ML20054M088
| ML20054M088 | |
| Person / Time | |
|---|---|
| Issue date: | 07/07/1982 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1115, NUDOCS 8207090235 | |
| Download: ML20054M088 (78) | |
Text
..,.
.. ~
..r
=. :.,,
- v.
.:,+-S
. ~
- , ~ ~,: ~.
v...
/
NUCIZAR REGULATORY COMMISSICN 1
gD b
g [t uL I
In t:he Mat::ar of: ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON EXTREME EXTERNAL PHENOMENA DATE:
July 7, 1982 PAGES:
1-71 AT:
Washington, D.
C.
[fl I
C
.Q
}
J &'
ALDERSON REPORTING
- r. y
=
400 vi_T d a Ave., S.W. Washin g==, D. C.
20024 Talaphc:a: (202) 554-2342 E
4, w,.
M e n',902:iS 9 7; n 'd J
.H A: h' !i t
11If F I '"
'- 4 g
.+
1 1
UNITED STATES OF AMERICA 2
NUCLEAR REGULATORY COMMISSION 3
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
SUBCOMMITIEE ON EXTREME EXTERNAL PHENOMENA 5
6 Room 1167 7
1717 H Street, N.W.
8 Washington, D.C.
9 Wednesday, July 7, 1982 10 The maetLng was convened, pursuant to notice, 11 at 4: 55 p.m.
12 PRESENT FOR THE ACRSa 13 DAVID OKRENT, Member O
14 M. BENDER, Member 15 J.
C.
MARK, Member 16 J. EBERSOLE, Member 17 DESIGN ATED FEDERAL EMPLOYEEa 18 RICHARD SAVIO 19 20 21 22 23 OV 24 25 O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
2 O
i esacstatsas 2
MR. OKRENTs The meeting will now come to 3
order.
/}
4 This is a meeting of the Advisory Committee on 5
Reactor Safeguards Subcommittee on Extreme External 6
Phenomena.
7 I am Dr. Okrent, Subcommittee Chairman.
8 The other ACRS members present, apparently, 9
are Hr. Ebersole, and Mr. Mark.
10 The purpose of this meeting is to discuss the 11 proposed FY84 and FY85 research programs and funding l
12 levels in areas of extreme external phenomena.
l 13 This meeting is being conducted in accordance O
14 with provisions of the Federal Advisory Committee Act 15 and the Government in the Sunshine Act.
h 16 Dr. Richard Savio is the Designated Federal 17 Employee for'this meeting.
18 The rules for participation in today's meeting 19 have been announced as part of the notice of this 20 meeting previously published in the Federal Register on 21 Friday, June 18, 1982.
22 A transcript of the meeting is being kept and 23 will be made available as stated in the Federal Register 24 notice.
It is requested that each speaker first 25 identify himself or herself and speak with sufficient O
l l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
3
()
i clarity and volume so that he or she can be readily 2
hearc.
(}
3 We have received no written statements or 4
requests for time to make oral statements from members 5
of the public.
6 We have an agenda for the meeting, which I 7
hope we can adhere to fairly well.
If we do, we will 8
finish at 7500 o' clock.
I aon't ~
you all, but know about 9
some of us have a lot of reading to do this evening.
10 So, while we do want to here what is important, try to 11 omit things that are descriptive and of less importance r
12 from what you tell us.
13 I think that our first speaker is to be Mr.
)
14 Be r a ta n.
So why don 't we begin.
15 HR. BERATAN:
Following the agenda, I would 16 like to bring you up to date on what is happening in the 17 budget of the Earth Sciences Branch, going back to the 18
'82 budget and bringing you up to date.
19 The Office of Research received from the 1
20 Comptroller's Office about $3.2 million of which the 21 Earth Sciences Branch is sch'eduled to get about a half 22 million dollars which will be used to sup'plement some of 23 the geological programs.
I have a breakdown of where we
(
24 are planning to put,this money.
25 We plan to let a contract for the regional O
l ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
1 4
()
1 seismic network, the Memphis Area Network, additional 2
work in the State of Maine, geologic work, some U.S.G.S 3
stress measurements in Maine, and some seismic work in
(}
4 the Loomis area in Connecticut, which will consume the 5
supplemental.
6 A new figure for the 1983 budget has been 7
increased by from $5.5 to F6.5 million, and that will i
l 8
enable us to further enhance the program.
1 1
9 We did a rather thorough review of the 10 existing geological / seismological program and we have 11 passed' out some changes that you can read at your i
12 leisure, which we are planning to implement.
l l
13 J.ust to give you a brief capsulation of what O
14 we are planning to do, in the New England area, the new 15 emphasis will be on crustal structures, a study of the i
16 crustal structures and geophysics, measurement of 17 crustal stress and crustal strain, and geomapping will 18 be limited to the target areas such as the Loomis, 19 Connecticut area.
20 In the New. Madrid area, there will be new 21 emphasis on recent strain and geophysical and geologic 22 studies to the determine the nature and extension of the l
l 23 New Madrid seismi: zone.
This redirection is not as 24 extreme as the redirecton of the New England program.
25 In Southeastern U.S., the Charleston area, we t
l 1
l ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
5
()
1 will extend that investigation to include the 2
investigation of possible seismogenic structures, 3
including some stress measurements.
{)
4 That is how we plan to expand the program.
We 5
will be cutting out some of the on-going.
l 6
MR. MARK:
Will the New England Program allow t
l 7
you to set f oot in New Brunswick?
8 MR. BERATANs Yes, or at least as close to it
~
9 as we can get.
to MR. MARKS Are you allowed to go across the 11 bo rde r?
12 MR. BERATAN We have an information exchange I
13 agreement with tha Canadians.
We are working very
! ()
14 closely with then right now.
As a matter of fact, they 15 have some of our instruments up th ere.
16 MR. MARK At least that region is included as 17 part of your interest?
18 MR. BERATAN Yes.
As a matter of fact, we 19 just bought some new strong mo tion instrumentation which 20 will be placed up in that area and other areas.
We just 21 came into $50,000 just the other day, and we spent it on 22 strong motion equipment.
So we will be expanding that 23 network to include strong motion instruments.
24 MR. OKRENTt Before you leave this subject, a 25 question of some interest is how do we ascertain the one O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
6
()
1 in a thoussad, the one in ten thousand, and the one in a 2
hundred thousand earthquake for the central stable
[}
3 region, so called.
Is there anything that we are doing 4
or plan to do in this regard that in any way relates to 5
geology and seismology?
6 MR. BERATAN:
Yes.
Under the topical studies, 7
we have a program to determine recurrence intervals, 8
analyses of strong ground motion and that sort of 9
thing.
We think that that is covered.
10 MR. OKRENTs In other words, another way of 11 saying is, would you have more work in Michigan and 12 Wisconstn is a magnitude of 5.8 occurred there, and 13 nothing had occurred in New Brunswick?
14 MR. BER AT AN 4 I will defer to Andrew Murphy.
15 HR. MURPHY:
Andrew Murphy, Seismologist.
I l
16 an afraid that I don 't quite understand the question.
17 MR. OKRENT:
Mr. Ryder must understand the 18 question because I can see him smiling.
19 Do you want to explain it as you understand 20 it, Mr. Ryder?
21 MR. RYDER:
Leon Ryder, Earth Science Branch.
1 22 I believe Dr. Okrent is referring to the 23 apparent lack of seismicity in the upper Midwest, and
(
24 the staff 's use of that data to draw licensing 25 conclusions, and how that would be affected if we had a O
ALDEhSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
-- -. ~
7
()
1 magnitude of 5.8 occurring there, how much of a surprise 2
would tha t be as compa red to occurring in New Brunswick
/~T 3
which was a surprise, and what kind of impact would that V
4 have upon us.
Would we, therefore, have more studies.
5 I Juess the bottom line is, what kind of effect would 6
that have upon the licensing conclusions that the staff 7
draws eventually.
8 MR. OKRENT:- You certainly have hit much of 9
the thrust of the question.
I guess I am trying to 10 ascertain whether or not we are placing enough priority 11 in examining earthquake magnism to the central stable 12 region so that va can judge the one in ten thousand, the 13 one in a hundred thousand, or if this is an impossible O
14 task, or whatever.
15 Do we have to wait until an event occurs that 18 f orces you to place a lot more attention there, but 17 belatedly?
l 18 HR. HURPHY:
I think what we are doing now is l
l 19 looking at methods for determining ear thq ua ke s 20 recurrence intervals in a somewhat generic sense, not 21 necessarily per se at New Brunswick or Michigan.
22 We are looking at the process for determining l
23 that, and looking to see in the program whe ther or not
()
24 there are special areas, special case areas that may be 25 aseistic for a long periods of time such as Wisconsin, l
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
8
()
1 or areas that have low seismicity, and where you come up 2
with an unexpected earthquake such as New Brunswicks 3
In those areas, as I understand the program,
{}
4 we are looking at that type of probles.
So that when 5
this is finished, if there are special problems 1.-
thi7 6
area, special types tectonic problem areas, special 7
styles of tectonic problem areas, that can be then 8
looked at and the initial problem has been solved -- not 9
necessarily solved, but examined in detail.
10 MR. OKRENT:
My recollection of some of the 11 stress patterns that we saw at that two-day meeting in 12 Reston were that there are considerable differences 13 whether you are east of the Adirondaks or west.
I am I ()
14 just trying to ascertain whether you should have some.
15 kind of a program whose aim it is to look at improbable 16 earthquakes in the central stable region, or can you 17 think of one that is practical?
18 I am not recommending a specific program.
I 19 as trying to see whether you have given it the same kind 20 of focus tha t you would have if a magnitude of 5.8 21 occurred there, in which case I know very well it would 22 receive a lot of your attention.
I 23 MR. MURPHYs I think that it is a question
()
24 that the earthquake has not occurred, and where to focus i
25 in Michigan or in Wisconsin would be difficult.
i l
i ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
9
()
1 MR. OKRENTs I don't want you to focus on any 2
specific location.
(}
3 MR. MURPHY:
No, sir.
In the less seismic 4
staa of the United Sta tes, such as the central stable 5
region, there is acre that we can learn about where the 6
earthquakes are occurring, and if there is any clue in 7
this expected earthquake in New Brunswick that may be l
8 applicable to the less active regions at this time.
9 MR. BERATANs For the '84-85 budget proposal, 10 we show for seismology and geology about $4.7 million 11 for
'84, and about $5.6 million in
'85.
This is broken 12 up between regional studies, which include Southeastern 13 United States, Northeastern, New Madrid, Nemaha, the O
14 Pacific Northwest, and some improvement of l
15 instrumentation analysis.
Is In some of the topical study area, we will l
17 initate some studies in site dependent response spectra, 18 and analysis of strong ground motion, generic 19 attenuation problems, and the ascertainment of 20 uncertainty in ground motion estimates..
21 Some of the generic studies are the LG 22 sttentustion and near ground field effects, and 23 recurrence.
Some of the strong-motion seismograph data 24 reduction will require and install an array of strong 25 motion instruments in the Eastern U.S.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
10
()
1 The USGS topical studiesinclude such things as 2
strong ground motion, falling studies, assessment and 3
evaluation of probability of occurrence.
(}
4 Some of the geo-technical engineering studies 5
are some soil structure and settlement studies, and some 6
liquefaction studies.
