ML20210E773
| ML20210E773 | |
| Person / Time | |
|---|---|
| Issue date: | 07/15/1999 |
| From: | NRC |
| To: | |
| Shared Package | |
| ML20210E754 | List: |
| References | |
| REF-10CFR9.7 NUDOCS 9907290029 | |
| Download: ML20210E773 (242) | |
Text
.
e'f T i I I
C ,u
/g y
g
- _~
,o m %f4 ,
. a_ - e.m _.,gN p,
w ppg,wpW g g, -ny A v%)4 enpqqdhy;g%aWJh g gpM&Mf g;m p w@ T- masg k
- 3 c M eve sv-
) 9:p:
- 1. ; cm J'%
u,p bW Y n ege <
';m W: w<e px ag#gqwn%Ld:m fM ww w%yM un -3 ,7 %@
3;m & e Wp9 p .fn 4-p syWR~ <
~
o$uamamm%mpsmaemsam 1 nrunns m+ mow u, L
.j. ,
Mw p' %p t Q&g y-m % a pg, w u p kstd w$,w@&
wv WG gggggg mwwww n4 yd m 4,Qg g g%
n pp W
%u g
h @ &,n
, +%
- n. h@me v
Fy a' p w nn s M9%m w$$g@sy$#W$an@g@m@y@MW dem M a aann mw MQ b k .
i n u _h_h __f h_hh h ff NN_g#%g@m@hyhgM_
t g
[h_ RW_pS_b
,'4%q>;&hQ:rQq p wyQiQv:;s;ayyh 1
s e g g M, w e Q gm e kp . s v&--
ega..o
%.1<,
.s-)c-nh,Hm>s;g.
( s, p hhk_g%m e;% ew%a_L;y e
,j\v :v y o .a L w
n we u ,
a J M
,f<;p,<
h e 4
,- . " !;m, <
,'.m
- a
, w*
r A
M
"*'e Q1
,t m
A p.p gn ge C wg SQ@W ind[.a <af g.yu; n
~y $p%,s, -v u e m v.n y. . .
w a o s : t
'='k<
t.
c WD
,'ls %W&: 'e Sn 7
d qc q
. m.Jg vxheyp y.
a a b ph MyMk hWfWpy y,- ?g"*;g i zyc!ggM g, w
g4 . Sv+p a g J! _ _
f aM n i'
Q un " Qmww F N V Q,w w& j xa PAG
~
M ~$m ;n y a $ $ u% w ,g S $p @p @%
p t
~sL Wmgp Wgghgg ggng fhkQ3f A % p* f kgAg MV4% q#%lMQVF W rp#mG#?M'~a% V 'I$N ON W "G H
Qb&
- QQf N A Qt WNk%%Q$!y%Qh?k W MB%fW4 4 myk n wnn 7Qb
$ M e.cm hg u n%
%FerE@
y{M M$ @qwdpNy$es%q @
gomooa Wgenh@g n wagy p a>
,~
m acrR PDR WW4 pfhdQQf &MMM %?
2 M- W4" n Q MA '
lfif b. J '
yc $s. u$u.M 4
7MQ
'%n"'WQf?ppyw$y ~o S 4 Q, M~T~?lP wwg qbJ 9 q m;gdg g y.d N u.v.
% Rgn.e. m$y
- , s
(. e, pp,M 3
n
, n- ,m g.
s_, w g
- 8
~4
I ORIG NA_
OFFICIAL TRANSCRIPT OF PROCEEDINGS '
'O UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION
Title:
REACTOR DECOMMISSIONING RISK WORKSHOP O.
v Work Order No.: ASB-300-857 LOCATION: Gaithersburg, MD DATE: Thursday, July 15,1999 PAGES: 1 - 240 ANN RILEY & ASSOCIATES, LTD.
1025 Connecticut Avenue,NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
- i]' 72'9r 900727 PT9 '
pop L J
1 1 UNITED STATES OF AMERICA
[ \ 2 NUCLEAR REGULATORY COMMISSION V
3 ***
4 5 REACTOR DECOMMISSIONING RISK WORKSHOP ,
l 7
l 8
9 Holiday Inn Hotel 10 2 Montgomery Village Avenue 11 Gaithersburg, MD 12 13 Thursday, July 15, 1999 14
() 15 16 The above-entitled workshop commented, pursuant to 17 notice, at 8:08 a.m.
18 19 20 21 22 23 l
24 25
(/T ANN RILEY & ASSOCIATES, LTD.
s
,) Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
2 1 PROCEEDINGS
() 2 -
[8:08 a.m.]
3 MR. CAMERON: Good morning everybody. My name is l 4 Chip Cameron. I'm the special counsel at the Nuclear 5 Regulatory Commission for Public Liaison. And I'd like to 6 welcome you to our public workshop on risk issues related to 7 the spent fuel pool at the decommissioning plants. And the 8 staff has -- the NRC staff has 0.sked me to serve as the !
9 facilitator for today's session. And in that role, I'm 10 going to try to assist all of you in having a more effective 11 meeting. And some of my goals, in that regard, are:
l 12 One, to help you keep the discussion relevant and 1
l 13 focused, and there's a number of aspects to this goal. We l 14 sant to try to keep the discussion focused on the agenda
'O
(_,/ 15 topic that we're on at the particular time. There may be 16 other items that surface that are more appropriate for 17 discussion in another part of the agenda or perhaps even in 18 another process that we might use. I'm going to keep track 19 of those items on that flip chart, which I'm going to call 20 the " paddock." You may have heard that referred to as the 21 " parking lot." But, we'll circle back, from time to time, to 22 make sure that we pick up those issues at the appropriate 23 time.
24 Another aspect to the focused and relevant goal is 25 to try to develop what I call discussion threads on a ANN RILEY & ASSOCIATES, LTD.
\ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
l 3 1 1 particular topic, rather than an unrelated monologue, where '
2 we just jump from one issue to another. So, we'll do that 3 to-the extent practical. That may be -- that may be a 4 challenge.
5 Now, my second goal would be to make sure that our 6- discussions are as clear as possible. So, I will be asking 7 you, from time to time, to explain your statements and also l 8 to provide a rational for your conclusions, so that l
9 everybody can understand the basis for what you're saying.
10 A third goal is to develop solutions to the issues j
l l 11 that we're going to be talking about. So, I would urge you q l
12 to listen to each other and try to develop constructive l 13 comments on the issues today. And there are going to be )
14 several places on the agenda that we're going to summarize 15 any actions that should be taken by the NRC, by the 16 industry, possibly by others. And I will be keeping track 17 of those on these flip charts up here. And I'm hoping that 18 we'll have these charts -- we'll have something on these 19 charts by the end of the two days.
20 And a final goal is I want to make sure that all l 21 of you have an opportunity to speak, including the audience.
22- And we will be going out to those of you in the audience, 23 who may have a comment on the discussion topics at the end 24 of each topic. The focus of the discussion is going to be 25 .up here at the table. And I think it's particularly O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l_ Washington, D.C. 20036 (202) 842-0034
l 4
l 1 important to note that we do have NRC presentations on es
(\s) 2 issues; we do have some folks from the industry that are l 3 going to be making presentations on the issues and, of l i
1 4 course, they're going to be in the discussion. But, we do 5 have representatives of the public, citizen groups here with 1
6 us today, and we want to make sure that their thoughts, '
1 7 their perspectives are reflected in our discussions today.
8 In a minute, we'll be going around and doing some i 9 introductions and you'll be able to get to meet everybody. l 10 Ground rules are fairly simple. You have name 11 tents. These are, you know, name tents. That's something 12 that is sort of an obscure fact. But, you have those in l l
13 front of you. Hopefully, it has your name on it. But, even 1 14 if it doesn't, if you want to say something, okay, put your g^N q,) 15 name tent up like that. That relieves you of the burden of 16 having your hand up all the time and it lets me know who 17 wants to speak. And what I'll do is I'll call on you and at 18 the beginning, if you could just give your name, until John, 19 our stenographer, gets to know you and where you're seated 20 around the *able. This will help us get a clean transcript 21 of this meeting. We are keeping a transcript. That 22 transcript will be available to the Commission for their 23 deliberations. And in a minute, we'll be talking about how 24 information from this meeting is going to go to the 25 Commission.
O/
ANN RILEY & ASSOCIATES, LTD.
's Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
- ]
5 1 And when I go to the audience, if you have IL )\ 2 something to say, you can either go to one of these standing 3 mikes or I'll bring you this talking stick and just state 4 your name and affiliation, if appropriate; let us know where 5 you're from.
6 We have a little bit of a challenge that I'm sure 7 we can work with at the front table. The black mikes are 8 the mikes for the PA system here and the other microphones 9 you see only go into the stanographers recording. So, you 10 need to speak into the black microphones. lau'll see that 11 there aren't really that many mikes around the table, then.
12 So, we're going to have to be doing some mike passing. And 13 John suggested that we pass the hat for contributions in the 14 audience to get some more money to get some more t'
(%,) 15 microphones. So, if you see a hat in front of you, just 16 throw, you know, five or ten dollars in, whatever you want.
17 But, at any rate, just to give you an agenda 18 overview, before we get t.o introductions, we're going to 19 start off with some opening remarks from Stu Richards and 20 from Mike Meisner. Gary Holahan, who is up in the front 21 here, is going to give us a few minutes on how we get to 22 this -- to this meeting. And then, we're going to get into 23 the first substantive discussion area, which is called 24 sequences for long -- sequences for long term and 25 intermediate events. Glenn Kelly from the NRC staff is
( ANN RILEY & ASSOCIATES, LTD, Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
6 1 . going to do a presentation for us and we're going to have an
( 2 industry perspective on this from Ed Burns. Then, we're 3 going to go to open discussion after a break.
4 And this beginning session is to talk about the 5 NRC study and methodology, assumptions, over arching issues.
6 And when we get there, Rich Barrett from the NRC is going 7 to, what I call, tee the issue up for us and just describe 8 what that session is supposed to encompass, in terms of 9 discussion points.
10 At 1:00, we start a session called mitigation of 11 events. In other words, once you've -- once we've talked 12 about sequences, enen what types of mitigation activities 13 could be applied to these particular events. And we're 14 going to start with &n industry presentation on that. And
() 15 we're going to have open discussion for most of the 16 afternoon and then get to a summary about 3:30. And then 17 we'll have some closing remarks at 4:00 by Stu Richards and 18 also maybe a preview of the next day for you.
19 And as you can see, day two, we're going into some 20 specific areas: seismic. We're going to do two breakout 21 sessions tomorrow afternoon, one on heavy loads, one on 22 thermal hydraulics. And we do have a summary and follow-up 23 session at the end of the day, and this is where we might 24 discuss how the NRC is going to be using the information 25 from this workshop; how people who comment will be able to O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034 i
7 1 I see how their comments have been evaluated and dispositioned
() 2' by the NRC; what are the regulatory applications of this 3 information; and how is this going to be integrated into a 4 larger decommissioning regulatory context.
5 With that, let me go out to you and -- for some 6 introductions. And if yvd could just tell us your name and 7 where you're from and one -- what your concern or interest 8 in this particular issue is. And let's start with Gary --
9 well, let's start with Vonna, I'm sorry.
10 MS. ORDAZ: I'm Vonna Ordaz with the NRC and the 11 leader of the technical working group.
12 MR. HOLAHAN: I'm Gary Holahan, the director of 13 the Division of System Safety and Analysis in NRC, and most 14 of the technical people working on the working group work 15 for me.
16 MR. ZWOLINSKI: I'm John Zwolinski. I'm the 17 director of the Division of Licensee Project Management.
18 We're responsible for actions and activities moving forward 19 in the decommissioning area, including possible rulemaking 20 and integrating decommissioning activities.
21 MR. RICHARDS: I'm Stu Richards. I'm with the NRR 22 projects and I'm primarily interested in getting a lot of 23 feedback in the session here, because we're working on 24 rulemaking in the future and trying to consolidate rules 25 that apply to decommissioning and establish a framework that O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
8 l' ought to make it easier to go through the decommissioning
() 2 3
phase.
MR. MASNIK: I'm Mike Masnik. I'm the chief of i 4 the Decommissioning Section in Projects and I'm responsible ,
5 for the day-to-day oversight of those reactors assigned to 6 NRR that are in the decommissioning process.
7 MR. HANNON: I'm John Hannon, Plant Systems branch 8 chief. I work with Gary Holahan, NRR.
9 MR. GUNTER: My name is Paul Gunter. I'm director 10 of the Reactor Watchdog Project for Nuclear Information and 11 Resource Service here in Washington. I would just take a 12 moment just to submit a request from two citizen's groups 13 that have been principle in litigating the decommissioning 14 issue, that's the New England Coalition on Nuclear Pollution 15 and the Citi ens Awareness Network. Their request is that 16 future public meetings on the decommissioning subject, they 17 would certainly appreciate timely invitations to participate 18 in these -- in this dialogue.
19 MR. LOCHBAUM: Dave Lochbaum, nuclear safety 20 engineer for the Union of Concerned Scientists.
21 MR. NELSON: Thanks, pass the baton. Alan Nelson, 22 NEI, looking forward to working and resolving many of the 23 solution -- finding many of the solutions and risk informed 24 to provide decommissioning regulations.
25 MR. MEISNER: Hi, Mike Meisner. I'm with Maine ANN RILEY & ASSOCIATES, LTD. ;
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
9 1 Yankee and I chair the NEI Decommissioning Working Group.
() 2 MR. BURNS: Ed Burns from Merin Engineering, 3 supporting the industry review of the draft NRC report.
4 MR. PARRY: I'm Gareth Parry. I'm an advisor for 5 Probabilistic Risk Assessment in Gary Holahan's division.
6 MR. CHEOK: I'm Mike Check. I work for the'NRC 7 and I work on the Risk Assessment for Spent Fuel Pools.
8 MR. KELLY: My name is Glenn Kelly. I'm in the 9 Branch in NRR. I work for Rich Barrett.
10 MR. BARRETT: I'm Rich Barrett. I'm the chief of 11 the Risk Branch in NRR. And what I'd like to achieve in 12 this meeting is a sound technical basis for future decisions 13 that all of our stakeholders can understand and accept.
14 MR. CAMERON: Thank you, Rich, on a good note.
() 15 Keep in mind that we are going to have other people coming 16 u m to the table for some of the other discussions, people 17 with expertise in that area, and we'll introduce them when 18 they come up. We already have one action item brought up by 19 Paul Gunter, which is to involve Citizens Awareness Network, 20 Paul, and New England Coalition on Nuclear Pollution in any 21 further discussions we have on this particular issue.
22 Right now, let's go to Stu Richards for some 23 opening remarks, and then to Mike Meisner. Stu, do you want 24 to use this one or do you want to try that?
25 MR. RICHARDS: Oh, no, I'm confused.
ANN RILEY & ASSOCIATES, LTD.
/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
)
I 10 1 MR. CAMERON: Try that one. That one works.
() 2 3'
MR. RICHARDS: Does this work? Oh, all right.
I'm Stu Richards. I'm the NRR projects director 4 for Region 4 and decommissioning. Can you hear me all 5 right? I've been asked to make a few comments and then pass 6 it on to Mr. Meisner.
7 For those who haven't been involved, this process 8 has been going on for several months, so I'd like to give 9 you a little background. I'd like to thank everybody for 10 being here for next two days. I think this is going to be 11 an important workshop, to hopefully establish a framework i
l 12 for regulatory oversight in the decommissioning area for the '
13 years to come.
14 A little history: back in March, we met with the
() 15 Commission, as part of the NRR reorganization. We told the 16 Commission that we wotid take a pause and take a look at l 17 where we have been in the decommissioning area, as far as i
18 regulatory oversight. There were a number of rulemakings 19 that were underway and we put a hold on those rulemakings, 20 the intent being to make sure that we've considered all the 21 experience we've gained over the years with decommissioning 22 plants, take a look at the various rulemakings and make sure 23 that we were going to integrate that effort to some kind of 24 a common goal in the future.
25 So, what we told the Commission that we would do )
I h
v ANN RILEY & ASSOCIATES, LTD.
Court Reporters l 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 l
(202) 842-0034
11 1 is that we would undertake kind of a two-prong approach.
(a) 2 And as such, we put together two groups -- two working l 3 groups, if you will. One working group is under my 4 responsibility. That working group, being led by Bill 5 Huffman, has the job of taking a look at all the regulations 6 that apply in the decommissioning area and the way they are 7 laid out in the 10 CFR; to make recommendations as to how we 1
8 could better package those regulations, to make them clear 9 to the industry and public; and hopefully make it a simpler, 10 more efficient process to deal with, as plants go from the l 11 operation phase to the decommissioning phase. That 12 particular group's efforts is not a topic of this workshop 13 today. So, that's a little background material.
14 The second working group is under Gary Holahan, a
() 15 so-called technical working group being led by Vonna Ordaz. l l
16 Their job was to go -- or is to go out and look at the risk 17 associated with storing spent fuel at spent fuel 18 decommissioning plant and, hence, the workshop here today.
19 We told the Commission we would try and do this in 20 a public way. Up to this workshop, we've had four or five 1 21 public meetings. They've been much smaller than this. Many l
I 22 of the people at the table here have been at all or most of 23 the meetings, and I think are familiar with that 24 information. But, this workshop is kind of a cumulation of 25 these four or five meetings, working towards trying to come i ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 i
t
p l
I 12 1 up with ideas on the preliminary results of the technical
() 2 3
working group and their thoughts on risk.
What's the goal today? We've asked the technical 4 working group to provide us -- us and Projects a product, 5- which is basically their assessment of the risk associated 6 with the decommissioning plant. They've produced a f
l 7 preliminary report and we're looking for feedback on that.
8 Ultimately, what we want to do is we're going to get the 9 product from the technical working group, and the people in 10 Projects that work for John Zwolinski and myself are going I l 11 to take that and factor it into rulemaking in the future.
12 So, that's the ultimate goal, I hope, for this workshop 13 today.
l l 14 We did send up to the Commission a SECY paper, 15 dated June 30th. I think there were copies made available 16 at the front door, when you came in. If you don't have a 17 copy of that paper, it's99-168, and it gives a summation of l
l 18 pretty much what I've just said. I think it's a pretty 19 read. The front of it is only eight or nine pages. So, if 20 you're looking for some background on what we're doing, 21 that's a good place to go.
22 I'd like to point out a few of the people, who are 1
23 here with us today. My boss, John Zwolinski, is here.
24 John's boss, Brian Sharon is in the audience. We have Sam 25 Collins, the director of NRR, who intends to be here ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 I
l 13 1 tomorrow afternoon for the wrap-up session. I think the
( ) 2 decommissioning area has gotten the attention of senior 3 management in the NRC. I think it's gotten the attention of 4 the NRC Commission, as a whole. So, this workshop and what j 5 we do here today is getting a lot of visibility and I sure 6 appreciate all your participation.
7 Mike?
8 MR. MASNIK: Thanks, Stu. I think one thing we 1
1 9 can all agree on, everyone in the room, is a common goal 10 associated with this workshop. And regardless of what your l
11 leanings are and your point of view, I think we all need to j 1
12 have a better risk informed understanding of 13 decommissioning. That's very important for licensees and it 14 really goes to the issue of resources. We all have scarce O
( ,/ 15 resources, as does the NRC. The more we can focus those 16 resources on what's important to safety during 17 decommissioning, the better off we all are. That's kind of 18 the litmus test or benchmark we will be using, as we go l
19 through this workshop, to see how we're progressing.
1 20 Up to this point, as Stu indicated, we've had a 21 number of meetings with the staff. And, you know, on some l
22 cases, those meetings were a little contentious. But, in 23 looking back over what we've accomplished so far, I think 24 it's real important to point out that the framework -- the 25 analytic framework that the staff has put together in their l;
f;
\/
ANN RILEY & ASSOCIATES, LTD. l Court Reporters i 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
14 1 draft report is actually an excellent approach and way to 2 go. When we look back over the discussions we've had in the (O}
3 past, what we really come down to, some area of disagreement 4 on only two issues: one is how do we calculate the 5 probability of a catastrophic seismic event; and, two, how 6 do we treat and deal with the potential for human error in 7 evaluating these sequences.
8 I think that's something that we're going to focus 9 most on in this workshop. From discussions I've had this 10 week and past week, I think we've got a real opportunity 11 here to come to some constructive resolution of both those 12 issues. And I'm not minimizing the fact that there are 13 other issues to discuss, as well. But, I think in the minds 14 of the industry, those are the two key ones. So, I expect .
(~}
(_j 15 today and tomorrow, we're going to have a very positive 16 constructive dialogue.
17 And I just want to finish by saying that we've got 18 a lot of industry folks at the table and out in the 19 audience, and a lot more that have been participating all 20 along. And one thing I don't think we've ever mentioned is 21 that the majority of those participants aren't going to 22 benefit -- their plants aren't going to benefit at all from 23 what we're going here today, from what the NRC will do in 24 the future in rulemaking. We're trying to make a positive 25 contribution to the future of the industry and make things
~% ANN RILEY & ASSOCIATES, LTD.
[k -
m Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
15 1'
1 go easier for the next guy down the road. Because, frankly,
() 2- most of our plants, the ones that are actively 3 decommissioning now, are well beyond the issues that we'll 4 be discussing today.
5 What we do hope to accomplish, in investing our 6 time and effort in this, is to hopefully come up with a 7' model for doing the same thing for other areas of 8 decommissioning. Today, we're treating the front end, you 9 know, the issues that plants and regulators have to face in 10 their first year of decommissioning. There's the back end 11 of the process, too: terminating the license, site release 12 criteria, dry cask storage and the like. And we're hoping 13 that this kind of approach can serve as a model for risk 14 informing those areas of decommissioning, as well, and r\
(_,/ 15 perhaps even provide some benefit for doing the same thing 16 for operating reactors.
17 Thanks.
18 MR. CAMERON: Thank you, very much, Mike. Let's 19 go to Gary Holahan now, who is going to give us some 20 background on the study and how we got here today. Gary?
L 21 MR. HOLAHAN: I think both Mr. Meisner and Stu 22 Richards have laid out a pretty good understanding of why 23 we're here already. I think Stu covered historical reasons 24 and I think Mike began to touch on some of the technical 25 issues.
l O' ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 L
16 i
)
1 What I'd like to do is spend just a few minutes f
(
v 2 talking broadly about how we decided to approach this issue, 3 After this point, we'll sort of -- we'll sort of break up 4 the technical issues and sort of try to digest them almost 5 one at a time, and that's the way most of the agenda is set 6 up. So, I'll spend a few minutes talking about how we got 7 here in a technical sense.
8 If you turn the clock back six months or so, what 9 we found ourselves -- the position we found ourselves in is 1
10 the being asked to make technical decisions about issues 1 11 related to spent fuel pool decommissioning and whether it's 12 emergency planning or other issues. It required an l
13 understanding of the safety of spent fuel pools during I 14 decommissioning. And what we noticed was that there had C\
() 15 been lots of decisions made before, but they were made 16 really on an ad hoc basis. Each licensee had asked for 17 relief or for changes in some context, under their special 18 circumstances, for the issues they had a particular interest 19 in. Information was brought to the staff. It appeared 20 reasonable or there was a discussion and technical decisions 21 were made. But, they were very much ad hoc and it wasn't at 22 all clear that there was a systematic approach to making 23 decisions that would really provide a broad technical basis 24 for going ahead and doing rulemaking or having a really well 25 understood set of ground rules for both licensees, for the O'
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
17 1 public, and for the staff, to understand how to deal with 2
I
\
[\_)\ future issues. <
I 3 So, as Stu mentioned, we decided to take a step 4 back,' rethink where we technically, to put together a sound 5 technical basis for decision making. And what we realized 1
6 is we are, you know, surrounded by other issues going on in l l
l 7 the Commission. .The Commission is really in the process of 8 risk informing many of its activities. We wrote a policy 9 statement a few years ago. We put out guidelines on how to 10 deal with operating reactor license amendment type issues.
11 The Commission recently approved us to go through and modify 12 the reactor regulations, as a whole, to put more risk 13 information in that process. The Commission has just 14 started a pilot project to deal with operating reactors in
() 15 our oversight inspection and enforcement activities, to use 16 more risk information. So, it seemed ripe to take on the 17 issue of decommissioning and to use the tools, the 18 techniques, the knowledge we have of risk assessment 19 approaches and to see how they could be applied to 20 decommissioning.
21 Well, what sounds good in the philosophical way, I
22 in fact, sometimes turns out to be rather difficult, because 23 we find there are a lot of challenging technical issues. A 24 number of the questions that you'd like to answer 25 technically have really not been, you know, discussed and ANN RILEY & ASSOCIATES, LTD.
-O Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
18 1 studied as extensively for decommissioning phase, as they
() 2 3
have been for operating reactors.
of difficult questions.
And so there are a number You know, one that Mr. Meisner 4 mentioned, and I think we'll probably spend some fair amount 5 of time on it, is how reliable are human actions over, you 6 know, long periods of time, days. For operating reactors,
? you know, days are not the issue. Hours, sometimes minutes, 8 are important in making -- in deciding whether operators are 9 making correct and reliable -- taking correct and reliable 10 actions.
11 And so, when you go back and look at the research 12 and the literature and the experience that's available, it's 13 available for dealing with human reliability, in the context 14 of operating reactors. How do we extend this? How do we
() 15 make decisions that are relevant for spent fuel pool and 16 decommissioning? Well, it's not so straightforward. And 17 so, I think that's part of the reason that these issues are 18 somewhat difficult.
19 But, I think what we realized is drawing risk 20 information into the mix is the most rational way of making 21 decisions for spent fuel pool decommissioning. And so in 22 order to do that, we rea]ly step back and say, you know, 23 where should we start. And, in fact, we decided to start 24 basically with a clean slate. And when you decided to do a 25 risk assessment from a clean slate, you start out by looking O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 ,
(202) 842-0034 1
f l
19 l
1 1 for sequences: what can happen, what can go wrong. Then, 2 you go on to look at probabilities: more probable events 3 need more protection; less probable events, less protection.
l 4 And at some point, the events seem so improbable, so remote, 5 so speculative, that, in fact, no level of protection is 6 required.
l 7 And so at this first stage, there is a lot of 8 consideration of, you know, what's relevant. So, there's 9 sort of a completeness issue. And you'll see that we've 10 raised issues that some people were uncomfortable with. Why 11 are you worried about -- about reactivity accidents? You 12 know, we didn't worry about that before. The reason we're 13 dealing with that issue is for completeness. We want to 14 have consideration of a lot of issues, many of which won't 4 15 turn out to be significant in the end, but you have to deal 16 with them.
17 So, I think you'll -- if you've looked at our 18 draft report, you'll see it deals with a spectnim of issues 19 broader than we've seen in the past on many decommissioning 20 decisions, and likely, in fact, to be broader than decisions 21 in the future, because this is our opportunity to have a 22 broad consideration of technical issues. But as part of 23 that, we have to decide what's most important, what's least 24 important, where do we need to provide protection, where is 25 it not important, so, in the future, maybe we don't have to h
ANN RILEY & ASSOCIATES, LTD.
Court Reporters
, 1025 Connecticut Avenue, NW, Suite 1014 I
Washington, D.C. 20036 (202) 842-0034
]
20 1 go through all these issues. Again, we can focus on the
, 2 things that are most important. So, part of this process 3 is, in fact -- and I think Mr. Meisner used the word 4 " focus," sometimes we even use the definition of 5 risk-informed regulation, as a definition of focusing the 6 NRC, its licensees, on those things that are most important 7 to public health and safety. So, that's part of what this 8 process is. This is the lens. We take all the issues, all 9 the light we can shed on this, and focus it down to the 10 things that appear to be most important.
11 In this process, we looked at lots of scenarios.
12 .In doing this, we draw on operating experience, on logical 13 relationship among issues, analysis. Sometimes, we're 14 challenged to do analyses that haven't been done before in 15 their realistic way. But, we try to put all of these 16 together. So, that's the stage that we are at here. It's 17 our first shot at a complete look at spent fuel pool safety 18c issues, with our first attempt at the probability and 19 consequences of those events.
20 What we'd like to get out of this meeting is a 21 review, assessment, input, evaluation of what we've done in 22 terms of completeness, have we missed issues; in terms of 23 probability and consequences, have you overestimated or 24 . underestimated the consequence of issues; are we about right 25 or are we wrong in judging the probability of issues. And ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
l I
21 l
1 in a related context, sometimes you can deal with issues not
(} 2 by just finding out what's the likelihood of them; but, in i 3 fact, doing something to intervene.
l And I'm hoping to learn l 4 something from the industry today, whether in terms of 5 information or commitments, that says some of these issues 6 can be dealt with by either showing that there are already 7 training and procedures and instrumentation or whatever 8 available to assure that good and reliable decisions are 9 made by people, that's the people part of the issue; or 10 perhaps where it's necessary to focus on important issues, 11 additional training, instrumentation, procedures, whatever 12 is appropriate to the issue, can be brought to that issue, 13 to make what might otherwise be a questionable or a 14 difficult decision a lot easier. So, I'm looking to learn
() 15 more about where'we are now or what might be done to make 16 safety issues either less significant or certainly better 17 understand in this context.
18 I think that's about where we are in a general 19 context. And I think as we go on to the sessions, these 1
20 sort of issues will be dealt with on a -- you know, on a 4
21 sort of issue by issue basis. So, Chip, I'd like to turn it i 22 back to you.
23 MR. CAMERON: Thanks, Gary. I think that was an 24 excellent context for our discussions and we'll be coming 25 back to many of those points. I think we should get right O ANN RILEY & ASSOCIATES, LTD.
' Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
~
o I'
22 1 to it, in terms of the details of the study and go to Glenn
() 2 3
Kelly for his presentation -- oh, Rich,' sorry. Rich is going to tee it up for us.
4 MR. BARRETT: You're using a golf analogy.
5 MR. CAMERON: Right.
6 MR. BARRETT: Can everyone hear me?
7 SPEAKER: Yes..
6 MR. BARRETT: Okay. I would like to also extend a 9 welcome to everyone, on behalf of the technical staff, who 10 have worked very, very hard over the last couple of months 11 to get us to where we are today. We really appreciate 12 everyone's participation in today's workshop and we look 13 forward to a very successful day, based on what we've been 14 able to accomplish so far and the discussions we've had in
( 15 previous meetings. I 16 I'd like to point you -- bring your attention to 17 the staff's slides, which we'll be using in the first ,
)
18 session, which originally was envisioned as being two 1
19 sessions. And I will be briefly talking to the first two l 20 slides, slide number two and slide number three. Some of 21 what I plan to say is the predators of what Chip and Stu and 1
22 Mike and Gary have said. But, if you live in Washington 23 long enough, you'll know that that's not at all unusual.
24 Some years ago, Congressman Mo Udall stood to speak in the i
25 House on an important issue and he was one of the last '
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025' Con,tecticut Avenue, NW, Suite 1014 Mashington, D.C. 20036 l (202) 842-0034 1
I
1 23 1 speakers and he started his remarks by saying, "Everything 2 that needs to be said on this subject has already been said, 3 but everybody ain't said it yet." So, please forgive me, if )
4 I repeat some of what's been said before.
5 The NRC, in cooperation with NEI and other 6 stakeholders, has embarked on a process for resolving the ,
1 7 risk issues associated with decommissioning. And I think '
8 it's important to understand the process -- the thought 9 process that we're going through here, and I think this is 10 somewhat repetitive to what Gary said, what we wanted to do 11 was a complete analysis of all the risk implications of 12 decommissioning and put that on the table. That was step 13 one. The second step is, based on that information, to have 14 a full and open discussion of these issues by all parties, 15 and that's where we are today. That's where we've been for 16 the last several weeks. And our hope is that the third 17 phase will be that this results in a sound technical basis 18 for future decisionmaking, that all of our stakeholders can 19 understand and accept.
20 Now, the NRC has started this process by 21 performing what I'11 call a scoping analysis. It's a 22 preliminary, and I emphasize the word preliminary, risk 23 analysis intended to determine which accident sequences are 24 credible, which ones have the potential to be contributors 25 to risk. And we've identified four general types of ANN RILEY.& ASSOCIATES, LTD.
Court Reporters l 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 I
24 1 accidents that we're interested in and that -- those
() 2 accidents are, first of all, loss of heat removal accidents, 3 which we are referring to here as long-term accidents, 4 because they evolve very slowly. The second tend to be what 5 we call loss of inventory type of accidents, and those are 6 what we're referring to as intermediate term accidents, I
7 because they tend to evolve somewhat more quickly, but not i 8 very fast. The third are accidents associated with heavy 9 loads. These are accidents, which can cause damage to the 10 pool. And the fourth major category are seismically induced 11 accidents. And as you know by now, this entire workshop is 12 organized around these four categories, pretty much.
l 13 We have documented our assessment for all the !
14 participants to examine. We documented it in writing in our
) 15 preliminary report, and we've also had discussions on at 16 least two occasions in public meetings in the last month or l l i 17 two. The preliminary report documents our methods, our '
! 18 assumptions, and our results. We have conducted what we 19 think is a useful risk analysis, in a very short period of l 20 time. And I appreciate the kind words that Mike said 21 earlier about the work that the staff has performed over the i
22 past two to three months. A lot has been accomplished.
23 Nevertheless, we recognize that our results are based on 24 assumptions that may not represent the reality for future 25 plants going through decommissioning. And I want everyone
( ANN RILEY & ASSOCIATES, LTD.
kJ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
25 1 to focus on that. What we are concerned about is future 2 plants going through the_ decommissioning process.
3 Our report points out that there is no --
4 currently no plant out there, which matches the assumptions 5 that we have made. We have based our assessment --
6 preliminary assessment on what we saw at existing plants.
7 But, we're not focusing on existing plants. What we want to 8 look at is, what is the basis for decisionmaking for future 9 plants.
10 We have requested feedback from industry and other 11- stakeholders, to help us to refine our assessment. We've i
12 .already received some very useful feedback from industry in l 13 the seismic area, and we expect to get other very good )
l 14 feedback today on all of these areas. We, also, anticipate 15 today a free and open exchange of information and of views 16 for these next two days. We are not here to defend the 17 preliminary analysis. The preliminary analysis has already 18 served its purpose. We are here to refine and improve this 19 assessment, based on the best technical information that all j 20 of our stakeholders have to offer. That's our goal.
21 Our expectation from the workshop is that it will 22 give us the basis for finalizing the assessment, in a way 23 that incorporates these insights. We intend to use this 24 information to derive a consistent and predictable basis for 25 future plant-specific decisions for decommissioning plants, O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
26 1 including exemptions as appropriate.
() 2 We want to produce guidance in two areas. We want 3 to produce guidance on how a licensee should analyze the 4 risk associated with decommissioning. We want licensees to 5 understand what constitutes an acceptable analysis for 6 future decisionmaking purposes. And secondly, we want to 7 come out of this workshop with an understanding of what 8 design and operational features a future decommissioning 9 plant should consider -- should consider, in order to 10 justify exemptions or other regulatory actions.
11 Let me not dwell on this. Let me finish by saying 12 that an important goal of this effort is to assure that all 13 stakeholders understand the basis for our decisions; very 14 importar.: to us. We are going to be discussing some highly
() 15 techni. cal issues over the next couple of days in some highly 16 specialized areas. Not all of us are specialists in all of 17 these areas. So, if anyone here finds that you need 18 clarification, in order to follow the discussion, I would 19 like to invite you to ask questions during the 20 presentations. We don't want anyone to get lost because of 21 something we said that isn't understood by everyone or 22 because of some acronym we've used or because of some arcane 23 reference we've made to some technical document that two or 24 three of us might understand, but not everyone is familiar 25 with. So, we invite you to ask questions. And if you're
(
O
\_)
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
\
l 27 1 not able to de that during the session, we wculd invite you
() 2 3
to approach members of the NRC staff during the breaks, to get any clarification you need, in order to follow the 4 subsequent discussions.
5 Based on all of the preparation we've had over the 6 last couple of weeks, I think that we have laid the 7 groundwork for a successful workshop. We're prepared for 8 success today and tomorrow. Again, I want to thank all of 9 you for coming and I look forward to a productive two days.
10 Glenn Kelly is going to start the discussion this 11 morning of the long-term and intermediate sequences with a 12 description of what the NRC did, how we approached this, 13 what we found to be important, and what result we got in our 14 preliminary analysis. Earlier on, we had organized this (I 15 discussion, such that we would discuss the long-term 16 sequences first and then go into the intermediate sequences.
17 But, when we got to the point of preparing our slides, we 18 realized that there was so much commonality in the two; 19 there were so many common issues and so many common 20 technical questions and assumptions that we decided to 21 combine the two. We're not going to go into all of the 22 details this' morning and we're not going to make an attempt 23 to justify the results. W- going to describe what we 24 got. I think the importL' .ng today is to hear what the 25 industry and other stakeholders have to offer now, based on O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 L . .
28 1 all of the discussion that we have made in the past meetings
() 2 3
and based on the report that we've documented in the public record.
