Protecting People and the EnvironmentUNITED STATES NUCLEAR REGULATORY COMMISSION
Official Transcript of Proceedings
NUCLEAR REGULATORY COMMISSION
Title: Advisory Committee on Nuclear Waste
131st Meeting
Docket Number: (not applicable)
Location: Rockville, Maryland
Date: Wednesday, January 9, 2002
Work Order No.: NRC-166 Pages 1-315
NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers
1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005
(202) 234-4433. UNITED STATES OF AMERICA
NUCLEAR REGULATORY COMMISSION
+ + + + +
ADVISORY COMMITTEE ON NUCLEAR WASTE
131ST MEETING
+ + + + +
WEDNESDAY,
JANUARY 9, 2002
+ + + + +
ROCKVILLE, MARYLAND
+ + + + +
The subcommittee met at the Nuclear
Regulatory Commission, Two White Flint North,
Room T2B3, 11545 Rockville Pike, at 8:30 a.m.,
George M. Hornberger, Chairman, presiding.
COMMITTEE MEMBERS PRESENT:
GEORGE M. HORNBERGER Chairman
RAYMOND G. WYMER Vice Chairman
B. JOHN GARRICK Member
MILTON N. LEVENSON Member
STAFF PRESENT:
HOWARD J. LARSON, Special Assistant
RICHARD K. MAJOR
LYNN DEERING
LATIF HAMDAN
SHER BAHADUR
AMARJIT SINGH
JOHN T. LARKINS
RICHARD P. SAVIO
CAROL A. HARRIS
ALSO PRESENT:
JAMES ANDERSON
BILL REAMER
JIM WINTERLE
BRET LESLIE
NEIL COLEMAN
RANDY FEDORS
JEFF POHLE
PAUL BERTETTI
JOHN BRADBURY
ROBERTO PABALAN
GOODLUCK OFOEGBU
MYSORE NATARAJA
TAE AHN
ALSO PRESENT: (cont.)
BIS DASGUPTA
TIM McCARTIN
BANAD JAGANATH
WES PATRICK
DENNIS WILLIAMS
. A-G-E-N-D-A
AGENDA ITEM PAGE
Opening Statement by Chairman Hornberger . . . . . 5
Management Overview, Bill Reamer . . . . . . . . . 7
Risk Insights, Bret Leslie . . . . . . . . . . . .19
Total System Performance Assessment. . . . . . . .53
and Integration, James Weldy
Igneous Activity, Brittain Hill. . . . . . . . . .81
Structural Deformation and . . . . . . . . . . . 115
Seismicity, John Stamatakos
Container Life and Source Term . . . . . . . . . 132
Gustavo Cragnolino
Unsaturated and Saturated Flow Under . . . . . . 161
Isothermal Conditions, Jim Winterle
Thermal Effects on Flow, Randy Fedors. . . . . . 177
Radionuclide Transport, Paul Bertetti. . . . . . 199
Evolution of the Near-Field Environment, . . . . 229
Roberto Pabalan
Repository Design and Thermal-mechanical . . . . 250
Effects, Goodluck Ofoegbu
Preclosure, Bis Dasgupta . . . . . . . . . . . . 267
Summary of Issue Resolution Status, Schedule . . 287
Lessons Learned, James Anderson
. P-R-O-C-E-E-D-I-N-G-S
(8:35 a.m.)
CHAIRMAN HORNBERGER: The meeting will
come to order.
This is the second day of the 131st
meeting of the Advisory Committee on Nuclear Waste.
My name is George Hornberger, Chairman of the ACNW.
Other members of the committee present are John
Garrick, Milton Levenson, and Raymond Wymer.
Today the committee will receive an update
on the status of key technical issues. Howard J.
Larson is the designated federal official for today's
initial session.
This meeting is being conducted in
accordance with the provisions of the Federal Advisory
Committee Act. We have received no written comments
or requests for time to make oral statements from
members of the public regarding today's sessions.
Should anyone wish to address the committee, please
make your wishes known to one of the committee staff.
It is requested that the speakers use one
of the microphones, identify themselves, and speak
with sufficient clarity and volume so that they can be
readily heard. I want to emphasize the fact that we
need to have speakers, including those in San Antonio,
identify themselves clearly, because this is being
recorded. And also, here, for people on the other end
of the video, please make sure -- this includes all of
the speakers -- to clearly identify yourselves.
The key technical issues have been, as we
know, a focus of the interactions between DOE and NRC.
Key technical issues were the term that the NRC came
up with to guide their evaluation of DOE's work on
Yucca Mountain. There have been, as we know, many
technical exchanges and agreements with -- between the
NRC and the Department of Energy.
And today -- well, back up just a minute,
the Commission has expressed interest in having the
ACNW give them their views on the status of key
technical issues, and in particular where the most
difficult parts in reaching agreement between the --
reaching agreements between the Department of Energy
and NRC lie.
And also, the committee has, for many
years, been interested in all of the discussions of
KTIs as to how the NRC is risk informing their work.
And, in fact, we are -- we continue to be interested
in where the NRC sees the most risk significant issues
in the whole KTI system.
We've also, of course, been interested in
the past as to how KTIs cover the whole landscape and
have asked -- continually asked questions as to
whether there are gaps between the KTIs and how the
KTIs interface with each other, and how, in fact,
everything gets integrated at the end.
And so the committee has typically
questioned the NRC staff on these topics and others.
Today we're going to get a -- I think some fairly
detailed updates on the issue resolution -- the KTI
and issue resolution process, and we have a whole
series of presentations.
So Bill Reamer is going to kick us off on
the overview.
MR. REAMER: Good morning. Bill Reamer,
NRC staff.
We will try to accomplish each of the
items that you mentioned today. We've I think got
everyone here, although I know that the weather has
been a problem for some of the staff folks. We'll go
through the key technical issues issue by issue, give
you the status, describe our path forward with the
Department of Energy.
I can't give you a date when there will be
a Department of Energy license application. First
off, we don't know what the outcome will be of the
site recommendation process, when and whether the
Secretary will make a decision to recommend the site.
We also don't have the Department of Energy schedule
for possible license application as well, but I don't
think that that needs to necessarily impact our
discussions today.
I think this will become more clear in the
March timeframe, but today I don't have a date to give
you on when to -- we would expect a DOE license
application if there is one.
We do want to answer the committee's
questions completely on the relationship of the KTIs
to risk, but you're not going to find, at least in my
presentation and the presentations that follow, the
five top KTIs from a risk standpoint.
Someone said yesterday, "Today you will
hear everything that you need to know about KTIs, and
maybe something that you don't need to know." And I
think if you hear something you feel like you don't
need to know, you should ask, because our view is you
may need to know that.
So, just briefly, my agenda is to start
with the result for issue resolution as our goal and
to describe our approach to get there, where we are in
a programmatic sense, what we have planned in terms of
the path forward in a program sense, how we're
supporting that from the standpoint of using risk
information, program uncertainties that affect our
path forward as well.
So the NRC's role -- the NRC staff role is
to -- under the law is to be able to review a
Department of Energy license application, if there is
one, and make a decision on a possible construction
authorization in three years. Our prelicensing
activities with the DOE have generally been focused on
gaining confidence that any license application that
we get will be sufficient for us to commence a review
and write a safety evaluation report and make a
recommendation.
And so issue resolution and our path
forward on the KTIs takes its cue from this. If DOE
submits a license application, it should be
sufficient. It needs to be sufficient for the NRC to
commence a review. Generally, therefore, we would
take the position as the staff that if a license
application addresses some of the KTIs and not other
KTIs, that's not going to be sufficient for us to go
forward with the review.
Now that's obviously an extreme case.
It's never going to occur. But if some KTIs are dealt
with well and other KTIs are not dealt with, that's
putting a problem for us into the license review
process. That's saying that additional information
that we may need is going to have to be gained after
the license application is received, and that's not
our preferred course.
Our preferred course is to resolve the
KTIs with the Department of Energy before any license
application is received. And the key here, again, is
the three-year clock that the Nuclear Waste Policy Act
puts on the Commission to make a decision.
I think what I've just said is consistent
with what you'll find in the Commission's preliminary
comments to the Department of Energy submitted on
November 13.
Basic approach that we're following is to
identify gaps in the DOE supporting information. By
this we're considering the DOE story, the DOE
argument, the DOE performance assessment, the safety
assessment, what we need to review -- that story --
and reach conclusions.
We're taking into account information not
only that we get from DOE but information in public
literature. We're taking into account the views of
other stakeholders. We're taking into account the
views of this body, Nuclear Waste Technical Review
Board, the state.
We're looking at risk information and how
that impacts the identification of gaps. But
remember, we need to have confidence that overall
there is an understanding on the part of the
Department of Energy and ourselves that we understand
the full system, the total system, and that's why each
of the KTIs -- that's one of the arguments why each of
the KTIs is important. It's important to gain that
understanding of the whole system.
The reasons why we identify information as
being needed, that's important as well, and that's
what we'll be talking about with you in more detail
today. We also are taking the approach to get the DOE
agreement on what we see as the information gaps, and
by that we -- we get a signal from the Department of
Energy that they understand what we're saying and that
they are prepared to follow up and do something in
response.
We'll be documenting the technical basis
of issue resolution periodically, and we involve the
public throughout the process.
Where we are now is that we've identified
over the past two years what we think are the
remaining gaps for DOE to address with respect to the
key technical issues. I cannot overstate the effort
that the staff has made, the staff of the NRC, the
staff of the Center for Nuclear Waste Regulatory
Analysis, to evaluate the DOE supporting information,
and identify what information is needed and to state
the reasons why.
The results have been the 293 agreements
that are the product of the 18 technical exchanges
that we've held with DOE. I think people should
resist the temptation to say 293 agreements is an
indication that the process is not working. I think
it's a strong indication that the process is working,
that we are focusing, that we're reaching a
convergence, that the process of refining the
information gaps and what additional information is
needed is working, and that we have a reasonable basis
to proceed.
We'll be issuing the integrated issue
resolution status report in the spring of '02, and I
think we're on your meeting calendar in April to talk
about that. The path forward is to continue to
actively monitor the Department of Energy response to
the agreements.
We're working on two fronts. We're
specifically working on the -- from the standpoint of
tracking DOE schedules with respect to specific
agreements, to provide a response, reviewing the
responses that we receive, and providing feedback to
DOE. Looking to set up our next round of meetings
with the Department of Energy to further refine where
we are on the DOE information gaps.
But on another front, we're also looking
programmatically at the process, how close is DOE
getting in terms of their response to what we think
we've agreed to, and looking at ways that we can
improve the process so that responses are, from our
standpoint, on target.
We have a meeting tentatively planned with
DOE on February 5, I believe it is. I know that that
potentially conflicts with the committee's meeting,
but I think we ought to try to find a way that there
can be some coverage, some involvement, some
observation at least from the committee or committee
staff as to that meeting, because I think that may
provide some additional information that can --
CHAIRMAN HORNBERGER: What's the focus of
that meeting, Bill?
MR. REAMER: It's more in the nature of a
meeting on meetings. It's a meeting to plan our next
set of technical exchanges with DOE to get an
understanding of how the process needs to work in
order to make those meetings effective and to provide
feedback that we have to provide on the process, how
close is DOE meeting the target on the responses that
we've seen thus far, suggestions as to how we might
improve the interaction that exists in order to get
closer alignment.
DOE is working on a plan to -- of their
work from 2002 to a potential license application, and
we think that will be available in the March timeframe
from DOE.
Now, generally supporting the path forward
and the planned activities are all of the activities
we're doing with respect to risk informing the
process. We won't talk a lot about the Yucca Mountain
review plan, but it is an important tool to risk
informing our review.
The Yucca Mountain review plan really in
many ways reflects experience with the KTIs and the
criteria and the acceptance criteria and reasoning
that we've developed in the context of addressing the
KTIs. So I don't think it's going to be a major
impact.
It does address and move forward on the
basis of what we call the integrated subissues, which
is our device for assuring -- one of our devices for
assuring that the key technical issues are risk
informed. And that's also the format we will be using
in the issue resolution status report, so I think that
will be helpful to give a transparency to how the KTIs
are integrated with the overall performance of the
repository and our efforts there.
The presentation that follows mine will
get in heavily to our efforts in the risk information
area. But generally what we're trying to do is using
sensitivity analyses, examine the important risk
contributors, quantify the uncertainties associated
with those important risk contributors, and use them
as a basis to assure that the PA is routed in the
evidence and not in something else.
Next slide is really more for our
colleagues at the Department of Energy, to remind them
-- remind everyone that really the burden to address
or not address the key technical issues, the
regulators, the key technical issues or the
regulators' issues here, is on the potential
applicant, the DOE. The staff's role I think is to
actively focus on refining what's acceptable and why.
The KTIs are the logic that we are using
to do that. The activities I hope the committee will
feel at the end of the day are risk informed. I have
every reason to believe that this process can work and
will work, that if a license application is submitted
that it will be sufficient for the NRC to conduct a
review and reach a decision.
I did mention earlier the plan that DOE is
working on, which we view as the -- providing the
information to address this -- what I call this
program uncertainty.
So the gist, then, is the issue resolution
is progressing. We're monitoring the agreements, the
performance on specific agreements in the process as
well. We are factoring in risk insights continuously,
continuing the technical exchanges, the next round of
technical exchanges with DOE, and the -- what I
mentioned, the LA plan, the planned LA is an important
input that will be coming.
That's all I had.
CHAIRMAN HORNBERGER: Thanks, Bill.
MR. REAMER: Any questions for me?
CHAIRMAN HORNBERGER: A couple. Well, one
-- a comment and then a question for you. I'm really
happy to hear that you are moving forward with plans
to continue technical exchanges. My personal opinion
is that they have been extraordinarily helpful, the
technical exchanges between DOE and NRC.
My question related to that is I know that
you're moving forward with an integrated IRSR. And
the question, then, is the previous technical
exchanges were focused on KTIs, and in moving to an
integrated IRSR there is not a total move away from
KTIs but really taking a more integrated view. Do you
have any ideas on how future technical exchanges will
be structured in light of this integrated IRSR
approach?
MR. REAMER: Well, they are going to be
structured to basically accomplish two things to
assure that the KTIs, which are the logic for the
staff's prelicensing activities, get addressed in a
way that meets the goal that I described. So the KTIs
will continue to be a major player and focus.
But to do so in a way that's consistent
with the way we will do the license application
review, which is the Yucca Mountain review plan and
the integrated approach that we will take there. And
so we want both.
CHAIRMAN HORNBERGER: Okay.
Questions from the committee? Milt? Ray?
VICE CHAIRMAN WYMER: Yes, one. Will the
integrated approach tend to focus on the most risk
important issues? Is that what will emerge from the
integrated approach?
MR. REAMER: Well, the integrated approach
will surely make sure that that information is
available and known and assure its integration into
our approach.
VICE CHAIRMAN WYMER: Will it identify
them clearly or --
MR. REAMER: I think you will today get a
good sense of the -- from the presentations of what's
important from a risk standpoint.
CHAIRMAN HORNBERGER: John?
MEMBER GARRICK: Well, you partly answered
that. You said up front -- and I appreciate that --
that we weren't going to get an importance ranking of
the KTIs out of today's proceedings and --
MR. REAMER: No. I said you wouldn't see
a list --
(Laughter.)
-- in my presentation and the other
presentations.
MEMBER GARRICK: Yes, okay.
MR. REAMER: But I'm sure I -- I'm
satisfied that you came to the meeting with an
expectation, and that when the meeting is over your
expectation will be satisfied.
MEMBER GARRICK: Okay. Well, that's
encouraging.
CHAIRMAN HORNBERGER: Staff? Anyone from
the audience?
Okay. Well, we are going to move along.
We have a full schedule. Thanks very much, Bill.
Next up is Bret, is that correct?
MR. LESLIE: Let's see if I've got the
right angle to be able to push the slides down as I
talk through this.
During our preparations for this meeting
I got the short straw to make this presentation. And,
really, what I'm trying to do right now is to provide
an overview to directly address the concern. I mean,
Bill said that we're not going to have a list of what
are the most important KTIs, but I'm hoping to provide
enough of an overview on some of the ways we've tried
to look at prioritizing the KTIs so that you'll have
the necessary information to come up with whatever
conclusion you want to.
I'm going to expand on a few -- several
things that Bill has already addressed. So some of
this is a repeat, but I'm trying to provide a little
more detail, so that the following presentations on
each of the individual KTIs you'll try to -- see the
picture.
Okay. It's a little slower than me.
As I said, this is an overview. And what
I want to do is lay out the framework, see if I can
make some of the connections for you. And the
specific applications for each KTI will be later, so
if you start to ask me a question about a specific KTI
or an integrated ISI, I'll defer your question.
The idea is I want to do really three
things. I want to give an overview again of the issue
resolution, what that overall goal is, explain in some
respects, given the regulatory constraints, how we use
risk insights, then move on to what are the risk
insights we have gained from our performance
assessment.
And, finally, something that was just
briefly crossed -- touched on by Bill is a risk
insights initiative. And this is something that we
had planned on presenting to the committee in March,
and what you're going to see is a couple slides of
snapshots. We're just starting that process -- but
where we think we can do a better job.
Moving on to the risk -- issue resolution
overview, as Bill indicated, the goal is in terms of
a -- for a potential license application. The
information that we receive from the Department of
Energy must be sufficient and of high quality. So
that's the overarching goal, and we're guided by
what's in Part 63. In particular, when we're looking
for post-closure and performance assessment, we're
guided by Section 63-114, which are the requirements
for performance assessment, which say what is
necessary for any performance assessment for post-
closure.
So that's kind of in the background of
where all these agreements are coming for. If you
look at detail in the agreements, some of the same
words in Part 63 provide a technical basis. Those are
the types of information that we're trying to get in
closing those gaps.
In identifying the gaps, again, we're
identifying the gaps in the Department of Energy's
approach, and we're looking at the data and analysis
and models that they're using in their performance
assessment. And, again, how we identify those gaps
are based upon the requirements in 63-114, what is
required for a performance assessment in any potential
license application.
But also, it plays into risk insights.
And I'm kind of jumping the gun, but recall that the
risk triplet is answering three questions. You can
have risk insights not just about consequences but
what can go wrong, its likelihood. Okay? You can't
say what is most important in terms of risk by only
focusing on consequences because if DOE has left out
a technical basis for a process that could impact the
consequences, then how good is that assessment of the
overall performance of the system?
So, really, you have to have the
information for each of these to get the overall
importance of -- in terms of risk insights. And what
you'll see -- what can go wrong is really the
features, events, and processes. And this is a non-
negligible portion of our agreements. Many of the
agreements are talking about provide the technical
basis for screening something out.
And, again, you have to know what to put
in the performance assessment or provide a basis for
screening it out for removing it from the performance
assessment.
Those gaps, as Bill indicated, are also
identified by other people, gaps identified in
performance assessments by EPRI, or insights from the
ACNW or the NWTRB, or the State of Nevada. What
things are they focusing on?
Because, again, we're thinking in terms of
down the road for a potential license application we
need to have those gaps identified and addressed.
And, finally, our risk insights come not
only from our own work but from the Department of
Energy, from both their performance assessment and
what their safety case will be, because, again, we're
looking for a sufficient, high-quality license
application, but the safety case is the Department of
Energy's to make. So we have to focus on what they
are asserting or proposing to use in their performance
assessment and their licensing case, potential
licensing case.
So how have we been applying the risk
insights? Well, in terms of the issue resolution
meetings, early on, you know, we learned -- as Bill
indicated, we've learned things as we've gone through
these technical exchanges. And one of the first
things we realized is that we needed to have a proper
perspective on what the DOE considered in terms of
each KTI, in terms of performance base.
So we requested that after that first
meeting each subsequent key technical issue, technical
exchange, have a presentation on the Department of
Energy's performance assessment, so that we could try
to have this conversation in a risk informed manner.
Prior to these meetings, both the Center
and the NRC staff conducted numerous evaluations, both
at the system and the process level, to gain insights,
so that when we went into the meetings we were focused
on the things that were most important. With that in
mind, the agreements that we -- came out of those
meetings are a function of the risk insights that went
into them.
And so we believe that the agreements are
risk informed, again remembering in a broad sense risk
informed in terms of also any potential license
application identifying the gaps and providing that
information.
As Bill indicated, our products are
applying those risk insights. The Yucca Mountain
review plan, which is due out later this spring, and
the integrated issue resolution status report, which
will be out this spring, will show how we've applied
those risk insights.
But also, in the presentations that follow
you'll see in each of the KTIs how the independent
efforts that they're doing, the independent
investigations, are -- how they've applied those risk
insights to what information they're trying to assess,
what they've assessed last year, and also in the
upcoming year. So they're applying the risk insights
to figure out what additional information is needed.
Moving on to the second part of this
presentation, we have used the iterative performance
assessment approach, and the committee is well aware
of that. This iterative approach has been used
approximately for the last 10 years by the NRC and the
Center to focus on gaining risk insights and
information, and the value of the iterative approach
is that it allows us, as we go along, to refine what
the insights are.
Suppose the first iteration of a
particular portion of the code is coarsely
represented, and we do an analysis and say, "This area
is important." So when we go back and revise our
code, we may add additional complexity to the code to
see which part is causing the importance.
And this iterative approach is not only
for the post-closure, it's for the pre-closure. So
when we talk about identifying the parts important to
waste isolation, there we're talking about the TPA
code and the post-closure assessment. And when we're
identifying the parts important to safety, this is in
terms of pre-closure and the pre-closure safety tool
that we're using.
This iterative approach causes the staff
to focus our review and the prelicensing documents on
the data -- on quantifying the uncertainties. What
are the most important uncertainties? And so that
guides us in terms of, what are the gaps -- some of
the gaps that the Department of Energy should address?
And, finally, this iterative approach
allows us, if data are very scarce or very uncertain,
conservative approaches can be adopted in the first
cut. But as the data are gathered, conservatives are
reduced and realism is increased as we go on.
One of the outcomes of this iterative
performance assessment process is it allows the
identification of the risk important features of the
repository system. And this has been called the
flowdown diagram. And this is an important diagram to
understand, because these are integrated subissues.
This is the format of how we're doing our
reviews in the Yucca Mountain review plan. It's how
we'll be documenting things in an integrated
resolution status report upcoming. But it's really
the integration of information in KTIs.
And I put this post up -- it's slides 16
and 17, and this is kind of the road map. On the left
side of those two slides are the integrated subissues.
The titles are ENG1. I -- let me back up. Across the
top are the integrated subissue -- ENG1. On the left
side are the listing of all of the KTIs and their
subissues.
So from an integrated performance
assessment standpoint, if you look at ENG1, which is
degradation of engineered barriers, which I believe
the committee believes is highly important, you can
see that multiple key technical issues -- the thermal
effects on flow, the near field KTI, the container
life and source term, the TPA -- TSPA KTI, and the
repository design and thermal-mechano effects -- all
provide input. It's that integration of the
information of each of the KTIs that is important in
evaluating the performance assessment.
So I've kind of left this crosswalk up,
and I think it may help you, as you go through, or
asking a KTI, well, how do you fit into the
performance assessment? Well, this -- these two
diagrams are the key to understanding that when we're
talking about resolution of KTIs, you have to look at
where that information is fed in terms of performance
assessment.
As Bill indicated, we have been using
sensitivity analyses to understand, in terms of the
previous slide, what are the most important areas. So
here we go. Degradation of integrated -- degradation
of engineered barrier pops up as -- when we run the
TPA code as being one of the more risk significant,
integrated subissues. Okay?
And you can get to the point where you're
identifying specific parameters that are important to
the overall system performance. And you can also
address the barriers.
So we're using these various sensitivity
analysis techniques and total system performance
assessment code to gain insights on which portions are
important to waste isolation, which parameters matter,
and we can gain insights on the different barriers'
performance.
So what are -- in an overall picture, what
are some of the results? And what drops out? Now,
the committee has heard this information before, and
I'm just trying to summarize it in a different way.
What are some of the significant features
and processes that our sensitivity analysis tell us
are important? And this is a list. I'm not really
going to go through it. You should hear in the
subsequent talks how the KTIs are contributing to
things. Direct release of volcanism -- Brit will walk
you through that. But from an overall system
performance assessment, this is what our code is
telling us.
Okay. But -- those are the things that
our code is telling us are important. Well, you can
also get insights on what's less important. Less risk
significant features and processes in our own
assessment and our code, how we've seen -- well, we
suggest -- our results suggest that unsaturated zone
flow and transport is relatively unimportant.
This means that the two integrated
subissues -- flow paths in the unsaturated zone and
radionuclide transport in the unsaturated zone --
those integrated subissues are relatively less
important, at least in our view. That's only part of
the question -- part of the picture, and we'll get to
the rest of the picture in a second.
Also, direct mechanical disruption by
falling rock and direct mechanical disruption by
faulting are also of relatively low importance. But
those two items are focused and figure -- feed into
that mechanical degradation integrated subissue.
If you only focused on this, you'd say,
then, why are we spending any time on these items?
Okay? Well, you have to put it in the regulatory
perspective. Our focus of issue resolution is guided
by the identification of what is risk significant, and
also what is DOE's potential safety case. What is DOE
relying on?
So, again, the focus of the issue
resolution, as we went into these meetings, we knew
what were the most important things. These were the
things we tried to focus on in those meetings, in each
KTI, keeping in mind KTI might be unsaturated zone
flow and flow and transport under -- I mean, excuse
me, unsaturated zone, saturated zone, under isothermal
condition, USFIC.
Well, what portion of USFIC is important?
Well, it turns out the quantity of water. So
understanding the seepage is important, and so if you
look at the KTIs you'll see that there is a focus on
those things that are important.
In the latter part, in the Department of
Energy's approach, mechanical disruption of engineered
barriers is important for its absence in its
performance assessment. They're relying on -- they're
relying on screening out mechanical degradation.
Okay. So if you're going to screen out this process
that we suggest may have some impact, you've got to
have an adequate technical basis to provide that
screening out.
And our concern here, really, is as -- as
rock fall damage could lead to stress corrosion
cracking -- which, again, the waste package is an
important barrier that we know impacts the overall
system performance. So, really, we're asking for the
technical basis for them for screening out, and I
think the --
MEMBER GARRICK: Bret, when you say
quantity and chemistry of water, do you in the
chemistry mean also the quality of the water?
MR. LESLIE: Yes. The chemistry is the
quality. What --
MEMBER GARRICK: So the composition issue
is covered in the chemistry of --
MR. LESLIE: Exactly. The chemistry of
the water means the composition.
Also, again, the Department of Energy, in
their supplemental science and performance analysis
report, relies on the unsaturated zone quite a bit for
performance. Our own assessment suggests that it's
not very important. Okay? Well, DOE wants to rely on
it. They are under the -- they have the burden to
support that. And so what we're asking for is
information to support their assertions of significant
performance in the unsaturated zone.
So, again, issue resolution -- the big
picture is guided by what the Department of Energy is
proposing, us identifying the gaps and using our risk
insights.
Now moving on to the slide that you
probably want to know -- prioritization of the key
technical issues. What I just said is summarized in
this first bullet and what Bill indicated. All of the
KTIs are needed for a high-quality license
application, and that's, again, constrained by our
limit by law of a three-year review period.
So the answer is all of them are needed.
That's the first answer.
What is needed is a function of what the
Department of Energy proposes for their safety case.
If they had a simpler safety case or wanted to rely on
things that were less -- had less gaps, let's say,
then our request for information or our agreements
scope might be smaller.
So the agreements are, again, a function
of what the Department of Energy is proposing, our
identification of the gaps, and our request for DOE to
close those gaps.
And I'll give you an example. We can only
be as risk informed as the Department of Energy.
There might be things that they can screen out by
doing a simple consequence analysis. Criticality may
be one. Okay?
Well, the Department of Energy has chosen
-- that's their choice -- not to screen out on showing
what the consequences are. Instead, they've adopted
a different approach. We have to respond to what
their approach is. We don't decide what they do.
They decide what they want to do and the approach they
take.
If the approach that they're taking isn't
very risk informed, we can't make them become risk
informed. We can show the risk insights, but it's up
to the Department of Energy to decide how they want to
put their case together or address a particular issue.
The results from the performance
assessment, I went through that in the previous
slides, slides 6 through 10. We've also tried to get
at the overall complexity of the subissue agreements.
And we added these slides at the very end. And if
you'll kind of look at 32 and 33 -- these are in your
backups at the very end -- what we did is -- is -- and
this is a fairly recent -- well, not so recent effort.
It happened before the site recommendation, so this
table is somewhat dated.
The first thing you'll note is 282 does
not equal 293, and that's because we didn't include
pre-closure and we didn't include subsequent technical
change. So don't focus too much on the numbers. But
what we did is we tried to break out by subissue in
each KTI the number of agreements first.
And we polled the staff and said, "Okay.
Based upon what you're expecting in the scope of the
agreements, do you categorize the information need as
minor, moderate to major, or major?" So this is
another way of focusing within a key technical issue
where is the dominant focus.
And so I'm not going to discuss this
anymore. I just wanted -- it's information for you to
have in the backup.
And now I'd like to move on to the third
part. We realize -- and the committee has -- has
written -- that it's not always clear that we are
using risk insights or risk informed in the issue
resolution process. And there could be a couple
reasons. One, we're not communicating it. Okay? Or
we're truly not risk informed.
And so Bill and I and a couple others have
brainstormed and said, "Well, what can we do to
improve that communication and make sure that our work
is truly risk informed?" And we've come up with this
project that I'm going to be working on over the next
month, six weeks, and this is anticipation of the
DOE's rebaselining out in March, that we need to be
prepared for that discussion with the Department of
Energy.
But, really, it's to document the insights
and tie it to the resolution of the KTIs. And the
idea is that we need to be better communicating to the
ACNW and others what are those risk insights tied to
those agreements. Why are we asking this information?
Both from a regulatory perspective, but also from a
performance assessment perspective or a risk
perspective.
We also want to make sure that we're
properly implementing risk insights into the program.
The timing is near term. Again, the Department of
Energy, in the March timeframe, is likely to come out
with a rebaselining of their activities, what they
think are most important to complete.
And so in the near term we expect to have
some -- make some -- make some observations and
document them and create some guidance. But at the
end of the year, the total system performance
assessment KTI is planning a major intermediate
milestone that would document all of these risk
insights.
But the idea is that we get our staff
prepared for that exchange with the Department of
Energy. Again, the idea is that we document very
clearly in terms of the agreements, why are we -- why
do we have this agreement? What risk insights can we
apply to this agreement? What are the drivers for
this agreement?
And, finally, the outcome -- the other
outcome is guidance from staff. Are there areas where
we -- if we find that there are ways we can better
communicate this risk information, we'll try to
document that and make sure it's implemented.
So who will participate? There's a core
group that's leading it. It includes Chris Grosman of
the PA staff, myself, and a couple people from the
Center. Everyone is going to participate. We're
going to be working together with the issue resolution
staffs.
We're going to sit down with them and kind
of go through what information can they bring to the
table, what can the PA people bring from -- to the
table in terms of the overall system -- you know,
overall and -- and system level, and detailed process-
oriented level analyses, what are the risk insights.
And so we'll be going down and talking to
the key technical issue teams. Again, we're getting
the risk insights from and for each key technical
issue. We're going to be going through the issue
resolution agreements, each of the agreements, to see
indeed that they are risk informed.
And also, we may try to get some of the
insights from the recent licensing activity such as
PFS. What have they found in terms of their risk
informed analysis? What insights might they give to
us to help us along on our path as well?
So I think by now you understand what the
purpose of this presentation was. It's, again, to
reiterate what the overall goal of issue resolution
is. It can't -- it's informed by risk insights, but
the overall goal guides our work. In other words, a
sufficient and high-quality license application, it's
dependent upon what the Department of Energy's case
is.
I think you have seen that the risk
insights are derived from a variety of techniques and
from different sources, and you'll hear more about
those risk insights from each of the key technical
issues. And we're also interested in improving our
program through the risk sights initiative.
That's kind of a summary of my
presentation, but I want to kind of lay out the
groundwork for the rest of the presentations. This
might not be the right order. This was the order that
we had proposed a couple days ago; management had
agreed.
But basically we want to start off with
the total system performance assessment perspective
with James Weldy, and then we're going to go to
igneous activity, Brit Hill. This is one of the areas
that the committee is keenly interested. We want to
get your enthusiasm up early in the morning. And the
other two that we had -- that I know Ray and others
are interested in are later in the afternoon, so we
can wake you up again.
(Laughter.)
But each of these presentations is going
to have the same format and feel, and this is -- this
is because of the short time constraints we had, we
have relied on presentations we have already given
elsewhere. And so it might not be the right title on
the top of the slide, but we think we're trying to
address the information.
And, basically, each of the following key
technical issues presentations will talk about the
status of the subissue resolution, discuss the risk
information insights which include what DOE needs to
provide before any potential license application, and
what the NRC needs to do before any potential license
application.
And the idea is that they are going to be
providing select examples of how those risk insights
have been applied to the Center's work. Why is the
center focusing on in the near field what salts
matter? Okay? And Bobby will go through why that --
why is that important and the type of work they've
done, and how they will apply risk insights in this
fiscal year coming up. And then each of the KTIs will
have a summary.
And we'll be flexible. If you're finding,
you know, after two or three or four of these that a
particular portion of the presentation isn't
scratching the itch and we're running short on time,
we'll be flexible and revise on the fly as needed. So
that's kind of -- yes?
MR. REAMER: Bret, Bill Reamer. Just on
the sequence. I think the order has been changed
slightly. The fourth item, unsaturated and saturated
zone flow, will be moved to the container life and
source term item, and it will be moved up to the
fourth item.
MR. LESLIE: Right. We had intended to
keep these two together. They're intimately related.
But one of our staffers -- I think Tae Ahn -- who is
involved in container life and source term can only be
here this morning, so we switched it around. So it
won't -- we'll only have half the punch in the
afternoon.
The backup slides, again, are -- I'm not
going to go through, but the first two are the
relationship to this chart. You know, you probably --
might not be able to see this chart, but you might
want to pull them aside, so as you're going through
you can see the relationship to the KTIs and the
integrated subissues.
And then the subsequent pages from
pages 19 through 30 are really out of our comments,
the sufficiency comments, the November 13th letter.
And they are a synopsis of what kind of information we
need from each of the key technical issues. They also
provide the reference for where our agreements are.
And, again, you might want to rip that
page out as you go along and have it in front of you
as we go to each of the key technical issues. And,
again, the final two of the backup slides are the
level of complexity.
With that, I'll stop. I know that's a
lot.
CHAIRMAN HORNBERGER: Thanks, Bret.
Questions for Bret?
MEMBER GARRICK: Bret, the key technical
issues have been around several years. Has the list
really changed?
MR. LESLIE: I think the statement stands
by itself. Yes, they've been around. The list hasn't
-- the key technical issue list hasn't changed.
MEMBER GARRICK: Yes.
CHAIRMAN HORNBERGER: Have the subissues?
MR. LESLIE: Have the subissues changed?
MEMBER GARRICK: Well, I guess what I'm
thinking of, when the key technical issues were first
generated, the discussion and thought processes were
very much geotechnical oriented. There wasn't nearly
the emphasis that now exists on the engineered
systems.
So I guess what you're saying is that they
are sufficiently global or high level that they
probably haven't changed, but to be sure the subissues
have been quite dynamic and affected by that -- by the
change in emphasis with respect to engineered
barriers, for example.
