbing materials 4 placed in the packing and invert.
5 All of these are being done in total 6 cooperation with design because we have learned through 7 experience that from a PA perspective, we're not smart l
I 8 enough to either define these or define their properties 9 over time. So we're looking at an integrated product that l
10 will fall out of this and that will be reflected in the
]
11 design that's finally decided upon.
12 MEMBER HINZE: Abe, are these results meeting l l
p_
13 the subsystem requirements of 60 as we now have them? j i i k/ 14 MR. VAN LUIK: Easily.
15 MEMBER HINZE: That's what is very apparent.
16 MR. VAN LUIK: Yes, it's very apparent that we l 17 would meet those requirements. However, that also makes 18 it very apparent that those are kind of useless 19 requirements. And those requirements again were based on 20 an intuitive conceptualization of a system that really 21 doesn't hold.
l 22 So don't catch me saying anything positive l
l 23 about subsystem requirements.
24 Since the SCP, advances have been made in (n) 25 design; understanding of the sites, process-level NEAL R. GROSS l COURT REPORTEFiS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N W.
j (202) 234-4433 WASHINGTON D C. 20005 3701 (202) 234-4433
183 1 modeling, and system level modeling. With a credible
-s 2 system level modeling capability, we believe it is both
/ 1, 3 possible and preferable to do system-level evaluations of 4 subsystem and component design changes and site 5 uncertainties.
6 And that's what I was trying to illustrate 7 there. Preparation for TSPA is proceeding as planned 8 including uncertainty evaluations, abstraction of process-9 level modeling, model testing, and external expert reviews 10 both in expert elicitations and a formal peer review.
11 Performance assessment in addition to this i 12 continually updates its system-level models to support 13 testing and design with scoping sensitivity studies to a s> 14 decision making both in terms of ongoing engineering, l
l 15 tunnelling, what materials to bring into the repository, 16 and also with the decision making as to what do we
! 27 actually put into the waste package -- or into the l
l 18 engineered system I mean. l 19 And that's all I think I was asked to address.
l l 20 But I'n certain that you have more things to ask 23 VICE CHAIRMAN GARRICK: Yes, I'm sure we do.
l 22 While we're talking about waste package design, most of 23 these alternatives seem to refer to coatings, claddings, 24 coverings, protection. Doec this, in any way, suggest
?~m
( j 25 that more fundamental considerations with respect to the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE IJLAND AVE., N W (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 2344433
184 1 design of the canister are not in the suite of options g 2 under consideration?
3 MR. VAN LUIK: I really wish you had asked 4 Jack this question because it is my understanding that 5 they have an effort in progress -- and Jack will correct 6 me if I'm wrong. But they do have an effort in process, 7 in fact, to look at alternative waste package sizes, a repository lay outs, and emplacement modes. l 1
9 Jack?
10 MR. BAILEY: Yes, that's correct, Abe. We're I l
11 working with what you see right there. We're really l 12 pushing on towards VA where we need a design that we can 13 work with. We are going to spend time and look at some of G
V 14 the alternatives and some of the sensitivities of those 15 types of things.
16 So no, we're not solidly locked at this point.
17 And that was one of the points i was trying to make with 18 the variety of some of the pieces.
19 VICE CHAIRMAN GARRICK: Okay, thanks.
20 Okay, we'll go around the table.
21 Marty.
22 MR. STEINDLER: Do you think there should be a l 23 subsystem --
24 (Laughter.)
25 VICE CHAIRMAN GARRICK: Why don't you stop NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
185 1 beating around the bush?
,, 2 MR. STEINDLER: I can't say that with a
()
~~
3 straight face. But, you know, would you address the issue 4 of whether or not there should be subsystem regulations?
5 MR. VAN LUIK: I believe from my own 6 experience that subsystem regulations that have a strong 7 nexus with total system regulations will be redundant in 8 terms of having exactly the same -- you know, they assume 9 a certain nexus by assuming a certain conceptual flow of 10 information and performance.
11 So they would just be a reflection, just a 12 mirror reflection, of part of the total picture which I 33 think buys you nothing in terms of adding public health
/-
(_,/ 14 and safety. On the other hand, if you then say well, if 15 there is uncertainty in the conceptual model and the 16 linkage in that nexus let's drop down an order of 17 magnitude or something like that, to be meaningful, you 18 have to drop an order of magnitude.
19 Let's make it more difficult then to pass the 20 hurdles of the subsystem requirements. Then, in fact, you 21 are regulating to that conservatism rather than to what 22 the Congress or the EPA has mandated as the focus on 23 public health and safety.
24 So I think from an implementer's standpoint, (mv) 25 our standpoint, it's a lose-lose situation to have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1J23 RHODE ISt>.ND AVE., N.W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 186 l
1 subsystem requirements. You either gain nothing or you es 2 add conservat-ism that really doesn't add anything to
' (~ )
,i 3 public health and safety.
4 That's my opinion.
5 MR. STEINDLER: Okay. Well, can I translate G that into saying that defense-in-depth basically is a 7 public relations issue?
8 MR. VAN LUIK: 1 do not see how you can make 9 that statement because what we show here as cur approach ,
1 l
10 to defense-in-depth is okay, here we have a system that i 11 meets the requirements, the total system requirement.
12 Here are the things we are adding to that system to give 13 us and you more assurance that that requirement will be
/ hi 5
V 14 met.
15 So how is that a public relations thing? I 16 showed that the -- you know, there's a pretty good chance l '7 if we have a 30 kilometer boundary for where we have the 18 RMEI or whatever and we can defend that, that already the 19 system, without any additions, just the waste packages 20 themselves, will probably do fine for that kind of a 21 standard.
22 But to assure ourselves that we have at least l
! 23 two orders of magnitude to go into licensing with, we will 24 probably add defense-in-depth features to the design. So Q 25 I can't agree with you. I cannot agree with that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W
{202) 234 4433 WASHINGTON, D.C 20005-3701 (202) 234-4433 i
187 1 statement.
,y 2 MR. STEINDLER: Fine, fine. That also implies t t
'v' 3 then that defense-in-depth, by your definition, is not 4 equivalent to subsystem regulations.
5 MR. VAN LUIK: No, I think my whole point was 6 that defense-in-depth and subsystem regulations, that 7 there is no clear connection between the two. For 8 example, the groundwater travel time requirement in 60 9 presumed a certain functioning of the total system that 10 just does not apply to this case.
11 So did it give us defense-in-depth? I don't 12 think so. Because if we can get 30% of the water to move l 13 very quickly through the mountain and really not contact 14 any waste packages, that's to our benefit and we only have 15 70% of slow drainage to deal with.
16 So you know, that's what I keep saying is that 17 it prese.mes a certain conceptual model that may not fit i
18 the situation that you're dealing with.
19 MR. STEINDLER: One other point. You l
20 presented in your penultimate viewgraph I think a whole
! 21 series -- four, seven statements on alternatives. At what 22 point and by what means do you folks intend to turn i
23 alternatives into a fixed design and when do you intend to '
24 do that?
/~N
(_ 25 MR. VAN LUIK: I think I-would need the help NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON D C. 20005-3701 (202) 234-4433
~_
188 l 1 of Jack in answering this question. But I think his
,_ 2 answer before was that we are looking to very quickly, k'j 3 like within this year's time frame or maybe very early 4 next year, fix the design for TSPA-VA. However, we have 5 alternative studies in progress now that are not expected 6 to mature until after the TSPA-VA.
7 So for licensing, you may see a very different a design that what we're considering here. But these l 9 studies are being done to support the TSPA-VA design to l 10 make a statement that other people can interpret as to 11 whether or not this project is viable or not.
12 MR. STEINDLER: Don't you get into a ccnflict 13 between the customer for your TSPA-VA looks at the thing q
.-) 14 and says gee, I'll make a decision on that basis and then 15 finds that your TSPA-LA looks a lot different?
16 MR. VAN LUIK: Looks a lot different; but 17 obviously, by that time, we will have a standard in place.
18 And obviously, by that time, we will have a design that 19 may not be as large as -- you know, as difficult to 20 emplace. It may have taken care of some problems at and 21 the same time still offers defense-in-depth a few orders 22 of magnitude safety margin for the licensing case.
23 I mean, that would be the best of all worlds.
24 If we came in with a design that was clearly, you know, n
( ) 25 less conservative, we wouldn't get a license. You know, w/
NEAL R. GROSS COURT REPORTERS AND lRANSCRIBERS 1323 RHODE ISt/sND AVE., N.W (202) 234 4433 WASHINGTON D C. 20005-3701 (202) 14 4433
189 1 so we have to -- you're right in a sense that TSPA-VA is j ,
.s 2 one customer and TSPA-LA is another customer, even a more l ls
)
l 3 serious customer.
4 But the NRC is charged with looking over the 5 TSPA-VA and making some statements on it. So you know, 6 it's not that loose a system. It's -- and there's only a 7 couple of years in between.
8 MR. STEINDLER: Okay.
9 MEMBER HORNBERGER: Abe, you mentioned that 10 your results were sensitive to the saturated zone flux of 11 .3 meters per year.
12 MR. VAN LUIK Yes.
13 MEMBER HORNBERGER: In looking at your graphs, j 3,
- U )
14 it strikes me that they're sensitive to that only in the -
15 - by virtue of an artifact of making a decision at 10,000 16 years. That is, does the peak dose actually change a 17 great deal with that flux?
18 MR. VAN LUIK: The peak dose at this point in 19 time, because of our modeling assumptions, does change.
20 If the water moves faster, it comes out before 10,000 --
21 you know, the peak moves to the left. But it also lowers 22 a little bit because you have a lot more dispersion with 23 the faster moving flow.
24 This all needs to be sorted out and that's why n,
()
w, 25 we have an expert elicitation and we also have, as Dr.
NEAL R. GROSS COURT REPCHTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
190 l
)
1 Brocoum described, extra testing scheduled for the I
,_s 2 saturated zone because this is an important issue. Slow I i
( )
\'/
3 moving water moves us past 10,000 years but gives us 1
4 higher doses about a half an order of magnitude higher is 5 there's an order of magnitude shift in the flux. l i
6 Because slower moving water has less mixing 7 and so there's less diluuion, so to speak, along the flow 8 path. So it's a double edged sword.
9 MEMBER HORNBERGER: Of course, that's also 10 making an assumption about dispersivity. And the real 11 issue -- or one of the real issues has to do with your 12 assumptions on dilution.
i 13 MR. VAN LUIK: Yes.
/~'N.
Y_.) 14 MEMBER HORNBERGER: And even dilution at the 15 well head which is probably a big uncertainty.
16 MR. VAN LUIK: It's a big uncertainty. And if 17 you have a well in a very large field that has, you know, 18 basically the same concentration of pollutants, you get no 19 changes at all. But if you're drawing from 50 meters and 20 your pollutant is in a surface layer, you get tremendous 21 order of magnitude differences.
22 And so when we get to the -- you know, we hope ,
1 l
23 that some of these things which are rather imponderable, l
24 that we will have encugh margin in our safety case that we
,,)
i 25 can be generous and say this is a qualitative conservatism NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-?701 (202) 234-4433
191 1 by not taking credit for it.
n 2 However, we have, you know, the EPA saying you s
)
3 shouldn't take credit for it, and the NRC advising us it's ,
4 obvious everybody that does any calculations of this sort 5 knows that there's an order of magnitude or so and you 6 need to investigate it.
7 So we will look into it, and we will make a 8 decision as to whether or not to take credit for it. But 9 even if we don't take quantitative credit for it 10 qualitatively, that will be part of the defense-in-depth 11 argument.
12 MEMBER HORNBERGER: One of the things that you 13 mentioned when -- to one of Marty's questions was that, if
/ ;
-> 14 I understood you correctly, was that you're assuming that 15 flow in fractures just completely goes around the 16 repository?
17 MR. VAN LPIK: No.
18 MEMBER HORNBERGER: That flow in fractures is 19 a good thing for you?
20 MR. VAN LUIK: What we're assuming is that --
l 21 no, what I was commenting on is that in order to take a l l
22 thousand year groundwater travel time as a serious l l
23 indication of how this system would perform, you could not l
l 24 make the assumption that, you know, some percentage of the )
,- x
. 1
() 25 flow would be very fast but totally inconsequential to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234 4433 l
l
i 192 1 your performance because it bypasses the system.
g- 2 If there is rapid fracture flow through
\)s 3 something that we stand off from, for example -- you know, 4 these are hypotheticals. But if there is rapid fracture 5 flow of perhaps up to 30% of the water coming through 6 fracture systems that we stand off from or that have no 1
7 way of impinging on the waste packages, then that's a good l 1
8 thing rather than a bad thing because it decreases the 9 total amount of flux available for radionuclide transport. l l
10 So what I was commenting on was that that l I
11 vision of a thousand year groundwater travel time is fixed 12 on a certain way that the mountain behaves -- that the 13 total system behaves. And that probably is not true. But
( i
\/ 14 the 30% is probably not true either.
15 But it's just to illustrate that that mindset 16 required a certain assumption of how the system works.
17 Where we don't think it works that way.
18 MEMBER HORNBERGER: The 30% not being true is 19 you've taken the high end? Again, you've don't what John 20 said, done something totally irrational here?
21 MR. VAN LUIK: No 22 (Laughter.)
23 The 30% kind of falls out of a mass balance of 24 when we look at the fracture properties and the matrix
,a k_) 25 properties in Bo's model. If we push as much water NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
193 1 through the system as we're pushing through on average, 2 then we would have to have about 30% of that flow moving 7-Y 3 very rapidly through a fracture system and the rest of it 4 moves very slowly through the matrix as we assumed before.
5 So, you know, that just picture of a model 6 alone shows that, if you're going to regulate us on how 7 fast the molecule of water moves through that fracture 8 system, it may have nothing to do with system performance.
9 That's the only comment I was making is on a 10 conceptual basis.
11 MEMBER HORNBERGER: The 30% -- I think that 12 I'm right -- that's a very uncertain number though, right?
13 MR. VAN LUIK: It is a very uncertain number.
f3 4 1
\/ 14 And it has changed from -- well, it hasn't been much 15 higher than that, but it has been much lower than that, 16 yes. Depending on your matrix permeability assumptions 17 which are basically moving away from assumption and going 18 back to using the data that we have.
19 VICE CHAIRMAN GARRICK: Bill.
20 MEMBER HINZE: Just a few brief questions 21 about the saturated zone again. Steve has indicated the 22 importance of this in a dose-based standard. Is your --
23 are your calculations to this point taking into account 24 the expert elicitation of the abstraction on the UZ flow?
/~~S .
( l 25 MR. VAN LUIK: Those are --
\~/
NEAL R. GROSS
?
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
194 1 MEMBER HINZE: And if they aren't, are they l ,, , 2 roughly in accord?
'~
3 MR. VAN LUIK: When we take those into 4 account, we will have a slight increase in the UZ flow 5 distribution. But not -- there was no implication that 6 this had anything to do with the saturated zone. The 7 saturated zone is fed by recharged much higher up.
8 You know, above 6,000 feet north of us and I 9 think just a little bit west of us. And so whether 10 there's more infiltration in Yucca Mountain itself than we 11 thought, it has really no real coupling to the height of ;
12 the water table. It's the recharge higher up that 13 determines the height of the water table.
-w
- 14 MEMBER HORNBERGER: But it has taken into 15 account the saturated zone abstraction?
16 MR. VAN LUIK: The saturated zone abstraction 17 did inform how to do this modeling. And in fact, it was a 18 recent regridding -- not regridding, but a recent -- what 19 is it you do with models when you match them to data?
20 Calibration.
21 The recent recalibration of the regional model 22 that gave us this .3 as kind of an average for flux.
23 That's not to say that .3 is an average for velocity, 24 however, If )j 25 MEMBER HINZE: So this came out of that work?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
195 1 MR. VAN LUIK: It can out of that work. But
~s 2 that work itself is not due until the end of this month.
I )
3 So there will probably be another recalibration of our 4 models at that point.
5 MEMBER HINZE: Steve also made the point that 6 much of the work is not going to be available from the 7 saturatied zone studies for the TSPA-VA. How are you going 8 to approach putting in values for the TSPA-VA?
9 MR. VAN LUIK: Well, it's in recognition of 10 that fact that he alluded to that we actually constructed 11 the expert elicitation on saturated zone flow and 12 transport to give us place holder boundaries on the l
l 13 uncertainties until we have the new testing done.
m I
( )
'u./ 14 So we expect that for the LA, we will have 15 that expert elicitation plus extra work in the saturated l
l 16 zone to give us a firmer story for that part of the l
I 17 argument. But for the VA, we will base it on what we know 18 now plus the expert elicitation, whatever input they give l 19 us.
20 MEMBER HINZE: Thanks very much, i
l l 21 VICE CHAIRMAN GARRICK: Paul.
22 CHAIRMAN POMEROY: I recognize that there are 23 only those two dose curves -- two slides essentially of 24 the dose curves where you show something like an order of
,o (m/ ) 25 magnitude difference for the boundaries at five kilometers NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 2344433
196 1 versus 30 kilometers.
p 2 MR. VAN LUIK: Yes.
N.] 3 CHAIRMAN POMEROY- I want to ask you a funny 4 kind of question perhaps. Does that correspond somewhat 5 to your intuition of what you might expect for a change of 6 that distance? Or do you really -- don't have an 7 intuitive feeling for that?
8 MR. VAN LUIK: I don't have an intuitive :
1 l
9 feeling except that I have looked at the assumptions that 10 feed the very simple model for basically diluting in the 1
11 saturated zone. And intuitively, that looks about right.
12 So I guess the results look about right. i 13 MEMBER HORNBERGER: What dispersivity did you
/~N.) ,
i (O 14 use, do you recall? I 15 MR. VAN LUIK: I believe the number ten comes !
16 to mind.
17 MEMBER HORNBERGER: Ten meters?
18 MR. VAN LUIK: Yes. But I'm not sure because 19 that is one of the parameters that got changed in this 20 last roll around with the saturated zone model.
21 MEMBER HORNBERGER: Ten meters fixed?
22 MR. VAN LUIK: Fixed, fixed, yes. And that's 23 why, you know, we don't claim that the saturated zone 24 modeling is definitive or defensible at this point.
( 25 CHAIRMAN POMEROY: Somebody -- and this is not HEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433 e,
1 l
197 1 -- I can't point to a document here. But somebody told me rx 2 at one point that when you held the UZ workshop, that 1
( )
gj 3 people were given the opportunity to suggest additional 4 data that you might need.
5 How does that feed back into the system or has 6 it? And does it result in somebody considering the 7 possibility of getting new data?
8 MR. VAN LUIK: Oh, you're talking the UZ?
9 CHAIRMAN POMEROY: Yes.
10 MR. VAN LUIK: Okay. What actually happened 11 there is that we have a change request in the system 12 that's supposed to mature I think on June 3rd that
_ 13 includes -- what we did was rack all of the suggestions
' ') 14 made and did a sort of a triage on, you know, if we can 15 afford just one or two, what would we do.
16 And we collapsed two or three of tham into 17 one. And this is going to result in several percolation 18 tests and niche -- what we call niche tests in the staff.
19 And also, in the longer term, it's being factored into 20 what we -- if we do an east west drift, what we will do in 21 terms of experimentation in the east-west drift.
22 So for the longer term, for the LA and maybe 23 even post-LA perhaps for the confirmation period, we will 24 be doing testing to look at, like Dr. Broccum mentioned, n.
(_,) 25 the 28-280 millimeter, you know, to get drips into the NEAL R. GROSS COURT REPORTERL AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
1 l
198 1 drift. We want to have an experimental basis for that 7s 2 kind of thing eventually. )
1
-( )
\'~j 3 And that's what some of these niche tests are i
4 going to be looking at.
5 VICE CHAIRMAN GARRICK: Abe, we heard your 6 answer on the issue of subsystem requirements, the lose-7 lose answer. Let me turn it around a little. Given that 8 we all are believers in the systems approach and an 9 integrated approach, can you imagine a set of subsystem 10 requirements that has a reasonable nexus with bottom line 11 results?
12 In other words, it's not just a question of 13 the subsystem requirements as they are defined in 10 CFR p,
14 60, it's also a question -- and given that this is a 15 specific site, we have a real advantage over other 16 licensing matters. Is there a set of subsystem 17 requirements that the Department would be comfortable 18 with?
19 MR. VAN LUIK: The Department would be 20 comfortable with?
21 (Laughter.)
22 VICE CHAIRMAN GARRICK: Well, or maybe t. hat l
l 23 you would be comfortable with. Let me take you off the 24 hook.
13 i / 25 MR. VAN LUIK: Oh, right. We have done some x_/
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE,, N.W.
, (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
i 199 1 1
1 playing around with calculations seeing, okay, for a i s
2 system to meet a certain goal, we run the system l 1 ! \
~'
3 calculation and say okay, what was the mean value of waste l 1
i 4 package lifetime, what was the mean value of things, you l l
5 know, starting to move from the waste packages.
i 6 And if we say well, for this string of 1
7 modeling, if we put a requirement there and there, then we l l
8 would be pretty assured of getting this result here. But l 9 then again, this is the situation where you have a very 10 strong nexus because you've backed it out from a l 11 calculation that gives you your compliance argument, you 12 know, if you do this in the final analysis.
13 And it's also totally dependent on all the
/g
-) 14 assumptions and everything that went into that. So I 15 would say again you would, at that point, have some very 16 reasonable and defensible subsystem requirements, but they 17 would be quite meaningless because they're tied to this la one overall conceptual model.
19 So we've struggled with it and see really no 20 win-win in having a subsystem requirement unless there is 21 something that the total system requirement completely 22 overlooks, you know. Like for -- well, I even hate to 23 suggest it, but for example, if you have a requirement for 24 a dose through all pathways or through all aqueous
( ,) 25 pathways and you say oops, at this particular site you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 t
1 200 1 might have a gaseous pathway, then you might have a 2 subrequirement, but it wouldn't even be -- it would be a i i . 1 O 3 system requirement for a gaseous pathway dose, you know, l
4 just to correct your oversight and in setting the other 5 one. )
l 6 The answer is no.
l 7 (Laughter. ) i 1
8 I would personally not be comfortable with any 9 subsystem requirements because you would have to one, show i
l 10 that it's somehow derived from and based on the total l l
11 system requirement or is less conservative than that 1
1 12 requirement.
13 And if it' s less conservative, then it likely
() 14 becomes the crux of the licensing case, which you don't 15 want it to be.
16 VICE CHAIRMAN GARRICK: Right. Now you 17 wouldn't object to being required to demonstrate subsystem 18 performance?
19 MR. VAN LUIK: No, not at all. In fact, we 20 love to. We just did that. And you criticized it for 21 being inanely conservative.
22 (Laughter.)
23 VICE CHAIRMAN GARRICK: Of course.
24 One other -- well, another question.
,a (j 25 Supposing the standard sort of came out that was based on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 201 I
l 1 peak risk and one and two and did have a time of 1
s 2 compliance associated with it.
lI L ,1
\
3 Based on what you have presented, there seems ;
l l
4 to be substantial evidence that you could design a waste 5 package system for which you could develop a pretty 6 convincing argument of containment over a 10,000 time of ;
7 compliance. Am I not correct in that?
8 MR. VAN LUIK: You're not correct in the sense 9 that what you're looking at is five kilometer 10 calculations. There is a-time hiatus between the time it l 11 leaves the waste package and gets to that five kilometers.
12 VICE CHAIRMAN GARRICK: Right.
13 MR. VAN LUIK: So you're a couple of thousand i 1
's I 14 years off. We can show thousands of years of 15 containments. Eut you will have this very slow and 16 gradual release from the system. So saying total 17 containment for 10,000 years would -- I think would be 18 close to impossible for us to ever demonstrate in a 19 licensing setting.
20 So don't give us that one.
21 VICE CHAIRMAN GARRICK: Well, the thing I was 22 getting to is, from an engineering standpoint, there's a j 23 lot of people think that you can do that. You can design 24 a 10,000 year canister, including exterior barriers. Now
.rx
() 25 if you do that, then of course -- and you're working NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
202 1 against a standard that's a peak risk standard, you have rx n 2 complicated the peak risk calculation; because with that i
'O 3 degree of containment, you're back to having pushed the 4 peak risk way out in time most likely.