4 MR. CAMERON: I wanted to thank you for those 5 remarks and perhaps just say one or two things before we go 6 to Glenn. Those were all important remarks, Rich, and 7 particularly the one on understanding and clarity. And I 8 would just remind people that if they don't understand a 9 particular point, to ask; and also to ask all of the people 10 around the table don't assume that everybody knows the 11 acronyms information, etc., etc.
12 And one point that you raised, Rich, that I want 13 to put over here in the paddock, because it may not be real 14 obvious to people, is the point about trying to get an 15 understanding of how licensees should perform the analysis.
16 -I'm not sure that really pops out on the agenda, so we'll 17 put it over there, so that we don't forget about it. And 18 before we go to Glenn, we do have one new participant at the 19 table. And Ray, could you just introduce yourself to 20 everybody and just, you know, tell them what your 21 affiliation is and what your concern is with this issue, 22 briefly.
23 MR. SHADIS: Briefly, my name is Ray Shadis. I'm )
24 with an organization called Friends of the Coast up in l 1
25 Maine. We're involved in the Maine Yankee decommissioning.
ANN RILEY & ASSOCIATES, LTD.
O-Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 i
29 1 I, also, am a staff person for the New England Coalition on
() 2 3
Nuclear Pollution.
the public sector.
And we're here representing, in essence, Thank you.
4 MR. CAMERON: Okay, good. Welcome. Gary?
5 MR. HOLAHAN: I'm trying out your tent rules. I 6- know it's not explicitly on the agenda, but I thought since 7 there are other stakeholders at the table and possibly in 8 the audience, before we go ahead and get specific about 9 individual issues, if there are -- if there are questions or 10 if there are different agendas or if there are, you know, 11 issues, in a general context, that people have a different 12 understand of why we're here or what needs to be covered, I 13 think I'd like to hear that from the stakeholders, in a sort 14 of general context before we go ahead and, you know, get
() 15 down to any of the details.
16 MR. CAMERON: Okay. Let's get some -- let's get 17 some feedback on what we've heard so far, keeping in mind 18 that we are going to be going into detail on some of these 19 subjects. Paul?
20 MR. GUNTER: All right. Again, my name is Paul 21 Gunter. I'm with the Nuclear Information Resource Service.
22 In reading over the materials, one of the -- one item is 23 conspicuously absent. And if we could clear this air right 24 from the beginning, I think it would be helpful.
25 The NRC has gone to great measures to look at
( ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025~ Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
)
30 1 accidental events. The question we have is what about I
() 2 intentional events. Obviously, with regard to -- it was 3 only a month or so ago that the Nuclear Regulatory 4 _ Commission, itself, experienced a bomb threat. And in the 5 course of evacuating I White Flint, II White Flint, there 6 were a number of issues that I think you all have become 1
7 aware of, potentially some people evacuating into a blast 8 zone. But, clearly, the issue of preparedness and drills, 9 it made itself apparent.
i 10 It's conspicuously absent from your review of risk 11 in this overall subject, that we haven't looked at the issue 12 of sabotage and terrorism. And from our position, we think 13 that this is a paramount issue and that it raises the 14 potential for the quickest time to fire; it raises issues
) 15 associated with the prevention or delay of mitigation; and 16 potentially raises your most serious events. But, it's 17 absent. 4 18 MR. CAMERON: Thank you, very much, Paul, and 19 we're going.to go to Gary for that. I did put that up on 20 the paddock and, perhaps, we can have a discussion of how, 21 when that might be treated. Gary?
22 MR. HOLAEAN: Let me speak a little on this issue.
23 It's an important point. And we said that completeness is 24 important and this is exactly a completeness issue. And 25 there may be others, who can speak to this, also.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 !
(202) 842-0034 !
I
31 1 There are a couple of considerations. One is --
() 2 one consideration is.~-- this subject is a little bit 3 difficult, because the Commission has told us, in 4 establishing their safety goals, that we ought to do risk 5 calculations, except for sabotage. So, it's not really in 6 the numbers. But, I think in a broader context, although we 7 don't analyze sabotage type events, we don't put 8 probabilities on what's the likelihood of a saboteur showing 9 up and doing x, y, and z. It doesn't mean that this 10 analysis is irrelevant from that point of view. In fact, 11 it's very relevant.
12 What this kind of analysis -- risk analysis, 13 including the consequence analysis and all that, it shows 14 you what the important safety issues are. And so, you know,
() 15 we don't mean this to be your guidelines for sabotage; but, 16 in fact, it's guidelines for protection against sabotage, 17 because it tells you-, you know, what'n important to protect.
18- The issues -- losing water from the pool; boiling water off 19 the pool; obviously sabotage doesn't mean causing seismic 20 events, but the fragility of the pool, itself, or its 21 structural strength -- the issues that we're dealing with 22 are exactly the same relevant issue from a security and 23- safeguard point of view.
24 If I use the analogy of what we do for operating 25 reactors, we don't -- we don't do a probabilistic analysis O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 L
32 1 of the likelihood of sabotage. But, the safeguard people,
() 2 3
the safeguard program looks at the information coming out of our risk analysis, in order to decide, you know, what's the 4 most important parts of the plant to be protected, decisions 5 about -- about, you know, what to protect and what are 6 relevant safety issues, I think will, you know, necessarily 7 be an integral part of the study.
8 So, I think the study will be relevant for 9 safeguard decisions. The piece that won't be is what is the 10 probability. The truth is we probably don't know much about 11 the probability of such things anyway. Is there something 12 technically that we will be missing, so that we can't make 13 good safeguard decisions in the future for having not done 14 it here, I think that's -- that's an appropriate question.
15 It doesn't come to mind that we're missing something; but if 16 it is, then maybe we ought to give some thought to it. So, 17 I-think if we complete the study, the probabilities in the 18 study won't tell you a lot about a truck bomb or some other 19 sabotage event, but the relevant nature of how do I keep 20 radiation and people separate, I think some of those 21 fundamental issues will be dealt with in this report. They 22 will be useful.
23 MR. CAMERON: Okay. Paul, let me ask you, before 24 we go to anybody else, about Gary's explanation on this. Is 25 that --
do you still have a point?
O ANN'RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C. 20036 (202) 842-0034
33 1 MR. GUNTER: Well, obviously, the point that I
() 2 3
want to make is that this whole issue is about building public confidence. And if, fact, there are events that 4 currently affect public confidence and safety at U.S.
5 nuclear power stations, they should be dealt with in this 6 arena. Clearly, the fact that the public is aware that 7 operational safeguard response evaluations have shown that 8 there are vulnerabilities at U.S. stations, that the agency 9 and.the industry are backing away -- from our perspective, 10 backing away from the -- for operating reactors, leaves a 11 big gap, in terms of what state of readiness we have at a --
12 for scaled down operations at U.S. reactors, with much 13 larger inventories of radioactive waste.
14 So, clearly, the question on -- the technical
() 15 issue is, I think the level of security left at stations and 16 the various measures of testing in qualifying those levels 17 of readiness and security clearly needs to be in your 18 evaluation of risk.
19 MR. CAMERON: Okay. Any other comments? Stu, do
-20 you want to respond to that?
21 MR. RICHARDS; Just very briefly. You know, the 22 idea of the security threat J have come up before in our 23 meetings. I think like Gary said, it just doesn't lend 24 itself to be risk informed. It does plant vulnerabilities 25 like you said. We told the Commission, in our paper, that O ANN RILEY &. ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
34 1 going forward, we would be integrating the various
() 2 rulemakings, which include security. I believe that's 3 mentioned in our SECY paper. So, as this security area is 4 considered, if nothing else, as allowed by the rulemaking 5 process, it will go through the public participation 6 process.
7 I'm not familiar how, you know, the security 8 safeguards aspect. impacts that to the degree available. As 9 we change the rules, you know, there will be public 10 participation.
11 MR. CAMERON: Is it -- before we go to Ray, is it 12 fair to say that -- I don't know if this -- I think it might 13 fit into the action item category, that security safeguards 14 considerations will be considered ultimately in the process (f 15 to do a decommissioning rule?
16 MR. RICHARDS: Sure. We've said that we're 17 looking -- as an agency, we're looking at the safeguards 18 rules, as they relate to decommissioning, with the intent of 19 being part of our integrated rulemaking, which, of course, 20 is a public process. But for this workshop, safeguards is 21 not really something that can be risked informed and, hence, 22 it's just outside the scope of what we've come prepared to 23 talk about today.
24 MR. CAMERON: Okay. Let's hear from Gary and then 25 we'll go to Ray. I think Gary has a point amplification on ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
35 1 Stu.
() 2 MR. HOLAHAN: I think -- it seems to me that the 3 relevant issue -- there is going to be safeguards 4 rulemaking, I understand, and on a plant specific basis, 5 there will be safeguards decisions made. This technical 6 study is not a substitute for any one of those. It seems to 7 me that the relevant question, in the context of this 8 meeting and this study, is: are we learning enough, as a 9 technical basis, to make good safeguards rules and to make 10 good safeguards decision making in the future; or is there 11 something-else we ought to be doing, you know, in a 12 technical context, so that good safeguards decisions can be 13 made later.
14 MR. CAMERON: Okay. So, that's -- do we need to f 15 -- you're sort of turning the relationship around, in a 16 sense. Is there. types of technical information that we need 17 now, in order to make good safeguards decisions later on?
18 MR. HOLAHkN: Clearly, we don't want to be -- put 19 ourselves in the position where -- we have this report, you 20 know, it's supposed to be the technical basis for rulemaking 21 and decisions in the future. And then we go on to make 22 those decisions, whether safeguards or anything else, and 23 say, well, now, I'm at this, you know, second stage, but I 24 find that the report, in fact, isn't helpful, because it 25 didn't do x, y, z. I think what Paul is suggesting is that
( ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
1 l
)
36 1 when we get to that stage, you know, we will have missed out
() 2 3
s on something that we needed to have, as support of the I foundation.
4 MR. CAMERON: Do you have anything further to add ;
5 on that, before we go to -- let's go to Ray and then we'll 6 come back to Rich. Ray?
-7 MR. SHADIS: Just a word about the issue of 8 sabotage and security. In a sense, there is risk informing 9 that is built into this. The problem, from our perspective, j l
10 is that it is subjective. It's based on best professional 11 judgment on the part of your safeguards people. But, it 12 doesn't get the test of truth in the public arena.
13 The safeguards people have designed a reference 14 bomb vehicle and, apparently, they've asc'Jned a budget to
) 15 terrorist, so that it can only purchase a limited amount of 16 explosives. And, therefore, given a certain mass of 17 concrete and rebar and so on, they really can't do any 18 damage to spent fuel pools. And, therefore, we don't really 19 need to protcct that fuel. So, we see this process, when we 20 talk to the safeguards people, that is, in some ways, 21 analogous to the process that you've used for the limited 22 number of subjects that you've chosen for your technical 23 report.
24 And that's bothersome to us. Looking at it from 25 the outside, we just see this void, when it comes to dealing O ANN RILEY & ASSOCIATES, LTD, Court Repcrters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20336 (202) 842-0034
37 l 1 with that particular subject of sabotage. It is real.
() 2 3
Somewhere along the line, somebody decided that -- within l
the NRC decided that the probabilities were there. Maybe, 4 they didn't quantify them, but they went right out and 5 ordered all -- as far as I know, all the U.S. nuclear 6 facilities to construct vehicle barriers, to provide, and I 7 hate-to use the word, some defense-in-depth, to keep people 8 away from critical areas. And that really needs to be 9 addressed, as far as the risk is concerned, to shut down 10 plants.
11 My principle subject that I wanted to raise really 12 has to do with the way that the focus of this whole process, 13 from the time that the first meeting was announced to today, 14 has continually narrowed down to what have become obvious
() 15 goals of satisfying an industry need, to get rid of their 16 insurance burden and their offsite planning burden as early 17 as possible. And I think that if you're going to do that, ,
18 all well and good. I think you've got a good start in your 19 report. But, there are a lot of scenarios. In 20 particularly, I'm thinking right now of different fire 21 scenarios: fire in -- resin container fire, for example; I 22 fire in a waste storage building; fire in a container 23 vehicle with waste stored in it.
24 Almost every plant has done some analysis on this 25 in their FSARs and has concluded that the offsite doses are O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025' Connecticut Avenue, NW, Suite.1014 Washington, D.C. 20036 (202) 842-0034 l
i
38 1 not going to, you know, trigger any emergency response.
/~%
(J
) 2 However, if you're going to go back and look again at 3 seismic risk, if you're going to go back again and look at 4 is there a fire risk and try to quantify what the 5 consequences might be, I think that we're missing something, 6 if we don't go back on these other scenarios and look again 7 to see if we can't more accurately quantify what those 8 consequences might be. I find it difficult to believe that 9 if we have a truckload of waste burning on a coastal 10 peninsular, that we're not going to have clouds of 11 particular crap floating across the bay at levels which, if 12 they're ueasured in particular places, would not trigger 13 those emergency response mechanisms. So, I -- what I'm 14 asking is that that be looked at again.
(~N
's ,) 15 The other thing that's very disturbing about this 16 whole process is that the NRC seems to have backed into it.
17 What we really have here is an initiative from the industry 18 seeking relief from having to pursue their exemptions. And 19 if this were a citizen initiative, not an industry l 20 initiative, it would be suggested to us that we enter a 21 rulemaking petition and that we specify what it is we want 22 in that rule change. And I understand that NRC staff has 23 offered to help citizens, who want to enter a rules change 24 petition. But, that really is the proper way to enter this; 25 then everybody knows, at the beginning of the game, what the l
[ ANN RILEY & ASSOCIATES, LTD.
\- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
)
39 1 . goals are, where we're headed. It's much, much easier to
() 2 judge the information, as it's brought forward, if we all 3 know where we're going at the beginning.
4 Thank you.
5 MR. C?.MERON : Okay. Thanks, Ray. We have a 6 couple of comments on that point; first, Glenn, and then 7 Stu. I did put up on the paddock board what else needs to 8 be considered, the point that you made. And perhaps that 9 will come up throughout the discussions and we can revisit 10 that at the end session of the day.
11 Glenn, you had comment, I think, that was --
12 MR. KELLY: My name is Glenn Kelly. I'm with the 13 staff. I wanted to say, Ray, you're exactly right about 14 that we did not look at being risk informed about the term 15 -- abo,t these design basis type events, such as burning a 16 building that's holding waste, because our analysis here was 17 specifically chosen to be one that was in severe accident 18 space. We were looking here to see whether there -- the 19 potential was high enough at these decommissioning plants 20 for very severe accidents that could potentially cause a 21 large number of fatalities. We wanted to look at that and 22- determine whether that was a sufficiently high potential 23 that we really were going to be concerned about it; or
-24 . whether the risk was there, it was low enough such that 25 exemptions could be readily allowed. So, wa did not attempt O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut. Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
40 1 to risk informed the deterministic aspects of the design e' .
2 basis type of evaluations.
(N) 3 In the future, that may be an area that we get l 4 into for looking at risk inform in that area. But for right 5 now -- because those are not things -- even though they may l
6 -- if you burn one of these containers, and I don't have the 7 detailed knowledge to know what kind of source term you're 8 going to get, but it's my understanding that it's not going 9 to raise to the level where you're going to get fatalities; j l
10 and, therefore, you know, in our particular situation here, l l
11 you know, we haven't looked at it, because we were very 12 focused in what we were looking at.
13 MR. CAMERON: Okay. We'll revisit this issue, as 14 we go along, and perhaps later on in the day -- or later on
() 15 in the meeting. Let's go to Stu and Gary and then let's 16 finish up with Ray and get onto Glenn's presentation. Stu?
17 MR. RICHARDS: Yeah, thanks. I welcome your 18 comments, Ray. You know, you've been at most of our 19 meetings, if not all of them, so I'm glad to see you made it R20 to this one. I do have to take some exception with some of j 21 your comments.
22 First off, I think you said that we backed into l
23 this and that the industry has put us up to it. I got 24 involved in decommissioning just a few months ago. My 25 knowledge is, is that it's not the truth. That's not what
[ \ ANN RILEY & ASSOCIATES, LTD.
\s- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036
( (202) 842-0034 l
41 1 happened. Under Chairman Jackson, you know, she challenged
() 2 the staff, generally, to work with the regulations and she 3 put out some pillars. She wanted us to be consistent, 4 predictable, efficient, public confidence. And when we 5 looked at decommissioning, we looked at an area, where it 6 seemed like each plant would come in on an individual basis 7 and would make, you know, requests for exemptions and they'd 8 all be handled on a case-by-case basis. It was getting to 9 where, you know, we couldn't very well compare what we did 10 two years ago to what we're doing today. The regulations 11 are kind of spread out throughout Part 50, which was really 12 written for an operating reactor.
13 So, our intent here is not to make the industry 14 happy. Our intent is to carry out a mandate given to us by
( 15 Chairman Jackson, which was to try and make the whole 16 process a more predictable and easier to understand. And, 17 in fact, I think we've gotten some positive comments from 18 members of the public, that's exactly what we should be 19 doing, trying to make this thing more clearer.
20 And as far as spelling out where we're trying to 21 go, we put it down in SECY paper to the Commission. I think 22 we've been very timely in making that available to the 23 public. They haven't voted on it, although we've asked --
24 you know,.the staff has asked for comments from the 25 Commission by tomorrow. They don't always meet those time i ANN RILEY & ASSOCIATES, LTD.
[~'/
\-- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
I
42 l l
1 lines. But, nonetheless, our SECY paper lays out very e'%
( ) 2 plainly where we're trying to head.
/
3 And we're not prejudging here, you know, what the 4 risk are. That's the whole part of the workshop. So to the 5 degree that you or any other stakeholder can demonstrate 6 where we're wrong about assigning a level of risk to 7 decommissioning activities, we welcome those comments. On 8 the other hand, I think, as others have pointed out, I don't 9 think we should be putting a lot of resources in areas where I
10 it simply isn't warranted. Those are resources better spent l 11 elsewhere.
12 MR. CAMERON: Perhaps the underlying point here, 13 including Ray's reference back to petition for rulemaking, 14 is that this process is open and transparent and information
-s
( 15 is available to everyone to be able to comment on this.
16 Let's go to Gary and then to Ray and then let's 17 get back on agenda.
18 MR. HOLAHAN: I think I'd like to make two points 19 and maybe one commitment. And if this starts a trend of 20 other people making commitments, that's okay with me.
21 First comment is there are a number of reasons why 22 safeguards isn't done in PRAs. Some of them are fairly 23 obvious. If you look -- pick up our study and you see when 24 we're worried about hurricanes or tornado missiles or 25 whatever it is, you'll see data, lots of information, okay.
lfD I \l ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
43 1 The heart of this sort of scientific activity is publicly
() 2 available information that people can scrutinize and argue 3 over. That sort of information isn't available and isn't 4 going to be made publicly available in the safeguards area, 5 you know. We don' t publicize -- and not just we, nobody in 6 safeguards area puts their information on the table in 7 technically available reports, so that it can be dealt with 8 in a sort of open scientific environment. And that makes l
9 this issue very difficult to deal with in that sort of way.
10 However, I started out saying that our goal was to 11 be complete in our addressing issues. And if we can't be 12 complete, we, at least, owe people an explanation of, you 13 know, why we're not complete or how otherwise these issues 14 should be addressed. So, it seems to me that we ought to
() 15 take away from this, that there should be some safeguard 16 sabotage section in our report. Even if it's not a risk 17 assessment or a probabilistic approach, we should explain to 18 people, you know, what we've done and what we haven't done, 19 you know, and the reasons for those things.
20 One other point I'd like to make is, Glenn 21 mentioned our focusing on high consequence events -- the 22 probabilities of high consequence events. We don't do that 23 arbitrarily. We' haven't arbitrarily said, well, let's look 24 at severe accidents and let's not relook at design basis.
25 We do it, you know, for what we think is a very good reason.
l ANN RILEY & ASSOCIATES, LTD.
l Court Reporters i 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
44 1 What we're interested in is protecting people from receiving
() 2 3
doses. And the reason we focus on the high consequence event is because the consequences are many more times more 4 -- many times worse than the difference in the 5 probabilities. Although severe accidents are less likely 6 than a fire in the residence, the consequences are more than 7 a thousand times worse. And so when you look at it in a I 8 probabilistic context, the severe accidents dominate the 9 likelihood of citizens getting doses.
10 But as with the previous discussion, it seems to 11 me the reason that we are addressing spent fuel pool, large 12 releases associated with oxidation of that fuel, and that 13 we've said, you know, nothing about these others needs to be 14 explained to people, okay; why are we addressing these
() 15 issues, why are we not addressing the much lower 16 consequence, but higher probability events, you know, and 17 what's appropriate to deal with those. I think we owe an 18 explanation of why we've chosen to do what we've chosen to 19 do. And I think we can take that away, also. These are not 20 arbitrary. There are reasons for them. To the extent that 21 we haven't explained to people, then I think we haven't done 22 the job very well.,
23 MR. CAMERON: Okay, thanks, Gary. And I'll put 24 both of those points up for action items. Ray, how about a 25 -final point on this issue, and I thank Gary for opening up ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters l
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 !
(202) 842-0034 J
45 1 this broad train of thought for us, and then we'll go to
() 2 3
Glenn for his presentation.
MR. SHADIS:
Ray?
Thank you. I'm glad to get that 4 prolific response. You have really two curves here, which l 5 meet at some point between consequences and probabilities.
l 6 You have a high probability for these different fire 7 scenarios, for example, and other scenarios, and perhaps the 8 consequences are less. Although,_from what we've heard in 9 the previous meetings, we're down to talking about' gap 10 releases and everybody has poo-poo'd the notion that this is l 11 of any great consequence. So, you know, maybe the 12 consequence curves are also beginning to approach one l
.13 another.
14 The -- well, you know, what I'm saying here is
() 15 that I appreciate your -- the position you're coming from, 16 in terms of what you're choosing to do. But, I don't agree 17 with it. I think that if you're going to put something out 18 under the title of a risk informed decommissioning rule, 19 then you ought to talk about decommissioning, not 20 specifically the spent fuel pool scenarios. And, you know 21 -- and even there, I mean, I'm not sure where we are with 22 some of the cask drop things. I guess we'll get to that. I 23 hope we'll get to that in these sessions.
24' MR. CAMERON: Yeah, we will and we'll -- so, we'll 25 be revisiting that. Gary, one response.
.h v
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
46 1 MR. HOLAHAN: I think in our commitment to explain
(} 2 why we're doing what we're doing and not doing it, we will 3 -- we will effectively, I thin, address this idea of higher 4 probability, lower consequence. This has been done for 5 reactors. There are probability consequence curves that 6 deal with more than just severe accidents. There's a whole 7 -- you know, it's been -- this is an old topic. It's been 8 done in the early '70s. It's called Farmer Curve for 9 operating reactor issues. And it seems to me, we can plot a 10 few of those points and, I think, give a better 11' understanding of what the lower consequence, higher 12 probability end of this issue is and what the high 13 consequence, low probability issue is. And I understand 14 that we should be putting that out, in a way that you can --
() 15 that the public can react to. ,
16 I think we're still going to show that we're 17 looking at the right end of the curve. Most of the public 18 protection needs to be at the high consequence, high 19 probability curve. But, we ought not to argue it up, 20 because we haven't given you the -- we haven't given you the 21 story. And I think that, you know, as part of the next 22 phase of this, we need to put that on the table. And, for 23 now, that's about the best we can do.
24 MR. CAMERON: Okay. Mike, before we go to Glenn, 25 you have a comment?
O ANN _RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
r 1 47 1 MR. MEISNER: Real briefly. As Ray pointed out, pi
- 2 the kinds of issues you're talking about now are within the
%)
3 design basis of the plants. We've got a lot of information 4 analysis. The industry would be happy to provide that 5 stuff.
6 MR. CAMERON: Okay, great. Well, that was -- I 7 think that was useful discussion. Glenn, are you ready to 8 talk about the study?
l 9 MR. KELLY: Okay. My name is Glenn Kelly and .'m 10 going to assume in my presentation here that you're at least 11 somewhat familiar with the report or with the information 12 that we've talked about here, in that I'm not going to be 13 going into a huge amount of detail. I'm going to be giving 14 you an overview of how we performed our analysis, in order
('N
(_,) 15 to look broad based-wise for what the potential risk were 16 for spent fuel pool. And I did want to take a second of why 17 we, also, focused on the spent fuel pool, because we looked 18 around at a decommissioned plant and we said to ourselves, 19 what else can cause early or latent fatalities to the i 20 public, if there was a problem that occurred at the plant.
21 And, basically, we came to the conclusion that the only 22 place that we really saw that was in the fuel that's being 23 stored in the spent fuel pool. So that's why our analysis 24 concentrated on that area. By the way, if you don't have a 25 copy'or aren't familiar with the report, there are copies of
/ %
ANN RILEY & hPSOCIATES, LTD.
b Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
48 l 1 the report that we performed the draft technical study at 2 3 'M door.
3 The primary thing that we were looking at, in our l 4 draft report here, was the estimation of the frequency of 5 the zirconium fire. And, again, why did we choose the 6 zirconium fire? And the reason is that, in order to get 7 consequences out to.the public. there had to be a 8 sufficiently energetic source drive the radionuclides j 9 offset and get them out. And te realized that, effectively, 10 in the spent fuel pool, because there's not a lot of energy 11 compared to what you have with an operating reactor, that 12 what you were -- the energy that you did have, in order to 13 create something that would take it offsite, you needed a 14 fire, in this particular case. And zirconium fire turned 15 out to be a potential concern.
16 And we looked at that from a deterministic 17 standpoint, to determine whether or not it could occur, and l
18- we felt that, yes, it could. So, then, we looked at it from 19 a probabilistic standpoint to say, okay, what's the chances 20 of a zirconium fire occurring. And if you look up the 21' equation on the slide, you'll see that the frequency of 22 getting a zirconium fire is equal to the summation of the 23 initiating events times the failure of whatever mitigating 24 systems there were -- that were needed to prevent that fire 25 from occurring. And we go in the report in a lot of detail O ANN RILEY & ASSOCIATES, LTD.
V Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 L
49 1 to explain exactly what these sequences are. We give the
() 2 event trees and the fault trees and we give the assumptions 3
that are associated with them and all the numbers. And it's 4 all laid out there in detail for you to look at. But, this 5 is basically how we went about doing it. It was the 6 . initiating event times the probability that either equipment 7 was going to fail or the operators made a mistake or they 8 just kind of lost track of what was going on.
9 Next slide, please. Now, a number of questions 10 have come up about how we came up with some of our numbers, 11 in particular the human error numbers. And I just wanted to 12 kind of basically just quickly go through this, and this is 13 a little bit simplified, but I think it will be useful for 14 you. For a lot of the long-term and intermediate-term
() 15 sequences, the initiating event frequencies were right 16 around one in a hundred per year. And then when we looked 17 at.the potential failures of mitigating systems, we found
[
18 that what would happen is that for operators errors -- and l
1 19 here, when I say operator errors, we're really talking about 20 organizational errors, where the organization has fallen 21 down, in following what's happening in the event here. The 22 organizational errors we're talking about were things that ;
-23 happened 9,999 times. After having this initiating event, 24 they would have gotten it right. It's that one time out of 25 10,000 that the organization misunderstood it.
l
/N ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 :
(202) 842-0034 i i
. I
50 l
1 And to kind of put that in context, you can think
() 2 of that at Three Mile Island, we had a one-for one. We had 3 an organization that looked at an even that was occurring 4 and they lost track of what was happening. They didn't 5 understand that the PORV block valve was open. And they 6 wondered down one path thinking that they were doing all 7 right, they were doing the.right thing, and it turned out to 8 not be the right thing. And it was a matter, in that case, 9 the operators didn't have the right information; they 10 weren't properly trained. And -- but that is a case where 11 an whole organization went down the wrong day. And that's {
i 12 the type of thing we're talking about there, that one out of l 13 10,000 times, that's going to happen.
)
I 14 And that's -- from our standpoint, there are I
() 15 people that would argue that maybe the number is 16 significantly lower than that. And we just don't believe
- 17. that there's adequate information out there to justify a 18 lower number. Now, in reality, it may be lower. But, I 19 can't tell you how much lower and I think it would be very 20 difficult to accurately quantify that.
-21 So, from our standpoint, what we've done is we've 22 looked our results and we said, rather -- we don't want sit 23 here and justify the bottom line numbers that we've come up 24 .with and say, you know, the NRC has come up with the exact 25 correct number or what the frequency of a zirconium fire is.
[~h ANN RILEY & ASSOCIATES, LTD.
(.s/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
51 1 What we have come up with is we have found that there are
() 2 sequences that lead up to have concerns about certain areas l 3 at the plants and what we're looking now is to understand --
4 have a better understanding of what exists out there in the 5 industry today; what will be existing in the future, which 6 is even more important; and why we should feel comfortable 7
about these areas and understand are they or are they not 8 potentially significant risk areas.
9 t
Next page, please. I wanted to talk about some of 10 the sequences that we felt were credible long-term 11 mitigation events. Now,-they're long term -- and this --
12 for this series of sequences, because you're talking about 13 days and days, that this event has to go along before you 14 end up getting fuel uncovery and zirconium fire.
() 15 The first one starts out with a loss of offsite 16 power from either a plant centered or grid related loss; the 17 diesel driven fire pump fails to start; offsite power isn't 18 recovered; and no recovery from offset sources, such as fire 19 engines. That lack of offsite recovery is really lack of 20 the operators adequately calling in the offsite resources.
21 It's not that you can't get offsite resources, in most 22 cases.
23 The next sequence was loss of offsite power from a 24 severe weather event. Let me just take one second to go 25 back and talk about that. .You can say to yourself, well, O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 L
I 52 1 how would something like that actually happen. I mean, why
() 2 would somebody be so foolish as to not do something, if they 3 see this. And, well, part of it may be that utility keeps 4 believing that they're going to get their power back.
5 Utility keeps thinking that they're going to be able to fix 6 this pump. And their focusing attention on getting the fire 7 pump to work, or getting offsite power, and they're not 8 looking at the overall process. And when we did these 9 analyses, we realized when I went and visited four 10 decommissioning plants, that they did not have procedures 11 that were specific that would say, for example, if the water 12 level got to be five feet over the fuel, that I'm going to 13 call the local fire department, tell them to bring their 14 fire truck out here, or anything like that. It was an ad
() 15 hoc type of decision, on the basis of the -- for the 16 operators. And so, we've taken that into account in our i 17 calculation of the human error rate.
18 The next set of sequences is the loss of offsite 19 power from a severe weather event, where you're not getting 20 any-recovery of offsite power in the time period. Diesel 21 fuels to start or run. This is the diesel driven fire pump.
22 .There are no -- we assume there are no diesel generators at 23 the decommissioning plant. And the operator does not call 24 in the offsite sources or they have difficulty getting 25 there, because of a -- if it had been a large hurricane or O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
I~
53 1 other type of severe weather event. And, again, that's
-m (d similar type evaluation to what we did before.
\
2 3 Go to the next set of slides. Loss cooling 4 initiating event, then failure of control room alarms, 5 operator fails to notice the spent fuel pool steaming and 6 levels dropping, during walkdowns. And another thing, gee, 7 why would an operator not notice something like that? Well, l
8 these are not operating reactors and the concern was that I 9 over time, might the operators become much less diligent in 10 doing their jobs. After a year, tt 3 is no onsite NRC 11 oversight at the decommissioning :s. Might this -- at 12 the decommissioning plants I visiced, they were working 13 four-day weeks. Might this has been an event that occurred 14 right at the beginning of a long weekend, where the
() 15 operators -- and take this with the understanding, the 1
16 plants that I went to, the operators -- the certified field 17 handle. were very qualified, they were -- I was very 18 impressed with the level of attention that they had, their 19 instrumentation and everything. We're looking for a plant 20 down the road that might not have the same kind of quality, 21 because it's not required currently by our regulations.
22 So, we're worried that maybe I set -- somebody set 23 their T.V. up in their control room and they're sitting 24 there watching the T.V. over the long weekend, and they just 25 -- because, there's no requirements that NRC has that you
(l
\ss i ANN RILEY & ASSOCIATES, LTD.
Court Reporters l 1025 Connecticut Avenue, NW, Suite 1014 !
Washington, D.C. 20036 (202) 842-0034 i
I
)
L 54 1 have to go in there every shift and check out and see what's
() 2 3
going on, and they just -- the guys were just watching the T.V. I mean, that's a potential type thing that you look 4 at.
5 How do you quantify that? It's very hard. Is it 6 possible? .Yes. But when we looked at it, we say, this is 7 the' type of thing that could happen. Doe PRA do a good job 8 of calculating those numbers? No. But, when we look at our 9 numbers, we realize that this is all -- I'm giving you one 10 potential scenario out of many potential scenarios. This is 11 an idea of how something like this could occur. But, I do 12 want to come back again that I was very impressed with the 13 plants that I visited, with the quality of the operators and 14 what they were actually doing at the spent -- at the
( 15 decommissioning plants today.
16 The next sequence is one where there's a fire 17 that's not suppressed in a building. It has the spent fuel.
18 And that fire fails either the offsite power or the cooling 19 pumps, and the operators fail to refill a pool using a i 20 diesel driven fire pump and we don't have recovery from '
21 offsite sources.
22 Let's go to the next slide. Now, based on --
23 based on these sequences, we've come up with a number of 24 staff concerns. One of them is operator attentiveness.
25 And, again, these are not concerns at current plants; these O ANN RILE 7 & ASSOCIATES, LTD.
1025 Connecticut Avenue, NW, Suite 1014 Court Reporters Washington, D.C. 20036 (202) 842-0034
55 1 are -- current decommissioning plants; these are concerns
() 2 3
potentially for future plants.
where they're not Operator attentiveness, currently, they're not required to 4 visit -- actually go out and look at what's going on in the 5 spent fuel pool. We don't -- as I said, we don't have that 6 daily NRC oversight after a year. The instrumentation in 7 the control room is not required to be operable. Even the 8 radiation monitors are not required to be operable, unless 9 the licensee is moving fuel. And if the -- so, if the fuel 10 is just sitting there in the pool, it's not being moved, so, 11 therefore, the -- even the radiation monitors don't 12 necessary have to be operators.
13 Now, we gave them credit for them being ope ~able 14 in our analysis. But, we note that they don't have to be
() 15 operable. There are.no automatic actions by any mitigating 16 systems. They are all -- everything is done by operator 17 action. All the responses are done by operators.
18 The procedures at plants deal with basically slow 19 leaks and how to deal with them. And I did not see 20 procedures that effectively said at certain type -- that 21 gave you information about, at a certain level, you're 22 supposed to seek additional offsite help or anything like 23 that. I might have missed that, but when I was there at the 24 four sites, I didn't notice that of any of them. Generally, 25 the procedures are not driven by the level of water in the O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
56 1 pool. I I"' 2 The next -- the fire pumps and the capability of V) 3 making up for the spent fuel pool is also a potential 4 concern. Under our regulations, the utility is allowed to 5 decide -- make a decision on what fire equipment it needs at 6 a decommissioning plant, in order to keep the plant save, 7 and that may or may not include, from NRC regulations, the 8
-- keeping the operability of the diesel-driven fire pump.
9 It's my understanding from conversations with various people 10 in industry that their insurance companies make sure that 11 their fire pumps are operable. And they have other reasons 12 to believe that their equipment is operable, and that's 13 something that we're interested in hearing, how the reality 14 of the plants may differ from the assumptions that we've
() 15 laid out in our analysis.
16 I was very impressed by the training and 17 experience of the current crop of certified fuel handlers 18 that I found at the four decommissionir.1 plants that I 19 visited. I thought they were very highly skilled. Most of 20 them were former senior reactor operators, but that's not in 21 a necessity. The utilities are allowed to propose a 22 training program for their certified fuel handlers and 23 theoretically. They could take any individual and train 24 them, whether or not they had a lot of previous experience 25 at the reactor and that person would then be a certified
N ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washingt un, D.C. 20036 (202) 842-0034
57 1 fuel handler. 4 l
f~') 2 And finally, we come to redundancy,
\/
\
3 defense-in-depth, diversity of equipment, margin of safety 4 type questions, because what I found that in spent fuel 5 pool, cooling systems, who are out at the plants, they were 6 primarily one train system, in the sense that they were 7 running off of -- as a matter of fact, most -- a lot of the 8 plants, all of the spent fuel for cooling equipment is 9 running off of one bus. There were no diesel generators 10 that were there to provide backup. There were no 11 uninterruptable power sources. And we felt that, at least, 12 it would be very important to address the -- what 13 defense-in-depth means for a decommissioning plant, what 14 type of margins to safety do they have, and how do they (O) 15 maintain redundancy there.
16 Now, I'm going to skip on to the next one. Just 17 to give you an idea about what kind of numbers we came up 18 with, and these are numbers that came out of our report, for 19 the frequency of fuel uncovered. Now, you notice here I'm 20 talking about frequency ~of fuel uncovery; but, earlier, I I
21 was talking about frequency of zirconium fire. And there's 22 a little disconnect there, and I'd like to explain that 23 disconnect.
l 24 It was relatively straightforward, in a certain 25 way, to calculate what was the frequency of fuel uncovery, 1
I ANN RILEY & ASSOCIATES, LTD.
