ML20138A609
| ML20138A609 | |
| Person / Time | |
|---|---|
| Issue date: | 04/22/1997 |
| From: | NRC ADVISORY COMMITTEE ON NUCLEAR WASTE (ACNW) |
| To: | |
| References | |
| NACNUCLE-T-0113, NACNUCLE-T-113, NUDOCS 9704280127 | |
| Download: ML20138A609 (460) | |
Text
. _ _ . . _ . . - . - . .. - _ - -.
Official Transcript sf Proce: dings
@ NUCLEAR REGULATORY COMMISSION ACWW7~ of B
Title:
Advisory Committee on Nuclea Waste 91st Meeting TRO8 (ACNW) '
RETURN ORIGINAL TO BJWHITE I M/S T-2E26 Docket Number: (not applicable) 41s-713o 1
. TnANra l Location: Rockville, Maryland o I 1
Date: Tuesday, April 22,1997 i
d Sv$$ NO hE Work Order No.: NRC-1089 Pages 1-339 ACNW OFFICE COPY - RETAIN FOR l J
THE UFEOFTHE COMMITTEE 2800J0 I NEAL R. GROSS AND CO., INC. ,.
Court Reporters and Transcribers -
1323 Rhode Island Avenue, N.W. a FMp l W[ Washington, D.C. 20005 I j g ', ' f (202) 234-4433
.f if y/ 34 l
.!!!! !!l ll[lll! , ll e -
_____m _ _ _ _ _ _ _ . _ _ _ . . _ . _ . . . . _ _ . _ . . . . _ . . _ _ _ _ _ _ _ . . _ _ _ _ _ _ _ _ _ _ . _ _ . _ _ _
i i
t O :
i
~
DISCLAIMER {
PUBLIC NOTICE !
BY THE !
- j. UNITED STATES NUCLEAR REGULATORY COMMISSION'S ;
l ADVISORY COMMITTEE ON NUCLEAR WASTE .
APRIL 22, 1997 i i
l-The contents of this transcript of the i i proceedings of the United States Nuclear Regulatory l
Commission's Advisory Committee on Reactor Safeguards Nuclear Waste on APRIL 22, 1997, as reported herein, is a record of the discussions recorded at the meeting held on the above ;
4 i date. l
- This transcript has not been reviewed, corrected i j
and edited and it may contain inacc'Iracies.
e f'
.t .
. O
- V NEAL R. GROSS COURT REPORTERS ANDTRANSCRil3ERS 1323 RilODE ISLAND AVENUE, NW (202)234-443 L WASilINGTON, D C. 20005 (202)234-4433
1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION GI 3 + + +++
4 91st MEETING l 5 ADVISORY COMMITTEE ON NUCLEAR WASTE 6 (ACNW) 7 + ++ + + 1 i
8 TUESDAY I 9 APRIL 22, 1997 l 10 + + +++
11 ROCKVILLE, MARYLAND 12 +++ + +
13 The Review Committee met at the Nuclear
/~T 14 Regulatory Commission, Two White Flint North, Room T2B3, 15 11545 Rockville Pike, at 8:30 a.m., Paul W. Pomeroy, 16 Chairman, presiding.
17 18 COMMITTEE MEMBERS:
19 PAUL W. POMEROY, CHAIRMAN 20 B. JOHN GARRICK, VICE CHAIRMAN 21 WILLIAM J. HINZE, MEMBER 22 GEORGE M. HORNBERGER, MEMBER 23 24
[~h 25
()
NEAL R. GROSS 00URT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
2 l l
1 ACNW STAFF PRESENT:
2 John T. Larkins, Executive Director
[-
\_/ 3 Michele Kelton, Technical Secretary l 4 Richard,K. Major .
l 5 Howard J. Larson 6 Lynn Deering 7 Andrew C. Campbell i l
i 8 Richard P. Savio j l
9 Michael Markley i l
10 Carol A. Harris 11 Sam Duraiswamy l
12 Theron Brown 13
,O i b 14 ACNW CONSULTANTS PRESENT:
15 Bruce Marsh 16 Michael Ryan 17 Kenneth Foland 18 19 ALSO PRESENT:
20 John Trapp 21 Chuck Connor 22 Brittain Hill 23 Tim Sullivan 24 Kevin Coppersmith (Aj 25 Gene Yogodzinski
! NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
! (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
i 3
1 ALSO PRESENT (cont.)
gg 2 Tim McCartin
~%.
3 Abe Van Luik l 4 Mike Bell 5 Wes Patrick 6
7 8
9 l l
10 11 12 13
/~~'s
'%- 14 15 16 17 18 19 20 21 22 23 24 (O
q,) 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
4 1 A-G-E-N-D-A l l
l g 2 Ac cmda Item Pace j
( i
%/
3 Opening Remarks by ACNW Chairman Pomeroy . . . . . . . 5 4 Discussion of the Status of Igneous ,
l l
5 Activity Related to the Proposed i I
6 Yucca Mountain Repository, W. Hinze . . . . . . . . . . 7 l l
7 NRC Introduction and Overview, J. Trapp . . . . . . . 12 1
8 Geologic Setting and Probability Estimates i l
9 C. Connor . . . . . . . . . . . . . . . . . . . 25 10 Overview and Status of DOE Activities, and Summary 11 of Probabilistic Volcanic Hazard Assessment 12 T. Sullivan . . . . . . . . . . . . . . . . . . 109 l
I 13 K. Coppersmith . . . . . . . . . . . . . . . . . . . 115 i
,m
\' '] 14 Summary of Volcanism Studies Related to PVHA 15 Gene Yogodzinski . . . . . . . . . . . . . . . . . 164 16 Summary of CNWRA Consequence analysis for Volcanic 17 Disruption, B. Hill . . . . . . . . . . . . 193 1
18 NRC Preliminary Performance Assessment Calculations 19 T. McCartin . . . . . . . . . . . . . . . . . . . 213 1
20 Incorporation of Volcanism into TSPA-VA 21 A. Van Luik . . . . . . . . . . . . . . . . . . . 253 l 22 NRC and DOE Agreements from Technical Exchange l 23 J. Trapp . . . . . . . . . . . . . . . . . . . 274
- 24 T. Sullivan . . . . . . . . . . . . . . . 277 ,
x I
(,)
i 25 Roundtable Discussion NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. !
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
5 I
1 P-R-O-C-E-E-D-I-N-G-S g3 2 (8:37 a.m.)
( ,/ l 3 CHAIRMAN POMEROY: The meeting will now come l l
4 to order. l 5 This is the first day of the 91st meeting of .
l l
6 the Advisory Committee on Nuclear Waste. During today's 7 meeting, the committee will discuss the status or studies j l
8 related to igneous activity at the proposed Yucca Mountain l l
9 Repository.
10 Ms. Lynn Deering is behind me there. Standing 11 up is the designated federal official for today's session, l 12 This meeting is being conducted in accordance with the 7s 13 provisions of the Federal Advisory Committee Act.
- / 14 We have received no written statements from l 15 members of the public regarding today's sessions. Should 16 anyone wish to address the committee, please make your 17 wishes known to one of the committee staff.
18 It is requested that each speaker use one of 19 the microphones, identify himself or herself, and speak 20 with sufficient clarity and volume so that he or she can 21 be readily heard.
i 22 Before proceeding with the first agenda item, l
l 23 I would like to cover some very brief items of current 24 interest. The first item of interest is, of course, as
(,/ 25 many of you know, on April 15th the Senate passed the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1 6
1 Murkowski Substitute Amendment to Senate Bill Number 104
,- 2 of the Nuclear Waste Policy of 1997 by a vote of 65 to 34.
i,' '/
3 That vote is two votes shy of the required l l
4 number to overcome, override, a Presidential veto, if that 5 should happen. And on April 10th, Representative Upton 1 6 introduced House Resolution 1270, the House version of the 7 high-level waste legislation. So, that's moving forward.
8 The second item is that Undersecretary of 9 Energy, Tom Grumbly, announced his resignation in late 10 March. Grumbly will join his former boss, Hazei O' Leary, 11 at ICF Kaiser in Fairfax, Virginia, where he will become 12 president of the Federal Programs Group.
13 The third, DOE told the Nuclear Regulatory n
14 Commission that it believes NRC can take over the 15 regulatcry responsibility for certain DOE nuclear 16 facilities at an annual cost of roughly 75 million 17 dollars.
ll 18 In late 1995, NRC estimated regulatory costs 19 to be around 300 million and an increase of work force of 20 1200 to 1400 employees. According to DOE projections, the j l
21 NRC will, at the end of a ten-year transition period, ,
1 22 assume regulatory responsibility for approximate]y 200 DOE j 23 nuclear facilities.
24 Those were all the items of current interest I
/N
( ,)
25 have. Do any of the other members wish to make any !
NEAL R. GROSS COURT REPORTERS AND TRANSCRlBERS l 1323 RHODE ISLAND AVE., N W. l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l l
7 1 opening comments? If not, let's turn to the first item of 2 our agenda. Namely, the discussion of the status of 7-(l 3 igneous activity related to the proposed Yucca Mountain 4 Repository.
5 Dr. Hinze is the lead member for this 6 discussion. And he will chair the sessions throughout the 7 day today. Bill.
8 MEMBER HINZE: Thank you, Paul. It's finally 9 arrived.
10 CHAIRMAN POMEROY: I'm appreciative of that.
11 MEMBER HINZE: To experts or laymen who visit 12 Yucca Mountain, volcanism certainly is one of the more 13 visible of the potential hazards to the repository site.
l\
(_ l 14 And this has been identified by the NRC in their program 15 of the vertical slice where volcanism igneous activities 16 is a KTI.
17 The problem of volcanism and igneous activity 18 in general as a potential disruptive effect is exacerbated 19 by the infancy of the science of the volcanic prediction. l l
20 The resulting uncertainty and the 21 interpretation and the incompleteness of the data sets
)
22 available has made this into one of the more contentious 23 issues of Yucca Mountain and also has led to the 24 consideration of this item in a probabilistic manner.
n
() 25 The potential problems and the prediction of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS )
1323 RHODE ISLAND AVE., N W. l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
8 1 igneous events and the potential serious consequences of
,-) 2 an event has made this one of the most highly studied and
\ l 3 certainly highly debated of the Yucca Mountain project.
4 Several groups have been investigating the 5 volcanism issue. Bruce Crowe, Greg Valentine and their 6 cohorts at Los Alamos National Lab have been studying this 7 problem for nearly two decades and have been advising the 8 DOE on this.
9 They have summarized their work in a final 10 report that has been made available to us. The state of 11 Nevada, Gene Smith, Gene Yogodzinski all have made 12 intensive geological studies and probability studies of 13 the potential for igneous activity at Yucca Mountain, f3 )
+
\~/ 14 The NRC also has been very active and has 15 spert a considerable amount of resources, time and money, 16 in this study of the volcanism issue. The Center for 17 Nuclear Waste Regulatory Analysis and particularly Chuck 18 Connor and Bruce Hill have spent a great deal of time 19 looking at the consequences and also the probability of 20 volcanism.
21 Finally, and most recently, DOE has conducted 22 through Geomatrix and Kevin Coppersmith an expert 23 elicitation on the probability of an igneous event. Their 24 report has recently been completed and is being acted O
's ,/ 25 upon, we understand, by the Nuclear Regulatory Commission NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 2344 433
9 1 staff.
2 Today, we will be hearing from all of these 7-3 parties. And it is appropriate and timely that we do so i 4 because the DOE wishes to close this issue muc and has i 5 terminated the data acquisition phase.
1 l
6 The NRC, we understand, is also considering I l
7 producing a staff technical resolution paper, that may not 8 be the exact wording, on the probability of the volcanism !
l 9 at Yucca Mountain.
10 These studies have led to probabilities that 11 are converging, probabilities of occurrence of volcanism )
l l
12 that seem to be converging. Nonetheless, there are l
13 differences. And one of the things that we wanted to
/"%
~ 14 determine today is what is the significance of these 15 differences to the dose that may be received by the l
l 16 critical group? '
17 The use of consequence models to predict the 18 hazard from volcanism that had been carried out by the 19 Center suggests a very minimal risk to the critical group 20 of the Yucca Mountain Repository. We need to understand 21 the assumptions, the sensitivity, the robustness of these 22 models.
23 In terms of the objectives of the working 24 group of our meeting today, frankly, I don't expect that
(~%
() 25 we will be hearing much in the way of new information.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
10 1 Perhaps some small parts here and there.
-~ 2 What we do hope to accomplish is we hope to V
3 receive a comprehensive overview of the results and what 4 they mean to the bottom line, to the exposure of the 5 potential critical group.
6 What we do hope to learn is the basis for any 7 differences remaining among the cognizant groups that I 8 mentioned, their importance, and what is being done, what 9 can be done to minimize these differences if they are 10 important.
11 I should also point out that the ACNW in its 12 next two meetings in May and July will also be conducting )
l 13 brief sessions on the volcanic activity, the igneous
/._.,s l
14 activity. In May, the committee will be taking up expert 15 elicitation.
16 And the probabilistic volcanic hazard analysis 17 undoubtedly will come up at that time. In July, the 18 committee will be looking at reviewing the performance 19 assessment in the high-level waste and the Nuclear 20 Regulatory Commission.
21 And we understand that the volcanic activity, 22 the igneous activity, will come up at that point. For the 23 presenters, Lynn Deering has prepared an excellent 24 background document for this meeting, which I hope
'O
\m,/ 25 everyone has received.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N3V.
WASHINGTON, D.C. 20005-3701 l (202) 234-4433 (202) 234-4433
11 1 And in that, she has included a long, long 2 list of excellent questions. I'm not going to repeat 7-3 those questions now, but I do want to emphasize a few that l
4 I would like to have us all be thinking about.
5 First, what are the principal causes of the 6 differences in the probability obtained by the different 7 methods? Can these differences be reconciled?
8 What is the significance of these? If a range l
9 of probability is the appropriate result, how will the l 10 range be folded into the performance assessment?
11 Second, dealing with consequences.
12 Consequences is an important element in determining the
! 13 risk from igneous events. How are the DOE and the NRC I
- s
["'T
'w-) 14 investigating the consequences and what are the results?
15 Is the Suzuki model which deals with tephra f
i 16- dispersion appropriate for this study? Are there any l
l l 17 other types of models that can be used to judge the Suzuki l
l 18 model?
19 Third, at what point will the NRC close out 20 the study of igneous events? And what steps will lead 21 them to this?
l 22 And finally, how are indirect effects of 23 igneous activity being studied and the risks determined?
24 I think that if we can answer all four of those questions n
( ,) 25 with a reasonable assurance, whatever that word means, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, O C. 20005-3701 (202) 234-4433
12 1 then we'll be in pretty good shape, i
,_ 2 We have a full schedule. And we're going to (s) 3 be hearing from the four different areas of study today.
4 We will conclude this afternoon with a round table which j 5 ve hope that everyone will be able to bring their l
6 questions so that we can have a good interchange. )
7 That does not mean that we won't be asking l l
8 questions as individual presenters move along. With that, 9 I would like to introduce the consultants that we're l
1 10 pleased to have with us today.
1 11 We have three distinguished geoscientists with 12 us. At the far end is Dr. Bruce Marsh, professor of 13 geosciences at Johns Hopkins; Dr. Mike Ryan, a research j
(
\s # 14 scientist'with the U.S. Geological Survey in Reston; and l
15 Ken Foland, professor of geosciences at Ohio State l l
16 University.
17 We are very pleased to have you. And I'm sure 18 that you will want to feel like an integral part in asking 19 questions and making comments. With that, do any of my 20 colleagues, with or without ties, nave anything that they 21 would like to add before we call upon our first speaker?
22 Okay. With that, we are going to be 23 introduced to this topic by the NRC lead on igneous 24 activity, John Trapp. John, the floor is yours.
(m,) 25 MR. TRAPP: I'm always glad to come here to NEAL R. GROSS
- COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433 t
13 1 these nice, intimate gatherings with a few hundred of my
, 2 friends and go from there.
I
)
3 CHAIRMAN POMEROY: Continue if you feel that 4 way for the next 30 seconds, John.
5 MR. TRAPP: I was asked by Lynn to give a few 6 introductory comments and basically to cover three basic 7 points, kind of give an overview of what to expect for the 8 rest of the day, give a few comments on some of the 9 technical feelings we've got on the PVHA report.
10 And I'll also talk a little bit about the 11 effect of the change in the, I'll call it the, EPA 12 standard and how that's affected our work.
13 Basically, most of what you're going to hear
,r m t
'/
)
14 today is going to be a summary of what was presented 15 during the meeting between DOE, NRC, the technical 16 exchange which took two days. I 17 As such, because of recovering this material l
l 18 in about three-quarters of a day followed by the round j l
19 table, we're going to be going through an awful lot of 20 material awful quick. I hope that most of you have had a 21 chance to read the NRC yearly report, etc.
l 22 A lot of the information that you're going to j l
l 23 hear today is summarized in that report. And I hope '
t l l
24 you've also read the PVHA report from DOE.
f'~s 4
( ,
) 25 In addition to this, you will hear some l
)
l NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
i
l 14 l
1 summary statements about work that's being planned by both f3 2 DOE and NRC, where re're going as far as this whole thing.
! \ l C/ 3 One of the big differences in this meeting from the 4 technical exchange is we will get a chance to hear some of l
5 the views of the state.
< 6 This was not present, and it basically offers l 7 both the political and technical opinion which needs to be
)
8 brought into consideration. We will not be talking about 9 a lot of the generic concerns with expert elicitation or 10 questions, this type of stuff. That's basically going to 11 be handled in your May meeting.
12 MEMBER HINZE: Right.
13 MR. TRAPP: And we will go a little bit into
14 TSPA, but not to a great deal because TSPA again will be 15 covered during your July meeting. So, a lot of these 16 questions actually will be partially covered today.
17 And hopefully, between now and July, we should 18 get most of the questions at least answered to a 19 reasonable extent that were raised by Dr. Hinze.
20 MEMBER HINZE: John, is it appropriate to ask 21 what are the staff's plans? Are you going to get to that 22 in terms of --
23 MR. TRAPP: The staff's plans will be covered 24 in basically the last presentation I'll be making today.
fh '
( ,)
?
25 MEMBER HINZE: All right.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
15 1 MR. TRAPP: So, I'll be going through what f S' 2 we're doing both technically and otherwise.
l
~
3 As you know, the EPA standard has been kind of 4 in flux for several years now. If you take a look at the 5 old standard, it really was very strongly probability-6 based, probability in relation to the accessible 7 environment.
8 And if you take a look at a lot of the 9 probabilities that have been tossed around in igneous 10 activity, you end up with an awful lot which are sitting 11 right at this magic number in the old EPA standard, 10-7 12 Because they sat right on that point and p- 13 because there were different opinions as to whether we're I l
> 14. above and below this point, it meant that we were having a 15 very, very strong focus on our probabilistic aspects.
16 If you take a look at the proposed 17 recommendations for the new standard and some of the I 1
18 things that are in the Senate bill, what we're now talking l
19 about is release, transport, and dose to population. When l l
20 I say population, all my health physicists get all upset. l 21 As such, the focus on probability has been 22 much less. But because before all we had to do was worry i
l l 23 about getting the material on up into the accessible 24 environment, we weren't worried about the transport
,f") l
( ),
25 aspect. j l
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W. l l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433 l j
16 1 And we weren't worried about going through all fs 2 of the heath physics calculations. As such, since we I h
\~~J 3 started getting the feel as to where the standard is 4 going, we have spent quite a bit of time going into the !
5 transport which is part of the Suzuki code, which was 6 mentioned, and trying to calculate how this would affect 7 population in some critical group.
8 In PVHA, I'm only going to bring up a couple 9 major points. We do have an ongoing concern on what is 10 the effect that new information is going to have all the 11 way through.
12 An expert elicitation really is kind o2 a 13 slice in time. And there will be new information. Now, m
( )
\-s' 14 it's very easy to take this new information and analyze it 15 from a mathematical standpoint.
16 There's a question, what also does it do to 17 things like the underlying conceptual model that the 18 people had? This is something that really cannot be 19 analyzed. And it's kind of hard to figure out how these 20 affects would be.
21 From a technical standpoint, our biggest 22 concern really is the zonation models which were placed or 23 used by many of the experts. Trying to basically not just 24 talk about their zonation model, but if you take a look at
(~%
() 25 most of the work that was done by the state, there also NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
17 .
l 1 was a zonation model there.
i
/
s 2 And the point is, by taking these different
\
U 3 types of zonation models, the assumptions that are laid i
4 within these zonation models can really, totally drive the 5 probability obtained.
6 This last statement on here really is kind of 7 self-explanatory except, like I said, we do have a full 8 range including the state numbers. Now, if you take a 9 look at the range that you've got that is in some of the 10 literature, you end up with things down to 10-" which comes 11 out in the PVHA up to some numbers which almost get up to 12 104 13 We consider the PVHA basically as one input.
im.
-- 14 We consider the state's stuff that was published in 15 various technical journals as another input, and we 16 consider the work that was done by the Center as another 17 input. We've got to take a look at all this to come up 18 with the final conclusion.
19 CHAIRMAN POMEROY: John, before you leave 20 that, let me ask a couple of questions. First of all, I
21 with regard to new information, we all, I think, recognize 22 that that's both the generic and the specific issue for j 23 this particular elicitation.
24 Isn't the first thing you need to talk about )
C)
( ,
, 25 or think about whether that information is of any 1 NEAL R. GIUDSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
18 1 importance or significance to the calculation? For fs 2 example, would one additional aeromagnetic anomaly have 3 any effect on this study, for example?
4 How do you propose to treat new information 5 tnat comes up as you get into looking at this? Do you 6 really see it merely as calling attention, calling DOE's 7 attention, to new information and say, " Find out whether 8 this is important?"
9 Or do you intend to actually make some 10 studies? Does the staff intend to make some studies to 11 determine its importance?
12 MR. TRAPP: It's more the latter. What we'd 13 be doing was if we get new information, take a look at it, n.
- 14 try to determine if we think it has some significance.
15 Of course, all of the information would be 16 passed onto DOE, but the emphasis that we'd put on it 17 really would be based on our opinion as to its relevance, 18 yes.
19 CHAIRMAN POMEROY: But you recognize clearly, 20 of course, the problem of getting inside somebody's head 21 to see whether it actually is important to a particular 22 expert or group of experts. ,
1 23 MR. TRAPP: It's almost impossible. And 24 that's really in some ways our concern.
A
(,)
25 CHAIRMAN POMEROY: Right. And I guess the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1 19 1 heart of the question I think is will every expert
,s 2 elicitation run into this same basic problem as far as NRC t
N~)i l
. 3 is concerned? '
4 MR. TRAPP: I'm not going to answer it for the 5 NRC because it's really too broad a question. And if I ;
I 6 did, I'd probably have the managers down here all over the i 7 place.
8 CHAIRMAN POMEROY: Plan to have several l I
9 questions they're going to try and do that. l l
1 10 MR. TRAPP: But, yes, I do see it as a problem )
11 because there are several areas of characterization which 12 basically we felt were not really fully carried out by )
13 DOS. It's mainly the reason why we went into some of the
/ \
\' # 14 geophysics work, some of which has been reported in the 15 last EOS article.
16 With the fact that this characterization 17 probably has left things in a much more uncertain state 18 than we'd like, there may be more of a problem in this one 19 than in some of the others.
20 MEMBER HINZE: Bruce, please.
21 MR. MARSH: In this regard, how do you plan to 22 keep abreast or understand theoretical developments? In 23 other words, you get new information or the existing 24 information may take on a different complexicn based on (h
( ,) 25 more understanding of magnetic transport, magnetic NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISt.AND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
20 1 processes, which is a rapidly developing field.
g-] 2 So, I just wonder have you thought about N- I 3 melding these together and what impact this will have?
4 MR. TRAPP: I think we're blessed by having a 5 very good support staff at the Center. Chuck and Britt 6 have been keeping well up on all the new material. Very 7 honestly, the best answer I can give to it is I depend on 8 those two to take care of it. I'm not sure how else to 9 answer your question.
10 MEMBER HINZE: Paul, did you have some follow-11 ups?
12 CHAIRMAN POMEROY: Yes. John, let's turn for
_ 13 a second to concern with the geologig basis of zones. I'd
like you to help me out a little bit there. What is the 14 15 nature of the concern, that you don't think the geologists 16 knew, or the volcanologists knew, what basis --
17 MR. TRAPP: The simplest form, if you take a 18 look at the PVHA, and this will come up in some of what 19 Chuck will be presenting, and it will all definitely come 20 up in some of the stuff Kevin Coppersmith will be 21 presenting.
22 There is a question as to the nature of the 23 transition from Crater Flat to Yucca Mountain. If you 24 attended the DOE /NRC meeting on structural geology,
/
(3_.) 25 exotectonic models last year, the thought processes that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
21 l
1 came out of here is that Yucca Mountain is part of the 1
I
.,-) 2 Crater Flat basin.
1 3 If you take a look at the DOE synthesis report 4 in geophysics and DOE synthesis report, i believe, Seismic 5 Tectonics is the correct title, both of those come up at 6 the same conclusion that Yucca Mountain is part of the 1 7 Crater Flat basin.
8 If you take a look at the zones that were 1
l 9 drawn within the PVHA report by many of the experts, there 10 was a boundary either hard or soft, etc., that was drawn 11 basically on the topographic difference between the two. l l
12 We see no geologic or geophysical reason to 13 draw this type of boundary. It's basically the reason l
(~N
! )
x/ 14 that Chuck and his models have gone the way they are.
15 So, from a technical side you will see how 16 we're handling it right now. And the concern, very 17 honestly if you take a look at it, is not that the numbers 18 are different, but the meaning of the numbers that are in 19 PVHA versus the numbers that we've got are slightly 20 different.
21 What we're talking about basically for our 22 probability numbers is volcanic eruption and dispersion 23 through the repository. If you take a look at the DOE 24 numbers, they're really talking more secondary effects, (n_,) 25 dikes, etc., and this type of thing.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
i l
22 1 If you get down to a number that would be very i 2 close to our number, it would be something like about 6x10' i /
3 . So, basically, you're talking about maybe an order, 4 order and a half magnitude lower than ours.
5 So, it's a disagreement in the geologic 6 interpretation, but it's really what thic interpretation 7 means to overall risk.
8 CHAIRMAN POMEROY: Certainly. And we're going 9 to get to that bottom line question also. I do want to 10 just get into this a little bit further. Do you feel 11 that, therefore, tl., ten experts on the' PVHA panel did not 12 consider data?
13 I think there's a specific statement in there k s/ 14 someplace that they decided that structural control was 15 not the issue that they were going to consider, 16 particularly in reference to that boundary.
17 MR. TRAPP: No, they considered all the 18 information that they had. Geomatrix did a very good job 19 of giving them the information that was available at that 20 time.
21 But from a technical, both from that and a 22 regulatory perspective, we're going to take a look at all 23 the information, etc., and try to figure out exactly what 24 it means, how we're going to take a look at it from a
()
m 25 regulatory perspective.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
23 1 We are basically being more cautious towards 2 the public by having these types of numbers. From an 7-Nn^ ,/ l 3 overall standpoint, when you get to overall risk, I am not .
1 4 sure that most of it really makes that much difference.
5 We need to find out. But we're talking, no 6 matter how you do it, extremely low probability numbers.
7 And if you're going to have risk, you got this l 1
8 multiplication factor. And there's a limit as to how far 9 you can go.
l 10 One in a million is probably about as far as 11 you can go in probability. It may be higher, but that's 12 reasonable. If you compare that to something that's got 13 uncertainty, you're going to have a hell of a lot more e i
\~/' 14 consequence to make the risk equal.
15 CHAIRMAN POMEROY: Let me ask you one last 16 question, John. It's a stated and avowed purpose of all 17 of these elicitations to try to bring forth the full range 18 of uncertainty that the community would express with 19 regard to any specific issue that the elicitation is 20 addressing.
21 And presumably, the experts were asked to try 22 to do that. Does the 10~' to 108 range that the NRC feels 23 is appropriate express the full range of uncertainty of 24 the community?
) 25 I note, for example, of course that it doesn't NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
24 1 include the 10-', 10-2 , nor in fact does it include the 2 states of higher probabilities.
l, ,\
O' MR. TRAPP:
3 But if you take a look at the 4 number, it basically includes the mr.an value of the PVHA, 5 and if you take a look at the values that come out of the 6 states, especially the last ones which are basically 7 ranging f rom 1x10-7 to I believe --
8 There was a value quoted in the report of 9 8 . 8x10-6, but I think the actual number that they were 10 talking about was 3x10-6, but the best estimate of just 11 over 1x10-7 So, our number basically includes the mean '
12 values.
13 And really when you start putting these type k--) 14 of things into a total system performance assessment, it's 15 the mean value which really drives the whole curve.
16 Therefore, if you start going from the 10-8, 10-7, I think 17 we are covering the mean of all areas that I've talked 18 about.
19 But no, we haven't covered the total range of 20 uncertainty. That would be, like I said, about 10-5 to 21 10' , five orders of magnitude.
22 CHAIP @N POMT"10Y : Yes, I'm sure that, for 23 instance, I know Atomi.: Licensing and Safety Board would 24 want to look at that whole range of uncertainty, would
/^\
U 25 want to at least know about the whole range of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, O C. 20005-3701 (202) 234-4433
25 1 uncertainty. Thank you, John. That's all I have.
rx 2 MEMBER HINZE: I think these first two items t \
U' 3 will be coming back to haunt us as we move throughout the 4 entire day. At least I hope they will. Is that it then, 5 John?
6 MR. TRAPP: That's it for me. And, therefore, 7 with no further ado from Chicago and on his way to Vienna, 8 as you may know, I'll let Chuck cover it.
9 MEMBER HINZE: Chuck, you're going to give us 10 a summary of the work on geological setting and 11 probability, right?
12 MR. CONNOR: Right.
,_s 13 MEMBER HINZE: And there is a handout.
( \
14 MR. CONNOR: Thanks. Is that on?
15 MEMBER HINZE: It's always good to have you 16 here. And we're interested in what you have to say.
17 MR. CONNOR: Okay. As you mentioned, I'll be 18 trying to condense a couple of presentations that a few of 19 you have heard before. I'd like to talk about the 20 geologic setting and probability of volcanism at the Yucca 21 Mountain Repository based on some studies that we've been 22 doing at the Center.
23 And, of course, I'm not going to discuss the 24 entire geologic setting, but just those that are (D
(_) 25 particularly relevant to volcanic hazard assessment and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D.C. 20'X)5-3701 (202) 234-4433
1 26 l l
l 1 also some data that we've been gathering out ther i
- e. 2 the past year or so which is roughly new information s
~} 3 Here's an outline for what I want to go 4 through. I want to talk about the regional structural l l
5 setting of basaltic volcanism near Yucca Mountain. j 6 People have recognized for a very long time 7 that small volume basaltic volcanism structure has a 8 relationship. And it's important to identify that 9 relationship and hopefully try to use that in constructing 10 probabilistic models for the volcanic hazard.
11 Second, I'm going to talk a bit about some 12 recent ground magnetic work that my colleagues at the 13 Center and I have been doing in the area around Yucca
('s-) 14 Mountain. Yucca Mountain is an area not only of basaltic 15 volcanism, it's also an area of a lot of tectonic 16 activity.
17 And there are subsiding basins there which 18 tend to obscure some volcanic features. I think it's 19 important to understand patterns in basaltic volcanic 20 activity between from the miocene through the pliocene and 21 quaternary.
22 In order to do that, we need to rely on some 23 geophysical data in order to identify the locations of 24 buried cones in the alluvium.
(m
(_) 25 And third, I'll summarize some of the geologic NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
l 27 l l
1 factors and go through some of the probability models that l
7- 2 we've been developing to try to get a handle on our l
( ,/
3 particular range, which has already been mentioned.
4 It's a critical step, I think we could all ,
l 5 agree, it's a critical step to understand what we're 6 talking about looking at this volcanic field. The Yucca I 7 Mountain Repository, proposed repository, is located here.
8 This is a rather camplicated map for this 9 presentation. But I just want to point out a few features 10 on it. Here is the quaternary Crater Flat alignment.
l 11 Lathrop Wells volcano is down here, also quaternary 12 volcano.
13 Several aeromagnetic anomalies located in the
'-# 14 Amargosa Valley first identified by Langenheim and her 15 coworkers at the USGS. There are five here. We've if recently done some surveys in here which I delineated 17 three volcanoes at this location.
18 MEMBER HINZE: Are any of t'ose r not shown on 19 the aeromag map?
20 MR. CONNOR: Sorry?
1 el MEMBER HINZE: Which of those are not shown on l 22 the aeromag map?
23 MR. CONNOR: Well, of these anomalies, they 24 are all identified by Langenheim on the aeromag map. I'll
((~~,) 25 go back to it, but this one in particular is a little l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS
- 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
28 1 obscure on the aeromagnetic map because of its relatively 2 low amplitude.
\)'
3 This one, for example, Anomaly B is actually a 4 big volcano. And that covers about 16 square kilometers, 5 pretty thick accumulation of lava flow for this kind of 6 system.
7 So, these are the large volcanoes that we can 8 readily identify in the alluvium out there from the 9 aeromagnetic data. There is pliocene volcanism and some 10 quaternary volcanism up to the Northwest, the funeral 11 formation, pliocene volcanic field, 20 volcanoes or so 12 down here.
13 I think Gene Yogodzinski is going to talk a r
\ >'I 14 lot more about this, the geochemical and isotopic setting 15 of this volcanic field. But he and Gene Smith discuss the 16 Amargosa Valley volcanic isotopic province, which 17 basically encompasses all of this basalt to the east and 18 north of Death Valley.
19 I'm going to zoom in here a bit and take a 20 look at again the location of the proposed repository.
21 This is the area of Crater Flat with the quaternary basalt 22 and so on. And I want to point out that people have tried l
l 23 to identify the system over time in a couple of different 24 ways.
/^;
(_) 25 I think it was in 1989 that Frank Perry and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
29 1 Bruce Crowe defined the Crater Flat volcanic zone which is 2 this north / northwest trending zone which encompasses the
,7, V
3 quaternary pliocene basalt -- in Crater Flat, heading up 4 toward~the quaternary little cones in Black Peak.
5 And that's later been extended to the south to 6 encompass these anomalies. So, the idea is that the 7 volcanism is concentrated in a long, thin zone like this.
l 8 And of course, defining a zone in that way tends to 9 decrease the probability for volcanism at the repository l 10 site.
11 The state of Nevada proposed something called 12 the Area of Most Recent Volcanism, which was extended to 13 include Buckboard Mesa, and thus the repository location p_
l
\) 14 itself. ,
i 15 And Gene Smith and Ho paid a lot of attention j 16 to this kind of prominent northeast trending alignment 17 %ere in quaternary Crater Flat, also the northeast l
18 alignment of Little Black Peak and Hidden Cone and l 19 suggested that this aspect of things was important. l l
20 They proposed a structural zone extending, !
21 very narrow extending from Lathrop Wells up to the 1
22 repository which yields high probabilities of volcanism.
l l l 23 So, between the, I'll just call it the Crater l 24 Flat volcanic zone model in this northeast trending model,
's 25 defining the structural zones gives you a lot of variation NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
30 1 in probability.
7
- 2 I think it's quite important to look at the O 3 geology relative to these zonal models. And these are 4 three cross sections through Bare Mountain. Let me put l 5 this one back up for a second. l 6 These are cross sections basically taken 7 through Bare Mountain and across the proposed repository 8 through Crater Flat. These two are balanced sections put l
9 together at the Center at the USGS.
10 This one is an internally deformable crustal 11 block model put together by DOE. And basically, up here, 12 we've got the Bare Mot:ntain f ault bounding at half -- and
,_ 13 which encompasses Crater Flat and Yucca Mountain and 14 extending east into Jackass Plat.
15 This model is essentially the same except that 16 it sort of bounds the depth of this detachment fault here.
17 In this case, it's about ten kilometers. Down here, it's 18 about five kilometers. There's some debate about that.
19 And then, again, the deformable block model is 20 a little bit simpler. But I think for our purposes today, 21 we can say that these models are similar in that they're 22 relatively high-angle faults close to the surface.
23 At any rate, extending down to some depth, 24 maybe five kilometers, maybe deeper beneath Yucca Mountain
/'%
k ,) 25 and Crater Flat. It's basically an extensional setting, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
31 1 robin or half robin kind of situation.
s 2 Maximum extension must be somewhere over in
! ) -
%^J 3 Crater Flat. And there isn't really much evidence for 4 some prominent structure between Crater Flat and Yucca 5 Mountain, prominent structure.
6 That is, crustal structure, which would 7 prevent magma from rising through Yucca Mountain rather 8 than Crater Flat, something like that. So, that's as far 9 as I want to take that structural model, but we can come 10 back to it if you want.
11 It's also possible to look in -- view of this 12 structural setting. And I'm going to use some of the 13 gravity data put together by the U.S. Geological Survey O.
i 1 l
\/ 14 and others in the Yucca Mountain area, and NTS, the NTS 15 area over the last 40 or 50 years.
16 There's a tremendous amount of gravity data.
17 Maybe the best data set in the world is collected in this l
18 area. This map consists of about 8,000 gravity, 19 individual gravity stations.
1 20 The -- quaternary basalt in Crater Flat is 21 shown here in yellow. The Amargosa Valley anomalies are j 22 just shown by the yellow dots here. And again, the i l
23 repository is shown there in red with gravity varying 24 across the region, of course, due to lateral density
(~N l
(_) 25 variations in the crust.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
32 1 What you can see here is that the Bare 7- 2 Mountain fault is quite a prominent feature on this map.
! a 3 And that volcanism, quaternary volcanism, is tending to 4 , occur east of that fault.
5 There is some debate about where the Bare 6 Mountain fault goes as it trends into, actually out of 7 southern Crater Flat. You can see a relative gravity low 8 extending south of the area.
9 This feature has been called the Amargosa 10 trough. And the pliocene basalts, actually all their ages 11 aren't known. But these aeromagnetic anomalies in 12 Amargosa Valley fall within that trough.
13 On the west side up here, that's bounded by i'3 Y-) 14 something termed the Gravity Fault in the alluvium, and 15 its extent north is somewhat more difficult to sort out 16 what's happening here. And if you recall, the structural 17 cross sections, certainly the west side of the system, was 18 pretty well defined by the Bare Mountain fault.
19 And that series of faults out here make the 20 eastern bcundary a little bit less distinct. This huge 21 anomaly up here is the Timber Mountain Caldera complex.
22 MEMBER HINZE: Have you ever considered -- the l 23 alluvial anomalies?
24 MR. CONNOR: I'm sorry?
r~x
() 25 MEMBER HINZE: Have you ever considered i NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
33 1 stripping off the alluvial anomalies to really get at the s 2 bedrock gravity?
/ \
\'~~~]
3 MR. CONNOR: Yes, I have. And in fact, I had 4 a conversation with Vicky Langenheim about this. And she 5 mentioned that they are doing that, or at one point were 6 doing that on a regional case.
7 Not only for the subset of data I've been 8 using, but for the entire region. So, that's something 9 that would be really worth doing of course. And it's 10 apparently in progress. But I haven't done that myself.
11 MEMBER HINZE: Is that part of the Yucca r
12 Mountain project of the U.S. Geological Survey?
l 13 MR. CONNOR: It's not clear to me because my )
7\
\ )
x' 14 understanding is they're basically not funded to do that l 15 work right now. So, I don't know if she and her 16 colleagues are doing that on their own somehow or what.
17 But that was something she planned to do. And 18 she actually published a paper about that and the high-19 level management waste meeting notes which was sort of 20 getting started on that.
21 MEMBER HINZE: Well, let me ask you. Do you 22 feel that you have sufficient amount of data on the 23 alluvium to prepare a strip map?
24 MR. CONNOR: I think so. There's a lot of p.\
(_) 25 well data, of course, in Crater Flat. And the gravity NEAL R. GROSS COURT REPORTERS AND TRANt4fdBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
34 1 data, of course, are great. So, I don't think that's
,-s. 2 going to be too bad a problem to do that.
i /
3 MEMBER HINZE: It certainly would make your 4 arguments I think a lot stronger because you have these 5 superimposed effects -- surface is killing you or 6 accentuating.
7 Before you take that off, let me ask you 8 another question. The first of your conversation here is !
l 9 towards the faults and the volcanic centers. There is, to 10 my knowledge, only one instance in this area where there 11 is any igneous activity in a fault. l l
1 12 That's the Solitario Canyon up there at the 13 north end of Yucca Mountain. I think that there are i
(')
E/ 14 structural geologists that will tell you that more 15 important than faults to this structural control on 16 volcanism are the joints.
17 Can you justify the concern here about the 18 steep gradients and about the faults that they represent i 19 in this discussion of the geological setting?
20 MR. CONNOR: Yes. I think that joints and 21 faults are both important in transporting the small volume 22 basaltic magma in the shallow crust. And as some of you 23 recall, some of the work we've done involves the dilation 24 tendency of faults through joints to a particular stress
/~N
() 25 orientation.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISI.AND AVE., N W.
(202) ?34-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
35 l
1 So, in the Yucca Mountain area, minimum l
< x, 2 horizontal compressional stress oriented about like that,
( )
x 'e 3 so I would imagine the faults or joints of that 4 orientation would have a slightly higher tendency to I 5 dilate than faults of other orientations.
6 MEMBER HINZE: If they're vertical. Right.
7 MR. CONNOR: Yes, exactly, if they're l 8 vertical. So, that's an important consideration, the 9 orientation of these faults.
10 Obviously, maybe not obviously, but I think 11 there's less of a tendency for a fault like the Railroad 12 Valley fault which is an orientation like that to dilate
,, 13 and host magnetism perhaps than faults of other 14 orientation.
15 MEMBER HINZE: Is there a decent map, is there 16 an appropriate map, of the tectonic joints of this area?
17 MR. CONNOR: Not that I know of. I don't 18 recall anybody -- maybe I'm sure on Yucca Mountain itself 19 there is a map, joint sets in the Yucca Mountain block 20 itself.
21 MEMBER HINZE: Well, I was wondering if in 22 considering the dilational aspects of the faults which 23 trivets through and around Yucca Mountain, did you 24 consider the joints at all?
t
/^N 1
(_,/ 25 MR. CONNOR: No.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
36 1 MEMBER HINZE: Should they be considered?
- 2 MR. CONNOR: It's possible, but I think l t )
3 what -- for instance, Paul Delaney has done a lot of work 4 on joint sets and dike intrusion on joint sets. And this 1
1 5 is something that's happening on a scale of hundreds of 6 meters.
7 And once these probability models we're 8 dealing with are dealing with scales of kilometers, I 9 hesitate to say that joint patterns, say in the Paintbrush l 10 tuff, are going to have an important influence on whether 11 volcanism is going to occur here or over here.
12 MEMBER HINZE: But if the action is in the 13 hundred meter level range, if that's what really is n
's >) 14 controlling it, maybe it has to be broken down into that 15 kind of a scale in order to solve the problem.
1 16 MR. CONNOR: Yes. I mean, you know --
17 Let me back off and say that I agree with you 18 that joint or fracture orientation is important with 19 respect to the regional stress field. There are some good 20 examples of that.
21 We were out in the San Rafael volcanic field 22 with Paul Delaney a few weeks ago and saw great examples 23 of that. I am not certain how to incorporate that into a 24 probability model at this point. But we do to a certain
/~%
(_,). 25 extent, looking at this dilation tendency on the faults NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
37 1 themselves.
,3 2 Yes?
k 3 MR. MARSH: Yes, Chuck in this context of what 4 Bill is asking about, the joint density across the map may 5 be very interesting to look at.
6 MR. CONNOR: Right. I'd like to get into this 7 a little bit later when I am showing some specific 8 probability models, but a serious problem in incorporating 9 structural data into the hazard analysis is alluvial 10 cover.
11 If I made a probability model based on mapped 12 joint density all the volcanism is going to occur right 13 here because that is where the mapped joints are. There
!3
\~- 14 is this big alluvial basin over here.
15 MR. MARSH: But the difference is that you do 16 hav> a structural map and you know the basic units. In 17 over words, pre-Cambrian sediments, et cetera that run l
18 through the area. l 19 So, you know basically the jointing 20 characteristics in these from the bedrock and the areas I 21 with outcroppings. So, it actually may give you by having 22 some joint density map -- as you are talking about it now, l
23 you are assuming sort of an equal uniformity in texture 24 and joints probability across here. But it may not be
() 25 true. And that may be specific to the various rock types.
NEAL R. GROSS COURT REPORTERS A.ND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
38 1 So, in other words, since you do have 2 outcroppings, you do have a decent structural model, you
(, w)
U 3 could make a first attempt at this probably.
4 MR. CONNOR: Yes, especially comparing 5 something like Bear Mountain to the Yucca Mountain. l l
l 6 MR. FOLAND: Chuck, just one more point before ;
i 7 you move on from the gravity.
8 MR. CONNOR: Sure.
9 MR. FOLAND: What is the scale? What is the 10 dimension of the smallest young basaltic feature that 11 shows up in terms of gravity anomaly.
12 MR. CONNOR: In terms of gravity, the gravity 13 does a very poor job of identifying basalts and I wouldn't
(~N
\
~ ') 14 use it for that task.
15 There are 8,000 data points over here which is l 16 a lot of back-breaking work but that is not going to begin l 17 to resolve any of these kinds of things. ,
l 18 And I am not using the gravity data to help 19 identify basalt.
20 MR. FOLAND: Okay, in terms of magnetic, then.
21 MR. CONNOR: I will get to the magnetics in a 22 second. But the magnetics, it turns out, is pretty good 23 at identifying the basalts in the alluvian core based on ,
1 l
24 the magnetization contrast.
p)
(, 25 MEMBER HINZE: I am not going to let you get NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISt.AND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
i
39 1 that off yet.
7y 2 For the benefit of the Committee and the N 3 consultants, while you have that map up there, could you 4 tell us, if I am not asking too much, what we have in 5 terms of mantle velocity anomalies or any indications? I I i
l 6 think this is an appropriate topic to make certain that we i
7 are all together.
8 MR. CONNOR: Sure. That is a good point.
9 I guess it was several years ago, at the 10 suggestion of Linda Kovach who was a project manager at 11 the time in research, suggested that we take a close look 12 at the tomographic data for the Yucca Mountain region.
13 Also, Evans and Smith had published a paper about
/ \
14 tomographic data in the Yucca Mountain region.
I 15 I wrote a report with Chris Sanders who was !
I 16 then at ASU on tomographic anomalies and their use in this l
17 kind of problem.
18 Evans and Smith basically identified several l
19 scales of tomographic anomalies in the area, though they 20 admitted that they weren't using the best data sets; they 21 used what they could.
22 They had a relatively large, deep anomalies 23 which I think would cover most of this map area, cutting 24 out a little bit of the NTS over here. That was one slow
,9
(_) 25 zone in their model.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
40 1 They tried to identify an anomaly. They
~x 2 thought they identified an anomaly, if I recall correctly,
, I s_/
3 at sort of a mid-crustal level beneath the crater flat 4 area and I believe that has been disputed to a certain 5 extent based on some data coming out of the University of 6 Nevada at Reno.
7 They also pointed out some features of the 8 small scale structures beneath Yucca Mountain at the edge 9 of Crater Flat and pointed out that Crater Flat 10 structurally tended to extend beneath the repository based 11 ont he shallow tomographic data.
12 Recently, there has been some work presented.
,, 13 For instance, at the last GSA meeting, there was a paper
\
' '/ 14 presented but I canno" remember the author's name 15 unfortunately, from the University of Nevada at Reno. By 16 Glen Biozzi.
17 He had more data of course and was looking at 18 several tomographic slices, level slices through the area.
19 What I thought was interesting, and I asked 20 him a couple of questions at his talk, what I thought was 21 interesting was that he was identifying a fairly broad, 22 what I would call north-turning but it was somewhat 23 diffuse, zone at 40 to 60 kilometers. So, a slow velocity 24 zone at 40 to 60 kilometers extending across this region.
rN k ,)
m 25 He felt that the data weren't sufficient to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 1
41 1 completely resolve that anomaly; specifically, he was 2 worried about slo'o velocities creeping up into that layer j 7 s s a V
3 depth in his model. I 4 But I thought that was an interesting point i 5 because based on Fran Perry's work that is the depth of 6 last equilibrium for these basalts is 40 to 60 kilometers.
7 So, - you have a slow velocity zone that is kind of nice.
8 On the other hand, I would say that zone was 9 broad compared to this map. It was hard to do that 10 comparison exactly and I am not going to do it here, but 11 it basically covered this zone.
12 MEMBER HINZE: But he had a lot of data, a ,
13 good coverage. 1
(~~\ ,
14 MR. CONNOR: Yes. He felt that the anomaly l
15 identified by Evans and Smith at mid-crustal levels for )
l I
16 example, was probably a near-surface effect. I carried l 17 that away for his discussion, also. l 18 When I questioned him about it he was fairly 19 tentative about this 40 to 60 kilometers thing because he 20 thought some of the slow velocity zones depth were leaking 21 up, and I am not an expert on that data processing 22 technique. But that was something that he was running 23 into.
24 I want to point out one thing about the 7s
(_,.) 25 tomographic and that is there has been some work done, for NEAL R. GRO3S COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
i 42 1 example in the East African rift by a German group who s 2 identified some really nice little velocity zones i \
~
3 associated with active basaltic volcanic fields.
4 There has been some slow velocity zones 5 identified beneath, say, the San Francisco volcanic field, 1
6 But I am not convinced that the kinds of 7 volumes of partial melt that generate this kind of !
8 volcanism are necessarily going to correlate well with a )
1 9 seismic tomographic anomaly.
10 MR. RYAN: I was wondering if you recollect 11 what the difference was in velocity at 40 to 60 12 kilometers.
13 MR. CONNOR: I am not going to tell you, I
( )
L>' 14 can't remember.
15 MEMBER HINZE: I think the velocities are very 16 poorly constrained. It is just a time delay of .2 to .25 I l
17 seconds. l 18 MR. CONNOR: One per cent comes to mind but l 19 don't hold me to it. It was small compared to what you l
20 see beneath Caldera or something.
21 Okay. I just want to quickly go through. I 22 think we have beat this to death a little bit but, this is 23 the horizontal gravity gradient and what I have shown here l
24 is the amplitude of the horizontal gravity gradie High l l
r~%
{ ,)
, 25 amplitude gradients are dark here and the light areas are j NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISt.AND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 734-4433 I
43 1 low gradients.
,3 2 I have also plotted on here the outcrop extent
( ') 3 of Miocene basalt. One of the things that intrigued me 4 about the gravity data and the structures asscciated with 5 these data is that a lot of the basalt does occur near 6 relatively high gravity gradient areas.
7 For example, at Crater Flat we have this 8 Miocene basalt here up against the Bear Mountain fault. A 9 normal polarized magnetic anomaly that I will talk a 10 little bit more about in a minute.
11 The Little Cones are up here against the 12 fault. One of the things that I have speculated about is 13 that if the dip of the Bear Moantain fault, the shallowing o I
's) 14 northward, that may explain why these volcanoes, if they 15 are breaking out of that fault zone at a given depth, it 16 would explain the displacement of these volcanoes away 17 from that fault zone.
18 In the Miocene there is also good correlation 19 with gravity gradients with, for instance, the Beatty 20 basalts up along Fluorospan Canyon fault. There is a 21 pretty good correlation, I think, between these large 22 crustal features revealed by gravity data and some of the 23 basaltic volcanism.
24 Of course not in every case though. For
/
r~s (j 1 25 instance, you have Miocene volcanoes out here away from a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
44 1 steep gravity gradient a2. e a . So, it is a geophysical data rx 2 set that I think should be considered.
)
L/
3 MEMBER HINZE: Any questions regarding this?
4 If not, then I would like to ask a question.
5 What is the significance of the pink area that 1
l 6 extends south from the proposed repocitory and then shifts 7 over towards Lathrop Wells?
l 8 MR. CONNOR: This area? That is a relatively 9 high gravity gradient area and down here it correlates l
10 roughly with the Stagecoach Fault. You can see that it l
11 extends up through here. l l
12 Well, it is a slightly higher gravity gradient l l
13 area and I think it fits rather well with the structural
' ')
14 edge of the eastern part of the basin, being a little less 15 distinct than the west edge, but extending beneath Yucca 16 Mountain and the site of the proposed repository.
17 I guess some of the structural geologists have 18 called it the Bow Ridge Fault, which would be the eastern 19 extent of the basin. I am sure there is disagreement 20 about that, if anyone wants to comment about that.
21 But there is a high gravity gradient there.
22 And that, to a certain extent, shows up in some of the 23 aeromagnetic data that is on the deconvolution of that 24 data that was done by DOE.
r~s k ,) 25 MR. AARSE: Is this kind of information NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
45 1 entered into the probability calculation?
gx 2 MR. CONNOR: Well, in a minute I will show you
! )
q,i 3 some attempts that I have made to do that.
4 MEMBER HINZE: If we could strip off the 5 alluvium in Crater Flat, would the density of faults be 6 approximately equivalent to that of Yucca Mountain arda?
7 MR. CONNOR: That would be my guess. I think 8 that would be approximately equivalent to Yucca Mountain.
9 MEMBER HINZE: There has been an attempt to 10 map some of those, not just by seismic, which raises all 11 kinds of hackles.
12 But in terms of the magnetic data, you have i 13 looked at the magnetic data of this area probably more 1 4
' 14 than anyone else.
15 Does the magnetic data support in any way, the 16 occurrence of faults in Crater Flat?
17 iS . CONWOR: Sure. They're good.
18 MEMBER HINZE: How about that for a straight 19 line?
20 CHAIRMAN POMEROY: Awfully good, awfully good.
21 MEMBER HINZE: It does support it though?
22 MR. CONNOR: Yes.
23 CHAIRMAN POMEROY: Chuck, when you use the 24 terms, and I am trying to quote exactly, "a pretty good
() 25 correlation, I think, " --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
46 1 LAUGHTER f-s 2 MR. CONNOR: You are going to nail me down
(
3 here, Paul. Okay.
4 CHAIRMAN POMEROY: What does that imply to you 5 and is that a conclusion that say all volcanologist would 6 derive from that kind of a map?
7 LAUGHTER 8 MR. CONNOR: There are --
9 CHAIRMAN POMEROY: Answer the first part.
10 MR. CONNOR: Well, a pretty good correlation -
11 - I'll drop the 'I think'. It is a pretty good 12 correlation. It means that some of the volcanoes lie 13 along these faults and they don't in every case.
14 So, the data should be considered but it is 15 not the silver bullet. We can't say that all volcanism 16 will occur on these high gradient areas.
17 MR. MARSH: But you could quantify that by 18 distance correlations from gradient, et cetera.
19 MR. CONNOR: Right. And you can quantify that 20 and I could talk about that a little bit more in a minute.
21 But basically, if you take these data into 22 account when you put the probability models together, the 23 locations of past volcanic events are better predicted by 24 that model than if you don't consider the locations of the yy
() 25 faults. Okay?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 47 l 1 MR. RYAN: Is there any reason to think that l
gS 2 in the alluvian covered area that the faults would differ i
\'~')
3 in density significantly or in orientation from their more 4 northerly neighbors?
I 5 MR. CONNOR: I think I would speculate that 6 they don't change density or orientation locally. In l 7 other words, I feel pretty comfortable over here and 8 actually the aeromagnetic data over here support the idea 9 that the faults are north / south trending and seem to have ;
1 10 a comparable density.
11 I think it becomes a lot more speculative, for 12 instance, down here.
13 I would like to talk about some ground
+
\Y 14 geophysical data that we have collected at the center in 15 May, July and August of last year. And I have a few 16 reprints available from an EOS article related to these, 17 if you are interested. 1 18 I want to talk about a ground magnetic survey 19 we did at Northern Cone, that is the closest volcano to !
23 the site. And Southern Crater Flats in the Little Cones 21 area and close to a magnetic anomaly identified by the DOE 22 and the USGS. And then a survey we did over at Amargosa 1
23 Valley, Anomaly A, also identified by the USGS.
24 This is a map of the area around Northern ;
y) 25 Cone. I kind of stretched this image a little bit because l
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISt.AND AVE., N.W. 1 (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 I
J
48 l
1 the amplitudes around Northern Cone itself are quite l
f3 2 large. So, this is the extent of Northern Cone.
i l
- O 3 Again, I changed scales on you; this is two 4 kile. meters just to reference you and we are about 8 5 kilometers west of the center of the repository block to 6 Northern Cone. i 1
1 7 I made this map because we are interested in !
I 8 the relationship between structure and the basaltic 9 volcanism near Northern Cone. l 1
I 10 As you can see some of these prominent faults 11 coming through the area and extend on into the map area in 12 the northern part; these are taken from a Frizzel and 13 Shulterz map.
,TT 4
-s# 14 To make a long story short, we basically 15 identified these gradients across here, I believe, are l l
16 associated with normal faults, extending through Northern 17 Cone and Northern Cone is a good correlation with the 18 north / south trending fault.
19 So, I would say in a shallow crust, this 20 basalt exploited a structure on its way to the surface and 21 that structure happened to be north / south trending which 22 is a good, high dilation tendency orientation in this 23 area.
24 I guess a second point is that there is no iO
( ,) 25 evidence from the ground magnetic data that there is a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
49 1 shallow northeast trending structure along the alignment i
n
( )
2 itself. Remember there is the quaternary Crater Flat v
3 alignment at least in the shallow crust doesn't come to 4 light, even though the north / south trending structures are 5 quite clear.
6 So, one possibility is that there is a master 7 dike at some depth like 10 kilometers and these things are 8 exploiting structures above that master dike of different 9 orientations. Another possibility is that there is no 10 master dike; these are individual events.
11 MR. MARSH: What about these lines going off 12 in the northeast direction?
13 MR. CONNOR: These guys over here?
/ \
14 MR. MARSH: Yes.
15 MR. CONNOR: What I see on this map is 16 basically a fault basically defined by these inflection 17 points. But I am not sure exactly what you are referring 18 to.
19 MR. MARSH: Those little bulls eyes.
20 MR. CONNOR: Oh, these guys?
21 MR. MARSH: Yes.
22 MR. CONNOR: It is pretty noisy alluvium up ,
1 23 there and it could just be a tuff block, a larger tuff !
l 24 block sitting at the surface or something like that. I l
I
[()T 25 wouldn't attempt to interpret those beyond that.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON. D C. 20005 3701 (202) 234-4433 j
50 1 Okay. I am going to take you down to just rx 2 south of Lathrop Wells, Amargosa Valley Anomaly A.
( )
v 3 We got interested in mapping Amargosa Valley 4 Anomaly A because it is very close to Lathrop Wells which 5 is the youngest volcano in the area. This map is about l 6 three or four kilometers south of Lathrop Wells itself and l l
I 7 is an area identified by Langenheim as a possible basalt 8 basin aeromagnetic data, but it was a very complicated 9 anomaly, so we decided to take a close look at that.
I 10 We mapped this over a three day period and 11 there are about 33,000 data points on here on something l
12 about 60 kilometers of survey line. You can see here that l 13 we actually identified three prominent reversely
'- l 14 magnetized bodies.
15 I got nice positive on the north side and 16 large negatives and interpret those to be shallow, l
17 reversely magnetized rocks well-explained by a plate-like 18 model, i.e., small lava flows or edifices.
19 I think the important thing about this map is l
l 20 it has this nice northeast trend to it. This distance is l
21 about 4.5 kilometers.
22 It looks like Amargosa Valley Anomaly A really 23 consists of three very volcanic structures. That fits in 24 with the idea that northeast trending alignments are
,q l
g ,/ 25 important and ought to be considered in probability NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
51 1 models.
2 MR. MARSH: What kind of surface here?
b,eg 3 MR. CONNOR: It is a flat alluvial surface.
l 4 MR. RYAN: Pardon me.
5 MR. MARSH: How deep is it?
6 MR. CONNOR: Two hundred meters, plus or minus 7 50, something like that. They have magnetizations of I l
8 about one to three ampmeter is what I am using to derive l i
9 that depth.
10 MR. RYAN: If 200 meters is the depth to the 11 top, what about the depth to the overall?
12 MR. CONNOR: I am modeling them as thin l 13 blocks. i l
14 MR. RYAN: Okay. And what is the sense of 15 crustal rotation in this part of the study area, clockwise 16 or counter-clockwise, if it could be reckoned?
17 MR. CONNOR: People have argued for some 18 clockwise rotation.
19 MR. RYAN: This pattern is consistent with 20 that.
21 MR. CONNOR: Yes.
22 CHAIRMAN POMEROY: Just so I understand that, 23 Chuck. I read it but --
24 MR. CONNOR: Excuse me.
i s 25 CHAIRMAN POMEROY: The total extent of the NEAL R. GROSS COURT REPORTERS AND TAANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
52 1 three anomalies is something in the order of 1,000 meters?
,- 2 MR. CONNOR: This is about 1,000 meters here.