7 In meteorology and hydrology, we will be 8
spending about $1.7 million in 1984, and about $1.4 9
million in 1985.
These are a variety of topical 10 studies, and the winding up of the severe storm work, 11 also the last of the atmospheric dispersion field tests, 12 and the atmospheric model data evaluation from the field 13 tests.
4 O
14 In the siting concepts are, althought the 15 heading is " flooding," it is really just one flooding.
16 study, which is a research study that we are 17 collaborating with the Corps of Engineers in at the 18 University of Florida, which is a study of wave run up i
19 as a result of hurricanes.
20 He have initiated this year a study of 21 groundwater contamination as a result of the Class 9 22 accidents.
The purpose of this study is to investigate 23 mitigative techniques, to examine the generic site 24 conditions for the control and reduction of groundwater l
[
l 25 radionuclide contamination, and identify those factors C) l l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D C. 20024 (202) 554 2345'
11
()
I which affect the off-site release and possible 2
interdictive options.
p**
3 That is the program for the Branch in 1984 and 4
1985.
5 MR. OKRENT:
If you had to cut $1 million?
6 MR. BERATANs In what year?
7 MR. OKRENT:
In 1984 8
'MR.
BERATANs I would probably cut the 9
atmospheric dispersion test, which*is about a 5900,000 10 item.
11 MR. BENDER 4 What would you lose thereby?
12 MR. BERATANs The validation of some of the l
l 15 models used for emergency preparedness.
CE) l 14 MR. MARKS These are tests, and not code 15 development?
16 NR. BERATAN That is right.
These are field 17 tests.
They are actual releases.
18 MR. MARKS In a variety of terrains?
19 MR. BERATANs Yes.
20 MR. MARK What can you release that will be 21 detected?
22 3R. BERATAN:
It is some kind of sulfu 23 hexafluoride.
()
24 MR. MARK It doesn't sound very helpful.
25 MR. BERATAN4 It is not toxic, so I am told, I ()
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE S.W., WASHINGTON D.C. 20024 (202) 554-2345
12
()
1 and Iry, you can reinforce me on that.
2 MR. SPICKLERa It is non-toxic and it is 3
detectable in the parts per billion range.
It is not
{}
4 normally present in nature, to it is a wonderful tracer 5
material.
6 MR. MARK 4 You have not given any thought to 7
refrigerated methane?
8 MR. BERATAN:
No.
9 MR. MARK:
You can get that from Los Alamos.
10 They make it for this purpose.
It has a molecular 11 weight of four more than the other stuff, and you can 12 pick it and you can handle it.
It is also obviously 13 non-toxic.
O 14 HR. BENDER:
You wouldn't have the data --
15 MR. BERATANs We would probably be a little 16 more conservative.
17 MR. SPICKLER:
I wonder if I could answer 18 that.
19 I think the bottom line is to try and 20 determine what is the minimum program that is required.
21 I know that industry may scoff at what I have just said, 22 but the bottom line is to minimize the programs that are 23 absolutely needed, and that is what the purpose of 24 diffusion tests is:
What minimum instrumentation is 25 needed, and wha t models work and what models don 't O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
13
()
1 work.
Without the test, we would probably have to go to 2
more of a burden on industry with regard to more
(]}
3 conservatism.
4 MR. BENDER:
It seems to me that in the past 5
we have discussed this matter.
Things like the 6
postulated plume would come up and how that might i
7 influence the content of the plume, the cha racteristics 8
of the aerosols.
They all enter into the question, i
l 9
ESSEX doesn't model very much of that, if any of it.
10 MR. SPLOKLERa It does a very good job for the 11 gases.
12 MR. BENDER That is the least troublesome.
l 13 ER. SPICKLER:
If we are talking about some
)
14 micro-particles, the aerosols would behave like a gas.
l 15 IF you are concerned about iodine removal, there is 16 nothing that that will model iodine other than iodine.
l 17 MR. OKRENT:
If you had a million more, where 18 would you put it?
19 MR. BERATAN If I had a million more, I think 20 I would expand the geology and seismology programs, and 21 make them a little more intense.
22 MR. MARKS What is a single stress 23 measurement?
24 MR. BERATANs It depends on whether they are 25 shallow or deep stress measurements.
If you are talking O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE, S.W., WASHINGTON, D.C. 20024 (202) 554 2345
I 14
()
1 about deep stress measurements, you may be talking about 1
2 a couple or three thousand dollars per hole.
We are 3
going to go into a cooperative program with the GS on
)
4 some shallow hole and we expect only to spend in the 5
neighborhood of $50,000 for those.
6 MR. MARKS This is an area where you might l
7 easily use up money and also get inf ormation.
8 MR. BERATANs Yes, we plan to do this 9
cooperative study in New England because there is a 10 region up there where the stress fields apparently have 11 reversed themselves or turned about 90 degrees from 12 where we expect them to be.
We feel that some stress 13 measurements up there may be would do us some good in O
14 understanding what is happening.
15 MR. OKRENTs Any other questions at the 16 acaent?
17 M'R. BERATAN:
The detailed descriptions are in 18 the handout.
19 MR. EBERS01E:
When you make a stress 20 measurement, how well do you know its dispersion?
21 MR. BERATAN:
I will have to defar to my 22 stress expert, Tom Schmidt.
23 MR. SCHMIDT:
Usually a stress measurement is
()
24 at both fracture ring and over a boring, and a suite of 25 seasurements is taken.
Commonly there is a dispersion O
ALDERSoN REPORTING COMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554 2345
15
()
1 of about 10 degrees in the direction of maximum 2
depression.
But a suite of th ree, five, 10, 15
(}
3 individual measurements are averaged together.
Commonly 4
those in the industry maintain that they are five or ten 5
degrees absent of a direction.
6 MR. OKRENT4 Why don't we go on to the next 7
item.
8 MR. RICHARDSON:
My name is Jim Richardson.
I 9
am head of the Mechanical Section of the Mechanical 10 Structural Research Branch in the Office of Research.
11 I would like to have you refer to the last 12 page of the handout, since this is the same handout that 13 will be used later by Mr. Guzy when he presented the O
14 details of the SSERP.
But for the purpose of presenting i
15 an overview of the SSMRP, I would like to refer to the l
16 budget sheet, which is the last sheet.
17 Our budget ever since 1980 has been on the 18 order of $3 million per year.
In 1982, this has been 19 decreased to $2.8 million, and in FY-83 it again will be 20 about $2.8 million, and in FY-84, which is the scheduled 21 com pletion of the program, it has been reduced to $1.9 22 million due to some internal budgetary decisions within l
23 the Office of Research.
24 We are a,1so funding, in cooperation with NRR, 25 a seismic hazard characteristics project that is being O
l ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
16 (O
,j 1
managed by Mr. Berstan's organization.
We have provided 2
the funding for that out of the SSMRP of $110,000 in 3
FY-82, and $205,000 in FY-83.
4 The handout gives a breakdown of the SSMRP 5
funding for 1982, 1983, and 1984.
You will notice that 6
we have changed the emphasis from the past organization 7
of the program, during the time we were developing the 8
methodology, when the projects were along disciplinary
(
9 lines.
Now tha t the methodology has been developed, and to we are in the study stage, we have reoriented the 11 program toward products as we focus in on'the completion 12 of the program in 1984.
13 The San Onofree effort that we were assisting O
l 14 NRR with was terminated early this year.
We are
{
l 15 continuing with our sensitivity studies presently, and 16 are completing the Zion analysis, which will be 17 completed at the end of this year with the report due 18 out early next yest.
We are in the process of 19 developing simplified methods, and proceeding with the 20 validation task to validate the methodology.
We will 21 then apply that validated methodology to the PWR plants, 22 to make sure the methodology is generic in its nature.
23 We are also involved in technology transfer, 24 tha t is making the technology available to the NRC 25 staff, as well as the general public.
Of course, we
(:)
l ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345 t
l
17
()
I have the general management efforts that are associated 2
with the program.
3 This is a summary of the budget.
I would like 4
to add as part of the overview that the program has been 5
significantly reoriented as a result of meeting with our 6
senior research review group, which was instituted at 7
the suggstion of the ACRS.
We met with that group 8
twice, in November and in January.
We will meet again 9
with them in August.
10 As a result of discussins with them, the 11 program was reoriented to be more responsive to NRR 12 needs, and the emphasis has shif ted in the SSMRP from a 13 program that was originally focused on determining O
14 margins to more risk oriented methodologies that would 15 be used to validate incoming PRAs from licensees, and 16 will be used to gain whatever insights we can to direct 17 further research in the area of seismic problems.
18 We have one product program that is now 19 a va ila ble, the SMACKS program, and the users manual has 20 ben written.
It is available now for use by the NRC 21 staff, and the public.
22 Last month, we met with our senior 4
23 consultants, and are in the process of digesting.
Those
()
24 comments, along with the interaction we have had with 25 our senior research review group, have gone into the l
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
I' 18
()
1 construction of the program plan for the remaining part 2
of the program.
We anticipate that this program plan 3
vill be in its final form by next month, where we will
{
4 review it, hopefully for the last time, with the senior 5
research review group, and then proceed to a new 6
program.
7 I believe that this is all that I would have 8
to cay on the overview point of view.
Mr. Guzy will be 9
presenting more details later in,the agenda.
10 MR. OKRENT:
From sn overview point of view, 11 does there exist or will there exist a critical 12 avslua. tion of what is included in the current SSMBP 13 methodology?
What are the strong and weak points of O
14 wha t is included?
What is not included of probably 15 significance?
16 MR. RICHARDSON:
We rely heavily on our senior 17 consultants to give us guidance in this direction, as 18 well as the senior research review group, and we also 19 try to be ss self-critical as we possibly can, to 20 identify where our weaknesses are and where our 21 strengths are.
22 Wo see s very definita need to sivartise what 23 a re the bounds of the program, and what are the
()
24 limitations of the program, last anyone misuse the 25 program.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
19
)
1 MR. OKRENT4 Is there such a document now?
2 MR. RICHARDSON:
No, but there will be.
()
3 MR. OKRENT4 Was such an item part of the i
4 discussion of the senior review group?
5 5R. RICHARDSON By item, do you mean the 6
publication of a document?
7 MR. OKRENT No, sort of a very thorough 8
discussion of the strengths, weaknesses, and gaps in the 9
methodology?
10 HR. RICHARDSON:
Yes, I believe.
11 MR. EBERSOLE:
May I ask one question?
12 Does that review group contain any 13 instrumentation that look at the high frequency end of O
14 the spectrum, and the susceptibility of a great many 15 pieces of apparatus in the plant that are absolutely 16 vital to the continuity of power, and so forth?
17 MR. RICHARDSON:
No.
The research review 18 group, internal of the NRC, is made up of top management 19 people, who are basically the division managers from NRR 20 snd the Office of Research.
Our senior consultants are 21 primarily systems oriented, probabilistic oriented, and 22 structures oriented, and geology / seismology.
23 MR. EBERSOLE:
I raise the question because 24 the consultants ough t to raise the question of whether 25 or not the proper input spectrum has been going into O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
20
()
1 i t.
I think that this is a justified concern due to the 2
arbitrary cut-off of some of the higher frequencies
(~)
3 which may result in a shortfall of such apparatus.
V 4
HR. RICHARDSON:
Yes.