[))
\m Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 i Washington, D.C. 20036 l (202) 842-0034 i L !
58 1 because you know how fast the water is going to ball off,
() 2 3
based on the decay heat levels that you're assuming. So, could do that calculation pretty well.
we But the -- what 4 actually happens when you start uncovering the fuel and you 5 start getting your zirconium -- you know, how quickly you 6 get to zirconium fire. We don't have as much information in 7 those areas.
8 It was also assumed, and that was partially a 9 simplifying assumption, that it would be impossible for the 10 operators to get in there and affectively mitigate the 11 accident, in the event that the fuel was uncovered. And 12 this was a -- assumed that this was true, based on the 13 potential dose rates associated with the spent fuel pool 14 being uncovered. That's something we're also interested in
() 15 understanding better, in case that we've greatly 16 misunderstood that concern. So, we're assuming that once 17 the fuel gets uncovered, effectively, you can't get in there 18 and mitigate it and, therefore, fuel uncovery effectively 19 leads to zirconium fire. So, these frequencies here give 20 you -- tells you what the initiating event is. And then on 21 the right hand side, it tells you what is our chance of 22 getting fuel uncovery.
23 Now, you notice it says for case one -- if you 24 look in our report, we had three different cases. Case one 25 was kind of the as found case, based on how I found the four ANN RILEY & ASSOCIATES, LTD.
O' Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
59 1 decommissioning plants to be staffed and operated. Case two
() 2 looked at that same reactor, but assumed that instead of the 3 last fuel being moved one year previously, assumed the last >
4 fuel was moved one month previously. And case three was 5 whe14 we looked at if a utility went in and kind of went 6
bare bones with minimum or a little bit better than minimum 7 amounts that are allowed by the regulation, what would their 8 configuration look at, and we analyzed that particular case.
9 But, we feel case one is the one that's most applicable 10 here.
11 Next slide, please. Okay. These are the 12 intermediate mitigation events. And these are events where 13 --
the intermediate mitigation events, what's happening here 14 is that we're getting a pipe break or a hose has fallen into
() 15 the spent fuel pool and then we get a siphoning event, so 16 that somehow, we're siphoning water out of the pool. And ;
17 we're looking -- we assumed that the water was effectively 18 siphoned down to, oh, about eight feet above the fuel. And 19 then at five feet, as assumed that the -- five feet, does 20 evacuation begin, Mike, or we start telling them about 21 evacuation?
22 Five feet was the time of evacuation. So, that's 1
23 the difference. These events occur faster than the previous l 24 events, which took days to occur. These occur in a day or 25 less, okay.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 i Washington, D.C. 20036 '
(202) 842-0034 i
J
60 1
1 The first event there, you have a small loss of r
V) s 2 inventory event that occurs and the data for this came from 3 actual data at decommissioning plants. Spent fuel pool 4 level is not restored either by the makeup pumps or the fire 5 pumps, and that's because either the operators made a i 6 mistake or they had hardware failure and they didn't call an j 7 offsite recovery.
8 The second one was a large loss of inventory. The 9 difference between the small loss and a large loss is that a 10 small loss can be handled by the standard makeup pumps that 11 the spent fuel -- of the spent fuel pool cooling system.
12 For the large loss of inventory event, you need the fire 1 13 pumps, because they have significantly greater capacity.
14 And, again, it's the same basic scenario, where they're not
/"
(_)) 15 able to make up or they don't make up and then they don't 16 call in offsite resources.
17 And the last one is a loss of -- small loss of 18 inventory event. Control room alarms don't go off and the 19 operators fail to notice the condition during their 20 walkdowns.
21 Now, we have the same basic staff concerns for the 22 intermediate, as we did as the long term. But, the other E'
23 staff concerns that are additional here is that we're 24 concerned about vacuum breakers, because we've found 25 instances in the literature, where vacuum breakers have not h\~/
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 )
J
61 1 been adequately sized or tested and there have been actual
() 2 experience with that at plants. It's our understanding, at 3 this time, that there are probably not any large pipes 4 penetrating this spent fuel pool near or below the fuel 3 level that are not capped off. But, we would want to have 6 that type of thing confirmed when someone comes in. And 7 similarly, there are no intermediate size or large size 8 pipes, either permanent or temporary, that extend 9 significantly down into the spent fuel pool and it could act 10 as siphons.
11 So, that's -- that's really what I wanted to tell 12 you about our analysis for these type of events. And, 13 again,- the key is we wanted you to not focus on the numbers, 14 but to focus on the insights that we got out. What are our f 15 concerns that we see from the analysis? The analysis tells 16 us that there are certain things that kind of stare at us 17 when we look at the PRA results and tell us that maybe this 18 is an area that we need to look at a little bit closely.
19 And, hopefully, we can hear more about that.
20 MR. CAMERON: Okay. Thanks, Glenn. I know 21 there's going to be lots of comments and questions on the 22 presentation. What I'd like to.do is to go to Ed Burns and 23 get his information on the table and ther, we can take a 24 break and then come back for full discussion. And I would 25- note that there is a penalty for knocking the microphones ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
62 1 into the well here. But, you --
[ \ 2 SPEAKER: Only the live ones.
V 3 MR. CAMERON: Only the live ones, okay. I'm glad 4 we clarified that. At any rate, Ed, how would you like to 5 do this? Would you like to use this microphone and stand up 6 here and do your presentation?
7 MR. BURNS: This is an effort that was the result 8 of a NEI-sponsored review of the NRC draft probabilistic 9 analysis. And included in that review, we looked at seven 10 other related NRC-sponsored evaluations that also looked at 11 spent fuel pool risks. And this effort, as was pointed out 12 earlier, to provide really a constructive feedback on the 13 draft assessment and to provide any information that the 14 industry might have or any perspective that we may have O
(j 15 relative to what's important, as we view it.
16 Some of the major topics that I'd like to talk 17 about is the general purpose of the draft document and of 18 this discussion that we're providing. Some of the previous 19 probabilistic perspective have been performed by the NRC and 20 their contractors. Talk a little bit about consistency of 21 analysis assumptions and then the design of the decision 22 criteria that are used, basically, the success critiria for 23 each or the N states; then realism in accident sequences, 24 particularly the initiating event frequency, the crew 25 response, and then the time allowed for operator actions.
ANN RILEY & ASSOCIATES, LTD.
O' Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
63 1 And then, we'll talk a little bit about what the dominant
() 2 contributors might be, if we -- if we would requantify those 3 sequences. And clearly any requantification presentation 4 that I would give you would be very superficial, based upon 5 we've only had a short time to look at this; and then 6 finally, some risk insights that we've identified, as part 1
7 of our review of the report.
8 We certainly agree with the draft staff report 9 assessment that the quantification that was presented in 10 that document was a near an upper bound. And we'll discuss 11 what the implications of that upper bound might be. And, 12 secondly, we certainly agree that the framework that's ;
13 presented in the draft report is -- appears to be very 14 usable and applicable, as a tool and framework for
() 15 characterizing the risk perspective.
16 MR. RICHARDS: Ed, can you give us a little 17 background on your background?
I 18 MR. BURNS: Let's see. I've been doing risk f t
19 analysis since 1976, mostly in the BWR area, for at power R20 risk analysis. Some of the original work that we did was 21 started with some -- actually, some safeguard work for 22 Sandia that we did originally, to try and see whether 23 safeguard work could be performed in a probabilistic 24 framework. I've done five. full at power PRA developments 25 and participated with the BWR owner's group industry and in ANN RILEY & ASSOCIATES, LTD.
\- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
i 64 1 various probabilistic risk assessment and arguments relative
() 2 3
to regulatory requirements.
What we want to do is we want to make sure that we 4 can effectively use PRA arguments in a risk informed 5 environment. So, risk informed _ regulation really uses PRA l 6 input to optimize the allocation of limited resources. And 7 as stated in the' draft report, it's to reduce unnecessary 8 conservatisms associated with current regulatory 9 requirements. So, those are the kinds of purposes that 10 we've set up, to look at reviewing this document.
11 Previously probabilistic analysis of spent fuel 12 pool and severe accidents at decommissioning plants have 13 looked at -- primarily considered seismic-induced spent fuel 14 pool, as one of the primary mechanisms of causing risk.
15 Additionally, the cask drop accident has been identified 16 previously, although NUREG 1353 identified that resolution 17 of the generic issue 836 resolved that issue and made it a 18 non-dominant contributor to severe accident risk.
19 If you examine some of the seven or so previous 20 documents that have been identified, have been developed to 21 look at risk under severe accidents at spent fuel pools, 22 you'll see that, in general, they come to the conclusion 23 that seismic induced spent fuel pool failure is one of the 24 dominant contributors. And that's the one area that I'm not 25 going to cover today and we're going to cover that tomorrow, i
l ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034
E l 65 1 Tom O'Hara will be here to discuss that contributor to
() 2 3
potential risk.
You'll see that there are several other 4 contributors that have been identified by various NUREGs.
5 The currently draft report that we're discussing discusses
'6 -- or identifies those -- virtually each of those ones that 7 have been identified previously, as risk significant. So, 8 it says that each of the ones that we've identified, that 9 have been talked about and identified before are all risk 10 significant.
11 One of the things that triggered the concern, I 12 think, at NEI was that in -- with the numerical numbers that 13 were presented in the report, involved the fact that in 14 NUREG 1353, which looked at some of the risks that might be
() 15 associated with this spent fuel pool, the risks were 16 generally considered to be, in terms of draining the pool, 17 relatively low, a 10 to the minus eight, the 10 to the minus 18 range, except for the seismic event. Whereas the draft 19 report that was given to NEI for review indicated that there 20 were a large number of contributors that could be in the 10 21 to the minus seven t' 10 to the minus six range.
22 Admittedly, there's a difference between what 1353
- 23. identified as best estimate and what the draft report 24 identified as upper bound or you might call it worse case i 25 analysis. So, we're looking for some consistence in those 1
l l
ANN RILEY & ASSOCIATES, LTD. !
-- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
66 1 calculations and what exactly 'a would like the characterize 2 the risk with.
3 The consistency argument revolves around whether 4 we can identify a well recognized measure of risk that we're 5 all comfortable with. In other words, if we can't identify 6 something that is consistent with the safety goal, what 7' other surrogate measure that we have that we could use in 8 place of that. And it's difficult currently to correlate 9 that in-state -- an in-state in a spent fuel pool with the 10 analysis that we've done thus far in PRA relative to core 11 damage frequency. We have a feeling for what core damage 12 frequency means and we have a feeling for what large early 13 release means and we have a feeling for what those 14 relationships are to the quantitative and qualitated health
( 15 effects on the public. But for the spent fuel pool, there's 16 a much less comfort regime relative to any surrogate measure I
17 that we have. j 18 So, to try and get a clear technical basis for 19 evaluating severe accidents at spent fuel pools, we want to 20 make sure that there's a clear connection between any 21 deterministic analysis that's performed, i.e., the zirc fire 22 analysis, and what the probabilistic analysis is saying, so 23 that there's a connection or, if you will, a direct 24 . relationship between a probabilistic analysis that's 25 performed and the deterministic offsite health affects that ANN RILEY & ASSOCIATES, LTD.
O)
\~ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
67 1 are developed. And that would be consistent with -- and the
() 2 fact that we want to make that a best estimate calculation 3 would be consistent with reg guide 1174, the several 4 accident policy -- or the safety goal policy statement and 5 the way that the current ASME PRA standard, regarding PRA as 6 being developed.
7 So, connecting risk measures in end states with 8 what we know or what the probabilistic analysis, that's 9 definitely a difficult area. Different sequences result in 10 much different potential challenges, in terms of timing, 11 complexity of action, and resources that can be brought to 12 bear. An that's one of the areas that seems to be missing 13 from the draft report, is that connection. The draft report 14 uses frequency of fuel uncovery, as a risk measure, but that
( 15 is a little bit --
it's different, in terms of draining the 16 pool versus a boil-down scenario. So, you've got -- and the l 17 eventual inference from the report is that the frequency of 18 fuel uncovery does equal what's called a release to the zirc 19 fire. So, that connection -- or that inference is not 20 convincingly made in a probabilistic framework.
21 Alternatively, what we could --
22 MR. KELLY: I didn't understand what you were 23 saying there. Could you -- the last part, could you explain 24 what you mean by that?
25 MR. BURNS: Normally, what we would do, if you ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticuc Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
68 1 wanted to draw an. analogy between a level one PRA and a
() 2 3
level two PRA, you'd have some kind of a bridge tree or you'd have a connection between the consequences or the 4 progression or the mitigation of the severe accident, once 5 you had determined that there was some challenge to the 6 fuel. So, if you wanted to declare the end state of the 7 first part of your analysis to be failure of frequency of 8 fuel uncovery, that would be fine. And then, you'd want to
'look at what the mitigation measures are subsequent to that,
~
9 10 that you still had available. So, you may still have 11 resources that you can be -- can be brought to bear to 12 prevent a circ. fire, even though you've reached the point of 13 the top of the active fuel.
14 Alternatively, though, we could use the frequency (f 15 of fuel uncovery as a measure to show that risk was below 16 what was credible. So, if you could demonstrate that all of 17 the frequencies that could get you there are so low that you 18 are not concerned about it, then that would still be -- then 19 that would provide an adequate measure of risk, I think.
20 But, that's not currently what the draft report says.
21 The other area of potential disconnect between the 22 end state and the probabilistic analysis, that previously, 23 as we mentione6, the probabilistic analyses that had been 24 performed indicated that the loss of pool water was the 25 dominant cor tributor, mainly through a seismic event. And O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
69 1 the deterministic calculations that had been performed, I T 2 regarding the zirc fire analysis and the off-site doses, V
3 were generally consistent with that. So, we don't have a 4 direct correspondence between the consequences of a 5 boil-down event, which is now one of the more dominant 6 contributors in the analysis, and what the off-site doses 7 would be.
8 This just is to reiterate that there does appear 9 to be a need to make a stronger correlation between a front 10 end probabilistic analysis and the consequence event for 11 release to the public. And just another reiteration of the 12 same topic is that even the recent reports, INEL report, 13 upon which post of the data -- or a lot of the data is used, 14 is derived for use in the draft NRC report. The INEL report
() 15 was really for an operating plant and it, also, had a 16 different end state. It used near boiling frequency, not 17 the frequency of fuel uncovery. So, the end states that 18 were in the two reports were different and, yet, the data, 19 the timing, those kinds of things, appear to be taken to be 20 the same.
21 So, if there is an inconsistency in the end state, 22 that might be considered a flaw in the analysis. We think 23 that the framework is still excellent and usable. And as 24 part of the review, we've tried to requantify some of the 25 sequences, to see if they can be identified as still below
[ ANN RILEY & ASSOCIATES, LTD.
\-- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
70 1 what would be considered a credible accident and, therefore, i.
() 2 3
allow the use cf frequency of fuel uncovery to be a surrogate measure that we could use to show that while we're 4 below -- we're below this value and, therefore, the risk 5 could be considered very low.
6 This is my pitch for best estimate accident 7 sequence frequencies. Really, in the risk informed 8 regulation, we depend a great deal on the ability to 9 charncterize on a best estimate basis the accident 10 scenarios. This best estimate characterization can then be 11 used to prioritize resource allocation. I think that's 12 consistent with the direction that the Commission is heading 13 and the ACRS has pointed us. The use of upper bound or 14 worst case assumptions to demonstrate the importance of an f 15 issue is difficult to fit into that risk informed, unless 16 you can demonstrate that the upper bound or worst case tells 17 you that there is no risk.
18 MR. KELLY: Could you point out -- this is Glenn 19 Kelly, could you point out where you feel the analysis uses 20 upper bound or worst case assumptions, as we're going 21 through here, so we can help understand what it is 22 specifically you mean?
23 MR. BURNS: Sure. I guess I was going by the fact 24 that the report said you used upper bound analysis.
25 MR. KELLY: No , it doesn't say that.
l O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
l Washington, D.C. 20036 (202) 842-0034
71 1 MR. BURNS: Maybe I misinterpreted what I read.
() 2' The conservative analysis that I -- the 3 conservative ideas that I found were that conservatisms 4 included -- the reasons that we want to take out 5 conservatisms are to make sure that we have a level playing 6 field for addressing, on a realistic basis, what are the 7- accident sequences to be concerned with. We want to ensure 8 that the analysis is not an upper bound and we want to avoid 9 conservative bias. The contributors, then, could be 10 compared equally across the board.
'11 These are the -- in the limited time that we had 12 to look at the report, these are the items that we've i 13 identified as potentially conservatisms. And I certainly 14 don't want to mislead you and say there are no
) 15 non-conservatisms in the analysis, because I think that 16 there may be some areas that -- particularly, with HRA 17 dependency, that may not be treated, But, in terms of 18 conservatisms, the HEPs, we believe are relatively high for 19 the accidents that we're looking at. The LOOP initiators, 20 AC power recovery probabilities, the diesel fire pump 21 reliability, the times that are being used as the time i 22 frames appear to be relatively short, and the temperature of l 23 the zirc ignition for the consequence analysis.
24 As far as the HRA goes, the analysis appears to l l
25 ' cite INEL document as the basis for the HRAs. The INEL !
Ci ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
72 1 document, in turn, says that the HRA methodology that was
() 2 3.
used was a simple approach. It was established in the draft report, which doesn't appear to have been reviewed. One, 4 I'd like to see people like Alan Swain or Gareth Parry try 5 to review that methodology before it was used. We -- it was 6 described in the INEL document as a relatively quick and 7 conservative methodology to estimate HEPs, and it appeared 8 to not be sensitive to whether -- what the operating 9 characteristics of the procedures were.
10 Also, the time windows that were used in the INEL 11 report were very short, based upon they were full core 12 off-load and in an operating plant, as I said, earlier, and, 13 therefore,.they were much shorter than the time frames that 14 we're discussing.
() 15 The kinds of information that the HRA -- you would 16 like to see in an HRA is credit for self checking from the 17 crew, the second crew member check, additional shift 18 attention in recovery, additional cues causing increased 19 attention -- that would mean multiple spaced alarms; in 20 other words, the temperature alarm, the level alarm, and the 21 radiation alarm are all staggered in time relative to most 1 22 of these sequences -- and the long reaction times.
23 Particularly, there is also, because of the multiple days 24 involved, there's a management oversight question of whether 25 that would add some level of additional benefit to the O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
73 1 operating crew and the organization response.
!/ k 2 The desirable' features of an HRA would include V
3 considering the performance shaping factors of the
! 4 complexity that he has to deal, in terms of the accident, 5 the time available that he has, what management oversight 6 over multiple days occur, and the shift changeover. Those 7 are all things that would affect the performance shaping
- 8. factor of the response to the accident. They don't seem to l
be addressed in the -- at least in the draft that we looked 9
10 at.
11 What we were looking for were things like in PRA, 12 the Swain handbook and the EPRI cause-based methodology; 13 identify things related to shift changeover, as a beneficial 14 effect, in terms of reducing the probability of failure of
( 15 an operating crew or organization to respond to an accident.
16 And not very much credit is given for that shift changeover 17 for -- and this is really related more towards valve 1
18 dispositioning or annunciators that are out of commission or 19 such things. But, in this particular case, if you just use 20 that relatively conservative approach, you get very low 21 numbers, in terms of what the -- what the shift changeover 22 should benefit you. In addition, the annunciation response, 23 those numbers are very low and the diagnosis by the control 24 room personnel for a one-day event are extremely low. So 25 those are the kind of numbers that one might expect to see.
( \s./
l-O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 l (202) 842-0034 J
I 74 1 But, when I tried to look at some of the numbers
() 2. if found in the report, it appeared that the recognition, 3 for example, of loss of cooling had very long time, maybe 15
)
1 4 shifts of people looking at inis information, and the HEP 5 assigned to it was relatively high. You'll notice that the 6 ranges of HEPs that are identified here are from three to a 7 minus three, to one minus two, a pretty narrow range, 8 considering the long -- very long time frames and the j 9 different actions that are being taken. So, each of these 10 -- so this was an area that we thought might be desirable to 11 have a second look at.
12 For comparison, if you look at what's used at at 13 power PRAs, you'll find that back in 1978 and 1980, when we 14 did the ATWS rulemaking, which was a risk -- one of the risk k 15- informed applications of PRA, at least in my judgment, that 16 we looked at what kind of operator actions could be taken 17 under ATWS conditions, under very short time frames, and at 18 power PRAs. And we used human error probabilities in the 19 range of ten to the minus two, to ten to the minus three, 20 under relatively short conditions. And in at power PRAs, we j 21 used, under the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> mission time, if he has long times 22 to take actions, we credit - generally credit numbers in the 23 10 to the minus five to 10 to the minus six probability l 24 range for those kinds of actions. So, there is a precedent 25 for using relatively low numbers.
l O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 i
75 1 For the loss of cooling event, which is identified
() 2 in the draft report, this is one of the most -- this is the 3 most straightforward one.for the operating crew to respond 4 to and it's also one that has a dominant contribution in the
'S draft report and one that's relatively -- I'll say 6 relatively new, in. terms of the past 12 years in risk 7 analysis of the spent fuel pool.
8 The characteristics of the event scenario, we have 9 alarms that occurring not sequentially and not closely 10 spaced in time, in terms of level of temperature and 11 radiation. I may have a camera that are observing that 12 condition in the spent fuel pool. Shift lock down of the 13 area, which should provide a substantial benefit, in terms 14 of identifying the problem and a substantial amount of time 15 for offsite recovery repair and bringing creative conditions 16 to beer.
17 The loss of offsite power event, which is in the 18 same category as the loss of cooling event, is just caused 19 by the loss of offsite power, in this case. In this case, I 20 know I have a problem. I don't need an alarm. I know I 21 don't have any cooling to the spent fuel pool. So, the HEP 22 could be argued to be even lower, in this case. In fact, 23 the stress level might be -- if you remember the swing curve 24 of optimum stress, you might be under optimist -- it's not a 25 routine situation and it's not an immediate life threatening O ANN RILEY t ASSOCIATES, LTD.
Cou.; Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
76 1 situation. So, the HEPs that one might assign could be
() 2 3
argued to be even lower than strictly just the loss of cooling event.
4 I don't want to go down a primrose path, though 5 the PRA methods really are very poor in this area. We have 6 examples of where we include recovery in PRA over 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
7 They're usually procedurally directed, sometimes will 8 include the exponential repair model, which is included in 9 the draft report and sometimes will invoke off-site 10 resources. But, that's rare.
11 If you go beyond 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, I don't think it's 12 unreasonable to expect that ad hoc actions by the crew and 13 the offsite engineering would be applied, nor that corporate 14 resources could be brought to bear. So -- but there really
() 15 isn't a methodology that's been developed, and I think 16 that's been said earlier, to cope with such a problem --
17 methodology problem.
18 What I would -- what are the implications of all 19 of this, in terms of HRA? I think that the draft analysis 20 is inconsistent with past PRA practices regarding best 21 estimate HEP quantification. Again, they appear to me to be 22 more upper bound calculations. The long duration of events 23 is not explicitly incorporated in the quantification of the 24 loss.of cooling event, at least in my judgment. Reductions 25 of factors of ten to a thousand in the HEP's could be
[~' ANN RILEY & ASSOCIATES, LTD. ,
\s Court Reporters I 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 !
J
77 1 considered consistent with current practice. In fact, Swain
() 2 3
acknowledges that some of the HEPs can be slow low as not being needed -- need to be considered further, in the 4 analysis.
5 The present analysis provides -- because of this f I
l 6 sort of uneven conservatism, which affects mostly the lost 7 of cooking events., could lead -- mislead the decision makers 8 by saying that the loss of cooling events are extremely 9 important in the context of spent fuel pool analysis, since 10 the HEPs dominate many of those accident sequences.
11 I will skip over the next few slides. These are 12 other examples of conservative bias. The LOOP initiated 13 event frequency, if I use the latest NRC data from offsite 14 power, I'd get a factor of two reduction in LOOP frequency.
() 15 If I use that same data, the offsite AC power recovery goes l 16 down by a factor of 10. And I can argue that the frequent 17 -- total frequency of fuel uncovery is down by an order of l 18 --
two orders of magnitude, if I run through the HEPs on 1
19 that analysis.
20 The next is the curve that I used to grab those --
21 it's based upon the NUREG analysis. The cask drop and 22- accident analysis, again, in 1355, it was identified to be l
l 23 three times 10 to the minus eight and the draft analysis is 24 identified to be two orders of magnitude higher.
25 The diesel fire pump, the value that's used for l
l l ANN RILEY & ASSOCIATES, LTD.
Court Reporters I 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 i
78 1 the diesel fire point is 20 percent of the time, it's not
() 2 3
working. You probably should rip it out of the plant, if it's only 20 percent. Eighty percent reliable, it's 4 probably not worth even having there.
5 The time frames are one of the crucial items, 6 especially in the loss of cooling event. In the draft 7 analysis, it appears that the first 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, which is the 8 pool heat up time, which is the time when I would expect to 9 find the problem, which -- I need to find the problem when 10 the pool is heating up to boiling. It appears that the 11 draft report gives me credit for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />; whereas on a 12 realistic basis, I would expect that pool heat up to occur 13 over 90 hours0.00104 days <br />0.025 hours <br />1.488095e-4 weeks <br />3.4245e-5 months <br />, which is still -- which is a substantial 14- difference. The boil down time, for the purposes of this
'15 diagram, I just used the same value that's in the draft 16 report. But, again, I think thats a relatively 17 conservative value. If you add up all these numbers, how 18 long this event is occurring, it's occurring over seven 19 days. So, this is a seven day evolution that's happening.
20 The last item that I'll identify as a conservatism 21 is the ignition temperature for the zirconium. The draft 22 report clearly states that the information indicates that 23 the zirc condition could be higher than 16,000 degrees l 24 centigrade. So, the onset of oxidation would appear to be a i
25 more conservative estimate of what that -- what that O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
79 1 temperature is.
() 2 We would advocate that the draft report continue 3 with the framework that exists and that some consideration j
4 be given to requantifying the sequences, maybe eliminating 5 some of the conservatisms that have been identified and may l
6 be identified in the future. Basically, to reassess the 7 operator actions that would be logical to credit and ensure )
8 that best estimate evaluations of the initiating events and 9 equipment response be included.
10 My cut at that, and this is certainly juct a very 11 -- very preliminary analysis, would say that if you did l
12 that, you could get a best estimate evaluation of the l
13 sequence frequencies that are around four times 10 to the 14 minus seven, versus the one time seven minus five. This
) 15 does not include seismic, so that number of four minus seven 16 would change, based on whatever the seismic evaluation would 17 indicate.
l 18 Now, if I was -- if I was imaginative, I would 19 have had the two pie charts together, with the relative size
-20 as a function of the area. But, I don't have that. This is 21 a distribution of accident sequences, as a result of the 22 draft report. And you can see that other than the seismic 23 event and the cask drop event and the loss of inventory, 24 these other events, these other contributors are loss of
( 25 pooling events that make up the rest of the pie.
l l \ ANN RILEY & ASSOCIATES, LTD.
-s# Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 L
80 1 For the reanalysis that -- requantification, this
() 2 indicates the distribution, where the loss of offsite power 3 due to severe weather was the dominant contributor in the 4 revised frequency that I calculated. The table that's 5 present here gives a very brief description of some of the 6 rational that was used to change some of the numbers. I'll 7 skip that and go directly to what risk insights could be 8 derived and they are very similar, in some cases, to the 9 ones that were identified by Glenn. We would want to make 10 sure that the reliability of the gate seals in the spent 11 fuel pool is sufficient to make sure that they're not going 12 to leak; that the spent fuel pool cooling system has 13 adequate siphoned break valves; that any temporary pumps 14 that are used as part of the spent fuel pool also have
() 15 controls on the suction of discharge and the siphon breaks.
16 And then we would want to make sure that we had the 17 capability to align the diesel fire pump to the spent fuel
'18 pool without actually access to the refuel floor. Those are 19 the basis for a lot of the numbers -- or some of the numbers 20 that I used as part of the requantification.
21 I guess I would summarize by saying that the 22 approximate requantificution indicates that there's a 23 substantial conservatism in the NRC estimates and that a i 24 requantification of the risk analysis to reflecting past NRC l
25 analysis and current PRA HRA practice, and best estimate l
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
81 1 versus worst case analysis would be appropriate.
'() 2 3
MR. ZWOLINSKI:
MR. BURNS:
Say, Ed, John Zwolinski?
Yes, sir.
l 4 MR. ZWOLINSKI: Could you go back to your risk 5 insight slide, just before your summary? Are these the 6 kinds of things that you would expect to see in a plant in l 7 the future adopt, as common practice, or ensure exist, along 8 with addressing some of the human performance issues you 9 were talking earlier to?
10 MR. BURNS: What I personally would expect, yes.
l 11 MR. ZWOLINSKI: Okay, I'm trying to get the j 12 context for your --
l 13 MR. BURNS: Yes, exactly. I tried to incorporate l
14 those in the analysis.
) 15 MR. ZWOLINSKI: Okay, l
16 MR. MEISNER: John, let me expand on that just a l
17 bit -- just to expand real briefly. We all knes coming into 18 the workshop that what would be most constructive is to come 19 prepared with a set of commitments that if, in fact, those 20 commitments would be credited in the reanalysis for the PRA 21 and demonstrate that lower risk level, that the industry 22 would be willing to make. These are some examples. Don't 23- take this as a complete set. The next sessions we're going 24 to be getting into will deal more with the procedural 25 aspects of things that also might go to lowering human error O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 t
82 1 probabilities. So, you need to look at -- this is more
() 2 design related and the next session we'll be talking about 3 more procedural oriented commitments that the industry is 4 willing to put into FSARs, for instance, given the 5 sufficient credit in the staff's reanalysis.
6 MR. CAMERON: Okay. Those are good 7 clarifications. And I'd like to let Ed finish up here and 8 then we're going tc take a break. And I have a question for 9 all of you to consider on that break, in terms of our future 10 discussions. But, go ahead, Ed.
11 MR. BURNS: So, I'm down to the last slide.
12 Frequency of the zirc fire sequences, at least for the 13 non-seismic events that I've looked at, appear to be below 14 what I would call a credible range, 10 minus six range. )
l 15 Accidents that have higher frequencies might be more 16 considered as part of future analysis involving fuel 17 handling mishaps, for example. They would have higher 18 frequencies and may be more appropriate to be considered as 19 part of the severe events that could occur and cause offsite 20 doses.
21 MR. MEISNER: And if you don't mind, I'll just 22 expand on that a bit. And it goes back to what Eay Shadis 23 said earlier, that once you correctly characterize some of 24 these oddball events, what you're driven back to is what can 25 really hnppen. And what the industry is proposing is to O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
83 1 start cunsidering now things like the real probable events,
() 2 like fuel handling accidents. Do your calculations to 3 determine when those accidents, for instance, offsite 4
consequences, drop below the one rem EPA PAGs for emergency 5 planning, as an example, and use those kinds of insights on 6 probable occurrences, to drive the kind of rulemaking and 7 relief fcr exemptions that we're all focused on.
8 MR. CAMERON: Okay. Thank you. We're going to 9 take a break in a minute. But, as you can see from Ed's 10 presentation, there is a lot of information here and, of 11 course, we'll put this in the transcript, attach a copy of 12 this. Glenn, also, went through a lot of information, 13 including staff concerns.
14 What we need to figure out how to do, to use our O) q 15 time most efficiently, is how do we organize our discussion 16 when we come back from the break. Some of the material in 17 Glenn's presentation, some of the material in Ed's 18 presentation may be more appropriate for the mitigation 19 sessions, okay, rather than this general session. We could 20 spend the whole time just reacting to some of the slides 21 that Ed put up, and I'm not suggesting that that would be a 22 bad way to do it. But, we really need to figure out are 23 there major organizational chunks that we can proceed 24 through in our discussion, when we come back.
25 And I see Mike has his tent up and that's good. I I) k/ ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
84 1 just remind everybody when we do come back, if you want to
( 2 talk, put your name tent up. But, I don't -- you know, I 3 sort of hesitate to jump off onto the organizational 4 discussion, before we give everybody a chance to, you know, 5 get out of here for a little break. But, Mike, do you want 6 to wait until we come back or do you have a --
7 MR. MEISNER: Two quick things. One, I just want ;
8 to indicate that in addition to the presentation, we've got 9 a draft report that goes into more detail in all this. We 10 need to flush it out and we'll provide it to the staff.
11 MR. CAMERON: All right.
12 MR. MEISNER: The second things is my 13 recommendation for after we come back from the break is to 14 focus on the area that dominates the whole analysis and
() 15 that's human error.
16 MR. CAMERON: Okay, that's one suggestion for 17 everybody to think about. And I guess I would like to talk 18 to some of the NRC staff during the break on any of those 19 suggestions. John, do you want to say something? !
20 MR. ZWOLINSKI: Just for clarification, the draft 21 report you're alluding to, this would be something you would 22 docket sometime in the future?
23 MR. MEISNER: Yes.
24 MR. CAMERON: Okay. All right, let's go for a 25 break and, you know, let's be back here at 11:00. That O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
85 1 gives you 20 minutes. That's a good break. And think about
() 2 3
the organizational issue I posed.
[ Recess.]
1 4 MR. CAMERON: Okay. Just two administrative 5 announcements. One is if ycu haven't signed it out front, 6 please sign in and leave us your address, so that we can get 7 any information on this workshop or future discussions of 8 this issue or related issues out to you. And secondly, just 9 for all of the NRC staff here, the NRC is. going to have a l
10 small caucus 15 minutes before we are scheduled to come back 11 from our lunch break, whenever that is going to be.
12 I talked to a couple of people about the
.13 organizational issue and I think David may have some 14 comments on that, and we'll go to him in a minute. But, it 15 .makes sense to, I think, deal with any methodology issues, 16 any issues on initiating events, sequences, generic PRA 17 issues. The HRA issue, as we've been calling it, has a lot 18 of discussion-that's going to come up in the mitigation 19 sections. But, I think that it what Mike was referring to 20 before is that there's a generic issue there, in terms of 21 assumptions that are made. I think he wants to discuss 22 that.
23 So, I would like to see if we can -- whatever we 24 agree or -- agree on, in light of the items that we're going 25 to discuss to finish out this session on sequences, to see
[~)
(_/
IdRJ RILEY & ASSOCIATES, LTD.
Court Reporters 1025. Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
86 1 what should be placed up here on the action items, in terms
( 2 of next step. that NRC, the industry, whomever, are going to 3 take, to try to close out those issues one way or another.
4 Okay, that's my two bits on this subject. David, 5 do you have some ideas of where we should go on this now?
, 6 MR. LOCHBAUM: Not really. I was just going to --
l l 7 I have three comments,' based on this morning's session.
8 Should I go into those or do we want to discuss 9 organizations?
l l 10 MR. CAMERON: Let me see if anybody else has some 11 organizational thoughts. I don't want to discuss 12 substantive issues now, okay. But, does anybody have a 13 disagreement? Does everybody understand my vague 14 description of where we're going to -- what we would like to
() 15 do in this -- to finish off this session and does anybody 16 have any disagreement or anything to add on that?
17 [No response.]
18 MR. CAMERON: Okay. Let's do that. Now, David, 19 I'm assuming that, you know, your comments are probably 20 going to be relevant to this particular session; but they 21 may'be relevant to something else, too. But, that's fine.
22 Why don't you go ahead.
23 MR. LOCHBAUM: Well, the first comment addressed 24 that directly. There's been some discussion about future 25 safeguards activity in a decommissioning area. Without a O ANN.RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
87 1 program, it's very difficult to figure out how this effort
() 2 fits into the overall big picture of what the NRC is doing 3 on decommissioning. So, I don't know if comments are --
4 should be spoken at this session, or if there's something 5 coming up down the road that it would be better to address 6 it in that forum. We don't know. This is -- as far as we 7 know, this is it. This is the only -- the only one ever.
8 We don't know that there's going to be something down the 9 road. We've heard some illusions to it, some implications, 10 but, again, without a program, we don't know.
)
11 MR. CAMERON: Okay. That's an -- you know, that's 12 an important point and I think that probably that would fit 13 into tomorrow afternoon's ending session. But, maybe we can 14 get that off the time now, in terms of -- we had some 15 discussion about safeguards. And as Gary pointed out this 16 morning -- a couple of things pointed out by the NRC this 17 morning, is: one is that this report should explicitly 18 address how the safeguard issue is being handled or not 19 handled, why it isn't in here, some of the discussion from 20- this morning. And, Gary, if I mischaracterized that, please 21 correct me.
22 We, also, heard, I think from Stu, that the 23 safeguard issue is going to be dealt with in a future 24 rulemaking. Does the NRC have anything more explicit to 25 add, at this point, in terms of a score card? In other ANN RILEY & ASSOCIATES, LTD.
O' Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
88 1 words, when and how is the safeguards issue for
] 2 decommissioning going to be addressed? Or will the NRC say 3 that yes, we're going to make a commitment to address it in 4 such and such a way? I guess there's some supporters out 5 there for something that I just said; I'm not sure what.
6 But, Gary, Stu -- and I don't know, Stu, this maybe more in 7 your bailiwick. I don't know. But, can you give anybody an 8 idea about when and how the safeguards issue is going to be 9 address?
10 MR. RICHARDS: Not much beyond what I've already 11 said. And the reason I'm somewhat hesitant to speak to it 12 is that I don't -- you know, I don't have any personal 13 experience with the peculiar rules that may go along with 14 dealing with safeguards information in security rulemaking.
() 15 So, I can't really sit here and say though, you know, there 16 will be an opportunity for public comment on that; and then 17 later find out that because of the safeguards information, 18 that's not the way it's done. I just don't know.
19 I do know that there has been some discussion i 20 internally that the -- you know, the safeguards issues need 21 to be addressed. It's part of our integrated approach.
22 We're talking with the security people on that. And I'm 23 sure to a degree that the process allows public involvement 24 in rulemaking, we'll allow it.
25 MR. CAMERON: Well, if you -- can the NRC, at ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
r.
89 1 least, commit to specific evaluetion of what type of process
() 3
~~
MR. LOCHBAUM: Before you --
4 MR. CAMERON: -- might be used and when? Go 5 ahead, David.
6 MR. LOCHBAUM: That wasn't what I was looking for.
7 MR. CAMERON: All right, okay. Go ahead.
8 MR. LOCHBAUM: I was using safeguards as an 9 example of -- there's this decommissioning meeting on spent 10 fuel accidents. There's also some talk about the safeguards 11 in the decommissioning arena. All I was asking for was some 12 program, as to what the NRC's big picture plans are for 13 decommissioning, whether it's one task, three tasks, nine --
14 whatever, safeguards being an example of something else
'( ) 15 that's going on.
16 MR. CAMERON: Okay. Broader issue. Do you want 17 to address that now or do you want to do that tomorrow l
18 afternoon?
19 MR. RICHARDS: Maybe, I can talk with David 20 separately. As I mentioned before, we put our broad plan in 21 front of the Commission in the SECY paper that's available i
22 as a handout. You know, I don't know what kind of detail 23 you're looking for, but that SECY paper lays out what we're i 24 looking for and we're still waiting for feedback on the 25 Commission -- from the Commission on whether we ought to do l
ANN RILEY & ASSOCIATES, LTD.
k Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
90 1 it or not.
2
( MR. LOCHBAUM: Did we -- you did mention the 3 safeguards. I assume that it was complete, but it lacked 4 that' element.
5 MR. RICHARDS: SECY didn't mention safeguards?
6 MR. GUNTER: No.
7 MR. RICHARDS: I think it's in there.
8 MR. LOCHBAUM: Can you point it out for us?
9 MR. CAMERON: Okay. I have a suggestion here on 10 this one. While we're looking for that, let's explicitly 11 have this discussion again about the plan for 12 decommissioning tomorrow afternoon, okay.
13 MR. LOCHBAUM: Sure, a6 Part of the closing.
14 MR. CAMERON: All right.
() 15 MS. ORDAZ: It's under background on page two.
16 That's where it's first mentioned. And Diane mentioned it's 17 also on page six as a bullet.
18 MR. CAMERON: Okay. Let's, you know, specifically 19 flag those and point those out for David. But, we are going 20 to revisit -- we'll specifically revisit this issue and, you 21 know, hopefully,'you will, you know, be able to be with us.
22 But even if you're not, I think we know what your concern 23 is, so we'll do that.. Now, your second point.
24 MR. LOCHBAUM: The second point was the NRC -- the 25 discussion today seemed to be focused on not killing members O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
(
91 1 of the public, which we're all in favor of. But the NRC,
() 2 also, has a rule or an obligation to protect plant workers.
3 And the focus on severe accidents, or specifically spent I
4 fuel pool accident, is only -- or its primary function is to 5 protect the public. There are less severe accidents, 6 including things that lead to spent fuel uncovery without 7 zirc fire, that can pose a very severe threat to plant 8 workers. And that seems to be totally ignored in this 9 decommissioning arena. We think it should be covered, 10 because the worker need to be protected, as much as the 11 members of the public do.
12 The third point was on human performance. In the 13 last 15 years, 23 operating nuclear power plants have been 14 shut down for a year or more due to safety problems. They
() 15 weren't safety problems caused by workers making mistakes, 16 they're not following procedures. It's basically bad 17 management that led to these problems. After a plant has 18 been shut down for a year or more, the NRC does -- takes 19 away its resident inspector. The plant no longer produces 20 any revenue. The NRC is no longer providing the same level 21 of oversight as it did when it was operating. So why should 22 the public have any belief that the management would be any 23 better at a shutdown plant than it's been at operating 24 plants in the past? So, I think the true human performance 25 factor that should be looked at is plant management, not l
ANN RILEY & ASSOCIATES, LTD.
Court Reporters l'025 Connecticut Avenue, NW, Suite 1014 Wasnington, D.C. 20036 (202) 842-0034
92 1 just where the workers follow procedures or respond properly D 2 to an alarm.
[U 3 MR. CAMERON: Let me ask the NRC staff, first of 4 all, plant worker implications for plant workers, how is 5 that or will that be factored into this? And I guess that 6 at the same time, the management issues, how will that be --
7 fus it being factored in? Will it be factored in? And is 8 there some session that's coming up this afternoon or 9 tomorrow, where this would be relevant to have more of the 10 discussion on this issue, rather than just noting it for 11 future discussions. Stu?
12- MR. RICHARDS: My answer may not be very 13 satisfying. Mr. Lochbaum and others have brought up some 14 topics that I think are valid points to consider. I'd like 15 to respond to some, after having had a chance to read our 16 paper. As far as what's being covered in this workshop, I 17 think we're starting with what my boss calls the tallest 18 tree in the forest, which has been identified in the past as 19 the zirc fire issue.
20 The other issues, within design basis issues, I 21 think Gary Holahan already mentioned that we need to -- you 22 know, we need to. discuss how we're going to deal with that.
23 I think the worker protection issues is just like -- well, 24 all the other parts of our regulation.still apply to 25 decommissioning plants, for instance, Part 20.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
93 1 You know, I don't want to see this workshop
( \ 2 broadened in scope. I think we have a pretty full agenda O
3 dealing with -- what's perceived to be the higher risk 4 events right now. But, I will say that I believe our paper 5 does discuss the fact that going forward in the future, we 6 intend to have an integrated rulemaking, as far as the 7 rulemaking process. Of course, there's a lot of public 8 input. So, I think there's going to be ample opportunity to 9 address all of these issues before we reach any kind of --
10 we're talking years here to complete the rulemaking process.
11 MR. CAMERON: Does -- let me ask -- in light of 12 that, let me ask Gary, Gary there were two issues that you 13 thought that it would be useful for the study, to at least 14 explain why they were not dealt with or how they are going (A) 15 to be addressed. Does -- do these two issues that David 16 brought up, do they -- would they merit benefit from the 17 same treatment in the study?
18 MR. HOLAHAN: I think there are a little -- the 19 two of them a little different. The organizational element 20 and its effects on human performance and the accidents, 21 itself, I think need to be discussed, as part of the 22 mitigation system -- mitigation discussion after lunch.
23 Because, I think even if you look at Glenn's view graphs. I 24 mean, we are talking about organization breakdowns, rather 25 than individual operator failures. And I think that's an e' ANN RILEY & ASSOCIATES, LTD.
k -)'
s Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
i 94 1- essential element of the study. And the reliability numbers 2 that we talk about, you know, have to reflect that.
3- The worker dose issue, I guess I'd have to think 4 about a little bit. It might fit into the category of 5 things that we ought to explain. I think I need to think 6 about that a little bit, to see whether, you know, other 7 requirements, like Part 20 and, you know, maintaining a low 8 risk to public health and safety or, you know, adequately 9 address that issue. So, if you're writing -- if that noise 10 as writing on the board that you're doing, I think, the word 11 " consider" is okay. But, I'm not sure that I'm committed to 12 another section of the report, yet.
13 MR. CAMERON: I said -- I said possibly consider 14 on that one.
( 15- MR. HOLAHAN: Okay.
16 MR. CAMERON: And I think what you're saying, and 17 David, you can, you know, when we go to the mitigation 18 sessions, we'll make sure that we try to explicity address 19 what Gary called organizational breakdowns, okay, which I 20 think has maybe -- your sense was the management confidence 21 issue. But, we'll try to work that in. Mike?
22 MR. MEISNER: Just to address maybe -- or suggest 23 a framework for the worker does issue. If I understood what 24 you said, David, it was a -- it was a situation maybe where 25 you had -- you had it drained down and then covered the ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
1
\ 95 1
1 field, but it didn't progress to a -- to a zirc --
() 2 MR. LOCHBAUM: That's just one scenario. There's 3 -- all the things that go on during decommissioning, if they 4 go wrong, could present a threat to the workers, as well as 5 the public. Both hands of that equation need to be properly 6 dealt with and sure that both parties are protected.
7 MR. MEISNER: Okay. I just wanted to suggest, 8 except for draiwndown situation, all the rest of that is 9 within our license basis. So, we have to consider it under 10 the regulations.
11 As far as the draindown that doesn't progress to a l
12 zirc fire, I would also suggest that exactly the same things 13 we're talking about here, and that's the probability of the l
14 draindown or loss of cooling or whatever associated with the
() 15 mitigative actions would not only protect public health and 16 safety, but would do the same for the workers. In other 17 words, what you want to do is prevent the water getting down 18 below the top of active fuel. Having done that, you assure 19 that the dose is from the spent fuel pool, itself, or not 20 accessible with respect to worker dose. I just want to say 21 I think it's pretty much one in the same.
22 MR. CAMERON: Gary, do you want to amplify on 23 that, or follow it?
24 MR. HOLAHAN: Yeah. I think David has raised an 25 important point. For certain periods of time, when you are l
l ANN RILEY & ASSOCIATES, LTD.
L Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 L__.
96 1 providing extra protection because a zire fire is possible,
() 2 and we talked about things like alarms or training or 3 whatever, I think those measures do provide some level of 4 protection to the workers, as well. They are the same 5 issues.
6 However, it's not clear that, let's say, later on, 7 after three years or five years or whatever it is, when zirc 8 fire is no longer the driving issue, whether the same level 9 of, you know, procedures and instrumentation and all of that 10 would still be kept in place. And if it were to be kept in 11 place, it might be needed not because you're worried about 12 driving material offset, but because of the on-site element i
13 of it. So, it might be a reason for maintaining some of the 14 things that you do early in the shutdown for later purposes.
() 15 So, you would shift from needing that protection for public 16 health and safety, to needing those for worker safety. But, 17 I don't think we thought about that very carefully, yet.
18 MR. CAMERON: But, it sounds like that we're going 19 to, because of these remarks.
20 MR. HOLAHAN: It sounds like we're going to 21 consider what we should do about such a thing.
22 MR. CAMERON: All right. Paul, I take it that I
23 Gary's points sort of was what you were going to say?
24 MR. GUNTER: No. I'm going to defer to Ray.
25 MR. CAMERON: To Ray? Okay, Ray.
i ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
l l
97 1 MR. SHADIS: A lot of stuff went by us this
() 2 morning here, a lot of details -- point by point details.
3 And I want to know if we're going to be addressing them l 4 again this afternoon, with an opportunity to comment point 5 by point?
l 6 MR. CAMERON: Yeah, I think that we're -- that's 7 our -- our general objective is to take a look at all the 8 specifics that were talked about within the context of the 9 proper agenda item and to deal with that. Right now, I l 10 think to finish out this morning, what we'd like to do is to 1
11 see -- to consider the points that were made, that are on 12 initiating events; the sequences, see if there's agreement, 13 disagreement on those. Look at the generic -- what I'm 14 calling a generic PRA methodology issue that Mike has
,Q
( ,/ 15 alluded to on human reliability assessment, HRA. So -- and 16 if you don't -- you know, I mean, it's all of your 17 responsibility around the tatle, if you see a point that you 18 want to -- in the proper session, if you see a point that 19 you think needs to be discussed, let's put it on the table 20 and we'll discuss it, okay.
21 What I'd like to do is to see if we can get those 22 -- any points that relate to this front end, rather tha 23 mitigation, that we clear up now. And I might ask Mike to 24 start us off with his issue. Go ahead, Mike.
25 MR. MEISNER: Okay. Well, to cover both your
[~
ANN RILEY & ASSOCIATES, LTD.
%- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l f 1
t
98 1 points, in the interest of time efficiency and initiating
() 2 3
event frequency, and there I'm referring to some of the things that Ed brought up, like lots of outside power 4 frequency, AC recovery, the amount of time to boil, diesel 5 driven pump failure probabilities and the like. We're 6 simply looking for the staff to take that back and mull it 7 over and see if it makes sense. I'm not sure that those 8 things are worth much discussion time today. And if that's 9 acceptable, I'll be happy to kick off an HRA discussion.
10 MR. CAMERON: Let's discuss what Mike just said.
11 And I guess, Mike, for everybody's benefit, you ran through 12 those rather quickly. Is there one page or two pages or a 13 section of the -- of Ed's viewgraphs that deal with that, so 14 that we can be clear about what we're talking about it?
f 15 MR. MEISNER: Yeah. Look primarily at pages 26 --
16 oh, wait, these are all numbered different. Well, the pages 17 are numbered inconsistently. Starting with page 26, that 18 has conservative bias examples on it, through the next 19 roughly six pages, most of those deal with initiative event 20 frequencies.
21 MR. CAMERON: Okay.
22 MR. MEISNER: I skipped over cask drop, because we 23 are going to talk about that us a separate issue.
24 MR. CAMERON: Okay. Conservative bias examples 25 and the next several pages after that, with the exception of ANN RILEY & ASSOCIATES, LTD.
(-}
-- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
n 99 1 cask drop, which, of course, is going to be talked about.
() 2 NRC staff, Gary?
3 MR. HOLAHAN: Let me start off in general to say, 4 we certainly will take your suggestions and go back and look 5 at them. I think what's important for today is that we
! 6 understand what it is that you want us to look at. And so l
7 -- I mean, I can't judge that, at this point. I just ask 8, the staff to acknowledge or nod or something, do they 9 understand what you're suggesting well enough to take it as 10 a comment, to make a judgment about whether it's helpful or 11 we agree with it or don't agree with it.
12 Personally, I didn't quite understand the rather 13 large differences in the heat up analysis. And I though 14 heat up analysis was sufficiently easy, so that even I could (O
_/ 15 do it. And to think that there were big differences in our 16 calculations versus a realistic, I think -- I think I'd like 17- to hear something about what those differences are, so that 18 we know what it is that we're going to go off and look at.
19 MR. CAMERON: Yeah. I think that, you know, in 20 light of that comment, I think we need to, even albeit 21 briefly, have a discussion of this. Because, it may be one 22 thing for the NRC staff to say they understand it, but we 23 also need to get the input of others on those issues, so 24 that the NRC, if any, so that the NRC staff can take that 25 back and include it in their evaluation. So -- .
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 .
a
100 1 MR. HOLAHAN: I'd like to take anybody's input.
() 2 But, we need to understand what it is that's being sent.
3 MR. CAMERON: Go ahead, Ed, I think you're on.
4 MR. MEISNER: The graph that shows the time line 5 merely shows what my interpretation of what the draft NRC 6 document says about pool heat up, which was that 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> 7 was assumed -- greater than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, I think, were the 8 words that were used, was assumed in the -- as the time for 9 the pool to heat up from ambient temperature to pool 10 boiling, and then 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> was used as the time that was 11 allowed in the HRA analysis for any operation actions to 12 take place, as a result of that.
13 But, if you look at what typical -- if you look on 14 the next page, there's a set of tables that talks about what
() 15 typical fuel pool load, as far -- in terms of mega BTUs per 16 hour1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br />, and that fuel -- if you use those kinds of loadings on 17 the pool, you end up with times to boil in the range of 90 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br /> for most plants. Even if you drop -- even if you 19 drain the pool to the bottom ot the canal -- the transfer 20 canal, you'd end up with times in the range of 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> to 21 time boil. So, there's a substantial difference in the head 22 load that's being used. I think, 10 meta BTUS is being used 23 as part of the analysis versus what we see here. That's the 24 difference. j j
25 MR. CAMERON: Comment from NRC staff on -- you ANN RILEY & ASSOCIATES, LTD.
v Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
p 101 1 understand the difference now, I guess.
2 MR. STAUDENMEIER: I'm Joe Staudenmeier, Reactor 3 Systems Branch, NRC. It looks like your heat service for
( 4 the pool is under a megawatt or something like that. I 5 mean, we calculated decay heat loads based on a full pool, i
l 6 at a specific time after shutdown. I don't know what the 7 basis is for your decay heat loads. Once you put the --
8 specify the volume in the pool and the decay heat load, it's '
9 just a straightforward calculation. And I guess we can --
10 ours are for a full pool, I think, one year after shutdown 11 it may be. I don't know what your numbers are based on.
12 We're open to using realistic numbers for the pool under 13 consideration. I don't think there's any real disagreement 14 about how to calculate, it looks like, it's just that your
() 15 decay heat load is difference.
16 MR. BURNS: Yes, I think that's right. These j 17 decay heat loads are based on existing fuel pools with one 18 year -- one year after the last fuel offload.
19 MR. STAUDENMEIER: I guess it depends -- that's a 20 plant specific type of thing and how close you are or how 21 far your pool is, because --
22 MR. BURNS: Right, that's why I gave three 23 separate examples. I don't have examples for everybody.
24 MR. STAUDENMEIER: Yeah, because just in things 25 that we've reviewed, I've seen pool heat loads that are
/ ANN RILEY & ASSOCIATES, LTD.
k%s Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
102 1 signficantly higher than what you have down there for
() 2 numbers for plants that we've looked at for decommissioning.
3 MR. CAMERON: Okay. Glenn, any comment on that?
4 MR. KELLY: When we performed the deterministic 5 analysis to try to see whether or not there were concerns 6 that we had to worry about, about spent fuel pools going to 7 zirconium fire, when we looked at the draindowns, we assumed 8 that the pool was configured in a manner that would be 9 hopefully fairly enveloping for the industry, so that --
10 because when we started out, we wanted -- we were hoping 11 that we'd be able to say, you know, with an enveloping pool 12 configuration and heat loads, that we'd be able to say it 13 wasn't a problem.
14 It didn't turn out that way. We were doing all of f 15 these things in parallel, the deterministic, as well as the 16 PRA analyses. And so our deterministic analyses and looking 17 at the heat loads in that looked at a full pool and with 18 significant heat load, based on that full pool. And there 19 are some plants that have -- may come in for decommissioning 20 that are less than that. But, when we did it originally, 21 and, you know, because we did it in a short period, we 22 attempted to envelop, as far as the deterministic analysis 23 concerned. We didn't try to envelop it, as far as the 24 probablistic analysis.
25 MR. CAMERON: Okay. Gary, does that satisfy your O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
103 1 need for information on that spent fuel heat load -- is that f 2 right, spent fuel pool head load?
3 MR. HOLAHAN: I think I understand why the numbers 4 are different now.
5 MR. CAMERON: Okay.
6 MR. HOLAHAN: What are the right numbers to go in 7 the study or whether, in fact, a spectrum of possibilities 8 should go_in the study is scmething that I think we ought to l
9- address. We don't just want to put in a few typical l 10 examples and say, these are okay; when, in fact, you now, 11 plants have licenses to allow them to put more fuel in the 12 pool. I think, you know, we want to be in a position of 13 saying, you know, any and all spent fuel pool situations ,
14 would be safe, not just a few typical examples.
() 15 But, this may very well be the sort of situation, 16 in which you say, I understand that, you know, realistically 17 the heat loads are a lot lower than the maximum allowed.
18 There should be some mechanism, either in a rule or in the 19 licensee's analysis, that they could do an analysis of what 20 actually exists, you know, at the time. So, maybe time is 21 not the parameter, but an analysis to show that there's 90 22 hours or something.to that effect. But from the technical 23 point of view of understanding how long it takes to boil, I 24 think -- you know, I think we know that.
25 MR. CAMERON: Okay.
I T ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
104 1 MR. HOLAHAN: What -- you know, what to do with it
/%
(V ) 2 is a little different question.
3 MR. CAMERON: Gareth, do you -- Mike, are you 4 going to -- do you want to respond to that?
5 MR. MEISNER: Well, I just wanted to say, take it 6 with the spirit with which it's given. And we're trying to 7 come up with the best estimate approach here, as opposed to 8 an upper bound approach. These are -- when you say typical, 9 typical doesn't mean a few; typical means typical. And 10 that's the spirit with which we provided it. So, I'm just 11 asking you to take a look at it.
12 MR. CAMERON: Okay. Gareth, do you have a 13 comment?
14 MR. PARRY: Yeah. In the spirit of trying to come r
( ,h) 15 up with the best estimates, we have to be, I think, a little 16 cautious about just taking numbers from various sources and 17 characterizing as better than others. And just as an 18 example, I'll talk about the loss of outside power frequency 19 and recovery that Ed showed in his graph, which is on page 20 26. We've got to be careful that the date that he quotes 21 from the NUREG is, in fact, for operating nuclear power 22 plants, which typically have two or three incoming lines.
23 And the recovery is likely to be a priority for 24 reestablishing the grid. It's not so clear, for example, if 25 those conditions are going to apply to a shutdown ANN RILEY & ASSOCIATES, LTD.
s Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l 1
105 1 decommissioned plant.
2 Now, this is -- this comment, I think, is 3 generally broader than this particular issue. I mean, we'll 4 see the same thing when we talk about human error 5 probabilities. It's the conditions at the plant that are 6 going to drive what -- are going to drive the numbers and 7 they're going to determine whether they're reasonable or 8 not.
9 MR. CAMERON: Okay. Let me ask -- go ahead, Stu.
10 MR. RICHARDS: As kind of a process, just so, you 11 know, you know where we're coming from, I think Rich made 12 the comment, we're looking at plants down the road. Most i 13 people hear their plants are passed the point of which this i
14 is going to impact them. You know, we can continue to deal
) 15 with plants on a case-by-case basis, but that's what we're I 16 trying to get away from. That's not an efficient way to do 17 business. So, we need to come up with, you know, a set of 18 -- hopefully, a set of regulations that envelops by and 19 large the industry and allows things to happen without a lot 2C of industry submittals and without a lot of staff reviews.
21 So, that's -- you know, that's what we're looking for.
22 I mean, you're talking about it all depends on the 23 plant. Frankly, that doesn't help me at all, because we're 24 looking to come up with some regulations that are enveloping 25 here, not regulations thet tell us to go and look at each j i
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
106 1 plant on a case-by-case basis, how we're doing now.
( 2 MR. CAMERON: Would the information that you have 3 developed, Ed, and I think Mike said -- used the word 4- " typical," will still -- would that still help to -- help 5 the NRC to develop the envelop that Stu is talking about?
6 MR. BURNS: No. I think the thrust of that 7 diagram was not to say that the specific fuel loading, in 8 terms of mega BTUs per hour, was two versus four mega BTUs 9 per hour. The question was that in the analysis -- the 10 draft analysis, it appeared that the time to pool boiling 11 was set at 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, even though it's greater than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
12 So -- and for these loadings, it was 90 hours0.00104 days <br />0.025 hours <br />1.488095e-4 weeks <br />3.4245e-5 months <br />. Well, if 13 it's four mega BTUs per hour instead of two, well, maybe 14 it's 40 hours4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> instead of 90 hours0.00104 days <br />0.025 hours <br />1.488095e-4 weeks <br />3.4245e-5 months <br />, but it's still different f 15 than what was used in the analysis, which was 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. So, 16 I think there's a difference between trying to be enveloping 17 and trying to be excessively conservative. That was the 18 only purpose of the slide.
19 I think the other point that I would make is that 20 --
is that in terms of risk informed input from PRA, we're 21 looking for best estimate, I think. I mean, always, that's 22 what I've heard is what the Commission wants, is what's a 23 realistic examination of what kinds of accidents can occur 24 and what their relative relationship to each other is. If I 25 don't know that and I have things that are biased one way or ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
l 107 1 the other, then I don't know how to make a decision, based
[% 2 on that information. That's my interpretation of what risk 3 informed means.
4 MR. CAMERON: So, you're saying that the 5 information that you're presenting will -- goes to that 6 point of best estimate?
7 MR. BURNS: There are areas that would seem like 8 they're logical _ areas to reconsider, to make sure that 9 they're'not excessively conservative.
10 MR. CAMERON: Okay. Gary, do we -- is there any 11 other points that we need further information on, in order 12 for us to evaluate, and is there anything that we should put 13 up as an action item for the NRC, in terms of these 14 initiating. events and sequences and the information that's 15 been provided?
16 MR. HOLAHAN: Well, I think we will review the 17 information given and the report that was referred to 18 earlier, in these areas. But, I think it's worth -- and you 19 can put that on the board, if you like -- I think it's worth 20 clarifying one point. My understanding of the desire to be 21 realistic in. risk informed regulation means for each plant, 22 when we make a safety decision relevant to that plant, it 23 should be done on a best estimate basis; not that it should 24 apply to half the plants and not to the other half of the 25 plants. You know, we want to make a safety decision that's O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 I
i
108 1 relevant, saying that each plant is safe; not that on the
() 2 average, they're safe, but there are a few that are not 3 safe. That's not my concept of good safety decisions.
i 4 MR. CAMERON: Okay. The report that NEI is 5 talking about, and I believe Mike, you said -- someone said 6 two weeks or --
7 MR. MEISNER: Two to three weeks.
8 MR. CAMERON: -- something like that. Now, is 9 that report available for others to comment on? Ed?
10 MR. BURNS: It's coming in from NEI, sure.
11 MR. CAMERON: All right. So, the report will be 12 out there for everyone to see. Go ahead, Stu.
13 MR. RICHARDS: yeah, I think we've made a pretty 14 good effort to try and make sure that these involved parties 15 get all the documentation. Dick Dudley is our primary 16 contact on that. So, if somebody here is looking to be on 1
17 an e-mail list for notification or get things, please see 18' Dick. Dick, can you -- and for instance, they mentioned j 1
19 there's a couple of groups in New England, who want to be 20 notified when we do things. We'll put them on our list.
21 MR. CAMERON: I think Ray being on the list will 22 cover one of them and we'll make sure that the other one is 23 on the e-mail list.
24 MR. RICHARDS: Yeah, when we get this report in 25 from NEI, of course, it will be publicly available and we'll
[~\
\-
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
109 1 make sure notification goes out.
Okay.
( 2 MR. CAMERON: And maybe I will put that up 3 there, just so we have a complete list. Gary?
4 MR. HOLAHAN: I think there was one thing that 5 wasn't mentioned yet that's part of our plan, and that is we 6 are subjecting our technical report to a review by National 7 Labs and other experts under contract, as a sort of 8 technical peer review, in addition to, you know, collecting 9 comments from stakeholders. We thought we would -- we would 10 try for independent technical input. And because that's 11 going to take, you know, several months -- to the end of the 12 year, it seems to me that if NEI or other interested parties 13 are sending us input over the next several weeks or even a 14 month, that we'll have the time available and the resources (f 15 available to be looking at those comments. So, this is --
16 you know, written reports in this time frame, of the next 17 month or so, I think, would be useful.
18 MR. CAMERON: Okay. And at some point, people may 19 be interested in commenting on that schedule, also, and what 20 steps we're taking before this issue is resolved.
21 All right, maybe we should go to the generic PRA 22 methodology issue on HRA. And Mike, I'll let you frame that 23 for us.
24 MR. MEISNER: Okay. Well, let me start by saying, 25 I'm no PRA expert. I thought I'd just begin by speaking in O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014
' Washington, D.C. 20036 (202) 842-0034
110 1 plain terms. What we've -- what we've heard from the staff
() 2 is that in the course of five days -- well, in justifying 3 why we wouldn't be successful in the course of five days 4 recovery. We've heard some fairly vague things. You know, 5 we've heard that the operators will be watching T.V. And 6 not just watching T.V. for a shift, eight hours, but 7 watching T.V. for 15 shifts in a row and not recognizing 8 that there's steaming in the pool. Or we've heard that 9 because there's not daily NRC oversight out at the sites, 10 that somehow, again, over five days, the operators will fail 11 to recognize something is going on or even worse, that over 12 five days, they're going to fail to call the fire 13 department.
14 I'd like to just start by bringing some common
() 15 sense approach to this. Let's assume that instead of ten to 16 the minus two or ten to the minus three or even ten to the 17 minus four failure rates that the staff has assumed in their 18- draft report, but let's assume that it's a ten to the minus 19 one. There's a one in ten chance that the operator will 20 fail to perform an action that he's supposed to. Let's give 21 a little twist to it. Let's say that that one in the ten 22 chance applies to a shift, eight hours, okay. Usually, in 23 PDA space, at least that I'm familiar with, you're talking 24 about human errors associated with taking an action, like, 25 you know, starting a pump or closing a valve. Here, we're O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
111 1 talking about a long-term continuous failure to perform.
() 2 But, it seems to me that we could break it up into discrete 3 chunks, eight-hour period, one shift, and then a new set of 4 faces come in and deal with the problem. So, let's say it's 5 ten to the minus one for that one shift. Let's say it's ten 6 to the minus one for that next shift. And let's keep going 7 throughout this five days. That, as a practical way, seems 8 -- and not being a PRA expert, seems to be a reasonable 9 approach to take, in evaluating these problems. And when 10 you do that, of course, I mean, you get driven right away to 11 kind of these very low numbers that nobody believes, which 12 is probably why, in many PRAs, that after a certain period 13 of time, they just assume success.
14 The other thing that we're missing in all of this
() 15 is that -- it's not one operator on shift; it's a couple.
16 There's a management team there. There's support personnel.
17 There's engineers. And, you know, there's even the NRC.
18 Eventually, the NRC will be notified. We don't have it 19 quite right, I'm sure you all will tell us what the right 20 thing is to do. The point is, though, we have five days, 21 seven days, if you believe Ed's calculation. I, as a 22 non-technical person in that area of PRA, I just find it 23 hard to understand, and I'd like to -- I'd like to kick it 24 off with just that plain sense point of view.
25 And I guess just to finish, I would then jump off O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
112 1 from what Ed presented and -- like on the overhead about key
[
V} 2 HEPs expected, we -- the staff has assumed one error number 3 for this whole period. Since it appears that in the staff's 4 view, that's the dominant area to evaluate, shouldn't you 5 get into something like a shift lead basis evaluation and 6 flush out that HRA model? Consider things like shift 7 changes, annunciation response, and the like, and model that 8 maybe one sequence in more detail, to convince yourself as 9 to what would be a more accurate representation.
10 MR. CAMERON: Okay, thank you, Mike. Before we go 11 to the NRC, let's hear from Paul Gunter on this issue.
12 MR. GUNTER: Well, obviously, from a public 13 interest point of view, we want to see conservatisms 14 incorporated has that interest. But, Mr. Meisner, there are
) 15 real events that multiple shifts can commit the same error.
16 Clearly, one -- particularly when they are cost driven. The 17 unheated fuel handling building at Dresden 1 where service 18 water pipes were allowed to freeze suggests that a number of 19 shifts weren't maintaining the technical specifications for 20 that building.
21 The fact that Nine Mile Point 1 had water in the 22 basement. And where -- it was the rad waste building, 23 suggests also that multiple shifts at that facility were not 24 attentive to an issue.
25 But, clearly, the concern is that as reactors L ANN RILEY & ASSOCIATES, LTD.
l Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 i
i 113 1 become -- as they back out of a service sector and become
( 2 more of a drain than a faucet on the facility, that you have 3 an economic driver that needs extra vigilance. And I think 4 that that is what we are looking to see from the Nuclear 5 Regulatory Commission is a provision for that conservatism.
6 MR. MEISNER: I need to respond to that. First of 7 all, the examples you used don't relate to the spent fuel 8 pool management directly, Keep in mind --
9 MR. GUNTER: They do, too.
10 MR. MEISNER: Keep in mind that the operators that 11 we have on shif t have no other job than to watch the spent 12 fuel pool monitor level and temperature and boiling 13 concentration.
14 The other thing is, underlying your premise, is a
() 15 ' suspicion that licensees will act differently under cost 16 pressure. You can believe that if you like or not, but let 17 .me inform you that that is not the case in decommissioning.
18 I hope everyone knows that the cost associated with 19 decommissioning is not borne by the owner of the plant, it 20 is not borne by the management of the plant, that is a 21 direct pass-through to the public. And what we are trying 22 to do is be good stewards of the money we spend in 23 decommissioning. But the fact that we might have too spend 24 a little more money to get some safety benefit is surely not 25 going to inhibit us from doing it, from your suspicious ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
114 1 point of view, because it is not our money, it is the
() 2 3
public's money.
We go through a rate-making process, at least with 4 ng/ plant, Maine Yankee, through FERC. Other plants may go S through FERC or they may go through the local public utility 6 commission. But the one thing that is clear in all this is 7 that it is not the owner's money.
8 MR. CAMERON: Paul, follow-up before we got to 9 NRC?
10 MR. GUNTER: Just briefly. Again, you know, the 11 issue of multiple shifts committing the same problem, 12 whether it relates directly to the fuel issue, what it 13 suggests is a human behavior pattern and that is the context 14 that I bring it up. In regards to the only job you have to
() 15 do, if we look at fire watches on duty, that is all they 16 have to do, we find them building nests in some of these 17 plants and going to sleep. j 18 Actually, that sole source of responsibility can 19 lead to such tedium, as we have seen in that particular 20 arena, that it lowers human response values. But -- so, I 21 mean my concerns remain along that avenue, that I think are 22 justified.
23 But with regard to, you know, that is a total 24 pass-through, I am still not convinced that with the rising 25 costs of decommissioning as projected over the last several ANN RILEY & ASSOCIATES, LTD.
O\ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 ,
Washington, D.C. 20036 :
(202) 842-0034 i
115 1 years, that you are not going to look for some kind of cost
( 2 containment strategy. And I think that is what we are 3 really here about is how to -- how the utility is looking to 4 develop a cost containment strategy for decommissioning 5 operations, and this raises -- this is the context that I sq 6 raise that concern.
7 MR. CAMERON: Okay. Let's --
8 MR. MEISNER: Okay. And that is a valid point.
9 But keep in mind that what we are trying to do is not waste 10 resources. Not waste resources for the rate payers, not for 11 the owners. You don't want us out there having 25 people on 12 shift when we only need two, and the only thing standing 13 between us and having two people is an NRC regulation that 14 isn't applicable to decommissioning facilities. We would be
() 15 open to a whole lot of criticism on the other side of being 16 poor stewards of the money that we are holding for the rate 17 payers. You know, it is a damned if you do, damned if you 18 don't.
19 But I will tell you, the folks I know in the 20 decommissioning industry are pretty responsible people.
21 They don't let people make nests in the corner of the 22 buildings, and they sure as hell are very responsible in 23 dealing with the resources they have. I think it is very 24 clear to everybody around this table that we wouldn't be 25 here if the regulations were well constructed for ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
! 116 1 decommissioning plants. If you don't believe that, then we
() 2 3
should just quit the conference now, it is not worth having.
MR. LOCHBAUM: Sounds good to me. I second that.
4 MR. MEISNER: We are here because the majority of 5 people here believe that, in fact, the regulations are --
6 and are clearly demonstrable to be inappropriate for 7 decommissioning plants and cause excess cost. Yeah, cost is 8 the issue, but the underlying suspicion that you are 9 suggesting, that somehow it pads the pockets of the owners, 10 doesn't -- is not true. We are trying to responsible 11 stewards for our rate payers.
12 MR. CAMERON: Okay. Let's go to the NRC for -- j 13 going back to the assumption that we started out with here.
14 MR. HOLAHAN: What I would like to do is to go
() 15 back to something to Mr. Meisner said early on, which is 16 discussion of how can things on shift after shift and not be 17 discovered. And he used a word that I think is important 18 and he used the word " discrete" -- discrete shifts. Well, I 19 think in a probabilistic world you might say -- in an 20 analysis you might say independence, or you give credit in 21 analysis for this shift being different from the last shift 22 when there is a reason to think that they would bring, you 23 know, different eyes to the issue.
24 And I think even though there are lots of argument 25 around and across the table, we all want the same thing. We ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
117 1 all want operators who do their jobs well. You know, we
() 2 3
want them to be monitoring the things that are important to monitor. And we want them to, you know, reliably deal with 4 issues.
5 One of the difficulties is making judgments about 6 how independent is this operator from that operator or this 7 shift from that shift. The TMI event has interesting 8 aspects to it that I think are relevant to both of these. I 9 think it was mentioned earlier that, obviously, you had a 10 mindset in the control room when the event was going on, you 11 know, multiple people thought they were doing the right i 12 thing. But it is also interesting to note that that 13 situation didn't go on for more than -- was it 2-1/2 hours?