3 It is about 4.5 kilometers.
4 CHAIRMAN POMEROY: Okay, so 4.5 total?
5 MR. CONNOR: Yes. One of the anomalies is 6 about 1,000 meters across.
7 MEMBER HINZE: The dilational tendency when 8 those volcanoes were produced was in what direction?
9 MR. CONNOR: Well, do you know how old these 10 are? I don't. They are at least 7,000 years old, right?
11 And if you correlate them with aeromagnetic anomaly B, 12 then'they are on the order of 3.5 million years old.
13 So, if they are indeed Pliocene, then it is w- 14 essentially the same orientation, is my understanding.
15 MR. MARSH: Well, Bill, aren't the positive in 16 a little peculiar position? Are you getting at the fact 17 that they may have been rotated from where they were 18 placed originally?
19 -
MEMBER HINZE: They're much more off to the 20 northwest than you would anticipate.
21 MR. CONNOR: No, I actually disagree with 22 that. I think the positive wraps around like that.
23 MR. RYAN: To get at Bill's question, the 24 dilational tendency could actually be a function of depth 7N
( ,
) 25 and three dimensions. One could make a case for a north-NEAL R. GROSS COURT REPOP.TCRs AND TRANSCRIBERS 1323 R'tODE ISLAND AVE., N W. ,
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 I l
l
53 1 northeastward dilation at the depth at which these
,3 2 anomalies express themselves. And perhaps a north-south l '
~
3 or north-northeast, south-southwest dilation at very, very l 4 shallow levels.
5 MR. CONNOR: Well, these are very shallow. i l
6 MR. RYAN: Even more shallow than the top of 7 the anomaly?
8 MR. CONNOR: Well, yeah. I don't know Bill; I 9 don't really think there is much evidence of rotation l 10 there. l l
11 MEMBER HINZE: Well, you would have to look at 12 the paleomags before you could sure, j i
13 MR. CONNOR: Sure, yeah. f
/N, l l 14 MEMBER HINZE: But it does seem to be tilted. )
I 15 We could argue.
16 MR. CONNOR: Well, if you look at the map, for 17 instance, what I did was I compared this with the 18 anomalies that I saw at Red Cone, ror example. And as you 19 know the positive associated with Ped Cone is sort of 20 messy and it is out there somewhere; I wouldn't push it 21 too hard.
22 The third area that we mapped through May and 23 July last year is this area around Little Cones. The area 24 around Little Cones are two volcanoes located in Southern
, -~s
( ,) 25 Crater Flat.
NEAL R. GROSS COUP r REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
I F1 1 If you have been there, you probably noticed ,
l
,S
- 2 that these are really small-looking features. It turns Nj 3 out that because Southern Crater Flat is actively 1
4 subsiding, Little Cones lava flow have been buried since i i
5 they erupted about a million years ago. This lava flow is 6 revealed extending about two kilometers south of Little 1
7 Cones, beneath the alluvial pavement, it is at a depth l 1
8 that varies from about 15 to 20 meters beneath the 9 surface, which John Stematikos used to compare with other I
10 estimates of subsidence in the area.
11 There are also some other features on this 1
1 12 map. This is a normal polarity anomaly located south of 1 l
13 Little Cones and we interpret that to be a salt body l
/ 's t
\2 14 buried at ta depth similar to those at Amargosa Desert on 15 the order of 150 to 200 meters.
16 I think it io important to realize that that 17 is a normal polarity anomaly which means that although we 18 don't know it's age it is certainly a different age that 19 Little Cones basalts and others in the area which means 20 that there has been volcanism through time in this area 21 and possible along the Crater Flat alignment.
22 Finally, there is a third volcanic feature 23 here. This is what we interpret to be a shallow dike 24 extending north-northwest of a outcrop of Miocene volcanic n
(_,I 25 basalt at the very southern end of the map.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 23&4433
55 1 Although it is quite short, only about 500 g-wg 2 meters long, it is parallel to the trend or suspected Gl 3 trend of the Bear Mountain fault in this area.
4 So, that is what we gleaned from those data.
5 MEMBER HINZE: Were you able to get a dip on 6 that?
7 MR. 70NNOR: I haven't actually modeled it, I 8 have to admit.
9 MEMBER HINZF- Let me ask you a question. Are 10 you suggesting that if we went and did the high-density 11 survey on the ground throughout Crater Flat and Amargosa 12 Valley, that we would end up with a number of hits that 13 would in some way change the probability? Is that what-
- 14 you are suggesting?
15 MR. CONNOR: First, I wouldn't wieh it on 16 anyone to make a map of that kind; it's a lot of work.
17 But I guess my answer to that is, based on 18 these and similar data, what I think is, is that there are 19 something like 10 to 20 aeromagnetic anomalies that bear 20 further investigaticn. And this is a very efficient way 21 to do it.
22 So, if I had my druthers, I don't see any 23 reason not to make those maps; it seems like a good thing e4 to do at this point. You can do the calculation yourself, j i
rN
(_) 25 but I imagine that would af f ect probabi. a ty, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. l (202) 234-4433 WASHINGTON, D.C. 200050701 (202) 234-4433
56 1 But I think there is another aspect which John s 2 Trapp alluded to and that is what we really need to be i s L) 3 careful about here is that we understand the process. The 4 geology process is controlling volcanism in the Yveca i
5 Mountain area.
l 6 It is not completely understood, i t. is not l l
7 completely understood anywhere in the Western Greet Basin.
1 8 We have an opportunity to learn more about the structure 9 of volcano interaction here, using these relatively new !
l 10 techniques; this very rapid data collection kind of 11 technique.
12 I think it is worthwhile exploiting that, but i 1
13 not in the whole region. I would instead, choose these 10 I
,a l
(' -) 14 to 20 anomalies that are pretty clear from perusal of the 15 magnetic map of being potential targets.
16 MEMBER HINZE: Have you conducted surveys of 17 any of those target areas?
18 MR. CONNOR: We conducted surveys about a week 19 and a half ago at one of those areas and that data set is 20 pretty preliminary right now.
21 MEMBER HINZE: What does that mean? It is not 22 going to change, is it?
23 MR. CONNOR: Well, we are in a formal meeting 24 and I hate to present data that I collected a week ago l A
(_) 25 that hasn't been reviewed.
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS !
1323 RHODE ISLAND AVE., N W, I (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
57 1 MEMBER HINZE: Let me ask this question. Is 2 it worthwhile looking at these anomalies on the basis of p-V 3 the most recent results?
4 MR. CONNOR: Absolutely.
5 MEMBER HINZE: Okay, that answered my 6 question. l l
7 CHAIRMAN POMEROY: Well let me -- can I pursue 8 that? ,
l 9 MEMBER HINZE: Sure, j l
10 CHAIRMAN POMEROY: Why -- Certainly it is l 11 important, everybody would agree, to understand as much of l 12 the geologic process as you can. Yet in a regulatory 13 environment you are probably never going to have the funds
/ g r
\> 14 and the time to do that, I suspect.
15 So, my question would be, we have talked about 16 three different volcanic areas here. Is there something 17 in any one of these that the more detailed aeromagnetic 18 mapping has provided to you --
19 MR. CONNOR: These are ground magnetic maps.
20 CHAIRMAN POMEROY: Sorry.
21 MEMBER HINZE: Okay.
22 CHAIRMAN POMEROY: Is there information 23 contained here that, in your opinion, is not contained 24 within the PVHA uncertainty?
tO
(_,) 25 In other words, the uncertainty that people NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
58 1 have given in terms of occurrence and the numbers, it 2 would certainly not, perhaps, be the mean number that they 7-~
i NJ 3 used but is there an importance to this?
4 MR. CONNOR: You have two questions there.
5 One is there information contained in here that is not in 6 PVHA and the answer is of course, yes, they didn't have 7 all this information. There are three anomalies here. No l
8 one in the PVHA treated those as three volcanoes.
9 Now, the question is, does that make any 10 difference in the long run and I will reiterate.
11 This teacheL us about process and just to give 12 you sort of an indication of that, when this came out in 13 EOS, George Walker who is on the PVHA panel wrote me a O
\s/ 14 letter and said that this is very important because it 15 demonstrates that the northeast trend is very critical in 16 the model.
17 Now, he didn't say that he changed his model :
1 1
18 or that his model would be any different or his answer !
)
19 would be different. But it certainly was an indication 20 that these data are helping us close in on the geologic 21 process. Because I believe that, ultimately, it is the 22 geologic processes that are going to drive up our comfort 1
- I 23 level in deciding whether to put this thing at Yucca j i
24 Mountain or not or whether volcanic hazards are important
(~ % !
i,. 25 or not.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 i
59 1 Now, does it change a probability number?
,r-) 2 Well, of course that depends on the model that you are l
'\ ,1 l 3 using. I would think that if it turns out that these ten l l
l 4 anomalies are actually Pliocene volcanoes or can be l 1
5 interpreted to be Pliocene volcanoes based on this then 1
6 that would have a small impact on probability, on the 1
7 order of maybe a factor of two or three, would be my guess 8 based on some very preliminary things that I have done.
9 I would be surprised if it changed more than 10 an order of magnitude. I am speculating about that.
11 Whether that is ultimately important in the 12 context of PA or total system performance is not a problem 13 that I have addressed at this point in time. My hope is (3
\' '/ 14 that we have the opportunity to better understand the 15 processes before having to reach that decision.
16 But of course, that is just my perspective.
17 MR. FOLAND: Chuck, you mentioned a few 18 minutes ago, rapid data collection. Could you 19 characterize how much time is involved in doing one of 20 these surveys?
21 MR. CONNOR: Yes. This particular survey we 22 did in three days. Basically what we use is a cesium 23 vapor magnetometer manufactured by Geomatrix Geophysical 24 Instrument Company. That is an adjustable sampling rate, f3 (j 25 but we tend to sample every two seconds or four seconds.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
1 60 1 We walk across the ground and if you are rx 2 moving at a meter and a half per second, that gives you an I 4 V
3 idea of the amount of ground that we cover.
4 Interfaced with that is a differential GPS.
5 The differential GPS has a resolution of 20 centimeters, 6 something ridiculous like that. The survey accuracy due 7 to time updates and stuff like that might be on the order 8 of two meters. Of course, in high gradient areas, that is l 1
9 10 or 20 nanotussels and in low gradient areas like this 10 one that is a couple of nanotussels. l l
11 I guess we did that in three days, something 12 like 60 kilometers to traverse across there.
,s 13 Because we get feed back very quickly or
/ %
\ )
14 actually while we are collecting the data in terms of 15 graphical output and that sort of thing, we tend to survey 16 these areas more densely. You find out that there is 17 nothing going on out there, it is a long way to walk 18 through the sand, so we tend not to collect data as 19 densely as we do anomalies we can identify.
20 A good example of that is this survey because 21 we knew that we wanted to understand the extent of this 22 lava flow but we didn't want to map all the high frequency 23 variation in this lava flow. So, there is lava here but 24 you can almost begin to pick out some tracks here but we r'%
(_) 25 didn't want to map this noise on top of the flow.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
61 1 We also changed our survey design when we
,s 2 identified this anomaly. We focused in on that anomaly l (x- )
3 and mapped that much more than some other areas. Okay?
4 So, it comes out to be something like 30 to 40 5 thousand data points. They are all drift corrected and ,
6 the IGRFs removed and that is all the processing that has 7 been done to those maps.
8 MEMBER HINZE: Bear with me one moment. I 9 would like to follow up on Paul's question.
10 It is my understanding from listening to some 11 of the experts involved in the PVRA that they were driven 12 away from Yucca Mountain, I think that is the proper term, 13 by the recurrence of volcanoes in the southerly portion of
(~'\
- 14 Crater Flat.
15 Finding another few volcanic events down in 16 that area will not have a profound affect on what is 17 happening in terms of probability up in the Yucca Mountain 18 area.
19 As a result of that, it seems to me that the 20 critical anomalies to investigate are any that are in the 21 proximity of Yucca Mountain. Are there any of these 10 to 22 20 subtle anomalies that you have observed in the Yucca 23 Mountain region, Midway Valley, et cetera?
24 MR. CONNOR: Again, all I want to do here is r~m
( ,) 25 speculate because you are asking me about stuff we are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2M-E33 WASHINGTON, D C. 20005-3701 (202) 234-4433
62 1 doing right now which is fine. But I want you to
,-) 2 understand that we are doing it right now and I haven't
's) 3 had a chance to review it and that sort of thing.
4 But, yeah, of those 10 to 20 anomalies there 5 are two if, I recall correctly, in Jackass Flat for 1
6 example, that I would imagine if those turned out to be I 7 Pliocene basalts that would have an impact on probability i
8 models that drew a hard boundary between Crater Flat and 9 the repository. Okay?
10 MEMBER HINZE: Yes, sure. 1 I
11 MR. CONNOR: So, I agree that there are )
l 12 definitely anomalies that have a higher priority than l I
13 others. But I would also reiterate that if we are going '
tn\
5- / 14 to understand the process, more data gives us more )
1 15 confidence, I think, in understanding not only where the 16 volcanoes are and when they erupted but also changes in )
i 17 volcanic activity in the region through time. It also l 18 increases our understanding of interaction of structure 19 and say rates of extension and episodes of volcanism.
20 I think that given that it is relatively 21 straight forward to collect these data, it is worth that 22 kind of attention.
23 CHAIRMAN POMEROY: But just to clarify it in 24 my mind, Chuck if you can, every incremental bit of data rm At some point you have
() 25 helps us understand the process.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
63 1 to say either that I have sufficient data to make a 7, S 2 decision or I don't.
(v) 3 MR. CONNOR: Sure.
4 CHAIRMAN POMEROY: It is my understanding of 5 this problem that the geologic processes may not be fully 6 resolved in my life time and perhaps not in youre. So, 7 how doyou determine where the cut-off is? Where should we 8 decide if we have enough data or not? How do ve make that 9 decision? How do you cut off at some point and say I have 10 sufficient data to make a decision?
11 MR. CONNOR: Yes.
12 CHAIRMAN POMEROY: That is something that 13 scientists usually don't like to do.
V 14 MR. CONNOR: Sure, sure. I would answer your 15 question this way, and this is only my personal view, I 16 have studied the Yucca Mountain system for five years now 17 and these data improved my understanding of volcanism in 18 the area.
19 They gave me confidence in understanding the 20 northeast trending alignment which Gene Smith talked about 21 in 1990 and it has been dumped on and what not. E' 22 basically, it reaffirms the importance of that.
l l 23 I think it is important that we see a 24 clustering of volcanism in Southern Crater Flat; that is
( 25 certainly not the repository site. The more volcanoes NEAL R. GROSS
! COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1 l 64 1 that you identify there it reaffirms our understanding of 2 volcanism clusters.
b,e~g 3 I think it is interesting that there is a 4 a crmal polarity anomaly along that alignment, the 5
quaternary Crater Flat alignment. All the other volcanoes 6 are reversely polarized which makes me suspect that that 7 feature may have hosted volcanism over time and that is an 8 important consideration in the probability models.
9 Some of the probability models coming out of 10 PVHA used changes in volume eruption rates through time to 11 get a handle on recurrence rates.
12 j Well, if you don't know what was going on in
,a 13 the Pliocene because half these volcanoes are buried or i \
'k ' 14 something like that., then you have trouble understanding 15 the change in recurrence rates through time based on 16 volume model in a volcanic field like this one where the 17 eruption rates are extremely small compared to a lot of 18 basaltic volcanic fields.
19 So, we do learn a lot about it. And, if I may 20 be so bold, you are interested in the bottom line. That 21 is something that I am really interested in', too.
22 It is not such a great exercise to trudge 90 23 kilometers through the desert to get the data and --
l 24 CHAIRMAN POMEROY: I understand that. l 7N l
! ,) 25 MR. CONNOR: And I guess that is an indication NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433 l
65 1 of how important I think it is to get that kind of data.
7S 2 I think that will help us understand issues.
3 MR. HILL: I would like to add a point, if I 4 may. This is Britt Hill at the Center.
5 One of the real well-defined characteristics 6 for other volcanic fields in the Western Great Basin is 7 that when they're long-lived, five or six million years of 8 volcanic activity, there tends to be a regular spatial 9 shift through time.
10 If you look at the Coso field, Reveille Range 11 lunar crater field, Big Pine, just about any big field out 12 there, and the locus of volcanism displays a regular 13 spatial shift.
t )
14 We hwee not uncovered that kind of spatial 15 shift on the Yucca Mountain field. There doesn't seem to 16 be a real change from Pliocene to quaternary. None of the 17 probability models that have been developed for the Yucca 18 Mountain field have tried to accommodate a regular spatial 19 shift through time in the locus of volcanism.
20 It may be, and I am just speculating and one 21 of the things we are going to do is we are going to 22 evaluate the significance of this for that kind of a 23 probability model, is that if we had ten additional 24 Pliocene volcanoes in the Armagosa Desert, we may have to
/n 4 think about models that account for that spatial shift
( ,/ 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (?O2) 2344 433
l 66 1 from a Pliocene locus in the Armagosa Desert to a l
1 2 quaternary locus in Crater Flat.
fS L) 3 Now, whether that will make a bottom line 4 difference in the absolute value of probability of the l
5 repository site I can't evaluate because our models are 6 not capable of dealing with that spatial trend through 7 time.
8 But I do think we have an obligation to test 9 that hypothesis and then move forward if we feel that a 10 class of probability models needs to be accounted for to 11 accommodate this new data and the potential spatial shift. ,
l 12 MR. RYAN: What is your sense of shallow to 13 intermediate crustal storage recions for these magmas as
~
A 1
( ?
\/ 14 you might reconstruct them, and what would the evidence be j 1
15 for that.
16 MR. CONNOR: I guess I would let the 17 geochemist answer that question. Go for it, Britt.
18 MR. HILL: It is indirect evidence right now 19 b'ased on the mineralogy, about three per cent olivine is 20 about the only phenocryst in there. There is an absence 21 of plagioclase in part due to what I believe is a fairly 22 high water content in the magmas, about 2.8 per cent 23 water.
24 There are phenocrysts of tightened partosite O)
(_ 25 to carosotitic amphibole in some of the Crater Flat magmas NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
67 1 in trace amounts. But they are texturally phenocrysts not
,s 2 xenocrysts or megacrysts.
3 So, the interpretation would be that we are 4 dealing with something that is probably 10 kilobars or 5 greater as the last equilibration depth.
6 There is very little, if any, evidence for j l
7 crustal assimilation. Buckboard Mesa, the 2.8 million 1 8 year volcano has some granitic xenolith to it but it is a l l
9 small contamination trend. By and large we are not seeing 10 any evidence of shallow mineralogy, eight to shallower 11 kilobar mineral assemblage or crustal assimilation as a 12 significant process.
13 MR. RYAN: So then, Britt, are you suggesting i )
\' 14 that these magmas came directly from depths on the order l
15 of 30 kilometers?
l 16 MR. HILL: I would push it more to 30 to 60 17 kilometers was there last equilibration depth and pretty 18 much rammed up. ;
19 MR. RYAN: That might be higher than 10 20 kilobars then.
21 MR. HILL: Well, it certainly would be higher 22 than 10 kilobars. Ten kilobars to me is a petrologic 23 convenience about where the experimental data usually are.
24 MR. MARSH: What are you basing this water 7-~
(,) 25 estimate on?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
68 1 MR. HILL: Experimental work by Green back in 73 2 the early Seventies on Hawaiite magmas of almost identical
- 4 L.)
3 composition to this where he had 2.8 per cent water at 10 l 4 kilobars and did not have amphibole on the liquidous, but 5 did find a tightened partosite with 5 weight per cent 6 water on the liquidous.
7 MR. MARSH: That is quite a high water 1
8 content.
9 MR. HILL: We have had a lot of discussion and 10 speculation on interpreting very indirect data. We tried 11 to get some ion probe data on melt inclusions but we were 12 unable to find any suitable melt inclusions in a 13 reconnaissance study to get direct measurements. I
(~'\ l
14 MR. RYAN: I think I would agree with Bruce 15 that this is a surprisingly high water content.
I 16 MR. HILL: I think in the last -- l l
17 MR. RYAN: If it is systematic. l l
18 MR. HILL: It it is systematic. But I think l
19- we have seen quite a bit of work in the petrology over the 20 last five years going for hydra-basaltic magmas, 21 especially when they derive from a lithospheric mantle.
22 MEMBER HINZE: What is the implication of that 23 if it is high?
24 MR. MARSH: Well, if this whole magma was an
('N
(_,) 25 amphibole, you could crystalize it all with two per cent NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
69 1 water into an amphibole and none of the basalts that we 2 know in any detail gets anywhere near two per cent water o)
(
Q ,/
3 in basalt.
4 MR. HILL: Certainly some of the work of Grove 5 and Sissom up in the Medicine Lake area would support 6 that.
7 MR. MARSH: Yes, but there is a lot of 8 contaminated stuff in those lavas.
9 MR. HILL: Sure.
10 MR. MARSH: Lots of junk in it so it is very 11 hard to know what you are dealing with in those things.
12 And they have big crystals that are probably restitic 13 chunks from the wall rock and things. If you look at IL') 14 their crystals they have a real history.
15 MR. HILL: Sure.
16 MR. MARSH: What I have seen of these, they 17 look pretty clean as you said.
18 MR. HILL: Very much.
19 MR. MARSH: It would be very nice to have ion 20 probe.
21 MR. HILL: It would be nice but we tried in 22 reconnaissance but we didn't have the resources to go a 23 full blown study.
24 The important point for this question would be O
(,)i 25 that if we have lower water contents, the absence of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
70 1 plagioclase would imply that we are seeing a deeper magna
,rs 2 system rather than a shallower magma system, wouldn't it?
V 3 MR. MARSH: Well, the major implication I 4 think would be that the magmas would be a lot more 5 explosive. When you think about the delta V, the change 6 in volume from in solution to out solution can be a factor 7 of 10,000 sometimes.
8 So, on the surface they look rather normal in 9 terms of cinder cones and the way the craters are spread 10 around. I would expect to see them much more explosive 11 and almost diotrene eruptions with a lot of xenolithic 12 material in the flows coming out near surface.
p_ 13 MR. HILL: I can quickly draw from direct
)
i 14 analogy at Cerro Negro volcano in Nicaragua where we do 15 have ion probe data with 3.5 weight per cent water in the 16 melt inclusions in the olivine and, for all intents and 17 purposes, looks like Lathrop Wells.
18 MR. MARSH: And you have to realize that these 19 melt inclusions in the crystals have been stripped of a 20 lot of their major elements and these actually often 21 happen from melt being struck between crystals and they 22 have been annealed together, so that is an absolute upper 23 limit.
l 24 MR. HILL: Sure.
(_ / 25 MR. MARSH: So originally, if you take into l
NEAL R. GIUDSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
71 1 account the point of crystallization, it may have only g 2 been a half of a weight per cent of water.
e i 3 MR. RYAN: With the possible exception of 4 Kimberlites I would be very surprised, frankly, to find 5 magnas of this composition ascending directly without 6 pause from 40 kilometers depth. One would expect, j 7 actually, some development of shallow and intermediate 8 hostel storage prior to eruption. This is the garden 9 variety case worldwide.
10 MEMBER HINZE: Is there any consequence to 11 this in terms of probability or the consequences of the 12 effects?
13 MR. RYAN: Well, depending on the subsurface l f)
I 14 structure there could be consequences in terms of the 15 combinations of dynamics of dynamics and pathway. If, for 16 example, the subsurface storage region is substantially 17 offset in plan from the ultimate erupt event, and if that .
18 storage region is at, let's say three to six or three to 19 eight kilometers depth, and it will offset I suppose 20, 20 30 kilometers, or 10 kilometers, interesting and perhaps 21 as yet unanticipated consequences for the overall dynamic 22 system and the overall pathway.
23 MR. MARSH: Especially for degassing. Because 24 it doesn't have to stay there a long time but it has to f~y
( ,) 25 stay long enough to nucleate and coalesce bubbles and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 ,
1 72 1 degas itself to come out in a more normal eruptive
.s 2 fashion. If it came up in one shot, I think as Mike says, i #
)
3 you would see a much more dramatic effects on the surface.
4 MR. 'imu: But doesn't that depend on the 5 ascent rate?
6 MR. MARSH: Well, these things are accompanied 7 -- in terms of the small columes they have to come fairly 8 rapidly in your mana, if you are going to come up with 9 that chilling entirely.
10 MR. HILL: Right.
11 MR. MARSH: But you can hold them at 12 intermediate depths in the crest when the crest is still 13 fairly warm then they can form a chilled margin and s
'/
14 protects them from thermal and degass there. But, so 15 those are limitations also, 16 MR. RYAN: If one accepts the very large water 17 contents you are suspecting, then the exclution of 18 volatiles is going to radically change the momentum 19 environment in which magnas would then reaccelerate once 20 exclution takes place.
21 MR. HILL: Sure.
22 MR. MARSH: And retrodive, of course, would be i
23 carbon dioxide release and water doesn't usually come out 24 until extremely shallow pressures, as you know. But the (3
Do
(_) 25 CO2 reconstructed contents would also be of interest.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234 4433 l
l
73 l 1
1 you happen to have a feel for that?
2 MR. HILL: Not at all.
p-~3 l 3 MR. MARSH: And the other factor that you l
4 release this much water is the liquidous and colidous 5 actually rise up through the system very fast and you 6 increase crystallization rapidly. And the fact that these 7 are low crystal entities, the phoenecous contents, crystal 8 entity contents in terms of pre-eruptive crystals looks 9 really low. It means to me that the system is not 10 degassing strongly in terms of water. And so the water --
31 probably doesn't saturate until shallow depths. Otherwise 12 the liquid would sort of sweep right up in the system very 13 fast and you would get incredibly strong nucleation
%- 14 trends, leaves of nucleation through the whole body, as 15 Rutherford and Kashwin have shown in experiments.on the 16 Mount St. Helen's eruptive.
17 MEMBER HINZE: I think we will go back to this 18 discussion at a round table.
19 MR. HILL: I'd like to have a sidebar with you 20 on that and how we get anthro liquidous then on these 21 mats. But that's --
22 MEMBER HINZE: Yes, we can pick --
23 MR. HILL: -- something we can do around 24 lunch.
3 through magnetic mapping.
4 Probability models should wind up with higher l l
\
5 values than Southern Crater Flat, for instance, than in, 6 you know, at the Yucca Mountain Repository, or something 7 like that. That's an important aspect of the models that 8 needs to be considered.
9 Second, there is an association between j l
10 volcanoes and faults. As evinced by the Northern Cone 11 Magnetic Dataset. It's important to consider that somehow 12 in probability models.
13 Third, is a prevalence of this northeast -- of
-' 14 northeast trending vent alignments. There is, again, two 15 maps and we've now identified a third based on the 16 geophysical dataset.
17 And four, there is a low and persistent rate ,
18 of volcanism in the area. Apparently, through time and, 19 of course, additional identification of volcanoes for 20 instance, may change that. But basically this is a 21 volcanic field that has been going for eight million years 22 at a very low rate of magnum production. I think that's 23 safe in anybody's book.
24 Okay, I'm going to try to move through this.
p
() 25 This is stuff that most of you have seen at least twice NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
76 1 before. Basically, I try to incorporate this kind of data
-- 2 in probability models in a -- using a series of steps. In 3 the first step, basically these boxes, excuse me, these i
4 boxes represent maps and in the first step I tried to map l 5 a probability distribution based on a non-homogenous 6 model.
7 There are several different kinds of models 8 that can be used to do this. Near neighbor models, 9 colonel models like the Knott Epanechnikov Kernel, is 1 10 basically estimating some sort of a special recurrence 11 rate based on where volcanism has occurred in the past, 12 and perhaps when that volcanism occurred. It could be a 1 13 spaciotemporal model.
(~~h .
t : 1 N' 14 I want to fold into that dataset an estimate J
15 of how structure, for example, faults revealed by the
]
16 gravity data or map faults may influence magna ascent and 17 hence the distribution of volcanoes. So I'm going to try l
18 to i:;1d some structure into that somehow. l l
19 Third, you can look at an estimated dike 20 geometries that can intersect the repository, dike links 21 along which vents may occur. So, for example, you know, 22 what's the distribution of buds along the dike or 23 something like that.
24 Finally, we need to fold in an estimate of the I
(,) 25 regional recurrence rate of volcanism. So for the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
77 1 volcanic field, if you need to find isotopically or 73 2 whatever, what's the regional recurrence rate? And these NJ 3 things give you some estimate of the hazard.
4 Okay, I'll just do this fairly quickly. This 5 is an example of a non-homogenous probability model 6 superimposed for the Yucca Mountain area. This particular 7 model is a near neighbor model. It's based on eight near 8 neighbor volcanoes. Blue here is where cyan is relatively 9 low probabilities, increasing to red to relatively high 10 probabilities.
11 What I've done is I've normalized this map so 12 that the probability of an event in this map region is 13 equal to one. Okay? So if a volcano occurs, where will
('/ )
14 it occur in the Yucca Mountain area, for example. And 15 this is a result based on the timing and distribution of 16 past volcanic events. Certainly in Crater Flat you get ]
i 17 the highest probabilitie". Out here, this border between 18 green and yellow is about 1 x 10-5 per square kilometer.
19 That's the probability of volcanism in a given location l 20 given a volcanic event in the area. And, of course, it l 21 drops off with this from Crater Flat.
22 So, this kind of captures a sort of, think of l
23 it as a big filter. It captures some aspects of volcano 24 distribution. But, this is a model that I would guess we p)
(_j 25 first proposed in 1993, it doesn't capture some things NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
78 1 that may be important in volcano distribution. For s 2 example, Bear Mountain here is an uplifted paleozoic I \
LJ 3 block. Many people think that the probability of 4 volcanism at Bear Mountain, because of its structure and 5 possibly it's elevation, some people would speculate would 6 prevent volcanism from occurring there. Volcanism 7 probability is high there, say as high there as it is over 8 here or over here.
9 So, you know, there are things that you could 10 speculate about that may change that distribution. This 11 is just a straight model.
12 MR. MARSH: What's the background values?
p_s 13 Just in the countryside, numbers?
I i i /
14 MR. CONNOR: Out here it drops off to about 15 10 -' .
16 MR. MARSH: So that's a baseline.
17 MR. CONNOR: Yes.
18 MR. MARSH: So you have worked from there up?
19 MR. CONNOR: Yes. Given an event in this 20 region, not folding in anything with a recurrence rate 21 here. Yes, the highest values are about 10 -- a little 22 higher than 10-4 in this region. So we are getting orders 23 of magnitude variation across the plot.
24 MR. MARSH: Over what period of time would ir'\
( ,/ 25 that be?
HEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
79 1 MR. CONNOR: Well, remember this is just if a fg 2 volcanic event occurs.
L] 3 MR. MARSH: Given an event, okay.
4 MR. CONNOR: Given an event, this is --
5 MR. MARSH: Spatial?
6 MR. CONNOR: This is only spatial. Okay.
7 Now, as a way of addressing this possible structural 8 control on volcanism, what I did was I actually folded 9 together the horizontal gravity gradient data with the 10 probability map. So, and there are various ways to do 11 this. I'm not going to go through it in a lot of detail.
12 But essentially what we've done here is said
,_s 13 that if I normalize the horizontal -- amplitude of the
/
1
\/
\' 14 horizontal gravity gradient as a probability density 15 function, then I can multiply that together with the near 16 neighbor map, renormalize that and look at the resulting 17 PDF.
18 The idea there is that as we saw earlier, the 19 horizontal gravity gradient was giving us amplitude of the 20 horizonal gravity gradient and it was giving us a good l
21 indication of where prominent structural features are i 22 based on volcano distribution. Sometimes we think -- or i
i 23 at least often, it looks like volcanoes correlate with the 24 locations of those faults. So fold those two datasets F'%
(_j! 25 together and see how that changes the probability map. l l
! NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE , N W.
f (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
80 1 Well, one thing it does it is makes it a 7- 2 little bit more complicated, of course, areas of (3) 3 higher / lower probability. I wouldn't pay too much 4 attention to the noise out here. But what it really does 5 is it elongates this anomaly in a west / northwest direction 6 parallel to the Bear Mountain fault. Okay?
7 So, if you think that Bear Mountain Fault is a 8 major feature bounding the west edge of the grabin and you 9 think that volcanism is related to extension of this 10 grabin, then it seems like -- it seems reasonable that 11 this kind of influence would take place. So the whole 12 probability surface is extended in a west / northwest --
13 northwest direction. And to a certain extent mimics the
,.m 14 extent of the Crater Flat Volcanic Zone proposed by Krone-15 Perry.
16 Okay, there is --
17 MEMBER HINZE: Are you saturating -- excuse 18 me, but are you saturating your maximum values? It 19 appears -- if you look at the gradient map, it appears 20 that the horizontal gradients are much more intense along 21 the Bear Mountain Fault, the southern portion, and then --
22 MR. CONNOR: In the center of the map. That's 23 right --
24 MEMBER HINZE: Well, and also along the (A),
25 eastern boundary of that anomalous area. So, why do they NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
i 81 j l
1 show up as the same? l l
,_s 2 MR. CONNOR: Well, I haven't changed the 1 3 scale here. So this boundary between yellow and green is )
l 4 1 4 still 1 x 10 per square kilometers. Is that what you are 5 asking?
6 MEMBER HINZE: Well, it appears that what I'm 7 concerned about is that you may have saturated the 8 coloring on this, have a low cut off so that you pick up l
9 the eastern margin of that and also the intense values in 10 the center of Crater Flat where you don't have a high l
11 horizontal gradient.
12 MR. CONNOR: Well that o true. But the --
1 13 actually if I draw a profile across here, the high values
- \
(_) 14 are still in the center of Crater Flat. j 15 MEMBER HINZE: Okay.
16 MR. CONNOR: Okay? There is no doubt about 17 that and actually, now that you ask that question I 18 realized I planned to bring that profile, but I didn't.
19 But I -- the highest values remain in the center of Bear 20 Flat. In fact, what we've done is looked at horizontal 21 slices all along this thing and basically this axis is the 22 highest probability zone, decreasing as you go northwest, 23 okay?
24 The contour plot for this is kind of, you
/ ~ %.
'(_,) 25 know, gets a little messy out in here where people tend to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
82 1 focus. So I, you know, think this is a reasonable way to n 2 show it.
)
(-"/ 3 MEMBER HORNBERGER: How does your weighting 4 factor affect this?
5 MR. CONNOR: Yes, let me -- I'll address that 6 point in two slides. I 7 MR. MARSH: Is this normalized to one?
8 MR. CONNOR: This is normalized to one, 9 exactly.
10 MR. MARSH: It looks like -- the proportion of 11 rate area is very large, compared to the last one.
12 MR. CONNOR: It's normalized the same way.
13 MR. MARSH: So the value changed, basically.
\
P'
't) 14 The background value, I guess, is lower.
15 MR. CONNOR: That's a good question. I'm not 16 positive. But I don't think it's changed significantly.
17 I think the proportion actually is about the same. So we 1 18 look at that.
l l
19 MR. RYAN: Chuck -- seemingly, depending on l l
l 20 the combination of the radial distance from the Crater 21 Flats area and the proximity of gravity gradients to that 22 area, the yellow regions essentially leak off, if you I l
1 23 will, from this red dominated area, despite the fact that 24 they might actually be normal to the map fault system on ;
/^\
() 25 the surface. Just as an example.
1 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
. 83 l l 1 MR. CONNOR: Yes.
7\ 2 MR. RYAN: This is not necessarily a ix'j) 3 criticism, it's simply an observation.
4 MR. CONNOR: Okay, right.
5 MR. RYAN: Okay, could you comment on your 6 selection of the radius of influence that -- or your R i 7 factor --
8 MR. CONNOR: Sure.
9 MR. RYAN: -- basically your area of Crater 10 Flats in relation to local structure?
11 MR. CONNOR: Yes. The -- well, there are 12 several levels to your question. First is in a near 13 neighbor model you choose the number of near neighbors you
\- # 14 use to make your estimate. I tend to use models with 15 relatively large number of near neighbors, eight, you 16 know, which is a relatively high proportion of volcanoes 17 in the field, to smooth thing out. Because I feel like 18 the few number of events would even, particularly with 19 this few number of events, we should look at a relatively 20 smooth picture.
21 And something like an Epanechnikov model you
! 22 actually have the smoothing distance parameter, it's been 1
23 called, I think amply during the PVHA study. And that is 24 a very important influence because if you choose a very (N
's ,) 25 short smoothing parameter like --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 2344433 WASHINGTON, D C. 20005-370* (202) 234-4433 l
84 1 MR. RYAN: So the radius itself is coming into rs 2 the rating factor?
! I v'
3 MR. CONNOR: Right, right. Then if you use a l l
4 small smoothing parameter, then you lump your probability I 5 next to where volcanoes have occurred in the past, of 6 course, very tightly. If you use a large smoothing 7 parameter, you smooth that out and spread that probably 8 density function out over a bigger area.
9 Now, the -- I can think of a -- I did some 10 cluster analysis work to try to sort out, get an idea of 11 what kind of cluster r*dius we are dealing with and that 12 came out in the Yucca ;iountain area to be on the order of
,_, 13 16 to 20 kilometers, something like that. So, I prefer
/
\
\ '! 14 using models with smoothing factors of about that scale.
15 Other people use tighter smoothing models. Tighter 16 smoothing parameters are used in PVHA.
17 And, finally, how does that compare with the 18 density of faults, for example? The fault density tends 19 to, you know, where there mat at Yucca Mountain and not an 20 alluvian, fault density is quite high compared to those 21 kind of smoothing parameters. You know, so over hundreds 22 of meters of fault spacing.
23 Okay, this just -- I'm just going to show this 24 to show that there are a variety of ways to fold these Ch
(_) 25 data cets together using different assumptions about the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
85 1 structure. This is the set of maps I just showed you f3 2 where I took the near neighbor model and multiplied that
( )
%.J 3 together with the horizontal gravity gradient. Here, 4 giving these two maps equal weight to produce this 5 probability map.
6 On the other hand, you can do things like try 7 to use fault density. And you know, it's tough using 8 fault density because, obviously the map fault density 9 changes depending on how much alluvial cover is present.
10 If you assume that the density of faults in Crater Flat is 11 the same as it is as Yucca Mountain, just make that 12 assumption, then I inunt through the density of faults 13 this way. And actually folded in some additional
's- 14 information. I was really concerned with the distribution 15 of high dilation tendency faults, that is faults that have 16 -- or the proportion of high dilation tendency faults.
17 That's faults with a north or northeast trend through the 18 area.
19 So that tends to shut off areas like uplifted 20 paleozoic fault blocks or the Timber Mountain caldern in 21 terms of fault density. So even though there are a large 22 number of faults there, they also have very little to do l 23 with current stress data.
24 And so if you fold that together you get a
/"'s
( ,) 25 probability surfage like that and what I point out is that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i
1323 RHODE ISLAND AVE., N W.
l (202) 2344 433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
l
86 1 these are comparable in shape and so using different gs 2 assumptions about the structure I've come up with a l' '/
1 3 relatively similar looking model. Again, it indicates 4 this north / northwest trend is important and that the 5 probability is highest in Crater Flat.
6 Okay, so if we get specific to the repository 7 --
8 MR. MARSH: What other kinds of -- have you 9 done this using other fields? For example, topographic 10 stress from just topographic that --
11 MR. CONNOR: Yes, I did it with topographic 12 data. I'm not a big believer in topographic control on 13 cinder cone distribution. But I went ahead and did it El 14 anyway.
1 15 If you make a PDF from the topographic field 16 and you use the elevation of the crest of the mountain, l l
17 then the probability of disruption after a repository i
18 changes very little. It goes down by, you know, from -- I i 19 can't remember the exact number, but from say, 2 x 10-8, or I
i 20 1 x 10-8 or something like that. You know it drops by a l l
l 21 factor of two or less.
22 If you use the elevation of the repository 23 itself, rather than the crest of the mountain, it tends to 24 go up a little bit. But again a trivial amount.
rh
( ,) 25 MR. MARSH: And when you do your fault NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISt.AND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (?O2) 234-4433
87 1 probability, do you, because you have alluvium over a lot
,3 2 of things, but you know the bedrock basically what's there
'~
3 from the structure models.
4 MR. CONNOR: Right.
5 MR. MARSH: Can't you propagate through normal 6 fault densities that you would see in various kinds et 7 bedrocks over various ages, because you know the 8 structural history, and have a, you know, have a map for 9 the whole area. Is that what you've done here?
10 MR. CONNOR: I haven't quite done -- gone 11 quite to that extent. What I've done is I've looked at 12 the distribution of alluvium using the GIS covers from 13 fizzel & Shulterz where there is dense alluvium, and of
( i <
's /
14 course there is a few faults. What we've done is we've 1
15 taken a look at increased -- or changed fault density in l 16 those alluvial cover areas from basically the distribution 17 of map faults, in which there is no faults in alluvium, to 18 the density of faults is about twice that of the maximum i 19 density of faults on Yucca Mountain.
20 So rather than make the assumption I've just 21 explored, the effect of that change in the parameter on my 22 bottom line, which is the probability.
23 MR. MARSH: But you have put something in for 24 Crater Flats, I guess, or --
(T
'( , ) 25 MR. CONNOR: Sure.
l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1 88 1 MR. MARSH: And so you weighted it, I guess, 2 to make the worst case. Probably assuming there is a 7~3 V
3 certain fault density there.
4 MR. CONNOR: Well, as you increase the fault 5 density in Crater Flat, that tends, and it is in a 6 structure model, that tends to decrease the probability at 7 Yucca Mountain.
8 MR. MARSH: Right.
9 MR. CONNOR: So, yes, that's right. And 10 we've, in this particular model, I've assumed the fault 11 density in Crater Flat is the same as the maximum density 12 of faults in one square kilometer area on Yucca Mountain.
13 MR. MARSH: And you've normalized all these to
- 14 one again?
15 MR. CONNOR: Yes, it is all normalized to one.
16 MR. MARSH: So, once you know a regional 17 probability, you have to plug this into a regional 18 probability to actually get absolute probability, right?
19 MR. CONNOR: Yes, yes. Which is what I've ,
l 20 done in the next couple of slides. Here, for example, is i i 21 a range of probability of intersection at the repository l 22 for a regional recurrence rate equal to one that just l 23 means that if the volcano occurs, it has a probability of 24 intersection with the repository. )
,e x
(_)h 25 What I've done here is I assumed a uniform NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
! (202) 2344433 WASHINGTON. D C. 20005-3701 (202) 234-4433
89 1 distribution of dike half link from 100 meters to 2,000 r~g 2 meters. Okay? Assuming that if an event occurs that i
3 close to the repository, then that's a disruptive event.
4 So that's going to be important when we start comparing 5 apples and oranges between this and PVHA. So please keep 6 that in mind.
7 And here is the kind of curve that develops 8 for a number of different near neighbor models or kernel 9 models in different weights assigned to fault density.
10 Okay? So if you weight structure data quite heavily, you 11 are out here on the graph. If you give structures no 12 data, you are back to the straight non-homogenous Poisson 13 model, you are back on this axis.
J 14 And you see that all these curves go through a 15 maximum between, you know, if you assign a little weight 16 on the order of one up to a factor of two, for these 17 structural models, you tend to go through a maximum. And 18 that's the kind of range and probability we get with, it 19 shows these curves to represent the range in probabilities 20 we get for intersection given a volcanic event in the 21 region for these different models.
22 So I want to point out that at this point I i
23 don't think there is a very good geologic basis for 24 distinguishing between a lot of these models. And that's p).
\m, 25 one reason we are trying to do this kind of a bounding l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
90 1 analysis.
s 2 Another set of curves, not using map
/
)
i '~
3 structures but rather using the horizontal gravity 4 gradient data, you get a very similar kind of 5 distribution. And then finally if we just cast that as a 6 probability, this is given a regional recurrence rate 7 varying between two volcanoes per million years and ten 8 volcanoes per million years. Okay?
9 So that's -- some of the uncertainty in here 10 of course is what the regional recurrence rate of 11 volcanism is going to be. And you can see this is the 12 kind of range we are talking about here. None of these 13 models gave me values of less than 1 x 10 8 annually. And
\
(~/
'x- 14 I would say that you only get into the upper part of this 15 curve, up above 1 x 10" annually with some pretty strong 16 assumptions, worst case models. So, I tend to call the 17 upper limit 1 x 10" where you can get up to 2 x 10" in the 18 worst case models.
19 Okay, so I just want to summarize a little bit 20 with some conclusions. I guess the language in the first 21 conclusion is probably a little vague for you, Paul, but, 22 you know, this is the Lest I can do right now, is that
! 23 plio-quaternay volcanoes near Yucca Mountain correlate 24 well with high dilation tendency faults, and regionally
/~T
( ,) 25 with vertical displacements in basement indicated by the l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
91 1 gravity data.
. 2 What does that mean? It means that some of
'~'i 3 the volcanoes lie along these high gravity gradient areas.
4 We can find faults going through some volcanoes.
5 Certainly not always the case, as far as I can tell.
6 Second, when include structural control in 7 volcanic hazard models, probability tends to increase.
8 And that increase is by about a factor of two or so.
9 Rarely more than that. So, for example, if we follow one 10 of these curves, this is not including structural control.
11 As I give increasing weight to structural control, the 12 probability goes up a bit.
13 I don't see order of magnitude changes in s
[/)
\_ 14 probability by including or excluding structure. Okay?
15 Things change --
16 MR. MARSH: Why is there a maximum in that?
17 Why does it come down so fast?
18 MR. CONNOR: Well, actually it doesn't come 19 down as fast as it goes up. But what happens is, let's 20 take a look at the -- if I can find the slide, let's take 21 a look at the horizontal gravity gradient data. For 22 example, when I give structure a lot of weight compared to 23 volcanism, then stuff, for instance way over here, is 24 going to start, you know -- if I only included structure ,
I r~s
( ,) 25 and not the past distribution of events, this fault would I
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. {
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
l 92 1
1 1 have a lot of weight compared to this fault.
l 7s 2 MR. MARSH: So you keep taking it away from --
i )
\' '>
3 MR. CONNOR: So I take it away from the Yucca !
l l
4 Mountain area. Right.
5 MR. MARSH: Concentrated in the -- l 6 MR. CONNOR: Yes, and it just happens to go
)
l 7 through a maximum which is convenient --
8 MR. MARSH: But you have to net sum to one, so 9 it --
10 MR. CONNOR: Right, right. Okay. And these 1
11 models indicate a range of probability in volcanic l l
l 12 disruption from about 1 x 10-7 to 1 x 10-8 And again, if i
13 you assume very low regional occurrence rates, if you
(~T l s- 14 assume that the recurrence rate of volcanism is one event 15 in a million years which is lower than it has been in a 16 region, or if you assume very high recurrence rates of 17 volcanism, you know, you can get above that value.
18 Based on these models, this is the range we've 19 come up with. There are some differences in how events 20 are defined in this case compared to PVHA, so it's 21 important to keep those clear in your mind. But, I guess
- 22 I would say there is overlap in the mean are near overlap 1
l l 23 in the mean value with this range.
24 And these models are a little low compared to eT
( ,)
25 those values proposed by the State -- by Ho and Smith, i HEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 l
l
93l 1 MR. MARSH: But yet this doesn't take into
~s 2 account the probability of volcanism occurring in this I
)
3 area of Nevada or the western United States?
4 MR. CONNOR: Well this does, this final value 5 does because I fold it in as a last step that regional 6 recurrence rate. Right so --
7 MR. MARSH: So how big is the regional rate?
8 I mean, when you say regional, what is that?
9 MR. CONNOR: Well, if we go back to --
10 MR. MARSH: Oh, you can just tell me.
11 MR. CONNOR: Okay. Well, I can't just tell 12 you. It's a little tricky because if -- what I'm doing 13 here is looking at volcanoes, the aeromagnetic boundaries r
(m\ ') 14 on the south, sort of the southbound of the map, going up 15 north to about Buckboard Mesa, going from the Bear 1 16 Mountain Fault over to Yucca Mountain. I'm taking that i
17 slice and I'm saying well, if the volcanoes I see now l 18 represent the amount of pliocene and quatery volcanism in 19 this region, then there are, say, 30 volcanic events in 20 that area over the last five million years.
21 MR. MARSH: But you compare that to the whole 22 state, this area that you chose right there, for example, l 23 is it higher or lower for example, than a volcanic --
24 MR. CONNOR: Well, sir, if you choose a very
/~T 1
(_) 25 large area, what you are saying -- this is kind of like a NEAL R. GROSS l
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
94 1 homogenous model. I'll choose my area and get the r- 2 recurrence rate in this large area. If you choose a large
- /
3 area then you, of course your recurrence rate goes down.
4 MR. MARSH: Well, if you choose the whole 5 earth.
6 MR. CONNOR: Right.
7 MR. MARSH: But if you actually choose, you 8 know, if you choose an area where we've had --
9 MR. CONNOR: Well --
10 MR. MARSH: -- cinder cones through basin 11 range or an area of province, I'm just interested in how 12 the base level is set, actually.
7, 13 MR. CONNOR: Well, if you look, this is a 14 calculation that I did a wl.ile back. But if yov '
ai 15 the area as a whole, I come up with about 10~" per year as 16 sort of a generic number that if you just wanted to throw 17 a dart at the Western Grade Basin, that's the probability 18 of volcanism over about a eight square kilometer area, 19 which at the time I was using for the footprint or 20 repository.
21 So these are on the order of, you know, two --
22 maybe two orders of magnitude higher than those kinds of 23 numbers. nd if you, you know, look at a map like Lutken-24 Smith, for example, you know, the locus of volcanism in
\s- 25 the quarternery has occurred in very specific areas in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
95 1 Western Great Basin and it's had a high tendency to occur
,s 2 where pliocene volcanism occurred.
(m j) 3 MR. MARSH: So if you add that into it, what 4 happens then?
5 MR. CONNOR: Well, with a non-homogenous 6 model, and a nice advantage of a non-homogeneous model is 7 I think I've gotten away from that edge problem which is 8 an issue you are dancing around. So that's one reason to 9 go with a non-homogenouc model.
10 This model is not going to change as I go to 11 bigger and bigger areas. The map might be, of course, 12 become more complicated on the edges. But at Yucca 13 Mountain the probability surface won't change a lot.
7-i )
\> 14 MR. RYAN: You mentioned that, when you were 15 discussing the graph on page 17, that you had selected 16 dike half lengths of 100 meters to 2 kilometers. Did I 17 hear you correctly?
18 MR. CONNOR: That's right.
19 MR. RYAN: Can you talk about that? Can you i
20 explain why you selected those numbers?
21 MR. CONNOR: Yes, I was looking, I basically 22 used, I was interested in disruptive events. So, I used a 23 value -- uniform random distribution sampling in there. l 24 And it actually, if I cut the half length down to 1,000,
,/ 3 s_) 25 it decreases my values by about a factor of two. Okay?
NEAL R. GIUDSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, O C 200C9-3701 (202) 234-4433
96 1 What I was concerned about was the area likely
,y 2 disruptive, disrupted and with the potential for extrusive E 1 V
3 volcanism as a result. And if you look at, for instance, 4 eruptions of Perikoten volcano, Horruio volcano, Tolbachik 5 volcano, modern cinder cone eruptions, extrusive vents 6 associates with those vents are distributed up to several 7 kilometers away from the main vent. Horruio is a good 8 example where an alignment of something like reven cones 9 formed over a period of 15 years extending about five 10 kilometers.
11 MR. RYAN: So you, if, am I correct in l 12 thinking then that you have used the surface expression ,
i l
13 basically in analogue settings --
(~ l 14 MR. CONNOR: Right. l l
15 MR. RYAN: -- as a guidance and rule of thumb l 16 for selecting dike dimension of depth. .
l 17 MR. CONNOR: And I'm not particularly enamored l l
18 with this approach. I think it's a good -- you know, 19 there is merit in it. As I mentioned earlier, I was 20 recently in the San Rafael Field where Paul Delaney and 21 Ann Gardner have done a lot of mapping. And there the buds 22 expend over comparable or probably slightly smaller 23 distances. And you know, you have Sase beautiful 24 intrusions eroding out of the Entrada sandstone, so it's
(~N
( ,) 25 very obvious what these geometries look like.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
97 1 I'd like to go back and do some of these
<w 2 calculations using those kinds of geometries. And, again,
( )
3 I think that what that does it is makes the models robust 4 but it's likely to change the final answer by a factor or 5 two or less, using that kind of thing.
6 MR. RYAN: Do you think that these analogue 7 dike dimensions that you have used have, one can translate 8 them down in the third dimension some considerable 9 distance?
10 MR. CONNOR: No. These might correlate with 11 dike segments, for example, rather than a large dike. And 12 let me give you an example. During the 1975 Tolbachik 13 eruption, there was eventually an~ eruption about nine i
'N '
14 kilometers from the first breakout, they call it the 15 northern and the southern breakout. And you know, that 16 indicates to me some dilation and intrusion. Possibly a 17 master dike. But those would be separate events. If I 18 used a dike segment model where I had, you know, a four 19 kilometer upper limit.
20 MR. RYAN: The other think I was wondering 21 about is with respect to the integrated volcano structure 22 probability model on page 15, particularly plots C and E.
l 23 Plot C you derive by looking at the gravity gradient 24 dataset as input.
(3
(_,) 25 MR. CONNOR: Right.
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
! (202) 234-4433 WASHINGTON, D.C 20005-3701 (202) 234-4433 l
98 1 MR. RYAN: Plot E you derive by looking at gg 2 mapped fault distributions as input. There seems to be an i
(_)4 3 apparent rotation of the major axis of that red high 4 probability region clockwise somewhat, maybe five, eight, 5 ten degrees, possibly pushing it. Is that rotation 6 apparent or real?
7 MR. CONNOR: I think that I would prefer not 8 to push these models that hard.
9 MR. RYAN: Because the gradient that seems to 10 he influencing Plot C has a nocth/ northwesterly trend, but 11 the mapped structures that seem to be influencing Plot E 12 have a distinct north by northeast or north / south trend, 13 if I'm correct in recollecting.
(_ )
N' 14 MR. TRAPP: Chuck, just a point. I don't know 15 if you've seen the literature on some of the rotational 16 values that they have mapped out at Crater Flat, but there 17 is going from northern part of Crater Flat to the southern 18 part, a very definite clockwork rotation in the structural 19 field.
1 20 MR. RYAN: In what sense? '
l 21 MR. TRAPP: Clockwise.
22 MR. RYAN: Yes, okay.
23 MR. CONNOR: Yes, in terms of the probability 24 map, I wouldn't want anybody to try to extract that kind I \
(_ j 25 of detail. But in terms of the basis for what emerges, I 1
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 2344 433
99 l
i 1 you know, from what goes in comes out, that not using it 1 l
gy 2 as part of a hazard analysis. And after this
( ) 1 xs .
3 consideration I haven't really looked at it in that l 1
4 respect. That's a nice idea.
1 5 MR. MARSH: Are there any signs of dikes and ]
6 sills of similar age in the mountains around?
l l
7 MR. CONNOR: No.
I 8 MR. MARSH: No dikes and sills in the ;
1 9 mountaine? l l
10 MR. CONNOR: Not that I know of. If you go 11 over to Nye Canyon and Pahute Mesa and Piaut Ridge, there 12 pliocene dikes and sill complexes, but not at Yucca I
,_ 13 Mountain. For example, there is the 10.5 million year old I 1
\- ' 14 Solitario Canyon dike which is basically half a kilometer, 1 15 the southern end is about half a kilometer from the 16 repository. But's 10.5 million years old, not pliocene.
17 MR. MARSH: But, what did you say, Nye Canyon 18 or something?
19 MR. CONNOR: Yes.
20 MR. MARSH: The same age?
21 MR. CONNOR: No, actually I think those are 22 six -- six or seven, I'm sorry.
23 MR. MARSH: Okay, so there is no tendency that 24 style of emplacement of sills, for example, at this age of (3
( ,/ 25 one, two million, in the area.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WAF,HINGTON, D.C. 20005-3701 (202) 234-4433
100 1 MR. CONNOR: That we can see.
fx 2 MR. MARSH: Right. But that's significant,
- vI 3 also, the fact that you don't see dikes, transport dikes 4 in the mountains around, the structural highs. That may 5 be a feature to put in, a small tweaking feature to put 6 in.
7 MR. HILL: We don't see the -- what you are 8 saying the transport features at Yucca Mountain. But 9 certainly up in the Sleeping Butte complex the igneous 10 activities located on the topographic high in that region 11 and very clearly controlled by structure based on the work 12 of Scott Minor and his colleagues survey.
1 13 MR. MARSH: But there is still no sills
(s V) 14 anywhere?
15 MR. HILL: Not exposed, but we are still l
16 seeing the four and a half million year old original flow 17 surfaces preserved up there. A very low erosion rates and 18 uplift rates relatives to coming down 100 meters below the l
19 surface.
20 MR. MARSH: Why I ask about sills is that it 21 tells you something about the, perhaps, temporal storage, l
22 intermittent storage on the way to the surface. Sills are l
23 very, very common in complexes that are erupting in 24 intermittent depth. They are very easy to place,
(-
( ,) 25 especially in trains like this where you go from bedrock l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
l l 101 I 1 into an alluvial region and they are starting to lose i
-~ 2 their hydrostatics. It's very, very common all the way j 7 '
, v 3 through thousands of kilometers in Antarctica and all ;
l 4 over.
5 MEMBER HINZE: John, you had a question?
6 MEMBER GARRICK: Yes, one of the issues that's ;
7 always involved in probabilistic calculations is the issue 8 of the end states and a well set of defined end states.
9 Are these results that you presented in view graph number 10 19 associated with a specific definition of disruption, or 11 a specific definition of an end state? And is that 12 translated into a result in terms of consequence?
13 MR. CONNOR: Well, first I haven't presented
(~% 4 k/ 14 anything here taking this to consequence. This is just 15 hazard.
16 MEMBER GARRICK: Well you do talk about 17 disruption. So you --
18 MR. CONNOR: Well, right.
19 MEMBER GARRICK: So you must have some sort of 20 a definition in mind as to what constitutes the reference 21 --
22 MR. CONNOR: Right.
1
! 23 MEMBER GARRICK: -- for disruption.
i 24 MR. CONNOR: Right. So this would be an
(
(_,) 25 igneous intrusion which has a center, an orientation and a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
102 l
1 length to it. And that length is -- that half length is l l
w 2 500 or excuse me, 100 to 2,000 meters. So, and that (v) 3 results in disruption of the repository. Is that --
4 MEMBER GARRICK: Well, yes --
5 MR. CONNOR: Well, now the answers here change 6 depending on your definition of those events. .
7 MR. GARRICK: That's right.
8 MR. CONNOR: And PVHA, the different experts, 9 I think it's fair to say, use different definitions of 10 events. As I mentioned before, I'm not wed to any 11 particular definition of the event. I tried to develop 12 this one based on analogy where volcanologists have seen 13 these things form, you know, what the areas disrupted look V 14 like.
15 But, it's an important issue to explore and 16 again, we've explored it to a certain extent. We've 17 varied the length quite a bit and so forth. But, you 18 know, geologic analogy I think is the only way to go on 19 that. Right?
20 MEMBER GARRICK: Okay, well, probably be more 21 on this later.
22 MEMBER HINZE: Paul?
23 CHAIRMAN POMEROY: Chuck, we are going to be 24 comparing, in a few minutes, I hope, the apples and 73
( ,) 25 oranges that you referred to earlier. Ari I just wanted NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
103 1 to clarify, for the record essentially, this model that r~s 2 you presented is one of several. But was this model f
i v
3 presented to the PVHA experts? And was it available to 4 them for their consideration?
5 MR. CONNOR: Not entirely. This model was 6 presented to the PVHA experts and at that time we hand 7 folded this other information in. But I did present to 8 them, I guess in two talks, some information about the 9 fault dilation tendency, how I thought that fault 10 orientation and dilation tendency might affect the 11 probability of distribution in a qualitative way.
12 So, these maps were never presented. Numerous 13 tariations of this one were, and again, this idea of, or a
/ \
N h 14 concept on how to fold instructural control was.
l 15 CHAIRMAN POMEROY: Okay, thank you.
l 16 MEMBER HINZE: Chuck, there are many l l
17 conclusions I think we can reach from your presentation, 18 your excellent presentation. One of the conclusions that 19 I would draw is based upon the second bullet to your ,
l; 20 nineteenth transparency, you've got instructive control in l l
21 a factor of only two. I would conclude that the issue 22 related to the PVHA raised by John Trapp concerned with 23 the geological bases of zones can be certainly modified or 24 eliminated.
/~N l
(_,/ 25 MR. CONNOR: Actually I disagree with that 1
i NEAL R. GIUDSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. l (202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433 l l
j
104 1 because what I found is that if I take my -- this
,r- 2 conclusion refers to comparison of a non-homogenous model O] 3 with a non-homogenous model with structure folded in.
4 Many of the PVHA models, and Kevin can probably put a 5 number on this, put a barrier, or a zone definition 6 between Crater Flat and the repository. And what that 7 does is it essentially creates a step function or some 8 variation on the step function of high probability on the 9 west and lower probability on the east based on the 10 occurrence of a structure, or a structural zone of some 11 sort.
12 I don't see any evidence in the structural 13 data that I've got or the kinds of processing I did that
<,_,i
( /
14 you can decrease probability at the repository site based l 15 on structural control.
