ML24331A156
| ML24331A156 | |
| Person / Time | |
|---|---|
| Issue date: | 11/07/2024 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| NRC-0096 | |
| Download: ML24331A156 (1) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Docket Number:
(n/a)
Location:
teleconference Date:
Thursday, November 7, 2024 Work Order No.:
NRC-0096 Pages 1-81 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1716 14th Street, N.W.
Washington, D.C. 20009 (202) 234-4433
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com 1
1 2
3 DISCLAIMER 4
5 6
UNITED STATES NUCLEAR REGULATORY COMMISSIONS 7
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 8
9 10 The contents of this transcript of the 11 proceeding of the United States Nuclear Regulatory 12 Commission Advisory Committee on Reactor Safeguards, 13 as reported herein, is a record of the discussions 14 recorded at the meeting.
15 16 This transcript has not been reviewed, 17 corrected, and edited, and it may contain 18 inaccuracies.
19 20 21 22 23
1 UNITED STATES OF AMERICA 1
NUCLEAR REGULATORY COMMISSION 2
+ + + + +
3 720TH MEETING 4
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5
(ACRS) 6
+ + + + +
7 THURSDAY 8
NOVEMBER 7, 2024 9
+ + + + +
10 The Advisory Committee met via 11 teleconference at 8:30 a.m. EST, Walter L. Kirchner, 12 Chair, presiding.
13 COMMITTEE MEMBERS:
14 WALTER L. KIRCHNER, Chair 15 GREGORY H. HALNON, Vice Chair 16 DAVID A. PETTI, Member-at-Large 17 RONALD G. BALLINGER, Member 18 VICKI M. BIER, Member 19 VESNA B. DIMITRIJEVIC, Member 20 CRAIG D. HARRINGTON, Member 21 ROBERT P. MARTIN, Member 22 SCOTT P. PALMTAG, Member 23 THOMAS E. ROBERTS, Member 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
2 ACRS CONSULTANTS:
1 DENNIS BLEY 2
4 DESIGNATED FEDERAL OFFICIAL:
5 KENT HOWARD 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
3 C-O-N-T-E-N-T-S 1
PAGE 2
Meeting Called to Order, Chairman Kirchner 3
Introductions & Opening Remarks......
4 4
ACRS Meeting Guidelines & Logistics....
5 5
TerraPower Natrium Topical Report on Plume Exposure 6
Pathway Emergency Planning Zone Presentation &
7 Discussion.
8 Remarks, Subcommittee Chair, Tom Roberts 7
9 NRC: Candance Messieres..........
9 10 By TerraPower: Ian Guilford, and 11 John Biersdorf
.............. 11 12 Presentations & Discussion w/NRC staff
..... 47 13 By Mallecia Sutton, Michelle Hart, and 14 Phan Hanh 15 Public Comments................. 78 16 Adjourn 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
4 P-R-O-C-E-E-D-I-N-G-S 1
(8:30 a.m.)
2 CHAIR KIRCHNER: Good morning. The 3
meeting will now come to order. This is the second 4
day of the 720th meeting of the Advisory Committee on 5
Reactor Safeguards, ACRS. I'm Walt Kirchner, Chair of 6
the ACRS. The ACRS members in attendance in person, 7
are Ron Ballinger, Greg Halnon, Craig Harrington, Bob 8
Martin, Scott Palmtag, Dave Petti, and Tom Roberts.
9 ACRS members in attendance virtually via 10 Teams are Vesna Dimitrijevic and Vicki Bier, and I 11 believe that Matt Sunseri will join us later this 12 morning or this afternoon.
13 ACRS consultants Dennis Bley and Steve 14 Schultz are also with us this morning. If I missed 15 anyone, members or consultants, please speak up at 16 this point.
17 Not hearing anyone, Kent Howard of the 18 ACRS staff is the Designated Federal Officer for this 19 morning's meeting. No member's conflicts of interest 20 were identified for today's meeting. And I know we 21 have a quorum as well.
22 During our session today, this morning, 23 the Committee will discuss TerraPower, the Natrium 24 Topical Report on Plume Exposure Pathway Emergency 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
5 Planning Zone.
1 The ACRS was established by statute and is 2
governed by the Federal Advisory Committee Act, or 3
FACA. The NRC implements FACA in accordance with its 4
regulations.
Per these regulations and the 5
Committee's bylaws, the ACRS only speaks through its 6
published letter reports. All member comments, 7
therefore, should be regarded as only the individual 8
opinion of that member and not a Committee position.
9 All relevant information related to ACRS 10 activities, such as letters, rules for meeting 11 participation and transcripts, are located on the NRC 12 public website. And can be easily found by typing, 13 about us ACRS, in the search field on the NRC's home 14 page.
15 The ACRS, consistent with the Agency's 16 value of public transparency in regulation of nuclear 17 facilities, provides the opportunity for public input 18 and comment during our proceedings. We have received 19 no written statements or requests to make any 20 statement from the public. However, we set aside time 21 at the end of this meeting for any further public 22 comments or input.
23 A transcript of the meeting is being held 24 and will be posted on our website. When addressing 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
6 the Committee, the participants should first identify 1
themselves and speak with sufficient clarity and 2
volume so that they may be readily heard. This is 3
important for our Court Reporter, who is capturing our 4
proceedings for the record.
5 If you are not speaking, please mute your 6
computer on Teams. If you are participating by phone, 7
press *6 to mute your phone and *5 to raise your hand 8
on the screens. Please do not use the Teams, Chat 9
feature to conduct sidebar discussions related to the 10 presentations. Rather limit its use to that of 11 reporting IT problems.
12 For everyone in this room, please put your 13 electronic devices in silent mode and mute your laptop 14 microphone speakers. In addition, please keep sidebar 15 discussions to a minimum since we are working with 16 live ceiling microphones.
17 For presenters, we just remind you that 18 these table microphones are rather unidirectional and 19 you need to speak into the front of the microphone to 20 be heard online.
21 And finally, for anyone with feedback for 22 the ACRS about today's meeting, we encourage you to 23 fill out one of the public meeting feedback forms on 24 the NRC website. While not anticipated, portions of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
7 this meeting may be closed to protect sensitive 1
information, as required by FACA and the government in 2
Sun, in the Sunshine Act. Attendance during the 3
closed portion of the meeting will be limited to the 4
applicable NRC staff and its consultants.
5 And with that, I will now turn this 6
morning's deliberations over to our TerraPower 7
Design-Centered Subcommittee Chair, Tom Roberts. Tom.
8 MEMBER ROBERTS: Thank you, Walt. Good 9
morning, I'm glad -- good to see the people here in 10 person.
11 As Walt said this morning, the full 12 committee received a briefing on the Topical Report 13 and the staff draft Safety Evaluation Report from the 14 staff, for the Natrium Topical Report that's entitled 15 Plume Exposure Pathway Emergency Planning Zone Sizing 16 Methodology, which they've numbered NAT-3056.
17 This Topical Report was reviewed by our 18 Subcommittee, the TerraPower Natrium Design-Centered 19 Subcommittee on September 19th of this year. And I'll 20 provide some brief background regarding the Topical 21 Report. I'll summarize the discussions that we had 22 two months ago in a call on energy staff management to 23 kick off our presentations.
24 The Emergency Planning Zone, it's called 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
8 EPZ, is the area for which emergency planning is 1
needed to ensure prompt and effective actions can be 2
taken to protect the public in the event of 3
radiological incident. Judgment is required to 4
determine what accident scenarios should dictate the 5
size of an EPZ. The bases for these judgments, this 6
judgement rather, dates back to the 1970s and evolved 7
over the years, culminating in a new Regulatory Guide 8
1.242. It was issued late last year.
9 This Topical Report is the first use of 10 the new Regulatory Guide. To review allows us to 11 understand the choices made by the Applicant. Why the 12 staff accepted them with the limitations and 13 conditions that they've put on. And determine whether 14 this application of the new reg guide is suitable for 15 the revisions used in the Topical Report, or the reg 16 guide should be considered.
17 And looking not just at the TerraPower 18 proposal, but also what that means in terms of the 19 usefulness of the Regulatory Guide, and whether some 20 revisions may be warranted.
21 During our Subcommittee meeting, three 22 issues were identified where both we and the NRC staff 23 concluded further discussion was warranted. These 24 issues included the bases for various accident 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
9 frequency values used as part of the screening 1
assessments. Justification for why the ground 2
accelerations used to determine the limiting seismic 3
scenarios were probably conservative. The language in 4
one of the staff's limitations and conditions that did 5
not seem consistent with a non-light-water reactor 6
probabilistic risk assessment standard.
7 The Applicant and the staff will cover 8
these issues in more detail this morning. It's worth 9
noting that the Applicant has revised the Topical 10 Report since the Subcommittee meeting. And the staff 11 is in the process of revising the Safety Evaluation 12 Report, also based on the discussions that we had back 13 in Subcommittee meeting.
14 And with that being said, I'll turn this 15 over now to Ms. Candace De Messieres, who is the 16 Branch Chief of Technical Branch 2 and DANU -- I'm not 17 going to attempt to give you what that means -- in the 18 Office of Nuclear Reactor Regulations. Candace.
19 MS. DE MESSIERES: Thank you very much.
20 And I'll give it a try, I'll give it a try. Thank 21 you, Chair Kirchner and Member Roberts for the 22 opportunity to present today.
23 As was mentioned, I'm Candance De 24 Messieres, Chief of Advanced Reactor Technical Branch 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
10 2 in the Division of Advanced Reactors and Nonpower 1
Production and Utilization Facilities, or DANU in the 2
Office of Reactor Regulation.
3 So as was already highlighted, during this 4
meeting TerraPower representatives will provide a 5
summary. And NRC staff will discuss its review of the 6
Topical Report describing a risk-informed methodology 7
for determining the Natrium sodium fast reactor plume 8
exposure pathway Emergency Planning Zone or EPZ size.
9 I'll now highlight a few items related to 10 this review. First, this review is interdisciplinary 11 in nature. Bringing together expertise from across 12 the Agency in areas of emergency preparedness, 13 consequence
- analysis, and probabilistic risk 14 assessment.
15 Second, like several other Topical Reports 16 the ACRS has considered in support of Natrium, staff's 17 review has focused on acceptability of the 18 methodology, with reviews to ensure appropriate 19 implementation performed as part of separate licensee 20 actions.
21 Third, the report incorporates approaches 22 described in alternative emergency preparedness 23 requirements that were effective in December of 2023, 24 colloquially described as the EP for SMR and ONT rule.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
11 Though this Topical Report was originally submitted 1
prior to plan promulgation, it was later revised to 2
incorporate reference to the plan overall.
3
- Finally, also this was mentioned, 4
following the ACRS Subcommittee meeting on this topic, 5
TerraPower submitted a revision of its Topical Report, 6
and the staff revised its Safety Evaluation to further 7
describe and enhance clarity in areas of the selection 8
of the limiting seismic release scenario, hazard scope 9
and uncertainty treatments.
10 Mallecia Sutton, Michelle Hart and Hanh 11 Phan of the NRC staff will highlight the updated areas 12 in the staff presentation in more detail. Thank you 13 for the opportunity to present today and we look 14 forward to your observations and feedback.
15 MEMBER ROBERTS: Okay. Thank you, 16 Candace, we'll turn it over to TerraPower.
17 MR. GUILFORD: Thank you very much, good 18 morning. My name is Ian Guilford, I'm a Senior 19 Manager on the licensing team at TerraPower. And I'm 20 joined here in person by my colleague, Chris.
21 MR. COURTENAY: Yes, good morning. My 22 name is Chris Courtenay. I'm principal licensing 23 engineer with TerraPower.
24 MR.
GUILFORD:
We appreciate the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
12 opportunity to present today to the ACRS full 1
committee. We appreciate staff's time and expertise 2
reviewing the Topical Report and providing valuable 3
feedback during the audit process. And we also 4
appreciate Subcommittee's comments.
5 As mentioned earlier, we did revise the 6
Topical Report so that new revisions should be 7
available. We believe that that provides additional 8
clarity into the screening process that we use for the 9
events that feed into the EPZ methodology.