14 And, in fact, it was an independent person coming in saying,
( 15 this doesn't look right, why is that valve -- that valve 16 doesn't look like it is closed.
17 So, I think, you know, in real experience, we see 18 both of these things. We see problems that are carried over 19 from one person to the next, and I think absolutely -- Paul 20 is absolutely correct, we have seen situations carried over 21 from one shift to the next. Well, if those -- the last 22 shift thought it was okay for that pipe over there to be 23 dripping, I guess it is okay.
24 So it seems to me the solution to this issue, we 25_ want the probabilities to be low. We are not arguing that l
ANN RILEY & ASSOCIATES, LTD.
( Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
118 1 we want.the numbers to be high. Okay. Nobody wants the 2 operators to be unreliable, we want them to be good. The 3 question is, what do you need to do to make them reliable?
4 What do you need to do to give us confidence, to give the 5 public confidence that operators are reliable people? And 6 to the extent that, you know. 1 out of 10, or 1 out of 7 10,000 that they get it wrong, that the next shift is, in 8 fact, going to have, you know, an independent set of eyes, 9 so that, you know, we can confidently say, yeah, well, it is 10 not one in a thousand, it is one in a million.
11 I think we need to step away from the argument 12 that says I think they are great guys, or I have seen them 13 building nests in the corner and get down to the point of 14 saying -- what is it that we expect of the utility, of their 15 management on-site, of their procedures, you know, of the 16 alarms available? You know, concrete things which at this 17 stage are just technical issues, which later on can be put 18 in the FSAR, or put in a rule, or a license condition or 19 wherever they need to be, so that people have assurance 20 that, you know, that this is more than, you know, comments 21 made one Thursday morning, and is something that we can be 22 . comfortable with the in the long run to cover, you know, 20 l l
23 years from now, you know, anyone who is decommissioning, you j i
24 know, under whatever economic conditions they are at.
25 You know, our first concern needs to maintaining ANN RILEY & P'90CIATES, LTD.
O- Court .seporters l
1025 Connecticut Avenue, NW, Suite 1014 i Washington, D.C. 20036 ;
(202) 842-0034
119 1 safety. Okay. And so we need to have assurance through
() 2 3
.some mechanism, other than just talking about it, that we are going to have reliable operators.
4 I think part of this is an issue that we haven't 5 dealt with. Okay. And I think, in fact, some -- you know, 6 we may need to go back, and a bunch of people may need to go 7 back to think about what kinds of things. Ed Burns this 8 morning suggested that the fact that alarms occur at 9 different stages of an event may be a mechanism for 10 realerting people that there_is something that they ought to 11 do. Well, I think that is an interesting thought. Okay.
12 You may, in fact, be able to write shift turnover 13 procedures, okay, that alert people to, you know, to do 14 things that in a way, you know, reset their clock. You may
( 15 be able to put up a sign on the wall that says, by the way, 16 if the water drains out of that pool, you are the first guy 17 that is going to be dead, to remind the next shift that, you 18 know, these are real serious safety issues.
19 I think if you have done a calculation of the 20 shine from the pool, lethal doses are pretty easy to get.
21 So it is the operators who ought to have every incentive to 22 knowing that things are working pretty well.
23 So, clearly, there is lots of uncertainty how to 24 do.this. Clearly, there is lots of suspicion and anecdotes.
25 I think we need to get to the point of putting on the table O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
120 1 some concrete steps. I don't know today that we know what
() 2 3
those steps are. Okay. In one of Ed's vieugraphs, he suggested that Gareth Parry look into this. Well, we 4 brought him to the meeting.
5 But I think this idea that experts need to think 6 about this issue is a good thought. But what the experts 7 need to think about is not just whether the number is 10 to 8 the minus 4 or 10 to the minus 6, they need to think about 9 what kinds of things can you do to make operators extremely 10 reliable for long -- long-term type events. And since I am 11 not one of those experts, I would really like to hear 12 someone who thinks they are jump in.
13 MR. CAMERON: Just let me ask Mike, and I think we 14 need to hear what Ray has to say, too. On this last issue 15 that Gary brought up, what we need to do to have reliable 16 operators, I think this is the type of thing that the 17 industry was prepared to offer perhaps during the sessions 18 this afternoon about what procedures are in place or what 19 you might be willing to put in place.
20 MR. MEISNER: Yeah, that's right. And we will be I 21 talking about that in great detail this afternoon. We are 22 more than willing to make commitments.
23 I need to say a couple of things, though. First 24 of all, before we cast too much aspersion on operators, you i
25 know, I got my ops manager here and he is probably biting ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
i 121 1 his tongue. NRC approves the operator training program.
()' 2 NRC comes in and inspects it, you know, just like you do in 3 an operating facility. I am not sure why you have this high 4 level of suspicion that operators won't do what they are (
5 supposed to do and somehow you need a sign up on the wall 6 that says you are going to be the first one to die. You 7 know, I think that is just silly.
8 We will go into details of commitments, but I was 9 trying to generate a discussion that would -- of issues that 10 would try to meet a straight face test. Is there anybody 11 around this table that honestly believes that if they were 12 on shift, that somehow over five days they would fail to see 13 steaming in the pool and subsequent major draindown to the 14 top of the active fuel? I mean that is what I am trying to 15 present here, that it doesn't make any rational sense at-16 all.
17 You can see it on fast-acting events for operating 18 reactors, you know, when they have got two minutes to 19 respond and they will make a mistake, or -- you know, and 20 you could probably see it on fire watches who are minimally 21 trained people who are working for minimum wage and they are 22 just sitting there in the plant waiting ft tire to 23 happen. But, my goodness, these are trained individuals 24 that follow a program prescribed by and approved by the NRC, 25 who are dealing with simples systems and very few things can i
\ ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
I I
122 1 wrong. You can lose cooling, you can -- you know, you are
() 2 3
getting a slow draindown.
issues.
They know how to deal with these 4 And here we are talking about them as they are 5 complete dummies that won't be able to do anything for five 6 daya. That is what I am trying to get across here. We will 7 get into the commitment side of it later. But I was just 8 looking for a practical approach that says, why don't you 9 model this as it really should be? You know, you have got 10 different people coming in at different times and it is not 11 just operators, you know, our security people find a lot of 12 things before our operators do. There is all kinds of 13 things to take into consideration that we will be talking i
14 about this afternoon.
) 15 MR. CAMERON: Okay. I think that Mike pretty well 16 characterized what his concern is, and there may be a 17 context that the NRC can offer about the common sense aspect 18 of this. And I see that we have a bunch of cards up here..
19 But let's go to Ray first 20 And, also, I just want to tell -- Mike mentioned 21 one of his people out here. I just want to say that we will 22 go on to people in the audience for a brief period of time j 23 before we end this session, if we ever end this session.
24 But before we end this session, to hear what you have to I 25 say.
Ih ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
I l
123 i I
1 But go ahead, Ray.
(f 2 MR SHADIS: I just want to give a little bit of 3 perspective on the economic driver issue that Paul Gunter 4 raised and that is that there are a number of companies out 5 there who are eagerly pursuing decommissioning contracts 6 around the country, and they are did it on the basis of I 7 can do it cheaper than you. That is an economic driver that 8 is independent of the fact that the citizenry are going to 9 be dunned for the bill at the end. And it does have its 10 effect.
11 Additionally, you know, this is not the place for l 12 it, but we could go through the line item budget for Maine 13 Yankee's decommissioning and see whether or not all the 14 funds are appropriately applied to areas that have the k 15 safety significance, and I think maybe we would find they 16 are not.
17 So, you know, that particular argument that it is 18 the rate payer paying doesn't necessarily carry through. It 19 may in Maine Yankee's case, it may not in the case of 20 another facility. And the same thing is true for the 21 training of these fuel handlers. It may well be that we 22 have a team at Maine Yankee that has the training and the 23 esprit de corps and so on to maintain vigilance. It may not 24 be at another plant.
25 But even in the best of facilities, we find that O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington. D.C. 20036 l (202) 842-0034 )
1
l 124 1 over time with tedious tasks, that workers do tend to fail.
()
p 2 Maine Yankee had that experience in the spring of 1997 when 3 their fire watch teams falsified their logs. So, you know, 4 it can happen to anyone. And I think that Paul is 5 absolutely correct, you really need to take a conservative 6 view of this and be cognizant of what is generally called 7 the human condition.
8 MR. CAMERON: Thank you. I guess that I would 9 just caution people, just for terms of, you know, perhaps 10 efficiency, -- because we can get into a lot of arguments 11 about did this happen at this particular place, or did that 12 not happen at that particular place, -- that we try to use 13 examples that are meaningful, as that is, but not put 14 anything that might be in dispute on the table.
C
( )\ 15 Glen.
16 MR. KELLY: I was really just -- two areas that I 17 wanted to quickly point out. First is that several people 18 mentioned how they were proposing at the -- they like the 19 idea that the risk assessment should be conservative, so 20 that when they apply that, that they are going to assure 21 that they are getting good results, that the public will be 22 protected.
23 And from our standpoint, when we are performing 24 risk-informed decision-making, what we want to do is use the 25 PRA to give us the best information that we can get, which
[ } ANN RILEY & ASSOCIATES, LTD.
\/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
125 1 is a best estimate analysis, and then from that we can
() 2 decide what additional conservatism we need to lay on top of 3 it in order to protect the public health and safety.
4 And that is in particular what we did here in our 5 risk assessment. The PRA was performed on a best estimate 6 basis, The words that Ed got out of the report near upper 7 bound apply to one particular deterministic input that was 8 performed on the basis of a near upper bound. And as I 9 remember, that was one of the heatup times and that was 10 based, because we were trying to look at the pool 11 configurations that would be bounding, such that, if 12 possible, we could envelope most of the plants that were out 13 there.
14 But when we do our PRA itself, it is, to the best
) 15 extent that we can, we are going to do it on a best estimate 16 basis and that is what we did.
17 And the second point that I wanted to make was 18 about Mike's common sense argument, and his description that 19 he was not a PRA expert, and I would agree with him in this 20 particular case. It is -- as a mathematician and a person 21 who has been doing PRA for quite a few years, you cannot 22 merely use your judgment or your common sense to make 23 decision-making. If we did that we wouldn't need a PRA in 24 the first place, we wouldn't need the kind of insights that 25 the PRA gives us. What has turned out a lot of times is j i
ANN RILEY & ASSOCIATES, LTD.
Os Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
126 1 that things that seemed to common sense when we did an
-s
( ) 2 analysis, gave us insights into new areas that we hadn't 3 appreciated were potentially problems.
4 And I think in this case what happened is -- I 5 will speak for myself. When I went into this evaluation, 6 performing this risk assessment, I thought we were going to 7 come out with very low numbers. It didn't turn out that 8 way. That was my expectation. But we let the numbers come 9 out where they came out, doing it on a best estimate basis.
10 In particular, in the area of human performance, I 11 think it is very important to understand that most of the 12 time when we did our evaluation it was not on the basis of 13 assuming that these events were independent and that a lot 14 of times has to do with we have seen out there, based on the j
() 15 kind of procedures, the things that we have, and I will let 16 Mike and Gareth talk in more detail about that. But we had 17 good reasons for coming up with our system and evaluating it
'18 the way we did, and we don't want to -- we don't want to 19 justify our numbers.
20 Again, I want to come back to the key that what we 21 want to do is use our analyses to point out where we want an 22 insight that something could be a concern. And then we want 23 to understand what does the industry have that makes that 24 concern go away. And that is really where we want to focus, 25 not about what is the exact correct number. l ANN RILEY & ASSOCIATES, LTD.
O* Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
)
127 1 MR. CAMERON: Gareth.
l
() 2 MR. PARRY: Yeah. I would just like to endorse 3 what Gary said. I think he had it right when he talked 4 about what we need to do is understand the conditions, the 1
5 plant practices that would enable us to make a judgment that 6 a particular human error probability was low.
7 HRA has been struggling for many years to become a 8 science rather that a black art. And we have learned quite 9 a lot recently about the things that influence human 10 performance, and I think it is important to put those in a 1
11 structured format for a technical basis document, rather 12 than just make arguments that, well, these guys are 13 reasonable, they have got five days, it will work out fine.
14 We need a structured process by which we can make those 15' arguments for low probabilities, not precise numbers, just i
16 whether we can believe that the likelihood of success is 17 correct or not. And it stretches through things like 18 characteristics of the administrative procedures at the 19 plant, the state of the instrumentation, response 20 procedures, training, staffing levels. And we have to 1
21 address all those issues and understand the potentials for 22 dependen,y across the different crew members, across the 23 different shifts before we can definitively say that we 24 believe there is no human performance problem.
25 MR. CAMERON: Okay. I think we are going to aim O ANN RILEY & ASSOCIA'rES, LTD.
Court Rcporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
128 1 for quitting at 12:30 here. Okay. And I guess that -- I
() 2 3
don't know, Mike, if you heard anything that answered your question about -- it seems like it is just common sense that 4 this wouldn't be ignored.
5 MR. MEISNER: I think what I heard was, let's get 6 into the next session where we can start talking about what 7 kind of procedural controls and other things are in place.
8 MR. CAMERON: Okay. Jon, the stenographer, did 9 you get that?
10 THE REPORTER: Yes.
11 MR. CAMERON: Okay. You got that.
12 And is there anything more that can be said or 13 should be said about this particular five-day, one-day, 14 outside of let's get into looking at this point, before we
) 15 just ask the audience? Gary? l 16 (No response.)
17 MR. CAMERON: All right. We are going to have 18 about 10 minutes out here for anybody who wants tc comment 19 from the audience on what they heard this morning.
20 Yes, sir. Just state your name and affiliation.
21 MR. CANAVAN: Ken Canavan. I am with GPU Nuclear.
22 And I guess I was a little confused this morning -- excuse 23 me -- about the purpose of the generic analysis.
24 Originally, I had assumed that the purpose of this generic 25 PRA analysis was to risk-inform future regulations for O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 )
(202) 842-0034 I
129 1 decommissioning. But I am a little concerned about some of
() 2 3
the statements that were made, such as you don't want to focus on the numbers, you want to decide where improvements 4 can be made. But I think the numbers are all part of that.
5 I think it was clear from some of the information 6 Ed. Burns had put out that it can drive where you would 7 allocate your resources, depending on what you consider to 8 be the human errors, what you consider to be the hardware.
9 I tried finding some of the numbers in some of the 10 references and had trouble doing that, and also disagree 11 with some of the values, some of them seemed quite high, j
-12 both in the human error area and loss of off-site power and 13 recovery, and in some hardware cases. So I think that is an 14 area that could be looked at. And I am not going to comment
() 15 on the HRA staff, I think that has been talked about enough.
16 MR. CAMERON: Okay. Thank you. Do you -- was 17 there an outstanding question on what the objective is here, 18 or was that just a comment? ,
J 19 MR. CANAVAN: Well, if I had it right, then I am 20 okay. In other words, if the purpose was to indeed make 21 future regulations risk-informed, then I think we are all 22 set and I don't really have a question. But if that is 1
23 incorrect, I would like a clarification. j 24 MR. CAMERON: Okay. Thanks, Ken. j 1
25 MR. ZWOLINSKI: That is correct.
1 1
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
130 1 MR. CAMERON: It is correct. Okay.
() 2 3
Anybody else out there?
MR. HENRY:
Yes.
Bill Henry, engineer manager at Maine 4 Yankee. One thing I wanted to try to tie where we had a 5 question about the BTU content in the pool, in the 10 6 million versus the 3 million numbers Ed had. I think what 7 -- I think both sides may be correct. When we did our 8 analysis, we used the ANS methodology and had numbers up in 9 the 6, 7 million range. We did a pool heatup test and they 10 were 40 to 50 percent lower. I think the tie may well be 11 that, yes, you can base it on a 10 million BTU per hour 12 content based on ANS, but a real number is probably just 60 13 percent of that. So if you are looking for a best estimate, 14 maybe that is where you tie can come in.
15 MR. CAMERON: Okay. Thank you very much.
16 Anybody -- anybody else? Yeah, Gary.
17 MR. HOLAHAN: I would like to follow up on that 18 comment because a 60 percent difference in decay heat makes 19 an enormous difference in anybody's analysis.
20 MR. HENRY: Especially a 60 percent upper bound, 21 40 percent lower.
22 MR. HOLAHAN: I understand.
23 MR. CAMERON: We can't hear you.
24 MR. HOLAHAN: A 40 percent change is --
25 MR. HENRY: I believe there were at least three O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washingcon, D.C. 20036 (202) 842-0034
131 1 full heatup tests done in New England, and they all showed 2 similar numbers.
3 MR. HOLAHAN: I think we would like to see that 4 information-to see if we could use it in some fashion.
5 MR. CAMERON: Can you get that information to him?
6 MR. HENRY: Sure.
7 MR. CAMERON: Yeah, he is going to get that 8 information to you.
9 Okay. It is about 12:25 on my watch. Let's be 10 back here at 1:30, and we are going to start with an 11 industry presentation on mitigation. All right.
-12 [Whereupon, at 12:23 p.m., the meeting was 13 recessed, to reconvene at 1: 36 p.m., this same day.]
14
) 15 16 17
'18 19
'20 21 22 23 24 25 O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
132 1 AFTERNOON SESSION 2 (1:36 p.m.)
3 MR. CAMERON: Okay. Good afternoon, everyone. We 4 are going to move into the next discussion segment, which on 5 your agenda is called " Mitigation of Events," and we are 6 going to start out with an industry panel led by George 7 Zinke. And in a minute I will just ask the new people to 8 introduce themselves to you.
9 George is going to present some overview material, 10 and I think, if I understand this correctly, George, you are 11 going to then go to specific people for specific areas to 12- say some stuff. Then George was going to come back to the 13 concerns that were express by the NRC. And I think that 14 that is what we will use to kick off the discussion period
() 15 with George stating one of the concerns. He will say 16 something about it, let's open it up for discussion, then 17 let's go to the next concern. And I don't need to -- we 18 don't need to be rigid about that, but at least that will ,
l 19 give us some type of organizational structure for this. !
l 20 So, George, I am going to turn it over to you. l l
21 And you may want to introduce your team or have them l l
22 introduce themselves.
23 MR. ZINKE: I guess we need to go ahead and 24 introduce all the new players. I am George Zinke, I am the 25 director of regulatory affairs at Maine Yankee.
I' ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
133 1 MR. BALL: I am Bill Ball, I am the ops manager in 2 Maine Yankee.
3 MR. CURRY: Jim Curry, decommissioning engineering 4 manager at Oyster Creek.
5 MR. HASELTINE: I am John Haseltine, the director 6 of Connecticut. Yankee.
7 MR. WITHROW: I am Greg Withrow, I am from Big 8 Rock Point, the manager of engineering, licensing and soon l 9 _to be operations. j I
10 MR. CAMERON: Okay. Thank you. George, go ahead.
11 MR. ZINKE: Yes. We are going to go through these l
12 slides fairly quickly, but I will explain as we go along.
13 You can put the next slide up.
14 As we have already stated this morning, and this
) 15 is kind of a repeat, you know, the dominant sequences we 16 have been looking at, internal fire, loss of power, loss of 17 spent cooling, for intermediate term, loss of coolant 18 inventory. Next slide, seismic and heavy load. So as we 19 looked at mitigating factors, we were looking at that 20 grouping of events.
21 Next slide. As we looked at these events and how 22 they eventually relate to public health and safety, the 23 concern with internal-fire is that at some point in time, 24- you know, with the fire, you get a loss of cooling, with 25 loss of power. It is the same kind of thing that the real ANN RILEY & ASSOCIATES, LTD.
~s/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
134 1 issue of concern is the loss of cooling, which would happen A
i 2 immediate -- immediate when you lost power. And, of course, 3 the loss of cooling was event trees, you know, modeled by 4 itself. And one of the concerns with loss of cooling is 5 that you eventually lose inventory, and then we get into the 6 zire fire scenario. So all of the loss of cooling are not 7 an event, the end in themselves, but they drive to something ;
I 8 else.
9 The seismic, you know, concerns with the seismic 10 could be that either you lose cooling, or you lose 11 inventory. And the heavy load drop, the concerns ends up of l .12 a losing of inventory because the drop has ruptured the 1
1 13 liner or somehow caused a flow path that you could lose 14 inventory, f 15 The differences between these, as far as operator l 16 mitigating actions, they are generally the same. The )
l l 17 procedures are generally going to be the came. With things l 18 like internal fire and loss of power, there are some extra 19 actions that may be taken upfront. But, ultimately, the 20 kinds of questions that were brought up this morning that 21 deal with, you know, operator error or management systems, l 22 of culture, those are going to be the same for any of the l 23 scenarios.
24 One thing to keep in mind is the event initiators, 25 next slide. But there are several of these that are i ANN RILEY & ASSOCIATES, LTD.
l Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
135 1 self-revealing or very obvious when they occur. A loss of f)
J 2 power is hard to miss because the lights go out. And the 3 seismic event, in order to get a seismic event that is going 4 to rupture, you know, an operator is going to know that --
5 you know, it is something he feels, you know, it is pretty 6 obvious.
7 Heavy load drop is going to be obvious because if 8 you are handling heavy loads, there is somebody there, so he 9 is going to see that some event has happened.
10 The other three, internal fire, loss of cooling, 11 loss of inventory, as we have seen from some of the 12 discussions this morning even, that detection becomes 13 important because they aren't self-revealing.
14 The next slide. What I am going to do now is 15 discuss very briefly the fire and the loss of power, and 16 then I am going to move into the loss of spent fuel pen 1 17 cooling. And at that point we will get into the detail of 18 all of the training and the procedures. So, really, what I 19 am going through in these first few slides is those things 20 that might be a little bit unique to fire and loss of power.
21 Dealing with internal fire, there are prevention 22 characteristics at a decommissioning plant. There are still 23 work controls, the control of combustibles. So it is not 24 like we are going to decommissioning and all the controls 25 are off.
ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
136 1 And these are regulatory driven. Detection, 2 there's operator rounds, there are security rounds, fire 3 watches, detection devices where the fire protection devices 4 are tested because of the regulatory framework they are in.
5 When we go into decommissioning relative to fire 1 6 protection, we do come under a NRC rule, and it specifies I 7 have to maintain fire protection, but the purpose has 8 changed. The purpose has changed to decommissioning kind of l
9 criteria. And, indeed, the regulation itself is not '
specific, saying how many smoke detectors I need or how many
~
10 11 of this. But a real key in the transition from operating to 12 decommissioning is that there is a change mechanism and a 13 criteria threshold that I have to meet, that I have to not 14 decrease the effectiveness of the fire protection plan.
() 15 So some of the things that we will talk about this 16 morning of, well, maybe this equipment is not required or 17 this equipment gets removed, before a decommissioning plant 18 can do any of that, it has to meet the threshold of not 19 reducing the effectiveness of the plan by regulation.
20 The mitigation of a fire, there's pre-plan 21 strategies. What I mean by pre-plan is that if one of the 22 strategies has to do with hooking up hoses to various 23 locations or piping or system, that that -- you know, the 24 hoses are either there or that we have got the right heses 25 are pre-staged, you know, so that everything is going to
[~'
\-
ANN RILEY &. ASSOCIATES, LTD.
Court Reporters 1025 Conneccicut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
137 1 fit. So you don't end up in like a fire scenario and l [%) 2 saying, well, now, what am I going to do? You know, these 3 have been pre-thought-out.
4 Trained on-site personnel for their fire 5 responsibilities. We train the off-site firefighters. They 6 come to the site and this is one of those things that 7 changes throughout decommissioning. So if we are changing 8 the responsibility of the off-site firefighters, we bring 9 them on, train them, tell them this is what you would have i 10 to do, this is where you would have to hook up. Find out 11 what problems they would have and make sure that, as best we 12 can, we would have a successful event.
13 -There are redundant fire pumps that carry over 14 from the operating plant. It is like I said before, in
() 15 order to get rid of that, you would have to show that it 16 doesn't reduce the effectiveness. So, fire suppression is 17 routinely tested, and, again, because of a required fire 18 protection plan.
19 During this time, one of the big differences 20 between a decommissioning plant and an operating plant is 21 that the concern, the primary concerns is the fuel pool..
22 And you don't have lots of systems and lots of complicated 23 cable runs. You are pretty well limited in what you have in 24 quantity compared to an operating plant, which makes a lot 25 of mitigation efforts much less complicated, given that you l
I /'D ANN RILEY & ASSOCIATES, LTD.
Iss l Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
138 1 are only trying to get cooling back or inventory versus the
() 2 3
complexities associated with recovering in an operating plant. .So temporary repairs of damage to cooling systems if 4 more feasible because you are not dealing with complicated 5 controls and interlocks. You are just trying to start a 6 pump and get some water somewhere else. So there become a 7 lot more options and easy things you do because the task 8 itself is easier.
9 We end up with the water required. There are a 10 lot more options available, even off-site, availability of 11 bringing in portable pumps that can be real small, nothing 12 special, hooking up hoses. The kinds of strategies that we 13 can follow, you know, it is many more and much simpler than 14 what it would have been for an operating plant of -- how do
() 15 I get water into the vessel?
16 MR. ZWOLINSKI: George.
l 17 MR. ZINKE: Yes.
1 18 MR. ZWOLINSKI: For clarification, help me 19 understand, are you speaking in the context of the spent 20 fuel pool and associated systems being an isolated island of j 21 some sort?
22 MR. ZINKE: I would say either. There are some 23 differences. If it is an isolated island, then you have --
24 things are more together. If it is not in an island and you 25 are taking those pieces that were associated with the
~
-/ ANN RILEY & ASSOCIATES, LTD.
\ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
139 1 operating plant, you know, of course, you may have longer
() 2 cable runs and they may go through various rooms and, 3- because of that, your fire protection may have to be larger 4 as far as, well, wehre do I do fire rounds?
5 So there is difference relative to that. But what 6- I am speaking to is either, although there would be some 7 difference. Both have to be effective.
8 MR. ZWOLINSKI: So the actual fire protection 9 program plan is modified as systems are taken out of 10 service, but what you were talking about was the fire 11 protection plan for anything that would support the spent 12 fuel pool and associated systems?
13 MR. ZINKE: Yeah. There is a number of 14 regulations that get tied in there and that is -- to be real i
() 15 precise, we have the 50.48, the evaluation of the fire 16 protection plan.
17 MR. ZWOLINSKI.: Right.
18 MR. ZINKE: And then, if we are crediting systems l 19 for emergency preparedness. So let's say I am crediting the 20 fire pump for emergency preparedness, then when I would 21 change the emergency preparedness plan, I have another 22 regulation that is also tied to effectiveness. And the same i
23 way it is in the SSAR for a variety of reasons. If I am 24 going to change it, then I have the 50.59. So there's i 25 various regulations, but bottom line, to get rid of them,
> ANN RILEY & ASSOCIATES, LTD. I Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034
140 1 you have to through these and make sure that all of their
[ 2 functions, that you are not reducing effectiveness of those 3 critical areas.
4 MR. CAMERON: George, can I just interrupt you one 5 second?
6 MR.'ZINKE: Yes.
7 MR. CAMERON: John's comment reminded me that I 8 probably should check an assumption I was making with the 9 group. You know, obviously, there is a number of ways to 10 conduct discussion on this presentation. And I was a little 11 afraid that if we just start asking questions or having 12 discussion on each slide, that we may never get to the --
13 get it a3L in, get to the end of this. I was sort of 14 assuming that when you got to the discussion of addressing 15 tha concerns, that'any questions or comments about these 16 previous slides would probably be revised, so that we could 17 wait for a discussion then.
18 But'I also know that there may be instances where 19 someone doesn't understand something that is on a slide. So 20' do you want to just wait till we -- the group, I am asking, 21 do you want to wait till we get to the concerns to open up 22 the discussion and deal with all questions then, or do you 23 want to have clarifying questions ~as we go along?
24 MR. ZINKE: I know from my point, if I am asked a 25' question and I know I am going to be covering it later, I ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
p 141 1 will say that, so that we kind of keep the flow. But I
() 2 3
think it is going to be easier if there is some -- just a clarifying question of what do I mean. Better to take it at 4 the time before everybody forgets.
5 MR. CAMERON: Okay. If there is -- if you need to
! 6 know, let's interrupt and ask. But if it is starting to get 7 into the commentary, dialogue, discussion, let's hold it.
8 Okay. Thank you.
9 MR. ZINKE: Next slide. A loss of off-site power 10 event. Again, when an event happens at a plant, the 11 operators will not just follow one strategy. They won't 12 say, okay, I have got to get off-site power back, so I will 13 work at it a few days, and if that doesn't work, then I will 14 try something else. As part of their training, that they
) 15 are expected to have multiple success paths going at the 16 same time. And this is the same kind of effort that happens 17 in operating plants, which may be difficult to model, but it 18 is a fact that they -- that is what the expectation is for 19 the operators.
20 So, parallel mitigations strategies pursued, 21 restoration of the off-site power. There is going to be 22 pre-planned off-site assistance. When we have planned for a 23 -- What if you lose power for X number of days? Where could 24 we get a portable generator? Where would we get fuel?
25 Would this be reasonable if there is snow outside? You ANN RILEY & ASSOCIATES, LTD.
U- Court Reporters l 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
142 1 know, those kinds of things are pre-thought-about such that
() 2 3
the operators are not having to invent things at the time.
MR. CAMERON: George.
l 4 MR. ZINKE: Yes.
l 5 MR. CAMERON: We have one clarifying question from i 6 Glen Kelly.
l 7 MR. KELLY: Yes. George, earlier you made a 8 comment that -- I hid my comment here. You were talking i
9 about the -- first, that operators wouldn't follow one l
10 strategy. Is that actually proceduralized? Does it say in 11 the procedures that you are supposed to follow parallel 12 paths, or do you just leave it to the operator to have the 13 common sense to do that?
14 MR. CAMERON: Can we show people which slide we
() 15 are on to that deals with this question?
16 MR. ZINKE: Let me delay the answer to that one in 17 a second, because --
18 MR. CAMERON: Okay.
19 MR. ZINKE: You know, it is going to tie -- the 20 same answer is going to tie on the next slide.
21 MR. CAMERON: Great.
22 MR. ZINKE: So we are going to get into that in 23 much more detail.
24 MR. CAMERON: Good.
25 MR. KELLY: My second part here is you said when O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
143 1 you decommissioned, you don't decrease the effectiveness of
() 2 the fire plan. And I guess my question is -- let me throw l 3 out an example. My fire plan may require me to be able to l 4 put out fires in a spent fuel pool area, but there is not a 5 lot of things to burn in a spent fuel pool area. I probably 6 could handle it with some fire extinguishers, you know, a
.7 battery of fire extinguishers, and, therefore, I might be 8 able to argue that I don't need fire pumps because my fire l 9 extinguishers are adequate to put out any fires that I would 10 have in there.
11 If I did that, then I could get rid of my fire 12 pumps, because I have not effectively decreased my fire plan 13 effectiveness. That is not the same as saying that I have 14 not decreased my ability to put water into the pool or to A
(),
15 cool that. So, that is what I need to understand, because 16 you talked about it from the point of view of the fire plan, 17 and not against capability of providing makeup to the pool.
18 MR. ZINKE: Right. And that is the clarification 19 I was trying to make on the different regulations involved.
l 20 So, you are right, relative to its fire plan function, that ;
21 gets evaluated under 50.48, under that criteria, relative to 22 the fire pump's other functions, you have to evaluate it 23 under 50.59, which would get into probability and 24 consequences of an accident or an event. So --
25 MR. KELLY: Well, 50.59, I will debate 50.59. My O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
r 1 144 1 understanding of 50.59 is that it deals with design basis
! ~
(3) 2 events, and we are in beyond design basis space, in 3 emergency preparedness space, and, therefore, 50.59 would 4 not apply, as I understand it. But I could be wrong. l 5 MR. CAMERON: Could we -- this is a generic, this 6 is an issue to discuss, so we will come back and discuss 7 this. And then, Mike, you can make your comment on it at 8 that time. Okay.
9 MR. MEISNER: You can see me moving, right.
10 [ Laughter.)
11 MR. CAMERON: Go ahead, George.
12 MR. ZINKE: All right. Next slide. This is the 13 second section we are going to get into more detail of the 14 same kinds of things that are also applicable on the fire i
s f g),
15 and the loss of off-site power. In general, what do the 16 operators do and what is their training? From a detection 17 standpoint, we recognize -- in your draft report it credited 18 operator rounds once or twice per shift, on 8 to 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> 19 shifts. We do have control room annunciation, pool 20 temperature level, irradiation monitors.
21 The draft report made a distinction that there are 22 times when these are not required operable, which is not 23 exactly right. The term " operability" for an operating 24 plant most often applies to those things that are in tech 25 specs. And these things are not in tech specs. However, l
l
-[)
N/
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
F 1 145 l
1 for an operating plant and a decommissioning plant, we have
() 2 3
a SSAR which does have requirements, and includes requirements on when something has to be able to function.
4 It won't necessarily use the word " operable," just because 5 of some fine points on definitions, but, in practicality, if i
6 they are in the SSAR, they are required to be operable, and '
7 they can't be just igr.ored. So there are controls, it is 8 just different than the tech spec controls.
9 And then again, like an operating plant, much of 10 the things that an operating plant has to maintain and keep l
11 working are in the SSAR, not in the tech specs.
12 Operator turnover process, we are going to talk 13 about that in just a minute.
14 And then, additionally, we have security rounds 1
( 15 that -- in detection. People are in the plant. They would 16 observe certain obvious things. And that is not to say that 17 a security i officer would observe, you know, a half inch 18 change in level. But this is relatively long-term accidents 19 or events that we are talking about, such that in the 20 long-term, they would recognize large changes in levels, 21 certainly temperature.
22 From our experience with security people, they 23 recognize a whole lot of things. They will recognize if a 24 pump is off and they are used to it being running. And they l 25 bring these things up.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
146 1 Next slide. Mitigation. The first bullet, detail
() 2 3
procedures used following detection. Part of the discussion this morning dealt with level of detail in a procedure.
4 Indeed, generically, our procedures are not such that for a 5 fuel pool event, that it is going to say if it drops one 1
6 inch, do this; if it drops a foot, do this; if it drops 18 7 inches, do this; if it -- you know, they are not 8 prescriptive in that sense, that would be covered more under
]
9 the training and the experience of the operator.
10 They are prescriptive that they provide that they 1
11 provide the options that are there and the "how to" -- how 12 to implement an option.
\
13 At this point, what I want to do is turn this over 14 to Bill Ball, who can go into a lot more detail from an
() 15 operations standpoint about the kind of rounds and the 16 expectations, the expectations that are in our off-normal 17 procedures, and even the turnover process.
18 MR. BALL: My name is Bill Ball, I am the ops 39 manager at Maine Yankee. Let me -- I had a' lot of things I 20 wanted to say, so now is my opportunity. Let me just go 21 through and, hopefully, I will cover the stuff George wants 22 to me say -- go over.
23 But one of the things I want to bring up is these i
24 operators are highly skilled individuals. They are not the l 25 normal craft labor type individuals. They only have one ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 ,
Washington, D.C. 20036 (202) 842-0034
147 1 thing to do and that is protection of the spent fuel pool,
() 2 3
and they take it serious.
We are required to take local rounds once or twice 4 a shift. Also, during the turnover process, they are'also 5 required to collectively, together, check different 6 instrumentation. That is another check, or self-check, so 7 to speak. They are also required, when they take these logs 8 out there, they bring these logs back-and the shift manager 9 is supposed to -- or he does, he reviews these logs and 10 approves the numbers, so to speak, what the actual level is, 11 to'make sure it is in the acceptable criteria. So it is 12 another check.
13 The other thing I wanted to mention, which was 14 kind of touched upon a little bit this morning, was how long I 15 of a shift an individual can stand. That is all controlled 16 by tech specs. An individual normally can't work more than 17 a 40 hour4.62963e-4 days <br />0.0111 hours <br />6.613757e-5 weeks <br />1.522e-5 months <br /> work. He normally doesn't work more than an 8 or 18 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> shift, and it is all controlled by tech specs that 19 the NRC has approved.
20 And the training aspect, although the quantity of 21 training has been reduced, the quality has -- you know, the 22 quantity being there is less systems out there that the 23 operator needs to worry about, but the quality is still the j 24 same. He still gets trained on the design of the system, I J
25 the operations of the system, the mitigation accidents or l
, ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025' Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
148 1 actions. He is still examined once a year. He still needs
[Y\1 2 to take an initial examination. And, overall, the training 3 program is approved by the NRC. That is all submitted for 4 your approval.
5 I don't know if I touched upon everything you 6 wanted me to there, George.
7 MR. ZINKE: Talk about the rounds.
8 MR. BALL: The rounds, I did a little bit. The 9 rounds are done once or twice a shift. There are specific 10 logs that'they need to take. They are, like I said before, 11 they are second checked by the shift manager.