5 And of course, as you push it out in time 6 further and further, the confidence that one has and the 7 ability to calculate that decreases.
8 MR. VAN LUIK: It's a two edged sword.
9 VICE CHAIRMAN GARRICK: Yeah, it's a two edged 10 sword. And I guess the issue here is what is the best 11 solution. From the standpoint of the public, I would 12 think there would be a high level of security in knowing 13 that this stuff wasn't going to even depend on the natural
(\' I 14 setting for many thousands of years -- that you had 15 engineered into this repository an extremely high degree 16 of protection.
17 MR. VAN LUIK: Can I --
18 VICE CHAIRMAN GARRICK: Yes.
19 MR. VAN LUIK: I think, you know, that's one 20 way of interpreting what the results are that we're 21 showing. Actually, that engineered system would not work 22 that way if it were not for the natural setting. So you 23 can't divorce the two.
I 24 VICE CHAIRMAN GARRICK: Well, that's true.
[ \ l
(,) 25 Sure, they're coupled. I NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l l
l
j 203 1 MR. VAN LUIK: And the other thing is that the i
7-2 happy coincidence of unhappy things here is that we have a j
('# )
3 more robust waste package in a site that has more flux l
4 than we thought maybe a year and a half ago. However, as j I
5 you saw from my dose histories, the peak dose now occurs 6 before 20,000 years within basically, you know, an average 7 pluvial cycle.
8 So I think now we have the best of both 9 worlds. We show very good chance of having no doses at 10 five kilometers for 10,000, an excellent chance. And we s 11 how that the peak dose and the long term tail off of that 12 dose which goes out indefinitely and does not increase 13 anymore because of the natura. ;tems properties --
h k/ 14 VICE CHAIRMAN GARRICK: Yes.
15 MR. VAN LUIK: -- that, you know, we have the l 1
i 16 peak dose problem within one pluvial cycle which gives us i 17 a lot more credibility as far as making a case to the i
18 public. When it used to be 800,000 years in the future, i i
19 we ourselves laughed at it.
20 Now it's looking like a more credible thing l 21 that we can actually convince people not only do we mee*
l 1
22 the 10,000 criterion; but for the very long time, this is 23 a very safe place to be.
24 VICE CHAIRMAN GARRICK: Yes. But there are 25 other trade offs. At the 800,000 peak risk, you had (S)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RH:!DE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 2344433
l 204 1 pretty well limited yourself to having to worry about a 7
2 single radionuclide.
( )
'~'/
3 MR. VAN LUIK: Correct.
l 4 VICE CHAIRMAN GARRICK: As the percolation l 5 flux increases through the repository, that advantage will 6 disappear. !
i 1
7 MR. VAN LUIK: We still show iodine, 8 technetium, and neptunium as being the three, and then !
I 9 carbon close behind them as being the three major 10 radionuclides that we have to consider even up to the ,
l 11 100,000 time period. l 12 When those other things start coming in, they ,
l 1
13 never come in even close to the levels of those three. So l c., l l
( )I 14 basically yes, you do have a very long -- turn into a 15 million years. But -- and those will have different 16 constituents making up that dose.
17 But the dose will never be as high as it was 18 in that first 20,000. And that does presents an 19 acceptable risk. That's basically the crux of the 20 argument, I think.
21 VICE CHAIRMAN GARRICK: George.
22 MEMBER HORNBERGER: I thought of another 23 question on subsystem performance as John was speaking.
24 What's the groundwater travel time?
! 25 MR. VAN LUIK: What is the groundwater travel v
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i
1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
1 205 j 1 time?
I fs 2 MEMBER HORNBERGER: Yes, for your --
I )
L./
3 MR. BAILEY: It is highly variable. And --
4 MEMBER HORNBERGER: Give me a range.
I 5 MR. VAN LUIK: Well, you can see a range, an l 6 implicit range, in the viewgraphs that I showed. When we !
7 start losing containment for some packages, we start 8 immobilizing radionuclides. We actually have some finite 1
9 number of radionuclides coming out within the first three l 10 or four thousand years.
11 So there's probably an average of about a 12 thousand years to get to that five kilometers. However, 1
13 given the variability in our modeling, we have some things l I
(~')T k- 14 coming out within hundreds of years and taking 10,000 15 years to get that far.
16 So the -- you know, the modeling for TSPA-VA 17 will lay out those uncertainties. But, you know, to 18 impose an average radionuclide travel time would make a 19 lot more sense than just say groundwater travel time.
20 However, again, that is key to an assumption 21 as to how the system is configured and how it works. And
- 22 the bottom line is public health and safety. If we can't l
l 23 demonstrate that, none of these subsystem requirements 24 mean a hill of beans. And I think Pigford did a wonderful f~'\
( ,) 25 job in '82 showing there is a case at Yucca Mountain where l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
t (202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433 l
206 1 you can meet the subsystem requirements and still fulfill l
~ 2 the EPA.
%-l No, where you can -- which way was it? It was 3
4 one or the other. There was no good nexus is what he was 5 showing, yes. And if there was a good nexus, they would 6 be meaningless.
7 VICE CHAIRMAN GARRICK: Well, I think that in 8 the risk field we see this all the time. In the nuclear 1
9 power field, we've discovered that there's way so reducing i
10 the core met frequency and, at the same time, increases 1
11 the risk. l 12 Because you've increased the scenarios under l 13 which you have offsite consequences even though the threat l
\)
14 to the containment may be less. So yes, that's why we --
15 that's why you have to take a total systems --
16 MR. VAN LUIK: Oh, that's why you were asking 17 that question.
18 VICE CHAIRMAN GARRICK: Yes.
19 MR. VAN LUIK: If you push out the peak, ys u 20 make things worse later.
21 VICE CHAIRMAN GARRICK: Right.
22 MR. VAN LUIK: Okay.
23 VICE CHAIRMAN GARRICK: Yes.
24 MR. VAN LUIK: And I think the answer is given
,r's,
(,,! 25 the new realities, that . sed to be true when we had very l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
l 207 1 low flux. You would just have this forever increasing A 2 dose. But now, because of the higher flux, we wash it out c
'L/I 3 and then bring it back to a steady state basically.
4 VICE CHAIRMAN GARRICK: Right. I'm suggesting 5 to the members that there's no reason that we have to ask 6 all our questions to Abe. And I see that Steve and Jack 7 are still here. They're fair game as well.
8 Any other questions?
9 Go ahead, George.
10 MEMBER HORNBERGER: I'd just like to hear I 11 guess once more from Jack an answer to this question of 12 getting to something like either a small number of designs 13 or a working design. I guess I'm still perplexed at how
/. ,1
'# 14 this plethora of design options is going to be sorted 15 through in a reasonable manner such that all of the 16 supporting data and testing and everything else can be l i
17 done to actually get a feasible solution here.
I 18 VICE CHAIRMAN GARRICK: Are we going to get a l 19 panel here?
20 MR. BROCOUM: A couple of weeks ago the 21 engineers came in with 14 cases. And that was when I was 22 alluding to the cost, the most expensive case among these 23 14, which was most of those things Abe was showing added i 24 on finally at $1.8 billion.
i n
\
(/ 25 We sat back because, you know, we're concerned NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
208 1 about where EPA is going, and we are concerned as it goes
- p. 2 into rule making what our positions are. So we thought i !
3 about the cases we ought to be working on. And we came up i 4 with three basic cases and three alternates to each of the 5 three basic cases.
6 I'll tell you what they are. And the 7 engineers are going to then come up with the engineering 8 that would meet those three cases by two orders of 9 magnitude. And it's the engineers who have to decide 10 what's practical from a design perspective -- drip shield, 11 backfill, all this stuff 12 And it's the PA people's job to run these i
13 cases. That's how we're working. The three cases that we ex l )
(_/ 14 came up with is 30 kilometers, 100 millirem, 10,000 years; 15 20 kilometers, 100 millirem, 10,000 years; five i
16 kilometers, 100 millirem, 10,000 years. Then, knowing l 17 that we have peak dose at about 20,000, we'll do those 18 three cases going to peak dose also and see what -- how 19 that -- what different engineering you need to get to that l
l 20 case.
I P
l 21 So in a sense, we're doing six cases which l 22 gives us a menu in the rule making we're going into with 23 EPA to work with. And we'll understand what engineering 24 or what features it takes to meet those cases, what the 25 costs are because they're going to cost those out, and (A\s)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
209 1 what the performance is.
,_ 2 So that's how we're approaching it. We've o l \;
\
'~' 3 broken it down into those six cases. But it's the last 4 one I'm going to make was five kilometers, 15 millirem, 5 10,000 and the peak dose. So that's how we're working.
6 And they're going to come back, both the engineers and the 7 PA people, and you know, show us what it takes to reach 8 those and is there really significant differences among 9 those cases in terms of cost.
10 There may not be. See, that's what we're 11 trying to find out. So that's kind of how we're laying 12 out the next several months as we expect sooner or later 13 the EPA rule to go on and be for interagency review.
,m k_j 14 VICE CHAIRMAN GARRICK: Do any of those cases _
15 cover the existing situation of the population at Amargosa 16 Valley?
17 MR. BROCOUM: Yes, the 30 kilometer case would 18 be Amargosa Valley population. The 20 kilometer case is 19 essentially the site boundary at the intersection there at 20 Lathrop Wells. And the five kilometer case is the 21 precedent that EPA has set with other r<gu.stions. So 22 that's where those three cases come from.
23 It's a big issue with us. Always has been in 24 the past. We're trying to quantify it now, see how big it p
L,v) 25 really is. There's no much difference -- for example, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS N 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C 20005-3701 (202) 234-4433 l
l
210 1 they've already established in PA there's very little --
, 2 there's a half order or less magnitude difference between
(,-)
3 20 and 30 kilometers.
4 But there's an order of magnitude difference 5 between 20 and five. So these are the kind of insights 6 we're trying to get. And we're trying to get them also on 7 the design side to see what the real trade offs are and 8 what the cost benefits are.
9 So we expect that to come up in the next --
4 10 they already have their early versions of these and we're 11 expecting to refine these over the next several months.
12 CHAIRMAN POMEROY: And Steve, just so I 13 understand it, then say for the 30 kilometer, 100 millirem f~s,
(_/ 14 case at 10,000 years, you're actually shooting for one 15 millirem?
16 MR. BROCOUM: That's correct. That would be 17 our goal.
18 CHAIRMAN POMEROY: All right.
19 MR. STEINDLER: Do you consider looking at 20 peak dose at any point in time?
21 MR. BROCOUM: Yeah, the three cases -- oh, at 22 any point in time?
23 MR. STEINDLER: Any point in time.
24 MR. BROCOUM: Well, you remember when I was in
<s (xe ) 25 front of the committee -- it must have been a year ago --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
211 1 that I was saying it was beyond human imagination at 400-
,s 2 600,000 which was the numbers I had in those days. Since
( % i w/
3 the new percolation fluxes have changed our PA and now 4 .
within the first 30,000 years, I think it's reasonable to 5 look at the peak dose cases in that time frame. l l
6 We're running these calculations out to 7 100,000 years, so I guess anywhere in the first 100,000 8 years currently we are looking at them. I think it's 9 unreasonable and it's also a design goal not to have a i
10 design for a period of time that, once that period of time l 11 is over, the design deteriorates quickly.
12 In other words, performance deteriorates 13 quickly. And that's what's happening now at 10,000 years.
! i
\/ 14 Performance is deteriorating quickly. So I think it's 15 difficult for DOE to get up in a public forum and say 16 we're for public health and safety when we have a 17 situation crossing -- and arguing for 10,000 years and 18 crossing it and arguing on here that 10,000 year standard 19 -- and at the same time, having the performance l
20 deteriorating so rapidly across that boundary. I 21 So we have a real complex issue of how to 22 handle that. So I would say within the first several tens 23 of thousand of years, that is correct. Beyond that, I i I
24 think it's a policy call we haven't made. But we are ex !
( ,) 25 looking, as you can tell from Abe's viewgraphs at all, i
NEAL R. GROSS !
I COURT REPORTERS AND TRANSCRISERS 1323 HHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
212 1 we're up to 100,000 years.
7', 2 MR. STEINDLER: Yeah, in light of the
\_.
3 uncertainty -- pardon me, John -- with which we view the 4 EPA actions, the notion of sticking to 10,000 years may be 5 unwarranted.
l 6 MR. BROCCUM: And I am familiar, of course, i i
l 7 with your letter that the committee wrote last -- I forgot j 8 what it was now, last year sometime. There are different I 9 ways to handle this. You can stick with 10,000 years but i 10 do something else beyond the 10,000 years. I mean, you 11 can have a goal and not necessarily a requirement.
12 We can look at it in EIS. There's a lot of 13 ways to handle it. But I think we cannot ignore it. In I l k' 14 that respect, I agree with you.
15 MR. STEINDLER: That's my point.
16 VICE CHAIRMAN GARRICK: Thanks, Steve.
17 Any other questions?
18 I want to thank the present ers. That was a 19 very interesting morning. And we expect to continue this 20 discussion into our next meeting. And we look forward to 21 seeing some of you there. And with that, I turn it back 22 to the Chairman.
23 CHAIRMAN POMEROY: Thank you, John.
24 I'd just like to add my appreciation. Both at
) 25 this point and this afternoon I'll reiterate we are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
213 l
1 appreciative of the support that DOE is providing to us i
I p._ 2 here by committing the time and resources to make these I
'# l 3 presentations. We hope that we'll be able to cont.inue l 4 this cooperation because it works very well for us. l 5 With that, I' d like to adjourn this meeting 6 until 1:00 this afternoon. We'll come back then and we'll 7 turn to the question of the use of expert elicitation.
8 (Whereupon, the proceedings recessed for lunch 9 at 12:01 p.m.) ,
l 10 11 i
12 13 l f ~y k 14 15 16 17 18 19 20 I
21 l
l i
22 23 24 fm
(,
~s j 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
214 1 A-F-T-E-R-N-O-O-N S-E-S-S-I-O-N g 2 (1:17 p.m.)
! 1 i
3 VICE CHAIRMAN GARRICK: I think we'11 bring 4 the meeting to order. Our Chairman is expected 5 momentarily. But I think in view of the agenda and our 6 ability to take the time up, probably it's wise for us to 7 get underway. And I'm sure Paul would support that view.
8 This afternoon we're going to talk about 9 expert elicitation. And this is a subject that Paul is 10 personally very interested in and involved in and probably 11 had a few introductory remarks to make, which I won't 12 make. We'll let him sandwich those in later on. So I'd 13 like to just move ahead with the presentations and --
r 1
_d' 14 CHAIRMAN POMEROY: My apologies.
15 VICE CHAIRMAN GARRICK: We were about to 16 introduce the first speaker, but if you have some opening 17 remarks, Paul, it's your --
18 CHAIRMAN POMEROY: Please go ahead.
19 VICE CHAIRMAN GARRICK: You want to go ahead 20 with the first presentation?
21 CHAIRMAN POMEROY: Yes.
22 VICE CHAIRMAN GARRICK: Steve Brocoum.
23 (Slide) 24 THE USE OF EXPERT ELICITATION IN THE REVIEW OF rh
( ) 25 A HLW REPOSITORY v
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 2000 5 3701 (202) 234-4433
215 1 MR. BROCOUM: Expert elicitation. My name is,
,s 2 of course, Steve Brocoum for the court reporter.
I. \
3 (Slide) 4 MR. BROCOUM: Here's the outline, what the 5 purpose is, which ones we're planning currently, our 6 perspective on expert elicitations, -- there are some 7 issues there that I think we have with the NRC staff --
8 and a summary.
9 (Slide) 10 MR. BROCOUM: We feel that expert elicitations 11 are particularly useful in the following situations: when 12 empirical data are not reasonably obtainable or the 13 analyses are not practical to perform.'
(_) 14 There are a lot of issues in our program that 15 we're not going to have the complete information. It's 16 impossible to characterize every square inch of a site.
17 Certainly if the point of measuring is 5 kilometers or 30 18 kilometers is really impractical in the sense of getting 19 data from every unit along every location.
20 Where uncertainties are large and significant 21 for demonstration of regulatory compliance. Of course, we 22 are doing something unprecedented. When we briefed a 23 couple of weeks ago within DOE a Tara o'Toole, who is head 24 of ES&H and she is now the person that we report to, at
,r~3
() 25 least for now, in the new administration on EPA standard, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
216 1 when we were talking about 10,000 years, her comment was,
,, 2 "That's twice recorded human history. It's ridiculous."
3 1 mean, she thought that was way beyond what we ought to 4 be doing, to just give you an examp1.e.
5 Anyway, it is twice recorded human history.
6 And I like to think that if you go back 10,000 years, 7 you're at the beginning of the Bronze Age. So that's for 8 a little bit to put it in perspective.
9 In most cases in the natural system, different 10 conceptual models are permitted by the data. It's very 11 difficult to eliminate all the various conceptual modele.
12 And if you do, someone else will dream up another one. So 13 there are always new models to deal with.
i V 14 And there are interpretations that are 15 required to assess uncertainty in the data and processes
)
16 and models. Those elicitations are important in bounding 1 i
i' 17 uncertainties, again especially with the unprecedented 18 time frames we're being evaluated.
19 We're doing characterizations for a period of 20 10 or 20 years, performance confirmation for a period of 21 50 years. And we're trying to extrapolate that kind of 22 information to how the site may perform for thousands or 23 tens of thousands of years.
24 (Slide)
(v ) 25 MR. BROCOUM: Expert elicitations can be used .
1 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234 4433
217 1 to help us reach conclusions so we could move forward.
73 2 We're never going to have all of the information. We will
'\[
3 always have, as it states in 10 CFR 60, gaps and
< uncertainties. I think it is the quote. Okay?
5 So we can quantify the range of information 6 and quantify the informed technical opinions bearing on 7 technical issues by getting experts together. We can 8 integrate the diverse technical input, the conflicting 9 type of input, and document also conclusions within the 10 parameters of the evaluation.
11 (Slide) 12 MR. BROCOUM: We are planning to conduct 13 several expert elicitations. And they are in various I\ ') 14 stageo of completion.
15 The purpose -- I think I'm going to go through 16 sort of quickly because I think I'm repeating to some 17 degree what Abe said -- is to facilitate our development 18 of our key process models, which are the key models that 19 feed the TSPA. We characterize the knowledge and 20 uncertainty ia technical issues most important to these 21 process models.
22 (Slide) 23 MR. BROCCUM: These expert elicitations 24 generally follow the guidance provided by NRC. And the D.
(s,) 25 reason the word " generally" is here, for example, the
- NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
218 1 volcanic hazard one did not exactly follow the guidance 2 because we started it before the guidance came out.
7-3 In one case, we gave information to the 4 experts that DOE compiled and the NRC thought may be in a i 5 sense biasing the experts. So in the following ones, i 6 we're not taking that particular step because that's not 7 in the guidance.
8 In any case, we generally follow and 9 essentially follow the guidance provided by NRC. That was 10 issued, as you know, in August of '96.
11 (Slide) 12 MR. BROCOUM: These are the ones that are 13 underway or completed: force of probablistic volcanic
\)
m 14 hazard analysis. That was started before the guidance 15 came out. It was completed in June of '96. And basically 16 it was because there was a disagreement between us and the 17 NRC on the probability estimates in volcanism.
18 It was 10 experts over a period of 20 months 19 and at a cost of $1.8 million. So they are very resource 20 and time-intensive. They take up a period of time. We're 21 trying to get more efficient. You'll see as I go through 22 what we're planning to do. he're trying to cut it back to 23 6 months and about 6 or 7 hundred thousand dollars each.
24 The seismic hazard one, which is due to be
()
c.
25 completed this summer, involved a total of 25 experts, 13 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D C. 20005 3701 (202) 234-4433
}
219 1 months, and $3 million costs.
,3
, 2. (Slide) 3 MR. BROCOUM: The unsaturated zone, this one l 4 is due to be completed if it hasn't already been 5 completed, this month, unsaturated one flaw one, the key 6 parameters, 7 experts, 8 months, about $700,000.
7 The waste package degradation, important I l
8 parameter for the TSPA, will be completed this summer.
9 That's 6 experts, 6 months, and approximately 4650,000. I 10 And the others that are coming up, we're 1
11 trying to also do them more or less within that time 12 frame, 6 months, more or less 6 experts, and about 13 $650,000. l
,e m e i
(-.) 14 (Slide) 15 MR. BROCOUM: The saturated one flow and 16 transport, again very important for TSPA, will be 17 completed in September.
18 Thermohydrology is being planned for Fiscal 19 Year '98. So it will start after October 1st, and it will 20 be completed by July, very important issue here in our 21 design in our TSPA.
22 (Slide) 23 MR. BROCOUM: Waste form dissolution, one of 24 the key parameters in our waste containment and isolation
,, m 25 strategy, will be also undertaken after October 1st and be (u s)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
220 1 completed by the Fall of '98. It will be about a $650,000
,, 2 study, we expect.
/ \
3 MEMBER HINZE: Steve, if I could interrupt 4 you?
5 MR. BROCOUM: Yes.
1 l
6 MEMBER HINZE: Cutting these down to the half 7 mil to a mil area, is it primarily resulting from a 8 decreased number of experts or where --
l 9 MR. BROCOUM: Well, I need to ask some people 10 in the room. That's where it seems to be because there i
11 are about 6 or 7 experts each, instead of 10 in one case 12 and 25 in another for the seismic hazard. So as you add 13 experts, of course, you probably add time and you probably
( ,/ 14 add costs.
15 MEMBER HINZE: But is there any other way that l
16 you're saving money within these programs?
17 MR. BROCOUM: I don't know. I think it's a 18 timing factor. The six months are very important. But 19 the trade-off is you can do fewer -- you know, if you do 20 more comprehensive, in the long range you're going to do 21 fewer of them. And so the question is these are the key 22 process models, the key parameters.
i j 23 We want to cover them all. We may have to 24 revisit some of these after the VA for the license (n) 25 application. We would not exclude that possibility, i NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433 l
r
221 1 especially if we get more information.
,y
,- 2 (Slide)
Y. 3 MR. BROCOUM: Our perspective. These 4 elicitations are a vehicle for enhancing -- so my total 5 cost is 7 expert elicitations and about $8 million, more 6 or less the total bottom line so far. Okay?
7 These elicitations are a vehicle for enhancing 8 internal and external confidence in the technical program.
9 They help to eventually support DOE's compliance arguments 10 that we will have to make to the standards that are 11 eventually promulgated.
12 They don't eliminate the need of judgment on 13 the part of decision-makers, but they certainly give you g
P\> 14 information in a structured form. And they try to pull 15 together all the uncertainties.
16 And, as I've said several times, because of 17 the process you go through and the technical expertise you 18 need, you need to get real experts, they are expensive and 19 they are very time-consuming. So even the fastest one 20 we're planning to do with take six months.
21 (Slide) 22 MR. BROCOUM: Now I want to talk about the j 23 volcanism expert elicitation because that's the one we 24 have the most experience with in dealing with the staff.
A
( ,) 25 We had a meeting on the 25th of February. And l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 1
222 1 this Is now how it appears it appears to us. This is how
,s 2 we feel. I don't want to say what the NRC staff did. I'm
( )
v 3 telling you how we feel. It's very important because we 4 have some negative feelings. Those are feelings. Okay?
5 So I am trying to be very clear.
6 It did not appear to us that the staff agree 7 with our results. It seemed to us that the NRC staff took 8 the opinion of one of their experts and weighed them 9 higher or the same as the ten of ours because basically 10 all of their arguments were presented by one expert in 11 terms of the probabilities.
12 They did not give us a technical basis for l
13 such an opinion. They never showed us a probability i /~N
( )
x- / 14 distribution function. They never showed us a study that 15 is similar to our study. Although they disagree with the l
l 16 results of our study, some of the results, -- I won't say 1 17 all of the results -- they suggested a probability. And 18 that was one person's opinion. It was only an opinion.