MR. LESLIE: Right. While the emphasis
might have changed, we believe that to describe the
overall system these -- this is the information you
still need to do.
MEMBER GARRICK: Right.
MR. LESLIE: You can't just do an analysis
on a particular barrier. The idea is that the
emphasis and the resources and the focus may have
changed, but overall to describe repository
performance you still need that information on all
those issues.
MEMBER GARRICK: Yes. I guess the other
thought here is, how complete is the set? In the
course of time, at least at the KTI level, have we --
if we had our druthers, would we add or subtract any
KT -- one of the things you said was that all of the
KTIs have to be addressed I think you said by law.
MR. LESLIE: Well, I didn't say by law,
but --
MEMBER GARRICK: And so as far as the
overall and global considerations are concerned, there
hasn't really been any change in several years, since
the list first pretty much came out. And that's --
what's that been, five years or some time --
MR. LESLIE: Something like that.
MEMBER GARRICK: Yes.
MR. LESLIE: You could bend this a lot of
different ways.
MEMBER GARRICK: Yes.
MR. LESLIE: And what we're focused on is,
do they have the necessary information? You could
bend it by integrated subissue, and then you'd have 13
or 14. The information I believe remains the same
however you bend them. I -- we have worked by KTIs.
I don't think we're planning on changing the number of
KTIs. It's the information that matters.
MEMBER GARRICK: Okay. At one time we
were trying to --
MR. PATRICK: Bret, could I interject?
MR. LESLIE: Yes.
MR. PATRICK: Bret, if I could, please,
just interject, I think Dr. Garrick did make an
important point, that the KTIs are quite broad in
their scope. We continue to reexamine, as we conduct
process level modeling, to assure ourselves that they
continue to cast a net that's sufficiently broad that
nothing is missed on the one hand, and on the other to
eliminate those items that are no longer in some cases
even relevant, frankly. The part where the evolution
has occurred and is most visible is at a much lower
level, down in the subissue level.
A good example, you mentioned, Dr.
Garrick, the emphasis on container life. Even more
specifically, you can point out that as the selection
of metals that DOE is proposing has evolved over time,
there are new phenomena that we did not consider in
1996 for instance that we now see as very important,
because different classes of materials are being used.
Likewise, there are some phenomena that we
were examining back in those days that are no longer
relevant for these new materials that are being
considered by the Department of Energy. So at that
level, there has been a good deal of evolution, but
it's still -- in that example case, it's still all
within the broad container life and source term key
technical issue.
MEMBER GARRICK: Yes.
CHAIRMAN HORNBERGER: For the record, that
was Wes Patrick from the Center.
Wes, you weren't here this morning when I
asked everyone to make sure they introduced
themselves.
MEMBER GARRICK: Thanks, Wes.
MR. PATRICK: Thank you.
CHAIRMAN HORNBERGER: Raymond?
VICE CHAIRMAN WYMER: Yes. As you know
very well, we've been interested in the degree to
which evidence supports the DOE case. And so you've
been very interested in that, too. But my question
is, then, one you probably can't answer very
precisely, is what criteria or formal process or what
mechanism do you have for deciding how much evidence
is needed?
MR. LESLIE: Part of that depends on how
the Department of Energy -- for instance, in screening
out, they must provide a technical basis for screening
out. Well, it talks in terms of you can screen it out
if it -- or you do not need to examine it in detail if
it does not make a difference in the magnitude or time
of dose. Okay?
If DOE came in and did a calculation that
it did not matter in terms of times and dose, and we
could agree that that was a robust calculation, that's
sufficient. If they want to make arguments -- and
I'll give the example of rock fall, that -- that they
are suggesting that they would design away rock fall.
Okay?
Without having given us a design, okay, we
need to ensure that the information that would go into
the design decision is complete. So that means how
many rocks, how big rocks. You know, it depends,
again, on how they frame what they want to do.
VICE CHAIRMAN WYMER: It's a very
judgmental thing.
MR. LESLIE: Yes.
CHAIRMAN HORNBERGER: Milt?
MEMBER LEVENSON: Yes, I've got I guess
two questions, one a very broad one and one a very
narrow one. The broad one -- the objective of the KTI
program, as I understand it, is not to resolve
specific issues. That gets done when you review a
license application. It's to assure yourself that
you've identified all of the issues to be addressed in
the license application. Is that correct?
MR. LESLIE: It's not just the
identification. It's that sufficient --
MEMBER LEVENSON: No. What information
would be provided to back up those issues? That
doesn't mean that just because you've identified that
information that it will be acceptable. You still
have to review it all --
MR. LESLIE: That's absolutely --
MEMBER LEVENSON: -- the license
application.
MR. LESLIE: -- correct.
MEMBER LEVENSON: Okay. My question is:
Isn't there a loose end in your ability to identify
everything you're going to need, since you have not
seen the formal safety case that will be presented by
DOE in its license application?
MR. LESLIE: I'll answer this -- that
this --
MEMBER LEVENSON: I mean, what you're
looking at is everything you need to know to review
not the generic question but specifically a license
application and the safety case it makes.
MR. LESLIE: We define issue resolution as
in terms of what we know now, and we've emphasized
that to the Department of Energy. As they make
changes in design, or provide additional detail, there
may be additional questions.
So in a way, yes, you're right. We're
hoping, you know, that whatever DOE decides they stick
to it, because if they change in a big sense and bring
in a lot of new issues, we're going to have to review
it to make sure that whatever they propose is, you
know, supported.
MEMBER LEVENSON: But what I'm saying is
that you're not in a position to say you have
identified all of the necessary KTIs until you've seen
their actual formal safety case.
MR. LESLIE: I would disagree with that,
because this is what is needed in terms of knowledge.
Whatever their design is has to take into account what
the reality of the situation is, that certain
processes occur, you know, groundwater flow, and so on
and so forth.
How they emphasize things in their
potential safety case is up to them. The emphasis
might change, but we believe we have everything that
is necessary to describe the system.
MEMBER LEVENSON: Regardless of what their
safety case is.
MR. LESLIE: I believe that's where we're
at.
MEMBER LEVENSON: Okay. The other is
very, very detailed. What is subissue RDTME4?
Because it's on your slide but not on your chart
there. And, secondly, while it's on your slide, it
isn't relevant to any integrated subissue?
MR. LESLIE: Where is it on the slides?
MEMBER LEVENSON: And it has major impact,
right?
(Laughter.)
MR. LESLIE: Well, I --
MEMBER LEVENSON: I just want you to know
I looked at your slides. It's number 18.
MR. LESLIE: Yes. Part of this -- and I
didn't state that when I showed this in the backup
slides. This is a crosswalk. We have -- you know,
we're evolving, and as we've written the Yucca
Mountain review plan, and as we're developing the
integrated resolution status report, the boxes might
change, the overall perspective might change.
When we first started KTIs we didn't have
a Part 63. RDTME number 4, repository design and
thermal mechanical effects, subissue 4 was seals.
Okay? Well, seals was in Part 60. Part 60 is no
longer relevant. So this is an old slide, and you'll
see that I -- I actually caught it here and bend it --
you know, there's no RDTME4 on this thing. I just
didn't have an opportunity to change the slide last
night after I saw that, so thanks for --
MEMBER LEVENSON: Okay.
CHAIRMAN HORNBERGER: Bret, I guess I have
a question that follows on Ray's and Milt's question
I think. It's hard for -- I know I want to ask you a
general question and not a specific, but just to give
the flavor I'll just refer to one specific. But my
question is general.
There is plenty of room for disagreement
among scientists on these -- the kind of issues we're
dealing with here. So as an example, you showed in
your slides that your analysis of the unsaturated zone
is significantly different from DOE's analysis of the
unsaturated zone. And there may even at the end of
the day be room for disagreement among scientists.
And my question is: are you comfortable that you have
a process, either through -- well, somehow I guess
through the issue resolution process, to deal with
these kind of disagreements.
MR. LESLIE: That's a great question,
because if you look -- for instance, we'll give the
example the conceptualization of the unsaturated zone.
Part 63 says evaluate alternative conceptual models.
And, in essence, that's what you're saying. The
Department of Energy has a conceptual model where
transport times are very slow in the unsaturated zone.
Okay. If you look at what the unsaturated
folks have asked for in information is adequate
information to support evaluation of both conceptual
models. In other words, give us your results of the
chlorine 36 study, which suggests that the DOE's
conceptual model may not be correct. We need
sufficient information to evaluate both alternative
conceptual models. There's a requirement to -- for
the Department of Energy to do that.
So, yes, disagree -- we're not saying that
a particular model is correct. But if there are
alternate conceptual models that can address the same
issue, that adequate information -- and that those
alternate models are considered in the performance
assessment.
CHAIRMAN HORNBERGER: Questions from the
staff? John?
MR. LARKINS: Quick question. I'm just
curious. What constitutes major, moderate to major,
and minor? Are these levels of uncertainty? Timing?
Data needs? Or what types of considerations went into
this categorization?
CHAIRMAN HORNBERGER: Tim McCartin will
give us a presentation on that next time.
(Laughter.)
MR. LESLIE: No, that's not the answer.
(Laughter.)
I'm not intimately involved with what --
this was an effort put together by the Center, and it
incorporated the KTI resolution staff here. But the
idea was, okay, take a look at the agreements you've
asked for. What is the scope of information? Are you
talking about long-term testing? How large are those
uncertainties?
We'll go to the engineered degradation
barriers -- or degradation of engineered barriers and
container life and source term. Of those subissues,
what is the key driver? Okay. And how much
information is needed to support that?
So it's qualitative. It wasn't
quantitative. This was an informal type of analysis
to try to get -- another way of trying to get a handle
on how much information is really required. So the
answer is there aren't specifics. It was staff's
effort to try to assess what level of effort was
really needed or the complexity of the information
needs.
MR. LARKINS: Okay.
CHAIRMAN HORNBERGER: Okay. Thanks very
much, Bret.
Let's see. I think the schedule says
we're due for a break, and I think that we'll take a
break. Let's reconvene at 9:50.
(Whereupon, the proceedings in the
foregoing matter went off the record at
9:40 a.m. and went back on the record at
9:53 a.m.)
CHAIRMAN HORNBERGER: The meeting will
come to order again.
So the schedule that I have says that
we're next going to hear from James Weldy on
performance assessment. Is that right?
MR. WELDY: That's correct.
Good morning. I'm James Weldy, and I'm
going to be talking today about the status of
resolution of the TSPA and I key technical issue. The
outline of my talk is we're going to talk about status
of issue resolution, what DOE needs to provide, what
NRC and CNWRA are going to be doing before license
application to prepare for the review, talk about what
we've done in fiscal year 2001 and what we're planning
on doing for fiscal year 2002.
Current status of issue resolution for all
four key technical issues, subissues, is closed
pending. We've reached agreements with DOE to provide
the appropriate information by the time of license
application, so all four subissues are closed pending.
None have reached closure at this point.
And, of course --
CHAIRMAN HORNBERGER: James?
MR. WELDY: Yes.
CHAIRMAN HORNBERGER: I just want to
check, is the microphone adequate? Any need to move
it closer? Is it okay for the recorder? Oh, it is
okay. Okay.
MR. WELDY: Are there problems in the back
of the room hearing me? Is this better? Okay.
And I just wanted to point out the
limitation within the overall performance objective
subissue that the achievement of closed pending -- the
closed pending status is only related to methodology
issues, not related to any sort of determination that
-- as to whether DOE has met the standard or will meet
the standard or not. It's not a comparison of the
dose limits and the groundwater protection limits.
So going into detail about what DOE needs
to do, based on our agreements before license
application, within subissue 1, multiple barriers, DOE
has presented a number of different methodologies,
different techniques that they have in mind to make
their case for -- that they have multiple barriers.
But they really haven't put together an overall case,
how they're going to use these techniques to address
all of the issues and concerns within multiple
barriers.
The issues such as barrier -- variability
of performance of the barriers, independence and
interdependence of the different barriers, and
parameter model uncertainty considerations -- how are
they going to put all these techniques that they have
together and make a case to say we do have multiple
barriers in the repository system?
And to -- their schedule for addressing
these issues is that in fiscal year 2002 they will put
together the approaches, and by the license
application they'll present the results. And we -- it
resulted in two agreements for which the NRC and DOE
came to agreement, and we have not, of course,
received any information yet due to the late date of
the TSPA technical exchanges in August 2001.
This will be a common theme throughout,
that we haven't received much information to resolve
the agreements at this point.
MEMBER GARRICK: Can you just say
something briefly about the -- the content of the
agreements -- not a big elaboration, but just to give
us some sense of your own classification of minor,
major, etcetera?
MR. WELDY: I would say that they are
relatively minor in the sense that they really do have
the techniques already that they've put together, and
it's just really explaining how they're going to make
that determination as for what they need to reach
closure on the issue.
Since they have a lot of the information
put together, all of the -- a lot of the techniques
put together already, and the techniques seem pretty
reasonable, and they're the types of things that we
would consider using as well to make that sort of
argument, we believe that it shouldn't be too
difficult for them to put together all of the
different techniques and put it together in a
comprehensive argument that they do have multiple
barriers.
MR. BAHADUR: May I ask --
CHAIRMAN HORNBERGER: Go ahead, Sher. You
need to use the microphone and --
MR. BAHADUR: This is Sher Bahadur of ACRS
staff, ACNW staff. The total system performance -- I
was looking at the number of agreements in the list
that was provided before, and 42 elements are
classified as major. And so it -- most of these have
not been resolved yet, or do you have any feel for
those?
MR. WELDY: As for status resolution, like
I said, we had a late technical exchange where we
reached closure, so -- a closed pending status. It
was just in August, so DOE hasn't provided a lot of
the information -- much of the information yet to
close those agreements.
As for the number of agreements, the 42
classified as major, I'd --
MR. WHITMEYER: May I interject? This is
Gordon Whitmeyer from the Center. I don't know
exactly which chart you're reading from, Sher, but I
believe what is shown or should be shown, if there's
the word "major" next to one of the subissues it may
be TSPA and I number 3.
MR. BAHADUR: Right.
MR. WHITMEYER: And there are 42
agreements. That's a summary that's saying that if
you look at the total number of agreements, and the
amount of information that's required of DOE to meet
those agreements, it requires a fairly major effort on
their part. Doesn't necessarily mean there are 42
major agreements. I don't know if that helps clarify
things or not.
MR. PATRICK: And that also deals with
model abstraction and --
MR. WHITMEYER: That's the model
abstraction --
MR. PATRICK: -- multiple barriers --
MR. WHITMEYER: -- for subissues.
MR. WELDY: That's subissue 3, which we'll
be getting to shortly.
MR. BAHADUR: Thank you.
MR. WELDY: Okay. Within subissue 2,
scenario analysis, DOE has completed FEPS database,
which they have made the argument is comprehensive and
complete. And, in our opinion, it seems to cover
everything, but the biggest problem we've found with
it is more transparency and traceability issue.
The real concern is that they really have
different levels of broadness for their different
FEPS, and for a number of the FEPS it's difficult to
tell features, events, and processes. I apologize for
the acronym.
For a number of the different features,
events, and processes that they've defined, it's
difficult to tell exactly what falls within the scope
of that features, event -- that FEP. It's just some
things are so broad that they can lump any issue that
they -- that gets brought up into that FEP and
indicate that it will be covered -- that they will
address it during the modeling rather than during the
screening argument.
So it's -- it's hard to tell whether they
have put together a completely comprehensive list, if
you can't tell what's in with -- within each of the
FEP, and what they actually thought of when they were
developing it.
So the -- that's the major theme of the
issues that we've brought up in the agreements that
we've made with the DOE -- is that we really need to
be able to tell what's in the FEP and what they're
doing with it and how it's getting propagated into the
performance assessment.
Of course, there are additional agreements
on clarifying their screening argument or providing a
more acceptable screening argument for some of the
FEPs, but we don't think that there is anything -- any
huge issues that they've missed that wouldn't be --
couldn't possibly be covered by some of their
currently defined FEPs. It's just not very
transparent or traceable to find out what's going on.
Of course, this information is required to
demonstrate that the -- they have considered
everything that's important within their performance
assessment. And we reached seven agreements within
the subissue to achieve closed status, and we have not
received any information from the DOE on those
agreements at this point.
For subissue 3, model abstraction, this
one is a very broad category, and the agreements cover
a number of different issues. Primarily, we tried to
focus on methodology issues, but there are also a
number of specific technical issues that -- for which
we had to reach agreements in this category.
But they tended to boil down into some
major categories, such as how they've represented
uncertainty within the system, and have they done it
consistently from abstraction to abstraction, have
they represented the uncertainty from detailed process
level models appropriately within the abstraction when
they went from the detailed model to the abstraction.
Other issues include the integration of
the different process level models. Are they
consistent between the different models? And
transparent explanations of how they've simplified
models, how they've used conservatism, how they've
represented parameter uncertainty.
There were a number of consistency issues
that they found that while not necessarily wrong or a
real problem just were not consistent between
different models and was not clearly explained why
they were inconsistent, which we didn't -- we thought
definitely needed a little more explanation within
their safety case.
And, of course, the information on model
abstraction is necessary to ensure that data
collection, model development, and treatment of
uncertainty are adequate to provide a basis for their
performance assessment. And we've reached 43
agreements within this subissue, and so at this point
they have provided information to complete one of the
agreements, and that information was sufficient to
provide the information that the NRC needed. So it is
considered resolved by the NRC staff at this point.
CHAIRMAN HORNBERGER: Are these -- again,
just to get a general feeling for the nature of these
agreements. Are these agreements for DOE to either
modify or augment their AMRs to provide the
traceability to the model abstractions? Is that the
flavor?
MR. WELDY: That's probably a reasonable
characterization of a lot of the agreements. Based on
the wide range of the model abstraction, there is a
mixed bag, of course. There's a number of very
detailed, specific technical comments that we included
in here. Everything on the biosphere was included
under the model of the TSPAI KTI, since it doesn't
have a KTI of its own to reach closure on. So there's
a number of detailed technical comments related to
that within this group. So it's a fairly wide range.
Now, has that addressed your question?
MR. BAHADUR: Yes.
MR. WELDY: Okay, good.
CHAIRMAN HORNBERGER: Yes, Jim?
MR. FIRTH: James Firth, NRC staff. I
guess I would point you back to the presentation that
we gave back to the committee where we laid out some
of the specific agreements that we got, and this was
the August 29th presentation.
That goes through -- a lot more specific
in terms of what the agreements were. And to
reemphasize what James had said is that a number of
these things address similar concerns to what was
being addressed by the KTIs in their meetings, but
reflect what was the information in the TSPA SR
documentation which was not available at that time.
So there's a range of modeling questions
and concerns -- questions that we had that were being
addressed through the agreements.
There were also cases where, as we started
doing things from the ISI perspective, to make sure
that everything was covered in the KTI meetings
earlier, that there were a couple things that we felt
that we didn't have the information, to make sure that
we have the information consistent with what we would
see in the eventual Yucca Mountain review plan, and to
show -- to meet the acceptance criteria. So we'd have
the information to be able to do a review.
MR. WELDY: And, finally, subissue 4,
overall performance objective. The agreements that we
reached -- one of the bigger issues that we found and
had with them is their strategy and implementation of
their software verification and model support or
validation that the DOE has been doing for their
software.
We really need to get some more
information and get improved results from that from
DOE to have confidence in the results of their models.
Other things that we've reached agreements
on within overall performance objective is looking at
more details on their -- how they are running their
TSPA, how they're demonstrating convergence of the
results, how they're demonstrating that they've
addressed variability sufficiently with their time
steps, with the level of discretization, and make sure
that they're not running into incorrect results just
due to the way that they set up the models.
And this information is required, of
course, to provide confidence in the results of the
TSPA and any decisions that might be made on -- as a
result of the TSPA calculations. And within this
subissue, we reached seven agreements and have not
received any information on these agreements yet.
Well, that's what DOE is going to be
doing. What are NRC and CNWRA going to be doing up
until the time of license application? Of course, one
of our primary jobs is going to be monitoring what DOE
is doing, reviewing the information that they provide
us to meet the agreements and make sure that the
information is acceptable and provide them feedback
where we find that it's not acceptable.
And we're going to continue to participate
as an observer in the DOE quality assurance audit to
monitor issues such as validation, verification,
transparency, and traceability within their documents,
and make sure they are moving forward in the right
direction.
We're going to complete development of our
own TSPA, the TPA code, perform sensitivity analyses
and visualization tool to help us get insights into
how the system is behaving, focus our review and
enable us to really focus on those things that are
most important to the DOE's safety case based on our
own independent modeling.
And we're going to conduct a thorough
review of the consideration and propagation of
uncertainties from detailed models to the TSPA model
and look at where you could run into problems, where
you could run into issues of risk dilution, so that
we're able to do a good review of the DOE work.
In fiscal year 2001, the CNWRA and NRC
staff outside of the issue resolution activities,
primarily we've been working on developing tools to
help us review any DOE potential license application.
We've been updating the TPA code, which is, of course,
our own tool for assessing the performance of the
system. We've been performing uncertainty and
sensitivity analyses to identify the scenarios,
processes, models, and parameters that influence the
results the most, so we can really focus our review of
the DOE work.
We've been working on approaches to
understand and gain insights on overall performance of
the system, and we've been doing some confidence-
building efforts on our own tools. So we've been
starting to develop a software validation test plan
for our own TPA code, and also the TPA code was
subjected to an external review a couple of years ago,
and we're continuing to work on responses to the
external reviewer's comments and have just about
completed that activity.
And those sorts of activities will help to
provide confidence in our tools, so we are focusing on
the right thing, make sure we're focusing on the right
things for -- of the DOE TSPA.
We've also been working on pre- and post-
processors for the TPA code to make it easier to use,
make it easier to extract results, and make it easier
to understand what's going on in the system.
Some more details on the sensitivity
analyses that we've been doing over the past fiscal
year, we've been looking at distributional
sensitivity. Some of the distributions defined within
the TPA code may be defined on just a few data points,
and we have to assume what distribution corresponds to
those data points. And so we're looking at what would
the effect be on the results if we -- instead of a
normal distribution we selected a log normal? What if
we shifted the mean up a little bit? How much would
that affect results?
We've been looking at alternative
conceptual models and how that could affect results.
If we replace the model of one repository system with
an alternative, such as the dissolution rate of the
spent fuel, how much will that affect performance?
And we've been doing barrier analyses to
gain insights on barrier capabilities and importance
to the performance of the repository, and also to get
insight into barrier capabilities through the --
looking at intermediate results, how barriers affect
the transport of materials through the system.
For 2002, we plan on continuing some of
the similar activities and adding some new ones. We
plan on continuing the issue resolution process,
reviewing DOE work, providing comments, resolving
issues, narrowing down those agreements and seeing if
we can reach closed status on some of the TSPAI
subissues.
We're going to be developing Version 5.0
of the TPA code which will be another major step
forward in the TPA code, replace -- evaluate some of
the simplifications in the code, try and improve the
models, and get an idea of what some of these
uncertainties will have on the performance of the
system.
We're going to be doing more uncertainty
sensitivity analyses, the quantitative evaluation of
the TSPA and TPA code results. We're also going to be
looking at detail at the DOE's TSPA code. We've
acquired the software, and we're going to be looking
at the implementation of how they're doing things and
see if we can -- if we identify any concerns with
what's going on, that you can't do just by looking at
a document but you actually have to look at the code
itself.
And we're also going to be starting off on
a risk insights project where we document those issues
that -- the risk insights that we use to help develop
the agreements with the -- within TSPA and with the
other KTIs. And what we did and what we looked at to
identify those things that needed agreements and were
important to the performance of the system, and also
work on a way to communicate those risk insights to
members of the public and other NRC stakeholders.
And brief summary -- we achieved closed
pending status on all four of our subissues in the
last fiscal year, and we're going to continue to
monitor DOE work to ensure that they are keeping up
with the agreements and submitting high-quality
products that meet the needs of -- the informational
needs that we identified to write the agreements.
We're going to continue to work to improve
our understanding of the repository system, and we're
going to put some more effort into identifying,
documenting, and communicating risk insights that we
have with other NRC stakeholders.
And with that, I'll open up the floor to
questions.
CHAIRMAN HORNBERGER: Thank you, James.
Questions from the committee? Milt?
MEMBER LEVENSON: Since today is
January 9th, it's time for me to ask the two questions
I ask every six months. Since you're presenting on
the overall assessment integration, does the system
current -- have they added conservation of mass to the
overall analysis and system, or is it still lacking?
MR. WELDY: Within the DOE TSPA code?
MEMBER LEVENSON: Yes.
MR. WELDY: My indication is that they
have not done a full conservation of mass at this
point. Randy, would you like to explain?
CHAIRMAN HORNBERGER: You have to come to
a microphone and identify yourself and -- for the
record.
MR. FEDORS: Randy Fedors from the Center.
I believe there's a technical agreement on those that
they will look into that, and I -- it's a closed
pending because they'll develop the approach.
MEMBER GARRICK: So you can ask the
question again in six months.
(Laughter.)
MEMBER LEVENSON: I think this is now the
fourth year, so --
(Laughter.)
The second question is a similar one. We
got a presentation a few months ago at which one of
the analyses -- they just removed a few of the extra
so-called conservatisms. It changed the result by a
factor of 104 on the dose. It's my feeling that we
have no idea how many places things are either
overestimated or underestimated in the name of
conservatism, because it seems to be done by modelers,
by data input people, by abstracters, all along the
way.
Are you people making any effort to try to
not necessarily correct all of this but get a feel for
how far from realism the analysis has been pushed in
the name of -- I don't like the "conservative" because
overestimates and underestimates are not necessarily
conservative, even if you think they might be. But it
seems to me it would be fairly important in the final
analysis, before you can identify what is the real
risk, you have to have a handle on all of these
perturbations.
MR. WELDY: Yes. And, specifically, to
answer your question, we do have an agreement in the
TSPAI subissue, under subissue 3 -- I don't know the
number off hand -- but that specifically relates to
DOE doing a much better -- a better job of documenting
how they're using conservatism, where they're using
conservatism, the guidance they're giving to their
modelers and staff on where and how to use
conservatism.
So we are looking at that issue and
ensuring that they are doing a consistent job across
it of at least documenting -- across the different
KTIs and different technical areas of at least
documenting where they are claiming conservatism and
why they believe it is conservative, or, if you don't
like the word "conservative," will lead to higher dose
estimates rather than not -- lower dose estimates.
Of course, when we're developing the
agreements, we look at -- if we find an issue that we
-- a technical issue that we have concerns about, we
do think about, well, will this lower dose or increase
dose at the critical group location? And if it's
likely to lower dose, we -- if the issue -- if the
problem results in a high -- is likely to result in a
higher dose, we generally still submit it to them but
at a lower priority level.
We do it as an issue of clarity and
documentation and generally won't get it -- a specific
agreement written for it. But if it's something that
results in an overly conservative -- or a result that
lowers dose, then we'll generally write an agreement
for it.
MEMBER LEVENSON: Our concern -- my
concern is that such complex systems -- it is seldom
obvious where it would lead to a higher or lower dose.
The classic case is, why don't we make the containers
a lot smaller so that as you analyze each container it
contributes a small amount to the risk. But a large
increase in the number of containers does not
necessarily reduce the final risk.
MR. WELDY: Right. And that's actually
the third category is where we just can't tell we --
MEMBER LEVENSON: Well, in a significant
fraction of the cases in -- especially non-linear
systems, but even when it is linear but complex, it is
not obvious what the effect of overestimating,
underestimating, which is why we keep coming back to
let's do the most realistic job and then add the
safety margins.
MR. WELDY: Yes, and there are certainly
areas where you can tell for sure, such as in igneous
activity. If you assume that the waste package
provides no protection and all the waste is -- that is
contacted by the magma is extracted from the mountain,
that's going to lead to higher dose results than if
you do more complex modeling and are able to defend a
greater level of protection.
MR. FIRTH: James, could I add a few
things?
MR. WELDY: Okay.
MR. FIRTH: James Firth, NRC staff. I
guess, Milt, going back to your original question, I
guess one of the important aspects is, has DOE
provided a sufficient basis to show that the
repository would be safe in terms of the long-term
performance? And that's slightly different than
knowing what the actual performance would be, and we
have regulations in place that we will be evaluating
what DOE is coming in with, and we'll be evaluating to
see that it is sufficiently safe.
That's a different question, and we don't
necessarily have to push DOE to find out what the
exact level of risk is. And I guess the second point
is on your follow up question -- we recognize that it
is very difficult in some cases to evaluate whether
something really is conservative in terms of leading
to higher dose estimates or not.
And that's why we do have an agreement
with the Department of Energy, so that they will be
more rigorous in terms of providing the evidence and
the basis to show -- to support their assertions that
this is the effect of this assumption or this way of
analyzing it. So that it's not quite as liberal in
terms of the use of asserting that this is
conservative, that it does get more evidence-based.
MEMBER LEVENSON: I just want to make one
thing very clear, and that is that I am very much in
favor of it being conservative. I just want to make
sure that what we think is conservative really is
conservative.
MR. FIRTH: Yes, and we would -- James
Firth, NRC staff. We would agree that we want the
basis to support that they are not being non-
conservative or leading to lower dose estimates
because they've not looked at things in enough detail,
that they are backing up their assertions with some
evidence.
CHAIRMAN HORNBERGER: Raymond?
VICE CHAIRMAN WYMER: Yes. Well, there's
quite a few subissues that are still to be resolved.
Some of them are not very important with respect to
the overall risk of the repository. Some handful are
important.
And in order to answer some of the
questions with respect to these subissues there's been
a request for a lot of detailed work to be done by
DOE, in some cases experimental work that will require
some time to complete.
My question is: with respect to timing,
and this may be kind of an unfair question, but are
there some agreements that are not likely to be
resolved by DOE within the time available for -- by
the time the license application is submitted?
And the kind of thing that comes to mind
right off the top is some of the corrosion issues like
alloy 22 that it's kind of hard to determine what the
10,000 year behavior will be of an alloy in a six-
month experiment. So are there issues like that?
CHAIRMAN HORNBERGER: What we might do is
actually defer that question to the rest of the
presenters, and perhaps ask you, in fairness, if there
are any with respect to TSPA itself.
MR. WELDY: Within TSPA itself, most of
them we expect will be resolved by the time of license
application. I mean, it's primarily modeling issues
and issues related to the DOE safety case itself, so
there is not really much there that can be deferred
until later.
There might be some model validation work
that will continue to be done to just confirm that
they have completed the work and confirm the
confidence in the model. But I believe that the
majority of it we would expect to be completed and
needs to be completed by the license application.
Jim?
MR. FIRTH: James Firth, NRC staff. I
just -- in terms of the performance assessment, like
demonstration of the overall performance measures,
multiple barriers, and scenario analysis, those things
-- the agreements there deal with things that we would
need in time of a license application.
Some elements of model abstraction --
there becomes more of a line in terms of what's needed
to make the decision for -- in terms of what DOE is
provided in terms of demonstrating that it's safe.
That may be confirmed through a performance
confirmation that would come later, but the premise of
all the agreements is that this is information that we
believe we would need to review a license application.
We're not establishing what DOE's schedule
is, so --
CHAIRMAN HORNBERGER: Yes, we know that,
and I don't -- you know, just to interrupt you, James,
I think what we're looking for is not necessarily a
long discourse on what your agreements expect. I
think Ray's question was we are interested in your
gut-level feeling as to whether there is -- might be
a potential problem in terms of delivering information
that you think is necessary in time for a license
application.
So pretty clear cut, the answer is no, we
anticipate that all this information will be
delivered. That's fine. We're just interested in any
insights that you might have, and maybe if Gustavo
wants to go out on a limb and say, well, I don't think
they can get this information in three years --
(Laughter.)
-- that's what we're interested in.
MR. CRAGNOLINO: Why me?
(Laughter.)
CHAIRMAN HORNBERGER: Did you have a
follow up?
MEMBER LEVENSON: Well, I was just -- you
know, inherent in the question Ray and you have asked
is an implication that these guys have some
information which I don't think they have, which is,
when are they going to get -- how long is it to
license application -- if you can estimate how long to
get the data.
CHAIRMAN HORNBERGER: Well, we realize
it's a question that puts people on the spot. We
recognize that it may be a question that people don't
want to answer directly, and yet it is a question that
occurs to us that as you go forward with these
agreements to provide information, as scientists you
might have a good sense as to how long this might take
and whether or not it's doable within a reasonable
licensing timeframe.
You may decide you don't want to answer
that question, but that's the thrust of our question.
John?
MEMBER GARRICK: I found several of my
questions answered in the backup slides, which are
very helpful.
MR. WELDY: Glad to hear it.
MEMBER GARRICK: And several issues that
you have planned to deal with in this following year,
in this year.
But I would like to ask, on the -- your
review of the supplemental science and performance
assessment, do you consider this a part of the TSPA?
MR. WELDY: It's additional information
that DOE has collected. It's certainly something that
we need to look at and understand what they've done
there, what new information they've found and what
sort of sensitivity analyses they've done to help us
get an understanding of the system.
And if there's new information in there
that wasn't in other TSPA documents that they've
provided us, we need to consider that and make sure
we're aware of it for a review of other documents.
I don't think DOE intends that to be any
formal part of their licensing case, but I can't
answer that for DOE. But there is certainly useful
information in there that we need to review and
understand what -- what they've collected and what
they're doing.
MEMBER GARRICK: Well, I had kind of a
different impression. I had an impression that it was
very much -- very well could have been just an
amendment to the TSPA because they sharpened up a lot
of the assumptions in the TSPA model, particularly
having to do with things like solubility, and some of
the -- what they identified as very conservative
assumptions in the TSPA SR were -- took a much more
realistic scope in the supplemental report.
MR. WELDY: And I think it also gives us
an opportunity to prepare for where DOE might be
going. So we can start looking at the issues --
MEMBER GARRICK: Right.
MR. WELDY: -- that they are trying to
sharpen the pencil on and make sure that we understand
any additional concerns we might have related to those
issues.
James Firth, what --
MR. FIRTH: James Firth, NRC staff.
Anything that DOE is going to be carrying forward from
the SSPA is going to be working its way into the AMRs,
the TSPA, so we're going to be focusing on their
changes to all of those documents.
And that's where we're going to be
focusing on our review, so we'll consider the
information there to help inform the questions that we
ask, but we're really focusing on the safety case that
they may be developing if they go forward with the
license application.
MEMBER GARRICK: Thank you. That's all.
CHAIRMAN HORNBERGER: Okay. Thanks very
much, James.
MR. WELDY: Okay. Thank you.
CHAIRMAN HORNBERGER: Okay. We are going
to move on to it looks like one of our favorite topics
here -- igneous activity.
MR. HILL: I always wonder what you mean
when you say "favorite."
(Laughter.)
Well, since we had our last interactions
in July and August, there have been a few new things
that we can talk about on igneous activity. We can
have a new update on issue resolution and also put the
risk information into a little more straightforward
path on what are the most risk important topics and
how are we going to move forward with the Department
of Energy on these topics?
We'll also use that information to show
how we're focusing the work last year and this year on
the most risk significant aspects for issue
resolution.
Now, probability -- the first main
subissue for igneous activity -- has been closed
pending since August 2000. We have one agreement
that's fairly minor, one that's probably minor but has
the potential to become a larger level of effort --
I'll talk about that in just a moment. But,
basically, probability is well in hand.