10 Today's presentation will be by John 11 Biersdorf. John Biersdorf is participating remotely.
12 He is a principal engineer at TerraPower. And I'll 13 turn it over to John.
14 MR. BIERSDORF: Morning, everyone. Can 15 everyone hear me or -- it looked like you could see me 16 for a hot second. When everyone on --
17 CHAIR KIRCHNER: We can hear you John, 18 just speak up a little bit more.
19 MR. BIERSDORF: Oh yes. Is this better?
20 CHAIR KIRCHNER: Yes.
21 MR. BIERSDORF: Okay. So my name is John 22 Biersdorf. I'm with TerraPower. I'm a PRA engineer.
23 I'm going to turn off my camera just to save 24 bandwidth. All right, so we'll get started. Yes, so 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
13 what we're going to do is just kind of lay down the, 1
going through the overview of our EPZ methodology. So 2
kind of high level, we'll just kind of talk a little 3
briefly about the guidance used, some of the 4
methodology. And then kind of go over the conclusions 5
that we have.
6 We're going to present essentially the 7
same material that we provided for the Subcommittee 8
meeting, except we've added a few additional slides to 9
add some clarification over some of the aspects that 10 needed some additional information from the previous 11 meeting.
12 So first when we developed this 13 methodology, we used some regulatory guidance.
14 Primarily, the guidance we used was Reg Guide 1.242.
15 And then we developed some of our criteria, using 16 NUREG-0396.
17 And then for our high-level overview of 18 the EPZ methodology as a whole, we basically start 19 with a compiled list of all our PRA event sequences.
20 We then use some specific selection criteria to parse 21 those events out into non-seismic events and seismic 22 events. As there's specific selection criteria 23 outlined in this presentation that kind of outlines 24 the categorization of both.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
14 And then, from there we'll utilize some 1
additional input we take into account, such as civic 2
meteorological data, and then some of the rad 3
consequence source term information. We quantify our 4
dose results. We'll get our EPZ dose results at a 5
specific distance that we're quantifying.
6 And then establish an EPZ boundary if all 7
of our EPZ dose criteria is met. And if not, we have 8
like an iterative process outlined where we can kind 9
of reassess the specific events. And then kind of 10 establish the EPZ boundary when our criteria is met.
11 Next
- slide, please.
So for our 12 non-seismic events, once we have our compiled list of 13 events, and we broke them down to the non-seismic 14 events, we'll have a, essentially we'll look at it.
15 And we'll say, okay, is this event a greater than 1E 16 to the minus seven?
17 If it is, is it screened, or if the 96th 18 percentile is greater than 1E to the minus seven, it's 19 screened in for evaluation. If it's not, then we'll 20 look at the mean frequency. If it's greater than 1E 21 to the minus eight, we'll then retain it for cliff-22 edge effects. And if it's neither of those, it'll be 23 screened out.
24 So this methodology will give us 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
15 essentially all DBAs for our analysis. And then a 1
list of events with a 95th percentile greater than 1E 2
to the minus seven, correctly screened in. And then 3
events greater than, with a mean value greater than 1E 4
to the minus eight are screened in for cliff-edge 5
effects.
6 And then our next slice we kind of walk 7
through some specific examples to try to illustrate 8
that point. So if we had -- and these are just for 9
illustrative purposes -- so if we had a mean frequency 10 of 2E to the minus seven, and the 95th percentile of 11 6E to the minus seven. So again, we have that 95th 12 percentile greater than 1E to the minus seven. So 13 this would be directly screened into our EPZ 14 evaluation.
15 Our next example is a mean frequency of 8E 16 to the minus eight. The 95th percentile of 2E to the 17 minus seven. So that mean frequency again is below 18 our cutoff threshold. However, that 95th percentile 19 is above. So again this would be screened into the 20 evaluation.
21 A third example would be a mean frequency 22 of 2E to the minus eight with a 95th percentile of 6E 23 to the minus eight. So the 95th percentile we've just 24 missed the threshold for our screening into the EPZ 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
16 evaluation. However that mean value, we sit at 2E to 1
the minus eight. So we'd screen it in and look for 2
cliff-edge effects.
3 And then with the final fourth example, we 4
have a mean frequency of 8E to the minus nine, with a 5
95th percentile 2E to the minus eight. So this one, 6
both the mean frequency and the 95th percentile are 7
below our cutoffs. So we've screened them out of 8
evaluation.
9 MEMBER ROBERTS: Hey, John, this is Tom 10 Roberts. I wanted to ask about that last case, and I 11 think I'm going to start my question with an excerpt 12 from Reg Guide 1.242 and want your comment on that.
13 This Section B.3 in Appendix B, it says if 14 based on PRA, the use of a low-frequency cutoff should 15 consider uncertainty. And it goes on to say the PRA 16 results should retain event sequences with frequencies 17 below the cutoff, and analysts should use them to 18 confirm that there are no cliff-edge effects and that 19 there is adequate defense in depth.
20 So I read that and there are similar words 21 in NEI's 18-04. There's no discretion of a low 22 frequency cutoff for the cliff-edge effect or defense-23 in-depth determination. There are if you, you know, 24 parse those two sentences carefully, there's three 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
17 concepts in those two sentences.
1 One is uncertainty, the second one is 2
cliff-edge effects, and the third one is defense in 3
depth. And your evaluation would appear to be 4
intended to cover all three.
5 And when you talk about uncertainty, you 6
know, I could see where you look at things like 7
grammar uncertainty, it makes sense having a limited 8
cutoff, because going much lower wouldn't get you any 9
additional information. Because you already have 10 enough information to reliably capture all the 11 sequences that could possibly exceed 10 to the minus 12 seven from, in America's perspective.
13 From a cliff-edge effect perspective and 14 defense in depth for that matter, you're getting into 15 more of a crossover between, you know, probabilistic 16 and deterministic pattern criteria. And if you look 17 at the cliff-edge effects in NEI 18-04 it's not very 18 detailed, but there is a companion document that has 19 a frequently asked question.
20 And if I could read from that real quick, 21 it talks about, the LMP methodologist has specifically 22 identified whether specific non-LWR design -- I muted 23 myself momentarily, so I'll start over. The LMP 24 methodologist does not specifically identify whether 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
18 specific non-LWR design exhibits cliff-edge effects, 1
except to the extent that if it exists, it may be 2
apparent in the results of the PRA in which then 3
provides risk insights, prevents sequence blood and 4
cutoff, for design, beyond design-based events at five 5
to 10 minus seven per plant year.
6 Yet the existence or lack thereof of 7
cliff-edge effects is really part of the safety 8
characteristics of a nuclear plant. And not merely a 9
property of the PRA models.
10 So I look at that, my interpretation of 11 that is the real purpose of not screening in the PRA 12 before doing the cliff-edge effects, is because the 13 only reason you're using the PRA is to make sure you 14 haven't missed something. That's more of a 15 completeness check.
16 There's a
sequence that the PRA 17 identifies, even below the cutoff, that says well, 18 maybe my safety margins aren't adequate. And, you 19 know, then they'll want to go look at that. Not 20 necessarily from a safety analysis perspective, but 21 from plant design perspective. But you kind of have 22 a choice of either you do something to increase the 23 safety margin and eliminate the cliff-edge effect 24 concern.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
19 Or you do something else, like you've, 1
maybe in a safety analysis you assume the cliff-edge 2
actually exists, and verify that the design is still 3
adequate, assuming the cliff-edge effect does exist.
4 So it's not really tied to a frequency. It's tied to 5
completeness of the design and adequacy of safety 6
margins.
7 Then I go on, the third piece of that is 8
defense in depth. And if you look at defense in 9
depth, the NEI standard, you know, at least my 10 opinion, isn't really clear on what you do with 11 sequences below five to 10 minus seven. Because it 12 really focuses on LBEs, licensing basis events and 13 verify you have defense in depth for the LBEs.
14 But there's another reference, NUREG-1860, 15 which one of our consultants, Dennis Bley, actually 16 was co-author of, back in about 20 years ago. And it 17 describes, if I could please read that briefly. It 18 says uncertainty associated with limitations in 19 knowledge, such as unknown or unforeseen failure 20 mechanisms, or unanticipated physical and chemical 21 interactions among system materials, cannot be 22 identified by PRA.
23 Defense-in-depth measures to address this 24 type of uncertainty can be established from repeatedly 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
20 asking the question, what if this barrier, measure, or 1
safety feature fails? Without a quantitative estimate 2
of the likelihood of such a failure, as well as by 3
ensuring consistency with established defense-in-depth 4
principles. This approach to defense-in-depth invokes 5
specific deterministic provisions to compensate for 6
the unexpected.
7 So I look at that, and I compare that to 8
the words of the NEI 18-04 standard, it appears 9
somewhat analogous to the cliff-edge, you know, 10 principle. Of course you're not using the PRA so much 11 as to determine the frequency, and say am I adequate, 12 but am I missing something?
13 You know, is there a barrier that has a 14 very high modeled reliability, such as if I was wrong 15 about it for whatever reason, because I don't really 16 understand all the phenomena that could happen, then 17 what's left in terms of safety?
18 Kind of a long question, but really it 19 gets down to, when you look at the words in the Reg 20 Guide 1.242, they don't have a lower frequency cutoff 21 for cliff edge and defense in depth, as such, but 22 there seem to be a good reason for that. So I wanted 23 to --
24 MR. BIERSDORF: Right but --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
21 (Simultaneous speaking.)
1 MEMBER ROBERTS: -- you know, long kind 2
of, just asking it, so what is the rationale based on 3
that for having low frequency cutoff for cliff-edge 4
effect?
5 MR. BIERSDORF: So our frequency threshold 6
for cliff-edge effect overall analysis for an LMP, we 7
don't have one. So we actually look at cliff-edge 8
effects down to our truncation threshold. But for the 9
inclusion in the, or to the EPZ analysis, we have the 10 1E to the minus eight threshold just to try to 11 incorporate essentially looking at events that we 12 understand that, you know, we can make a good 13 assumption on emergency response on.
14 So we cut, basically made a cutoff where 15 we include events that we feel are adequate 16 representations of how the facility will respond.
17 They are adequately covering the spectrum of events 18 requested in 1.242. And that we weren't inadequately 19 missing anything from our analysis.
20 So there wasn't anything specifically 21 associated with 1E to the minus eight, other than we 22 wanted to ensure that we were including the events 23 that we, or we were including enough of the events 24 that we have adequate confidence that we will properly 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
22 assessed our analysis and that it showed that within 1
our analysis.
2 So in our LMP, we do have a cliff edge 3
that goes down below that to show that we have 4
adequate defense in depth, but we haven't had those 5
two separate --
6 (Simultaneous speaking.)
7 MEMBER ROBERTS: Can I -- so what that 8
seems like is that if you've done the assessment and 9
beyond vague down to, you know, with no threshold, 10 then there are no cliff-edge effects remaining to 11 consider for this criteria. Is that a fair, fair, 12 John? I'm not quite saying how, if you've addressed 13 cliff-edge effects all the way down to, you know, no 14 limit. Then what residual sequences could possibly 15 exist that would screen in for the EPZ determination?
16 MR. BIERSDORF: I'm sorry, I'm not sure if 17 I caught that.
18 MEMBER ROBERTS: Well, if you're doing an 19 assessment of cliff-edge effects, with no limit for 20 LMP, then your plant design has already been assessed 21 to either be free of cliff-edge effects or presumably, 22 if you left cliff-edge effect potential in for 23 whatever reason, it would be adequately modeled in the 24 PRA.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
23 So it's just not clear what's left, you 1
know, once you get down beyond pica kinds of effect 2
determination. What could then possibly screen in 3
with or without a de minimis criterion?
4 MR. BIERSDORF: Yes, it would be just more 5
of that, you know, if we had something that is 6
somewhat on the cliff edge, down in the 1E to the 7
minus 12 from an LMP standpoint, we wouldn't have that 8
included into our EPZ analysis. It would be part of 9
our LMP, but not within the EPZ.
10 MEMBER ROBERTS: It seems like you would 11 have made a judgment either that you have the adequate 12 safety margin because of the perceived more likelihood 13 or whatever it is. And you choose to leave it in the 14 design, in which case it would then seem like defense-15 in-depth assessment would have you model that as 16 failed.
17 Because you, you know there is not as much 18 margin as you would like for that particular, the 19 aspect of design, you've, and because it was such a 20 low frequency, to say that's okay. That would seem 21 like you would maybe in all your models, not just the 22 EPZ, assume nothing is failed. Does that make sense?
23 MEMBER ROBERTS: I think so.
24 MEMBER ROBERTS: In power 10 to minus 12, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
24 you'd say, fine. If I got a cliff-edge effect at, I 1
see a 10 to minus 12 frequency path and assuming that 2
cliff edge occurs, and I fall off that cliff. Then 3
that would seem fine.