16 So what I'm saying here as it increases it, it 17 increases it frankly by a fairly trivial amount. l 18 MEMBER HINZE: That's what I'm saying. That i 1
19 it increases it certainly, but not by an amount that is --
20 MR. CONNOR: In comparison to the near 21 neighbor models. Now the event center, if you just look 22 at the event center, I believe that, at least in, I guess 23 it was the third PVHA meeting a number of models were 24 presented -- a number of histograms were presented. The es
(
(_ /) 25 probability of an event center at the repository was zero l
l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE fSLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 105 l 1 in many of those models. Is that correct? !
,x 2 PARTICIPANT: No.
! )
U 3 MR. CONNOR: Okay, no, what was it?
l 4 MR. COPPERSMITH: (Off mike. Inaudible.)
5 CHAIRMAN POMEROY: I'm sorry, but you can't do i l
6 this without getting to a microphone. Kevin can -- l l
7 MR. CONNOR: Well, Kevin will come up, but the 8 point is that many of the models have a step function j 9 across a zonal boundary. And I'm not particularly 10 satisfied with the idea or use of structural zones with 11 step functions across them in a probability surface. You 12 know, probably surfaces, you've got to be able to 13 differentiate them, I think. You've got to be able to --
/ T
\
'l 14 you have to be very careful about how you use that.
15 Certainly in this case, in my view, there is 16 no geologic evidence for such a zone boundary, even if you 17 wanted to use one. And hence, you know, structure does 18 not decrease probability. l l
19 MR. TRAPP: There is a very important point )
l 20 and I disagree with you very much, Dr. Hinze. Because if 21 you take a look at the models that we are dealing with, l l
22 what we are trying to do is concentrate on those models in '
23 which we are talking about an extrusive component. If you l
24 take a look at -- and I'm going ahead to the first slide l O 25 that Kevin will be talking about, you notice that his
(,/
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
106 l
1 models and the probability he is talking about have got l l
g- 2 both the intrusive and extrusive component. l 1
j
%.)
3 Therefore, an awful lot of his models 4 basically end up with a probability -- and it's really 5 just a probability, of a dike, not an extrusion. l 6 Therefore, you are talking a tremendous difference in 7 possible consequence when you take a look at the actual 8 numbers that you are dealing with. I've stated, as close 9 as I can figure, from the PVHA report, the comparable 10 number would be about 6 x 10-9 for their mean value for the 11 model which would compare to ours.
1 12 MEMBER HINZE: Okay, then another conclusion 13 which obviously I'm going to be incorrect about is that
( )
's / 14 adding another dozen hits to the aeromagnetic study or to ,
l 15 the ground magnetic study in Amargosa Valley or southern 16 Crater Flat is not going to have any effect, any l l
17 significant effect upon the probability.
l 18 Is that --
l 19 MR. CONNOR: I really don't want to speculate 20 about that, but of course if you add --
l l 21 MEMBER HINZE: Have you run sensitivity l
l 22 studies to determine how robust this is?
23 MR. CONNOR: By adding 10 or 20 suspected 24 events? No, I haven't done that yet.
(x- ) 25 MR. HILL: That's something we're planning on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344 433
107 1 doing fairly shortly to come up with a reasonable
,s 2 location, evaluate possible age scenarios and look at the ixj )
3 impact of those on existing probability models.
4 MR. CONNOR: I would -- I hate to speculate 5 about data that hasn't been collected, but in the interest 6 of moving forward, what I would say is that the 7 probability will, of course, increase if you add events to 8 the system and the probability at the repository site will 9 increase.
10 Of course, the distribution of those events is 11 quite critical to answer your question.
12 MR. MARSH: One question. In that context, is 13 that if you have more from the aeromag or the ground mag,
,-\
(x' )
14 if you have more areas that are possibly eruptive centers 15 and you have your probability normalized to one in this 16 area, wouldn't you draw the probabilities over to that 17 area and heighten it and reduce it with the repository or 18 do you increase the probability just for this geographic 19 region for the state as a whole? That way it's a little 20 bit artificial because they're not doing a ground survey 21 for the whole state also or the whole region. How do you 22 actually handle that?
23 MR. CONNOR: Well, the nonmodulous models, I 24 think, are advantageous because they don't require that
/ )
( ,/ 25 you define some region.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
108 1 MR. MARSH: Right.
2 MR. CONNOR: And that's nice. So if I find 7-)
V 3 some -- I mean obviously the regional occurrence rate may l 4 increase, if you find more events, right, because our )
5 estimate of the regional occurrence rate is based on past 6 events, so in that sense, of course, it will increase.
7 And of course, there might be some change in i i
l 8 the structure of the probability surface itself based on i 1
9 past events.
10 MR. MARSH: If you find more in one place in i 11 some area in your math, does it reduce the probability of 1
i 12 the repository or not?
l i
13 MR. CONNOR: Well, it would reduce the l
[) )
' 14 probability in some areas of the math and increase it in 15 others, definitely. What it does at the repository I 16 don't know.
17 MR. MARSH: Well, if they're far enough away.
18 MR. CONNOR: Yes.
19 MEMBER HINZE: Well, we thank you very much, 20 Chuck, and we'll take a modicum of blame for wrecking the 21 schedule of your presentation.
22 MR. CONNOR: Well, thanks, Bill.
23 MEMBER HINZE: It was extremely interesting.
l 24 Well, let's see, we're now at 10:15 and Tim St. .ivan and l
/~' l
(_)g 25 Kevin are up to discuss the overview and status of DOE NEAL R. GROSS ,
I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
109 1 activities and summary of probabilistic volcanic hazard
,r ms 2 assessment.
N.Y 3 Tim? We appreciate you and your colleagues 4 joining us today. We really will find it extremely 5 interesting and useful.
6 MR. SULLIVAN: So good morning. This is two-7 part presentation shared by Kevin and myself. Based on 8 what we've seen in the last couple of hours, I frankly l
9 don't expect we're going to catch up on any schedule.
10 (Laughter.)
l 11 CHAIRMAN POMEROY: We'll help you with that.
12 (Laughter.) '
13 MR. SULLIVAN: I'm going to give you an 7
( )
14 introduction and some background and the status of the 15 volcanism program at DOE.
16 I'm a geologist with the Department and have 17 been the lead for our vulcanism program for the last 18 couple of years.
19 You should have copies of the handouts. The 20 volcanism status report was issued in 1995. This was a 21 document produced by the project, by Los Alamos, that 22 provided the most comprehensive synthesis of data 23 available at that time.
24 I noticed it's not in your workbooks here, but p
k_, 25 I know the NRC staff has that and I'm sure the ACNW has NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
110 1 that as well.
,-, 2 It turns out that was the first of several i 4
~
3 synthesis or summary reports that the Department has 4 produced. The LBL part of the M & O has produced the 5 geophysics synthesis report which was referred to by Chuck 6 and John and I believe you have that as well. And the 7 USGS has produced the seismotectoric framework report 8 which is a comprehensive discussion of our tectonics 9 program and seismic hazards program.
10 An update to that report is planned for later 11 this year which will incorporate additional data analysis 12 which was completed in 1995 and 1996.
13 And then finally, at the end of 1997, a part (3
kl 14 of a larger document called the Project Integrated Safety 15 Assessment, the PISA, Chapter 2 called the site 16 description will be completed and this will provide an ,
17 integrated discussion here of tectonics, seismic hazard, 18 regional and site geology, geophysics and igneous 19 activity. This should be available to you early in 1998.
20 MEMBER HINZE: Did you say when the vulcanism 21 report --
22 MR. SULLIVAN: End of the fiscal year.
23 MEMBER HINZE: End of the fiscal year.
24 MR. SULLIVAN: Right. DOE decided early in
/~ l
(,g) 25 1996 to close out site characterization data collection NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHWJTON, D.C. 20005-3701 (202) 234-4433
111 1 for the volcanism or igneous activity program.
2 This was based on several factors. The low 3
i Q~J hazard probability results from DOE's earlier work and 3
4 from the PVHA which you'll see in a moment.
S In our view, the PVHA results are insensitive 6 to new data collection and we'll support that view later 7 today.
8 DOE performance assessment results, which will 9 also be reviewed in a later presentation, have 10 consistently shown that system performance or risk is not 11 sensitive to volcanism. In fact, in our view, volcanism 12 is not a key technical issue for performance at Yucca x
13 Mountain.
/N
- 14 MEMBER HINZE: So there.
15 (Laughter.)
16 MR. SULLIVAN: The Department recognizes its 17 responsibility to evaluate new information and to conduct 18 additional analyses as appropriate to insure that we 19 maintain a sound, defensible position in the license 20 application in 2002, 21 Additional consequence analyses are planned as 22 a part of an upcoming total system performance assessment i 23 and these will be discussed in a later presentation.
24 The main focus of our presentation today will i
(r' ~,)N 25 be on the probabilistic volcanic hazard analysis. This 1
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON. D C. 20005-3701 (202) 234-4433 I
112 1 was an expert elicitation that involved ten subject natter l
7
- 2 experts in volcanism that was conducted over approximately I
\- / !
3 a one year period.
l 4 The experts formulated their judgments based 1
5 on information provided by DOE, provided by the Center, l 6 provided by the State, provided by the USGC and based on 7 their own experience both in the U.S. and world-wide.
8 The final report was completed in June of 1996 9 and I presume you all have that as well.
l 10 Following that, in the fall of last year, we 11 held an informa) meeting with the NRC to discuss the 12 expert elicitation process that was used in PVHA and then j 13 in February we had a two-day technical exchange to discuss
, a
\~ ' 14 the PVHA results and we're going to discuss those again 15 here this morning. They're reproduced at the bottom there 16 and Kevin will provide a full description of those 17 results.
18 At the technical exchange conducted in 19 February, the NRC and the Center presented data and 20 analyses that were conducted after completion of the PVHA.
21 DOE agreed to evaluate this new data through hazard 22 sensitivity studies, specifically we have evaluated the 23 significance of the increased number of events associated 24 with Anomaly A and we have evaluated the significance of O)
( ,
25 the revised volume estimate for Little Cone through hazard NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
113 1 sensitivity studies. Kevin will present those results 2 here shortly.
3 / i
'~'
3 At the technical exchange and again here this l 4 morning, Chuck presented the basis for the NRC's 5 conclusion that the probability of future volcanic events 6 ranges between 10 to the minus 8 and 10 to the minus 7. i 7 As Chuck and John have acknowledged, the information that 8 they have used to formulate these models was also l l
9 available to the experts for their use in determining l l
10 appropriate source zones at Yucca Mountain. This range 11 differs from the PVHA results, but is included within the l
12 bounds of the full probability distribution.
13 It remains DOE's position that the PVHA
(~) ,
14 results, the meaning and the uncertainty distribution l 15 provided defensible basis for characterizing the l l
16 disruption probability.
l 17 At the technical exchange, DOE committed to 18 providing a description to the NRC of how we will use the 19 results of the PVHA in performance assessment and Abe will 20 provide that for you early or later this afternoon.
21 This should support closure of the hazard 22 probability subissue.
23 Finally, as I think we'll hear again later 24 this morning, at the technical exchange, the NRC and the
(_) 25 Center presented dose calculations for volcanic eruption NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
114 1 through the repository with associated tephra and f
-~ 2 radionuclide dispersion.
V 3 The risk result using the hazard probability, 4 the upper bound hazard probability of 10 to the minus 7 5 you heard described earlier, not DOE"s mean value of 1.5 6 times 10 to the minus 8 indicated an average annual dose 7 of a half a millirem per year.
8 We view this as a low result. We have also 9 concluded that the risk of volcanism is not significant.
10 It appears that DOE and NRC results are converging which 11 should lead to closure of consequence subissues.
12 Kevin Coppersmith, under contract to the 13 Department of Energy has been the manager and facilitator 7_ \
14 for the PVHA process. As I said earlier, this extended 15 over a year over most of calendar year 1995. He'll 16 describe both the process and the results of the PVHA 17 hazard probability results for you now.
18 MEMBER HINZE: Thank you, Tim. I think we're 19 missing at least I'm missing the last several slides that 20 you had and I'm sure the Committee and the consultants 21 would be interested in having a copy of those.
22 (Pause.)
23 May I ask a question in terms of the 24 clarification. Your second overhead or your third n
( m, / 25 overhead concluding the cover, the reasons for closing out NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
115 1 the same characterization, in no way mentions es 2 consequences. Is this strictly because zero times any
( \
GI 3 number is zero or why --
4 MR. SULLIVAN: In part, we have also closed 5 out the consequence analysis work. WE feel we have an 6 adequate basis there now to perform the consequence 7 analyses needed for TSPAV.
8 MEMBER HINZE: Are there any documents which 9 present the results of your consequence studies?
10 MR. SULLIVAN: Not in terms of dose 11 calculations, but there is consequence information in the 12 synthesis report that you don't have.
13 MEMBER HINZE: We will be getting that?
O 14 MR. SULLIVAN: Right, the close out of 15 Valentine's work is reported there.
16 MEMBER HINZE: I see. Is -- could you 17 summarize his work for us since we are not going to see 18 that for six months?
19 MR, SULLIVAN: I can't do that here, but I 20 could do that this afternoon.
21 MEMBER HINZE: Okay, great. Let's -- further 22 questions?
23 Okay, Kevin? That's Coppersmith with a C and 24 not a K. It's spelled phonetically on there.
f)y
(__ 25 MR. COPPERSMITH: Thank you. I appreciate the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
g (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
116 1 opportunity to be here. I do not intend, I always bring a
~s 2 large stack of vu-graph. I don't intend to go through all 8
~
3 of those. In fact, this ie very difficult to do. As Tim 4 said, I'm going to give a very brief overview of a large 5 multi-expert elicitation, involved, occurred over a period 6 of a year, involved multiple workshops, field trips, a lot 7 of interaction and discussion. The report gives an 8 indication of the level of detail of the study. All I can 9 do at this point is to give a basic overview of the 10 process and some of the conclusions and sensitivities.
11 And that's pretty much it. I'd be happy to answer 12 questions. l 13 Detailed technical questions, I feel like I
\~s' 14 should have a panel of 10 experts up here answering the 15 questions in terms of the details of their particular 16 models. We have, in response to the questions in the l 1 17 past, the technical exchange, for example, try to 18 highlight the interpretations that remain, for example, 19 the issues of structural control and whether or not l
l 20 faults, fault density was considered and we will be happy 21 to try to answer those types of questions today as well.
22 Maybe I should move to the other side. In 23 terms of objectives of the study, again, defining some 24 terms here to assess the probability of disruption by
(~%.
(,) 25 volcanic event of the proposed repository. Disruption is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
117 !
I 1 defined as the physical intersection of magma with the 1 l
73 2 potential repository volume. That can be viewed as the '
i i l 3 very tip end of an ascending dike as well as a hit, direct 4 hit by the repository. Any type of intersection of a dike l
1 5 with the volume represented by the proposed repository.
6 A second key objective is the quantification l l
l l
7 of uncertainty. The basic concept of having a multi-8 expert study is to characterize and quantify 1 i
i 9 uncertainties. So that's the focus not only in the 10 individual evaluations made by the experts, but in the 11 fact that we are using multiple experts to define the full l
12 probability distribution and as I'll talk about later on, 13 that's the basis for a belief that the results are l
.n 14 essentially robust to most types of new findings that 15 could become available.
16 Looking at both eruptive and intrusive 17 features, everything is couched in terms of the 18 probability here is an annual frequency, annual frequency 19 of intersection. This is as those who are involved in 20 risk who are dealing with the probability of intersection 21 which is a hazard probability, that will be multiplied 22 times the consequences given an intersection of various 23 types and that the product of those two will represent the 24 risk associated with igneous activity.
('T,
'q_) 25 At risk in the parlance of this Yucca Mountain NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
118 1 is the total system performance assessment. Some of the
-~ 2 aspects of the PVHA study, I just want to emphasize that LJ 3 the quantification of uncertainty is a central focus.
4 We're trying to capture all the important elements, 5 including the expert to expert diversity of 6 interpretation. We ask the experts themselves to note l
7 represent particular proponents of views, that the l
8 evaluators of alternative views and to apply weights to 9 alternative models. They're exposed to all the pertinent 10 data, to the research themselves and the methods for 11 conducting hazard analysis that were available at that l
12 time. This is an opportunity, obviously to use all the 1
! 13 Yucca Mountain data that have been gathered by the various I/N
14 groups as well as the their own experience bases and l
l 15 having done work in other volcanic fields throughout the i
16 world.
l 17 Ultimately though, the reliance that an extra 18 place is on a particular data set, say a geophysical data i
l l 19 set is a function of their own experience and their own l 20 prerogative to use the types of data that they feel are 1
- 21 important. I 22 They're encouraged to consider a full range of l l
23 methods and data. Many technical experts are used to 24 being proponents of particular views, particular models.
,/-
(_ ,/ 25 I have been in that position myself and I'll give a talk i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASH NGTON, D C. 20005-3701 (202) 234 4433
119 1 at American Geophysical Union about a particular model.
2 I'll be a proponent of that model. I'll have published
(~3
)
3 on that model. This is not the role that we ask these 4 experts to take. In some cases, we ask them to discuss 5 their published views. We're asking them to consider the 6 uncertainty in their model versus other models and to 7 apply weights to alternative models. So we're not looking 8 for preferred estimates or bounding estimates or 9 conservative estimates. We're asking them when they're in 10 doubt, between two alternative models or parameter values, 11 to not choose one or the other, but to weigh the 12 alternatives.
13 Throughout a formal process of expert
(,, )
14 elicitation was followed. Only for those that are in this 15 particular, interested in this particular area, I'll only 16 show these as examples of the technical guidance for 17 expert elicitation that were followed throughout the 18 course of the study.
19 NRC has published guidance of branch technical 20 position. The Department of Energy as well as combined ,
l 21 efforts for guidance and recommendations on how expert ;
i I
22 elicitation studies should be carried out, the process l I
23 followed the PVHAs is consistent with that guidance.
24 Members of the expert panel are shown here and ;
/~x i
( 8 I w_/ 25 I won't go into the experience and capabilities of these, i 1 l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234 4433
120 1 They're well acknowledged experts. I should point out that g~3 2 they include individuals who have a detailed knowledge of k.) 3 the Yucca Mountain data sets like Bruce Crowe. A majority 4 of the Panel have not had, did not come to the Panel with 5 an experience of Yucca Mountain. They gathered the 6 information. It was disseminated. They spent a lot of 7 time in workshops and field trips to come up to speed on 8 the particular data sets. The advantage of having the mix 9 of both is obviously the site specific experts are 10 familiar with the local data, but the regional experts can 11 come in with their experience data bases having worked in 12 other parts of the world.
13 CHAIRMAN POMEROY: Were these selected from a
,/ x
) 14 larger pool, Kevin?
16 CHAIRMAN POMEROY: And was the reason for 17 their selection based on essentially the FAC principles or 1
18 other principles that are laid down in some of that other i
19 guidance? l 1
20 MR. COPPERSMITH: yes. It's not only, we have i
21 individual selection criteria, but you're also looking for 22 a balance and disciplines of capabilities and institutions 1 1
23 throughout the Panel. So those come into play.
1 24 I wanted to point out though that this is a l r~N
() 25 very large study. I've been involved in many multi-expert NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344 433
l 121 1
1 hazard studies. This is one of the larger studies I've 2 been involved in and it involves not only a large panel, i , r~s.
I \ /
i 3 but also many individuals who participated as presenters l 4 in workshops or are actively involved in the field trip.
i 5 This is an example of some of those who are involved in l
l 6 work shop presentations, can just go down through the
! 7 list, people like Allin Cornell, experts in probability 8 and uncertainty characterization; people from the Center, 9 talking about the studies that they had done locally in 1
10 the area or Paul Delaney, talking about dike geometries 11 and distributions and so on. So a large number of people, j 1
12 at one point counted up on the order of about 75 1 13 individuals involved in one aspect or another of the l
'"N l l
' >)
~-
14 study.
\
l l
) 15 This type of expert elicitation process l l 16 derives much of its power from interactions of experts.
t 17 There are many opportunities for the experts to come 18 together in workshops and make presentations as well as a l
19 couple of field trips, and an opportunity to see field 20 relationships. To proceed through the process that's 21 outlined in the branch technical position as well as other 22 guidance on expert elicitation processes from discussions 23 of the data sets, observations in the field of those data, i
l 24 dealing with alternative hazard models, volcanic hazards 1
25 itself or probabilistic volcanic hazard analysis is not, NEA'. R. GROSS COURT REPOR ERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
122 1 does not embrace or involve all the aspects of basaltic
,f-)
2 volcanism that are possible and certainly the range of
('~) 3 expertise in this Panel goes well beyond what's needed to 4 carry out a PVHA.
5 So it's an opportunity to talk about things l
6 like nonhomogeneous spatial models, nonhomogeneous 7 temporal models and the data that are required to 8 characterize those.
9 Alternative interpretations are obviously what 10 we're after. We're looking for alternatives in 11 ir"erpretations of models that are consistent to various 12 degrees with the available data. Individual elicitation !
1 13 interviews were held. The feedback workshop to look at p) i.
'/
- 14 the preliminary interpretations and the calculated results 15 of those interpretations. We're trying to foster debate l 16 among the experts. We're trying to get them to examine the 17 technical basis for their interpretations and there is a 18 certain amount of technical challenge that we try to 19 encourage throughout that process, 20 I'll just show a couple of generalized logic 21 trees. The process that we're looking at for uncertainty b
22 characterization is essentially shown here. Alternative 23 models or alternative parameter values for a particular 24 aspect of the hazard model is shown diagrammatically with (3
( ,) 25 alternative branches in logic tree.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
123 1 Now I'll just step through some of the 7-sg 2 assessments that the experts went through to give a feel V 3 for the types of evaluations that were made.
4 The first aspect to deal with the concept of a 5 volcanic event, its length and azimuth. In this case, as 6 Chuck mentioned, the experts themselves were allowed to 7 identify or to define a volcanic event in any way that 8 they wanted to. That ranged basically from an individual 9 dike as represented by an individual cinder cone to a 10 collection or a dike set that would have more substantial 11 length to it. Usually event definitions took into account 12 the age of the particular volcanic center as well as its i
13 geometry. Proximity, for example, of two or three cones I )
\/ 14 that appear to be the same age that are say three 15 kilometers apart. There will be some probability of that 16 being a single event, as opposed to three distinctive 17 events.
18 So the length distribution, I can show some 19 examples of that, largely comes from dike geometry lengths 20 as well as dike set geometries. Azimuth is driven to a 21 large extent by the orientation of the maximal horizontal i i
1 22 stresses in this area, almost all the experts agreeing -
23 that the predominant azimuth in the northeast quadrant.
24 The types of temporal models that are
/' \
( ,) 25 considered, the homogeneous types of models or NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHOC5 ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 124 l 1 nonhomogeneous models, most of the experts preferred the gx 2 homogenous type model over different type frames. In v
3 other words, events would be counted and used for a rate 4 calculation over a particular time period, say the lat 5 million years, the last 5 million years and so on. It 6 would be assumed that that rate was homogeneous over these 7 different time splices. But of course, that's start time, 8 that time period of interest is uncertain and there would 9 be various probabilities assigned to a particular time 10 period.
11 There were, in the case of Rick Carlson, used 12 a volume predictable nonhomogeneous time dependent model, 13 taking into account the waning volcanism over the last --
/
i
\
14 over the post-10 million year time period. The time 15 period of interest, I wantioned uncertain and given a ,
l 16 particular weight across for each individual expert, the l 17 region of interest was that this is the overall area that ;
1 l
18 would be considered to be the area of interest for the ;
1 19 PVHA analysis. In general, the Amargosa Valley as a topic 20 province, that Gene Yogodzinski will talk about in a 21 minute was used as that region of interest. In other 22 words, it's an area within which the volcanic hazard j 23 begins. It's this area that may be the overall geometry 24 of the source at depth within the lithosphere, it may
(,/ 25 represent an area within which just simply a spatial l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l l (202) 234-4433 WASHINGTON, D.C 20005-3701 (202) 234-4433 l
l
125 1 background rate of volcanism is applied. The question
-m 2 came up earlier, what is the rate density of volcanism for V 3 this area. In general, they would start with the region 4 of interest and then segment areas within that that might l
5 have a lower or higher rates.
1 6 Spatial models. In general, there were i 7 several spatial models that were aesumed. One was a 8 zonation model where areas are identified and those areas )
9- identified have a rate density of a number of events per 10 year that comes largely from the counts of the observed !
11 volcanic centers over various time periods.
12 The zonation boundaries are uncertain and that 13 uncertainty was characterized either by alternative zone
\--
14 geometries or by a boundary that had essentially a fuzzy 15 boundary that allowed for uncertainty in the location of 16 the boundary.
17 Another approach, the spatial aspect, kernal 18 smoothing, which is basically running the smoothing 19 operator of th observed pattern of volcanic centers.
20 These type of approaches that were advocated and discussed 21 by Chuck Connor were used to a large degree by these 22 experts with some changes. For example, they would have 23 changes in the actual kernal itself whether or not they 24 used Epanechnikov of some sort of Gaussian smoother, The l'h
() 25 smoothing distance over which the smoothing occurred was NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
126 1 uncertain. It could be variable, and other approaches 73 2 which allowed for smoothing within particular source
( )
A. /
3 zones.
4 Another zonation or another approach which was 5 used to get at the spatial aspect was advocated by Mike 6 Sheridan, used by a few of the experts. He's gone around 7 the world and looked at the pattern, the geometry of 8 volcanic fields within basaltic volcanic fields and sees 9 that in general the geometry of the field shape is 10 parametric. It tends to follow by Gaussian distribution 11 and that type of field shape was used by some of the 12 experts to define the greater flat region and to define 13 its characteristics.
()
'-- 14 So these characteristics of the various 15 spatial models are important. Volcanic hazard, like 16 seismic hazard and other hazard is a function of two 17 things. Where things occur, the spatial aspect and the 18 frequency with which they occur, the temporal aspect.
19 So we are looking for as much uncertainty 20 characterization in those two as we can. We want 21 alternative spatial models and their uncertainties, 22 alternative temporal models and their uncertainties. Much l 23 of this deals with the spatial part of the problem, in 24 addition to the particular temporal model that is used. ]
(3,) 25 When it comes to characterizing the temporal models, much I NEAL R. GROSS j COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE.. N.W. l (202) 234-4433 WASHINGTON. D C. 20005 3701 (202) 234-4433 l
l
127 1 of that comes from counts, event counts within a
,s 2 particular region. And it's just shown diagrammatically
~'
) 1 3 here and move around across all of the various centers, 4 Lathrop Wells, Northwest Crater Flats and so on and 5 develop an uncertainty distribution in the number of 6 counts, event counts and using their event definition at l
7 that particular location.
8 Going back a little bit, some of -- these are 9 all source specific. We're dealing with a particular lo source zone. They become specific for spatial smoothing.
11 They're also used to define for a vivariate Gaussian 12 parametric shape, the number, the pattern of these centers 13 on the map can help define the location axes of that t
'd 14 distribution.
15 The issue of abrupt and gradual boundaries, it 16 became clear after the first round of the elicitations 17 were over, we did calculations and we went back to the 18 feedback workshop and identified key issues. One of the 19 key issues was the nature of the boundary between Crater 20 Flat and Yucca Mountain. That's an important 21 characteristic . Uncertainty needs to be quantified in 22 that assumption and so this is an area. We gave them 23 various options for dealing with the nature of the 24 boundary. I would say in all cases uncertainties were r's
( ,) 25 characterized in the nature of that boundary.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
128 1 We can talk more about it, but there's many 2 ways this is done. For example, the source zones that are
!n 4 LJ 3 identified by a particular expert represent the pattern or 4 the rate density difference between that zone and areas 5 outside of it.
6 Within that zone, particular events occur that 7 have a geometry, have a particular length and orientation 8 and those lengths, as I'll show in this slide range up to 9 10 kilometars or more in length and these, for example, 10 here's the maximum dike length shown here and the 11 orientations. This is again, an integration across all of 12 the experts.
13 These events are allowed to extend out of l lO t 14 those zones into adjacent regions. All the experts i
15 allowed that to happen. So in fact, much of the hazard 16 that occurs here in terms of the number of the frequency i 1
17 of intersection of dikes with the repository comes from 18 dikes that occur, whose centers are in greater flat of the l I
19 extent of the dike, but allow an intersection within the 20 mountain block itself.
21 MR. MARSH: Now these are just estimated dike 22 positions? These are not observed? these are not based l 23 on observed orientations of dikes in the area today?
24 MR. COPPERSMITH: that's right. There's no f3 i ) 25 dikes --
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 2344433 l
l 129 1 MR. MARSH: There's very little seen except l r's 2 for some of the old --
.w]
3 MR. COPPERSMITH: Basically, it would be like 4 voting on what you would think the distribution would be.
l 5 You can see the dike orientation, for example, 6 it's based almost entirely on in situ stress measurements 7 coming from the focal mechanisms of earthquakes, in situ 8 hydrofrac measurements. There are a number of stress 9 indications in the local Yucca Mountain area, has been the 10 focus of study by the program.
11 MR. CONNOR: Could I make a quick comment on 1
12 that, Kevin? ;
/_si
\~'/ 14 MR. CONNOR: This is Chuck Connor. First you )
1 15 see the dike in Solitario Canyon which is 7.5 million !
l 16 years old. that's in the Solitario Canyon fault and also 17 in pliocene Crater Flat which is very near surface, but 18 you can map dike segments in the near surface .in pliocene 19 Crater Flat. That includes numerous dike segments in a 20 zone that's 750 meters wide, that kind of thing and 21 extending for someth'ng like 3 or 4 kilometers through l
l 22 pliocene Crater Flats.
l l
23 So those are really the two dike sets closest 24 to the repository that are best exposed. they both have
/O F \
() 25 north trends.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
130 1 MR. COPPERSMITH: Most of the younger centers,
,f- 2 the dikes are not observable.
(_3 ) 3 MR. MARSH: But that data isn't in here, 4 right? That's just used to make the opinion, right of the 5 experts?
6 MR. COPPERSMITH: Sure. These are their 7 judgments based on available data.
8 MEMBER HINZE: Kevin, if I can interrupt you 9 for just a moment. Going to the NRC's concern about the 10 PVHA and specifically with the geological basis of the 11 zones, and you've processed models which have abrupt and 12 gradual and also there are some models that include Yucca 13 Mountain. My recollection of all the experts at least --
O
'- 14 MR. COPPERSMITH: All the experts allow for 15 finite probability of volcanism or the intersection 16 probability to be is finite. It is above zero. So in all 17 cases the site lies within a region that has a finite l
l 18 probability of being hit by a volcanic dike.
19 The issue is rate density.
20 MEMBER HINZE: Did any of the experts include 21 Yucca Mcantain in their -- at the same rate density that 22 they did for Crater Flat?
24 MEMBER HINZE: And would those results of
/~T
( ,) 25 those processing, give us a clue as to the degree of NEAL R. GROSS COURT REPCRTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1 131 I 1
1 concern that we have to have with the basis of the zones? l 1
l gs 2 MR. COPPERSMITH: Well, to me when I go back l i )
\_/ ,
3 and I look at and I think ultimately all the experts i 4 should be involved in this discussion, perhaps in l
5 licensing in a few years, if you look at the basis for ;
I 6 their interpretations of their interpretations of the 7 spatial distribution of future volcanism, however they did 8 it, parametric field shape, spatial smoothing, source 9 zones. It was based on a combination of things. The 10 combination of their knowledge and experience in 11 association of various types of structural elements, 12 either large deep-seated structures or shallow faults and 13 the observed distribution of say post-500 million year
~- 14 centers.
1 1
15 For example, the issue of association with 16 volcanic events and faults is obviously known by and it 17 was a basis, there was a lot of discussion about that 18 topic. Some of the individuals, George Thompson, Alexa l l
l 19 McBirney have published on this particular issue. It's l 20 something that obviously they're aware of. l i l l
i l 21 There was a lot of discussion about the deep l
]
! 22 seated structure, the structural trends, Walker-Lane l
23 trend, other more northwesterly trends versus shallow 24 seeded structure, what happens during the assent of magmas
,in
(,) 25 in the upper few kilometers of the crust, the co-existence NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
132 1 of extentional faulting with basaltic magneticism, j s 2 obviously it's an issue for everyone who has worked in the
<-l 3 basin and range.
4 I think, in general, though the driver, I i 5 think everyone is aware of the fact that there are normal 6 faults in Yucca Mountain and there are inferred faults 7 beneath Crater Flat. I think the driver though in terms ;
8 of the rate difference between what's going on or is 9 predicted for it to go on in Crater Flat versus Yucca ]
10 Mountain is what's happened over the last 20 million 11 years.
1 12 MEMBER HINZE: Is the results of this PVHA l 13 consistent with what we heard from Chuck Connor that the l
/T l
\' '/ 14 affected structure would change the probability by a 1
15 factor of 2, roughly 2? l l
l 16 Is that --
17 MR. COPPERSMITH: I don't understand the 18 notion because I know that almost all of these models are 19 based on structure. The issue is what scale? Are we 20 dealing with individual fault will be the location of the 21 dike or are we dealing with regional scale structure? I 22 think all -- I think everyone involved in this issue of 23 small volume basaltic magmas is aware that regional 24 structure plays a role, but it's an issue, I guess, of how
,r m,
(_,i 25 it plays into the probability models themselves, to be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
133 l
1 able to say it's a factor of 2 or 3. l 1
l rx 2 My feeling at this point is that those issues,
)
N_/
3 I think you can boil this all down to, we're dealing with 4 the rate, and I think Chuck was trying to say the same l
1 5 thing, the location of events and the rate. In order for !
6 the results, the probabilities annual frequencies of l 7 intersection to be significantly different, we need to 8 find an order of magnitude more events or we need to have 9 a much higher rate density in the site area than has been 10 - than has been assumed before.
11 MEMBER HINZE: In the ra11ge of the experts' 12 view about probability, what was the -- was there a 13 primary factor in all of them that made the spread? And
/ \
' )
14 if so, what was that or what was the view, either a number 15 of different things, dike orientations, etcetera.
16 What gave the tails?
17 MR. COPPERSMITH: the tails? I can tell you 18 what drives the mean.
l 19 20 MEMBER HINZE: What gave the range?
21 MR. MARSH: Sensitivity analysis here, the 1
22 parameterization would really be useful. j i
l 23 MR. COPPERSMITH: Yes, we have in the -- we i 24 have about 40 or 50 pages of that with plots in the l
(~'N I
(_) 25 report. Basically, the issue deals -- it's fairly simple. l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
134 1 You've got the spatial models which deals with the
,ew 2 location of events. Where will future events occur.
V Right now, most of the experts believe that where you see 3
4 them is where you're going to get them with uncertainty.
5 Okay, so the fact that you've got distributions over the 6 last three and a half million years or so or maybe the 7 last 5 million years including Amargosa Valley in a 8 particular area, that's where they're expected, that's 9 where the rate density is highest. That's where they're 10 most likely to occur.
11 That is uncertain and the uncertainty in that 12 is either couched in terms of variations in that geometry 13 or variations in uncertainties in the length of dikes
/_.x
( )
14 themselves as they might extend in Yucca Mountain. That's 15 one issue. the second is the rate issue, the frequency 16 issue. How often, what's the rate per year of volcanic 17 events in this area? That's where event counts are used 18 and there's uncertainty in those event counts. One of 19 those uncertainties is maybe we don't have all the events 20 identified. And this was called the hidden event factor, 21 essentially, how many more might we find and in general ;
1 22 you can see people will add ten to 50 percent more. In 23 other words, take that event count, multiply it by 1.1 or )
I 24 1.5. That's to account for the fact that number 1 in i
/^' f
( ,s/ 25 Amargosa Valley, we have quaternary fill that might be l l
NEAL R. GIUDSS '
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE . N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
135 1 hidden beneath that. I have the aeromag information they fm 2 had at that time. Or lost underneath the actual lava i
U 3 sheets themselves or otherwise not identified these 4 individual events.
5 So that is the rate part. And I think when we l 1
6 look at what drives the hazard right now, it's that where 7 you allow the future distribution to be, and what that 8 rate will be, so you need a significant difference, j 9 significant change in those two.
10 Someone asked the question about the rate 11 density in this area per square kilometer for a typical 12 volcanic field. This has very low event counts. It's been 13 pointed out for a long time the average recurrence N-] 14 interval here is about 200,000 to 500,000 years between 1 15 events. So this has a very low rate density and that's 16 the problem, right? If we had a field with 1,000 events i
17 on it, we'd have much higher confidence in the spatial 1
18 distribution and the frequency. We have very low events, 19 so that's always been the paradox of this particular 20 field.
21 So we have, I would say that the types of l I
22 things that need to change radically would be much higher !
23 rate density in Yucca Mountain which means that the 24 spatial distribution would need to change significantly r~x ]
( ,) 25 from what's been judged by this Panel and the rates need l
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
136 1 to go up significantly also.
-~s 2 So we must -- and I say significantly, for i
- V 3 these types of models, factors of ten, that we haven't 4 seen. So I think Chuck is right when he says this change 5 is by a factor of 2. That's right, and that 2 is on the 6 10 to the minus 8. So when we deal with -- I deal with a 7 lot of work in the seismic area, we're dealing with 8 usually annual probabilities of exceeding some design 9 ground motion of 10 to the minus 3 or 10 to the minus 4.
10 these are on the order of 10 to the minus 7, 10 to the 11 minus 10 is the range we have for this.
12 MR. RYAN: Kevin? A perspective on this 13 figure 362 with respect to the dike orientation plot and
/ s V- 14 the maximum dike length plot, both of those complete 15 spectrums are often observed in the same single volcano.
17 MR. RYAN: So it's in one sense --
19 MR. RYAN: Can be interpreted as variation 20 within vis-a-vis variation between --
21 MR. COPPERSMITH: Exactly. I hate to use the 22 words, but m going to half to Paul. These are the 23 dif f erences between aleatory and epistemic uncertainties.
24 CHAIRMAN POMEROY: I knew that would get in i
v
) 25 there.
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1 1323 RHODE ISLAND AVE , N W. i (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 j i
j
137 1 MR. COPPERSMITH: Yes. They were told that.
,- 2 The first would be the randomness, the aleatory component
~'
3 that is allowed and you'd expect it for a given set of 4 events. The second is their uncertainty in what that is.
5 We separated the two, the epistemic part goes in the logic 6 tree, the aleatory, the random component is embedded in 7 the hazard directly.
8 I had to say it. These issues of aleatory and 9 epistemic are ones that the seismic experts are learning 10 about daily.
11 MR. COPPERSMITH: Let me just show some of the 12 results. Like I said, the PVHA report has much more 13 sensitivity than I am able to show here. This just gives e s
(
\' ) 14 a feel for -- these are all, again, the annual frequency 15 of intersection of a volcanic dike with the repository for 16 each of the experts. And in general, you can see that 17 their range -- the reason these are discrete values is the 18 logic tree is a discrete process. There are particular 19 paths in that tree that also is associated with 20 probability. So if we say we deal with a dike that is 6 21 kilometers long, is oriented north, 40 degrees, east and 22 it's within this particular source zone, it has this rate 23 density of occurrence and that leads to this probability 24 of intersection. That's one particular path associated
/^
( ,%) 25 with a probability which is the product of all those NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l l
138 1 branches leading to that point.
f- 2 When you put those all together, these logic (3 )
3 trees have tens of thousands of branches or more that take 4 into account their uncertainty in all of the spatial and 5 temporal components. So for a particular expert, the 6 amount of uncertainty that's captured is on the order of 7 about two orders of magnitude on the annual frequency of 8 intersection.
9 This gives a feel for that within expert 10 uncertainty.
11 MEMBER GARRICK: How independent were these 12 arrived at in terms of the individuals?
,\
13 MR. COPPERSMITH: They are not independent.
/
~' MEMBER GARRICK: They are not independent, but 14 15 obviously they weren't independent in the sense that the 16 panel discussed the results, but were they independent in 17 the actual construction of the distributions in terms of 18 the actual construction? Did they go off by themselves 19 and --
20 MR. COPPERSMITH: There were individual 21 interviews with each of the experts, but then we had a 22 feedback workshop where they shared their preliminary 23 assessments and then they were finalized after that.
24 MEMBER GARRICK: Now what happened between (Q
i
,/ 25 those last two steps? Was there much convergence?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WAShlNGTON, D C. 20005 3701 (202) 234-4433 1
139 i
l
r~x 2 at all. Actually, that's been fairly common where we've l
l 3 carried out a feedback process, it's usually relatively l
4 small amount of actual change that occurs. There's much 1 l l l
l 5 more that goes on in terus of the thought process and l
6 ultimately much better documentation of the technical 7 basis for the assessments.
8 They're challenged. For example, when someone 9 stands up there and George Walker will have an event, a 10 dike length distribution that goes up to 55 kilometers and 11 George will say, are you kidding, for these low volume, 12 what are you talking about? George will say well, I saw 13 it in Iceland this is what I saw there and then he'll (si 14 think about his weight, whether or not he wants to give 15 that lower weight, higher weight, whatever. But there's a 16 technical challenge that goes on in that feedback process.
17 They then go back and finalize on their own, 18 their assessments. The old issue of they're either 19 completely independent or as interdependent as you can get 20 them, in between is where you don't want to be. This is a 21 very interactive process.
22 MEMBER GARRICK: What information do you have 23 in the report that indicates the supporting evidence for l
l 24 the distribution by individual?
25 MR. COPPERSMITH: There's an elicitation l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 2344 433 WASHINGTON, D C. 20005-3701 (202) 234-4433
140 1 summary that's given by individual and signed by each 7- 2 individual. It's their documentation.
N )3 3 MEMBER GARRICK: Yes, is it clear in the 4 process what the individual's distribution is based on?
5 MR. COPPERSMITH: Yes, they do not calculate 6 the distributions. They provided all the input to those.
7 They gave us the model to do the calculations. They see 8 the resultc. It's a -- all of the input components in 9 terms of the alternative models that were used and so on 10 are explained in the report.
11 MR. MARSH: Is there a simple explanation that 12 you've looked through sensitivity analysis of why they 13 differ like they do in the mean?
\
)
15 MR. MARSH: Can you give a few significant 16 factors? j l
18 MR. MARSH: What are they?
19 MR. COPPERSMITH: In general, they have to do 20 with, I'll be going into each individual, but they have to 21 do with the spatial distribution of what they think is 22 going on in Crater Flat versus Yucca Mountain area and the 23 rate density, the number of events per year that they 24 assume will occur. And thirdly, the event geometry, what
('
( )N 25 is the length and orientation of events.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
141 1 MR. MARSH: So it's basically a difference in fs 2 philosophical approach to the weights you give to the i t V 3 three things?
4 MR. COPPERSMITH: That's right. For example, 5 George Walker, in event geometry, had a bimodal 6 distribution. He felt that you have 50 percent 7 probability that it would be in the northwest and the 8 northeast sectors and it was aleatory that was the actual 9 range that he expects to see. A bimodal distribution 10 given 1000 dikes. You go around and count, half of them 11 have this distribution and half have the other.
12 There's a different probability of 13 intersection given those two cases, given the geometry of k- 14 the particular center. So those could all be dissected 15 and what we did in the study was to show for a given dike 16 geometry or dike azimuth or hidden event factor or event 17 counts at some particular, Crater Flat, and so on, the 18 variation due to different time periods that are assumed 19 or different spatial models that are used, we show how the 20 results change for those different models.
21 MR. RYAN: It's entirely possible that 22 George's response to that was influenced by his mapping of 23 the Kulao volcano which, in fact, shows that kind of 24 bimodality.
gs
(,) 25 MR. COPPERSMITH: He gave elicitation summary, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
142 1 he gave three or four examples. I've never -- in his s 2 elicitation over the course of a couple of days we must i r 3 have visited 30 or 40 sites around the world in vivid 4 detail. The issue of rafting, he can give you an example 5 of when he got on a rafted piece of lava and observed 6 velocities and everything else. It's an amazing group of 7 people who looked at -- represented over 300 years of 8 cumulative professional experience. It's quite a group.
9 MR. FOLAND: One point of clarification to 10 follow up on Bruce's question, the opposite of that 11 effect, the models that turn out to have high probability, 12 are there some factors which are common among the various 13 experts which might lead one to think that there's 14 something systematic in the high estimations?
15 MR. COPPERSMITH: I would say their event 16 counts are very similar to each other. They range from say 17 -- total accounts in this area may be 8 to 20 events and I 18 think that's a relatively small, maybe a factor of 2 to 3 19 variation in event counts leads "o essentially the ;
20 frequency part of this is very comparable.
21 Also, in general, have a fairly high reliance 22 on the spatial location of observed centers over the last i 1
23 5 million years. Most of them use the last 5 million f l
24 years of record. They do not want to go into the miocene l t"' l
( ,N) 25 aspect when you had calderas and silicate volcanism stage.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
- 143 1 The tectonic history tells me to stay out of that and to l
7s 2 begin at 5 million years.
\
V 3 So they're basically using the same time 4 period, the same number of events and in general, the same 5 overall distribution of events that again comes from the 6 observed centers. I think that's what drives the mean.
7 So to get the mean to change those are the things that 8 would have to change significantly.
9 MEMBER HINZE: Did you have a major change 10 from abrupt to gradual boundaries?
11 MR. COPPERSMITH: We had, for those who went 12 to the feedback workshop, we spent a day on this issue of 13 what is that boundary abrupt or not between Crater Flat
/ \
14 and Yucca Mountain.
15 The difference is in belief. What you're 16 dealing with in abrupt to gradual is you're saying how 17 does the rate density of events change. It's similar to a 18 seismic source zone in that sense. Within this area you've 19 got an A value that is a certain amount and in this area, l 20 there's an A value that has a certain amount. What l
21 they've done here though is to allow for events to extend i
l 22 out, number one. Their event geometries, their average, 23 their mean estimates of event geometries are 2 to 5 24 kilometers. You saw the maximums go up into 10 to 20. So r3 k) 25 they all allow, number one for it to extend out. That's HEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 2344 433
144 1 like the uncertainty in the epicenter. And secondly, they
(-
%J 2 explicitly, some gave uncertainty in that boundary either 3 through like Bruce Crowe, here's a bunch of alternative 4 source zones or like George Thompson, here's my 5 uncertainty, my fall off. I've got a 5 kilometer wide 6 zone where that boundary could actually be. So if they !
7 explicitly dealt with that issue, now but they do believe 1
8 that, in fact, because when you go up to Yucca Mountain, 9 the last 10 million years there have been no observed 10 volcanic centers. In a bedrock environment you should be 11 able to observe them. In fact, we do see one up at 12 Solitario Canyon, at 10.5 million years, versus Crater
,_ 13 Flat where the action has been.
I
\,-)
14 Now when it comes to details then, well, j l
15 there's faults at Yucca Mountain, couldn't that localize 16 volcanism? Yes. But it hasn't. It hasn't very often, so !
l 17 they allow for finite probability event to occur. All the 18 site lies within a zone for everyone. There is a finite 19 probability, but it's low and the reason it's low is 20 because it hasn't in the last 10 million years.
l 21 So I think it basically boils down to their 22 prerogative to the weight they'd like to put on that l
23 particular model.
24 CHAIRMAN POMEROY: Kevin, before you take that g
(_) 25 away, can I just ask a question kind of somewhat for the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N W.
1 (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
145 1 record?
2 I've heard the statement made informally that
%J 3 the range of probabilities resulted from one or two 4 people's particular opinions or derived opinions. My 1
5 interpretation when I looked at this slide a while ago was 6 that that wasn't true. Is that a fair -- !
l 7 MR. COPPERSMITH: Number one, you have to l
l 8 think of the personalities. If you've been through these ,
9 before. These people easily are swayed by one or two 10 dominant personalities. I've never seen a panel of more 11 dominant personalities. They are all very sure of i
12 themselves. In fact, we were warned beforehand that hey l 13 watch it, this is a contentious bunch. And in fact, they
/\
\ >
14 are very confident. I see no evidence that they were 15 swayed by a particular view in an undue way. I think they 16 are obviously all of us if we're sold by the technical 17 arguments for a particular idea, we will agree with it.
18 For example, at some of the methodology issues, the 19 spatial smoothing, for example, many of the technologies 20 for doing that didn't exist before Chuck did his work.
21 And when it was presented, they said this is a good way 22 because they do derive a lot of comfort from the pattern 1
23 of post-5 million year centers. Well, here's a way to use 24 that. And so that spatial smoothing process was used.
m
(,) 25 Then they put some changes on it and said maybe I'll draw I
i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (2P2) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
146 1 a source zone and say that 90 percent of the probability
,3 2 density lies within it, but will allow 10 percent to trail
( )
~
3 out. And within that I'll do special smoothing to get the 4 local probability distribution.
5 I think that things like that, you could say 6 well they all jumped on the smoothing band wagon or they 7 all jumped on the Amargosa Valley isotopic provence, not 8 without consideration. I think they thought about the 9 issue and there are certain ideas that they've felt 10 comfort in.
11 I think the dike orientation in the northeast 12 quadrant, they're swayed very much by the pattern of and 13 it was discussed of the maximum horizontal compressive
'/
- 14 stresses that are a whole bunch of stress measurements 15 that have been male in the area.
16 The next vu-graph just summarizes that 17 probability distribution. This is the PVHA probability 18 distribution function, the PDF that ranges from 10 to the 19 minus 7 up to 10 to the minus 10 per year. You can see 20 that that full range of about three orders of magnitude is 21 somewhat more than any particular expert which is on the 22 order of 1.5 to maybe 2 orders of magnitude. That's l 23 obviously the expert to expert uncertainty.
! 24 In general, we were looking at about a third n
( ,) 25 of the uncertainty coming from expert to expert l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
l l
147 1 differences, two-thirds from within, what's called from 7s 2 within expert differences and that also is not uncommon
('- 3 for these types of studies. Particularly if we're dealing 4 ,with say more than five experts or so.
5 But that probability distribution then is 6 something that has a mean value and has a number of 7 parameters that describe it. But I think when we deal 8 with changes then, we should be looking at what would 9 change this distribution and particularly since we know 10 the risk is often driven by the mean value, what would 11 change the mean value of that probability distribution.
12 In the discussions that Chuck was having, the l
13 10 to the minus 7, 10 to the minus 8 area, falls within
\
[I Y- 14 this distribution. But the mean value is about 10 to the l
15 minus 8. j i
16 MEMBER GARRICK: You say the risk is driven by )
17 the mean values, but that's if you're using mean values as 18 your propagation parameter?
19 MR. COPPERSMITH: Yes, exactly. Again, I've 20 bene involved in PRAs for seismic and I do hazard 21 analysis. We provide the full hazard distribution that 22 can be broken down in fractiles. That would be the intent 23 here and I guess Abe will talk a little bit about the 24 sampling, but right now we start with PDF and then there's A
( ,) 25 interaction that goes on with the risk analyst in terms of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
148 1 what components they feel of that PDF will be most 73 2 important to risk. I'll try to stay out of that. I know U 3 a little bit about that problem.
4 Just in general, and conclusions, really a 5 heavy emphasis on uncertainty characterization. That's 6 really what we're after in this type of study. We tried 7 to get them to consider as many alternative models, 8 spatial and temporal as possible. This summarizes the 9 components of uncertainty. These are the significant 10 contrf ors to uncertainty. One of the other I didn't 11 talk aDout, but one of the other sensitivity analyses that 12 are done, one is just to turn knobs and looks at the 13 variation and the mean result or in the full probability
'- 14 distribution. The other is to look at that total PDF and 15 look at what contributes most to the uncertainty, what's 16 the contribution to variance in the PDF and we've done 17 that and looked at basically the rate parameter and the 18 choice and spatial model, some of these issues are the 19 ones that drive the uncertainty characterization. So this 20 is the description of that PDF.
21 Now what we've done recently is to evaluate 22 some of the new data that Chuck talked about and I have to 23 begin with a premise that the PVHA was focused on the 24 knowledge and uncertainty and the frequency of dike
,/ m
( ,.) 25 intersection. It's believed to be robust and quantitative NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
149 1 and then we'll now look at it as new data or gathered,
,r~. 2 we'll look at their effect, the significance of those data G
3 to the PDF is the important part of the problem.
4 One of the reasons that multi-expert studies 5 are done is so that they will have some robustness to 6 them. Rather than have an individual expert give you a 7 single value which is subject to change with whim or with 8 the addition of new data, these individual experts 9 provided a probability distribution themselves. Across a 10 panel we have still broader probability distribution.
11 So we're hoping for an inherent robustness 12 that's come from this and again multi-expert studies are 13 usually motivated by the need for something that will be 14 somewhat long lived, some of the seismic hazard studies 15 done for the Eastern U.S., for example, were done in order 16 to calm things down and to characterize uncertainties that 17 would be robust for some period of time.
18 They were given all the -- devoted a lot of 19 effort looking at the existing data, testing alt'enative 20 hypotheses and so on, so now we come to new data that have 21 been gathered and we need to look at their influence on 22 the PDF. I would say that in general we should be looking 23 at changes in the mean of that distribution, but we can 24 look at recalculation of the full distribution, if needed.
(p_,) 25 New data as they come up, first we look at NEAL R. GROSS COURT F.EPORTERS AND TRANSCRIBERS 137.3 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
150 1 their implications to PVHA. There are geologic data, even
,-)
, 2 volcanic data that can be gathered that, in fact, don't
\'~] 3 have any influence on the input parameters to PVHA, 4 compare it to the assessments made by the experts and then 5 do a quantitative assessment or calculation and compare it 6 to the PDF.
7 So the two new data sets that we looked at and 8 were discussed earlier. There's increased volume estimate 1
1 9 of Little Cones that was based on the ground mag that l l
10 Chuck Connor talked about and an additional very volcanic 11 feature in Amargosa Valley, again based on modeling on
.2 ground magnetics.
13 In terms of the volume at Little Cones, it is 6
\- 14 included as a potentially significant. It does affect the 15 assessment that one of the experts made. He used the 16 volume predictable approach to the temporal aspect of his 17 problem, Rick Carlson, and therefore the volume, the 18 integrated volume within Crater Flat which was used to 19 make an assessment of the time periods or the frequency of 20 occurrence of volcanism within Crater Flat could be 21 affected by the revised volume estimate.
22 The bottom line when you go through that and 23 make the change is a very small change in the total volume i
l 24 estimate which is the way that he carried the problem gs l
(j 25 through and results in a very, very small change in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
151 1 mean annual frequency of intersection.
f~s 2 In the case of the Amargosa Valley anomalies, i
s /
\
3 I need to clarify a little designation of what we're 4 talking about. This aeromagnetic map which may be a very 5 poor copy in your packet, it's the aeromagneite map that 6 was developed by Jean Langenheim at the U.S. Geological 7 Survey and she presented it at two workshops. It was 8 provided to the experts for their consideration. This is l
9 the Yucca Mountain area here. This is the Amargosa Valley l l
10 and area of deep quaternary fill and these are the l 11 anomalies that exist within Amargosa Valley. These two 12 Anomalies F and G, are the two that were identified in 13 Langenheiin's map at the time. This small anomaly here now
/ \
i 4
' 14 apparently is the one that has been evaluated with ground 15 mag and seemed to also be very similar in character to F 16 and G. So we're dealing with the potential addition of 17 another event here or another particular anomaly and 18 looking at whether or not the experts considered F and G 19 themselves to be potential events. And so we've gone 20 back, given this information, gone back and looked at the 21 characterization that the individuals made, not only in 22 terms of their consideration of what's going on in 23 Amargosa Valley, but also their event definition. For 24 example, we're dealing with a northeasterly trend, a total
/
(T) 25 length that's on the order of about 4 kilometers, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
152 1 consistent with the event definition for many of the
- 2 experts. So with the addition of another event here b, doesn't necessarily mean that we add another event 3 ,
1 4 depending on their event definition.
5 We also, for those that gave a low probability 6 to either F or G being an anomaly, feel that the new data 7 do support more highly the existence of those features 8 being, in fact, buried volcanic centers.
9 So the event count, the basic bottom line in 10 terms of the effect of this is the event count definition, 11 being increased by 1 to 3 new events.
12 What that does in terms of their mean or 13 average event count in Amargosa Valley, again, there's
(~%
( }
' \/ 14 uncertainty characterized here so the number of events 15 that an expert might consider for Amargosa Valley would be 16 the possibility, let's say there are three events there 17 with a particular weight, four events with a particular 18 weight, five events with a particular weight. So the l
19 average of those is the number of events times those 20 probabilities. And so t he average across all the experts 21 at Amargosa Valley before this information was 4.7 events i
22 in Amargosa Valley. We went back and looking at the event 23 counts and made a revision based on the new information 24 and have essentially 6.1 events in Amargosa Valley.
( ,) 25 And that -- let me just -- there are many, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 734-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
I 153 I l
1 many'02 these source maps that we could look at. This is )
i f
2 just an example from Bill Hackett who considers one
('-) 3 particular interpretation of -- that includes Crater Flat.
4 This is one source zone within which you will have a I 5 certain rate density of volcanic events. Another zone 6 that looks like this that would include some of the older I 7 events, his zonation, locations were based on the age of 8 particular centers. Presumably B is the only anomaly out 9 there that has been drilled and dated on the order of 10 about 3.8 million years old. Presumably, these based on 11 the depth of the quaternary cover and polarity and so on 12 can make some arguments that they also were in that age, 13 this is in the post 5 million year category, would be t }
\/ 14 included in this particular source zone.
15 So I'm going through and looking at those 16 definitions, revising the event counts and we've just --I 17 guess 1996 isn't particularly old, but just to make the 18 point that was the assessment that was made previously and 19 this is the updated -- my 9-year-old daughter thinks I'm 20 very old, so this is --
21 MEMBER HINZE: Is that in our pile of 22 materials?
23 MR. COPPERSMITH: Should be. If not, I'll 24 take the blame.
/'
(S,) 25 MS. DEERING: Yes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
154 l
l 1 MR. COPPERSMITH: It should be the second to l
l l ,q 2 last.
(V l
3 MEMBER HINZE: Fine, thank you.
l 4 MR. COPPERSMITH: And this is the calculated l
5 difference. I mean it's obviously, this change, you can 6 see how it works. You have your added particular account l
7 out of the increase, say a 20 percent increase in the 8 number of events in Amargosa Valley. That plays into a l 9 particular spatial distribution. If it lies within a l 10 source zone like Bill Hackett's did that includes the site l 11 that would lead to increased hazard and there is a slight 12 increase in the hazard related to this. If you're using a j ,_ 13 spatial smoothing approach that smooths those with a short
\'] 14 smoothing distance, the operator has a short smoothing 15 distance, keeps the events in Amargosa Valley, they will 16 not extend and intersect the repositories so there's no 17 difference for those particular scenarios.
18 So when you take it, put this revised input l
19 and turn the crank on the whole PVHA, the difference in 20 the cumulative distribution is shown here. That's the 21 dashed line. So essentially it's about a 3 percent
, 22 increase in the hazard related to this additional event.
l i
l 23 Actually, in some cases it's more than one 24 event for some of the experts it ends up because F & G are
/~N
(_,) 25 better defined, they end up adding up to three additional NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
155 1 events.
73 2 So the bottom line in terms of this new h 3 information is it doesn't have a significant impact on 4 hazard and I think that in general the types of things that would lead -- we need to see for those again who are I 5
6 involved in risk, I think ultimately the test of l
7 significance should not be one at the hazard level at all.
8 It should be at the risk level. What would lead to a 9 significant difference in the risk measure and since so 10 far at these probability levels the PVHA or the hazards 11 result from volcanism has no effect on the risk at all in 12 any way. It's difficult to see what types of changes in l l
13 it would lead to significant differences in risk. l
("h l
\ >
'# 14 Now changes in the mean hazard, the mean 15 volcanic hazard would come again from I think very 16 different rate densities, counts that would be increased 17 by an order of magrJcude essentially, coupled with very 18 different spatial models than were assumed here, 19 essentially allowing for rates that are comparable to 20 Crater Flat, if not higher, occurring in the Yucca 21 Mountain block. I think this panel of experts because of 22 the absence over the last 10 million years of volcanic 23 events within the block would be very reluctant to allow 24 for the types of numbers we see in Crater Flat on the
Now you've got this surface l
14 is seen in the mantle?
1 15 expression of perhaps a mantle signature. Partly getting 1
16 at this, what is the driving force in the mantle for 17 producing these melts? And what does one see seismically 18 that might correlate with your province?
19 MR. YOGODZINSKI: They're so small in volume 20 I'm not sure you could see it seismically. A geophysicist 21 could answer that. I think to melt this mantle, which has 22 been cold and isolated for a long period of time, it has 23 to be hydrous in the first place. And there is every 24 reason, based on the trace elements, to believe it is
,o
(_) 25 hydrous. There are amphiboles at Red Cone. Britt talked NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
182 l
1 about those earlier.
2 Amphibole is not a common mineral phase in 7-~
i
~ these rocks. But then again, an amphibole breaks down 3
4 very quickly as you bring it toward the surface. So I 5 think there is very good reason to believe a priori, 6 because of its isotopic signature in an indirect way, that 7 it is hydrous, and some sort of lithospheric thinning in 8 the Western Basin Range. That is, some decompression with 9 a hydrous mantle.