In fact, there is a 5
correlation between the SSMRP and other seismic programs 6
in their relationship to our equipment qualification 7
programs.
l l
8 HR. EBERSOLE:
That is the one we are related 9
with.
(
10 HR. RICHARDSON: We see a very close 11 relationship and we are looking to the SSHRP to help us 12 define and characterize the seismic input f or equipment 13 qualification.
O 14 HR. EBERSOLE:
If we have the wrong input, we 15 will have weaknesses all across the board.
16 HR. RICHARDSON:
I agree with that.
17 HR. OKRENT:
You have a topical valida tion, 18 and the question I was raising relates to that.
I am 19 not quite sure what validation means.
What does the 20 term "ve.lidation" sean?
21 HR. RICHARDSON:
Certainly, there is no way 22 that we can validate the whole program with one 23 experiment.
I know of no way of doing that.
We can
()
24 validate subparts of the progrsa.
We can validate the 25 data as best we can.
A validation effort, in some ALDERSON REPORTING COMPANY,INC, 400 VIRGINtA AVE., S.W., WASHINGTON. O C. 20024 (202) 554-2345
~
21
()
1 areas, is based on experimental data and other places it 2
will be based on comparison with other known 3
methodologies.
)
4 MR. OKRENT:
It takes a large part of the 5
money.
6 MR. RICHARDSON:
Yes.
7 MR. BENDER:
Could I get a little better 8
understanding of how the consultants have advised you.
9 What things have they' suggested you do that you were not l
10 doing before, or vice versa ?
11 MR. RICHARDSON:
They recommended more 12 emphasis on validation.
13 MR. BENDERa I think that is too broad a
()
14 term.
Which validation of what?
15 MR. GUZY Dan Guzy of the Structural 16 Validation Branch.
17 Yalidation, as we were directed or as was i
18 suggested to us, is a broad all encompassing term and it 19 is not any specific items or overall type of thing.
It 20 is a combination of looking at what physical data ve 21 have, referring to the assumptions and the data bases 22 used, and the methodology.
To look at really what is 23 important in the calculations.
24 We are depending heavily on the sensitivity 25 studies to tell us what parameters would assess, what O
ALDERSON REPORTING COMPANY,INC.
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
22
()
I components are important to the final uncertainty 2
numbers.
What is driving the risk numbers off, which 3
systems, wha t methodology?
It is an overall view of 4
what'is dominating your answers, and then concentrate on 5
those areas, and try to get as much physical benchmarks 6
and data with that program.
7 We cannot make an earthquake occur, or a 8
series of earthquakes and get some statistical data 9
validated.
So it is sort of a catch-all and that is the 10 direction we have been going, it is to look at the 11 various research that is going on, the results in this 12 country and other countries, and to compare that against l
13 the assumptions and methods that have been used.
14 MR. BENDER:
Has the emphasis been placed on 15 soil structure behavior, or on the intermediate i
l l
16 structure, or on the components themselves, or is there I
17 a distribution?
18 MR. GUZYs I think we are depending on the 19 sen sitivity studies to tell us where to concentrate.
I 20 Right now we know the piping responses dominate the l
21 total response of the chain of the earthquake, and the 22 building response.
Right now that is where we would 23 look at it, in terms of what is available, and compare
()
24 that against physical data.
25 MR. BENDER:
Dr. Bush suggested that we try l
ALDERSON REPORTING COMPANY,INC.
1 l
400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) $$4-2345
23
()
1 strain control evaluation methods, does that make any 2
sense?
3 HR. GUZI We would have to have the data
/)
4 base.
5 MR. BENDER:
In the absence of it, what do you 6
do?
7 MR. GUZY:
In the absence of it, you ao better 8
analysik, looking a t non-linear building response, 9
non-linear soil structure interaction, and looking at l
10 the ductility factor that is used to understand your 11 linear analysis in the non-linear range for items 12 identified right now, and things to look at now and in 13 the future.
O 14 Again, we want to base our strongest part of 15 our effort on the things that are shown to be important 16 both through the sensitivity study results and the final 17 numbers out of the Zion calculations, which were 18 completad on sensitivity bounds and will be used in our i
19 final models.
20 MR. BENDER Are you trying to get the damping 21 characteristics defined better as a function of the 22 degree of non-linearity that is being anticipated in the 23 structures?
()
24 HR. BONDS Mike Bonds.
What we are trying to 25 say is to take a look at each of the different areas O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
24 1
that are in the calculations, and we have proposed 2
validation in each of those areas.
f])
3 In some cases, we would be comparing our codes 4
against other accepted or equivalent codes.
In some 5
cases, we would be comparing building response and 6
natural frequency, and damping characteristics, as you 7
have suggested, against experiments that have been run 8
at MIT to get an idea of whether the damping ratios at 9
the higher accelerations were correct.
10 MR. BENDER:
You are talking about the 11 question I was asking.
I asked about whether or not 12 linear behavior --
l 13 MR. BONDS In the area, we have proposed two l ()
14 major topics -- three actually.
The first is to look at 15 non-linear soil structure interaction.
In the SSMRP, we 16 have the tools, and we want to oxercise them and compare 17 them.
l 18 MR. BENDER Stop right there.
Do the 19 consultants you have think that it is a good approach, I
20 have they concurred with it?
21 MR. GUZY:
Yes.
They were very suptor ting of 22 doing this type of research.
23 MR. BOND:
The other ares that Dan mentioned 24 was the reduced ductility approach which is being used l
25 in piping and building.
O ALDERSON REPORTING COMPANY,INC.
400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554-2345
25
()
1 MR. GUZYs We are trying to evaluate all the 2
physical data we can.
In our experimental program, we 3
vant to benchmark it as much as we can against what is
[}
4 svailable.
5 MR. BENDER:
You mentioned the Zion analysis.
6 MR. BOND:
The effort now is to do the final 7
calcula tion in the September / October time f rame for 8
presentation next year.
9 MR. BENDER:
Who is doing that?
10 MR. MARKS Are there any buildings in Turkey 11 or Italy that make some strange pipes that you can 12 calculate?
13 MR. GUZY:
Unfortunately, most of the data, O
14 lite the El Centro earthquake, my understanding is tha t l
l 15 right now it doesn't look like there is a lot of i
16 earthquake that we could use.
Naybe Mike can answer 17 that better.
18 MR. MARKS I think of validation as really 19 requiring something of that kind.
Your finishina the 20 Zion calcula tions doesn't mean that you validate 1
21 anything. it just finishes Zion.
22 MR. RICHARDSON:
We would not consider that l
l 23 validation.
()
24 MR. GUZY:
Our finding is that the piping l
25 f ailure domina tes.
It is much more important than
()
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
26
()
1 findings in soil structure.
We are looking to the 2
questions for the piping.
We are looking into the 3
available data.
We are using the analyses, which a
perhaps suggest redirecting the other programs for data 5
and to look into the areas of what is really dominating 6
the final answer.
7 MR. MARK That I understood.
One of the nice 8
things that is going to come out, one hopes, is that we 9
vill know where the soft spots are.
10 MR. GUZY:
The sensitivity studies and the 11 Zion results will tell you where are the uncertainties, 12 and those are the areas that we would like to 13 concentrate on during the last two years:
To validate O
14 and make sure our answer is right, and see what we can i
15 do to reduce the uncertainties that were introduced in 16 our modeling by finetu. ling them, and concentrating our 17 efforts
~-
18 MD. OKRENTs Do you have a pretty good feeling 19 for what are the weak points and the omissions in what 20 SSMRP is doing in the Zion analysis?
21 MR. GUZY:
Not at this point.
22 MR. OKRENT4 If you don't, how will you be 23 able to tell that the output of the analysis in fact
(
gives you the things on which you concentrate?
For 24 25 example, they might find a high probability of a large O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON. D.C. 20024 (202) 554 2345 l
1
27
()
1 LOCA, and they might find a very small probability of 2
any LOCA, but in either case the answer might depend on 3
some assumptions concerning how they find you get a 4
LOCA.
5 MR. GUZYs The censitivity studies will look 6
at the various parameters and tell you if you fragility 7
curve moved so much, the pipe was moved so much, and 8
your probability of failure went up so much.
9 MR. OKRENT:
Sensitivity studies are useful if to you have an idea that you are in the right range, as it 11 were.
But if you don't know if you are in the right 12 range, you can be fooled by a sensitivity study.
You 13 can end up by being led to the wrong conclusion.
O 14 I will not belabor the point except that I
(
l 15 would urge you, sooner rather than later, to get a list l
l 16 of what you think is good or bad about the existing 17 method, and also its specific application, because it is 18 one thing to have a method and it is another thing to 19 know how many pipes you have modeled and how well.
20 We had better go on to the next part of the 21 overview.
22 MR. BURDICK:
I am Gary Burdick.
I have no 23 handout, and I will be very brief.
()
24 What I would like to bring to the l
25 subcommittee's attention is an area that my branch, the
()
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
28
()
1 Reactor Risk Branch, will get involved in starting in 2
the next fiscal year.
We plan to do an IREP s.udy of 3
the BWR with a Mark II containment, and using the newly 4
developed COMCAN III code do a very thorough common 5
cause failure analysis.
Using the same approach, we 6
will try to get into the seismic analysis.
7 We believe that this will benefit the SSMRP 8
people.
It will give them a different approach that 9
they could compare theirs with.
It could possibly be 10 that a marriage of.the twc approaches might be in fact 11 the way to go, although we are not sure.
12 Ihis is a research effort, and we plan to do l
13 quite a bit O
~
in-house.
COMCAN III will start working for 14 the NRC, for our division, on the'19th of this month.
i 15 This is all I have.
16 MR. OKRENT:
We will go on to the next item, 17 which I guess'is the one related to RES plans for 18 res'earch directed 'at a probabilictic characterization of 19 seismic and flooding hazards and hazards from internal 20 and external flooding.,
21
.HR. JOHNSON:
My name is James Johnson.
I am 22 with the Dsta and fethodology Branch, Division of Risk 23 Analysis in the Office of Research.
I, too, will be
()
24 very brief.
I_ vill try tofsummarize in.a very general 25 manner what Research's plans are in the flood risk ALDERSON REPO9 TING COMPANY, NC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345 s
29
()
1 assessment area.
2 I would like to just begin by briefing saying 3
that the Division of Risk Analysis has been involved in r~ )
4 1 flood risk assessment, and we have had basically two 5
efforts on-going.
The first one had to do with the 6
problem of estimating flood occurrence probabilities.
7 Before we got into this effort, we realized 8
that for many years individuals have tried to predict 9
flood occurrenc'e probabilities from a purely statistical 10 manner, just by basically fitting distributions, and i
l 11 trying to extrapolate the detail.
12 We felt that in order to estimate the small 13 probabilities, there had to be some additional input.
O 14 So our effort was to explore the use of the so-called 15 "BEJA" method.
What this effort resulted in.vas a 16 computer program that would allow one to calculate 17 extreme flood probabilities.
18 I must say that this effort did not proceed 19 f ar enough along for us to be convinced that it provided 20 a viable alternative for forecasting floods.
Additional 21 work would have to be done to determine the parameters 22 of the model, which would necessitate various l
23 assumptions.
()
24 MR. BENDER What parameters?
25 HR. JOHNSON:
In addition to the historical ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
30
()
1 records, estimates of what the probabilities were at 2
particular discharge levels.