12 MR. ZINKE: Tell them specifically how they would 13 -- how they look at level, what they look at.
14 MR. BALL: Oh. On level they have to -- on level,
) 15 there is a couple of things they do. They have, in the 16 control room they have a computer based monitoring where 17 they check off on that, and then they also go out locally 18 and check level on a mechanical dipstick that we have. And 19 I am sure most decommissioned plants have dual function 20 ' level indication. There is other checks out there that they 21 check like anti-siphoning devices, they need to check them 22 clear.
23 All this stuff is in the DSAR and then we turn it 24 -- operations turn those into logs, and then those logs need 25 to be taken and they need to be reviewed by management. All ANN RILEY & ASSOCIATES, LTD.
[\~'/I Court Reporters i 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034 i
149 1 that happens.
() 2 3
MR. ZWOLINSKI: Bill, for clarification for all of us, have you just shared with us the way business is 4 conducted at Maine Yankee? The advice -- are you sharing 5 that is the way you see it generically across the industry?
6 So help us with that understanding, if you could.
7 MR. BALL: I think I share it across the industry.
8 I haven't done a formal, you know, go and ask every single 9 decommissioned plant out there, but I have got a very high 10 confidence level that all decommissioned plants operate like 11 that. But these guys could probably jump in.
12 MR. HASELTINE: John Haseltine from Connecticut 13 Yankee. We had some discussions on this yesterday just to 14 make sure that what Bill is saying is accurate, and we all
() 15. concurred. But, basically, we do the same at Connecticut 16 Yankee. We make our rounds once a shift in our case. They 17 are required to record level and temperatures, discharge l
18 pressures, check certain things, write them down and bring l i
19 them to the control room and they are checked, and they do i 20 the same thing on the next round. i 21 The training is very similar. In fact, I think 22 the training is even better on spent fuel than it was when 23 it was an operating plant because now it is quite a lot i 24 larger than it used to be, and I think the operators are a 25 lot more familiar with it because it 9.s the only thing in ANN RILEY & ASSOCIATES, LTD.
04 Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 i
150 1 town. And it is a formalized training program. It requires
() 2 you to get a grade, it requires you to get approval. It is 3 audited and inspected by the NRC and it is something, you 4 know, you have to do in order to have an operator look over 5 the pool.
6 MR. ZWOLINSKI: This is a regular Part 55 program.
7 MR. HASELTINE: Yes.
8 MR. WITHROW: Greg Withrow from Big Rock. I think 9 actually what they are stating is the minimum requirements 10 we do. I know that they do more than that, and so do we. I 11 mean we have alarms on level, alarms on radiation, that sort i
12 of thing. We also use beepers on people's belts that are l 13 electronic beepers, so that if they get an alarm on level, 1 14 it goes off. You know, so there is a lot of things.
(7S,) 15 We record rounds twice a shift. We lot hourly 16 those readings and all of our stuff is alarmed, you know, so 17 -- and that is our primary responsibility, and the training 18 is very similar. And in your first year or two, you 19 typically retain the same operations people you have in the 20 past. If you go out five, ten years, then I think some of 21 your questions come into the realm a little bit more. But I 22 think if you are looking in the one to two to three year 23 timeframe, which is concerned relative to zirconium fires i 1
24 and other things that are really important, then I think you I 25 really have to look at the present operating staff being l
I
[') ANN RILEY & ASSOCIATES, LTD. ,
V Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
151 1 carried through for at least a couple of years. At least in
()
,~
2 3
all of our cases that is exactly what is happening.
MR. CAMERON: Okay. I would just note that John's 4 point on generic versus plant-specific, in this case, an 5 individual plant, I think it is a question probably that 6 applies to all of these areas, not just the loss of spent 7 fuel pool cooling. So just something to keep in mind.
8 Do we have some clarifying questions? All right.
9 Go ahead.
10 MR. RICHARDS: Just some questions to try and gain 11 on, you know, your experience. I guess we have what, four 12 plants represented? I think the training probably isn't 13 Part 55, because that is licensed operator training and 14 these are certified fuel handlers. But, nonetheless, it is p) q_ 15 submitted to us for review and approval. Is that a 1
16 continuing training or a one time deal?
{
17 MR. WITHROW: Continuing.
18 MR. BALL: It is a continuing training program.
19 It is very similar to senior reactor operator continuing ,
20 program. In fact, we benchmarked it off our operator 21 training program and massaged it, and that is what we 22 submitted.
23 We are required -- the crew is required to attend 24 training on a quarterly basis and take annual written exams 25 and annual operating exams. l O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
152 1 MR. RICHARDS: Is zirc fire potential covered as
() 2 3
part of that training?
MR. BALL: No per se. The mitigation -- the 4 mitigation action for a lower level or loss of cooling is 5 covered.
6 MR. RICHARDS: Could you tell us a little bit 7 about the instrumentation available? I mean you have do you 8 have one level switch or two? Do we have a TV camera? How 9 many devices or redundant indications of level do you have?
10 MR. BALL: I can speak of Maine Yankee and maybe 11 each one needs to speak of their own, because there may be a 12 little bit of a difference there. We have two remote level 13 indications that provide input into the control room. In 14 addition to that, we have two local level indications that 15 the operator can use, and he does use when he goes up and 16 takes local readings. And in addition to that, we have a TV 17 camera from the control room that can basically zoom right 18 in on the level, indication level -- level indication also.
19 MR. RICHARDS: How much of that is required by 20 regulation?
21_ MR. BALL: I couldn't tell you. George might be 22 able to help. Tech-spec-wise.
23 MR. CAMERON: George.
24 MR. ZINKE: As far as the details of how much is 25 -- you know, what -- how many indicators and how often you O
\-
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
153 1 look, that level of detail is not tech spec -- you know, 2 generically, isn't going to be tech-spec-based, it is not 3 going to be driven by specifics in a regulation. That level 4 of detail is typically in a SSAR. Now, from plant to plant 5 it may differ. You know, one plant says I am going to have 6 two remote and one local. One plant may say I am going to 7 have two local and one remote. But it is going to start 8 out, in moving into decommissioning, it is going to start 9 out with whatever you had an operating plant. In order to 10 reduce that, then you are back into some kind of an 11 evaluation, depending upon if we are talking about 12 indication or if we talking about fire equipment, or if we 13 are talking about plant equipment.
14 But the specific details are not going to be in d 15 actual regulations or guidance.
26 MR. CAMERON: Can we continue our canvas in {
17 response to your question? In other words, Jim and then 18 John, and Greg, whatever you have to offer on Stu's 19 question.
20 MR. ZINKE: In just a second, my advisement, I 21 wasn't real clear. That level of detail isn't required for 22 an operating plant either. Yeah, so we are basically 23 staying with the same level of detail that was required when 24 things were very complicated and complex and massive.
25 MR. CAMERON: Okay. Thank you.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
)
Washington, D.C. 20036 l (202) 842-0034
1 1
l 154 1 MR. BALL: Can I just mention one more thing?
() 2 3
Just to compare what we were when we were an operating plan, compared to what we were for indications in a decommissioned 4 plant, at Maine Yankee, what I had in the control room was 5 one high level -- one high and low level annunciator and one 6 high temperature annunciator, compared to all kinds of level 7 pressures, temperature indications and alarm just for the 8 pool. So at least at Maine Yankee, the focus has really 9 been on just the pool and to get the best detection 10 capability as we can. And it has actually improved from 11 what we had as an operating plant 12 MR. CAMERON: Okay. Thanks, Bill. ;
1 13 Jim. I 14 MR. CURRY: Yes. I am from Oyster Creek, so we n
i) s, 15 are a plant that hasn't yet shut down, hasn't yet made the l
)
16 decision to shut down, but we will commit our kind of design 17 philosophy is to put in the appropriate instrumentation and I 18 system redundancy and system quality characteristics to 19 assure that we have an adequate reliability to demonstrate 20 that the draining of the fuel pool is not a critical event.
21 So we are listening for the success criteria. We certainly 22 take clues from our peers in terms of instrumentation that 23 they think is appropriate, the BWR standard tech specs, or 24 Owners Group work in that area.
25 So I think George has got some words up there
/ \ ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
155 1 about pre-plan strategies commeneurate with event speed, 2 complexity. That is really -- that is really our key as we 3 listen to these groups and come up with, gee, what is -- how 4 good is good enough? How low of a probability do you need 5- to feel comfortable that this event is not going to happen?
'6 We are committed to designing our systems to that 7 level of confidence. So we will have whatever redundancy is 8 appropriate to achieve that.
9 MR. CAMERON: Okay. Let's go to John. And, you 10 know, keep in mind that we going to be revisiting these 11 areas, too. So, you know, what we don't get now, we will j 12 get out later on. John.
13 MR. HASELTINE: Yeah. At Connecticut Yankee the 14 tech specs do require us to have a level high and low in the 15 control room alarms, as well as temperature alarms. We have 16 since added continuous level indication and temperature, but 17 we didn't have it when we were operating. And it is only a 18- single track.
19 MR. WITHROW: Yeah, Big Rock is similar to Conn.
20 Yankee, We have it in our tech specs it requires 21 instrumentation, single channel for both level temperature 22 and we also have radiation detection for area monitors.
23 But, you know, being sort of simple guys from the 24 Midwest, we sort of like just going up and looking at the 25 fuel pool and seeing what the temperature reads on a ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
156 1 thermometer and what it shows on a little ruler up there and (j 2 that seems to work pretty good when the instruments aren't 3 in service.
4 But, you know, the other things that are real 5 critical is to think about what those level indications tell 6 you and where those alarms are. Those alarms are probably, 7 in our case, five, ten days before you could ever get to a 8 particular problem. And what I mean is they are set at the 9 very top level and just inches below normal operating level 10 of the fuel pool, where to get down to fuel you have to go 11 ten -- you know, 10-15 feet. So you have lots of time 12 between the time you get an alarm and the time you have to 13 do something relative to the type of issues we are talking 14 for significant releases.
,~
15 MR. CAMERON: Okay. I think we have -- before you 16 probably were going to go to loss of inventory next.
17 MR. ZINKE: Yes.
18 MR. CAMERON: I think we have some clarifying 19 questions over here with Glen. j 20 MR. KELLY: There were a couple of people that 21 mentioned places where people were -- and I want to 22 emphasize the word " required" -- required to take local 23 rounds, required once or twice a shift to go someplace, were 24 required to take logs. What -- in the context at your 25 plant, what does " required" mean? Is that something that O w' ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
157 1 -you feel is an NRC requirement? Is that something that you
( ) 2 have an administrative requirement that you set on your 3' operators? What does requirement mean?
4 MR. ZINKE: I will take that. Generally, in the 5 terms that'" required" have been stated here, if they said a 6 tech spec required it, you know, of course, that is much 7 more closer to regulatory. A number of the details like, 8 you know, -- How often do you do rounds? What specifically 9 are they going to look at? -- part of that is driven by the 10 SSAR. So if it is in the SSAR, then it, -- you know, using 11 that term " required" because it requires it in the SSAR.
12 Beyond that, there will be things that, from plant 13 to plant, they will chose to do just because they think it 14 is the wise thing to do. Also part of the answer in that, 15 in this discussion we are having today, the very last slide, 16 which I will just draw you -- you know, if there are things 17 that we are talking about that would significantly change 18 the PRA, then we are ready as an industry to commit to those 19- kinds of things like putting in, if it doesn't already 20 exist, putting in the SSAR that you would have rounds.
21 MR. WITHROW: So, do keep that in mind. Our 22 difficulty right now is not knowing, well, you know, are 23 these things you have already credited and we just can't 24 rel' from the report, or they are things -- so that is why 25 Uhin open discussion.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
158 1 MR. CAMERON: So this whole issue of what is
() 2 3
required at one plant, what is required industry-wide, what should be required is all connected to the problem, too, or 4 the issue of -- you can't really tell what has been credited 5 in the study in terms of requirements. And I don't even 6 know, if we talk about what is required or what you would 7 commit to, do we need to worry about what this in the draft 8 study. But we can talk about that. j 9 MR. ZINKE: Yeah. Just as a point of reference, 10 because I know, as we are going through this and look at the 11 -- well, what is assumed? A key thing was Case 3, and Case 12 3 was presented as being slightly above regulatory I
- i. l
)
13 requirements. But as we read through nearly all of the l 14 assumptions on Case 3, they look to us that they are below
() 15 regulatory requirements, that there is now way you can say l 16 you are complying with regulations if you have gotten into 17 that kind of a condition.
18 MR. CAMERON: The issue is still important in 19 terms of what the NRC does with the final study about are 20 these requirements below or above.
21 Glenn, do you have another clarifying question?
22 Go ahead.
23 MR. KELLY: Two things. First, it was important 24 to us to understand whether some of these were 25 administrative procedures that the utility itself had l
[v ANN RILEY & ASSOCIATES, LTD.
Court Reporters i 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034
159 1 imposed on itself, because that adds rigor to something as 2 against I just. choose to do it because it seems like it is a
- 3 good idea and maybe I will do it this time, and maybe I i
4 won't do it next time. There is no real penalty to me as an 5 operator for not doing that. It is important for us to 6 understand to what extent it is formalized, that this is 7 expected that you are going to do this.
8 MR. HASELTINE: Let me try -- this is John 9 Haseltine. One of the things that is in our tech specs is 10 Reg. Guide 1.33, and you have to follow requirements in the 11 Reg. Guide, and the Reg. Guide requires you to have normal 12 operating procedures for, in this case, spent fuel cooling 13 and the associated systems. It requires you to have 14 annunciator procedures for annunciator alarms. It requires 15 you to have off-normal procedures for any types of 16 accidents. And, also, you know, in this case, rounds, et 17 cetera, it comes under procedures.
18 So it is not something that you can just say I am 19 not going to have -- I am not going to have these because it 20 is not a requirement. It certainly is a requirement, and we 21 do have all those, they are all formalized and approved by, 22 .in our case, a PORC, and whatever the equivalent of PORC is 23 for other plants here. So it gets a detailed review and 24 every time we change it, it gets the same review. So this 25 is a requirement.
ANN RILEY & ASSOCIATES, LTD. I Court Reporters j
, 1025 Connecticut Avenue, NW, Suite 1014 '
l Washington, D.C. 20036 (202) 842-0034 l 1
160 1 MR. BALL: And Maine Yankee is also tech spec'ed 2 to'the same requirement.
3- MR. CAMERON: Okay. Thank you. Can we go to loss 4 of inventory then?
5- MR. ZINKE: Let me move to just a few more 6 subjects on here, because they are essentially the same as 7 the inventory. We have talked about the pre-plan 8 strategies. Another key in an event is that personnel 9 resources do get supplemented. Now, there can be a variety 10 of things that drive that. You know, one that is regulatory 11 driven will be the_ emergency plan. Once you get to the 12 thresholds in the emergency plan, you will make a lot of 13 notifications. If you got to threshold of an NRC 14 reportability, you are going to be making some 15 notifications.
16 But another thing that drives that, I will let 17 Bill talk about, but I mean just because staffing is low, 18 and the operator in charge knows staffing is low, it seems 19 that they are very aware of their limitations and it seems 20 less likely that they are going to hot dog it or just try to 21 handle it on their own than call for help.
'22 Bill, can you expand upon the expectations for an 23 operator, particular on back shifts?
l 24 MR. BALL: Yeah. I think George pretty much said j 25 it. The expectations that we have explained to the
- ANN-RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 ]
I I
J
161 1 operators is there is almost nothing that you have to do
() 2 immediately any more, and they have been trained like that.
3 We only have -- we have one additional operator besides the 4 person in the control room. And the expectation is anything 5 out of the ordinary, they are supposed to call ops 6 management and we will get them help. It doesn't mean they 7 can't take immediate actions for things that need to be 8 done, but number one on their list is get help.
9 MR. ZINKE: The next bullet, the multiple options 10 available. Again, as we look at all the plants, the options 11 are different. But what is consistent is that there are 12 multiple options. There is going to be multiple water 13 sources. They may be different at every single plant, but 14 there are water sources available. Water transport O
yj 15 mechanisms like some plants may be able to transport it via 16 pipe or hoses or buckets, or whatever, but, again, the 17 details may differ from plant to plant, but the fact that 18 there are multiple ways of transporting water is consistent.
19 The same way with power sources. Some may keep a 20 diesel, some may bring in a littler diesel. Some may depend 21 upon the ability to get a generator on sight, you know, 22 within a day. Some may keep more than one off-site source, 23 some may not. So the differences in design, again, are 24 going to differ from each plant, but the fact that there are 25 multiple options available seems to be consistent between
[~ ANN RILEY & ASSOCIATES, LTD.
\- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
162 l 1 the plants.
2 Now, inventory, next slide. The rest of these 1
3 slides are going to go pretty quick now because, in general, j l
4 everything that we have talked about --
5 MR. CAMERON: Let me just interrupt you for a 6 second. I think Paul has a clarifying question.
7 MR. GUNTER: Yeah. Do I need the microphone or 8 can I -- can you hear me over there for the transcript?
9 MR. CAMERON: Can everybody else hear him? All 10 right. Go ahead.
11 MR. GUNTER: I am wondering what kind of analysis 12 goes into common mode failure for, let's say, service water 13 systems and how that is picked up in risk analysis. And how 14 is your level of confidence in looking at common mode 15 failure, where is that documented? You know, I can hear you 16 say that you hm a got multiple sources, but, obviously, if 17 they are all frozen, that raises an issue.
18 MR. ZINKE: I am trying to think of how to -- as 19 far as analysis and PRA analysis, I would leave that to the 20 NRC on how they would have modeled something like that. For 21 our purposes of determining that there is redundancy or that 22 we have multiple ways of doing this, sometimes there may be 23 a common mode failure, sometimes not. However, for a great 24 deal of what we are talking about in mitigation, it becomes 25 not a concern in the sense that particularly things like l
l
/O ANN RILEY & ASSOCIATES, LTD.
O Court Reporters 1 1025 Connecticut Avenue, NW, Suite 1014 i Washington, D.C. 20036 (202) 842-0034 j
1631 1 off-site are pretty well independent of things on-site.
() 2 Generically, across the industry, some may have 3 two sources of water and maybe there is one common pipe 4 between them. However, if that froze, there is some 5 connection that could bypass it with a hose. So we don't 6 have any specific analysis that analyzes the single failure, 7 but, in general, because of the kind of event we are talking 8 about, even if you have single failure modes like a pipe 9 freezing up, there are options of getting around that.
10 SPEAKER: George, those are just initiating 11 events?
12 MR. ZINKE: Yes. l 13 MR. CAMERON: Any other elaboration from -- Jim, 14 go ahead.
) 15 MR. CURRY: Yeah. The consideration of single 16 failure of the PRA has to be done, so if you do have a 17 single failure point, you basically that failure 18 probability. I mean George talked about a pipe, a common 19 pipe. The probability of a pipe failing, a passive failure 20 is what it is. When you look at the system, if you end up l 21 taking out two components, you end up taking out two 22 components.
23 But, absolutely, a PRA would be -- it is a common 24 wisdom that I think the industry has grown up with, don't l
25 -overlook the common mode failures. That is the great
, , ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
164 l' weakness and that is what you have to be careful with when
() 2 you start playing with these numbers, because usually that 3 -- once you design a system with adequate redundancy, it 4 usually is the common mode failures that will actually limit 5 how good you can be.
6 We are taking a look at -- we are doing an 7 analysis of the probability of draindown, as many other 8 folks are. If you like, afterwards we have our resident 9 expert here and he can give you a detailed discussion of how 10 he actually treats the common mode failure. But I think the 11 key point for this group is you can't overlook it in any 12 PRA.
13 MR. CAMERON: Okay. Could we just get a, 14 hopefully, simple statement from the NRC folks on common
() 15 mode failure of whatever it is.
16 MR. CHEOK: For the staff's risk analysis, what we 17 did was we had assumed common mode failures for the 18 hardware, the pumps and the valves, but not from the 19 sources. We assumed that from on-site and off-site sources, 20 they would be worse enough that the source common mode 21 failure itself will not be a problem, but for hardware 22 failures we have modeled that.
23 MR. CAMERON: Glenn, any further -- anything to 24 add on that?
25 MR. KELLY: Well, the other thing was that the O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
i I
l 165 1 system that we modeled was so simple, it had quite a few
() 2 single failure points in it that the common mode failures 3 wouldn't have been the dominant contributors. There are so 4 many other places that a single failure would stop the 5 cooling system that you had to -- and then it will be 6 expected that they would have had to go to a fill mode where 7 they just' add water to the spent fuel pool if they are l 8 unable to fix a failure point.
9 MR. CAMERON: John.
10 MR. ZWOLINSKI: If we can back up, in the entire 11 process when we essentially review a licensee's application l 12 for an operating license, these plants are analyzed for 13 common mode failures. And we take that through their 14 license life to assure that events haven't occurred in the
) 15 industry in which we have identified potential common mode 16 failures, and that is when we might go into our retrofit l
17 type activity.
l 18 MR. LOCHBAUM: Can I respond?
19- MR. CAMERON: Sure.
I 20 MR. LOCHBAUM: You are stating that during the 1
21 operating license review, the common mode failure reviews 22' are applied for safety equipment, not non-safety-related 23 equipment like the spent fuel pools are for most plants.
24 MR. ZWOLINSKI: It is primary safety-related 25 equipment, that is right.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
166 1 MR. LOCHBAUM: So they would not be reviewed for
() 2 3
the equipment we are talking about today?
MR. ZWOuINSKI: I know we have done it for safety 4 grade spent fuel cooling systems, but many of the spent fuel 5 pool cooling systems are not safety-related.
6 MR. LOCHBAUM: So that review would not apply.
7 The Tarc reviews would not apply to the four going forward 8 decommissioning cases.
9 MR. ZWOLINSKI: Right. And I think that is where 10 the PRA folks went beyond the typical review that'is 11 performed when we issued a license.
12 MR. CAMERON: Rich, maybe you had better --
13 MR. ZWOLINSKI: The issue that Paul raises, is 14 common mode failure an issue that the staff should be (O
,/ 15 mindful to, so on and so forth, is valid and we have 16 attempted to get after it.
17 MR. CAMERON: Rich, why don't you try to add a j i
18 final point on that, and then we will go back to George. l 1
19 MR. BARRETT: I am not sure how mach I can add, I
20 but I do want to say that there are some terms that have !
l 21 been used here that are -- that perhaps interchangeably that !
22 might not -- might not have been the best thing to do. .j 23 It is true that, I believe it is true that for 24 these systems that are put into the plants during the 25 decommissioning phase, they are not reviewed for single l 1
0, ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
167 1 failure. And in the licensing arena, in the design basis
() 2 arena, that is really the standard we go to, is the single 3 failure criterion, as opposed to in the risk arena where we 4 try to make a realistic assessment of availability, 5 reliability and single point vulnerabilities and common mode 6 or common cause failure situations.
7 So, in the -- I would say at that at the current, 8 in the current situation, we do not have single failure 9 proof systems for decommissioning plants. However, in the 10 risk analysis that we did, we would have taken into account 11 all types of failures, including common mode failures. But 12 as Glenn said, common mode failures were not as important as 13 single point vulnerabilities because of the simplicity of 14 these systems.
(f 15 MR. CAMERON: All right. George, do you 16 want to --
17 MR. KELLY: Mike showed me some numbers and he 18 said that there are a couple of cases where you have l 19 parallel, like you have two pumps coming off a single line, 20 and there were enough places where we had parallel lines 21 there where the common cause failures were high enough that 22 we can say that it was a no-never-mind. So we did look at l
23 it, it is modeled in there, but it is not orders of 1
i 24 magnitude different from what we found for the single point i
25 failures, but we did consider it, and it is modeled in what l
l l
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 i (202) 842-0034 )
168 ,
1 we did.
I 2 And I just want to say, I don't think common cause 3 failure is going to be an important point here so much that 4 -- because the PRA already does look at it, and I think we 5 have got that area pretty well handled.
6 MR. CAMERON: Okay. Great. George, you will see 7 that have a summary slide up there. I am not sure you are !
8 ready for that yet, but --
9 MR. ZINKE: That's all right, because we are going 10 to -- the next couple of slides had to do with loss of 11 inventory. And the way we have looked at loss of inventory 12 is, basically, from all the subjects we have talked about, 13 it is the same, so there is nothing new on the loss of 14 inventory, but a major difference is the speed of the event,
() 15 that it is a quicker event. And we have maintained that the 16 quality of people that we have, the training, you know, i 17 particularly if we talk about minimum requirements, that is 18 sufficient to respond to these. It is not all these extra i
19 things that we have done, but the regulations as they are 20 right now.
21 There was one thing said this morning as far ar, 22 you know, the ad hoc decisions which gets into what level of 23 detail is in this procedure that I am required to have, and 24 we feel that with the required training and required 25 qualified people in these positions, and the fact that when O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
169 1 they get into an event, they are going to supplement that f( )
2 with a lot of second-guessing type people, that the ad hoc 3 decision versus very prescriptive, do this at this level or 4 this at this minutes, that we think from a probability of 5 success, we don't see the issue.
6 MR. CAMERON: Okay. A clarifying question here, 7 Rich Barrett.
8 MR. BARRETT: Yeah. I understand the similarity 9 between the loss of inventory, mitigation strategies, and 10 the ones that you have just described for loss of cooling, 11 but I am curious, has any of your plants ever gotten into an 12 evolution where you, for operational reasons, deliberately 13 manipulated the level of water in the pool?
14 MR. CAMERON: Okay. Bill Ball.
15 MR. BALL: We did I would say about a year ago, 16 but PH was too low and the pool was getting out of spec in 17 chemistry, so we have deliberately added and drained water 18 out of the pool. Now, when I -- when you say deliberate, we 19 are talking probably a four inch differen7e changes in the 20 water, four or five inches of change, but it was to bring j 21 chemistry back into spec. And it was done under an approved 22 procedure with an approved 50.59.
23 MR. CAMERON: Jim, John, Greg, anything on that?
24 MR. HASELTINE: Nothing that I can remember in 25 Connecticut Yankee.
i l
- ~h ANN RILEY & ASSOCIATES, LTD.
s- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
170 1 MR. CURRY: Nothing significant that I can recall.
() 2 3
MR. CAMERON:
All right.
Okay. Greg is shaking his head no.
Go ahead, George.
4 MR. ZINKE: With regard to that question, I mean 5 there may be evolutions that maybe move water or, you know, 6 that maybe have some potential that, you know, something 7 wrong happens, and that is why we get -- you know, I am not 8 emphasizing in this human performance part, but you get into 9 issues like seals on gates and siphon protection, so, you 10 know, we recognize that this particular presentation wasn't 11 really aimed all the hardware things that are particularly 12 important for preventing inventory losses.
13 MR. CAMERON: All right.
14 MR. ZINKE: Now, we are at summary.
( 15 MR. CAMERON: All right. Summary now time.
16 MR. ZINKE: To us, the things that seem the most i
17 important, that make a difference, that better explain why,
)
18 you know, the people would actually, you know, be successful 19 in their mitigation efforts, the whole event scenarios that 20 we are dealing with are much simpler than in an operating 21 plant. We are talking about something you can see. You can 22 see the fuel pool, you can see the water. You can see water 23 drop. Very different than when you are dealing with a 24 reactor and it is all closed up and you are trying to figure 25 out, well, what is going on inside of it, and you make lots ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 i
.]
i l
171 1 of assumptions and you have multiple systems, and one has to 2 work in 10 seconds and one has to work after that 10, but 3 not before something else.
4 Mhat we are dealing with is much simpler than 1 5 that. Detection is assured. We have talked about the 6 operator rounds, the various methods of instrument 7 detection, the turnover process that, given a long-term 8 event, it is going to be detected. 'And, again, the long h 9 for the long intermediate events, the mitigating actions are !
10 simpler than they would have been in an operating plant. So 11 when we talk about, well, maybe we won't have a former SRO, 12 but most of the knowledge that former SRO needed, you don't I 13 need anymore.
14 I mean we don't have all those complicated
)
15 systems. So the knowledge that an operator needs, the 16 things that he has to pay attention to.while on shift, he 17 has got the fuel pool versus having to make decisions of, 18 you know, J5 things more important than the fuel pool and 19 which do I do first, he has got the fuel pool.
20 We move to the last slide, conclusion, and then we 21 will go to the staff concerns list from this morning. We 22 believe the industry approach assures mitigation, but we 23 will commit to actions discussed, provided appropriate 24 credit is given in the PRA. And then we have talked a 25 little bit about that earlier, e ANN RILEY & ASSOCIATES, LTD.
C Court Reporters i 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
172 1 MR. CAMERON: So that sort of gets around the
() 2 3
problem of whether Case 3 is above or below, because -- or at least starts to get around it because you would commit to 4 the action discussed previously, or all the things that you 5 had on your slides, and then there would be credit given for 6 those in a PRA. And I guess the question for the NRC would 7 -- you know, are there other actions that should be -- do we 8 need other actions?
9 But maybe you should go into the concerns and then 10 we can get into -- let's get into a discussion of the 11 concerns and get into a discussion generally.
12 And I am sorry, Paul, go ahead.
13 MR. GUNTER: Just a clarifying question. On page 14 6 of the mitigation activities, under parallel mitigation
() 15 strategies pursued, you have got here personnel resources 16 supplemented on detection. Example given, emergency 17 planning. Okay.. Am I incorrect to state tnat -- part of 18 what is going on here is to eliminate emergency planning, 19 right? Incorrect. Okay. Or to downscale -- scale down?
20 Scale down at least?
21 MR. MEISNER: I will clarify it as soon as you are !
22 done.
23 MR. GUNTER: But I mean it seems -- I just want to 24 be clear where we are in terms of -- I mean it could be 25 viewed as elimination of effective -- of evacuation or O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue,-NW, Suite 1014 Washington, D.C. 20036 ,
(202) 842-0034 '
173 1 emergency plan. If, in fact, you don't have enough
/~ \ 2 personnel on hand, that, in effect, is an elimination. And NJ 3 what we are interested is on what -- where is the level of 4 scale where you lose an effective plan if you -- when you 5 start cutting meat off of the bone of this plan, how far do 6 you pare it done before you can't work it?- And then why do 7 you include in here that you would supplement it?
8 I mean I am just trying to get some sense of what 9 the point of that is. j 10 MR. MEISNER: Okay. When go into decommissioning, 11 there comes a point in time that all of the events, 12 accidents that you analyze cannot result in an off-site dose i 13 of 1 rem, which is the EPA protective action guidelines for 14 evacuation of the public. So the focus of the emergency
() 15 plan exemptions and reductions have dealt with the division 16 between an alert and a site area emergency which has to do 17 with the off-site activations and the starting to evacuate 18 interface with the state and preplanned evacuations.
19 The two things that lead into that, one, I have 20 just mentioned that if you can't -- if after analyzing the 21 events and accidents, that you can't get to that rad level, 22' then that indicates, well, why do I have to keep planning 23 for something that I can't get there,
\
24 The other thing that factors in, and I believe it 25 is in the -- you know, in the NRC, I know it has been in the i
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
l l
174 l
l l
1 SERs, is that there may be some, maybe -- emergency plans
( 2 were generally written for short-term events, short-term 3 response. So there comes a time that if you have long 4 periods of time to react, that even if you have gotten to a 5 scenario where you want to do something off-site, you have 6 got so much time available to get that to happen that the 7 kinds of preplanning that the regulation required for real 8 fast events doesn't have the same value.
l 9 With regard to on-site emergency planning, it l
10 stays basically the same. The thing that drives some of the l ,
! 11 changes in on-site planning is the fact that, again, 12 scenarios are less complex, so you may not need as many l 13 engineering people to figure out how to get water in a pool 14 as you would on how am I going to recover from an ATWS in an (h 15 operating plant. So they are relative to the complexity of 16 things you would expect, but you still have rad protection 17 support, you still have engineering support, ops.
18 The point on the slide as far as personnel 19 resources supplement, one of the points in the PRA, the l 20 staff's PRA had to with, okay, you have skeleton staff on l
21 back shifts, and that is absolutely true. You know, you 22 ' don't --
r L 23 MR. GUNTER: Off-site or --
l 24 MR. MEISNER: On-site. On-site staff. There is 25 not much working going on, so the operations staff, there ANN RILEY & ASSOCIATES, LTD.
ON Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
175 1 won't necessarily be any maintenance staff. You have a
() 2 3
skeleton staff. So if something happens during that timeframe, in order to ensure that -- you know, it is not 4 the expectation that now you sit around until the day shift 5 comes in. You go ahead and supplement the numbers of people 6 that you have. You call people. You may get help via the 7 phone, you may bring them in. And e-plan is one of those 8 things that if you ever got to that threshold, you 9 automatically call people in to assist.
10 MR. CAMERON: Rich, were you going to say 11 something relative to Paul's or something else?
12 Paul, do you have anything else to add or ask 13 George?
14 MR. GUNTER: I think that gives me some insight 15 into --
16 MR. CAMERON: Okay. All right. A clarifying 17 question from Rich.
18 MR. BARRETT: Actually, I would like to ask a few 19 questions here about how -- where do we go from here?
20 Because I find this presentation extremely interesting and 21 very, very useful. Very valuable insights in here, and not l
22 only from these slides but also from the comments of people 23 from other sites who have actual experience on how j 24 decommissioning can be implemented in a safe manner. And so 25 I have really three questions. And I say that so that I can O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014
. Washington, D.C. 20036 l
(202) 842-0034
1 176 1 keep my little tower up here until I am done.
2 First of all, I think it would be very useful for 3' us to understand, in what manner do you plan to document 4 this presentation? You know, we have these slides and the 5 slides are very useful. I think the comments that were made 6 by all of you gentlemen, you know, especially insofar as it 7 relates to your direct experience and to what actually, a 8 little more of the detail behind some of these things. I 9 think that that would be very useful for us to understand, 10 and not so much to understand how current plants are 11 operated, but to understand what it is that is important as 12 we begin to look at future plants, future decommissioning 13 plants that are going to go through this window that we are 14 concerned about.
) 15 So my question is, to what extent do you plan to 16 document this in a more complete way r -e can have it and 17 refer to it as we go forward with trying to resolve this l 18 issue?
19 MR. MEISNER: I guess I will respond to that with 20 a question. To what extent do you intend to credit it?
l 21 And, you know, documentation, you have got the overheads 22 here, you have got the transcript. We don't know what the 23 staff's thinking is on this, and I would hope that maybe 24 they will take up the rest of the afternoon, then we can 25 talk about what will be documented and where.
[' ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
177 1 We are not going to put anything on anybody's
() 2 3
docket that is not going to go anywhere as far as getting credit and risk reduction. That really wouldn't be 4 worthwhile.
5 MR. BARRETT: Yeah. I would certainly agree with 6 that. I think that, really, the whole purpose for being 7 here is to understand those operational and design 8 characteristics that are important to impacting what our 9 perception of risk is, and that was the whole spirit in 10 which I asked the question, because that is where we want to 11 go with this thing. And so I would certainly hope that this 12 would -- that this kind of information, this real life 13 information about the way plants are designed and operated 14 today, and how we would expect them to be designed and
() 15 operated in the future, would be the crucial information for 16 driving future NRC decisions about decommissioning.
17 MR. CAMERON: And when you talk about industry 18 will commit to actions, I take it that that would be generic 19 actions across the board for the industry generally, rather 20 than any plant-by-plant?
21 MR. MEISNER: That's right.
22 MR. CAMERON: All right.
23 MR. BARRETT: Let me ask the second question then, 24 and I understand that it might be very useful to wait until 25 the end of the day or the end of the workshop to make O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
178 1 decisions about what gets documented and how. But the other
() 2 question I have, and, again, I don't think it requires an 3 immediate answer, but we heard two very interesting 4 presentations today from the industry. One is a risk 5 analysis and the other is this very useful information, and 6 both analyses, both presentations have been very useful and 7 interesting to us. It would be very interesting for us to 8 understand the relationship between this information and the 9 risk analysis that you have presented. ,
i 10 One of the bottom -- actually, the very last ]
11 conclusion that you have drawn here is that you are willing i
12 to commit to some operational and design characteristics to 13 the extent that they can be credited in the staff's 14 probabilistic analysis.
e' I
()s 15 So I think it would be useful for us, as we go 16 forward and try to resolve this issue, to understand the 17 relationship of these very real factors that you have 18 brought to the table with this presentation. The 19 relationship of these to the risk analysis as you have 20 presented it, and to the risk analysis as it might be 21 impacted further by things which have not yet been credited 22 in your risk analysis, I think that would be, again, very 23 useful for us to understand the relationship between the 24 risk analysis you presented and the operational assessment 25 that you have presented.
( ANN RILEY & ASSOCIATES, LTD.
A Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 642-0034
179 l 1 MR. MEISNER: In the discussion this morning, you 2 heard a rough requantification, using the staff's model, but 3 not having all the data. That rough requantification 4 reflects in general the assumptions of this afternoon.