4 That's
- almost, you
- know, there's 5
intersection between the cliff-edge effect and defense 6
in depth. That would seem like a kind of defense-in-7 depth measure would be used in that particular, you 8
know, SSC fails. And then, so that on that assumption 9
your model still produces acceptable lower results.
10 MR. BIERSDORF: Yes, I guess, I'm not, are 11 you basically stating that would be enough and that we 12 should drop the cliff-edge? I'm kind of not following 13 at this point.
14 MEMBER ROBERTS: It seems like once you've 15 come to the cliff-edge effect determination and you 16 concluded that either there are no cliff-edge effects 17 of concern or if, the cliff-edge, you know, shut 18 itself, the resulting action frequency is still low.
19 You know, below these kind of, your regulatory limits, 20 then that would seem to be adequate.
21 And I guess, I'm not seeing the scenario 22 that's left that would be in this 10 to the minus 23 eight, to 10 to minus seventh range that would come 24 out, once you've gone through the assessment that the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
25 cliff-edge effects are acceptable.
1 MR. JOHNSON: Hi, this is Brian Johnson, 2
PRA manager. And I think I'm going to take a crack at 3
this, because I think I agree. We have separately, 4
from the EPZ topical, our LMP application of the PRA.
5 And they are actually parallel. So we don't directly 6
use the LBEs in the EPZ event selection.
7 They both, you know, come from the same 8
PRA. They are evaluated similarly. When we do the 9
defense-in-depth evaluation of cliff edges, down below 10 the bottom of the PRA and our defense-in-depth report, 11 which is separate from this report, we do evaluate, 12 you know, for large potential releases and other cliff 13 edges.
14 It is very similar to what's done here for 15 the purposes of the reg guide application for EPZ 16 topical. It is parallel though. So we don't do the 17 one-event selection that LBEs and EPZ events at the 18 same time.
19 They're done from the same source of the 20 PRA and so, you know, both the EPZ methodology and the 21 LMP methodology consider cliff edges somewhat 22 similarly. And we have the LMP that shows we don't 23 have a cliff edge of concern. And then the EPZ, when 24 looking at what events are in the PRA, does something 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
26 similar.
1 And if there was something large that 2
maybe for some reason the LMP says is acceptable risk 3
but would maybe resize the EPZ. We would include it 4
in our EPZ sizing.
5 You know, the LMP LBEs are a 30-day dose 6
at site boundary and don't, especially in the BDBE 7
range, have any specific limits. Well, sorry, below 8
the BDBE range, the once you're under 5E minus seven.
9 And so it's a judgment. The dose does go up to a 750 10 rem at the site boundary at 5E minus seven, which may 11 be causing potential prompt issues.
12 So that's why we have two different 13 evaluations. They don't have the same acceptance 14 criteria and they aren't the same application of the 15 PRA.
16 MEMBER ROBERTS: Hey, Dennis you had your 17 hand up a couple minutes ago. I can't see on the 18 screen if you still there, but --
19 DR. BLEY: I did.
20 MEMBER ROBERTS: -- do you have a 21 question?
22 DR. BLEY: I thought your question took 23 care of what I was going to say, but now they've 24 confused me a bit. It seems that if they've handled 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
27 it in the LMP part and identified any cliff-edge 1
effects, we should have a record of those to consider 2
when we get to the second evaluation.
3 I almost understand why they're doing it 4
again, but if they're doing it again, putting a cutoff 5
on the cliff-edge effects, doesn't seem to make sense 6
to me. I don't see any basis for it.
7 And it would seem, and I don't have the 8
details, so it's not clear, but anything that had 9
surfaced as a remaining cliff-edge effect in the LMP 10 analysis, at least those probably ought to be 11 considered regardless of frequency over here for the 12 very reasons they were just discussing.
13 That, you know, maybe it doesn't strongly 14 affect overall risk, but it might slightly change the 15 EPZ. But it doesn't seem logically consistent, I 16 think. So do we really --
17 (Simultaneous speaking.)
18 MR. JOHNSON: I'm going to maybe one more 19 time. The LMP evaluation of cliff edge is maybe going 20 to accept things that are not problematic, and we 21 evaluate them here as well. This is a more stringent, 22 potentially, limit at the end of the BDBE range and 23 below the BDBE range.
24 So I do think it's logically consistent of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
28 trying to apply that more stringent evaluation to a 1
reasonable set of events in that low frequency area.
2 DR. BLEY: I guess the only, I would be 3
more comfortable if you said you were going to this 4
cutoff and you were also looking at any cliff edges 5
that survived in the LMP analysis. I can't imagine 6
anything, I'm kind of like Tom, I can't imagine 7
there's anything that will really make a difference, 8
if that's the case. But that would seem a much more 9
consistent approach to me, but that might just be me.
10 MEMBER ROBERTS: Yes, thanks Dennis, I 11 tend to agree with you. There's an aspect of this 12 that the EPZ is part of defense in depth. And so when 13 you look at defense in depth, there may be scenarios 14 that you would consider to be low enough frequency to 15 not worry about in LBE space. But maybe you would be 16 concerned in the EPZ space because that's the last 17 layer of defense in depth.
18 So just seems logical that if you had a 19 low frequency cutoff in either the EPZ or the LMP, 20 will be in the LMP, free EPZ screen. So I agree with 21 Dennis. That seemed to make logical sense to have the 22 low frequency cutoff in the EPZ and yet not in the 23 others. It seems almost backwards.
24 Vicki, I can't see the screen, but I 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
29 understand you have your hand up.
1 MEMBER BIER: Yes, thank you. I'm going 2
to be brief about this because I went on it, like to 3
bat on it in Subcommittee. I continue to be a little 4
bothered by slides like this that seem to take the 5
mean and the 95th percentile as being of, as kind of 6
equal importance in determining what's screened in and 7
screened out.
8 And I understand screening is kind of a 9
zero-one choice. It's like they're in or out. But it 10 sort of doesn't make a lot of sense to me. You know, 11 the 95th could have been the 80th or the 99th. And, 12 you know, it would result in a different set of events 13 and seems kind of arbitrary.
14 I don't think it's a huge safety issue 15 just because there are, you know, it's in a sense 16 conservative that you're screening more things in.
17 But it just seems a little bit illogical from my 18 perspective. But anyway, I don't think it's a reason, 19 you know, that this should not be approved or 20 whatever. But it just seems awkward to me. Anyway, 21 that's my only comment right now.
22 MEMBER MARTIN: This is Member Martin. I 23 do have a maybe a clarification question, discussing 24 sequences. And in regard to EPZ we were earlier 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
30 talking about release scenarios or the consequences 1
that lead to a particular plant-damage state that have 2
similar, you know, release characteristics.
3 When you talk about these sequences like 4
on this slide, really talk about the probabilities 5
associated with sequence, families. Clearly a branch, 6
you could have multiple branches that have similar 7
plant-damage end states. And otherwise, similar in 8
the sense of the same kind of release characteristics.
9 It wouldn't be about a particular, you 10 know, non-seismic event sequence. It's likely 11 multiple ones. What are we talking about here? A 12 particular branch in the PRA? Or is there an exercise 13 to identify event sequence families, and that's the 14 metric being used here?
15 Really, because that relates more to a 16 release scenario than -- which is applicable here.
17 MR. BIERSDORF: So are you asking if we 18 both consider event sequence families, as well as 19 individual events sequences?
20 MEMBER MARTIN: So the clarification I'm 21 looking for is, as you were discussing the slide, you 22 were talking about sequences. You didn't mention 23 sequence families, which is really I think the more 24 important terminology here. You wouldn't want to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
31 have, say, you know, two or more events that, you 1
know, that may be, look like they screen out. But 2
ultimately, they are more or less the same release 3
scenario. Because collectively, it would cross over 4
the threshold.
5 Do you account, is there an exercise that 6
you look at these event sequences and then consolidate 7
them by the plant-damage end state and ultimately the 8
release scenario. Is that the metric you're using?
9 Or are you just point blank looking at the branches 10 and doing a cutoff exercise from that?
11 MR. BIERSDORF: So we look at the families 12 that's aggregated it, within the PRA when the events 13 come into the analysis.
14 MEMBER MARTIN: And do you feel like 15 there's any subjectivity to that exercise? So this is 16 where some of the, another kind of uncertainty weighs 17 into this.
18 MR.
BIERSDORF:
You know we have 19 uncertainties associated with particular sequences.
20 But there is a, what do you call it, systemic, just a 21 human contribution to uncertainties, which I've seen 22 hitting the realm of making folks like us more 23 nervous. How methodical, how objective is that 24 process?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
32 MR. BIERSDORF: Well, I mean we follow the 1
PRA standard for, you know, the event sequence 2
analysis. We have all of the, the steps. And we take 3
it into account with the uncertainty and stuff like 4
that as we feed into the events there.
5 So we obviously know that there's going to 6
be some subjectivity as you combine sequences. But, 7
you
- know, we try to understand where those 8
uncertainties are coming from. That's why we try to 9
come up with more of a, you know, this conservative 10 approach is where we're trying to capture as many of 11 the events as we can, into the analysis, to ensure 12 that we have a larger breath of scope, if you will.
13 Just to make sure that we know what these 14 sequences -- or the families of the events and stuff 15 like that we have in the analysis. And that we're 16 properly accounting for all it.
17 MEMBER MARTIN: So when you consider --
18 hang with me for a second, right. The domain of all 19 events represents 100 percent of events. With a 20 cutoff of anything, one times 10 to the minus eight, 21 you don't really know whether you're capturing, you 22 know, 50 percent, 90 percent, 99 percent of that 23 domain. Can you tell us what fraction of the domain 24 of events that you are otherwise capturing, using a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
33 cutoff like this?
1 MR. BIERSDORF: So can I --
2 MR. JOHNSON: Brian Johnson PRA manager, 3
again. I know for our amount of releases, it's over 4
99 percent. Our actual PRA integration cutoff studies 5
I believe, depending on which metric we're using, 6
right, because we have different integrated risk 7
metrics including, you know, probably exceeding 100 8
millirem dose, latent QHO, prompt QHO.
9 I believe our last study was around 10 10 minus 12 for total convergence. And 10 minus eight is 11 over 99 percent of all releases. I don't think that 12 convergence study is part of the EPZ topical directly, 13 but of course it would be available from the 14 evaluations of PRA.
15 And we can look at integrated risk, you 16 know, because we -- our objective is not to split 17 sequences 100 times to make them less frequent. It's 18 to understand how the plant operates and make sure 19 that we are grouping together events that are 20 functionally similar. And have the same dose 21 consequence.
22 MEMBER MARTIN: So you mention a number 23 like 99 percent coverage. You know, that's, at least 24 to me, more meaningful, right. Had me on various 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
34 uncertainties with, you know, risk models. You know, 1
to state a cutoff doesn't really provide any 2
benchmark, any context.
3 But you come back and say, oh, well, the 4
method actually results in greater than 99 percent or 5
something like that, you start making it plain though 6
that you've covered credible events. But in the 7
context of just a cutoff, you don't have that. It's 8
just a number.
9 There is some precedent for that. There's 10 the Reg Guide 1.216 related to LWR containment 11 integrity that uses a metric like that. And it might 12 have been something -- I mean this also comes as a 13 question for the staff. That there are other ways to 14 look at this, that, you know, might be a more 15 palatable metric to trying to find, whether it's a 16 compromise or, you know, some kind of definition of 17 where can you stop? That's all.
18 MEMBER ROBERTS: Yes, Tom here. This 19 seems to be related to a long-term issue that I know 20 that Vesna's been pushing, which is when you look at 21 these events, reactive to their extremely low risk 22 profiles, and is there some criterion that would be, 23 you know, based on the risk profile of that plant and 24 not necessarily based on frequency cutoffs? Because 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
35 the frequency cutoff might screen everything out.
1 So that's really, it's related to that.
2 I think that's probably unrelated to this discussion, 3
because that's, you know, almost more philosophical 4
than this. This does, it's a mighty attractive 5
precedent in terms of using, you know, perceived 6
likelihood as a basis for screening for the EPZ 7
determination.
8 But then that's certainly associated with 9
something that I think it is being worked and would be 10 related to that. The overall effort of whether the 90 11 percent of the sequences for example from the 12 particular reactor, becomes that reactor's cutoff 13 because, you know, that's something that makes sense.
14 And that's the theory, really.
15 I don't see any hands up. Vicki, I saw 16 you had your hand up, but you might have had it up 17 inadvertently. So I'm giving you a chance to speak 18 now if you want to.