For example, suppose you
{' )
had a dis =harge level of 4,000 cubic feet per second, 3
4 you would have to input, in addition to the historic 5
record, bounds on that to exceed the probability at a 6
particular discharge level.
7 NR. BENDER:
I ha ve a ha rd time relating to 8
that, but I will pass for now.
9 HR. JOHNSON:
In addition to the forecasting 10 procedure, we also had an effort underway to analyze the 11 effects of firading.
12 MR. OKRENT:
Before you go to the next topic, 13 in a sense, the method you were just discussing was in c
O 14 the form where you could fit in expert opinion.
15 HR. JOHNSON:
That is correct.
16 MR. OKRENT:
That is not radically different 17 in principle from what the SSMRP program did with regard 18 to seismic hazard and with regard to what is called 19 seismic fragility.
20 Although it is not done in an identical 21 fashion, the SSMRP program decided apparently to rely 22 rather heavily on this expert opinion approach.
In 23 fact, I think the regulatory staff relies fairly heavily 24 on the expert opinion approach when it gets consultant 25 reports on different sites on which they give opinion, O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON, D.C. 20024 (202) 554 2345
31
()
I however the question is phrased that this group 2
answers.
3 I as trying to understand why it was or is 4
that the expert opinion approach seems to have been sort 5
of adopted on the one hand in the seismic a rea, and why 6
it seems not to have flowered for flooding.
Do you have 7
any thoughts on that?
8 MR. JOHNSON:
I would not go so far as to say 9
that it did not flower.
We essentially stopped work in to this parti:ular area because of a number of reasons.
I 11 think that a lot of additional work would have had to be 12 done in this particular area before we would have had 13 anya co'nfidence in the results using the procedure.
b
'd There is a lot of problem with using it 14 j
15 because the output that you get from the code is very l
16
- dependent upon the discharge level at which You select 17 to provide these subjective inputs.
So there are a lot 18 of sensitivity studies that would have to have been done 19 before we would have called the results believable or 20 credible.
It is just that it wasn't pursued far 1
21 enough.
22 MR. OKRENT:
I am not offering a judgment that l
23 one is not believable.
()
24 MR. BENDER:
Just as a point of clarification, 25 the historical records that you use go back to when ?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
32 '
()
1 MR. JOHNSON:
Typics11y, it would be less than 2
100 years.
We were.using a historical record of about 3
87 years.
In ganarsi, they probably sverage 50 to 60
)
4 years.
You can get some records that are 100 year.a.
5 MR. BENDER 4 I am trying to get some 6
exceedance levels at a higher level of probability, and 7
you might normally look for some place where there is a 8
longer historical record.
Is that never done, even 9
though you don 't have to move from one place to 10 snother?
11 MR. JOHNSON:
Yes.
There is a concept called 12 regionslizstion where we try to transfer information 13 from one location to another.
O 14 NR. BENDER:
Is that in your existing 15 program?
16 HR. JOHMSON:
No.
17 MR. BENDER:
Why isn't it?
18 MR. JOHNSON 4 We just cut things off before we l
19 proceeded, before we got very far.
This was an in-house 20 effort, and we were doing it with a small staff.
We did l
21 not put a lot of time and dollars into it.
22 MR. EBERSOLE:
I take it that floods can come 23 from dam failures which may be in cascades, and you do
()
24 look at that, or am I wrong?
25 MR. JOHNSON:
We had a mechanism in there that O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
~
33
()
1 was sort of a first order effect to try to account for 2
the effects of das breaks, but it was by no means a very 3
sophisticated procedure.
4 MR. BENDER:
You are currently relying on 5
probabilitistic assessments?
6 MR. JOHNSON:
That is correct.
7 In addition to the flood occurrence 8
probability, we have had a program on-going to analyze 9
the flooding hazards.
We basically assumed that the 10 water was inside the plant.
There we looked at how the 11 systems can respond given the depth of water.
12 For FY-82 and FY-83, we have no research 13 planned in the flood area.
! ()
14 MR. MARKS Did you say 1982 and 19837 15 MR. JOHNSON:
Right.
16 NR. MARK:
You are going to tell us about 17 1981, 1982 and 1983, or 1984 and 1985?
18 MR. JOHNSON:
I w as telling you about 1981.
19 In the current year, nor in 1983, are we 1
20 planning to do any flood research.
Again, this is due l
21 to budget limitations.
For 1984 and 1985, we are 22 planning on spending about $500,000 each year.
l 23 Of course, one of the primary tasks will be to
)
research this area of flood frequency versus severity.
24 25 We plan to establish a fairly comprehensive program.
It i
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
33
()
1 will be different from the RFP that we have on this.
2 It will not be quite as ambitious, but we will 3
look at as many known promising procedures as we
{
4 possibly, including procedures that are primarily 5
statistically based.
We will also look at procedures 6
that are based on physical models to try and determine 7
the things that contribute to these extreme floods.
We 8
will also look at extreme flow precipitation to try to 9
extend the current records, and any other procedures 10 that appear promising.
11 In addition to this f requency eff ort, we will i
12 also do some flood risk screening to try to characterize 13 reactor, sites by the type of threats posed by floods or
[
)
I 14 severe storms.
We will also develop additional 15 probabilities of floods because of certain important 16 plan t saf ety functions.
17 All of this information will be pulled l
18 together to try to develop methods for incorporating 19 flood risk procedures in the reactor risk assessment.
20 The basic objective is to determine those plants at 21 which off-site floods or storms are contributants.
22 The actual details of this research plan will 23 be worked out with, hopefully, what will be a research
()
24 review group that we will assemble to make 25 recommendations on directions in which the program O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
i 35 i
()
1 should proceed.
2 MR. OKRENTs What fraction of the $500,000 per 3
year is goes into things that contribute to yoir
(}
4 improved knowledge of the likelihood of increasingly 5
higher floods as contrasted to the effects on the plant 6
given the flood?
7 HR. JOHNSON:
I would say the majority of it 8
would'go into the problem of coming up with frequency 9
versus severity, or'at least 75 to 80 percent.
10 MR. OKRENT Do you think that by the end of 11 the two years, you are reasonably likely to have a 12 methodology to include an IREP like study on whose f
13 uncertainties are not so large that it is not too
()
l 14 meaningful to include it?
15 I know that it is crystal-balling, but what is 16 your expectation or hope?
17 MR. JOHNSON:
I would think that at the end of 18 the two years, we will have a methodology that will be 19 probably much better than anything tha t is available 20 now. But whether it will give us some uncertainty levels 21 on the order that we would be comfortable with is a 22 whole new question.
23 ER. MARKS You mentioned $500,000 for those
(
24 Years on flooding.
What is the difference between that 25 and wha t I find in this Earth Science research, where it O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
36
()
1 is $150,000 on flooding this year?
2 MR. JOHNSON:
I am not sure what the $150,000
(}
3 is.
4 MR. OKRENTs The $ 150,000 is from Leon 5
Beratan 's program, I think.
I think Mr. Johnson is 6
talking about a different program, but both within the 7
Office of Research.
8 MR. MARK:
In the Office of Research, but not 9
in the Branch of Earth Sciences.
10 MR. JOHNSON:
Right.
11 MR. MARK Why is it advantageous to have 12
$150,000 on flooding over here for these guys as opposed 13 to you?
7,V 14 MR. JOHMSON:
That is a question I cannot 15 answer.
I l
16 MR. EBERSOLE:
I heard you say that you are 17 going to look at the effects of the floods on the 18 plants.
19 MR. JOHNSON:
Yes.
20 MR. EBERSOLE:
This will be how e xpensive?
l 21 You mentioned characteristics, and where you 22 get a floci can be an important aspect of the flood.
23 When it gets to a certain point, there is a cut-off and 24 it doesn 't much matter how much the plant gets.
25 MR. JOHNSON:
We will be looking at it as a O
l ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W, WASHINGTON, D.C. 20024 (202) 554 2345
37
()
1 function of flood.
2 HR. EBERSOLE:
Will you be looking at
{)
3 forecastability of floods to determine whether there is 4
any advantage in a set-up which enables the operator to 5
anticipate trouble in advance and do something about 6
i t?
7 MR. JOHNSON:
Yes, we will be doing some of 8
that, too.
9 MR. OKRENT:
Thank you.
10 The.next item, I guess, is Ea rth Sciences, 11 again, if I understand the way the agenda is arranged, 12 but I believe Mr. Beratan has left.
13 Let's go with Mr. Guzy.
)
14 MR. GUZY:
I will be talking from the same 15 slides that Jim passed out.-
The first slide pretty much 16 talks about the development of the planning this year, 1
17 an what is going on in the planning.
I think Jim 18 explained that pretty well.
19 I guess the point of being redirected to 20 pursue validation was at NRR's request.
They asked us 21 to emphasize the project in that area.
There was also 22 an express need for making the methodology simplified, 23 generic, and available as soon as possible for internal 24 use.
A lot of things that we have done to the program l
25 are now directed toward that.
O l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554-2345
38
()
1 ER. MARK 4 You call that technology transfer.
2 MR. GUZY:
Technology transfer, also the
/~T 3
simplified methods, and the new project we have cut out, V
4 the seismic characterization project which makes seismic 5
hazard data for the Northeastern United Sta tes available 6
for use in the PRA type analysis.
7 Dur work in the past has concentrated on Zion, 8
and the spin-off from the work that was done on the site 9
specific project, and this will round up the Eastern 10 United States as far as the data to be used by the 11 SSMRP, and perhaps for other industry PRAs.
12 MR. OKRENT:
There as a certain information 13 letter on the results of some work done in Germany where O
14 they tried to predict strains and frequencies for 15 control tests, HDR tests.
How is that specific research 16 factored into the SSMRP program, and how decisional can 17 it be?
18 MR. GUZYs Perhaps Mike can address this 19 better, but I think it is factored both as an input, and 20 also as part of the validation.
I think the findings of 21 the SSMRP have shown that the uncertainties in piping we 22 are sponsoring tend to dominate the total response 23 chain.
I think the kind of numbers of the validation
()
was the direct input.
24 25 MR. BOND:
The work was done in two phases.
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINTA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
l l
39
()
1 The first part was done essentially blind, and there 2
were discrepancies.
There was a closer evaluation, and 3
the factor of two was still there.
As input to our
)
4 piping calculation, we used a coefficient of variation 5
of about 0.77, which is not inconsistent with that.
We 6
were aware Of that when we were formulating our input 7
coefficients for variation.
8 We would like, in the validation phase, to go 9
back and look at the HDR test.
Right now there is a 10 shaker test going on at that facility, which we would 11 like to input into our validation program for piping.
12
- 58. MARK What was the general nature of the 13 test?
O 14 MR. BOND:
It was a decommissioned plant.
15 They had problems with their fuel, so they 16 decommissioned the plant after only several hundred 17 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br /> of operation, and turned it into a research 18 facility.
It is the secondary piping that they mounted 19 shakers on.
20 MR. MARK:
And they excited the piping at 21 different points.
22 MR. BOND:
There was a low level explosion 23 outside, bat there was little level input.
()
24 MR. MARK They were trying to make a 25 calculation of those, and that is what you find as a O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
40
()
1 factor of two in the data, or what?
2 MR. B3ND:
Compsring the calculation with the
(])
3 data.
4 MR. MARK:
Did you calculate twice as much or 5
half as much?