5 Now, maybe a fruitful thing to do would be to take 6 the staff's report and look at all the decision points, 7 essentially, like out of Table 3.1-2, which has event i 8 probabilities and frequencies used in the spent fuel pool 9 risk analysis, identify those that are associated with the 10 things we have been talking about this afternoon, and talk 11 about a change in approach. 3 12 Now, that may be the simplistic way to go. We 13 also discussed this morning the fact that I am not sure that 14 a simplistic single assumption over a five day period is 15 appropriate, and maybe the staff needs to do some more 16 detailed work on determining what a real human error 17 probabi]ity is. But if you want to take the simple 18 approach, I think we could use -- pick out some of the key 19 things in that table. ;
20 MR. CAMERON: Let me ask a formatting question 21 here. Where does the discussion of the NRC staff concerns 22 fit into trying to get to the point that you are talking 23 about? I know that George was going to go through the 24 concerns. Basically, if you put the -- if you address the 25 concerns, each concern, does that get us to where we want to ANN RILEY & ASSOCIATES, LTD.
d Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
180 1 be?
() 2 3
MR. MEISNER: I think it gets us there depending on what -- how simple you want to be.
4 MR. CAMERON: But at least it is a way to start, 5 maybe the simplest way to start.
6 Okay. Well, why don't we -- you don't you state, 7 go through the -- you know, start with the first concern.
8 And these are in the staff's presentation, okay, from this 9 morning. Say something about that, and then let's have --
10 you know, let's have a discussion of each one and then let's 11 see where we go with that. Is that fair?
12 Rich, do you have more to say? Okay. Go ahead.
13 MR. BARRETT: Yeah. I thin). that is a very good 14 way of proceeding. I am very anxious to hear how you plan !
(h 15 to address the staff's concerns, and I think that would be a 16 very good next step from where we are now.
17 But the third point I wanted to make, and I think 18 this is important for all us to keep in mind as we go 19 forward here, and to some extent I am echoing a comment that 20 I think Paul made a while ago. What we have heard today is 21 real life experience at real plants as they exist today, and 22 the existing regulatory framework and the existing -- the
{
23 way things are as we sit here today.
- 24. I think as we go forward, what we are talking 25 about is a new set of rules, a new set of technical O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 *
(202) 842-0034
183 1 specification requirements, possible exemptions to some of
() 2 the rules that are out there. We are moving into a 3 regulatory environment that we cannot fully anticipate what 4 it will look like as we sit at this table.
5 As we go forward, it is important for us to sort 6 out for ourselves what it is that we are taking credit for 7 here and make sure that the things we are taking credit for 8 here are not those things, among those things which might go 9 away as a result of exemptions, or go away as a result of 10 new rulemakings.
11 So I would just caution everybody to keep that in 12 mind as we go forward. With that, I am very anxious to hear 13 the discussion of the staff's concerns.
14 MR. CAMERON: Okay. George, do you want to lead 15 us off with the first one?
16 MR. ZINKE Yes. The first item had to do with j J
17 operator attentiveness, and as Bill has stated, I mean, 18 obviously, this is a concern in the sense that -- a concern 19 for a utility in that the operator has to be attentive and 20 we have to do whatever things are necessary to make the 21 operator attentive. This morning, in addition to this 22 narrow term " operator attentiveness," we talked about 23 management systems, and, indeed, it is important that a 24 management system at a plant, the management culture is that 25 it promotes this.
ANN RILEY & ASSOCIATES, LTD.
04 Court Reporters l 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034
182 1 Before I turn it over to just general, I would 2 like Bill to talk about some of the kinds of things that we
[V) 3 have to do in just -- in giving concern to operator 4 attentiveness.
5 MR. CAMERON: Okay, this is Bill Ball.
6 MR. BALL: Yeah. Some of the things that we do at 7 our plant is we actually go and assess the operators. We go 8 on rounds with them on a quarterly basis, things like that.
9 It is awful hard to say these are the things that are going 10 to make sure an operator is attentive. You are talking such 11 a subject term, it is very difficult for me to say that that 12 is going to guarantee his attentiveness.
13 We make sure he has things to do. I mean he has 14 logs that he has to go out and take and take specific p
i) s 15 readings that are centered around the safety of the spent 16 fuel pool. I don't know if there is much more than I can 17 say to the subject.
18 MR. CAMERON: Mike Meisner.
19 MR. MEISNER: I guess I would add, too, I am not 20 sure -- if what the staff is going to is, geez, we need 21 rulemakings on operator attentiveness or something of that 22 order, I suggest that you also need the same thing for 23 operating plants. We all know that these are long-term 24 acting events in decommissioning, and operator 1
25 inattentiveness for a period of minutes or even a few hours
[ '\ ANN RILEY & ASSOCIATES, LTD.
(-) Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
183 1 really has no bearing one way or another on the progression
() 2 of the event. Yet, operator inattentiveness on the part of 3 an operating reactor, over a few seconds, may make a big 4 difference.
5 I guess I can gut-wise get a feel for what the 6 staff concern might be, but I would say that you need to 7 rely upon the general regulatory framework that requires 8 staffing, training, a course of study, and then the inherent 9 management structure that makes those things happen without 10 being subject to NRC violations.
11 I am not sure where else you can go with this 12 particular concern.
13 MR. CAMERON: We have a number of people who want 14 to comment here, but I was just reminded of one possible
() 15 disconnect here that we may be able to do something about.
16 We just went through a whole list of industry responses to 17 various events. Okay. Now, we are into NRC staff concerns, 18 operator attentiveness. Now, it would be nice, in theory, 19 if you could correlate the operator attentiveness concern 20 with some of the things that were brought up in the industry 21 presentation, and I know that we haven't done that 22 cross-walk, and it may be difficult to do, but I think that 23 if we can talk about addressing the staff concerns, some of 24 that may be through what you have in your presentation, 25 George. Some of it may be through other things, or maybe we ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
184 1 don't need to do that at all. But I would just state that.
2 And, Greg, I want to get Greg on here because he 3 is part of the panel. Why don't you say what you were going 4 to say?
5 MR. WITHROW: Just a little comment here, and I 6 think David asked the question earlier, and I am not sure we 7 are going down the right path here, is that what timeframe 8 are we really dealing with for this particular issue? And I 9 think we are dealing with two to three years, right, after 10 shutdown, basically. And what David and the other guys are 11 asking here is, well, we want to understand the full picture 12 of decommissioning. Well, for this particular issue we are 13 talking about, we are only talking about the first three ;
14 year timeframe, let's say, for argument purposes.
() 15 If you look at it in that light, I think it 16 operator attentiveness is very similar to what we have today 17 with operators that are trained and qualified to operate the 18 plant. Those people mostly likely will be there for the 19 first two to three years. We will have change-over after 20 that time period, I am pretty sure of that, you know. But 21 in the time period of concern for this particular issue, I 22 think operator attentiveness is very similar to what it 23 would be for an operating plant.
24 Secondly, you know, to correlate the information 25 that we gave you to your concern about how do you know they O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0Ce4
185 1 are going to be out doing stuff, well, the bottom line is we fh O
2 have alarms in our control room. We have beepers at -- this 3 is specifically to Big Rock, we have beepers on their belts 4 that tell them if they have an alarm. We require them to go 5 on rounds twice a shift, and usually they alternate those 6 rounds, so that we have two people on shift, each one takes 7 an alternate shot at it.
8 We log our readings hourly, okay. And then beyond 9 that, we have security people that are making rounds around 10 the plant. And even beyond that, most of the plants, and 11 this is not across the board, but most of the plants have 12 moved their monitoring stations, i.e., control rooms don't 13 have a control room anymore, they have a monitoring station, 14 down into the security area. And the reason we did that
) 15 was, (1), to get it out of where we wanted to take stuff 16 apart, but (2), so that the operators would have other 17 people to communicate with to keep their attentiveness up.
18 You kncw, that was very purposely thought out of why we did 1
19 we move those people there, that was one of the reasons. l 20 So I think, just following up with David's l
21 question, I mean we need to look at what timeframe we are i 22 really doing this analysis for. It is not for five, ten 23 years out when you would really be worried about turnover of 24 personnel.
25 MR. CAMERON: Okay. Before we go to NRC, let me
[~\ ANN RILEY & ASSOCIATES, LTD.
(s- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
1 186 )
1 get David on the record here. And, Greg, thank you for that
() 2' 3
little bit of a cross-walk, and there is also the time sequence that you brought up.
4 David, you may want to use this.
5 MR. LOCHBAUM: I think the way we look at operator 6 attentiveness, it is not hand-eye coordination tests, or 7 anything like that, that is not what we think would be 8 appropriate. But it is how often the operator rounds are 9 done, things like that. I realize Reg. Guide 1.33 requires 10 you to have operator rounds and alarms and things, but it
)
11 doesn't necessarily say you have to do it every shift versus 12 once a week. So I think -- and you could put all that stuff 13 in the decommissioning safety analysis report, but with 14 50.59, you could take it out the next week and return it to 15 once a week.
16 So I think if the NRC were to set certain 17 guidelines that said, based on having operator rounds once a 18 week, or whatever else the nice things they wanted was, and 19 in the decommissioning plants, if they choose to meet those 20 guidelines or that criteria, then they would get a certain 21 operator response success rate or something like that. And 22 without that, absent those kind of checks, it would be less 23 of a success rate because you don't have the nice bells and 24 whistles you have with those criteria.
25 So having that kind of criteria and that kind of O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
187 1 industry response would be one way to approach operator j
() 2 3
attentiveness and probably many'of the others on that list as well. But just having Reg. Guide 1.33 doesn't 4 necessarily assure that rounds are done once a shift, or 5 once a day for that matter.
6 MR. CAMERON: I Vonna, why don't you address that !
'7 and whatever else you wanted to.
8 MS. ORDAZ: Okay. I just had a few questions.
)
9 Reg. Guide 1.33, I know that is included in your tech specs 10 for operating reactors. Is that translated into your defuel 11 tech specs generically? You are shaking your head, is that 12 for everybody?
13 MR. HASELTINE: It is for Maine and Connecticut.
l 14 I am not sure about Big Rock.
f 15 MS. ORDAZ: So it is not fully applied generically 16 to all plants, two out of the four.
-17 MR. CURRY: Well, Oyster Creek hasn't developed l l
18 its tech specs. We are closely watching the BWR Owners ;
19 Group work in that area.
20 MS. ORDAZ: Okay.
I 21 MR. CURRY: So that is why -- )
22 MR. HASELTINE: The standardized tech spec draft 23 that the staff put out for PWRs does have it in there.
24 MR. WITHROW: Reg. Guide 1.33 is committed to in 25 our QA plan.
I O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
188 1 MR. HASELTINE: Okay, So.they have equipment 2 another way.
3 MR. CAMERON: Let's go to this gentleman here.
4 MR. FORD: Brian Ford, I am with Millstone. I 5 just happen to be working the BWR generic tech specs. And, 6 yes, it stayed in there because that was one of the things 7 .that didn't change, go away when you changed your plant t 8 state. So I have a high degree of confidence that what the 9 industry is going to propose is generic tech specs for 10 decommissioned plants, still has it because I put it in 11 there.
12 MS. ORDAZ: Okay.
13 MR. FORD: And the committee has already voted on 14 it.
(/ 15 MS. ORDAZ: Okay. So that might be another 16 outcome of the commitments.
17 MR. CAMERON: Okay. Thanks, Brian.
18 Go ahead, Vonna.
19 MS. ORDAZ: Okay. Yeah, that is what I was 20 wondering because, you know, that is what gets you into all 21 these procedures that we are looking for -- to see if you ;
22 have for level and temperature, et cetera.
23 The other question I had was on management 24 oversight of operators. We are talking about operator 25- attentiveness as one of our concerns, and it gets back to 1
ANN RILEY & ASSOCIATES, LTD.
A Court Reporters j 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
i 189 1 another question that is in your tech specs, I guess in the
() 2 3
admin. section, and this is a generic question -- is that consistent throughout the level of management that you have 4 on shift? And security, like you mentioned security does 5 rounds, and all the other individuals, is that -- is there a 6 minimum set of individuals that is always there and is that 7 consistent throughout?
8 MR. ZINKE: Let me answer that. From a generic 9 standpoint the people on shift, operations, is specified in 10 tech specs, so that it will have a minimum amount in the 11 tech specs for each facility. For security, the people, it 12 is going to be in the security plan, and so it will be 13 specific. Now, that may vary from plant to plant, I don't 14 know, because, as we talked this morning, sometimes that
( 15 senurity stuff being safeguarded, you don't know what 16 estrybody else is doing, but that is where it is specified.
17 With regard to oversight, of what kind of 18 oversight do I have of people doing activities, that wil]
19 tend to generically not be specified in the same manner.
20 One reason is, you know, I know even in the plan it talks 21 about, you know, you have -- we have an Appendix B QA 22 program. However, the regulations that are in such, the 23 Appendix B QA program, deals with safety-related items, and 24 the oversight that would be related to a fuel pool, even 25 radiological issues, tends to be a commitment by the utility O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
190 1 to augmented QA, which may differ from plant to plant.
() 2 3
But, generally, there will be some commitment about oversight, but it may be a little bit different.
4- MS. ORDAZ: Okay. Thank you.
5 MR. CAMERON: Can I ask one question relative to 6 what David suggested? His suggestion, I believe, was that l
? if the NRC set guidelines, however those guidelines would be 8 institutionalized, that this would give some assurance. Is 9 this any different than the industry making various 10 commitments that we were talking about in the NRC giving 11 appropriate credit for those commitments?
12 I am trying to figure out whether that would 13 accomplish the concern that you have, David.
14 MR. LOCHBAUM: Does it make a difference to me, or
( 15 does it -- you are talking about --
16 MR. CAMERON: Well, no, I am just wondering, is it 17 -- you know, are we talking about something that is 18 completely different in nature, or would the industry 19 commitments to this satisfy what you were talking about in 20 terms of guidelines?
21 MR. LOCHBAUM: It probably would, but from a 22 perception standpoint, right now, there is a general public 23 perception that the NRC is not a regulator, it is basically 24 doing whatever the industry wants, after the Senate 25 hearings. So if the NRC were to establish criteria that the i ANN RILEY & ASSOCIATES, LTD.
%- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 1
l
i-
{
191 1 industry met, that looks a lot better to the public than the
() 2 3
NRC just doing whatever the industry wants, even though it may be the same thing.
4 MR. CAMERON: Okay. But I wondered whether there 5 would be a difference there. Mike.
6 MR. MEISNER: Yeah. Chip, I would guess it is 7 more a matter of optics like David, than anything else, and 8 maybe a cart before the horse. You know, we set up this 9 workahop so that the industry could give the NRC a basis to 10 change their assumptions. In fact, what I think will really 11 happen after that is done is the staff will put together a 12 rulemaking package that rests on certain presumptions or 13 criteria, that if a licensee wants to take advantage of the 14 new rule, it is going to have to meet.
r
( ,h) 15 So, you know, I think either way, it is both 50.59 16 and maybe some other sort of unchangeable commitments if we 17 want to take advantage of new regulations. I 18 MR. CAMERON: Okay. l l
19 MR. LOCHBAUM: So I don't think it matters either 20 way.
t I
21 MR. CAMERON: But we need to have this starting l 22 point before we can figure out what the exact vehicle is to 23 put the requirements on the industry.
24 All right. Operator attentiveness was the first 25 concern. Mike.
I g ANN RILEY & ASSOCIATES, LTD.
i \ _
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l j
192 1 MR. MEISNER: Yeah, I think what we haven't heard 2 yet is from the staff. I am not sure we understand exactly 3 what they mean by operator attentiveness, nor is it clear --
4 I can't trace a path back to their analysis as to what they 5 means, like I can, for instance, with a procedure for 6 restoring water.
7 MR. CAMERON: Okay. I think that is a fair 8 question. Do we have Glenn or Rich? All right, this is 9 Glenn Kelly.
10 MR. KELLY: I would change that, I would rewrite 11 that to be organizational attentiveness, but that the 12 concern was that we are worried about the very type things 13 that also sound good, about the systems being simple and the 14 operators having only this to do. That works very nicely 15 one way in the sense that the operator is not overburdened 16 with things. At the same time it can become exceedingly 17 dull and boring and basically kind of put you to sleep about 18 what is going on if you are just doing this day after day 19 and nothing is happen. You know, the pool level stays the 20 same, you know, yeah, the pool is still there. Okay, I can 21 go back and check it eight hours from now. The pool is 22 still there. Okay. I am okay there.
23 As an organization, I am not being challenged like 24 I am on an operating reactor. So we are worried here that 25 is really easy, we see, for an organization to get a little L
I h
ANN RILEY & ASSOCIATES, LTD.
Court Reporters i 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 l (202) 842-0034 1
j
l i
193 1 soft if they are not used to dealing with these hard 2 problems now and'they are -- particularly if the 3 expectations are that nothing bad can really happen with the 4 spent fuel pool, it is just basically a big swimming pool.
5 If I look at the procedures, all I can do is, you know, it 6 takes about maybe I can use 5 gpm pump to take care of any 7 leaks. There is no expectation I am going to get a 150 gpm 8 leak or what I am going to do with something like that.
9 So, generally, this just had to do with the fact 10 that organizations may not be prepared, really prepared for 11 having a significant problem with a spent fuel pool. And in 12 the event that there is an accident, without having -- we 13 saw that the potential was that an organization could end 14 up, as we talked about happening at Three Mile Island, they 15 start going down one path, and they miss seeing what is 16 happening over here. Or they are concentrating on fixing 17 that diesel-driven pump because they just know they are I 18 going to get it done in another day, it is going to get 19 fixed and this whole thing will go away, and so it is okay 20 in the meantime while the pool is heating up, because they 21 just know it is going to -- they are going to get it fixed.
22 And every plant has probably had that. But they just knew 23 they were going to get a piece of equipment fixed, and, oh, 24 we didn't quite.make it, which is why we have NOEX.
25 MR. CAMERON: I guess there is a process question
[ ANN RILEY & ASSOCIATES, LTD.
( Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 l (202) 842-0034 1
194 1 here in the sense of, one, what has the staff heard from the
() 2 3
industry presentation that would make them feel more comfortable about now organizational attentiveness? And the 4 other thing is is that does the industry have a good enough 5 idea from our concern about what they might do, or what they 6 might commit to, to satisfy our concern? Mike.
7 MR. MEISNER: Well, first of all, I think you are 8 mentioning apples and oranges here. You started out talking 9 about lack of attentiveness and being bored and somehow drew 10 the conclusion that that leads to somebody fixating on 11 restoring a pump to the exclusion of anything else.
12 You know, I think you cautioned us earlier against 13 a common sense approach, and it sounds to me like you are 14 dabbling in that yourself. I think the issue of
() 15 attentiveness is not the issue, but rather, is there a 16 procedure in place to direct that the operators do those 17 actions.
18 What I am trying to get to is, where in your 19 model, where in your analysis,-do you credit or discredit 20 operator attentiveness, as opposed to all those other things 21 up there? In other words, what can we do? What is it you 22 are looking for? What commitment of substance would there 23 be that would make a difference in your conclusions?
24 MR. CAMERON: I think that is a fair question, and 25 before we go to Paul, I would like Rich -- Rich Barrett to O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
195 1 answer that.
() 2 MR. BARRETT: I would like to answer two questions 3 here. First of all, address Chip's question regarding what 4 is it that we might have heard in the most recent 5 presentation from the industry that would lead us to feel 6 more comfortable, let's say, about organizational 7 attentiveness and the sense of alert and readiness, the 8 sense of urgency, the sense that this is really an issue 9 that has to be dealt with day after day, shift after shift.
10 And I think that taking into account the 11 perspective that this is a situation that will exist for 12 three years, and some of the thermal-hydraulics indicate l l^
13 that it might be less than three years, that you have an 14 organization that is s'oming into this from the experience of
() 15 operating a nuclear 51 ant, which is to say an organization 16 that is a highly tuned organization, that is a situation in 17 which, by and large, the operational staff are, in fact, the l
l 18 same people who were operating the reactor up until that l
19 point.
20 I think there are a number of factors such as l 21 those that would lead us to believe that the organizational l l 22 attentiveness would be high, in spite of the tedium that 23 would certainly -- you would certainly expect to set in at 24 some point. I think also in the first couple of the three !
25 years there is probably, I would imagine, more than the !
ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
196 1 average amount of activity, ancillary activity at the plant, 2 associated with the decommissioning process itself,
{
3 associated with the movement of fuel perhaps. Then there 4 might be, in'the longer-term, after you have passed the 5 window during which a zirconium fire might be a concern.
6 So, to answer your question, Chip, I think we 7 heard a lot in this presentation, and I certainly -- I don't 8 know that I have been exhaustive in coming up with examples 9 of what those things were, but I think when we go back and 10 we think about this, we will find a lot in there that would 11 give use a sense, a better sense of about organizational r 12 attentiveness.
13 With regard to where in the model this is modeled, 14 I think it would probably be asking too much of this model
() 15 to say that we have -- you know, we have gone in at that 16 level of detail, and specifically have gone into performance 17 shaping -- human performance shaping factors. Maybe I 18 should let Glenn speak to that. But this is -- I don't 19 believe it is that sophisticated a model.
20 MR. KELLY: Well, Mike and I were just talking 21 about that and rather than my paraphrasing what Mike told 22 me, I will let Mike go ahead and speak directly.
23 MR, CHEOK: In the model, how we modeled this was 24 in the HEPs pretty much, and I think Ed alluded to this 25 earlier this morning in the dependencies between HEPs. When O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
197 1 someone fails to do something, an earlier action, how does q 2 this affect his later actions? Are all these actions 3 independent? I think that is how, you know, we modeled the 4 organizational attentiveness. How does the organization 5 respond to -- sort of generically.
6 MR. CAMERON: Has there been -- well, go ahead, 7 Stu.
8 MR. RICHARDS: I think in order to keep the 9 dialogue moving -- thank you.
10 SPEAKER: Go ahead, Stu.
11 MR. RICHARDS: I think he asked a fair question.
12 You know, what we are supposed to be doing here is talking 13 about mitigating actions. We haven't gotten past number one 14 yet, I think he asked -- and I am thinking, too, the PRA
() 15 was supposed to highlight areas of concern from a risk 16 perspective, and I think the purpose of the workshop was to 17 talk about, well, what can you do to address those risk 18 concerns? So we are kind of talking around the issue here.
19 The question is, what can we do to make things 20 better in operator attentiveness? ad the industry has kind 21 of -- like, hey, I don't know what you are talking about.
22 And what was the question to us. Do we have any idea what 23 we want them to do to address that? If not, if it is kind 24 of a philosophical thing, why don't we move on to number 25 two. .
l l
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW- Suite 1014 l Washington, D.C. 20036 (202) 842-0034
198 1 MR. MEISNER: Thank you.
() 2 3
MR. CAMERON: And I guess that one suggestion here is that the industry came up with a number of suggestions of 4 things that they might to commit to, and their suggestions 5 are organized in regard to certain events. Will the staff 6 take a look at that to see what credit should be given in 7 the PRA, including -- and that would include taking those 8 suggestions and running them against these concerns. The 9 industry could be required to do that, but I think that, as 10 we are finding out, with some of these concerns, the 11 industry is not exactly sure what the nature of the concern 12 is. Is that correct?
13 I am just looking for -- I know we are going to 14 have a discussion of a bunch of other things, but if we want
() 15 to come out of this afternoon with something that the staff 16 is going to do, to commit to move this ball ahead, we need 17 to be thinking about that.
18 Rich, go ahead.
19 MR. BARRETT: If the question is, does the NRC 20 intend to take all of this information into account in order 21 to change our perception of the risk associated with 22 zirconium fire?, the answer is absolutely. That is why we 23 are here.
24 If the question is, does the staff intend to go 25 back and requantify?, I am not so sure about the answer to O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 i Washington, D.C. 20036 !
(202) 842-0034
199 1 that. I am not sure that that is necessary, a necessary
() 2 3
step toward resolving this issue.
possible either.
I am not sure that it is 4 What we are doing here -- where we are here is in 5 a region -- a regime where probabilistic safety analysis is 6 not normally done. We are not generally dealing with 7 accidents which are totally dominated by human error 8 probabilities. We are not normally in a situation where we 9 are talking about 10 to the minus 4 human error 10 probabilities or five dayn to take human actions. So what 11 we have tried to do here ia to use probabilistic risk 12 assessment, to the extent that we could, to highlight those 13 areas, those types of sequences which we have a concern 14 with, and to challenge the industry to come in and tell us,
( 15 deterministically, what features can best address those 16 sequences, so that we can come to a bottom line, on a 17 qualitative basis at least, that risk has been addressed, 18 the. risk of these sequences has been addressed. l 11 9 MR. CAMERON: Are you suggesting that with the 20 consideration of what the industry has put forward, and a I I
21 consideration of what we have heard from other people, that i i
22 when we take that back, we may put the probabilistic -- the l 23 probabilistic study may have served its purpose and may not 24 .be the driver of this? For example, it could be we think 25 these industry commitments, deterministic information, will l
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NN, Suite 1014 Washington, D.C. 20036 (202) 842-0034
200 1 satisfy us?
() 2 3
MR. BARRETT:
MR. CAMERON:
That is correct.
Okay. You know, we are getting 4 close to a break time again, but I want to hear Paul and Ray 5 and Mike, and I want to hear from the audience. So let's do 6 that, and then let's come back. I think it should take us 7 an extraordinarily long time. Paul.
8 MR. GUNTER: Thank you. Maybe this is an 9 oversimplification, but it seems to me that the issue --
10 with-this particular issue of operator attentiveness, it is !
11 something that could be dealt with by a level of enforcement 12 commensurate to the risk that would keep the utility.
13 attentive. And I am sure there are people in here from the 14 Navy, but what is the penalty when you fail the requirement
( 11 5 of your watch? I mean does anybody -- can somebody tell me?
16 I didn't serve in the Navy, but what, you know, what is the 17 penalty? And -- it may be.
18 But the issue is, I mean, you know, you assess the i
19 situation that this watch has been assigned, commensurate 20 with risk, and then you assign an appropriate enforcement 21 level to deal with. Now, I think that is how you get a !
l 22 utility's attention. And, frankly, I think that this l
23 exchange back and forth really needs to -- we need to 24 establish, you know, who is running the ship here.
25 MR. CAMERON: All right. Ray.
l O
\-s/
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
201 1 MR. SHADIS: Yeah, I just want to take a minute 2 to confess that you all have me thoroughly confused.
)
3 Individually, I seem to be able to understand each one of 4 you, but then when I try to put the conversation together, 5 it fails. I am very pleased to hear about the different 6 programs that these different power plants have in place to 7 . deal with their spent fuel. And it is always nice to see 8 the industry doing more than is required by regulation. But 9 I am confused about the issue of the industry making 10 commitments to this or that. I don't know how it is 11 relevant -- honest to God, I don't -- to regulation.
12 I wonder if before there were speed limits, 13 anybody broke the speed limit. I mean if there wasn't some 14 idiot out there going faster than everybody else. And you
() 15 are in the business of establishing regulation. This is a 16 fairly serious matter. I am not sure that the question 17 really was answered as to what the NRC means by operator 18 attentiveness, or how, if they can define that, how they 19 intend to get there so they have operator attentiveness. I 20 really don't think that was answered yet.
21 But I would like to hear that at least, and then, 22 you know, I can tell you, representing my own constituency 23 out there in the public, that we don't want to depend 24 entirely on the professionalism and the good will of people 25 working in the industry. We want to know that there is a ANN RILEY & ASSOCIATES, LTD.
O. Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036-(202) 842-0034
! l l
i 202 1 body of regulation there that requires our protection.
[ ) 2 And I think before you even get to that, and this i 3 is -- I am sorry, but you have been doing this for a while 1
4 and it has got me really confused, before you even get to 5 that, you need to tell me if it is 100 days that you are l 6 talking about or 700 days that is a window that is going to 7 require this operator attentiveness in order to a oid an 8 off-site impact that will, you know, in turn, require 9 evacuation. Where are we with it? I think you need to get 10 that physical portion settled first and then we need to talk 11 about how you avoid those accidents.
12 MR. CAMERON: At some point before we go to Mike, 13 I think that, you know, we do need to have a clarification 14 on the time element here, and I understand your confusion
() 15 about where we are going with this. And I guess that where l 16 I thought we were going was that the industry actions were 17 going to form, eventually form the basis for a regulatory 18 regime. And let me go to Stu to talk about that.
19 MR. RICHARDS: I agree with you to a degree, Ray, 20 that there is a little confusion here, which is probably a 21 good reason to take a break.
1 1 22 But beyond that, I heard what, you know, what our l
l 23 staff said as far as operator attentiveness. This is the 1
l 24 idea you go from an operating plant with a lot going on, 25 and, clearly, people are generally, you know, have good ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
203 1 reason to pay attention because they are a lot of activities 2 to this passive system, you know, a large pool of water 3 with, you know, a reactor core sitting in it, which you 4 could argue lends itself to tedium, I think was the word 5 that was used.
6 I heard the industry respond to that, saying, 7 well, we require our people to go out and make rounds. We 8 require them to take logs. We train them, we give them 9 periodic tests and retraining. So they ditt bring to the 10 table a couple of things here to say that, you know, we 11 don't just hire people and put them in the control room, we 12 train them, we require them to do activities. We relocate 13 them to ensure that they have other people to interact with.
14 I think where the disconnect is coming is the 15 industry has asked, how are our comments affecting the way 16 we view our draft PRA results? And you know, we need to 17 make that connection. I think maybe this particular topic 18 doesn't lend itself well to that, as some of the others 19 will, such as having procedures in place.
20 But I think your idea of taking a break to talk 21 internally about how we proceed is a good one.
22 MR. CAMERON: And I think you are right about what 23 you just said.
24 Mike.
25 MR. MEISNER: Okay. I am going to take a couple h
%/
ANN RILEY & ASSOCIATES, LTD.
Court Reporters l
I 1025. Connecticut Avenue, NW , Suite 1014 84 b3
204 1 of minutes here. We have got several very fundamental
()
N_e 2 issues. First of all, this notion of operator 3 attentiveness, what I was trying to get down to was, Mike 4 said very clearly he modeled this in his analysis. What we 5 need to know is, what he modeled and what would make a 6 difference in the model? If we are not going to get down to 7 those answers, we are not going to get anywhere.
8 Secondly, the bigger picture, on this notion of 9 requantification, and I am going to answer Ray's question in 10 the process as to why we are talking about this commitment 11 process and all. What we have here, from the viewpoint of 12 the industry, is a self-created issue on the part of the 13 staff. They have put together a model that ensures that 14 human error is the sole and only focus of the model, and in O
t s ,) 15 our point of view they have done it incorrectly and it is 16 fatally flawed.
17 The reason we are getting into commitments is to 18 try to break an impasse here. It is not something we 19 normally do. We would much prefer that the objective 20 technical approaches prevail here, but we don't think that 21 is going to be the case. So our band-aid cpproach is to 22 identify areas where the industry can provide assurance and 23 commitment to cause the staff to change their model. If the 24 staff is unwilling to change their nadel and, clearly, they 25 are by unwillingness to requantify, then we had better take
('
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Waw ington, D.C. 20036 (202) 842-0034
205 1 a significant break and everybody go back to the drawing (v ,; 2 boards.
3 The model has to be requalified if we are going to 4 have a risk-informed approach to decommissioning. That is 5 what the Commission asked for. I heard that clearly on 6 March 17th. You can go -- and I have read the SECY that the 7 staff just issued, too. Clearly, the staff is intending to l 8 develop a rulemaking in the context of the backfit rule.
9 You are never going to be able to demonstrate compliance 10 with the backfit rule unless you have got some level of 1
11 quantification of what the level of risk is and in order to )
l 12 turn that into some cost benefit. If you do it some other 13 way, I think you are not playing by the rules.
14 So our expectations is that we are going to r
( 15 provide you concrete commitments if you will tell us what it 1 16 is in the model that makes a different. And having given 17 you those commitments, it is our expectation, and I thought 18 we went into this workshop with some level of agreement, 19 that the staff would take that and factor that into their 20 model and credit it. And it as simple as that.
21 So I would hope that we are going to get some 22 answers specifically on attentiveness, what it is in the 23 model that makes a difference, and what in the staff's view 24 would make a difference, and is a straight commitment, 25 because, for the life of me, I haven't got the vaguest idea I
/"'i ANN RILEY & ASSOCIATES, LTD.
(,,/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034 i
l
206 1 what it is. J
() 2 3
MR. CAMERON: Okay. We will address that as I think a threshold question when we come back, at least 4 starting off with generally what the staff is going to do 5 with the information. And I think that we can put a much 6 finer point on that.
7 Let's just go for anybody out here who wants to l 8 say anything, and then let's take a break.
9 MR. CANAVAN: Yeah, I will also keep it quick.
10 Ken Canavan, GPU Nuclear. I thought this was a real 1
11 appropriate time to say that I thought that this was an 12 historic day. Making history is kind of traumatic 13 sometimes, but it is historic in that the NRC is here with 14 all the stakeholders. They are here with industry, they are
() 15 here with the public, to make new rules, to optimize safety 16 and resources. The question is -- how? So I know it is 17 late in the day and everybody gets a little testy, so I 18 thought I would throw that out. We are really just working 19 at "how," and it is a common goal.
20 But to bring us all back just one step, I wanted 21 to talk a little bit about the operator actions and the PRA 22 study. I am a PRA guy and I do take issue with some of the 23 human error probabilities that are in the current generic 24 study. And the reason why I do is because I think some of j 25 the HEPs and some of the things that shape a human error O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
207 1 probability weren't really considered.
( 2 We talked earlier today about loss of spent fuel N
l 3 pool cooling going on for five days. Well, this is a big 4 Olympic size swimming pool that evaporates into the building 5 over these five days, condenses inside the building and, 6 basically, it starts raining in the building. All that 7 water goes into the sumps and drains and goes down and some 1
8 pumps start, along with all these annunciators go on, and 15 )
I 9 shift crews come in and out and don't notice it is raining j 10 in the building.
11 So as far as an HEP goes, we can basically say 12 that that is probably a really low value. Now, I don't know 13 what value you stick on it, and I don't know what method you 14 use to get there, but the current value in there is not what (3
(_) 15 I would consider a really low value, and it is just to a 16 point that some of the HEPs or some of the performance 17 shaping factors may not have been considered like it is 18 raining in the building.
19 And that was the contrast, what had been going on 20 earlier in the day. That is how the action is different 21 from all the other actions that are typically in a Level 1 22 online PRA. And earlier everybody was contrasting 23 similarities -- pointing out the similarities between other 24 actions and how things can be missed. I am pointing out how l 25 it is a little bit harder to miss this particular one.
('
\/
) ANN RILEY & ASSOCIATES, LTD. j Court Reporters l 1025 Connecticut Avenue, NW, Suite 1014 I Washington, D.C. 20036 (202) 842-0034 l
)
208 1 And there are some other issues, and the problem
() 2 3
with conservatisms in the PRA, you might draw the wrong conclusion, you might draw number 1, you might draw that one 4 of the biggest concerns is operator attentiveness, where 5 maybe that isn't the concern, because you were way 6 :onservative on one part of the model. You need -- what I 7 Chink we need to do as a group is get a little less 8 conservative on some of the p&rts of the model, be realistic 9 through the whole thing, so we have a really good pointer as 10 to what the issues are. And then more important issues get 11 more important rulemaking and more important enforcement, as 12 Paul pointed out earlier. Less important issues, less 13 important items, less attention. And this allows us to 14 focus our resources and focus on safety.
g,) 15 And the last thing -- the last thing I had written 16 down was that there were some insights in the generic study, !
17 specifically, numbers 3 and 4, that didn't seem appropriate 18 to draw from what was in the study. I read the whole thing 19 and couldn't see how you got 3 and 4 out of it. All of a 20 sudden we jump to having a release and evacuating people.
21 And it was unclear as to how numbers 3 and 4 were actually 22 drawn out of the study, and I guess that is more or less a 23 question.
1 24 MR. BARRETT: Could you remind everybody of what 3 25 and 4 were?
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 '
(202) 842-0034
209 1 MR. CANAVAN: Right here. Three and 4 are --
() 2 3
basically, there is a point in here that it says, it appears that there would only be five to seven hours available for 4 local emergency response. This may be too short for an 5 effective evacuation to preclude some early fatalities. And 6 I was unsure, had anybody done any release work? Had 7 anybody looked at release categories? And had anybody 8 looked at the dispersion of the fission products outside the 9 buildings? Because it really depends on the building and 10 the configuration on how fission products get out.
11' And I guess the last one was number.4, was similar 12 to that, that action levels for the e-plan higher than alert 13 might be needed. I was also wondering, had anybody done any 14 fission product transportation, or even any fission product
() 15 release type work?
16 MR. KELLY: This is Glenn Kelly with the staff.
17 The answer is yes.
18 MR. CAMERON: Okay.
19 MR. BARRETT: Unfortunately, in the report we did 20 not include the consequence analysis, and so I could see 21 where that would be -- would appear to be a non sequitur, 22 yeah.
23 MR. CAMERON: Let's take a break and come back at 24 4:00 and see if we can get to some fundamental bottom lines 25 about where we are going. And I think that probably the NRC l l
i J
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 I
l Washington, D.C. 20036 (202) 842-0034
210 1 staff has to -- might want to caucus at this point.
2 All right, 4:00.
)
3 [ Recess.]
