ML20128J005
| ML20128J005 | |
| Person / Time | |
|---|---|
| Issue date: | 09/26/1996 |
| From: | NRC ADVISORY COMMITTEE ON NUCLEAR WASTE (ACNW) |
| To: | |
| References | |
| NACNUCLE-T-0108, NACNUCLE-T-108, NUDOCS 9610100152 | |
| Download: ML20128J005 (500) | |
Text
{{#Wiki_filter:. . l'
@fficial Transcript of Procacdings O NUCLEAR REGULATORY COMMISSION
[QfWF 6 IO<9
Title:
Advisory Committee on Nuclear Waste 86th Meeting TRO8 (ACNW) RETURN ORIGINAL TO BJWHITE Docket Number: (not assigned) 7(!.g2g2s ; THANKS! I I Location: Las Vegas, Nevada l O Date: Thursday, September 26,1996 hhk h N N LE T-0108 PDR Work Order No.: NRC-859 Pages 1-298 i j \[ , () i NEAL R. GROSS AND CO., INC. Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W. Washington, D.C. 20005 (202) 234-4433 WOFFICECOPY- RETAIN FOR L ELIFE0FTHE00MMITT!!E
- j
- . . . _ . . . . _ - - . - _ - _ -- . - _ . _ = _ _ - - . . . . - . ..
l l DI8 CLAIMER PUBLIC NOTICE BY THE i UNITED STATES NUCLEAR REGULATORY COMMISSION'S j ADVISORY COMMITTEE ON NUCLEAR WASTE l SEPTEMBER 26, 1996 i The contents of this transcript of the proceedings of the United States Nuclear Regulatory i Commission's Advisory Committee on Nuclear Waste on SEPTEMBER j 26, 1996, as reported herein, is a record of the discussions recorded at the meeting held on the above date. A 4
' N)
This transcript has not been reviewed, corrected l and edited and it may contain inaccuracies. O NEAL R GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVENUE, NW (202) 234 4433 WASHINGTON, D.C 20005 (202) 234-4433
.)
1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 1 3 +++++ 4 86TH MEETING 5 ADVISORY COMMITTEE ON NUCLEAR WASTE l 6 (ACNW) l 7 +++++ . l 1 8 THURSDAY, l 9 SEPTEMBER 26, 1996 10 +++++ 11 LAS VEGAS, NEVADA 12 +++++ 13 The Advisory Committee met at the Hotel San /'% 14 Remo, 115 East Tropicana Avenue, at 8:30 a.m., Paul W. l 15 Pomeroy, Chairman, presiding. I 16 17 COMMITTEE MEMBERS: 18 PAUL W. POMEROY Chairman 19 B. JOHN GARRICK Vice Chairman 20 WILLIAM J. HINZE Member 21 GEORGE HORNBERGER Member 22 23 24 f) Q 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 23& 4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 J
i l l 2 l l 1 ACNW STAFF PRESENT: 1 l
-s 2 Michele Kelton, Technical Secretary 3 Richard K. Major 4 Howard J. Larson 5 Lynn Deering 6 Andrew C. Campbell l
7 Richard P. Savio 8 Carol A. Harris 9 Virginia Colton-Bradley 10 11 l l 12 1 13
,0
(/ 14 15 16 l 17 18 19 l 20 l 21 22 23 24 l O t i 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 t
l 3 1 A-G-E-N-D-A gg 2 AGENDA ITEM PAGE O 3 Opening Remarks by the ACNW Chairman 4 l 4 Welcome by Yucca Mountain Project Manager, 6 l l 5 Wes Barnes l 6 Flow and adionuclide Transport at Yucca Mountain j l 7 Introductory Remarks, G.M. Hornberger 9 8 Evidence for Fracture Flow at Yucca Mountain 9 June Fabryka-Martin 12 10 Bruce Robinson 37 11 Deep Percolation and Flow Modeling at Yucca 55 12 Mountain, Dr. G. Bodvarsson 13 Insights for Yucca Mountain from Fracture Flow l [ h ) \'s# 14 Studies at Apache Leap Research Site 15 Dr. Randy Bassett 98 16 Geochemical Effects on Transport, B. Glassley 140 17 Role of Fracture Coatings, B. Robinson 171 18 Role of Colloids in Fracture Transport, 19 J. Kessler 189 20 Integrated Transport Modeling Coupled Flow for 21 Yucca Mountain, B. Robinson 220 22 Integration into TSPA of Process Level Modela 23 for Flow and Transport, Dr. Luik 255 24 Discussion and Meeting Summary q(_) 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
4 1 P-R-O-C-E-E-D-I-N-G-S 2 (8:33 a.m.) O 3 CHAIRMAN POMEROY: The meeting will now come 4 to order. This is the first day of formal presentations 5 of the 86th meeting of the Advisory Committee on Nuclear j l 6 Waste. Today the committee will first concentrate on the 7 topic of flow and radionuclide transport at Yucca 8 Mountain. Today's entire session, in fact, will be , 9 devoted to this topic. And in a few minutes, I will turn 10 the meeting over to Dr. Hornberger. 11 We also have with us today two ACNW 12 consultants, Dr. Martin Steindler and Dr. Andrew Bassett. 13 Dr. Andrew Campbell, there he is, is the designated O 14 federal official for today's meeting. This meeting is 15 being conducted in accordance with the provisions of the 16 Federal Advisory Committee Act. 17 We have received no written statements from 18 members of the public regarding today's session. Should 19 anyone wish to address the committee, please make your 20 wishes known to one of the committee's staff. 21 It is requested that each speaker use one of 22 the microphones, identify himself or herself, and speak 23 with sufficient clarity and volume so that he or she can 24 be readily heard. () 25 Before proceeding with the first agenda item, 1 NEAL R. GROSS ! COURT REPORTERS AND TRANSCRIBERS ! 1323 RHODE ISLAND AV5., N.W. l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
5 1 I would like to cover some brief items of interest. First
,f-2 of all, it's a great pleasure to announce that the Nuclear 3 Regulatory Commission has appointed Dr. George Hornberger, 4 second on my right, who is a professor at the Department 5 of Environmental Sciences at the University of Virginia, l
6 to a four year term as a member of the Advisory Committee 7 on Nuclear Waste. 8 Secondly, for the first time in several years 9 there are now five sitting commissioners on the Nuclear 10 Regulatory Commission. Commissioners Diaz and McGaffigan l 11 were sworn in several weeks ago and have assumed full-time l 12 duties. Other personnel changes of interest, Bill 13 Russell, the Director of the Office of Nuclear Reactor n s 14 Regulation, has announced his retirement effective 30th of 15 September of this year. Frank Muraglia, current deputy 16 director of NRR, the Office of Nuclear Reactor Regulation, 17 will become the acting director until the position is 18 permanently filled. 19 Finally, Dr. Virginia Colton-Bradley, who is 20 on the end in the red jacket, rotational assignment with 21 the ACNW has been extended through the end of September, 22 which is upon us now, at which time, Ms. Lynn Deering will 23 return to the ACNW staff from her assignment with 24 Commissioner Rogers. () 25 I'd like to turn, now, to the next item on our NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
6 l 1 agenda which is a welcome from Wes Barnes, the Yucca fs 2 Mountain project manager of the site characterization
'~
3 office of the Department of Energy. While we find Wes, i 1 4 here, I'd like to publicly acknowledge the outstanding ' 5 efforts of Wes, of Russ Dyer, of Carol Hanlon, and of many 6 other people that contributed to a highly successful field I 7 trip yesterday. And we, the members and staff of the 8 ACNW, are deeply appreciative of those efforts and want to 9 publicly thank you for them. 10 MR. BARNES: You're welcome. 11 CRAIRMAN POMEROY: With that, Wes, the floor 12 is yours. 13 MR. BARNES: My name is Wesley Barnes. I'm 14 the Department of Energy's Yucca Mountain project manager. 15 Welcome. I understand this meeting was put off for a long 16 time. You were supposed to be here last year. I'm sure 17 it was much more impressive this year than seeing the TBM 18 sitting out on that pad. We got to go down and see the 19 TBM in operation. 20 I often say that I run the project. My 21 secretary runs me. In this case, Ms. Hanlon runs me. She 22 told me to say welcome, we're glad to see you, and 23 describe to you a new organization we're going to put in 24 place. She seems to think that that's important that you () t y_,/ 25 focus on that for a moment. [ I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 2M-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
7 1 What I said to my people was this. I think 2 every company in America should ask themselves every year (~h t
' /
3 what business are we in. And the example I always use 4 that I got from a textbook was that, if the railroads had 5 done that in the '50s and the '60s, we'd probably have one-tenth the truck traffic we have on highways today. 7 They would think they've railroads but in the 8 transportation business, if they had focused on that, we'd 9 have public trains from Washington to New York, et cetera. 10 But what am I supposed to be doing here as e 11 project manager moved from a very academic program to 12 project status? And a project, to me, is something that 13 has, making up a number, 6 million discrete activities.