6 MR. BOND:
I can't tell you.
7 MR. RICHARDSON:
In general, it tended to be 8
under predicted, but there were cases of both.
9 HR. MARK The prediction means that the 10 calculation is lower than the actual.
11 HR. RICHARDSON:
That is what concerned us, 12 and we were consistently conservative in our 13 calculations, but that was not the case.
O 14 MR. OKRENT:
I did not get a clear feeling 15 before on what steps have resulted in a validation of 16 the SSMRP.
17 MR. GUZY:
What we tried to do is to look at i
18 what is very important.
The second steps is to look at 19 wha t physical data exists.
I think we are trying to 20 identify 21 MR. MARK:
What is this test?
22 MR. GUZYs That is the end-point test tha t is 23 done by EPRI.
The Division of Reactor has a shaker test 24 on the feedwa,ter line, and it is the same kind of thing 25 as HDR, but more data, more comparison.
EPRI is also O
I ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
41
()
1 doing some shaker-table test of the piping systems.
In 2
a lot of cases, it is looking at what is available now, 3
and how are their results, and how are their findings.
4 MR. MARK:
I understand comparing with the 5
tests.
6 ER. GUZY:
For the results of an earthquake, 7
you have to look at everything in the chain.
If you 8
have a test on a certain type of a pipe, you compare 9
your input.
If you have one earthquake and one result 10 on a plant, it doesn 't give you a whole lot as far as 11 statistics, if you are not looking at it in enough 12 detail.
13 MR. MARK:
You don't have any messurement.
\\
~
14 MR. RICHARDSON:
I wanted to add some new I
15 information.
I have just returned from a three-week 16 trip in Japan.
We had some very interesting discussions 17 with both the Minister of International Trade and 18 industry, and the Safety Technology Agency which is 19 responsible for safety research.
l 20 Japan is embarking on probabilistic risk l
21 analysis for the first time.
They are very interested 22 in joining with us in some cooperation and using the 23 SSMRP.
They are also interested in supplying some test l ()
24 data to us to help validate the SSMRP, primarily shake 25 table data, probably not the large shaker table at O
ALDERSON RE"oRTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
42
()
1 Tadatsu, primarily because the Tadatsu shaker-table is a 2
demonstration test and not a research facility.
They 3
are using that to give the public confidence that large
[}
4 reactors can survive earthquakes there.
5 They don 't have a lot of instrumentation nor 6
a re they interested in themselves or anybody else going 7
in and tinkering very much with the test specimen or the 8
data that comes out of it.
This is more show and tell.
9 However, there are many other, interesting experiments 10 that have gone on in Japan and to be conducted over the 11 next few years that I think we will be able to use in 12 validation as part of the SSMRP.
l l
13 They are interested in cooperating because
)
14 they are embarking on the PRA path for the first time, l
15 and they are struggling.
They are looking to us to give 16 them some help.
17 NR. MARKS It helps me a lot.
18 MR. GUZY:
As f ar as building tests, the 19 program at Los Alamos for structural response, both l
20 response and du=tilities, that is another area of l
21 physical test where we will benchmark the responses for 22 fragility.
23 By the way, it was not brought up before, but
(
24 we plan to have an overall validation plan review by th e 25 end of the summer.
It is incorporated in the results O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
43
()
I that we get from things we have identified as important 2
to the sensitivity studies, and assessing what the other 3
labs are icing.
I think the things that we will develop 4
over the next two years will fit into that.
5 Do you want me to discuss what happened last 6
year, or do you want me to discuss what will happen next 7
year?
8 MR. OKRENT:
We can read, so just indicate 9
what you think is important.
10 MR. GUZY:
In a nutshell, what happened in 11 1982, at the beginning of the year the major work effort 12 was on the SONGS project, which was terminated.
The 13 program has been through the peer review groups, and O
14 management review group.
There has been quite a 15 redirection of program, quite a program planning change 16 as the year has gone through in the final products.
17 The next page of the handout shows the current 18 projects as we see them, as they stand in 1983 and 19 1984.
20 The sensitivity studies, which we have 21 mentioned before, will be terminating soon, essentially 22 at the end of this fiscal year, but there will be some 23 carryover.
The Zion studies will be completed then
()
24 also.
25 After that, as part of our effort to make the O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554 2345
44
()
1 methodology available, economical, and easy to fly, we 2
will concentrate on developing simpler modeling methods 3
and making general methodology more applicable to
(}
4 plants, other than just Zion.
5 As a consequence, the first project will 6
concentrate on will be generic PWRs.
This will involve 7
such things as looking at places where we can simplify 8
both our structural modeling and our systems modeling.
9 Actually, there are a number of accident sequences to 10 look at, which will be benchmarked against design 11 calculations.
i 12 If we see that piping response throughout the 13 plant has a certain range of uncertainty, and certain O
14 characteristics, then we can calibrate a few piping 15 analyses and extend that through the rest of the overall 16 systems analysis.
17 The systes analysis, there are 148 accident 18 sequences that are being used in the full Zion analysis, 19 which will probably be cut down to 20.
It is a matter 20 of looking at what is important, and simplifying the 21 methodology.
22 Another important component is the seismic l
23 hazard data, which we get at ss a separate project,
()
24 which will supply essentially input for hazard, or input 25 for the hazard curve any place east of the Rockies.
O ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W, WASHINGTON, D.C. 20024 (202) 554-2345
45
()
1 MR. OKRENT4 Will the Zion study, in your 2
opinion, give a reasonably good estimate of the 3
likelihood of small, small LOCA or small LOCA as a
(}
4 function increasingly severe ground motion?
5 MR. GUZY:
Yes.
We look at the various 6
elements of earthquakes both through the fragility curve 7
and the design systems.
It is done essentially on a six 8
level of earthquakes and for each level.
9 MR. OKRENT:
I an interested in knowing 10 whether you think the plant modeled in sufficient detail 11 and that you have a sufficiently good basis for judging 12 when a small, small LOCA, or a small LOCA will occur 13 from seismic action.
Do you think your answer will be O
14 reasonably good, given the degree of shaking and leaving 15 out the uncertainty of whatever the shaking is?
16 HR. GUZY My feeling is that it will be 17 reasonably good in the range where the risk is coming 18 from.
It is my feeling that in a very high earthquake 19 range, a lot of things can happen, but the overall 20 probability of that earthquake occurring dominates the 21 final numbers.
22 I think the level of detail where risk is 23 coming from in my opinion is sufficient for looking at
()
24 the probability of LOCA th e re.
But perhaps in the 25 higher thaa 1 G range, or whatever, for ground motion, O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
46
()
1 perhaps the modeling is not that refined to be able to 2
respond, but the response may not be that important.
3 Do I make sense, Mike?
4 3R. BONDS In a high level earthquake, so many 5
things would begin to fail that a small LOCA or a small 6
pipe break would not be of consequence in and of itself, 7
there would be enough other things that would be damaged 8
that it would not dominate the accident.
9 MR. OKRENT What is the answer to my question to when you are below the point where the building falls 11 down, whataver it is that is a more consequent event 12 between the SSC and the point G.
Do you think you will 13 be able to make a f airly good prediction on the O
14 likelihood of a small, small LOCA, or a 10CA, with the 15 modeling that you are able to incorporate in your 16 analysis as of now?
17 MR. ANDERSON:
If I understand your question, 18 you are asking, do we have different fragility curve for 19 one-inch pipes, three-inch pipes, six-inch pipes, and 20 12-inch pipes, etc.?
Is that the nature of your 21 question, so that we can say that a one-inch pipe fell 22 out at this level?
23 MR. OKRENT:
That is certainly one of the
()
24 things that would be included in an answer to the 25 question, but there is also a question of, have you j
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
47
()
1 modeled enough of the piping to know with confidence 2
that you are handling all of the one piping and the
(}
3 two-inch piping, or have you done it by a few 4
representative pipes, so you are not sure at a certain 5
level above some shaking.
Do you understand what I am 6
getting at?
7 MR. ANDERSON:
Yes.
Do we have the 8
fragilities for those, and have we modeled enough.
9 MR. OKRENT:
I am trying to understand whether 10 with the funds and the time, and the knowledge, do you 11 think you are in a f airly good state with regard to 12 predicting the likelihood of a small, small LOCAs, and 13 small LOCAs, given the level of shaking in the ground, O
14 or whether in fact the state of methodology, time, and 15 knowledge, leaves that a fairly uncertain prediction 16 still.
17 MR. BOND:
I would like to answer that.
If 18 you go down to two-and-a-half inches in modeling the 19 reactor coolant system and associated piping, that gives 20 us a pretty good handle on the small LOCAs.
For 21 calculating responses for pipe smaller than that, we 22 don't actually calculate a response.
23 What we have done is in similarity with other
()
24 risk analysis approaches.
We have computed 25 probabilities of failures versus pipe size, taking O
ALDERSON REPORTING COMPANY,INC.
400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
48
()
1 account of actual calculations of pipes down to 2
two-and-a-half inches.
We have extrapolated below that r~g 3
to get a probability of failure of small pipes.
This V
4 approach induces more uncertainty automatically in the 5
process, and that is the way we handle it.
I believe 6
that this is as good a calculation as we can get, and I 7
feel that it is an adequate one.
8 MR. EBERSOLE:
Do you find in doing this that 9
the probability of failure of small pipes is higher than 10 that of large pipes?
11 MR. BOND:
Yes.
12 MR. EBERSOLE:
Taking that as a case in point, 13 it is an unfortuate fact that there are in the field O
14 right now many designs that use the one-inch type 15 instrument hitters at which point they connect 16 mitigation systems for small LOCAs, such as you will 17 have for a one-inch pipe break.
18 So you lose the mitigating system coincident 19 with having created a small break or an aperture in the 20 primary, and you have no mitigation functions.
So you 21 have one channel which is then denied by the arbitrary 22 assumption that you have a random failure in it.
23 Regarding your question on small LOCAs and
()
24 primary loops, you would be much better off to have a 25 small 10CA there than you would have in one out of two O
ALDERSON REPORTING COMPANY,INC, 400 VIRGlNIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
49
()
1 service systems, which is a myriad of small pipes and 2
only two service systems available to maintain constant 3
cooling of the core.
}
4 If it were a small or large failure, you would 5
lose one of the continuous service systems which is 6
maintaining coolability of the core, and then it only 7
takes the implication of a random failure of the 8
remaining support, and you are gone.
9 If you want to have failures, have them in the to primary system.
Don 't have them in the service system.
11 MR. B3NDa We calculated responses in the 12 service system.
For example, the system that we feel is 13 sost important from a risk point of view, is the aux O
14 feedwater system, and we completely model that, all the 15 back into the --
16 ER. EBERSOLE:
Historically, the staff has 17 never permitted one to take a failure in a service 18 system.
You could, if you wanted to, take one in the 19 primary loop because you always had redundant 20 mitigation.
If you took one in the service system, you 21 had no redundant nitigation.
So the concept of a 22 service system being failed by an earthquake has been 23 outladed as a kind of policy matter.
(
24 Do you follow me?
You don't have a random 25 f ailure which is permissible anymore.
Your other system O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
50 1
1
()
1 must be perfect.
You can picture the system as a 2
daisy-chain composed of ten elements, two of which
{)
3 support the plant.
One minor f ailure leaves the plant 4
hung on the other element of the cooling, and you cannot 5
have any f ailures in that.