On consequences, in September we had a
very productive technical exchange with the Department
of Energy. We reached four major agreements that are
going to give us a path forward for closing the
igneous activity consequences subissue.
The remaining agreements that we had
reached, the previous technical exchanges, and
including the September technical exchange, there's a
number of minor agreements that appear to be very well
on track to completion in this fiscal year and the
following fiscal year. So I think we're in very good
shape for them.
The four major agreements, however, it's
really been too soon to tell. That only occurred in
September. We've got ongoing discussions with the
Department and plans for updates regularly throughout
the year to keep track of progress.
Hello? Okay. Just to bring you up on
probability, the risk insights for probability is that
there are a number of published alternative models
that give probability values anywhere from a factor of
10 to a factor of 100 higher than the DOE mean value
of 1.6 times 10-8 per year.
Now, the way that we've resolved to go
forward with the Department is that DOE has agreed to
conduct a sensitivity analysis in any licensing
application that should arise using a probability of
10-7 per year. And that will give us a way to
evaluate whether these differences between the
Department's preferred value and the value that we
would prefer, whether those differences are truly
significant to risk or not.
DOE has also agreed to evaluate some new
aeromagnetic data that's been collected by the U.S.
geological survey. This aeromagnetic data will give
us insights as to whether there are additional buried
features in the Yucca Mountain region that could
represent basaltic volcanoes.
We received the technical update impact
report right around Thanksgiving that has a
preliminary interpretation of these aeromagnetic data
that say there could be 13 additional igneous features
in 30 kilometers or so of the proposed repository
site. That is a preliminary analysis that's subject
to ongoing interpretation by the Department and their
contractors.
We're going to be meeting with the
Department some time this spring and will be holding
a teleconference in February to go over some of this
information. But if that interpretation stands or if
there are additional features that haven't been
incorporated into existing probability models, the DOE
has agreed to evaluate the impact of that new
information on the probability models.
We'll also need to determine if there are
additional basaltic igneous features in the Crater
Flat Basin that may be present but undetected by this
new aeromagnetic technique or any existing geophysical
technique that's been used to characterize the site.
Well, consequences -- to focus on the four
major technical agreements that we have with the
Department. The first of these is probably the most
important to understanding risk on -- it's about magma
repository interactions. We talked about this quite
extensively in the July meeting.
But, basically, how can repository
structures affect normal igneous processes? As magma
continues -- or if magma is rising, it interacts with
the subsurface system. How could alternative flow
paths develop as opposed to a normal volcanic conduit?
The risk significance of all this is these
alternative flow paths may increase the number of
waste packages that we have potentially disrupted
during an igneous event.
Now, we have received, again, a very
preliminary scoping calculation from the Department in
its technical update impact report that says now
they've looked at just potentially some of these
alternative flow paths. They don't agree that there's
a technical basis to support these flow paths. But if
these flow paths occurred, they would have a risk
impact of about one millirem per year with an event
probability of 1.6 times 10-8.
Now, it is important to note that using
that scoping calculation and using a probability that
would be sufficient for prelicensing issue resolution
-- 10-7 per year -- that would change that risk
significance up to about 10 millirems per year. So
you can see that alternative flow models, in addition
to alternative probabilities, have a truly
demonstrable risk impact on the total system
performance assessment.
The way forward here is that the
Department has agreed to evaluate how repository
structures can affect igneous processes, in particular
how magma may be affected during its accent and rise
in the subsurface, how conduits -- the pathway to the
surface, how those conduits localize in drifts or in
pillars or away from the repository itself, and how
that conduit and flow system may evolve for the
duration of an event.
MEMBER LEVENSON: Before you leave that,
how does the probability affect the dose?
MR. HILL: It affects the risk.
MEMBER LEVENSON: Yes. But you're quoting
a dose.
MR. HILL: No, I was quoting the risk.
MEMBER LEVENSON: Not the probability of
a dose, which is --
MR. HILL: No, that was a risk. The DOE's
risk was one millirem per year at a probability of
10-6 for the event.
MEMBER LEVENSON: Yes. Now, you change
the probability, but might be much more likely
somebody would get that dose, but how does it increase
the dose?
CHAIRMAN HORNBERGER: It's a weighted
dose. It's a probability weighted dose. You need
to --
MR. HILL: Yes. It's the expected annual
dose. Now, if you change the probability of the event
by a factor of 10, the risk would increase by a factor
of 10. The risk is the probability times its
consequence. So we have --
MEMBER LEVENSON: Yes. But millirem is --
MR. HILL: -- two components. Pardon?
MEMBER LEVENSON: But the millirem per
year is only half of that.
MR. HILL: I'm sorry. I -- half of it?
MEMBER LEVENSON: Well, okay. Never mind.
MR. HILL: No. The probability weighted
expected annual dose is one millirem per year. That
includes the dose consequences multiplied by a 1.6
times 10-8 event probability. If that probability
increased by an order of magnitude, the resulting
expected annual dose would also increase by an order
of magnitude.
MEMBER GARRICK: If it were not weighted,
it would be one big dose.
CHAIRMAN HORNBERGER: Yes. So if a
volcano occurs, the dose is much higher than one
millirem.
MEMBER GARRICK: Yes.
MR. HILL: That's been documented in the
issue resolution status report for a number of years.
Okay. The second area of major agreement
would be on magma waste package interactions. And,
again, how could potential igneous activity affect
waste packages? The reason this is risk significant
is that controls the source term for both volcanic
release and intrusive release or subsequent hydrologic
flow and transport.
There's really very little technical
information on how the alloys that would make up the
waste package and drip shield -- how all of these
would behave under igneous conditions. So we're
having to extrapolate from fairly limited information.
The Department recently agreed to evaluate
waste package response for a realistic range of
igneous conditions. That would include the
temperatures and pressures and dynamic pressure
variations that would be representative of typical
basaltic igneous events, and also consider the
potential effects of magmatic gas affecting corrosion
processes -- waste packages located some place away
from the direct impact of magma.
We also need to make sure we're looking at
a duration of an igneous event, because these events
can last for days to weeks to potentially a month of
duration. And so these thermal, mechanical, and
chemical loads can be impacting the system for many
years following the igneous event.
Also, after the event we have to worry
about seismicity, because one of the characteristics
of alloy C22 is embrittlement after exposure to 1,000
degree Centigrade temperatures and cooling back to
ambient conditions. So even if we had an intact waste
package following exposure to these sort of magmatic
conditions, any sort of shaking could potentially
cause a fracture to occur in the waste package. So we
have to look at the post-event impacts on performance
as well.
The third area is magma waste form
interaction. So if waste packages are disrupted, what
is the likely process for waste entrainment? This is
a very difficult problem because there is no natural
analogs to this. We don't have experience with
10 gram per cubic centimeter material being
incorporated into basaltic igneous events before.
But the risk significance is that if there
isn't effective incorporation of this material we
could be reducing the source term for volcanic
disruption or also affecting the waste form for
subsequent hydrologic flow and transport.
So we really have to understand, and the
DOE has agreed to evaluate, how a potential magma flow
can affect the cladding of the waste, changes in the
waste form, and also mechanically how that affected
waste and waste products can be incorporated into a
flowing igneous system. And that, of course, is going
to have to build on the knowledge from the magma
repository interaction studies as well, to get the
appropriate flow conditions.
And, finally, the last part is on tephra-
fall remobilization. So if we have a repository
penetrating volcanic event, how could the tephra from
that volcanic eruption erode and be remobilized on the
surface in the years after the event? The reason
that's risk significant is that the rate of erosion
and deposition strongly affects the long-term risk
calculation.
The risk really can be broken into two
main components. First is the risk of an event in the
year of an eruption. That would be weighted by the
probability of 10-7 of having an eruption in that
year. There is also an accumulation of risk from
prior events that also have, in each year, a 10-7
probability of occurrence.
And when you look at calculations for an
expected annual dose in any given year following
closure that expected annual dose for most years is
dominated by the accumulation of risk from prior
events and not by the event occurring in the year that
you're calculating.
So this process of remobilization by wind
and water, where potential tephra is mobilized off the
slope surrounding the volcano on the Yucca Mountain
where you have a fairly steep topographic gradient,
transport down the Fortymile Wash basin by both
flowing water and by wind and deposition in the area
that would affect a reasonably maximally exposed
individual.
These processes are key to understanding
the expected annual dose that's going to need to be
used for compliance with 10 CFR 63.
What do we need to do? What does the NRC
need to do before license application? Again, we're
focusing on evaluating the DOE models and reasonable
alternatives for these four most risk significant
issues. We have to confirm the numerical and analog
experimental investigations for flow dynamic on
sustained magma repository interactions.
Again, there is no data, there's no
analogs, there's no simple way of doing this except
through numerical modeling and some analog experiments
to really understand the flow dynamics of a repository
penetrating event.
We also have to evaluate the risk
significance of long-term remobilization by wind and
water. The reason that's difficult is because the
youngest eruption in the Yucca Mountain region is
80,000 years old. The deposits from that volcano, the
distal tephra deposits, the ones that are most
analogous to this situation, they've all been eroded
away and have been deposited down in other basins.
So we have no site-specific data to
evaluate how tephra erodes and is transported from the
Yucca Mountain system. And analogs have some very
important limitations in terms of their eroding and
transporting processes.
We can get some good insights, though,
from analogs and field investigations of magma ascent
processes. How do conduits localize? What are the
controlling stresses and rock mechanical properties
that can affect conduit localization once you get a
dike going? And I think that's one area that we're
going to be investigating in the next couple of years
to come up with a good basis for reviewing DOE work in
this area.
And, finally, we need to review -- review
new information on the responses of engineered
materials to igneous events, and this is an area we
see a lot of work coming in from the Department trying
to explain how igneous temperatures and conditions can
affect the engineered materials proposed for the
repository systems.
So our technical accomplishments last year
really were the bases to reach the closed pending
agreements. We talked about these quite a bit at the
July ACNW meeting. And, very quickly, it was the
numerical models for initial magma repository
interactions, the Bokhove and Woods paper if you will.
Some of the preliminary models for sustained magma
flow will redevelop the alternative flow paths. That
was the Woods et al. paper. And some of the initial
evaluations for long-term tephra remobilization, where
we talked about rates of remobilization being
important.
So in fiscal year two, the emphasis is
going to be on sustained magma flow, not the initial
interactions that we looked at last year -- but how
does a conduit develop and flow develop occur during
a potential igneous event? Really, it comes down to
two options. Do we have a normal vertical conduit
that can come up through a drift where there's a
potential for circulation and wall erosion through
time?
Could we have some entrainment of waste
away from the conduit? We're going to be trying to
answer that question this year. But also for the
second flow path where a breakout occurs at some
distance away from the point of interaction, how could
that occur?
What would be the potential for a
segregation of flow and waste entrainment during this
inclined part of flow in the system? And could high-
level waste be concentrated in a low velocity zone?
Or would the normal convection, back pressure, and
churning in the system essentially give you waste
incorporation?
We're going to be doing numerical and
analog experiments this year to try to answer some of
those questions on sustained flow in different
geometries.
Also, we'll be continuing the
remobilization work and parameter evaluation, complete
the Yucca Mountain petrology studies. Those are the
studies of the origin of the basalt. They give us
some very important insights for evaluating the
probability of the event.
Finalize updates to PVHView. This is the
computer code that puts in the probability models. It
will allow us an independent tool to evaluate impacts
of new volcanic centers and new repository footprints.
And also to continue to work with NRC staff for some
alternative high-level waste incorporation models
during flow.
The model we're currently using is just a
simple mathematical relationship. We're trying to get
a better mechanistic understanding of how high-level
waste could be incorporated into an erupting volcano.
So to wrap this up, moving igneous
activity from closed pending to closed is going to
require substantial effort by the Department of
Energy. The reason is there are very few data, there
are limited analogs, and it's a complex physical
process. We also have the need for model validation
and alternative conceptual model requirements, and
these are things that are going to need to be done
before licensing.
Our current risk calculations
unfortunately do not capture these uncertainties in
the intrusive and extrusive event source terms for
these alternative flow paths and also for waste
incorporation. Also, we're not looking at the range
of uncertainty for long-term remobilization.
These uncertainties directly understand --
directly affect our understanding of risk to health
and safety, which is currently on order of a millirem
per year in terms of expected annual dose, the
probability weighted dose. DOE is going to need to
resolve these uncertainties before we get to the
license application stage, if that stage arises.
There is a little supplemental information
in your package about the links to sufficiency, and a
few more details of the technical work for this fiscal
year.
I'm sure you've got some questions for me.
CHAIRMAN HORNBERGER: Thanks, Brit.
Perhaps we could start with Ray's question. Do you --
as a -- your personal view as a scientist, do you see
any data that is absolutely essential to have that is
unlikely to be forthcoming in a short time period?
MR. HILL: I think the Department can meet
the technical needs to support its models by licensing
in all of these four areas in the time remaining.
CHAIRMAN HORNBERGER: Thank you.
John?
MEMBER GARRICK: You mentioned earlier
that it was very important to consider the post-magma
events with respect to things like seismicity, and
what have you. Now, that's not a combined load
situation. That's a sequential load kind of
situation.
MR. HILL: Yes.
MEMBER GARRICK: Doesn't that really
introduce some rather serious modeling challenges?
MR. HILL: I think we can abstract the
process to a very simple level. Part of this is going
to depend on the engineering results, but a lot of the
need right now is looking at the embrittlement that
would occur in C22 following an event, where the
impact toughness drops to a very low level. I think
it's on order of 10 foot pounds is the impact
toughness from an original approximately 250 foot
pounds impact toughness.
So even the sort of ambient seismicity,
the things magnitude four or so that would be part of
base case scenarios, could have a potential impact on
the canister performance quite significantly. So I
think we could look at a generalized model for --
given the amount of -- extent of magma flow given a
likely failure distribution, what would be subsequent
to the event, the risk significance of that sort of
canister failure, without having to get into a complex
mechanistic model.
We'll just start up -- you know, we run
for various times of the initiating event, run
seismicity with the appropriate failure fraction, and
see what the risk significance of that is. So I think
we can evaluate that quite easily.
MEMBER GARRICK: Yes. Now you mentioned
here on the consequence end of the scale, the risk
scale, that there is considerable work yet to be done
with respect to the uncertainties, the quantification
of uncertainties.
Now, on the probability side, I assume
that -- and I seem to recall that the uncertainty
analysis is -- has been done. Otherwise, you wouldn't
really be able to identify those numbers as true
means, because means have to come from a distribution
-- a distribution that would manifest uncertainties,
I assume.
One strategy that is often employed is to
not necessarily try to quantify the uncertainty in the
consequences but quantify it in the context of the
probability in terms of specifying a consequence and
asking the question, what is the likelihood of that
consequence occurring?
And I'm just asking what kind of strategy
might you employ here. Will you be identifying a
specific consequence and asking the question, what is
the likelihood of that consequence occurring? Or are
you really going to try to do an uncertainty analysis
of a range of consequences, for example?
MR. HILL: I think part of this comes down
to how we're going to treat alternative conceptual
models. And right now the cleanest way of looking at
this is evaluating them independently and not trying
to assign probabilities to alternative conceptual
models.
MEMBER GARRICK: Yes.
MR. HILL: I think it's a big enough
technical challenge just to find the consequence
events -- excuse me, the consequence model that we're
going to have to abstract and evaluate. Trying to
assign a probability for a range of different
consequence models I think is, in the time available,
would be extraordinarily difficult to accomplish.
MEMBER GARRICK: Yes, and that's why I
asked the question, because you indicate on there some
commentary about identifying the uncertainties.
MR. HILL: The way I'm looking at it is
the probability of an igneous event, either volcanic
or intrusive, impacting the proposed repository site.
And I think that's as far as we're going to be able to
get to probability in order of having a technically
defensible way of assigning probability is the -- if
you will the initiating event. I --
MEMBER GARRICK: Yes.
MR. HILL: -- realize we don't quite say
it that way. It's the initiating probability is about
all we can get to.
MEMBER GARRICK: Okay. That helps. Thank
you.
CHAIRMAN HORNBERGER: Raymond?
VICE CHAIRMAN WYMER: Yes. It looks like
this whole issue of the effects of a volcanic eruption
right in the vicinity of the waste repository is going
to be kind of a focal point of a lot of discussion and
a lot of dispute for the -- in the time of the license
application.
And also, the -- you're still trying to
zero in on the probabilities. And from your viewgraph
-- the probability of such an event. And from your
viewgraphs it looks like in the one extreme that the
dose that somebody might get is flirting with the
limit of the dose all by itself, without any other
contributors that would be tolerable from the
repository, given all that, then.
Is there any serious study, assuming that
maybe things will come out adversely, of the use of
backfill as an ameliorating mechanism for reducing the
effects of the volcanic eruption?
MR. HILL: That's certainly within the
Department of Energy's realm, to consider any design
option that they wanted to implement for --
VICE CHAIRMAN WYMER: The question is --
MR. HILL: -- potentially adverse impacts.
VICE CHAIRMAN WYMER: -- is there any
serious study being made of it?
MR. HILL: By us or the Department?
VICE CHAIRMAN WYMER: Anybody.
MR. HILL: I think the Department
seriously considered the effects of backfill in the
first round of the analysis model reports that we
received.
VICE CHAIRMAN WYMER: By considering --
MR. HILL: When they had a backfilled
repository design, and that the Rev 1 updates were
looking at the no backfill option.
VICE CHAIRMAN WYMER: I know when we were
down at the Center recently there was a fairly good
presentation on the potential effects of backfill.
MR. HILL: Right.
VICE CHAIRMAN WYMER: That you've done
down there. And I wondered if that sort of preempts
or supersedes or agrees with what DOE had done.
MR. HILL: Well, I think we look at --
we're evaluating what is the currently proposed
design. If that design changes, we'll evaluate the
impacts on that. But as we talked about in August,
clearly having backfill in the drifts would
significantly affect the ability of igneous magma to
flow down the drifts.
How far and the reduction in risk impact,
we don't know quantitatively, but it's logical to
assume it would ameliorate some of the adverse
affects.
Okay. So if we got to the scenario that
you were introducing of having a risk that could be
approaching a very uncomfortable number, then there
are mitigation strategies that could ameliorate some
of the igneous impacts.
VICE CHAIRMAN WYMER: And you are poised
to deal with DOE's efforts to ameliorate these --
MR. HILL: I think we have the ability to
respond to whatever design changes the Department
proposes.
CHAIRMAN HORNBERGER: Before going to
Milt, I just -- just a clarification. I believe that
the result that you reported, DOE's calculation of
about a millirem per year on that consequence, is an
all pathways analysis. It's not only volcanic
activity. It counts everything. It's a probability-
weighted dose, right?
MR. HILL: I believe it was portrayed as
just a disruptive igneous event and didn't really
consider the hydrologic impacts.
CHAIRMAN HORNBERGER: So that was just a
one off analysis, then. They didn't use their TSPA.
MR. HILL: It was essentially looking at
an increase in the volcanic source term. Say that
your volcano had an alternative flow pathway. And,
again, I want to emphasize the Department has not
agreed that that pathway would occur.
CHAIRMAN HORNBERGER: No, no, right.
MR. HILL: It's in response to the models
that we have put out there in the Bokhove and Woods
and Woods et al models as a scoping calculation.
CHAIRMAN HORNBERGER: Oh, okay. So that
was a scoping calculation.
MR. HILL: And if your source term for
volcanism --
CHAIRMAN HORNBERGER: I didn't understand
that, then.
MR. HILL: -- incorporated that
alternative pathway, it would give a probability-
weighted dose with the Department's model on order of
a millirem per year.
CHAIRMAN HORNBERGER: Okay. Thanks.
Milt?
MEMBER LEVENSON: Yes. You mentioned that
there --
MEMBER GARRICK: Microphone.
MEMBER LEVENSON: Sorry. You know, both
new boys and old men forget these things.
(Laughter.)
You mentioned that there is no natural
analog, but we come close to having a manmade one
called Chernobyl where a couple hundred tons of molten
UO2 flowed across some floors. All of the models that
I'm aware of, all of the computer analysis, all had
predicted that that would just go right through there
a la The China Syndrome, etcetera.
In reality, of course, none of it did. It
flowed across the floors and poured down stairwells or
other places that were existing openings. Do you know
whether DOE is looking at that type information? And
were -- as a follow up to that, it has been going on
almost continuously in Sweden at the Royal Institute
of Technology, on the matter of interaction between
things like molten UO2 and various materials, is
anybody following that?
MR. HILL: I'm not aware of any Department
of Energy investigations in looking at the flow
analogies for UO2 at Chernobyl. And I'm not aware of
any investigations we're looking at either.
MEMBER LEVENSON: The basic thing is that
the common wisdom at the time was wrong.
MR. HILL: Yes.
MEMBER LEVENSON: Are we using that same
common wisdom? is my question.
MR. HILL: Depends whose perspective of
common wisdom.
MEMBER LEVENSON: It's an unfair question,
because we're talking about what --
MR. HILL: Well, it depends on whose
common wisdom we're talking about here.
MEMBER LEVENSON: -- DOE is doing --
right, yes. Okay.
CHAIRMAN HORNBERGER: Thanks.
Latif?
MR. HAMDAN: Brit, of the consequences
that you went through in your presentation -- and
there are quite a few -- can you tell us how many of
those are already incorporated in one way or another
in the TSPA?
MR. HILL: In terms of consequences?
MR. HAMDAN: Yes, the consequences that
you went through. Are they incorporated in the TSPA
right now, or are they not?
MR. HILL: Do you mean the four areas that
we have agreements --
MR. HAMDAN: Yes.
MR. HILL: -- in?
MR. HAMDAN: Yes.
MR. HILL: Aspects of those already
incorporated.
MR. HAMDAN: Okay.
MR. HILL: For example, we currently
assume that waste packages fail when they're put into
a conduit. The Department makes the same assumption.
We recognize there is no mechanistic basis behind
that, and nobody has done a full-blown engineering
analysis to look at the impacts. But it's a
reasonable interpretation that's in the performance
assessment.
If new information comes out that shows
there is resiliency, we can modify that appropriately.
The same thing for waste incorporation. There is not
an efficiency factor for incorporation, because nobody
has proposed a technical basis for that. But it would
be a very simple way to incorporate that new
information into any of the performance assessments.
So the critical aspects are there, and we
can evaluate them.
MR. HAMDAN: So that's not going to be a
big -- it will not take too much time to modify the
code, or what have you, when we --
MR. HILL: I think most of this can be
evaluated with simple parameter distributions and some
code modifications we're doing right now in TPA5 for
remobilization. But the problem isn't in making the
code do what you want. It's to come up with a
technical basis that justifies why you're taking
80 percent credit for a certain process.
MR. HAMDAN: Thank you. Thanks.
CHAIRMAN HORNBERGER: Other questions?
John?
MR. LARKINS: Just a quick question. I
notice that you're going to do an FY2002 continued
remobilization modeling and parameter evaluation. How
sensitive is the -- those calculations to the
composition of the material that's deposited either
aerosol or other material from a volcanic event?
MR. HILL: The volcanic material doesn't
have any impact on dose. It's only the waste
incorporated in it.
MR. LARKINS: Yes. But what I meant is
the form of that material.
MR. HILL: Okay.
MR. LARKINS: Or the composition.
MR. HILL: I believe we're just assuming
the form is the standard or high-level waste form.
There has been no change to that waste form as part of
the process.
MR. LARKINS: No, to the chemical
composition.
MR. HILL: There has been no oxidation
effects, no other chemical reactions during
entrainment and transport.
CHAIRMAN HORNBERGER: But you do make an
assumption about particle size.
MR. HILL: We do make the assumption on
particle size.
MR. LARKINS: Yes. Well, I know it's
sensitive to particle size. I just -- also, it seems
like there may be some changes in oxidation, which may
increase or change the -- how soluble the material is
in terms of reentrainment.
MR. HILL: Well, the real -- the two
leading isotopes of concern are americium and
plutonium, and under the range of fairly rapid
transport from igneous we don't think there is much
instantaneous effect, if you will, from the minute to
hour of transport, because the igneous system itself
is very reducing.
It's only when you get atmospheric mixing
at temperature do you have any potential oxidation
effects. And we think those are fairly small for
americium and plutonium isotopes of concern.
Of course, for the overall waste form,
following the event, in the years following the event,
at the surface you certainly think that we'd have to
get a better handle on the kinetics of oxidation and
other sorts of chemical processes.
CHAIRMAN HORNBERGER: Milt?
MEMBER LEVENSON: Yes. You just mentioned
that the principal isotope is plutonium, americium, so
that must mean that the external dose is not the
primary aspect here. It's an inhaled dose?
MR. HILL: Correct. It's an inhalation
dose is 90 percent of the total dose.
MEMBER LEVENSON: Okay. Does your
calculations for resuspension and the inhaled dose,
etcetera, include the mass of everything there? The
context of my question is some years ago an analysis
of -- near Rocky Flats of a motorcyclist driving
around contaminated soil and steering up a cloud
indicated that in order for him to have gotten maximum
allowable dose his lungs would have contained between
two and three pounds of dirt, and the dose was
insignificant in health effects.
MR. HILL: Right. We --
MEMBER LEVENSON: Is such an analysis
being done here? Are you including the mass of the
carrier when you calculate inhalation?
MR. HILL: The answer is directly no.
It's not explicitly put in there, but we have done
scoping calculations to show that the amount of mass
per year per lung for total particulate -- it's not
just respirable particulate, but the total suspended
particulate -- is small compared to what the lung and
tracheal/bronchiole system can absorb.
So we're not talking about putting two
pounds of dirt into somebody's lungs in an unrealistic
sort of inhalation model. The concentrations we're
using on average are one milligram per cubic meter of
total suspended particulate. And of that there is a
much lower concentration of high-level waste.
MEMBER LEVENSON: Well, you know, I come
back to my -- one of my pet hangups. If your code
doesn't contain conservation of mass, I don't know how
you -- if you're distributing this plutonium over
hundreds and thousands of tons of magma --
MR. HILL: Yes.
MEMBER LEVENSON: -- how you get
significant amounts in somebody's lungs without having
massive amounts of magma.
MR. HILL: Well, it comes down to the dose
conversion factors that are being used and the
efficiency. This is why we're backed down for a lot
of this being from the inhalation dose, because of the
dose conversion factors that are being used for
americium and plutonium for inhalation dose.
You do not need a large mass of high-level
waste to effect those levels of total effective dose
equivalent impact.
MEMBER LEVENSON: Well, you don't need a
big mass of spent fuel, but -- but it's -- well, I
guess it comes down to your assumptions about how much
dilution occurs.
MR. HILL: I'm not positive if you want to
talk about a mass or a particle number basis, but I
believe we're talking 10-2 grams per square centimeter
for ash loading and about 10-6 grams per square
centimeter of waste loading. So about a four order of
magnitude ratio between the mass of ash and the mass
of waste at 18 kilometers.
MEMBER LEVENSON: Okay.
MR. HILL: I could be corrected --
MEMBER LEVENSON: What was the ratio of
plutonium to magma in your masses?
MR. HILL: Plutonium or --
MEMBER LEVENSON: Of fuel.
MR. HILL: -- of all waste, of high-level
waste.
MEMBER LEVENSON: Yes.
MR. HILL: The waste form.
MEMBER LEVENSON: Yes.
MR. HILL: Is about four orders of
magnitude smaller, off the top of my head, and a
number that should be checked. I'm trying to go from
what I remember from the TPA outputs on aerial loading
at 18 kilometers.
MEMBER LEVENSON: Okay. So for the
plutonium it's maybe six orders of magnitude.
MR. HILL: I believe somebody who knows
the inventory better than I do would put that --
MEMBER LEVENSON: Yes.
CHAIRMAN HORNBERGER: Tim?
MR. McCARTIN: Yes. If I could just --
Tim McCartin, NRC staff. We do conserve mass with
this calculation, and there is an explicit amount of
ash and radionuclides at specific locations, and it is
accounted for. But I can assure you it is -- you
know, what Brit's referring to, there isn't an --
there aren't some explicit numbers you'll see in the
output of the code, but in -- within the code it is
determining how much mass is getting there.
We keep track of it. There is a dilution
mechanism of the radionuclides getting mixed into the
soil to a certain depth, how much of that is ash, how
much is radionuclides, and a certain percentage of the
overall dust in the air is comprised by looking at how
much of the stuff there, let's say in the top 15
centimeters, which I'll just throw out -- I don't know
the exact number -- is radionuclides.
But we do -- it is -- there is an
accounting for that, but it --
MEMBER LEVENSON: In your analysis.
MR. McCARTIN: Yes.
MEMBER LEVENSON: Okay. So --
CHAIRMAN HORNBERGER: I think we should
move on. I mean, I think that we can explore this
further. As Brit says, we could check the numbers,
but the question has been posed. Any other --
remaining issues? Okay. If not, let's move on.
Thanks, Brit.
Our next presentation is on structural
deformation and seismicity.
John, you can introduce yourself, so I
don't mispronounce your last name.
(Whereupon, there was a long pause while
the next set of slides were being set
up.)
CHAIRMAN HORNBERGER: So what happened?
Did we have an earthquake that wiped out this
presentation on the computer?
(Laughter.)
MR. STAMATAKOS: It wants to start halfway
through, wants me to be done quicker.
CHAIRMAN HORNBERGER: Well, we don't mind
that.
(Laughter.)
MR. STAMATAKOS: Won't let me get to the
introduction.
I'll go to Plan B.
CHAIRMAN HORNBERGER: Plan B.
(Laughter.)
MR. STAMATAKOS: Okay. Okay. How's that?
My name is John Stamatakos, pronounced
originally Stahmatacos (phonetic), so you had many
options.
(Laughter.)
You had many, many choices that you could
have -- I'm going to talk about structural deformation
and seismicity. And I'm going to apologize from the
outset in that in my haste, even though I was told to
do it, I forgot to include an outline slide, although
my outline follows very closely with what everybody
else has done.
The other point I would just make in terms
of introduction of this particular subissue is
recognition that of all of the KTIs I think we're the
-- one of the ones that could be considered most
upstream. A lot of the things we do really feed into
a lot of the other issues in terms of the description
of the site and site characterization. So at times
it's actually most difficult for us to sometimes see
how what we do ultimately comes out in a risk
calculation.
Also, I think you'll see that a lot of the
work that we're doing now is really well integrated
with many of the other KTIs, not just in issues that
were originally identified under structural
deformation and seismicity.
And, finally, I think a lot of what we do
contributes directly to pre-closure, so there are some
instances where I'm going to talk a little bit about
how the work that we do relates to pre-closure.
In the SDS KTI we had four subissues that
were originally identified -- tectonics, faulting,
seismicity, and fractures. I've lumped the faulting
and seismicity together because they contain many very
similar agreement items. The first of our issues is
closed, and tectonics -- it has been closed for -- for
some time now.
In faulting and seismicity, we have a
small number of agreement items. Most of them I would
classify as minor. Some of them are moderate and have
the potential to be more significant, depending on the
outcome of the agreements.
Of all of the remaining subissues that we
have, the fracturing and structural framework one is
the one that is directing most of our attention now,
and that one is one where we think most of the
remaining work needs to be done by the Department of
Energy.
So I'm going to go through them in stages.
I'll first talk about seismicity and faulting. I've
prepared my organization in a way of first trying to
identify what the risk is, looking at a DOE approach,
and then where the status is. I'll follow these with
slides that give some specific information about the
things that we need to do and the things that we think
DOE needs to do to reach closed status on the
subissue.
The risk for seismicity and faulting
mainly has to do with the degradation of the
engineered system, damage to waste packages, damage to
the drifts themselves. Up until probably last fall,
the DOE approach has been to screen these disruptive
events by looking at or promising to develop a robust
enough design so that any impacts would be able to be
handled by the design of the waste packages and the
repository.
So we didn't have a direct link in TSPA in
that way, because everything was evaluated as
something that was essentially screened out of the
TSPA. In answering our questions from the technical
exchange and in our agreements, DOE now proposed to us
in a letter that was sent in October to do an
alternative approach to looking at specifically
seismicity, and that is that they now propose to do a
full sort of seismic risk assessment in order to
quantify inputs that they're going to incorporate in
TSPA-LA.
We've only seen the plans for that. We
haven't seen exactly how they're going to implement
that, but that's a pretty significant change in the
approach, I think pretty significant change in the
approach that they have in looking at seismicity issue
for -- for post-closure.
So in terms of status of the -- we had six
agreement items that were outstanding with DOE, plus
some additional items that are integrated with the
other KTIs -- pre-closure, RDTME, and some of the flow
issues. It looks to me like this new DOE approach, if
it's what it says it is, may satisfy many of the
agreements in seismicity and faulting.
What DOE needs to do prior to LA is most
importantly we need now to see the details of how they
propose to do their seismic risk approach. They say
it's a sort of standard type of seismic risk approach,
just propagated out to the large time period. So I
think that will be the biggest challenge is, how do
you take -- how do you do that kind of an assessment
for a 10,000 year repository?
One of the other issues that I think has
the potential for some effort on DOE's part is that we
have some questions about their -- when they did their
seismic hazard calculations, they did it as an expert
elicitation. During that elicitation, we've raised
some questions about the procedures that were used,
particularly one component of the expert elicitation.
And we're asking for some documentation,
and we have some ongoing discussions with DOE on
whether or not that documentation exists or what form
of documentation would be required to sort of satisfy
the completeness of their expert elicitation process.
CHAIRMAN HORNBERGER: John, could you just
say a couple words about why it's a challenge to do
this for 10,000 years? It seems to me it should be
straightforward?
MR. STAMATAKOS: Well, I think the trick
part has to be in coming up with fragility estimates
for components out in the -- not the seismic part.
The seismic part -- you've got to involve it with the
fragility to come up with the risk.
And then a large part of what's left in
the seismic has to do with site-specific information,
and this really relates to pre-closure and surface
design questions more than post-closure. But DOE has
collected information, but we don't yet have all of
the information for developing a site response model
and all of the things that come with that to develop
design ground motions that will be inputs for the --
MR. LARKINS: Quick question on the expert
elicitation.
MR. STAMATAKOS: Yes.
MR. LARKINS: Is there a standard or
something that you're using for --
MR. STAMATAKOS: Yes, there's an NRC NUREG
that they agreed to follow, plus their own
documentation that they agreed to follow in -- in
conducting and documenting --
MR. LARKINS: And that's where the
difference is.
MR. STAMATAKOS: That's part of where the
differences are. There's kind of an interesting twist
to that, because where we see the differences actually
leads to potentially some of the ground motion experts
having inputs into the process that led to what we
might characterize as very large ground motion.
So when you look at the -- this is the
seismic hazard curve for Yucca Mountain, and the
ground motions at the low probabilities is 10-6 here,
10-7. I don't even have the -- extrapolate 10-8. So
ground motions are quite large and the uncertainties
are quite large.
And so there's sort of an interesting
twist in the sense that, you know, what's come out of
the ground motion part are some of these very large
uncertainties that lead to very large ground motions.
And I think it speaks to the questions about, you
know, where they may be potential very large
overconservatisms in the analysis.
What we need to do prior to LA, most
importantly I think is we need to -- our continued
interactions with DOE. We're very anxious now to see
what their new approach is going to entail. We would
like to meet with DOE, and DOE just yesterday or day
before yesterday sent us some documentation on
proposals to resolve our expert elicitation questions.