10 I think that that is mostly what most 11 petrologists believe has allowed you to melt this cold 12 mantle at relatively cmall percentages. It is not a lot 13 of magma, not at all. So the extension, in combination r
14 with a hydrous mantle. That's my view.
15 MR. FOLAND: Well, I mean, someone mentioned 16 the tomography. But if I remember, Evans & Smith years 17 ago, they actually found the slower velocities actually at 18 the end of the entire trend. So going all the way down to 19 sema was the slow velocity, hotter mantle, more 20 aesthenispheric mantle. And then if you go up to Lunar 21 Crater, that's the other analog.
22 And Yucca Mountain sits right in between these l
i 23 two --
24 MR. YOGODZINSKI: Right.
( ,) 25 MR. FOLAND: -- not hot spots. Maybe hot lines .
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
183 1 MR. YOGODZINSKI: Right. But there is -- it few 2 appears, based on the chemistry, that none of that
. (J
)
s 3 aesthenispheric mantle has mixed with -- has been really 4 involved and can be seen chemically in the magmas at Yucca 5 Mountain.
6 MR. FOLAND: At the surface.
7 MR. YOGODZINSKI: Yes. Yes.
8 Okay. Any more questions about --
9 MEMBER HINZE: Just one quick question. Do 10 you have any further comments about the source zones that 11 were used in the PVHA? You've shown us Richard Carlson's 12 source zones. Do you feel that the AVIP was understood 7
13 and properly --
I s
T i
14 MR. YOGODZINSKI: I believe it was understood, 15 and I believe it was largely accepted. But I believe it 16 had relatively little impact, because the event count was 17 largely unknown, things like that, the ages are largely 18 unknown.
19 And soon after I gave this talk last time, I 20 left UNLV, and this has really been stagnant ever since, 21 to be honest.
22 Ckay. Now let me go on to something that Gene 23 Smith asked me to present. And I think it actually 24 connects nicely with a lot of the discussion this morning.
f
(_,/ 25 Citadel Mountain is an analog for Yucca Mountain, an l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
184 1 analog with respect to the locations of the cinder cones e 2 on the moutitain.
!m L.))
3 Citadel Mountain is a tilted fault block cored 4 by Oligocene ash flow tuffs. It's located up in the 5 Pancake Range north of the Yucca Mountain area. I'll show 6 you a map, a rather poor map, in a second. It's an uplift 7 of the fault block. It's mostly on the northwest and 8 southeast with a dip slope off to the northeast. So in 9 this sense, it's a little different than Yucca Mountain.
10 Most of the uplift of Yucca Mountain is on the west, the 11 dip slope off to the east.
12 A very simple, but I think relevant point that 13 I'd like to make is that their cinder cones erupted along
/,_h t s
\~' 14 the entire range crests, as well as in the adjacent basins 15 in this part of the world.
16 Now let me try and go to the map. That is 17 actually not as bad as I thought. Actually, very quickly, 18 if I put this up, I can get you located. Yucca Mountain 19 is in this part of the world. This is the Reveille Range 20 and the Lunar Crater volcanic field, and Citadel Mountain 21 is here in the Lunar Crater volcanic field.
22 Now I'll try and talk about this. This is a l
l 23 xerox of the 1972 published map by Sullivan & Ekrund and l 24 others. And what I've outlined here in blue are the lava t's l
'( j) 25 flows associated with cinder cones produced in the Lunar
\
I i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
185 1 Crater volcanic field. I have gone over -- I needed to 73 2 make this figure because I wanted to highlight the l f )
%J 3 structural features.
4 The mountain -- there is a dip slope on the 5 mountain off to the northwest, relatively steep i
6 topographic gradients in the southeast -- excuse me, the 7 northaast, relatively steep topographic gradients in the 8 west and northwest. The mountain is cored by ash flow 9 tuffs outflow, very much like those at Yucca Mountain.
10 And from this point, dotting all the way down 11 the dip slope of the mountain, there are cinder cones.
12 Now, if you've driven around out there, you've seen cinder l 13 cones perched up on the ridge crests. This is much less l
/N.
- 14 true in the immediate vicinity around Yucca Mountain, but 15 you don't have to go very far before you see this.
16 Lunar Crater - the reference is right there.
17 So if you keep in mind where Lunar Crater is here, and 18 this point down here, you can keep oriented on the 19 following two figures.
20 This shows the distribution of lava flows in 21 green and cinder cones in red. Lunar Crater is right 22 there.
23 Again, steep topographic gradients here and 24 here, a dip slope there, the Citadel Mountain, and a
(~h
(_,/ 25 series of cinder cones over this avaa.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
186 1 Now, Lori Dickson has worked on mapping this fs 2 area and dating -- mapping out especially the relative I
)
3 relationships, age relationships among the lava flows, and 4 they are in the process of obtaining new 4039 ages on the 5 lava flows. And I'll show you her preliminary map, and 6 I'll --
7 MR. RYAN: I'm sorry. What's the distinction 8 between the brown and the green fields there?
9 MR. YOGODZINSKI: The green is the lava flows, 10 and the red are the cinder cones. Now, some of the cinder 11 cones produced big lava flows, and some of them are 12 nothing but cinder. And so I'll now show you a map that 13 connects the cinder cones with the lava flows that they t'~'T v 14 produce, 15 MR. RYAN: And the lightest green is?
16 MR. YOGODZINSKI: This here is -- I think 17 those are lake sediments in these tuff cones. This is 18 Lunar Crater. It's an explosive crater, and there is 19 apparently lake sediment at the bottom.
20 MR. RYAN: But they're compositionally 1
21 cquivalent?
l l
22 MR. YOGODZINSKI: Yes. Everything here is .
l 23 basalt. Actually, the composition of these basalts are 24 very different from basalts in the Yucca Mountain area.
l f~~%
(,,) 25 This is the stratigraphy as it has been worked NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 187 1 out by mapping. The red cone here is associated with this g- 2 cone. All of the cones are shown in red, and they are V 3 sitting on top of the lava flows that they produced.
4 The oldest one is Pl. There is actually no 5 cone associated with that. There are a series of dikes 6 and scoria down here. This is thought to be the location 7 where the P1 lava flow originated. It is the oldest one.
8 It has been dated at 3.84. I'm not even writing the 9 numbers down herre. These are done at New Mexico Tech.
I 10 It's preliminary data done on ground mass concentrates.
11 Slightly younger than that, you can see it ;
12 geologically, that it's slightly younger, the 13 relationships, are the H-cone, which originated at the
[~T
-- 14 crest of Citadel Mountain. And we know that the 15 topography was well in place at this point, about 3.8 16 million years ago, because in this part of the world the 17 lava actually cascades down across the existing 18 topography. So the topography was there when the cones 19 erupted.
20 Oc-cone is the large one up here, which 21 erupted on the lower flanks of mountain. C-cone and R-22 cone there are the youngest, but the relative ages between 23 them is not known. R-cone, again, produced the lava flow 24 that cascaded down off existing topography, about 1.24 tO
(,,) 25 million years ago in the case of this volcano.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
188 1 So the very simple point that --
,g 2 MR. FOLAND: What's the gray?
\.)
3 MR. YOGODZINSKI: The gray is other cones that 4 did not produce significant lava flows. In some cases the 5 gray cones are older, and in some cases the gray cones are I
6 younger.
7 Lunar Crater is about 1.26 or 1.24. These 8 cones, Gene tells me, are actually older than this lava 9 flow. But some of these cones, like these two, are !
10 clearly younger than this lava flow. But there is 11 virtually nothing there to date. ;
l i
12 The other point that Gene asked me to make was ;
l 13 to be sure that I said that on the published geologic map
\ '} 14 of this area, there are some cinder cones that are easy l 15 enough to map, yet they do escape the scrutiny of the l l
16 mapper. This cone does not appear on the published l
17 geologic map of this area, and these two small cones here 18 do not appear on the published geologic map of this area.
19 Ken?
20 MR. FOLAND: Yes. Well, you're presenting 21 somebody else's work, so I won't hold you accountable.
22 MR. YOGODZINSKI: Okay.
23 MR. FOLAND: But the ring around Lunar Crater, 24 those are flows, those are large flows, and those are --
,ip
(_,/ 25 MR. YOGODZINSKI: Well, wait a minute now. I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (20-2) 234-4433
189 1 have been here. I haven't worked here. I have been here.
gS 2 Now, there are --
%-)
3 MR. FOLAND: Correct me if I'm wrong, there 4 are some flows exposed in the walls of the crater. Those 5 are older flows.
6 MR. YOGODZINSKI: About 3.8.
7 MR. FOLAND: 3.8.
8 MR. YOGODZINSKI: Right.
9 MR. FOLAND: So there is a whole series of 10 flows.
11 MR. YOGODZINSKI: And then there is -- but the 12 rim is made of tephra that is younger than that, about
,., 13 1.24. Is that --
c' )
14 MR. FOLAND: No, I think that perhaps it's 15 100,000 years or something. That is pretty young. There 16 is a 1.24 flow that comes from the north.
17 MR. YOGODZINSKI: Okay. Okay.
18 So, yes, in this case that cone is younger, 19 but I think that Gene explicitly pointed out that these 20 cones were older than at least the C-cone.
21 MR. FOLAND: Okay.
22 MR. YOGODZINSKI: So in the PVHA process, 23 there were magma boundaries, structural boundaries to the 24 flow of magma from the Crater Flat Basin up into the
.x
( \
(,/ 25 mountain. And this analog, in a very simple-minded way, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W.
gW 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
190 1 the analog study of citadel Mountain suggests that these
,s 2 should be looked at very carefully.
! )
\J 3 It gets very easy to go out there and find 4 something that is geometrically, geologically similar to 5 what we have in Yucca Mountain. And a cross section 6 through Citadel Mountain, and a cross section through 7 Yucca Mountain, a comparison there might shed light on 8 this magma barrier concept. I think I'll just leave it at 9 that.
10 MEMBER HINZE: This is a topographic barrier?
11 MR. YOGODZINSKI: Structural barrier, magma 12 barrier. Someone else is undoubtedly in a better position 13 than I to comment on those ideas as they came into the n
f )
's '
14 experts' models.
15 MEMBER HINZE: Questions?
16 MR. FOLAND: I actually don't follow that, in 17 terms of the magmatic barrier, because it doesn't seem to 18 me that there is a magmatic barrier. You can just go 19 right across -- what, is it Route 75 or 375, whichever one 20 it is.
21 MR, YOGODZINSKI: No. I think the point here 22 is that there is no magmatic barrier. In a mountain, 23 Citadel Mountain -- this is the way I understand the 24 reasoning -- in Citadel Mountain, geologically, it's very (m
(_) 25 similar to Yucca Mountain, and there has been no magmatic NEAL R. GRG55 !
COURT REPORTERS AND ' To OERS l 1323 RHODE ISLAND n, " i (202) 234-4433 WASHINGTON, D C. 20005-37u i (202) 234-4433
191 1 barrier.
s 2 MR. FOLAND: Okay. So it's the topographic i
+ i 3 argument, which I think is what Bill was suggesting, is 4 that here is the mountain and the lava certainly came up 5 in the elevation.
6 MR. YOGODZINSKI: Certainly came up at high 7 elevations.
8 MEMBER HINZE: At one time, DOE was 9 multiplying their results by .15 because of the height of 10 Yucca Mountain. I'm wondering, Kevin, what role did 11 topography play in the PVHA elicitation?
12 MR. COPPERSMITH: This is Kevin Coppersmith.
13 Let me make a point on the barrier. Number 1,
,m c i
.J 14 none of the experts had a barrier, in the sense that a 15 dike that was assumed to have its origin within Crater 16 Flat could not extend into Yucca Mountain. There was not 17 a hard barrier at any of the source boundaries.
18 So the way these models work is that, first, 19 you look at the spatial distribution, you represent events 20 as points, so that you're looking at spatial clustering of 21 individual points. Then, you associate each point with a 22 dike geometry, a length, an azimuth, and whether or not 23 that dike is centered on the point, or, in fact, there's a 24 probability distribution on where it will be relative to
,q
) 25 that point.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
192 1 So for points that originate within Crater r~s 2 Flat, all of the experts allowed them to extend into Yucca i
O 3 Mountain. Okay? So there is no barrier from that point 4 of view.
5 On the issue of topographic control, it was 6 discussed quite a bit, because everyone can bring to bear 7 an analog that shows dikes in upland areas, as well as 8 within basins. I think George Thompson made the point 9 that he expects, because of the configuration of the Basin 10 Range of opposing fault blocks and opposing faults, 11 dipping towards each other along the Basin margins, he 12 would expect more volcanism within the basins.
13 But it was allowed for, and it was never -- it bx) 14 never came through as a strong determinant. In other 15 words, I'm going to outline topography or the amplitude of 16 topography to tell me about the likelihood of volcanism.
l 17 MR. YOGODZINSKI: I think that part of Gene l l
l 18 Smith's reason for asking me to present this was in part 19 because this is something that had been looked at l 20 subsequent to the PVHA process, and he would probably be 21 in a better position to defend it than I am.
22 MEMBER HINZE: Thank you, Kevin, for those 23 remarks. And certainly, you are quite appropriate to put l 24 that caveat on the presentation.
/%
(_,) 25 And we will thank Gene, and you pass our NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
193 1 thanks on to Gene as well, and thank you for your g-r 2 presentation on the AVIP.
'w.) We now move to the next part of the equation, 3
4 and that is the consequence analysis. And Britt Hill, 15 5 minutes? No.
6 MR. H1LL: 15?
7 CHAIRMAN POMEROY: Just put up the number, 8 Britt.
9 MR. HILL: Oh, okay.
10 (Laughter.)
11 Well, most of what I would like to talk about 12 this afternoon, since we are rather far behind, I'll try 13 to just touch on the salient points and hopefully get some t'%
tx -) 14 feedback from people on areas that aren't clear from 15 previous publications.
16 MEMBER HINZE: Don't short sheet us. We want 17 the story.
18 MR. HILL: I'll try to be succinct.
19 There is really two things that we're trying 20 to look ab that I'll try to address in this presentation j 21 that are very relevant to developing consequence models.
22 How much waste material can be entrained by an erupting l
23 basaltic volcano, and how far across the countryside can l 24 this stuff be scattered? And those are the two areas that IT j
(_) 25 I'm going to be focusing on during the next couple of i NEAL R. GIUDSS ,
COURT REPORTERS AND TRANSCRIBERS j 1323 RHODE ISLAND AVE., N.W. '
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
194 1 minutes.
g^N 2 I've given this presentation to many different
!v! 3 audiences and found that it is usually useful to try to 4 give people an understanding of what is magma. And I 5 realize these are not going to be the gnat's eyelash of 6 reality, but it's a good starting point. This is the 7 physical conditions of this material that is going to be 8 impacting on the repository system.
9 You have temperatures of roughly about 1100 10 degrees C, densities of about 2,600 kilograms per cubic 11 meter -- so you can put that on your desk as a really 12 wonderful paperweight -- viscosities on the order of 10 to 13 100 plus pascal seconds. For comparison, I looked it up (3
f f
\/ 14 in the CRC handbook, sucrose is about 1000 pascal seconds 15 at 100 degrees Centigrade.
16 So these are not the big, sticky magmas like 17 you're going to see at day sitcc 31ke at Mount St.
18 Helen's, but really fairly fluid material. If we ever get 19 to the process stage of how this impacts on a canister, 20 we'll need to be accounting for viscosity effects.
21 But the magma, when it comes up into the 22 engineered system, is going to have an assent velocity on 23 the order of about a meter per second to start with. And 24 then, as the eruption slowly builds up during the course f'%
(_) 25 of days, we can be having material flowing through the NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
195 1 system at about 100 meters per second during the eruption.
7-2 And all of that can be translated -- how much
\J 3 vesiculation yo:1 want into the actual forces being applied 4 to the engineered system.
5 We are also evolving some fairly acidic gases 6 at these temperatures, and we have every reason to believe 7 that the degassing from the volcano will adversely affect 8 canister performance in the repository setting, rather 9 than enhancing performance.
10 A lot of these are secondary issues that I'm 11 juot going to be talking about today. We have lumped them 12 under the term of " indirect effects." What we're 13 primarily concerned with, and what the calculations I'll t i
\/ 14 be presenting later are all about, is how much of the 15 waste is actually getting entrained by the eruption and 16 scattered out into the accessible environment. That's the 17 key point.
18 The secondary effects of -- well, we have 19 degassing, we have lava flows in the tunnels, or whatever 20 scenario you wish, that's a separate analysis.
21 We really don't know what is going to happen 22 to a waste package when it gets hit by basaltic magma. We 23 have taken a very preliminary look at the thermal effects.
24 Given these conditions that heat capacity and the physical n
( ,) 25 characteristics of a waste package, it is not going to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., NVI.
(202) 2344433 WASHINGTON. D C. 20005-3701 (202) 234-4433
196 1 last too long.
,- 2 It is going to fail thermally on the order --
(~'/ 3 anywhere from seconds to about a year under these ambient 4 temperature conditions. So we're starting from a base 5 assumption that the waste package is going to fail under 6 magmatic conditions.
7 We are certainly open to additional analysis 8 that shows how robust or non-robust a waste package will 9 be under these kinds of eruption conditions. But again, 10 our statting assumption is that the package will fail 11 during an igneous event.
12 We also don't know or don't have a very good 13 sense of how the waste itself is going to behave during an
.rx i !
\- ' 14 eruption. We're dealing with essentially a pressed pellet 15 that has been thermally and mechanically stressed, that 16 has a density of around 10 grams per cubic centimeter.
37 We do know that it is fractured and that these 18 originally about centimeter size fuel pellets will be 19 broken down into finer fractions. There is considerable 20 debate going on to what the grain size distribution of i
1 21 those fractions will be. But right now, we are starting 22 off that the particle size is not going to be at the one l l
23 centimeter size -- the intact pellet -- but the pellet l 24 will be fragmented into various smaller components.
r^s
! ,) 25 And part of the calculations I'll present will NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
197 1 show different assumptions about if the waste is this gx 2 diameter, with that 10 gram per cc density, this is how it (v) 3 will affect its transportability in magmatic eruption.
4 One other source of sort of the big unknowns 5 or the uncertainties that we have right now is that we're 6 very uncertain about the dispersal capaLilities of most of ;
1 7 the Yucca Mountain volcanoes. We don't have the tephra 8 deposits preserved, except for some very faint remnants at 9 Lathrop Wells. Everything else has been stripped away.
10 We look at the cinder cones, and we interpret 11 how dispersive they could have been. You can go anywhere 12 from essentially a small event at Northern Cone that 13 probably had no distributed material, except some g
j 14 elutriated ash blown out of the fire fountain, all the way 15 down to Lathrop Wells, which is 120-odd meters of totally 16 non-consolidated broken scoria that had to have been 1
1 17 produced from an eruption that had a sustained column on i 18 the order of kilometers.
19 But we can't get a probability distribution l 20 function from the limited information we have right now.
l 21 So we're taking a modeling and analog approach to better l l !
l I I
22 understand how, potentially, these Yucca Mountain type l 23 basaltic volcanoes could transport material into the 24 accessible environment.
O l
y ,j 25 The whole point of this is that in order to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
198 1 get to these dose numbers, the risk numbers, and the 73 2 significance that everybody seems to want, we have some
!v) 3 data, we have some models, and we have a number of 4 assumptions. And all of these have to be mixed together 5 into some sort of a final answer right now.
6 va've talked in many different forums about 7 analogs and what is and is not analogous to basaltic 8 volcanism in the Yucca Mountain region. One of the 9 volcanoes that we pay particular attention to is the 1975 10 basaltic eruption of Tolbachik volcano in Kamchatka that 11 is, we believe, very relevant to understanding eruption 12 processes for Yucca Mountain type volcanoes.
13 Both of these systems -- Yucca Mountain and q
t
\ 'i 14 the specific setting for Tolbachik -- occur in extensional 15 tectonic settings, with no obvious evidence of shallow 16 crustal reservoirs. They are basaltic, fairly non-evolved 17 in the scheme of getting up into the basaltic range of 18 things, derived from hydrous mantle lithosphere.
19 And I'm not here to argue petrogenesis, but l 20 really to talk about the eruption process. And in terms 21 of process, what happened at Tolbachik is very analogous 22 to what we would speculate has happened at volcanoes like l
l l 23 Lathrop Wells. The Tolbachik eruption -- we had a j I
24 northern breakout and a southern breakout. We're focusing )
q
() 25 on aspects of that eruption.
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
199 1 The northern breakout has -- the first cone f- 2 has a volume of cone plus lava plus tephra that is N,3) 3 identical to what Bruce Crowe and his co-workers have 4 calculated for Lathrop Wells. In the Yucca Mountain 5 region, we have volcano volumes that range for quaternary 6 from about .06 to about .2 cubic kilometers, when you 7 convert it all back down to magma volumes.
8 And then you go into the pliocene. That gets 9 up above a cubic kilometer very easily. The total 10 eruption at Tolbachik is about .45 cubic kilometers, so 11 we're seeing, at the first pass, very comparable eruption 12 volumes to compare the processes at Tolbachik with the 13 processes we'd expect at future eruptions at Yucca i \
14 Mountain.
15 And the whole thing is that we can look at 16 Tolbachik, look at what has happened there, and try to 17 understand the clues that we have at Lathrop Wells and 18 other volcanoes in the Yucca Mountain region, and 19 constrain beyond speculation but try to turn some of this 20 into real data to constrain the process models for dose l
21 calculations.
l
- 22 The reason I'm harping about Tolbachik is that l
23 for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> at the very end of the cone 1 eruption, that 24 basaltic volcano dispersed roughly three million cubic
(~h
( ,) 25 meters of shallow subsurface rock up to the surface. We
! NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l l
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
200 1 have some pretty good geologic controls about what the 7s 2 subsurface structure is in Kamchatka, and in some specific
- 1
\ /
'~'
3 area of Kamchatka, where we have roughly 800 meters or so 4 of basaltic rock, then a range of uncertainty.
5 We can't say if the base of the basalts are 6 800 to about 1,300 meters, and then a dominantly 7 sedimentary section. So we've got a real nice layer cake 8 stratigraphy to look at. We have examined the xenoliths, 9 the wall rock fragments that come out at cone 1, and also 10 looked at another part of the deposit I'll speak of in 11 just a second, to constrain this volume.
12 And using just simple geometric arguments on 13 the proportion of volcanic to sedimentary rock at
'- 14 Tolbachik, with the total volume of material erupted, you 15 can say that, as a minimum, the conduit in the late stage 16 of this eruption widened to at least 37 meters in 17 diameter, assuming a cylindrical geometry, down to a 18 maximum, given this range of uncertainty, of around 60 19 meters in diameter.
20 Trying to get an estimate on the uncertainty 21 gives us an average widening of the conduit to around 49, 22 plus or minus seven. Let's just call it 50 meters in 23 diameter -- is what happened during this eruption for the 24 last 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> when the conduit went from about one to two
()
n 25 meters in diameter and widened out to 50 meters in l
NEAL R. GROSS
! COURT REPORTERS AND TRANSCRIBERS j
1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
b
201 1 diameter, f- 2 The reason this is important will become
\-) Let me just say that we have to 3 apparent in a second.
1 4 facies associated with this event. First, you have a 5 fairly abundant amount of xenoliths. These wall rock 6 fragments are scattered about on the surface of the cone.
7 But even with this extreme brecciation event i 8 occurring, you're still only seeing about one to two 9 percent of the total surface area is covered with these 1 10 wall rock fragments. So they're not hugely abundant.
11 Twenty percent of the cone is just covered with pieces of l
12 the subsurface.
13 What's more important is that an ash layer was !
,g)
\
'd 14 distributed for about 10 kilometers around the volcano 15 th..c consists of finely pulverized subsurface rock. And l
16 this occurs anywhere from about a 25 centimeter thick down i 17 to a core vanishingly small thick, white ashy clay section i
18 that represents the bulk of this three million cubic 19 meters of pulverized wall rock.
l 20 So it's not even so much what you're seeing on l
21 the cone itself, but the very late stage of the tephra 22 fall is just blanksted with this white ash deposit. And 23 just estimating, in the past 20 years I'd guess about 20 24 to 30 percent of that deposit has already been removed by O)
( ,
25 erosion. It's just the same as kicking up, if it's dry, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N.W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
202 1
1 it's fairly dusty, it's very easy to remove this deposit.
2 There is one other key bit of geologic
's_) evidence associated with this brecciation event, and that 3
l 4 is with these xenoliths, you can find a very unusual type 1
5 of volcanic bomb. That is, a mixt .a of different I 6 compositions of wall rock, each of which has a range of l
7 thermal history to it.
8 So you can find pieces of older basalt that 9 are very intact -- you can't see any thermal effects to 10 them whatsoever -- all the way down to ones that have a 11 very nice quenched margin to disruption fracturing and a I
12 little bit of invasion to them. l 13 And also, the sedimentary -- more importantly, f7 ~) l
\/ 14 the sedimentary xenoliths that you see within these bombs 15 range from completely unaffected thermally -- something l
16 that was just thrown into this melt and quenched -- all 17 the way to ones that you'd mistake for pumice, they are so 18 melted and inflated during this brecciation event.
19 So what we believe these xenoliths are showing 20 is a sampling of this brecciation, where we're entraining 21 different depths or different distances away from the 22 conduit during this brecciation event, taking pieces of 23 the distal wall rock if you will, the part that isn't 24 adjacent to the conduit, that hasn't been heated up and
,r3
( ,) 25 melted, and mixing those fairly cool pieces of the conduit NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 203 1 in -- or cool pieces of the wall rock in with the rock l
gg 2 that is adjacent to the established magmatic conduit.
'wl All of this is in matrix support and a quench 3
4 basalt matrix. So this isn't an agglutination of layers 5 through time, but rather a single instantaneous event that 6 records this white ash process, this conduit widening 7 process, at Tolbachik volcano.
8 Now, to tie in to why you're paying attention 9 to this for Yucca Mountain region is that we find the same 10 sorts of remnants at Lathrop Wells, and to a lesser extent 11 at Little Black Peak. First of all, the proximal 12 xenoliths, the xenoliths on the cone at Lathrop Wells, are 13 unusually abundant.
/~~g i s
\' 14 When you go up to Lathrop Wells and the people 15 that we've taken to that have a wide range of experience 16 there in, say, the basin and range, you come up and you 17 look at that and you go, " Wow. There are a lot of wall 18 rock fragments, the xenolith fragments, sitting in Lathrop 19 Wells' cone." Unusual for a basaltic volcano. The same 20 sort of abundances occur at Little Black Peak.
21 More importantly, we find the same kind of 22 polylithologic/polythermal bombs at Lathrop Wells that you 23 find up here at Tolbachik.
24 These occur only on the southern flank of
/ T
, '( ,) 25 Lathrop Wells, at what we are to interpret would be the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 HHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
204 1 last gasp out of the volcano. They are not distributed rm 2 throughout the volcano. You can only find them in one
(
%) very specific area, which is exactly what you'd expect if 3
4 this was solely a late stage event.
5 So we're dealing with two volcanoes at 6 Tolbachik and Lathrop wells where we have a similar 7 eruptive volume, a similar sort of tectonic setting, 8 coming up through a crustal section that has been 9 disrupted in the upper kilometer or so.
10 And based on the distribution s L lithologies 1
11 for Lathrop Wells and the structural controls that we have 12 available out there, we'd say that Lathrop Wells disrupted 13 anywhere a crustal section, a dominantly tuffaceous
- i 14 crustal section, of about half a kilometer to two 15 kilometers deep. l 16 So, right now, we are thinking that the i 17 subsurface -- the late stage disruption at Lathrop Wells j 18 was comparable to what we have observed at Tolbachik, and l 19 thus we should be taking into consideration that late 20 stage conduit widening under a future volcanic event may i 21 disrupt an area of 50 meters in diameter. l l
22 Given the average thermal loading on the 23 repository of roughly 83 metric tons uranium per acre, 24 depending on how you want to put your spacing of waste C'i
(_j! 25 pa:kages and how much waste is in a vaste package, that
' NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. l (202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
205 1 would translate to roughly four to 10 waste packages by l
l p 2 widening the conduit to that 50 meters in diameter.
~.)
i 4 l 3 So when I talk a lit.tle bit later about the l 4 dose calculations, and we go from one waste package up to 5 10 waste packages, this is the geologic basis for that 6 assumption. l l
I 7 MR. FOLAND: Britt, this is Lathrop Wells.
8 MR. HILL: Yes.
9 MR. FOLAND: Are you proposing that other 10 volcanoes in the Yucca Mountain region were like Lathrop 11 Wells?
12 MR. HILL: No, I'm not. As a matter of fact, 13 t he.re is little evidence to support that.
,9,
\ t U 14 CHAIRMAN POMEROY: Did you say, though, Little 15 Black Cone is -- ,
1 16 MR. HILL: Little Black Peak has the same --
l 17 there is a few of the polylithologic bombs. It is quite a 18 bit more eroded than Lathrop Wells. But the relative, to 19 say, going out to Black Cone, Red Cone, or a number of 20 other volcanoes that would be even less eroded in the 21 Western Great Basin, there is a remarkable abundance of 22 xenoliths there.
23 And still, it's not Lathrop wells, it's not ,
l l 24 roughly one percent on the surface. But when you go out
(
/m
! It) 25 to most cinder cones in the western U.S., maybe every 10 ,
1 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
206 1 square meters, 20 square meters or so you find something l
1
,s 2 that is an obvious xenolith from shallow crustal, if you i \
(~) 3 get lucky. 1 4 Ken, you've seen a lot of cinder conen out 5 there. Would that be your observation?
6 MR. FOLAND: Yes, I think based on that. i 4
7 MR. HILL: Well, when you go and you throw a 8 one square meter square down on Lathrop Wells or Little 9 Black Peak, and you start picking up 10, 15, 20 centimeter 10 size and greater pieces of wall rock, that is unusual.
11 And certainly, you don't find that on little cones or the 12 quaternary Crater Flat cones. But you do see it at Little 13 Black Peak. l r% a
\/ 14 I haven't done the statistics to say whether 15 Hidden Cone has it or not, but Hidden Cone is j I
16 contemporaneous with Little Black Peak. They're both I
17 about 350,000 years.
18 But Lathrop Wells is not isolated. There is 19 elements of that, but the trend is kind of interesting.
20 If you look at eruptive volume from pliocene, even through 21 the quaternary, we're getting distinctly more explosive in ,
22 every quantifiable measure that you can see through time.
23 So I think the trend would support that future 24 volcanoes would be capable of Lathrop Wells' style
/~~
( ,h) 25 eruptions, rather than Northern Cone or pliocene Crater NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 200C&3701 (202) 234-4433
207 1 Flat.
i f s., 2 MEMBER HINZE: Is there anything in the water
] content of these rocks to differentiate Lathrop Wells and 3
4 Little Black Cone?
5 MR. HILL: For all intents and purposes, from 6 the preliminary work we have done, from the work that 7 Frank Perry and other people in the Los Alamos program 8 have gone through, there is no gross geochemical 9 distinction between Lathrop Wells or anything in the 10 quaternary out here.
11 There are subtle petrogenetic variations, but 12 nothing to say that while Lathrop Wells is anomalous )
1 13 petrogenetically, relative to the other quaternary, they i
/~T l 1
- t. )
N/ 14 are extraordinarily similar. In most places, you'd call 15 them identical, until you really started expanding your 16 scale.
17 So we've got waste. We disrupted part of this 18 repository. Where is it going to go? I just show these 19 two plots to give a sense of scale for given the range of 20 volumes that we see at Yucca Mountain, what would an 21 analog volcano look like if it erupted through the 22 repository.
23 We'd start off with Tolbachik. Again, about 24 .45 cubic kilometers, a pretty good size quaternary event.
(^h
's_) 25 We'd be having eruption through here. You can see our one NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
! 208 l 1 centimeter isopach would come down about 50 kilometers L
l g-~s 2 south. I have rotated the isopachs around as though the
! )
N/
l 3 wind is blowing from north to south, because these black 4 dots are at critical group locations, at 20, 25, and 30 5 kilometers south of the repository.
6 So we had a Tolbachik kind of event. This is 7 one guess of what the isopachs would look like. You'd i
I 8 have roughly five centimeters to about three centimeters 9 covering your critical group location. In contrast, if we .
l 10 had a very small eruption, something like Cerro Negro, ,
l i
11 that has a magmatic component of .008 cubic kilometers, !
l 12 which is about an order of magnitude smaller than the 13 smallest event we've got preserved out at Yucca Mountain, 14 this is what the isopachs would look like. j 15 A one centimeter isopach would only go down 16 about 12 or 14 kilometers south of the volcano. We 17 wouldn't be getting more than a millimeter at 25 18 kilometers, and at 30 kilometers it would just be a 19 dusting of ash. So we're factoring this in when we do our 20 consequence models to san.c a range of volumes that would 21 give us this kind of a distribution to accommodate the 22 scale of future eruptions.
23 Now, in order to calculate -- in performance l
24 assessment models, in order to calculate the thickness of
/~N
( ,) 25 ash, we have been using a model that was developed by i
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C, 20005-3701 (202) 234-4433
209 1 Suzuki and published back in 1983. I realize.that there 2 are a large number -- maybe not a large number -- but l (~'S
! V 3 certainly a variety of tephra dispersion models that can 4 be used to simulate the transport of material from a 5 volcano.
6 Many of those models, however, are really 7 developed for large silicic eruptions. That transport is 8 predominantly in the fines component range, the ash 9 component, down into the hundred micrometer or finer 10 range. Also, for eruptions that go up to a level of 11 neutral buoyancy and develop a very large umbrella cloud 12 through them, that stagnate at very high altitude -- 12,
,_, 13 15 kilometers or so.
( )
14 These are not characteristics of basaltic 15 volcanoes. They tend to form fairly low eruption columns 16 on the order of five kilometers or so, and get blown over 17 fairly rapidly by the wind.
18 The Suzuki code gives us a better handle. It 19 doesn't account for an umbrella cloud, and also it is a 20 little bit better suited for modeling the course material 21 rather than the distal fines. And we're really worried 22 about this area -- 20 to 30 kilometers south of the 23 volcano.
1 24 We are also examining the possibility of
/~T l (_ l 25 modifying this dispersion code using some of the work done l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
1 210 i
1 by Andy Woods. I think his model may be a better process
~s 2 model than the processes captured by Suzuki. But right i
\
O 3 now, we are going -- we are reviewing our preliminary 4 calculations using the Suzuki model.
5 MEMBER HINZE: Do you have any plans to 6 exercise those codes, bring them up, or any of the other 7 codes otner than the Suzuki?
8 MR. HILL: We're evaluating Woods' model right 9 now. It is too early to say whether it's going to be a 10 superior product to what we're using with Suzuki or not.
11 I'm very comfortable with Suzuki. I 12 understand where a lot of the limitations are. We have 13 tested sensitivity of the model through a number of input
! \
14 parameters, and I talked about that in the annual report.
15 We have been evaluating this model using data 16 collected from the 1995 eruption of Cerro Negro, a very 17 small volume basaltic volcano down in Nicaragua. The 18 reason we used that is that we collected the data and have 19 all of the parameters we need to evaluate the model. It 20 you go into the literature for basaltic eruptions, you'll 21 see that eruption duration is usually well represented.
22 But nobody talks about the time that the eruption actually 23 sustained a column.
24 Most of these basaltic eruptions have an early
,r h
( ,) 25 phase where there is very little convective activity, and NEAL R. GROSS !
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4 433 WASHINGTON. D C. 20005-3701 (202) 234-4433 l
J
211 1 then a main tephra dispersing phase that's usually a
,3 2 smaller portion of the total eruption, nor are wind speeds
( )
'~
3 and wind directions ever really mentioned. It is very 4 difficult to get a robust data set that you can actually 5 test for a basaltic volcano, but we developed that data 6 set specifically from Cerro Negro in order to evaluate 7 these tephra dispersion models.
8 So it's one where we have a high confidence in 9 the quality of the data set used for accuracy, but 10 wouldn't presume to say that this validates the Suzuki 11 model. It's a reasonable test, and let's leave it at 12 that.
13 What we found is that the Suzuki model is very
,m
! i 2 14 sensitive to wind speed. If we use the wind speeds that 15 we observed during the '95 Cerro Negro eruption, we 16 underestimate by about 50 percent the deposit thickness in 17 the 20 to 30 kilometer range.
18 Now, this is the difference between 19 performance assessment and geology. When we first came up 20 with these numbers, it was, "Oh, boy, you know, we're 21 underestimating by about 50 percent." And when I talked 22 to the performance assessment people, it's, " Wow, we're 23 within 50 percent. This is a fantastic model."
24 (Laughter.)
(3
( ,) 25 Really.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l l 212 l
1 (Laughter.)
l f3 \ 2 VICE CHAIRMAN GARRICK: If they're within 500 l i
'w) 3 percent, we'll be happy. i 4 (Laughter.) l l
5 MR. HILL: We also found, in addition to wind j 6 speed, there is a moderate sensitivity to the column 7 height, particle diameter, and particle density. But we 8 feel we can constrain column height reasonably well based 9 on mass flow and total eruption volume. Particle diameter 10 and particle density you can only vary within certain 11 reasonable limits.
12 Other parameters, such as shape parameters, 13 sorting parameters, really don't have a significant effect
(_ \
~
14 on the deposit thickness in the range of distances that 15 we're trying to consider.
16 So the bottom line is we have evaluated 17 Suzuki. We feel we are within about 50 percent of the 18 thicknesses of 20 to 30 kilometers, and that that gives us 19 a good basis to evaluate, from a more probabilistic sense, 20 the risk associated with dispersion of basaltic volcanoes.
21 And we have folded all of that into our total 22 performance assessment code, version 3. The module is 23 called ASHPLUME. And that is based on the Suzuki model, l
24 and we're using that to sample the range of input (3
(_) 25 parameters more stochastically, such as wind speed, NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
213 1 volcano volume, eruption duration, to get a sense of the g'z 2 dose numbers that I'll present in a little bit.
8 I w' i So are there any questions on the approach 4 we're taking or the reasons we're taking this approach for 5 tephra dispersion or for subsurface entrainment?
6 MEMBER HINZE: Why don't you go on and let's 7 see what we pick up as --
8 MR. HILL: Okay. I'd have Tim come on up and 9 talk for a few minutes about the non-disturbed repository 10 performance, to give us a context for the volcanic doses 11 that I'll be talking about afterwards.
12 MR. McCARTIN: Okay. As Britt said, I will
,_s 13 give a slight introduction to actually the next portion of 14 Eritt's talk, and that is to talk about some preliminary l
15 dose calculations that have been done for the total system .
l 1
i 16 to put the doses that Britt will be presenting for just 17 volcanism into perspective in terms of the overall system i
18 performance, f l
l 19 And very briefly, I think it is important to 20 understand the doses I will be presenting. I will focus 21 just in fairly short order, really, on TSPA '95 and some 22 NRC staff calculations that we did in support of our 23 evaluation of the NAS recommendations. And it is 24 important to understand why these calculations were done.
/3 i ,) 25 Certainly, I'm not trying to denigrate the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
214 1 calculations. They were done certainly using the best 2 information we had at the time, but the purpose for the f-'s y./
3 calculation I think needs to be kept in mind when you see 4 what the doses represent. And for TSPA '95, they were 5 coming to closure with their TSPA results as the NAS 6 recommendations were coming out.
7 They included a peak dose calculation at five 8 kilometers, but it was done at the very end of the study, 9 and so you certainly don't want to interpret these results 10 as the best that they would do if they did the calculation 11 today.
12 Likewise, for the NRC calculations, the NAS
,_ 13 recommendations came out. We quickly went to do some t
1
'~'/ 14 calculations to see, in terms of the peak dose and a i
15 reference biosphere critical group, looking at a million 16 years, what did that mean in terms of implementation. We 17 certainly used parameters and models that we felt were 18 appropriate.
19 However, we did not go into any great detail 20 to define the critical group, parameters for that critical 21 group, etcetera, but trying to see the mechanics of 22 actually doing the calculation out to a million years, 23 were there any special kinds of pitfalls that we wanted to 24 discuss with EPA in development of the standard and in
'q,/ 25 terms of consideration for our own regulation.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i
1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINCTON, O C. 20005-3701 (202) 234-4433
215 1 CHAIRMAN POMEROY: Tim, as you go through,
,s 2 could you just -- I don't want to stop you now, except to ,
l (R/ ) 1 3 say when you're talking about this, can you indicate any l 4 point where you've used very non-conservative values 5 versus upper bound values, so we can perhaps get a clearer .
l 6 understanding of where -- l 7 MR. McCARTIN: Sure.
8 CHAIRMAN POMEROY: -- these -- !
l 9 MR. McCARTIN: Yes. I'll try to point out a 10 few areas where -- yes.
11 CHAIRMAN POMEROY: Thanks, Tim.
12 MR. McCARTIN: First, the TSPA '95 results, 13 once again, it was undisturbed performance only.
'\ # 14 Volcanism was not considered quantitatively, although 15 other TSPA efforts did coneider volcanism. They looked at 16 a drinking water dose at five kilometers. That was 17 something very quick for them to do.
18 And, once again, we would suggest that, you 19 know, it was done because they were calculating doses to 20 the old standard, which had a five kilometer compliance 21 boundary. There wasn't any look to see, well, where is 22 the critical group, certainly lifestyle of a critical 23 group, etcetera. And so it was done at the end of their 24 particular exercise, and there really wasn't time to go to (q ,/ 25 any of the details that are important in a dose NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE _, N.W.
q (202) 234-4433 WASHINGTON D C. 20005 3701 (202) 234-4433
216 1 calculation.
fs 2 But in terms of the TSPA '95 results, anyone
( \
L) 3 familiar with the report knows there are many, many 4 different plots in that. I selected one. I'm not trying 5 to suggest this is the most typical, but you can see the 6 doses range anywhere from the microrem out to around 7 10 millirem. And I guess one thing to bear in mind, that 8 it was a 10,000 year calculation and it was cut off.
9 They did have calculations in there for longer 10 time periods. However, the down side with the longer time 11 periods was it was a point value calculation. It was 12 basically one -- what they called an expected value, where !
13 they used the mean values of all of the parameters. It l I) k/
i 14 was just one realization. )
1 15 And generally, we're more comfortable l l
16 presenting results where you can see the spread of how the I 17 dose varied with the variation in parameters. But that 18 was, once again, the five kilometer drinking water only 19 dose.
20 CHAIRMAN POMEROY: So here's the 10,000. This 21 is cut off at 10,000 years, though.
22 MR. McCARTIN: Yes.
23 CHAIRMAN POMEROY: So is there a -- I guess 24 the question is: is there a significant dose immediately l
(") 25 beyond --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (20Q 234-4433
217 l
1 MR. McCARTIN: If you look at the other
[ ,s 2 calculation they did to dose, they did show some million
, i r
,/
3 year calculations. But, once again, it was a single 4 realization. It would suggest that the doses at the high 5 end might go as high as, let's say, 100 millirem for a 6 perspective.
7 But it is always dangerous, in my mind, to 8 select a point value calculation. It is hard to get the 9 perspective of where it sits when all of these things are 10 varying, but --
11 MEMBER HINZE: We're with you 100 percent.
12 MR. McCARTIN: In evaluation of the NAS 13 recommendations, we did, as you know, some calculations
,a i a
\/ 14 for our deliberations with EPA. We analyzed the nominal 15 base case, which we're assuming the probability was near 16 unity. Peak dose, we went out to a million years. We 17 looked at a couple of different compliance points. One, 18 obviously, for traditional aspects is at five kilometers,
)
19 and we looked at a drinking water pathway only.
20 We also looked at the dose at 30 kilometers.
21 That considered all pathways. That is more like the 22 Amargosa Desert area, where you would potentially have a 23 farmer and have the other ingestion pathways.
24 And we tried to use the average member of the
/%
! is_j 25 critical group approach. I will say we used sort of a l
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
218 l
1 representative person in the population. I don't think we l fw 2 went into any great detail in terms of looking at what is j (v) i the lifestyle of this average member of a critical group,
! 3 4 but we used a representative person. And as you know, l
l 5 we're doing a lot of thinking, soul searching, as to what 6 that lifestyle characteristic should be.
7 And so the calculataan we have is sort of a 8 population person average, if you will. It is still an 9 annual individual dose, but we sort of averaged all of the 10 characteristics in the region into this one person. Is 11 that appropriate for the average member of the critical 12 group? That is yet to be determined.
13 We also looked at volcanism and human
(,)
intrusion, which was analyzed separately. For volcanism, 14 15 we looked at a 20 to 30 kilometer pathway, once again, 16 using that same average person if you will. And we were 17 looking at the ingestion pathways for a farmer at that 20 18 to 30 kilometer location.
19 Once again, what did our doses look like? And 20 you can see for the two -- this is the CCDF of our peak 21 dose out to a million years for both the Amaragosa Desert 22 and the drinking water. And you can see doses vary. They 23 are around -- at the Amargosa Desert, the high end is 24 maximized out to about 100 millirem. It is about a rem at
,a (x,) 25 the drinking water location.
I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
219 1 Now, once again, I caution everyone to realize r-~s 2 that there was not a lot of work done to determine what (v) 3 the appropriate characteristics of the five kilometer 4 location, as well as the 30 kilometer location, in terms 5 of dilution, which is very important, and the 6 characteristics in lifestyle.
7 Now, the characteristics in lifestyle, 8 certainly for Amargosa Desert, is more important than at 9 five kilometers where we were doing merely a drinking 10 water dose only. But, once again, we made some simple 11 assumptions for dilution, etcetera.
12 MEMBER HINZE: What did you use for your model
,_ 13 for the dispersion of this? What did you use for the
('~') 14 dispersal of the ash? What did you use for this?
15 MR. McCARTIN: Actually, Britt Hill will 16 talk ~~
17 MR. HILL: These are for undisturbed only.
18 MR. McCARTIN: Yes. I'm sorry. This is for 19 the nominal base case. This is not --
20 MEMBER HINZE: Okay. Fine.
21 MR. McCARTIN: -- undisturbed, if you will.
22 I'm just trying to give the base case, and then Britt will 23 talk through the volcanism calculation. But this is more 24 or less to give some context for the doses that you'll
,0
(/ 25 present.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 2344 433 i
220 1 In terms of how they compare, you can see
,g 2 looking through the different numbers, if we look at the
() 3 median and mean doses, you can see sort of where they sit 4 numerically. And the range was, as noted before, it 5 varied from one to around at the five kilometer drinking 6 water dose only. Very dependent, however, on certainly 7 assumptions made with respect to dilution and what that 8 group did at five kilometers. Likewise, .2 to 118 at the 9 30 kilometer location.
10 You know, I guess a couple of the big 11 differences that might be noteworthy, although we're 12 considering all of the pathways here, not just drinking 13 water as here. Part of the reason I believe the doses are
/
(' 'h) 14 lower are you get a little more dilution in terms of at 15 the Amargosa Desert location, but also you have alluvium, 16 which you can take credit, if you will, for some 17 retardation in the alluvium.
18 Generally, we have taken no credit for 19 retardation within fractures. At five kilometers, you are 20 still in a fractured rock regime with the plume, and so 21 there is a couple of reasons there for the differences in 22 the doses.
23 VICE CHAIRMAN GARRICK: Tim, what was the 24 infiltration rate for this?
(q) 25 MR. McCARTIN: Oh. Yes, that's a good point.
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 2344433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
221 8 1
1 For these -- well, we varied them, and we used, I'll say l 7s 2 -- it was about a year ago, but generally, our range has
! )
~#
3 been one to five millirem -- or five millimeters per year.
4 These calculations were done prior to some of the later 5 DOE estimates. That might suggest a slightly higher l
6 range, you know, as high as 10 millimeters per year. But I 7 these we were varying between one and five.
1 8 VICE CHAIRMAN GARRICK: And that number seems l 9 to be going up a little?
10 MR. McCARTIN: Right.
11 VICE CHAIRMAN GARRICK: Yes.
12 MR. McCARTIN: Right. Yes. Although a lot 13 depends -- you know, there is many parts of the modeling (3
5- 14 that we're looking at closer, certainly with respect to 15 the phase 3 modeling, in terms of the way water will get 16 into the drift, how much will it impact the waste package.
17 This model was using our phase 2 approach, a little more 18 simplistic.
19 But with that, I guess the main thing to leave 20 you with is what the dose -- the median doses, mean doses 21 are, and then Britt will -- which is more the purpose --
22 talk through the volcanism results in more detail.
23 MEMBER HINZE: Thank you very much, Tim.
24 And, Britt, we have a handout from you?
f~h (s_, ) 25 MR. HILL: Yes. It's the second part of the NEAL R. GROSS 1 COURT REPORTERS AND TRANSCRIBERS ,
1323 RHODE ISLAND AVE.. N W. l (202) 234-4433 WASHINGTON, D C- 20005-3701 (202) 234-4433 i l
222 l
1 one you have. '
2 CHAIRMAN POMEROY: And, Britt, while you're
'~
3 being wired up, can I just say that it's even more 4 important, so that we don't have to go through the 5 discussion that you and Kevin had at the technical 6 exchange, that when you go through the model itself, can 7 you indicate where there are bounding assumptions, where 8 there are reasonably conservative consumptions, and where 9 there are non-conservative consumptions?
10 MR. HILL: Sure.
11 CHAIRMAN POMEROY: Assumptions, rather.
12 MR. HILL: Thank you for the lead in.
13 Basic assumptions -- we're assuming the n
x- 14 volcanic eruption occurs through the repository anywhere 15 from 200 to 10,000 years after closure. This way we're 16 not getting the first couple of hundred years where we've 17 got a fairly high buildup to worry about.
18 The base assumption, based on some of the 19 arguments I outlined earlier, is that one canister has 20 failed and that all waste, about 10 metric tons, is l 21 available for magmatic transport. When we are doing our 22 total system assessment, we are varying the area so that 23 we are having smaller amounts of waste being considered.
24 But again, as a base assumption, we say one canister has
(
v) 25 failed.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
223 1 We know from the site data that wind is a 2 blowing to the south 14 percent of the time. That has
, }
3 been factored in for our critical group locations in the 4 doses that we're calculating. We're saying the critical 5 group is located 20 to 30 kilometers south of the 6 repository. We have not calculated any five kilometer i
7 doses. Let's just leave it at that.
l e And we're also using the current dose l 1
9 conversion factors that were used in the dose calculations l l
10 for the NAS recommendations, which are also the ones that l 11 are being used in the current version of our TPA code.
12 These calculations were run last year, not using the i 13 complete TPA code, but were using the modules that are
( /
v 14 incorporated into TPA. So these numbers would be 15 reproduced if you ran TPA right. now using these input 16 parameters and assumptions.
17 Some of the areas where we --
18 MEMBER HORNBERGER: Excuse me, Britt. Is the 19 dose primarily inhalation?
20 MR. HILL: The dose is driven by ingestion 21 through plant uptake and meat uptake. There is an 22 inhalation dose. One of the areas that we're doing some
.: .3 thinking about is the resuspension factor for volcanic 24 ash.
(3
() 25 Right now, we're using a resuspension factor.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 2000S-3701 (202) 2344433 l
t
1 1
224 1 I think there are five available. The closest --
r3 2 MEMBER HINZE: What is a resuspension factor?
N.
3 MR. HILL: It's on the ground and it gets 4 kicked up. We're using one for sand, and that is not very 5 accurate. If you've ever walked across a sand dune versus 6 volcanic ash, you know that you get a lot more volcanic 7 ash in the nose than you would on sand with medium grain 8 size particles.
9 We're thinking about that and doing some 10 sensitivity -- will be doing some sensitivity studies to 11 see whether the inhalation dose is going to be critical.
12 For five kilometers in, we'd, of course, be 13 looking at inhalaticn and ground shine as the pathways, in
/~S
\)
14 addition to ground water from the other release.
15 Paul, you were asking about critical processes 16 and assumptions. Well, here they really are. First, the 17 behavior of the ascending of the magma in the disturbed 18 geologic setting. We would like to do some better process 19 modeling to really understand what happens when this 20 ascending dike hits the drifts at 300 meters below the 21 subsurface. i l
22 We have taken primarily an analog approach to 1
23 start with, and we have been asking around. Nobody has l l
24 been able to come up with a good 300 meter tunnel down
(~) 25 there that a basaltic volcano of continental type has come
(_/
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
225 1 up through. So our base assumption on this is it is
,- 2 behaving just like the undisturbed geologic setting in
/
3 terms of wall rock entrainment. These are the basic 4 assumptions that the DOE has used as well.
5 We have a big unknown on canister response to 6 ascending magma. That is one of the areas where we are 7 doing more thought. I try to evaluate the available 8 information. But, quite frankly, igneous activity was not 9 a design criteria, and it hasn't really been addressed for 10 what happens under these mass thermal and chemical loads 11 that an igneous event would place on a canister relative 12 to the normal response from the ambient environment.
13 So, right now, we feel we have a good
(
' 14 justification for making the assumption the canister will 15 fail as a reasonable and conservative measure to do these 16 calculations.
17 The waste particle size distribution -- since 18 we're dealing with transport of dense materials 19 subaerially, how big that material is is going to 20 critically impact the dose that you get from that 21 material. We're starting off with waste pellets that are 22 roughly centimeter size that will broken down into smaller 23 size fractions.
24 The calculations I'm presenting will show
.n
( ,) 25 three different size ranges, a median diameter of 10 NEAL R. GROSS COURT REPORTERS AND TRANSChlBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
226 1 millimeters, one millimeter, and 10 microns. And we'll 2 evaluate dose based on those ranges. That range is based j
3 on literature information and evaluation of the best 4 information we have on the fractured nature of these 5 particles.
6 Right now, based on some work that we've 7 uncovered from the NRC's NUREGs coming out from the 8 reactor program people, we would skew our bias down --
9 skew our bias? That's wonderful.
l 10 (Laughter.)
i 11 We'd say that the finer grain fraction is more i 1
12 prrbable than the one centimeter fraction, would be more 13 supported by the available information.
,m i )
\~/ 14 Again, we're kind of in the dark about --
15 MEMBER HINZE: What is coarse here? The 16 Suzuki model is better for coarse textured materials? Did 17 I understand that?
18 MR. HILL: The Suzuki model accommodates all 19 range from I think it's --
20 MEMBER HINZE: Okay.
21 MR. HILL: -- was it 15 microns or so?
22 MR. CONNOR: About, yes.
23 MR. HILL: 15 microns in diameter on up to --
24 we're cutting it off at 10 centimeters in diameter from e-(x) 25 grain size distribution. It is really -- you're looking NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. l (202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234 4433 l
227 1 at gravitational settling.
73 2 MEMBER HINZE: Okay. The validity of it is
( )
3 not greater in one size range than in another?
4 MR. HILL: No. No.
5 MEMBER HINZE: Okay.
6 MR. HILL: It's not that.
7 We're looking at how is the waste actually 8 affected by the ascending magma, whether it's incorporated 9 into the particles or operates as discrete particles. How 10 it's actually being entrained, again, is a big unknown.
11 We don't have volcanic rocks coming up through dense 12 friable material. But we're making the assumption that 13 the waste is adhering to the tephra particles, and that
( )
k '~ 14 the particles have to be a certain diameter greater than 15 the waste.
16 So when we say it's a 10X incorporation ratio, 17 _ hat means the particle, the tephra particle has to be 10 18 times the size of the waste particle in order to transport 19 that waste particle into the accessible environment.
20 This gives us a transport cutoff. If we have 21 a very fine grain eruption, it just haven't the 22 transportability to move this material down range. We're 23 hoping to get a better process handle on that, if it turns 24 out that this is critical. But right now, we feel this is C
( ,)h 25 a good starting assumption, that the waste is being NEAL R. GROSS :
COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE., N.W.
(202) 2344 433 WASHINGTON, D C. 20005 3701 (202) 234-4433 .
l
228 1 transported.
fg 2 And finally, the dispersal capability of the
! )
x_/
3 volcanoes is like I had outlined previously from analog 4 volcanoes and is effectively modeled using the Suzuki 5 code. We have the uncertainty of about 50 percent. But 6 given the range of numbers that I'll show, and the range 7 of uncertainty in the entire total system assessment, that 8 is deemed accessible by everybody I talk to, really, 9 except those pesky geologists.
10 Okay. For each one of these dose 11 calculations, we did 300 simulations with one canister, 12 10 metric tons available for transport. We're getting a 13 mean annual peak dose, and the second number is the
( )
\/ 14 standard deviation about those realizations. The reason 15 things are a little odd is you end up with log normal 16 distributions.
17 We're sampling stochastically the eruption 18 power and duration. Simply put, we're accommodating for 19 column height, eruption rate, and total mass of the 20 system. So we're varying those parameters between what 21 we're seeing for Yucca Mountain volcanoes.
22 The eruption rate, of course, we don't know, 23 but we can make reasonable assumptions from analog 24 volcanoes about, given this volume, how long the eruption e-(_)3 25 could reasonably go on, and control column heights in that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
229 1 roughly five kilometer column height range.
73 2 We're also varying a little bit the shape of V 3 the column, whether it's sort of getting blown over or is 4 mushrooming out. It turns out that is not very sensitive, 5 but we're sampling it stochastically anyway. Time of the 6 eruption, which is going to be mass controlled, wind 7 speed, and tephra diameter.
8 The real bullet is that the grain size of the 9 waste, as you might have guessed, is a critical assumption 10 for determining dose. If we have an average diameter of 11 the waste at 10 microns, with a range of waste particles 12 plus or minus one log unit, so we're sampling that from 13 one micron up to 100 microns, and the incorporation factor 14 is the tephra must be twice the size of the waste in order 15 to transport, at 20 kilometers we're having a dose of 50 16 millirems per year. If you move to 30 kilometers, that 17 drops down to seven millirems per year.
18 And you can see that the incorporation ratio 19 for fine waste grain sizes really doesn't matter too much.
20 You're getting a very small variation.
21 So if we say the waste is fine, we don't 22 really have to worry too much about the transport
\
23 mechanism relative to the tephra. It's when we get down 24 -- you jump all the way down to the bottom, about saying
/^s kj 25 that it is one centimeter in diameter, then, of course, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
230 1 we're getting a cutoff in the amount of material that can i 2 be transported quite dramatically.
,-)
'~
3 You can see at 20 kilometers we're about 10-3 4 millirems per year at two times incorporation, but you 5 notice how much that decreases when you go to a 10 times 6 incorporation ratio, just because there isn't that much 7 large material in a basaltic eruption. So we're getting, 8 you know, down to -- I've taken some flack for that, but 9 it's essentially an incalculable dose, rather than saying 10 zero for there.
11 All of these, again, are for one waste 12 package, for the event within that specific timeframe.
13 MR. RYAN: Britt, the waste density, again, in
\- 14 grams per cc was?
15 MR. HILL: Roughly, 10 grams per cc.
16 MR. RYAN: That's what I thought I had heard 17 earlier.
18 MEMBER HINZE: How did the sand dunes form out 19 there?
20 MR. HILL: They are Eolian, blown by the wind.
21 MEMBER HINZE: Do you think that might happen 22 to this tephra?
23 MR. HILL: Very easy to imagine that tephra is 24 going to be redistributed through sheet-wa- processes as ia
() 25 well as Eolian processes out there.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
231 1 MEMBER HINZE: Might be concentrated in --
2 MR. HILL: Certain areas, it is definitely (7-
)
3 going to be eroded, even though we have all come to agree 4 that there is not much of an erosion rate out in the Yucca 5 Mountain region.
6 (Laughter.)
7 We also have come to understand that Lathrop 8 Wells, if the tephra sheet is 100,000 years old, it is 9 completely stripped. So we wouldn't expect to have 10 unconsolidated tephra centimeters to 10 decimeters thick 11 remaining around on the countryside.
12 We are evaluating that. We are doing some 13 initial calculations. If we had small volume eruptions
(" %
- 14 that dispersed materials, say, 10 kilometers or so to the 15 south, how much of tha' mass would be washed down into the 16 40 mile wash drainage, and potentially transported into 17 the accessible or the critical group locations I 1
18 30 kilometers or so down range, and trying to get a handle 19 on whether we need a module and a detailed conceptual 20 model for remobilization of ash or whether, really, the 21 significant process is the direct dispersal rather than 22 remobilization factors.
j 23 MEMBER HINZE: Our friend Tim just talked 24 about the frightening experiences you can have with spot
,23 25 calculations in performance assesement. I was wondering
!xs)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
232 .
1 1 if you've calculated this at 21 kilometers. Do you have 1
,_ 2 any idea about gradients? Do you --
3 MR. HILL: Well, that's the reason we did.
4 Even though we don't have an explicit standard, we're l 5 pretty much getting towards the 30 kilometer standard.
I 6 But we did the 20 kilometer critical group location to l
7 give a sense of spatial variation given the variance in 8 these parameters. It's not simply linear, but it falls 9 somewhere -- at 25 kilometers, somewhere roughly between j 10 t h." se five and seven, for example.
11 CHAIRMAN POMEROY: But it's not linear, did I 12 you say? Is that correct? It's not linear?
13 MR. HILL: Yes. You wouldn't just make a two-rh c a
'\ / 14 point line and go back to the repository. First of all, 15 because the critical group assumptions on the ingestion 16 pathway, we'd have to go back from the non-farm input, and 17 that depends, well, where north of highway 95 you say i 18 there's going to be no farming. Those are for people 19 other than I.