/G
(-) 14 And when you're finished, something is standing there. I 15 don't think I'm there yet. And it's not even Ms. Hanlon. 16 When you're finished with 6 million discrete activities 17 something is standing there. I don't think I"m in a 18 project status yet but I think I'm in the transition from 19 that program to a project. 20 To do what I think we're supposed to do, get 21 to a license application from the NRC by 2002, I want to 22 focus completely around products. And these are the 23 things that I've got to accomplich in the next six years. 24 Coming up soon, a viability assessment, site O( ,j 25 recommendation report, a license application, so we're NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
8 1 going to form our troops around that. And everybody'that 2 works for the project is going to be involved with some O 3 product. 4 Another is simple organization. Those four 5 boxes that's up with one exception just have one person in 6 them. Mainly, it's the line function that I'm concerned 7 about. 8 Project control is also something new. We've 9 had that on this project but in a very moderate sense. We 10 are going to have a great deal of project control as we 11 move into the project part of Yucca Mountain's life. 12 I'm watching the rest of them read that chart. 13 Questions? 14 Yesterday went well. 15 VICE CHAIRMAN GARRICK: No. Go ahead. Go 16 ahead. 17 MR. BARNES: Yesterday went well. Everybody I 18 talked to this morning said it was a good day. 19 CHAIRMAN POMEROY: Everybody felt it was 20 outstanding, Wes. 21 MR. BARNES: Mr. Pomeroy, Dr. Pomeroy, I 22 appreciate the offer of you wanting to be the one 23 subsistence farmer. I'll try to get the contract 24 together. 25 CHAIRMAN POMEROY: I'm looking for a highly NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 2R4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
9 l 1 lucrative contract. ) 2 MR. BARNES Being the DOE official here, I I7 .T ('/ 3 think we can Work that out and we'll try to put that 4 contract together. We need that one person. 5 Anything else? 6 Once again. Welcome. Hope you have a good 7 two days. If you need me for something, please call. l 8 CHAIRMAN POMEROY: Thank you very much, Wes. 9 At this point, I would like to turn the 10 meeting over to Dr. Hornberger and we'll proceed with the 11 working group meeting on flow and radionuclide transport 12 at Yucca Mountain. 13 The floor is yours, George. (\
\~s/ 14 MEMBER HORNBERGER: Thank you, Paul.
15 Today's working group on fracture flow and 16 radionuclide transport will investigate the status and 17 results of studies of both the saturated and unsaturated 18 zones at Yucca Mountain. Specific areas of focus include 19 the ingress cf water to the repository horizon as 20 indicated by chlorine 36 studies, factors that may control 21 the transrort of radionuclides out of the repository via 22 fracture systems and the Department of Energy's modeling 23 strategy. 24 This will encompass three main issue ares. () 25 One, flow and transport through an interconnected network NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 ?433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
I 10 1 of fractures such as may occur at Yucca Mountain. Two, I l f3 2 the role of geochemical effects on transport, including 1
- \
3 studies of site geochemistry, sorption, solubility, and 4 colloids. That, three, approaches to integrate 5 geochemical data and modeling with hydrologic models and 6 the role of integrated models in iterative performance 7 assessment. 8 One goal of the working group is to determine l l 9 the status and strategy of investigations and whether , l l 10 their integration and application to performance 11 assessment are timely and appropriate. Over the past year 12 there have been, and continue to be, dramatic budgetary 13 and programmatic changes in the HLW activities for both (h 14 DOE and NRC. 15 The DOE program is now concerned with 16 developing a viability assessment by 1998 which includes a 17 waste containment and isolation strategy, synthesis of 18 existing data, refining process level models, possibly 19 addressing model uncertainties through expert 20 elicitations, and conducting a total system performance 21 assessment for the viability assessment. 22 The NRC program, which is continuing to 23 evolve, is now focused on addressing a number of key 24 technical issues through a vertical slice approach. The ('N (_) 25 working group will aid the ACNW in reviewing and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
11 1 evaluating NRC's key technical issues and their importance ,s 2 to performance. t') 3 The standard for compliance of the proposed 4 HLW repository at Yucca Mountain also is undergoing 5 significant changes. The National Research Council 6 reviewed the technical bases for the Yucca Mountain 7 standard and recommended that a risk based, site specific 8 standard be developed. Efforts are currently underway to 9 develop risk based standards and regulations by both EPA 10 and NRC. 11 The ACNW has been reviewing a number of areas 12 within the scope of the changing DOE and NRC programs. 13 One of the key areas of concern ins a risk based approach ,/ m '\ /) 14 is the transport of radionuclides from the facility to the ; l 15 critical group and the factors that affect the attenuation 16 of radionuclide concentrations. The new HLW standard l 17 being developed by EPA likely will focus on risk or dose I 18 to an individual member of a critical group and may not l 19 have specific subsystem requirements which are currently 20 required under 10 CFR Part 60. 21 Nevertheless, the site's ability to isolate or 22 contain waste is a significant contributor to overall 23 safety. Radionuclide transport via ground water, which is 24 one of the key components of the DOE waste containment and I (D_,/ 25 isolation strategy and was one of NRC's key technical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4431
1 12 I 1 issues, is likely to be the most significant pathway in l
,-s 2 terms of dose or risk. Therefore, it is important to N
\ l
l
'o establish the important processes and mechanisms for j 4 retaining and retarding the release and transport of 5 radionuclides from the repository.