We're working on developing some
additional technical bases to help evaluate the DOE
ground motion results, and we'd like to continue to
work on that.
The last two items really pertain to
issues more in line with pre-closure. DOE is going to
provide information in what's going to be called
Topical Report 3 with a companion seismic design input
report, and those reports will summarize the
information that we need for the site-specific
information that will get us from the generic hazard
curve to the design basis inputs that will be used for
the pre-closure design issues.
The other subissue that we're looking at
has to do with fractures, and this is a much broader
issue and is very diversely integrated with lots of
the other KTIs.
Basically, we recognize that fractures and
faults and other parts of the geologic setting control
many aspects of groundwater flow, and also those
features are very important in developing accurate
pictures of the rock mass properties that are used for
modeling in rock fall calculations or drift stability,
things like that.
The DOE approach is sort of two-fold. For
their flow calculations, DOE mainly relies on bounding
assumptions or modeling results to constrain the
fracture properties. And so our work has mainly been
trying to see how -- sort of validate those modeling
procedures against what the fracture information and
characterization tells us on the site.
For the rock fall analyses, DOE is using
direct measurements that they got from the ESF and the
ECRB, but we have some questions about the
applicability of that data directly, and so we've
raised some questions related to how that data is
used.
But it's been the process of sort of an
active, ongoing interaction between DOE and ourselves.
We have about four agreements with DOE, plus, as I
said, some other additional agreements that are
integrated with the other KTIs. DOE is considering
some of our responses, and we're currently considering
some of their responses. So we are involved in an
active discussion of this particular issue.
In particular, some specific things that
DOE needs to provide us is the -- a response to a
review that we had of one of their major documents on
fractures. This is sort of the focal point of a lot
of the discussion about the fracture database and the
fracture characterization.
And in addition to that, there are some
things that DOE has to do that we're working on but
are really related to items that -- from the
unsaturated flow KTI and from the NT KTI. These have
to do with characterization of the stratigraphy and
sedimentology in the alluvium and valley fill portion
of Yucca Mountain as that relates to flow models.
And we've been working on the issue
related to the potential for hydrothermal activity,
the whole fluid inclusion debate. We've done some
independent studies related to that as well.
What we plan to do before LA is to
continue to establish our technical basis for
identifying which parts of the fracture database are
really critical to performance and which ones aren't,
and that's a highly integrated effort with a lot of
the other KTIs.
We're working on developing some
independent stratigraphy and stratigraphic studies
that we can use to help the flow people in the
alluvial part of the aquifer. We've done a lot of
work on the fluid inclusion work, and we're about
ready to submit a paper of our own independent
evaluation of some of the secondary mineralization as
it relates to this hydrothermal question in the ESF
and ECRB.
CHAIRMAN HORNBERGER: Can you tell me the
bottom line of your analysis of the fluid inclusions?
MR. STAMATAKOS: I think the bottom line
of fluid inclusions is that there are still a lot of
contradictory interpretations of those results. I
think ultimately what we would argue, along with I
think probably what the -- parts of the GS would say,
is that we think that a lot of the evidence for that
is that they are quite old and not modern.
We don't necessarily agree with some of
the interpretations of the sort of long-term heating
that led to some of the younger ages in the fluid
inclusion studies, but it's still a very I think
technically challenging issue.
One of the things that I would point out
is in terms of potential developments, it has to do
with modifications that might come about in the
design, or in this case if DOE proposes an alternative
operational mode in which there are other parts of the
repository that are their proposed use rather than the
original primary block.
We've done some work to evaluate what
significance that might have on characterization, and
what we would say is that there is the potential that
-- for a significant amount of additional work, or at
least some work on DOE's part to how they would
provide a technical basis to justify the applicability
of existing characterizations of the details in this
particular part of the repository to other elements of
the repository.
So in 2001, a summary of some of our work
and accomplishments. Obviously, we've done a lot of
work trying to continue our discussions with DOE on
our -- all of our agreement items. We've worked on
developing some technical assessments for that
potential alternative load -- thermal load option in
the repository design.
We've completed some work on issues
related to some of the other KTIs. In particular,
we've done a lot of work on looking at the issues
related to hydrology in the alluvial part and these
fluid inclusion questions.
In FY2002, we're going to continue much of
that work, although we are now focusing more and more
on looking at how the site characterization relates to
pre-closure safety as well. We're going to continue
to work with the other KTIs on issues related to their
agreements.
We certainly want to try to continue to
understand how aspects of faults and fractures are
most important to performance, and we're going to
continue to work on issues related to the seismic
hazard, both as they relate to pre-closure and post-
closure.
So, in summary, I would say that for SDS
we've reached a significant amount of progress. Our
tectonics subissue has been closed. It is closed for
some time. We see substantial progress on many of our
agreements related to faulting and seismicity. Most
of our remaining work is going to be related to the
fracture characterization and how those fractures and
faults -- what role they play in both groundwater flow
and in repository design and drift stability
calculations, how that integrates with the other KTIs.
So, in summary, I'd say that our work is
highly integrated with much of the agreements reached
in the other -- on the other KTIs.
And with that, I hopefully got us back on
schedule, and I'll take your questions.
CHAIRMAN HORNBERGER: Thank you, John.
Milt? No? Ray?
VICE CHAIRMAN WYMER: Yes. You've already
touched pretty heavily on something I'm going to ask,
but I'm going to ask it in -- the question in slightly
a different way than you have dealt with it in your
viewgraphs. And it relates to how sensitive are the
results that you obtain with respect to detailed
knowledge of the site characteristics?
MR. STAMATAKOS: Which results?
VICE CHAIRMAN WYMER: The calculations of
the --
MR. STAMATAKOS: Of ground motion?
VICE CHAIRMAN WYMER: Yes, and of the
impact of this on the ultimate dose.
MR. STAMATAKOS: Well, that's a hard
question to answer because the -- the approach that's
been taken over the last several years has been one in
which the seismic hazard has been sort of an input
into design decisions to say we will maintain waste
packages and drip shields, so that they can withstand
any rock fall size that the repository can deliver
during any size earthquake. So in a sense they've
just become sort of a benchmark.
VICE CHAIRMAN WYMER: So what you really
said is all of this work is meaningless.
MR. STAMATAKOS: Well, I don't think that
it's meaningless. I think the new approach is going
to be -- is going to be able to directly answer your
question in the sense that now what DOE proposes to do
is to actually incorporate the seismic results
directly in a TSPA-type calculation.
VICE CHAIRMAN WYMER: You're going to
challenge the statement that everything is designed so
that it won't make any difference?
MR. STAMATAKOS: Well, their response to
some of our questions is now -- is that instead of the
approach that it won't make difference, they're going
to incorporate it into a sort of risk-type calculation
by doing this seismic risk approach.
VICE CHAIRMAN WYMER: Is the site well
enough characterized to do everything you need to do?
MR. STAMATAKOS: In terms of seismicity?
VICE CHAIRMAN WYMER: Yes.
MR. STAMATAKOS: I think so. We haven't
seen the final specifics. You know, we have a generic
-- that hazard curve I showed you is a generic hazard
curve for some hypothetical site, and what we have to
now do is propagate that seismic energy through the
soil and rock column to the surface and do those kinds
of calculations.
But I think in terms of the overall
seismic hazard results, yes, I think that the DOE is
well characterized, minus that one question about --
VICE CHAIRMAN WYMER: Extending the site?
MR. STAMATAKOS: -- the ground motion and
extending the site, yes.
And in answer to your generic question, I
think that in terms of our KTI we're well on the way
to being able to have almost or nearly everything that
we need well before license. I don't see any reason
why all of the information won't be available for --
before licensing.
CHAIRMAN HORNBERGER: John?
MEMBER GARRICK: Well, that was going to
be my question. Most of the other speakers -- I guess
all of them have expressed considerable optimism that
the path forward is pretty optimistic relative to
resolution of the agreements. So I assume you feel
the same way?
MR. STAMATAKOS: Yes. Yes.
CHAIRMAN HORNBERGER: We haven't heard
from --
MEMBER GARRICK: Well, that's right.
Okay. No, that's all.
CHAIRMAN HORNBERGER: John, I have a quick
question. I know that -- let's see, the DOE saturated
zone model uses, if I recall correctly, a fracture
spacing on the order of 20 meters or something on
average. Do you think that's about right?
MR. STAMATAKOS: It may be right in that
ultimate abstraction, but the -- what we've found in
our fracture studies -- and my fracture guy is not
here, so I'm going to reach a little bit -- but is
that -- that the -- it's much less uniform and
homogeneous than that.
So that fracture spacing may be correct
only on a small scale, and there are clusters of
fractures that -- of large fractures that occur very
close together, and there are areas that are not
nearly as heavily fractured. So it's much more
heterogeneous than that, but we don't -- I don't -- I
can't answer whether or not that heterogeneity
ultimately plays a significant role.
On the large scale of the saturated zone,
we know that the large faults certainly play a role in
the flow path, and that the flow is not just
controlled by the surface of the water table, but that
there is an anti-satrapy that's induced and -- by the
faults or fractures, and DOE has now incorporated that
into their saturated zone flow water.
CHAIRMAN HORNBERGER: Phil?
MR. JUSTICE: Phil Justice, NRC staff.
DOE doesn't use a single fracture spacing in all of
its flow or transport models where fractures are
needed. And 20 meters, I haven't seen that one in a
while.
In the ECRB testing of facilities, for
example, such as alcove 8, niche 3 test, the spacing
in the middle non-lithologic -- lithoficile unit
that's being used is on the order of roughly 1.3
meters.
The question to us is not so much the
number, but the representativeness of the source of
data that allowed that generalization, and whether
that generalization for the particular test or portion
of the repository being evaluated is representative of
the entire repository system of interest.
CHAIRMAN HORNBERGER: Thanks.
Thanks very much, John.
MR. STAMATAKOS: Sure.
CHAIRMAN HORNBERGER: We have a
substitution, according to -- or a modification in the
original agenda, and I think Gustavo is going to go
next and talk about container life and source term.
MR. CRAGNOLINO: Okay. Good morning. My
name is Gustavo Cragnolino, and this is all of the
people that have made this presentation possible. And
I want to emphasize the name of our project element
manager at NRC, Tae Ahn.
My outline is, as you have seen before,
similar to the previous presentation, and I want to
emphasize this. And I am going to go directly to the
status of the CLST subissue resolution. The several
subissues, six in our case, are listed here. The
first one is effects of corrosion processes on the
lifetime of containers, and, as the other one, is
closed pending.
You can cover additional detail in the
last of my backup slides, what was presented before by
Brit, in terms of the scope of the CLST, the type --
the nature of the agreement with the DOE and what is
expected. And I'm not going to go into detail of
that.
Let me go to the main point that you are
interested to know about our view regarding what DOE
needs to provide before LA. And this is not only
analyzed in terms of specific subissues, but the text
in blue indicates precisely the main scope of the
subissues.
And this is relating particularly to the
case of subissue 2 and is what we consider the -- it's
necessary for part of the DOE event evaluation of
premature waste package failure, because this has an
important impact essentially in the problem of
criticality, in-package criticality, and we need a
reassessment of the probability of early failure
arising from an effectual defect, and, in particular,
closure welding and post-welding operations.
However, the subissue where we have a
special concern is what is reflected in subissue
number 1, and what we need is a better technical basis
to resolve uncertainty in container life prediction.
And the point -- and I'm not reflecting exactly the
agreement. There are 17 agreements here, but I'm
focusing on what are the important ones because seven
of the agreements are subsidiaries to this.
And one is the definition of the aqueous
environment in contact with the waste package, and
this is an issue that Bobby Pabalan is going to deal
with.
What we have is -- providing alternative
measure for passive corrosion rates. Data and
modeling support to establish that you have really
long-term stability of the passive film, because this
is the main issue in which you would like to have
long-term container life.
Here is an issue that in some way is an
answer to the question which arose before regarding
something that has scientific relevance and is in the
study of this. We think that the repassivation
potential is the basic criteria for the decision if
you have a container that is susceptible or not for
localized corrosion.
By that other point of view, in the
scientific community, in the corrosion community, and
this discussion -- and DOE favors this type of
context. This is an issue that has to be resolved.
And they are dealing -- but we need a more technically
accepted criteria for the selection of their
initiation potential for localized corrosion.
And this is very important, because we can
talk in theory about the behavior of the material, but
when you start to fabricate a container you have
potential problems related to welding, post-welding,
and this is important on corrosion and stress
corrosion cracking.
And to illustrate the point, this is
experimental data showing that this repassivation
potential has some sort of threshold potential for the
initiation of localized corrosion in our criteria,
because it's a lower bond, decreased significantly is
the material, alloy 22, and this solution is heat-
treated at 870 degrees Celsius for only five minutes.
And this is something that would happen
during the processing, with the slow cooling from the
annealing temperature, the fabrication of post-welding
treatment. Therefore, if you have corrosion potential
values in this range you have the potential for the
localized corrosion of the container material.
Defective welding appeared to be less
pronounced, but we need additional data in which the
defective welding plus this type of treatment is
evaluated. And this is a matter of concern, and this
is something that we expect that we've solved.
For the other subissues, -- we need a
better estimation of the rate of radinuclides released
from the waste, but this is essentially related and
focused on what is the definition of the in-package
aqueous environment. And what is the impact of this
environment of corrosion or stress corrosion cracking
because, as you know, DOE put a lot of emphasis and
gave a lot of credit to the container.
Finally, this is an issue that concerns
us, too, because it has to do with the role of the
titanium drip shield, and we believe that DOE should
provide more data on stress corrosion cracking
testing, technical basis for the choice of the
hydrogen pickup fraction, and also justification for
this value that they have adopted.
They can move from 400 ppl to 1,000. I
think that they are moving in the right direction, but
not with this quote, because this could have an impact
in the evaluation as has been shown very clear in this
figure.
A slight increase in the fluoride
concentration to 10-3, increased the anodic current
density, and, therefore, the corrosion rate by three
orders of magnitude, and this rate, obviously,
significantly reduces the lifetime of the drip shield
and was, even if you have sulfate and nitrate or
nitrate plus sulfate, you cannot inhibit the
detrimental effect of fluoride.
And this is a point of concern that DOE
will need to progress.
What we should do to go and move ahead in
this path flow, and try to find resolution of these
issues that I mentioned. What I am talking about is
that we should very closely monitor the DOE progress,
because it is all viewed as a risk progress, and this
has to be very clearly emphasized.
They say that movement in the right
direction from the part of the DOE, and we need to
review and respond to the revised AMRs and the dated
information.
And although this is a mechanism that we
find more appropriate, this is related to what I
mentioned before regarding the importance of
publication process, and virtually any change in the
design, and we have to be prepared to assess.
And this is a range of uncertainty because
even though everything looks fine, we can have
surprises. Obviously, we need to do in order to be
able to tackle what is shown before now with an
independent evaluation of issues and we will focus
experimental program and the model program.
And this is the approach, and I don't need
to insist, but we have two other areas; uncertainly
that still exists in the environment, and the site and
component fabrication, material procurement, closure
welding and post-welding treatments. And I forget to
say that you can interrupt me at any time to ask
questions.
When we come down and what we have done in
this previous year, and we prepared Revision 3 of the
CLST IRSR, and that was the basis of our input to the
Revision 0 of the Integrated IRSR, and essentially in
the area of engineering environment, and where we
provide input also to other of the different type of
model areas of the differing -- presented by breaking
-- regarding engineering, too, that is degradation by
mechanical disruption, or the ones related to the
direct contact of the radioactive ash as a result of
organic eruption.
But participating in several technical
change, and this is what we have been doing all this
time through different types of reports and
publication and addressing pending issues on corrosion
of Alloy 22, and Titanium Grade 7, with the purpose of
evaluating the accuracy of the DOE data, and the
methodology for predicting early failures.
And we have been trying to follow very
closely any advances that have been done in the
evaluation of the source model, and this is a
secondary activity, but also was important for us to
support the evaluation of in-package criticality for
Navy spent fuel.
Finally, this is something that we have
paid attention to, and something that we have been
pushing forward because we consider the assessment of
the approach and the methodology for waste package,
and engineering barriers performance confirmation is
something that has to be kept in mind for the
evolution of this program.
We initiated a effort culminating, and
very soon we are going to have a report for the
evolution of natural industrial analog for Alloy 22,
and obviously we would like to support as much as
possible you, and the Nuclear Waste Chemical Review
Board, and participating in several meetings.
This will be our work plan for 2002. It
is important to conduct our independent evaluation of
the DOE status and follow all the information that we
will receive, and we will provide input for the model
abstractions of TPA Version 5.0 in our two related
modules for EBSFAIL and EBSREL release.
And as I mentioned before, we have
completed practically this, and continued activities
related to performance confirmation. Now, this is
what I wanted to call your special attention to.
We believe that the activities of the DOE
in order to obtain resolution of the issue has a broad
scope, and that in some ways is good, but there is not
clear prioritization.
And this is not our own words. This is
one of the reasons why the waste package proponent was
called by the DOE, and to help them to define the
product decision of the work.
And to follow in some cases a very
aggressive schedule, and may not have sufficient
resources. It is not my problem, the resources, and
I can't throw away the last one, but I give you
something to think about.
An evaluation of the radionuclide release
continued to be affected by significant uncertainties
in the definition of the aqueous environments, but
there is important progress there.
Finally, after many years the focus is
very clear and we expect to have more research. These
affect corrosion of the waste package and drift
shield.
By the way, in the black and white copies
that I have drift instead of drip, and I clarify that,
and knowing the design, I don't want to cover a drift
with a shield, and maybe wood, but I would not want to
do it.
And this is a problem for radionuclide
release because the cladding corrosion is still an
issue. And DOE knows this very well. This is an
important point.
If you look in the SSPA, there are
significant modifications in the model abstraction,
but they base it in very insufficient data, and this
is where we have to be concerned.
If there is a modification that has an
implication for a substantive reduction in the dose
because it is less concerned about drip, but has no
basis. And I don't want to dwell on the example, but
I can give it if you want.
And this is what we need in terms of
performance confirmation. We think that this is an
integral part of the program, and can be considered of
something to be done in the future, because
performance confirmation planning is integral to
design. And this is experience in any important and
significant industry that has had failures.
In summary, performance estimates are
affected by uncertainty in container life predictions,
including effects of fabrication and welding. This is
very important.
And a good evaluation of premature waste
package failure, and always as we mentioned before, we
have some uncertainty in the package embridlement that
has to be solved, and the drip shield case.
And I think from our part that active
monitoring of DOE progress, and I emphasize again
there are clear progress in the resolution of
agreements is needed to close all the pending
subissues at the time of the proposed license
application. And this finishes my presentation, and
I am open to any questions.
CHAIRMAN HORNBERGER: Thank you very much,
Gustavo. Raymond, do you want to start?
VICE CHAIRMAN WYMER: What I am going to
ask you, you have included by implication, Gustavo,
but you have not addressed it exclusively. A big deal
in recent years has been the effect of trace
impurities, such as lead, mercury, and things like
this.
What have you come up with recently, or
what are you planning to do, or where do you stand on
it?
MR. CRAGNOLINO: Well, we are exploring
these. We had difficulties in the study, for
instance, and one impurity that has been claimed that
is very important, and that is lead, to obtain
reproducible conditions to do these experiments.
And the preliminary work that we have done
up until now doesn't indicate a potential risk unless
you have extraordinarily high concentration of leads.
It is preliminary, but this is what we
think. I think we have to look more in the range of
comparison of environment that DOE is exploring now,
and going beyond the famous or infamous high
concentrations type of environment, and looking to the
variation that tends to go to the alkaline side or a
little bit to the acidic side.
But we are looking at the effect of trace
impurities, but my opinion, and the opinion that we
have, is that we are going to continue looking for
that mostly important period to be before. And this
is our preliminary type of conclusion.
VICE CHAIRMAN WYMER: Okay. I have one
additional question. On your next to the last view-
graph, you say prior experience with sensors and
monitoring devices in further or in other applications
reveals an extensive development and prolonged periods
of testing and verification, perhaps for many years,
are necessary.
Just how troublesome will that be with
respect to approving a license application?
MR. CRAGNOLINO: Well, it is not
troubling, but for the time of the license as a part
of Part 63, it is very clear that there has to be a
plan, a performance confirmation plan, and this plan
cannot be a series of generalities.
It should include provisions for these
types of issues, because this is an indication that
any type of potential problem that arises, there is
the advice or the sense of an approach to take over
and resolve the problem, or at least the possibility
to attack the problem.
And for this reason, it is not necessary
for us to be completely solved at the time of the
license, but you know there has to be a plan, and
there has to be a clear indication -- as a part of
this program, and certain experience in this type of
sense.
VICE CHAIRMAN WYMER: Well, if those
periods are necessary, and then the carrying out of
those extensive tests, to discover that the
performance that was assumed isn't there, that is bad
news isn't it?
MR. CRAGNOLINO: I guess so.
CHAIRMAN HORNBERGER: This is Tae Ahn of
NRC staff. I believe we have a dual purpose for
pursuing these types of testings. One is as Gustavo
stated, to be prepared to evaluate DOE's confirmation
plan in the licensing application.
They will attach the performance
confirmation plan. That is one purpose. The other
purpose is these tests will validate to a certain
extent the elaborate testing that we are conducting
right now. Most of these were being done on the
capacity condition with rock block and so forth.
And so we probably have two different
purposes for doing this type of testing.
VICE CHAIRMAN WYMER: So the implication
is that by the time of the license application, you
would have progressed far enough down the road in
verification that you have reasonable expectations
that there are not going to be any clinkers. Okay.
Thanks.
CHAIRMAN HORNBERGER: I am not sure that
I followed your question or your comment. I mean, a
performance confirmation plan is a plan isn't it?
MR. CRAGNOLINO: It is a plan,yes.
CHAIRMAN HORNBERGER: It does not require
that you have the results.
MR. AHN: That's what I added, and that we
have another purpose; to validate the lab testing
result in the real posture environment.
MR. CRAGNOLINO: This -- let me give you
a simple example for your satisfaction. It is very
simple. That product has nothing to do, and the plan
cannot say that we are going to measure corrosion
potential. No way.
We have to say that in order to measure
corrosion potential, if this is an issue, we have
decided or we know that there is a potential problem,
and the way to address this is to develop the specific
instrument that can be prepared in such a time, in
five years, and we have a plan for this. This is the
response.
What is important, too, and this is part
of the plan for my confirmation, has to be a clear
indication that in the confirmation period will be a
verification basis for continuation of the type of
tests they have been doing now.
And even we have a provision in the
agreement, specific agreement, that said that in order
to complete, for instance, this type of information
that we are requesting here in the measure methods for
possible corrosion rate, or even this, that this
information will be provided during the course of the
eventual license application, because this is
information that will take time.
But we have to make sure that at the time
of the license -- and I am correcting you. You said
six months of testing, and they have far more than six
months.
VICE CHAIRMAN WYMER: I was biased.
MR. CRAGNOLINO: They have several years
of testing.
VICE CHAIRMAN WYMER: I know that, yes.
MR. CRAGNOLINO: And this is important,
but we are going to gain more confidence if they have
a clear plan to continue this, because this is one
table in which all these types of clarifications
should appear in order to evaluate the license.
We need sufficient data, but a clear plan
as to what are they and different times.
VICE CHAIRMAN WYMER: I just wanted to be
sure that we brought out the clarification of that
point, and so we fully understood it.
CHAIRMAN HORNBERGER: Yes, and I would
make one comment on that. I think the words on your
slide give a little bit of a false impression that
they are generic, and you refer to all monitoring and
measuring, and my home, the thermal couples in my
furnace, are 45 years old, and still functioning very
well.
And if you are talking about sophisticated
instrumentation to measure corrosion films, you are
right. But I don't think it is correct to use the
general thing that all monitoring and sensing
instruments need development. There is a lot of
things that function very well for long times. But --
MEMBER LEVENSON: And to make one further
question --
CHAIRMAN HORNBERGER: Please go ahead.
MR. CRAGNOLINO: I have to review my --
well, I have to put the condition as closed there and
say that not thermal couples, but a few others. But
for simplification, you need this type of thing.
MEMBER LEVENSON: Yes. Well, the generic
statement of sensors and monitoring devices, and other
application is much too broad.
MR. CRAGNOLINO: Well, you know, my
manager told me that I have to remove the application
that I have in mind, and one of them is a RST
generator in a nuclear power plant; and the other one
is reactor core.
MEMBER LEVENSON: I have one question, and
that is in doing your studies on release source term
from inside the canister, how many of the types of
things that will be -- that are scheduled or might be
there, are included in your review?
I mean, we have the spent fuel, and there
is the vitrified waste from Savannah River, the two
obvious ones. Did you also look at -- and I don't
know if this month it is in or out, but the Cannon Can
Program for disposing of weapons grade plutonium by
putting massive amounts of plutonium in the middle of
vitrified logs, and then the aluminum waste program
from Savannah River?
MR. CRAGNOLINO: No. We don't look in
that detail, but we have looked precisely in Navy
fuel, and release from Navy Fuel, and all the
potential problems.
We have been looking at the way that DOE
has characterized in general the different types of --
because there is a large number of spent fuel as you
know that has been used in many different applications
for certain reactors, and that belong in many States
and so on that have been put together, and are going
to be disposed of.
And we have been looking in great detail,
but we have a clear understanding of the most
important type of fuel and the way that they are going
to be disposed by.
MEMBER LEVENSON: Well, it is not clear to
me that massive amounts of almost pure plutonium in
the middle of a vitrified log is a less significant
source than the fuel.
MR. CRAGNOLINO: I am sorry to interrupt.
The point is that we evaluate what DOE presents to us
in a document, and we cannot move forward to some
potential scenario related to something else.
What my concern is that there could be
design changes as you mentioned of this nature. We
are not in the condition of evaluating now until the
DOE puts this in the table.
But it is this type of thing happening and
in the event of putting us in the situation that it
would be difficult to complete certain aspects of the
evaluation and the documentation.
And therefore we will prepare in the same
way for the license, but this is something that should
be evaluated, depending upon the nature of the
problem.
I agree with you, but you have to realize,
too, that this is mostly is commensurate with spent
nuclear fuel and this is not our concern, because it
mainly contributes to the goal, and this is what we
are putting a lot of our resources to evaluate.
CHAIRMAN HORNBERGER: I don't think we
want to try to resolve all possible design changes
here.
MEMBER LEVENSON: No, no, no.
CHAIRMAN HORNBERGER: The real question is
where do we stand with KTIs.
MR. AHN: I would like to add what Gustavo
explained to you, the force and theory of TSPA, as
well as the NRC TPA. We concede that there are three
different types of inventory.
The first one is the commercial spent
nuclear fuel. The second one is the high level waste
clad. The third one is the 270 different types of
fuels, and that is called DOE owned-spent nuclear
fuel, including Navy fuel, and plutonium disposition
fuel as you mentioned.
The third type of category has a very
small amount of inventory compared with the first two.
Therefore, it does not directly affect the dose,
except for the criticality.
And as you mentioned in the criticality
assessment, there are fissile material in plutonium
glass contributes to that. That is separately
treated, and in Subissue 5.
So therefore we believe that we consider
also in actually the assessment that DOE uses
artificially and very conservatively in this scenario.
The actual radionuclide release is based
therefore on the solubility limits. We even did not
discuss about their approach because they chose the
very conservative approach in the last package
changes.
CHAIRMAN HORNBERGER: Okay.
VICE CHAIRMAN WYMER: I wanted to say one
other thing. I think this whole issue of container
life and source term is probably the most complex of
all of the issues involved in the repository. It
certainly is extremely complex.
MR. CRAGNOLINO: Yes, I would say so.
CHAIRMAN HORNBERGER: And I want to say
that I think the center is doing an outstanding job in
dealing with all these issues considering the
complexity of them. I just wanted to be sure that I
didn't give the wrong flavor to my discussion.
CHAIRMAN HORNBERGER: John.
MEMBER GARRICK: On your status table,
Viewgraph Number 3, I guess it is. Would you comment
on which of these subissues gives you the greatest
concern, or which 2 or 3, in terms of resolution?
MR. CRAGNOLINO: I would say that this is
the one that has the most significant impact.
CHAIRMAN HORNBERGER: That tallies with
what Bret presented, because that is the one that you
had major, and you had 2 through 5 being moderate to
major. And you had number six being minor.
MEMBER GARRICK: Right. Right.
MR. CRAGNOLINO: In the way that we relate
it, because originally we decided this subissue --
that there were more options in the DOE program. Now
it is reduced to drip shield, and several since are
tackled together here.
But you are completely right. This is the
dominant one, and there are others that we have to
obviously keep track of. In this area, the DOE has
done very good work, and there is a lot of progress.
In this area the problem has been
accelerated in the last few years, and now we know
that DOE has put in a lot of research, and this is
reflected here in the technically updated biyearly
report.
And one of the more solids in thermal
sizing is to provide additional new information in
Appendix Number E, and that is the waste package.
That means that obviously we have to concern ourselves
to be very clear to express the concern, and at some
point at the time of the license that we not be
completely clarified.
But the duration is right, and I think
that we are in the condition of evaluating the
progress. We need to monitor this very closely and
help to rectify. If we see that the priorities are
going to some issues that are not so relevant as
others.
MEMBER GARRICK: Okay. Thank you. I have
no further questions.
CHAIRMAN HORNBERGER: Gustavo, I just had
one question. You had mentioned in your work plan for
2002 that you were going to complete the assessment of
natural archeological and industrial analogs.
Are there any requirements -- do any of
your KTI agreements go to natural analogs, or is this
something that you are doing because you think that it
is important information to bring to bear on the
analysis?
MR. CRAGNOLINO: Yes. There was not
specific agreement in the CLST KTI, but this turned
out to what is called multiple lying of evidence
issue, and I think it is important.
It is clearly something that we overlooked
at the time of the CLST agreement, because we were the
first kids on the block really. We didn't pay
attention, and also we have to recognize ourselves
that the fact that the industrialist period was just
very limited, and we did not give it the importance.
And that we realized that there was
concern on their part, and there was concern on the
nuclear waste technique review board and the title is
a little bit out of the issues, because it is very
difficult to cover completely this type of issue.
And the industrialist period in Alloy-22
is 20 years, but we tried to combine as much as
possible information for other industries, and also
made the fundamental case of specific localized
corrosion for natural or archeological analogs that
can't have something relied. This is the only thing
that we can say.
MR. AHN: This is Tae Ahn, and I would
like to add one more thing. We did not specifically
write any agreement.
CHAIRMAN HORNBERGER: Okay. You did. And
as you pointed out, Gustavo, you were the new kids on
the block. Have other kids followed you on to the
block?
MR. CRAGNOLINO: We don't have too much
integration to people. Well, the center has a
tradition in this. You have to realize that many
years ago there was precisely in the center that we
were sharp to this kind of issue of natural analog for
all of your geochemical type of aspects in particular,
and in relation to source term.
And the center has the very good feeling
of what you call in NOPAL 1, the face of --
CHAIRMAN HORNBERGER: No, I know all of
that, but my question was a bit more specific, and
that is do you know if DOE in particular is doing any
work on analogs related to the current materials?
MR. CRAGNOLINO: As far as I am aware, no.
In the case of metallic containers.
CHAIRMAN HORNBERGER: Right. Thank you.
MR. CRAGNOLINO: I'm sorry, for the source
term, you know -- you are aware of that?
CHAIRMAN HORNBERGER: Yes. No, I am aware
of all of that.
MR. HAMDAN: Gustavo, since you --
CHAIRMAN HORNBERGER: Use the microphone.
MR. HAMDAN: You seem to -- I mean, you
seem to be allowing for the option of doing some of
this work during the performance confirmation,
correct?
MR. CRAGNOLINO: Well, no, no. Repeat the
question again?
MR. HAMDAN: The question is that in the
things that you think need to be done --
MR. CRAGNOLINO: Yes.
MR. HAMDAN: -- that some of this work can
be postponed, or will be done during the performance
confirmation.
MR. CRAGNOLINO: No. There is a different
level. I see that the confirmation -- DOE is
conducting a program in which they tried to measure
corrosion rates in a series of embriddlement that are
in principle relevant for Yucca Mountain.
At the time of the license application,
and I don't know when it is going to be, but let's
assume that the license application is going to be 2
years down the road, or 3 years down the road.
Then we have 3 years or 5 years down the
road where they are already done for several
environments, and they provide the time that they will
be used for the license application.
But the series of embriddlements and where
they started the tests very early, or very late,
modification that came from the other studies that
maybe they have only 1 or 2 years, or just one year of
testing.
So that means that 3 years will be 4
years, and this is a very limited time. For this, we
need to have a clear plan that they are going to
continue this, because there could be surprises in
this new type of new environment that they are
testing.
MR. HAMDAN: But the question I want to
ask is because of this option being available with the
first confirmation, is it correct or incorrect to say
that really the license application is an artificial
midline, and it is because if there is something that
they do not finish, or they do not have enough data
for, they can always take it from the prelicense
application column, and put it in the performance
confirmation column?
MR. CRAGNOLINO: But to put it in --
CHAIRMAN HORNBERGER: Let me interrupt and
ask if Bill Reamer would like to give a short answer
to that question.
MR. CRAGNOLINO: Sure. That is not my
answer.
MR. REAMER: Thank you. Bill Reamer. I
am not sure that it is quite that simple. I think
that there probably are examples within specific KTIs
-- perhaps this KTI -- of certain agreements that
involve testing, and where data could come in after
the license application under a performance
confirmation label, or as part of the performance
confirmation plan.
But to say that information that the staff
needs for license application, if it is not available,
could simply be moved to the performance confirmation
column is probably not correct, because there is
certain information that we need and we have provided
that information to DOE. And we expect that it will
be provided to us.
MR. HAMDAN: I just wanted to make one
comment.
CHAIRMAN HORNBERGER: Okay. Really short.
MR. HAMDAN: Very short. This question
was basically to the container because their life is
10,000 years.
CHAIRMAN HORNBERGER: Right. Yes, but it
is a more general question, and Bill, I think, had
answered it, and we don't want to confuse the issue by
saying that it is arbitrary, that license, and
everything can be moved to performance confirmation.
CHAIRMAN HORNBERGER: Tae Ahn.
MR. AHN: Yes, I would like to add to
Gustavo's answer to your question on whether the
theories are working on the analogs. I believe they
do to a certain extent, because they are implementing
the agreement they had made with us.
I reviewed several presentations that
analyzed the passive film to a certain extent.
CHAIRMAN HORNBERGER: Okay. Thank you.
Any other questions or comments? Okay. Thank you,
Gustavo. We are going to break for lunch, and we will
start at one o'clock.
(Whereupon, at 12:15 p.m., a luncheon
recess was taken.)
. A-F-T-E-R-N-O-O-N S-E-S-S-I-O-N
(1:06 p.m.)
CHAIRMAN HORNBERGER: All right. The
meeting will come to order. This afternoon, again all
afternoon, it is a continuation of our being updated
on the key technical issues.
And we are just going to continue on.
This is the presentation that was postponed from the
original 11:20 presentation on unsaturated and
saturated flow. So, go ahead.
MR. WINTERLE: Okay. I guess the mike is
working. I am Jim Winterle from the Center, and the
KTI is unsaturated and saturated flow under isothermal
conditions.