1 1
20 MR. CONNOR: Well, the ash blankets also have 1
21 basically an exponential decay with distance from the 1
22 vent, so just the ash itself is non-linear. And, of ;
1 l
23 course, any incorporation is going to create non-linear on 24 top of that, so it's very non-linear.
ry i
) 25 MR. HILL: Yes. I'd caution people not to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 233 '
1 extrapolate forward or backwards from these dose points.
,3 2 MEMBER HINZE: Is there any indication that ;
i <
\ ) i 3 these models -- the Suzuki, or Andy Woods, or whoever --
I 4 has a constant percentage of error as you go away from the 5 source?
6 MR. HILL: Well, the Suzuki model has been 7 reasonably tested a couple of times in the literature.
8 Lori Glaze and Steve Self, for example, modeled the -- was 9 it the '81 Lascar eruption in Chile using Suzuki. It has 10 also been applied to St. Helens and found to be a pretty 11 good indicator of these larger volume silicic eruptions.
12 I'm not sure -- this is something we've 13 discussed internally quite a bit, why are we getting the
,i-s 1
\/ 14 underestimation in Suzuki? Is it something inherently 15 different about basaltic eruptions, or just conceptually 16 at a very fine scale, because again, we're using very low 17 column heights.
18 Our mis-match is at the distal end. The 19 proximal end works really well with Suzuki. So it's 20 almost like material is falling out too soon, or something i
21 is not quite working right that for a large eruption, you 22 don't catch the error for a large fine grain eruption, you 23 don't catch that, but just down to these sensitivities, 24 the limitations of the model become apparent.
( 25 But again, there has been nothing in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N W.
! (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
234 I i
1 literature to indice this is a real problem. I don't 2 think it's a real problem for the calculations we're t
r~s\ )
\
~'
/
3 presenting right now.
4 MR. RYAN: Were your calculations using a five 5 kilometer column height in this case?
6 MR. HILL: No. The calculations are -- the 7 column height is sampled based on our eruption power and 8 duration. Volume and duration based on historical 9 basaltic eruptions. So that feeds back into some of the i 1
10 empirical relations that Walker and Wilson have developed. j 11 MR. RYAN: Have you used, observed basaltic l l
I 12 eruption column heights as guides?
13 MR. HILL: Yes. I have gone back and dug
'n
' )
14 through the literature on I believe it was 11 historical 15 basaltic eruptions that we feel are analogous in that they l
16 have sustained convective columns that have observed ;
17 column heights. I think it was 10 out of 11 matched i
18 extraordinarily well. One of them I believe is the 1968 !
I 19 Cerro Negro eruption. The duration of that eruption was ;
1 20 very long compared to the ash blanket produced. I think 21 it's just a reporting error that the actual time of temper 22 dispersion was very short, about a week, relative to the 23 months of eruption that are reported in the literature.
24 MR. RYAN: Your isopach maps that you showed
!q,) 25 for Cerro Negro and Tolbachik showed a piling up around NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHIN'3 TON, D C. 20005-3701 (202) 234-4433
235 1 the vent.
em 2 MR. HILL: Sure.
i 1
%"/ 3 MR. RYAN: Do you always find that? Let me 4 back up just a little bit and mention that as we all 5 remember, one of the remarkable things about the 1890, 6 1980 Mount St. Helens activity was that in fact the bulls-7 eye was not near the rolcano itself, but was at Ritzville 8 several tons of kilometers downstream. So that the plume 9 mass distribution from St. Helens was very much like 10 taking the garden hose and sticking it vertically in a 11 very strong ambient wind. You had lots of mass that was 1
1 12 reaching out far away from the volcano. I 13 MR. HILL: Yes. This gets again, into my f) 14 understanding of why we're getting that secondary fallout 15' so far down range for St. Helens was an agglutination 16 process and an adhesion process of these very fine l 17 statically charged ash particles in a wet environment, l
18 which is something we would not be worrying about for the 19 basaltic tephra, which is predominantly not ash-sized and .
20 coarser.
21 We can and have varied this. In the Suzuki 22 model, there's a beta factor that affects column height, 23 or excuse me, column shape, which can give you that 24 secondary fallout.
,9
() 25 MR. RYAN: Sp eading around the trajectory.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
236 1 MR. HILL: 4 little bit more.
, _s 2 MR. RYAN: On either side of it.
!' I 3 MR. HILL: Yes. We vary that within all 4 reasonable ranges. That doesn't accommodate the variant 5 with the deviation from observed deposit thicknesses for 6 Cerro Negro.
7 The first part of your question is not even so 8 much St. Helens, but most basaltic volcanos show a two-9 slope line in thickness versus distance. Within the first 10 five kilometers or so, you have got a fairly steep fall 11 off. Then you get around five clicks, and things level 12 out. It's not truly exponential, but it's alike a two-13 part exponential trend.
p kl m 14 Tolbachik doesn't show that except the 15 proximal data within a couple of kilometers we don't know 16 how thick it is because it's tens of meters thick, and 17 nobody has done ballant. But I suspect it's a similar 18 sort of process.
19 MR. RYAN: You mentioned that the wind was 20 blowing, was it southerly 14 percent of the time?
21 MR. HILL: Yes.
22 MR. RYAN: In your assumptions. What's the 23 other 86 percent of the time? What's the azimuthal 24 variability?
/ \
25 MR. HILL: It's the southern quadrant of I
()
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N W. I (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
i l 237 l
1 think it's 15 degrees sector that encompasses from the 2 repository if north is up. That's where the wind is
- l. 'h
\~'J 3 blowing in that quadrant 14 percent of time. I think it's l 4 not a uniform distribution. It's skewed more towards the 5 north, but other points of the compass are fairly well 6 represented. We have the data for that.
7 But again, it is important to remember we are 8 calculating dose at a specific point, not over an area. )
9 That point is strategically located at that specific 10 place. So we have to when we go through the dose 11 calculations to get the risk, we have got to account for 1
12 the probability of occurrence of different processes. l l
13 MR. TRAPP: Britt? :
/~N l
! )
N' 14 MR. HILL: Yes. I i
15 MR. TRAPP: Just one thing. It kind of goes a j l
16 little bit to Bill's question. Going through the Suzuki j 17 model and comparing it to the distribution of known ash 18 blankets, we appear to have a fairly good correlation. j 19 There's a little bit like you said, 50 percent under 20 estimated.
21 What we are doing right now is assuming the 22 waste basically is following proportional according to 23 these type of factors, the same thing as the ash. Now 24 because of the difference in the particle size, the (O,) 25 particle weight, et cetera, all this other kind of thing, NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
238 1 it's something we really haven't looked at in total 2 detail. It's something we need to take a look at. It fS O 3 will be carried forward in further work.
4 MEMBER HINZE: John, that really gets to a 5 question that I had of Britt and perhaps you. What are 6 the weak points here, and how are you going to go about 7 solving them? It seems to me that one of them is the 8 diameter of the particles, but the other is the 9 incorporation. How do you solve those problems?
10 MR. TRAPP: Let me try one thing that we I 11 know are going to try. This may end up getting there. If 12 you assume, in this case I'm talking really assume some of 13 the worst case estimates, assuming the extreme small I h
\ 14 particle size and not this case, the 10 millimeter, but 15 talk about the one micron size.
16 Assume that this stuff is uniformly 17 distributed through the ash blanket. Take a look at some 18 of these analyses, find out what the answers are. If the 19 answers by the time we get done with them and give them to 20 the performance assessment people and then the people who 21 decide whether they like performance assessment or not, if 22 these don't bother them then why worry about it any more.
23 If the numbers do get to be high enough that 24 we do have to start considering it, then we need to 73
(_) 25 analyze those factors much more critically. So yes, we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
239 l 1 will be doing more sensitivity studies and this t/pe of I l
l g 2 stuff to try to figure out where we have got to sha. pen l
'J 3 our pencils.
l 4 A similar question is this is for one 5 canister, so we're talking about four to 10 times, if you 6 have another Tolbachik. Right?
7 MR. HILL: Right. And that's where we get to 8 the final slide.
9 MR. FOLAND: Britt, one final question. Did 10 you actually modify the particle densities?
11 MR. HILL: Yes.
12 MR. FOLAND: In this code for your --
13 MR. HILL: To accommodate for the mass of the
?\
'- 14 waste, yes.
15 MR. FOLAND: What you said confused me, that 16 it wasn't incorporated.
17 MR. HILL: Okay. It is in there.
18 MR. RYAN: Just curious. Sorry to interrupt.
19 What is the diameter of these particles, the waste 20 particles?
21 MR. HILL: Of the waste particles?
22 MR. RYAN: Yes.
23 MR. HILL: Well, that is what we were varying.
24 It could be anywhere from a starting assumption of one
,/-
q,) 25 centimeter in diameter all the way down to 10 microns in i
NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISJND AVE., N.W.
( WASHINGTON, D.C. 20005-3701 (202) 234-4433 (202) 234 4433
240 1 diameter for the median diameter.
I
~3 2 CHAIRMAN POMEROY: One other just comment. l
] i 3 After the technical exchange, a few of us were very 4 interested of course in that rate of change of those as a i 5 function of distance, if you will. If there are any 6 possibility of getting more information from these models, 7 even though you do have to change the parameters that go 8 into them as a function of distance away from the 9 repository. We're nor _bsolutely sure where EPA is going 10 to end up with this. It would really be nice to know 11 something about that sensitivity, given these assumptions 12 and calculations.
13 MR. HILL: I think we need clear guidance on
[)
\~ / 14 that, rather than something that we would go doing off on 15 ourselves because there is a great sensitivity to 16 calculating any number outside of 30 kilometers.
17 CHAIRMAN POMEROY: I am aware of that.
18 MR. HILL: I am saying we have the ability to 19 do it. It's a policy decision on whether or not we do 20 that. But we can implement it if we would like to.
21 CHAIRMAN POMEROY: Very good. Thank you.
22 MR. HILL: So everybody likes CCDFs. I was 23 presenting mean values bef^re. Of course you wonder is 24 the mean conservativt or would you even put that term to a
/G.
( ,) 25 mean. For these dose calculations, I think the mean is by NEAL R. GIUDSS j COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
241 1 any objective measure, reasonably conservative. Let's
-~ 2 forget the exceedance probability for just a moment and
( \
'.' ') 3 just look at the base case of volcanism, where it does 4 occur.
5 Probability is one here, and we're dropping 6 down on our CCDF. The mean, as represented, leave off the 7 zero for just a second. From our previous example of 50 8 millirems per year, the mean represents the 94th 9 percentile of that data. So it's hard to see how the 94th 10 percentile as your mean is not a conservative measure of 11 the dose that we're expecting here. So that's the point 12 of the first.
13 VICE CHAIRMAN GARRICK: So this means that YY 14 your PDF's are very skewed. If this were made up of a 15 series of scenarios, if you had represented this as a 16 family of curves and a cut curve across the CCDF would 17 indicate a highly skewed PDF.
18 MR. HILL: Yes. You have got a source of 19 anxiety down at this region that's driving your mean. You 20 do have large -- sometimes you are going to get a large j 21 dose out there. I'm not saying risk, I'm saying the dose 22 can sometimes be troubling. That's just what the 23 distribution is saying. That there are certain scenarios 24 where you are not going to of course get much of a dose, 1
25 but there are going to be ones that are going to give you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
~~
242 ;
i 1 a large effective dose out there.
~, 2 I would like to emphasize that these r )
' ' ~ ~
3 calculations are not what I would characterize as worst 4 case. These are reasonable bounds on our understanding of 5 the processes at this time. We have not gone out to try 6 to find worst case scenarios to evaluate at this stage.
7 It's calculations presenting a range of understanding, if 8 you will. Not a worst case scenario. I would urge people 9 to not present these as the upper bound of the scenario.
10 It's the upper bound of the current parameters.
11 CHAIRMAN POMEROY: I was hoping maybe we 12 couldn't go into that, but since you brought it up, you 13 are talking about 10 canisters.
A
( )
's J 14 MR. HILL: Yes.
15 CHAIRMAN POMEROY: Do you think that might be 16 100 canisters?
17 MR. HILL: No. I don't.
18 CHAIRMAN POMEROY: Thank you. What are you
)
19 talking about in terms of a volcanic volcanism probability 1
20 number?
l 21 MR. HILL: Ten to the minus three on 10,000 22 years or 10 to the minus seven on an annual basis.
23 CHAIRMAN POMEROY: It's not the upper bound of 24 your --
(~%
(_,) 25 MR. HILL: Of our current models and current NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
243 1 understanding, I am comfortable with that range as the
~s 2 upper limit. I do need to acknowledge that there are
( )
U other models in the reviewed literature that would get an 3
4 order of magnitude greater than that, or that there are 5 some other processes that we need to be considering about 6 spatial shifts through time on a regular fashion that may 7 impact that number.
8 But based on the recurrence rate that we have 9 out there in the quaternary, it's hard to see that getting 10 up into that 10 to the minus six range.
11 CHAIRMAN POMEROY: Of the other pieces that 12 are built into this, as listed over there on the lefthand 13 side on the bottom, are there ones that you can clearly 14 identify as not being upper bound?
15 MR. HILL: Waste grain size?
16 CHAIRMAN POMEROY: Yes.
17 MR. HILL: Dose point, the closer in, and 18 depending on your scenarios, that's going to affect the 19 dose quite a bit, as we're all aware.
20 Depending on are we talking about a single 21 dose point or a range of dose points, it gets back into 22 the wind and the distribution on that. These are more of 23 standards that we don't control. But I think you are 24 going to see a large difference if I said 50 waste n
(_,) 25 packages versus 10 waste packages. That's not 100.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
244 1 That's not an order of magnitude. It's only 50 percent,
,f 3 2 depending on which side of the equation you are operating
\' )
3 on. But I think that would start giving you significant 4 increase in dose.
5 MR. TRAPP: Britt, on your previous slide, 6 there was a very important point brought up. I think it 7 kind of addresses some of this, or maybe it wasn't one of 8 the previous ones.
9 Anyway, the point is that we are making a 10 basic assumption in all these models that we are dealing 11 with what we call an undisturbed case or a case where the 12 repository isn't there. We're not totally sure what the 13 effect of the stress reduction, et cetera, this type of
'- 14 thing will have to the possibility of any magma going 15 through there. You can hypothesize both ways. One, that 16 it will basically fizzle out in the repository horizon.
17 The other, that you end up with kind of a coke bottle 18 effect. You shake the thing up and blow the top off the 19 whole thing.
l 20 We don't know right now. It's an area that we 21 would like to take a little bit more look at and find out 22 if it really does effect any of these numbers.
23 MR. RYAN: One of the things that happens, 24 depending on whether or not -- depending on the level of
,q
(_j 25 the heat capacity of an object that's potentially going to !
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 123 RHODE ISMND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 245 l i
1 be encased by basaltic lava, depending on its heat 7s 2 capacity and its volume, and the amount of gas say is
\ ,]
3 within that, is that if the heat capacity is sufficiently 4 high and if the volume is sufficiently large, and if the 5 included gas phase is sufficiently small in volume, that 6 the enclosing basalt will in fact chill a rind around that 7 object, very much like the streamlining on a volcanic 8 bomb, or in fact the so-called tree molds that are often 9 observed in the field which are passive markers of points 10 of prior chilling of basalt.
11 Once chilled, of course, the rock is a very 12 poor conductor, and can be rather insulating. So I think 13 what I am saying is it's by no means a foregone conclusion i
/'Ni
l 14 in my own mind that contact of a waste canister by basalt 15 mandates a rupture of that canister and mandates a release 16 of the radioactive waste contained within it into some 17 ambient flowing basalt stream. This may well happen, but 18 I'm not completely convinced it's a guaranteed fait 19 accompli.
20 MR. HILL: I share your sentiment exactly, 21 that I think we have a very poor mechanistic l
22 understanding. We do not have I should say a mechanistic 1
23 understanding of what happens with magma heating the waste 24 package. Certainly to have it in contact with magma, (O) _,
25 versus putting it into the eruption column or into the 1
i NEAL R. GROSS I l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
i
246 1 widening the vent and in training it during one of these
- 2 brecciation events though, may be a very different
~
3 scenario to worry about.
4 I would say that just given the mass and the 1
5 load, the physical load that you are going to put on it i l
6 during that process, it's hard to see it not failing. So l 7 I would agree with you that the dynamics of contact of the 1
8 magma and the small conduit during initial ambient or 9 initial ambient flow may well not be thoroughly fracturing 10 the canister. But when we get to the later stages of the 11 eruption, they may well be weakened, certainly by that 12 stress, if not breached by it.
13 Then we put this additional mechanical stress i
' 14 on a thermally stressed system, that's one of the basis, 15 why we are saying our starting assumption. I want to 1
16 emphasize that assumption, is that the canister has 17 failed. That's to us reasonably conservative. If we can 18 come up with a mechanistic basis that shows like I said, 19 the canisters are robust against certain or all eruption 20 conditions, then we have got a solid basis for evaluating 21 performance models. We don't have that, it's not coming 22 out of the canister engineering grot ,s. So we have to go 23 forward with an assumption that isn't going to under 24 estimate risk.
TT
( ,) 25 MEMBER HINZE: What's the quality between one NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234 4433
247 1 and ten of the communication between your modeling and 7- 2 what DOE is doing in terms of modeling?
l
\
)
3 MR. HILL: After the technical exchange, it 4 sounded like it was pretty good. DOE wanted to use the I
5 ash plume modeling to better simulate the dispersal of l 6 material. Before with the older standards, we weren't 7 concerned about a critical group or really dispersing it.
8 It was waste on the ground constitutes the basis for doing l l
9 the dose. We didn't have to worry about dispersion.
10 It sounds like there has been an l
11 acknowledgement that there are different ways to approach 12 incorporation ratio, what the sub-surface area disruption l 13 might be. Greg Valentine has one approach. I have a ;
(~N
- 1
'x 14 different approach based on the way I am looking at 15 basaltic volcanos.
16 MEMBER HINZE: You have got a 50 meter radius?
17 MR. HILL: Fifty meter diameter.
18 MEMBER HINZE: Diameter, 50 meter diameter.
19 What does Greg have?
20 MR. HILL: He is looking at a percentage of 21 entrainment based on work he has done at the Lucero field.
22 There's a Journal of Geoloav paper by Valentine and Groves 23 last year. When I last talked with Greg at the IAVCEI l
24 meeting, that's pretty much what's in the synthesis (G) 25 report, is the bulk of the work that was done out in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE lELAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
248 1 Lucero field.
<s a 2 So he's looking at a stratigraphic section out l
\_)
3 there for xenoliths in a spatter cone, by his own 4 admission in a spatter cone, and using that to say well, 5 given that we have a 10 meter thick repository interval 6 versus a crustal section of X kilometers being sampled.
7 Therefore, that's the percentage of material that's l 8 available for transport.
9 Then he uses the average thermal load per 10 acre, I believe, to fold that into the amount of material 11 for dispersion.
12 MEMBER HINZE: So there's no problem in data 13 communication between you and DOE in terms of canister or I j
\/ 14 canister strengths or alterations?
15 MR. HILL: I wouldn't be presumptuous on that.
16 MR. TRAPP: I would say there's no problem, 17 because really in that area it's an area that there hasn't 18 been communication.
19 MR. HILL: Quite frankly, previous TSPAs have 20 always assumed that if magma touches a canister, it's 21 failed. So there hasn't been a disagreement on that 22 point.
23 MEMBER HINZE: One of the problems that we 24 have had that apparently has been present in the past is
() 25 this communication problem. I am trying to see whether NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
249 1 that -- you know, we have really been pleased with the r'y 2 improvements in that. I am just curious as to how it is a
s V continuing.
3 4 MR. HILL: In a way, it's hard to evaluate 5 because there's such a difference in performance models 6 now with the dose standard versus the release standard.
7 Tim presented what was done in TSPA 95, but the volcanism 8 scenario is really based on TSPA 91 and TSPA 93, which was 9 designed around a whole different really series of 10 problems. So it's not really fair to compare whether they 11 are going to or have going to or did going to incorporate 12 this when they haven't done the analysis yet.
,_ 13 Certainly everything we heard at the technical
(
)
14 exchange is a willingness to consider it.
15 MEMBER HINZE: Great.
16 CHAIRMAN POMEROY- . And just for the record, 17 Britt, would you kind of read for me what that second to 18 the last bullet says? l 19 MR. HILL: That's where we're getting up for 20 the final stage here. I have gotten to the CC, the mean i 1
21 is conservative. There is from the previous slide, the l 22 mean is less than 50 millirems per year. We talked about 23 the geological basis for saying it may disrupt 10 24 canisters. So again assuming that we're just doing
\s ,) 25 something simple on this, that we're, dealing with 500 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
250 1 millirems per year. The probability models range up to 10
- 2 to the minus three for 13,000 years. So there's risk. 10
~'
3 to the minus 3, 10 to the minus 4, 500 milirems per year.
4 That's where the multiplication comes through for less 5 than or eaual to 0.5 millirems per year.
6 Now I would like to put one particular spin on 7 that because we're always talking about significance. We 8 have significance relative to an undefined standard or 9 release standard. I don't think anybody is saying that 10 the release standard is going to be a millirem per year.
11 A volcanism of itself is not going to be causing the 12 repository to fail from a release standard of around that 13 range.
,/m
- i
'x 14 The significance in terms of the program is 15 even harder to evaluate. Tim showed a range of numbers 16 here. We don't have a 30 kilometer dose from the DOE that 17 we can directly compare to for this. From the NRC's 18 numbers, comparing five kilometer drinking water with 30 19 kilometer Amargosa Valley, it's about an order of 20 magnitude decrease. For the mean dose from the NAS 21 recommendations is 14 millirems. Now how significant is 22 another milirem to all of that? In one sense, it's about j 23 a 10 percent contributor. In another sense, it doesn't l l
l 24 push us over the limit.
h 25 But this is not the real evaluation of l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 2344433
251 1 repository performance for undisturbed settings. T:lere
,- 2 are many scenarios being developed that show that the
- )
3 expected dose from the undisturbed repository is 4 significantly less than 10 millirems per year. In that 5 case, this might be the only significant release process 6 for the entire repository system.
7 So there are, I would emphasize, different 8 ways to look at significance, absolute versus relative.
9 VICE CHAIRMAN GARRICK: Just looking at this 10 one for a moment longer, and assuming you are addressing 11 the 10 to the minus three volcano.
12 MR. HILL: Yes.
13 VICE CHAIRMAN GARRICK: If you were asked to
('%,
il 14 draw a whisker diagram of the 90 percent confidence 15 inter il, where would that appear, to your best judgement?
16 In other words, if you drew a fiva percentile, where would 17 you put the five percentile and 95 percentile at the 10 to j 18 the minus three probability? I am just trying to get an j 19 idea of your estimate of the uncertainty about the mean.
20 Do you know what I mean by a whisker diagram?
21 MR. HILL: I know what you mean. I am just 22 very reluctant to do that.
- 23 VICE CHAIRMAN GARRICK
- Yes. Well, you must 24 know where it is if the mean is 94 percent, because that's
(\ 25 got to come from that kind of information. So it's
(
s.s
)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
252 1 somewhere.
s 2 MR. HILL: I think that's something I would s
( 't 3 rather follow up accurately rather than giving a back of 4 the envelope --
5 VICE CHAIRMAN GARRICK: Yes. Well that would 6 certainly answer the question of why the mean is that, if 7 you had the whisker diagram above the 10 to the minus 1
8 three ordinate. l 9 MR. HILL: The other thing is that these l I
10 calculations have all been for a direct release only. We )
11 haven't been evaluating the indirect release. There is l l
l 12 considerable debate within the program about what the 13 significance of that could be. If we have essentially the
/T Y 14 igneous dike, the intrusion penetrates the repository 15 farther than .he area of direct disruption, it will cause 16 X number of additional canisters to fail. But those l
17 additional failed canisters will not be released by 18 volcanism, so what's the performance impact of failing 19 additional canisters on the order of maybe 100 more 20 canisters at a specific time.
21 Again, that comes back down to certain models 22 are saying that by 1.000 years you've got 50 percent 23 failed, in which case another 100 doesn't make too much 24 difference, or that through coupling processes and others, r,
(,) 25 that the canisters stay in tact through the lifetime of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
253 1 the repository. It's hard to find a straight decision
, e3 2 point to say this is or is not significant.
4
\_/
3 MEMBER HINZE: Britt, we thank you very much.
4 We apologize for keeping you up there, but it's obvious 5 that there was a great deal of interest in what you had to 6 say. We appreciate it very much.
7 I would like to suggest that we take a 12 8 minute break and come back. I apologize to our colleagues 9 that are coming up next, Abe Van Luik and Tim Sullivan and 10 John Trapp, but we'll have those next.
11 (Whereupon, the foregoing matter went off the 12 record at 3:28 p.m. and went back on the 13 record at 3:46 p.m.)
! )
\' 14 CHAIRMAN POMEROY: Let's reconvene right on 15 schedule.
16 MEMBER HINZE: Our next -- what schedule are 17 you on? Our next speaker will be Abe Van Luik, who will 18 be talking about the incorporation of volcanism into TSPA-19 VA.
20 MR. VAN LUIK: Don't pay any attention to 21 April 24 on the first viewgraph. This is no longer my 22 title. I am actually -- but it's still me.
23 MEMBER HINZE: It's still the Department of 24 Energy?
( ,) 25 MR. VAN LUIK: Yes. What they have done is l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
- (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 1
254 1 made me the Technical Manager of Performance Assessment, 73 ) 2 so I don't have to worry about the other types of details
(
v 3 that come with management.
4 What I want to do is summarize the volcanic 5 scenarios we have used in previous TSPA analyses, going 6 over some of the same territory that was just covered, 7 quickly however. I want to do a summary of the volcanic 8 scenarios that we want to consider in TSPA-VA. We want to 9 incorporate alternative interpretations of the probability 10 of occurrence, and incorporate alternative models of 11 direct and indirect effects and consequences.
12 Let me move right along here. This is what we 13 hsve done previously. We looked at basaltic volcanism.
6 4
'd' 14 We looked at the left side of this tree. The intrusion 15 acts directly on the repository. A dike forms. There may 16 be transport of waste. There could be indirect effects 17 also from no waste magma contact. TSPA 91 looked at this.
18 TSPA 93 looked at this. We have before looked or Link, et 19 al, has looked at basaltic cone forms. You notice here 20 that there is a point probability and the PVHA that you 21 have heard about already gave us some new information 22 there. But this is where we were up until we started 23 ' working on TSPA-VA.
I 24 If you remember TSPA-91, we looked at direct l
/s l
(_,) 25 entrainment of waste to the surface. The amount of the NEAL R. GROSS ,
COURT REPORTERS AND TRANSCRIBERS I 1323 RHODE ISLAND AVE., NV!.
(202) 234-4433 WASHINGTON, D C. 2000M/01 (202) 234-4433 l l
255 1 entrained waste was a function of the volume of the dike,
,s 2 extent of wall rock erosion, area of the repository
('~' )
3 intersected by the dike. We had random dike orientations, 4 lengths and amount of wall-rock erosion.
5 One thing about TSPA-91, we had two separate ,
6 groups doing TSPA-91. We had Sandia and the Pacific 7 Northwest lab. If you recall the presentations we made at 8 that time, Sandia used all the information gathered by the 9 project. PNL went off in its own direction, and went to 10 the literature and came up with a more conservative 11 calculation, but the bottom line was comparable, which 12 gave us some amount of feeling that we had a pretty good 13 case.
l3
14 TSPA-93, we looked at indirect effects, 15 looking at intrusive events, accelerating waste package 16 and waste form degradation, the acidic gases that you were 17 talking about a minute ago.
18 What were the results? Well, in TSPA-91, we 19 felt the models were conservative. The probability that 20 we picked was 2 X 10 to the minus four over 10,000 years.
21 We lorked at the old EPA sum. If you recall 40 CFR 91.
22 You can see that this point right here is defined by the 23 probability of occurrence. This is a very similar plot to 24 what you were just shown by the center staff.
( ,) 25 Then our point right here, if you look at the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., kW.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
256 1 compliance points for the old EPA standard, we were a 2 couple of order magnitudes away from that in any direction 7-~
t 1 i/ ,
3 that you want to go in.
4 So basically, we figured that in TSPA-91, we 5 had covered the issue of direct entrainment and movement 6 of waste to the surface.
7 TSPA-93, again looking at the same standard, 8 but this time we did some calculations way out to a l 9 million years. You can see that given that your l 10 probabilities were fixed but increased your time period, 11 that this line just keeps on marching up as you extend 12 your time. But even at a million years, we could still l 13 meet as it shows here, a 10,000 year standard. i rx
! \
\/ 14 Basically what this shows right here for the l
15 10,000 year case is that the indirect effects had no real 16 effect. The accelerated waste package corrosion a little 17 bit, and that's it. The rest of it was pretty much a 18 normative case.
19 Now we're going to improve on these because 20 now we have the PVHA which gives us credible estimates of 21 probability of occurrence and uncertainties for 22 intersection of a dike with the repository. We have the l 23 Los Alamos national laboratory's volcanism synthesis 24 report. We have also the modeling that you have been (x
) 25 hearing about from the Center.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433 l
l 257 l
1 At this point, I would like to give a few 1
2 seconds to Tim Sullivan. He can have a couple of minutes l 7--
( /
3 if he would like, to give us a little overview of the 4 volcanism synthesis report.
5 MR. SULLIVAN: I hate to do that. Abe, you ,
1 l
6 are on a roll, and you haven't been interrupted yet. !
7 In response to your question, Bill, I had a l
8 little bit more information. Britt is probably as l
9 familiar with this as I am.
10 There's some information provided in the 11 volcanism status report, the 1995 report, that presumably 12 you all have. In the synthesis report itself there will 13 be significant updates to the consequence analysis (n
\-) 14 chapter, that's chapter 5.
15 There are four main topics that Abe has 16 covered there. The first is the final detailed results of 17 studies of eruptive effects and sub-surface effects, 18 including lithic abundance studies that can be used then 19 to constrain the estimates of waste entrainment for 20 various eruption styles. That is based of course on 21 analog studies that Greg conducted in 1995.
22 Second, interpretations of the alteration of 23 silicic host rocks, reduced by basaltic dikes and sills 24 for both vitric and zeolitic tuffs. This will support the O)
(_ 25 indirect effects, model updates that Abe is going to refer NEAL R. GROSS COURT REPORTERS AND TRANSCRlBERS l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 2344433 l
l
258 1 to shortly. Again, these are analog studies.
rs 2 Thirdly, also reported in there are studies of IU) 3 factors that influence shallow intrusion geometries.
4 Then finally, preliminary modeling and 5 theoretical studies of the sub-surface effects of 6 intrusions will be reported in the synthesis report.
7 In terms of that communication question that l 8 you asked --
9 MEMBER HINZE: That was four.
10 MR. SULLIVAN: Psur, correct. I would just 11 make this comment. The contractor's technical staff for 12 the volcanism program now consists of Frank Perry part-13 time. So there's little opportunity really for (3
\
' -)
14 communication except at professional meetings presumably 15 as Britt referred to before.
16 MEMBER HINZE: Tim, one of my concerns about l' the communication is that the Center and the NRC are 18 moving with dispatch with the Suzuki model, and trying to 19 put some constraints on it. What I was trying to get at 20 is there any information that's coming out of your 21 consequence studies that might be useful to them? Or is 22 this information in their hands or just not in the public 23 arena?
24 MR. SULLIVAN: For the dispersion models,
()
,r %
25 there is no new information, with the exception of the NEAL R. GROSS COURT REPORTE9S AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(2C2) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
259 1 lithic abundance studies that would contribute to waste fx 2 entrainment estimates.
I )
<./
3 MEMBER HINZE: Journal of Geoloav article.
4 MR. SULLIVAN: Well, updated from that, yes.
5 We'll get you that information as soon as we can.
6 MR. VAN LUIK: Thank you. I think a point on 7 communication also is that we are counting on being able 8 to cooperate with the Center, with the NRC's permission of 9 course, in obtaining their models and evaluating them and 10 vice versa. If there's anything that we have in the 11 performance assessment world that you want to look at, I 12 think we can make that available to you.
,,, 13 So what are we proposing to do in terms of I \
' 14 scenario analysis for TSPA-VA? The same diagram as 1
15 before. We now have the PVHA instead of a point 16 frequency. We are going to do in tact waste entrainment I i
17 and distribution at the surface. We are also going to 18 reevaluate what we did in TSPA-93, given new information, 19 looking at no waste magma contact directly, but the waste 20 is -- the source term is modified.
21 We are going to look at the basaltic cone, 22 waste entrainment and distribution at the surface. Then j 23 we have already done the work by' Link, et al, in 1982. We 24 are going to seriously consider the NRC's Tephra
,a
(_,) 25 dispersion analysis that you have just been exposed to.
! NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344 433
260 1 Then of course as Tim explained, the Los Alamos volcanism 7- 2 synthesis report will support this effort and also look at t
)
~'
3 the intrusion acting indirectly on the repository through 4 altering sub-surface flow, either through lateral 5 diversion in the Calico Hills or creating fast path flow 6 along a dike alignment.
7 So these are the things that we are proposing 8 to handle in the TSPA-VA which is due next year.
9 MEMBER HINZE: Before you remove that. You 10 are not going to get by that easy. Why the light line, 11 the thin line around magmatic alteration of waste? It was 12 that way in the other one too.
13 MR. VAN LUIK: I believe that you need to go C\
t !
\/ 14 to the document that describes these particular ideas. I 15 thought that this one was basically through discussion in 16 the document, said to be really an alternate view of this i 17 one.
l l
18 So this should be the magma modifies -- the ;
1 19 nearby magma modifies the source term. This is really the 20 same thing as the alteration of the waste. So I think l
I 21 this is something that could be cleaned up.
22 So what are we looking at for alternative 23 models in terms of probability and disruptive events. We 24 want to use the elicited values from PVHA. We think they
/"N
() 25 have a good pedigree.
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
261 1 We want to employ importance sampling to s 2 assure that events are sampled and weighted appropriately.
( )
3 As someone pointed out this morning, with the 10 to the 4 minus eight probability, doing 1,000 runs doesn't assure 5 that you are going to hit an event.
6 We want to use fixed best estimate, direct 7 effects consequence models. Direct effect is considered 8 to have about a 10,000 times greater consequence than 9 indirect effects to do this important sampling.
10 We want to do consequence analyses using peak 11 dose to average members of the critical group located at 12 five and 30 kilometers up to 10,000 years and then up to 13 100,000 years.
/~~';
Yl 14 You don't see a million years here like we did 15 in TSPA-91 and 93 because since TSPA-95, we have 16 accelerated the movement of water in a mountain 17 con,iderably. We believe that actually our peak doses 18 will show up before 100,000 years.
19 We want to look at sensitivities to 20 alternative probability distribution functions, still 21 honoring the elicited PVHA frequency of intersections. We 22 will create alternative CCDFs at peak dose so that you can 23 have a -- what did you call them, a horse tail.
24 What are the alternative PDFs that we're going
/~T is ,/ 25 to be studying in the sensitivity analysis? Looking at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
262 1 fixed 50th and 95th percentile values, for example.
,_s 2 Sampling from the raw probability distribution function,
/ \
i /
' ~ '
3 where we are transforming it to a log normal that honors 4 the 50th and 95th percentiles. Sample from a uniform 5 distribution of elicited median values from a log-uniform 6 distribution encompassing the loth and the 99.9th 7 percentiles.
8 Discarding the one outlier. You saw the 9 probability distribution function. There is one outlier 10 at the low end of the dist;ibution, and recompiling the 11 PDF. We will do these as sensitivity analyses. Then we 12 will document these results in the TSPA-VA.
13 This is of course addressing the question of C'\
$- / 14 how sensitive is all this to the shape and the properties 15 of your probability distribution function.
16 Looking at direct effects and consequences, we l 17 will review the appropriateness of the conceptual l
l 18 assumptions in the Center's model and compare it to models l
! 19 used in TSPA-91 and recent information from the volcanism 20 synthesis report.
21 Assuming that the Center model is l
22 representative, we would like to use it and define for it 23 reasonable ranges for key input parameters. What we would 24 like to do is use DOE experts and a subset of the
( ,) 25 probabilistic volcanic hazard assessment external experts.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
263]
1 This is not a formal expert elicitation, however.
2 We would like to make sure that the bounding i
7-1
\'~/
3 parameters selected by the Center for the analysis they 4 just showed you are represented within the range of 5 parameter values that we select. Examples are the 6 magnitudes of ejected material, depth and percent of wall 7 rock entrainment, dike length and width, the eruptive 8 material characteristics, and there's many more factors 9 that go into that model.
1 10 We would like to conduct conditionEl 11 simulations of consequences, meaning that we set the 12 probability of occurrence equal to one, and sample ov3r 13 the whole range of the parameters that we will fix using
/_N r a
\> 14 this small subset of experts and document these results in 15 TSPA-VA.
16 MR. TRAPP: I do have to make just one l
17 comment, Abe. That word bounding in there, we would 18 object to strenuously.
I 1
19 MR. VAN LUIK: Yes. That's why we put it in !
20 quotation marks. It's our view that you did some 21 bounding, and it's your view that you did some very 22 sensible estimates. You know, that's --
l 23 AUDIENCE COMMENT: (Inaudible.)
24 MR. VAN LUIK: We would love to have one of 25 your people as our expert, but I think it's illegal for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
264 1 you to work for us. I think it would really mess up the
.s 2 licensing process too. So we need to I' spect each other V) 3 and our independence I think.
4 Alternative models for indirect effects and 5 consequences, we want to review the previous work that we I l
6 did. The indirect effects on TSPA-93, we think was pretty 7 good, but it did not go far enough.
1 8 We would like to revise those models or 9 develop new models, including models to bound the effects 10 of the enhanced degradation that we looked at in TSPA-93, 11 enhanced degradation of cladding and waste form. Revise 12 the solubility of radionuclides, given the new chemistry 13 that will be imposed on the system. Modify transport
, 4 k/ 14 characteristics along likely flow paths. Look at a 15 revised saturated flow field if necessary.
16 Conduct conditional consequence modeling.
17 Again, this is conditional probability one, of indirect 18 effects using bounding models of indirect effects. This 19 time the bounding is not in quotes because this is our 20 bounding, not your bounding.
21 Reconfirm that the consequences of indirect 22 effects are significantly less than those due to direct 23 effects. We can't reconfirm that of course. That will be 24 a different story. But this is our expectation. Again,
,q
( ,) 25 document this in TSPA-VA.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4432 WASHINGTON, D C. 20005-3701 (202) 234-4433
I 265 1 MEMBER HINZE: And you will have hydrologists, 2 et cetera, working with you in putting this all together 7_
V 3 then?
4 MR. VAN LUIK: Of course. PA no longer works 5 in a vacuum on our project.
6 MEMBER HINZE: Great.
7 MR. VAN LUIK: TSPA-VA will utilize recent l l
8 results from the PVRA, the synthesis report, and the 9 Center work. The sensitivity analysis, the probability of 10 occurrence, direct and indirect effects and consequences 11 will be conducted and documented. We will document i
12 everything that we do in TSPA-VA. l 13 If either consequences or risks are k/ 14 significant, then volcanic scenarios will be included in 15 TSPA reference case. In other words, if our result also 16 comes out as the Center's result did with a 500 millirem 17 per year result, we will include it, even though the risk 18 at that point is still .5 millirem per year.
19 But if both consequences and risks are 20 insignificant, then we will document it in TSPA-VA, but we 21 will not include it in the reference case, and probably 22 will not go forward with that case in licensing. We will 23 say this issue is closed at that point once the NRC of 24 course agrees to that.
CT I have
( ,
) 25 So this is our plan for the TSPA-VA.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
1 266 1 just donated about 20 minutes of time because you asked no
,s
, 2 questions. I think the bottom line is look forward to a
()
3 good product and a comprehensive treatment of the subject.
4 VICE CHAIRMAN GARRICK: Before you leave, Abe.
5 MR. VAN LUIK: Yes, sir.
6 VICE CHAIRMAN GARRICK: You went through this 7 awfully fast.
8 MR. VAN LUIK: By the chairman's request.
9 VICE CHAIRMAN GARRICK: We'll have other 10 opportunities to talk to you. But there are some issues 11 that as a risk analyst that concern me a little bit. One 12 of those is that you would allow yourself to get into a 13 trap of talking about either consequences or risks.
l'D kl 14 If you talked about consequences or 15 likelihoods that may be a rational process. But to talk 16 about consequences or risk or to put yourself in the 17 position where you might be measured on the basis of 18 consequences is really asking for trouble. l l
19 MR. VAN LUIK: I would agree with you to some 20 extent. In fact, my recommendation was to go strictly for ,
l 21 risk. If the consequence was high, but the probability 22 made the risk low, to just leave it out. l 23 However, at the technical exchange that we ,
l 24 just had with the NRC, one of the agreements was that we
(_) 25 would look at either consequence, in other words, set the j NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
267 1 probability equal to one and see what the consequence is,
,- m 2 or risk. If those are high, we will continue forward with
\
3 the analysis. So this is part of our agreement with the 4 NRC staff I believe.
5 Is that correct, Tim? You are going to talk 6 next with John on the agreements, but that was my 7 impression. If not, strike those words and we'll just go 8 for risk.
9 VICE CHAIRMAN GARRICK: Yes. I think that is 10 very dangerous. The time that we have gotten in real 11 trouble in the reactor risk field is when we made any 12 gestures at all towards decoupling of risk into components 13 that in this case don't even make sense. It doesn't make I /
14 sense to talk about consequences or risk given that 15 consequences are an inherent part of risk.
16 MR. VAN LUIK: That's true, yes.
17 VICE CHAIRMAN GARRICK: You have just muddied 18 up the language of the business a lot. So it is just a 19 suggestion that you may want to reconsider what that's all 20 about.
21 MR. VAN LUIK: I will definitely reconsider.
22 VICE CHAIRMAN GARRICK: The other thing I 23 wanted to comment about a little bit is every once in a 24 while some of us sort of reflect on what we would really 7-m j 25 like to see come out of the performance assessment NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHING foN, D C. 20005-3701 (202) 234 4433
268 1 business. You have an enormous risk communication problem 2 with the performance assessment.
>f~,T 3 MR. VAN LUIK: Yes.
4 VICE CHAIRMAN GARRICK: So given that this is 5 entering into a public arena, I would think that you would 6 want to do everything you possibly can to make the results 7 and the presentations as understandable and comprehensive 8 as possible.
9 I don't know for sure how to do that, but one 10 thought that occurs to me is that out of all of this 11 analysis, it would be very nice if in the end, you came up 12 with what you would consider to be a certain number of 13 scenarios that have been so defined that they encompass
(' )
14 all of the issues and questions that have been suggested.
15 Let us assume that that might be 20 different scenarios.
16 You could somehow present those scenarios in 1
17 such a manner that you would be able to address each one l 18 on the basis of its own merit and to some extent take a 19 first step towards making the kind of commitment that the 20 public will be looking for from the scientists. Namely, 21 which of these scenarios do you think in fact we should 22 worry about the most.
23 In the end, what you would really like to do 24 of course is to appropriately weight the scenarios such
/~5
( ,) 25 that you really do answer the question of what is the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
269 1 risk. If you can't do that, at least structure it in such fs 2 a way that you can talk to each one of them and I \
'm ! communicate as much as you can about your confidence of 3
4 which scenarios are the most important in terms of risk.
5 Are you planning to do something like that?
6 Because right now, you are suffering from the severe 7 problem of conditional risk assessment rather than an 8 unconditional risk assessment. You present information on 9 the basis if we have this infiltration rate or this set of 10 initial conditions, this is the set of CCDFs we get. I 11 don't think we're doing the job when we do that.
12 I think we need to go that next step and say 13 well, this group is what we think are if not a direct 1
\2 14 result of combining all of them on the basis of the 15 evidence that's available, at least on the basis of our 16 best shot, our best judgement, these are the one or two or 17 three scenarios that we're convinced from a technical 18 standpoint, maybe even not even from a compliance
- 19 standpoint, but from a safety standpoint, are the most 20 importanu.
21 That would enhance, it seems to me, the whole i
! 22 communication issue of what I nate come to call PPA is all l
23 about, probabilistic performance assessment. Because up 24 to now, it has not really been that.
f3
( j l 25 MR. VAN LUIK: I fully concur with what you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
270 1 are saying. In fact, so does my management. Because one 7_ 2 of the things that we have been loathed to do up to this 3 point is exactly what you have said, which is to actually 4 make a pick out of all these choices and say this is the 5 most likely path in our judgement. It would be nice to 6 have everything in a PDF so that you can do that 7 probabilistically.
8 What our management did last week was an 9 unconscionable thing. Lake Barrett had to brief Tara 10 O'Toole yesterday on the latest results from PA and on a 11 couple of strategic things that are important to her.
12 Lake finally said, you guys, give me an expected value 13 case now because I am tired of hearing it could be this,
,/~m
-- 14 it could be that. Give me what your gut instinct says, 15 even if you can't do it.
16 You know, we made the argument that we 17 couldn't at this point do a fully probabilistic 50th 18 percentile or expected value case. He said well give me 19 your best shot. So we actually created an expected value 20 case with plus or minus a standard deviation because he 21 said it is impossible for him to brief someone who is not 22 steeped in this whole folklore of CCDFs and statistics and 23 have them understand what the actual expected performance l
24 is. So he is pushing us in that direction. I believe
(_) 25 that from now on, we will show those types of results.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D,C. 20005-3701 (202) 234-4433 l
I 1
1 271 l
1 For sure, in the executive summary of TSPA-95, and that 2 will go to Congress and other people.
/oT/
t.
3 We will show what we as an integrated program 4 expect the mountain to perform like. We will try to do it 5 in ways that the TSPA-VA executive summary that actually 6 goes up on the Hill and everywhere else, can be read by a 7 politician and they will understand these are basically 8 the risks that we are facing.
9 So those are the marching orderc that we have, 10 is to make it understandable and what's that other word, 11 traceable and transparent.
12 VICE CHAIRMAN GARRICK: Well, the only reason 13 I made the speech is because as you went through this, it (ni V 14 seems like you were trying to respond to so many groups 15 and so many different requirements that it wasn't clear to 16 me that you the analyst really believed what you are doing 17 any more. You were kind of a victim of the process. The 18 process is being determined by expert elicitation 19 activity, ou t h.. organizations, and what have you. That j 20 has to be considered.
1 21 But I think in the end, what we want to know I l
22 is what the real experts that are pulling all this I 23 together really believe and can defend with fervor and 1 l
l 24 conviction.
rx 25 MR. VAN LUIK: Well, that's another reason why NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
272 1 if you look on the viewgraph, they are a little bit sneaky 7- 2 because they keep saying these will be sensitivity studies
'~'
/
3 and documented in TSPA-VA. But the bottom line is only 4 those things that have high consequence and reasonable 5 sized risks will be reported in the actual nominative case 6 of the TSPA-VA and be carried forward into the executive 7 summary.
8 So the sensitivity studies will satisfy every 9 client that we have ever had or thought we had, but the 10 actual executive summary will sum up the case that we want 11 to make to the public.
12 VICE CHAIRMAN GARRICK: Okay. I still wish I 13 had never seen these last two bullets.
(~)/
4
'/
~ 14 MR. VAN LUIK: There were a couple of 15 discussions on that. I thought that we were bound by the 16 agreement that we made, but we'll find out about that 17 next, I suppose.
18 MEMBER HINZE: Further questions?
19 MR. VAN LUIK: I appreciate the suggestion i
20 then. I'll go home and take them off if nobody objects.
21 MEMBER HINZE: Let me ask you a quick l I
22 question, Abe.
23 MR. VAN LUIK: Yes, sir.
24 MEMBER HINZE: What about the issue resolution 1
()
("%
25 reports that are being prepared in the process by the NRC.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
I 273 l l
l 1 What role do they play in this, if any? '
2 MR. VAN LUIK: They play a role. We are (g w)
/
3 expecting those I believe, at the end of this calendar 4 year. Is that correct? All of them are scheduled to come 1
5 out in time for us to address them in the TSPA-VA. For 6 each one of those reports, we will do our analysis and 7 then address the content of the resolution report.
8 If they come in much later than the end of 9 this calendar year, it's going to have to wait until the 10 gap between VA and licensing to address them individually.
11 But we're hoping in each case everyone that applies, 12 somehow the TSPA, that we will be able to address them in i 13 the upcoming TSPA-VA report.
i \
d 14 MEMBER HINZE: I guess that's been made loud 15 and clear to the NRC.
16 MR. VAN LUIK: I believe so. I believe that 17 we have commented on their schedule and applauded them for 18 getting them done this calendar year.
19 MEMBER HINZE: Well thank you. Once again, a 20 great job. We really appreciate it.
21 We'll move directly then to John and Tim 22 Sullivan, who will be talking about the agreements from 23 the technical exchange. I think perhaps we can lead right 24 in then to the future activities, if that's all right with
,~
(,) 25 the two of you. How are you handling this?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
274 1 MR. SULLIVAN: We don't have this 2 choreographed. This is one of the things we didn't sit
,f -)J
~%.)
3 down and --
4 MR. VAN LUIK: Can we take both ties and clip 5 one thing to both ties?
6 MR. SULLIVAN: Is it all right if I -- I'll 7 chime right in here.
8 MEMBER HINZE: Take Bruce's chair, if you 9 don't mind.
10 MR. TRAPP: Well, I could say that along with 11 everything else, this obviously is a first of a kind 12 because how many times have you seen the DOE and the NRC
,~
13 on the same viewgraph?
\ /
x' 14 What I really was kind of thinking about doing 15 and I'll stand out this way, is basically read the 16 agreement and maybe give a comment or two. Let Tim give a 17 comment or two, and then go onto the next agreement.
18 The first agreement basically is that NRC and 19 DOE agree that the rate of volcanism is relatively i
I 20 constant for the last 5 million years and basically we can l
21 use this kind of constant rate to carry on to the period ,
l 22 of performance. I 23 What this really addresses is a question of 24 waxing, waning volcanism, et cetera, this type of thing.
(~'h l
(,,/ 25 It has been several models which have been proposed that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
275 1 go one way or the other. What we're saying, and this is
,e-3 2 based in large part on the studies dealing with the LJ 3 Amargosa Valley isotopic province, looking at what happens 4 with probabilities and different events when you start 5 taking a look at this. It appears to be a constant rate 6 which we feel should be used in performance assessment.
7 MEMBER HINZE: John, if I may. That certainly 8 has been an assumption and was used in the PVHA for 10,000 ,
i 9 year time period. Question, what happens when we go to 10 100,000 years?
I 11 MR. TRAPP: I really don't see any difference !
12 myself. I would be glad to let Britt comment or anytning, I 1
~
13 but I really see no difference.
S2 14 MEMBER HINZE: Does anyone have any comment on l 15 that, our consultants, anyone? l 16 Has DOE looked at this, Tim?
17 MR. SULLIVAN: Well, you know, given the long 18 period of record and the long return period for volcanic 19 events, that seems a reasonable assumption to me.
20 Kevin, do you want to comment any further?
21 MR. COPPERSMITH: Yes. Kevin Coppersmith. We 22 asked the experts that question in the last workshop of 23 the future forward time period. They saw no difference 24 between 10,000 and 100,000 years. If we went out r~N
\
x_j 25 1,000,000 years, they begin to say hey, we may need to NEAL R. GROSS COURT REPORTERE AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
i 276 1 think about a time history that might be different from 73 2 the constant. But out 100,000, you are looking back 5 I i
\ /
i 3 million. They felt fairly comfortable in that forward 4 time window. There would be no significant change.
5 MEMBER HINZE: Thanks so much.
6 MR. TRAPP: Tim, you got anything more on this 7 one?
8 MR. SULLIVAN: No, this has not been an issue 9 of contention.
10 MR. TRAPP: Well, this next one was probably 11 the easiest issue to agree on. The reason it's brought up 12 is really there was this one piece of information, and we 13 weren't sure which was talking about a potential six
(.-
kl 14 million year old silicic volcanism. Through a whole bunch 15 of codes into some of the different models, assumptions, 16 what happens, the center went, took a look at this stuff, 17 got some material and age data that it's roughly about 9.5 18 million years.
19 MR. HILL: Yes, it's 9.3 -- or 9.1 plus or 20 minus .3.
21 MR. TRAPP: So basically, what it amounts to 22 is we don't have to worry about this all of a sudden 23 silicic episode sitting in the middle of these basaltic l
f 24 episodes, and it makes our projections, our understanding
/%
- (_) 25 of the system a little bit easier to handle.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
277 l 1 Tim?
,3 2 MR. SULLIVAN: I have no comment on this. )
\ l 3 MR. TRAPP: Number three, DOE agrees to l
l 4 consider evaluating new data such as the size and volume 5 of Little Cone, the number of events through Anomaly A !
l 6 through hazard sensitivity studies. l l
i 7 I guess the point I would bring out here is l l
l 8 the -- such as it's really a question whenever you saw I I
9 some of the analysis as new information comes into play, l l
10 we would expect to see additional analysis, etc. as it l l
11 goes through. And again, we will be taking a look at --
12 it's not a question of hey, we found one new rock, please l l
13 analyze, let's make sure that this is something that might f}
k/ 14 have some effect.
15 MR. SULLIVAN: As you saw in Kevin's 16 presentation, we've gone ahead and evaluated the impact of 17 this new information on the hazard estimate, and the 18 impact is insignificant. And this suggests to us that new 19 data of the type that's being collected by the center is 20 not going to be significant to hazard assessment.
21 That is, that data that refines the volume 22 estimates of the buried centers or that information which l
l 23 contributes to refinement of the event counts. And DOE 24 has done this reevaluation as an example, but we don't f%,
() 25 intend to be continually updating the results of this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
278 1 study,
, ~3 2 We will document what Kevin presented in the I )
'V letter that's mentioned here at the end of the agreement.
3 4 MR. TRAPP: The only other thing I would say 5 along this line is where we're really concerned more than 6 anything else is -- I guess Norm would describe it as i
l 7 paradigm shifts; are we really changing the conceptual 8 model that we're dealing with versus just tweaking l
9 numbers.
10 CHAIRMAN POMEROY: Tim, excuse me; but I 11 assume you mean though that you will continue to look at j l
12 new data as it comes forward, and that you'll do something l l
13 to look at its relative importance. I mean, I suppose I
'w/ 14 could see some situation where new information came l l
15 forward that was so startling that you would want to i 16 evaluate it.
17 MR. SULLIVAN: Yes, Kevin identified at least 18 those types of new information that the experts thought 19 might lead to them reevaluating, and they were rather 20 dramatic occurrences. We will evaluate developments in 21 the field of volcanism and additional information 22 collected in the Yucca Mountain area between now and the 23 license application and update the PVHA results as 24 required.
f)
( ,) 25 That's our plan.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
! 279 1 CHAIRMAN POMEROY: Thank you.
,rx 2 MR. TRAPP: This agreement was probably the
'c
$ )3 3 longest one to try to write and the longest one as far as 4 time to put together in simplest form.
5 We presented what we consider our -- and I'm l
6 not going to call them the range; I'm going to call them 7 our best 6 mate of values f rom 10-7 to 10-8 DOE is 8 basically going ahead and saying that they want to use the 9 PVHA results.
10 My basic concerns here or comments dealing i 11 with this are related to the fact that again, if you take 12 a look at all the stuff that's in the literature -- for 13 instance, the material that has been put out by Gene Smith l l i i \ 14 on this -- you are talking values which really are ranging 15 f rom 10-" all the way up to about 10-5 16 If you start putting this into a value and try 17 to calculate these whole things, weighed everything, what 18 you end up with is median values which are slightly i
19 greater than 1 x 10-8 up to about 5 x 10 8 You end up with 20 mean values that are just about the same.
l 21 I will caution you though, and I'm not sure 22 that this totally gets across, is that again, the numbers 23 that we're using and that the state are using are for the 24 volcanic disruption. The numbers that you're getting from i
rx \
(_) 25 PVRA are volcanic disruption and indirect effects from the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
280 1 dike.
g~\ 2 So while the numbers appear to be converging, L] 3 there's a wider disparity than appears just looking at the 4 first glance.
5 Tim?
6 MR. SULLIVAN: Well, as has been discussed 7 several times today, the information that was used that 8 formed the basis for the state's estimate and the 9 information that has formed the basis for the center's 10 estimate was available to the PVHA panel. And they 11 accorded it, you know, the weight they thought appropriate 12 in developing their source zone models.
13 And it's on that basis that we intend to
. p
! \/. 14 proceed then with the hazard estimates from the PVHA.
l 15 MEMBER HINZE: You're saying there's a l
16 difference between the NRC, the center, and the PVHA 17 because of the lack of consideration of diking sills in 18 the center's and the NRC's work?
19 MR. TRAPP: Yes, our number basically is to I
j 20 represent the volcanic vent itself, the possibility of a 1
21 vent forming and dispersing material. The numbers that i
! 22 you get out of PVHA are the vent or -- and this is where 23 the whole structure and everything else starts getting I 24 important.
f5
(_) 25 Because when you take a look at these various ,
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 j
281 1 structural models that are used in PVHA, they have the f- s 2 vent occurring within the zone or within the zone and the
~
3 boundary. And what actually comes out and hits the 4 repository is not the vent, but it's a dike.
5 So you're getting the combination of those two 6 probabilities combined.
7 VICE CHAIRMAN GARRICK: John, are you 8 suggesting that maybe if you both defined the probability 9 the same that there might not be this difference? It I 10 sounds like --
11 MR. TRAPP: I'm saying if we both define the 12 vents the same and took a look at the PVHA, there would be )
i 13 a slightly larger difference in numbers than you -- )
i
- i I
\)
- 14 VICE CHAIRMAN GARRICK: Well, it sounds like i 15 one's a joint probability and the other one is not. One's 16 a combinatir.n. l I 17 CHAIRMAN POMEROY: John, -- l l
l l 18 NEMBER HINZE: Go ahead, Tim.
19 MR. SULLIVAN: I would certainly agree that's 20 a contributor, John, but I don't think it's the major i
l 21 contributor to the difference between the full range of 22 these estimates. As Kevin said earlier, it has to do 23 mostly with the rate density that you assume.
, 24 And as long as you include Yucca Mountain in a
,m.
( ,) 25 high rate density area, then you're going to get a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4 433
1 282 l 1 slightly higher annual probability. j l
m s
2 Perhaps, Kevin, you need to review that again.
c s
\ /
3 MR. TRAPP: But if you take a look also, and 4 Kevin can go through it, Yucca Mountain whs not included 5 in the high rate density. It was included in the low rate 6 density in all cases.
7 MR. COPPERSMITH: Let me go back to the event 8 definition problem.
9 MR. TRAPP: That's where it stems from.
10 MR. COPPERSMITH: We basically have -- we're 11 defining an event or defining an intersection as a dike 12 intersecting the repository. So it's not indirect effects l
13 of, say, a nearby dike, one that's up gradient, down
()- 14 gradient, close enough to have a thermal impact or l l
15 degassing, there are corrosive effects and so on. !
16 It is physical intersection of a dike with the i 17 repository. Okay, so the indirect effects would be a l
18 different issue. Now, whether or not the tip end of a 19 dike intersecting the repository has sufficient energy to 20 lead to release is not considered here.
21 Now one way to look at that, we also have the 22 probability of the actual center point of that dike l
23 intersecting the repository, and that is about a factor of 24 three less. We also have a model that right now we allow 7-() 25 for them to define a probability distribution -- if this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
283!
1 is the ascending dike, here's the center point of that.
-~ 2 They define tie probability distribution that G'
3 it is offset from that center. If we use -- we assume 4 that it is actually the center point, we have that 5 probability of distribution. That's about a factor of one 6 and a half, maybe two less than the integrated 7 probability.
8 So we can use that information to look at the 9 effect of this. If, in fact, it needs to be at the center 10 point of the dike to have sufficient energy or release, 11 then basically you can either look at those centers, 12 factor of three, or look at the event centered probability 13 distribution. And those -- all three of those PDF's are
,c3
-- 14 given in the report. l 15 I think -- going back to what Tim said, I l 16 think that is not -- it does contribute to a difference.
17 But I think it really is the issue of rate density that 18 was assumed in the Yucca Mountain area versus the Crater 19 Flat area. The rate densities are lower gene rally within 20 the Yucca Mountain area because the observed centers are 21 zero in the post five million year time frame.
22 Therefore, rate densities are assumed that 23 come from like the background zone. Just like in 24 seismicity, the Amargosa Valley isotopic provence, O
( ,) 25 whatever the background zone was, that rate density, for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBER 3 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
284 1 many of the models, is the driving rate density for i
i r- 2 intersections in the repository.
k.__S) 3 MR. TRAPP: All right then, the point that I 4 was trying to make is the rate density that was assigned 5 to Yucca Mountain basically because of the way the lines 6 were drawn was low rate density.