6 Today's speakers include Dr. June Fabryka-7 Martin from LANL; Dr. Bodvarsson from LBL; Dr. Randy 8 Bassett from the University of Arizona; Dr. Bill Glassley l 9 from LLNL; Dr. Ines Triay from LANL; Dr. John Kessler from 1 10 EPRI; Dr. Bruce Robinson from LANL; and Dr. Abe Van Luik i
)
11 from DOE. We also, as Paul mentioned, we have with us 12 Marty Steindler serving as an ACNW consultant and 13 Professor Randy Bassett, the University of Arizona, r\ > s \~ / 14 serving as an invited expert. 15 We have many speakers and an ambitious agenda, 16 so with that, I'd like to get started. 17 Our first speaker today is Dr. June Fabryka-18 Martin from LANL. 19 DR. FABRYKA-MARTIN: Is there a microphone I 20 can use here? 21 CHAIRMAN POMEROY: We've run into our first 22 problem. 23 MEMBER HORNBERGER: First problem. Yes. 24 DR. FABRYKA-MARTIN: I'm June Fabryka-Martin
~
( ,s) 25 from Los Alamos National Lab. Can everybody hear me okay? NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3'01 (202) 234-4433
13 1 And, I'd also like to point out for those of you who are
.f s 2 seated on the other side of the pillar, there are hand
- s V 3 outs so vou could follow along in the talk that way and I 4 can kind of prompt you on when to turn the pages, even.
5 I was -- the topic I was asked to address was 6 to update the ACNW on the chlorine 36 results that have 7 been obtained this year. We started an ambitious sampling 8 program in January of this year and have about 100 9 analyses that I'll be talking to you about today. And 10 then secondly, Carol Hanlon reminded me that really the 11 bottom line is what does this mean about radionuclide 12 transport at Yucca Mountain. And that's, as a result of 13 that, I'm sharing the floor with Bruce Robinson, my ,q
's / 14 colleague from Los Alamos. After I present the data and 15 our qualitative interpretation, he'll show you the solute 16 transport modeling results he's gotten and what that means 17 for flow at Yucca Mountain.
18 The outline of my talk is as follows. First, 19 I'll say a few words or one overhead, about chlorine-36 as 20 a hydrologic tracer just to remir<d people who may not be 21 familiar with it what the sources of this nuclide are at 22 Yucca Mour.tain and what the approximate concentration 23 values are for it. 24 Secondly, I'll talk about the objectives of ()j (~ 25 the ESF sampling program for chlorine-36 that we started NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
14 1 in January.
-s 2 Thirdly, talk about the data from that 3 program. Our interpretation of those results and the 4 uncertainties that we see and that we're addressing, how 5 we're addressing them. And I've broken that part of the 6 talk up into two sections because you'll very quickly see 7 when you look at the data that they fall into two 8 populations, those that have highly elevated chlorine-36 9 concentrations and those that are less elevated, some 10 background levels. So, I'll address those separately.
11 Then, Bruce will talk about the solute 12 transport calculations, comparing the simulations of 13 chlorine-36 movement from the surface out to the ESF,
\_- 14 comparing those simulations against the data themselves, 15 then finally, some conclusions.
16 This is a very old schematic of Yucca 17 Mountain. I'm sorry about that. But the general features 18 still hold true. It's still alternating sequence of 19 welded and non-welded tups. And it serves as a frame work 20 for me to point out what the hydrologic issues that we 21 were trying to address in the ESF sampling program were. 22 Can you see? Maybe I should stand on this 23 side. 24 First of all, it's understood on the project
-~
j 25 that the infiltration rate across the surface is highly l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. l (202) 234 4433 WASHINGTON, D.c. 20005-3701 (202) 234-4433
i 15 l 1 variable in a systematic fashion. That the lowest
,s 2 infiltration rates are where the alluvial cover is deep
( \
\,/
3 enough that evapotransportation is able to decrease the 4 infiltration rates to negligible values and that the 5 infiltration rates are higher, invariably higher, where 6 there's little or no alluvial cover on ridge tops and side 7 slopes. 8 We also, it's generally understood or 9 accepted, that water does penetrate through that top most 10 welded unit, the Teva Canyon welded unit through 11 fractures, to get down to at least the top of the PTn and 12 there's lines of evidence for that. But then once we get 13 to the PTn, that's where the real serious questions about 14 water flow begin. And particularly, for example, it's ? 15 been proposed that the PTn, because of it's much higher 16 porosity and low degree of fracturing compared to the 17 overlying welded unit, that the PTn acts as a barrier to 18 further downward movement. It slows down the water 19 movement and limits its entry into the Topopah Spring 20 welding unit. But that's an open question. That's one of 21 the ones we wanted to address in the ESF study. 22 So, the questions are, once it enters the PTn, 23 to what extent is water movement, or the flux, slowed 24 down? To what extent do we have lateral diversion along O. () 25 the contacts? And then, what happens when it hits a fault , NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-443'
16 j 1 or a fracture? Does the water continue moving down and so l
-s 2 forth? So, those were the issues that we were focusing 3 on.
4 And chlorine-36 is a hydrologic tracer. Just 5 to make a couple of major points here. I won't go into l 6 any details. But there's two major sources of chlorine-36 7 in the hydrologic environment at Yucca Mountain. There's l 8 anthropogenic sources and particularly global fall out 9 which is dominant in any young waters, let's say waters 10 younger than 40 or 50 years. We talk about chlorine-36 l 11 usually by ratioing it to the stable chloride isotopes. 12 So, it's chlorine-36 to the chloride with units of times l 13 10 to the minus 15. So, in those units, global fall out b k/ 14 may have reached peak values of as much as 200,000 on the 15 surface soils in the late '50s, early '60s. 16 The other major source is natural atmospheric 17 sources from the action of cosmic rays acting on argon 18 isotopes primarily. And that contributes a ratio of about 19 500 times 10 to the minus 10 at the present day. But, 20 evidence, particularly evidence we've gotten over the past 21 year, indicates that that ratio may have been as high as 22 1,500 over the past -- well, even over the past 50,000 23 years. And also over the past -- a longer period of time 24 as well. r" ( )N 25 In addition to those two major sources, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
17 ' 1 there's also two less important sources but ones that we 2 still have to be aware of looking at the data at Yucca - O 3 Mountain. And that's in situ production in rocks and 4 minerals near the surface by the reaction of cosmic rays ! 5 on potassium and calcium, and chlorine-35. That's a l 6 variable. The importance of that signal is variable. We 7 think it's a function of exposure, age, and elemental 8 composition. And we think it's probably usually negligible 9 just because the weathering rate of those surface rocks is 10 so slow. The possible exception of that is the case of 11 calcite. And I'll mention that later on in my talk. 12 And then, finally, there is another source 13 that definitely is negligible for the EFS samples and 14 that's everywhere there's uranium and thorium, at least at 15 low levels, and therefore, everywhere there's a neutron 16 flux. That neutron flux creates chlorine-36 by neutron ; 17 capture in chlorine-35 at low levels, relative to the 18 signals that we're seeing. 19 There are three objectives to the ESF study 20 that we started collecting samples in January. One was to 21 evaluate the frequency and distribution of preferential 22 flow paths by looking for evidence of bomb pulse chlorine-23 36. Secondly, we wanted to provide bounding estimates for 24 the travel time of water in the matrix of the Topopah () 25 Spring welded unit at the potential repository horizon. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
18 l l 1 And finally, we wanted to evaluate the extent to which the 1 1 1
,y 2 Paintbrush non-welded unit could reduce vertical fluxes or
> i l 3 increase groundwater travel times.