19 That's how it seems to us. You can almost say it was 20 arbitrary from our perspective.
l 21 We recognize that the NRC staff needs to l
22 conduct independent evaluations of our work, but we think l
l 23 that when they differ with us, they have to be accountable 24 to explain to us why they differ. I mean, they're as l C\
() 25 accountable as we are in terms of technical work when they NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433 l l
l 223 1 have a different opinion.
l 2 (Slide) 7_
- )
~#
3 MR. BROCOUM: So we have a disconnect, if you 4 like, on the volcanism seismic hazard analysis. In 5 general, to me it's not the volcanism one. It's the whole 6 concept of expert elicitations. It's a bigger issue 7 because we're doing seven. We'll probably do more by the 8 time we get the license application or we'll repeat them.
9 So it's an important issue for us.
10 We believe that an appropriate level of weight 11 should be given to such elicitations. Otherwise, it puts 12 the licensing value of expert elicitations in question.
13 And what I am talking about in this particular
() __ 14 case -- I think you were given a detailed presentation of 15 this last month -- is the NRC staff told us that they 16 could agree that the probability of volcanism intruding 17 the repository was -- they showed us a bar from 10-7 to 18 10" We felt that the NRC staff wanted us to ignore this 19 information.
20 They try to convince us to move to this range.
21 We have a very well-structured and very well-documented 22 expert elicitation. If there's a problem with it, we want l 23 to know what the specific problems with the elicitations 24 are.
25 If the NRC did a separate study and they have l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASH!NGTON, D C. 20005-3701 (202) 234-4433 1
i
224 )
l 1 a probability distribution function that looks something l I
,~s 2 like thic, we would like to see it and see the basis for r i l
L } \
3 it. We have not seen any of that. So it comes across to i 4 us as an opinion, a technical opinion of an individual.
5 That's the only reason I'm showing this. !
i 6 (Slide) ]
7 MR. BROCOUM: So, in summary, we do have an I l
8 unprecedented program. We all know that. Expert ,
l 9 elicitations should play a key role. I inserted the word ;
1 10 "should." l I
11 We plan to use the results as part of the 1 12 basis for licensing. These results will be 13 well-documented. They will follow the NUREG. We expect I
\_/) 14 that these elicitations will be given the proper weight 15 during the licensing review.
16 We feel for us to proceed successfully, both 17 us and the staff, we need to agree on the value of such 18 elicitations with NRC management because they are very 19 time-intensive and resource-intensive.
20 So that's basically my point, my presentation.
21 CHAIRMAN POMEROY: Thank you, Steve.
22 I could, as I have privately, call your 23 attention to my remarks yesterday to the --
24 MR. BROCOUM: Yes, I have --
rx
( ) 25 CHAIRMAN POMEROY: -- Commicsion and to the v
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
225 l 1 transcript, the very end of the transcript, which you also
,- s 2 ought to review, of our meeting a month ago. !
- ' ) i i
~'
3 Are there any questions for Steve from my l
4 colleagues? Marty?
5 MR. STEINDLER: Yes. It strikes me that your ;
i 6 comment about one versus seven or whatever, ten, ir a body l l
7 count argument. And so do I assume, therefore, that an l I
8 expert elicitation result derived from 20 people is better l l
1 9 than one derived from 10 people, which is better than the '
l 10 one derived from 2? i 11 MR. BROCOUM: Not absolutely, but generally 12 there is a weight of the evidence. And when you do an 13 expert elicitation, you do get a group of experts,
'/'~;
k )4
- 14 recognized experts. And you have a process you go through 15 to solicit them. And the results are a combination of 16 that process.
17 If one person outside the process has some 18 information or has done a little bit of analysis, it's 19 incumbent on that person to not only state his opinion but 20 present the basis of an --
21 MR. STEINDLER: Okay. That was not my point.
22 Let me ask you a different kind of question.
l 23 Does DOE believe that there exists one set of protocols 24 for expert elicitation that is demonstrably more likely to e~s (v) 25 lead to an acceptable answer than others?
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
226 1 MR. BROCOUM: I would say not necessarily so.
,s 2 Since the staff has issued a NUREG and we have had some s i )
3 meetings on that, we have agreed to follow that NUREG. As 4 I pointed out for volcanism, that elicitation started 5 before the NUREG came out. So in some instances it 6 doesn't quite follow it, but it does in most.
7 I think you need to have some sort of a 8 process.
9 MR. STEINDLER: I'm not sure I know what l
10 you're saying. Are you saying you happen to be following l l
11 that NUREG because it's a NUREG and you've got to deal 12 with the NRC?
13 MR. BROCOUM: Oh, we have to deal with the h
i (~~J
\~ 14 NRC. And usually when you follow NRC's procedure, it's 15 easier to get through the process. If you have a 16 different procedure, as you know, the burden is on the 17 applicant to not only convince them you're right but that l
18 your process was good.
19 VICE CHAIRMAN GARRICK: Steve, is your beef 20 with the process or with the result?
l 21 MR. BROCOUM: My beef is I think we followed i
i 22 the process, almost followed it. And I think they didn't 23 like the result. And they never gave us evidence as to 24 why their -- if they want to close out volcanism views in A
() 25 their probabilities, tlat's fine. Then let's close it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 227 l
l 1 out. But volcanism has been churning for years. And the
,3 2 reason we started is we were disagreeing two years ago.
( )
' ~
3 I mean, so I think we honestly followed the 4 process. We had experts. We've documented. I'm not 5 saying there are no flaws with it, but if there are flaws 6 with it, that would throw off our distribution by an order 7 of magnitude. We want to know what they.are so we can 8 correct it.
9 If they've done an analysis that gives them a 10 different distribution, we would like to see that analysis 11 so we can see -- you know, usually when you compare two 12 methodologies, you get some insight.
13 We haven't seen the methodology on the NRC l )
\~' 14 side. So we can't get that insight. We only know they 15 don't like what we did. We know there is an implication l
l 16 we ought to throw off the data from the lower half of our !
17 probabilities. And that doesn't help us become smarter on 18 the issue. So that's the issue.
19 VICE CHAIRMAN GARRICK: Given the 20 uncertainties that are involved here, do you really see 21 that there is that big a difference between the two 22 numbers that have been proposed as the basis for a l 23 calculation if you must use a point estimate?
24 MR. BROCOUM: There's not that big a rx kj 25 difference, but that's not the issue because we're going NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS
! 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
l
228 1 to use it in other places. So I think there's a 2 fundamental issue here.
/-~sT t/
3 The point estimate they wanted to agree with 4 was beyond 99 percent of our curve. If we take that in 1
5 every case, we may not be able to design and build a l 6 repository. So there's a bigger issue here.
7 VICE CHAIRMAN GARRICK: Thank you.
8 CHAIRMAN POMEROY: Marty, do you have any l 1
9 other? George?
10 MEMBER HORNBERGER: Yes. Steve, I just had 11 one question. On the wording on one of your slides 12 talking about the purpose of expert elicitation, you said l l
13 that such "elicitations are important in bounding l
,e~\ l U 14 uncertainties." I just wondered if you would clarify the 15 word " bounding."
16 MR. BROCOUM: I should have said understanding 17 uncertainties. I didn't even notice that. I would maybe 18 say understanding.
19 CHAIRMAN POMEROY: Bill?
20 MEMBER HINZE: A couple of things, Steve.
21 This term " proper," " DOE expects that these elicitations 22 will be given the proper weight during the licensing 23 review," that's not a very definitive word. I'm sure 24 we're all in agreement on that.
25 (v) That's in the beholder's eyes. Could you be a i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
l l
1 229 l
1 little more specific about what you mean by " proper"? '
7._x 2 MR. BROCOUM: We submitted the expert t \ l
\ /
3 elicitations to the staff. If they have issues with them, ;
4 we would like to see their issues in writing. We would 5 then maybe address those specific issues.
1 6 I don't know how we would proceed, but i 7 normally with the staff you submit something. They ask 8 you questions. They give you a report back. Then you l
9 kind of go back and forth until you reach closure.
10 What's happening here is we submitted our l l
l 11 report. They did their analysis, and they got their i 12 probabilities. We don't know what the base of the 13 analysis is. We have no way to dialogue I p
I s/ 14 MEMBER HINZE: So you need formal 15 documentation in a timely manner?
16 MR. BROCOUM: Well, I don't know if we need 17 formal. We don't have a way to dialogue. We could be 18 talking past each other, which I think we are to some 19 degree.
20 MEMBER HINZE: A second point is a comment as 21 a result of attending one of the UZ expert elicitations.
22 I thought it was really excellent in the sense that not l
23 only did the experts provide their opinions and their 24 guidance on the uncertainty, but they also provided I)
%J 25 information on how additional data could be obtained in e.,
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISI AND AVE., N.W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
230 1 whatever reasonable is, reasonable manner.
1 p- 2 I thought that was the right people at the
('/ 3 right stage coming up with some guidance for the program l 1
4 that was really very valuable and I hope will be shown to I i
i 5 be valuable in the future. i 6 And I would encourage the DOE to consider that l
l 7 type of approach with your elicitators as you move through I 8 the program. It really was a great --
9 MR. BROCOUM: That's an excellent suggestion.
10 MEMBER HINZE: -- thing to do. )
11 CHAIRMAN POMEROY: Steve, I have any number of i l
12 questions, but I think for expediency, you know that, of l 13 course, I at least Ehare some of your concerns about the r^N) ,
's_/' 14 problems of communication. And we'll look forward to try 15 to improve those --
16 MR. BROCOUM: Thank you.
17 CHAIRMAN POMEROY: -- to whatever extent our 18 advice carries any weight. But we du appreciate your i 19 coming over and giving us your views.
20 MR. BROCOUM: Thank you.
21 CHAIRMAN POMEROY: Thank you.
l 22 Mike, I believe you're up next. Mike Lee from 23 the NRC staff will talk some about the same subject, the 24 use of expert judgment.
r i (v >) 25 MR. BELL: Dr. Pomeroy, before Michael starts, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W l
?34 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
231 )
l 1 Michael Bell. l
, 2 I guess Mike is going to give a more generic (m) '
3 presentation on the BTP, how DOE appears to be i
4 implementing it overall from our observation of some of l 5 the other explanations that are going on. But afterwards I 1
l 6 I think we'd like to come back and have a general l
l 7 discussion on both issues. 1 1
8 CHA1RMAN POMEROY: That would be fine, Mike.
9 MR. LEE: So welcome to Point / Counterpoint. I i 10 promise to share any residuals with anyone in here if we 11 ever get on TV. How's that?
1 12 (Slide)
I 13 MR. LEE: As Dr. Bell pointed out, I'm going
!O
! i G' 14 to speak to the staff observations generally with regard 15 to our monitoring of the elicitations that are taking 16 place right now as well as kind of bringing the Committee 17 op to speed on things that have happened since the BTP was 18 published back in November.
19 (Slide) 20 MR. LEE: In terms of discussion pointa, here l 21 are the things that I'd like to talk to today, Just for 22 the purposes of discussion, when I say "BTP," I also mean 1
23 NUREG-1563. It's kind of an abbreviated version of the 24 same thing. Rather than go through these, -- these speak
,m v
) 25 for themselves -- I'll just cut to the chase.
f
! NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON D C 20005-3701 (202) 234-4433
232 1
l l 1 CHAIRMAN POMEROY: Mike, just let me interrupt 2 you for one second.
1 73 i
s
%J l 3 MR. LEE: Sure.
l 4 CHAIRMAN POMEROY: Does somebody have a copy l
l 5 of Mike's viewgraphs? I don't seem to have a copy up 6 here. Thank you.
7 (Slide) 8 MR. LEE: Just as a matter of history, the 9 public comment period on NUREG-1563, which is this little 10 orange thing here, ended back in May of '96. We got three 11 sets of comments, as some of you may remember. One was 12 from the State of Nevada. One was from the Department of 13 Energy. And one was from the Nuclear Waste Technical
(\ \
k -) 14 Review Board.
1 15 I believe the last time I briefed the ACNW on j 16 the BTP before we went final, I acknowledged that I l
17 thought we had an alignment of the planets because I 18 believe the collective thinking was that the BTP was on 19 target and that the level of detail in the document was 20 what everyone was looking for. And, more importantly, 21 everyone could live with it.
22 So, with that endorsement, we briefed you all l
l 23 I guess in August of '96. In September of '96, we were 24 invited to engage with DOE in an Appendix 7 meeting on n
) 25 comparison of the guidance in the BTP with the recently NEAL R. GROSS Ci .f REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
233 1 completed PVHA. And we did that in September.
,,, 2 Just to contradict Dr. Brocoum a little bit, I
\
)
/
3 my recollection is that we felt that the BTP and the PVHA 4 process were pretty much on target. We did have a couple ,
1 5 of comments and questions and suggestior that we shared 6 with the DOE staff, but overall we thought we saw a pretty I
7 good coincidence between what the guidance asked for and j 8 how the elicitation was conducted. {
9 Granted there were a few things that we were a j i
10 little squirrelly about, but I think, all in all, staff 11 was pretty happy.
12 CHAIRMAN POMEROY: What were those? Can you 13 just --
(~
-) 14 MR. LEE: I'll get to those in a minute.
15 CHAIRMAN POMEROY: Okay. Thank you.
16 MR. LEE: NUREG-1563 was finally published in 17 November of '96. Four hundred and fifty copies were 18 printed. And as of this week, it's my understanding that 19 fewer than 35 are still in stock. So there's been a lot 20 of requests for the document. So I guess it has been 21 pretty popular.
22 Subsequent to the printing of the BTP, the NRC 23 staff developed three piece of correspondence related to 24 the document or the pub 12shing of the document. The first :
i
[)\
( 25 one was a letter to DOE in December which proposed closure NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
234 1 of six open items that were related to the SCA or study
,- 2 plans that addressed the use of expert judgment.
('~' )
3 The second letter, which was kind of a 4 follow-on to the Appendix 7 meeting that we held in 5 September with DOE, brought to DOE's attention three 6 issues we thought they should begin to think about as they 7 implement the guidance in the BTP. l 8 I'll get to those later on, but, very briefly, 9 the first issue was regarding a level of detail and 10 documentation based on feedback. You know, the 11 elicitations are kind of a give and take type of thing.
l l 12 And I think both we and DOE agree that there's some ;
13 minimum level of detail that's acceptable, and we just )
f'~)
\_ / 14 dialogued a little bit about that. I can get into that a 15 little later.
16 The second issue concerned the use of the 17 technical facilitator/ integrator. And I'll speak to that 18 a little later as well. You may recall that DOE in the 19 conduct of the PVHA relied upon the SSHAC report. That's l
l 20 the Senior Seismic Hazard Analysis Committee report, which l
21 is a jointly sponsored NRC, DOE, and EPRI product, which I 22 understand is now in press.
23 And the third item concerned the need to 24 consider updating the results of an elicitation should the es (J) x-25 conditions warrant it. That issue was kind of a broader NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
235 1 issue in our view because as both NRC and DOE go down this
,s 2 track towards viability assessment, site suitability, and
(' ~ '
)
3 potential license application, we think we need some 4 flexibility in the system to recognize that there may be 5 the need to go back and reexamine issues that have already 6 been closed. I was happy to hear Dr. Brocoum acknowledge i
7 that, and I think we're in pretty good shape.
1 l
8 And then in January of '97, following a staff )
i 9 monitoring of the PVHA, we understand that DOE wanted some 10 information on how to document and disclose conflict of l l
11 interest. So those are the three pieces of documentation j 12 that we're now kind of tracking as well as implementation ]
13 of the BTP.
,9
~/ 14 Since late '96, the staff has been monitoring 15 the implementation of the BTP. And we heard Dr. Brocoum 16 just speak about the number of elicitations that are 17 planned and underway.
18 And then this past April, we were invited to 19 brief the Atomic Safety Licensing Board on the use of 20 expert judgment in the High-Level Waste Program. I can 21 speak to that in a little bit if you don't let me forget.
22 That will probably be on Slide 5.
23 (Slide) 24 MR. LEE: On this slide, what I've attempted
/~N 25 to do is just kind of summarize what the BTP says. Dr.
(s_-)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
236 1 Brocoum also hit some of these points. So I'm not going
,q
, 2 to really go through the slide, but I'll just cut to the (sl 3 bottom line.
4 You may recall that DOE came in here about a 5 month and a half ago or so and was briefing the Committee l
6 on the PVHA, and I think the staff did as well. At the 7 time, Kevin Coppersmith got up here and was talking about l
8 why they were doing elicitations. And one of the points 9 he made, which still kind of sits with me and I think with 10 many of the staff, is that the use of formal structured 1 11 elicitation process is expected to contribute to the l l
l 12 robustness of the licensing case that DOE is going to l l
l 13 make. I think everyone in this room would agree with
\
)
% d' 14 that. I 15 At least if you think about how NRC has been l l
16 conducting its program over the years, one of the mantras i
17 that we kind of live to is document your decisions, make 18 the process transparent, be robust, make the best case 19 that you can, and be good.
20 In many respects, we see that the l
21 implementation of the BTP in the context of these 1
22 elicitations hopefully will bolster DOE's case at the time 23 of any potential license application submittal. We can't 24 say that it's going to guarantee that the staff is going t
n 4
() 25 to agree with the results in every instance, but we think NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
237 1 that DOE is going to be better off by applying the BTP as e~x 2 it would another measure of QA. So I don't think there's (v) 3 really much disagreement between the staff and DOE in that 4 regard.
5 So that's all I've got to say there. So let 6 me just move on.
7 (Slide) i 8 MR. LEE: In the next slide here, I just 9 wanted to point out a few things that NUREG-1563 says. j 10 You may recall that we lay out some recommended guidelines 11 on how to conduct an elicitation, under what circumstances l l
12 it might be appropriate to conduct an elicitation.
13 In that regard, we had this nine-step process j t \
'\~ 14 that we tried to, for lack of a better word, recognize 15 based on a review of the literature. It seems every time 16 we picked up a book on expert elicitation, one or more of 17 those nine process steps were identified. So we thought 18 that was pretty good.
19 We also recommended that we have a provision 20 for updating, which I've also talked about should the 21 results of an elicitation, should some work later on in 22 the future, suggest that there's a need to revisit the 23 elicitation results.
24 The BTP also pointed out that we don't n
t_) 25 preclude DOE from using its own experts. I think it has l NEAL R. GROSS
! COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433 l
t
238 1 to be recognized that DOE, as any other NRC licensee, is 7-s 2 going to make its licensing case the best it can. And it L'] 3 can rely on anyone and everyone it can to make that 4 licensing case. So I think that's an important point to 5 point out.
6 I think it's an unrealistic expectation in 7 some instances that there are going to be unblemished l 8 experts out there that have never been involved in the 9 program in one way, shape, or form., And we think that !
10 there are ways of handling that. l 11 One of the ways we think that that's going to 12 be dealt with is by having conflict of interest and bias 13 procedures worked in, and I'll speak to that in a little
(~/ 14 bit. It gets to one of my letters or our letters.
15 Lastly, that's kind of a given. You can't use 16 experts in place of data collection, which kind of leads 17 me into where the tenor of the discussion with the 18 licensing board went.
19 I don't know if you're familiar with how they 20 operate, but I guess every year they have a little 21 workshop and pull all of the people in from all of the NRC 22 programs and huddle for a couple of days and talk about l 23 various things that are going on in the programs. And I 24 guess my number came up, and they asked me to talk about fs k j) 25 expert judgment in the context of the High-Level Waste NEAL R. GROSS l COURT REPORTERS AND TRANSCRl8ERS l 1323 RHODE ISLAND AVE., N W.
(202) 234 4 433 WASHINGTON, D C. 20005-3701 (202) 234-4433
239 l
1 Program, as I mentioned before.
c 2 Some of the observations I walked away from i
( :
x~>
3 after talking -- it's not a free-for-all, but it's a i
4 pretty open give and take type of discussion -- are I 5 think that the licensing board recognizes that the l 6 High-Level Waste Program -- and this is just my opinion.
7 I can't speak for the licensing board or the licensing 8 board staff, but I believe they recognize that the !
1 9 High-Level Waste Program is kind of a unique type of 10 activity in regulatory activity space.
11 You're dealing not with a facility that is 12 going to have a lifetime of one or two generations. It's 13 going to be many, many generations based on a 10,000-year
)
J 14 performance objective as it currently stands.
15 So, in that regard, I think they recognize 16 that there is going to be a fundamental change in how 17 decisions are made regarding compliance with the 18 requirements and things like that. They understand that 19 there is going to be a probability-based standard of some 20 shape or form based on the ongoing deliberations we have 21 with the EPA as well as our own independent rulemakino 22 efforts to conform to the EPA standard.
1 23 In that regard, I think they understand that 24 because you have a probability-based standard, there are l
) 25 going to be probability distribution functions, as we all NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C, 20005-3701 (202) 234-4433
240 1 know. And, in many respects, expert judgment is going to r_ 2 help us to develop some of those distribution functions.
I 3 The licensing board, though, still wants DOE 4 -- and they didn't tell me to sa y this, but I think they 5 still expect DOE to make the licensing case the best they 6 can based on available data. And to the extent that 7 expert judgment helps in that regard, so be it.
8 They did have a few things to say about the I l 9 staff technical position, one of which I think is they 10 would have liked us to have gone in a little bit in the 11 position about how the use of expert judgment might play 12 out in the context of a licensing hearing.
13 And we told them that, our response was that:
~s I
m 14 Well, that's kind of your decision to make, but I think 15 they wanted -- as the licensing board, they're the final ;
1 16 arbiters of what's used and what's not used in terms of l 17 demonstrating compliance. They might the final decision.
i 18 However, I think they would have liked to have had us 19 explore that a little bit in the context of maybe some 20 background material and things like that. l 21 The other issue they would have liked us to 22 explore a little bit is a point that I think Dr. Pomeroy 23 or Hinze or maybe Dr. Steindler just brought up was this 24 business about one expert versus collection of experts. I
/m
( ) 25 think the board would have liked to have heard the staff NEAL R. GROS $
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
241 1 explore how you would weigh the various opinions based on
,y 2 an expert judgment, as opposed to one individual expert's
\ j v
3 opinion, so we said, well, maybe next life. If we ever 4 go back and rewrite the BTP, that's something we can 5 pursue.
6 The one impression I did walk away with, 7 though, was also recognition -- I think they do, too --
8 that the rules and evidence in 10 CFR Part 2 .aeed to be 9 fixed because right now things are geared towards just 10 single experts and not multiple experts. And they realize 11 that there's a disconnect there. I'm sure sooner or later l
12 that's going to get worked out.
13 (Slide) l I i f
'd' 14 MR. LEE: Just in closing, though, we also in 15 the NUREG talked about the BTP is going to be the standard 16 against which we judge any future elicitations that may l 17 become part of a license application.
l
! 18 And we also included some information on an l
l 19 expanded definition of peer review. NUREG-1297, which is l
l 20 I guess about a 1988 staf f position on the elements of an i
21 acceptable peer review.
22 We weren' t very comfortable with the a
23 definition of peer review in there. There wasn't much to 24 go on. So when we put the BTP out for public comment, we p
() 25 expanded that definition.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, O C. 20005-3701 (202) 234-4433
242 1 (Slide) g~s 2 MR. LEE: The next slide here talks about what
( )
'w/
3 the NUREG-1563 does not say or another way to say it is 4 here are the things that you can't do or what the BTP 5 hasn't attempted to do.
6 NUREG-1563 has not attempted to prescribe the 7 specific technical issues for which expert judgment should 8 or should not be applied. We agree that that's the 9 prerogative of DOE to make as potential applicants. So we 10 tried to stay out of that area. And, happily, DOE agreed )
11 with that position.
12 The other issue that we tried to point out for l I
13 everyone's benefit is that following the guidance does not i I~h l k/# 14 necessarily guarantee that the staff will accept the i 15 results. And I left that -- there should be a "cannot" 16 there, a typo. You can pencil that in.
17 What we were saying in this second bullet here 18 is that we expect that both the staff and the licensing 1
19 board will judge the results of elicitation or any 20 demonstration of compliance on its own merits.
21 And, as everyone knows, the DOE is going to 22 prepare a safety analysis report. And we'll produce a 23 safety evaluation report based on that information that's 24 provided in the license application.
t
'(_j) 25 However, we hope that by following the process NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASH;NGTON, D C 20005 3701 (202) 234-4433
243 1 that's laid out in the BTP for the reasons that I
,_ 2 mentioned earlier, that DOE's licensing case chould they
/ >
l
( / !