19 MEMBER BIER: Yes, I think my comment has 20 been addressed. I just wanted to expand on Bob 21 Martin's comment that, you know, our, we wouldn't be 22 concerned about so much, two events that collectively 23 exceed the threshold. But more like, you know, 100 or 24 1000 events. And that's already been discussed. So 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
36 I'm fine with where we're at.
1 MEMBER ROBERTS: Okay. Let's proceed.
2 MR. BIERSDORF: Okay. So our next -- so 3
we had the non-seismic events. So the evaluation for 4
the seismic events, this is the second bin of the 5
events sequences that we -- event sequence standards 6
that we assess.
7 What we have is we have two essentially 8
stages for our analysis. We establish our PGA 9
limiting, our limiting PGA value, which I'll discuss 10 in the next slide. But for our CPA portion, what we 11 do is we develop a bounding-seismic-EPZ event that we 12 use as more of a conservative surrogate that would 13 establish, or that would identify whether or not the 14 seismic events would meet the EPZ criteria.
15 And by the time we have OLA, we'll have a 16 full seismic PRA that can give us those events for 17 those families that we can evaluate against the 18 criteria overall. So the limiting PGA value will 19 carry through. But we'll have two sets of events that 20 assess. So for CPA we have a bounding seismic event.
21 And by OLA we'll have the full gamut of events coming 22 out of the seismic PRA.
23 Excuse me, so for development of our 24 limiting PGA, we choose the lower of the two. So 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
37 either 2x times our GMRS value or 1g. So, again, 1
that's just the lower of the two. So if our 2x GMRS 2
is less than 1g, we would use that. If it's greater 3
than 1g, we use the 1g threshold.
4 And this just ensures that we have a range 5
of seismic hazards with an incredible range of ground 6
motions. It kind of acknowledges some of the 7
limitations identified within the PRA. And then some 8
of the uncertainties in the emergency response 9
infrastructure.
10 MEMBER MARTIN: Just a question, real 11 quick, 1g seems actually pretty conservative. You 12 feel like you can do that without kind of 13 overstraining the design or other decision-making 14 aspects? Trying to think, you know, the more limiting 15 size may be half that. Any thoughts on that? It 16 seems pretty high that acceleration and such.
17 MR. BIERSDORF: Yes, I mean it's pretty 18 high. We just wanted to make sure that, one, we were 19 capturing a level of conservatism within it. So that 20 we were ensuring that we covered the full range when 21 we assessed. And we also have the 2x GMRS, if you 22 just felt, you know. So we're not upper bound at 1g.
23 So, you know, if our 2 GMRS is lower, then 24 we can use that. And then that would still provide us 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
38 with a frequency threshold that provides us some 1
flexibility.
2 MEMBER MARTIN: And you think that's going 3
to happen more often, then let's say, the 1g 4
threshold?
5 I see a nod over here, but you obviously 6
don't want to say something.
7 PARTICIPANT: He'll, he'll let him --
8 MEMBER MARTIN: I think it's fine. Then 9
more likely than not, what are you looking at, maybe 10
.6 as, is that range or maybe a particular site and 11 Wyoming.
12 MR. BIERSDORF: Yes, that's, yes, I think 13 that was in our PSAR. So our 2x GMRS in our PSAR was 14 around.6g. So that's kind of where we're expecting 15 our threshold to roughly be around. At least for this 16 site. And then, you know, some sensitivities that we 17 did with just kind of generically, we felt that this 18 threshold was adequate for what we felt we needed for 19 the seismic events.
20 MEMBER MARTIN: All right. Thank you.
21 MEMBER ROBERTS: Can you clarify what 22 return frequency you're calculating the GMRS for, 10 23 minus four, 10 minus six?
24 MR. BIERSDORF: I couldn't hear that.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
39 MEMBER ROBERTS: The question is, what 1
return frequency are you looking at for this GMRS? Is 2
this a 10,000-year earthquake, 90-year earthquake? Do 3
you know?
4 MR. BIERSDORF: Yes, that's going to 5
depend on the site, obviously. But I don't think 6
it's, I can't give the specific, but we're looking at, 7
you know, the lower end of the BDBE range for our 2x 8
GMRS.
9 MEMBER ROBERTS: That might be about 10 to 10 the minus four, then? Okay.
11 MR. BIERSDORF: Yes, I believe. I don't 12 have the number off the top of my head for our actual 13 GMRS. But our 2x GMRS, we're sitting at that end of 14 the BDBE range.
15 MEMBER ROBERTS: Okay. So if you were to 16 compare to the lower end of the BDBE range of the 17 hierarchy, you know, intensity of an earthquake, how 18 does the 2x GMRS compare to, say, a 10 to the minus 19 sixth, you know, seismic event?
20 MR.
BIERSDORF:
In terms of like 21 magnitude, or impact?
22 MEMBER ROBERTS: Yes, in terms of, you 23 know, PGA.
24 MR. BIERSDORF: So you're looking at, like 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
40 something like what's the frequency for 1E to the 1
minus four? Is that what you're saying, versus the 1E 2
to the minus six?
3 MEMBER ROBERTS: Yes, I'm trying to 4
understand the margin of the GMRS. If you're using a 5
BDBE at the lower end, 10 to the minus four. Saying 6
then if you were to look at BDBE at the other end of 7
the range, would 2x GMRS bound that?
8 MR. BIERSDORF: I believe so, yes. Like 9
because then you'd have a lower seismic, I mean, I 10 don't know -- like we haven't done the full analysis 11 for outside of Kemmerer. So I can't really give you 12 a spectral range of where things land outside of that.
13 So I can only kind of really tell you 14 where we got from our site specific --
15 MEMBER ROBERTS: We understand this is all 16 site specific, and we're talking the methodology.
17 Just trying to get a sense of the margins. So it 18 seems like the justification for the margin is really 19 site specific and this would be the methodology.
20 And as I think, Candace said, during her 21 intro, if you apply the methodology and the 22 application is not conservative, then that would, you 23 know, be a separate discussion. That, but if you 24 agree then that makes sense to me.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
41 MR. BIERSDORF: Right, and I think that 1
was part of the, out of limiting condition, correct?
2 MEMBER ROBERTS: Right, and Seth will get 3
to that. But they did add a limitation condition that 4
says that. So I think that all ties together, it 5
makes sense to me. I just wanted to make sure I 6
understood it.
7 MR. BIERSDORF: Okay. Sorry about that.
8 So once we have our event sequences collected -- one 9
second. Sorry, once we have our event sequences 10 collected then we can establish our EPZ sizing based 11 off of the criteria listed here.
12 So we've developed these specific 13 criterion using the outline in Reg Guide 1.242. So 14 Criterion Alpha is pretty much identical to 1.242, 15 where you're just looking at projected doses from 16 DBAs. And then we use the PAG limits of 1 rem for 17 your mean, and 5 rem for the 95th for a 4-day TEDE.
18 Criterion Bravo, there in 1.242 they look 19 for most radiological release sequences. What we did 20 is we've made a frequency threshold, so any event 21 that's greater than 1E to the minus six, we'll look at 22 as, again against those PAG limits, which 1 rem for 23 mean, and 5 rem for the 95th percentile, again for 24 that 4-day TEDE.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
42 And then Criterion Charlie, 1.242 is 1
looking at immediately life-threatening doses from the 2
worse-case radiological release sequences. And we 3
used a selection where the frequency would be below 1E 4
to the minus six and greater than 1E to the minus 5
seven. And then this would exceed, it would not 6
exceed a 24-hour, 200 rem red-bone-marrow acute 7
effective dose.
8 So it just kind of goes over the 9
justification for using the red bone marrow. We used 10 specific sources, so NUREG-0396 identifies it as a 11 good surrogate for acute whole body. And then, you 12 know, the MACCS best practices and technical basis 13 NUREGS that's also identified in there.
14 And then in NUREG-4214 it kind of 15 identifies the hematopoietic syndrome as the dominant 16 cause of early fatalities, which is based off of the 17 red bone marrow. And so we just felt it was 18 identified as the most limiting and confirmed with 19 some of the sensitivities that we ran for our 20 analysis. And then we also utilized it, as it's 21 available for quantification within our MACCS 22 software.
23 So then in conclusion, we based, we just 24 want to reiterate that the EPZ size is based off of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
43 the smallest distance from where all or our criteria 1
is met. And then if the criteria is not met at our 2
desired size, the determination is made to make design 3
changes if we're still in that portion of the 4
development stage.
5 Otherwise, we make refinements to our 6
specific events, our quantification, and evaluate an 7
EPZ size. Otherwise again, we'd have to expand that 8
size out until we meet the specific criteria. And 9
that was identified a few slides prior. So that's the 10 EPZ methodology.
11 MEMBER ROBERTS: Okay. Vicki.
12 MEMBER BIER: Yes, a quick question. The 13 discussion about the bone marrow et cetera, makes it 14 sound like this is really being based on avoidance of 15 acute fatalities or acute health effects, whatever.
16 And I think for the, you know, current fleet of large 17 reactors, we have over time, concluded that acute 18 fatalities are really quite unlikely. And should not 19 be the main driver of protective actions.
20 That we really should be focusing on 21 cancer instead. And I just wanted to get a comment, 22 am I interpreting that correctly? And is this being 23 done because acute fatalities are actually a realistic 24 possible consequence of an accident? Or am I 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
44 misunderstanding something in this process? It just 1
seems strange to me that we're coming back to what 2
seems like an emphasis on acute fatalities.
3 MR. BIERSDORF: It was more essentially 4
just to capture the identified criterion in 1.242 for 5
that Criterion Charlie. And then we just tried to 6
align with some of the stuff that we saw in NEREG-0396 7
to see if we had the proper effects assessed and that 8
we could evaluate it accordingly.
9 MEMBER BIER: So would this be a big 10 driver of EPZ selection or just kind of for 11 completeness and compliance purposes?
12 MR.
BIERSDORF:
In terms of like 13 evaluation, against Charlie, or specifically the red 14 bone marrow?
15 MEMBER BIER: I guess I'm not sure which 16 I'm asking. Sorry.
17 CHAIR KIRCHNER: Vicki, this is Walt.
18 Maybe I could jump in here and say that what we've 19 seen now, TerraPower, we're having a lot of questions 20 because you're the person coming
- through, 21 implementing.
22 MR. BIERSDORF: Yes.
23 CHAIR KIRCHNER:
But with another 24 applicant, I just would observe that the Criterion 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
45 Charlie sets the distance in terms of the methodology 1
in the analyses we've seen with other applicants.
2 That becomes the limiting factor in setting that 3
distance.
4 MEMBER BIER: Yes, okay.
5 CHAIR KIRCHNER: I'm not saying that we're 6
emphasizing --
7 MEMBER BIER: Great 8
CHAIR KIRCHNER: - acute exposure, but 9
from the methodology standpoint, what we're seeing is 10 that that Criterion Charlie, is the one that sets the 11 distance when you go through the exercise and look at 12 the other criteria.
13 MEMBER BIER: Thanks.
14 MEMBER BALLINGER: This is Ron Ballinger.
15 Just a note, those are not NUREG-CRs.
16 MR. BIERSDORF: Oh, sorry.
17 CHAIR KIRCHNER: No, 0396 isn't a --
18 (Simultaneous speaking.)
19 MEMBER BALLINGER: 0396 is not, but the 20 others are, have to be CRs.
21 VICE CHAIR HALNON: The numbers are too 22 high.
23 MEMBER BALLINGER: Yes, they are too high.
24 MEMBER ROBERTS: Okay, any more questions, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
46 I guess, and so I'll address Vicki's question that the 1
methodology when applied will determine for this 2
plant, whether or not Criterion Charlie is limiting.
3 I guess, if you don't know, I don't know. But I could 4
make guesses, and it could go either way.
5 But the -- and Vicki, I think the real 6
answer to your question is in the reg guide. And if, 7
you know, from experience in other work, like Double 8
Tree PRA were to point to a, you know, a need for a 9
different criterion that would really be, you know, a 10 question back to the staff for the reg guide.
11 It just seems like TerraPower is tracking 12 pretty closely to the reg guide on these criteria. So 13 the reg guide has raised a different question.
14 Any other questions from members or 15 consultants?
16 Hearing and seeing none, TerraPower, thank 17 you for the presentation. We'll turn it over to the 18 staff. We're at less than an hour. I guess we'll 19 keep going. If you push the staff presentation, we'll 20 take a break.