)
4 Our approach was to specify three general 5 sampling criteria as we walked through the tunnel. This 1 6 is the status of that sampling program as of the end of 7 last month and, well, it's still true right now, actually. I 8 First of all, we collected samples every 200 meters. This 9 is what we called a systematic samples. Every 200 meters 10 through the tunnel from the portal back to station 57 is 11 where we stopped so far, we collected a sample. j 12 Secondly, we had what we called feature based 13 sampling that was focused on the Topopah Spring welded 19 (_/ 14 unit from which we collected samples from faults, some of 15 the fracturer, some lithophisal cavities. Also, breccia 16 zones, broken zones, and so forth. That was the bulk of 17 our samples. 18 And then finally, we sampled subunit contacts 19 in the PTn. Usually three samples per contact above, at, 20 and below the contact to look for evidence of lateral 21 movement by seeing an abrupt change in the signal. 22 Here are the results. What I've plotted 23 here -- 24 MEMBER HINZE: Excuse me. Are these contact O)
's , 25 sitec that we might expect ponding on those?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 1
19 1 DR. FABRYKA-MARTIN: Yes, that's right. We 2 talked to Dave Bush. He went through the tunnel with us 3 and we selected, I guess it was five different echunit 4 contacts, each at which there was an abrupt change in 5 porosity or hydraulic conductivity. And I won't really 6 talk about those although I do have an overhead to show 7 it. Because we didn't see anything that was obvious. 8 What I've plotted here is a function of 9 distance from the north portal at the right-hand side, 10 station zero, going in to station 45. And for those of 11 you who don't know, each station represents 100 meters. 12 So, station 1 would be 100 meters from the north portal. 13 So, all together, this is 4 and a half kilometers in from A 14 the north portal. 15 Plotted versus station location, the measured 16 chlorine-36, the chloride ratio. And the solid symbols 17 represent the systematic samples. The shaded symbols 18 represent the feature based samples. 19 Also, on here, I have shaded -- well, there's 20 a dashed line that represents the present day meteoric 21 background ratio of 500. And the shaded area represents 22 our current estimate of the total range over which the 23 meteoric signal has varied over the past 50,000 years. 24 When you look at these data, what should jump (Oj 25 at you right away -- I shouldn't even have to tell you f NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
20 1 that. But anyway, there's two populations should be very 2 apparent. And I've arbitrarily decided there would be 7_
/
)
3 populations above 1,500, what I would consider unambiguous 4 elevated ratios. And I even venture to say unambiguous 5 indicators of bomb pulse chlorine-36 versus those that are 6 less than that value of 1,500. So, the rest of my talk is 7 going to focus on those two populations, starting, first 8 of all, with the ones that are elevated. 9 One thing that very quickly became apparent to 10 people was that if you looked at the locations of these 11 elevated signals on a map of the surface faults, then 12 several of the -- or, all of them, really, ultimately, 13 appear to be associated with faults mapped at the surface. A k--) 14 It's, I think -- well, particularly the Bow 15 Ridge fault. Well, I should say, I should qualify that. 16 All the samples where we had multiple hits of a bomb pulse 17 signal were associated with major faults that were known 18 to cut all the way up to the surface. And specifically, 19 we saw four or so samples at Bow Ridge Fault, another four 20 or five or so samples at Drillhole Wash Fault, and then 21 another I forget how many, six samples or so, at Sundance 22 Fault. And then these other locations usually had one or 23 two occurrences only. When you're down in the tunnel, 24 however, they are near a fault, just not at one with a (oj 25 name yet. I I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.-- .- . . - . _ _ . ~ ~ . - . - - . , - _ - - . . . - _ - ~ ~ _ ~ .
21 1 So, that was our first suggested evidence that 2 faults may be involved. But a more meaningful plot is to O- 3 look at the locations of these bomb pulse signals relative i 4 to faults mapped at the tunnel level. This is a ; l 5 simplified version of a cross section provided me by the 6 U.S. Bureau of Reclamation showing some of the major -- 7 well, I guess all of the major faults but not all the ! 8 minor ones in there. 9 Again, I've marked -- the ratios that are 10 elevated are marked in red here on the overhead and just 11 in open squares on the copies. And the squares in black l 12 are the ones that are less than 1,500. That are not -- do 13 not have an obvious component of bomb pulse in them. O 14 And again, you can see that especially where r 1 15 there's multiple hits or multiple occurrences of bomb l 16 pulse, they're always associated with these major faults, 17 the Bow Ridge Fault, Drillhole Wash, or Sundance Fault. 18 And in addition, at least -- I would say, at least 60 19 percent of the samples with elevated chlorine-36 to : 20 chloride ratios are within 100 meters of one of those 21 three major faults, and at least 81 percent of them are ' 22 within 100 meters of any fault. And I venture to say that ; i 23 after we look at these sampling sites in more detail, 24 we'll find that essentially 100 percent of them are within j 25 100 meters of some fault. t NEAL R. GROSS . COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 ,
22 1 That may say more about the occurrences of 2 faults in the tunnel than about the distribution of the O 3 bomb pulse signal, however, because there's no shortage of i 4 offsets in there. 5 MEMBER HINZE: June, could I ask, what about 6 4,400 there? j l 7 DR. FABRYKA-MARTIN: We didn't have fracture ; i 8 mapr" that far as of this point. So -- I 9 MEMBER HINZE: Well, where those samples that 10 are above the threshold, are they associated with a more 11 highly fractured region? 12 DR. FABRYKA-MARTIN: From the broken limb 13 region? 14 MEMBER HINZE: Yes. 15 DR. FABRYKA-MARTIN: No, I think the broken 16 limb region actually started a little bit further on, 17 although I'm not positive of that. 18 Oh, you know, I do have an overhead that might 19 answer that question, though. Hold on. 20 Am I doing okay on time? 21 MEMBER HORNBERGER: Yes. i 22 DR. FABRYKA-MARTIN: Here it is. Oh, no, it's l l 23 not going to answer your question, either. l 24 This is a plot of the frequency of offsets as O V 25 a function of location. I'm sorry, we just didn't have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234#M
. _ . . _ _ _ . . _ _ _ _ . . _ _ _ . . _ _ . . _ _ _ . _ _ _ - _ _ . . . _ . . - . . .____ _ _ _ . _ _ _ . _ _ _ ...m. _-
, 23 1 any data available at that time and I still haven't looked ; 2 at it. lO i 3 MEMBER HINZE: Well, let me ask the question 4 4 this way. Are your comments related to the faulting? And l 5 higher values of the chlorine-36, do they apply to the one i I 6 at 4,400, to the samples? 7 DR. FABRYKA-MARTIN: Yes, I did -- we have an i 8 indication of -- all it says is a note is fault at where J l 9 the samples were collected. SO, I don't have any details A l 10 about that fault, whether it's like -- what the timing of ? l 11 it, whether it was likely to extend to the surface. l 12 That's in our program for this coming fiscal year which 4 l 13 I'll mention later on. i lv 14 MEMBER HINZE: Thank you, June. i j 15 DR. FABRYKA-MARTIN: Of course, it's a i 4
- 16 critical question whether or not this really is bomb pulse 1
l 17 chlorine-36 and not some other source of elevated ratio. i j 18 What we've started to do, and we'll continue this next 1 4 19 fiscal year, is look at other bomb pulse nuclides in the s' 20 tunnel at the same locations where we see these highly 21 elevated signals. Specifically, we looked at iodine and 22 technetium-99 which are both fission products present in 23 global fallout. They're expected to be present as anines J 24 in the subsurfaces. So, like chlorine-36, they should () 25 move conservatively, not be retarded by absorption, for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