6 MR. BOND:
I am not exactly sure what you are 7
referring to.
8
.1R. EBERSOLE:
There are about ten systems in 9
the plant that support the shutdown cooling function, 10 each one of these is a daisy-chain of functions, one 11 interdependent on the other.
Break the chain at some 12 point, and it becomes non-functional.
You do it with a 13 seismic event, there is only one other daisy left'.
14 MR. B,0NDs As far as the pipes in the 15 secondary system, to the best of our knowledge, and to 16 an unprecedented degree, we have modeled those 17 interactions including the piping itself.
18 The sensitivity studies as of now, even though 19 we predict fairly reasonable, believable pipe failure 20 probabilities, we do not find high failure probabilities 21 dominate the risk.
In fact, we find electrical 22 components, some of the smaller valves that have 23 off-center operators, the safety relief valves turn out 24 to be very important to some of the control systems.
25 HR. RICHARDSON:
If I could express my
()
ALDERSON REPORTING COMPANY,INC, 400 VIRGIN!A AVE., S.W., WASHIN3 TON D.C. 20024 (202) 554-2345
51
()
1 personal opinion of what I think your question is, which 2
i s, is the SSMRP capable of predicting various size 3
LOCAs for a given input.
My personal opinion is that 4
the methodology has been developed to do that.
I don't 5
think the f ragility curves that have been developed to 6
date no into enough detail to answer your question.
7 I think the capability is there.
I don't 8
think that we have gone into the detail fracture 9
analysis of all the piping, all the materials, to fully to answer that question.
I think the capability is there.
11 He have not done it yet, nor would we do it between now 12 and 1984, but I think we have developed the capability 13 to do it.
14 HR. GUZY You have to look at what is more 15 important.
If transients are more important than LOCAs, 16 that is where we would look at, if that is what is 17 contributing to the risk.
l 18 MR. OKRENT:
Right now I am just trying to 19 understand whether it was thought that the ability to 20 predict small, small LOCAs or small LOCAs was quite good 21 with the available abthodology.
22 MR. B3ND:
The small, small LOCA, no.
From 23 the point of view that we have used extrapolated
()
24 responses and failure curves, there seemed to be a very 25 high degree of correla tion between pipe size and average O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
I 52
(])
1 response, af ter giving you a 150 different earthquakes, 2
so that the extrapolation did seem valid.
{
3 MR. OKRENT4 Okay.
)
4 MR. GUZY:
In getting the simplified generic 5
ps=kage out, we are aiming for May of next year.
We 6
will have the computer codes available and up and 7
running on the public availble system.
We will have 8
procedures out for the simplified methods, for the 9
justifications, and the benchmarking and calibration 10 that goes along with that, and tThe seismic hazard data 11 will be disseminated through the technology transfer 12
- project.
Our target for everything for use will be May 13 of next year.
O 14 A new project that we have now entered into is t
15 a BWR methodology development, ess en tially at the 16 direction of the senior research review group.
We will i
17 be developing methodology that will be equivalent to the i
18 simplified PWR methodology, and we will use as much of 19 what we have developed and extend it to the BWR.
20 Hyd ro-d ynamic load s, and things such as that, will be 21 assessed in the development process.
22 Essentially, the procedures, and the 23 guidelines will be equivalent to the level of what will
()
24 come out in the simplified methods.
25 We plin to apply this PWR analysis to a plant, O
ALDERSON REPORTING COMPANY,INC, 600 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
53
()
1 and having applied simplified methods to models and 2
design of plan ts, there will be some availability for 3
comparison, this is ascuaing that there are somewha t
{}
4 equivalent plants.
5 We plan to have a simplified BWR analysis, as 6
will as a PWR analysis.
The BWR will begin later.
It 7
will probably begin in the first quarter of next year, P
8 and will end in the middle of 1984.
9 The validation we have discussed.
We plan to 10 have a plan out on that that will go to the end of the 11 project.
In some ways, the plan on that will be 12 dependent on other things.
13 Essentislly, this year and in 1983, the I
)
14 sensitivity studies, design analyses, and simplified 15 methods will be completed.
The BWR will be completed in 16.the middle of 1984.
Technology transfer and validation 17 are the two projects that will be on-going.
18 MR. MARKS Will your PWR model ha ve 19 instrumentation?
20 MR. GUZYa I think that this should be 21 considered as part of the generic package.
Our final 22 product should be applicable to all PWRs.
23 MB. B3ND:
The assignment of release
(
categories given a core melt, the SSMRP has from the 24 25 inception based on the WASH-1400 methodology.
These O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., W ASHINGTON, D.C. 20024 (202) 554-2345
54
()
1 have been the guidelines of the prog ra m.
2 We have identified a small project to look 3
into more recent research of the containment failure
{}
4 modes to take into account the recent research that has S
been done.
This work is scheduled about a year 6
downstream.
We are not specifically considering it 7
right now.
8 MR. OKRENT:
What is the interaction between 9
the SSMRP and Mr. Burdick's efforts?
10 MR. GUZYa I think we are beginning more and 11 more to interact, as he mentioned.
We ara considering 12 including seismic in their IREP type studies, and they 13 are looking at what is going on in the SSHRP.
O 14 I think the intent in our division is to get a 15 product out which can be used by them.
We will be 16 becoming more involved with them.
I think there are 17 still some policy decisions about how and when the SSMRP 18 will be used by Mr. Burdick.
19 MR. OKRENTs Have you gotten involved enough 20 with them that they have provided detailed comments on 21 your simplified method which say that it seems' 22 reasonable, or here are the things that should or should 23 not be in it.
I am trying to understand the degree of 24 interaction.
25 MR. GUZYa They have studied the review O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE, S.W., WASHINGTON, D.C. 20024 (202) 554-2345
55
()
1 systems analysis, the event tree / fault tree analysis 2
used in Zion.
They have been involved somewhat in the 3
PRAs.
I don't know of a comment per se on what we are 4
doing in the simplication and the structural area.
It 5
is our field, and not theirs.
6 I guess we have not had any specific comments 7
on the specific simplification, but they have looked at 8
the overall methodology, the fault tree / event tree 9
development, and I think both divisions are looking at 10 themselves as having some sort of merger in the next 11 year.
12 MR. OKRENT4 Within the next year could be in 13 a month or 12 months.
If it is 12 months, it would be 14 sort of late for them.
15 MR. GUZY:
Our schedule is to have the 16 methodology out in May.
How that fits into their needs, 17 I don't know.
18 MR. ANDERSON:
They have had somebody assigned 19 to this program from their group from day one, to review l
20 and assist in the technical review of the SSMRP.
They 21 recently assigned somebody, haven't they, to work with 2;
us on the simplified method.
So there is a formal i
23 interchange, or formal commitment of participation in 24 the review.
25 Does that answer your question?
O ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
,m s.
e 56 g
.x O
4 sa oraest ' :
es uoe ere that ther a a 2
commented on the SSMRP, program prior to the-last several 3
mon ths.
We never had the benefit of hearing or seeing O.
4 any of the comments that they have provided, if they 5
have provided them.
So I will have to answer your 6
question that way.
7 HR. GUZYa The senior research review group, 8
Mr. Berners'is one of the members on it, and we have 9
input through him on his needs for thh program.
10 MR. CKRENTs My' experience,with review groups 11 at that - level is tha t-these are all very, very busy
/:,
12 people.
Even your outside senior reviev' group is made g
13 up of a bunch of very busy-Diople.
, N O
I hs_ve read what
~.
14 they have written, and of gourso you coilld play one recommendettion agsinst' an'othec,.ind choos[whatever path 15 18 you want to go on several important issues.
s,.
w.
,.y, 17
' I dont knos t'h~4t'they have\\ tise to. sit d'own'
,s
~. _ _.
and arEive at wbEt' I 'woia[d13all committde deciisions on 18 e.
19 Saturday, as ' this committee is usually~ forced to do.
3 20
'M it. BON D s.
Just before lefc the NRC,. Bill.
s 21 Vastly wa.s persuaded to come out 'an'd\\ spend several days e~
n 22 with us aniU"look over the SSMRP as a shole.
We made 23 somE presento.tiocs, and we had a feemeetings after tha'tt '-
~
24 meeting.
(
5 8. _ G U Z Y.t We have had input in the past.
My 25 Q
l' g 3 -
e,
's-
+
s
.z..
s
,s
.1 ALOERSON nEFORT4NG CogPANY,INC, 5 MO VRGANIA AK., S.W., WAbHING10N, D.C. 20024 (202) 554-2345 u
y
..=
57
()
1 feeling is tha t we are getting more interaction and more 2
input from that division, and I think that we will see (J~)
3 more in the next year at the staff level.
4 MR. OKRENT:
Let's try to finish up.
5 MR. RICHA3DSON:
I was going to mention 6
precisely what Mr. Bond mentioned.
7 MR. MARK:
I read your program without f ully 8
digesting it.
There are eight project numbers, or 9
something like that.
Do you really expect to finish,all 10 of those on schedule?
11 ER. GUZY:
I think when you recognize that 12 there will always be gaps in validation, and validation 13 vill be on-going beyond 1984.
Technology transfer, of O
14 course, will also be on-going.
The others have definite 15 scopes and dates.
16 HR. MARK:
So you expect to see finished?
i 17 MR. GUZY:
Yes.
18 MR. OKRENT:
I understand that Mr. Ryder is 19 willing to provide us thoughts on the program that we 20 heard briefly described on earth sciences.
Is that 21 correct?
22 MR. RYDER:
I wanted to talk a little bit 23 about the seismic input.
()
24 Again, this is a joint program, essentially 25 being administered by Research.
It will be administered
()
ALDERSON REPORTING COMPANY. INC.
400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554-2345
58
()
1 through the Earth Science Branch.
It will also be 2
supported by the SSMRP and by the NRR to develop seismic 3
hazard codes, namely, the capability to estimate uniform
(}
4 hazard spectra at diff erent return periods for any of 5
the Eastern Unite! Ststes.
6 Essentially, the idea is to use expert opinion 7
to develop the input parameters, and incorporate the 8
results.
This is essentially the next stage or a 9, reworking of programming that had been completed a year 10 or so ago, which was used extensively by the staff.
11 This methodology, with expected past results, 12 will incoporate additional experts, new information such 13 as the recurrence of the New Brunswick earthquakes, new O
14 ideas such Anna Ohio and Charleston.
It is going to i
15 extend the work done to all parts of the United States, 16 particularly Southwestern Mississippi which was not 17 done.
18 Also a serious attempt will be dade to upgrade 19 and improve the methodology, particularly with respect 20 to ground action.
Consistency with past studies will be 21 checked.
The results of incorporating new expert 22 examination f or use of expert opinion will be f urther 23 scrutinized.
24 In the previous program, this methodology was 25 extensively by the staff as a p rime tool in setting O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
& ng 59
()
1 ground motion spectra for the systematic evaluation 2
program.
It has also been used in the on-going work of 3
the staff for testimony in hearings and in writing the
)
4 review, and writing operating license reviews.
5 We expect to use the next phase of the SEP 6
program.
We expect to use this in the on-going reviews 7
of new plants, and also as a tool in evaluating new 8
information, such as the earthquake that occurred in New 9
Brunswick, and changing concepts such as Charleston.