The outline is pretty much the same as
what all the other KTIs have been. I will try to add
in a few extra remarks based on some of the questions
that I have heard earlier today. But I will just go
through that.
The KTI subissues are as follows. They
are climate, and which was a category that the first
two subissues fell under; present and future climate,
and the hydrologic effects. That issue has been
closed for some time based on the recent technical
exchanges.
Shallow infiltration is another one that
is closed-pending. Deep Percolation, that is flow that
moves below the root zone and the unsaturated zone,
and that is also closed-pending based on agreements
that I will talk about in a few minutes.
Saturated zone flow, and that is also
closed-pending. And matrix diffusion is a cross
between flow and a transport topic, but it has managed
to find its way up through or under this KTI that is
also closed-pending.
Let's see. What DOE needs to do. First
of all, they need to decide on a repository design or
set of design alternatives, and make sure that their
model is consistent with that, and I bring this bullet
up because of the low temperature operating modes that
were illustrated in the SSPA, supplemental science and
performance assessment reports.
And they showed some designs that diffused
would necessitate changes to their unsaturated flow
models. And that is the second bullet; that they
would need to update the models to be consistent with
the design alternatives.
And the third one, which is mostly what I
am going to talk about, is that they need to provide
the additional information related to what was agreed
to in the recent technical exchanges.
And I will break those down in a general
sense without going into each, but the shallow
infiltration sub-issue has two open agreements, and
both are due this year.
And if that information is acceptable,
that should move to a closed status. The deep
percolation subissue has six open agreements; one due
this year, and five are due in '03 fiscal year. For
the saturated zone, there are 12 open agreements, and
two of them rely on USGS and Nye County data dealing
with Death Valley regional models, and the Nye County
drilling programs.
And so we are not too sure if DOE can
commit to the date that we will receive that, but
other than those two, the one is due in October, which
is not yet received, and we will talk more about
overdue agreements in a minute.
And nine are due this year, and the matrix
diffusion, there is one that is not yet received, and
two due this year. Let me back up to a previous one.
I will just generally go over the topics
of the information that we are looking for. The
shallow infiltration subissue; we are looking for them
to document the Monte Carlo approach that they used to
bound the uncertainty in infiltration.
The high and the low range, and the
probabilities, and receiving certain distributions of
infiltration, and the mountain surface was based on
this, and it was not entirely clear to us how that
process worked.
And then we also want them to justify the
values of the parameters that were used in that
analysis, and we expect that information this year
again.
For deep percolation, there is an
extensive ongoing field testing to justify the seepage
estimates in the underground, and then we are also
looking for some geochemical data to help substantiate
their predictions of where flow will go below the
repository.
For the saturated zone, we are looking for
well data and analysis. Some of that from the C-
wells, and the testing has been long since finished,
but we have not seen all of the final results and
analysis related to that.
And some of that is related to the Nye
County wells, some of which are still being drilled.
We have asked them to look at some alternative
conceptual models for flow, and there is also an
update to the regional flow model for the Death Valley
ground water basin.
And we have not seen the update to that.
That is a USGS report, and so we expected it last
October, but it is just a little behind schedule. For
matrix diffusion, we are looking for tracer test
results, and we asked them to do some sensitivity
analysis to help us gain some risk insight to how
important that process is.
The overdue agreements I talked about, and
three were due in October, and two of those three, the
first and third ones, 501 and 604, had to do with the
C-wells test, and we are still waiting for that final
report.
And one had to do with the water level
analysis with both the new wells that have just been
drilled. In November, we were expecting some
additional information regarding how the regional
sites get models and interface, and are constructed.
I am not too worried at this point about
any of these overdue agreements. All indications are
that it is in the works, but it is something to keep
an eye on if the backlog of overdue agreements gets
big. We might want to bring that to a higher level.
A lot of the agreements are related to
ongoing field studies and one of the questions before
were if we were at all worried about information that
might not be able to be completed before a license
application could be submitted.
And I would say if there is anything that
we are worried about, it would have to do with these
ongoing tests in the unsaturated zone, and the tests
dealing with the Nye County wells.
Presumably they should be able to finish
this work by then, but as scientists, and as you
yourselves all know, that the analysis of those tests
can often take a long time.
The interpretation of those analyses can
often be the subject of debate. We might be surprised
by the results of the tests that we see, and offer
alternative interpretations where the results of those
tests might not be consistent with the current
abstraction.
So those are potential areas where we
might need to go back and request more additional
information, or determine which fraction of the
additional information we need or interpretations of
those tests can be relegated to a performance
confirmation period.
CHAIRMAN HORNBERGER: Jim, have they
resolved the problems that they had with the ECRB
passive tests?
MR. WINTERLE: That is a good example of
communication between the NRC and DOE staff. They had
found -- I think my next slide deals with that. Yes.
In the close off, it was almost a two kilometer
section of the east-west cross-drift, and to our
surprise it seemed like there was a lot more water
observed in there on a recent entry than was thought.
And to our further surprise, it seemed
that tests were going to be terminated before there
was a good handle on where that water came from and
its source.
And I think Neil Coleman deserves a lot of
the credit for this, and for getting with the
Department of Energy people and insisting that we
really need to understand where that water came from.
The early indications are that it is
formed from condensate, but that raises the issue does
condensate dripping need to be considered in the
performance assessment extraction. Currently, it is
not.
So where it has gone from there is the
test plans for that ECRB were extended, and
significantly enhanced, and cameras were installed,
and sensors were installed, and air flow monitors were
installed, and humidity monitors, and the period for
the test now to my knowledge is specified as
indefinite basically until they find the answers to
the questions.
CHAIRMAN HORNBERGER: Can't they analyze
the puddles and just figure out very quickly?
MR. WINTERLE: They have taken some
preliminary samples and that is -- the early
indications are that it is condensate. Of course, it
lands on the conveyor belt and there is all kinds of
crud on there.
So they have rigged up collections systems
underneath rock bolts, and plastic sheets so that they
can try to collect water in a more scientifically
robust fashion.
MEMBER GARRICK: Was that the section that
they thought they had closed off and in fact the
ventilation duct went through the seal and was not
sealed? Do you know?
MR. WINTERLE: I don't think so.
MR. LESLIE: Brett Leslie from the staff.
I think you are talking about the heater test.
MR. WINTERLE: Yes, this one has a double-
bolt cut head over a large portion of it, and so I am
sure that there is some escape through the cracks
around the bulk head, but they have done a good job to
minimize that.
MR. COLEMAN: Excuse me, Jim, before you
go on. Neil Coleman, NRC staff. One other item worth
pointing out here that DOE needs to do before LA is
resolving the chlorine 36 paradox, where there was one
study that indicated no bump posts chlorine 36 at
repository level,and all of the previous studies
indicated numerous occurrences. We have an agreement
related to that.
MR. WINTERLE: Thank you, Neil. What the
NRC and CNWRA need to do before a license application
would be to continue reviewing this DOE data
collection associated analysis that they are going to
use to validate the process in abstracted models, and
review their modeling activities used to validate
process in abstracted models.
And maintain familiarity with DOE methods
and assumptions is a big part of what we do. There is
a mountain of reports. And continue to develop the
sites from our own process modeling, and I will
mention some of that in a few minutes.
And interact with DOE as we have been
doing over the past few years to make sure that they
understand what our underlying concerns are. In 2001,
some of our accomplishments were -- well, I guess I
will start with the first bullet.
We identified and prioritized our tasks by
those with the greatest technical and regulatory risk
significance, which is what led to the KTI agreements,
and we reviewed a mount of AMRs, and participated in
the technical exchanges.
We have done our own independent
unsaturated zone modeling and field studies. We have
evaluated DOE's field studies that are ongoing. We
have developed a saturated zone flow model that we can
use in our own performance assessment, and to test
alternative models for how geologic structure might
affect flow.
And we are supporting the TPA code
development and we have provided input for the Yucca
Mountain review plan. In 2002, we have got a few
milestone reports that we will provide to the NRC
dealing with saturated zone modeling, and the
hydrogeology of the Valley-Fill based on what we are
seeing from the Nye County wells.
And that will be in integration with the
structural defamation and seismicity KTI. A lot of
that work. And we are going to do some of our own
independent -- document some of our own independent
modeling, and review DOE field studies in a report to
NRC.
CHAIRMAN HORNBERGER: I am interested in
that last bullet because you are talking about field
studies, but you are going to do modeling and not any
field studies of your own.
MR. WINTERLE: We are going to do --
document some of our modeling of flow in the paint
brush tough layer. There is -- this might be a good
point to bring up changes that the Department of
Energy made to their unsaturated zone model for the
SSPA.
And we are not sure if that was just an
aside from the TSPA SR, or if that is what they are
planning to go forward with in the license
application.
Those changes allow for significant
lateral diversion of flow in the PT, and up to 40
percent of infiltration can now be diverted away from
a repository area.
We don't necessarily believe that there is
data to support that and so we are probing with
independent models of how that might work. We are
also looking at effects of heterogeneity.
And then in addition to that, we will have
a separate chapter of the report that summarizes what
we know so far based on the early information from the
seepage studies.
And, of course, we are going to continue
to review all the technical documents that come out of
the Department of Energy. And this is more repeat,
but we are going to continue our own independent
modeling which improves our understanding of risk
significance and preparedness.
I think it makes us better reviewers when
we occasionally attempt to model some of these things
ourselves. Review results of ongoing in situ niche
and alcove studies and this provides feedback into the
milestone reports that we will provide.
And on to the summary, where we have
obtained agreements with DOE that we need to close or
closed-pending status for all of our subissues. There
is a review and is now incorporated into seven risk-
informed integrated subissues.
I don't know how familiar the committee
members are with the integrated subissues, but there
are seven of those that this specific KIA provides
input to.
I guess that relates to Bret's table here.
The independent modeling and technical investigations
that we do continue to provide us insights for the
review of DOE models, and technical basis for our own
TPA development.
And our review of the DOE field studies is
vital for closure of these KTI subissues. There is
also a table in the back that summarizes how all of
this boils down into the sufficiency review.
CHAIRMAN HORNBERGER: So in Bret's table,
Subissues 3 and 6, you had classified as minor, the
level of complexity of the agreements, and 4 and 5
were moderate to major.
So I guess it is pretty clear that you
don't anticipate that DOE has any truly monumental
hurdles to get over here?
MR. WINTERLE: Well, Subissue 3, that was
closed before and they made some changes to their
infiltration estimates. So we reopened that one. So
we are expecting since the changes weren't drastic
that that is a minor level.
The matrix diffusion one, that is Subissue
6. I was actually surprised recently to find out how
important that seems to be to their transport
calculations in the unsaturated zone.
Yet it doesn't seem important for the
saturated zone, and one of our agreements is for them
to clarify why that is, and to make sure that all of
the numbers are plugged in there correctly, because it
doesn't seem to make sense.
But I think if they can answer that
question, then it is minor in terms of the level of
effort that it is going to take to close it. It is
not necessarily in terms of importance.
MR. LESLIE: Bret Leslie from the staff.
Just to remind you that this was a snapshot done
approximately back in September. So that is as much
as the information is worth.
CHAIRMAN HORNBERGER: As I said, we will
get a presentation quantitatively what moderate to
major means. Milt, do you have any questions?
Raymond?
VICE CHAIRMAN WYMER: Only kind of an
observation. We keep -- these keep being added into
the models, and not only by you, but by everybody
else. And then there is the question of updating all
these models and running new tests.
It seems to me what is happening here, and
what could be happening, is that you are generating
mountainous piles of information that need to be gone
through and examined for differences to find anything
that is significant that changed.
What kind of or how practical is this, and can it be
done?
MR. WINTERLE: Well, that is a good point.
I am still finding things in the SSPA document that I
didn't catch before. One example is the range of
uncertainty for saturated flow, and it is from a
factor of 10 to now a factor of 3. We are still
considering whether that needs to be brought up as a
problem.
VICE CHAIRMAN WYMER: That is you and
everybody else, you know.
MR. WINTERLE: It is a difficult
situation, and we can achieve closure with one certain
model, but does that mean DOE shouldn't try to improve
if they can their model? So I guess at some point
they are going to need to stop and say this is what we
are going forward with the license application with.
VICE CHAIRMAN WYMER: There is not much
evidence of that so far.
MR. WINTERLE: I agree, that is really
something that we are trying to stay on top of.
CHAIRMAN HORNBERGER: John.
MEMBER GARRICK: Can you just make a
couple of comments about the scope and schedule of the
DOE field studies?
MR. WINTERLE: The scope and schedule?
MEMBER GARRICK: Well, I will give the
disclaimer that everybody gave, that that is up to
DOE. But I think it is an ambitious undertaking.
MR. WINTERLE: But it is part of the
agreements isn't it?
MEMBER GARRICK: It is part of the
agreement.
MR. WINTERLE: The PTI agreements.
MEMBER GARRICK: I think the testing
schedule that they have done, the tests certainly can
be completed by the last I heard was 2004 for the
license application. Don't anybody quote me on that.
But I think that is reasonable.
CHAIRMAN HORNBERGER: This is being piped
directly to --
(Laughter.)
MR. WINTERLE: But I think the testing can
be finished, and the Nye County wells can be drilled.
It is just a matter of whether the information and
analyses that come out of that become controversial,
or counter the current model abstractions. You just
don't know what you are going to get.
Early indications with -- there is one
test called Alcove 8-Niche 3, and they have an Alcove
one 10 meters above the other, and I guess things in
the field never go as smoothly as you would hope.
And they hoped that the water was going to
go straight down on the fault pass and it looks like
some of it is coming out into the access drift, and
you can see it on the walls.
That wasn't really in the plan, but that
does not necessarily mean that you can't get useful
information from the test.
MEMBER GARRICK: Thank you.
CHAIRMAN HORNBERGER: Anything else?
Staff. Okay. Thank you, Jim. And I guess we will go
from isothermal to thermal.
MEMBER LEVENSON: Things are heating up.
CHAIRMAN HORNBERGER:
(Brief Pause.)
CHAIRMAN HORNBERGER: Randy, if you can
introduce yourself for the record, please.
MR. FEDORS: For the record, I am Randy
Fedors. I just recently took over the TEF part for
the Center. I have listed the contributors, and most
of them from last year contributed various portions of
this.
And Jeff Pohle and Asas Chowdhury, too,
are the managers on either end. It is the same
outline that you have seen all along. The status and
what DOE needs to do, and what we need to do, and what
we have done in the past year, and this coming year.
This was unintended on how fancy that came
out, and I didn't know what I was doing. It w as a
template from somebody else. Subissues. There are
two subissues for TEF.
One of them is related to the features,
events, and processes, and that just says the two
technical agreements on the reports in the database
are closed-pending.
The other subissue deals with all the
topics for thermal effects, and can people hear me in
the back by the way? Okay. I will assume everybody
can.
CHAIRMAN HORNBERGER: Nobody responded and
so that can mean yes or no.
(Laughter.)
MR. FEDORS: Okay. A thumbnail sketch of
the different technical agreements for the TEF and
determining temperature, humidity, saturation, and
flux, and in numerical order here, and in no order of
priority for us.
The first one is losses through the drift
scale thermal test bulkhead. There was a question
that started to allude to that topic, and the notion
that they put heaters into a drift to mimic canisters,
and then realized after they had been going for a
while that there was quite a bit of heat loss out, and
quite a bit of water loss out.
And on the order of two-thirds of the
vaporized water is what they are predicting right now
is exited through the bulkheads. So there is a
technical agreement to come to some resolution of
that.
The next one, the cold trap effect, is
what I am lumping together, and the process where you
have well known unsaturated zones, and ambient
conditions or otherwise, that you should expect near
99.9 percent relative humidity in the core space.
When you get that in a drift, as thermal
pulses, you are not going to get -- as you are not
drying out the zone, and you will start wrapping up
that relative humidity to that near a hundred percent.
And you might start redistributing the
vapor in the drift, and temperature readings are one
reason that you would look for that. So if there is
an edge effect in the repository, you might be driving
moisture to the cooler zone, and condensing there.
So there would be an evaporation
someplace, and then the question is how much movement
of that vapor to another location were you then
condensing.
By the way the passive test in the last
presentation was what really brought that back to the
forefront, and now the Department of Energy is dealing
with. The next topic is the ventilation model, and
two agreements. One was the ventilation test plan and
the other one was document the ventilation model.
Since I might not discuss that later, I
will just quickly summarize it; that the DOE model was
a linked model. Here is our ventilation and here is
our thermal hydrology, and we will just apply a flat
rate heat loss reduction to mimic ventilation.
The Center turned around and linked it
directly with multi-flow and a ventilation model that
entwined, and came up with a DOE approach that seemed
to be reasonable.
So I may not mention that again the rest
of this presentation. The next three, parameter
uncertainty, and full uncertainty, and model
uncertainty, there are four or five agreements that
deal with this, and essentially we are looking for
-- there is a lot that we are not sure of.
There is a lot of parameter uncertainty
and model uncertainty. We want to take a stab at
incorporating a full ensemble of model and parameter
uncertainty in the TSPA, or come up with a
justification for why it is not there. That is the
gist of those agreements.
MEMBER GARRICK: Did you specify in any
way how any of these things should be done? I am
thinking of something like model uncertainty, which
there is lots of discussion, debate, and controversy.
MR. FEDORS: Well, in a later slide where
I tell you what we are trying to do in Fiscal year
2002, I would have said that we are going to try and
take a stab at ourselves first, and see where it is
going to go so that we have some background if DOE
comes up and says something, and that we will have a
little bit more understanding of what is going on.
It is something that you don't look in the
published literature for here is how to do it.
MEMBER GARRICK: That's why I asked the
question.
MR. FEDORS: One way to do it, I think, is
pure brute force and spend lots of manhours, but I
don't think that is practical.
MEMBER GARRICK: I was just curious if you
had developed some interesting and ingenious approach
to that problem. Proceed.
MR. FEDORS: I am not that quick. The
progress on the technical agreements, I put down that
there are three that are considered satisfactorily
addressed.
I think you can fluctuate that number
between 2 and 4 depending on who you talk to and what
the programmatic decision on what satisfies an
agreement.
For the ones that we are going to be a
little more concerned with, one technical agreement
required a path forward, and that was the one that
treat model and parameter uncertainly literally
written into the technical agreement, stated that this
will be discussed at the TSPA technical exchange.
Well, we needed a path forward, because
that is not a path forward, and that's -- well, Jeff
Ponle was telling me today that the NRC letter pending
is not -- well, it is going through concurrence right
now on how -- you know, what the path forward will be
for that technical agreement.
We did not specify where it is going to be
documented, but here is the essence of what we were
looking for. There is one technical agreement waiting
for a DOE response, and that has to do with the drift
scale heater test. There was a white paper on that
prepared by the Department of Energy.
And NRC responded, and then part of the
technical agreement said that the Department of Energy
will address NRC's comments. Seven technical
agreements had some documents due in Fiscal Year 2002.
So those are coming up.
Three agreements had delayed documents and
the Department of Energy sent letters in July and
October stating here is when we think these things
will happen.
And also my last bullet notes that the
approach that the thermal effects group has taken in
the past year was that as things were delivered that
letters were sent to the NRC either reviewing it or
acknowledging that things were completed.
What does DOE need to provide before
license application? With the flexible design and
operating mode, one of the first thoughts that would
come to my mind is, well, if they go with the low
temperature operating mode, I don't think we have
enough information for that.
It is a case where it is not only where
they are expanding the repository domain and there is
a characterization issue, but there is also maybe the
cold trap effect that is going to become even more
important for that operating mode.
And that is the second bullet and there is
some data to support the cold trap effect. That is
not to say that the cold trap is not going to occur
during the high temperature mode, but for example, the
low temperature mode, if there is no ventilation, you
will be near a hundred percent relative humidity the
entire duration.
With the high temperature mode, you get to
60 percent relative humidity after 2 to 3,000 years
approximately, and the canisters don't to below a
hundred degrees C until 3 or 4,000 years, or something
like that.
Anyway, there is a much shorter duration
when the cold trap effect will be important for moving
fluids around along the drift axis, for example.
The other thing that we are looking for
before LA is some approach for the model and parameter
uncertainty, and as was already noted, that is not
clear how that is going to be done yet, and I think
how one would do that.
I think it will end up being a combination
of here is what is reasonable to do, and we will
justify what we don't include, and we will expand
uncertainty and account for things, and I will leave
that one alone.
And then the last bullet is just a catch-
all for the rest of the technical agreements. The
first bullet where I just wanted to reemphasize that
there is going to be water in these drifts at some
time or other, and how it redistributes during the
10,000 year period is what we really want to look at.
The passive test brought to the forefront
really the cold trap effect, and that is an example,
and we are concerned about that one. The data that
has to be collected by license application, the
importance of the Cross Drift Thermal Test is becoming
more prominent here, but it is a test that is not
projected to be started until 2003, and I have heard
2004.
It is down the line for some reason, and
I say that in terms of in light of the drift scale
test problems, we are not sure what conclusions that
we would be able to support coming from the drift
scale test given all the water loss out of that test.
And some of those observations I would be
referring to are when water starts -- the possibility
of water going down fractures and through the dry out
zone, the fingering, the reflux approach.
CHAIRMAN HORNBERGER: Do your agreements
require our call for results from the cross drift
thermal test prior to an LA?
MR. FEDORS: No.
CHAIRMAN HORNBERGER: Okay.
MR. FEDORS: And my feeling on that was
that it was a combination that was practical and there
is some information that you can get out of the drift
scale heater test still.
But I think our position on that, if I can
speak for others, is that there may be difficulties in
quantifying that loss out there and what effect it
has, and you had better include it in the model and
parameter uncertainty.
The field and laboratory data to support
the cold trap model. That one we don't even know what
the DOE cold trap model is going to be, and that is
not documented and presented to us yet.
I know that people are working on it, and
that is about it. And the characterization and
heterogeneity of properties in the lower lithophysal
zone, that is the whole reason that they created the
enhanced characterization repository drift block or
block drift, was to get at the lower lith (sic).
That is proceeding along I think and it is
not going to be an issue. Another bullet, coming
forward with the mismatch. This is kind of playing up
to what I alluded to before.
Depending on the operating mode, other
certain processes are going to become more important.
So we may be caught in a bind at the end and we may
not have this prior knowledge that the bind is going
to be there.
And the last bullet on the model is
parameter uncertainty, and we talked about that a
little bit, but I might also add the point that this
is a difficult one to handle because of the
transparency issue with the multi-scale thermal
hydrologic model.
There is an assemblage of 4, or 5, or 6
numerical models linked by abstractions, and scan
lines as they call it now. I think that may be going
under some revision, but in essence how do you trace
at least your parameter uncertainty through that whole
chain? It is not straightforward.
What does the NRC need to do before LA?
Well, you have probably seen a few of these same
things here. I might just summarize them as a little
bit of our job is to kind of anticipate what we are
going to run into down the line, and that the meetings
that we have with the Department of Energy certainly
facilitate us being a little bit more up to speed on
things, rather than waiting for a document to come out
sometime later.
Fiscal Year 2001 work. The major
activities, and all the technical exchanges are on the
top of that list for a reason, I guess, time wise.
The ventilation modeling I alluded to
earlier. The Mountain-scale thermohydrological
modeling studies, and the main focus of those were
primer sensitivity, and edge effects, and how that
might dovetail with the cold trap effect.
I have the cold trap listed here and the
technical support for the TPA code. In other words,
providing temperature and relative humidity histories
for the duration of the 10,000 year compliance period.
And I have one example to present, and
that's the cross-drift thermal test. Some modeling
was done prior, and obviously prior modeling, and
predictive modeling.
The importance of this was that this niche
test is in the lower lithophysal, and the other
important factor is the problems with the drift scale
test.
So the main objectives are put in thermal
heaters and see if we can predict the sheding of the
water, and how water is going to flow through the
fractures, and then analyze that water.
The modeling that I am showing here, this
is a saturation, a three-dimensional model. The blue
you can tell is where the dry out zones are,and that
is the low saturation.
And the higher saturations are the yellows
and oranges, and you can see some of the shedding
going on there. The bore hold locations where there
would be some collection, you can kind of pick that up
here.
There is some little blue spots there.
And in the modeling, they turned out to have a low
saturation because it is difficult to get water into
those. One of the conclusions of this study was that
looking at the flux of the water that got in there, it
was that it was condensed water.
It was water that evaporated and it was
transported into there, and then condensed in the bore
hole. That would have a major implication for how you
are interpreting the chemistry of this test and the
drift scale thermal test.
Work plan for Fiscal Year 2002. I will
jump quickly since we are almost up to the 15 minutes
here. The intermediate milestone reports are the
focus of what we are doing, and that is, one, review
all the technical agreement documents that we have
gotten, and that will be organized around dealing with
the model parameter uncertainty for the most part, and
the cold trap effect, and dealing with the drift scale
thermal test results.
I will also plug the cold trap modeling
and that we have created a little laboratory
experiment to try to understand a little more fully
with an analytical solution to start up and design the
experiment.
We have gone through some initial testing
just to see and make sure that water is condensing,
and that all of our sensors are reading stuff. I am
not aware that the Department of Energy has a
laboratory or a field test in the plans at all to
support any of a cold trap model.
And to summarize, there are three
technical agreements essentially completed, and seven
more should be done this year. And the three have
been delayed, and with that, I will take any
questions.
CHAIRMAN HORNBERGER: John.
MEMBER GARRICK: I don't think I have any
questions.
CHAIRMAN HORNBERGER: Okay. Raymond.
VICE CHAIRMAN WYMER: Well, this question
is going to tell you very clearly that I don't know
anything about these kinds of processes. But what is
more important with respect to humidity in the
repository; the water that is brought in by the
ventilation or water that drips in from percolation?
MR. FEDORS: It would be the other way
around for the ventilation. The ventilation -- I
think it was initially brought in to remove the heat
load, but it is very effective at keeping the humidity
way down low.
So as you soon as you shut off the
ventilation -- for example, like during the low
temperature operating mode -- the relative humidity
shoots up immediately to a hundred percent.
VICE CHAIRMAN WYMER: And this gets
outside of your scope, but the plan is that as each
drift gets filled, they will close it?
MR. FEDORS: That is something that I have
not sorted out entirely, and I thought there was some
confusion because some discussion of ventilation,
natural ventilation, so they would leave the drifts
open to take advantage of any -- you know, they have
got the infrastructure down there and there might be
additional ventilation over and above an ambient
ventilation through the mountain that would help them
out.
But as I understand from the igneous group
that they have -- that we need bulkheads in here to
help out. So there is a conflict there.
VICE CHAIRMAN WYMER: Okay. Thanks.
That's all that I have.
CHAIRMAN HORNBERGER: Milt.
MEMBER LEVENSON: Yes, I'm not sure that
you used those very clearly defined words like
important and less important, and more important, and
likely.
But since water is the 800 pound gorilla
in this entire thing, and there appears to be water in
the drifts, and there isn't any corrosion, et cetera,
how would you categorize the importance to the overall
program of this particular KTI group?
MR. FEDORS: Well, if you want me to rate
it, I would say it is 799 pounds. Do you want it more
likely or less likely?
MEMBER LEVENSON: No, no, I think you
would agree that it is one of the most important ones
probably.
MR. FEDORS: It is an important issue, and
I think the focus should be on the duration of time
over which it is going to be left in there.
MEMBER LEVENSON: Are you comfortable with
the modeling of thermal effects on flow vis-a-vis
what's going on in the rock? In other words, do you
think that we have a good enough grasp of the
phenomena so that the questions of hot versus cold
repository, et cetera, can be accurately modeled?
MR. FEDORS: I think the thermal part --
if I had to categorize things, the isothermal part is
very complex, and if you give me that part of it, I'd
be a lot more confident in getting the thermal part
correct. Is that kind of answering your question? I
believe it's very complex how water flows through
fractures.
MEMBER LEVENSON: Yes, but it's much more
complex than just how water flows through fractures in
that if you have a hot repository for a few hundred
years, you're going to be having evaporation in the
pores. If you have a cold repository, presumably you
have almost none. Are those effects part of what
you're looking at when you talk about thermal effects
on flow? If you're evaporating water for 500 years,
is there any porosity left when it finally cools down,
et cetera? It seems to me there are some major
issues.
MR. FEDORS: Then my point would be how
quickly does it re-wet. We know that it will -- water
will be supplied to the drifts through the fractures
rather quickly to pass the test, things like that, but
filling the matrix --
MEMBER LEVENSON: Yes, but filling your
tea kettle up with water -- filling your tea kettle up
with water doesn't redissolve the deposited stuff.
MR. FEDORS: Doesn't redissolve the
depositing stuff.
MEMBER LEVENSON: Yes. For 500 years, if
you're boiling or evaporating water out of these
pores, at the end of that time, those pores are not
automatically going to just open up.
MR. LESLIE: Milt, this is Bret Leslie
from the NRC staff. This is really a question for the
near-field where we take into account and evaluate
those things.
MEMBER LEVENSON: Okay. I withdraw it.
I was trying to find out where it's being covered.
MR. LESLIE: It's covered.
MEMBER LEVENSON: Okay.
MR. FEDORS: Thermalhydrologic chemical
near-field. So your question had to do with you've
plugged up the pores of the matrix.
MEMBER LEVENSON: Just the whole -- all of
the things that are different between the hot and cold
repositories. There are many things.
CHAIRMAN HORNBERGER: On one of your
slides where you listed -- the title was "Concerns."
And the bottom bullet talking about incorporation of
ensemble model and parameter uncertainty in TSPA and
it says something about the path forward is the
subject of a pending NRC letter to DOE. Can you tell
me a little bit about that letter?
MR. FEDORS: Yes. Actually, I noted that
in two slides. In the other slide, I mentioned that
Jeff Pohle had mentioned today that that's going
through concurrence right now. The letter basically
said, "We realize that there's no path forward for
this agreement. Here's what NRC was trying to get at
with this techno agreement -- incorporate model
parameter uncertainty or justify not including it in
the TSPA. Please put this in some document, either
one of those."
MR. POHLE: Yes. This is Jeff Pohle, NRC
staff. It's basically a paragraph within the letter
that covers a number of KTIs coming out to Jim
Anderson. Basically, what we do in that paragraph is
tie it into a number of items the Department said they
were going to do. At the technical exchange, we got
this thick handout, so we called out of there, "This
is what you said you're going to do, and we think if
you're doing this way to this agreement, that
agreement, that agreement, this ensemble, if you did
this, would be the path forward to resolve this
particular item." I don't know the numbers well right
now.
MR. FEDORS: I probably would say it's not
a worry either. There was some discussion whether we
even had to come up with a path forward, because these
other technical agreements separately address
different aspects of it.
CHAIRMAN HORNBERGER: Okay. But it's --
MR. FEDORS: So we just want to be clear
on it is all.
CHAIRMAN HORNBERGER: So it's basically
just clarifying the path forward.
MR. FEDORS: Yes.
CHAIRMAN HORNBERGER: Okay.
MEMBER LEVENSON: George, I have a
question.
CHAIRMAN HORNBERGER: Sure.
MEMBER LEVENSON: On your backup slide,
Number 14, you give the technical support for the TPA
code. Starting at time zero and going out for close
to 100 years, the high temperature mode has the waste
package temperature below 100 degrees, and that seems
awfully low since the reason for going to the low
temperature was to keep the rock temperature below 100
degrees. And the waste package has got to be hotter
than the rocks since it's the source of the heat, so
it just seems that for a high temperature to be below
100 degrees seems awfully low.
MR. FEDORS: Okay. These results contain
a ventilation module, and you'll see the first spike
straight up as, what, 50 years about? So this assumes
forced ventilation for 50 years, and that's why I was
alluding to it's very effective at removing heat and
moisture. And in this particular modeling approach,
they also played around with natural, quote, "natural"
-- you know, what if natural ventilation kept going
after the 50 years? So we'll ramp the ventilation
down in the model to some fraction of the forced
ventilation, and they did it in two steps. That's
where these other spikes are coming from, just to
explain the graph.
MEMBER LEVENSON: One other question: Do
you know when they talk about natural ventilation, are
they talking only about the chimney effect or are they
taking into account the breathing of the Mountain?
The USGS made some rather extensive measurements on
Yucca Mountain Proper which indicates that there's
very, very significant air flow in and out of the
Mountain even if you have no drifts and tunnels. Was
that taken into account in the natural ventilation
analysis?
MR. FEDORS: No. This was more scoping
analyses -- what if ventilation was -- I don't think
the Department of Energy has any -- I don't know what
their stance is on ventilation after forced
ventilation period is. We think that might be
important, and we are looking at models now to first
let's get the natural ambient condition right with
drifts, and then if there's a thermally perturbed
through buoyancy effects, that you would have
ventilation from that in between the ambient and the
thermal -- rather the force ventilation at the onset.
MEMBER LEVENSON: Well, the USGS also
measured barometric pumping --
MR. FEDORS: Yes.
MEMBER LEVENSON: -- and a number of other
phenomena, and it was very extensive.
MR. FEDORS: Yes. We're looking into some
of those very reports right now to kind of help us
bound -- create boundary conditions and tell us
whether it's reasonable for the air flow, because it's
difficult to get air flow out of that, you're just
monitoring the pressure.
CHAIRMAN HORNBERGER: Questions from
staff? Anyone else? Thanks very much, Randy. Let's
see, where are we? We are on the near-field --
evolution of the near-field environment. Oh, did I
skip one? Oh, yes, okay. I didn't mean to skip you,
Paul. I didn't mean to skip you, Paul, but, again,
I'll let you introduce yourself for the record.
MR. BERTETTI: Okay. Hopefully everyone
can hear me just fine. My name again is Paul
Bertetti, and I'll talk a little bit about
radionuclide transport. Part of this project is
experimental in nature too, so this morning I
conducted a couple of experiments on the slipperiness
of the ice and the hardness of the concrete, and my
preliminary results are that the ice is slippery and
the concrete is hard.
MEMBER LEVENSON: That's project
confirmation.
MR. BERTETTI: That's right. Well, I'll
do some more of that tomorrow morning.
Well, my outline is the same as many of
the previous ones. We'll talk a little bit about
status of subissue resolution here with respect to
transport -- what we're doing, what we think DOE and
we need to do in the future.
Radionuclide transport just looks at
processes that control radionuclide migration, both in
the unsaturated part of the system and in a saturated
zone, outside of the area that's influenced by what we
call the near-field environment.
We have four subissues: transport through
porous rock, which is essentially that unsaturated
material; transport through alluvium, essentially all
saturated material; and transport through fractured
rock, which incorporates both the unsaturated zone and
the fractured volcanic top that is saturated below the
repository, and the fourth one is the nuclear
criticality issue, which people haven't said much
about. That's kind of -- there are several KTIs that
have this criticality component. They're kind of all
addressed together in one lump group, so I won't say
a whole lot about that.
We have several agreements. I'll discuss
them in detail in the following slides. Essentially,
they're focused on getting information that we need to
be comfortable about the processes used to model and
represent radionuclide transport.
So we'll start with Subissue Number 1,
which is the radionuclide transport through porous
rock. We have five agreements. One of these is
partially complete. What that means is we received
some of the documentation necessary to satisfy that
agreement, and that's undergone review. And we've
requested a little bit of additional information.
And, specifically, what we had requested was a
technical basis for screening criteria of
radionuclides, and I think maybe the way we
communicated what we wanted to satisfy that agreement
wasn't really understood by the DOE, so we've kind of
tried to clarify that position so that we get a little
bit more information in that respect. Four have
documents that supposedly will be delivered in fiscal
year 2002, given DOE's plan of action and their
current resources.