3 choose to use expert judgment as part of licensing would l 4 be bolstered.
5 (Slide) l 1
l 6 MR. LEE: And, of course, there is the SSHAC 7 report, which we have talked about in the past and in the l l
8 halls and places when the subject of expert elicitations 9 come up.
10 It's our understanding that, in addition to 11 implementing the BTP in the elicitations that DOE is l 12 undertaking right now, they're going to rely on the use of 13 the technical facilitator/ integrator as a complement to f-~s
't-) 14 the generalist that we have developed in the BTP.
15 We haven't done a detailed review of the SSHAC
'6
. report. In fact, it's my understanding that when the 17 volcanism elicitation was underway, the SSHAC report was 18 otill being documented and there were a number of 19 iterations of it.
20 It also went out to the National Academy of 21 Sciences for comment. And I understand the academy's 22 report is gcing to be published as an appendix to the 23 SSHAC, which, as I mentioned earlier, is now being 24 printed.
()
25 That being said, though, we believe that the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
244 i
1 SSHAC recommendations are generally consistent with l l
- 7. s 2 NUREG-1563. However, in our BTP, we didn't go so far as l i ) I 3 to evaluate the relative merits of the TF/I.
l 4 And one of our comments, DOE in December of 5 last year regarding the PVHA and consistency with the BTP l
6 is that we recommended that DOE begin to give ;
l 7 consideration to some guidance, internal guidance, on how 8 they would use the TF/I because the SSHi. report was 9 silent on how that person or whoever would fulfill that 10 role.
11 (Slide) 12 MR. LEE: On the next slide here, I just want 13 to talk about current staff activities, but let me get a q
\ _/ 14 drink. Thanks.
15 As Dr. Brocoum said, there's a number of 16 elicitations that are now underv. The ones that the 17 staff has been tracking are the .. .A , the unsaturated zone 18 hydrology elicitation, and the waste package canister 19 degradation elicitation.
20 Thus far, the staff haven't observed any fatal 21 flaws. We're pretty satisfied, more or less, that the BYP 22 is being implemented as we had hoped. However, as with 23 the PVHA, we do have some questions and comments that I 24 understand that the staff has been sharing with DOD in
- p).,
25 real time. And next week, we hope to have a more formal NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(2 - ) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
245 l l
1 exchange over the telephone to share with DOE what our
,, 2 specific views are.
I ') !
~
3 Generally, the views that we do have, at least l
! 4 at this stage, tie back to two of the three letters that 5 we wrote to DOE. This is the December '96 letter that I 1
6 relates to the documentation, TF/I, and basis for l l
7 updating, as well as the conflict of --
8 CHAIRMAN POMEROY: Which two of the three?
1 9 MR. LEE: The December letter.
10 CHAIRMAN POMEROY: You say here the 11 " questions" and " comments are consistent with two of the 12 three recent staff letters." i 13 MR. LEE: Right. If you go back to Slide 1
(V 14 Number 3, --
15 CHAIRMAN POMEROY: Right.
16 MR. LEE: -- the issues that we're picking up 17 on relate to the December 31 letter. We still see some 18 things there we'd like to talk to DOE about. Well, we 19 have talked to DOE about them, but we need to have a more 20 formal exchange of information as well as the January 7 21 1stter.
22 We want to pursue with them how they're going l
I 23 to deal with conflict of interest and bias.
24 CHAIRMAN POMEROY: You were aware before (j 25 December 31st, though, 1996 that a TF/I person or function l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
246 1 had been identified in the probablistic volcanic hazard l
.,, s 2 assessment? ,
)
3 MR. LEE: Yes. And we don't have a problem 4 with that. We're just anxious for DOE to develop its own 5 procedures on how that -- I'll get to that slide in a 6 minute.
7 CHAIRMAN POMEROY: Okay.
8 MR. LEE: Yes. We don't have a problem with 9 the TF/I. We think DOE needs to do what the SSHAC report 10 didn't do. So I'll just go to the next slide.
11 (Slide) 12 MR. LEE: Now that we have been talking about 13 these letters a little bit, why don't we just speak to r\ 4
'% ) 14 them briefly? In a December 26 letter, the staff i 15 recommended a path to resolution of SCA Comment Number 3 ,
l l
16 but on somewhat different grounds than those proposed by l l
17 DOE. l 18 Moreover, we proposed a basis for closing five 1
1 19 other open items. And the staff waits confirmation that l l
20 DOE agrees with the course of action described in Appendix !
l 21 E or an equivalent course of action. l 22 It's my understanding that right now DOE is l
1 23 working up a letter to kind of tie up a number of items 24 related to the PVHA technical exchange and a number of
,a
(,) 25 other interactions with the staff. After they get to that l 1
l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
247 1 letter, I understand they're going to follow up with these
,7-s. , 2 other letters, which I'll get to.
\
)
3 So that was the first letter. In December, as l
4 I pointed out, in our December letter, which is a 5 follow-on to the September technical exchange, our 6 Appendix 7 meeting, on FVHA and the BTP, there were three 7 issues that we picked up on.
8 Once again, one of the issues was 9 circumstances under which the results of an elicitation 10 would need to be reevaluated.
11 The second point was documenting changes to 12 initial elicited judgments. In our review of the PVHA, i
l 13 which I would personally comment I thought was a very j
\
\ ') 14 extensively documented product, -- there was a lot of 15 information in there, and in many respects it was easy to 16 follow what was going on -- we did pick up on just a minor 17 point that following the initial elicitations, judgments 18 were changed based on this feedback, which is kind of part 19 of the TF/I SSHAC process, wnich doesn't bother us.
20 We just need to see the basis for that. And 21 DOE pointed out that it was documented administratively 22 offline, but we think it might be useful just to include 23 some of that information in the final documentation just 24 as kind of having a nice, neat package.
,m
( ,) 25 And the third issue, which I talked about a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
248 !
1 few minutes ago, was the need to develop internal 7- 2 procedures or guidance on the use of the TF/I. In our i \
\"J 3 limited review of the SSHAC, we didn't find any l 4 information on how that person or individual would conduct l 5 their functions or perform their roles.
6 And inasmuch as DOE is going to be using this i l
7 as part of the elicitation process, we think it needs to )
1 8 have some kind of internal procedure that it can work 9 within to make up for that lack of instruction or 10 guidance.
11 (Slide) 12 MR. LEE: And the third letter, which was 13 written back in January '97, DOE requested information or TN i t
\_J 14 background with respect to how it might deal with issues 15 related to documenting and disclosing potential conflicts 16 of interest.
17 And, as I noted earlier, NUREG-1563 does not 18 prohibit the Department from using its own experts in the 19 conduct of any formal elicitation. However, the guidance 20 does recognize that the Department needs to address the 21 potential for bias and COI in any expert selection.
22 So, as I noted earlier, it's our interest to 23 make, going back to that mantra that the staff constantly 24 repeats, we think it's important for DOE as it conducts (8 ) 25 these elicitations to have some kind of procedure in place
(
w/
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
249 1 and incorporate that as part of the elicitation process so
,- 2 everyone is working on a common playing field.
\] 3 (Slide) 4 MR. LEE: So, in closing, the future staff 5 actions, the staff will encourage DOE to prepare a --
6 well, the staff is going to continue to monitor the 7 implementation of NUREG-1563. I think a lot of the staff 8 here at headquarters would like to be involved in 9 observing any and all of the elicitations. However, 10 within the limitations of our resources, we can't send 11 everyone out there.
12 So, at a minimum, I believe there's always 13 been at least one person monitoring elicitation. In some
./ ~
k_)T 14 instances I think we have had as many as three.
15 Moreover, we're going to encourage the timely 16 response to the staff's previous letters. And, as I noted 17 just a second ago, we understand that DOE's getting 18 prepared to do that. And hopefully a telephone call to 19 them next week on issues that we have just talked about 20 will encourage that.
21 And, to the extent that resources permit, I 22 think it's important to point out, though, that the staff 23 will conduct its independent work in the form of 24 investigations and analyses in those areas of the program
,,m
- i g ) 25 that it believes are the most important to the public l NEAL R. GROSS
, COURT REPORTERS AND TRANSCRIBERS
! 1323 RHODE ISLAND AVE., N W (202) 234 4433 WASHINGTON. D C. 20005 3701 (202) 234-4433 I
250 1 health and safety.
2 We think it's important for the staff to have
)
'~#
3 an independent understanding of the site, potential 4 phenomena that may affect performance. And this is l 5 consistent with the staff's strategy that will allow for 6 the early identification resolution at the staff level, 7 potential licensing issues. l 8 We think that by having our independent review I 1
9 capability, although it may be modest, it will allow the 10 staff to remain informed and knowledgeable when it 11 conducts reviews of the DOE pre-licensing and licensing 12 activities.
13 So how this relates to expert judgment is that r's G';
14 we're not going to have the staff doing independent work 15 in all areas that DOE is going to be conducting 16 elicitations, but for those areas that we think we don't 17 know enough about, we may do some limited scoping types of 18 works.
19 That's all I had. I'm ready.
20 CHAIRMAN POMEROY: Thank you, Mike. Come on.
21 We're not that bad. You know that this Committee has 22 strongly supported the issuance of the branch technical l
23 position and worked very hard, formally and informally, to 24 achieve a useful document.
p i
V j 25 I think what we're dealing with now are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
251 1 questions of implementation. So far the NRC staff has put
,-s 2 all of its concerns in a generic format. And, yet, we see
( )
RJ 3 problems in the elicitation-specific situation.
4 MR. LEE: Right.
5 CHAIRMAN POMEROY: And I guess, for example, 6 in your future staff actions, your last slide, you say 7 that the staff "will" -- this is a future action -- " meet 8 with cognizant DOE staff and identify the new 9 questions / comments / observations regarding the elicitations 10 now underway." ;
11 I guess I have two questions. One is: What 12 do you mean by "new" versus any that you have identified l 13 previously?
(~N s i
\ -) 14 MR. LEE: Okay.
15 CHAIRMAN POMEROY: In other words, are you l
l 16 referring back to the letters of December --
l 17 MR. LEE: Yes.
18 CHAIRMAN POMEROY: -- and January by that?
l l 19 MR. LEE: Right. To the best of my knowledge, i
l 20 there are no new issues that have come up subsequent to i
21 the issuance of those letters. We met internally about a 22 week and a half ago to discuss the issues that were of 23 concern to the staff.
24 And the three themes that kept on coming up p
i
%-)
1 25 were this business about documentation following feedback N EAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
252 1 sessions; how are we dealing with conflict of interest or gS 2 bias, not so much in terms of how the elicitation teams
- /
% s' 3 have been composed, but is there a recognition of 4 associations and things like that that would help to --
5 for lack of a better word, wouldn't undermine the
- 6 credibility of an elicitation after it has been done. And I 7 then the other issue was the business about the TF/I.
8 We want to see how DOE intends on implementing 9 that function given the tact that the SSHAC report didn't 10 document or provide guidance on how that function is to be 11 performed.
12 So there may be some parochial issues with 13 regard to a particular e:_icitatior , not unlike the types
/
+ s,
\'l 14 of things that we dial.ogued with DOE about last September 15 at the Appenoix 7 meeting.
16 I mean, there's that usual give and take. l
)
l 17 However, to the best of my knowledge, there aren't any '
18 showstoppers or surprises hanging out there.
19 CHAIRMAN POMEROY: I guess I have a further 20 problem. I would certainly agree with your first 21 statement here that the staff wants to " encourage a timely 22 response to" its " letters." Certainly.
23 Do you think that would apply equally, for 24 example, to the NRC staff providing some comments c.
() _
25 regarding the existing elicitations?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
253 1 MR. LEE: We're getting into PVHA.
7.s 2 CHAIRMAN POMEROY: I'm phrasing that as a kt )
~'
3 generic question. If you get a submission from an 4 elicitation that has been done over the past close to two l 5 years, you have had it a long time. Do you think there is 6 any obligation of the staff to respond to that in some 7 way?
8 That's an excellent example, for instance, of l l'
9 the way that DOE staff intends to implement the TF/I 10 concept. ;
I 11 MR. LEE: Sure. Mike Bell. )
12 CHAIRMAN POMEROY: Right. Thank you, Mike.
13 MR. BELL: Michael Bell from the NRC staff.
^ i
, s k) 14 And maybe I'm actually going to toss this to l 15 Steve, but my recollection is they transmitted the PVHA to 16 us for information, not for review and comment. And so, 17 therefore, we never planned to do a detailed review and 18 comment of the document.
i 19 CHAIRMAN POMEROY: But you do feel that if l
i 20 there were problems with that elicitation, that the staff 21 should identify those in some prompt way to the DOE?
l 22 MR. EISENBERG: This is Norman Eisenberg from 23 NMSS.
24 We did. We had an Appendix 7 meeting with DOE g
25 (v) on expert elicitation in general and especially on the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
254 1 experience in PVHA. So we did that. And we had a
,, 2 technical exchange on the issues covered in the PVHA. So
]
3 I think that's timely feedback.
4 MR. LEE: If I could just follow on to Norm, 5 subsequent to the Appendix 7 we had in September, we wrote 6 the December letter to DOE raising the three issues, which l
7 I think at the close of the Appendix 7 meeting, although l 8 these meetings aren't documented, there was a general l l
9 understanding that those are the three issues we would be 10 tracking future elicitations on. ;
11 So I think in the context of how the process 12 was being implemented, the DOE had timely insight or 13 notice as to what was causing the staff to have some
_J 14 concern.
15 In terms of the technical sufficiency or 16 technical comments with regard to the PVHA, I think that's 17 a more complex dynamic because -- correct me -- I think at 18 one time there was to be a Fall '96 technical exchange on 19 the PVHA from a technical standpoint where the staff would 20 present its views or its work and DOE would review its 21 work. But, for one reason or another, that- technical 22 exchange didn't come to fruition.
23 MR. EISENBERG: Until February. I 24 MR. LEE: Until February.
~
) 25 CHAIRMAN POMEROY: I have a lot of other x-l NEAL R. GROSS CCURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
255 1 questions, but let me ask my colleagues if they have nny 7 2 questions. Marty? George?
+ \
\
3 MSMBER HORNBERGER: Just a quick one. Steve 4 Brocoum started I think by saying he had the feeling of 5 something. So my question to you is: Do you have the 6 feeling that there is disagreement with the PVHA amongst 7 the staff?
8 MR. LEE: Well, let me quote some great 9 Americans. I think there are mixed feelings. Can I speak 10 for myself or -- I'll speak for myself.
11 MS. FEDERLINE: You can speak for anyone, 12 Mike, please.
13 (Laughter.)
/~
l s/ 14 MR. LEE- In the context of the PVHA, I think 15 some staff missed the boat in terms of what DOE was doing.
16 It's my opinion DOE was not doing a peer review of its 17 volcanism program. It was doing an elicitation to develop 18 probability and consequence estimates that it could use in 19 total system performance.
20 ; think where frictjan began was in terms of 21 -- and I'll call it a weakness, not a deficiency, but a 22 weakness of the PVRA I think could have been in improving i 23 the disclosu issues related to conflict of interest or 24 potential conflict of interest and bias. I think that
,/m
', ,) 25 would have probably made the staff feel more comfortable l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433 1
1
256 1
1 1 with what was done. ;
,y 2 However, at the same time, I think some staff l l \
1 1
'~'/ 2 3 have not recognized the fact that as a potential i
4 applicant, DOE is free to use whomever it needs to use to '
1 5 make a licensing case. I think in reactor space, I don't 6 think there has ever been a prohibition placed on a 7 licensee in terms of who they could or couldn't use as 8 experts or witnesses to make their licensing cases. And I 9 think some of the staff need to come up to that 10 recognition.
11 In terms of feeling, to answer your question, 12 I believe that we have a good feeling inasmuch as DOE has 13 done an extensive amount of work regarding probability and
,a
( )
'w / 14 consequence at Yucca Mountain. I'm not the expert in 15 this, but I understand that based on the technical 16 exchange, there's been a recognition or commitment to do 17 some more work. And I think the staff, having done its 18 independent work down at the center and here, have given 19 us a good feeling about where these numbers would lie.
20 So I think in this issue because it's very 21 contentious or potentially contentious, I think the staff 22 has benefitted from doing the work. I think in the long 23 run DOE is going to be in a better licensing position 24 because the staff has independently looked at this issue.
(a,) 25 And we're not going to have to rely on DOE's work alone.
I NEAL R. GROSS 1 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234 4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
1 257 1 So I think, really, where the rubber hits the gg 2 road is in the development of an issue resolution status t i Q/
3 report and subsequently in the development of a license 4 application review plan, which would go into the 5 acceptance criteria that the staff would use to judge the j l
6 acceptability of DOE's compliance demonstrations. This )
l 7 point may be moot, though, if PACS and FACs are expunged 1
8 from the future rulemaking for Part 60. I 9 But, all in all, I think we could say we feel 10 good because DOE did a lot of work. We've done a limited )
1 11 amount of independent work. And we're in a position now l 12 to move ahead.
l 13 MS. FEDERLINE: Mike, could I just add one
/ \
/
i N 14 thing? I 15 MR. LEE: Sure.
1 16 MS. FEDERLINE: One thing I sort of want to l 17 comment on is what the focus of the staff's pre-licensing la program has been, and that's to provide early feedback to 19 DOE and look at things in an integrated perspective.
20 We have been trying to avoid the approach of l
l 21 disaggregating the issues and putting them into little l \
l
! 22 compartments, as opposed to looking at them in an l
23 integrated fashion. And we felt it was very important to i 24 look at probability and consequences together for f%
() 25 volcanism.
HEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
258 1 And at the soon as possible opportunity when
,~s 2 we could look at those things together, I think we did
(J' s
3 have a technical exchange and laid out all of our 4 technical work to exchange with DOE.
5 We're certainly willing to have more 1
6 dic"assions with DOE, but I think it's very important to
'l put the context -- and, as a matter of fact, we agreed in 8 that technical exchange that DOE needs to do some more 9 consequence work. I 10 And at that point, we will be able to put it j l
11 in the proper perspective, which is for viability l l
12 assessment: Are there any vulnerabilities at this site l 13 that would keep us from moving forward?
ms 14 CHAIRMAN POMEROY: I don't think you would 15 find any disagreement with the fact that we think that 16 further consequence studies need to be carried out by DOE. j i
17 I guess -- )
i 18 MEMBER HINZE: I'd like to ask one more
]
1 19 question.
l 20 CHAIRMAN POMEROY: Yes. Go ahead.
21 MEMBER HINZE: Briefly and from a generic 22 viewpoint getting to this point of the number of experts, l 23 I realize that the BTP does not give a number of experts 24 required, but could you remind me of what it does say
/ \
() 25 regarding the coverage that is required, which really gets NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 2344433 WASHINGTON. D C. 20005-3701 (202) 2344433 I
1 259 1 to the point of the number?
l
,. .s 2 MR. LEE: I think my recollection is in the 3 BTP, we recommend that for any elicitation topic, you try '
l 4 to strive for a diversity of experts who are knowledgeable j 5 in the subject area who would represent a broad range of 6 opinions regarding a particular subject to be evaluated. l l
7 MEMBER HINZE: Covering the spectrum; right? l 8 MR. LEE: That's correct.
9 MEMBER HINZE: Do you and your colleagues in l 10 the NRC see any problem with the more recent expert i
11 elicitations by DOE and these are to save money, to cut 12 down on the number of experts? ]
13 MR. LEE: The opinion that I have gotten from
()
(
/ 14 the staff is with the exception of those three areas that 1
{
15 we talked about, documenting and the TF/I and things like ]
16 that, they have been pretty satisfied with how the 17 elicitations have been conducted.
i l
18 I understand, though, that they have some j 19 issues they'd like t. talk, but I don't believe that the l l
i 20 issues are tied to the number of personnel that are i 21 involved in the elicitation, more just fine points.
22 MEMBER HINZE: Thank you.
l 23 VICE CHAIRMAN GARRICK: We've said many times 24 that in a license proceeding, of course, the record is the n
( ,) 25 critical piece of evidence for granting a license. And we l NEAL R. GROSS COURT REPORTERS AND TRANSCRIDERS l 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005 3701 (202) 234 4433
260 1 have also talked about the importance of documentation on s 2 this whole elicitation process. And you have mentioned )
/ \
\'~' ]
3 that it's the desire for staff to monitor all of these 1 4 elicitations. ;
5 What are you doing specifically in the 6 documentation area to give yourself high confidence about 7 the quality and consistency of that documentation from l 8 elicitation to elicitation?
1 9 You don't have much to go on right now, but we i 10 do at least have --
11 MR. LEE: Well, I believe internally I have I
12 seen trip -- ;
l 13 MR. BELL: Well, this is Michael Bell, Dr.
/m 14 Garrick.
15 Actually, the purpose of our doing these 16 observations is to observe the process and give feedback 17 to DOE so they can document their case. It's not for the l 18 NRC staff to go and develop the record that then would 19 support the outcome of the next elicitation in the 20 licensing proceeding. Therefore, you would be in a 21 position where the staff's documentation would be used.
22 VICE CRAIRMAN GARRICK: Well, I'm thinking of 23 this in the context of guidance for that dccumentation, 24 much like the guidance provided for a safety analysis
('"i
( ,/ 25 report or what have you.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C, 20005-3701 (202) 234-4433
261 1 MR. EISENBERG: Dr. Garrick, there ia guidance 73 2 in the BTP on documentation. It's general guidance. It's
~~'
3 not a format and content guide. But I think with the few 4 exceptions that Mike has alluded to, the documentation l
5 that DOE has produced so far has been adequate. So I ,
1 6 don't know that there's a need to produce another guidance 7 document for documentation. !
1 8 VICE CHAIRMAN GARRICK: Thank you.
9 CHAIRMAN POMEROY: Can I just return to Mike 10 Bell for . ment? You'll be pleased, Mike Lee. ,
I 11 . just wanted to ask, Mike -- you know, we 12 strongly support the staff's being at all of these expert 13 elicitations. We believe that the staff should be there. l o.
('s -)
14 But I'm a little confused here perhaps.
15 In the PVHA, which has been a process that's 16 developed over a period of something like two years, where 17 we have seen the process almost from the start and we have 18 seen the use of a TF/I, for example, and, yet, we've --
19 apparently am I wrong that we have first conveyed our 20 problem with that in December of 1966?
21 MEMBER HINZE: '96.
22 CHAIRMAN POMEROY: Yes, '96. I wish it were 23 1966.
24 MR. LEE: I see Janet Kotra here. My (n) ,
25 recollection is Janet probably covered just about all of N EAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
262 1 the elicitations or all of the sessions with the except of g3 2 one. But she has a lot of corporate knowledge regarding N_,)
3 the PVHA from the staff's perspective.
4 If I could prevail upon Janet to --
5 MS. KOTRA: Dr. Pomeroy, I think I remember we 6 jointly attended one or more of those sessions. Janet 7 Kotra, Division of Waste Management staff.
8 I think it's important to recollect that both 9 the branch technical position on expert elicitation and 10 the SSHAC report were works in progress at the time PVHA 11 got underway.
12 We were informally aware. Although staff and 13 Division of Waste Management was not leading the effort
/_%
()
14 with regard to SSHAC, we were in consultation with our 15 colleagues in the Office of Research, who were cosponsors 16 of the SSHAC effort and were attempting to keep current 17 with regard to the evolution of that thinking.
18 But that was being written and the wholesale 19 being rewritten while this was going on. So it was not j 20 clear to us whether what we were observing in real time at 21 the PVHA was or was not going to mesh with what came out 1
22 in the final SSHAC report, nor was it our intention to 23 provide guidance that specifically addressed that other 24 document.
,o I i U' 25 It was to basically pull together what our .
1 i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W. 1 (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
263 1 institutional memory and familiarity with the literature 2 would suggest or good housekeeping and good practices for (7 ,)
' ~ '
3 a sound process in a very generic way. It was not to 4 provide a basis to critique a specific elicitation.