4 MR. CAMERON: Okay. If we could get everybody 5 back to the table. All right. Thank you.
6 We have got -- let me just -- we have got -- the 7 staff is going to start off and try to state its intent in 8 the context here, so that we can see if we can all get back 9 on the same page. And I would propose that we try to do 10 that, because that is the fundamental starting place where 11 we need to be to go forward on the details. And I would 12 suggest we will run till -- we will run till 5:00.
13 We will adjourn at 5:00. We are going to try to 14 get this fundamental issue resolved to everyone's
() 15 satisfaction, and then we can begin with details of how we 16 are going to -- of any of the details tomorrow.
17 But I guess, John, do you want to say -- do you 18 want to start before Rich gets started.
19 MR. ZWOLINSKI: Thank you. One of the things we 20 are constantly challenged with from our office director is 1
21 what process are we in and how do we see ourselves going 22 forward, regardless of topic. And one of the things that 23 has evolved over the last few hours is.an apparent little !
24 bit of shaking of which processes are we following and where 25 do we see ourselves going.
ANN RILEY & ASSOCIATES, LTD.
O~ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 i
I-t 211 1 As we have said in the SECY paper going to the
() 2 3
Commission, our technical working group has an important l role to play, but there are other corollary activities that 4- we are sponsoring. The fundamental -- the activity of today 5 and tomorrow was to put much of our shoulder towards the 6 technical-working group's activities that have been l
7 completed to date, and move forward with advancing the ball i 8 in that' arena.
9 I am sensing that there has been some degree of 10 frustration that you all are sharing information and you are 11 not quite sure where it fits, and so on and so forth. I 12 kind of felt it was important to maybe re-zero, going back 13 to where we were this morning, and some of the words that I 14- thought -- some of the presenters were all on the mark as
() 15 far as our hopeful thinking, wishful thinking, that we could 16 collegially at least have a common understanding of what we 17 were attempting to do with the workshop.
18 With that as a little bit of a backdrop, I have 19 asked Rich, as the key technical sponsor of the technical 20 working group's report, and efforts today, to kind of share 21 our vision of how we see all the information that is being 22 shared to us today, being transcribed, being available to us 23 as we move forward, ask Rich to kind of like ensure that 24 everyone understands the context of the information and how 25 we intend to take our next few steps in moving -- advancing O ANN RILEY & ASSOCIATES, LTD.
Court Rn.orters 1025~ Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
212 1 the ball.
l
[Jl' 2 So with that, Rich, would you mind re-zeroing the 3 clock for us, please?
4 MR. BARRETT: Okay. I would like to start by 5 saying that, as often happens, sometimes the most important 6 part of a meeting is the break, and I think what we just had 7 was a very good break. I think we all had an opportunity on 8 the NRC's side to share notes and share perceptions, and get
! 9- an understanding of where we are and where we perceive we 10 are, and what we have been saying and what we have been 11 hearing. And I think it is a very good thing at this point 12 to stop and go back and rephrase some things so that there 13 are no misunderstandings, or, hopefully, no 14 misunderstandings about where we are and where we go from
() 15 here.
16 So let me just review the bidding of where we have l 17 vome from, where we are, and where I think we go from here, 1
18 and what I think we can aueomplish, and what the next steps 19 are.
20 As we said so many times today, the analysis that 21 we put on the table was a preliminary one and it was put on 22 the table for the purpose of initiating a dialogue, 23 stimulating your thinking and giving you a basis for coming 24 in to tell us what you think is important for us to 25 consider. That was the purpose and I think that the O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 l (202) 842-0034
I 213 1 preliminary analysis accomplished that,
()
o 2
3 This afternoon we have heard a lot about the way in which you operate your plants. I think it is going to be 4 very important for us to take a look at your slides and at 5 the transcript to get a better idea of what you do and what 6 you think is important about what you do to ensure safety 7 for a decommissioning plant. And for us, that is the most 8 important of our four focuses, and that is to maintain 9 safety.
l 10 How close are we to resolution here? I have made 11 the mistake on many occasions of making predictions, but I l 12 will do it again. I think that our preliminary view is the 1
13 types of things that you have discussed today, th* types of 14 safeguards, the types of care that you take to enst.ce human
/^s
(_,) 15 reliability, and the types of equipment that are provided at 16 your plants are exactly the types of things, of operational l 17 and design features that we believe address the issues that 18 are raised by these intermediate and short-term -- I'm sorry 19 -- intermediate and long-term sequences. So what we have 20 heard is very promising. l
)
21 What I think we need to do next -- and, by the 22 way, we want to hear more. We want te :'ve you briefly go 23 through and talk about the things as they relate to our 24 concerns. But what I think we need to do next is for the 25 NRC to take this information, go back to the offices and l
\
i h
\- '
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 1
I
1 I
214 1 spend some time, and do a thorough job of bouncing this
() 2 information against our model, and asking ourselves, which 3 of the factors that you have told us about, in combination, 4 have the greatest impact on those factors that are driving 5 our analysis? And to understand what combination of these 6 safeguards can allow us to say that the issues that are 7 raised in our analysis have been properly addressed, and 8 that allow us to say that sequence, this sequence, this l 9 sequence, this sequence, that the risk from those sequences 10 has been addressed to a level that we believe is 11 appropriate.
12 To the extent it is possible, we would go in and 13 show specifically how these factors affect the analysis. To 14 the extent it is possible, we would even go in and try to l
() 15 requantify the sequences. We don't think that is a 16 necessary step. We think that we can reach resolution 17 without doing that, but we understand that is a concern of l 18 yours that we try to requantify it, and we will try to
- 19. requantify it. Keep in mind, however, that we have a 20 concern about pushing PRA beyond where we feel the NRC might 21 lose its credibility and putting down numbers that we can't i
22 stand behind.
l 23 But, in any event, we feel that the information l
24 you have given us, we can go back and we can go into our 25 model and we can show how your information addresses the l .
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 f' i L J
215 1 concerns raised by our model, and addresses the concerns
( 2 related to the sequences we have raised here. And, as I 3 said, hopefully, we can do some requantification as well. '
4 At that point, we will come back to you, we will 5 come back to all of our stakeholders with a proposal on what 6 are the kinds of operational and design safeguards that we 7 believe are prudent and necessary to address the sequences ]
8 we have raised and to bring the risk to an acceptable level.
9 And then that, we would hope would be the basis for a 10 further dialogue.
11 So with that, I hope what I said here serves to 12 address your concerns rather than raise further concerns.
13 But if you would like, I can -- I would like to hear your 14 reaction and I would like to talk about it further.
() 15 MR. CAMERON: Mike.
16 MR. MEISNER: I am not quite sure of my reaction 17 yet. I wanted to ask a few questions. At the end you 18 talked about coming back with those safeguards that would, 19 in your mind, be necessary to bring risk to an acceptable 20 level. What is an acceptable level?
21 MR. BARRETT: Well, that is a good question. I 22 think that what is an acceptable level is going to depend on 23 the decision you are making, and I think that is a question 24 we will have to address. As you know, there is a lot of l
l 25 guidance out there on criteria to be used for regulatory --
ANN RILEY & ASSOCIATES, LTD.
O-Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l Washington, D.C. 20036 (202) 842-0034 :
1
216 1 for risk-informed decision-making.
/ 2 And Regulatory Guide 1.174 is certainly not the l 3 definitive answer on all of these issues, but it is 4 certainly one that is typical. It is a set of guidelines 5 that are typical and that are consistent with other 6 guidelines that exist in the industry for other purposes, 7 and exist in the regulatory process for other purposes.
8 If you are talking about, again, -- and, by the 9 way, I would say Ed brought up some very good points earlier 10 about defining how this type of an accident relates to the 11 kind of accidents that we talk about in Reg. Guide 1.174, 12 because there we talk about core damage accidents, which 13 this is not, and we talk about large early releases, which 14 this is not. In some aspects, this is perhaps something in 15 between the two. And so we don't have any hard and fast 16 criteria.
17 But, as you know, in that document we talk about 18 core damage estimates that are in the 10 to the minus 5 19 range -- or deltas, rather. Changes in core damage in the 20 10 to the minus 5 range are typically of concern. And 21 accidents which lead to large early releases, that lead to 22 changes in that probability, tend to be in the 10 to the 23 minus 6 range.
24 Also, in the regulatory analysis guidelines, there 25 are similar criteria which may or may not trigger -- which I ANN RILEY & ASSOCIATES, LTD. l ks Court Reporters I 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 l (202) 842-0034 l
n l
217 1 tend to trigger whether the staff initiates a backfit 2 analysis.
3 So I think that the question -- the answer to your 4 question depends on what decision you are making. If you 5 are talking, for instance, about a change to the -- an 6 exemption, for instance, you would have to go one step l 7 beyond the analysis we have done already, because what we 8 have analyzed here is the baseline risk.
9 MR. MEISNER: Well, Rich, I am just talking about 10 rulemaking.
11 MR. BARRETT: Well, for rulemaking, if the staff 12 -- the staff has many options to go in, and I can't say that 13 I know exactly which way we are going. But if you are 14 talking about rulemaking that is justified by cost benefit 15 analysis, then there is a threshold of 10 to the minus 6, it 16 is an informal threshold, but it is used, and it is in the 17 regulatory analysis guidelines for a large early release, 18 and 10 to the minus 5 for -- typical of a core damage, delta 19 core damage frequency, and then you would combine that with 20 a cost benefit analysis.
21 If you were talking here about a rule that was 22 being passed as an option for the industry, in other words, 23 the industry could adopt this option, or stay with the 24 current regulations, as modified with the possible 25 exemptions, there might not have to be a regulatory analysis
( ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW , Suite 1014 Washington, D.C. 20036 (202) 942-0034
218 1 at all.
() 2 Similarly, if the staff felt that -- made the case 3 that we were going to impose a new set of decommissioning 4 regulations on the industry and that this was to be an 5 adequate protection issue, there would also not be a risk 6 threshold for making that decision. That would be a 7 decision that would have to be made in absence of a 8 regulatory analysis.
i 9 S,o I don't know if that answers your question.
10 MR. MEISNER: I was asking a real simple question 11 in response to what you said, and that is that you are going 12 to identify those safeguards. It is not our decision, it is 13 yours. We were going to identify those safeguards necessary 14 to bring risk to an acceptable level. I just wanted to know
( 15 what that meant.
16 MR. BARRETT: Right. And I think we are going to 17 have to have some discussions within the staff as what kind 18 of decisions are going to be driven by this.
19 MR .. CAMERON: Mike, you had some other -- I mean 20 you have --
21 MR. MEISNER: I am just starting.
22 MR. CAMERON: All right. Good. I j
23 MR. MEISNER: Well, you also said -- I mean the 24 way I read the recent SECY that came out, the staff was )
25 going to develop this rulemaking in the context of the l l
i ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
219 l- backfit rule. Is that right, John, or Stu?
() 2 3
SPEAKER:
think that was --
I can take it out and quote it, but I 4 MR. RICHARDS: No, I think when it talked about 5 the backfit rule, I think it was talking about the 6- Commission's direction that we apply the backfit rule to 7 decommissioning plants and we are going to roll that in. We 8 haven't gotten into the rulemaking far enough to address, 9 you know, --
- 10. MR. MEISNER: We will find it. Never mind.
11 MR. RICHARDS: Yeah, but we have to live by the 12 rules, Mike. But I think what you heard here from Rich is 13 just an option for going down the line. We either have to 14 comply with the backfit rule, because the Commission has I
() 15 already told us we have to do that, or we write a rule that 16 is an option for you'to follow and you can stick with the 17 way we have been doing business.
18 MR. MEISNER: What I am trying to get to -- now I 19 understand exactly what you are saying. What I am trying to 20 get to is so many of these things revolve around 21 quantification. And you said at one-point that you were j 22 going to requantify, and then later you said we can do some 23 requantification as well. What are we talking about here?
24 And here is why I am concerned. Okay. We started 25 out with a risk-informed approach for decommissioning. And I~ ANN RILEY & ASSOCIATES, LTD.
ksl Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 s.,,.. . . . . . . . . .
_l
220 1 that is something that, by its nature, at least for this
() 2 3
kind of issue, needs to be quantified so you know the order of magnitude of what you are dealing with. If I am going to 4 apply resources, I am going to apply them differently to a 5 10 to the minus 5th sequence than to a 10 to the minus 7th 6 sequence. Otherwise, I have got no basis on which to apply 7 those resources.
8 Similarly, you now have put out a draft report 9 that, if it is never updated or quantified, will be used as 10 a target for every intervenor on every w. commissioning plant 11 down the road, regardless of all the nice words you say 12 about it afterwards. Our point is, it is essential, it is 13 not even negotiable, that that report be finalized, and we 14 would hope it would be finalized using the industry input.
() 15 And that is what the purpose of the workshop was, I think.
16 MR. BARRETT: I agree with you 100 percent, we 17 have to finalize that report. We can't leave that 18 preliminary risk analysis out there.
19 MR. MEISNER: I am not talking about twisting 20 words here. I am talking about -- when I say finalize the 21 report, I am saying requantify the model and rewrite the 22 report. We have told you right from the beginning, we 23 reiterated again this morning, the model you put together is 24 good. We think it captures anything that could conceivably 25 be important to decommissioning, spent fuel pool safety.
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
221 1 The only thing we have a concern about is we think
() 2 3
you have gone way overboard on the numbers and assumptions that you-have put into the model. So no one has argued 4 about the model, but we do think it is very important that 5 that model be requantified and done so in an even-handed, 6 not an upper bound type of approach, which we think it 7 characterizes now.
8 And if we are in agreement with that, I think we 9 can proceed.
10 MR. BARRETT: II think -- what I can say is that 11 we will make -- we are going to take the information that we 12 have, we are going to go back into the model and we are 13 going to address the issues that are in the model. And, as 14 I said, we will make every effort to requantify. There may
() 15 very well be -- I think we need -- let me say this, I 16 understand your concern. I understand your concern that you 17 don't want to have numbers out there that are only qualified 18 by words.
19 And what I want to do is I want to try to -- I 20 want to address that concern. Okay. And whether that means 21 republishing or finalizing the report with requantified 22 numbers, if that is possible, that is a possibility. If it 23 means finalizing the report in such a way that it doesn't !
24 reflect the' preliminary numbers, I don't know, but I will --
25 we will do everything we can to address the concern you have i
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 l
Washington, D.C. 20036 1 (202) 842-0034 l
222 1 about the possible perceptions that are raised per that 2 analysis.
3 MR. MEISNER: Perceptions are only half the 4 concern. We need to have a usable piece of work here that 5 can tell us where to focus our resources. What is important 6 to decommissioning safety? And having -- pulling back the 7 model and relying on a range of staff judgments is not 8 risk-informed in my mind.
9 MR. BARRETT: I do want to take issue with this 10 concern about what is risk-informed and what is not 11 risk-informed. We have used -- we have resolved issues, 12 licensing issues in a risk-informed manner without actually 13 quantifying anything. For instance, we have looked at 14 issues, licensing issues, in which we have had a concern
( 15 about whether a change in a license commitment for a plant 16 might affect the ability of operators to take actions in a 17 severe accident situation, even though severe accident was 18 not part of that analysis.
19 And we went into the analysis and we asked 20 ourselves, which operator actions might be important? And 21 how much time -- how much less time might this operator have 22 to do that operator action? And we came to a qualitative 23 judgment that it didn't make any difference to have 5 24 percent less time to do an action that already had plenty of 25 available. I think that is risk-informed but it is not O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
)
223 1 quantitative.
( ) 2 MR. MEISNER: Yeah. I whole-heartedly agree with 3 you, Rich, there are a lot of situations, and we working 4 with NMSS on several where, you know, the potential risk is 5 in the 5 millirem range, and they want to do a PRA. Why?
6 You have already got the whole insight out of it. You are 7 already risk-informed.
8 The problem is that here is an area where we are 9 not talking about something like that. It is not a narrow 10 issue. You have raised all_ kinds of concerns that, in fact, 11 if you used your worst case, decommissioning plant risk I
12 exceeds every plant CDF that is operating in the country. 1 13 We can't leave that stand.
14 We need to have a scrutable type of process where,
() 15 if the staff says that you need to commit to X because it 16 provides risk reduction, we need to know how much. We are 17 already to the point -- this model was put together in a 18 l Very biased fashion. You overloaded entirely on human 19 error. We don't believe that any objective reviewer would 20 agree with your approach. How do we get around it? By 21 saying, okay, staff, here, you go put together a set of 22 commitments, and we will go along with it and we will throw 23 away the models?
24 We need to understand what level of risk reduction 25 is associated with something. The only way you can do that ANN RILEY & ASSOCIATES, LTD.
O' Court Reporters
.1025 Connecticut Avenue, NW, Suite 1014 j Washington, D.C. 20036 (202) 842-0034 1
j 1
224 1 is to requantify the model. And the only way then to go !
l 2 into rulemaking and validly meet the backfit rule is to have I
[Q')
3 a quantified model.
4 And I need to say something else, too. You noted 5 earlier when you were talking that one of the reasons you ;
6 may not want to requantify, if I got it right, is that you 7 have a concern that you are pushing PRA too far, and that 8 you wouldn't feel comfortable standing behind such low 9 numbers. I think I interpreted that correctly, did I? 1 1
10 MR. BARRETT: I think -- that is close enough. )
11 MR. MEISNER: All right. So my point is then, 12 that the numbers we are talking about are in the 10 to the 13 minus 6th and 10 to the minus 7th range. And what you are 14 saying is that, for any backfit issue, the staff will always O
( 15 default to not doing numbers and we will never reach the 16 conclusion that something is a backfit because you will 17 refuse to quantify down into the range where those numbers 18 apply.
19 MR. BARRETT: No , that is not the case. In 20 general, when we do a PRA for an operating plant, we 21 generally have a mix of operator actions, along with l
22 automatic equipment responses. And so we are generally 4 23 giving a certain amount of credit for the equipment to 24 operate automatically and a certain amount of credit for 25 various types of operator actions.
l ANN RILEY & ASSOCIATES, LTD.
f'%_)\ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
225 1 In this particular case, there really are no 2 automatic actions. And so the analysis is dominated by this 3 question of whether -- you know, whether the organization ;
4 will fail to respond effectively one time out of 10,000, or 5 one time out of 100,000. So that is what I -- that was the 6 concern I have, is that we are out in an area where PRA 7 generally doesn't have to deal. And as you pointed out, we 8 are out at five days, which is another area where PRA 1 9 doesn't have to deal.
10 But, no, absolutely not. In cases where we are l 11 doing backfit analysis for an operating plant, where you ,
1 12 have both automatic and operator activated actions, where l l
13 you have defense-in-depth, where, you know, you have the 14 containment, you have all of the safeguards that have been 15 built in, and, admittedly, you also have accidents that are 16 happening a lot faster, the PRA methodology is more 17 comfortable in that regime.
18 But, Mike, let me say this, we will make every 19 attempt to requantify.
20 MR. CAMERON: Okay. I wanted to let John say 21 something here, but I think that Mike has indicated the need 22 for requantification and taking the information they have.
23 Rich has said that we are going to do that.
]
24 MR. MEISNER: I haven't heard that yet, but --
25 MR. CAMERON: Rich, do you --
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 i
i I
226 1 MR. ZWOLINSKI: It is our intent, to the extent
() 2 practical, to do some requantification. But, fundamentally, 3 we want to take much of the information t. hat we have heard 4 today and apply it to the various sequences that have been 5 developed. And some of these may go away by inspection.
6 And it is not clear to me that it bearc a lot of 7 intellectual curiosity to devote resources to 8 requantification when, by inspection, some of these go away 9 with simple commitments, whereas others in which -- and 10 which we will take a look at, may arguably be worth the 11 time, energy and effort. So to make a uniform statement 12 that we will requantify the entire product, I am not sure 13 that is the best use of our resources, Mike.
14 And if in going forward, we can make use of a lot
() 15 of excellent insights that I felt I gained from the four i
16 plants, folks representing the four plants, things of that 1 17 nature can be adopted across the industry, it seems that 18 goes a long way towards the next step in the process, and
-19 that is the independent review that we have had advertised i
20 from the very beginning. And one of the things we felt was l 21 very important was to, essentially, get outside stakeholders i
22 beyond this room'to take a hard look at the work, in total, l
23 which means the work our staff has done, the work we will do 24 in folding in what you all have shared with today, what that 25 results in, the requantification to the extent we can do ANN RILEY & ASSOCIATES, LTD.
O- Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Wac ington, D.C. 20036 (202) 842-0034
227 1 some.
() 2 And I just don't want to say it is 25 percent, 50 3 --
I am a little nervous to commit that we want to redo 4 everything when several of these may essentially be 5 dispositioned essentially by inspection. So I think there 6 are some issues in that report that are maybe a little bit 7 taller trees that others, that you may be talking to 8 specifically. Others, I am not sure really warrant the 9 vigor or the rigor of the full test. And that is where I 10 think I sense a little bit of a disconnect.
11 But, fundamentally, this has been excellent 12 information to help structure where we see ourselves going 13 with our technical working group report. But it also has 14 give many of us on the process side additional insights as
() 15 to, when you start going into rulemaking space, what are the 16 kinds of things that you would look for as a set of 17 conditions that you would expect to essentially see at a 18 plant, in a decommissioned phase of operation, and for what 19 kind of a period of time? And we are still working through 20 that. But we are talking a few years or in that general 21 neighborhood. But that would be an underpinning, I would 22 think, statement of considerations most likely, that we 23 would essentially put in that new Part XX that we referred 24 to in the Commission paper. So --
25 MR. CAMERON: Mike, is this enough to allow you to O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 ,
Washington, D.C. 20036 l (202) 842-0034 l
228 1 go forward with a discussion tomorrow on a cross-walk 2 between the presentation that George did today and the NRC 3 concerns? If that indeed is one of the things that is going 4 to be the most valuable for this consideration of 5 requantification, I mean is this -- you know, can we proceed 6 on that road?
7 MR. MEISNER: I think we need to talk a little 8 more. And I am not sure what the staff's reluctance is to 9 fully requantify. My simplistic view is, if you have got a 10 model, you change the numbers and you requantify it. We 11 have got some industry experts out here. I don't know, is 12 that -- are we talking about hundreds of man-hours or tens 13 of man-hours? Ken.
14 MR. CANAVAN: I don't know the particulars of that
) 15 model. But depending on how it is built, and who built it, 16 and are they still around, it can range dramatically.
17 MR. MEISNER: Well, I know when I was at Grand 18 Gulf, we requantified the entire Level 1 model on all PCs 19 overnight, and it was not a big deal. Maybe I am missing 20 the point. But the issue still is, and perhaps, you know, I 21 talked about the staff being suspicious about our 22 organizations and operators before, but there is -- I don't 23 know how we are ever going to have comfort in the results 24 without seeing the model requantified. How are we going to 25 know that the staff gave us some credit for a range of
/~'\ ANN RILEY & ASSOCIATES, LTD.
\~s/ Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
229 1 commitments? As opposed to the staff taking a menu of
) 2 things that they would like to have and just throwing it
-3 into a requirement stack?
4 MR. CAMERON: Can we get some succinct answers to 1
5 that question? Because then maybe we can agree to where we 6 are going to go from here. Rich.
7 MR. BARRETT: I think they are basically -- I will 8 just try to answer your last question, and that is, how can 9 you get some level of assurance that the commitments that 10 you want to make address the concerns that the staff has 11 raised in our analysis? And I will say that I think there 12 are three ways of doing that.
13 Number one, to the extent that it is possible, and 14 credible, we will requantify it. To the extent that it is
() 15 possible and credible. Where that is not possible or 16 credible, we will go back into the analysis and show 17 explicitly how the commitments that we are talking about, 18 that we think are important, address the factors that are 19 driving the model to the results that we got earlier, and we 20 will come to the conclusion explicitly that in our -- from a 21 qualitative perspective or from a quantitative perspective, 22 that the issues are dealt with.
23 Failing that, and I don't think we will fail that 24 in any cases, or, certainly, very few cases, we would 25 default to qualitatively addressing the issue and appealing
[~D ' ANN RILEY & ASSOCIATES, LTD.
\ss) Court Reporters 1025 Connecticut Avenus, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
230 1 to engineering judgment and to the experience of the PRA
() 2 analysts as a last resort in assuring all of our 3 stakeholders that, from the NRC's perspective, the 4 commitments that we are talking about address the issues to 5 our satisfaction and that we can walk away from the 6 sequences of interest.
7 So that, those are the three steps that I would 8 say we would go through to -- in that priority, to address 9 your question.
10 MR. MEISNER: I am not sure why it would not be 11 credible to recalculate a portion of a PRA. I am not sure 12 what you mean by that. But let's try a specific example.
13 We talked before the break about operator attentiveness.
14 Mike very clearly said that is modeled in the PRA. Let's --
f 15 why don't we have a discussion about how it is modeled and 16 what could possibly be done to make a difference in that 17 model and how that might change the results.
18 MR. CAMERON: I am not sure you, you know, want to 19 have it now, but you want to have it, right? And that might 20 illustrate the difficulties or the ease, is that what you 21 are suggesting?
22 MR. MEISNER: Chip, I am merely trying to get 23 something concrete. We are talking all around this. You 24 know, well, we will requantify to the extent we consider 25 credible. We will, you know, --
we came in with specific 1
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
231 1 suggestions here, concrete things that we are willing to do,
( 2 and we were expecting to get some feedback on what 3 difference that would make. What things do you feel are 4 important in the model?
5 You know, before the break, I tried to ask, well, 6 ,
shat is it? And I am still -- I would just like a straight 7 answer.
8 MR. BARRETT: Mike, I think we have -- we may have 9 organized this session -- we may not have organized this 10 session along the lines of what you want. We went through 11 our analysis a couple of months back when we had the meeting 12 at'the ACRS. We. talked about what we did, what was the 13 basis for it, and what results we got. We documented that 14 information in a preliminary report, and I can tell you that O( ,/ 15 I was very reluctant to do that, given the preliminary 16 nature of the information, but at your insistence we did i 17 document it, we made it available to everybody.
18 And we did so because we were convinced that that 19 was the best way to go in terms of efficiently and 20 effectively resolving this issue. That it was better for us
.21 to put the information out there in its preliminary form 22 rather than to have it reviewed for quality before it is put 23 on the public record, because we were convinced by you and 24 by others that that was the way that we could most 25 expeditiously find out what were the strategies that could ANN RILEY & ASSOCIATES, LTD.
Court Reporters ;
1025 Connecticut Avenue, NW, Suite 1014 i Washington, D.C. 20036 ;
(202) 842-0034 ,
232 1 resolve these issues and get on with it. And we did that.
() ^2 3
So we put that on the public record and we came back on the 21st of June, and, again, we described the 4 analysis and the basis for the analysis, and the results of 5 the analysis.
6 So we came in today feeling that, having done all 7 of that, our best strategy for today was to describe our 8 analysis in a very -- in an overview fashion, and then turn 9 the podium over to the industry to allow the industry to 10 come in and show us your analysis, and to talk to us about 11 what it is that you feel is important. And so we felt that 12 today we were in a discovery mode, 13 And I think that, my personal opinion is that the 14 best way to continue to conduct this meeting is to continue
) 15 to find out what kind of experience we have from these 16 gentlemen in the various plants that addresses these 17 sequences, that addresses -- tomorrow, addresFes the heavy 18 load sequences, to talk about the analysis that you have 19 already brought forward regarding the seismic, which we 20 think is a good analysis, and to continue with the agenda as 21 it is. And then we have a basis for going back and for us 22 to understand.
23 I feel that it is difficult to take a shred of 24 information that has been handed to us and begin to try to 25 reanalyze our sequences based on that. I think we need O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
l l 233 1 time. This is an area that requires thought and care.
I~ 2 MR. MEISNER:
D} I agree with you, Rich. And you hit 3 it exactly right, that when we got together with Sam and f
I 4 others, the purpose for the workshop was to identify the 5 strategies that would resolve the issues. So you put out l 6 your repcrt, we read it with care. We identified a number 7 of strategies. We come into the meeting today and we find 8 the number one staff concern -- the number one concern is 9 operator attentiveness. And we are simply asking, how does 10 that fit into the model, because it is not in the report, 11 and what can we do to reduce its impact? That -- it is a 12 real simple question.
13 MR. CAMERON: Can we start to try to answer that 14 simple question? Okay. Starting tomorrow with each one of
( 15 these one concerns, saying, okay, here is what the model 16 said about operator attentiveness. Here is what impact the 17 procedures, commitments, the industry has put forward would l 18 have on that concern. And that would at least begin to see 19 about taking us down the road towards whether we could 20 requantify that. Isn't that -- is that true? Can we do 21 that?
22 MR. BARRETT: No, we can do that. I think that I 23 would qualify that a little bit. I think that what we could 24 do is to say, here are the various -- here are some of the l 25 things we heard yesterday that would clearly have : r. impact
[\ ANN RILEY & ASSOCIATES, LTD.
i()
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 l
234 1 on that. I don't think we would want to come out tomorrow
() 2 and'say the impact would be X or Y. But these are the 3 things that would have an impact on that particular factor, 4 and these are the things that would not. That is something 5 that we could talk about.
6 MR. CAMERON: And that is at least a first step 7 towards knowing whether you are going to be able to 8 requantify?
9 MR. MEISNER: No, we are missing the point here.
10 I am not sure that operator attentiveness is even modeled in 11- the study. We sure don't see it in the report, yet it is 12 raised as a large general concern. I want to know what it 13 is and how it fits into the model.
14 MR. CAMERON: Can we come back tomorrow prepared?
(' ) 15 I mean someone brought up prioritization before. In other 16 words, which -- we have all these concerns. Which ones are 17 more important than the others? Which ones can we 18 specifically document in terms of their contribution? Can 19 we do some homework tomorrow? And operator attentiveness 20 was at the top of the. list, not necessarily because it was 21 the most important, but can we come back tomorrow and say 22 here~are the more important -- here are the most important 23 ones? Okay. Here is what we heard from the industry that 24 we'think is applicable to a revision. And then -- and take 25 it from there in a discussion? I mean does that make sense?
l L
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034 1 i
235 1 Mike or --
() 2 3 that.
MR. BARRETT: We can do that, we can certainly do But I think what we are doing if we do that is we are 4 going back to previous meetings in which we are talking 5 about what the NRC has done. I personally am much more 6 interested in hearing more about what the industry -- what 7 information, real information the industry can bring to the 8 table, the type of things we heard this afternoon.
9 We have had an ample opportunity to go back and 10 forth about the NRC's analysis. I think it is time -- I 11 think it is a better use of our time tomorrow to continue 12 along the lines we have been on for this meeting, and allow 13 us to take that information back and, in a more systematic 14 manner, in a professional manner, deal with the information
() 15 in a way that we are proud of and we can stand behind, 16 rather than continuing meeting after meeting after meeting 17 having the NRC present information and then -- and justify 18 the information in a model that we fully expect is not going 19 to be the final basis for the rulemaking.
20 MR. CAMERON: It may not, it may not be. But we 21 presented our information. The industry came back with 22 their information. And I think what they are seeking is !
23 some discussion at least about from -- a discussion between I 24 everybody around the table about how their information might 25 affect what we have done. Is that correct, Mike?
ANN RILEY & ASSOCIATES, LTD.
O. Court Reporters i
1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
236 1 MR. MEISNER: That is half of it. We are looking
() 2 3
to understand the staff concerns where it has never been put on the table, either in the report or our previous meetings.
4 I have never heard personally about operator attentiveness 5 as a major staff concern. If that is the case, we need to 6 hear about it. I have never heard about lack of automatic I
7 actions by mitigation systems as a staff concern. I don't 8 have the vaguest idea about how that could fit into the 9 model.
10 We are ready to make commitments that will resolve i 11 the risk -- those issues that are risk significant in the 12 staff's mind. We don't believe, though, that they are risk 13 significant to begin with. So we would like to be fairly 14 carefully seleative on what it takes to change the staff 15 model. Is that clear? We don't -- we believe tne staff 16 model is fatally flawed, but we are willing to step up and 17 be constructive and provide those commitments that will 18 resolve those specific issues in the staff model. And we 19 have to have an agreement to completely requantify that 20 model, not just piecemeal.
21 MR. CAMERON: Gareth, can you --
22 MR. PARRY: Yeah, I think we have got hung up on 23 the word " attentiveness." I don't really think -- I think 24 what it is, it is a surrogate for the fact that what we are 25 concerned about is human performance in response to O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
\
237 1 disturbances at the plant. That is intimately tied in with I\ 2 human error probability. I think you heard us say this V
3 morning that the things that we would like to understand are 4 things to do with what types of procedures, what types of 5 training, what types of instrumentation, some of which you 6 have addressed this afternoon. And those are the things 7 that will ultimately allow us to be able -- allow any HRA 8 analyst to be able to say whether he believes that the 9 probabilities assigning to a particular action is or high.
10 So that is the issue, I think. It is just -- and 1
11 the primary things I believe are instrumentation, equipment j 12 availability, staffing, procedures, training, all those 13 issues. And so, given that we understand all those 14 conditions that are present at the plant, I think that we (O)_ 15 can make more rational decisions about the right levels of 16 human error probabilities to include in the model. But that 17 is -- I think it is not an easy analysis because, as you 18 probably know, there aren't any universally accepted models 19 that will give you the right answer. It is going to be 20 ranges. But I think we could make more rational decisions l
21 once we understand the conditions that impact human l 22 performance.
23 MR. CAMERON: Okay. I am going to make a 24 suggestion here. That we come back tomorrow prepared to --
25 the staff prepared to specify what these various concerns ANN RILEY & ASSOCIATES, LTD.
ss Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
238 1 mean in terms of the model, to hear from -- to get a
() 2 3
discussion with the industry about what commitments they were talking about today that go to mitigating those 4 concerns, and use that as a step for the staff going back to 5 see about requantifying on each of those concerns.
6 I mean does that make -- I mean I don't know what 7 other path forward we have at this point.
8 MR. MEISNER: I would only add that if there is 9 something that is hidden in the model that isn't apparent 10 from the report, the staff needs to bring that out if it is 11 going to make a difference. And I hope that we can stop 12 qualifying quantification and just say we are going to 13 requantify it, in mass, across the boards, and reissue.
14 MR. CAMERON: Okay. Yeah, I think that we need to
() 15 be specific about where it is in the model. So, let's be 16 back here at 8:00. We are going to start off with -- maybe i 17 we should pick something other than operator or 18 organizational attentiveness. But let's start with that and 19 let's move through it. And it is not just the staff 20 listening to what the industry is going to bring forward, 21 but really informing the industry about exactly what the 22 nature of that concern is, and where it is in the model.
23 And, hopefully, by the end of the day, we will 24 have something that we are satisfied with. And, Ray, we 25 will see you tomorrow.
ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington, D.C. 20036 (202) 842-0034
i 239 1 MR. SHADIS: Yes. I would be pleased to see some
() 2 of.these initiating events requantified in terms of their 3 probability with specific analysis of changed plant 4 conditions over the years. The increased frequency, for 5 example, of air flights over some of these plants. The 6 changes that have been recorded and are predicted by the 7 U.S. Weather Service in terms of severe weather phenomenon.
8 There is a number of things that I think the staff 9 may be relying on dated information for, including seismic 10 risk analysis, and we would really like to see those not 11 only requantified generically, but also with some 12 site-specific examples.
13 For New England, we have some recent analysis on l 14 what is called the Miramichi fault, which runs down through O)
(, 15 the New England states, and, you know, we would like to see 16 that reflected.
17 Additionally, the industry is very concerned that 18 their comments have somehow been buried and not paid 19 attention to. I would like to see the comments of some of 20 the public participants reflected in some of the work of the 21 staff as it is brought forward also.
22 MR. CAMERON: Yeah, I think that is -- a watchword 23 is that whatever is done with this, whatever product comes 24 out with it, we need to show how we dispositioned people's 25 comments. We plan to do that. l 1
O ANN RILEY & ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 Washington,.D.C. 20036 t
(202) 842-0034
240 1 Okay. We will be here at 8:00 tomorrow and thank
() 2 3
you all very much.
[Whereupon, at 5:08 p.m., the meeting was 4 recessed, to reconvene at 8:00 a.m., Friday, July 16, 1999.]
'S 6
7 8
9 10 11 12 13 14
() 15-16 17 18 19 20 21 22 l 23 24 25 !
O, ANN RILEY.& ASSOCIATES, LTD.
Court Reporters 1025 Connecticut Avenue, NW, Suite 1014 4 Washington, D.C. 20036 (202) 842-0034 2
REPORTER'S CERTIFICATE This is to certify that the attached proceedings (G
G
) before the United States Nuclear Regulatory Commission in the matter of:
NAME OF PROCEEDING: REACTOR DECOMMISSIONING RISK WORKSHOP CASE NUMBER:
PLACE OF PROCEEDING: Gaithersburg, MD were held as herein appears, and that this is 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.
fVn mt,$
q) - -
(y~~
Mn Hundley Official Reporter Ann Riley & Associates, Ltd.
O v