Just specifically, one of the most
important agreements that we have is kind of a very
generic one and that is provide analysis and
documentation of transport parameters. Essentially,
a lot of the values used for KDs throughout the
transport models in the system were based on expert
judgment, so the particular values are derived from a
very limited group of experts who kind of made
estimates on what those KD values should be. Yet
there's not a specific set of documents that provide
traceability and justification for those values. And
so what we've asked, and DOE's agreed to, provide that
information to us.
We also have asked for providing results
of the in situ field testing in the unsaturated zone,
and that alludes to the stuff that Jim Winterle talked
about earlier which are the alcove 8 niche 3 testing
and information that we can get from unsaturated zone.
I will add also here that we did a lot of
work in trying to come up with the language of those
agreements and what we were going to ask for, and I'd
say the majority of our agreements, the vast majority
of our agreements, are essentially just a request for
information that the DOE had already indicated that
they were going to produce.
CHAIRMAN HORNBERGER: Paul?
MR. BERTETTI: Yes, sir.
CHAIRMAN HORNBERGER: The last bullet:
Plutonium, uranium and?
MR. BERTETTI: Correct.
CHAIRMAN HORNBERGER: You mean --
MR. BERTETTI: Protactinium.
CHAIRMAN HORNBERGER: Yes. So
protactinium is an issue?
MR. BERTETTI: Well, this agreement comes
from DOE documents that indicated that they needed to
do more sensitivity studies to confirm that the data
that they had for these was adequate. So they
mentioned that there was some concern that the value
for the unsaturated zone of porous rock, that these
numbers that they had were adequate. It kind of
relates to this documentation. What they had is some
limited number of experiments. So what we asked for
is just to continue that work and provide the
sensitivity studies to see, well, do you really need
to worry about this or not. And so that's what we're
asking is can you provide some risk information to
justify what you're using or not.
VICE CHAIRMAN WYMER: But that's not an
indication that you really think that's a problem.
MR. BERTETTI: Well, we really don't know
unless we have some sort of information to gauge that.
Well, I mean that's part of the problem is we can make
a lot -- radionuclide transport is an interesting
issue, because a great many people think they know a
whole lot about it in the sense of, "Well, we can
model this using a KD." And the problem is there
aren't too many real examples of where a KD model has
been successful at predicting that actually has
happened. And so what we'd like to do is get some
information so we can check off the box. And I would
agree with you for the unsaturated zone in this sort
of transport, you're probably correct.
Subissue 2 is radionuclide transport
through alluvium. We have a number of agreements
related to this. One of them is complete. That was
kind of a request for an updated features, events and
processes document. One was due in October 2001.
Specifically, we had requested pre-test predictions
for tests conducted at the alluvium testing complex.
And the idea behind that was if we could have some
indication of what DOE expected to get from their
field experimental results, then we would have a
better way of assessing whether or not their
conceptual model and their process model, on which
their TSPA model was based, was adequate. We have not
received that yet, but they're making progress in the
testing, and ideally we'll get a testing plan in the
near future. Five of them are due next year, there's
another one due in the following fiscal year.
The alluvium is an interesting part of the
transport path in that it's not very well
characterized, and this kind goes to the question of,
well, how much information is enough? And I guess
that depends on how important that part of the system
is and whether or not you feel that your
characterization is sufficient to justify the way
you're modeling the system.
And as Bret mentioned earlier, this is
kind of way of -- an area where DOE and NRC have
differed in the way they model the system. The
saturated alluvium and saturated zone transport is
somewhat important in the NRC TPA code, at least in
terms of sensitivity studies, but is not very
important and the unsaturated zone component is the
most important for DOE.
So there's a little bit of disagreement in
terms of importance and sensitivity and as a result
you might see a little bit of difference in how much
characterization might be needed. So we worked hard
to get a set of agreements that satisfied both parties
in the technical exchange. I'll talk about some
specific information when I show -- when I discuss the
work that we've done over the last year.
Our Subissue 3 is radionuclide transport
through fractured rock. We have a number of
agreements here again. One of them is under review.
That's essentially the fractured, unsaturated zone
test plan that we've looked at. We'd like to get more
information to provide plans for characterization and
results of the in situ testing and to get information
from the fractured saturated zone testing.
Essentially, that's the testing that was
done at the Sea Wells complex. So the document that
we're waiting on is essentially the Sea Wells AMR that
will provide us with the results and the testing
process that was done at that location. And the
reason that's important is that's maybe the only field
test that's done in saturated, fractured volcanic
rock, so it's kind of an important, basic
characterization of the system.
I have a bullet on for colloids so I'll
mention them. Whether or not colloids are very
important to dose or very important to the performance
of the system is of some question; however, there's
field evidence that colloids may have been transported
at the Nevada Test Site, DOE has incorporated colloid
modeling in their TSPA model, we are going to
incorporate it in the next revision of the TPA code to
help us evaluate the importance of that. So what
we're asking for is a sensitivity analysis to kind of
test the importance of the parameters used to develop
that colloid model.
As I said, the criticality issue is
essentially kind of combined over a number of KTIs,
and it essentially involves delivery of numerous
topical reports to hopefully close out that issue.
What does the NRC need to do before
license application? Well, as mentioned over and over
again today, we need to monitor DOE's progress and
hopefully inform ourselves as to how they're
conducting their work and whether or not it's relevant
to what we think is important. And I think it's
particularly important for radionuclide transport,
because there's still a number of characterization and
field tests ongoing. And there's a lot of basic data
being collected, so we have to be particularly careful
about our ability to understand what's going on, how
things may or may not change during the collection of
that characterization data. I think everything else
is pretty self-explanatory there.
While this slide is not particularly
useful for this meeting, but it does -- the importance
here is that we spent a significant effort, like I
mentioned earlier, trying to understand the risk
significance and the documentation that DOE had
provided before our technical exchange. We were kind
of like the middle of the line, essentially, so we
spent a particular amount of effort to try to get
things as risk informed as possible.
Other things that we've done over the last
year, we've collected and analyzed some actual samples
from the Early Warning Drilling Project program. One
of the problems with data collection and
characterization is that there's a significant span of
time between collection of the sample, analysis,
reporting of the data so that it can be used in part
of people's interpretations and models.
So we have to take a very proactive role
collecting our data, assembling the data that has been
collected and may not be readily available. For
instance, one of the things that we did is we took all
of the Nye County data, the data that Nye County had
collected as part of their program, assembled it and
put it into this document so that it could be used by
CNWRA and NRC staff as part of their work.
CHAIRMAN HORNBERGER: Paul, but you also
actually ran some samples? What's the purpose?
Cross-lab confirmation testing or are you measuring
something else?
MR. BERTETTI: Well, two things is that I
can get results in a very short time period so that I
can actually start working on an analysis of water
chemistry from our standpoint. The other thing is
that we can try to confirm whether or not the results
are meaningful. Remember because these things are
ongoing, QA of samples is very important. We have a
way of -- it's kind of like an oversight and
confirmation process.
The other things that we're doing that I
think are pretty important are that we've tried to
apply some of the process level models, especially
using a surface complexation approach to evaluate an
alternative mechanism for modeling radionuclide
transport in the system. I think we've been pretty
successful at that, and to that end we've conducted
focus experimental studies to build up our own
database for that and to help us provide a robust
model. And we're actually -- the next revision of the
TPA code should incorporate, in a limited basis, the
ability to use results from our surface complexation
modeling approach in our TPA code. And we'll kind of
do that on look-up table basis, at least that's the
indications so far. And some of what we've done there
is laboratory studies of neptunium on calcite. I'll
talk about that in a minute.
And a good example is some analysis of
technetium uptake on clinoptilolite, which is kind of
a material that can substitute for alluvium. And the
fact was at one point DOE had claimed some minor
credit for technetium sorption. Our results indicated
that that was probably in the experimental noise, and
DOE, on further analysis, acknowledged that, yes, that
was essentially experimental noise.
Okay. So here are some examples of
information that we can gather from our own analysis
and collection of materials. This, for instance, is
a sample of alluvium collected from one of the Early
Warning Drilling Project wells, showing that there's
maybe some fine-grain codings on some of these grains.
And some recent studies have indicated that maybe with
these kind of codings they actually drive the sorption
drive process, and they might be actually fairly
generic and that they actually may be very much like
clay. And it turns out that recent DOE results also
indicate that the sorption of alluvium is kind of
driven by the presence of clay above maybe anything
else. And you can also start to group well water
chemistries, not only with respect to depth in the
saturated zone but spatially as well. So that's some
important things that we can do, and then we can be
prepared for the next step of characterization and
modeling.
VICE CHAIRMAN WYMER: What's the size of
that grain, Paul?
MR. BERTETTI: I'm sorry, this slide right
here is about two millimeters across.
VICE CHAIRMAN WYMER: Oh, it's that big?
Okay.
MR. BERTETTI: Yes, sir. So here's an
example of the experimental results for neptunium
uptake on calcite. This is just the distribution
coefficient, or KD, for neptunium and plotted against
the pH or the solution. What I don't want to say is
neptunium sorption on calcite is not necessarily the
most important aspect of transport in the alluvium,
but what's important about this is that DOE
assumptions of transport of the alluvium for the TSPA
VA were based on these limited results with water from
UE25-p#1, which is that deep carbonate aquifer well.
And, in fact, the processes that control this sorption
and the magnitude of sorption are not really related
to the process controlling sorption in the alluvium.
In fact, now DOE has conducted some data with alluvium
samples, but we're still limited to a very narrow
range of pH. And so I think we have to be careful
about potential invalid assumptions regarding
mechanism and the magnitude of sorption.
And the reason that's important is that we
might be able to get the magnitude right, and that
might be okay. The problem is the real uncertainty
and variability, both in chemistry, and the kinds of
variability that we want to add to our modeling to
incorporate real levels of uncertainty and realism we
should be able to do a better job at the experimental
part.
MEMBER GARRICK: Speaking of variability,
and I realize you're outside the near-field, how are
you accounting for the source term -- the variations
in the source term, because that's certainly going to
effect the radionuclide transport, is it not?
MR. BERTETTI: Variations in what respect?
MEMBER GARRICK: Well --
MR. BERTETTI: If you mean chemical
variations in terms of a plume, those are not
accounted for.
MEMBER GARRICK: Okay.
MR. BERTETTI: And they're not accounted
for in either model.
MEMBER GARRICK: So it's kind of
idealistic. It's not really very representative of --
MR. BERTETTI: That's absolutely correct.
And one advantage that a surface complexation model
approach rather than the -- rather than providing a
distribution to sample from, a distribution of KDs
from which to sample, that are geared at one pH, using
a distribution of KDs that have a range of real system
chemistry, like CO2 and pH and other factors, well,
that at least you can evaluate the uncertainty and
make an argument of whether or not your plume may have
a distinct chemistry from the surrounding environment
or whether dilution will essentially make it into a
background sort of transport system, in which case you
might have a pretty realistic approach to the
uncertainty.
MEMBER GARRICK: Yes.
MR. BERTETTI: By picking a KD model
that's based on one pH and a distribution that's based
on some experiments that don't have a really good
variability in chemistry, no, I don't think that's a
very effective way. But that's my opinion.
VICE CHAIRMAN WYMER: I'm disappointed you
haven't said anything about valence.
MR. BERTETTI: Well, you know, I'd like to
do that. I mean to me that's -- the redox state of
the system is very important, and that goes to the
characterization. Some people might make an argument
that specific mineral surfaces might provide a micro
redox environment that really would enhance sorption.
The problem is if you have four wells over ten
kilometers, to try to characterize that, you're not
going to be able to make a defensible argument to
support that.
The other thing I would add is that, Neil,
I've added these little dots between the red and the
green, so those of you who are red/green challenged
should be able to see that graph.
MEMBER LEVENSON: Before you take that off
--
MR. BERTETTI: Yes, sir.
MEMBER LEVENSON: -- I'm intrigued by,
looking at your red circle, the influence of time --
seven days, 14 days, 21, 31 days -- because all of
those are infinitesimally short in the time constant
of what we're looking at.
CHAIRMAN HORNBERGER: It keeps going up
for 10,000 years, you think?
(Laughter.)
MEMBER LEVENSON: Well, the question is
how valid are the yellow ones if they were instantly
measured?
MR. BERTETTI: Well, the factor is that
these, over the same time period of the red dots, stay
constant. I didn't mention that specifically, and
that's in a difference in the mechanism of sorption.
In this system, we start out with the water that's
essentially saturated with respect to calcite, so you
have a surface reorganization of the calcite and some
precipitation that goes on, and that kind of enhances
the uptake of neptunium. And we've kind of confirmed
that with our own co-precipitation experiments.
Whereas, if you have a system that's
undersaturated with respect to calcite, like the
alluvium -- saturated alluvium is now, you have
essentially kind of this stable sorption --
equilibrium sorption curve that you get with silicate
minerals and other minerals too.
So I mean it's a big -- it's not
insignificant that we have a similar sort of sorption
curve shape with a carbonate mineral that we do with
aluminum silicate. So the problem is over a long
period of time we could have a significant change in
chemistry and mechanism of uptake and processes, and,
frankly, that's not incorporated. Temporal changes
are not incorporated.
So what are we planning to do? Well,
obviously, we'd like to keep up with DOE documents and
the products that they contribute over the next year.
We'd like to continue to prioritize our modeling and
analysis so that we're looking at hopefully the right
issues and the risk prioritized features. To that
end, what we're going to do is we're going to continue
some focused experimental studies, we're also going to
look at our field characterization activities and
evaluations.
One thing I would mention is DOE has a
major effort and going down to the natural analog site
at Pennyblanca, and we're going to participate in that
in terms of observation and maybe some sampling and
confirmation activity as well. We'll continue with
these neptunium results to include some modeling and
hopefully build a more robust database for ourselves.
We're doing some sensitivity analysis and
colloidal transport modeling on our own to look at the
kinetics of that system, and maybe we find that it
might be actually driven by kinetic process. I think
at the recent MRS meeting some German investigators
looked at neptunium sorption on human colloids and
found that it was very much kinetically controlled.
And we'll also report some additional results on some
detailed modeling of processes.
In summary, we've just looked at risk
informed. I think we do a good job providing some
independent technical investigation, and we're trying
to maintain our risk-informed approach to the work
that we do. And that's all I have.
CHAIRMAN HORNBERGER: Thank you, Paul.
Milt?
MEMBER LEVENSON: I asked them along the
way.
CHAIRMAN HORNBERGER: Raymond?
VICE CHAIRMAN WYMER: I had a couple,
Paul.
MR. BERTETTI: Yes, sir.
VICE CHAIRMAN WYMER: One, with respect to
the KDs, you said that a lot of it was expert
elicitation of derived -- that's where DOE got a lot
of results. There weren't many really experimental.
MR. BERTETTI: Correct, especially for the
volcanic top, porous material.
VICE CHAIRMAN WYMER: Then you said that
DOE's going to tell you how they got their results.
Does that mean they're going to tell you how they ran
their expert elicitation or does that mean they're
going to go out and get more results?
MR. BERTETTI: Well, I would characterize
it like an expert judgment, and I think the idea is
they have expressed a desire to provide documentation
information about how those expert judgments were
derived.
VICE CHAIRMAN WYMER: So that's what it
meant.
MR. BERTETTI: And if they cannot do that,
then that could become an issue. So that issue is
probably minor unless that documentation doesn't
exist, and then it might be a major effort for them to
provide the adequate documentation.
VICE CHAIRMAN WYMER: Documentation on how
you do an expert elicitation isn't nearly as
convincing as documentation on experimental --
MR. BERTETTI: That may be correct.
VICE CHAIRMAN WYMER: Well, that's one
point. You know, I'm sure, and I know you know, that
NRC has --
MR. BERTETTI: Well, I wouldn't be so
sure.
VICE CHAIRMAN WYMER: You'll know you know
in a minute.
(Laughter.)
NRC has a very nice research program on
mechanisms of radionuclide transport going underway.
You were at the working group meeting where they
discussed that. I gather from what you said in the
course along the way that it is the result -- those
results are those kinds of results that are going to
be used to check the results obtained by KD values.
Did you not imply that in one of your viewgraphs?
MR. BERTETTI: Well, I think what we'd
like to do is for nuclides like neptunium that might
have a small but variable retardation coefficient is
that to actually incorporate our surface complexation
modeling results and put them into the TPA code, and
then have the TPA code sampled the natural variable
parameters like pH and CO2 to produce an output and
then compare that to the sampling of KD in which you
have a distribution of KD that's not dependent on the
real -- the advantage of that is we can go out in the
field and measure pH and pCO2 and determine a real
system variability, which is something that we can
measure, and it's kind of hard to measure the
probability distribution and function of KDs
independent of that, which is what is essentially done
now.
VICE CHAIRMAN WYMER: And, finally, with
respect to protactinium, it probably, in the entire
periodic table, is the element most prone to
hydrolysis. I've worked some with it; it's terrible,
it's a terrible material. So if you're going to
discuss protactinium and its transport, you're
probably talking about a colloid other than any sort
of ionic format.
MR. BERTETTI: I would agree with that
entirely. Every time we do a neptunium experiment
with neptunium 237, we essentially do a protactinium
experiment as well, because the immediate daughter is
protactinium 233. And so we kind of get a little bit
of information about how protactinium is behaving in
the system, and I would say that my very preliminary
indications are that protactinium is sorbed
significantly, and that's very consistent with our
studies of actinides to show that when they start to
hydrologize that's when you start to see a significant
sorption.
VICE CHAIRMAN WYMER: Right.
MR. BERTETTI: So except for competing
anions and other complexes, I would expect
protactinium to have a significant sorption potential.
MR. BRADBURY: Yes, this is John Bradbury,
NRC. With regard to your questions concerning expert
judgment, first of all, it was our -- DOE and their
labs have collected a lot of sorption data for man
years. I was our understanding that the use of the
expert judgment was mainly with regard to establishing
distributions to be applied in performance assessment,
and so we were looking for the explanation concerning
how the distributions were --
VICE CHAIRMAN WYMER: I guess I don't
really understand what you just said, because
distribution is the whole ballgame on KDs.
CHAIRMAN HORNBERGER: Probability.
VICE CHAIRMAN WYMER: Probability
distribution.
MR. BRADBURY: Probability distribution as
opposed -- since there's a limited number of
experiments that are done and how are those
represented out spatially and temporally?
VICE CHAIRMAN WYMER: Oh, okay. I
understand now. All right. Thanks.
MR. BRADBURY: So that was one thing I
wanted to --
MR. BERTETTI: I would agree with that,
and I'd also temper that, that there are some nuclides
that are not -- weren't well studied. I mean there's
a little bit of minor nuclides that may not be
important that didn't have a lot of experimental basis
behind them.
MR. BRADBURY: Yes. Can I expand on that
a little bit? The three radionuclides that have been
mentioned, the plutonium, uranium and protactinium, I
believe in our earlier issue resolution status report
and the agreements that we came up -- we established
in those, there are certain assumptions that have to
be shown to be valid for KDs to apply to
radionuclides. And for plutonium, uranium and
protactinium, there were certain aspects of those
experiments that created problems, and so they are
going to go -- for example, with regard to plutonium,
one of the assumptions should be that the sorption
process should be fast, and there were indications
that their sorption experiments that the KDs were
changing with time, that kind of situation.
CHAIRMAN HORNBERGER: John?
MEMBER GARRICK: I'm not a chemist, as
you'll find out in a minute, or a geochemist or
something, but I guess I'm wondering why this is a
KTI. There's no question --
MR. BERTETTI: DOE has mentioned that as
well. DOE had a suggestion that the radionuclide
transport KTI would be rolled into one large KTI that
would essentially be called flow and transport,
primarily because that's how they have organized their
system. I think it's important -- why is it a KTI?
I think there a lot of aspects of radionuclide
transport that take a lot of resources to address the
--
MEMBER GARRICK: Well, don't get me wrong.
I think radionuclide transport is important. What I'm
confused by a little bit is the decoupling of the work
that's going on from what I would call a
scientifically based source term that has tremendous
impact on the radionuclide transport. And the absence
of temporal effects makes me wonder, this program, as
it's outlined, what its relevance is, because things
are really going to be very different in a couple
process sense.
MR. BERTETTI: Well, I'm not sure I can
answer that. What I would say is I think that there
are temporal changes in terms of the magnitude and the
value of the source term. I don't think there are
temporal changes applied with respect to chemistry,
and I think that's one of the things that was a
comment from our TPA review, the peer review of our
TPA code. We tried to address those.
I think there is a -- personally, I think
that there is a disconnect between whether or not you
try to envision transport within some sort of plume
that has an isolated geochemistry or whether or not
you have mixing and how temporal effects on flow
change chemistry. So I really don't know how to
answer that. I would say that I kind of agree that
there seems to be a disconnect there, and I really
don't have another answer other than that. Bret is
eager to pipe in.
MR. LESLIE: Partially, I think Paul --
this is Bret Leslie from the NRC staff -- Paul was
careful when he said where RT kicks in. Beyond the
point of coupled processes.
MEMBER GARRICK: Right.
MR. LESLIE: So we're talking far field
now.
MEMBER GARRICK: Yes. He did say that.
MR. LESLIE: And I think that's part of
the answer to your question is RT is looking at the
ambient conditions in the far field.
MEMBER GARRICK: Yes, and therefore my
reaction is so what?
VICE CHAIRMAN WYMER: The implication is
that even if you have a coupled effect and you produce
something different from what you're going to find in
the far field, that like a valence change, that that
will then occur as you move away from the near-field,
that you'll get back to the species that you're
looking at. That's the implication of what you're
saying, I believe.
MR. LESLIE: Well, I know. The other
thing is, again, we had that flow-down diagram, and
you have to answer the complete system. We have to
represent the natural system, and radionuclide
retardation is part of that natural system.
MEMBER GARRICK: Well, I agree with that.
I agree with all of that, that radionuclide
retardation is something that probably isn't accounted
for to the extent that it should be. And all I'm
suggesting is that any time -- I've had lots of
experience with source terms of a different type
having to do with nuclear power plants and what have
you, and we couldn't get anywhere until we had done a
pretty good job of defining the source term, because
you have no idea of what kind of release dates you're
dealing with. And until you can define the release
dates, you have no real good technical basis for
calculating any off-site consequences, et cetera.
And I think there's a similarity here,
even though the mediums are very different, and I even
know in WIPP that there was a tremendous amount of
effort, especially late in the program, in trying to
define the source term. And I suspect we're going to
hear about that when we hear about the near-field
work, but I do see a real disconnect in what we've
heard this afternoon with respect to radionuclide
transport.
MR. LESLIE: This is Bret Leslie again
from the NRC staff. I think Gustavo actually covered
it but in a minor sense because of the focus of the
container life and source term is both container life
and source term. That source term does define what
those downstream releases are and to the effect that
the near-field chemistry also influences that, but he
didn't dwell on it that much. And, again, in terms of
risk information, it's the container that is where
most of the insights derive.
MEMBER GARRICK: Yes. Yes, I understand
that. Okay. Thank you.
CHAIRMAN HORNBERGER: Just to make sure
that I'm clear on this, I want to make sure that it's
not a disconnect in a bad sense. I mean John is
saying it's a disconnect, but it's only a true
disconnect if in fact you're working off on the source
term here and you're working on radionuclide transport
completely independently of that, and I don't think
that's what you're doing, is it?
MR. BERTETTI: No. I guess what I might
add is that if you make a reasonable assumption that
there's a significant amount of mixing in the
saturated zone with respect to kind of equalizing a
distinct chemical signature that the source term would
have, then our approach is I think very sound. And it
is coupled in that nature, because we can account for
the chemical variability in the natural system outside
of that altered area of the near-field.
The problem might lie in that if you have
a very distinct near-field chemistry that perpetuates
itself in the natural system, then our current
characterization of the natural system may not
adequately represent the transport characteristics
from that. And results from our near-field studies
and modeling and from the source term should help us
identify whether or not the magnitude and the volume
of material is released essentially would have that
kind of characteristic.
So it's something that needs to be
evaluated as we learn more about how the near-field
chemistry responds, and right now that's one of the
large areas of uncertainty is what's happening in the
waste package in and around the waste package
chemistry? And I don't want to step on what Bobby
might say, but that's a high degree of uncertainty.
And so until we get a better handle on what that
chemistry is and how that responds as we have some
infiltration, then I think it's very tough to couple
anything.
CHAIRMAN HORNBERGER: I guess, just to
pursue this, not too much farther, but it strikes me
that if you have massive changes propagating
themselves, what you're basically going to have to
conclude is that the near-field extends all the way to
the saturated zone. I mean, to me, the far field, by
definition, you're into sort of trace amounts.
MR. BERTETTI: Right, right, right. I
just want to say that you could have a trace amount of
chemistry, and I don't know how that would effect the
system overall. I would agree with you on that, yes.
CHAIRMAN HORNBERGER: Okay. Milt?
MEMBER LEVENSON: I think, George, to some
extent, all of the KTIs are really independent. You
take the pieces and that's what the modeling and the
code -- just like the corrosion didn't discuss that
it's probably irrelevant if there's no water, from the
standpoint each KTI, to some extent, is completely
independent. They have to depend on the other
operations to put them together.
CHAIRMAN HORNBERGER: Right. And that's
what I -- in my introductory remarks, I pointed out
that one of the things we were interested in is how
the KTIs link together to make sure that in fact they
are being integrated. Anything else? Staff? Thanks,
Paul.
MR. BERTETTI: Sure. I'm not sure if I
would have wanted to be skipped or --
(Laughter.)
CHAIRMAN HORNBERGER: That was a -- well,
I guess at the GSA meeting the year before last, I was
chairing a session and did exactly the same thing. I
was going down the list and actually introduced a
speaker out of order, and I was accused by the speaker
I skipped of having a senior moment. So that's okay
if you do that. And, Bobby, I'll let you introduce
yourself, as I have for the other people, for the
record.
MR. PABALAN: My name is Roberto, or
Bobby, Pabalan. I have listed in my first viewgraph
the people who have contributed to this KTI of
evolution of the near-field environment, or ENFE. And
if you have any questions, please feel free to ask
them.
(Laughter.)
And I will also try to answer your
questions. The ENFE KTI -- this is the outline for my
presentation, which is pretty much the same as the
previous ones, so I'm going to skip this. The ENFE
subissues, there are five of them, each pertaining to
the effects of coupled processes. The first one
pertains specifically to the drift seepage and flow.
The second one pertains to the waste package chemical
environment. The third one is on the chemical
environment for radionuclide release. And the fourth
one is on radionuclide transport in the near-field.
And the fifth one is on the effects of coupled THC
processes on potential nuclear criticality in the
near-field.
At the start of fiscal year 2001, four of
them, the first four were open, the fifth one was
closed-pending as of the start of the fiscal year. At
the end of fiscal year 2001, all of these are closed-
pending as a result of the NRC/DOE technical exchange
that was held sometime in January of last year.
Now, we wanted to show what DOE needs to
provide before license application. Well, the most
important thing is they need to decide on the
repository design and the thermal operating mode. As
we all know, the materials -- the repository design,
the materials that go into the repository and the
temperature conditions under which these materials are
exposed to are the main drivers for the coupled THC
processes.
So, in essence, the DOE needs to update
the evolution of the near-field environment process
models and the TSPA model abstractions to be
consistent with the selected design and the thermal
operating mode, whether they go to high temperature or
the low temperature operating mode.
The analyses that we have conducted in the
past year are based on AMRs and PMRs that basically
relied on the high temperature operating conditions.
So if there's any change in the design or in the
operating mode, then we need to review any new
information or changed information that will come in.
The DOE also needs to provide additional
information as a result of the agreements we had at
the technical exchange. There are 41 agreements as a
result of the technical exchange. That's a long
laundry list. Fourteen of those have been received,
and we expect to receive 27 sometime in this fiscal
year. Although it's a long list, many of them are
relatively straightforward. For example, it ranges
from providing sufficient technical basis for some of
the FEPs that have been excluded from the abstraction
all the way to more complex requirements dealing with
data uncertainty, model uncertainty and model
validation.
We don't expect these agreements to
present a problem in terms of fulfilling the
agreements prior to the license application. I think
many of them are straightforward. I wouldn't be
surprised, though, that some of them may actually be
completed in the post-LA period, specifically with
respect to model validation and also those pertaining
to the analysis of the model uncertainties, the
implementation of those uncertainties and the
propagation of these uncertainties in the TSPA
calculations, as well as in the analysis of the
uncertainties in the data supporting these model
calculations.
I'm going to give just a few examples of
the more important agreements for each of the
subissues. For Subissue 1, which pertains to the
coupled THC processes on drift seepage and flow, the
DOE needs to address the various sources of model and
data uncertainty in its THC abstraction. One of the
agreements that we came to at the tech exchange was
that DOE needs to provide physical evidence that
supports the model of matrix/fracture interactions,
specifically the facts on precipitation processes.
The DOE is supposed to provide data on the post-test
overcoring. I believe this pertains to the single
heater test. And also they need to provide the
results of ongoing sidewall sampling for the drift
scale heater test. Some of the information I believe
has been provided. Most of these results are still to
be expected for this fiscal year 2002.
For Subissue 2, which pertains to the
waste package and drip shield chemical environment,
Gustavo pointed out of the need for the DOE to provide
a good handle on the quantity and the chemistry of
water contacting the drip shield and the waste
package. Because the chemistry and the quantity
determines the performance of the drip shield and the
waste package materials. As you all know, the
performance of the waste package and drip shield is a
key safety attribute of the DOE safety case. The DOE
needs to place bounds on the concentrations of the
minor and trace elements which are important to
performance, for example, fluoride in the case of the
titanium alloy drip shield. The DOE needs to evaluate
the effects of evaporation or salt deposition or the
interaction with engineered materials on the chemistry
of the water contacting the drip shield and the waste
package surfaces.
In addition, for Subissue 2, DOE needs to
document the data, including the uncertainties, used
to calibrate the models or to support model
predictions. And they also need to propagate the data
and the model uncertainties through the TSPA
calculations.
With respect to Subissue 3, which is the
effects of coupled THC processes in the chemical
environment for radionuclide release, again, as
Gustavo pointed out and that Leslie alluded to, the
DOE needs to reduce and/or quantify the uncertainties
in the chemistry of water inside the waste package,
because it affects directly the degradation of the
cladding as well as of the waste forms. In addition,
our review of the DOE reports indicate that the DOE
colloid concentration model is extremely sensitive to
the in-package chemistry. The DOE also needs to
provide analysis to verify that the bulk-scale
processes dominate the in-package chemical
environment, which is the main assumption in the DOE's
in-package abstraction.
In addition, again, they need to document
the data, including uncertainties used the support the
in-package chemistry calculations as well as provide
stronger technical basis for the radionuclide and
waste form types selected for the colloid release
models.
For Subissue 4, pertaining to radionuclide
transport in the near-field, I think the things that
DOE needs to do are relatively straightforward. They
just need to provide the technical basis for screening
out coupled THC effects on the radionuclide transport
by diffusion of colloids. They also need to
demonstrate the suitability of the colloid models and
parameters for conditions in a perturbed near-field
environment, because there are abstractions basically
based on conditions under -- basically based on
ambient condition assumptions. If DOE also implements
retardation in the waste package or in the engineered
barrier system, as appears to be the case in the SSPA,
then they will need to provide the technical basis for
the transport parameters they use for the in-package
-- for the waste package transport and EBS transport.
For Subissue 5, on the potential
criticality in the near-field, all they need to do is
they need to close the open items remaining in the NRC
safety evaluation report with respect to the disposal
criticality analysis methodology topical report. And
they also need to revise the FEPs screening arguments
concerning criticality. So those are fairly
straightforward.
So what does the NRC and Center staff need
to do before LA? Obviously, we need to keep on top of
the DOE progress with respect to fulfilling the
technical exchange agreements. We need to review the
DOE data collection and model activities used to
support its process and abstracted models. We need to
continue independent evaluation of issues through a
focused and experimental modeling program. I'm going
to talk about some of those activities in the next few
viewgraphs. We need to maintain in-depth familiarity
with DOE methods and assumptions, models and model
abstractions, and we also need to be prepared to
respond to changes in the DOE safety strategy and the
repository design or in the process abstracted models.
And we also need to validate our own codes that we use
to review the DOE safety case.
For FY 2001, we have basically spent most
of our time conducting an intensive and critical
review of DOE documents related to a near-field --
various reports, technical basis documents, SER, SSPA
and AMRs and PMRs. We conveyed -- we contributed to
NRC reviews of these documents and helped convey NRC
concerns to the DOE staff in various technical
exchanges and in numerous pre-tech exchange
teleconferences, and also in addition to the
teleconference and sufficiency review. We helped
identify risk-formed paths through resolution of NRC
concerns, resulting in the status of the first four
ENFE subissues being changed from open to closed-
pending. We also documented our review and the status
of subissue resolution in the integrated IRSR report
that's going to come out sometime this year.
MULTIFLO is a coupled THC model that's
being used by NRC and Center staff for various KTI
activities, including near-fields, thermal effects on
flow, USFIC and TSPAI KTIs. For the past fiscal year,
development of version 2.0 was initiated. Version 1.5
was released, which includes new features, including
free-drainage boundary condition and improved
description of the fracture-to-matrix flow processes.
We also developed a graphical user interface for
MULTIFLO and a workshop training was held here at
Whiteflint for NRC staff.
In addition to the MULTIFLO work, we
conducted focused application of process-level models
to test DOE hypotheses pertaining to the deliquescence
points of salt mixtures that can deposit on the drip
shield and waste package surfaces. We also looked at
the chemical evolution of our Yucca Mountain
groundwaters that can result due to evaporation. We
also conducted focused experimental studies to study
deliquescence points of mixed salt systems, as well as
studies relating to uranophane solubility.
This last bullet on the uranophane
solubility is a preliminary step towards conducting
neptunium core precipitation experiments. This is one
of those activities that we stopped two years ago,
because in a TSPA-SR report it was apparent that the
DOE would not claim credit for secondary phase
precipitation.
It is now evident from the SSPA reports
that they might actually take credit for such
processes. So, again, doing a risk-informed type of
approach, we determined it was important to reinitiate
or redo -- start again the uranophane solubility
experiments and the follow-on work, the neptunium core
precipitation work.
The next two viewgraphs basically show the
results of those process modeling work. This
particular study was designed to test the DOE
hypothesis. The DOE, in its abstraction of the
chemistry of water on the waste package and drip
shield surface, assumed that aqueous corrosion begins
when the relative humidity reaches the deliquescence
relative humidity of pure salt. In the TSPA-SR
report, they assumed -- they used the sodium nitrate
deliquescence humidity as a lower bound for the
deliquescence point of salts that could form in the
waste package drip shield surface.
Our thermodynamic calculations are shown
in these figures. These calculations were done for a
temperature of 90 degrees centigrade, a temperature
which can be sustained for about 1,000 years or more
based on RTEF calculations. These systems are for
brine or mixtures of two salts. The Figure A is for
NaCL plus KCL; the second one is for NaCL plus
magnesium chloride, and the third one is for KCL and
magnesium chloride. The solid curves are the
calculated deliquescence humidities for the mixed
solids. The dashed line is plotted as a reference.