5 And I think that to the extent that we want to 6 avoid that, we want to provide general guidance. And we 7 have literally said, both before, during, and after the l
8 generation of that document, that it is not our intent to ;
1 l
9 prescribe how DOE is to make its licensing case on any 10 particular issue. l l
l 11 I don't know if that answers your question. l 12 CHAIRMAN POMEROY: Thank you, Janet.
13 Any other questions for Mike?
,m i \
V 14 (No response.) I 15 CHAIRMAN POMEROY: Thank you, Mike. We 16 appreciate it.
17 The Committee may want to consider whether we la want to write a letter with regard to this issue at some l 1
19 time during the course of this meeting.
20 I'd like to move to the next item on the 21 agenda, which, if I can find an agenda, is project 22 integrated safety assessment, PISA. l 23 MR. EISENBERG: Dr. Pomeroy?
24 CHAIRMAN POMEROY: Yes?
I t'h. !
y) 25 MR. EISENBERG: Mike Bell had asked that if we j NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISt.AND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433 i
264 1 could have an opportunity to make a few comments about
,~ .s 2 both presentations. Can we do that or are you in a big s
i 3 hurry to move forward?
4 CHAIRMAN POMEROY: No. No, I'm not. And I 5 apologize. Please proceed.
6 MR. EISENBERG: I have a few points that might 7 be worth making. Steve Brocoum indicated that there was 8 no basis provided for the staff's estimate of the 9 probability of volcanism. I think we did articulate a 10 basis f or our estimate. We published it in the peer 11 review literature. And we certainly documented it at the 12 technical exchange. So I think we did provide a basis.
13 Embedded in the PVHA were various models by i
)
C/ 14 different experts. Our approach used a somewhat different 15 model with different assumptions. But all of that was I 16 think clearly articulated, especially at the technical 17 exchange. That was the purpose.
18 We did agree on the upper bound. And this is 19 an important point. As Dr. Garrick has said, -- I've j 20 heard him say it on more than one occasion -- agreement by 21 experts on the value is not needed if the agreement on the l
22 upper bound demonstrates that risk is acceptable. And 23 that's the case here. So we shouldn't minimize the degree 24 of agreement between the two groups .
(~\
,i 25 We clearly said that DOE could use their NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D 0 20005-3701 (202) 234-4433
265 i 1 expert elicitation results in making their case. I don't
,, 2 think we have decided on our position in licensing.
/ \
However, our current view is sufficient to resolve the
~
3 4 volcanism issue for us. And that's what our goal was. As 5 Margaret said, in our pre-licensing mode, we're trying to 6 give timely feedback and early warning on vulnerability.
7 CHAIRMAN POMEROY: Can I interrupt you there 8 for just a second?
9 MR. EISENBERG: Sure.
10 CHAIRMAN POMEROY: There's something there 11 that is somewhat- elusive to me. If DOE had agreed to use 12 the 10 number, it seems to me that they've said, "All 13 right. Our elicit.ation isn' t worth anything nor will any O 14 of our future elicitations be worth anything." It's not 15 worth exposing the range of uncertainty if indeed we're 16 going to use an upper bound number.
17 Why not just have the NRC pick a number and go la ahead? Isn't that a result that one could derive from an 19 acceptance of the 10'7 number from a DOE perspective?
20 MR. EISENBERG: I think it's an attitude. I 21 think it's a misplaced attitude.
22 CHAIRMAN POMEROY: A feeling.
23 (Laughter.)
24 MR. EISENBERG: Clearly the licensee can make
(-
Q 25 the best case that it wants to. And I should point out NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 FlHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
266 1 another one of my points is that the goal of what the 2 expert elicitation was, which was to bound the range of a,,x 1 t, <
3 views in the scientific community and to quantify with 4 subjective probabilities what those views are, that was 5 the goal of the expert elicitation.
6 The work that the center and the NRC staff did 7 was to come up with our estimate of the probability.
8 That's a different objective. And it's not a surprise to 9 me at all that our range was smaller and the numbers are 10 not quite coincident.
11 No, I don't think that's a reasonable thing 12 for the licensee to conclude that they should take the 10-7 13 value, nor is that what the staff suggested. What we
(\
i
's >i 14 suggested was they could use their entire range in their 15 performance assessment.
16 The NRC staff, as it typically does, would 17 take a reasonable conservative upper bound or some 18 parameters and do its assessment. And if our assessment j 19 bounded or if our estimate of performance bounded theirs, 20 that would give us a lot of confidence that they were on
1 the mark in their 2acessment.
l 22 And I thought that's the way the system worked 23 a lot of the time. So I'm not sure what the problem is.
24 CHAIRMAN POMEROY: It certainly is the way,
,y
() ,
25 but you haven't addressed the problem of offering to close NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
267 3 out the issue at the staff level, granting all of the gm3 2 caveats that go with that if DOE accepted the 10-7 number.
tv] 3 MR. EISENBERG: Well, I think that we're 4 moving towards closing out the issue whether DOE accepts 5 the 10-' number or not . I think what our conclusion is 6 based on our estimates of consequences and our estimates 7 of probability is that it's not a significant safety issue 8 absent any new information.
9 CHAIRMAN POMEROY: Right, although I have 10 heard new information in the past few days, rumors 11 thereof.
12 Do you want to go on with your other points?
13 MR, EISENBERG: Okay. We state clearly in the
(~\
\
N-) 14 BTP and I'll emphasize again following the expert I
1 15 elicitation procedure doesn't guarantee the result any j 16 more than following, say, QA procedures guarantees that 17 you'll get a license. It's merely a procedural approach. i i
18 The licensing board is going to decide the l l
1 j 19 value and licensing of the expert elicitation of the
- 20 staff's views if they happen to be different from the 21 expert elicitation and whatever intervenor views might be l
22 brought to the table. So that's where the value is going 23 to be attached to the expert elicitation. i I
24 Just because we disagreed in this one instance l gx I
( ,) 25 doesn't mean that we want DOE to go to the 99th percentile NEAL R. GIUDSS i COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE., N W. l (202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433 )
268 1 in all their distributions of all thei. .rameters. I 9 think that that is not correct and certainly not the
[. )
~'
3 position of the staff. So I certainly wanted to clarify 4 that point, too.
5 Did I leave anything out?
6 CHAIRMAN POMEROY: Very good. Thank you, 7 Norm.
8 Do our colleagues have any questions for Norm?
9 MEMBER HINZE: Well, I might support one of i
10 Norm's point, and that is regarding the peer reviewing. I I 11 happen to be knowledgeable about the pere reviewing of i
12 that article, and that had six of the toughest referees )
13 available, including the head of the DOE's volcanic s-
\m) 14 activity program, Bruce Grower, at the time.
15 MEMBER HORNBERGER: I heard the editor was a 16 cream puff, though.
17 (Laughter.)
18 CHAIRMAN POMEROY: Okay.
19 MR. EISENBERG: Thank you, Dr. Pomeroy.
20 CHAIRMAN POMEROY: Thank you. Thank you, 21 Mike. And thank you, everybody.
22 Let's now, then -- Carol is waiting eagerly in 23 the wings here someplace, I believe. Carol Hanlon from 24 the DOE is going to talk to us on projected integrated l"
(v)) 25 safety assessment.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISiAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 '1 2) 234-4433
269 1 Carol, it's a great pleasure to have you here.
l
-s 2 We really are a group of cream puffs, as you know. And we r i
( "' /
3 really look forward to hearing your presentation.
4 MS. HANLON: Thank you, Dr. Pomeroy.
5 (Slide) 6 PROJECT INTEGRATED SAFETY ASSESSMENT 7 MS. HANLON: I am Carol Hanlon with the 8 Department of Energy Yucca Mountain Site Characterization 9 Office. And it is my pleasure to be able to be here today 1
10 to speak with you on the subject of the project integrated 11 safety assessment, just to show you I do have a cover 12 slide.
13 (Slide) t l
(/ 14 MS. HANLON: The purpose of my presentation 15 today is to introduce you to the project integrated safety 16 assessment. A version of this presentation was briefed to 17 the NRC managers on April 30th, 1997 at one of our 18 bimonthly DOE-NRC managers' meeting. We also briefed this 19 information to our own project manager on May 5th. And we 20 have made some conside"ations and some modificationa to 21 our approach in light of that.
22 I wanted to let you know that we intend to 23 complete the management plan, including the content guide, 24 in May and deliver it, distribute it, for information to
/~N 25 both the ACNW and its staff and the NRC sometime during
( -)
s_.
NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
1 270 1 the month of June.
2 (Slide)
?~'\i L
3 MS. HANLON: A little bit of background on the 4 project integrated safety assessment. You will all recall l
l 5 that in Fiscal Year '96 we were doing a great amount of 6 comprehensive replanning in response to budget constraints 7 we were all facing.
8 As part of that, as part of this replanning 9 effort, the Civilian Radioactive Waste Management Program i 10 plan was reviced to streamline and focus the program. And 11 in this streamlining and focusing the program, we felt ;
I 12 that the development of an integrated safety assessment i l
13 was considered essential. Therefore, the assumptions for ,
(7, l V 14 this safety assessment were built into our multi-year ;
1 15 planning base and carried through the next several years.
16 One of the reasons this became important was 17 that at the time we were looking at various things that 18 were important. Site description was important. It was 19 important to design. It was important to PA. It was 20 important to environmental health and safety.
21 Designs were important and interactive with 22 site description and PA. The envircnmental considerations 23 were important in the public health and safety, but they 24 were all on independent time frames, sometimes one waiting
(% ,/
) 25 for another. And we felt that the best way for efficiency NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON. D C 20005-3701 (202) 234-4433
i 271 1 in streamlining and to also focus what we were doing was 2 to integrate it in a single safety assessment.
(~\ 5
,%)
3 (Slide) !
4 MS. HANLON: You may have noted or will soon 5 note that I have changed the order of the presentation a ;
i 6 little bit. I think it won't throw you off.
1 7 So what, after all, was the purpose of the !
l 8 project integrated safety assessment? Again, it was to i 9 integrate the ef fort of the scier:tific programs, the i
10 design, performance assessment, and the health and safety I i
11 effort to assemble and begin to evaluate our safety case, l l
12 to assess the information that was currently available, i 13 and identify strengths and weaknesses.
i n
')
- 14 (Slide) ,
l 15 MS. HANLON: What is the project integrated 16 safety assessment? It is not, as Andy has mentioned to me 17 previously, Italian for license application. It is our 18 internal vehicle for presenting integrated information 19 currently available on the results of our scientific 20 programs, design, performance assessment, and health and
, 21 safety. It's not a part of the viability assessment, l
I 22 although i'll mention one small exception to that later.
23 However, sections of the PISA will be 24 available in the same time frame as the viability
()
25 assessment to provide support and additional information NEAL R. GROSS COURT REPORTERS AND TRANSCRIS n3 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASH!NGTON, D C. 20005-3701 (202) 234-4433
272 1 if it's appropriate.
- < 3 2 Later on the PISA will support the
(,]
3 environmental impact statement and the site recommendation 4 process. Here again I'd like to clarify the slide.
5 The PISA may be used to initiate preliminary 6 sufficiency comments, and that will be in conjunction with 7 information exchanges with the Commission staff. That's a 8 process that we haven't entirely defined yet. The 9 sufficiency comments arise from Section 114 of the Nuclear 10 Waste Policy Act. We haven't exactly clarified or defined 11 that process, but we expect that the PISA will provide 12 information for it.
13 MEMBER HINZE: Carol, if it's not premature
[
x /
N' 14 for me to ask, what is the relationship of the PISA to the 15 synthesis reports?
l 16 MS. HANLON: I'm not sure there is a direct i l
17 relation to it. The information that might previously I l
18 have been put in the synthesis reports will now be 19 basically contained in Chapter 2, either summarized and 20 presented in Chapter 2 or the references will be 21 summarized in Chapter 2.
22 So I guess, although I don't see a direct 23 causal link, in essence, you might say that they replace 24 the synthesis reports
,o k_,) 25 MEMBER HINZE: Is there any cutback on the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
273 1 number of synthesis reports that are going to be made
,. m 2 available?
i
)
3 MS. HANLON: I think that we don't any longer 4 plan to do synthesis reports. We are providing that 5 information in Chapter 2, the site characteristics, and in 6 the references that will support site characteristics.
7 And the first appearance of this will be in the PISA. I 1
8 MEMBER HINZE: We heard from the technical l
9 exchange on the igneous activity issue that DOE was 10 planning a synthesis report on igneous activity to be 11 released before the end of the fiscal year. Is that still i
12 online?
13 MS. HANLON: Let me defer that to Steve.
('~'\I
'V 14 MR. BROCOUM: Stephan Brocoum.
15 Yes, we have a report on volcanism coming in 16 the end of this fiscal year.
17 MS. HANLON: Does that take care of your 18 questions, Dr. Hinze?
19 MEMBER HINZE: Fine. Thank you.
20 MS. HANLON: So in evaluating how we would put 21 the safety case together, we looked at a number of ways we 22 might present the information. And what we have come up 1 l
l 23 with is an 11-chapter outline.
24 (Slide) ,
l (a) 25 MS. HANLON: That 11-chapter outline is the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 1
274 1 development that we feel allows for presentation of a
,_ 2 concise -- and sometimes we laugh about concise because
\
'~') 3 concise may be about this big -- but a concise, coherent, 4 and integrated safety case.
5 Basically, the elements of our 11-chapter 6 outline correspond to the elements of the safety analysis 7 reports, which have been successfully prepared by the 8 industry over the last 20 or so years. In looking at it, 9 we give strong consideration to the format and content 10 guide for a high-level waste repository application and we 11 gave strong consideration to 10 CFR 60.21 and the 12 considerations there. And we actually looked at different 13 ways we might order and organize this information. So the
(-) 14 outline we've developed we feel does the best for us in 15 presenting a coherent case.
16 And I'd like to go into this a little bit for 17 you just to give you a feeling of what the chapters do 18 indeed include.
19 (Slide) 20 MS. HANLON: First is the introduction and 21 general description. In addition to being an introduction 22 and a general description of the document itself, it will 23 include, per the request of our program manager, Mr.
24 Barnes, a summary of conclusions. And we're going to put (n) 25 that in a separate executive summary.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
- 275 l 1 So, from the rest of the document, the
- 2 additional chapters, where we find indications of '
t < 1 3 compliance or conclusions, we will capture those in one l
4 place in the executive summary.
l 5 Chapter 2 is kind of of interest. It's a road i
1 6 map to the rest of a project integrated safety assessment.
l 7 And we feel that this is really a pretty important part of 1
8 the document.
l l
9 It's patterned generally after Chapter 3 of 10 Reg Guide 170, that chapter being the design of 11 structures, components, equipment, and systems. And 12 that's the reg guide for the standard format and content l 13 guide for safety analysis for nuclear power plants. l A
- i
\_) 14 We viewed that analogous document as providing 15 a road map working through this document, understanding 16 why certain decisions were made, giving a synopsis. And 17 so it will contain an overall purpose and summary and 18 conclusions. It will talk about conformance of the 19 natural barriers and systems to general design criteria, 20 10 CFR 60, Subpart E. It will contain a discussion, a 21 summary discussion, of waste containment and isolation 22 hypotheses. And it will discuss systems, structures, and 23 components, including their classification and their 24 treatment.
m
( 25 So Chapter 2 is getting to be a pretty NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
276 1 important part of our document. We're put a lot of stress
, ~s 2 on it. And I guess I'm stressing it a little here because i \
\/
3 very often when we have a large document that's circulated 4 for review it's split up and it's given to different 5 authors.
6 An author gets one section on repository 7 design. He might not be aware that much of the 8 information he needs is either in the engineered barrier 9 system or in the waste package or perhaps even in the j 10 design description or pre-closure performance.
11 So the intent of this chapter really is going i 12 to be a road map. And hopefully when people look at it, .
l 13 whenever it's split up, a chapter that will go with it is )
\- / 14 not only the introductory chapter, Chapter 1, but l
15 definitely this Chapter 2. So as you're reading it, 16 please always ask yourself the question: Where is my 17 Chapter 2?
18 We originally made it Chapter 3 to be 19 analogous to the safety report, but because it does 20 pertain to the rest of the document and it is a road map, 21 it talks about natural systems as well as engineered 22 barriers, we moved it back. That's one change you'll 23 note. If you have received copies of my presentation from 24 the NRC managers' meeting, you'll note that we have
/~T
(,) 25 changed that.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., H W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
277 1 (Slide)
,. m 2 MS. HANLON: " Site Characteristics," Chapter (O 3 3, is also an interesting chapter because it's really 4 composed of seven subchapters: geography and demography, 5 nearby industrial transportation and military facilities, 6 climatological and meteorological systems, hydrologic 7 systems, geochemical systems, and integrated natural 8 systems.
9 The thing I think is of importance with this 10 is that geography comes first. As a physical geographer 11 myself, I think that's a very important section.
12 Our design chapters are the repository design, 13 waste package design, and engineered barrier design
< ,,T
\ )
v' 14 system. Also, the engineered system falls over into new 1
15 Chapter 7, " Radiological Safety Assessment." i i
I 16 That's one area I might point out where even i 1
l 17 development of the PISA content guide has been very 18 important to us in integrating our effort, creating l
1 19 understanding, clarifying our roles. 1 20 That was originally considered to be basically 21 an adjunct of the performance assessment, and the first i
22 cut of it was done by our performance assessment people.
1 23 But because it includes, among other things, doses to i
24 workers and because design basis events, any normal l
,m (j)
/
25 emissions are not now considered design basis, one event NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
l 278 !
I falls more into the category now of systems engineering I
73 2 and other engineering. So we have integrated between i )
3 those sections, created I think some clear understanding, 4 and now moved it more into the systems.
5 Chapter 8 is the one basically exception. I'm 6 not sure " exception" is the right term, but that's a 7 strong correlation with the VA. In actuality, Chapter 8 8 is the TSPA-VA. When Abe is done with this peer review 9 procecs and the whole process and feeds it off to the VA, 10 he'll also drop it into our document. So that chapter has l 11 a little bit different treatment. ,
1 i
l 12 Chapter 9, " Radioactive Waste Management,"
13 deals with how we're managing the different waste streams.
N. 14 Chapter 10 is " Radiation Protection" to the 15 worker. And 11 is " Conduct of Ops." So that may be more l 16 than you ever thought you wanted to know about this 17 outline.
18 MEMBER HINZE: Help me. Where is the geology )
19 now?
20 MS. HANLON: It's in Chapter 3, Dr. Hinze.
21 MEMBER HINZE: Where are the geological 22 characteristics of the site?
23 MS. HANLON: It's been subsumed by physical 24 geography.
25 VICE CHAIRMAN GARRICK: Carol thought of that.
%-)
NEAL R. GROSS COURT REPORTERS AND TRANSCRitlERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
279 i 1 MS. HANLON: I thought of that. j l
2 It's under Chapter 3, " Site Characteristics,"
\ l
'# \
3 and it's broken down into geography. I think there's an 4 error on this slide. I actually think this slide -- and I 5 can check for you. I think this slide ought to have ,
l 6 geology. 2.3. You caught me. I l
7 There is an error on this slide, Dr. Hinze, l
8 It should have been the third. ;
1 l
9 MEMBER HINZE: Okay. Thank you very much.
10 So there, George.
31 MS. HANLON: That's what I love about this 12 group is your careful reading.
13 MEMBER HINZE: We stay awake occasionally, l
(~'N i l I ;
V,/ 14 too. l 15 (Slide) l 16 MS. HANLON- I wanted to call this topic to 17 your attention. It doesn't exactly fit in anywhere in my 18 presentation. So I decided to put it in here.
19 In order to provide guidance for preparing the 20 project integrated safety assessment, we have developed l 21 the PISA management plan. And that management plan will i
22 discuss the overall process of preparing the project 23 integrated safety assessment. It includes the content 24 guide, the writers' guide. And of most importance to us im
! 25 here is the comment guide, the content guide, which
\.s)
NEAL R. GROSS COURT REPORTERS AND TRANSCR:BERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005 3701 (202) 234-4433
280 1 defines the contents for the various chapters and l l
2 sections.
\~~'l \
3 And, again, in developing this content guide, 4 we had some really specific intents in mind. We wanted to 5 make sure that the document gave clear guidance to 6 preparers of the PISA. We wanted to make sure that the 7 information was presented in a focused manner with clear 8 purpose.
9 This is where my people got sick of me asking 10 them, "So what? What is significant about this? Why are 11 you telling me this? Why would the reader want to care 12 about this? What's your purpose?" So over and over again 13 we worked very hard to get into the content guide.
,m i \
\/ 14 As directions for the author, the "So what?
l 15 What does this mean?"; we want guidance on specific l
16 details. We don't want just the vague generalities or 17 topic lists. But, in fact, we want to furnish the l l
18 guidance on what should be included and why it's 1
I 19 important. We want the relevance of the information both 20 to later on design discussions and system performance to 21 be established.
22 And that's especially true in the new Chapter 23 3 because that provides the physical description that does 24 feed our designs and our performance assessment basis of (y) 25 our process models.
v NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
l i
281 1 Another big "So what?" was the indications of I
7m 2 compliance. What this means is that we want to give i; \
e 3 indications of how what we currently know. In light of 4 what ve currently know, we feel we're able to comply with 5 applicable regulatory requirements. But so what? We're 6 telling you about this. In light of what we know, in 7 light of what we studied, what does this mean to us?
8 This does not mean that the PISA is going to 9 contain the details of the compliance case. That will 10 come later in the license application. And, just to give 11 you kind of a feeling for the level of compliance here, 12 when asked by our authors "Do you mean a couple of 13 sentences? Do you mean a topical report? Do you mean
,m I )
L/ 14 pages or paragraphs?"; we said we mean in the paragraph i
15 range, but we felt it was important as our authors began 16 to accumulate this information, assimilate it, and write 17 it up that they were thinking about the ultimate outcome l l
l 18 that it would be used for. So the document will have i
19 these indications of compliance. l i
20 We wanted to ensure that there were i 21 correlations between the individual chapters and sections 22 of the PISA and the document as a whole. Again, we wanted I
23 our authors to be aware of the logic of the entire 24 document and the interrelationship among various chapters q) 25 so that if something was needed in a site description for NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE , N W (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 282 1 a design chapter or performance assessment, they made sure 2 that it was there. And a follow-on to that was inclusion L_) l 3 of thorough cross-references among chapters and sections.
4 (Slide) l l
S MS. HANLON: A number of other things existed 6 at the time we were preparing this PISA content guide, and !
1 1
7 we wanted to make sure that we were aware of appropriate l 8 things. So the framework for the PISA will include 9 definitely a discussion of the waste containment and !
10 1 solation hypotheses.
11 The PISA will also include a consideration of I 12 NRC's key technical issues. They're being built into the 13 framework. Abe mentioned earlier that both of these ;
- \ l
- /
14 things, the waste containment and isolation strategy 15 hypotheses and the key technical issues, would be included I l
l 16 in Chapter 8 or TSPA-VA.
l I
17 They will also in the PISA be included in the 18 specific site description or design chapters. Where it's 19 appropriate that they be handled, they're being handled in 20 the body of the document.
21 The considerations that we're able to give to 22 the key technical issues will be contingent to some degree 23 on the information we have been able to get from the issue 24 resolution status reports. And, of course, our g
(_) 25 consideration of these key technical issues will evolve NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
283 1 with the issue resolution status report.
.- s 2 Again, a key correlation that we wanted to l l \ l
\"' l l
3 include was with the NRC's format and content guide, with i l
l 4 LA review plan, with 10 CFR 60.21. And, in fact, in 1 5 Chapter 1, when the document comes out, you will see a l 6 matrix that correlates the contents of the PISA with 7 appropriate contents of these other documents.
8 So there will be two, one taking the contents 9 from the PISA to the format and content guide and one 1 10 doing the reverse format and content guide to the PISA, to l
11 give us two cross-checks. And where things are not l 1
12 appropriate for the PISA, we'll identify where in the l 13 license application it will be handled. So we wanted to
,~\
V 14 make sure we were correlating as appropriate.
15 (Slide) 16 MS. HANLON: By now you must all have asked 17 yourselves the question: What has this got to do with the 18 license application? The project integrated safety 19 assessment will provide the basis for the safety analysis 20 report to be included in the license application. l 21 The license application will include vastly, 22 probably expanded detail and substantiation. But the PISA I
23 content guide is being incorporated into the technical 24 guidance document for the license application.