. - - - . - - . . - . . ..- . _ . - - . - . - .-.- . . - - .~ - .--. -,-_._.-. 24 l 1 example. 2 And, indeed, in every sample where we saw O 3 elevated chlorine-36 -- well, we only had a small suite of 4 samples, I should say that. Just three samples in the 5 case of technetium, four in the case of iodine-129. At 6 every sample, we saw iodine-129, technetium-99 as well. l 7 Specifically, we saw both of those fission products at I 1 8 station 2 where the Bow Ridge Fault is, which is a 40 l 9 meter depth below the surface. And we saw them in bore ) i 10 hole N55 from cuttings from a depth of 53 meters at the ! 11 base of the PTn. In addition, we saw iodine-129 in a 12 sample from station 34 + 71 which is near the Sundance 13 Fault. 14 We also looked at cesium-137 and plutonium l l 15 simply because it was easy to do, not because we expected 16 to see anything. And sure enough, we only saw them in the 17 surface cells and not in any of the deeper samples. So, 18 all of those distributions are consistent with what our ; 19 understanding of the geochemistry of those nuclides is. 20 In addition to those, the GS sample tritium, 21 or collective samples for tritium, for several locations 22 where we saw the elevated chlorine-36 signals, they did l 23 not see tritium in any ct those five samples that they 24 looked at from the ESF tunnel walls. There is also 25 horizontal bore holes at alcove 3 which is located at the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
25 1 top of the PTn. There was some tritium above background f- 2 in those bore hole samples. ( 3) 3 This was not an unexpected result, to not see 4 tritium. There's a number of explanations for that which 5 I can address if there's questions about that. So, in 6 other words, we do not consider the lack of tritium as 7 being an indication that that's not bomb pulse chlorine-8 36. 9 CHAIRMAN POMEROY: June, before we leave that, 10 for the non-chemists among us, which include myself, I 11 guess that statement in the fourth bullet implies that the 12 lack of -- the non-presence of iodine-129 and technetium-13 99 in other places where there are elevated chlorine-36 is (7 > 1 \-/ 14 consistent with your understanding of bomb pulse -- 15 DR. FABRYKA-MARTIN: There were not samples 16 that we didn't see those two on. But I should qualify my 17 statement about this. The only purpose of this small set 18 of samples, I only picked three or four for each of the 19 nuclides plus surface soils, the only purpose was to see 20 whether we could do it, develop the chemical procedures. 21 It wa snot to prove the point undeniably or beyond a 22 question of doubt that this confirmed the bomb pulse. 23 That's what we plan to do this next fiscal year. 24 So, I have no null results which would make (D. (_) 25 any scientist feel uneasy. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
26 1 CHAIRMAN POMEROY: Thank you very much. 2 DR. FABRYKA-MARTIN: So, this is the final 3 slide on the elevated signals which we are interpreting as 4 bomb pulse, telling you what the ares of uncertainties are 5 and what we're planning to do to address them. 6 First of all, there's source term 7 uncertainties. We're constantly checking for sources of 8 contamination. We do not believe any of the samples that t 9 represent our database of bomb pulse signals in ESF have 10 contamination at any significant level. 11 We also are looking at the potential of 12 production of soil calcite as contributing to these high 13 signals. I think it can be ruled out on mass balance O kJ 14 arguments. We'll be taking that a little bit further. 15 There's no question you can get ratios up high enough to 16 match or even exceed the ratios we're seeing in the 17 tunnel. However, the reason the ratios are so high is 18 because the total amount of chlorine is so darn low that 19 you're ratioing something over zero, of course, is going 20 to be a massive ratio. And so, that's why I think it's 21 not a feasible source for these highly elevated signals. 22 And finally, we're working on reconstructing 23 the past chlorine-36 to chloride ratio in the atmosphere. 24 And here the question is, what is the maximum upper bound O) ( 25 for that variation? I've suggested at 1,500 and we're NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 l (202) 234-4433
. _ _ - . _ _ _ _ _ _ _ _ . . _ _ _ _ . _ . . _ _ . ~ . _ _ _ _ . _ _ _ . _ . . _ _ , . _
27 1 working on getting a more firm scientific basis for that. 2 I believe that boundary is going to stay pretty firm, even O 3 as we go and gather more evidence for that. But we are 4 continuing on it. 5 Nonetheless, this is an extremely important 6 issue and so we're acquiring additional corroborating 7 evidence regarding the fast path interpretation. And now 8 when I'm Faying we, I don't mean Los Alamos. I mean as a 9 project as' a whole, we're working increasingly closer with 10 our fellow scientists in the USGS and the U.S. Bureau of 11 Reclamation for example. For example, we'll be looking 12 more closely at correlating those signals with what's 13 happening up at the surface, with topography, with surface 14 infiltration, and down in the tunnel with the structural 15 setting of each of those sample locations to look at the 16 orientation and look for systematic trends in that sense. I 17 We'll continue the program with measuring ; 18 other bomb pulse nuclides. GS will be continuing the 19 tritium work and we'll be continuing the fission product 20 work. We'll be measuring chloride pore water 21 concentrations because this is a surrogate indicator of 22 percolation rates insofar as the degree of concentration l 23 of chloride as a direct measure of the degree to which the 24 water has been concentrated by evapotransporation at the 25 surface. So, if you receive extremely low chloride NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
28 1 concentrations, that's an indication of a fast path. ! i 2 And finally, we've already done a lot of
- O l 3 solute transport modeling of the bomb pulse signal to show 4 under what conditions would be feasible to get bomb pulse i
l t l 5 at depth. And secondly, what that would imply in terms of ! 6 a flux. Bruce will mention this briefly during his part 7 of the talk. 8 Now flipping to the other samples, the ones 9 that are less than 1,500 ratios. If you remembe.r from 10 that data slide, most of those, the majority of those, the 11 vast majority of those, were well above the present 12 background ratio of 500 but still below 1,500. We're 13 believe those can be used at least as qualitative 14 indicators of groundwater travel times and to help to 15 provide some level of confidence in our modeling, in our 16 hydrologic and site transport modeling. And we're taking 17 multiple approaches to using those data to establish 18 estimates of groundwater travel times. 19 First of all, we're establishing upper limits 20 for those travel times by assuming a value for the maximum 21 possible signal in the past, which we're currently saying 22 is about 1,500, and assuming that the sample has decayed 23 down to its present value from that starting point. This 24 gives us ages in the order of 100,000 years or so. 25 Skipping to the third bullet, we also NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
29 1 establish a lower limit in this case by matching the peaks l 2 in the reconstructed signal -- I'll show you this on the 7s i 'I 3 next plot -- to see how recently could the signal have l 4 been as high as what we see in the sample. And like I 5 say, I'll show you that in the next plot. 6 And then thirdly, we're getting our best 7 estimate of travel times. And here we have upper limits, 8 lower limits. Here's our best estimates of travel times 9 from the signals by doing transport simulations using the 10 reconstructive signal in the atmosphere. And this is what 11 Bruce will show you. 12 As I mentioned a couple of times about the 13 reconstructed ratio, this is what I'm talking about. The t'
\- 14 chlorine-36 to chloride ratio in the atmosphere has 15 changed over the past due to two major processes. One, 16 changes in the chlorine-36 production rate in the 17 atmosphere in response to changes in the geomagnetic field 18 intensity. When the field intensity is high, cosmic rays 19 can't get in and the chlorine-36 production rate goes 20 down. And then when the intensity is low, it's the 21 converse.