It gives the deliquescence humidity for pure sodium
nitrate at 90 degrees centigrade.
Basically, the bottom line of the
calculations suggest that, okay, if you have sodium
chloride plus potassium chloride mixture, then it's
okay to use a pure sodium nitrate deliquescence
humidity as your bounding point for the deliquescence
point of the mixture. On the other hand, if you have
a magnesium chloride mixture with NaCL or KCL, and
also in essence with calcium chloride, which is
another salt that has very low deliquescence humidity,
then it's not appropriate to use a pure sodium nitrate
salt as a lower bound for the deliquescence part
because of the salt mixture.
In the SSBA, it indicates that DOE is also
looking at using the magnesium chloride and calcium
chloride properties as an estimate for the
deliquescence point for the salt mixtures. So I think
there is improvement in the DOE approach that's
evident from the more recent information.
So the question then is, okay, what kind
of salts can form on the waste package and drip shield
surface? One mechanism by which we can form these
salts is by evaporation of Yucca Mountain pore waters
that drip into the drift environment. What we've done
here is we've taken a few chemical compositions
published by Yang et al. for Yucca Mountain pore
waters. Basically, they're plotted up in the first
what we call a pie per diagram, for the initial
composition and the final compositions are plotted in
the second pie per diagram.
The initial compositions really range in
terms of the cationic composition -- you know, it has
a very well distributed range for the calcium and
sodium, potassium, magnesium chloride concentrations.
And it also varies with respect to sulfate chloride
and bicarbonate carbonate. When you expose these
Yucca Mountain pore waters to evaporated
concentration, what you form are basically two types
of brines.
Some of the initial Yucca Mountain pore
waters evolve into what can be called a calcium,
magnesium, sodium chloride brine. These brines, if
you take it to full evaporation, will form salt
deposits with very low deliquescence points, basically
because of calcium and magnesium. But what is also
important is that you form very low concentrations of
the corrosion-inhibiting species.
The nitrate -- this is the free nitrate
concentration accounting for the -- this is the
uncomplexed nitrate concentration. It's less than
0.01 molar. The total nitrate concentration in
solution can be very large, but the free nitrate
concentration is constrained by the formation of
calcium nitrate ion pair complexes in solution.
Sulfate is constrained by precipitation of calcium
sulfate during the evaporation process. On the other
hand, you can have Yucca Mountain pore waters evolve
into potassium, sodium chloride nitrate brines.
Now, the deliquescence points of salt
mixtures that can form from these brines will have
relatively higher deliquescence points than those that
form from the calcium magnesium brines. But on the
other hand, they're still going to be lower than those
of the pure sodium nitrate salt. These salts have
much higher free nitrate concentrations, hence you can
have possible waste package corrosion inhibitor. On
the other hand, where you have some amount of fluoride
initially in the solution, upon evaporation, in this
case I think it evaporated to about 15,000 times
concentration factor, you can have very high
concentrations of fluoride. In one particular case,
about 0.014 molar, which is much greater than the
0.001 molar threshold for accelerated general
corrosion of titanium alloy, which is observed in some
of the Center experiments.
For FY 2002, our plan is basically to
review the DOE documents relevant to near-field
subissues. We will continue those process-level
modeling work that may have been conducted. Those are
supported by some experimental studies because there's
really no experimental data on the deliquescence point
of salt mixtures at elevated temperatures. I have a
backup slide, I believe Number 2, that demonstrates
some of those results. We will continue to develop
MULTIFLO version 2 with enhanced capability to
simulate coupled THC processes. MULTIFLO will be used
to predict the quantity and chemistry of seepage water
and to help quantify uncertainties associated at the
process level and PA simulations for complex coupled
processes. We plan to validate these process-level
models that we have used in reviewing the DOE
analysis, and we will provide input to the integrated
IRSR rev 1.
So in summary, for the past fiscal year,
like I said, we've spent quite a bit of work in
conducting a critical evaluation of the DOE analysis
relevant to near-field subissues. We have interacted
with the DOE to resolve NRC concerns, resulting in all
ENFE subissues becoming closed, pending confirmation.
We continue to develop MULTIFLO, and we've conducted
some process-level modeling and experimental studies
which are designed to probe DOE assumptions regarding
processes judged to be the most risk-significant.
That's all I have, and I'll take your questions.
CHAIRMAN HORNBERGER: Thanks, Bobby. Ray,
you want to start?
VICE CHAIRMAN WYMER: Yes, I'll start. I
think you'd probably agree that the chemistry of the
near-field and especially the chemistry in the package
is extremely complex and especially with regard to
potential solid phase formation. And I just wondered
if the thermodynamic database available for your
modeling is anywhere near adequate to deal with all
the potential phases that could form and tie up some
of these things?
MR. PABALAN: With respect to the pure
phases, there might be enough information to allow us
to do the evaluation. But there are certainly more
complex solid solutions and things like that where we
don't have any information. Can we use better
thermodynamic data? Certainly.
VICE CHAIRMAN WYMER: But you feel what
you have is adequate to the task.
MR. PABALAN: It will allow us to evaluate
trends in the evolution and the chemistry of not only
water outside the waste package but also inside the
waste package.
VICE CHAIRMAN WYMER: Okay. The second
question I had is I was -- just kind of an
observation. I was a little bit surprised to hear you
say that you thought that the calcium nitrate ion
pairs are sufficiently strong that they would
significantly reduce the nitrate ion -- free nitrate
ion concentration.
MR. PABALAN: Yes. You're not the first
one to ask me that, and I've done a literature search
on calcium nitrate. There are a couple of what we
consider multiple lines of evidence. There are some
indirect measurements of water activities or osmotic
coefficients and some more direct, like x-ray
diffraction analysis of calcium nitrates, which
suggest that you do form strong calcium nitrate,
especially at the higher concentrations which is of
interest to us. That's a review that I'm still doing.
I also have some random spectroscopy
literature that I just got precisely to address the
potential for -- whether it's true, whether calcium
nitrate does form strong complexes. But the
indications are, yes, it does. I don't know if it's
greater to the extent that is suggested by the
thermodynamic modeling. That's something that we will
pursue in the next few months.
VICE CHAIRMAN WYMER: Yes, I'm skeptical.
Okay.
CHAIRMAN HORNBERGER: Milt?
MEMBER LEVENSON: I've got, I guess, two
questions, and it's more maybe to determine whether
they're included in your scope or not included in your
scope. In the near-field, you've discussed the
chemistry. Where are the mechanical assumptions
reviewed? And by that I mean things like the
corrosion of the waste container is going to occur
here and not there. The thing at one of the modelings
that we came across said the minute you penetrate and
the container is no longer vacuum-tight, you make the
assumption that 50 percent of the surface of the
container is gone. There's a lot of mechanical
assumptions in the near-field that are in the
modeling. Is the review of those things under your
area or is that somewhere else?
MR. PABALAN: We help review the DOE
approach to these kinds of calculations. We don't do
independent analysis, whether experimental or
modeling, with respect to that, although we provide
some input in the TSPAI type of analysis, which kind
of puts these things together. We're more focused on
the chemistry aspect versus the mechanical type of
information.
MR. LESLIE: This is Bret Leslie from the
staff. If Bobby moves a little bit that way, it's
important to remember that when we developed our
sufficiency comments, we took -- we may have five KTIs
reviewing the same AMR, and they're reviewing it under
their KTI, but we're providing integrated comments.
So the answer to your question is, no, the mechanical
things aren't necessarily reviewed by the near-field,
but they're either reviewed by the container life and
source term or the repository design and thermal
mechanical effects.
MEMBER LEVENSON: My question was were
they included here and the answer is no, right?
MR. LESLIE: That's correct.
MEMBER LEVENSON: Okay. The second
question is a similar one, whether it's included here
or not, and that is the assumptions in the modeling in
the near-field of how water behaves. In some of the
models, there's kind of strange behavior for water in
that water vapor moves independent of -- pressure
moves only on a gradient of temperature, so you don't
lose any water from the rock into the drift. It all
moves away. Now, there are similar things that are
assumed assumptions without evidence in the modeling
but in the near-field. Is that also somewhere else?
MR. PABALAN: Subissue I pertaining to
seepage and flow addresses specifically that issue,
but it is not -- well, it is but not in the -- it's
not considered in the modeling activity that I
discussed with respect to the --
MEMBER LEVENSON: Somewhere else.
MR. PABALAN: Yes, it's done somewhere
else.
CHAIRMAN HORNBERGER: John?
MEMBER GARRICK: Well, just a comment. By
the number of agreements, one might get the impression
that as far as the near-field goes, DOE did very
little right. And yet I've always had the impression
that this is where a lot of effort was given. Is it
possible that this is a product of what one might call
the lamp post syndrome, because there was quite a bit
of work done there is an opportunity for searching for
more details? I'm just struck by the amount of -- the
number of agreements that are involved here. It's
just --
MR. PABALAN: I would disagree with your
characterization that the DOE did very little. In
fact, it could be the lamp post effect. It's evident
since the publication of the TSPA-VA and subsequent
reports that the DOE has significantly improved their
THC models and model abstractions for the near-field
environment.
MEMBER GARRICK: Well, maybe that's the
problem. Maybe it's the timing is what -- what's the
cutoff time for these 41 agreements, I guess is what's
--
MR. PABALAN: The tech exchange was held
January, second week of 2001. So we -- I believe we
had some teleconferences prior to that, so I suppose
we completed our reviews a month before the tech
exchange. So that was the cutoff. The 41 agreements,
as I pointed out, most of them -- many of them are
straightforward, just asking for technical basis for
excluded FEPs, give us the FEPs database, give us the
ET-36 input files. So, essentially, most of them are
relatively straightforward to fulfill.
MEMBER GARRICK: Okay. Well, the timing
answers part of it, because you had on there a lot of
information about propagation of uncertainty, and I
know there's been quite a bit of that sort of work
done in the last year.
MR. PABALAN: Yes.
MEMBER GARRICK: Yes. Okay.
CHAIRMAN HORNBERGER: Other questions?
Staff? Comments? Okay. We're going to take a 15-
minute break.
(Whereupon, the foregoing matter went off
the record at 3:08 p.m. and went back on
the record at 3:25 p.m.)
CHAIRMAN HORNBERGER: The meeting will
come to order. If I have my schedule right now, we
have three remaining presentations on KTI, and the
next one is on repository design and thermal-
mechanical effects. Don't run away. You ready? And
I'll let you introduce yourself, for the record, as
everyone else.
MR. OFOEGBU: My name is Goodluck Ofoegbu.
I'm going to present -- do the presentation on
repository design and thermal-mechanical effects.
The presentation follows the same outline
as others have followed, but some of the issues of
resolution talk about what needs to be done before
license application. There are actually -- there used
to be four subissues, but the fourth one has been
resolved, and we've essentially taken it off the list.
All of the components of Issue Number 1 is resolved,
and the components of Issue Number 2 also have been --
is closed-pending. Issue Number 3 is where I'm going
to -- well, Issue Number 2 actually is an input to the
first component of Issue Number 3. So I'm going to
concentrate my discussion on these three.
There is a component that deals with
repository design which really is part of the
preclosure that has been discussed along with other
aspects of the -- possibility aspects of the
repository, and that's why I happen to be talking
about it. And then there are two components of
Subissue 3 that led to the post-closure repository
performance that is rockfall and drift collapse and
their longtime hydrological properties. And all of
these issues -- all of these components are closed-
pending because of the technical exchange meeting --
the results of the technical exchange meeting that was
last week.
On repository design, we have agreements
that deal with essentially the inputs, the design and
analysis. The first one is about seismic topical
report 3, which is going to be the time histories of
ground -- which -- that should be applied for the
design of subsurface facilities and surface
facilities. That report is supposed to be -- they've
postponed a number of times. It's supposed to be
delivered finally in January of this year and is
probably on its way as I speak, I believe.
The properties of ground support, the only
issue here is that -- the only concern, I should say,
is that DOE wants to exclude corrosion of ground
support -- possible corrosion of ground support
material during the pre-closure period. And the
argument for that is that we don't want that to cause
cooling corrosion or obvious corrosion. All they need
to consider is dry air oxidation and to get rid of
that is very slow for -- and will not cause
appreciable corrosion for 300 years.
And we asked them to produce a technical
basis to support this position on humidity. They have
in fact submitted a report on that. We've looked at
the report, and we are in the process of getting back
to DOE on it. The one thing I can say is that we are
not convinced that a satisfactory case has been met to
exclude corrosion of ground support materials for the
200 or so years of the pre-closure period.
On rock properties, DOE approach is to use
the Yucca Mountain fracture data to obtain values of
rock mass in this list that can then be used to get
the mean values of rock mass properties that are used
in design. And that is a valid approach except that
we are not convinced that DOE has done enough to
define the uncertainties. For example, Yangs Modulus
controls the stress. It is possible on the analysis
to go from a favorable analysis result to highly
unfavorable analysis result by changing the value of
the Yangs Modulus.
So one of the things we want to know is
within which range is it -- I mean what are the range
of things of Yangs Modulus for each quality value.
The same thing applies to the rock's strength
parameters. Of particular concern here is also that
quality in the -- the practice of using quality
indices to characterize rock mass actually tries to
account for the face of fractures and not the face of
spherical discontinuities like the lithophytes which
are caused over more than 70 percent of the proposed
repository horizon.
So we asked DOE to provide an analysis of
what the approach is for dealing with lithophyzing.
And DOE has agreed to do this, along with all the
analysis that they need to do for our properties.
They are going to look at the information they have,
analyze it and then decide that that is necessary and
then compile all of this in the design analysis -- I
mean design parameter analysis report that should be
submitted sometime within this year. And we believe
that if they do all that, that the information
presented is likely to be sufficient for review.
Then the DOE intends to use rock mechanics
analysis to demonstrate that the -- all the ground
openings will be stable through the pre-closure period
to support the pre-closure and all the things that are
required. And that is, again, a good approach. It
could be supplemented with a maintenance plan, but
that hasn't been discussed.
But in the analysis there are several
things that -- several analysis approaches that we
found unsatisfactory. For example, to specify
boundary conditions, how they are specified. For
instance, in the seismic loading, they were using it
in a certain way that is based on the zero
acceleration. But the ground motion is characterized
by a time history that contains multiple frequencies.
And they argued that this approach is satisfactory,
except that we have not seen the basis on some kind of
testing or some kind of modeling that says that these
two systems of loading I approve of them. Then
conduct some sensitivity analysis and submit a report
of the analysis by I think it's 2003, so that's not
coming this year.
On post-closure, the first is rockfall and
drift collapse. Not included in performance
assessment, the point is cleaned out using a
combination of design and probability. For rockfall,
DOE wants to design -- DOE wants to select a design
basis rock impact and design the drip shields and
waste packages to withstand such rock impact, and then
argue that the probability of larger impacts will --
what's happening? I thought you could hear me.
PARTICIPANT: You didn't turn it on.
MR. OFOEGBU: Oh.
PARTICIPANT: You've been talking very
load.
MR. OFOEGBU: Oh, really? Okay. The
problem is I can get too loud now.
(Laughter.)
Okay. So they want to design the drip
shield and waste package to withstand the design basis
rock impact and then argue that larger impacts -- the
probability of larger impacts is below the regulatory
limit. And we believe that this approach is valid.
There are a number of things, the way that they define
the design basis rock impact, both the size and
frequency. We had problems with that. We had problem
with how they conduct analysis to demonstrate a drip
shield and waste packages will withstand impact. We
also have problem with the probability of larger
impacts. But the approach that DOE has selected
should lead to a resolution of these if things are
carried out satisfactorily. And there is no reason
why they cannot be carried out satisfactorily.
Where we have considerable doubt about the
DOE approach is in drift collapse. DOE wants to prove
that the placement drifts are not going to collapse
for 10,000 years. And all of the analyses that we
have done at the Center, all of the information
available from underground rock engineering experience
and in fact even advice provided by DOE's own expert
panel on emplacement drift stability lead us to
believe that the emplacement drifts are likely to
collapse soon after the cessation of maintenance.
And because of that we are not able to
anticipate that a satisfactory technical basis can be
developed to support the DOE position that emplacement
drifts will not collapse. Instead we anticipate a
satisfactory case can be made by considering the
magnitude of these potential effects of drift collapse
and maybe making an argument based on the magnitudes
of propagation of those magnitudes where necessary to
the ultimate performance measure, which is dose. And
that such an argument will leave considerable room for
resolution of the concern. But an argument that is
based on the drifts maintaining the integrity for
10,000 years, even without ground support, is going to
be very difficult to assert.
On long-term hydrological properties,
again, the main concern here is that there is -- the
thermal loading will induce rock failure in the pillar
areas, and there are analysis results that indicate
that this is possible. And such rock failure would
cause the dilation, the opening up, of the horizontal
fractures, and that may cause diversion of water from
the pillar to the drift area. DOE has agreed to look
into this and the analysis that we expect will resolve
this concern will be submitted sometime in 2003.
Well, what the analysis intends to do to
support all this? We'll review DOE documents and
continue interactions with DOE. There is one meeting
that we were going to hold sometime about October or
November last year to look at their plan for
implementing some of their agreements, but the meeting
got postponed, and we are hoping that that kind of
meeting will be held in the near future.
We also expect to conduct our own
independent analysis to support our review of the DOE
documents and to assess risk significance. We are
doing these analysis to examine -- to look at the
effects of rockfall impact on the design of drip
shields and waste packages and to look at the effects
of drift collapse on the drip shields and waste
packages, on seepage into drifts and on waste package
temperature. We will also do some scoping size --
block size distribution analysis that will evaluate
DOE's design basis rock size and the frequency of
beyond design basis rock size, if I might use that
kind of expression.
This is one of the analysis we conducted
to evaluate the previous drip shield design. At that
time, the design basis rock size was supposed to be
13,000 kilograms, and this analysis tested the design
to look at the response of the drip shield on the
design -- on the rock impact of 8,000 kilograms.
Okay. And what we found was that the drip shield --
that particular design wouldn't be able to perform its
functions on the rock impact of -- an impact by an
eight-kilogram rock. And we found that this result
actually can be modified by changing the boundary
conditions.
The analysis that the DOE did previously
met certain assumptions about it -- connections of how
the base of the drip shield is connected and all that.
And by looking at the impacts of this, we were about
to decide among us that that particular design would
not be good to satisfy the rock impact that was the
design basis at that time. But I think DOE people are
aware of these and we have been told that the
assumptions is maybe in the design or maybe in the
design basis rock, but we haven't seen where the
change will be.
Okay. We're also going to do analysis to
try to evaluate block size distributions that will
enable us to do an evaluation of DOE's design basis
rock size. We're also going to look at the assumption
of our seepage. Currently, what is happening in DOE's
performance assessment is that 95 percent of the
worker that contacts that would go through the drift
footprints, which is an ares of 5.5 meters times I
think 5.23 meters in the drift direction, that the
amount of moisture that heats that area 95 percent,
almost 95 percent of that moisture is directed around
the drift because of the capillary barrier around the
drift opening.
The capillary barrier arising from -- if
water heats an opening -- water tries to intersect an
opening around such a medium, the water is going to
flow around -- flow along the surface of the solid.
So the tendency to flow into the opening is reduced
and only about five percent of the water is actually
going through. On the other hand, if you look at this
block size, the solution comes from a simulation that
was done at the Center. This is just one realization
that could be obtained from the Yucca Mountain
fracture that -- they are all realizations, so this is
just one example.
But if you apply the same concept of water
flowing parallel to the surface of a solid, you are
going to quickly conclude that instead of moisture
diverting around, that in fact a lot of the moisture
are going to be diverted along fractures and
eventually find their way to the opening.
So the capillary barrier assumption that
is currently in the DOE TSPA model and also in the
analysis TPA needs to be modified to account for the
potential effects of drift collapse on seepage. We
are going to be looking at that. It's not going to be
easy, but we hope that it can done.
There are other potential effects of drift
collapse. This was just shows that there is a
variability of the kind of results that may be
expected, different types and degrees of drift
collapse. This one, for instance, would be at the
early stages of collapse or maybe a collapse in the
relatively stable rock that has very little fracture
control. This will be early stages in vertical
fracture control, final stages of that. And the kind
of thing we can expect from horizontal fracture
control, the collapse.
Each of these is going to effect the
mechanical load on the drip shield or waste packages.
They will also effect the temperature. In this case,
for instance, you see that heat -- there is room for
-- the only way that heat is transpired from here to
there is by radiation, but here it's going to account
as some kind of insulating effect from the rock around
it. So we think that these effects need to be
evaluated, and if it is necessary carried forward to
performance assessment.
In 2002, there will be many reviewing
DOE's reports, at least those that I expected. We'll
be trying to make input to the TPA 5.0 code on the
mechanical failure of drip shields and waste package,
input to IRSR on the independent analysis for
DECOVALEX. Analysis participation in DECOVALEX is in
important. In fact, it is true DECOVALEX that we'll
be able to decide with confidence that the thermal
properties of in-tact rock can be used to characterize
conductive heat flowing in a rock mass. So that's one
valuable contribution, and that's why we continue to
support that.
In 2001, most of our accomplishments are
really related to the technical exchange and to the
participation in DECOVALEX. Thank you.
CHAIRMAN HORNBERGER: Thank you very much.
Milt, questions?
MEMBER LEVENSON: Yes. I have one.
MEMBER GARRICK: Microphone.
MEMBER LEVENSON: I have one question on
the -- where you asked for the providing of seismic
loads, the NRC is sending out for public comment 10
CFR Part 72, which is geological and seismological
characteristics for siting and design of dry cask
independent spent fuel storage installations. Is
what's being done here consistent with that since that
seems to be a very advanced stage?
MR. OFOEGBU: I'll let Raj handle that
question.
MR. NATARAJA: This is Raj Nataraja, NRC
staff. Part 72 is a very focused rulemaking just for
the ISFSI and NRS, and they had full options under
consideration, one of them being consistent with what
we are doing in Part 60 and 63. And they changed
their approach, and now they're going with a single
2,000-year return period earthquake, which they
believe is consistent with the risk level for the
ISFSI. However, for the repository, we have the
topical report approach where we have agreed on two
different levels of earthquakes -- Category 1 DBE,
Category 2 DBEs -- which it corresponds to 10,000-year
and 1,000-year earthquakes. So Part 72 is not going
to be doing what Part 63 is doing.
MEMBER LEVENSON: I guess my question
really is I view the Part 72 as recognizing that the
potential risk from a spent cask storage facility is
not the same as that of a reactor.
MR. NATARAJA: That's right.
MEMBER LEVENSON: And, therefore, the
general thinking and ground rules that go with a
reactor, safe shutdown, earthquake, et cetera, are
being set aside, and we're saying, "Let's tailor the
seismic requirements to the potential consequences."
I guess really my question is which philosophy is this
based on, reactor philosophy or the dry cask storage
facility?
MR. NATARAJA: It is neither. It is based
on the repository philosophy, which is a third
category, I would say.
MEMBER LEVENSON: But that -- okay.
CHAIRMAN HORNBERGER: Ray?
VICE CHAIRMAN WYMER: I had just a couple
of things. I couldn't help but wonder why if they're
designing the drip shield to withstand a rockfall
impact, and that's part of their argument, why they're
also doing that for the waste package or visa versa.
If they're designing the waste package to withstand
rockfall, why the drip shield?
MR. OFOEGBU: Well, actually we have asked
the same question. It's not yet clear whether the
drip shield is being designed to protect the waste
package from rockfall impact. What it appears is that
at this point the function of the drip shield is to
protect it from dripping water, and so in analysis
then we are going to look at the possibility of a drip
shield actually contacting the waste package during
the deformation episode. We are going to look at the
waste package carrying some static load and then
vibrating on the seismic load, a number of things like
that.
VICE CHAIRMAN WYMER: Second question is
how is it assumed that the rocks fall? Do they fall
flat? Do they fall on an edge? Do they fall on a
point?
MR. OFOEGBU: Well, those are all
possibilities. The previous DOE analysis actually
assumed they fall flat, and it's one of the things
that we were not happy about, and there is an
agreement item under the technical seal list area to
consider point impact.
VICE CHAIRMAN WYMER: Point impact is what
you're using.
MR. OFOEGBU: That's one of them. Point
impact, line impact, surface impact are all
possibilities, so those have all been --
VICE CHAIRMAN WYMER: And you're looking
at all of them; is that what --
MR. OFOEGBU: Yes, yes.
VICE CHAIRMAN WYMER: And, finally, you
said at one point toward the end there that heat
transfer is only by radiation from the drip shield to
the drift wall. What happened to convection?
MR. OFOEGBU: Well, convection too, but
it's mostly radiation.
VICE CHAIRMAN WYMER: The whole thing
doesn't hit you too hard that those temperatures --
MEMBER LEVENSON: It's hard to believe
it's mostly radiation at these low temperatures.
MR. OFOEGBU: Okay. The point that -- let
me now go into it. The point there is that when the
drift if filled with broken rock, the role is
radiation is going to be removed and you're going to
have conduction through broken rock and maybe
convection through the pore space that is still
available.
VICE CHAIRMAN WYMER: That's all I have.
MR. AHN: Can I --
CHAIRMAN HORNBERGER: Tae Ahn?
MR. AHN: Yes. Tae Ahn, NRC staff.
Regarding the rockfalls, CNSTE is coordinating with
IDTME. The staff member at the Center who is working
on is Dr. Goody, and I believe DOE threat is to avoid
any waste package failure by rockfall. The primary
barrier to prevent the rockfall failure is drip
shield. Therefore, they tried to design drip shields
to withstand the rockfall. Nonetheless, Dr. Goody
raises several detailed questions, including rockfall
effect on the waste package.
PARTICIPANT: That's sort of go back.
MR. AHN: Yes, that's what we did. The
second one is in his analysis by Dr. Goody he also
raised the point of contact. We raised that issue in
the preclosure tech exchange as well as IDTME tech
exchange with DOE.
CHAIRMAN HORNBERGER: John?
MEMBER GARRICK: Tae Ahn just answered my
question. I'm in good shape.
CHAIRMAN HORNBERGER: Okay. Other
questions, staff? Thanks very much Goodluck.
MR. OFOEGBU: Okay. Thank you.
MR. NATARAJA: Let me just add one more
thing to the question related to the seismic design.
The Part 72 considers a 20-year design life, whereas
we are talking about 100 years preclosure right now
for design purposes. That is one major difference.
And the quantity of waste is another difference
between the two. And because we had this topical
report agreement with DOE, we are not going to make
changes based on what Part 72 is doing. We are
expecting Part 72 to be doing what we are doing
originally, but because of some of the exemption
requests that they had, they took this approach. And
we are working with them closely. We are not
inconsistent with what they are doing. In terms of
risk space, we are quite similar.
CHAIRMAN HORNBERGER: Okay. We're going
to hear about preclosure.
MR. DASGUPTA: Good afternoon. My name is
Bis Dasgupta. I'm here to present to you the
preclosure aspects of the RDTME KTI. I would like to
first acknowledge the contributions of the NRC and the
Center staff. From the number of participants, you
can guess that there has been quite an increased
activity on the preclosure in the past year.
This is the outline of this presentation.
The format of this presentation is similar to all
other KTIs that you have been hearing so far, so I'm
not going to go through this list again. The summary
of issue resolution status, staff had their first
technical exchange and management meeting with DOE on
the preclosure safety. It was held on July of 2001.
Preclosure safety was divided into ten
topics consistent with the Yucca Mountain review plan,
and each of those topics were further divided into
subtopics. Selected subtopics under some of these
topics were discussed at the technical exchange.
Based on the discussions, NRC and DOE have reached a
number of agreements on selected subtopics.
The next three slides actually summarizes
the ten preclosure topics that DOE needs to address
before the license application. The subtopics and key
concerns and the number of agreements that has been
reached. The total number of agreements were about
nine, and in addition to these agreements there were
three position papers that were developed by NRC. I'm
not going to go through the details of these three
slides, but I would like to, however, point out that
the identifications of these subtopics and concerns
were based on limited, focused, risk-informed review.
And some of the concerns were of higher level or
general in nature, and some of them were specific.
More concerns will be identified with
ongoing review, and those concerns will be discussed
in the coming technical exchanges that we plan to have
with DOE. And all topics -- we will have concerns on
all topics, including those listed over here.
What does DOE need to provide before LA?
The list that you will see in this slide and the next
slide is actually based on the July technical
exchange. First of all, DOE will revise the flow
diagram that defines the preclosure safety analysis
methodology that incorporate and consider both
internal and external events in the preclosure safety
analysis process.
As regards the naturally occurring and
human-induced hazards and initiating events, NRC has
discussed only two hazards in this technical exchange.
One was aircraft crash hazard, and the other is
tornado missile impact on the waste package. NRC has
pointed out that the exclusion of aircraft crash
hazard as a potential hazard in the preclosure period
is kind of premature. NRC stated that DOE should
provide detailed analysis taking into account all
types of aircraft flying in the vicinity, including
the reasonable projections of the future activity.
Flight modes of military aircraft and combat training
aircraft should also be included in their analysis.
DOE will assess these hazards and will provide updated
reports on the aircraft crash hazard analysis.
DOE will also assess their evaluation of
the tornado missile impact on waste package. These
were the two agreements that were reached on the
hazards during the technical exchange.
In the future technical exchange, we will
discuss the operational hazards in nearby industrial
and military hazards, fire hazards and any other
hazards for which we've completed a review and have
the concerns ready to discuss with DOE.
The third bullet is the justification of
screening and categorization of event sequences. DOE
plans to eliminate event and event sequences from the
preclosure safety analysis based on the design of
structure system components important to safety. NRC
indicated that the DOE process of elimination of
events and event sequences must be consistent with the
risk-informed performance-based philosophy and should
be screened based on probability consequence.
The second point on this one is that DOE
presented their event sequence analysis with a point
estimate of probability of component failure. NRC
indicated that DOE probability should consider
uncertainty distribution in their event sequence
analysis. However, the mean value of the event
sequences can be used for categorization of these
event sequences. Categorization means Category 1 and
Category 2 even frequencies.
The consequence analysis of Category 1
event sequences were also discussed in this technical
exchange. NRC has reviewed DOE evaluation of those
calculations to public for Category 1 and Category 2
event sequences, developed the position paper,
accepting the overall approach of the methodology for
the public dose calculation and the compliance with
the performance objective.
In addition to the analyzed dose approach,
which DOE is using for performance objective, for
showing the compliance with the performance objective,
NRC has pointed out that DOE should also look into the
individual event dose limits, and that they should
comply with the regulatory dose limits. And also
consider multiple events occurring in a single year,
and that multiple event dose should also be complied
with the dose limits. One of the important outcomes
of the preclosure safety analysis and also requirement
of Part 63 is that the structure system components
important to safety should be identified.
Number 2, topics that were discussed
during the technical exchange. One is the Q list, and
the other is the quality level categorization that was
presented by DOE. Two agreements were reached in this
area. One was that DOE will modify the procedure QAP
2-3 to identify and categorize structure system
components involved with the safety, to include the
risk insight gained from the preclosure safety
analysis. DOE will also provide guidance document for
conduct of preclosure safety analysis. These were the
two agreements reached under the structure system
components important to safety.
The design information of structure system
and components important to safety, there were three
aspects that were discussed. One was a preclosure
criticality issue, burn-up credit, waste package drop
of numerical modeling and waste package fabrication
process. DOE will provide updated preclosure
criticality analysis and resolve the burn-up credit.
There was an agreement on that. As regards the waste
package, DOE will provide additional information on
final modeling of waste package drop analysis, for
example, the mischaracterization, boundary condition,
failure criteria used. This was one agreement that
was reached.
The waste package fabrication, staff had
several questions on the welding and fabrication of
waste package performance, and there were three
agreements that were developed during the preclosure
technical exchange meeting.
What does NRC need to do before LA? One
aspect of work is to assess DOE issue closure. DOE
will provide the data model analysis as part of the
agreement, and our job is to review that DOE has
fulfilled the agreements or not. Complete development
of PCSA tool. This is a tool that we are working on
to review DOE preclosure safety analysis. Develop
confirmatory hazard identification and failure rate
database. Review DOE documents on preclosure safety
analysis. We will perform limited analysis using the
preclosure safety analysis tool or any other modeling
that we're required to do in order to satisfy us that
the DOE's calculations are in the right direction.
Conduct NRC and DOE technical exchange on remaining
preclosure topics. Those concerns, as I said, will be
raised as the review progresses. Prepare preclosure
section of integrated issue resolution status report.
That's an ongoing process.
Overview of fiscal year 2001 activities
and accomplishments. As I just said, NRC and DOE has
just completed -- not just -- but in July they
completed the first technical exchange. We discussed
12 topics, subtopics, and nine agreements were
reached. Preclosure preliminary design basis
document. Integrated issue resolution status reports
is under development. There were two sections that
were already developed. One was on the site
description, other is the design of SSC, structure
system component, for subsurface facility. Currently,
we are working on five other sections, and those will
be available sometime during this year.
Preclosure tool development, this provides
the capability to review DOE preclosure safety
analysis, and this tool is consistent with the
regulatory requirements of 10 CFR Part 63, and it is
also compatible with the review matters of Yucca
Mountain's review plan. This tool also has the
capability to conduct independent analysis on all
aspects of preclosure safety analysis.
We have also reviewed and looked into the
incorporation of human reliability and software
reliability analysis in the preclosure safety
analysis. We reviewed the basic concept of selected
methodologies from considerable research and guidance
developed by NRC on human reliability analysis. We
explored the significance of software reliability and
reviewed basic concepts of methodology from research
and guidance again developed by the NRC. This
software reliability, I want to clarify, is that DOE
is planning to use remote operations in the preclosure
in the surface and subsurface facility operations, so
we'd like to look into that reliability of the
hardware and software in those remote operations and
what sort of hazards that they can lead to.
Last is the development of hazard
identification database. Staff is collecting
information on the nearby military and the industrial
facility to prepare for the review. The amount of
information is substantial, and it will be complex.
And this advanced activity will reduce the level of
FERP required during the review of the license
application. The component failure rate database that
we are developing, it is primarily for the PCSA tool.
The work plan for fiscal year 2002, we
would like to incorporate the human reliability and
software reliability capabilities in the tool,
primarily to review DOE's work, incorporate
probablistic safety assessment capability, improve on
the existing features based on the NRC and Center
staff feedback. This would be an ongoing process. We
would like to fix whatever problems that we have with
the tool, and this is therefore that we will devote
ourselves in this year. Review DOE documents and
preclosure safety analysis, including limited
independent analysis, prepare for NRC/DOE technical
exchange for an Appendix 7 meeting, including
identification of key concerns, continued preparations
of preclosure section of the integrated issue
resolution report, and continue development of hazard
identification database.
In summary, the main -- as I said, there
has been quite a bit of activity in the past year. We
had the first technical exchange with DOE. We
continued development of the preclosure safety
analysis tool, and we have also started working and
developing the preclosure section of the integrated
issue resolution status report. With this, I would
like to answer any questions you have.
CHAIRMAN HORNBERGER: Thanks very much.
Let me start with one that occurs to me. Has there
been any recent impetus to consider hazards due to
terrorist activities?
MR. DASGUPTA: Well, I'll have Banad
answer that, because we are really following the Part
63.
MR. JAGANATH: Right now there are -- this
is Banad Jaganath, NRC staff. Right now we are only
working towards the current Part 63, but I know
there's been going on about the hazards thing. Tim
McCartin does that because he's a Part 63 man.