,c (q ,'
) 25 (Slide)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
284 1 MS. HANLON: Just a clarification here. The 2 technical guidance document really is the management plan
('" )
3 for the license. And I think it was probably a good touch 4 that they didn't call it a management plan because then 5 we'd have about 30 management plans. So now you have the 6 management plan for the PISA and the technical guidance 7 document for the LA.
8 Again, we're working very closely with the 9 people that are producing that to ensure that what we're 10 doing with the PISA is entirely compatible and makes a 11 very even segue into the license application. And 12 comments we receive on the PISA will be reflected as we 13 develop the license application. l 4
'V 14 (Slide) 15 MS. HANLON: Just to note for you additional 16 chapters that will be included in the license application 17 but not the PISA. Of course, it will be an overall 18 introduction. There will be a thorough discussion of 19 quality assurance. There will be the performance 20 confirmation program, land ownership / control, 21 organizational structure, and the assessment of compliance 22 with 10 CFR 60.
23 If I have eliminated anything.significant, 24 please call it to my attention. It was not purposeful.
/~N (Slic e)
( ) 25 L/
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4 0 3
285 1 MS. HANLON: I wanted to also share with you 7, 2 our schedule for preparing the PISA. I started with June
( '
}
3 '97, when we will distribute the management plan with the 4 content guide.
5 And then I want to walk us back to December 6 1996 because in December 1996 based on a preliminary 7 content guide, we did begin developing Chapter 3. And, in 8 fact, we began developing geology and hydrology. So those 9 have been ongoing since 1996.
10 CHAIRMAN POMEROY: That covers the basis, i
11 MEMBER HINZE: What did you do the next five 12 minutes?
13 MS. HANLON: It took longer than five minutes, I i
(_/ 14 actually. l 15 In October 1997, we will in earnest begin 16 development of the remaining ten chapters. By early 17 Spring of 1998, the M&O will have a comprehensive review 18 of the individual chapters and sections prior to sending 19 that to DOE, where we will again have a DOE management 20 integrative, intensive, comprehensive review late in the 21 spring. That will happen in the late Spring 1998.
22 In Summer of 1998, we will consolidate the 23 document. That's a little more than a year away. We are 24 exploring the possibilities of how we'll get that document I j 25 out and distribute it. We're exploring the possibilities NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
- 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
286 1 of printing a certain number and perhaps even making it
,s 2 electronically available.
i )
v 3 The Fall 1998, we intend to complete the PISA 4 effort. And this is a point of information. The PISA 5 will not be iterative. It will be concluded 2n the Fall 6 of 1998. And in Fiscal year 1999, we'll begin preparing 7 the license application. That letc you know a little bit t
8 of the schedule.
9 (Slide) 10 MS. RAFiON: Just in conclusion, for us the 11 PISA effort has already proved to be an important 12 integrating and clarifying tool, as I mentioned earlier.
13 We expect it is going to be valuable to us in defining our
( )
\.J 14 safety case and in shaping the license application.
15 We feel it's going to be an internal and it's 5
16 an important vehicle, tool of communication. And we feel 17 it will afford opportunity for an early view of DOE's 18 strategy for approaching the license application.
19 May I answer any questions?
20 CHAIRMAN POMEROY: Thank you, Carol.
23 Questions from my colleagues? Marty?
22 MR. STEINDLER: If you're going to this much l
25 trouble, why don't you write the license application PSAR !
l l
24 in the first place? You're almost there.
p 3
(
w/ 25 MS. HANLON: Steve?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS I
1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
287 1 (Laughter.)
q 2 MS. HANLON: But I had a great answer.
\~) MR. BROCOUM:
3 That question gives me an 4 opening because I have to say something. As you can see, 5 Carol has done a lot of work planning this out. And the 6 way this came about was when we shut down our licensing 7 activities that year that they cut $85 million of our 8 budget, we invented a PISA because it was our stealth 9 application. We weren't allowed to work there. So we 10 invented a PISA so we could work on the side.
11 Now, last week I was briefing Lake right after 12 he gave his testimony to the Commission. And I was going la over all of the products we're expecting to get out next
('-
'l 14 year. We have the PISA, and we have the viability 15 assessment, which consists of four major products. That's 16 five major products, all coming out more or less the same 17 time. All have to be integrated. All have to fit 18 together, all coming out the ad of Fiscal Year '98.
19 Lake Barrett, who is the Acting Director, got 20 very nervous because his focus is now on the VA. He's 21 under a lot of pressure to get that VA out, and he's 22 nervous that we've taken on more than we can handle.
23 So I have promised Lake that as we do our 24 planning, our detailed planning this summer. if we cannot
'p q,) 25 do it all or having trouble, the PISA has t,e lowest NEAL R. GROSS CO'JRT REPOmERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1
! 288 1 priority of all of these products.
( ,, 2 So Carol has done a great job. We hope we can
( \
3 get it out because I think it will be a very valuable i
4 document to the staff and to the world. It will help us S get our safety case togetht:r. But in the overall scheme 1
6 of things, the viability assessment has the highest j l
'l priority. I l
8 So, in answering your question, Dr. Steindler, 1
9 we don't want to add more to the PISA at this point in 10 time because our plate is very, very full because we have 11 these five major documents all coming out within the month !
l l
12 basically. l 13 Thank you, ip)
V 14 MR. STEINDLER: Okay. I was kind of hoping 1
15 for a technically based answer, rather than a politically 16 based answer, but I understand what you're saying. Let me 17 ask a couple of other questions.
18 Who at the moment in the absence of this clear 19 and direct nexus to the license application is the 20 audience, the expected audience, for the PISA?
21 MS. HANLON: It's an internal document. It's 22 intended for us to use in putting our case together and 23 evaluating it and seeing where our strengths are, where 24 our weaknesses might be.
/m 25 And I just want to clarify that I feel that
( s) s.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 289 1 there is a very strong nexus between this document and the i
2 license application. This document, the PISA, will go i i \ \
3 exactly into the outline for the LA and become the safety 4 analysis report.
5 I think I alluded to that' fact when I said we i 1
1 6 were working very closely with the people doing the 7 technical guidance document. They will take these 8 chapters. They will be the chapters for the safety l 9 analysis report. And if they were to find we've asked 20 them to review it and if they would have seen any huge 11 problems, they would have said, "This is not going to i
l 12 work. You need to modify it." We have not heard that 13 from them, and they are well on their way to this. I n 4 i '
\v') 14 So we have tried to ensure that there is a 15 definite nexus. i 16 MR. STEINDLER: Do you intend to produce a 17 document which is fairly complete under technical 18 references, unc'er justification for the conclusions that 19 you come to, safety-related or otherwise?
20 MS. HANLON: Yes.
21 MR. STEINDLER: Okay.
22 MS. HANLON: But, again, it won't have the 23 same degree --
24 MR. STEINDLER: I can see why Lake Barrett was A
I,v) 25 nervous. I understand that.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
902) 234 4433 WASH lNGTON. D C. 20005-3701 (202) 234-4433
, 290 1 MS. HANLON: Just to add to that, Dr.
l ,- 2 Steindler, I just want to make it clear that it won't have
( )
3 the same level of conclusions that you would expect to see 4 in a license application. That's why we just went through 5 another internal integrating effort to make sure that it's 6 indications of compliance, it's not the compliance case.
7 That was try point about paragraphs, rather than topical 8 reports.
9 MR. STEINDLER: Well, it's up to you to decide 10 whether or not you' re going to have to revisit the whole 11 thing again when you finally get to the license 12 application stage.
13 But I can tell you my estimate is that if you
,m
_) 14 run a document out that is r. t fairly comprehensive in 15 technical detail on the first try, which is what the PISA 16 sounds like it's going to be, you're going to have to 17 revisit that whole thing again when you finally get the 18 license application because I assume the licensing board 19 is going to demand a fairly comprehen,ive --
20 MS. HANLON: Absolutely. And that's what 21 we're thinking of, and that's exactly why the PISA fits so 22 closely with the license application. We're creating a 23 framework, and we're trying to make sure.
24 In my mind, I always think about pigeon holes.
,m (w)h 25 We're creating the framework, and we're making sure that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C, 20005-3701 (202) 234-4433
l 291 i i
l dll the things that will ultimately have to be considered
. - 2 are considered. And perhaps at the present time when we l l
l (m) 3 discuss an issue, all the information may not be. And we l
I l
4 may have additional information the next couple of years. ;
1 5 We may have more things that we want to say.
6 We'll also have the acceptance criteria that NRC is going l
7 to give us if they do a license review plan or if they do 8 their issue resolution status reports. We'll have more I
9 information. 1 10 But we want to make sure as we build the PISA 1
11 that we haven't left out major things that pertain to our 1
12 safety case. So I view it as going back and revisiting l 1
l 13 this framework and putting additional infmrmation where l rN l ks] 14 it's appropriate. And we realize that we will have to do I I
15 that. I 16 MR. STEINDLER: One other question. Are you 1
17 going to have a sufficient amount of detail in, for l i
18 example -- I've forgotten which chapter -- 3, I guess, for i
19 example, on the EBS?
20 I'm reminded that Jack Bailey gave us a 21 diagram today that has associated with it not only 22 uncertainties but all kinds of alternatives. And there 23 wasn't any indication that there was going to be a focus 24 in of what exactly EBS is going to look like. If you're
/~'i 4,
, ) 25 going to do a safety analysis or a safety-related report, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
l i 292 l
1 you need some kind of detail.
g3 2 MS. HANLON: EBS is Chapter 6, I believe.
I, !
v ,
3 Yes? !
l 4 MR. STEINDLER: Whatever, yes. Undoubtedly, S right, j l
6 MS. HANLON: And Chapter 3 -- l 7 MR. STEINDLER: Sorry about chapter -- I meant )
l 8 Chapter 6, then.
l 9 MS. HANLON: Yes. I i
10 MR. STEINDLER: There is going to be enough 11 detail so that all of those unresolved issues that we 1
1 12 heard about this morning -- l 13 MS . 'HANLON : Actually, i'm fairly confident in l iv) 14 Chapters 4, 5, and 6 because Jack has taken a very l
j l
.i. 5 personal interest in it and paid a lot of personal I
16 attention.
1 17 So as we begin getting copies out of this, I i 1
18 have to say that Jack's engineering chapters and Chapter 8 19 were those chapters that made me feel most comfortable.
20 So of any of the areas I can say that I feel comfortable 21 in, I have a strong feeling that the engineered barrier I 22 system will be adequate.
23 Again, we may deline it later, but for a first 24 cut, I have strong confidence in it.
()
,fM 25 MR. STEINDLER: So there will be enough.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
293 1 Okay.
, 2 MS. HANLON: Jack?
( ')
3 MR. BAILEY: May I say something? Thiv 4 s 4 Jack Bailey.
5 MS. HANLON: Don't fail me now, Jack.
6 (Laughter.)
7 MR. BAILEY: I'll do my best, Carol.
8 Clearly we won't have everything evaluated.
9 As I said this morning, the chart that you're going to see lo is what we expect to have for documentation and thought 11 processes by the end of physical year '98. In order to 12 satisfy the needs of PA and again the documentation 13 process, I pretty much have to cut off design physical rs
, / )
V 14 year '97. I l
15 So what you're going to see in the PISA is 16 that which we selected for our base case and those items 17 which we added on that we felt we needed for the treatment j l
18 of uncertainties.
l 19 So you're going to see a limited case based on l
20 what the status of design, the status of science, the 21 status of knowledge were basically midsummer of this year 22 to the end of the physical year. That's what will be 23 incorporated in there.
24 And when we get to the license application, rx
) 25 you'll see the full treatment of all of that.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
294 1 MR. STEINDLER: Someone is going to discuss
.s 2 compliance and the indica'.. ion of compliance of that zero
( )
3 order baseline design that you're going to put in the 4 PISA?
5 MR. BAILEY- We're going to have a discussion 6 of that based upon the knowledge that we have at that 7 time, yes.
8 MR. STEINDLER: All right, And just a 9 comment. I thought it was very clever of you to move from 10 the license application to the PISA. I would congratulate 11 you on a good move, especially if you can pull it off 12 budgetarily, I mean. I know you can pull it off 13 technically.
m
( )
'J . 14 CHAIRMAN POMEROY: Okay. Maybe George is 15 going to pass.
16 MEMBER HORNBERGER: I'll pass.
I 17 CHAIRMAN POMEROY: Okay. John? l 18 VICE CHAIRMAN GARRICK: I was just going to 19 ask: Is there going to be a little PISA for public
! 20 consumption or maybe I should say for human consumption? I l
l 21 MS. HANLON: I think it's going to be Chapter 22 1, Dr. Garrick. That's intended not only to be the 23 introduction to the whole document but an executive 24 summary.
( ) 25 And that's what Mr. Barnes was looking at when
%J NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 295 l
i 1 he wanted to make sure that we had a chapter that
, 2 identified conclusions and gave really the more friendly
< s
(" )
3 view. j 4 Rather than making that a separate chapter, we l 5 wanted to make that an executive summary part of Chapter i 1
6 1.
I 7 VICE CHAIRMAN GARRICK: This Jooks like a 8 monster. You talk about dropping in the performance j l
9 assessment. l 10 MS. HANLON: Yes. That's going to be 5,000 11 pages itself, isn't it?
12 VICE CHAIRMAN GARRICK: That's like a l
13 1,200-pound gorilla on a bassinet. I
?~)
Cl 14 MS. HANLON: But it will be concise.
i l
15 VICE CHAIRMAN GARRICK: Okay. Thank you. l 16 MEMBER HINZE: No. No, thank you.
17 CHAIRMAN POMEROY: I have no questions either, 18 Carol, but let me compliment you on a great presentation.
19 It was very good.
20 MS. HANLON: Thanks, Dr. Pomeroy.
21 CHAIRMAN POMEROY: With that, I would like to 22 take a 13-minute break and reconvene at 15 after the hour.
23 And I'll be here this time.
24 (Whereupon, the foregoing matter went off the
,m
(%.y ) 25 record at 3:09 p.m.)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234 4433 WASHINGTON, O C. 20005-3701 (202) 234-4433
'(O)
CERTIFICATE This is to certify that the attached proceedings before the United States Nuclear Regulatory Commission in the matter of:
\
Name of Proceeding: 92" ACNW Docket Number: N/A Place of Proceeding: ROCKVILLE, MARYLAND were held as heroin appears, and that this in the original transcript.thereof for the file of the" United States Nuclear Regulatory Commission taken by me and, thereafter reduced to typewriting by me or under the direction of the court reporting company, and that the transcript is a true and accurate record of the foregoing proceedings.
) -
Md 144A ~
LCOR'B1TT'RI!(ER Official Reporter Neal R. Gross and Co., Inc.
i l
1 O
hTAL R. GROSS '
COURT REPORTERS AND TRANSCRIBERS 1323 RHODEISLAND AVENUE,NW (202)234-4433 WASilINGTON,D.C. 20005 (202)234-4433
'o O O YUCCA Y
MOUNTAIN PROJECT
_ _ _ A_. , _
DOE's Comments on Defense in Depth Presented to:
Advisory Committee on Nuclear Waste Presented by:
Dr. Stephan J. Brocoum Assistant Manager, Licensing * '^ -
Yucca Mountain Site Characterization Office
- ir May 21,1997 u.s. Department of Energy office of Civilian Radioactive Waste Management 1
) _ . _ _ _ _ - _ _ . = - - - _ - _ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
~
O O O.
Introduction Preclosure repository system amenable to traditional defense-in-depth approaches
- Prevention
- Mitigation
- Multiple barriers
- Conservatism, redundancy, diversity in design
- QA controls, emergency plans & procedures t t
i i
__,,,_,,..,2
O O O~
i Introduction (continued) t Defense in depth for postclosure repository sy~ stem includes -
- Prevent / delay radionuclide releases '
- Mitigate transport after release
- Multiple barriers: conservatism, redundancy, diversity in design
- Engineered barriers to compensate for uncertainties in natural barrier performance
- Natural barriers to compensate for uncertainties in engineered barrier performance
- QA Controls j Acnweaocount2s Perms-21-97 3
O O O Historical Perspective 1
1986 Environmental Assessment made conservative assumptions to compensate for uncertainties in site t and design information
- Assumed higher flux than expected at that time
- Downward flux assumed - no diversion
- Assumed lower bound on thickness of zeolites
- No credit taken for thermal dryout i r
i ACNWBROCOUM 125 PPUO5-2197 4
~
o O O
- Historical Perspective (continued) 1988 Site Characterization Plan presented objectives for repository system j
- Engineered barriers: limit release of radionuclides to natural barriers
- Ensure engineered barriers do not adversely impact natural barrier performance ;
- Natural barriers:
- Provide environment conducive to long-lived engineered l' barrier .
'
- Provide conditions where transport of significant quantities of radionuclides will take a long time i
ACNWBROCOUM 125 PPT 725-21-97 5
O O Of Historical Persoective (continued)
Issues related to defense-in-depth concepts We have stated support for a single risk-based quantitative criterion for postclosure performance Our proposed revision to 10 CFR 960 relies on total i system performance (natural + engineered barriers, each of which contain multiple barriers)
[
t
- - ,,, _ ,,.., e
o o O -
Current Safety Concept and Defense in Depth Previous design analyses and total system performance assessment
- Provided basis for developing waste containment and isolation strategy Waste containment and isolation strategy
- Establishes framework for allocating performance to natural and engineered barriers
- Provides basis for describing defense in depth for postclosure repository system elements AcNWBROCOUM125 PPTM5-21-97 7
i
O O O Evolving Waste Containment and Isolation Strategy h \@ Robust Waste Package '
hLimited Water Contacting Wast \ / l Packages ggggy
- nff a.
T f' F e j i ;
[e$$}h})}E@bYk h l5
{4 8
+I g,vsg "sr s %jj?njg-a
.::,9 i !
%p i i 3 Limited pie:gh(ll q / I
% 25 -
gyjg 4 !# ; brugg% i/g if '
Mobilization of '
'i @ gg g Radionuclides t
4;ys~. m_. ,
.r i
\ '
s ATransport Through Engineered and atural i Barriers Reduces Concentrations '
WSTTRB2.CDR,129.5-13-97 i
Evolving Waste Containment and Isolation Strategy Tk? Limited water contacting waste packages
! @ Robust waste packages
<$> Limited mobilization of radionuclides from waste form '
A Radionuclide concentrations reduced during transport through engineered and natural barriers l
l ACNWBROCOUM.125 PPDMS-21-97 g {
t I
O O O Refinement of Defense in Depth Strategy: Improved Site Understanding Recent evidence for higher percolation flux and better definition of site variability
- Percolation flux through potential repository host rock 1-10 mm/yr
- Seepage into repository drifts is likely to be less but variable in space and time
- Thermal effects will redistribute moisture with slow return to ambient conditions over several thousand years j
- Radionuclide concentrations will be reduced during transport Heterogeneities lead to uncertainity in predicting processes 10 !
______m _-_-._. -_.._-_____ -: _____ m.-- -____m___________m___-_.-.-- --- _-____-.--__ -__-_-_..__ ____.______-_.-_.._- -_____ -.__ -__.m.______-_..___m_ _ _ - _ _ _ _ _
O O U Refinement of the Defense-in-Depth Strategy (continued)
L Allocation of performance to eacn repository system attribute depends on effectiveness of component barriers snd their uncertainties Improved understanding of moisture conditions and better definition of spatial and temporal uncertainties j
- Used as input to sensitivity' analyses on total system performance j
- Identifies potential need for enhanced engineered 1
barriers to accommodate natural barrier uncertainties l
-- ,,m -s, .., ,, j i
ill i!{i,!! i! ' l l! j:j e e g
_ v e g m y
a 9 e
_O o o 1
2 -
5 r b a s m S U
t e d a p e r o t s
P P
5 2
s 1
_ a e e x e f M U
f u
i O r
Wg s t C i
l f t e
r d O
R B
m g s u g W
N c
A a i l
n n t s n de k s e
i t
i a
g i
o i
r e
t c d l a
a m t i
a i v o r e
n e
i mP o r
p r
e c
i e
r l
b e g
p r a Le t g a l a
- s n dr bsi a p a y n v e ha i t
t r oa e O pW e t w at i
s s s n l l
i d e f e e w o
i pn c o Dg o n
d aou n o
n z ey ccd e
ni t d me edloe l s r e swr i s
r e
ic e oik r v ea st s
r t
a r
sl s
i r
a i
v f
n a t
s i d
n n i e tu f oiB y oiwd r a pt d e e o n r
r a nr d r oel fC a s o oo i l
elac r re es e o mirb ei e B nt a itsi u r r s e wr f p na r
D l
adl t eo et et eiegf r a t
t r t u rf n ie a AiDr i i vpinif ngDuHiEb t d
r n emne N *
- E * * .
O
, .l! ; : i
@ Defense in Depth:
Robust Waste Packages ;
Use of corrosion resistant inner barrier and corrosion allowance outer barrier prolongs life of packages .
a Galvanic processes offer protection to inner barrier Potential for use of ceramic coating on waste packages may provide even longer life Use of backfill
- May offer mechanical protection for diversion system and packages
- Could limit advective flow to waste packages AcNWBROCOUM125 PPDM521-97 13
O~
~
O O -
I
~
4 Defense in Depth:
Limited Radionuclide Mobilization i
- For some radionuclides, solubilities limit I
mobilization j Cladding reduces waste form surface area exposed
- Long containment time limits alteration of waste forms t l .
Limiting impact of engineered materials on water
- chemistry may be useful to reduce mobilization i
. S AcNWBROCOUM 125 ppm 05-2197 g4 1
o o 0 .
A Defense in Depth: Radionuclide Transport Through Engineered and Natural Barriers Engineered Barriers Potential for additives to material beneath waste package (invert) to delay transport Use of backfill reduces potential for advective flow Natural Barriers j Matrix diffusion in both unsaturated and saturated zones reduces concentrations Sorption will be effective for some radionuclides j
=
Concentrations will be reduced when UZ flow reaches water table -
Mixing and dispersion during transport lead to dilution Additional mixing occurs at point of water withdrawal Acwweaocouus2s eenes-mi-er 15
Defense in Depth: Analyses of Disruptive Processes and Events Early site screening considered probability of significant disruptive processes and events ;
Current approach is to analyze !
features / events / processes on basis of likelihood and potential effects Total system performance assessment is used to evaluate consequences for limited number of features / events / processes ACNWBROCOUM.125 PPUM21-97 gg
lll1l ll !
g 7
n 9-i h
t 1
2 i M e
7
- d de T
P se w P
- ere r nv ts 5
2 1
M U
e e ei t ss O C
. e n f et a e h u nit l O
R B
W
- ni N C
- igg di l a mt epe s A
n n e e re dul i q ed t er
- do v e e n n ef o r mo sie
- y ch nt poh em ss r ap sst ns
- a h e r ed ofee r
- nd i r nf e s s m e n r a
- + a e sda bv on se ec i
m s s e u e g n l
at i i t r n
r a pu S af e a t
_ k e ui an msnm t t u
s ero cd nau syf
- a pe f
- d oq at ner ei v
- t so se eipa vt n mir t
_ ar n wp eot a cd u
- ei -
s ns v r ne mo-e t
sr uiee ph eut c s
brtr s mtp nhe o ayo e o ie gf l w
RbsCd Ch r
- a _
YUCCA MOUNTAIN PROJECT
- - i ... -
Defense-in-Depth for the Repository Engineered Barrier Systesa t
Presented w:
Advisory Committee on Nuclear Waste Presented by:
Jack N. Bailey r Deputy Manager, Engineering & Integration Operations ,,
Yucca Mountain Site Characterization Office #
U.S. Department of Energy May 21,1997 Office of Civilian Radioactive !
waste management i
~
0 O O~
Overview i
i Defense in depth for preclosure operational period provided by l
- Prevention
- Mitigation '
- Multiple Barriers t
- Conservatism, redundancy, diversity in design
- QA program
- Radiation dose standards
- Emergency plans and procedures / limiting conditions ~
for operation acNWBALEY.45 PPUM541~"
2
. \
O O O. ,
Overview (Continued) i Postclosure defense ia depth extends preclosure concepts .