7 MEMBER HINZE: A couple of questions.
8 Is there any chance of deconvoluting your 9 analysis to limit it to cones?
10 MR. TRAPP: That basically is about 6 x 10-' is 11 what their number came out to be.
12 MR. COPPERSMITH: Let me clarify.
, ,_. 13 For some of the experts, cones -- the
-' 14 alignment of cones would be combined into a single event 15 definition -- if they're on a proper alignment and they're 16 close spatial proximity, same age, etc. So these are the 17 event -- center for the event, not necessarily for cones.
18 MEMBER HINZE: But they had to have a cone to 19 have an event.
20 MR. COPPERSMITH: Except for the case of 21 hidden events. Okay, there's no location.
22 MEMBER HINZE: So they could have hidden l
23 events?
24 MR. COPPERSMITH: Yes, all of them had hidden
(_,/ 25 events ranging from about ten to fifty percent more than NEAL R. GROSS COURT REPORTERS AND TRANSCR:0ERS
[ '323 RHODE ISLAND AVE., N.W.
(202) 234 4433 "/ASHINGTON, D C- 20005-3701 (202) 234 4433 i
285 )
l 1 observed.
l 7~.s 2 MEMBER HINZE: Let me ask the question, how
(' ~ ' '}
3 much would the -- how have you considered the effect of I l
4 this upon these numbers? You know, you are connerned 5 about this event definition. Have you made some 6 quantitative ana.".ysis of how much it impacts the results?
7 MR. TRAPP: Basically, if you take a look at 8 the fact that the probability is derived totally separate 9 from the consequence -- therefore, if you're sitting with l
10 a -- anyway, they are not linked like you do a seismic 11 hazard analysis. You've got a probability for a magnitude 12 seven versus a probability for a magnitude eight versus 13 probability -- and you draw these lines.
r^x
- 1
\> 14 This is a probability which is kind of an 15 event definition, and it's really disruptive. It is 16 simply a multiplication factor. So if I say that the 17 probability is 10-7 versus 10-8, I'm dealing with an order 18 of magnitude difference in total consequence.
i 19 MR. COPPERSMITH: Again, for those who are 1
20 familiar with the PRA, reactor PRA, this is exactly '
j l 21 analogous. We're dealing with -- in this case, this l l t
! 22 probability of frequency of intersection is comparable to 23 the seismic hazard curve that expresses the frequency or 1
24 probability of exceeding various levels of ground motion 1
/ N
() 25 that has uncertainty associated with it.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
286 1 The consequence analysis is exactly analogous 7s 2 to the fragility analysis, fragility curves. They give
(
)
3 the probability of failure as a function of ground motion. I 4 So you're given the ground motion; this is the probability 5 of failure of a particular system or component. .
6 It's likewise here. Given an intersection, i 1
7 what does it do in terms of consequence? What's the i 8 probability of release? So when we talk about consequence ]
9 analysis, we're talking about conditional probabilities.
10 And I guess the point that John Garrick was making that it 11 -- when we talk about consequence space, it's just like 12 fragility space.
13 If we don't put a probability in front of it,
/~~N
-) 14 it becomes given that that occurs as a probability of one.
15 The most dramatic example in reactor PRA's that I remember 16 is meteorite impact. If you think of it in terms of 17 simply consequence, it's a very dramatic impact. You need 18 to multiply that times the probability of actual meteorite 19 impact at your particular location.
20 MR. TRAPP: And I would say that basically 21 the way that we've been handling the fragility curve in 22 this would be kind of the thing that is sampled such as 23 the wind speed, the waste package life, the grain size, 24 this type of thing. So there is a probability that way, j
()
I f8 25 but it is not -- once you set a point curve, etc., like I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C I M *,/01 (202) 2344433
l l 287 1 said, it's a straight multiplication.
, 2 CHAIRMAN POMEROY: John, just to clarify for
/
' l I '
3 me, the center model, your model, does not consider the 4 intersection of any dikes associated with this event?
5 MR. TRAPP: It's basically a number which is 6 the probability of volcanic disruptien, to be another way 7 of saying it -- better way of saying it.
8 MR. HILL: Paul, this is Britt Hill.
9 There is an area term associated with that.
10 That's what Chuck was talking about earlier to account for 11 the fact that you can have not just a central conduit, but 12 a distributed number of events. That's that 500 meters to l
13 two kilometer thing that he is sampling.
1
(_) 14 That would be to accommodate the area -- the 15 subsurface area term that is directly impacted by the 16 volcano. And I think there's been a little bit of 17 ambiguity about direct and indirect effects that's used 18 very differently. By direct, I think the way that we're 19 usually referring to that is that material is released 20 directly to the surface from a volcanic eruption by the 21 event.
22 Whereas, indirect, you still have direct 23 penetration of the dike into the repository, but no 24 material is being transported to the surface by the p-),
(, 25 igneous event. It's a subsequent transport through NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
288 l
l 1 normal, hydrologic flow and transport that is giving you 2 the release.
f 3,
nf 3 So just a small point of clarification.
4 CHAIRMAN POMEROY: Thank you, Britt; that's 5 very helpful.
6 MEMBER HINZE: Are there any plans for the 7 center to include longer dikes beyond the two kilometer --
l 8 MR. HILL: We've done some work on that where 9 we're looking again at classifying what is an event that 10 you can look at in Crater Flat, the four plus volcanos 11 there at one million years can constitute a single event.
12 And the end result is, you get fewer number of events, but 13 the events impact a greater area.
f It
\' ) 14 And so there really isn't much of an effect.
15 It sort of plays off one against the other. The same l 16 thing with increasing the number of dikes associated with 17 it. It's fairly speculative. It doesn't change it too 18 much until you start getting up into a very large number 19 of events or -- excuse me, dikes associated with a single 20 volcanic event.
21 But we're still faced with that problem of how 22 do you get material from a dike two kilometers away to l
l 23 flow laterally and come up into the accessible 24 environment. And there's just not a good mechanistic (m.,
(_,1 25 basis to make that assumption. It's really the area NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 l
289 1 immediately beneath the volcano at the surface.
2 By immediately, I'd say tens of meters as a
(~w V
3 good first guess that's going to be directly transporting 4 material to the surface. You're not having kilometers of 5 lateral flow from a dike where the volcano is two clicks 6 away from your repository boundary. It's just not 7 realistic to think you're going to get lateral flow in a 8 dike for Cwo kilometers and bring material up to the 9 accessible environment during that event.
10 MEMBER HINZE: When there are these -- the 11 fire forms that come associated with a dike opening up, 12 for example, in Hawaii, which you're the expert on, how 13 much lateral flow is there?
- i
~'
14 MR. RYAN: My ears perked up, Britt, at your 15 last comment. I think I would strongly disagree with it.
16 Let me simply say that in basaltic riff zone 17 environments worldwide, the pattern we see is migration 18 from the upper mantle starting out at depths that vary 19 between 100 and 60 kilometers winding up at shallow depths 20 that could be anywhere from ten to two kilometers depth.
21 And in that ten to two kilometer depth range, 22 a long period of stagnation and storage. Now, magna will l
l 23 then move out of that ten to two kilometer depth range 24 when it's over pressured by subsequent magma batch p
(_) 25 additions to the bottom of the system.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
i (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
290 1 And when it moves out, the dominant transport l
!py 2 mode is lateral, not vertical. This is simply a least 3 work argument that if the crust is sufficiently relaxed by l
i l 4 extension, the system can reduce its potential energy in a 5 very efficient way and do very little work in the process l
l 6 by a lateral migration.
7 We see this in spades documented seismically l
8 in Hawaii. We see it in spades in Iceland. And we think l
, 1 l 9 that there's very good geomorphic evidence for the same l
10 kind of thing going on in the earth's mid ocean ridge 1
1 11 system some 60,000 kilometers of riff zone in aggregate.
12 The eruption and cone building is really the l l
l 13 exception in one perspective, in one sense, to the rule. l
/s< !
1 t
N' 14 Some of the older Hawaiian volcano observatory staff l 15 members refer to dike intrusions as "uneruptions." So .
l 16 these uneruptions are dominantly a lateral, blade forming, l 17 intrusion mechanism.
18 MR. HILL: All right, but once you've 19 established a conduit at the surface, and this is what we 20 were talking about at lunch -- if you have established a 21 cinder cone at the surface two kilomaters away from the l 22 repository, even if you've got a dike in the subsurface l
23 extending into the repository zone, do you think it's 24 realistic that you would have two kilometers of lateral
()
/,
25 flow from 300 -- or at 300 meters depth to bring waste l
NEAL R. GROSS l COL"aT REPOR1ERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433 l
291 l
1 from that repository two kilometers away up the conduit?
,y, 2 And that's really the process that we're 4 3
~'
3 worried about, not the emplacement mechanism itself.
4 Certainly there's abundant evidence of lateral flow during 5 ascent. But once the vertical flow is established to the 6 surficial conduit, do we really think we continue at 300 7 meters depth in the subsurface to transport material for 8 kilometers away up the conduit?
9 MR. RYAN: I guess the answer is yes and no.
10 In the very short time frame, yes; I think
'11 many of the things you're suggesting would happen. Once 12 you develop a lubricated, high temperature conduit to the 13 surface, that would, for some several days to a few weeks, t'3
)
'- 14 represent the least work --
15 MR. HILL: Yes.
16 MR. RYAN: -- pathway to accommodate any 17 subsequent increase in magma in the system.
18 However, one thing that we see in spades is i
19 that these are pulsed activities that play themselves out l l
20 over a few tens of years in the migration of a large magma j 21 batch to the surface and its eventual intrusion l 22 interruption.
l l 23 So you will get, yes, perhaps the kinds of 24 things you're suggesting in a single pulse. But then, in
,a
( ,) 25 a few tens of hours or in a few weeks when that dike drops NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
j (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
292 1 below its solidus, when it becomes mechanically strong
,~ 2 with high, elastic moduli and high sheer strength relative 3 to its porous country rock, subsequent events will then 4 run parallel dikes past that.
5 So we call these sheeted dike complexes and 6 ophiolite suites and so forth.
7 You can get both kindc of things going on, 8 both kinds of scenarios are going on. The scenario you're 9 talking about is a subset really of a larger decadeal 10 pulse of activity that plays itself out over several 11 eruptions and several intrusions.
12 MR. HILL: Right, the event can have multiple 13 phases to it, of course; and that's what we're getting at.
A I
'x_/ ) 14 MR. RYAN: Sure, absolutely.
15 As an example, at Krafla, a volcano in 16 northeast Iceland, it started intruding and erupting in 17 December of '75. It. stopped, you know, September of '84.
18 There were 12 intrusions and nine eruptions. The 19 aggregate eruption zone was something like 13 kilometers 20 in length.
21 Similarly, in Hawaii, an eruption that's now 22 ongoing, started in January of '83 -- it's had well over l
l 23 50 pulses of activity, each one concluding with a die back 24 of fountaining activity and then cranking back up again, In) v 25 MR. HILL: Right. And conversely, at Cerro
! NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS l
1323 RHODE ISLAND AVE., N.W.
i (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
293 i
1 Negro in Nicaragua for the past 150 years there have been ex 2 I think 23 significant eruptions, each of which have
\
\~) l' 3 occurred within half a kilometer of the central conduit 4 and almost exclusively within the central conduit and no 5 evidence of any real significant bocca development beyond 6 the base of the cone. l 7 So there's plenty of examples.
8 MEMBER HINZE: This is a very good discussion.
9 MR. TRAPP: I wanted to ask one question 10 myself, and it's basically the bottom line that I would 11 suggest that's coming out of your statements would be the 12 suggestion that a larger amount of waste could become 13 entrained and dispersed than we've been assuming.
'- 14 MR. RYAN: It's hard to respond to that in a 15 succinct yes or no. Because these are pulsed events and 16 because each pulse can be several million cubic meters in 17 volume, and because all of the pulses would add up to a 18 ten or 20 year intrusion cruption scenario, you could get 19 all kinds of possibilities in terms of how magma would 20 interdict a repository depth.
1 21 For example, the first pulse would invade the l 22 free volume space of the repository. It may very well 23 fill it up floor to ceiling. That would then begin to 24 cool. And some period of a few tens of years, depending l
/~'N l lt,) 25 on how high the ceiling was from the floor, that would be l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
294 l
l 1 largely subsolidus. It would be intact.
i 2 But nevertheless, a decade later, another
,e 3 lO 3 pulse could come through and rip open this prior 4 consolidation episode. So there are a lot of different I
L 5 scenarios that one could envision with fairly garden 1
6 variety igneous process.
7 MR. TRAPP: I guess one way to describe it in 8 PVHA would be we're talking about a larger range of 9 uncertainty.
10 MR. SULLIVAN: Before you move on, please note 11 here that the last sentence of this agreement is a 12 commitment by DOE to explain how the PDF will be used.
13 And Abe outlined that for you this afternoon, and that p_
.\- } 14 will be included in the upcoming letter as well.
15 MEMBER HINZE: That was my question.
16 The last phrase of that is recognizing NRC's 17 comments, and presumably that will be taken into account 18 then?
19 MR. SULLIVAN: Yes.
20 MEMBER HORNBERGER: Could I just -- inst to 21 make sure I'm totally clear on this, am I to infer that 22 the NRC staff does not accept the PVHA?
l 23 MR. TRAPP: We've got concerns with the PVHA 24 would be a better way to describe it.
e n
(_)s 25 MEMBER HORNBERGER: Concernn? Concerns about NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
l l (202) 234-4433 WASHINGTON. O C. 20005 370t (202) 234-4433 l
295 1 the process, about the expert elicitation itself, about
,- 2 the --
~~
3 MR. TRAPP: The only ones I am --
4 MEMBER HORNBERGER: -- qualifications of the 5 experts?
6 MR. TRAPP: The only ones I am willing to talk 7 about -- because if I talked about any more than that, my 8 management would be on top of me -- are the two I've 9 already raised.
10 CHAIRMAN POMEROY: Your management takes the 11 chance, of course, that we will make an interpretation of 12 what that means.
13 MR. TRAPP: Well, this is why I said at the g)
'\ - 14 very beginning there will be a meeting between DOE and NRC 15 where they're talking about the whole expert elicitation 16 process, and this is basically the end of this month.
17 Wes? Roughly the end of the month.
18 Also, next month, you guys will be discussing 19 this. I am not going to get up and start going into that.
20 MEMBER HINZE: That was a prior agreement with 21 John, and understandably.
l l
22 VICE CHAIRMAN GARRICK: This is not an atomic 23 safety licensing board.
24 MR. TRAPP: Now if you want to get into that, A
() 25 I'll run upstairs, get Norm and a whole bunch of others, NEAL R. GRGSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE iSi.AND AVE., N W (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
296 1 and they could start talking about it; but I'm not going 2 to do it.
7-
~
3 MR. HILL: The agreement from the technical 4 exchange, I think, is the fairest way of putting it.
5 We're both -- if we look at the 10-7 as an upper bound, 6 we're pretty much in agreement at the current state.
7 There are differing views, however, on the lower bound.
8 I think that's the simplest way of putting it.
9 MR. TRAPP: And I'd point out again that our 10 upper bound number basically is approximately the same as 11 the best estimate used by the state.
12 MR. SULLIVAN: Let me respond to that.
13 DOE did not sponsor the PVHA expert
'13t
\> 14 elicitation in order to determine an upper bound. The 15 purpose of that elicitation was to define the mean value 16 and the uncertainty. And our intent is to use that full 17 distribution. We will be doing some sensitivity studies 18 as Abe described.
19 MR. TRAPP: Are we ready to move on?
20 MEMBER HINZE: Let's move on.
21 MR. TRAPP: Abe Van Luik was talking about the 22 agreements. This would basically be the agreement that he 23 was talking about, number five. It's slightly different 24 than he had stated.
()
g 25 We basically agree that volcanism is of NEAl. R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
297 1 regulatory interest, and its probability and consequence fs 2 will be considered. If determined to be significant to i )
3 repository performance, and now we've got to know exactly 4 what the standard is that we're dealing with, the effects 5 of volcanism will be included in the total system 6 performance assessment.
7 Right now, the only comment I'd make is it 8 appears to be one that we're going to have to carry 9 through.
10 MR. SULLIVAN: Well, my recollection is a 11 little faulty.
12 Abe, I don't see any commitment here on the 13 part of DOE to carry these calculations forward solely on
(~\
k-) 14 the basis of the results of consequence analysis and 15 isolation, do you?
16 MR. VAN LUIK: This is Abe.
17 I believe you're correct. It could be since 18 (a) mentions probability and consequences, and (b) if 19 determined to be significant, what's the subject of if, it 20 should be probability and consequences. A legalistic 21 rendering, which is what we gave ourselves, would be that i 22 we agree to both.
23 I would be happy to substitute risk because 24 probability and consequence is risk. So let us change it n
() 25 to risk and correct that in the letter that we're going to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) IM-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
298 1 send to the NRC.
, ~g 2 MR. TRAPP: Basically, as it says, the effects b 3 of volcanism, so you're talking about what you're 4 modifying. So if the effects of volcanism are determined 5 to be significant with respect to repository performance, i 6 you'll include it. l 7 MR. VAN LUIK: Exactly.
8 MR. TRAPP: So like I said, which standard --
9 what's the exact standard statement? And the only comment 10 I am making past that is the fact that it appears that l
l 11 it's going to have to be considered. And I think that 12 would help you out a little bit.
13 VICE CHAIRMAN GARRICK: Yes. l
/'^h I ,
' \ ') 14 MR. SULLIVAN: I think John agreed, Abe.
i 15 MEMBER HINZE: Volcanism pulled him out.
16 MR. TRAPP: The next one, the treatment of l
l 17 consequences outlined by DOE including extrusive magmatic 18 events (cone and dike formation) and intrusive magmatic 19 events (sill and dike formation) with both direct and l
l 20 indirect effects is generally appropriate at the level i
21 that we were discussing.
22 Two comments. If you take a look at what has 23 been done in TSPA-91, 93, etc., this type of thing, one af 24 the places we had a tremendous amount of problems as far
('N
( ,) 25 as the consequence analysis was the incorporation ratio.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
299 1 In other words, the volume of the waste that could get rS 2 out.
YY And we basically feel that these numbers were 3
4 several orders of magnitude too low.
5 In the discussion that was conducted today, 6 there was basically the statement that how they planned on 7 handling some of this extrusive material was to obtain the 8 center's model and run that one.
9 I personally get a little bit leery about that 10 comment just on face value because what we end up with is 11 the NRC licensing itself and the DOE working as a 12 regulator, and I think that one needs to be looked at in a 13 whole bunch of different perspectives.
p)
\~# 14 MR. VAN LUIK: If I may address that one.
15 This is Abe.
16 What we would like is to critically review l 17 your model, put in our own inputs so that all we have is 18 basically your computer framework, which is like borrowing 19 your Wordperfect to put in our Word, and I don't think i
1 20 there's any collusion at all that can be implied from 21 that.
l I 22 However, if -- you know, if, after you see 23 what we would like to do with it, you still feel that way, 24 we'll create our own model. It's not a problem.
,p
(_ 25 MR. TRAPP: Mike Bell.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE,, N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
i 300 1 MR. BELL: This is Michael Bell, part of that
,- 2 elusive NRC management.
~
3 I'd just point out that I'm sure people are 4 aware that NRC has any number of reg. guides that direct 5 applicants to use certain models, certain assumptions, and 6 this set of parameters; and we would find that an 7 acceptable way to meet an NRC requirement.
8 So I don't really see, you know, any fatal 9 flaw in DOE adopting the center's model. Of course, you 10 know, we'll look very closely at how they use it and make 11 sure they use it the way we intend it to be used.
12 I guess while I'm up, just let me go back to 13 the one point that came up five minutes ago about what --
/ \
~
14 is NRC rejecting the PVHA. I'm not sure if that's exactly 15 the way the question was posed. But I think John 16 addressed that very -- in his very first presentation this 17 morning on his four slide issues related to PVHA.
18 And basically, it was DOE is free to use PVHA 19 or whatever information they desire to make their 20 licensing case. And you know, we'll consider that and l 21 other information, I'd say, the way the NRC staff always 22 does when it reviews an applicant's submittal.
l 23 MEMBER HINZE: Thanks, Mike.
24 Is there anything more on this?
r~
(N) 25 MR. TRAPP: I don't have anything to add to l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l
l 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
301 1 that.
7- 2 The nev' one, DOE and NRC agree there's
') 3 uncertainty and consequence analysis for the magmatic 4 waste package / waste form interactions and will be 5 evaluated.
6 If you take a look at some of the stuff that 7 Britt's already shown, one of the things that does come 8 out is the effect of the assumption of exactly what size 9 we're talking about of the waste particles, exactly how 10 it's incorporated. Does the waste package last? It 11 really drives a lot of the results.
12 If we sit there and assume that everything 13 stays in the pellet form, let's face it, it's going to be p.
J 14 lucky if it gets a kilometer away. You break it down to 15 the average size of ten microns, etc. You start getting 16 these values on out to these critical groups that we're 17 talking about.
18 Based on some preliminary runs and some of the 19 things they've done, I suspect if we talk about the one to 20 two micron range which a lot of our waste package people 21 feel is the main number or best number to use, you'll end 22 up with higher doses. We need to take a look at it, and l 23 it needs to be evaluated to see really what its effect is.
24 MR. SULLIVAN: Without reopening the g~.s
(_) 25 discussion of whether the consequence analysis presented NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 1
302 1 by the center was a bounding value or reasonably
,_ s 2 conservative or something else, it's not clear to us that
? )
3 uncertainties in those parameters have been considered in 4 the analysis itself.
5 And that's what DOE intends to do as part of 6 the consequence analysis that Abe provided.
7 MR. TRAPP: Any other comment on that one?
8 Basically number eight, they're to provide a 9 letter which -- or DOE is to provide a letter describing 10 how they're going to handle the subissue resolution, as 11 specified in three or four above, for consideration. In 12 simplest form, last Friday, we did have a teleconference 13 with DOE in which we went over some of the preliminary o
x' 14 thoughts they had on this.
l 15 There is supposed to be a discussion tomorrow j 16 morning where we'll have another series of things going 17 through this. And all I can say is we're working forward l 1
l 18 on this issue.
19 MEMBER HINZE: Questions about that?
20 Let me ask a question, if I may, regarding --
21 MR. TRAPP: If I said you couldn't, what would 22 you say?
23 (Laughter.)
l l
24 MEMBER HINZE: I would say you were being l
' (^T i.
s ,) 25 normal. Glad to hear it.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
303 1 The NRC had a number of comments in the site 2 characterization analysis of the SEP. These were answered (v ~)
3 by the DOE as part of a letter from Steve Brocum which 4 accompanied the PVHA, as I recall. l 5 Where do the comments -- where do these 6 comments and concerns come into play at this time? Are !
l l
7 they being -- by the board, are they not being considered 8 any longer, or are they -- where are we? ,
l 9 MR. TRAPP: It's a very hard question to 10 answer for a couple of reasons.
11 If you take a look at the comments that were 12 raised in the SEP and the comments which have been raised 13 on the various study plants, there are a whole series of ia
, ! )
\' ' 14 them which are based on an assumption that certain things 15 would be done by DOE.
16 There's a whole series that basically we're 17 not sure exactly how to handle because the program has 18 changed. There's concerns that are raised with things i
1 19 like -- well, paraphrasing -- DOE, we really think you 20 ought to be going out and doing some more geophysics 21 specifically looking at these items.
l 22 We don't feel that that's been done. It's 23 really one of the reasons that we asked the center to take 24 a look and do some of their geophysics. It's obvious
,,-~s
( ,/ 25 they're not going to be doing any more work. I'm not sure NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE , N W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 234-4433
304 1 myself exactly how to handle them. It's a question that
,S 2 has been raised with our board, at least to a certain j
, s i N/ '
3 extent.
4 I understand there's going to be some type of ;
5 discussion going back. DOE has got also a statement or --
6 I'm not sure how to phrase this correctly, but they are 7 going to be talking about which study plans are -- let's 8 call them OBE -- so how to handle these things, I don't 9 know; except that we are supposed to address them in the 10 issue resolution report.
11 And until I know how to handle them, they'll 12 be covered there, but I'm waiting for direction.
13 MEMBER HINZE: Well, I was at the technical
' -) 14 exchange. This obviously was not discussed at the 15 technical exchange.
16 Is there any ongoing discussion with the DOE 17 on any of these points --
18 MR. TRAPP: Most of these --
19 MEMBER HINZE: -- worthy of that at this 20 point?
21 MR. TRAPP: There really isn't.
22 MEMBER HINZE: Is there -- the question is 21 whether there'c a reluctance to close them when --
24 MR. TRAPP: I would love to close them, r~T
(,,) 25 Mike's standing up. Maybe he's got --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
)
i
305 1 MEMBER HINZE: I was waiting for him to come
- 2 say something.
l 3 MR. BELL: Dr. Hinze, what you're really 4 asking about is a generic question. It's not specific to 5 volcanism.
6 But basically, there are a large number of 7 open issues from the time of the review of DOE's site 8 characterization plan that, as John said, anticipated 9 certain work was going to be done, certain programs were 10 going to be carried out that now, because of the cut in 11 DOE's program, much of this -- some of this work, at 12 least, won't be done.
13 The way we see addressing this is -- I guess
- 14 in our refocus program, as we go through developing issue 15 resolution reports and looking at the significance of 16 missing information, if it turns out that some critical 17 piece of information is needed to close some open issue to 18 resolve one of our KTI's, then we would go back to DOE 19 essentially and say, you know, this particular open issue 20 is going to remain open.
21 There will be some of these open issues that 22 presumably you'll now conclude is no longer significant to 23 performance and you know, that work didn't really need to 24 be done. It was one of those things that when DOE had a 77
( ) 25 very large program with lots of national laboratories and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
306 1 a large budget, you know, they planned to do the -- gain s 2 scientific understanding, but maybe it, when put in the
! \
\
'~'
/
3 context of total system performance, wouldn't have been 4 needed.
5 MEMBER HINZE: So it would be fair to say that 6 they are going to be selectively considered, but you're 7 not going to make an issue of keeping them open?
8- MR. BELL: Not if they're not important to 9 performance.
10 MR. HILL: I think a quick example would be a 11 lot of the number of comments that were related to the 12 Lathrop Wells, and I believe many people are familiar with 13 the technical disagreements about the origin of that p 6 V 14 volcano.
15 Several years ago, that seemed to be very 16 important towards probability, and now it seems to be a j l
17 background issue. So we're -- there's no reason to keep l l
l 18 on with open items regarding how many events were at 19 Lathrop wells. It's really not an issue that's 20 significant right now.
21 MR. TRAPP: And I would suspect there would be 22 a large number which would be handled to the effect that 23 because this thing hasn't been addressed, there is a 24 fairly large range of uncertainty. If you can live with (n) 25 it, that's fine.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
307 1 MEMBER HINZE: It just seems to me to be a --
, 2 the proper thing to do to identify those issues which are
> )
\/
3 no longer important, that are no longer of consideration.
4 And I don't know if the term is close them, but at least 5 tell DOE that they're not -- no longer being considered as 6 comments.
7 MR. TRAPP: Well, like I said, the spot -- the N
8 exact spot that that will be covered is in the mission li 9 ' resolution status report.
10 MEMBER HINZE: Okay.
11 Okay, any other comments about the technical 12 ;xchange and the agreements thereof?
13 VICE CHAIRMAN GARRICK: I just want a point of
(~\
Nl 14 clarification that's kind of a nit, but something that 15 John said earlier made me dwell on it a while.
16 John, if you have an applicant that you 17 required a point estimate calculation from, for example, a 18 conservative one or whatever, you wouldn't object if that i
19 applicant supplied you with a full distribution from which 20 you could pull any number you wantedi i 21 MR. TRAPP: No.
22 VICE CHAIRMAN GARRICK: No? Okay.
23 I just --
24 MR. TRAPP: No, basically --
(Q,j 25 VICE CHAIRMAN GARRICK: Especially in --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
308 1 MR. TRAPP: My personal point on this whole
,-s. 2 thing is I think they may be doing more work than they RJ i 3 really need to.
But especially (
4 VICE CHAIRMAN GARRICK: Yes. l 5 at a time when your Chairlady is pleading for movement 6 towards risk-informed performance-based analyses. And it 7 seems to me that's a very good example of doing just that.
8 MR. TRAPP: They can -- we have no objections j 9 to them doing it.
10 VICE CHAIRMAN GARRICK: Yes.
i 11 MR. SULLIVAN: Where are we doing too much 1
)
12 work?
13 (Laughter.) l l' ) f
\J 14 MR. TRAPP: Basically the spot that we're l 15 talking about.
16 MEMBER HINZE: Excuse me, could we turn off 17 the recorder?
18 (Laughter.)
19 MR. TRAPP: Our basic comment there is from 20 what we see and taking a look at volcanism the way it's 21 handled, we think it can be handled by a point estimator 22 rather than taking a look at the whole range of l
23 uncertainty and all these other kind of things.
24 Now, how -- if you want to go through the
(_, 25 whole thing, nobody's going to object a bit.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
309 1 CHAIRMAN POMEROY: John, that's a nice lead in f
-' 2 to my question because I -- at the technical exchange, it Q/
3 was my understanding that there was a statement something 4 like if DOE were willing to utilize the point estimate 10" 5 something, what was that something -- what was the rest of 6 that sentence?
7 MR. TRAPP: No, it's basically --
8 CHAIRMAN POMEROY: I'm just asking what the 9 minutes or, you know, the nonexistent minutes would say.
10 MR. TRAPP: It's basically that the 10", we l 11 feel, is a -- this is a case where we do have what we l l
12 consider a good bounding reasonable estimate. Yes, it's -
13 -
)
l l
\~# 14 CHAIRMAN POMEROY: This is important.
15 MR. TRAPP: It's a number which we feel 16 comfortable with. It's one -- and this is really kind of l
17 important. You talk about new information and the impact 18 of new information and all this other kind of stuff, and 19 you have one more event, two more events; you have a 20 little bit of difference in structure, and it makes this 21 number maybe go up a little bit.
22 If you take a look at some of the things that 23 we've done back of the envelope both on the computer, you 24 know, where you're doing mass of stuff and just sit and
/ i
(_) 25 scratching your head, we feel comfortable with this range NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
310 1 going from 10-8, 10".
fS 2 And we feel that if you use the 10" as a O 3 number, as a point estimator number, then we would have 4 something that would be able to be carried through with a 5 tremendous amount of confidence through the licensing.
6 We can see no -- assuming that we don't have a 7 volcano between now and licensing coming through the 8 repository, -- (laughter) -- we can see nothing that --
9 right now that would cause us to think that we'd end up 10 with a value higher than that for any type of meaning.
11 So we're really, like I said, talking about 12 the work that we're doing, etc., to support this issue 13 resolution report, we're taking a look at some of these 14 buried geophyoical anomalies and trying to find out their 15 significance. We're going ahead and starting to do some 16 work on the probabilities of the indirect secondary I
17 effects and processes. l l
18 We're looking in detail on controls and l 19 conduit locations. We're checking these things out 20 against and evaluating them against other volcanic fields.
21 For instance, one of the good ones would be the work that 22 was done at the San Francisco field where the models 23 basically were compared to what's there.
24 MEMBER HINZE: John, if I could suggest, it q
k,) 25 would seem to me that first item there, the significance NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
311 1 of buried geophysical anomalies, at least from my f- 2 viewpoint, I think it's clear from what we've heard today I
\'~')
3 that another event down at Amargosa Valley may not be of 4 any significance to the results.
5 But those that are further up in Crater Flat 6 and towards Yucca Mountain would be the ones that I would 7 give first priority to.
8 Does that sound reasonable?
9 MR. TRAPP: Yes and no.
10 There is a fear if we --
11 MR. HILL: Again, not knowing the age of these 12 potentially buried -- or the anomalies that are 13 potentially buried volcanos, we have to evaluate their
/~~'s 1 l 14 location in addition to the age and how that would impact i
15 the present probability models. ;
16 When we're saying investigate significance, ,
J 17 it's not a large, complicated exercise to do this, but a l l
18 reasonable selection criteria and taking a look, like l 19 you're saying, at the closer in ones being more important.
20 But I'd still like to reiterate a point we 21 made earlier about regular spatial shifts through time as 22 another potentially significant development that isn't 23 being adequately compensated for by current models. And 1
24 whether or not we think this is something that needs to be CN worked on more than just in passing, a couple of
- i. ) 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE. N W.
(202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
312 l 1 paragraphs in the annual report, depends again on how
,S 2 significant we think it is.
b 3 So it's very difficult to say whether these l l
4 potential anomalies that we haven't really considered I 5 beyond brief passing are significant or not is very i
l 6 premature right now. That's why this is a planned j l
l 7 activity. '
I 8 MEMBER HINZE: And this will be done and i l
9 analyzed and incorporated into the IRSR?
l l
10 MR. TRAPP: To the extent possible, it will be '
l 11 in the IRSR. It may be in the annual report. It may be l l
12 in the next year's annual report. It basically depends on i 1
1 13 a whole series of --
,/ 3 F
14 MR. HILL: It depends how significant it turns 15 out.
16 MR. FOLAND: Could I ask one point of 17 clarification, John?
18 I don't understand this -- that comment that 19 it may be significant to the comment that you said earlier 20 that you were willing to accept a 10-7 figure which seems 21 to be the bounding limit now in the present data base from 22 the PVHA.
23 MR. TRAPP: It's not the bounding limit on the 24 PVHA number. It's a limit which has been stated here.
y/ 25 But if you actually take a look at the number, the bound NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 2'd-4433
313 1 is not 10"; it's something less than that.
gy 2 MR. FOLAND: It's the determination most of O 3 the models go 10" to 10-", is that not true?
4 MR. TRAPP: It's a number that they have 5 thrown out saying that they do. But if you look at PVHA, 6 they don't. The numbers do not -- the models do not range I
7 to 10-7 that you'll see in the PVHA report. 4 I
8 MR. FOLAND: Okay. But having said that, what 9 you're saying is that you're going to collect more data l 10 that may impact the 10" figure. What you said was you j 11 were content with --
l 12 MR. TRAPP: In some ways, it's basically i
13 supporting the 10-'. There's also a question in some of I, i S- 14 these of basic assumptions going into the model. And f
15 there's a couple locations and a couple scratches of their 16 heads you can take a look at which would cause you to 17 totally change your whole model and the basis for putting 18 these together.
I 19 Britt has been talking about one, this whole l 1
20 possibility of instead of talking about a kind of uniform !
l 21 area where we are talking about is a shift in fields 22 through time. This could have an effect if you're talking 23 about a shift that's coming up, say, for the Amargosa l l
24 Desert area working its way on up north.
( ,1 25 The amount that we'll do on here, very NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
314 1 honestly, is going to be extremely limited because we hav> l 2 got very little time and very little money. We're going
(~}
%-)
3 to take a look at it.
i 4 MR. FOLAND: I guess I wouldn't argue with I l
5 that. The point I find incongruous is your acceptance of 1
l 6 10".
7 MR. TRAPP: Because taking a look at these 1
8 various models, we -- aside from seeing something 9 extremely major, we don't see, in going through these back l 10 of the envelope calculations, where these numbers will 11 change if you keep the basic models that we've got.
12 And all that you are doing to the basic models
_ 13 is adding events, for instance. No, it's not going to t
- 14 change it. It's going to stay there. If you tace a look 15 at these models and you find some information that says 16 the model itself that I've used in calculating 17 probabilities is totally wrong, then we may have a 18 problem.
19 I'm assuming right now on the 10", 10" that 20 what I think I know about the area, and it's damn little, 21 that they won't change. We think that if you use these 22 numbers, from everything that we understand or think we 23 understand, that it will be a robust number.
24 I could be sucking wind.
y
/ 1
(_/ 25 MR. PATRICK: Wes Patrick from the center.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 2344433 WASHINGTON, D C. 20005-3701 (202) 2344433 l
315 1 Are you saying, in other words, that you sea 2 the mean not shifting or however you choose to 7S t i v
3 characterize it; you're just typing up your distribution.
4 Is that what you're -- I'm trying to get it.
5 MR. TRAPP: Basically, yes; I see the mean as 6 basically saying between this range, yes.
7 CHAIRMAN POMEROY: But John, is there 8 something in the work that you are going to do that can l
9 result in a change like a factor of a thousand in the 10 first numbers? ;
11 MR. HILL: I think we're stuck with a basic 12 recurrence rate that precludes that unless you appeal to a 13 rapidly waxing system relative to the rest of the basin in !
I\I '
14 range. But I think it's also important to remember that 15 ,my position would be we're all having a better licensing i
l 16 position if we have a good understanding of the 17 uncertainty in our data in models that we've considered.
18 And this may turn out to be not significant, 19 but we need to go through the calculations because we have 20 a credible hypothesis that such things may occur. So 21 let's see if it's significant enough to do a detailed 22 investigation; or, in all likelihood, it's not necessary 23 to do more than just say well, we can reasonably guess an 24 age, reasonably guess the location, look at the models, g
i, ,)
25 consider it, and say we've considered it and it hasn't NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
316 1 affected our understanding much more than saying we've I
7- 2 given due consideration to available models and r
~
3 hypotheses.
4 CHAIRMAN POMEROY: But if you say you'll I 5 accept 10" and nothing else needs to be done, how is that 6 consistent then? I know you didn't say that.
7 MR. HILL: I think we're confusing an informal l
8 conversation with a position we're going to hold through 9 licensing.
l 10 MR. TRAPP: And I think we're also confusing l
11 the lod versus the actual PVHA range.
l 12 MR. COPPERSMITH: Can I comment on that?
13 CHAIRMAN POMEROY: Yes, please.
i )
\/ MR. COPPERSMITH: The calculated -- this is 15 Kevin Coppersmith.
16 The calculations, in fact, do go down to 10" 17 and do go to 10-" . That is again given the definition of a 18 dike intersecting. During the course of quality 19 assurance, if someone wants to evaluate the calculations, 20 there is a low weighted series of branches by a few 21 experts that lead to that.
22 Therefore, it has extremely low weight; but it 23 is the bounding calculation to the total set of 24 calculations which are added together using equal weights rn 25 across all ten experts.
(_ )
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLANC AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
i 317 1 MEMBER HINZE: Let's move ahead, John.
2 MR. TRAPP: Okay.
l 73i 6
~
3 CHAIRMAN POMEROY: John, you're sitting there 4 shaking your head. I can't let it go by. You're standing 5 there while Kevin was making that statement shaking your l 6 head.
7 MR. TRAPP: No, I wasn't really shaking my 8 head. There's a series of concerns that we've got, and 1
9 there are some things that we are doing to try to resolve 10 these concerns.
11 It really ranges more -- it's a combination of 12 PVHA and it's a combination of some of the stuff that l 13 hasn't been presented which is the, you know, Smith-type ,
\
(\ '
'- 'i 14 papers where you are taking a look at -- and the state l l
1 15 would probably object -- but here I would say worst case 16 estimates of probability. l 17 And you are getting numbers above 10-6 You're 18 getting numbers almost up to 10~5 And you are then having 19 to assume a structural control which basically almost 20 forces every event to come straight through the 21 repository, and you're basically having to assume a 22 strongly waxing system.
23 We don't see that, okay? Is there something 24 that could make us change? Yeah, it's possible. But no, n
We want to just take a look
(_) 25 we don't see these numbers.
l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
318 1 at a few of these things out there, these zones, and r3 2 reinforce where we're sitting and maybe hope that we can
.A 3 tighten up our numbers in the range a little bit better.
4 If you take a look -- this is also to support 5 the probability -- there's a series of planned peer review l
6 publications that are coming out; one that's talking about 7 this whole integrated volcanism structural probability 8 model that you've heard. We want to get this into the 9 literature and review it.
10 The San Francisco volcanic field, this is also 11 an area where we're talking about analog -- and there's a 12 petrogenesis of the Yucca Mountain basalt system. In 13 addition to this along the line of probability, we also
' -) 14 would like to put together an issue resolution status 15 report on the probability of silicic igneous activity.
16 This should be a very minor, minor thing. And 17 we're not really sure if it's even worth doing. That's 18 why it's really got a question mark if it's even going to 19 come out.
20 With our present schedule, we are planning on 21 writing an issue resolution report which basically would 22 be a year later down this line on the consequence of 23 basaltic igneous activity. Technical work is really going 24 into the secondary, indirect effects on performance. We
(_) 25 want to get a better handle on the area that can be i
l NEAL R. GROSS
! COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
1 1
319 l l
1 disrupted.
gm\ 2 We want to do some work -- a little bit more i
%/
3 work on this whole Suzuki model to test a few of the 4 assumptions in the Suzuki model. I think both Britt and l
5 Chuck talked about that earlier., We want to also take a 1 6 look -- and this could be the real joker. It's the 7 effects of the repository on ascending magma.
8 Are we correct in making the assumptior that 9 these analogs are the ones we should be dealing with, or 10 is there something about the repository which can totally 11 change this whole type of assumption? And we want to do 12 some more work on this waste package / waste form behavior.
13 VICE CHAIRMAN GARRICK: John, have you been
'- ') 14 given a specific design for a waste package?
15 MR. TRAPP: What we are dealing with right now 16 is the waste package that was in the CD advanced 17 conceptual design, and that's the one that we are using.
18 Peer review publications that we plan on 19 putting out: cooling and degassing of shallow basaltic 20 dikes which are getting to the secondary effects. There's 21 tephra dispersion and risk analysis on Cerro Negro.
22 There's a lot of work that needs to be put together there.
23 And especially the one I think is most 24 important is the development and evolution of the 1975 t
w_/ 25 Tolbachik eruption because we're using that a tremendous NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 320 l 1 amount as far as an analog.
,f g 2 In the area of data quality subissue,
]
3 basically we are going to be reviewing the DOE volcanism 4 synthesis report. I think I heard the day that now this 1
5 will be coming out at the end of this fiscal year. And 6 again, looking at the significance of these buried i 1
7 geophysical anomalies.
8 Now this last one probably is of more interest ;
l 9 to a whole bunch because now we start talking a few things 10 that cross cut. This first one, the 11 sensitivity /importance studies, volcanic system, total 12 system, we've done a little bit of that. We have got at 13 the present time the draft TSPA code in house.
14 We're trying to -- testing this ".hing out 15 which does incorporate the Suzuki model. I believe at the 16 end of this month we're supposed to get the final code, at 17 which time we will really start running these things, 18 doing a bunch of different sensitivity studies, vary 19 parameters, and see exactly where we sit.
20 We also will be -- well, that also would be 21 something, before I go on, to be discussed to whatever 22 degree the ACNW wishes during the July meeting. j k 23 Yes, we're going to take a look at the DOE l 24 TSPA-VA plan. Obviously we're going to have to review the n 25 final TSPA-VA. We've got to take a look -- and this
()
NILAL R. GIUO55 ,
COURT REPORTERS AND TRANSCR!BERS l 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
321 1 really isn't a straight geology, etc., but it's more in 2 the area of the performance assessment people, but we've 7~
('~' l 3 got integration we've got to worry about, and let's take a 4 look at the dose sensitivity to differing critical group 5 locations to other possible pathways.
6 For instance -- and I'm not sure if we're 7 going to, but we're planning, at least right now, on 8 looking at these five kilometers effects of igneous 9 activity which, you know, wouldn't be the well, but it 10 would be somebody around there and maybe a rancher kicking 11 up dust -- and really integrate all this stuff, waste 12 package / waste form, structure, PA, etc., into one coherent 13 package.
(,,)
\- I was asked very strenuously to talk about 14 15 reprioritization. You see as much as I'm.really going to 16 say on reprioritization at this time. We've got these 17 sensitivity /importance stuf.ies. We're going to be taking ,
1 1
18 a look at that. There's a whole series of budget l l
19 exercises which are going on right now.
20 And basically, when we get done with the 21 sensitivity /importance studies, we get done with the 22 budgeting, what I've shown you here may have been a total l
23 misleading representation. I don't think so, but it's 24 basically -- it's going to be reprioritized, so we'll be j0
( ) 25 putting the money where -- and the effort where we see the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
322 I
1 most benefit.
,es 2 MEMBER HINZE: Thank you, John.
r i Nj 3 Questions? Ambitious.
4 CHAIRMAN POMEROY: John, let me ask you a l 1
4 5 question about resources.
6 What's the level of FTE involvement combined 7 in this particular activity?
9 Right now, we are running at 1.4 NRC which l 10 covers basically myself as portion for project management, 11 all the managers, and the on-site reps; and roughly 2.5 l 12 people at the center which is basically Chuck and Britt, l l
13 some support from Conway, and the other support from their
()
14 agency.
15 CHAIRMAN POMEROY: So four FTE's is a rough 16 number?
17 MR. TRAPP: Four FTE's. Roughly, you've got 18 Chuck, Britt, myself -- we're doing it, for all practical 19 purposes, almost full time. And then --
20 CHAIRMAN POMEROY: And is that the planned 21 level of effort in the next fiscal year?
22 MR. TRAPP: As far as I presently know, and i
23 I'm seeing Mike shake his head yes.
24 I haven't been in the budget, so I'm glad to
,a 25 see that.
(J NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
323 1 CHAIRMAN POMEROY: Thank you.
n 2 MR. TRAPP: I've got a job next year.
i
(
3 MEMBER HINZE: Tim, if you could just give us 4 a brief review, that would be helpful.
5 MR. EULLIVAN: I'm goir.a t. 0 take a quick look 6 here at the two subissues. First, hazard analysis or 7 probability es*imate. As I've rmterated several times 8 here, the PVU was sponsored and is still intended to 9 provide a sound, defensible basis for licensing.
10 .As described by Kevin, DOE has fulfilled it 11 commitment to evaluate the new information, the new center 12 information; and we have concluded that these data do not 13 significantly impact the results of the PVHA. Based on
- 14 this and as well statements by the experts at the end of 15 the final workshop, we believe the results of the PVHA are I l
16 robust.
1 17 And new information, short of dramatic !
' l 18 paradigm shifts, are unlikely to change the disruption l l
l 19 probability document in the PVHA. As I've said earlier, l l
20 no further site characterization data collection is l l
21 planned for the reasons I already discussed. l i
22 As Abe discussed earlier today, we intend to 23 use the PDF from the PVHA and the reference case and do 24 selected sensitivity studies for TSPA-VA. We plan to n
) 25 address numbers three, four, and eight on the list of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE. N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
324 1 agreements that John and I went through and document that fy 2 in an upcoming letter which we anticipate will be coming 2
(_/
3 forth soon.
4 In our view, the next step is subissue 5 closure. Although the schedule is not clear to us any 6 longer if data collection is anticipated by the NRC and 7 the center through the end of FY97, we're no longer sure 8 what the issue closure schedule is.
9 In short, this indicates that the results of 10 consequence analyses that will be performed next year will 11 be documented in the TSPA-VA. Consistent with the 12 agreements from the technical exchange, the next step in
, 13 this should say after completion of the TSPA-VA an i) 14 associated documentation is consequence to subissue 15 closure.
16 Now I think this simply reiterates some points 17 that Abe made. We will, in terms of the letter and future 18 presentations, be rethinking and rewording that third 19 bullet following John Garrick's suggestions.
20 MEMBER HORNBERGER: How about the second --
21 MR. SULLIVAN: Yes, the second as well. So 22 let me leave it there, unless there's any questions.
l l 23 MEMBER HINZE: Thank you very much.
24 Any questio.is for Abe?
p
( ,/ 25 MR. SULLIVA17 : Tim. Or Abe.
l NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS
! 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
325 1 MEMBER HINZE: I'm sorry. It's early in the 7- 2 evening.
3 VICE CHAIRMAN GARRICK: One DOE is just like 4 another.
5 (Laughter.)
6 MR. SULLIVAN: Well, we hope we all give you 7 the same answer.
8 CHAIRMAN POMEROY: Very good.
9 MEMBER HINZE: I want to -- before we go any i i
10 further, I want to thank -- make certain this gets on the l l
11 record thanking Lynn Deering for putting together what I 12 have found to be a very helpful, helpful workshop with the 13 excellent presentations from the DOE and the NRC, as well
'k-) 14 as from the state and participation of the consultants.
15 I think it's been very useful to us. l 16 We are supposed to go into a round table at 17 this point. I feel as if I've been in a round room most 18 of the day. I frankly had 13 questions, and I don't know 19 if there's something ominous about the 13, but 13 20 questions that I could kind of isolate to trigger some 21 conversations during the round table.
22 And I've found, in going through them, that l 23 every one of them had been discussed probably to that. I 24 wonder if any members of the committee have questions that lD i
) 25 they would like to kick out into a round table for a l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234 4433
326 1 conference?
f~s 2 VICE CHAIRMAN GARRICK: I've heard enough
/ 1 L) 3 about volcanos for one day.
4 (Laughter.)
5 MEMBER HINZE: Well, I think it's all caught 6 up with us.
7 I'm wondering if our consultants would have 8 some general statements or specific concerns that they 9 would like to raise as a result of hearing this barrage of 10 information.
11 MR. FOLAND: Not so much burning issues.
12 Unfortunately, Bruce is not here, and we probably should -
13 - had a brief conversation at lunch, and I think the PVHA
,q
- 1 A/ 14 analysis is very, very useful and was quite successful.
l 15 But I'm reminded that about 25 years ago, a 1
16 colleague of mine stood up at a professional meeting and l 17 talked about water on Mars and never did publish the paper 18 on the water on Mars because all the expert opinion was l l
19 unanimous against there being any water on Mars. And 20 right now, we're trying to figure out when the water on 21 Mars was frozen into polar caps.
22 And so, with that comment, I hope that puts 23 part of it into perspective. And that is, consensus 24 doesn't mean that that's in fact the right answer. And
,a
( ) 25 this is what worries me, that the analysis is being driven NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 2344433
327 1 by the fact that there aren't many events that we know f ,. 2 about.
~
3 I see John Trapp and the center people saying 4 are there other unknown events, and then I hear the answer 5 is well, they don't make any difference. I guess that 6 troubles me a little bit that there may in fact be unknown 7 events and do they -- will they have some significance. ;
8 And I would be interested to know if the 9 experts in fact expressed an interest in seeing any more 10 data that would have been particularly useful for them to 11 be able to use to reduce their uncertainty. l 12 That's not a question for anybody. That's 13 just a comment. ,
/x i \
\~/ 14 MEMBER HINZE: Thank you very much, Ken.
15 Mike, are there any general comments? Any 16 concerns about the differences that we have between --
17 that we've seen between the DOE and the NRC?
18 MR. RYAN: I don't really have any large, 19 huge, general comment to offer by way of a conclusion.
20 I've been delighted in the forum of the meeting and being 21 able to nickel in, so to speak, as we've gone along from 22 virtually 8:30 this morning until 5:30 in the evening.
23 It's been my pleasure to be here.
24 I think I've either directly or indirectly
/,
( 25 hinted or simply explicitly talked about my concerns as NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., kW.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
328 1 we've gone along. I tend to think of volcanic systems as y 2 large, integrated, three dimensional structures with
\ ')1 3 processes and dynamics that are ongoing in them, and the 4 surface activity we see are really skin blemishes of 5 larger beasts at depth.
6 Of course, we don't have real time geophysics 7 in terms of real time micro-earthquakes that are -- that 8 would have been -- that were generated in this that would 9 help us map out, for example, in three dimensions what 10 these -- what the morphology of these bodies look like at 11 depth when they were actually operating.
12 We don't have the deformation fields that we 13 have at active volcanoes. So we're -- it's been very much
,r\
k )
\' I frankly applaud the 14 a post mortem kind of exercise.
15 folks that have been involved in this and the models 16 they've come up with to assign probabilities as based on, 17 frankly, exceedingly sparse data sets.
18 MEMBER HINZE: One question that has come up 19 here in the last conversation is the possibility of a 20 waxing volcanic regime. Is there -- are there methods --
21 are there protocols in place for looking at this in a --
22 other than in a reoccurrence interval among these 23 relatively poorly data'ed and sparse volcanic activity?
24 Are there other techniques that could be used
/~N
() 25 to give that warm, fuzzy feeling that we are in a static NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
329 1 situation and can use that for the next Lundreds of I
7~3 2 thousands of years? l j
- 4
\_/ !
3 MR. RYAN: What immediately comes to mind is 4 in a somewhat different tectonic context perhaps, the fact l l
5 that magma can come into a system sometimes aseismically l
6 and be resident without triggering any surface awareness, i 7 so tc speak, of its presence.
8 Now typically, this is not the case in 9 basaltic systems. More typically, it's the case in 10 silicic systems. And Mt. St. Helen's is a very good 11 example. The USGS was, of course, aware that St. Helen's 12 was going to kick off or do something potentially i l
13 explosive very early in the year 1980.
4 14 But we had no idea really of when that might 15 happen. And of course, once it did, then we developed a 16 series of models that used both seismicity and deformation 17 to rather closely anticipate when the next dome forming 18 eruption was going to occur and when the next ash plume 19 eruption was going to occur.
20 But these were, of course, very much after the 21 fact. Before the fact, there can be a lot of -- in the 22 case of the silicic system, a lot of aseismic creep going 23 on and aseismic magma migration. And of course, the trade 24 offs between deformation amplitude and depth of the q
\ ,) 25 pressure source are such that pressure sources of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4 433
l 330 1 significant volume can be five, six, seven kilometers deep
- 2 and really not have much expression on the surface.
U 3 MEMBER HINZE: Is the technology of crystal l
l 4 size distribution, CSD, in terms of looking at waxing and j 5 waning situations, is that something that's applicable 6 here?
7 MR. RYAN: Well, Bruce Marsh and Cathy Cashman 8 have been two of the key players in this, and it's been 9 basically Bruce and Cathy's child, so to speak, in terms 10 of petrologic application. Basically we're just talking I
11 about crystallization processes and igneous bodies which 12 are experiencing substantial cooling.
13 My sense is that -- is still in need of
(~)
O 14 laboratory quantification in terms of calibrating what 15 crystal growth rates would look like in olefines and 16 pyroxens and feldspars and so forth. Now some of that's 17 gone on in the last 25 years, but I think more needs to be 18 done.
19 MEMBER HINZE: I didn't mean to put you on the 20 spot here.
21 MR. RYAN: I would hate to hang --
22 MEMBER HINZE: This was a question for Bruce -
23 -
24 MR. RYAN: Sure, sure.
O Cl 25 MEMBER HINZE: -- and he escaped on us.
HEAL R. GROSS )
COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. j (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 ]
l j
331 1 MR. RYAN: But you can see even -- in
-s 2 exceedingly active systems like Kilauea in Hawaii, you can
('
)
3 see dike like bodies that have stagnated at depths of two 4 to four kilometers that have had really rather large 5 excursions of plagioclase growth so that the plagioclase 6 crystals are megascopically obvious from hand sample.
7 And yet, all else indicating that these 8 systems are extremely active with lots of through going 9 magma and so forth. So simply the chance intrusion of a 10 body slightly off axis, slightly off a conduit's axis, 11 will set the stage for stagnation.
12 MEMBER HINZE: Yes, understand.
13 Would Tim -- I'll get it right -- Tim, are kl 14 there any comments that -- or any concerns that you would 15 like to raise -- that DOE would like to raise?
16 MR. SULLIVAN: No, I'll pass.
17 Kevin, do you have anything?
18 MR. COPPERSMITH: One comment on your last 19 comment.
20 About the time history issue, there's a couple 21 of ways for hazard -- well, let me make an overall comment 22 that -- going back.
23 The issue of what's important is a very i l
24 difficult one. This is a hazard analysis, and not all !
((~h
~
,) 25 things are important to hazard. For example, the ;
l NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W. l (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
332 1 arguments that went on prior to the time that we came in
,- 2 to do the hazard assessment on Lathrop Wells and whether 1 i
\'~/ 3 or not that was polygenetic or monogenetic were downright 4 knock down, drag out fights.
5 Very important from a scientific point of 6 view. When you boil down -- it boils down to the basic 7 difference between whether or not there's one event there 8 or four, that's contributing to a total regional -- say 9 ten to 20 events. It doesn't have a major impact.
10 That isn't to say it isn't important; just 11 that its impact on hazard is not significant. But when we 12 talk about the difference right now of saying well, 13 whether or not there are buried events doesn't make a (h
s
\I 14 difference, that's in the very narrow world of hazard 15 analysis because there's only two things that matter to 16 hazard analysis: where future events will occur and what 17 their average rate is.
18 And those two things, unless they're i
1 19 dramatically different, don't have a big impact on the l l
l 20 result. Now the result is couched in terms of orders of 1 1
21 magnitude. It's 10-8, 10 ~ 7 orders of magnitude annual 22 likelihood. If we were talking about this happening three 23 times a year, it would be a totally different likelihood.
24 So we're dealing with extremely low likelihood
/
( ,! 25 of something occurring. And that's why this is a strange NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
333 1 hazard world of very low annual probabilities leading to l 7x 2 only certain things having any impact. l
\
\ ) l 3 I just wanted to make that point. i 4 Now the issue of time migration dealing first l
5 with the spatial migration issue, that is something that l 6 Bruce Crowe and the LANL people looked to in a lot of i
1 7 detail. Where do the -- where do the individual events 8 occur post ten million years? Where are those centers, 9 what's the distance to the next one going through a time 10 ser.ies, what is the azimuth to various events over a time 11 period?
12 We found no statistically significant 13 distribution on migration over that time period. So they l \
2 14 did look into that. In terms of the time history of 15 change of recurrence, there was waxing and waning which is 16 changing frequency. We did -- Dr. C.H. Ho from UNLV made 17 a presentation, provided papers on modeling using the 18 Weibul distribution.
19 And he has shown that he thinks -- show 20 indicate that there is a slightly waxing system that was 21 presented. The experts in general saw that in fact it's 22 very, very sensitive to your start time. You start right 23 after -- right before five million years, you get a 24 different result than if you go back, let's say, nine
() 25 million years or just after the oldest or the youngest NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
334 l 1 miocene volcanics, there's a big period of no production; j 2 and then you go into a period of production.
7-)
3 That's obviously going to be a waxing. With ;
4 few events, again we're plagued by the absence of events.
5 Some say that's a wonderful thing to have. Volcanologists 6 say it's a horrible thing to have. Those models were 7 available and were considered.
8 In general, I think the only experts who 9 really looked int 6 a model is Rick Carlson. He used a 10 volume predictable model that says that volume production 11 through time may change. What volume predictable says is 12 if you know the amount of time it's been since the last 13 one, you predict the volume of the next one. It's like a
( )
14 model for an earthquake.
15 Then you can use that and the average size of 16 an event or volume and you're basically decreasing the 17 time between them. So he used a time history of volume 18 production as an indicator or a change in the situation.
19 The volumes have changed little so that that model is very 20 close to a homogenous model, but slightly different.
21 They are -- I think spatial and temporal 22 migration was considered in this project. Models were 23 very simple as dictated by the absence of very few actual 24 volcanic centers.
k ,) 25 MR. HILL: If I may interject for just a NEAL R. GROISS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. l (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 1
335 i
1 minute?
2 Both of those statements, Kevin, really lay O
i 3 the groundwork for the potential impact of new 4 information. Not that we're saying that there are ten new
,- 5 volcanoes and it has this effect. But for the spatial f 3
6 migration, certainly Bruce in association with, I believe, 7 Kevin Walnum and Goldner Associates looked at migration l
8 through time at trying to take that vector-based approach 9 ano found that there was no regular spatial migration.
10 But the presence of new volcanoes and, it admittedly, one or two new volcanoes, probably won't
(_
1 12 change that. But if we're dealing with more than a couple 13 of volcanoes, we need to trka a close look and evaluate 0 14 whether that new informativa may give us a discernable 15 spatial pattern, especially if we look only within the 16 volcanic system of interest for PVHA, the five million k 17 year and younger, rather than the entire post-caldera E
18 episode.
19 By the same token, a lot of the experts had a 0
- 20 problem with Ho's model because of the picking the 21 starting time. But the new information that would also 1
22 make it difficult to support a waxing system would be 23 inclusion of the Funeral Formation within the Yucca 24 Mountain system.
25 Regardless of your startinc time, as long as NEAL R. GROSS COURT R50RTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. i (202) 2* i-4433 WASH:NGTON, D C. 20005-3701 (202, 7 4-4433 !
1
336 1 you're around five million years within the period that 2 most of the experts considered relevant, including the 7_
- 1
\~
3 Funeral Formation, gives you a steady state recurrence 4 rate even if you wanted to use Ho's definition of start 5 times, spacing, and how he manipulates the beta function.
6 And this would be the point of new information 7 may not set the world on fire with order of magnitude 8 differences in results, but it certainly gives us 9 confidence and an ability to make an informed decision 10 about how we ultimately use or discard models in the 11 licensing process and our absolute confidence.
12 You know, our comfort level is the term that's 13 been used quite often in how well these mathematical n
kJ 14 simplifications of a sprrne data set represent the system 15 we're trying to understand for a ten thousand year period.
16 So I'm not disagreeing with you at all; but 17 just that the new information needs to be considered and l 18 --
l 20 From my point of view, it's DOE's 21 responsibility to make the best quantification of 22 uncertainty knowledge about this issue as possible. The 23 regulator needs to feel as comfortable.