21 So we'll pause for a couple minutes to 22 switch out between the staff, or for the Applicant and 23 staff.
24 CHAIR KIRCHNER: Tell me when you're 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
47 ready.
1 MS. SUTTON: I'm ready.
2 CHAIR KIRCHNER: Okay. I'll turn it over 3
to the staff. Mallecia, you're going to start?
4 MS. SUTTON: Yes.
5 CHAIR KIRCHNER: Okay. Go ahead. Pull 6
that microphone, Mallecia real close to you, if you 7
would. I know it gets in the way of your notes, but 8
9 MS. SUTTON: Yes, that's fine. Thank you.
10 CHAIR KIRCHNER: Pull it a little closer, 11 more.
12 MS. SUTTON: A little bit closer.
13 CHAIR KIRCHNER: There.
14 MS. SUTTON: Can everyone hear me now?
15 CHAIR KIRCHNER: Yes.
16 MS. SUTTON: Good morning, hello. My name 17 is Mallecia Sutton. I'm a Senior Licensing Project 18 Manager at the NRC and the lead Project Manager for 19 the TerraPower Natrium Project and for the Plume 20 Exposure Pathway Emergency Planning Zone Seismic 21 Methodology, or EPZ.
22 I'm going forward. With me today for this 23 presentation is Michelle Hart, who is the NRC Senior 24 Nuclear Engineer and our lead technical reviewer for 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
48 the TerraPower EPZ Seismic Methodology Topical Report.
1 And Hanh Phan is Senior Reliability and Risk Analyst.
2 One of the principal authors of the SE, Safety 3
Evaluation.
4 Other technical reviewers that supported 5
and authored the Safety Evaluation are Edward 6
Robinson, Senior Emergency Preparedness Specialist and 7
Kenneth
- Mott, a
Senior Emergency Preparedness 8
Specialist. And they are from the Office of Nuclear 9
Security and Incidence Response, also known as NSIR.
10 And this slide, it describes the agenda of 11 NRC staff's presentation today. We'll provide the 12 purpose of the staff's review and the staff's review 13 strategy of the Topical Report, provide an overview of 14 the contents of the staff's Safety Evaluation Report, 15 and then summarize the NRC staff's conclusion 16 regarding the TerraPower's EPZ Sizing Methodology 17 Topical Report.
18 Now, Slide 4. Today we will be presenting 19 the changes made to the Safety Evaluation since 20 September 19, 2024, ACRS Subcommittee meeting. The 21 staff revised a draft to the Safety Evaluation to 22 reflect the Topical Report Revision 3, to clarify the 23 treatment of uncertainties in non-seismic sequence 24 screening, including cliff-edge effects.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
49 Staff imposed a new limitation and 1
condition related to the selection of the seismic 2
release scenario and updated limitation and condition 3
regarding hazards. We also made editorial changes to 4
provide further clarity to the draft Safety 5
Evaluation.
6 Now, I'll turn the presentation over to 7
Michelle.
8 MS. HART: Next slide, please. My name is 9
Michelle Hart, I'm a Senior Reactor Engineer in 10 Technical Branch 2.
11 So the Topical Report's purpose and our 12 review strategy, the purpose of the Topical Report is 13 to provide a methodology and criteria that will be 14 used to establish the site-specific plume exposure 15 pathway Emergency Planning Zone size for the Natrium 16 reactor. And it's a risk-informed approach to 17 determine the Emergency Planning Zone size by meeting 18 criteria in the 10 CFR 50.33(g)(2). Or that's 19 commonly called the alternative EP requirements for 20 SMRs and other new technologies.
21 The staff's review strategy was to review 22 consistency of the methodology with the technical 23 basis for the alternative EP framework in 10 CFR 24 50.160 and conformance with the guidance on some 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
50 exposure pathway EPZ sizing. And that's Reg Guide 1
1.242.
2 We also assess the acceptability of the 3
risk-informed approach and the interaction with 4
overall Natrium licensing approach using the Licensing 5
Modernization Project methodology or LMP.
6 Next slide, please. Okay. I'll now go 7
over the steps in the methodology in an overview.
8 There are three major portions to the methodology.
9 The first is accident screening to 10 identify the spectrum events to be used in the plume 11 exposure pathway EPZ sizing analysis. And then 12 there's, they perform a radiological consequence 13 analysis using those accidents that they had screened 14 in. And they determine the plume exposure pathway EPZ 15 size distance based on the sizing criteria with the 16 basis in the regulatory requirement.
17 In the accident screening, they do compile 18 release sequences from the probabilistic risk 19 assessment for all internal and external initiators.
20 Perform screening of non-seismic release sequences 21 based on frequency, and including uncertainty and 22 screening of seismic release sequences using a unique 23 set of selection criteria.
24 And in the radiological consequence 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
51 analysis, they would use site meteorological data, 1
perform a source term and radiological consequence 2
analysis and evaluate the consequences against the 3
dose-related criteria for Emergency Planning Zone 4
sites.
5 And so now, the next topic we'll discuss 6
is the accident screening portion and I will turn that 7
over that over to Hanh Phan.
8 MR. PHAN: Thank you, Michelle. With 9
that, good morning, ladies and gentlemen. My name is 10 Hanh Phan, Senior PRA Analyst in NRR/DANU.
11 In the next four slides, I will focus on 12 the first step in the TerraPower's proposed 13 methodology for determining the PEP EPZ distance.
14 This step involves identifying the release sequences 15 which will be made under the design and site-specific 16 PRA.
17 The PRA is --
18 MEMBER ROBERTS: Hanh, excuse me, please 19 get a little closer to the microphone?
20 MR. PHAN: Yes, sir.
21 MEMBER ROBERTS: Thank you.
22 MR. PHAN: The PRAs will be full scale 23 assessment that covers internal and external hazards, 24 all operational modes, and all sources of radioactive 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
52 materials. Previously, the staff has presented to the 1
ACRS a Draft Guide 1413, which outlines the process 2
for identifying licensing events with nuclear power 3
plants, including comprehensive search for initiating 4
events.
5 However, since this guidance is still in 6
the draft forms, first, it cannot be referenced in the 7
staff's safety evaluation. Therefore, the staff has 8
established and modified a limitation and Condition 2 9
based on the ACRS comments at the Subcommittee's 10 meetings.
11 Which states, an Applicant that references 12 this Topical Report, must justify the technical 13 acceptability of the PRAs performed for selected 14 hazards and modes. Prior to the initial fuel load, 15 PRAs supporting this methodology must include all 16 applicable hazards and modes. Next slide, please.
17 Slide 8, the PRA use for the EPZ sizing 18 calculation will be developed following the 19 requirements in the, provided in the ASME/ANS non-LWR 20 PRA standard. And will undergo peer review.
21 Key assumptions and sources of uncertainty 22 will be assessed to evaluate their impacts on the 23 calculation.
Additionally, a
qualitative or 24 quantitative evaluation of security events will be 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
53 performed and enclosed in the PEP EPZ size 1
calculation. Next slide, please.
2 MEMBER MARTIN: Just real quick, Hanh.
3 MR. PHAN: Yes.
4 MEMBER MARTIN: So on your previous slide, 5
I assume that kind of answers the question of what 6
does it mean to be a technically justified PRA. Now, 7
you, like that list is comprehensive, the four, or are 8
there others that you've kind of assumed are pretty 9
more minor?
10 And maybe in knowing what your next slide 11 is, can you kind of lead into the question of cutoffs.
12 And what is, you know, more of a sufficient 13 consideration of events below, you know, the two 14 cutoffs?
15 Now, would that be included in an 16 acceptance criteria for a technically justified PRA, 17 is kind of what I'm getting at? Is there more and 18 then specifically is there something related to say, 19 what's enough as far as, in cutoffs or, you know, when 20 are we done?
21 MR. PHAN: Yes, so, yes, I will briefly 22 discuss that. Next slide. And give you at least some 23 confidence why we should -- we -- be doing that, that 24 the cutoff values are acceptable at this stage.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
54 CHAIR KIRCHNER: May I just ask you to 1
elaborate on the last bullet. When you say, security 2
events here including potential sabotage.
3 MR. PHAN: Yes, sir.
4 CHAIR KIRCHNER: So, and that would be 5
done qualitatively or quantitatively in addition to 6
the numerical methodology. It's like an overlay, if 7
you will.
8 MR. PHAN: In addition to the PRA accident 9
sequences.
10 CHAIR KIRCHNER: Okay.
11 MR. PHAN: It's like night. Two processes 12 was designed. They screened, they did best release 13 sequences. One for seismic and the others are non-14 seismic sequences. For non-seismic, these sequences, 15 all sequences and families generated by the PRA will 16 be accepted for inclusion.
17 The very first screening criterion asks if 18 the mean frequency contributes 1 percent or more to 19 the total frequency. So this is the first criteria to 20 give us the confidence that important sequences and 21 families to be accepted.
22 If you look at the rate of powers of 10 23 months, these are, the total frequencies is very low.
24 So one percent of that protects a family of sequences, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
55 of families to be assessed with the policy.
1 MEMBER MARTIN:
So maybe I
don't 2
understand, this is Bob Martin, this completely. So 3
I can imagine you could have, say for instance, 100 4
sequences and families, just to pick a number, okay.
5 And if you kind of think of, this could be an 6
assumption, failure on my part. But, you know, maybe 7
the distribution kind of looks like a power log, so 8
you get more of them on the bottom end.
9 I could just see a scenario, and we're not 10 necessarily talking about TerraPower or Natrium type 11 of projects, but you can see the results there, but 12 you could have 20 less than one percent. And 13 collectively, they might represent 10 percent of the 14 total. Why is that good enough?
15 I mean, but, you know, I liked the answer 16 earlier obviously when I asked, is it, you know, Phan, 17 considering, you know, the whole domain is 100 18 percent. And, you know, we're looking really at more 19 than 99 percent. That's like, provides context, 20 meaning the cutoff doesn't mean anything.
21 But then, you know, this first gate here 22 seems to open, you know, if it's a general methodology 23 that the NRC is going to accept for other 24 applications, it doesn't necessarily block, you know, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
56 a different design or a different, you know, a 1
different PRA from, you know, from not letting 2
something through, you know.
3 So I noted the example from Reg Guide 4
1.216 where it focused in the whole domain and it --
5 it's, do have navigational rights, the containment, 6
Delta PR containment integrity.
7 And it looks at what they call more 8
likely, and such. It's a little bit different and 9
their acceptance criteria is 90 percent. Where I 10 think something like here, we're looking at 99 percent 11 of release scenarios that we'd be interested in.
12 That context I think if somebody else, 13 another applicant is looking at what's being done here 14 and this is of course an open meeting and there might 15 very well be somebody listening. You know, this sets 16 of gates wouldn't necessarily work in every case.
17 But a, you know, another method such as I 18 think the Reg Guide 1.216 can apply in here. It might 19 very well do that. Do you -- how do you feel about 20 other people looking at this? I know your obvious 21 answer is that well obviously we'll look at everyone 22 case by case, that sort of thing.
23 But if you communicate that something like 24 this will work in every case, which it's, I mean 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
57 you're not, you know, you're not really doing that.
1 But, you know, without having anything more than Slide 2
X and non-proprietary documents, someone else might 3
pick this up and think, okay, I can do it too. And it 4
have -- really be a completely different safety 5
profile.
6 Now, you all would probably come back and 7
say that's not justified or what have you. But it 8
seems like it might create turn or something might 9
slips by. Okay, anyway.
10 CHAIR KIRCHNER: Aren't these, these are 11 the wickets from the reg guide.
12 MEMBER MARTIN: No, the cutoffs. NO, 13 they're cutoffs, 10 to the minus seven, 10 to the 14 minus eight. Those are not the reg guides. And 15 that's their methodology.
16 CHAIR KIRCHNER: This is the NRC's?
17 MEMBER MARTIN: No, this is the --
18 CHAIR KIRCHNER: TerraPower.
19 MEMBER MARTIN: TerraPower's.
20 CHAIR KIRCHNER: TerraPower's.
21 MEMBER MARTIN: Okay.
22 CHAIR KIRCHNER: You stand corrected.
23 Thank you.
24 MEMBER MARTIN: So that's it. It's a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
58 unique methodology. And it's their interpretation.
1 But --
2 CHAIR KIRCHNER: Isn't too different from 3
what we've seen before.