- Prevention ;
i
- Mitigation
- Multiple Barriers.
- Conservatism, redundancy, diversity in design
- Engineered barriers to compensate for uncertainties in performance of natural barriers
- Natural barriers to compensate for uncertainties in
- performance of engineered barriers
- QA program
- Radiation dose standards '
- Limiting conditions for operation ACNWBAN.EY.125 PPT 705-21-97 3
i
~ ~
O O O i
- Design Goals for the Engineered Barrier System
- Engineered barriers
- Work in concert with natural site features
- Not adversely impact natural barriers
- Consist of multiple L'o.ciers to i
- Delay failure of the waste package i
- Delay release of radionuclides from waste package a Mitigate effects of radionuclide release i
i t
mm au.us mtrom.7 4
l
Engineering Goals for the EBS Meet preclosure requirements
- Packaging
- Handling
- Storage
- Closure Develop a design that provides acceptable performance for the expected postclosure case Use multiple barriers to improve confidence in the engineered system performance considering ,
- uncertainties in natural processes i
- uncertainties in response of design features 4
AcnwsAarv.12s eenes ai-s7 5 !
o
~
o o
-l i
Strategy for EBS Defense in Depth Develop design features for the expected case to provide desired performance j Systematically evaluate options for design features that could be used to improve performance
- Use performance assessment (PA) to analyze ;
performance contributions .
- Evaluate sensitivities to low probability events / processes Systematically evaluate the PA sensitivities to identify data uncertainties
- Document design features / options with regard to the effects of data uncertainties AcNWBALEY.125 PPDO501-97 g
O O O'
~
Strategy for EBS Defense in Depth (continued) t Systematically evaluate PA sensitivities to identify uncertainties in design feature / options response
- Document the design features / options with regard to the effects of uncertainties of their response .
Select appropriate design features to improve performance by desired amount and offset effects of major data uncertainties acuwe=cy.22seerr4>s.2, 7 7
o o I ~
O.~
Models to Determine Design input Unsaturated zone hydrology
-Infiltration, percolation and seepage Thermal hydrology
- Temperature, relative humidity, liquid saturation, seepage Thermal chemistry Unsaturated zone radionuclide transport
- Advective velocity, matrix diffusion, retardation Saturated zone hydrology and radionuclide transport
- Dispersion, dilution ACNWBMLEY.125 PPT 745-21-97 g
o O-Ambient Flux-Time Profile i
n 1000 Ambient Steady Flux l
,......, ------ Mobilized + Ambient Flux
/ 1
/ Pluvial ;
j periods
,/ 1 e
/ 1 !
100--- ! !
b .
D [ Wet Years I
E . !,
E ; 1 i
i s .
- i. l E *
. I to-
- l. !. ,
[\
t i
t i
l 1
i i i i i 10 !
1 100 1000 10000 100000 1000000 Time, yr !
e .
Preliminary i J9CHARifDAF.1235-16 9F ACNWBAILEY.125 PPTTMS-21-97 g i
O~
~
O O Episodic Flow-Time Profile Single fracture receives annual flux from 400 m2 area in 1 week.
2.5% of WP wetted each year 600 Wet Year f
M 400--
E '
i f
3 Ee 200--
i
- b. '
0 ; i i_ ,
i i ,
0 1 2 3 4 5 ;
Time, yr i
Preliminary J8Cl4WTCDRF.i2 MIS 4 F ACNWBAILEY.125 PPT 7A)5-21-97 g i
I o o O~
Temperature-Time Profile 83 MTU/ acre, ACD Layout, Point Load, No Backfill, Low Flux 250 10-yr-21-P 225 4 _
H DHLWP *
= I 200-
~
0
- ~
g 175 - -
g :
y 150 9 ,
O :
5 125 -
E :
j ,go _i ---
2 2 75 - ,'
E.
- 50 e ...
25 -_'
0 uj -
j "} "; "l 0.1 1 10 100 1000 10000 100000 Time, years Preliminary __,m....,
ACNWBAILEY.125 PPT 7A)5-21-97 gg
0 0 'O.~
Humidity - Time Profile '
83 MTU/ acre, ACD Layout, Point Load, No Backfill, Low Flux 1.0 -
i i
e 0.9 -7
- l
's**', ,.*
0.8 -? *.,,,,*
E 0.7 -1 o -
i u -
10-yr-21-P s
{0.6-i : -
H DHLWP ;
I 0.5 -1 i I
N u 5 j 0.4 -_ ,
l
$ i g 0.3 --- +
y :
{* 0.2 -- ;
0.1 --
- -x O.1 ^ ^
"l 'l '
'l '
"l '
-l 0 1 10 100 1000 10000 100000 '
Time, years Preliminary JCCHARTCDftr.12M-ts47 ACNWBAILEY.125 PPT 7A)5-21-97
. _ _ _ _ _ _ _ _ . _ . _ . . _ . . _ _ . . - . - - - - - - - - - - - - - - - - - - - - - - - - - - ^ - - ' - - ~ ~ ~ ' ^^
o o
~
0 ~:
l l Ph/lonic Concentration-Time Profi e 14 13 -
12 -
Normal l ii .
. Cement 10 - *
- Low pH R ,.[
, Cement l 6-5-
\
4-3- / PitCuAcid
! 2~ '
j i t
! 1 10 100 1000 10000 100000 i l Time, years lonic Concentration to 9- Condensate by I e Bolling off Water
(
o 8- i
$ 7-
- 6-l E
j
$5- I O 4- Coe sate 4 that nas not 3 Ambient t I equilibrated Equilibrated '
y 2-Water
. 0
......t, . --
Pre 1immary , ,0
,0.
. 00 .0 Time, years i i
6 -
~,
O O O.'
. t Mechanical Load - Time profile f
6000 5000__ Stress from 8 m drop of 6 T spherical rock 40--
= .;
o ,
f 30---
O i
E l 20-- i t
10--
i
- : : i 1 10 100 1000 10000 jococo !
Time, years Preliminary ---
i ACMVBALEY.125 PPT 705-21-97 g
[
___ m - - - - - . _ _ - . . - . _ _ . __ _ _ _ _ . - _ _ _ _ _ - . _ _ _ _ _ _ . _ _ _ _ . _ _ . . _ _ . _ _ _ _ _ _ _ _ _ - . _ _ _ _ _ . . _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
o o 0-
~
Major Features for TSPA Evaluation Subsurface layout i
- Drift size and spacing
- Thermal load
- Support and ventilation system i a
1 Engineered barrier system
- Invert materials
- Packing and backfill materials
- Flow diversion Waste package c - Size and thermal load
- Materials and fabrication technique ACNW3AILEY.125 PPT 705-21-97 15
DRA Design Options for Waste Isolation Surface Control Land Withdrawal -
Thermal Design f Ee Areal Mass Load (High, Low) g5 eDrainagructures ~
~
f 5e WP Spacing (Point, Line)
J+d$9 Altered Near-Field Rock SNF Assembly Blending Drift Lmer =
~
e To meet 18 kW limit e Normal Concret' To meet criticality limit e Low pH Concrete e Control Thermal Variability G eEmplacement Drift Wall O e Pre-Cast Concrete ,/ WP Sequencing 9 4-Drifts Open 0 e Cast-In-Place Concrete e Steel Sets Lag Storage e Subsurface ;
e Crown Joint * "'I"'#
O e No Crown Joint "
Upper Backfill ;
e Rock Fall Protection N Ge Limit Flow & Humidity Be Richards Barrier Oe ( p 1 ary Barrier) ..
f' ' N s > U8 Single Layer ce %, /
De
.7 / \ e Condition Water O
+ ga..t:p / g \ e4 ,
Cladding Credit e
g - ,-
khl as e Supported by Backfill f6 e Free Standing Waste Package g B Supported by W A "
A Corrosion Allowance Material Corrosion Resistant Material
~f" ^
7[
-m
, ne Lower Backfill A Galvame Protection 10 e Condition Water !'
A Ceramic Coating (inside or Outside) + Sorb Releases A Large, Small, Small-in-Large j(
e In-drift. Vertical Borchole, IIoriz. Borehole X
+ G e Pedestal i La',out of Emplacement Drift O WP Environment A Robust WP Levil + Additives (Non-Restricte O Limit Mobilization + Zeolites + Additives (Restricted)
O Radionuclide Concentration Reduction h + _
i t
16 l o
o O O~.
i Design Features Evaluation Matrix i
POST-CLOSURE GOALS POST-CLOSURE ENVIRONHENTS Delay Prolong time fran WP Mkigate breach of Water Relative RockfaN &
breach to release from Chemistry WP EBS Flux Humidity Drift Eneitteered Featums waste release Collapse e+-1-s credit X X X X conosen resistant material X Y X X X corrosion allowance material X Y Y ,
i;r. iFoie-;; sun X X X X X X Y ceramic conhng X X X
- v= ped age X X Y
' X y sme package Y
smd -in-4v-+edage x p etal Y X X wwert a*seves (non-hazardous) Y X X invert add *ves (hazardous) Y X backfie X X Y Y X support drip shield X Y X Y Y Y X rock fas protection X Y X X Y Y X limit flow & humidity X Y X X Y Y X X Y X Richards bemer X Y 1 Y X X Y X condition water X Y Y X X Y sorb releases X X Y Y X X Y X
- iP shield X
=w--; d by backfie X Y X X Y free sie..J s X
' X Y supported by WP X X Y aircap X dnft hner X
Y crownjoint X Y X no crown joint Y X X X normalconcrete Y y low PH concrete Y X 17 3 Y Y Y X X
ODasign Features @valuatiori Matrix O
~
(Continued)
POST-CLOSURE GOAL.S POST-CLOSURE ENVIRONHENTS Delay Prolong time te kma ~
bro of i"se from ,",' '
Chemistry ft ch to E
Engineered Features ***'"' " ;
pre-cast concrete Y Y Y Y X cast-in-place concrete Y Y Y Y X steel sets Y '
X e.. Aement drift wat Y X altered nearfield mck X Y X areal mass load - high X X X areal mass load - bw X Y WP spwjng- Wii bad X Y WP spacing -line bad X Y Y Y SNF assemth bierwhng X Y X X to meet 1skw limit X Y Y Y to meet criticalitylimit X Y Y Y control thermal varistnhty X Y Y Y WP sequenemg X Y Y Y 4 drifts ope 6 X Y Y Y lag storage X Y Y Y subsurface X Y Y Y surface X Y Y Y surface control X '
X land vnthdrawal (limit infiltrations) dramage structures X X -
es wedrift sloped '
level in drift emplacernent y Y Y vertical borehole horizontal borehole Note: "X" -ndicates prirr ary function of feature; "'" indicates a secondary fuaction 18 of the feat sre
O O O'.~
Summary of Desi~gn Process to Achieve Defense in Depth Design for the expected case Determine key sensitivities through performance assessment Evaluate uncertainties in data used for design
- Document design features / options Evaluate uncertainties in the response of design features l
- Document design features / options l
ACNWBAILEY.125 PPD 25-21-97 9g I
O~
~
O O Summary of Design Process (continued) l Consider available design features / options in the context of total system performance
- Conservatism t
- Redundancy
- Diverse barriers a
Select design features to provide desired level of performance ;
t ACNWBAR.EY.125 PPT 7A35-21-97 g
O O O YUCCA h
MOUNTAIN PROJECT
_ _ 4.x.im Status of TSPA Analyses in Support of the Viability Assessment Presented to:
Advisory Committee on Nuclear Waste Presented by-Dr. Abe Van Luik m , ,^
Yucca Mountain Site Characterization Office m.
U.S. Department of Energy i May 21,1997 rrice rcivilian naaioactive Waste Management i ;
r
o O O -
Outline
- History of uses of TSPA i TSPA status and plans
- Preparations for TSPA-VA j
- Preliminary scoping sensitivity studies supporting evaluation of alternative designs for defense-in-depth ACNWLUlK.125 PPT 7/0521-97 p I
o .
O D -
HISTORY OF YUCCA MOUNTAIN ,
PERFORMANCE ASSESSMENTS .
I. UP TO THE 1986 ENVIRONMENTAL ASSESSMENT A Total System Performance Assessment was Simplified modeling gave initial wamings conceming the done for the National Academy of Sciences in possibility of relatively high doses when discharging 1983 (Waste isolation Systems Panet Report) radionuclides into the groundwater of a closed basin ;
(disagreed with the appropriateness of the cumulative !
release performance measure being considered by the EPA in 40 CFR 191).
- Showed it possible to meet NRC 10 CFR 60 subsystem requirements and still fail the EPA cumulative re! ease standard.
Nuclear Regulatory Commission sponsored Simplified modeling of a repository in the saturated zone i System Assessment, by Sandia National Laboratory, suggesting solubility issues under potentially oxidizing to evaluate the draft EPA Standard (1983) conditions and the highlighting the importance of i understanding flow fields and transport parameters.
Environmental Protection Agency System Results of simplified modeling, including low-probability Assessment in support of its promulgation of disruptive events for the first time, suggested a Yucca :
40 CFR 191 (1985) Mountain repository capable of meeting the draft EPA l standards, with a waming that if unsaturated zone fluxes I were substantially higher than assumed, consequences ;
would be higher- these analyses were little changed in '
the 1993 update of this Background Information Document. -
Department of Energy system level evaluations Results of simplified modeling, making assumptions for the EA (1984,1985) thought conservative at that time, suggested a Yucca Mountain repository capable of meeting the standards.
Acuwuxus wrms-m7 3 l
0 O O -
HISTORY OF YUCCA MOUNTAIN PERFORMANCE ASSESSMENTS II. SINCE THE SCP: MAJOR CHANGES IN ENGINEERED BARRIER SYSTEM MODELING CONCEPTS L
Waste package design at the time of the Small containers, vertically emplaced in boreholes, Environmental Assessment (EA,1986) made of relatively thin outer barrier of stainless steel l and/or the Site Characterization Plan (SCP,1988)
Maximum capacity: 3 PWR assemblies,4 BWR assemblies, or 1 HLW canister 1
Current waste package design being Large multi-barrier container of nickel-based alloy evaluated within carbon steel, horizontally emplaced in drifts i
Maximum capacity: 21 PWR assemblies,44 BWR assemblies, or 5 HLW glass canisters, with 1 DOE SNF canister i
[
AcnwuRK.125 PPTM5-21-97 ;
4 i
.___________-_______________________-_____-_-____--______________-_______L
o O U,m ~
HISTORY OF YUCCA MOUNTAIN PERFORMANCE ASSESSMENTS 111. SINCE THE SCP: MAJOR CHANGES IN NATURAL
- BARRIER SYSTEM MODELING CONCEPTS Concepts regarding water flow and Based on short term observations, the not infiltration was ,
radionuclide transport through Yucca estimated to be less than a few mm/yr Mountain at the time of the EA and/or the SCP No estimates of net infiltration under wetter climate regimes were available, effects were thought to be minimal based on presence of the Sierra Nevada Mountains.
Solubilities largely based on literature, retardation estimates from preliminary laboratory work and literature.
1 Current concepts regarding water flow Second generation of flow and transport models incorporating and radionuclide transport through and calibrated to data from repository level have suggested a Yucca Mountain net flux under current conditions between ~2 and ~15 mm/yr i improved basis for estimating solubilities, site-specific database for representative radionuclide sorption parameters Flux range will increase to, perhaps, triple current values under
- episodic wetter climate conditions ACNWLutK 125 PPT 7/05-21-97 5
o O O HISTORY OF YUCCA MOUNTAIN
~
PERFORMANCE ASSaSSMENTS IV. SINCE THE SCP: MAJOR CHANGES IN ASPECTS OF TOTAL SYSTEM PERFORMANCE CAPABILITY Estimated health and safety impacts Using simple models, unsaturated zone flux was most of a Yucca Mountain Repository at the important parameter determining releases for 10,000 years, time of the EA and/or SCP and conservative estimates at that time indicated that releases were minimal Engineered system failure and disappearance was assumed after a specified time usual!y not exceeding 1,000 years The warning was sounded that higher fluxes wou!d mean 1 higher releases Doses were not addressed until late 1988: low doses then predicted because of low flux and optimistic Np solubility Current estimated health and safety Process-level models for engineered system and natural impacts of a Yucca Mountain Repository system nearing completion for TSPA-VA Process-level modeling being incorporated into TSPA through abstraction and testing, creating a more thorough and defensible TSPA capability All pathway doses are to be considered to the time of the peak dose, and for various hypothetical locations of potential dose recipients Preliminary sensitivity studies do not call into question the expectation that the TSPA-VA willlikely suggest that a Yucca mountain repository is a viable option for the disposal of the nation's spent nuclear fuel and high-lev 8ME2s rems-2"'
6
HISTORY OF YUCCA MOUNTAIN !:
PERFORMANCE ASSESSMENTS V. MAJOR CHANGES IN TSPA PHILOSOPHY FROM TIME OF SCP i
SCP assigned subsystem and component performance goals until complete TSPA-capability available ;
- Similar to NRC effort to include subsystem performance requirements in Part 60
- Used informal expert judgment to tabulate Estimated Partial Performance Measures (EPPMs) that drove system requirements down to subsystem and component level f
- EPPM tables in SCP assumed simplified and unsubstantiated conceptual models Site characterization and design process modeling and TSPA analyses have matured i since the SCP, and TSPA tools are used directly to support evaluation of data uncertainty, design assumptions, and alternative conceptual model significance ;
- EPPM and subsystem / component performance goals, valid for a specific conceptual model only, are obsolete
- Lower-level surrogates for the primary public health and safety concern are either derived directly from that primary concem, adding no assurance, or will be more conservative, adding no public health or safety i
ACNWLUtK.125 PPT 7/05-21-91 i 7
O O O
. t Outline
- History of uses of TSPA TSPA status and plans
- Preparations for TSPA-VA
- Preliminary scoping sensitivity studies supporting evaluation of alternative designs for defense-in-depth !
i ACNWLUlK.125. PPT 7A)S-21-97 g
1 I
^
Status of TSPA-VA TSPA-VA Plars is being implemented 9 of 10 workshops on abstraction of process models for input to TSPA have been completed
- NRC staff have observed each workshop Sensitivity analyses and model development are in progress j
Expert elicitations are being conducted TSPA-VA Peer Review has been initiated Technical exchange scheduled for July 1997 AcwwumO25 PPTM5-21-97 g
f 1
t r
Genera ' zed Schedule for TSPA-VA ;
i
= i
- : : : l Plan and Conduct
=
Construct and Document Abstractions i. i.
=
i i = =
i
- i i = !
l
\
Freeze Process Models j : j = i i , Freeze Data j j i .
=
=
i-Y ! !
=
l
- : i 1/30 i
, j i j Complete TSPA Reference j ,
- j j
Case y i ,
. j j j . 6/12 .
. ! i i . Complete Draft TSPA-VA Document i
- : : i l l l l l l l 1 i FY 97 FY 98 t
t ACNWLUlK.125 PPT 705-21-97 10
Cs cJ o-Conceptual TSPA Logic Process Models Expert Elicitation Peer Reviews TSPA-LA ,
Site . ug pro,
- SZ Flow & Transport
- TSPA Orientation +
- Preparation 4 g
- UZ Ilow
- Analyses L
- ThermalIlydrology
- Process Models
- UZTransport
- Scenarios
- Documentation
- SZ Flow & Transport
- WF Dissolution WP Degradation
- Abstractions
- Thermal flydrology
- TSPA-VA
- Near Field Environmen I Engineerine WF Degradation
- WF Mobilization
- WP Degradation Environmental
- Biosphere Y
Process Model m TSPA-VA
' TSPA-VA +
Abstraction / Testing Analyses "
VA 1 '
Documentation V
l i
ACNWLUlK.125 PPT 745-2197 gj
O O O Schedule of Process Model Abstraction / Testing Workshops and Summary Documents Workshop Workshop Deliverable Workshop Topic Dates Date Location t UZ Flow Dec 10-12,1996 Feb 27,1997 Albuquerque WP Degradation Jan 8-10,1997 Feb 24,1997 Las Vegas -
Thermohydrology Jan 21-23,1997 Mar 18,1997 Albuquerque UZ Transport Feb 5-7,1997 Apr 30,1997 Albuquerque WF Degradation Feb 19-21,1997 June 30,1997 Livermore Rn Mobilization Feb 19-21,1997 May 15,1997 Livermore NFE Mar 5-7,1997 June 30,1997 Berkeley Criticality Mar 18-20,1997 Sept 30,1997 Las Vegas SZ Flow Apr 1-3,1997 Feb 19,1997 Denver Biosphere June 3-5,1997 Aug 15,1997 Las Vegas ACNWLuiK_125 PPT 7A)5-21-97 g
- _ _ _ _ . ___.___ _ . _ _ _ ___ _ _ .____ _ _ _--.. _.__.-.-- - -...__--..-... .. ----.,-. . .- .-_-.. _ .~ _ . - - - - . . . . ~ . - - - - . - - - . . . . _ . . . . . . . . . . . . _ . . . . _ _ . . _ . .
O O O ;
Major Outcomes of Abstraction Workshops -
Identification of key performance-related issues I associated with each process model j
- DOE WCIS hypotheses
- NRC KTI's
=
Prioritize key issues using surrogate performance- '
related criteria as a temporary expedient i t
Develop approaches to address the key issues within the context of the VA
- Sensitivity analyses
- Abstraction plans i
i !
i l
Acwwwuors emeri sr :
33
Expert Elicitation Feeds to TSPA-VA '
- UZ Flow
- Confirmation of magnitude of infiltration flux i 1
- Uncertainty in infiltration and percolation fluxes Waste Package Degradation
- Range of likely galvanic protection percentage
- Range of degradation models for corrosion resistant '
materials and ceramic coatings SZ Flow and Transport
- Range of reasonable dispersivities
- Range of mixing percentages ACNWLU1K.125 PPTWD5-21-97 g
~
O O O ~
i TSPA-VA Peer Review Peer Review panel
- Chris Whipple (Chairperson)
- Bob Budnitz -
- Paul Witherspoon
- Rod Ewing
- Joe Payer j i
- Dade MoeIIer Kickoff meeting held in February,1997 Expect interim review comments in June,1997 i
t i
l AcNWuRK.125 PPUM21-97 15
i OUTLINE I
History of uses of TSPA TSPA status and plans
- Preparations for TSPA-VA -
- Preliminary scoping sensitivity studies supporting evaluation of alternative designs for defense-in-depth t
f, i
i i
i AcNWLUtK.125. PPT 7J05-21-97 gg i
[
O O O Other Performance .
Assessment Activities System-level and supporting models are updated and tested as new site and design information becomes available Models are used to support the ongoing process of evaluating the adequacy of construction and testing controls Models are used for scoping sensitivity analyses in support of design and testing '
~
ACNWLUtK.125 PPT 7A)5-21-97 g
o o 0 -
Scoping Sensitivity Analyses:
Effect of Defense-in-Depth Engineering Alternatives Updated TSPA 1995 case with current site model information to allow comparative analyses
- Infiltration map (Flint,1997)
- Percolation flux (LBNL,1997) avg 6 mm/yr (range 4 to 10 mm/yr)
- % of flux in fractures (30%)
- Fracture flux seeps into 10% of drifts j
- Revised thermal hydrology j
- Np solubility reduced by a factor of 100
- Saturated zone flux of 0.3 m/yr Acuwwna25.rerrEas-s718
O O O Scoping Sensitivity Analyses:
Effect of Defense-in-Depth Engineering Alternatives Updated TSPA 1995 case design assumptions
- 2 layer waste package (carbon steel and Alloy 625)
- Point loading of waste packages l - Galvanic protection (75% of carbon steel corroded l
4 before pitting of Alloy 625 commences?