24 MEMBER HINZE: I will call on John for any 73 last comments, (v) 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 234-4433
337 1
1 MR. TRAPP: Just a couple very minor points.
- 2 If you take a look at --
3 MEMBER HINZE: In fact, I'm wrong.
4 THE WITNESS: Most of -- I've probably said 5 what I had to say about ten times today.
6 If you take a look at the new information 7 issue, I'd like to stress one point there. The new !
1 8 information, in some ways, really has been better in 9 getting some of the integration and the integration of the 10 various models. For instance, if you take a look at what i 11 we gathered dealing with the Little Cones area, we started 12 being able to talk about the total basin subsidence or 13 relationship of volcanism to basin subsidence in the 10 I t
\~/ 14 faults in a better tie in of the structure of volcanism 15 and ability to put together a more comprehensive, tectonic 16 model that included all the things.
17 So the fall off of some of these things is i
18 * , > re -- and it has to be looked at more than just in the l 19 straight volcanism standpoint. The next one I would bring l
l 20 up -- it's my last point, and I'll say it kind of tongue 21 in cheek.
22 If you're looking at the geophysical data, if 23 you take a look at some of the microseismic, there's an 24 interesting warn and small microseismic at Lathrop Wels.
()
A 25 MEMBER HINZE: You have to keep those goats NEAL R. GROSS COURT REFORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D C. 20005 3701 (202) 2344 433 h
l 338 l i
1 out of there.
2 Well, with that, Mr. Chairman, I recommend
,3
'~']
\ l 3 that the committee just -- the information they've l l
4 received and try to bring it together in the future. l
\
5 And thanks to everybody.
l 6 CHAIRMAN POMEROY: Okay, I'd like to first of 7 all thank Bill and Lynn both for a r: .ly excellent 8 interchange of ideas and discussion today. I really think 9 it's been extremely useful.
10 I want to remind people that they -- we are 11 going to discuss the use of expert judgement in this 12 process and other processes at our May meeting. We'll try 13 to do it on May 22nd. And as I've said, we would very i
,~
\ i A/ 14 much like to have -- and we'll work with Carol Hanlon to 15 have both you and Kevin participate, as well as Steve, in ,
1 16 that discussion.
1 l 17 And we'll certainly try to have some l 18 appropriate NRC people here also to participate in that 1
19 discussions. We've looked at two concerns that John and l 20 management have brought forth. And I think we've gotten -
l l
l 21 some really interesting information both on the question 22 and the concern regarding new information and the second 23 concern that we discussed.
24 There clearly are other concerns . We will
()
,a 25 take those up with the appropriate people.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
(202) 234 4433 WASHINGTON, D C. 20COS-3701 (202) 234-4433
i 339 i
1 I's like to thank you all for coming and lasting until 10 minutes or 15 minutes of 6:00. We won't
,ey 2 l\ ) return until 8:30 tomorrow morning.
3 4 (Whereupon, the proceedings were adjourned at 5 5:53 p.m.)
6 7
8 9
10 11 12 13 7_.
N 14 15 16 17 19 19 20 21 22 l
23 24 ry
's.s] 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W.
CO2) 234-4433 WASHINGTOf', D C. 90005 3701 (202; ?34-4433
I l
I !
' (V 3 '
\
l CERTIFICATE This is to certify that the attached ,
l proceedings before the United States Nuclear j l
Regulatory Commission in the matter of: l l
1 Name of Proceeding: 91 8T ADVISORY COMMITTEE ON NUCLEAR WASTE (ACNW) MEETING Docket Number: N/A Place of Proceeding: ROCKVILLE, MARYLAND were held as herein appears, and that this is the original 1
transcript thereof for the file of the United States Nuclear fT G
Regulatory Commission taken by me and, thereafter reduced to typewriting by me or under the direction of the court 1
reporting company, and that the transcript is a true and accurate record of the foregoing proceedings.
t sua (C, ORB 5TT'RINER' l Vfficial Reporter i
Neal R. Gross and Co., Inc. ,
l I
! 4 i i l
l I
,/R.
NEAL R. GROSS COURTREPORTERS ANDTRANSCRIDERS 1323 RilODE ISI.AND AVENUE. NW (202)234 4433 WAS!!!NOTON, D C. 20005 (202)234-4433
O O P O '
United States
(~
\., ,[}< 'l*\Nuclear Regulatory Commission INTRODUCTORY COMMENTS Presented: ACNW APRIL 22,1997 By:
John Trapp, Senior Geologist Engineering and Geoscience Branch Division of Waste Management Office of Nuclear Material Safety and Safeguards (301) 415-8063 e-mail: jst@nrc. gov L - - - - - - - - - - - - _ - - - - - - . _ - - _ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
O O O ':
^ 7 l1 <y United States
\,,,,,,.I Nuclear Regulatory Commission l
INTRODUCTION
- FOCUS OF TODAYS PRESENTATION IS
SUMMARY
/HIGH POINTS .
OF FEBRUARY 25-26 DOE /NRC TECHNICAL EXCHANGE
- MUCH OF NRC PRESENTATIONS COVERED IN ANNUAL REPORT
- MAIN PORTION OF DOE PRESENTATION IS
SUMMARY
OF PVHA REPORT
- BOTH DOE AND NRC WILL SUMMARIZE PLANNED WORK
- STATE WILL PARTICIPATE
-MAY MEETING WITH ACNW AND PLANNED NRC/ DOE MEETING WILL DISCUSS EXPERT ELICITATION CONCERNS
-JULY ACNW MEETING TO GO INTO DETAILS OF TOTAL SYSTEM PERFORMANCE ASSESSMENT AND (AS NEED BE) RELATIONSHIP OF IGNEOUS ACTIVITY TO TOTAL SYSTEM 2 April 22,1997
O O O H l ,1 -.<1, United States
\., ,,,,) Nuclear Regulatory Commission IMPACT OF NATIONAL ACADEMY OF SCIENCE RECOMMENDATIONS ON IGNEOUS ACTIVITY KTI OLD EPA STANDARD BASED ON PROBABILITY OF SCENARIO AND RELEASE TO ACCESSIBLE ENVIRONMENT NAS RECOMMENDATIONS FOCUS ON DOSE OR RISK TO AVERAGE MEMBER OF CRITICAL GROUP WORK HAS SHIFTED FROM EMPHASIS ON PROBABILITY TO EMPHASIS ON TRANSPORT AND DOSE PRELIMINARY WORK SUGGEST PRIMARY EFFECTS
- IMPORTANT, HOWEVER, MAIN SECONDARY EFFECT MAY BE JUST SHIFT OF TIME OF PEAK DOSE l
1 3 April 22,1997
~
O O O n
!><~y United States
- %,,,,,/ Nuclear Regulatory Commission i i
i ISSUES RELATED TO PVHA CONCERN WITH EFFECT OF NEW INFORMATION CONCERN WITH GEOLOGIC BASIS OF ZONES STAFF RECOGNIZES DOE FREE TO USE WHATEVER INFORMATION THEY DESIRE TO MAKE LICENSING CASE i NRC WILL CONSIDER FULL RANGE OF INFORMATION t AVAILABLE t
'l April 22,1997 i
I
O . - . . , D. , . O." _
! GEOLOGIC SETTING AND PROBABILITY OF !
VOLCANISM IN THE YUCCA MOUNTAIN i l REGION AND DISRUPTION OF THE PROPOSED l l REPOSITORY l i l NRC DWM Program Element Manager - John Trapp j i
i CNWRA Element Manager - H.L. McKague !
I l
PRESENTED BY ;
l l CHUCK CONNOR l April,22,1997 ,
CNWRA Contributors l
l C. Connor, M. Conway, D.Ferrill, S. Greenon, B. Henderson, B. Hill, M. Jarzemba, P. La Femina, S. Magsino, L. McKague, R. Martin, J. Stamatakos 4 W WRA 1
O ACNW Wor iep D 1
OUTLINE l
! l l . Regional Structural Setting Of Basaltic Volcanoes i l
Near Yucca Mountain ;
4 l . Recent Geophysical Surveys And Their l
. . 4 l Significance :
l . Summary of Geologic Factors to be Included in !
Probability Models l i I
- l 1 l~
i l
l-i s t {
! I i l l i ;
l i i t r
e cNWRA 2 l
f l
1
!O EXTENT OF YVIR VOLCAhlC SYSTEM
- i. I - W 4p ar v W:.p.g, .,r; .
,, gp. jp/ m , \ c; b.4' ,
~ . ~ z ,. mp; r Q.
\ <
{ ;u;, y , .,. . .4..\ 12;,p)p ;,-
/ ,- Q (
. ., ,yl ;a, ..,,.; e s ..7, : .
U N. , ,. -.. - 1 4 . t.<1
,s,s , ,
- t(,
.( '.3/
.b,(.'h ,
At ~ ,'
c :' q o i V V; % nM) 7 .pd .; 1 Y;;
g.y )f)l,v'm ,T 9)p'f.T6.ff'l'?,'4)y~D;@'N mfsl f' 'vC""8d/qC$l L
m (
s h,_ , ..
l .
,, f.'fg g'. ;y , fi h t,.,y h k N 4
. e[p,f!h'h,h. .
i oh>. h 4 s,f amh.N v'1 k'*,d&(g$
sp w . g e %og
- z. .
! .cs.. u Mi
' m'd <$JLr E Y y yl v.6pA e.
r my fi- %
...,5.xllp..i h ia,
';\
q}Q(a 7
h .1 ~b ) CF Ms p, f)gs@m%yey,%sQ
/,
~.
w
!>..r4r-
>x 1
.., g, ,, '
, e .. g r\U,;&,,,'S;1 YM~ ~ s l /
1<yt, Non,',w?mn.
' f q's;:'.y:u Ls.f .\61 %
p i h h(
Q'%p li
. J. v.. ! ..>,- I
.~ -
\
g 's a f , .
e f /e iry . es .-
., {te s m,s,h. (gggg,_ s yA<.
l :/ vMn '.\g$s %,e 9;x. . \s:m...s 4. y f _. - ,, , vj c. e. s q<ggq,j[(, -) , 3 x
, s, #
. p ,
. . ,u .
(3 ;l 1(a C' 9,$; \ v ' if) ..-Q., P4.-. $\ ,@,,.g.m( :2:
~"
yA ,,j,Q,W'J %[,s,a % J~- -
v 4 e
Le s .. v',h
.. % s s ~
9s hsjCM g.S T ,' m'. ,
5l ~4),$y,_1'y * ,
'? "' ' t..
/ jlf' 1
,nh,M'dQqi e* f- ['~
S;f h A;h lf$ .j?@b fa.
h\ ! hk m
V.: , .. lM?jFkkR A, k KNon g. 4 p
x,sh'\ f ifsd . % g43 2~ E u( f - - , q3
?'qdg$U s
%l c [3hh]tlhh.lfj iYMR 1 1 f hh j.
. \, t' ' s.
i.Yh??.d:h.(....k(7b?
@ M
'Nh@NIk[ll l
'N14 I f% V
) I, . i l p S l, N rN - 8tg
'g'
- i $$gri; '< yc.siM rt4M) 's \ig(3.,
7 hk ]$6Shpu p ' <hw'.: c '"
/M $' ( \
\\ \ Mk9f li .
h:.w { )'
- 8. s lh,.,4 )ms mNu?g% Nh/M.be h'h.N.,Ihk'hM t'.
. .mgdg. l . @ ~
i '@p <2 Ma 9 2-6 Ma' * '
dj;n 51 \ ' ^V' S N.a ammmm h'lAs 14yb b % M $Y'/,. N E q, 20 km e-12 Ma Need a geologic basis for system definition: spatial, temporal, structural, isotopic C"- Funeral Fm. basalts = steady-state recurrence <6 Ma
. i l
i
. p Cross-Sections Throuah Crater Flat and Yucca Mountain -
- V Bare \
' Mountain WNw Fault Crater Yucca Ese N Mounta,in Mio9ene volcanics Flat o i
1 me R m: M.
c mc agasm # wan=
1 kw A
c ::. .a \
i 4,, . l*. .' ;' ,.,
q*gh m.g._n.., .. ' ;, { - ' f
-0 i
- m. m.:n, . . .
_ .:'.,.. -s
.. _ . . ., 1 h,l7'W i.',Ty ..
1.: -,6 w . ,
.,'..,"l 34 m . .
. .f.. . . . .
) p ', . 1% [ . ' . ,:'.( ...;'- ' ', , ' ~
.-l'..
EI
-10 km
! A A'
- Bare l
! Mountain Yucca
- wNw Fault Crater . ESE i
Flat Mountain 1
N o
~ * **
, .
; i t
4 d
-5 1
1 i , ... -10 < f//// / ////////
~ ~ ~
h 3 >>>>>>>>>>>>>>>>>>>>>>s >s r s 1 1 1 2 1 2 >s s s s 's iiiii ,, j A Bare A' t Mountain i Fault Yucca l Crater Mountain l hf{ __ _...- ~~
- y 4 j y &F t v.%;16 R7, , a".,-g, i ; n.,- p< ,;j
' --g p yp 'N
- ,, m
- na %- ' jeg}j,,g g /i nfs
~.
.yp f's y
w w.m mal ,.ek Yh A ?h >?.l:k? Efl W - buw ?,'?, f_f' Y 'U NE bTd: ih,~Y i$my.^~~ v G W'},f* i{ m ',imp m um' @~ o }. h ?: Miocene-glbt, hkk A $wim?{)y ;ffh?
;;;;,,;,,;;, $ ~ 3g;a. gf ,??rt;;;;;;;,,, , $h,Ybicanics
; n ; ny, ,
gn
~;;;;;;;;;; u 4 n ~- .
/
Crustal Block (internally deformable) (
- a g Aa pb wwm(e s+
%end(; e; sw Pm m
,f... w=4)aggw x en wxm w www a ,
wAggdhuwmqh9 vnmm wnn;en ' e a t j' , f k, m' n yh,M) *ys w ?l?w, ;l'f(( M l ffi ~~y, L O e d * ,"' ; , j [ ~ q, , _n TkY , g o,; v-
.n g p?(pt:y, 4' J;'3,'w' emq
[.9 r n u % pg 6 _. y.
----+'s e.- n #g:oy m,. .wrm . ua, cm? . us z*
lf .; f. e aQM- .44 g . y 7 wu,.+ C W , s 0;eY'
, ; n,., l/ fj _ ,
,a e .
n gh. , ac amps wn qw mm w m;p W QG&pAqf a e ,u v. 1fi% ND gt- . 4 n,c,; $ If . yh wn e {= e'Q$m;m;g ~>,$g% W hW W y
~
af E ag_eun
~ ag ng gs t lilHHHl!Hil IHE E ElH %
k ._ E.IEkkkE.!h~ . el '" w @;] f f kney
; f J ',)/ -
$M h&
y* - - ,fhdl)c@6h& Q
^
jf* ,A 3 L i g). a,, . Y, l'..-Qo
-Q,. h yl
- m Q,)lQ,cq fy\s
,f +[ a i%
m !\ '* y r w'i
':\f il m o a. ,f ) b h _ . ,/~f 'h f
~m %" mkp(u ll' -
3l 3 . L.; : I 'Q W M ? f } EQ N W :
\ ,
, l- - t b 'I) $\k ke ; ." -
),' -
n I Q y t
- N~M'kk%%
=
- W
[
- pm C- 4@S g' , ijS M., $. w1l'/y -
- a. i, ; , l -1 p 1, A
w 4 : , (/ / l lf<,j) , i
%u.
;i 2
M:, 0 N, %pg4ff f @y. &)A{ % Kh // 31]vm".nly// Q3% %{' 9L i x lL* , M s. m s ! J
,_ dA ' '7 ./>['s . . x
~-
#9 M y M i.
~
>=
b " k /q h[ f s-pg f n +\.N w . .- e 2 n. e . 4 ue i + m
, - x y B. .h. WGM%g < A s) f e% - i mC_ 4l a
,1 mDM mY h g 1 s ]i/ r
>4 )m%YP h PL"[ 'i ghh'hhf[I d lhNW. f& a. j & ru w w m;y.,3~;u 4r w' \ q l AJ e x d'
- n r9 [ykh'g'e - -
C
})k/',/-, ,M ] ,
n
, ,i
_h; , b , hx$k Ikh&l
- k. ,h w
fDj e a m.h l m e #w a pc J hk)[f hy f bm.,[f,-_..N @G W ='* + g,f@ I
@lk{
Qg [NK: ,,,%' gS y( Qf t\ q?
- e. } ,n.gjgg (f l y'f yl
.. . m , _
5%T" f y mg ,
, ,z _
/g;;f t, fa ' +:p
.+ j m ,w., l A +.
C W ^;&'m ff ! - C ? = hyk p y h
, lhf r ya..
agm o f , a tM _ z ., . e' _- G4 [h
'. ( ,,, 5:
d jk i
=
W, .,j mW- " J. ' j'v 1YYgt,Y M r.;Y Jjm[sl ;% w 2 .. M, (t ,n M< w m*- 29
- d. . d , ,
m.. =m wd D
' % d' .A
;/hj,Q ffd l t .i y l %,.
k d4 h b hh h ana a Ri g s m m a *4(r v m c bk w A s' mhw c:w M vh h a q% ,'
- " &&q$y/f,,Qy@g
~ > <~
d mpqM'pa
,, ? 8
>;c .pa
. ~ , W;y w n. ?
%%y&w@as
- air s >
),
+u
'3
_ q
? .
3.J[
my ggg.' ; ' f pgg;y' y;g3
*) ,- P" g r- - -
[ / 4_1,L. ,% - J gi,]-- ,c_- .,.3
; - . a'.g ,_ sg 7;( w.
; yy ; ,y}{lm ::. ,', j ;w. ;,Wp gy.pl yl '}{Mf_'.,3; yt); '
+
v.fw% s y 2 :% %y:;vgg}:fhy :' %p :43.; ;w[ , ;;p ;r%;% ; ;
- 3) , , > .
3.;p.: , _p , w; ww ; m.,1gg,, x, _g- : .ws 7,g w
> s; n
, ;x -
^
q '. . - l., t r lY r .. , b $l h?lli$' . ' , ';h;' ;,h j:i!: n r #
\
i0
.., . - l h;t?):; > wn W],511
^
Y A: ' Ti ' , > l C* ,
; 4t //;0~ g*J 'r .fjf[1
&[< , y,[':[- c,_. 7 4.g
}O(:
q,,,,,-
-yv '
m: NF5% I '
- o-.
., t v ~
4 5 J H :- ' t wi;co 1.d, G CD x :.k W;//: .ri.
'Ob i k'n h;Ca&dO. : . , s ',
i whon n y ~ mQ mg% 4:~m+ s% g g ~, 3 A f3 g4 _ . , n) - Q g
+
y .
-y c7 y k,';_$(6; gq/)
04 % ' '
- 3. i 1 o.
,~
j QC, ' Q(x-9 k -'? y)W~ C/ ( :E
- cr me a,-- m
- g/
-; ~ s j w e x , ..
T yM&,
~
3L go ) % (< 38- my, )
.3 3f t n - 7; g
g &a ;g n x.-
; / h,
.7e
- 1m :. g/
, /
a* -
.w ~
- lggl -
; - ,z
'i/T g
s (- ' s m;3
,o
'? y 2 0: s4* %J/,) ,i f c c3; > L M, )g v'
;"3 - -
- x. .) .
~ .O l .g> > L-t s_y g/d ( \..
\ ) , . q' O g : e
)gy / /,'
J ne. .- 3 c~ m 2)'. (
'o=
y
',,,,, r c
.~
_ ' F./-'\ - 3 [n&. ' l,_/, [ o ' /' -p EO: L ([ c, ('
,i cj ql, \
lf Off?
.. .o.
g
/ .
m a YY s . r . O.
.?
co n s
<m *A ,,
m JOl.o ,.e 9
.O e ,h
^
- O[ Y
. ^
h - - ( 9 NS' -i s,.e
- u. c v, -- .
so , + ,,
- w. .
^ N '
l' y < _~. , my se um r
' l /!. , 2:% K ,; #".-. ;,'
- b
.. 4: ,.
Lfj[ ' ; %, L ' < :M ' r #3 , -
<gs3,p p,' ,
,"' 09 " -
;, ,iT ,%;%' ;.G i-1 t O H:d ' i,l4l:y :h 'M NC ' 7 Gv0 4 ., ', , ' 4 E -- -
CNWRA Ground Magnetic Surveys Y\ 4100000 d y 'MY (' Y \~ 14( / [*M .jM. p\,1 i lp/ -
. w - .-
Northern Cone % -- _ i t A-e . .- .( f A' urveys
#$v(,! % }ld:,
WNU\[b$~zl'S t ji [.jf( I Southern Crater e u JProposed Repository Sitey j _ 4075000 - Q , . ., ~' y (. '
.h 4 Amargosa '
b - Valley Anomaly A { .
/
4
? .T. , /
4- fu" ,c i- Northern Cone L. N %' 1 t, ,, ,,y .9 u@. 'n Southern s4 Crater Flat \' ';p!k, athrop Wells Cone ,/.1 W- ' f ..- M C A o
, 7 4050000 -
Amargosa Anomaly A - o -
.. . i.
525000 550000 575000 0 25 km 7
!!!- ! ; !!!!;I: l!'!il l jf;l , I tl! I i O
s c
. te pi t 8
e ail d - nrha ots ss n Cu na a obSe - nl i l i - r t n o e Yla e n - s heee a ts rr vrwcl t - y etl u - e ou heo NStbVf a v r . u o o o 0 0 s 0 o s S s o s 1 5 1 0 2 2 2 5 2 - i c - & t e n b ]o ,f i; ^< i, g l - a ,, 0 0 - O M / L 0 2 4 S - d
- m k
- n 2 - u t' o 4 . 4 r G - A 4 R - f . W - N - C N(' lt 0 j q, p?: 47 I y' j ; 0 o 0 0 0 o 0 0 0 0 0 1 8 m 8 9 7 8 7 0 0 0 0 4 4 4 4 O j
O O O - CNWRA Ground Magnetic Surveys 4056000 nT
-80
-10 0 4055000 -
-12 0
-"o Ground Magnetic Survey of
- E Amargosa Anomaly A ,
200 ! 4054000 -
- " Reveals three distinct
-aeo reversely polarized
- anomalies
-320 4053000 -
* -s e Interpreted as buried basaltic volcano edifices 400
*A j8 Further evidence of the 405 m o importance of NE trending
.e9
;4eo alignments 4051000 -
Y Y Y i
, P"" , ,
544000 545000 546000 547000
O O O - CNWRA Ground Magnetic Surveys Southem Crater Flat: . l Three magnetic anomal.ies identified and mapped in 4o r**
~ southern Crater Flat 4069sm Little Cones comparable in
, volume to Red and Black 4069000 w Cones E Possible buried volcano at
=
150 to 200 m
=
5 Dike trending parallel to
= Bare Mountain Fault
.oeism
= i a
4067000 - E
*066s00 ,
o reo eoe eco soo soo 1t ss4ooo s54500 s3smo sme s36mo $3ssm L - - - - - - _ - - - - - - - - _ _ _ - - - - - - - - - _ - - - - - - _ - _ __ _ ________---------_--_____--_________-____3
O . m ,4 . . . , 0. . O -
! ; i 4'
SUMMARY
OF GEOLOGIC SETTING OF ! VOLCANISM i
! PROBABILITY MODELS OF VOLCANISM SHOULD ACCOUNT FOR
- THE FOLLOWING FEATURES: l i
l r l : l
; CLUSTERED VOLCANISM IN CRATER FLAT l l ASSOCIATION OF VOLCANOES AND FAULTS l ;
i PREVALENCE OF NE-TRENDING VENT ALIGNMENTS : l j LOW AND PERSISTENT RECURRENCE RATE ;
! l 1
I i
! I i r i
i i 1
,l i i
a CNWRA 11 i
r 1n
- fs
- 1d Estimate Estimate of structures Estimate of dike nonhomogenous that enhance magma geometries that can model ascent, weighted by w intersect repository
- 17 % P:Vo canic Disrua: ion ~ _
Estimate of the : regional recurrence ' rate of volcanism r i 12 [
.;.; m - 4 ..-,--h ,--_...A.4,,a _.- m.m J u 4 A Ks4-p>a
..sa-p,.6i4-wi+26 -_---4.M 9 .5,&6e__4
- 4 Ja_+-sLM_A*A4 -4A .A a --42 An y egw gmag,dnn n. eqvwy gggnamewypsam_pWpsyn_gsspaq '
m$hi.-jfg,ij ,$ $j m w s ig_ggg_g,id_%w;)jj a,M3 j;p J: j g gu Qt& ( M16k, r e e w g w; ww,de& m4 g M pA ig
%g g @n Q%5 wam wa WW w nat n
sg g
#w ut4-igy ejJ va3 u v'x ?
n s e fh%ge%e-a e Nys;[;+4M^ hhge];nfpm a7 . E$ hy bb N pg{,4 m snkn[s,a l 5h h b y b ege n$p g+pkh pm mpmy enn sa w 4 L I h l Vk W_m m e1 % a, m w i e> 'n m ) g n.x~ x -~-. unn pg. :. 1a w . a ,mv.a,cq
, .a
$.3 +Q w ~e w ~
+
8.;y fy c - _ . 2.-
. b b ,: 'N -
Y ? m$ %h : MW es%u %g msgy& 0 .< $q MessF 4 1 /
- WERM$j E isi ee m %.
.t ; & v- -
x ee . . . _ o!@uwpawss 3 g Ik k kkk hf *'- e6
~
pt . 1%
%g kphYe' e,S ~ $:$ $- '
2 -
.i d.8: pl$psegp4jh8hidUMh0kE d ,
tk k $$?%wNerp$D.7 Yh ' j
%U ,
kshkh .# .mk , hfkh! m_..~ . nm-~!id!Ek, ~.r m hhhha, . wu-kwkg - f -l dW i . N i ,,, -' -g g m, .(a r_
-( .i,'
y c.
! [tre
- a i ,
L - c" 3 NM Kal l fji 6 a l l l;% l lC L 6
~
i, f t y yg ' g4.. py ph]"' s vh n y, i 4 ! n w y .g p.7 S E; e L .m f b N g g;hk.khk% j k 1
; j ji i y 7 qwMWedq'
? 7p ;
e u.p&n s n n - u,e 3 1 >
~
mps n m u.+ a.v mgs s kb x >u m:ew~ Mp
- u
$"h i _
d; 8 Jg;g@ee$3ss+M@mhjygd?$0fs. m ,2M
- wry , w nype,gw .. w,;m .
y:, ,i yA. oy.~mw yy m.n.y < v .> .w mep m; p.v,*p '~ m,:: m.w n. ; - ~. , . .
. u n a ,.
, s a ., w> u -. w ,
. .,, t > w.,
, . + - m; -
$~ y h l'yU [ A}%,. 4. 1:
V'q{i . F J.. 5.p d 4 ',y ', G y y c .W-l. . = i- . . g( g ( ,' y j., - +< ., ~ : .. ;3 t,y,> g- . 1..
. -1.,a c 17,f,
.. ' 7.7 v , r m ; ., u. qq q~ ;% <g. n 9
,s.7(
'9 g.o .,s.W+s
>e ' s . 4 #-. s
~+ , .i.
[m[~ t 9 > a. u
- 7 s z y. -f ^) ~
h> lt Rf r ' T G d ~Q ,onn s c
\ ' ' \ A g .:
tu-W i, ". - ,i n' i N g *
< ;,i - ' ,e #
^
i[ , ,l
;w
\tj ' :p
.s >
~
- v. n, . v.
\ 'j- C' . llY}lv .(: ? l .I Y ' e /
l 3..w ,
.j..t.
, . _ s b ? , > h, s {.? i
.v., ._. , p_ ,I.. f
. , 6 e
,. * ;e t ,,
s , A
- 9
'g
< ,s .4..
6,. , .,b., i . , , y c' ., a
*" ~ i l 4 i $ $ S k j & 'l */. * ' -e 11g g l A , i ,-lh '(, 5 l f'
. . .' ' o .e *5 se
.tNr r-
- u. ; 1 t r' rs o > -
4
. / . W'i ; g . 1 f a f
j.-
- hi u w.
, j J '[f .y @.d,-.~..
re (..s.,g.: .m nme -mu i f.'.}d{ . ~l ' , s ; 1,.,.',dy
, n* ,, ws. .
Lh4
,e 1 ,p A
, t :, tc.;
,y,,,
,H
..[, x: , ? f,
a , - r * > J,,'
} l.' . , s .e
' .o . 3, ie
.. . ,. . _ ~ < c. .r ,
x < . , , . . n .. .. ,
, j* 4 S , ,'4 f
'] g 'i ~ ' g '
. .,# hla J j 2 [ f .; N 4I h ,, ,g - * , g Ig .
.J' +
s f ::9 7 A,d Wl FtW " r M. .',1..' p y s + t -
.,'g 4 + ' '
%y / q"e, u us- ' Agi ' +ux
_ .z s ' ',~ .A+ , , Wec-. ,DP' *
, s 4
(- : oM.
~ w a, i.r ,qw # . f. i t 2, ~,l' at? : . - . s 5
.t ig C$;
4, > +o, u- i
%, c* y,. 4* :.i,. f ,. 1
,,=.6 y dd.t . s t _ . .- #<
7*
.i > \
i a.3 3 ). _ ,3 6,c..-..f.a
.e ' i ,,1 l L* . y a s, k 4S ~, . , .,<1 %.,,
e C D, a Sc. lqy
+ t er- m :,. ,. - . ,
,g j + d s , .-< (p.
; 4 't ' 'g .r
,,, . ,. [4 v .f s( .3' 'y< c .;.e,4 f- .
(j i L G 4a'Q. ,f. A f :fg e ; ,,. n.y } *oj i . , g. Q. - _.. 4, l , , , , _ f 3, )a , j
%gm... . / 14 , . #, .;
ne s r*
. , *g r .
1 7,. * '- ' $ pgi .. t M\ 1.q .a r y, gc7
; . , , , ,, 5 p iggiJg; - +g,g
^~
, . ;* ! .. g , p! . '., , ,
- i = , y;t ~,c y%m . R .4 L .r . . , ;q f -1.+ .rT ;;- - ^ W, q , 4 ,,1 ,
A. s, ,- . l b k, f " , O "' SC"W k! h . ' > >' f, % w 1 ^$ " " ~
'l C! ,,1 %. [L Q } y ,u 4 p T f Q T t G y y m @ j g %v ; %a 4 yp% ?
pg/pMy4y;9 Ly o~ e,fQ+.y;Qf a :,. }g
;m m%r.y o, gggWn4
(
,y.. ' y, 9, . t: e y a J. p.
~ ,
,Qs nct &m;. w dA\ qw::uq -
m y.,.74 >.
,r.-~ c m. 7. G<
. qr Q a% s w ,.3
, r x r; g ty x y* yn jw m y;y ,Y g;< p ,
' i s ;& y q , . + w:.y.
- e. p ,a,y ig. ;
u . .
,a a : .
s~ x :. .
. ' n l, m. . .
14E. p) g- ,y M d31
- m. m,, ik ' ,M - s , t_ '9 3h * % 34 , ,,w '
- ,, N. n,,,, &" Qus 1-n
'w %.j %;z-s .p 9 %;%l..), y r , M C.J' s w @e a.>- fryMw Q,n c e p,$) s -
N L LLE.C, h M M m.a.D)j., 8 M 3.L M bdd y j;1k/MhdhN$h Aj - f ' wy [j, _ ,,, . , m-/f: .] m w m ww w n uw m a. w~y, ; a:~gmwonA. mown m~n . ~.,y g. .,w.~n, a. ym.,n(&, ~n t , w .nv.;s v s.w. c p. c, ,, yw w r e m ,,o~p,9... * ;4 o. ~ y, r , u,
- - . ..h.,a r. .w . .~m,g m ym wn
. , , . m. .L., ,a
, r o
c
. o . .gm . m,q w,...~ u. s .mmo. m:.m&m
' m -. Lp ,9
; u: <. .rp m- ,
. sh, ,
[o t,e-
~ , v 'sp v ,1:,;-6n i d ~ ~ yn Q.v l. >< ?s ~,% f ,.o .<m, . - ,. ,' s, o., -sQyl!*y' ' /. 3, cgy,f;&.,
l,- p b_ y i ,, . m;3#. p
.. A. ,~ +.;,,m,,*. t ;* .j e s ;, Q. 1
) N1.;
gr 3y
' , ,p; +a d(, _..
y us* v. , &. w%#' .,6 ,. , g.J /n ,t > " ,4
-J.2m <%.Nm.
4 # ,, , -
. .= - .: (
- y. * <
-t -.F'F,. ,t' r 9A O ._,l>t>
. ry . , ,/ < 3 ) s
. m. ' ' . . -
ge a.,g x3 .g :yd,m"J' . * % 'gu,7.A, 3 . w". s ; y' Mm- y t' y +{.+ g t .n U g: fe p r W e : y~3 ,- s t y_ pm. pa: - - , a - , , s;, rk* ;g, +
,y 3: n q- g%.n % . .a mg.q i
( ,. c. t n , nwm ,, e. - yt
, y s
r ?m(m, ~,s,.3 . .N ~ * ,. a%1 , g t M... hq. v. .1 py, e, y (,s
%d- -q 3 * ! ;_4, q '.q e. A o ,s,n j wj .y ..
4
. n . ;
f.. ., ,. 4 g; y a ay s
,e 9; 4, .
'a.
a.m <. g 4 ,k%
+ma L. yu, nr . w, n -
- r. .r , m,..
g %m;.p y y 7m$ ps cm o,,w , . o, , , , ,w u- ,...,a, (t.x.a,g
. ;i n
w .g ', r.2, v m.,g.. g 8 ,4 r. ;r.e.:-v w,, rWp a ... e
- ,- > ., n , v:. w , c ac n y .
- a. ,
- p. as . s n, f r as ..
.p.. y.; o .,
[-.
, , , . .-m f4 r_~.----<m 4o ,,
> ~,.._.sg.
v
. .a e , .
,2.; . .
a ,w w e:r 'uyCgyc f G i m %n k,q .~ W a y -
! g
.u. .
Wg.
- 1, c - u ; g ~ , . ,91 a< .:)r :y *;~?~L' A ir '
" .-' s' 4- > .
1 - A,c : v- -- >.. "w" ~, D. n
- f. jI j. eS3 Q [( R-r ' ' , A tf ( 3} a s t. , . : .
..,3 3; ,
+,d u;': _ f.!I,}< y ,j;.y@) o9 ';:g %, Jf M[e~ j kO Nf j , - Y. fp 91 m" . ;-.1aj,,- m,2 ~ +; i,' &m, .a N% e%
e n&q. : 2
, y a o v. !
n y s i m-s r a t y +fc y.o$g'
. : y -n" e : s . m u y e ,;g
.. a n Vh9 w @s - #.
\sr.mi. n, <g y ,
p;,ya m_.@b;.m%ay%. - re ,, ye - g:r. ,e ( e . : v; , n, ;&. ,
>^
.; . u c, .,
?
.\
.., N';, ; n m w .3 g.p 1,, y; . n a .': r I w u.%,W.d
. .i .
if v.w r. . :( c@ . -n . o ,,W&mw vagg- e
~ym m ?m ,..a .
1 4
- d. gg;.,Oy@v .. c..
a'J-w . l h h: . 1 lS {k} J s ![.k I ak ;. . )N8M y % [';su Nja ' m. n w'M x % *a m : % q::.p c4
).
I 3-u\ 'h, $4 [ N.. , . j. i n uwq" . 1 a %+. . . A.m v ,' -m s -y 7 c ,y> o My np - %o- V o c- n ;a m q. . s I 2 . , , m ,rav (p# ' 17 @ h;m.m. eMMM-M;*dJ ' q! & ' . , > y ~ : q . , e 1 N ,L[ h,p}'. ytj.) *h[ pWD
'1 *>
_- h^mnsk u.m%. ~A. 'Q.. m--( . 45TW s ..../e ,
- v*m L* ~ s w /? /-
d W^ > - M Q+ M ," 1 1' s i f.? ? j 4hc%W fv , .w-. A,;b.
.$ ?' ?_'
&j-T5u, '
G.
. )'t " 'f'l f ' l v
, pA; awlQ
.Ay ye- he
- w. _ . % s ,4 n .s3 bc < wvL,e.< % o 4 :
y; .. - t ,a, .qo b, w e- .. , 1 mm , r~cY ? b ,.co.;' .,h
- s. - , . <YN m
$gA.) 2. N.S %m % wk. D mA Y% W"- *~' ' '
QQ I bUa eeu u w.% V . V v.% *h h* o 4 .e
/= L ' J K<
4 ? l. s A. .h'is.; me y .q,s i $ [W ,>M JM.QyfKygE Q 4 mC,.gQ i
> a .. o : !
F j,jyb%'.M. ' h[.,d",.YG. k., , , VN# s,
~
,d p. M.L.gtj,d
*e 3,1.l s# * ,~.n 4--rK W( .
,,'i y 4
- s'"
mm ML .;b ^T .n. m% * - ,7 ' ) ~ .., .
"W ,, e , p .
T, a -1.w mc a s '. (y s - a
#'I k.}
?, t ;'
).
rU r 95 h p a&n b. d g (hh ci ps t,4ey
'.b . (y' h - I wn 79[ y T I, ' - b
- r;. Ube'W. - >.
^9 yq F q, i> n y es n p gm ,4j 4 ,
p: v,
, p. ,o d4 q ug- -
,eno .gmp gw" tA.s.
w g-gng , m 4 pa W
, ~
' ,', 4 yW _ E, . .j n-; J,
~
W; y. pm . RqvQ N.vwPu -. mS.-m$ : .- 7 g%u n.m AR .en:r W, % ~ x% ' ',v- k., h kf" h. j k k ,S w M, n p ibkCdh k hj r s u hhhf % J .A. . YND C .. @s F M., Ms,r, W M~, d)
,> ,J R..E+sQ3
, Gv: ,;
- ,s M , . %.n~mm s m, 2 ,a. ~. :: p_., --m..
m
~ ,r , .m_s
- a. u s>
;n .
~ c 1
a s. .
,s s .,
I
.,,W.<...e-
.n
, l I C + h i 'J. yt%.'ci , b~ ~ d I - pA-1 As q- as-
^
# 3 .
v 4 s ^'r- . ij t-m 3
.c i i
- s;4* yrs c, -
g g ,tW y ' , y' a t 3 u d:c 4 *o W (9 * - L g' < r ,,} pd 4 ;y; 3 i S /.. d ,
. , ~' /e.,t y e , s p ,., T+
}3 . , e f :,4 it' g>g*sl. /,,
g.. . ;;a , .J
--%J, . ., e .- N .o .m.%~ .s
;+37,e, m :q g.1 . c. c s.-- v 41 i.
p 4 y == y . s e,- -c >: ,- -y -
< 4 - - 1 9 L. .c..,a s .: .,7 . , -dl gY , , 4 .3 i 5' .<MM! 4 j %< d , J ..r .3,4 r@ N ,3
. , . , g = a %.
1 e.-ra ;.< - 7 e es L * "
. ,r .+& ,-
3'
.u -t.s" m a 1..,
- q. s>,s ML
- ,' vnn-, i. ;3r.
.. s i.
. s.e .. ,. f i - ;g, )i. ,., ., _%.'.e$ J , y> a^ -* {.--v.,,r~,,f.,~,.<.,a, .
q
. > p- ,
, ,,g .' 3. . a,.. 7w3.-, ,h,;,, %.4 m .
, i
>e.v, e,- :.
i
- wr. ,1s, a :n , n. .
.r,.~ , < 5...,;.-
n, ,q%, ay..kf^ .. , v , < < .>r ,p : z e
. e ,< t.
y
.t- 6.n} . #t . 3.y w -*. ea. ; . .W-- g *m' p,,,.b a
) ,:
i , .- I
).m g t9 $, , A i r t
' ,. ,} fL' , ,'?
"', ;' , .'~, ,.b
'.al
,' , ' "l,mf f .i . [.[
^ '
i l1 :
, .;.6' f
~
)~ u,E f. .4Q ',4 ,',d',v. ,
s
.A..' .* f,' n, ' i...Q.* , l :'I 4 .!, <k, W9,'. . . .1 _ . ,. __,
w , dy 1 ( ,\y wh, g ce , +'t^' g
.'t . - - - --- W + N'G ~:'s ' <r v : ; .,^,g ' - a !. .; Q,. q(;O, 3) ; ;' j '
.g a9.5
.s q, ,L, ~ . . . w; u .
mm
?,
m ymn m yvmm t
,,. .', y, 3; . ,
4n.Qp m e H. ,TL y r -t.~ qp.emmmmmvm, m VWJ d: q 6.s W llQY y)a w! w n a , o m " '+ nVf py<m'. f w.pp&w g g QQ w s &. n,jd:J,y;%yy'ym : g 6 ^ /,, .
- t. g. s. v< t.y c / 44 m '
y g. . .gmy g_ danaCg ;:4nn 1 g;s d y+ a rrpe p qm 6 ; % o v .. ~ ~s e;irJmm ,p.% w +6-s-w'% s w+n+ g :.n e++W.e r .,b A x*m a' s, -
,e s m - v s ,nr- www n v<
. ,pQn.s. W e :r .+ew' + .
L;. n,:m %. - . n. , " e.
~-
, . 6 . n.* . M.
s.w* . 4.
- ; m_.;
v..nd wei
> ~ H.
.r31 %g<
4e.gwt
. ,F t r: M,..q yaw.
-1 . d4g v.w' -s,,#,p , e,%y 3 yt wp i w.u,r. . .,y.
- 3,M, y33.
g u.4 3 ,mpt a . p A de., wu. . ae m..r .;a r. s.a ,s MAcgM. p
- 5 < s , o r : ,- L ~ 0~ +;
3
- g. ,.
i
,l',a .
n, .+; s<m, .- m > .w. .; -y, wn 7. - ,, q W. xy.,t .v . ,r,,p a v. ;q,4.
- s. +., v,
- - n.e, m ;.n ,- re, -.
, +.wp Am , g .s .,
6
>- 't ; M. 4 ' .i
- w. .- >
>',)rs -1
- ,,,- t .y.N:b,&. o5H \ Ma , s7 %. 4 c.,7e.P. ; p~ t'."Vv'csn., ,n. ,/[Lt'Q' i, a wp u.csf}' ':'* t, .w ,,
t ;n ,.
*r
; ) .m . .r. ,- , / ,f s
",u 1, . dWa l,,y.,
r' '
- 3l.- . ~ -Qn t LA%Ra
.) ybn.v.u lmA v) uQ'd' i-jta, t : : t ,, >r .c., y ' + , u I,Q&*,G f -' ,- ..j ;U 9' fe %'
.,, yl2 .o W9: i y , ,v,o ,l;hh'g D, %. : f7'w a rL , & %, y ,.
L
- R.
Q:' ) 4 , ', . ,e y/u% N< g?.i.4.ph' < f9<4?',e +k.%, ifl' , j.3o o' 3 m , . Ach & ( . y 4< y. ; y :w'E h .vm swJ?,.y ,. ) y ?c !/ + iL :<f l
.. $ t i H y p Q;.b g
- f dWi d s %A ;/pL" (w.di[m$d edi MMiji41,:f qwm.weg/p ? 'fig 4,6W, Y ii4ie o .
,' r-es-y.
3 x u p?,@l c y , tQ. f o ; Am,G , O. W> .18 L; v a. n . z.m[.,1,r dQ,MP +,
@% a'.f 'd sg;fi h,A b k {,@e N w m:4%.
. ~ ~
%m '.
- 5 O ,1 - -
^ . . '
4 4.d DN,[,j $ h.l? d's A'fh, g w* 'e .[ShNkk g p.u 2 t I E u s r [ m b;;'am u/gk h. Y b.d+u (e n. W N,IN, . 3 r
, .- " . u w .,-
; . 3 i wc s..aa,r+ f
.,n g e mL M: :p s*a,*n. ; 4 4. .;,. v : vr w .m 4 3,'t
~ >
3 - - u:^; - , y
. n:,d w *. a.c...
r p". 3 m 7p j 'g c ,,! r g. i r
.J.
y4-T n t.p-
} .nL s' ,/e. ". [ g s .. > ,r 4.G
.i p ~ .
- I l ., 4 +, "9 ".'*1...-
ni v' f< 3 s 3.,. ! I s i[M, f9. 4 h j i:, ' E W/m. h' y.. l. " .y .., v f ..V $ ; k',. }[ ' e n *(' d g~ ' '- 1. j~f. i< M @'.(,,i< e L s ,. 'TI 1
- r ^ o;hr'ny-s.< M C;Mlm
. m.
[e 3fNyW jd)4
/ral wb.' e 3- %
1
" w ' N M ,;0.~< P;a'W ,4sEi"[
* '?x JN' - '* i :: > < - - - 4
.,vs
,.A g .r r t
aj~ 3 y g m.m9g 3dwtA,<tu t e/ u .r.n,. r. x,,v va<r9nr-f ;qs ev..s - v n. %' a } ,t
- p f. , j , + Lap;now r>
s y ,.a , > 5
.x
%,/ -. y m a; 7 tes-
.A ;m.><',es.p.,> . p.,
y s, i> t 4 J ?1 i m, . ", - o
/
4.
,R,
.s
= . r s yf g ; g u:; . - , t -
- 3. L., O. .H .q. ) 5y p. . y 't y p,.,4p . e.L tt % a. . a. t,,+3..,..
, , t a g - .g4 - J ' v. W 4 e , u%y, L,i 3;=,.
) ;' , ., 6.%t"a.-
3
. a - y _ 4 4 4 w % b< -
w_
$h b[ \/ f- i gl_ NN 's.f, . ~ 3% E *
*N'< J ';
W t Q
.? L - b l , ?'p t
?,$'W? , *
']
'( i qs i V< c' l' iq . * '
'}b ,
.y p i h.;* jfl f *s ', yk 'lh '%
j.+ ", ,L g,, .,e' *R
)' ,, ,; g;
- c<g g js h a p4.
, ~ >
-t , .& ,t .
L e :: A' r .; * + ;b jd < u ['. s-
- '.+ ,43, .W, s4t a: ~:s %; ^'
r Ju <'y-s . p
,< > r!tu5 a.8 :: & M..M 7- aV c v a.
1-
>n
, ~ . ar>1 y,
k.4 $. i cdin.:2. h.,i L M 4.h M ,s hrJS.L M.i2MZa'.::d.A L - Adknio_m .A Lu .a L. -
idEGRATED VOLCANO-STbCTURE PROBABILITY O-N Gravity i i q 40 km l 4100 N_ 4080 l .< Y
, , s s p
i 4060M_ -
, c. _
525bN 550 N 575 m '- Faults
- i .
i i i 4 i Nonhomogeneous probability models (A) are convolved with the amplitude of the horizontal gravity-gradient (B), and the density of high-dilation tendency faults , (D) to make geologically dependent probability maps (C and E). i 1s l
O O o _
@ 0.015 ,,,, ,,,, -,,,,i,iii;iiii cQ T*
W Il - gy - ed -
~
5D Epanechnikov model (h=20 km) eO - Xt
-m near-neighbor model(m=8) _
0.01 _ Oe O. -
/ - >E ~
5W
~
.c
.O w m%
i
.O O _
i O E 0.005 i iiiiiiiiiiiiiiliiiiIiiii 5 0 1 2 3 4 Weight of Gravity-Gradient Structural model (w) [ [ 16 .
O O.02 O O. .
- iiiiiiiiiI'I 'i'- .
C - ~ : -
.9 e
0.018 T or : :
$ 11 0.016 uf q) :
E $ 0.014
-w _
e0 x .t 0.012 K - v) near-neighbor, m=8 T Oo :
- \ :
oh 0.01 N c Epanechnikov, b = 20 km :
>E . _
.t g> 0.008 .
/-
- / T
.O .C.+
- i y :
(U w 0.006 -
= _-
.O O = = = i
= -
O i
" 'i,IiiiiI'
O. 0.004 ,, 0 i 2 3 4 5 ' Weight of the high-dilation tendency fault model (W) 17 '
OcC 20 O ..
"A o
g - _ u) _ _
$mo 15 __
c v- - - g _ _
@ ~ ~
.x
.- w> _ _
QO 10 '
\
% .T OM 1 O f near-neighbor, m= 8 -
x O-
+-e IL -
Epanechnikov, h = 20 km 3 (I"
~~
@@ /
gg - O* uw Q. o - 5 5r 5 5 5 - 5 0 g _ M 0 iiiiliiiiiiiiiliiii iiii 5 0 1 2 3 4 5 y Weight of the high-dilation tendency fault model (w) t 18 e.
t CONCLUSIONS n
- Plio-Quaternary basaltic volcanoes near YM correlate well with high-dilation tendency faults ;
and regionally with vertical displacements in i basement, indicated by gravity gradients i
- Including structural control in volcanic hazard i
models increases the probability of volcanic disruption of the proposed repository compared to models that do not include structure l a
- These models indicate a range of probability of volcanic disruption of the repository between 1 x 10-7 and 1 x 10-8 /yr ,
l
~
19
O O
- O
;~
YUCCA : MOUNTAIN PROJECT A ^- E-_.._ _a Igneous Activity Program i i
~
Introduction P Presented to: 91st ACNW Meeting i Presented by: ; Tim Sullivan , Viability Assessment Team Leader n i Yucca Mountain Site Characterization Ollice r U S. Department ofEnergy April 22,1997 mrice on .wi an naa,oactive Waste Management m_or , I
- i VOLCANISM PROGRAM
- STATUS .
Volcanism Status Report was issued in 1995 Data analysis was completed in 1996 and will be reported in the Volcanism Synthesis Report, currently in final revision Site Description (PISA) will provide an integrated I discussion of our understanding of the regional and site Geology including all information relevant to igneous activity i
~ er l
\ - __ - _ _ _ - _ . . _ _ _ _ _ __ _-- - _ - _ _ - -
o o O
- i VOLCANISM PROGRAM STATUS .
\
DOE has decided to close out site characterization data collection for volcanism based on
- low disruption probability results from PVHA l
- insensitivity of PVHA results to new data
- DOE performance assessment results are not sensitive to direct volcanic disruption of the repository because of the low annual probability
- modeling of the direct and indirect effects of eruption and intrusion to date have indicated little effect on site performance armed 97
\! ,lt 'tI;!ti ;'f:iI[iiIt ; ii !Ij
!ll1
_ t _ M c u de n nf _ R A o w cer dr o to t r e a du ea sp l G n sn a a nei nobaa s n O t i o os isd se n e i d e RS t t adp y me o n b PU l i _ aa r rdle nl p c _ T foeb n ei s a s e in e er MA wse n c n a d e ST I e af e e n nsd uo b l NS e i l _ qt i te sd ea w ayn sm _ A l u a u oo ans l o cf nr s n an _ C va el al i n l po i L a n a t l l ni nleAAa - _ O ob VVt n i witoa t iaAAe t V EinildiPPs O di a e _ d advSSr DamAaTTp
O O O VOLCANIC HAZARD ,
' ANALYSIS
= \
DOE has completed evaluation of the probability , I of disruption of the repository and quantified the associated uncertainties based on elicitation of 10 subject matter experts in volcanism
- final report completed in June 1996 t
- intended to provide a sound defensible basis for licensing l -
PVHA results
- mean disruption probability -- 1.5 x 104 l
, - 90% confidence interval -- 5.4 x 1 -10 to 4.9 x 104
- bounds -- 10-10 to 10-7 i
, i f
i
1 IGNEOUS ACTIVITY l 1 TECHNICAL EXCHANGE I NRC presented data and analyses conducted after the completion of the PVHA DOE agreed to evaluate this new data through l hazard sensitivity studies i 1 Preliminary results will be presented today l 3
IGNEOUS ACTIVITY TECHNICAL EXCHANGE ' NRC presented basis for concluding probability of future volcanic events ranges between 104 and 10-7 This range differs from DOE's PVHA result but is included within the bounds of the full probability distribution l It remains DOE's position that PVHA provides a defensible basis for characterizing the disruption t probability i
)
i t {
r: IGNEOUS ACTIVITY TECHNICAL EXCHANGE DOE will describe how this probability distribution will be used in performance assessment
~
This should support closure of the probability subissue i i 1
- .7 _
lGNEOUS ACTIVITY TECHNICAL EXCHANGE - NRC presented dose calculations for a volcanic eruption through the repository with associated tephra and radionuclide dispersion. = Risk result indicated an average annual individual dose of 0.5 mrem /yr DOE has also concluded risk of volcanism is not significant DOE /NRC results are converging--this should lead to closure of consequence subissue acwn4_97
O O O' - PROBABILISTIC VO.LCANIC HAZARD ANALYSIS FOR YUCCA MOUNTAIN, NEVADA , i Kevin J. Coppersmith Geomatrix Consultants San Francisco, CA 91st ACNW Meeting April 22,1997 ; Rockville, MD I' t I t
?
=.. ~
O
~
.O O~
t PROBABILISTIC VOLCANIC HAZARD ANALYSIS (PVHA) FOR YUCCA MOUNTAIN, NEVADA OBJECTIVES OF THE STUDY:
- 1. To assess the probability of disruption by a volcanic event of the proposed repository i
- 2. To quantify the uncertainties associated with this assessment !
Disruption: the physical intersection of magma with the potential repository volume Volcanic crent: both eruptive and intrusive features ; Probuhility: annual frequency D i e t 6 F
- _ = - . _ - _ . - - - _ _ _ _ _ - - _ - _ - - _ _ - - _ - _ _ - _ _ _ _ _ - _ _ _ - _ _ _ _ - _ _ _ - - - - - . - . - - - - - - . - _ _ - - - - - - _ _ _ . _ _ . -
- _. ~
O 9 O~ '! ASPECTS OF PVHA STUDY . Quantification of uncertainty was a centraffocus ofthe PVHA study. hnportant aspects ofthe i PVHA uncertainty churucterization are thefollowing:
- Attempting to capture allimportant elements of uncertainty, including expert-to-expert ;
diversity ofinterpretation
- All of the experts were exposed to all pertinent data, researchers, and methods l
- Ultimately, the reliance that an expert places on certain data, methods, experience is their prerogative !
- Experts encouraged to consider full range of methods, data, analogue experience in ;
addressing issues l
- Focus is on uncertainty not " preferred" estimates; when in doubt, don't choose, weight the i alternatives 1
- Formal process of expert clicitation followed 1
i I i 6
. - - . . - . _ _ _ _ . . - . - _ - _ _ _ . . _ _ - - _ _ - - - - - _ _ _ _ _ _ _ _ _ _ - - . _ _ _ . _ _ _ _ - - _ _ . - - - - - - - _ - _ _ - .,-,--.w-, c ---- - - - - - - _ -
~
O O O~ PROCEDURAL GUIDANCE ON EXPERT ELICITATION The PVHA was conductedin accordance with allrecentguidancefor expert elicitation , studies, including: ' NRC Branch Technical Position on the Use of Expert Elicitation in the High-Level Radioactive Waste Program: Kotra, J.P., Lee, M.P., Eisenberg, N.A., and DeWispelare, A.R.,1996, U.S. Nuclear Regulatory Commission, NUREG-1563,35p. DOE Principals and Guidelines for Formal Use of Expert Judgment by the Yucca Mountain Site Characterization Project: Department of Energy, Ollice of Civilia:i Radioactive Waste Management, Yucca Mountain Site Characterization Office, May, 1995,10p. DOE, NRC, EPRI Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on Uncertainty and Use of Experts: Senior Seismic Hazard Analysis Committee (SSHAC), U.S. Nuclear Regulatory Commission NUREG/CR-6372,170p. ,
?
O O O~~~ MEMBERS OF EXPERT PANEL PVHA PROJECT EXPERT AFFILIATION Dr. Richard W. Carlson Carnegie Institution of Washington Dr. Bruce M. Crowe Los Alamos National Laboratory Dr. Wendell A. Duffield U.S. Geological Survey Dr. Richard V. Fisher Univ. California, Santa Barbara (Emeritus) . Dr. William R. Ilackett WRH Associates Dr. Mel A. Kuntz U.S. Geological Survey ' Dr. Alexander R. McBimey University of Oregon (Emeritus) Dr. Michael F. Sheridan State University of New York, BufTalo Dr. George A. Thompson Stanford University Dr. George P.L. Walker University of Hawaii I
O O O' SPECIALISTS (NOT ON THE EXPERT PANEL) INVOLVED IN PVHA WORKSHOPS AND FIELD TRIPS , r . Duane Champion USGS . Peter Morris ADA . Chuck Connor CNWRA . Paul Orkild USGS
- Allin Cornell Stanford . Frank Perry LANL
- Paul Delaney USGS
- Gene Smith UNLV
= Jim Fatilds U. orlowa e Richard Smith INEL
- Robert Fleck USGS . Jack Stewart USGS
. Chris Fridrich USGS . Brent Turrin USGS
- John Geissman UNM e Steve Wells UCR
- Brittain Hill CNWRA . John Wesling Geomatrix
. C.-I-l. Ho UNLV e Gene Yogodzinski UNLV . Bruce Judd SDG , . Vicky Langenheim USGS . Les McFadden UNM e Chris Menges USGS
- Scott Minor USGS
O O O~. WORKSHOPS AND ACTIVITIES PVHA Project ACTIVITY TOPIC / FOCUS DATE Workshop #1 Data Needs February 1995 Field Trip #1 Crater Flat kkirch 1995 Workshop #2 Alternative llazard Models March 1995 Field Trip #2 Sleeping Butte /Lathrop Wells April 1995 Workshop #3 Alternative Interpretations May 1995 Elicitations Individual Interviews June-July 1995 , Workshop #4 Feedback ofInterpretations December 1995 i L
/ \ /
s
%/ )
s I \ , i
%_/ %J CH
'n 8E C :r e a v2=g 3 n m, a s ES N%
a a Z$
- c. o ._ --
ho a =. O ". Etent Et ertt yg W'*" 9 Zonattan Zonation 7 spattal Q f* Length A :trrtuth gg04,g p,77,4 _ os riadet Afodel Boundanes o*- s$ Dtst Dist m te-r st ag w FJ 85' na c e=
-m 3 7 4 Homogen = ,
D e G o~ m M C
' Tc M n O3 U 3, 0,s* ' C_
6, 5 3 I Opt.en 8-7 1 oc m V ' MO Pos? ' Mc < 3 N1CE 2" 9'C E oC 2 20ee$ O C 3 Opt on g 2 5' Z n {},$9 2 Nonh0mogen - Post 5 ua j m g On o ,' On g Optiom i e C 5" "" N30E , Post 10 Ma c Option m
, n
' O w
D st 3 M Kereel smooses R E' Figure 3-15 General logic tree structure used to construct PVHA computation model.
n j"' \ f O& 3C RE c
- EE
- v. =
3-n - g 3 3h o y Z$ C 3a E h 5' n CE $ I' 3, 9 8 o-e< ses 80 OE f m' 5" "" "" "'
$ Age LW NwCF SECF 4i SB TM BM freid Otw -
~" **
$ Cafa Counts Cmss Cern a C unn Counts Cants Cesnts Para ner Counts g 7 ,
E3 Ya
-n 3Mn u>
N on C 88, 3
- c ..
" o, w
@ i see i i g -<
3 < % Se' 1 h i t 3 2 3 g i 1 set 2 i0 g abeup, CovMs m Source 1.0 J lI14 2 3 4 2 Set 3 2 12 O C 3 Z swe. 2 \2 20 n g a.
=
. se,2 n2 3 i s s s 0.5 Source B Grodao' RC'e
- 0. t se, s w
oo SL Figure 3-16 Logic tree structure for subtrees addressing the uncertainty in modeling the hazard from snecific sources. These subtrees are attached to the overall logic tree shown on Figure 3-15.
I
Title:
Probabilistic Volcanic Ilazard Analysis for Yucca Mountain, Nevada f Document No : BA0000000-01717 2200-00082 Rev. O Page: 3-114 of I15 j I Spatial Models l 1 i 08 O6 'I 0.4 - - f
- 0. - -
g-Uniform Zonation Kernel Smoothing Field Shape Temporal Models 0.8 - - 1 0.6 t 04- - O Homogeneous Nonhomogeneous i I Time Periods of Interest ; I ! 0.8 - t 0.6 I 0.4 - . _ = _ ! 0.2 - 8
# 0 I j siMa 52 Ma s4 Ma 55 Ma 59 Ma s10 Ma stiMa n
O f E Hidden Events Factor { 1 i 08-i 06 ' 0.4 - l 0.2 ) [ ; O r 1 1.1 1 2 1.21S 2 5 i Dike Orientations ! 1 I 0.8 ; j 0.6 ! 0.4 : 0.2 - 0 ~j l l N20W NISW NSE N10E N2SE N30E N40E N45E f l i Maximum Dike Lengths l 0.8 ; 0.6 , 0.4 - [ 0.2 5 0 5-10 km 11 16 km 17-22 km 23-28 km 29-34 km 35 40 km 4150 km >50 km f i I i l Figure 3-62 Summary of experts' assessments for components of the PVilA model. [ l I ! ! r l l 9i. t Civilian Radioactive Waste Management Sptem [' Manacement & Operating Contractor ,
,I
, t
. , . . . . - --. --.- . . , - . . - . . . - - - . ..- ..,_. ~ _ - , , - - - . . . . . . - . - . ~ . . - _ . - - - - _ .