22 But that's probably a fairly firm model. I 23 mean, there's some uncertainties a that. But the gist of 24 it, it's pretty firm. r-( ,N) 25 On the other hand, the other big player in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
-. .. -._.- ..-... - .-_ - - - - - . .. - - - - ..~. . - - -..-- - _ ..
30 1 changing the ratio is the chloride deposition rate. And 2 that's going to change in response to climatic factors. O 3 Where is the air mass coming from? What's temperature and ! 4 precipitation, and so forth? That was a little bit harder 5 to get a handle on. And we've taken a stab at it. l 6 And the result of that, our reconstruction, is
]
7 shown in this top plot where I have three curves. The j 8 middle curve, the solid line, is assuming that it's all i 9 tied to a present day ratio of 500 and the upper and lower l 10 bounds are assuming lower and higher initial values based l 11 on some packrat midden data that we have for the region. I 12 And the ratio I've shown here is decayed to the present i 13 day values, so that 300,000 years ago is one half life. 14 The present day value would be about 500. That means it i i 15 would have started 300,000 years ago at 1,000. I i 16 The lower plot is the one that's more easier . 17 to see some of these assumptions we've made and how we've 18 tested it. It's the same plot as above only now blown up 19 to cover a span of 40,000 years. And here you can see 20 what we've shown here is our assumption is that chloride 21 deposition rate is constant for the past 10,000 years. 22 And then throughout the Pleistocene, it was 33 percent 23 lower, driving the ratio up: 24 And we've tested this assumption by looking at 25 packrat -- well, looking at chlorine-36 in packrat middens l l' NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON. D.C. 20005 3701 (202) 234-4433 1
, . , , , , . , - . - - - - , - .- - , - - - - - - - - . . . ~
31 l 1 throughout the region. These are from four sites that are 1 l 7 2 within about ranging from 10 to 120 miles from Yucca l t i V 3 Mountain. And they seem to confirm that the ratio is 4 fairly -- well, it was just bouncing around a value of 5 about 500 for the past 10,000 years. And then it abruptly 6 increased 10,000 years ago. 7 And so, I was talkr;g about, before, about how 8 we establish a lower bound, for example. A lot of you l 9 noticed that a lot of those ratios we saw in ESF, we're l l 10 about 1,000 or so, those ones that are less elevated. You 11 can see the last time that there was a ratio that high l l 12 would have been about 10,000 years ago on back. SO, that l 13 would be a way of establishing a lower limit, lower I fs 2 ( 'l !
\/ 14 groundwater travel time, for those samples in the ESF.
15 Before I turn it over to Bruce, and we talk 16 about solute transport modeling, let me address again 17 where we see the uncertainties and what we plan to do 18 about them this year. 19 First of all, just like with the case of the 20 elevated signals, we're continuing to evaluate sources of 21 contamination in the samples and we haven't found anything 22 associated with the samples I've reported. I 23 In this case, surface calcite may be a source 24 that does contribute in a significant way to the
/
(3) 1 l 25 variability we see in those ESF samples. And so, we'll be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
32 l < l 1 looking at that. And then, continuing trying to get l l l 2 better support for the reconstructed signal. ' 3 As far as planned acquisition of corroborating 4 evidence regarding travel time, again, look for 5 correlation of the signals with what's going on at the 6 surface and with the major structures in Yucca Mountain, 7 comparing our signals and our groundwater travel times l 8 against those which are being obtained with uranium series i 9 dating of fracture minerals by the USGS. 10 Looking or checking for correlation between 11 carbon-14 and chlorine-36 signals in the perched water. 12 Just like the packrat midden samples, the carbon-14 13 results of the perched water support the reconstruction, j O 14 too, and I think Bruce will say a couple words about that.
)
I 15 Again, measuring the chloride pore water I 16 concentrations as another indicator of percolation rates I 17 to back up our travel times. And finally, continuing 18 solute transport modeling. 19 And with this, I think Bruce has a couple of 20 slides. And then we'll come back to conclusions. 21 MEMBER HINZE: ; before you leave that, 22 could I ask a question about the sampling? 23 DR. FABRYKA-MARTIN: Sure. 24 MEMBER HINZE: You're trying to focus in on () 25 the factors that control the presence of these anomalous NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
__ _ _. _ ___._ ___m. _ _ _ _ . _ _ . - _ . . . _ . _ _. . _ . _ _ _ _ _ _ _ . . _ __ 33 . I 1 values. Have you collected samples, multiple samples in 2 near vicinity of each other to see what kind of local O 3 variability there is and how they relate to the attributes i 4 of the formations? I 5 DR. FABRYKA-MARTIN: Some of the samples are . I 6 collected within very close to one another. Now, you're j 7 asking about reproducibility of the signals, essentially? 8 MEMBER HINZE: No, I'm talking, really, about 9 the local variability and its relationship to the 10 geological attributes of the formations. 11 In other words, if you collect the sample here 12 and you may have a fracture or two going through it. You 13 collect the sample over here where there's no fracture O 14 going through it. What kind of variable is it? l l 15 DR. FABRYKA-MARTIN: Comparing the samples? 16 MEMBER HINZE: In other words, what is the 17 significance of the location of these samples? You speak 18 about there's a fault within 100 meters. There's probably 19 a fault within a hundred meters of every point. 20 DR. FABRYKA-MARTIN: Exactly. 21 MEMBER HINZE: So, this doesn't do much for 22 me. 23 DR. FABRYKA-MARTIN: Right. 24 MEMBER HINZE: Help me with what you're plans 25 are for sampling these two, really get at the possible NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
- . - ~ _ - - - . - - - . . - . _ . - . -. . _ - - . _ - . - - - . . - _ - - - _ - . I 34 ! 1 relationship of these anomalous values to their pathways? 2 DR. FABRYKA-MARTIN: Well, we do plan to get O 3 smarter on our next sampling round, a lot more 4 sophisticated in our sampling strategy. This time, going j 5 in with the Bureau of Reclamation and the subsurface 6 mappers to identify sites to sample. It's been suggested, 7 for example, that there may be on certain orientations of 8 the fractures that are much more highly to be transmitting 9 the pulse than others. That would be simply related to a 10 faulting activity and in parallel to it or subparallel to 11 it. Or, they'll have a better idea what through going 12 fractures are, going all the way up through the PTn than ; 13 we do. At least all the way through the Tsw, for example. 14 As far as going back and resampling locations 15 were we saw bomb pulse to get a better idea? 16 MEMBER HINZE: Right. 17 DR. FABRYKA-MARTIN: Yes, we'll do that, too. 18 But we have done it to some extent, although we don't have 19 the results back yet. 20 MEMBER HINZE: Do you have any areas where you 21 have drill holes where you might above the ESF or the near 22 proximity of the ESF, and look at what the changes might 23 be going down the section? 24 DR. FABRYKA-MARTIN: We do. There's the NRG () 25 holes. The north ramp geology holes, and other holes, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