10% of packages assumed to have ineffective galvanic protection
- No cladding credit -
- No backfill
)
- No ceramic coating
- No drip shield ACNWLUtK.125 PPT 1;OS-21-97 g
O- 0 0 Scoping Sensitivity Analyses:
Effect of Defense-in-Depth Engineering Alternatives i
Case 1: Updated design Case 2: Updated design with cladding credit taken l for 90% of fuel pins j Case 3: Case 2 with ceramic coating on waste packages lasting between 5000 and 15000 years
- 10% of packages assumed to have ineffective ceramic coating
[
Case 4: Case 3 with quartz sand backfill and drip shield t
~
AcNWLutK.123 PPT 7/05-21-97 20 I
o O O e
Scoping Sensitivity Analysis 100,000-yr Drinking Water Total Dose (ICRP-30 Based) History i At Skm boundary 1,000 : - - , Casei
! Case 1+90% cladding i 100 -r - ------ - l -- ---- ---i--- - - Case 1+90% claddir:g+ Ceramic U[5,000-15,000 yrs] :
! l Case 1+90% cladding + Ceramic U[5,000-15,000 yrs]+ drip shield l
cx 1-10 -b- : I
, - - - + - - - - - -
d
~ - .
1-r--------;- - . y--- - c n E
CD : I :
u E 0.1 7:
g --+ - - - ' - - -
+ --
.....i........-...... 4.......s; i 8
O 0.01 -b-f-
- f
- t* --+--- -- --
i: i a o_oo3 4 _. i.
3.-i -+ - - - - . - . - . . . ;
t
~
- l. _ -
0.0001 - r- il l- --- Y - - - - - - - - - - -
--5 0.00001
'L L/
1 5
0 20,000 40,000 60,000 80,000 100,000-t Time (yrs) !
ACNWWIK.125 PPTM5-21-97 g L
t I
O O O -
t Scoping Sens.t. .ty Analys.is i ivi 100,000-yr Drinking Water Total Dose (ICRP-30 Based) History At 30km boundary 1,000 ; . .
.............e l Case 1 100 -r- 1 Case 1+s0% ciadding i 5 l - - - Case 1+90% cladding + Ceramic U[5,000-15,000 yrs]
- 1-.
10 -; . .. .] ase1+9 % i dding+ er mi uts.ooo-is,ooO yrsj+erip snierd ,
x ss i
l - _
_ ,.. .. j .-
_ . . . 5-e L.
5 I 5
I Q} _- .. l ..
{
m 7.. ... .. j... ..p i........ ..j 3
l %,._ -4 5
... f.... . .. . ..y....... ..
o 5 I -
O E 0.001 -
-[ - - -- l--- , [ /---- , . - v--- -
-- - - - - -! - = = ' - - - - i - - - -
-- j
- l- -- A 0.0001 ;
5 l ./
4 i-- - - -
_; i 0.00001 - I /" 7-5 I~- -
l 0 20,000 40,000 60,000 80,000 100,000 Time (yrs)
ACNWLUtK.125 PPT 105-2197 y '
_=.__m. _ _ _ . . _ _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ . - _ _ _ _ . .
o O O -
t Scoping Sensitivity Analyses:
i Other potential design alternatives being evaluated Waste package
\
- Alternative ceramic coatings
]
- Alternative degrees of galvanic protection
- Alternative corrosion resistant materials
- Alternative cladding degradation 4 :
Engineered Barrier Segment
- Alternative thermal loading designs
- - Alternative drip shields q
- Alternative backfill '
- Sorbing material placed in packing / invert ACNWUJIK.125. PPT 7A35-21-97 23
Conclusions a
Since the SCP, advances have been made in design, understanding of the site, process , and system-level modeling
- t With a credible system-level modeling capability, it is both possible and preferable to do system-level evaluations of subsystem and component design changes and site uncertainties a
Preparation for TSPA-VA proceeds as planned, including uncertainty evaluations, abstraction of process-level modeling, model testing, and external expert reviews i
Performance assessment continually updates system-level models~ to support testing and design with scoping sensitivity studies to aid decision-making !
ACNWLUlK.I25 PPT 7M21-97 g
O O O -
YUCCA MOUNTAIN PROJECT n - i.. . . -
E se of Expert Elicitations .
in DOE's High-Level R_adioactive Waste Program Presented to:
Advisory Committee on Nuclear Waste Presented by:
Dr. Stephan J. Brocoum Assistant Manager, Licensing 5
, y t Yucca Mountain Site Characterization Office 4 h 6 U.S. Department of Energy 21, IW fUce f Civilian Radioactive Waste Management
O O O -
~
I Outline
- Purpose of Expert Elicitations
- Summary r
E ,
4 i I
O O O -
Purpose for Expert Elicitations Expert elicitations are particularly useful in the following situations
- Empirical data are not reasonably obtainable, or analyses not practical to perform
- Uncertainties are large and significant for demonstration of regulatory compliance
- More than one conceptual model is permitted by available data
- Technical interpretations are required to properly assess the uncertainty in data, processes, and models Such elicitations are important in bounding
=
uncertainties, especially with the unprecedented time frames being evaluated
O O O e
Purpose for Expert Elicitations (Continued) i Expert elicitations can be used to help reach conclusions j
- Can quantify the range of information and informed technical :
opinion bearing on technical issues
- Can integrate diverse technical input and document !
conclusions within the parameters of the evaluation t L
i I i
4 i
, i
Current Expert Elicitations DOE is currently conducting several expert elicitations
- These elicitations are in various stages of completion ,
- The purpose of these elicitations is to facilitate '
development of key process models ,
- This is done by characterizing the knowledge and uncertainty
, in technical issues most important to these key process models 1
! i 5 i
Current Expert Elicitations (continued)
These expert elicitations generally follow the guidance provided by NRC
- Branch Technical Position on the Use of Expert Elicitation in the High-Level Radioactive Waste Program (NUREG-1563),
August 1996 P
l l
6 I
~
O O 6 Current Expert Elicitations (continued)
Expert Elicitation Why Being Done Status Probabilistic Volcanic Address disagreement Completed June 1996 Hazard Analysis between DOE and NRC on probability estimates and i significance of different data sources on such estimates Probabilistic Seismic Uncertainty in timing, Ongoing. To be completed i Hazard Analysis location, and magnitude of by August 1997 ground motion and fault displacement results in varying interpretations important input for design of facilities (preclosure) 7 i
l
O O O -
Current Expert Elicitations (continued)
Expert Elicitation Why Being Done Status '
Unsaturated Zone Flow Large uncertainty in Ongoing. To be completed percolation flux, resulting in by May 1997 differing interpretations by Project scientists Percolation flux is a key parameter affecting total system performance Waste Package Degradation Waste package containment Ongoing. To be completed is a key contributor to total by August 1997 system performance Limited bases for prediction of long-term materials behavior a
i Current Expert Elicitations (continued)
Expert Elicitation Why Being Done Status Saturated Zone Flow and Dilution and dispersion are Just initiated. To be Transport very significant to total completed by September system performance 1997 :
Limited data set resulting in differing interpretations :
between DOE and NRC ,
Thermohydrology Limited information in near Planned for FY98. To be term due to length of completed by July 1998 :
thermal testing ;
r Confidence in models questioned by T/H Peer Review r
9 r Help focus long-term testing '
O O O
~
Current Expert Elicitations 1 (continued) l l
Expert Elicitation Why Beina Done Status l
Waste Form Dissolution Applicability of lab data to in Planned for FY98. To be situ behavior is uncertain completed by September 1998 Limited data has resulted in i making overly conservative assumptions in pastTSPAs !
i i
r
?
10 :
t k
L
- O O O i
DOE Perspective on Expert Elicitations Expert elicitations are a vehicle for enhancing internal and external confidence in the technical program Elicitations help to eventually support DOE's compliance arguments Such elicitations do not eliminate the exercise of judgment on the part of the decision maker Because of the process and technical expertise needed, expert elicitations are expensive and time-consuming t
11
O O O .
DOE Concern a At the 2/25/97 meeting on volcanism, NRC staff did not agree with the results of the PVHA expert elicitation
- The opinion of one or more NRC staff personnel seemed to be weighed the same as our ten experts
- The technical basis for such epinion was not fully provided
= DOE recognizes that NRC staff need to conduct independent evaluations of our work However, DOE believes that an appropriate level of weight should be given to such elicitations
- Otherwise, it puts the licensing value of expert elicitations in question 1 12 i
Aggregate Results for Frequency of O Intersecting the Yucca Mountain Repository i Foot Print by a Volcanic Event e i 5 , ,,,o p ,
,,,,,m, , , ,
,,,,,c., ,
,,,,,,,, , , ,,uu , , , , , , , ,
- (a) 5th% 50th% Mean 95th% -
+ IncEdual Means s
4 - O Individual Medians -
,, 3, ,, ,, ,
) _ -.
$ ~
-3 -
l _
b s
.c W 1 y2 - _
8 - _
n.
1
- O 0000 _
- J. + ++ H++ + _
0 O , ,,,,o.., , , ,
, o ,,j , ,
,,,,o, , ,
, , o . ,; ,,,,,,,,; ,
,,,o,,, ,
,,,o,q AM - v -
(h)
BC -
Ot -
~
GT -
C+ + l GW - DH ~
MK -
00 -
E, MS -
W -
M W RC - 0-0 -
RF - W ~
WD 00 ~
WH - N# ~
l Aggregate - v '
, , , i i , nl , , , e n ni , , , , n ni , , , , , , , ,1 ,,,,,,nl , , , , , , , ,i , , , , , , , , ,
10 12 10 11 10 10 10 8 10-s 10 7 10 6 10 5 l Annual Frequency ofIntersection C) raoevot< :: 224,s.s..,7
O O O Summary 1
! Because of the unprecedented nature of this program, expert elicitations play a key role For certain areas, DOE plans to use the results of
- expert elicitations as part of the basis for licensing
- These results will be well documented
- DOE expects that those elicitations will be given the proper weight during the licensing review The value of such elicitations needs to be agreed .
to with NRC management j
- These elicitations are quite resource intensive 14 i
O O O -
(,.-~.,}
y United States p
%...../ Nuclear Regulatory Commission Formal Use of Expert Elicitation in the High-Level Waste Program:
Staff Observations to Date Presentation to the ,
Advisory Committee on Nuclear Waste May 21,1997 l
l Michael P. Lee 415-6677 l
DIVISION OF WASTE MANAGEMENT OFFICE OF NUCLEAR MATERIAL SAFETY AND SAFEGUARDS U.S. NUCLEAR REGULATORY COMMISSION
. _. _._ ~ _ _ . _ . _ _ _ . . _ . . _ . _ _ _ _ . . _ . _ _ . . _ . _ . _ . _ . _ .. . -.. . _ . . _ . _._ _ . _ _ _ . _ . _ .
O O O - .
OUTLINE .
- RECENT BTP' CHRONOLOGY I
i
- WHAT NUREG-1563 RECOGNIZES I
- WHAT NUREG-1563 SAYS....
- WHAT NUREG-1563 DOES NOT SAY
- CURRENT STAFF ACTIVITIES
- PREVIOUS STAFF CORRESPONDENCE ;
i'
- FUTURE STAFF ACTIONS i
i aka NUREG-1563 3
f
____-- _ ____ _ . _ _ - - _ - - - _ _ _ - . _ - . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _________________________________i
O O O -
RECENT BTP CHRONOLOGY .
- Public comment period on Draft BTP ended May 1996
- Appendix 7 Meeting on consistency of the Probabilistic Volcanic Hazard Analysis '
with the BTP September 1996 '
- Final BTP published as NUREG-1563 November 1996
- Subsequent staff correspondence to DOE:
- Proposed resolution of 6 open items December 26,1996
- 3 issues related to BTP implementation December 31,1996
- Documenting and disclosing conflict of interest January 7,1997
- Staff monitoring DOE's implementation l
of NUREG-1563 Ongoing since late 1996
- Staff briefed Atomic Safety Licensing Board April 1997 3
t
WH/ T NUREG-1563 RECOGNIZES .
Under appropria.e circumstances, it is acceptable to supplement data and analyses with the opinions of experts as part of the support for demonstrating compliance with NRC's geologic disposal regulation, and that, in some cases, these opinions are best obtained using a formal and well-documented process.
- Empirical data are not reasonably ( btainable, or the analyses are not practical to perform.
Uncertainties are large and significant to a demonstration of compliance.
- More than one conceptual model can explain, and be consistent with, the available data i
- Technical judgments are required to assess whether bounding assumptions or calculations are appropriately conservative.
Formal elicitation can reveal a wide range of scientific and technical interpretations, thereby exposing (and possibly quantifying) the uncertainties in estimates concerning repository siting, design, and performance attributable to limitations in the state of technical knowledge.
Formal procedures may also help groups of experts resolve differences in their estimates by providing a common scale of measurement and a common vocabulary for expressing their judgments.
Bottom line: Use of formal, structured expert elicitations are expected to contribt te to the robustness of DOE's licensing case. ,
4
O O O -
WHAT NUREG-1563 SAYS.... .
- Sets forth technical positions that:
provide general guidelines on those circumstances that may warrant the use of a formal process for obtaining the judgments of more than one expert (i.e., expert elicitation); and describe acceptable procedures for conducting expert elicitation when formally elicited judgments are used to support a demonstration of compliance with NRC's geologic disposal regulation, currently set forth in 10 CFR Part 60.
Does not prohibit the Department from using its own experts in the conduct of any formal ,
expert elicitation
\
The formal use of expert judgment should not be viewed as a substitute for available data.
...The subjective judgment of experts should be distinguished from both measured data or technical calculations based on accepted scientific laws and principles. It ,
should be viewed as an alternative, and employed when other means of obtaining requisite data or information have been thoroughly considered and it has been >
concluded that such means are not practical to implement...."
4 s
i r
I
o.;
o o WHAT NUREG-1563 SAYS;... ..
(continued)
BTP will be the benchmark used by the staff for judging the acceptability of any formal elicitations used to support a potential license application.
i e includes an expanded definition of " peer review" over that provided earlier in NUREG-1297.
L l
r 6
O O O .
WHAT NUREG-1563 DOES NOT SAY....
NUREG-1563 has not attempted to prescribe the specific technical issues for which expert judgment should (or should not) be applied. ,
The staff has viewed such determinations to be the prerogative of DOE.
DOE agrees with this position
- Following guidance in the BTP does not guarantee that the staff will accept the results.
...the use of a formal elicitation process, even when conducted in a manner consistent with guidance provided in this BTP, guarantee that specific technical conclusions will be accepted and adopted by the staff, a Licensing Board, the Commission itself, or any other party to a potential HLW licensing proceeding. Rigid adherence to a sound elicitation process, in and of itself, in no way guarantees that the <
resulting judgments will be sufficient to meet the applicant's burden of proof regarding ;
the substantive issues addressed by the elicitation. Nonetheless, expert judgments obtained through an evidently flawed or poorly documented process will weaken their ability to support demonstrations of compliance...." i L
t 7
WHAT f :UREG-1563 DOES NOT SAY....
(continued)
- NUREG-1563 does not take a position on the recommendations of the report of the Senior Seismic Hazard Analysis Committee (so-called SSHAC Report).
SSHAC recommendations are be..eved to be consistent with NUREG-1563 NUREG-1563 does not evaluate the merits of the use of a technical facilitator/
integrator (TFI) as an alternative to the use of a "generalist," as recommended in SSHAC ,
Although the SSHAC recommendations introduce the TF1 concept, little practical guidance in SSHAC on the conduct of the TFl in fulfilling the assigned roles and -
responsibilities t
b b
8
O O O -
CURRENT STAFF ACTIVITIES -
- Staff currently monitoring imp amentation of NUREG-1563 by DOE in ongoing elicitations:
Probabilistic Seismic Hazard Analysis Unsaturated zone hydrology Waste package canister degradation t
- No fatal flaws observed....
P However, the staff do have some questions / comments with regard to how the '
elicitations are being conducted -
These questions / comments are consistent with two of the three recent staff letters to DOE related to the use of NUREG-1563 t
- The staff believes that it is important for DOE to respond to its three recent letters so that a dialogue can begin on how these questions might be addressed.
9 !
l
O O O .
PREVIOUS STAFF CORRESPONDENCE
- In its December 26,1996 letter, the staff recommended:
A path to resolution of SCA Comment 3 on grounds somewhat different from those recommended earlier by DOE (i.e., Appendix E to NUREG-1563) t A basis for closing 5 other SCA and study plans open items Staff awaits confirmation that DOE agrees to the course of action describedin Appendix E or an equivalent course of action e
in its December 31,1996 letter, the staff identified some concerns with respect to how NUREG-1563 might be implemented:
Circumstances under which the results of an elicitation would need to be re-evaluated Documenting changes to initial elicited judgments - how, why, and to what extent l
Develop internal procedures / guidance on the use of the technical facilitator/ integrator Inasmuch as DOEis proceeding with implementing NUREG-1563, the staff awaits a response to its letter 10 I
r3 O O .
PREVIOUS STAFF CORRESPONDENCE ,
(continued) e in its Januuy 7,1997, DOE requested informationibackground with respect to how it might deal with issues related to documenting and disclosing potential conflict of interest (COI).
NUREG-1563 does not prohibit the Department from using its own experts in the conduct of any formal expert elicitation However, the guidance does recognize that the Department needs to address the potential for bias and COI in expert selection Because DOE has noted its desire to build a strong licensing case, the staff believes that it is in DOE's best interest to develop procedures / guidance on how it will address the notential for, the p_ercention of, or the mitigation of bias and COI '
i b
11 !
t
.__----_.-.____-_-_.---_-___-_____-_-______-_-_w--- . --_ _ _ . _ - - _ - - - . _ _ . _ - - ~ ~ - _ _ _ _ . - _ -
O O O -
FUTURE STAFF ACTIONS -
- Staff will contir.ue to monitor DOE's implementation NUREG-1563.
Encourage a timely response to staff's previous letters Formally meet with cognizant DL E staff and identify the new questions / comments / ,
observations regarding the elicita-ions now underway
- To the extent that resources permit, staff will continue to conduct investigations and analyses in those areas of the HLW program that it believes are the most important to '
public health and safety.
Important for staff to have independent understanding of the site and of potential phenomena that may effect performance Consistent with staff's strategy to allow for the early identification and resolution, at the staff level, of potential licensing issues Allows to the staff to be informed and knowledgeable when it conducts its reviews of DOE pre-!icensing and licensing activities (i.e., site suitability, the Viability Assessment, and the license application) 12
~
O O O YUCCA @
MOUNTAIN PROJECT
_ _ _ A .x. , <.
Introduction to Project Integrated Safety Assessment k
Presented to:
Advisory Committee on Nuclear Waste Presented by:
Carol L. Hanlon ..
Yucca Mountain Site Characterization Office -
T 1 U.S. Department of Energy !
g7 Office of Civilian Radioactive Waste Management j
O . O O .
Purpose of Presentation i
Introduce Project Integrated Safety Assessment to the Advisory Committee on Nuclear Waste
- Briefed to NRC Managers on April 30,1997
- Briefed to YMSCO Project Manager on May 5,1997 l
- Management Plan including Contents Guide scheduled for completion end of May,1997 t
I ACNWHANLOR125 PPUS5-21-97 g I
. I i
Project Integrated Safety .
Assessment: Background FY 96 comprehensive replanning effort in response to budget constraints Civilian Radioactive Waste Management Program Plan revised to streamline and focus the program Development of integrated safety assessment considered essential ,
Assumption and support built into multi-year planning base i
ACNWHANLONM25.PPTM5-21-97 3
What is The Project Integrated Safety Assessment?
Internal vehicle for presenting integrated information on scientific programs, design, performance assessment, and health and safety NOT part of the Viability Assessment process
- However, sections of the PISA may be available in the same time frame as the V!:Ality Assessment PISA will support the Environmental Impact Statement, and Site Recommendation Process in conjunction with information exchanges with
' Commission Staff q
PISA may be used to initiate preliminary sufficiency ,
l comments l
-- ,_ m ,
4
Project Integrated Safety Assessment: Purpose Integrate effort of scientific programs, design, performance assessment, and health and safety Assemble and begin to evaluate the Department's safety case j Assess information currently available Identify strengths and weaknesses j i
ACNWHANLON.125 PPT 7f05-21-97 5
Outline for Project Integrated Safety Assessment
- Allows for presentation of concise, coherent integrated safety case Corresponds to elements of Safety Analysis Reports prepared by the industry for successful reactor license applications
- Strong consideration of Format and Content Guide for the High Level Waste Repository License ~
Application Consideration of 10 CFR 60.21(c)
AcNWHANLON 125. PPT 7/05-2197 g
o
~
O O -
Project integrated Safety Assessment : Outline i Chapter 1: Introduction and General Description of Site and Repository Concept Chapter 2: Considerations Relating to important Fe'a tures of Natural and Eragineered Systems Chapter 3: Site Characteristics Chapter 4: Repository Design l
Chapter 5: Waste Package Design Chapter 6: Engineered Barrier System Design Chapter 7: Radiological Safety Assessment of the Repository through Permanent Closure j Chapter 8: Performance of the Repository after Permanent Closure !
Chapter 9: Radioactive Waste Management Chapter 10: Radiation Protection Chapter 11: Conduct of Operations ACNWHANLON.125 PPT 7A)S 21-97 L 7
i Project Integrated Safety Assessment:
Focus for Content Guide Information to be focused, with clear purpose
- Guidance on specific details furnished
- Relevance of information to design and system i performance established '
- Sections of Chapter 3 Indications of compliance included
- Details of compliance case to be presented in License Application !
Correlations made between individual chapters / sections of the PISA and the document as a whole -
j i
-Inclusion of thorough cross-references among chapters / sections ACNWHANLON.125 PPUMS-21-97 g
.i
Project Integrated Safety Assessment:
correlations Waste Containment and Isolation Strategy ,
Hypotheses
- NRC Key Technical issues '
- Consideration of KTis being built into PISA framework
- Consideration of KTis will evolve with the issue Resolution Status Reports Format arad Content Guide, LARP, and 10 CFR 60.21(c)
AcNWHANLON.125 PPT 7/05-21-97 g I
~
O O O .
Project Integrated Safety Assessment:
Relation to License Application PISA provides basis for Safety Analysis Report of License Application
- Detail and substantiation to be expanded in the License Application PISA Content Guide being incorporated into Technical Guidance Document for License Application Comments received on PISA will be reflected in License Application ACNWHANLON.125. PPT 7mS-21-97 10 !,
- _ - - - - - _ - - - - - - _ - - -- _- -_-L
~
- O O O -
i'
- Project Integrated Safety Assessment:
Relation to License Application (continued)
License Application will include additional Chapters
- Overall Introduction j
- Quality Assurance ,
- Performance Confirmation ,
- Land Ownership and Control
- Organizational Structure i
- Assessment of Compliance with 10 CFR 60
- I
{
ACNWHANLON.125 PPUMS-21-97 gj l
Project Integrated Safety Assessment:
Sections of Chapter 3, Site Characteristics Geography and demography Nearby industrial, transportation, and military facilities Climatological and meteoroligical systems Hydrologic systems Geochemical system Integrated natural system response to thermal loading ACNWHANLON,125. PPT 7A)5-21-97 g 7
Project Integrated Safety Assessment:
Schedule for Preparation June 1997 - Distribute Management Plan and Content Guide, Writer's Guide December 1996 - Development of Chapter 3, Site Characteristics began October 1997 - Development of remaining 10 Chapters begins j
=
Early Spring 1998 - CRWMS M&O comprehensive review of individual chapters / sections
- Late Spring 1998 - DOE Formal review of individual i
chapters / sections j
=
Summer 1998 - Consolidate document I
- Fall 1998 - Complete PISA effort FY99 - Begin preparation of License Application AcNWHANI.ON.125 PPUMS-21-97 13
_-_---_-----_----__-_----_--L
~!
4 O O O -
Project integrated Safety Assessment:
Guidance Developed for Preparation
- Management Plan
- Discusses process of preparation
- A - PISA Content Guide
.
- B - PISA Writer's Guide
- C - Section Managers List PISA Content Guide
- j
- Defines contents of PISA Chapters / Sections
- - m s m , m .. , y j
1 l Project Integrated Safety Assessment:
Conclusions PISA effort has already proved important as an integrating and clarifying tool PISA expected to be valuable in defining safety case and shaping License Application We expect the PISA
- To be an important vehicle for cornmunication ;
- To afford opportunity for an early view of DOE's strategy for approaching License Application ACNWHmLON.125MTMS-2197 15
- __ ____ _ - _ _ _ _ _ _ _ _ _ _ _ _ .