1
Title:
Probabilistic Volcanic Ilazard Analysis for Yucca Mountain, Nevada ! Document No.: BA0000000-01717-2200-00082 Rev. O Page: 3-114 of 115 i ? Spatial Models I 1 0.8 - 06 g-1
- Uniform Zonaton Kemel Smoothing Field Shape 4
i Temporal Models l 0.84 1 i 0.64 1 0.4 Ok. m j Homogeneous Nonhomogeneous i 1 j Time Periods of Interest i 1 { 0.8 . 0.6 - l 04- = _ _ . t 0.2
- I O j y 51 Ma 52 Ma s4 Ma 55 Ma s9 Ma s10 Ma 511 Ma l S
! E Hidden Events Factor i 1 ! 0 8 -- 1 0.c I 0 4 -- { 0.22 1 1.1 1.2 1.2 1.5 2 5 1 l Dike Orientations ( 1 0.8 0.6 - 0.4 0.2 1 0' N20W N15W NSE N10E N25E N30E N40E N45E Maximum Dike Lengths 0.8 0.6 0.4 0.2 - ', O i 5-10 km 11-16 km 17-22 km 23-28 km 29-34 km 35-40 km 4150 km >SOkm Figure 3-62 Summary of experts' assessments for components of the PVilA model. O Cisilian Itadioactise Waste Management Splem Management & Operating Contractor
Title:
Probabilistic Volcanic liazard Analysis for Yucca Mountain, Nevada Document No.: BA0000000-01717-2200-00082 Rev. O Page: 4-45 of 52 1 1 I t 10 i i . , i i , , , , j { g _ A. McDirney M. Sheridan e d C 6 - 0 l, 10 .i
, i .i .i i i i e i 6 i l $g _ B,Crowe _ R. Corison
- 8
$6
~
! 3 4 -
.I 1 .. i . , i i, i l
i l 10 ., ., , l 8g _ G. Thompson .. R. Fisher _ ! 36 -
. .J .. ..' _.. t 1 i E?
,0 g _
G. Walker
. W. Du f field 3 6 -
=
i j 4 - , c j R2 - 4 . ,ii l,i l. O
; a i i , i A i i i
10 . q . m, 4 ., , , ., , ., , , l 88 M. Kunty _ _ W. Hockett _ l {6 t
- 4 -
i i i i i i 0 i 10-'2 10 ~ ' 11- 0 10-9 10-8 10-7 10 ~h 0 ~510 "2 10 - ' ' 10 - ' 10-'10-8 10-7 10-50-5 f l Annual Frs ;a< ncy of Intersectwn <1n nual Frequency of Intersect,on
- j. l'igure 4 3I Comparison of indn idual expert distributions for frequency of intersection i
e'
- Cisilian Itadioactise Waste Management System Management & Operating Contractor
Title:
Probabilistic Volcanic flazard Analysis for Yucca Mountain, Nevada l Document Noc 13A0000000-01717-2200-00082 Rev. O Page: 4-46 of 52 0 5 i i , (a) S t h *; 50tn'; Mean 95tni
$ individual Means '
3 individual Medians 1 F Y YY Y ({ l 23 - M i 1
.ce 2 -
O L. k ! 1 0 0 0) o w -
+ ++ Me t 0
' -^"- -
'f ' ' '
i i i i ,
^*
(b) I ? - BC ? -- ci ::? cw -? y MK $ - kMS -
;? -
N Q RC -
? -
RF -:? wD ^? - wH - (? - Aggregate -
' ? -
i i i i i i 1 0 -12 10 -' ' 10-' 10-8 10-8 10-7 10-6 10-5 Artriual Frequency of In tersection Figure 4-32 Aggregate results for frequency of intersecting the Yucca Mountain repository foot print by a volcanic event. (a) Aggregate distribution for frequency of intersection. (b) Individual and j aggregate nicans, medians, and 90-percent confidence mtervals (horizontal bars) Ciulian itadioactis e Waste Management System Management & Operatine Contractor .
O O O~ 'l l PROBABILISTIC VOLCANIC HAZARD ANALYSIS (PVHA) FOR YUCCA MOUNTAIN, NEVADA CONCLUSIONS i e The PVHA was conducted to evaluate the probability ofintersection of the 'i repository by a volcanic event, including the uncertainty in the assessment e Uncertainties were quantified using a panel of ten experts and incorporating alternative models and parameter values e Consideration of available data, methods, analogues, and processes led to i alternative spatial models, temporal models, and procedures for their calculation
- Annual frequency ofintersection spans 1.5 - 2 orders of magnitude for each expert; 3 orders of magnitude for entire panel e Rate parameter, choice of spatial model, smoothing constant, and counts in NW Crater Flat are significant contributors to uncertainty 1
- Mean frequency ofintersection is 1.5 x 10-8 , with a 90% confidence interval of 5.4 x 10 to 4.9 x 10-"
~
O O O~ EVALUATION OF NEW DATA BY PVHA SENSITIVITY ANALYSIS PREMISE: Thefocus ofthe PVHA was to quanti & the knowledge and imcertainty in the annual frequency ofdike intersection. The result is a robust, quantitative assessment. The significance of new data will be evaluated using sensitivity analyses relative to the PVHA probal'iiity distribution function (PDF).
. Multi-expert studies emphasizing uncertainty quantification are inherently robust . PVHA experts devoted considerable elTort to evaluating the existing data, testing attemative models and hypotheses, and incorporating modeling and parameter uncertainties . Data collection for DOE's volcanism program has ended; new data gathered by other groups will be evaluated by sensitivity ane.iysis for their significance to the PDF (changes in the mean estimate) e New data are evaluated for 1) their implications,2) comparison to the assessments by the experts, and 3) their quantitative implications to the PDF.
~
~~
O O O EVALUATION OF NEW DATA BY PVHA SENSITIVITY ANALYSIS (cont'd) DOE has evaluated the significance of two new data sets provided by the NRC in the February Technical Exchange on Igneous Activity:
- 1. An increased volume estimate of Little Cones, based on modeling of ground magnetics data.
- 2. An additional buried volcanic feature in Amargosa Valley, based on modeling of ground magnetics data. -
l
. I
~
--.-----.--_--.a- - - _ - - . - _ - - _ _ - - - , - - - - - - - _ _ _ _ _ _ _
O O O-
. t VOLUME OF LITTLE CONES -
e 3 Volume estimate of Little Cones was used increased to 0.03 km . 3 Total volume estimate of the northwest Crater Flat centers-increased from 0.23 km to 3 0.26 km t . Revised fit of cumulative volume data results in change in volume production rate - 3 3 from 0.316 km /yr to 0.320 km /yr, a 1% change . Only impacts the assessment for Dr. Richard Carlson, who used volume-predictable ; approach for temporal distribution Calculated result is a <l % change in mean annual frequency ofintersection
. - - - . . . . . - . - . . . . - . . - . . - _ _ . , - - - - . . . - - . _ - . - - - . - _ _ _ . - - - - . - - _ _ _ _ - .--_-._-----..----...--.<-~~,n. + ~ wen,-e,*,~ . - - * = . - - ,m-.
~ '
AMARGOSA VALLEY ANOMALIES
- Ground magnetic data give higher credibility to anomaly A (near Little Cones) and anomaly "A" near F and G being buried volcanic cones e 8 of 10 experts gave significant weight to time periods that include the estimated ages of the buried anomalies (~3.8 Ma)
- Anomaly A: addition of another event, or increasing weight on the higher event counts
. Anomaly "A": event-count increased by I to 3, depending on event defmition by expert and number of events previously considered
- Mean event count in Amargosa Valley increased from 4.7 to 6.1 events
- Hazard recomputed using the revised event count
. Results in 4 percent increase in the mean annual frequency ofintersection . Revised assessment of mean annual frequency ofintersection remains approximately 1.5 x 10-*
l!,! j[ ;!l ; -il i !. ; ijli : O m m S' a.
+
u _
- m. ~ -
~
O 4 a - 6_ - 4 sa 4 s.v - m - a o ;' U o o 0 " o 0 0 0 0 0 ,
- '; a s s m -
9 - 8 1 0 3 8 . - a.
' ' i 1 2 - '
3 3- 3 . 4 m - a )j li[ ' i a .* a- - i a-e. 5 0 4 3 - m- 6 e _ 3
=
5 3 5 1 1 j~
- m. 6 - 6 1
a- 1 1 1
4
=.
m v / l, , g - d m
=.
- n. G -
Q't %ihf m a . i w - == i .
- u. s. s .
O u u . e 8_ a. 4 a i c - 3 t m
. e '
+ +
- i n
e. e.
+
g B
%9 4
+
a a ; . m. u. m s o j 4
. r Oe e
e. m gg C e
'0 A30 -
4 u m w 6
$ 4- F * . 3 6
== a 1 1 ~ 1
^
m m. A1 m =. m = w m s.a
+; +E e
4, e =
- b. /j e.
ji (
- I -
I
, ~ \r m
5 0
'4 3 6 .e 3 3 A..
O a j j 4, a e t a
=
+ e-a e
)
4 l l l l l
. 1 o i 4036000 l j nT l -80 l -10 0 l 4055000 -
-12 0 i -14 0 l -160 l -18 0 l l 1 -200 l 4054000 - ._: ' 8. q -220
~
-240 i ; 260 I i 280
! -300 i
-320 l 4053000 -
e4 340
-360 i
i O . r' 38o
-400 ;
l .
- o. k -420 l
' 4052000 - o -440
-460
-480
-500 1
4051000 -
. ' .e l 0 500 1000 1600 t i i 1 l l t i
- 544000 545000 546000 547000 llQ l o'lo ntillil Ilit lGile 'llt litt llI t >l \llit il ger ssti \ lit uillell\' . \ hllt bil'illgt llIlt't* IIllgIly>(l OlyInt(j/hss flHe IllGlI ls 'sn ufttfit sil litt lGile 'lls %IIII s 't t 'llle It /s lie A s!Ille fflin 'ill lllt ' llitlyllt 'fts t ity silltlllt *s. stit fi g ts lllqsti Ig el \ ply *gl Ille lyllt 'll., t llH oll f t llll %fe 't 'j t llIn lIlle llit alI% t ilit llllt ' t 'lc ollgr ift
- t llt lit ti It 'l c olllit lIl it lIltll t llit sflHjllt '% .%tH ll tlt ']till 11! llIt ' Ilk lQilt 'llt litt llI %II/Ij)t oll% llle ' Illfe 'lj elt 'lt ilit sfl llltll lll< 'st ' tilla s/II,lfh *% tilt ' filt st flh t 'tl lI\ ti lbillit 't l n lllQllllit 'll! t IlllIlt 't ' l t s/t tilit st 's ( 't pflfs oll! lijf, Il cllIs lIllil' 'l'ltl sg g lg ifgl g jirgllllg sy u sllq ;tg ply g ,19 lly o fg ,(lg ,j g ,yggy e
l Int ll>% i FlQine% i tnul iJ tne iliflt" t >llet terltunlllu Intellafin ufulliiet snulunelis Relelein e FieltI(ItIRi') 1% It 'lllt at t 'tl O 0
Title:
Probabilistic Volcanic Hazard Analysis for Yucca Mountain, Nevada '
, Document No.: BA0000000-01717 2200-00082 Rev. O Page: 1 10 of 10 r3 Ei ! -
,U.' 'l
.Q [ t ,
J. e N -T.: N - 4 p.. .\.J g
\ .7g ."g..; 7 h. . . .: __
;j ','. .,u,,,.," ;M-
,s
- a. . .s .. p e "I s ,. '
f,.$ . %d. e . ,. . .3t t.
.~.t c
.;rg NEEEFINGv' ,*' ~ ! . . . v
',. s ! ,., ;* '
EtTITE V . y' j .,-(
,a . . ' '. ] . F
~ . pg; 8,. 'i 4
/ ? ) ,@,
1 ( f/ . g, A,., h.& W .n d 4^ . '
;q,'
', Lrrrir. P:/ . .' . ,
' is*
I).
-l E .
'..J.
a FEAK. l . . , . r gl?..,, W 'U- ' j~ .plA4,Ml .V ' g'p ,'. .
'd '
F ..g .!
~
- h;g ' % M'\!2fl
- )g6',',,
. Qm . .. . . . . . .
Y'.. .fj(2,t ,; -irj.'gh
, ., . 3 y- e. , ; . ., pl%. . .
'. U. . s b . W .rak
;: u p ,.>; n
' ! * ,;,, 7 " , .~: ' ,
.a ,
c
! ?" ,. 't
"/ . ,, -
- 1, ,-
g l+d p 3~ U I;. I!e f
,,)
~^kh.k,
.i
( - [Ma CRATER 'Fidt c' , f ;y :p )
'~ . ' ; ,' '
i c, c
' '%gLATlido{&TEt.!p;'
, , . . 'ry . .
(04U-e s--' e o y sg%rc % , . .= s o o . . " , . .y >. ,.&. .- s l M -
+yp; w;,g,. )t.
t Scatt a mitoutrces '.i. ' , '4 . ,
, .-[?-: -- ,
's k S ' ;.! i 4
~
Q.L~d fh \, ? . 5%- ouaon l mno ".hh'i <L AltARGOSA!VAMEY '
.- h~
I
. 'f 8 ,
.qW. . Lu.3 M ,,.
' (]'"--[ . v: k"' ;
i ,
,j% ..
.. g
( _,,.,, m o,
,.,t.
- f. .] , _ .~J. 4
= '.. -
t
.l* ,, ,
g4 # , ',
! s '.-d s
B
- Aerornognetic Anornaly ' . *,%, 'IAJ
! T i ** b .
l >I &ropOSedNepOS, tory : ;..I.. %. }- f I* l
/.,, d. .l.T'. ,#,
P , 4 g ll t ::_;ur n:=in.. . t .......__ s ,. ... ..... ..._ c._.. .-. , o.,,, y% , . a.o.. ;- BASALTIC VOLCANIC CENTERS Y
, AND AER0 MAGNETIC ANOMALIES IN TiiE VICINITY OF YUCCA MOUNTAIN, NEVADA Qm ,
$ GEOMATRIX Civilian Radioactise Waste Management System Management & Operating Contractor ,
' ~
O O O'^ , Resulting increase in mean event count Expert Old Mean Counts Updated Mean Counts Alex McBirney 5.8 5.9 Bruce Crowe 6.3 7.6 George Thompson 5.2 6.6 George Walker 3.5 5.1 Mel Kuntz 3.3 5.3 Mike Sheridan 6.0 6.1 Richard Carlson 4.0 6.4 Richard Fisher NA NA Wendell Duffield NA NA William Hackett 3.3 5.7 Average over 8 experts 4.7 6.1
5 1996 Sth 7 50th Mean 95th *~
+ Ind;vidual Means ,
_ individuol Medians 33 --
] }
I 1 E2 .l
'ci k
1 !i .
$ ++ #M 4
^" "
0 5 1997 updcte 5th7 50th Mean 95 t h '"
+ Individual Means 1, hiduc! Medians v
23 - 2 5 b E2 - t E 1 1 :-
+ ++ #M #
0 _ _ , n_ Ptfd 100 L w 80 - -C, 1996 E 60 ---- 1997 update
.O
- s 9 40 -
-Z t-3~ 20 - O
~
L 0 1 0 -12 10-" 10-'O 10-9 10-8 10-7 10 ~6 10-5
- 1n n ual ['rujut ncy of Inic rscciion Q ,
e 4 9
O O
- O l
PVHA AT YUCCA MOUNTAIN STATE OF NEVADA THE AREA OF INTEREST FOR PVHA AT YUCCA MOUNTAIN
-Gene Yogodzinski (UNLV, Dickinson College)
- -Eugene Smith (UNLV?
- EVOLUTION OF MAFIC VOLCANISM
! AT CITADEL MTN., CENTRAL NV.
-Eugene Smith (UNLV)
, -Lori Dickson (UNLV) l
The Area of Interest for PVHA at Yucca Min.
- OBJECTIVE: Define the natural boundary
- that outlines the mafic magmatic system for the the Yucca Min. area.
- ASSUMPTION: The distribution of Pliocene-and-younger mafic volcanism is in some way tied to the distribution of melting anomalies in the mantle.
- This assumption p! aces importance on mantle source chemistry for basalts.
- The Sr and Nd isotopic systems provide the best source of information.
~
~~
O O O~ Distribution of Mafic Volcanism Yucca Mtn. Area l <6 m.y.-old} < I I 1170W 1160W , Mtn. Sleeping Thirsty / Buckboard Mesa Butte / ( T-
- I k ) 370N Timber Mtn.
Caldera Crater O proposed repository Flat \go e, at Yucca Mtn. g 360 45' N modified from Luedke & Smith (1981) 0 10 20 Lathrop Wells Km
DISTRIBUTION OF BASALT SAMPLES: WESTERN GREAT BASIN l l l 0 ' O j Lk.Tahoe Bl
-- 39N \
o 0 " ~ o O Pancake & Reveille ranges g - 8)
-- 37N Saline Ran b " "'
Big Pine - s g ,
~~
Death Valley Northwest o O B o
%\*
U caN s Lk. Mead
-- 35N Death Valley [ 0 Southeast g
o 0 g\ N -- t
\ Mojave g8g 120W 114W i
- O O O.
WESTERN GREAT BASIN BASALTS (all data n=277 5 outliers excluded) ' 11.0 . 9.0 --: ge _ -. 7-7.0 -.; - 5.0 ---
=~..*
-a ,
- .=. 3. m :
y 3.0 -- r, :- - z 1.0 -- -
= , -- -
- c. ; - - ma. - - - :
5 :-
.o
-1.0 -:.
- ~~ 3,m E. -3.0 - : ~~N' - i-e - --
-5.0 -i -
-7.0 -::
=- -
-9.0 -~ "p- i-
-11.0 -i -
i-
. - 1 -
-13.0
~
0.702 0.704 0.706 0.708 0.710 87Sr / 86Sr
o o O YUCCA MTN. AREA i 11.0 .
- ....:. DATA SOURCES
- Asmerom et al. (1994) 9.0 -: .
gg _ - - = Farmer et al. (1989) 7.0 -- -
- -- Ormerod (1988)
,- - Rogers et al. (1995) 5.0 - *- - Perry, F.V., pers. comm.
i 2- Tg UNLV unpublished data 3.0 -: _fN . Yogodzinski et al. (1996) - V . z 1.0 - c i - o -1.0 --
-35 --
= :
s - E. -3.0 -i N , Thirsty Mtn. i_ e .
-5.0 -i '
i-Other YM ;_
-7.0 -:: =-
.- area ;
-9.0 -:: -
-11.0 4 ]gU -
-13.0 - - .
0.702 0.704 0.706 0.708 0.710 87Sr/ 86 S r
~
0 0 O MOJAVE & RR-LUNAR CRATER 11.0 .- - - - - - - ;- ~ D M ojave
~
9.0 - ~ V DATA SOURCES C 7.0 -: Qg Farmer et al. (1989) Farmer et al. (1995) 5.0 ~ E . O O Glazner et al. (1991) ;
-~ - Glazner & Farmer (1992) 3"0 g : ,-- Dg Lum et al. (1989)
Z - - 1.0 4
"-o'es Yogodzinski et al. (1996) c :
=
o -1.0 -- N'h s - 2-e : g _" .U :
-3.0 -i Reveille Range i
-5.0 -i & Lunar Crater
?
-7.0 -:. "- --
-9.0 -i -*- i-f- -
-11.0 -i -
-f -
i-
-13.0 l
0.702 0.704 0.706 0.708 0.710 87Sr/ 86 S r
O O O DEATH VALLEY NW - BIG PINE - SALINE 11.0 .
~
[F ~_ - -- DATA SOURCES 7.0 -i -
- "_ coleman & Walker (1990)
O- w iker & coleman (1991) 5.0 -: - g - Hoffine (1993)
.c s.23. Ormerod (1988) 3.0 ; _Cb . , _ Rogers et al. (1995)
-h>
Z 1.0 -i - -- -- c -
- A O o Dp 0 -1.0 -: -
, A o -
E. -3.0 _} Death Valley g_ i_ e : iW\/ 9 :
-5.0 :
A :
-7.0 -i Big Pine i-
-9.0 _l & Saline Range _
j_
.A :
-11.0 -i _
, ,- i-
-13.0 0.702 0.704 0.706 0.708 0.710 87Sr/ 86Sr
_ _ __ - _._ _ . . _. _ . . . _ _ _ _ . . _ _ . _ _ _ _ . . _ . . _ _ _ _ . . . _ _ _ . . . . . _ _ _ _ _ . ~ . . . _ _ _ _ . _ _ _ . _ . . _ _ O O O DEATH VALLEY SE 11.0 .
..:....:. DATA SOURCES
- Asmerom et al. (1994) 9.0 --
gg _ -a Farmer et al. (1989) 7'0 2 - nner d (1988)
- -U _. Rogers et al. (1995) 5.0 2 *- - Perry, F.V., pers. comm.
- ,7. - T g UNLV unpublished data 3.0 -: r- Yogodzinski et al. (1996)
V - z 1.0 -1: '- - - -- c - - 88 - - : 0 -1.0 -i - _y~ - 35 --
$. -3.0 -1 e .
5'
-5.0 -i '
i-Death Valley i.
-7.0 d Yi_ ~~
SE
-9 0 _i YM Area -
x gg i
-11.0 -:
-13.0 . .
0.702 0.704 0.706 0.708 0.710 87Sr/ 86Sr
O O O
SUMMARY
OF ISOTOPIC DATA Basalts of the Yucca Mtn. area define a distinctive regior3al isotopic endmember. Isotopically similar basalts are found in the Saline Range and possibly in NW Death Valley, but these fall within data arrays that are highly variable. Pliocene-and-younger basalts in SE Death Valley are identicle to those in the Yucca Mtn. area. Basalts of the Yucca Mtn. area.and SE Death Valley form an isotopic province centered on the Amargosa Valley
- l i
THE AMARGOSA VALLEY ISOTOPIC PROVINCE g j AND THE MAGMATIC SYSTEM IN THE l YUCCA MTN. AREA Thirsty Min. Sleeping Butte na m"'
# % Buckboard Cones % Mesa l %
x /
/ I, I '
I g 370 N TM Caldera 7 1 f Proposed Repository l
, a Yucca Mtn.
Crater e I ' I Flat {Sd l 2 ~ 'd , n, Aeromagnetic g ,,/ I i
' anomalies I s g TA i AMARGOSA h sxgl
'I inNe '
Ty 1( ~ VALLEY ISOTOPIC PROVINCE i DEA'TH ; I VALLEY I x, , l I l g s Black & Greenwater l ; Mountains
/
4 Sample D85-48
,; , s (Rogers et al.,1995) p 1'7 g
j 36 0 N 8/[ g 88 j 117 0 W \ 116 0 W {
% me, me e/
4 i
; Cinder Hill ,
(Farmer et al.,1989) ' l - 4
- MOST RECENT VOLCANIC ACTIVITY IN THE l l' AMARGOSA VALLEY ISOTOPIC PROVINCE e
SLEEPING BUTTE CONES l ~ 0.30 Ma
- f *%
I i i k a t' s l l
! CRATER FLAT I 1 Ws 37 0 N i ALIGNMENT N [ Proposedat Repository Yucca Min.
l ~ 1.0 Ma ! i g o '
'3 I '
.. No j g,
, I , ! ,- 'n
'x 1 g
'_ g l l A <
g LATHROP WELLS 1 I 'O t ~ 0.10 Ma i ', s I
. s :
IL N ;, 8 1 I go I , CINDER HILL ( .\ i
~0.7 Ma x N
1 g ( ' [k '.. I i O \1 l 0 01 } 36 0 N I ocoi 117 0 W \ o's / 116 0 W
% a6 o/
N .
's l
~ .. .
O o , AVIP CONCLUSIONS The boundary around the Amargosa Valley Isotopic Province encompasses the magmatic system that has produced the Pliocene-and-younger basalts in the Yucca Min. area. l This natural boundary and the magmatic ; system it encompasses should be considered in probabilistic volcanic hazard assessment modeling at Yucca Mtn. i
Kn 5Y fYF i ,ii 4150 , bN g? 8 E" !f
- FE o g!' 4125 - - - -
, g; e , n h .S 4 /* s $
QI 4100 - - - -
,E.
a tz
- 2 @
a= ., .' v. 3 m 4075 \ *,. yus -
\ . yys -
Og a g - - ft 8
; N e
\ . .
\ .
6f g a-x g 4050 - N ,
- - N ,, e. -
ee
* \ \ ',
<R
)4025
~ -
x - - L N - E o
\ \ h_
\ \ '
4000 - - - - 3975 - - - - Post 1 Mo - AVIP Post 5 Mo - N. AVIP 3950 ' ' ' ' ' 500 525 550 575 600500 525 550 575 600 East (km) East (km) { i b
~
Figure 3-46 Regions ofinterest specified by Richard Carlson. Diamonds represent volcanic events for the post-I Ma and the post-5 Ma time periods. YMS refers to the proposed Yucca Mountain repository site and the dash-dot line is the Nevada-California border. O O O
O O O EO Od EE 5: $5 3x !?i ag' at b ? Er y E;i 4300 , , i 4 a i i i 6 i a e + h5-
>E E o<
5: 4250 b -' -
~
8E 93 N 8 E. - a 3: ?a
$B R 4200 - \ ~ -
e va;
- N q gy 3 3 -
~g g 4150 -
N gg
$1 '
Y N N e-
= -
3 4100 - N yus
~
vus O{. N N N _ Q$In 4050 - g o0 bL - N I N 8 { f 4000 N ? l 3950 - -
- ~
\ ~
\ N E[
3900 -
\ - - 5
) -
~
3850 -
\ -' l AVIP 200 km Rodius ' '
3800 700 750400 450 500 550 600 650 700 300 350 400 450 500 550 600 650 East (km) East (km) y 8
~
Figure 3-28 Alternative regions ofinterest used as background source zones in George Thompson's PVilA model. YMS refers to the proposed Yucca Mountain repository site and the dash-dot line is the Nevada-California bon' i
.m o o 4o ."
CITADEL MOUNTAIN AS AN ANALOGUE FOR YUCCA MTN.
- Tilted fault block cored by Oligocene ash-flow tuffs Uplift of fault block mostly on the NW.& SE w/ dip slope to the NE Cinder cones erupted along the range crest as well as in the adjacent basins
af . > ,)' u. & i,l,rp. ,,
," -i -; ,, W ,,
,(
$~ u i- . ,.,;/ i ., a ,
l
; .' 'Uf ~ p l g }b, m4. % 1 apg. sy:: 3,4 m m,, u, a s ' ,v . , n%
( )
) s. ' ] <
\ .-N v
.m ,,
') ,'C ',
? **
.) ,- l't _ -
<-i, f~
h . /#. ,d, ; g[f y'f 'x
. e n .~ ,
e *
,.~.,. L f j ~63 9 ) f ,, J 1 i ,/
,Yv..
- '/
s-Ljl - ,
't
+I
~
_j
- Tbst J. e ,
J 'a - ( # g)
/tf w,l'- :;
outh e,.
, ..i' < Tb - -
( idney Butte av' i
- 1.- o- \- ,
i c/ ~ N, ).W ,.i-r )L} . .:, 9
' 4',(y % .
I r# l I /fl_, ,- <T .
')
s
.. ).
\ hnlJ) ' I['%Y,1 Y'L] k QNl' $ o~ - hwbyl ,1L/cr A %g ! ,u ig&,, MMsg, m n ',- 7 y /n/- nn. s., e. p% .,
- f. .^'y3' Ni- % Kj.y i*(:
7 s -
+k.-
l ., i c p. lm 7 ; ,
, j'y ' n ,. ,! %==g l I n y-6g/_ ly;f 9 }; s; _ - ' po l qf L yj7 :!i / & 't '
~
-o gN'.- / w'u
- .g '.% 3a;i 'frc <# "
#=3 Y
1 'y' i af & R JgdQ f ^kjf Jg; \O?h%g&g'g:.<Ikib/}>>u_...g l __ Q To-
.s lgl' C .j%pq
(
.,D p
W, z..
,$QQ&f),
f .- .
.a C J),aw o
,, nyb),'.
, - ,p , '. .
, m .
$AI,,<2;)]1,-N
-r '
i - % &g>)' ,
/k, 'k
; s.~Q/g- -
" Tas n- Q)f
, ej , '
1 ,,,.. 1ye:- - v. ' m. m, oo, , J 1sb , (p'a , .
~-
-f,' '
w / ~~~ a). ..
~u ?? ,,.
n~%=m -
^3ll ,_, ; / ' , $.' ~.)
, ,. '~~. _ ,
.~,.~% .y -
( -
. r/ * ~ / +
.by . ,- a .- .
5, ,./
x_ x ~~ % 's %, % [.~ j 'O l i 5 O We Y. / / l IO Y
R-cone
& C-cone OC-Cone
& H-cone VI-Cone
(. l l l 7 i I $ o 3 KILOMETERS by Loretta f) Dickson,1997 l
.. l
o o o
~
CITADEL MTN. CONCLUSIONS ,
- Analogue study at Citadel Min, suggests that ideas about magma barriers west of Yucca Mtn. (ideas like those that arose in the course of the Geomatrix PVHA process) should be carefully scrutinized.
, 7:; ; O O O w i I I I I I I Q' i O Individutal Expert Means ! C) Aggregate Mean _, l i X o c d rN-y - g "
'x :
Q _ l :_ ~;- .1 Q) ~g . ._. I L _
,, l u .E ..
3
~.
1, , 1 n n fi rY : 1' @. 0 .J0:0J000n'lD:
?Ei"-4J'I =#' 'E I R'
* . ih , I t 10-8 10-7 10-6 10-5 10-4 10-3 10"2 10-1 Catastrophic Eruptions Per 10,000 Years from Science, v274 p913 '96
CNWRA INVESTIGATIONS OF IGNEOUS ACTIVITY CONSEQUENCES ON REPOSITORY PERFORMANCE
~
i . FIN D1035 l NRC DWM Program Element Manager- J.S. Trapp l CNWRA Element Manager: H.L. McKague Presented at the April 22,1997 91St ACNW Meeting :: by i Dr. Brittain Hill l Research Scientist, CNWRA (210) 522-6087, bhill@swri.edu CNWRA Contributors i i C. Connor, M. Conway, S. Greenan, B. Henderson, B. Hill, ! M. Jarzemba, P. LaFemina, L. McKague, R. Martin I I
-_._ _ _ - - _ _ _ _ _ _ _ _ _ _ _ . - _ _ _ . _ - - _ - _ - _ - - _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ -_ _ - - _ _ - - _ - . _ -__ _ _ . _ i
O O 'O-CONCEPTUAL MODEL FOR WASTE INCORPORATION INTO ERUPTION MAGMA: 1100 C, density 2600 kg/m , viscosity 10-100+ Pa s (sucrose 1000 Pa s @ 100 C), velocity 1 m/s initial to 100 m/s eruption, acidic gas. WASTE PACKAGE BEHAVIOR UNDER MAGMATIC CONDITIONS POORLY KNOWN Waste package fails between 10-2 s to 1 year from simple thermal effects of magma. This is the duration range for most basaltic eruptions. WASTE BEHAVIOR UNDER ERUPTION CONDITIONS POORLY KNOWN Waste particles incorporated into ascending magma. Volatile expansion, temperature, and shear may further reduce particle grain-size. DISPERSAL CAPABILITIES OF YMR VOLCANOES POORLY KNOWN Waste ejected ballistically from volcano vent or convectively in the eruption column and transported downwind in eruption plume. Data, Models, and Assumptions Are Necessary to Evaluate Dose Associated with Volcanic Eruption Through Repository 2
O O 'O- ' ANALOG VOLCANISM YlELDS YMR INSIGHTS , Avg LW Avg CF C1 Early C1 Late . 1975 Tolbachik eruption in
,, ' ,d Ys adama, %da, anabgws h Sio, 48.80 50.20 49.80 50.00 processes relevant to understanding TiO, 1.95 1.59 1.02 1.30 Quaternary YMR volcanoes. t Al,0, 16.81 17.30 13.48 15.32 i Fe,O, 2.33 2.15 3.06 3.47 FeO 8.37 7.72 6.99 6.88 Extensional tectonic settings with MnO 0.17 0.17 0.16 0.17 magma OQuilibration depths around 40 km, remarkably similar basalts derived C$ 7 $ 1 .60 3 .
Na,O 3.62 3.41 2.44 3.14 from hydrous mantle lithosphere ' K,0 1.80 1.76 1.03 1.62 i P,O, 1.22 1.10 0.25 0.35 *
- Cone 1 = 0.14 km DRE Estimated H,0 2 2 2 2 - Lathrop Wells = 0.14 km DRE Estimated T ( C) 1100 1100 1200 1200
- Quat YMR = 0.06-0.23 km DRE Density (kg m4 ) 2590 2570 2600 2640 Viscosity (Pa s) 17 30 4 6 '
Crystal Fraction 0.03 0.03 0.03 0.03 . Insights into poorly preserved YMR Viscosity with crystals (Pa s) 20 34 5 7 Oruption depos.ts i and processes YMR data from 11i11 and Luhr (unpub. res),1975 Tolbachik from Volynets et al. (1983). 3
SUBSURFACE AREA OF DISRUPTION I Scale Tolbachik Cone I - 12 hr at end of 1975 Tolbachik Cone 1
- ;;@g, s MA eruption,2.8x108 m of shallow
,cy,,f n w m. m . w - ~
- c. s g
m 4
- w.- ~
$/ subsurface rock disrupted.
- ,- <rY
~
E Mi"i" " ~~~ @*:iv i ni : 200 , d=37 m - Main contact between volcanic and U -
, 3 1
y m sed.imentary rocks 800-1300 m, water
- ' ^
400__ _; _ M,, .~ table at 500 m _,.%- <! % e ~
- - u --
N
~
^
W 600- I -con ~ tact 800 m" ,. Volume of subsurface rock equals n - - , - - q c d=so_47 m N g m
'=
widening conduit from initial 1-3 m 8 800 d.iameter to 49 7 m diameter eop_ e <p r m . esuL
" W e '
1000% c :##8M h M . M3 2 $
%Rangeof M" ~s ~
Late-stage re&ch.on in magma How rate M 4uncertaintyi gcontact 1300 m.i allowed dilation of fractures below 500-1200 M"#@ L M N YM m-deep water table, aiding brecciation to#M# U
.; y h o h.
1400 [ , N sedimentaryj Sparse rock-ash on cone, but abundant g s; (1-2%) xenoliths on cone fla iks 1600-i i _ _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ - _ - _ = _ _ _ _ _ - _ _ _ - - _ ____ _ - _ _ _ _ - _ _ . . .
'O-
~
O O SUBSURFACE AREA OF DISRUPTION r n:- r,: - l -
.a -
Wells and all other YMR volcanoes eroded w. [ Proximal xenoliths unusually abunc ant on outer l l
- - flanks of Lathrop and Little Black Peak Mhu ay... a t
* = , _3 L i ,1 , Same type of unusual polylithologic, polythermal
! i mm_____- ij!j
, volcanic pyroclasts at Lathrop and Tolbachik produced during a late-stage disruptive event (Figure)
Both volcanoes disrupted similar range of crustal sections, up to 0.5-2 km deep 1 - Scale of disruption for Lathrop Wells most likely l comparable to 1975 Tolbachik. May disrupt 3 4-10 waste packages if similar event occurs in
*p a m --
1-- 83 MTU/ acre repository. r
~~
s": -_ l 5 1 I l
a
. ,V C,., U -
TRANSPORT OF MATERIAL FROM BASALTIC VOLCANOES Size-range of analog eruptions superimposed on YMR site, wind to south. i, e,y~ y% e 'e
~'e
(- l5 *tj4-W*/ .x - i Y n, h y/ e - e j e e j4 *
\ -
.. e Je e 's e f /s e y e e
[c e 'e
/ .
. /I 1
/ x \' '
~7 // \, ! j
' i "
h hb g yf $ g~ e f Q'-\ e w/ +e e g e e ' .,e j y ,y e e G y
.e e v
e (' I e eQ n \\ , 1
'.,,) ~( b / e \e ',\/, v e.
J te / e e
~
.@ h e eQw l3/te e & s e e J
Y d y le i
'.. 7; j/ 17 e + e J, e ,I 4 e e
~'
el, S/ r e e
"! */ e e e e 4 kel e l q e[ 4 e e e 'F e -u. k e e /e e e ti ep e
- Le e f
- . m y ivy ~
/' - '
~
1 S. e \'*M@#e ; 3x w i ; /z s /r e e e - [5 e e e e \1M i
\M < >[\\ J rbh4M
. y ;-
\ ,/ \
e 6 e s) eV i - e b e a% e e e V.1 d g y j e e e e
'e,
- e e ,
ej e r J e e e e e t
.s g\\ ,)V'y TW/ fp4v
.W . . . A Y ..$...\.....
- W - - - [b - a .
YMR,5 km tics O Critical Group locations YMR,5 km tics e Critical Group locations 1975 Tolbachik analog, tephra isopachs in cm 1995 Cerro Negro analog, isopachs in cm 1 cm isopach shown in dashed line. Tolbachik (0.45 km ), Cerro Negro (0.008 km ) 6
~
O O 'O-MODELING TEPHRA DISPERSION WITH ASHPLUME e m e ge s e 7 f, L Dispersion model of Suzuki (1983)
% m e 10 N ' e g .' )
Accounts for vertical and lateral g 9- . l
*\ dispersion of particles t 1
i D. ' y#f - Modeling 1995 Cerro Negro,
, g, l calculated deposit thickness very
?
I/
~
\
/ M- ~^f '/ . ' 'I .
/ /
i sensitive to Wind Speed I- .1995 Cerro Negro 1 O Ix A [ 01 cm isopacn . Moderately sensitive to column height, particle diameter & density e e o 3 m e
~J 'Q - Yv g pgy i g Model reproduces measured deposit thicknesses within 50% at 20-30 km
~ N N'- y/ 21 N7;
<2 Ma \ W"q lJe jg' 2-6 Ma \ W j .
6-12 Ma
\ '
.i i y
- ASHPLUME used to calcul ate dose from volcan,ic eruptions YMR,5 km tics 9 oritical Group locations AsHPLUME model,1995 Cerro Negro eruption
.h .
i t],.}
;_ 7 CNWRA DOSLi CALCULATIONS FOR VOLCANIC DISRUPTION OF THE PROPOSED REPOSITORY t
FIN D1035 NRC DWM Program Element Manager: J.S. Trapp CNWRA Element Manager: H.L. McKague ! Presented at the April 22,1997 91St ACNW Meeting by Dr. Brittain Hill Research Scientist, CNWRA (210) 522-6087, bhill@swri.edu CNWRA Contributors C. Connor, B. Hill, M. Jarzemba, R. Manteufel . 8
CURRENT DOSE CALCULATIONS FOR VOLCANIC ERUPTIONS BASIC ASSUMPTIONS Volcanic eruption occurs through repository between +200 and +10,000 yr 4 - One canister fails and all waste (10 MTU) available for magmatic transport Wind blows to south 14% of the time Critical group located 20-30 km south of repository Current TPA approach (DCF, GENil-S, etc.) used for dose calculations CRITICAL PROCESSES Behavior of ascending magma in disturbed geclogic setting
- Under investigation, assumed non-disturbed behavior Canister response to ascending magma
- Failure assumed based on initial analysis Waste particle-size distribution
- Mean values 10,1, and 0.01 mm, log-trinngular range 1 log unit Waste incorporation mechanisms into magrna
- Tephra diameter must be 10x, 5x or 2x waste diameter to incorporate Dispersal capabilities of YMR volcanoes
- As observed at historically active analogs, using ASHPLUME t 9
o O O . - EVALUATE SENSITIVITY TO WASTE PARTICLE-SIZE AND INCORPORATION FACTORS RESULTS OF 300 SIMULATIONS EACH,10 MTU WASTE AVAILABLE FOR TRANSPORT l MEAN ANNUAL PEAK DOSE,1 STANDARD DEVIATION (MREM / YEAR) LOG-NORMAL DISTRIBUTIONS Parameters sampled stochastically: Eruption power & duration (column height, eruption rate, mass), column shape, time, wind speed, tephra diameter l l PARAMETER SENSITIVITY EXAMINED CRITICAL GROUP LOCATION _20KM 30 KM 0.01 mm diameter,2x incorporation 50, 300 7, 40 0.01 mm diameter,10x incorporation 40, 300 4,40 ' 1 mm diameter,2x incorporation 9, 90 0.2, 3 1 mm diameter,10x incorporation 0.3, 3 10-4, 10-10 mm diameter,2x incorporation 10 ,102 10-8,10-7 10 mm diameter,10x incorporation 10-'8, 10-'7 10-38 10
o ' ' ~ O O - CURRENT DOSE ASSESSMENT FOR VOLCANIC ERUPTIONS t Comparison of Current Undisturbed Repository Performance with Extrusive Volcanism CCDF shows mean values are reasonably le4 : conservative for these calculations i
- UNDISTURBED REPOSITORY Current calculations show this mean is
'*-'1 - NRC evaiuation of > s50 mrem /yr NAS Recommendations x - Amargosa Desert Location i /
i ie-2 ,
. Vent widening may disrupt 10 canisters, j
m . thus 50 x 10 canisters = 500 mrem /yr i 8 1e-3 , - ean= t 5 : EXTRUSIVE VOLCAN 0 rem /yr Current CNWRA probability models range E s 1e-4
- waste grain-size: 10 nm 1 - Waste incorporation: 2:1
' /
to 10-3 for 104 years '
- - Dose point: 20 km S of site .
- - wind to S 14.1% 4 l
Risk = [ Volcanism, 10 -10 ,500 mrem /yr] 1e-s = - "=30 - i
- 100 MTU release .
- - 10' yr Volcanism Probability: 10-g g g j 1 e ,
, , , , , , , , , , , , ,, . . , , , , ,,,i, ", -
jos incorporation mechanisms, system 104 10-3 10-2 i o-i 1 00 1oi 1 02 303 to4 response to thermal, chemical and l Peak Annual Total Effective Dose Egivalent in the Time Period of Interest (mrem /yr) mechanical loads from igneous activity, ; dose point locations. ! 11 I
O O + _ O ANNUAL INDIVIDUAL DOSE ESTIMATES FROM
" PRELIMINARY" PERFORMANCE CALCULATIONS U<g A,
so 4 ooc # i i Tim McCartin US Nuclear Regulatory Commission Office of Nuclear Material Safety and Safeguards i Presented at:
)
ACNW Meeting on Igneous Activity April 22,1997
.- - - .=. . _.. -.--.
o o o
~
CONTEXT OF ANNUAL INDIVIDUAL DOSE ESTIMATES FROM PRELIMINARY PERFORMANCE CALCULATIONS i TSPA '95 Preliminary inclusion of a peak dose calculation in response to NAS , Recommendation . L NRC Staff Evaluation of NAS Recommendations identification of implementation issues with long term dose estimates and exposure scenario (reference biosphere and critical group) i 1 t
. _ . _ _ _ ~ _ _ _ .._ _ __. _ _ _._ _. _ . _ . . _ . _ _ ._.__. _ . _ _ _ ... _ _ _ _
.._-_____._.7_~ ~
O O O - TSPA '95 I BASIC ASSUMPTIONS i [
- Undisturbed performance only :
1 volcanism not considered quantitatively TSPA 91 considered direct release to surface from volcanic event TSPA 93 considered indirect effects of volcanic event l
- Drinking water dose at 5 km I
f t i 2
DOE SPA '95 10,000-yr Total Peak Dose 1 Q .
.C_. .
O 0.1 = - b : m -
.O -
j _
-*- 25 MTU/ac, backfill O
w
-*- 25 MTU/ac, no backfill
. w -
0.01 ' ' ' " "i ' ' " ' "i '""i ' ' ' " "i ^""i: """i 10-7 104 10-s 304 3 o-a 10-2 10-1 10 Peak Dose to AE (rem /yr) CCDF of peak annual individual dose at 5 km down gradient (with and without backfill, high infiltration, 25 MTU/ acre) 3
n v rd NRC Staff Evaluation of NAS Recommendations () , BASIC ASSUMPTIONS e Nominal base case analyzed probability near unity peak dose (up to 1 million years) : compliance at 5 km (drinking water pathway) and 30 km (all pathways) average member of critical group approach for individual dose
- Extrusive volcanism analyzed ,
undisturbed performance, extrusive volcanism, and human intrusion (analyzed separately) compliance at 20-30 km (all pathways) average member of critical group approach for individual dose consideration of magmatic interaction with waste and deposition of ash 4
~ '
O O O CCDF's for Peak Dose, Millirem Amargosa Desert and 5 Km locations 1.0 - r - I's (
\
5 Km Location h] N Amargosa Desert
\
\
\
\
l 0.1 \ N - i
\ ' '
I
, I l l {
l
! l
\
h \
\
\
\
N I 0.01 -- ' ' ' 1 10 100 1000 10000 ; Peak Dose, Millirem i NRC Evaluation of NAS T Uncertainty analysis (100 realizations) of annual individual dose 5
~ ~~
O O O Individual Dose Comparison (analysis done to evaluate implementation of NAS Recommendations) CALCULATION MEDIAN MEAN 95th PERCENTILE TSPA 95 0.4 mrem 10 mrem (5 km, drinking water) NRC NAS Nominal 23 mrem 76 mrem 104 mrem (5 km, drinking water) (range: 1.5-1,060) , t NRC NAS Nominal 4 mrem 14 mrem 18 mrem (30 km, all pathway) (range: 0.2 - 118) i 6
i e U.S. DEPARTMENT OF ENERGY um OFFICE OF CIVILIAN RADIOACTIVE WASTE MANAGEMENT r ADVISORY COMMITTEE ON NUCLEAR WASTE
SUBJECT:
Planned Sensitivity Analyses of the Effects, Consequences and Risks of Volcanic Hazards at Yucca Mountain in TSPA-VA ! PRESENTER: Dr. Abe VanLuik PRESENTER'S TITLE Technical Synthesis Team Leader AND ORGANIZATION: U. S. Department of Energy Yucca Mountain Site Characterization Project Office Las Vegas, Nevada TELEPHONE NUMBER: (702) 794-1424 April 24,1997 1 I I t
O O O ~l Outline Summary of volcanic scenarios used in prior TSPA analyses Summary of volcanic scenarios to be considered in TSPA-VA -
- incorporation of alternative interpretations of probability of occurrence
- incorporation of alternative models of direct j and indirect effects and consec uences j
o o o
~~ ~
~
Scenarios - Summary of Prior Work Basaltic Volcanism intrusion Acts Directly intrusion Acts Indirectly on Repository on Repository o I E .N Dike Forms Surface Drainage Subsurface Freq. Altered Flow Altered l I ,i i
$ 2L
[ Freq.\ Transport of No Waste Magmatic Basaltic Waste intact Magma Alteration of Cone Sill Forms Contact Waste Forms
- i :
l TSPA-9 TSPA-9 Unk et al., 1982 Freq. Freq.
o o o
~
1 Previous Inclusion of Volcanic Scenarios in TSPA TSPA-1991
- direct entrainment of waste to surface
- amount of entrained waste a function of volume of dike, extent of wall rock erosion, area of repository intersected by dike 1
- random dike orientation, length, amounts of wall-rock j erosion j
TSPA-1993
- indirect effects ofintrusive event accelerates waste package and waste form degradation
- ___ - _ ___-__ ____- --__-____-_ _ _ -_ _ L
4 Summary of TSPA-1991 Volcanism . Results
- Probability of occurrence of igneous event = 2.4x10-4 over 10,000 years !n 10a 10-1 Release models felt $ o -'
to be conservative ] ,o .3 _ . . ,. 8 6 10 d 35 EPA remn . Maximum release E 20 s :"'" (expressed as old i %1-<-.~ m EPA sum) = 1 0 '" ' 10-7 10 4 10-5 10 4 10-3 10 2 10-1 10 0 10 1 10 2 EPA sum
e o o ~~1 Summary of TSPA-1993 Volcanism Results . Both probabilities and 3 consequences are low i "%^"". J 10-' om y rs '
. P '1 104 yrs' = 2.4x10-4 y, ,
'il P ~1 106 yrs] = 2.3x10-2 , 9 ., .
10 4 ' Maximum release at [ ., , 10,000 years ~ 10-4 ii. C 10 4 10-8 10-7 10-6 10-s 10 4 10-3 10-2 10-1 10 0 10 1 10.2 , EPA sum i i
o o o ~~ Applicability of Recent Interpretations PVHA
- provides credible estimates of probability of occurrence (and uncertainty) for intersection of dike with repository LANL Volcanism Synthesis Report - data on erupted lithics, intrusion geometry, geochemical alteration, vapor convection NRC/CNWRA - model of direct effects and tephra dispersion from basaltic eruptions
v od eJ Proposed Scenarios for TSPA-VA Basaltic Volcanism Intrusion Acts Directly intrusion Acts Indirectly on Repository on Repository I Dike Forms Surface Drainage Subsurface Freq Altered Flow Altered l I i
!A
[ Freq \ Magmatic Basaltic Sill Forms Lateral Fast-path flow W siein act Mag a Con aci Alteration of Cone diversion in along dike Waste Forms CH units alignment l Waste is Waste modifies Waste is entrained. entrained and source term and distributed at distributed at l surface surface <
\ fjuANEVolcaM5i$
l s a e
. ;:t.fokhtn!!iurface-::
. N kW*WbW
- gg.gygjg:ggg.g==~
3.y;;;NS,,CyTephras7:.i... gp, ,gngg
g 9 o Alternative Models for TSPA-VA: Probability of Disruptive Events Use elicited values from PVHA ' Employ importance sampling to assure events are sampled and weighted appropriately j Use fixed ("best-estimate") direct effects consequence model ~ direct effects considered to have about 10,000 times greater consequence than indirect effects ~ Consequence analyses use peak dose to average member of critical group located at 5 km and 30 km at times of up to 10,000 and up to 100,000 years Evaluate sensitivity to alternative PDFs honoring the . elicited frequency ofintersections (sensitivity evaluated with alternative CCDFs of peak dose) l !
g 9 Alternative Models for TSPA-VA: Probability of Disruptive Events i Alternative PDFs to be studied in sensitivity analyses:
- fix values at 50th and 95th percentile
- sample from raw pdf (or raw pdf transformed to log normal that honors 50th and 95th percentiles)
- sample from uniform distribution of elicited median values
- sample from log-uniform distribution encompassing the 0.lth and
, 99.9th percentiles
- discard the one outlier at the low end of the distribution and recompile the pdf Document results in TSPA-VA
C O Alternative Mod $s for TSPA-VA: . Direct Effects and Consequences Review appropriateness of conceptual assumptions in the CNWRA's model (compare to models used in TSPA-1991 and recent information from Volcamsm Synthesis Report) Assuming CNWRA model is representative, define reasonable ranges for key input paramei.ers
- use DOE experts and subset of PVHA external experts - CNWRA " bounding" parameters should be represented within the range of parameter values - examples: magnitudes of ejected material, depth and percent of wall rock entrainment, dike length and width, erupted material characteristics Conduct conditional simulations of consequences (assuming probability of occurrence equals 1) over the range of parameters Document Results in TSPA-VA
o o o 1 Alternative Models for TSPA-VA: Indirect Effects and Consequences Review previous models bounding indirect effects Revise previous models or develop new models, including bounding the effects of
- enhanced degradation of waste packages
- enahnced degradation of cladding and waste form :
- revised solubility of radionuclides
- modified transport characteristics along likely flow paths t
- revised saturated zone flow field Conduct conditional consequence modeling ofindirect effects using bounding models ofindirect effects Reconfirm that the consequences ofindirect effects are significantly less than those due to direct effects Document assumptions and results in TSPA-VA
O O O 1 Summary TSPA-VA will utilize recent results from PVHA, Volcanism Synthesis and NRC/CNWRA analyses / interpretations Sensitivity analyses ofprobability of occurrence, direct / indirect effects and consequences will be conducted and documented in TSPA-VA If either consequences or risks are significant, then volcanic scenarios will be included in TSPA-VA reference case If both consequences and risks are insignificant, , then volcanic scenarios will not be included in TSPA-VA reference case
o . O O O ,~.. 1- United States . _, ,i Nuclear Regulatory Commission DOE /NRC AGREEMENTS FROM TECHNICAL EXCHANGE WITH COMMENTS : I Presented:ACNW APRIL 22,1997 By: John Trapp, NRC (301) 415-8063 ! AND
" Tim" Sullivan, DOE (702) 794-5589 i
~
i sprit 22,1997 t
~
O O O , f* '""> i 1 United States Nuclear Regulatory Commission
%.)} .
AGREEMENTS FROM TECHNICAL EXCHANGE k
- 1. DOE and NRC agree that the rate of volcanism is relatively constant for I
- the last 5 million years and can be assumed to remain relatively constant for the period of performance.
- 2. DOE and NRC agree that based on current information, silicic volcanism need not be evaluated.
- 3. DOE agrees to consider evaluating new data such as:
- the size and volume of Little Cone
- the number of events at Anomaly A i through hazard sensitivity studies.
2 Ap gl22,1997
,...~.
- t. United States
] Nuclear Regulatory Commission AGREEME:NTS FROM TECHNICAL EXCHANGE
- 4. NRC believes that an annual probability of 10-7/yr is a reasonably conservative upper bound for extrusive events. There are differing views on the lower bound. DOE considers that the PVHA provides a defensible basis for characterizing the probability of disruption (includes both intrusive and extrusive magmatic events). The probability distribution function (PDF) has an upper bound frequency of 10-7, a lower bound of 10-10, and a mean of 1.5 x 10-8 per year. DOE agrees to explain how the PDF for the probability of disruption will be used in performance assessment, including sensitivity studies, recognizing NRCs comments.
3 April 22,1997
o 0 O~ ': y~. g: United States
. ,, j,I Nuclear Regulatory Commission AGREEMENTS FROM TECHNICAL EXCHANGE
- 5. DOE and NRC agree that:
t A. Volcanism is of regulatory interest and its. probability and consequences will be considered. B. If determined to be significant with respect to repositor; performance, the effects of volcanism will be included in the total system performance assessment.
- 6. The treatment of consequences outlined by DOE that includes extrusive magmatic events (cone and dike formation) and intrusive magmatic events (sill and dike formation) with both direct and indirect effects is generally appropriate at the level of detail provided.
l i 4 April 22,1997 i
~
~ ~~
O O O i ,..~. ,
'1 United States
'y}l Nuclear Regulatory Commission AGREEMENTS FROM TECHMCAL EXCHANGE '
- 7. DOE and NRC agree that there is uncertainty in consequence analysis for magmatic waste package / waste form interactions and will be evaluated.
- 8. DOE agrees to provide the NRC with a letter describing the DOE basis for subissue resolution, as specified in 3 and 4, for consideration in the development of NRC's Issue Resolution Report.
5 Apgg 22,1997
o., o o o . ., l w .1 United States
\, } .....
Nuclear Regulatory Commission NRC PLAXNED ACTIVITIES Presented: ACNW APRIL 22,1997 By: John Trapp, Senior Geologist Engineering and Geoscience Branch Division of Waste Management Office of Nuclear Material Safety and Safeguards (301) 415-8063 e-mail: jst@nrc. gov 1 April 22,1997
O O O'.'! 8 p.- ~m,,,,
~
! 1 United States
,,,,,/ Nuclear Regulatory Commission IGNEOUS ACTIVITY KTI SUBISSUES WORK ORGANIZED ALONG THREE SUBISSUES PROBABILITY CONSEQUENCE t DATA QUALITY t
t t I April 22,1997
,0~.
1 United States i 1 Nuclear Regulatory Commission 03 PROBABILITY SUBISSUE IRSR PROBABILITY OF BASALTIC IGNEOUS ACTIVITY EARLY FY98 TECHNICAL WORK
-INVESTIGATE SIGNIFICANCE OF BURIED GEOPHYSICAL ANOMALIES
-PROBABILITIES OF INDIRECT / SECONDARY EFFECTS AND PROCESSES
-EXAMINE CONTROLS ON CONDUIT LOCALIZATION
-EVALUATE MODELS WITH DATA FROM OTHER VOLCANIC FIELDS 3 April 22,1997
i
' o .. . '
'\ United States
^'
, ij Nuclear Regulatory Commission
%sso3 *'
l PROBABILITY SUBISSUE (CON'T.) . , PLANNED PEER-REVIEWED PUBLICATIONS l'
-INTEGRATED VOLCANISM-STRUCTURE PROBABILITY MODELS FY97
-TOLCANO CLUSTER DEVELOPMENT, SAN FRANCISCO VOLCANIC FIELD FY97
-PETROGENESIS OF YMR BASALTIC MAGMA SYSTEM FY98 IRSR PROBABILITY OF SILICIC IGNEOUS ACTIVITY TBD FY98?
TECHNICAL WORK COMPLECTED i 4 April 22,1997
\, United States i Nuclear Regulatory Commission 03 1
~ CONSEQUENCES SUBISSUE IRSR CONSEQUENCES OF BASALTIC IGNEOUS ACTIVITY EARLY FY99 TECHNICAL WORK
-EVALUATE SECONDARY / INDIRECT EFFECTS ON PERFORMANCE
-DEVELOP ADDITIONAL BASIS FOR SUBSURFACE AREA OF DISRUPTION '
-MODIFY AND TEST TEPHRA DISPERSAL MODELS
-MODEL EFFECTS OF REPOSITORY ON ASCENDING BASALTIC MAGMA
-MODEL WASTE PACKAGE / WASTE FORM BEHAVIOR ;
i 5 April 22,1997 i i
,f m 1 United States . N Nuclear Regulatory Commission CONSEO_UENCE SUBISSUE (CON'T) . PLANNED PEER-REVIEWED PUBLICATIONS
-COOLING AND DEGASSING OF SHALLOW BASALTIC DIKES FY97
-TEPHRA DISPERSION AND RISK ANALYSIS, CERRO NEGRO FY97
-DEVELOPMENT AND EVOLUTION OF THE 1975 TOLBACHIK ERUPTION FY98 6 Apri: 22,1997
;; United States ',,,,__}.) Nuclear Regulatory Commission i
DATA QUALITY SUBlSSUE l REVIEW RELEVANT ASPECTS, DOE VOLCANISM SYNTHESIS REPORT INVESTIGATE SIGNIFICANCE OF BURIED GEOPHYSICAL ANOMALIES l s 7 April 22,1997 i f
O~~ O O
\ United States
- V, j Nuclear Regulatory Commission CROSSCOTTING ACTIVITIES
-SENSITIVITY /IMPORTANCE STUDIES - VOLCANIC SYSTEM AND TOTAL SYSTEM
-REVIEW RELEVANT SECTIONS OF DOE TSPA-VA PLAN
-REVIEW RELEVANT SECTIONS OF DOE TSPA-VA
-EVALUATE DOSE SENSITIVITY TO CRITICAL GROUP LOCATIONS AND OTHER PATHWAYS
-EVALUATE DOSE CONVERSION AND PATHWAYS
-INTEGRATE - WASTE PACKAGE / WASTE FORM, STRUCTURE, PA REPRIORITIZATION TO FOLLOW SENSITIVITY /IMPORTANCE STUDIES AND BUDGETING 1
8 April 22,1997
YUCCA MOUNTAIN PROJECT
~ ~ -
Igneous Activity Program Path Forward I Presented to: 91st ACNW Meeting Presented by: Tim Sullivan Viability Assessment Team Leader Yucca Mountain Site Characterization OfTice
- U.S. Department ofEnergy '
OfTice ofCivihan Radioactive April 22,1997 waste Management i 2TERAM PPT ! i i
c. o o o PATH FORWARD HAZARD ANALYSES The results of the PVHA are intended to provide a sound defensible basis for licensing DOE has evaluated new CNWRA data and these data do not significantly impact the results of the PVHA The results of PVHA are robust; new information is unlikely to change the disruption probability documented in the PVHA 2TEPVHA PPT
O e O PATH FORWARD
- HAZARD ANALYSES No further site characterization data collection planned; new data are unlikely to change results Disruption probability estimates from PVHA will be used for the reference case and sensitivity studies as described earlier DOE will address TE agreements (# 3,4, and 8) for evaluation of new CNWRA data and description of the use of hazard results in TSPA-VA in an upcoming letter Next step is hazard probability subissue closure 2TEPW14 PPT Y - _ _ . _ _ _ _ _ _ - - - _ _ - . _ _ _ _ _ - _ _ ~ - J
l PATH FORWARD CONSEQUENCE ANALYSES Consequence analyses for direct effects will be developed by DOE for TSPA-VA
- For indirect effects DOE will review previous bounding effects considering new information and reevaluate the earlier conclusion that indirect effects are several orders of magnitude less than i i
direct effects i Document assumptions and results in TSPA-VA This approach is consistent with TE agreements (# 4,6, and 8) for consequence evaluation , Next step is consequence subissue closure 2TEPVHA PPT i t l
PATH FORWARD TSPA-VA
- Igneous activity scenarios wil be evaluated for TSPA-VA If either consequences or risk are significant, DOE will include volcanic scenarios in TSPA-VA reference case i
If both consequences and risks are insignificant, DOE will not include volcanic scenarios in TSPA-VA reference case and will document the rationale for this approach in TSPA-VA t 2TEPWHA PPT
------_-----_____ _ _ _ . - - - _ _ _ _ a}}