4 MEMBER MARTIN: Well, we haven't seen any 5
of this before. This is --
6 CHAIR KIRCHNER: We have.
7 MEMBER MARTIN: For cutoffs. no.
8 CHAIR KIRCHNER: Yes, I bet you we have.
9 MEMBER MARTIN: Cutoffs. Who?
10 CHAIR KIRCHNER: NuScale. Well, but 11 that's --
12 (Simultaneous speaking.)
13 MEMBER MARTIN: Well, that's not relative.
14 CHAIR KIRCHNER: Similar approach, not the 15 same numbers, but -- yes, thanks, Martin. Go ahead.
16 MEMBER MARTIN: Okay. Well, like I said 17 it concerns me that it's not a general methodology 18 because it doesn't have context. And I feel like, 19 that, you know, there's precedent in the NRC. I've 20 seen that containment reg guide where, you know, you 21 have the context. It's a little bit more clear where 22 cutoffs are. And, you know, maybe the worst thing 23 that comes out of accepting it, the methodology. It 24 looks like this with cutoffs, is that it creates turn.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
59 But nonetheless, why do that? Why not have a, you 1
know, a deeper discussion about, you know, context 2
with an eye towards what other people might otherwise 3
interpret from what you've got here?
4 And there's a question in this, I mean, 5
how much were you aware of, say, the other methods 6
like in Reg Guide 1.216? And it might be a new 7
thought, but, you know, so kind of catching it off 8
guard, that's not completely fair.
9 MR. PHAN: Thank you very much for your 10 concerns and feedback. But this has a big 11 application, because of their implementation of LMP.
12 We believe that when this particular application has 13 been (audio interference) is this figure.
14 MEMBER MARTIN: Well, like I said, I think 15 the answer I got from TerraPower was an acceptable 16 answer. But at the same time, it's not necessarily 17 something that obviously it comes out of the 18 documentation.
19 MR. PHAN: Yes.
20 MEMBER MARTIN: But you know it, but I'm 21 not sure the public would necessarily recognize it.
22 MR. PHAN: Thank you.
23 MEMBER ROBERTS: But if I understand what 24 you're asking, maybe I'll give just a kind of a 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
60 trivial example. If you had a sequence at 9 times 10 1
to the minus nine, it was, you know, a thousand times 2
the limit. Then you'd like to have something that 3
says well, yes, I looked at all the scenarios that 4
encompass some fraction of the dose consequence, and 5
so that would screen in this sequence because it has 6
so much more dose than you want to be screened in.
7 MEMBER MARTIN: Right.
8 MEMBER ROBERTS: But it all seems like if 9
you take away the low cutoff for the cliff-edge 10 effects and defense in depth, that you'll get a 11 similar, you know, view of things. That you would be 12 looking at those types of sequences and concluding 13 that there may be a cliff edge of sorts.
14 That just by getting a little bit worse, 15 you get significantly worse cutoffs. And that, you 16 know, came to a cliff edge and that would come out of 17 an assessment like that. So maybe related to the 18 question we asked earlier about why you would have a 19 low cutoff for cliff-edge effects. And it may be 20 another way to get at the same thing that you're 21 concerned about.
22 MEMBER MARTIN: Well, I think it comes 23 down to transparency really. You know, it seems in 24 the case of TerraPower that, you know, it meets any 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
61 kind of, let me say, a safety criteria that, you know, 1
I think we can all accept. But on the surface, it's 2
just not obvious, until you have this conversation.
3 And then that, it should be a little bit 4
more transparent. I think there are other methods 5
that are more transparent and, you know, we have 6
people from the public online. You know, that call 7
like this and, you know, they're not going to 8
appreciate all the wickets and how they assure, you 9
know, you get adequate protection.
10 But as you said, in this particular 11 application we're fine. But I'm not so sure going 12 forward, you know. We should, you know, can't say we 13 can't accept methods like this, but we should have an 14 eye towards improving the transparency of the 15 presentation of the information through the methods 16 topical. That's my concern. It's just not obvious.
17 MEMBER PETTI: Just a clarification, it 18 looks like the realm of cliff-edge effects, those sets 19 of events that are going to be looked at, are 20 frequency limited here, in this chart, right. But 21 then we also heard though that they have sort of a 22 broader cliff-edge in defense in depth evaluation 23 through an LMP, right.
24 But they're making a judgment here for 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
62 EPZ, they're going to limit that broader space based 1
on a frequency cutoff (audio interference), right. Is 2
that how you, that's what I thought I took away from 3
it.
4 MEMBER ROBERTS: Yes, that's what I heard.
5 You know, whether or not that makes sense to me is a 6
different question than what I heard.
7 MEMBER PETTI: Oh, yes, yes, okay.
8 Because I'm, yes, I'm always worried about the case 9
that's right out there. You know, I mean, yes, if you 10 go down lower in frequency, doses should be going up, 11 right. And all of a sudden is, as you quoted from 12 that report, you get low enough in frequency, there 13 really are no more values and safety functions. Just 14 you're left with the technology.
15 And there are characteristics of each of 16 the advanced technologies that show up when you get to 17 those incredibly low values. And those are kind of 18 the inherent hazards that you're designing the plant 19 to prevent, you know, from rearing their ugly heads.
20 Okay, no, I understand that. Thank you.
21 MR. PHAN: Thank you. The next question 22 is whether the mean frequency is equal to or greater 23 than 1 minus seven per reactor year? If yes, is 24 included. If no, then the next question is whether 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
63 the 95 percentile values is equal to or greater than 1
1 minus seven reactor year?
2 Again, if yes, it is included. If no, a 3
last question is whether the mean frequency is equal 4
to or greater than 10 to the minus 8 per reactor year?
5 If yes, the sequence or family, it further assesses 6
for any cliff-edge effect. If any identified, the 7
release sequence or family will be included.
8 Otherwise, it will be excluded from the EPZ dose 9
calculation.
10 Next slide, please. Slide 10. For the 11 seismic release sequences, a limiting peak ground 12 acceleration, PGA, will be used to define seismic 13 events for the PEP EPZ calculation at the CP stage.
14 This limiting PGA will be set to at least 15 twice the ground motion response spectrum and up to 16 the maximum of 1g. Based on the existing paper for 17 most U.S. sites, (audio interference) GMRS is less 18 than 1.0 g. But in some cases, or some locations, it 19 asks this question.
20 However, it's up above the cap 1g. So to 21 ensure a conservative approach, the staff established 22 a new limitation and Condition by adding the use of 1g 23 PGA and states, an applicant that references this 24 Topical Report will provide site-specific 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
64 justification for the use of the upper bound PGA of 1g 1
when exercising the methodology in the Topical Report 2
Section 3.7, Selection of Seismic Release Sequences.
3 Additionally, the site-specific scoping 4
level seismic PRA will be performed to provide other 5
insights at the CP stage.
6 With that, I return to Michelle for 7
continuing with the consequence analysis.
8 MEMBER PETTI: I just have a broader 9
question. Maybe it's too site specific to answer, but 10 from a probability and a risk perspective, is the 11 approach to seismic similar to the internal hazard 12 approach?
13 Or do you end up going -- because we don't 14 have data and we're going down to 10 to the minus 15 eight -- so you'd ask yourself, what's the 10 to the 16 minus eight to earthquake? To be consistent, but I 17 know that those aren't 10 to the minus eight, the 18 earthquake measures, they're much higher in frequency, 19 right. So there is this incongruity between external 20 events and the internal events.
21 Is that -- it's just sort of -- because 22 we're limited in terms of our knowledge, I would 23 imagine on seismic, when you get those really low 24 probabilities. Is that fair to, when you're 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
65 calculating?
1 MR. PHAN: At this stage because the 2
seismic PRA is not deep enough. So we cannot say that 3
1E minus eight or 1E minus six that is (audio 4
interference) sequence for seismic events. That's why 5
the applicant, they propose using 2x GMRS. Because 6
that's up to PRA during the seismic period. So at the 7
cutoff stage we decided PRA fully completes and 8
perform according to the PRA standard. That would 9
answer what you're asking.
10 MEMBER PETTI: Thank you.
11 MR. PHAN: Thank you.
12 MS. HART: Okay, thanks. So once you 13 determined your events and the families that you're 14 going to include in your radiological consequence 15 analysis, you would pass it on to the radiological 16 consequence analysis step.
17 And in the EPZ sizing analysis methodology 18 Topical Report, the methodology uses the outputs and 19 methodologies in related Topical Reports which are 20 currently under review. That being the radiological 21 source term Methodology and the radiological release 22 consequences methodology.
23 The source terms are developed as a part 24 of the safety analysis at PRA and used as input to the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
66 EPZ sizing analysis. In other words, there are no EPZ 1
specific or unique source terms that would be 2
developed, because these are the same scenarios that 3
you're using for your safety analysis and PRA.
4 Next slide, please. So to estimate the 5
dose, the Topical Report methodology identifies the 6
doses are first estimated at the proposed EPZ 7
distance, and then compared to the three dose-based 8
criteria. And that the plume exposure pathway, EPZ 9
will be established at the furthest distance for which 10 all three criterion are met.
11 And as TerraPower had described earlier 12 today, there is an iterative risk process to determine 13 the plume exposure pathway, or to establish that in 14 the first place. And if the result of the proposed 15 distance is unacceptable, they will either change the 16 plume exposure pathway EPZ size, or they will make 17 design changes if they're during the design phase.
18 Next slide, please. So for the dose-based 19 criteria, there are the three criteria for most 20 radiological release sequences. Criterion A, for 21 design-basis accidents and Criterion B, for those 22 sequences and families with a mean release frequency 23 greater than 1 times 10 to the minus six, per reactor 24 year.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
67 You would compare the mean, 96-hour dose 1
for the event, to the lower end of the EPA Protective 2
Action Guide range, that's 1 rem TEDE. And you would 3
compare the 95th percentile, 96-hour dose for the 4
event, to the upper end the EPA PAG range, and that's 5
5 rem TEDE. And that PAG range is for the early 6
response, or the early phase of radiological incident 7
for things such as evacuation or security measures.
8 Next slide, please.
9 VICE CHAIR HALNON: Michelle, this is 10 Greg, just I'm sorry to go backwards. Dose 11 estimation, I think it was Slide 12. I would assume 12 that the expectation is that you're going to see site 13 boundary being that first estimate. Because that's 14 really, if you look at it, the full piece of it is the 15 cliff effects, the cliff-edge financial, if you will, 16 aspect.
17 If they determine the site boundary is 18 adequate, is it then your expectation that they 19 continue to refine it down to, to get the actual EPZ 20 within the site boundary? Or are you just going to 21 allow them to stop and say okay, we're good at site 22 boundary, if you're good.
23 MS. HART: So for the purposes of 24 determining a plume exposure pathway EPZ size, it is 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
68 about protecting the offsite population. And so once 1
you have control of the area where you're, you know, 2
if you're within the site boundary, there's no reason 3
to have to --
4 VICE CHAIR HALNON: Okay.
5 MS. HART: -- refine it further.
6 VICE CHAIR HALNON: So we'll probably see 7
a lot of site boundaries, I mean you can actually 8
change the site boundary, you can buy more land --
9 (Simultaneous speaking.)
10 MS. HART: Correct.
11 VICE CHAIR HALNON: Instead of just space.
12 MS. HART: Yes, and that's another 13 refinement that could take, as opposed to refining the 14 facilities.
15 VICE CHAIR HALNON: Thank you, Michelle.
16 MS. HART: You know, it's up to them, but 17 yes, I think, you know, certainly the methodology is 18 not saying that this is to determine a site boundary 19 EPZ, but it is a goal that could be accomplished 20 through the use of the methodology.
21 CHAIR KIRCHNER: And we saw that, 22 Michelle. I'm sure you remember, we saw that with an 23 application for an early site permit.
24 VICE CHAIR HALNON: Right.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
69 CHAIR KIRCHNER: And they did a plant 1
envelope approach, and then they could show that for 2
one technology selection, they could meet these kind 3
of criterion within the EAB. For another selection it 4
would go a little bit over the EAB in terms of setting 5
the EPZ.
6 VICE CHAIR HALNON: And this is in effect, 7
an indirect plant envelope determination. So that's 8
why I wanted to see if there was expectation to get 9
down. And there may be a good reason, maybe there's 10 a high factory or something within the site boundary 11 that is using the process heat. You may not want that 12 part of it. So you may want to refine it down. But 13 for us and in most cases, site boundary is really the 14 goal, I think. Thanks.
15
`
MS. HART: So back to, where are we, Slide 16 4, 14, 15. Yes, 15.
17 MS. SUTTON: We'll move on, okay. Just 18 want to make sure.
19 MS. HART: Okay. Oh, no we missed one, 20
- 14. I'm sorry, let's go back. Yes.