35 1 too. The problem we run into in the bore holes is that 2 the samples are so finely ground by the ream bit that it O 3 releases a lot of chloride from the rock and dilutes the 1 4 signal. And it's very difficult to try to correct for 5 that dilution effect. We can do it, but there's a huge 6 uncertainty to the extent that you can't do a very -- that 7 was the first thing. We thought we saw the bomb pulse 8 signals go to the same depth in the NRG holes which are 9 only 100 feet away or a couple hundred feet away from the 10 tunnel itself. It just didn't work because they were too 11 diluting. t I 12 i MEMBER HINZE: There are no good core samples j 13 that, are -- V 14 DR. FABRYKA-MARTIN: Core's a possibility. 15 NRG holds, I don't believe -- well, that's true. Some of 16 them were -- I guess they were. We could try that. The ! l 17 problem is to get enough sample for our analysis. We'll 18 need a couple meters length of cord because we need to get 19 -- extract -- 20 MEMBER HINZE: It's also a good use of the 21 core, though. 22 DR. FABRYKA-MARTIN: We'11 give it a try. 23 We'll have to also find a way of extracting for it. 24 MEMBER HINZE: I don't want to hold you up. 25 VICE CHAIRMAN GARRICK: Well, you're getting NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
____.___..___.__.__._.______.__.._.._...___.._.___._y 1
'36 1 awfully close to a question that I wanted to ask as a non-l i
2 chemist and non-geologist, and non- a lot of other things. O 3 But, is your sampling procedure being driven i 4 by any particular standard or practice? Or , QA l 5 requirement or whatever? 6 DR. FABRYKA-MARTIN: We're working under a QA 7 program so these are so-called AQ traceable samples. But 8 for in this particular case, that means having -- 9 documenting everything in a notebook about how we proceed 10 on selecting the samples. 11 VICE CHAIRMAN GARRICK: I was curious as to 12 what was really driving the scope and the procedure with 13 response to the sampling. Sample is -- 14 DR. FABRYKA-MARTIN: As far as selecting the-15 site? 16 VICE CHAIRMAN GARRICK: Yes. Well, actually 17 collecting the sample and processing it. 18 DR. FABRYKA-MARTIN: We do have written 19 procedures that we do follow for that. Is that what you 20 mean? There's no standard procedure. If you mean like an 21 EPA procedure? Or ATSM? 22 VICE CHAIRMAN GARRICK: Whether it was a Los l 23 Alamos specific procedure and-- 24 DR. FABRYKA-MARTIN: Yes, it is a Los Alamos 25 procedure. Well, Los Alamos / Yucca Mountain project. And NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
37 1 it is written and documented. 2 MR. ROBINSON: My name is Bruce Robinson from O 3 Los Alamos. I'm going to be back a little bit later to 4 talk about radionuclide but what I wanted to do here was 5 just spend a couple of minutes trying to present some 6 hypothetical explanations for particularly the samples 7 that did not contain the bomb pulse but that contained 8 elevated levels of chlorine-36 from present day 9 conditions. 10 Back to this schematic of the flow system, 11 because of the degree of welding and fracturing of the 12 various units, the travel time of fluid to the water 13 table, we think, is basically controlled by the fact that 14 from the surface to the repository horizon, i.e., the ESF 15 horizon as well, the Paintbrush tuff non-welded unit would 16 be the unit that would exert the biggest control on the 17 travel time from the surface to the ESF. We think that 18 fracture flow in the Teva canyon would result in travel 19 times that are very short. The PTn being non-welded and 20 more likely to be matrix dominated flow would have much 21 longer travel times, on the order of thousands of years in 22 most of the simulations that we performed. 23 In the Topopah Springs, there might be a mix 24 of fracture and matrix flow down to the ESF level, but () 25 some of the fluid would be allowed to travel quite quickly NEAL R. GFH)SS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
38 1 in fractures in the TSw as well. So, it's the nature of fs 2 the PTn that's important to determine the travel time from (OI 3 the surface to the ESF. 4 This is another way of analyzing the data that 5 tries to bring into the mix some more information than 6 simply chlorine-36. This is out to 50,000 years. Thic is 7 the reconstruction that June showed you of the chlorine-36 8 signal. Present day values are about 500. We think that 9 the data support an increase of approximately 10,000 to 10 30,000 or 40,000 years ago up into the 800 to 1000 to even 11 higher range of chlorine-36. 12 However, we also have data from carbon-14. 13 Unfortunately not from the same ESF samples that June p i i
\/ 14 showed, but we have data in perched water samples that are 15 present at the site. And so, what I want to do now is to 16 try to tie these two measurements together to try to see 17 if the combination of measuring chlorine-36 in the perched 18 water and carbon-14 both give you a reasonable result.
19 And I'm going to move the plot up to show you that 20 analysis. 21 From the reconstruction that I showed above, 22 one can plot, say, the carbon-14 activity versus the 23 chlorine-36 activity for various points on that 24 reconstruction. These are data points shown in the non-l (~'% l ( ,) 25 filled points. And we sort of draw a band to say -- to l NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
39 1 give a kind of a range that one might expect from 2 measuring these two samples together. In other words, if 3 you go back to say greater than 10,000 years or on the 4 order of 20,000 years, the carbon-14 activity would be 5 quite low. That should correspond to an activity of a 6 chlorine-36 ratio that's greater than present day of 500. 7 It should be more like 1,000. Whereas, samples that are 8 much younger in age would have higher carbon-14 and 9 approach, maybe asymptotically, this 500 value that we 10 measure for present day chlorine-36. 11 These are the samples that were collected in 12 perched water in which we have both measurements. What 13 they show is a general consistent trend as-what we predict 14 in the sense that, again, samples that have carbon-14 that 15 suggest ages less than, say, 5,000 years, tend to be more 16 like the 500 chlorine-36 ratio. And as you go to the 17 older ages measured by carbon-14, they tend to go up into 18 the 600 and 700 range for chlorine-36. 19 What this suggests is that from two different 20 measurements, the perched water samples, anyway, seem to 21 have ages on the order of, say, 5,000 to 10,000 or 20,000 22 years old. If you do modeling of the system, as I said, 23 the PTn tends to control the predicted age. 24 This is a chlorine-36 simulation which would () 25 tend to corroborate and try to allow us to place bounds on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
40 1 what the average infiltration rate, or percolation rate, 2 through the system is, since the dual permeability
/,_h
~
3 simulation that attempts c apture rapid movement through 4 fracture systems and more slower movement through, say, a 5 PTn which is a non-welded tuff. One gets rapid travel 6 from the surface to the PTn in these sorts of predictions. 7 So, we would predict bomb pulse more often than not to be 8 found at either in the Teva Canyon or at the PTn where 9 travel times go up. 10 Now, if there are discontinuities through the 11 PTn, then that would be an explanation for how a bomb 12 pulse could get even deeper, say to the ESF level. But 13 the general type of numbers that you get for this O k)m 14 simulation for the chlorine-36 to chloride ratio are in 15 the range of 600 to 900 in this simulation for the 16 reconstructed signal that we assumed for this simulation. 17 And that's at an infiltration rate that is variable over 18 the mountain but averages about 5 millimeters per year. 19 So, this is our ways of trying to sort out the 20 chlorine-36 and the carbon-14 data in order to pin down 21 what, perhaps, is more important than an explanation of 22 bomb pulse levels of the ESF. It is the average 23 percolation flux through the mountain. 24 And, I had one more -- O ( ,) 25 MEMBER HINZE: Are faults on this? Have you l l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBER 3 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