MR. MCCARTIN: Yes. Briefly, in final 63,
the Commission did speak briefly to that, and right
now -- and it's really in its early stages. The
Commission is reevaluating fiscal security
requirements, et cetera. And right now there is no
planned changes to any regulations; however, it's
currently going on. We'll see what happens, if there
are changes, that it be done through a public
rulemaking. And it is possible. Right now there
aren't particular changes in mind.
CHAIRMAN HORNBERGER: Thanks. John?
MEMBER GARRICK: It's been a while since
we heard about the PCSA, that's preclosure safety
assessment; is that what that is?
MR. DASGUPTA: The tool you're talking
about?
CHAIRMAN HORNBERGER: Yes, the tool.
MEMBER GARRICK: Tool, yes.
MR. DASGUPTA: Yes, preclosure safety
assessment.
MEMBER GARRICK: And as I recall, that was
pretty much quite similar to and based on integrated
safety analysis method that's described in Part 70.
And the analysis is focused on the event sequences and
establishing likelihoods and consequences of specific
sequences. Can you tell me, and the thing that we
have been looking at in Part 70 is how ISA independent
safety analysis, the tool that's been developed, can
serve as a building block towards PRA, or QRA, if you
prefer, probablistic risk assessment. Has that been
kind of the strategy with respect to PCSA is to
structure in such a way that you could go to the next
level without having to do a lot of things over?
Number one.
Number two, is there an aggregation of
assembly component to the PCSA model; that is to say
do you aggregate or assemble the scenarios, the event
sequences to some higher measure of risk?
MR. DASGUPTA: Yes. I think there's a lot
of activities going on in that preclosure safety
analysis area, in the tool development area. First of
all, I'd like to say that this tool really combines
the methodology that has been used in the integrated
safety analysis concept and the softwares that are
used in the preclosure -- or PRA business. So we are
not doing anything new. It's only that we are sort of
keeping all these different gradients under one
umbrella.
And the main function the tool really does
is to follow the Part 63, or the Yucca Mountain review
plan, what the requirements are, for example, the
hazard analysis, the natural and human-induced hazard
analysis, operational hazard analysis. You know, you
have all those different techniques. So we have those
techniques that are already existing that pertain to
the tool. Then we have the even sequence analysis by
using the SOPHAEROS software that's been approved by
NRC. So we could develop those models and try to
analyze those even sequences and do a sensitivity
analysis and uncertainty analysis using those tools.
And then use the consequence analysis,
part of it we are using in our software, to use the
consequence analysis. Then integrate all these
results and try to see how the performance assessment
or the performance objectives are made. Because
safety assessment in preclosure has to follow the Part
63 Category 1 performance objectives and Category 2
performance objectives.
Now, answering your questions, we probably
have all categories or all aspects of PRA in it. That
means hazard analysis, we have even sequence analysis,
we have consequence analysis. And the safety
assessment with the risk insight comes -- has been
built as it is required in the Part 63. Now, in this
tool, we are going a little forward and trying to also
do a probablistic -- you know, add the probablistic
capability of the overall safety assessment in which
we would look into the frequency probability
distribution as well as the consequence probability
distribution.
MEMBER GARRICK: Yes.
MR. DASGUPTA: And try to assess the risk
as well.
MEMBER GARRICK: Well, I was really quite
interested in whether or not it has in it the feature
of being able to assemble the individual event
sequences into a total representation of --
MR. DASGUPTA: That's what exactly it is.
Yes, I mean we take all these things --
MEMBER GARRICK: Because the ISA doesn't
quite do that.
MR. DASGUPTA: No, no, no. But this tool
goes farther beyond that, yes.
MEMBER GARRICK: Yes.
MR. DASGUPTA: And it goes into it and
then collects all this information and then tries to
assess the risk insight of these Category 1, Category
2 event sequences.
MEMBER GARRICK: Thank you.
CHAIRMAN HORNBERGER: Raymond?
VICE CHAIRMAN WYMER: Just a point of
clarification. Under the heading, "What Does NRC Need
to do Before LA," one of the items there is, "Develop
confirmatory hazard identification and failure rate
database." And then under the next viewgraph,
"Overview of Fiscal Year 2001 Activities and
Accomplishments," you say, "Development of hazard
identification database and component failure rate
database." And then over on the next viewgraph, you
say you're going to develop --
MR. DASGUPTA: It says, "developed."
Well, it's an ongoing process, actually, because we
started this component failure rate database as well
as hazard identification database, and we thought that
we probably have reached, but there's a lot of
information out there. We are researching on it and
then putting things. So it's not complete. Yes, it's
an ongoing process.
VICE CHAIRMAN WYMER: Okay.
MR. DASGUPTA: And I think we would stop
before LA. And this database will be helping us in
reviewing the --
VICE CHAIRMAN WYMER: I do have a real
question to follow-up. I don't think I understand
what hazard identification database is. I don't know
where the data come in. It seems like that might be
a list, but I don't --
MR. DASGUPTA: Yes. Well, as I said
before, and I can have -- Milt is attending from there
can jump in whenever you feel like. This is
collecting -- this is actually collecting information
of the nearby military and industrial facility.
VICE CHAIRMAN WYMER: Sorry, of the what?
MR. DASGUPTA: Nearby military activity
and the industrial activities around the Yucca
Mountain area. I think there are a lot of information
that we need to process before the license
application, because DOE has to analyze that hazard,
taking all the data into consideration. And then we
are trying to do a proactive job to develop that
database also, because when the license application
comes, it will be very difficult within a short period
to process this data and to --
VICE CHAIRMAN WYMER: Okay, thanks.
CHAIRMAN HORNBERGER: So data sometimes is
more than chemical concentrations developed from
spectrophotometers.
(Laughter.)
Milt?
MEMBER LEVENSON: Yes. I have one
question and one comment, sort of, I guess.
Previously, we've been told that the KTIs are
relatively broad so that there's a good chance that
whatever safety case DOE chooses to make, you will
have it covered. You may have changed the emphasis.
For the preclosure, is what's going on adequate to
cover the case of whether DOE decides to use truck or
rail shipment, because that makes a significant
difference on the front end?
MR. DASGUPTA: Well, are you here looking
at within the facility, because what we are looking at
--
MEMBER LEVENSON: Yes, yes. Yes, yes.
MR. DASGUPTA: Yes.
MEMBER LEVENSON: Not the process of
transportation.
MR. DASGUPTA: Okay. Well, I really
didn't understand your question, but can you repeat
that again?
MEMBER LEVENSON: Yes. Previously, we
were told that the information being assembled -- the
whole KTI process provided you enough information so
that no matter which safety case DOE came in with you
would have adequate information to review that case.
Some cases you wouldn't use all the information, the
emphasis would be different. And my question is on
the preclosure, the information that you have
requested so far, would it enable you to assess a
preclosure system based on trucks, and much smaller
casks, therefore, or rail cars and the different
issues that come up with that?
MR. DASGUPTA: Yes. I mean it all depends
upon their operations and their structure system
components that they are using. So they would have to
make their safety case by whatever process --
MEMBER LEVENSON: Yes, yes. I know, but
the question is --
MR. DASGUPTA: Yes.
MEMBER LEVENSON: -- the information that
you have already asked for --
MR. DASGUPTA: Yes.
MEMBER LEVENSON: -- in the KTIs, would it
cover both cases?
MR. DASGUPTA: Right now? Yes, go ahead.
MR. LESLIE: This is Bret Leslie from the
staff. You have to remember that both DOE and NRC
have only recently begun preclosure in terms of
preclosure safety assessment, and I think Bis did a
very good job of saying of what has been provided this
is what we have, and we anticipate more information --
MEMBER LEVENSON: Okay, okay.
MR. LESLIE: -- and exchanges.
MEMBER LEVENSON: Okay. The next is a
comment, and that is you mentioned remote operations
as though it was something relatively new. There is
in Idaho some 40 years of experience of very high
quality remote operation -- the manufacture of reactor
fuel. The welding, the inspection and everything,
there's a very, very extensive database on failure
rates and remote operations and lots of hot cell data
all over the country, but that one is QA level that is
comparable to this. And I just wondered whether
you're accessing that.
MR. DASGUPTA: Yes. I think we are just
starting to work on this area and try to see what data
is available and what are the processes that are
available in this area.
CHAIRMAN HORNBERGER: Questions from
staff?
MR. LARKINS: Yes. I just wanted to
expand upon Milt's question. Is there really enough
information know about the above-ground facility and
stuff that you can do an adequate hazards analysis,
and how do you plan to treat things like fires without
knowing a little more about the design?
MR. DASGUPTA: Well, I mean right now we
have to depend upon the level of design that DOE has,
and DOE has, for instance, you mentioned fire hazards,
and they have reports on fire hazards, and we have not
reviewed it totally. It's in the process of review,
and we didn't discuss it in this technical exchange,
because we didn't -- but we plan to do it at the next
technical exchange.
Similarly, you know, I mean, whatever the
information DOE has, the level of details, and we are
going to look into that and try to see that the safety
case has been made based on that details. If there is
lacking in details, that we feel that DOE needs to
provide us more information, we will definitely ask
for it. And this is the whole process of the
preclosure safety analysis that we'll work through.
MR. LARKINS: It will be an iterative
process.
MR. DASGUPTA: It is iterative process.
MR. LARKINS: Yes. Do you have a model
for fire hazards calculating loads and --
MR. DASGUPTA: Well, this is currently
being looked into. I mean I'm not really in a
position right now to expand on that.
MR. LARKINS: Okay.
CHAIRMAN HORNBERGER: Okay. Thank you.
And now we come to our wrap-up, James Anderson.
VICE CHAIRMAN WYMER: Nothing personal,
James, but I have to leave.
MEMBER GARRICK: We've saved all the tough
questions for you, Jim.
MR. ANDERSON: Great. Can you hear me all
right? Good. Again, my name is Jim Anderson with the
NRC staff, and I'm going to try to wrap this all up,
and hopefully, like I've said, you've asked all the
tough questions already. Basically, what I'm going to
try to do is I'm going to try to pull a lot of
thoughts that people have presented and maybe pull
them all together for the whole issue resolution
process. And if you have any questions after that, we
can get into those specific ones.
Basically, the outline for my
presentation, what I'm going to try to cover is the
status of all the key technical subissues. That will
probably be pretty quick, because you were already
given most of them. I'm just going to try to
integrate them all together.
Schedule and status of the KTI agreements,
some current NRC staff activities -- planning future
issue resolution meetings, using the risk insights in
the issue resolution process, a little bit about the
integrated issue resolution status report, or
integrated IRSR. I know I'm scheduled to give you a
brief on that I believe in April, and hopefully that
document will be out by that point in time. And then
a brief summary.
As has been discussed in the previous
presentations, we've conducted technical exchanges in
all the KTIs, and we've also had one preclosure
meeting, and it, like this, mentioned additional
meetings are needed, because that was only like a
preliminary discussion of just some of the preclosure
safety topics.
As a result of the meetings, the 37 key
technical issue subissues are currently categorized as
either closed or closed-pending, and in the backup
slides I've included the definitions for closed,
closed-pending and open, and I've also provided a
table that lists all the key technical issues and the
subissues and their current status in one place. So
if you need to look at those.
The technical exchanges also resulted in
293 NRC and DOE agreements. And the agreements cover
a variety of issues -- documentation. DOE provided
some information at the meetings and all we needed was
the documenting of that information. We asked for
additional technical basis and justification, as
you've heard in a number of examples in the earlier
presentations.
In some instances, we've asked for data
files that we want to review as part of our analysis
of what DOE's doing. So I guess the point I'm trying
to make is I've heard a lot of discussions of the 293
agreements being 293 issues, and it's really not 293
issues. There's a lot less than that. Most of this
is confirmation type information or just data files
and things like that.
CHAIRMAN HORNBERGER: Jim, let me try to
get a little clarification on that. Yesterday, John
Garrick and I spoke with Commissioner Merrifield, and
he expressed an interest in knowing from the KTIs
which ones were potholes, speed bumps or road blocks.
On the 293 exchanges -- or agreements, rather, how
many of them would you say fell into this category of,
"Please give us a data file on this or give us your
data on this"? That's number one, and maybe they're
potholes, I don't know. And how many of them are
detailed requirements for, let's say, two years of
data collection?
MR. ANDERSON: I really haven't look at
them in that specific of detail. Probably a good way
to look at those would be the chart that Bret
presented in his presentation which had major --
CHAIRMAN HORNBERGER: Right.
MR. ANDERSON: I would say in those areas,
the ones that are listed as major probably have the
two-year testing things you're talking about there.
And the ones that are listed in more of the minor
category are probably just, you know, "We're
interested in the data files that back up some DOE
position on something like that."
CHAIRMAN HORNBERGER: Push a little
harder. Give me your gut-level feeling. Of 293, are
150 of them in this category of data files?
MR. ANDERSON: No. There's a lot less
than that are category of the data files. I would say
the majority of the agreements are -- I don't know if
it was Bret or someone else in the presentation
characterized them as, you know, if DOE said they were
exclude something in TSPA, we wanted additional
justification or technical bases for why they're
excluding it. I think most of the -- and if I'm
speaking, Bret, you can correct me -- but I think most
of them fall into that type of category more so than
the data files category.
CHAIRMAN HORNBERGER: Okay.
MEMBER GARRICK: Just carrying that
onward, I guess DOE hasn't indicated any ballpark
effort that's required to fulfill these agreements.
And then if one knew that, you could ask, well, how
many are above a certain -- some thresholds of level
of effort would be very helpful in kind of grasping
what we're really talking about right here, because
these numbers are being used abusively around the
country, that we converted a few KTIs into 300 issues
very easily.
And it's true that a large number of them
are just requests for reports and existing data and
what have you, but there's probably some number, 50,
that require a level of effort that maybe is multi-man
year, I don't know. Some sort of a comprehension of
what we are really talking about here would certainly
be helpful.
MR. ANDERSON: Right, and I think we'll
probably get closer to that in the next several
months. As Bret mentioned, he, with a number of other
people, are doing the risk insights initiative, which
will be some of our look at which KTIs, KTI subissues
and agreements might be more risk -- have more risk
than others.
And then also the DOE plan, when they
present their plan for going from fiscal year '02 to
license application and they were going to include in
that how they're going to address the 293 agreements,
I think that information might also add to our
understanding of which ones that are going to have the
most effort -- that are going to need the most effort.
MEMBER LEVENSON: John, we might ask the
other side of that coin: Has anybody in NRC made an
estimate as to how many man years it will take you to
review those 293? That you should have done for
planning purposes?
MR. ANDERSON: Well, I think we took a
very quick look at with Bret's table. I don't think
-- there's just too many variable, especially with the
possible design changes and things like that. Some of
the agreements could take a long time if they do one
design but could take very little if they do a
different design. So I think an effort like that on
the NRC side I think -- I don't know if it's fruitful.
MEMBER GARRICK: I guess the only point
here is that somehow we need to blunt the somewhat
facetious interpretation of what's going on here, that
we've converted ten key technical issues into 293
technical issues. And that's what we're wrestling
with. And I know they're not issues -- they're not
technical issues.
MR. PATRICK: Dr. Garrick, Wes Patrick
here at the Center. Perhaps a couple points I would
add to that. One of the problems we seem to continue
to bump into is semantics. You certainly, and I think
a number of others on ACNW, are more familiar with the
term of RAI, request for additional information, that
comes into play once a license application has been
received.
In my view, the agreements are much more
analogous to RAIs. They're items that are necessary
for the staff to receive and review to be able to
reach a determination that there is reasonable
assurance that the public health and safety is going
to be protected in accordance with the regulation.
The KTIs are considerably higher than
that, and Tim McCartin could speak to this, but we
worked very hard to craft Part 63 in a way that would
address things in those larger lumps, if you will, of
technical concern processes, groups of processes that
are important to understanding the risks associated
with pre- and post-closure performance of the
repository.
I think if people were go to back and look
at a typical reactor case, certainly we're very
familiar with it here with regard to independent spent
fuel storage installations. It's not at all uncommon
for there to be scores, perhaps hundreds of requests
for additional information.
What makes things a bit different here,
and this is the second point I would make, is that the
Nuclear Waste Policy Act specifically directed NRC and
DOE to get some of that work done ahead of time.
Hence, the introduction of terminology of agreements,
meaning pre-license application sorts of requests for
additional information. So I don't see these
particularly unusual when they're taken in that sort
of a context.
MEMBER GARRICK: I'm sure you're right,
and all we're trying to do is to help this process of
communicating with the outside world.
MR. ANDERSON: Thanks, Wes. All right.
That's about all I had to say about the status of the
key technical issue subissues. I was going to move on
to schedule and status of KTI agreements. As of the
end of the year, DOE submitted information pertaining
to 88 of the 293 agreements. Understand this is a
fluid-type process, so basically whenever they submit
something or whenever we review some documents and
send letters back to them, the numbers will change.
The DOE schedule for providing information
on the remaining agreements is as indicated right
there: 89 in fiscal year '02, 84 in fiscal year '03
and 32 by license application. And I'm sure a number
of you will add all four of those numbers up to see if
they equal 293.
(Laughter.)
CHAIRMAN HORNBERGER: They don't.
(Laughter.)
MR. ANDERSON: Moving onto the next slide,
for the fiscal year '02 agreements, and those I mean
the agreements which were due in fiscal year '02, DOE
submitted the information mostly on time. And by what
mostly on time I mean is within a month of the due
dates. We've reviewed probably about half of those
agreements, and additional information has been needed
on several of those. We're in the process of issuing
three or four additional letters, which will cover the
rest -- I shouldn't say the rest -- almost all of the
rest of those agreements. Those should be issued
hopefully within the next couple weeks.
For the fiscal year '02 and beyond, the
agreements which are due in fiscal year '02 and
beyond, DOE is currently preparing a plan to address
all the agreements in its fiscal year '02 and beyond
planning to LA, so we really haven't gotten too much
information in fiscal year '02 yet, and I'm sure that
once the plan is announced, we'll get a better idea of
what to expect in fiscal year '02, '03 and beyond.
And the current plan, at least which was
discussed at the management meeting we held with DOE
in December, was that they would have their initial
plan ready for discussion in the March time frame.
And I think they've indicated that the whole overall
process would probably go a couple months after that,
but we should have some initial information during the
March 2002 time frame.
Next area I was going to address was the
current NRC staff activities. Planning future issue
resolution meetings. I've been internally referring
to this as round two of issue resolution. In round
one, we kind of -- we've discussed all the KTIs, we
tried to identify the information gaps the NRC felt
needed to be filled by DOE. And during this next
round of issue resolution meetings, I hope to focus on
those agreements and any new information/design
changes that might have come up since the last
meetings and try to refine that information gap. With
293 agreements, I'm sure we haven't nailed down
exactly in words exactly what we want, so I think
there's room for discussions between NRC and DOE to
really focus on what exactly do we need.
And in those discussions, and I'm kind of
moving ahead to the red bullet there, we're planning
to have a meeting in early February to discuss future
meetings -- how we're going to conduct those meetings,
how we're going to use risk insights in the process,
priority -- which KTI or which agreements do we want
to discuss first. And also during that meeting I'm
hoping we can discuss some NRC/DOE communication
issues.
And by those I mean how we're
communicating the status of current activities just so
when we do get documents in, they're not surprising to
us. Or when we issue reviews and stuff, they're not
surprising to DOE. So that's what I'm hoping to
accomplish in the early February meeting, and I
already mentioned that hopefully in March we'll
discuss the DOE plan.
And in the remainder of fiscal year '02,
we'll start holding these meetings, and I think
there's going to be both Appendix 7 and technical
exchange type meetings. I think Appendix 7s, and I
believe DOE would agree with me, may be needed more
than the big formal technical exchange meetings. The
Appendix 7s, the key technical people can get together
and really discuss what the information needs are. So
hopefully there will be a number of each of those type
meetings in fiscal year '02 and as I move on to '03.
And in the preclosure area, like Bis
mentioned earlier, we still need to have some initial
meetings in that area to really iron out what the
concerns or information gaps are, and DOE needs to
provide us information on where they're going in that
area.
Bret really hit this one earlier, but I'm
just going to try to recap it a little bit -- risk
insights and the issue resolution process. As we've
discussed with you in the past, the NRC has used risk
insights in preparation for issue resolution meetings,
the round one meetings, as I've been calling them. As
we move forward into the next round with DOE, DOE has
been looking at these agreements, I'm pretty sure,
from a risk point. And using Bret's initiative and
some of the information we'll get out of that, we can
go into those meetings and really focus on the
agreements and see just to make sure we're asking the
appropriate RAIs, as Wes called them.
And the last bullet, future issue
resolution meetings will also use risk insights, and
the point I was trying to get there is, basically that
the bottom bullet, is that based on the discussions of
risk some of the agreements may change in scope or
possibly could be deleted all together if a DOE
strategy is changed or things like that.
MR. LARKINS: Would that also mean that
you may recategorize the agreements?
MR. ANDERSON: Categorize the subissues?
MR. LARKINS: Yes.
MR. ANDERSON: It could.
MR. LARKINS: Not the agreement, the
agreements. Rather than having what you call 293
agreements, have them recategorized.
MR. ANDERSON: You mean have less or --
I'm not sure what --
MR. LARKINS: Combined group package.
MR. ANDERSON: Yes. I mean, basically,
what -- I'm trying to think what we'd do in situations
like that. We've used a number of different things.
We could just say agreements complete superseded by
another agreement. We could do that way or in some
instances we've actually modified the wording of the
agreement to be more specific. I'm not sure if I'm
answering your question, though.
MR. LARKINS: I was going back to this
communication issue that John had raised earlier, if
there was a better way of categorizing the 293
agreements into something that was more readily
understood and less misunderstood.
MR. ANDERSON: I don't think we have any
effort at this point to do that.
CHAIRMAN HORNBERGER: I think you can take
that as a comment and just go on.
MR. ANDERSON: At that, I will. Moving
on, just a quick status of where we're at with the
integrated issue resolution status report. This has
been a work in progress for quite a while, and it will
document the status of issue resolution. It will
follow the Yucca Mountain review plan format, and we
expect to issue it sometime hopefully in the next few
months, spring of 2002, which is on the slide there.
The first version will discuss the KTIs
preclosure and quality assurance areas, and then
future revisions will discuss remaining areas within
the Yucca Mountain review plan. As the last
presentation in preclosure, there's a couple areas
where -- in the preclosure area where DOE hasn't given
us any information, so they won't be addressed in this
first version, but hopefully we can include that type
of information in future revisions.
In summary, we believe the issue
resolution process is progressing. The NRC staff is
actively monitoring the agreements, and like I
mentioned earlier, we're in the process of getting out
several reviews of agreements that DOE has already
provided information for. We're anxiously awaiting
DOE's plan to see how they're going to go from fiscal
year '02 to LA, and I think that will help both -- I'm
sure it will help DOE, and it will also help NRC in
our planning for products and how we want to handle
meetings in the future. NRC staff will continue to
refine the use of risk insights, and the next round of
technical exchanges will further refine information
gaps, which I've briefly discussed.
So I guess with that -- I guess I didn't
point out during my presentation, I mentioned the
subissue table in the backup documents. There's also
another table which outlines all the agreements, and
the slide before that has the five categories I used
to status the agreements. So that last table is
hopefully to capture the status of all the agreements
in one spot. So with that, if there's any questions
I can try to address or I can hand off to someone
else.
CHAIRMAN HORNBERGER: Thanks very much,
James. John?
MEMBER GARRICK: Jim. I attended one of
the technical exchange sessions and was reasonably
impressed with the efficiency with which it was
conducted, considering the size of the group that was
there and the somewhat formal process. You indicated
that you're having a meeting to kind of decide what
might be described as how to do the next round.
Do you anticipate any fundamental changes?
What kind of lessons have you learned? Do you find
that the system from the point of view of NRC staff is
working well? Or if there are problems, what are
those problems? What do you see as the major
difference between the second round and the first
round in terms of the process?
MR. ANDERSON: The process, at least the
way I envision it, and, again, that's one of the
reasons for the meeting in February with DOE just to
make sure we're all together on it, the way I envision
it is we'd focus to the most extent on the agreements
themselves and any new information that comes up. I
think to try to just keep refining what the
information gaps are, any additional information
letters we've sent out to discuss those, just to make
sure that DOE understands what the NRC's asking for
and that the NRC understands where DOE's going, just
to continue to refine that process.
One of the aspects of the first round of
meetings that I think we will definitely need to
continue is the number of pre-call or preparation type
phone calls with DOE just so we're all clear on what
we're going to try to discuss during these meetings.
I think that was the key all the preparation for those
phone calls and the phone calls themselves I think
really helped focus both the NRC and DOE staffs to
really have a productive meeting, especially when we
have that many people involved in the broad areas.
MEMBER GARRICK: Since these are open
meetings to the public, do you anticipate any change
in terms of how the public participates in the
process?
MR. ANDERSON: I think we'll try to
continue some of the things we did toward the end of
the first round of meetings, basically presenting kind
of an overview of the technical area we're going to
discuss to help the public understand the issues that
are going to be discussed during that meeting. We
will continue addressing questions the public has
during breaks or at different parts of the meetings we
open it up for public questions and participation.
Besides that I don't -- nothing comes to the top of my
head, but --
MR. AHN: Yes. This is Tae Ahn. During
our follow-ups of DOE's implementation of the
agreement, we learned that their official document
lacks many details of their accomplishment. In other
words, they could have closed certain agreements by
laying out properly, but they didn't do. We know they
implemented those aspects, but in the documents they
just simply did not write all of them. So in the
coming meeting, with face-to-face meeting, we'd like
to make sure they did they, they implement the
agreement. That may be an example of new things, at
least in our set of KTIs.
MEMBER GARRICK: Thank you.
CHAIRMAN HORNBERGER: Milt?
MEMBER LEVENSON: Yes. One question, Jim,
which is kind of a maybe rewording that's been asked
twice before, and I assume maybe you'll -- in the 293,
if you were to make a guess as to are any of them, and
if so how many, potentially road blocks, have the
potential to say, "We're in big trouble"?
MR. ANDERSON: You know, I would have to
give those questions to the KTI leads, because I'm not
technically smart enough to know -- I'm a process
person.
MEMBER LEVENSON: Well, you're the only
person I know that's heard all the KTI discussions, so
I just wondered whether you'd heard people say, "Oh,
boy, that's going to be a tough one," or whether you
have a -- are there any potentially that are real road
show stoppers, so to speak, do we know? I mean just
a feeling. I can't ask you to project the future
because --
MR. ANDERSON: I mean I would have to go
back again to the chart that Bret provided and which
categories are major efforts on DOE's part.
MEMBER LEVENSON: Well, maybe a major
effort. That's not really the question. A major
effort implies that if they put some people on it,
they can do it. That's not necessarily a show
stopper.
MR. LESLIE: Yes. And I think what you've
got now is if we felt that there was a road block,
then a subissue would have been open.
MEMBER LEVENSON: Right.
MR. LESLIE: Okay. If you look at how
issue resolution is defined, basically you come to
that conclusion. We've come to the conclusion that
are subissues are closed-pending, which we've outlined
what the gap is, DOE has agreed to provide that
information, and we believe that they will provide
that information.
MEMBER LEVENSON: If I put some words in
your mouth, would you say then that as of now you
don't -- you can't identify any among the 293 that are
potentially show stoppers? Okay.
MEMBER GARRICK: The record doesn't deal
with nods.
(Laughter.)
MR. LESLIE: I was only imitating my boss.
(Laughter.)
MEMBER GARRICK: And that was a vertical
nod, John.
(Laughter.)
CHAIRMAN HORNBERGER: Any other questions?
Latif?
MR. HAMDAN: Yes. I think I want to make
one comment that will clarify to Dr. Garrick and,
Milt, also question. The way to think, and this will
follow-up maybe on what Milt said. In other NRC
programs, we do what's called acceptance reviews, and
these reviews are done after a license application or
license application is received for the sole purpose
of determining whether or not the licensee has
submitted a complete application, complete enough to
conduct an objective review. In this case, if you
think of this as an acceptance review, then before the
license application in the pre-license space, this
will clarify a lot of things. You will not find road
blocks. All that's being done now is saying, yes, if
you honor these agreements, if you give us these RAIs,
then hopefully by the time the license application we
will have a complete application, we will accept the
application and start the review.
MEMBER GARRICK: Yes, we appreciate that,
and what you're really saying is that we have to keep
reminding ourselves that what we're talking about here
is what constitutes a sufficient amount of information
to be a reasonable license application.
MR. HAMDAN: So when you take --
MEMBER GARRICK: We're not offering a
license.
MR. HAMDAN: Yes. So the road block, if
they come, and hopefully they will never, if they are
ever going to come, they will come after --
CHAIRMAN HORNBERGER: Having said all
that, and we do appreciate that, and you know the
sense of the reason that we're questioning. But
having said that, we do have not just suspicions but
pretty well-founded arguments that all 293 are not
equal, right? And what we were simply looking for was
some accounting of which ones are the ones that will
require a major investment of person hours by DOE? So
that was the nature of our question had two aspects.
But we appreciate that.
Thanks very much, James. I wanted to ask
April Gill if there was anything that I had said that
she wanted to correct or if she wanted to comment on
anything that has been said. Anyone from DOE?
MEMBER GARRICK: Well, we just got an
offer. They might be willing to make a few comments
that offer a little perspective to this. I think we
ought to take them up on it.
CHAIRMAN HORNBERGER: Please.
MR. WILLIAMS: Dennis Williams, DOE.
There's a lot of ways to bin up 293 agreements, and of
course we've done -- am I coming across okay? And
we've done quite a bit of work on that, because we've
had quite a bit of discussion on what these 293
agreements entail. And I do have one work-up here
that has some categories and it has the number of
agreements and percentages in those particular
categories. One is perform testing with the
associated analysis. There's 11 agreements that speak
to that. There's one category that says basically
just perform some analysis. There's 41 agreements
that speak to that. There's a category that says
provide additional technical bases or documentation.
There's 188 in that particular category; that's 64
percent.
And in that category, some of those are
where we provided what we thought was appropriate, but
a lot of it was based on judgment, a lot of it was
based on -- I'll put it this way: Whenever some of
our technical people write a report, as all technical
people do, they understand it very well. But
oftentimes it doesn't come across very well to the
reader or understandable. So that's the traceability
issue, the transparency issue, and a lot of that is in
the category of documentation. And many times we have
an agreement that says, "Provide a technical basis or
the DOE provide a technical basis or basically other
evidence on how you are proceeding with regard to an
agreement."
Procedures or guides that are provided,
six. Provide a test or study documentation. These
are tests or study plans; 24. Provide existing data
and databases, databases that already exist but
weren't available; 13 in that category. And then
there's a miscellaneous category of about ten, and
that should add up to the 273.
MEMBER GARRICK: One number I missed, the
one that was the third one you gave, the additional
technical basis and documentation.
MR. WILLIAMS: Hundred and eighty-eight.
MEMBER GARRICK: Okay.
MR. WILLIAMS: Which constitutes 64
percent.
MEMBER GARRICK: Okay. So that --
MR. WILLIAMS: And I can give you, you
know, these --
MEMBER GARRICK: These others are the subs
of that.
MR. WILLIAMS: Right. Right, the numbers
after that. When we went into these technical
exchanges and we started deciding what DOE would or
could do, in the beginning it was, to a large extent,
fairly simple, because most of this work was already
in our plan. So it was just a matter of pulling it
out of the plan, communicating with the Nuclear
Regulatory Commission and then deciding on when we
would do it. When we would do it has become more of
the difficulty a lot of times than actually what needs
to be done, because I think there's a lot of agreement
on what needs to be done, because we have a technical
staff, they have a technical staff. They're pretty
close on the needs. So most of it was sorting out
when we would actually get the work done.
As far as addressing the potholes, I think
there are some potholes out there that will slow
things down a little bit. Are there any tar pits that
we will become embroiled in that we will never get out
of? I don't think so. I think that one of the terms
is, "this isn't rocket science." I mean most of these
things can be worked out. Some take a little bit more
time than others.
We have a pretty good idea that we can
work out these things, because we've ran the plans
out, we've ran the scopes out. We have a pretty good
idea of what scope we need to have to satisfy these
agreements. The only problem is it extends out for
quite a period of time, and now we're in the process
of figuring out how do we refine that, how do we run
some things in parallel, how do we make it come in a
little bit sooner?
The areas that are the potholes, in my
estimation, they're areas where we hadn't done a whole
lot of work on, like container life and source term.
We've delayed that a lot over the years. Probably the
other area that's going to take a lot of effort to get
it sorted out is igneous activity. Not that we didn't
know about a lot of these things. Our technical
people had noted that some of these things needed to
be done depending a lot on what kind of a design we
had.
And as we pulled back out of the design,
some of these things became more important. Igneous
activity, we've got a plan that our people have put
together. It's a two- to three-year program. There's
several million dollars involved in it, but it can be
done. Again, there's nothing there that is really,
really difficult. Biggest issues are associated with
when we can get it done based on the resources that we
can apply.
CHAIRMAN HORNBERGER: Good. Thank you.
MR. LARKINS: Can I ask a quick question?
CHAIRMAN HORNBERGER: Yes. Go ahead.
MR. LARKINS: Are you prioritizing these
now, going back, looking and establishing some
priority for these?
MR. WILLIAMS: We don't specifically
prioritize the KTIs or prioritize the agreements.
What is more, I think, important is doing -- certain
things you have to do first before -- okay, in some
cases, you have to do some testing before you can do
the analysis of it, before you can move it into the
abstractions and into the models. So I think we're
more concerned about that kind of a sequence than
specifically prioritizing one as being more important
than the other.
One other thing I just thought of, and
that has to do with whether or not we had agreed to
something that couldn't be done. And one case I
remember specifically, and that had to do with
drilling our undisturbed samples from the alluvium,
drilling to get undisturbed samples from alluvium.
And in my mind, that was in impossible task, because
I've tried to do that several times over the past 30
years. It just doesn't work out. We had a lot of
debate about that.
In the end, we agreed that it couldn't be
done, we would figure out a different way of doing it.
So based on that and some other discussions that we
had during the sessions, I don't think that we've
described anything that cannot be done.
CHAIRMAN HORNBERGER: Thanks very much.
Thanks, James. I didn't mean to keep you standing
there all that time.
MR. ANDERSON: No problem.
CHAIRMAN HORNBERGER: Okay. Well, I think
we are pretty much on time. That's quite amazing
after a full day, a very full day. I want to thank
not only all of the presenters but all of the people
who contributed to the material being presented, and
I know it was a lot of people. I know this was done
in a big hurry. I appreciate that you said that you
utilized some information that you had available prior
to it.
Nevertheless, this was a big investment of
time and effort, and certainly the ACNW and our staff
really appreciate the effort that people went to to do
this. It's been very good for us to get this update,
and after we have some discussion we may in fact get
back to you with any points that we need clarified.
With that, I think what I am going to do
is thank you for all of the Center people too down
there at the Center. Yes, I included the Center in my
thanks. I'm going to call a five-minute break. We
will end the recorded portion of the meeting. I ask
the members to come back in five minutes. There are
some things that we need to discuss, and potentially
make some headway on our research report. Five-minute
break.
(Whereupon, at 5:05 p.m., the recorded
portion of the ACNW Meeting was concluded.)