21 So for the worst-case radiological release 22 sequences, those are those sequences of the mean 23 release frequency between 10 to the minus seven per 24 reactor year and 10 to the minus six per reactor year.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
70 This is a comparison to a dose metric of 200 rem red 1
marrow acute effective dose for a 24-hour exposure 2
period.
3 It would generate a dose-distance chart 4
mapping the dose reduction as one moves away from the 5
Emergency Planning Zone boundary. And into the 6
analysis to ensure that the dose drops rapidly beyond 7
the plume exposure pathway EPZ size -- excuse me, 8
plume exposure pathway EPZ to ensure that there's a 9
low probability of early deterministic health effect, 10 which is consistent with what was originally developed 11 in NUREG-0396. And we do reference that kind of dose 12 aggregation evaluation in the guidance in Reg Guide 13 1.242.
14 MEMBER ROBERTS: If you can go back one 15 slide. There you go, right there. It says -- chance 16 to -- just to make a comment and then ask a question.
17 The header of this slide says, Criterion A DBAs and 18 Criterion B mean release frequency greater than 10 to 19 the minus six per reactor year.
20 What that actually says in the reg guide 21 is design basis accidents agree with that. And 22 Criterion B is projected dose for most core melt 23 sequences. So there is some precedent for defining 24 most core melt sequences in these, you know, SMRs and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
71 ONTs as 10 to the minus six per year.
1 So there's precedent for that but like 2
Bob's been reporting out, that's not in the reg guide.
3 And how you would determine what most core melt 4
sequences means in a particular reactor is something 5
that's kind of left up to the individual Applicant.
6 And so I would tend to agree with Bob.
7 If it's not in the reg guide, it's not 8
something that, you know, that says how to do it.
9 It's just there's a precedent that TerraPower is 10 following here, and the staff is, you know, accepting 11 based, you know, largely on precedent.
12 But it leads to a question, there's an NEI 13 document, an NEI 24-05 that was submitted a few months 14 ago, and I think the document is still proprietary, so 15 we can't really ask specific questions about it. Just 16 wondering in general, is this something the staff is 17 in the process of reviewing? And is that review 18 refining any thoughts in terms of evaluating reports 19 like this Topical Report?
20 MS. HART: So, certainly, yes. We have 21 received that NEI report, and we're in the process of 22 reviewing it, still early in the process. So, you 23 know, a lot of the tenets of this are generally 24 similar to, is that white paper. All of these things 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
72 are based on the similar, you know, the previous 1
applications we've gotten, all kind of using the 2
NUREG-0396 as a basis and maybe generalizing it or 3
using it in a less light-water reactor specific way.
4 And maybe bringing it up to date somewhat.
5 So as we go through that review, there may 6
be some things that we find out. And if, you know, 7
there are some major issues with this methodology, not 8
saying that are, it looks generally consistent at this 9
point, or some refinements that should be made, you 10 know, it may have to -- if it's important enough, we 11 would have to evaluate an implementation of something 12 that uses this methodology. But I don't see a major 13 issue at this point. But like I said, we're still 14 early in the review.
15 MEMBER ROBERTS: Okay, thank you. Yes, I 16 think I mentioned in the Subcommittee meeting that 17 we're likely interested in, you know, getting briefed 18 where you end up on that document, just to understand 19 what the current views are on that risk in defining 20 that basis. Thank you.
21 MS. HART: And I did want to mention the 22 quote that you have from the reg guide was, a 23 reference to what's in NUREG-0396. It's not 24 necessarily meant to be generally applicable for non-25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
73 light-water reactor designs or even new light-water 1
reactor designs.
2 You know, these are the core, phrase core-3 melt sequences. It's not necessarily meant to be 4
generally applicable. But you should do something 5
similar. You know, most release sequences I think is 6
more realistically an equivalent.
7 MEMBER ROBERTS: Okay. I agree. Thanks 8
for the clarification, Michelle.
9 MS. HART: Now, back to, where are we?
10 Slide 15. Yes, I think we finally got there. Any 11 more questions about the dose-based criteria?
12 CHAIR KIRCHNER: Well, just, Michelle, 13 since you'll get several applications in, already have 14 dealt with some -- that last, the previous slide. I 15 hate to, you're right, regress, but previous slide 16 talked about, and it drops off greatly. That was 17 0396. Is there any in 242, Reg Guide 1.242, is there 18 any definition of that? Or is that something that you 19 just evaluate and put a qualitative judgment on?
20 MS. HART: So the reg guide does point to 21 the evaluation that was done in Appendix 1 or I, 22 whichever way it's really supposed to be said, to 23 NUREG-0396. So there was that evaluation and you can 24 see it on the chart, that it does drop off rapidly.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
74 And so I think it's more of a qualitative discussion 1
within the reg guide.
2 And, you know, you kind of have to look at 3
the actual information for the specific site to be 4
able to really tell if you've got a reasonable 5
Emergency Planning Zone size so that you can prepare 6
for the majority of events that go down.
7 CHAIR KIRCHNER: Thank you.
8 MS. HART: So, Slide 15. So this is 9
discussing what we were just talking about, the 10 probabilistic dose aggregation and treatment of 11 uncertainty. That last criterion, the low probability 12 of exceeding an early deterministic health-effect 13 criterion considers scenarios with lower frequency 14 than used to determine the licensing basis events with 15 LMP.
16 Topical Report Section 6.3 does describe 17 the uncertainty and sensitivity analysis methodology 18 for this EPZ sizing analysis. It would go through a 19 Monte Carlo sampling on the PRA event frequencies.
20 There's source term and consequence 21 analysis uncertainty results are included, and those 22 are determined through the related methodology Topical 23 Reports. So there are, a lot of the outputs of that 24 is uncertainty in both of those.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
75 And the cliff-edge effect evaluations, 1
which we've talked about before, and that's for those 2
events down to what? Between 10 to the minus seven 3
per reactor year and 10 to the minus eight per reactor 4
year.
5 The primary focus is on single failures 6
that would dramatically change either the risk metrics 7
or the effects of the accident sequences such as 8
timing, plant response, source terms, or end states.
9 And this is, certainly within this 10 methodology, it's within the context of, would it 11 change your determination of plume exposure pathway 12 EPZ size?
13 Next slide, please. And then the last 14 criterion is a criterion in the regulation as well.
15 It's the necessity of predetermined prompt protective 16 measures. They would use the radiological release 17 timing to identify if there are any necessary prompt 18 protective measures for the event. Each event is 19 assessed individually and any identified protective 20 measures for those events would inform the emergency 21 plan and procedures.
22 Next slide, please. So with that that's 23 a description of the staff's review. Our conclusions 24 are that the staff determined that the Topical Report, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
76 subject to the limitations and conditions, provides an 1
acceptable approach for developing analysis to aid in 2
the determination of a site and design-specific plume 3
exposure pathway EPZs for the Natrium reactor.
4 And accordingly, we conclude that the 5
Topical Report can be used in establishment of a plume 6
exposure pathway EPZ size to support emergency 7
planning and preparedness in compliance with the 8
regulatory requirements as listed.
9 Are there any further questions?
10 MEMBER ROBERTS: Thank you, Michelle and 11 Mallecia. Are there any questions from the members 12 here or consultants? Dave.
13 MEMBER PETTI: Michelle, the Monte Carlo 14 analysis and the cliff-edge analysis, what struck me 15 is you're really just looking for single failures that 16 would cause an increase. And that would tell you that 17 perhaps you don't have also adequate defense in depth.
18 But these lower, let's talk about the 19 broader set of events we can see, is that we had heard 20 earlier that they're down much lower than 10 to the 21 minus 10, or 10 to the minus 12. Those most likely 22 would require multiple failures.
23 And so to me I guess that's a big 24 difference in terms of when you think about it from an 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
77 EPZ. You're really just looking at, you know, each 1
barrier, if you will, for a simple term, individually 2
and not in its collective sets that would push you 3
down into lower risk. When you look at it through the 4
EPZ lens.
5 MS. HART: Correct. But yes, this is not 6
intended to refine the design or reduce risk in 7
general. This is trying to determine a reasonable set 8
of events or likely events that you would have to have 9
prepared prompt protective measures ready to go in 10 case you have an event.
11 And so there's a little bit of a different 12 focus in the goal of what this analysis is trying to 13 do here.
14 MEMBER PETTI: For me, that's helpful. It 15 puts everything into perspective. Thank you.
16 MEMBER ROBERTS: Okay. Any other 17 questions from the members in the room or online? I 18 don't see any hands up. So, Vicki, Dennis, anybody 19 else out there with questions for the staff?
20 MEMBER BIER: Not right now, no.
21 MEMBER ROBERTS: All right, seeing and 22 hearing none, now is the perfect time to go out for 23 comments from members of the public. So anybody who's 24 been, you know, not been in the room but online, who 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
78 would like to make an observation, go ahead and raise 1
your hand. And see Ed Lyman has his hand up and then 2
I'll call you.
3 Please unmute, state your name and 4
affiliation if appropriate, and then state your 5
comment. Ed.
6 DR. LYMAN: Yes, thank you. This is Edwin 7
Lyman from the Union of Concerned Scientists. Can you 8
hear me?
9 MEMBER ROBERTS: Yes.
10 DR. LYMAN: Yes, right. So, yeah, my 11 comment is that I appreciate the comment, I believe it 12 was Member Martin, about transparency and public 13 understanding of these processes. And from my 14 perspective, I think the most important thing for the 15 public to understand is that any screening process is 16 not cooked up to cherry-pick or to cherry-pick the 17 event sequences to allow the Applicant to get the 18 answer that they want.
19 And so that's really I think should be 20 front of mind when you think about these processes.
21 And in that context, I'd like to at least go back to 22 the discussion of whether it's appropriate to screen 23 out for frequency events that may have a cliff-edge 24 effect. And I just want to offer one example, since 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
79 the discussion was very abstract. I just want to put 1
this forward, for why I think it isn't appropriate to 2
have that screening.
3 And so let's say you have a flood, a 4
design-basis flood where it's the probable maximum 5
flood height. And the likelihood -- so your defense 6
is, you know, you have flood protection against that 7
flood height.
8 So if that would, occurs, then the only 9
way that there would be, that would progress to a 10 core-damage accident is if the flood barrier fails.
11 And let's say that's a very low probability event, 12 right, since the flood barrier is presumably very 13 robust. So you might screen that out.
14 But that's exactly the example where a 15 cliff edge might turn that very low probability event 16 into a very high probability of core damage, if the 17 actual flood height exceeds the probable maximum by a 18 small amount.
19 So just looking at that example, it does 20 not seem appropriate to do that frequency screening at 21 the first pass, to eliminate those kinds of sequences, 22 where the cliff edge might increase the frequency of 23 core damage by many orders of magnitude. So that's my 24 comment. Thank you.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
80 MEMBER ROBERTS: And thank you, Ed. Any 1
other members of the public out there? I don't see 2
any hands up. Give it another few seconds for 3
somebody to raise their hand or speak up.
4 Okay, and with that, I thank both the 5
Applicant and the staff and the presentations were 6
very helpful. And with the changes, I think both the 7
Topical Report and the Safety Evaluation will help in 8
terms of a clarity of what's intended and what the 9
additional conditions are intended to do.
10 And I guess, turn this back to Walt it's 11 the next -- I would recommend we take a break and then 12 go to the --
13 CHAIR KIRCHNER: Yes, precisely. I agree 14 with you wholeheartedly. So we'll take a break at 15 this point. And we'll come back at 10:35 Eastern 16 time. And we'll proceed with discussion and --
17 MEMBER PETTI: Do we need a Court 18 Reporter?
19 CHAIR KIRCHNER: Hold on. We'll come back 20 at 10:35 and we'll proceed to discussion and then 21 letter writing. Tom has a letter drafted on this 22 Topic. And with that, I think we can excuse the Court 23 Reporter for the rest of the day. Is that correct, 24 Larry?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
81 MR. BURKHART: That should be correct, and 1
I shall be in touch, I believe, tomorrow, when you're 2
(audio interference) right?
3 CHAIR KIRCHNER: Okay. So tomorrow would 4
be, we continue letter writing for which we do not 5
need to make a record, correct?
6 MR. BURKHART: You are correct.
7 CHAIR KIRCHNER: With that, yes, so thank 8
you -- who's our Court Reporter?