10 We have had a lot of use of this in the past, 11 and we hope that we can improve it, and get still more 12 use out of this program in the future.
13 HR. OKRENTs Was there a prediction made for O
14 the Maine Yankee site, or a site like or near the Maine 15 Yankee site from this methodology prior to the New 16 Brunswick earthquake?
17 HR. RYDERs The closest site that was done 18 prior to the New Brunswick earthquake was Seabrooke.
19 But, Dr. Okrent, this is rather interesting.
20 The last time we did this, we did it twice and three 21 years apart, and between that there was a magnitude 5.3 22 earthquake in Kentucky.
We ran a second opinion around 23 to gauge the ef f ects of an earthquake with seismicity 24 somewhat similar to Brunswick, but not directly that 25 type.
It was a very interesting study, and we found O
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
- - ~, -
60
()
1 that after the estthquake was taken into account, it did 2
not affect as much as one might have expected.
We are 3
very interested in seeing what would happen to New
[}
4 Brunswick.
5 3R. OKRENT4 Taking the methodology as it 6
exists, could you take it to predict return frequencies 7
for the Maine Yankee site, or is it somehow identified 8
to specific sites and is not tranferable to any other 9
site?
10 MR. RYDER4 We could take past information 11 and, using the same criteria as we have used in the i
12 past, come up with estimates that "would have been 13 acceptable" in the SEP as they would be at Maine O
14 Yankee.
But that past input did not include the New l
15 Brunswick eartquake because the survey was done in 1978 16 and 1980.
17 MR. OKRENT It seems to me that it would be 18 worth doing, to try to take the methodology and grind it 19 through to see what you would have predicted for Maine l
20 Yankee, had.you applied it.
l l
21 ER. RYDER:
As part of the program, on* of the 22 things that we are interested in doing, we are comparing 23 sites that would have been predicted.
In other words,
(
24 we are going to compare the results coming out of this l
25 new iteration with what was done in the past, and check O
V l
1 ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554 2345
61
()
1 the consistency and change of opinions.
2 I think we will be applying the new results to 3
Maine Yankee, to see what we get out of there.
As you 4
are prabably aware, a consortium of utilities in the 5
Northeast has presented or is preparating to present to 6
us a very large scale program reevaluating the hazard in 7
the Northeast, both deterministically and 8
probabilistically, and the Maine Yankee is included in 9
that.
10 MR. OKRENT:
I was not aware of this, but I 11 will not say that I am surprised to hear about it.
12 Nevertheless, I myself think that it would be 13 of some interest to have people act as if the New O
14 Brunswick' earthquake had not occurred, take the 15 information that they have gotten from their experts 16 back in 1980, apply it to the Maine Yankee site, and see 17 what they would say the return frequency is.
18 What I have in mind is the recent letter from 19 the staff to utilities where it says, "The staff now l
20 estimates the existing design basis is on the order of 1
21 100 year earthquake."
I am sure you know the one I 22 sean.
l 23 MR. RYDER:
I think I wrote it.
That was an
()
24 estimate made at the time of the five power plant 25 shutdown.
It was, quite honestly, a very primitive ALDERSON REPORTING COMPANY. INC, 400 VIRGINtA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
62
()
1 estimate that was made by just comparing some curves, 2
and coming up with where we thought it might be.
3 The laval of sophistication in these kinds of
{}
4 efforts is really much more sophisticated, and I think 5
can give you much better answers, particularly of the 6
whole spectral estimates, not so much peak 7
accelerations.
8 MR. OKRENTa This'is a recent memo that I am 9
talking about.
10 MR. RYDER4 I am not aware of a recent memo.
11 I know the estimate of 100 years was debated before the 12 Commission at the time of the five plant shutdown.
13 MR. OKRENTs No, I as talking about a meno or O
14 a letter that went out to Maine Yankee in May, roughly, 15 of this year, in which it said something like "for most 16 of the plants we are looking at for operating licenses, 17 from the East to the West, the SSE spectrum tend to run 18 between one to a thousand per year, and one to ten 19 thousand per year, and we estimate that the Maine 20 Yankee," whose design basis" which I think is 0.1 G, "is 1
21 on the order of 100 year earthquake," or something like l
l 22 that.
l l
23 It did not say when the one in 100 year l
/~T l
\\_/
24 estimate had been made, or what the basis was.
I l
25 assumed that somehow that letter was related to the New
(
ALDERSoN REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON. D.C. 20024 (202) 554-2345
63 i
O i
aruaswic* e rthoemxe ia 1 1a oce ce-2 MR. RYDER:
Dr. Okrent, I as 99.9 percent sure 3
that that estimate came out of the original work that 4
was done at the time of the five-plant shutdown because 5
I know of no new estimates.
I know where that number 6
calle f rom, I am sure.
A new estimate would..tve come 7
through our branch, so I have the feeling that it was 8
taken from the estimates that were made at the time of 9
- the five-plant shutdown and presented as such, but it is 10 not a new estimate.
11 MR. OKRENT:
You mean the staff has had as its 12 best estimate since 1980 that the plant-shutdown 13 earthquake is on the order of one in a hundred year O
14 earthquake, and it is still, figuring that everything is 15 okay, or what is it that you are telling me?
16 HR. RYDER:
The staff made this estimate in 17 response to the problem with the five-plant shutdown.
18 At that time, a discussion of margins was presented to 19 the Commission, and the Commission was satisfied that 20 the margins were sufficient to account for larger 21 earthquakes.
22 H o we ve r, the staff identified the problem, and 23 we commissicled a study by Los Alames Laboratory to do a O
24 oeo1ooica1 study end esseme some of the oeo1ooice1 25 factors with respect to Maine.
It has been something O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
64
()
I which we indicated we have been working on, and we are 2
accelerating the work.
3 At the time we presented that argument, it was
)
4 not a very narrow argum-..c with low uncertainty.
It was 5
presented in conjunction with the argument with respect 6
to margins.
7 MR. OKRENT:
I have asked Dr. Savio to get me 8
the details of these various estimates.
I would still 9
be curious to see what the methodology would have 10 predicted before and then after for.a site affected by 11 the New Brunswick earthquake.
12 Do you have any suggestions or comments, or 13 things that you would like to see done that are not in
()
14 the resear:h program on seismology and geology, since we i
l 15 have you here?
16 NR. RYDER4 I think, in general, they are.
17 doing the things that we think are necessary.
They have 18 asked for our opinion, and very frequently we have given l
19 our opinion, and I think they are generally following 20 those recommendations.
21 Ihe staff's emphasis has been that more effort 22 should be put on ground motion, particularly now with 23 all the records we are getting.
We are learning a lot.
()
24 MR. MARK:
Do you think the program topics are l
25 well chosen?
What about the ra te at which you are l
(1)
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
65
()
1 getting things checked off?
I think we were told 2
earlier tha t somebod y would be glad to put $1 million 3
into the seismic.
[}
4 MR. RYDER:
I think we are realists with 5
respect to the results of investigations.
We realize 6
that this is a very slow process.
What we really need, 7
unfortunately, is for more earthquakes in the United 8
States, but we want to make sure that we have the 9
capability.
10 MR. MARK In the short term, you will take 11 stress measurement, and you are not going to have more 12 money.
13 MR. RYDER I think the increased emphasis on
)
14 ground motion.
We are getting now, from New Brunswick 15 and New Hampshire, ground motion recordings that will
- 16. have a very important effect on the evaluations.
We are 17 already asking the utilities to examine this information i
18 as it is just emerging, and as it has not reached the 19 publication stage.
We made requirements on the 20 Seabrooke utility that they evalua te the ef fect of all 21 the ground motion recordings upon their estimates.
22 MR. MARK It is a varying spectrum, and a 23 varying frequency from the standard one that is usually 24 used.
25 MR. RYDER:
That is right.
We have always I
(2) l ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
66
()
1 made certain assumptions about Eastern ground motions.
2 We have to see how well our original assumptions stack 3
up with the data of 1.6 G or even the other peak
[}
4 acceleration.
It is a very slow process.
I think the 5
whole idea we are trying to push forward is not only to 6
develop basic information, but also to develop 7
methodologies for dealing with the uncertainties.
This 8
is what we are trying.to pursue in conjunction with the 9
SSFRP.
10 MR. OKRENT:
I guess the one aspect that I am 11 interested in a comment from you is the one I addressed 12 to Leon Beratan.
Are there things that you think one 13 can do that have a reasonable chance of shedding light
( O 14 on the central stsble region with regard to low 15 probability of more severe earthquakes?
I am thinking 16 of the development of a paper by D.F. Kirkland, but I 17 don 't want to restrict myself to that, but it certainly 18 raises some points.
19 MR. RYDERs I think the =omment that we get l
20 out of Perkins' paper is indicative and what we are very 21 well aware of is that what is wagging the dog is the 22 tail, and the tail is the uncertainty in the attenuation 23 f unction.
That is why we are extremely interested in 24 getting actual records of data.
25 People come up with various estimates of what O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554-2345
67
()
1 the ground motion attenuation f unction should be based 2
upon their interpretation of how you modify Eidwestern 3
data.
These are various strategies, and what we need or
)
4 would like to have -- we want to make sure that when 5
these earthquakes occur in the United States, that we 6
are going to record them and have the right kind of 7
instruments to record them, and that we have strong 8
motion instruments in the right places.
9 I think the right instruments in the right 10 places, I think quite honestly that would help us the 11 most in getting at the problem of the variable.
With 12 the very long return period, and the very rare events, I 13 don't know how much progress we are going to make with O
14 respect to the causitive factors of earthquakes in these 15 low seismic areas.
16 I think we have made lots of progress.
We 67 have made mixed progress in Charleston.
We have to keep 18 trying.
19 With respect to the area that you are 20 concerned with, the upper Midwest, there are very few 21 earthquakes, and certainly at this juncture I cannot 22 toss off the top of my head something other than getting 23 a better hold on ground motion.
()
24 NR. OKRENT:
I guess I am wondering if in a 25 sense it is not f ashionable for the people in this field O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554-2345
68
()
1 to think about that area, and so you are less likekly 2
have it explored, and perhaps to gain insight that may 3
arise one way or another.
{)
4 I must confess that I am quite uneasy about 5
the use of the available historical data as the basis 6
for taking subsets of essential central stable region.
7 I know I am uneasy about that and among the reasons for 8
this being uneasy is my understanding that we don 't 9
always have forewarning of significant size 10 earthquakes.
Sometimes they come without a history of 11 activity.
12 So that you have reason to say, we are getting 13 a small event, when you may get a big one.
This is not 14 so infrequent a situation that one can rely on 15 historical seismicity with a high degree of confidence.
16 If I can't use the historical seismicity with a high l
l 17 degree of confidence, then I would like to be at least 18 reassured that stress conditions plus the knowledge of 19 the crust, or whatever it is, reinforces not only on a 20 very large microscopic scale, but on a necessary 21 sub-microscopic scale.
Whatever it is, I don't know 22 what we would be useful.
23 In listening to the things to be looked at in 24 this area are sort of unique by not being on the list.
l l
2.c I think th a t this should be on the list in some way.
I
()
ALDERSON REPORTING COMPANY,INC, j
400 VIRGINIA AVE., S.W., WASHINGTON D.C. 20024 (202) 554-2345
i l
l 69 j
()
1 guess I am just urging that you try to develop a way of 2
doing that.