9 MR. BURKHART: James.
10 CHAIR KIRCHNER: James, Thank you, James.
11 I think we're finished with your services for this 12 meeting. Thank you. And we are in recess.
13 (Whereupon, the above-entitled matter went 14 off the record at 10:18 a.m.)
15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1716 14th STREET, N.W., SUITE 200 (202) 234-4433 WASHINGTON, D.C. 20009-4309 www.nealrgross.com
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved Natrium reactor is a TerraPower & GE Hitachi technology ACRS Full Committee Meeting November 2024 Plume Exposure Pathway Emergency Planning Zone Sizing Methodology TP-LIC-PRSNT-0031
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 2
- Overview of NAT-3056, TerraPower, LLC (TerraPower) Natrium Topical Report:
Plume Exposure Pathway Emergency Planning Zone Sizing Methodology
- Guidance Used
- Methodology
- Conclusion Table of Contents
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 3
- Regulatory Guide 1.242, Performance-Based Emergency Preparedness for Small Modular Reactors, Non Light-Water Reactors, and Non-Power Production or Utilization Facilities was utilized in developing the overall methodology.
- Supporting information from NUREG-0396, Planning Basis for the Development of State and Local Government Radiological Emergency Response Plans in Support of Light Water Nuclear Power Plants was used to assist in the development of the evaluation criteria.
NAT-3056 Guidance Used
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 4
- Assesses all modes and all hazards
- Incorporates site specific meteorological data
- Assesses radiological consequences
- Identifies prompt protective measures
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 5
- Includes DBAs
- Includes events with a 95th percentile release frequency greater than 1E-7
- Includes events for cliff-edge consideration if they have a release frequency greater than 1E-8 NAT-3056 Non-seismic Event Selection
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 6
- Mean: 8E-9, 95th percentile: 2E-8 screened out of EPZ evaluation NAT-3056 Non-Seismic Event Selection Example
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 7
- Developed a bounding seismic event for the CPA
- Will generate a list of seismic EPZ events using a limiting PGA as threshold NAT-3056 Seismic Event Selection
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 8
- Limiting PGA is LOWER of the two:
- Ensures that range of seismic hazard within credible range of ground motions
- Acknowledges limitations of the SPRA
- Acknowledges uncertainties in the emergency response infrastructure NAT-3056 Development of Limiting PGA
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 9
- Criterion A: Projected doses from the DBAs would not exceed 1 rem mean 4-day TEDE and 5 rem 95th percentile 4-day TEDE.
- Criterion B: Projected doses from most radiological release sequences would not exceed 1 rem mean 4-day TEDE and 5 rem 95th percentile 4-day TEDE.
- Criterion C: Immediate life-threatening doses from the worst-case radiological release sequences would not exceed 24-hour, 200 rem red marrow acute effective dose.
NAT-3056 Criteria for PEP EPZ Sizing
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 10 Supported by guidance
- NUREG-0396 - Planning Basis for the Development of State and Local Government Radiological Emergency Response Plans in Support of Light Water Nuclear Power Plants
- NUREG-7009 and NUREG-7161 - MACCS best practices and technical basis in the SOARCA
- NUREG-4214 - Health Effects Models for Nuclear Power Plant Accident Consequence Analysis Identified as most limiting and confirmed by sensitivity analysis Available for quantification within MACCS NAT-3056 Red Bone Marrow Use
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 11
- EPZ size is set based on smallest distance criteria are met.
- If the criteria are NOT met at the desired EPZ size, determination is made if design changes or refinements can be made to reduce EPZ size. Otherwise EPZ is expanded to meet the criteria.
NAT-3056 Conclusion
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 12
SUBJECT TO DOE COOPERATIVE AGREEMENT NO. DE-NE0009054 Copyright © 2024 TerraPower, LLC. All Rights Reserved 13 CPA - Construction Permit Application DBA - Design Basis Accident EPZ - Emergency Planning Zone GMRS - Ground Motion Response Spectra MACCS - MELCOR Accident Consequence Code System PEP - Plume Exposure Pathway PGA - Peak Ground Acceleration PRA - Probabilistic Risk Assessment SOARCA - State-of-the-Art Reactor Consequence Analysis SPRA - Seismic Probabilistic Risk Assessment TEDE - Total Effective Dose Equivalent Acronym List
NRC Staff Review of TerraPower Topical Report:
Plume Exposure Pathway Emergency Planning Zone Sizing Methodology Office of Nuclear Reactor Regulation (NRR)
Division of Advanced Reactors and Non-Power Production and Utilization Facilities (DANU)
Office of Nuclear Security and Incident Response (NSIR)
Division of Preparedness and Response (DPR)
ACRS Full Committee Meeting November 7, 2024
Review Staff
- Michelle Hart, Lead Technical Reviewer - Senior Reactor Engineer, NRR/DANU/UTB2
- Hanh Phan - Senior Reliability and Risk Analyst, NRR/DANU/UTB2
- Edward Robinson - Senior Emergency Preparedness Specialist, NSIR/DPR/RLB
- Kenneth Mott - Emergency Preparedness Specialist, NSIR/DPR/RLB
- Mallecia Sutton - Senior Project Manager, NRR/DANU/UAL1 2
Agenda
- Topical report (TR) purpose and review strategy
- Safety evaluation (SE) overview
- Focus on updates since ACRS subcommittee meeting
- Conclusions 3
Revised Draft SE Following discussions during the September 19, 2024, ACRS Subcommittee meeting, staff made changes to the draft SE
- Reflected TR revision 3 to clarify the treatment of uncertainties in non-seismic sequence screening, including cliff-edge effects
- Imposed a new limitation and condition related to the selection of the seismic release scenario
- Updated limitation and condition 2 regarding hazards
- Made editorial changes to provide further clarity to SE 4
TR Purpose and Review Strategy
- Purpose of TR
- Provide methodology and criteria that will be used to establish the site-specific plume exposure pathway (PEP) Emergency Planning Zone (EPZ) size for the Natrium reactor
- Risk-informed approach to determine EPZ size by meeting criteria in 10 CFR 50.33(g)(2)
- Review strategy
- Review consistency with the technical basis for the alternative EP framework in 10 CFR 50.160 and conformance with guidance on PEP EPZ sizing (RG 1.242)
- Assess acceptability of the risk-informed approach and interaction with overall Natrium licensing approach using the Licensing Modernization Project methodology 5
Note - 10 CFR 50.160 was issued November 16, 2023. TR Revision 1 made changes to address the proposed final rule and guidance.
TR EPZ Sizing Methodology Steps Overview Accident screening to identify spectrum of accidents
- Compile release sequences from the probabilistic risk assessment (PRA) for all internal and external initiators
- Perform screening of non-seismic release sequences based on frequency, including uncertainty
- Perform screening of seismic release sequences with a unique set of selection criteria, including uncertainty Radiological consequence analysis
- Collect meteorological data and incorporate into the radiological consequence analysis
- Perform source term and radiological consequence analysis
- Evaluate the radiological consequences against the PEP EPZ dose-related criteria to determine if changes are needed Determine PEP EPZ distance based on EPZ sizing criteria 6
Development and Uses of PRA
- Design-and site-specific PRA will be used to identify release sequences
- The PRA will address internal and external hazards, all modes of operation, and all sources of radioactive material
- Updated Limitation and Condition 2: An applicant that references this TR must justify the technical acceptability of the PRAs performed for the selected hazards and modes (e.g., site-specific scoping level PRA). Prior to the initial fuel loading, PRAs supporting this methodology must include all applicable hazards and modes.
7
Development and Uses of PRA
- The PRA will be developed in accordance with the guidance provided in the ASME/ANS non-light water reactor (NLWR) PRA standard
- The PRA will undergo a peer review
- Key assumptions and sources of uncertainty will be assessed to determine their impacts on the calculation
- A qualitative or quantitative assessment of security events will be conducted and documented in the size calculation 8
9 All Sequences and Families Mean frequency contribute 1% or more to the total frequency?
95th percentile value 1x10-7/ry?
Screened out from PEP EPZ evaluation Mean frequency 1x10-8/ry?
Mean frequency 1x10-7/ry?
Any cliff-edge effects?
Screened into the PEP EPZ evaluation Yes Yes Yes Yes Yes No No No No No Selection of Non-Seismic Release Sequences
Selection of Seismic Release Sequences A limiting PGA will be used to establish seismic event(s) for use in the PEP EPZ calculation at the CP stage The limiting PGA will be two times the ground motion response spectrum (GMRS) or a maximum of 1.0 g A new Limitation and Condition was established related the use of 1.0 g PGA:
- New Limitation and Condition 5: An applicant that references this TR will provide site-specific justification for the use of the upper bound PGA of 1.0g when exercising the methodology in TR Section 3.7, Selection of Seismic Release Sequences.
The site-specific scoping level seismic PRA (SPRA) will be performed to provide additional insights 10
Radiological Consequence Analysis to Support EPZ Sizing
- Radiological Source Term Methodology
- Radiological Release Consequences Methodology
- Source terms are developed as part of safety analysis and PRA - used as input to EPZ sizing analysis
- NRC staff will review the development of source terms as part of its review of a related license application safety analysis report, including the implementation of the related source term TR 11
- Information on Natrium topical report reviews available at https://www.nrc.gov/reactors/new-reactors/advanced/who-were-working-with/licensing-activities/pre-application-activities/natrium.html
Dose Estimation
- TR methodology identifies that doses are first estimated at a proposed PEP EPZ distance and compared to the three dose-based criteria
- The PEP EPZ will be established at the furthest distance at which all three criteria are met
- Iterative process to determine PEP EPZ
- If the result at the proposed distance is unacceptable, change the PEP EPZ size or make design changes (during design phase) 12
Comparison to Dose-related Criteria for Most Radiological Release Sequences Criterion A (DBAs) and Criterion B (mean release frequency > 1 x 10-6 per reactor year (/ry))
- Compare mean 96-hour dose for event to lower end of EPA Protective Action Guide (PAG)* range (1 rem TEDE)
- The EPA PAG levels to recommend evacuation or sheltering of the public during the early phase of a radiological incident are 1 to 5 rem (10 to 50 mSv) projected dose over four days (2017 EPA PAG Manual)
Comparison to Dose-related Criterion for Worst-case Radiological Release Sequences Criterion C (mean release frequency 1 x 10-7/ry to 1 x 10-6/ry)
- Compare to dose metric of 200 rem red marrow acute effective dose for a 24-hour exposure period
- Generate dose-distance chart mapping the dose reduction as one moves away from the EPZ
- Analysis to ensure that the dose drops rapidly beyond the PEP EPZ boundary 14
Probabilistic Dose Aggregation and Treatment of Uncertainty
- Criterion C considers scenarios with lower frequency than used to determine licensing basis events with LMP
- TR Section 6.3 describes the uncertainty and sensitivity analysis methodology
- Monte Carlo sampling on PRA event frequencies
- Source term and consequence analysis uncertainty results are included, as determined in the related methodology TRs
- Cliff-edge effect evaluations
- Events with frequencies between 1 x 10-7 /ry and 1 x 10-8 /ry
- Primary focus on single failures that would dramatically change either risk metrics or the effects of the accident sequences (timing, plant response, source terms, or end states) 15
Necessity of Predetermined Prompt Protective Measures Criterion
- Radiological release timing used to identify necessity of prompt protective measures
- Each event assessed individually to determine if timing supports necessity of prompt protective measures
- Identified protective measures will inform emergency plan and procedures 16
Conclusions The NRC staff determined that the TR, subject to the limitations and conditions, provides an approach acceptable to develop analyses to aid in the determination of a site-and design-specific PEP EPZ for the Natrium reactor.
Accordingly, the NRC staff concludes that the TR can be used in establishment of the PEP EPZ size to support emergency planning and preparedness in compliance with the regulatory requirements in 10 CFR 50.33(g) and 10 CFR 50.47(c)(2), as applicable, for prospective TerraPower Natrium reactor CP or operating license applications under 10 CFR Part 50.
17
Abbreviations ANS American Nuclear Society ASME American Society of Mechanical Engineers CFR Code of Federal Regulations CP construction permit DBA design basis accident EP emergency preparedness EPA Environmental Protection Agency EPZ emergency planning zone g
gravitational acceleration GMRS ground motion response spectrum L&C limitation and condition NLWR non-light water reactor PAG protective action guide PEP plume exposure pathway PGA peak ground acceleration PRA probabilistic risk assessment RG regulatory guide ry reactor year SE safety evaluation SPRA seismic probabilistic risk assessment TEDE total effective dose equivalent TR topical report 18