()
3 MR. RYDER I want say that is has been on the 4
list and it is still on the list.
For example, funded 5
research natwork in upper-Michigan, Minnesota and Iowa, 6
certainly can be classified as some of the lowest 7
seismic areas, I know our commission was dropped because 8
of the absolute lack of events.
am not sure, I think they still are on the program.,
10 I guess what it boils down to is that 11 seismologists have found out that most information is 12 where earthquakes are occurring.
It is very hard to 13 come up to meaningful results.if you don't have the O
14 basic information there in terms of earthquakes.
15 MR. OKRENT.
Maybe you should not go at it 16 from the seismological point of view.
17 MR. RYDER:
The other methodologies which have 18 been done in the past, essentially are not from the 19 seismological point of view, but are purely from 20 geological point of view, and they really have not been 21 that fruitful.
22 We are trying to make stress measurements, but 23 I think we can't o ve rem ph a size that perhaps the most 24 meaningful stress measurements that we are making about 25 earthquake generation are those which are determined O
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE S.W., WASHINGTON, D.C. 20024 (202) 554-2345
70
()
1 from earthquakes themselves.
2 The stress seasurements in the ve ry upper part 3
of the crust are useful, but they always should be taken
(}
4 in conjunction with the kind of stress m'easurements we 5
are getting from earthquakes.
When no earthquakes 6
occur, it is very hard to make stress measurements.,
7 I guess the answer is that ve have been 8
looking at that region.
I think Research is supporting 9,
it.
At this point, I think we cannot get a lot of 10 information, and what we have been getting is really in 11 the form of speculation.
It is very hard to separate 12 that out, unless you have some data that you know is 13 relevant.
0 14 MR. OKRENT:
Unfortunately, I don't have an 15 earthquake source to give you.
16 (General laughter.)
17 ER. OKRENT:
Are there any other questions for 18 Mr. Ryder?
1 19 (No response.)
l l
20 MR. OKRENT:
Are there any other questions for 1
21 any of the people here, or any comments that anyone 22 would like to add?
l 23 (No response.)
24 MR. OKRENT4 If not, I would like to.thank 25 everybody for coming to this late subcommittee meeting.
! ()
l l
ALDERSON REPORTING COMPANY,INC, 400 VIRGINIA AVE., S.W., WASHINGTON, D.C. 20024 (202) 554 2345
t i
L 71 i
.i i O i
1 111 aect ce the eti 2 easour ea-1
,2 (Whereupon, at 7:10 p.m.,
the mee ting 3
adjourned.)
4 i
I 5
l i
l 7
8 9
10 l
l l
11 12 13 O
14 15 16 17 18 19 20 21 22 23 24 25 O
ALDERSON REPORTING COMPANY. INC, 400 VIRGINIA AVE., S.W WASHINGTON, D.C. 20024 (202) 554-2345
...-_m.__
NUCLEAR REGUIATORY COMMISSION This is to certify that the attached proceedings before the in the matter of:
ACRS/ Subcommittee on Extreme External Phenomena
- Date of'Froceeding:
July 7, 1982 Docket llumber:
Place of Proceeding:
Washington, D. C.
were held as herein appears, and that this is the original transcript thereof for the file of the Commission.,
Patricia A. Minson Official Reporter (Typed)
-O W d 74Nw Official Reporter (Signature)
O
/
\\/
O i
O O
0 82 DEVFinPENT OF PLANNIfiG FOR SSMRP CCFPIETION ACRS EC0FTENDATIONS (NUEGS 0795 #0 0864) e SIMPLIFIED fETHODOLOGY FOR GENERIC USE e
APPLY METHODOLOGY TO A BWR e
REVIEW OF PROGRAM PLAN BY NRC STAFF AND CONSULTANTS SSMRP SENIOR ESEARCH REVIEW GROUP RETINGS (11/3/81 AND 1/26/82) e MEMBERS: D. ROSS (CHAIRMArd, G. ARLOTTO, R. BERNER0, D. EISENHUT, S. HANAUER, R. VOLU4ER, R. SAVIO (ACRS) e IDENTIFIED SSMRP USER NEEDS AND REVIEhED PROGRAM PLAN OurLINE e
EMPHASIZEDNEEDFOR:
- 1) VALIDATIONOFSSfW
- 2) SENSITIVD'Y STUDIES
- 3) SIMPLIFIED, USER ORIENTED PETHODOLOGY
- 4) SSMRP AND IR9/NRP MERGER e
REC 0 MENDED THAT ONLY A SIMPLIFIED AFPROACH BE USED FOR A BhR 4/18/82 LEITER ON NRR ESEARCH f1FFnS FOR SEISMIC NMLYSIS EIR0DOLOGY SSMRP SENIOR CONSULT #1T ESEARCH REVIEW GROUP WETING (6/2/82 #0 6/3/82) e MEMBERS: S. BUSH, A. CORNELL, W. HALL, W. HAYS, AND D. WORLEDGE e
GAVE PEER REVIEW OF PETHODOLOGY AND 5/25/82 DRAFT PROGRAM PLAN l
REVISED ES/LINL PROGRAM PLAN SCEDUIFn FOR SRRG REVIEW NEXT MONTH
7/7/82 FY 1982 SSMRP ACTIVITIES DEVELOPMENT OF PROGRAM PLAN FOR COMPLETION IN FY 198f4 e
INPUT FROM ACRS, NRC MANAGEMENT AND PEER REVIEW PANEL e
PROGRAM RESTRUCTURED TOWARDS PRODUCTS RATHER THAN METHODOLOGY DISCIPLINES SONGS PROJECT TERMINATEll e
ENDED PER 1/13/82 RECOMMENDATION FROM NRR SEISMIC HAZARD CHARACTERIZATION PROJECT STARTED e
PROVIDE GENERIC E.U.S. SEISMIC HAZARD DATA e
JOINT FUNDING BY RES (FROM SSMRP BUDGED AND NRR MANAGED BY DHSWM/RES SSMRP EMPHASIS IN FY 1982 0
1.
SENSITIVITY STUDIES e
EFFECTS OF VARIATIONS IN MODELS AND PARAMETERS FOR SEISMIC INPUT, SSI, STRUCTURAL RESPONSE, AND SYSTEMS ANALYSIS DOMINANCE RANKINGS 2.
COMPLETION OF ZION ANALYSIS e
FINALIZE DATA AND MODELS UNCERTAINTY BANDS 3.
SIMPLIFIED MODELS/ METHODS DEVELOPMENT e
REDUCED STRUCTURAL AND ACCIDENT SEQUENCE MODELING O
I N ANALYSIS AND SENSITIVITY STUDY RESULTS PROVIDE 8
BASIS FOR JUSTIFICATION AND MODEL CALIBRATION
7/7/82 SSMRP PLAN FOR FY '83 '84 SENSITIVITY STUDIES 0 STUDIESRELATEDTOSEISMICHAZARD,SSI,ANDSTRUCTURALRESPONSECOMPLETED BY10/84 0 SYSTEMS ANALYSIS STUDIES COPPLETED BY 1/85 ZION RISK ANALYSIS _C0FPLETED 0 CALCULATIONS CCMPLETED IN TIME FOR PRESENTATION IN MII>-0CTOBER 0 FINALREPORTDUE2/83 SJtPLIFIED IEELS/FERODS 0 IlEm0DSANDPROCEDURESFORPWRDEVELOPEDANDAPPLIEDTOZIONBY5/83 0 E.U.S. SEISMIC HAZARD DATA AVAILABLE IN 5/83 IECliN0 LOGY TRANSFER 0 ALLCOMPLITERCODES, DATABASES,ANDPROCEDURESDISSEMINATEDBEFORE5/83 i
0 ONGOINGUSERIRAININGANDASSISTANCE BWR E BODOLOGY DEVELOPW NT 0 EXTENSION OF PWR SIMPLIFIED PETHODS/MODELS 0 BEGIN IN LATE FY '@ OR EARLY FY '83 0 DEVELOPMEm0DSANDPROCEDURESANDAPPLYTOABWRBYMID-fY'84 VALIDATION O ANOVERALLVALIDATIONPLANTOBEDEVELOPEDBEFOREFY#83.
INCORPORATE RESULTS OF OTHER RESEARCH PROJECTS AND BASE VALIDATION PRIORITIES ON ZIONANDSENSITIVM'YSTUDYRESULTS 0 ONGOINGEFFORTTHROUGHFY'84 O
1
. ~.
- - - ~.. -
(1)
C:)
(37/7/82 SSMRP PROJECTS FOR FY 82-84
' TARGET
~
~
PROJECT COMPLETION DATE FY 82 FY 83 FY 84 SONGS AFWS STUDY 2/82 (ANALYSIS)
$ 306K 0
0 8/82 (FINAL REPORT)
SENSITIVITY STUDIES 10/82 (STRUCTURAL) 445 45 0
1/83 (SYSTEMS)
COMPLETE ZION 10/82(ANALYSIS) 459 86 0
ANALYSIS 2/83 (REPORT)
DEVELOP SIMPLIFED 5/83 232 320 0
(PWR) METHODS VALIDATION 10/84 294 974 982 BWR RISK ANALYSIS 1/84 30 517 233 TECHNOLOGY TRANSFER 10/84 174 298 126 GENERAL MANAGEMENT 10/84 867 555 595 TOTAL 2807 2795 1936 (FUNDING FOR SEISMIC HAZARD CHAR, PROJECT)
(110)
(205)
(0)
l O
O
.O EARTH SCIENCES BRANCH - RESEARCH BUDGET PROJECTIONS FY84 FY85 Reguest Request GEOLOGY AND SEISMOLOGY 4690 5590 REGIONAL STUDIES 3075 3975 Southeastern U.S.
(including Charleston)
Northeastern U.S.
New Madrid Region (plus Anna Ohio)
Nema ha Ridge Pacific Northwest Improvement of Instrumentation + Analysis TOPICAL STUDIES 1615 1615 Site Specific Spectra Generic Studies Strong-Motion Seismograph USGS Topical Studies Geo. Tech. Eng.
METEOROLOGY AND HYDROLOGY 1710 1410 Topical Studies Severe Storms Atmospheric Dispersion Atmospheric Model Evaluation SITING CONCEPTS 250 250 Flooding Groundwater Contamination From Class 9 Accidents 6,500 7,100
()
EARTH SCIENCES lANCH - RESEARCH BUDGET PROJECTIONS FY84 FY85 Request Request 4690 5590 GEOLOGY AND SEISMOLOGY 1Q11 3975 REGIONAL STUDIES 1000 1400 Southeastern U.S. (including Charleston) 925 1225 Northeastern U.S.
790 990 New Madrid Region (plus Anna Ohio) 150 150 Nemaha Ridge 21 0 21 0 Pacific Northwest Improvement of Instrumentation + Analysis 1615 1615 TOPICAL STUDIES 250 250 Site Specific Spectra 590 590 Generic Studies 150 150 Strong-Motion Seismograph 475 475 USGS Topical Studies 150 150 Geo. Tech. En9 1710 1410 METEOROLOGY AND HYDROLOGY 300 200 Topical Studies 60 60 Severe Storms 900 800 Atmospheric Dispersion 300 200 Atmospheric Model Evaluation 250 250 SITING CONCEPTS 150 150 Flooding l
Groundwater Contamination i
From Class 9 Accidents 100 100 l
6,500 7,100