41 1 considered faults, though? 2 MR. ROBINSON: On that simulation, there were O 3 no faults since, therefore, there would be no tendency to 4 predict bomb pulse at the ESF level. We're taking.a look 5 at the percolation rates and travel times away from faults 6 in that sort of a simulation. 7 This is just one more way to break down the 8 simulations. This is a breakthrough percentage. It's a 9 travel time distribution, if you will, at the ESF for 10 different values of the average percolation rate. This 11 drives home the point of what types of infiltration rates 12 on an average basis one -- are consistent with the data. 13 If you go between 1 and 5 millimeters per 14 year, now the 50 percent point, which is what we would 15 probably want to focus on here for an average infiltration 16 rate, would be way up here. And at 5 millimeters a year, 17 it's on the order of a few thousand years. At 1 18 millimeter a year, it's up in the 10,000 to 20,000 year 19 range. 20 So, this is just another way of presenting the 21 point that I made on the last slide. Namely, that 22 infiltration rates on the order of, say, 1 to 5 23 millimeters per year would be consistent with the travel 24 times that we're measuring. () 25 Lower infiltration rates have also been NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 'NASHINGTON, D.C. 20005-3701 (202) 234-4433
42 1 proposed and we can't rule these out. But in order for , 2 them to be true, say, a .1 millimeter a year, that would
~'
2 be about a 300,000 year travel time. What you have to 4 assume in order to accept an average of .1 millimeters a 5 year is that we've got the reconstruction of the chlorine-6 36 signal wrong, basically, and that it really was 300,000 7 years or so ago up at 200,000 and it has, by radioactive 8 decay, come down to about the 1,000 level that we measure 9 in ESF. That's an alternate hypothesis. One that can't 10 be ruled out at this point. But there are various lines 11 of evidence which I think drive us towards the 1 to 5 12 millimeter a year range. 13 One other -- this is backtracking just a /3 k_ 14 little bit. For a 5 millimeLer a year infiltration rate, 15 what we do in order to simulate the movement of that 16 portion of the flow which is very rapid, say the portion 17 that represents a bomb pulse measurement in the ESF, is to 18 basically increase fracture permeabilities through the PTn 19 unit in a modeling exercise. And what we find is that 20 when you do that with a 5 millimeter a year infiltration 21 rate, one can predict very rapid movement, even through 22 the PTn, to the point where it's on the order of 10 years 23 of travel time to the ESF. 24 So, there's a consistency in the models if one /\ i ,) 25 were to assume that near faults it's more highly fractured NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
43 l 1 and there are continuous pathways. ; 7- 2 One other thing to point out is that most of () 3 these models predict a very low proportion of the flow as l l 4 being due to the bomb pulse. If you measure bomb pulse, l 5 it could be a very small fraction of the total flow, the 6 total flux, that's really reached the ESF. And because of 7 the way we sample these things, it's left a residue of l 8 bomb pulse chlorine-36 which we then register in one of i 1 1 9 June's measurements. But it could -- it doesn't 10 necessarily, in fact in most model, is not 50 percent of 1 l 11 the flow at that location. It's more like less than a 12 percent or even a tenth of a percent. 13 I'm going to let June conclude and be back n f )
\/ 14 later.
15 DR. FABRYKA-MARTIN: From the ESF study so 16 far, we have the following conclusions about the 17 implications of those data for flow in the unsaturated 18 zone. 19 First of all, that bimodal distribution of the 20 ratios demonstrates that there are isolated fast paths 21 leading from the surface down to the ESF. Secondly, 22 penetration of recent water, meaning bomb pulse water, 23 into the Topopah Spring welded unit is indicated by bomb 24 pulse chlorine-36 in the ESF fractures. However, the bomb (~% () 25 pulse signals by themselves say nothing about the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
44 1 magnitude of those fluxes. And as Bruce made, the point 2 he made in the previous slide, is it's probable that the (_s) 3 flux that represents the bomb pulse component is extremely 4 small relative to the total flux. It's a very small 5 component. 6 Thirdly, our working hypothesis is that fast 7 paths that carry water into the TSw may be associated with 8 major fault zones that cut through the PTn. And this 9 would be in conjunction with areas of higher than average 10 infiltration, that both of those factors would have to be 11 present to get the bomb pulse down. That's our working 12 hypothesis although there are other hypotheses. 13 Fourthly, transport calculations indicate that r~'s e (_ 'n 14 arrival of bomb pulse chlorine-36 at the ESF is consistent 15 with the hypothesis of increased fracture permeability 16 into PTn such as one would expect to be associated with 17 faults in combination with infiltration rates of 1 18 millimeter per year or more. You don't get bomb pulse 19 using those same parameters for infiltration rates less 20 than 1 millimeter per year. 21 And then finally, the transport simulations 22 with the base case hydrologic properties indicate that 23 infiltration rates on the order of 1 to 10 millimeters per 24 year, rather than .1 millimeters per year, are required in
/~~
(%) 25 order to yield chlorine-36 to chloride signals that are in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
- _ ~ . -
- ~ _ _
45 1 a range of 500 to 1,000 times 10 to the minus 15 at the 2 ESF level. O 3 And with that, I conclude a talk and am ready j 4 for questions. 5 MEMBER HORNBERGER: Thank you, June. 4 6 Do we have any questions? 7 MEMBER HINZE: June, I don't want you to be 8 neglected. ( 9 What are you doing to get at the magnitude 10 problem? 11 DR. FABRYKA-MARTIN: Modeling. Dissolute 12 transport modeling. You mean magnitude of flux? 13 MEMBER HINZE: Is there anything that can be 14 chemistry-wise or using different isotopes? 15 DR. FABRYKA-MARTIN: Well, the other way of 16 doing it is the chloride mass balance. I think that has 17 the most promise. That isotopes, I don't think, you can 18 do it that way. 19 MEMBER HINZE: Are you doing the chloride mass 20 balance? 21 DR. FABRYKA-MARTIN: Yes. Yes, we will start 22 that. Up to now, it's -- Alianni has been producing data 23 of chloride concentrations in pore water squeezed from 24 core but that's almost entirely only the non-welded units. ( 25 We hope to expand that by going into the welded units. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
46 1 The difference between what we'd be doing and what he had 2 done is that he was interested in getting the total O 3 chemical composition of the pool water, therefore he was 4 limited to those types of units where you could get 5 sufficient volume to actually do the entire analysis. So, i l 6 it may say 5 to 10 mils, 20 mils or more. l l r l l 7 In contrast, because we'd only be focusing on I i 8 chloride and bromide, also, we can get by with 1 mil. 9 MEMBER HINZE: There has been a considerable l 10 amount of interest in the possibility of water moving down l 11 the Solitario Canyon Fault and then moving horizontally, 12 or at a low gradient, into the repository area along 13 changes in permeability. Is there any effort being made l 14 by you to collect samples that may help to determine the j 15 degree to which that is a possibility? Are you collecting 16 any samples? Running any samples from core holes that 17 approach the Solitario Canyon? Or, that you have several 18 bore holes that have variable distances from the Solitario 19 Canyon? < 20 DR. FABRYKA-MARTIN: I believe there's only 21 bore hole, one of the SD holes, that comes close to that 22 and we do have samples from there. Whether or not they're 23 of good enough quality to work with is an open issue. I 24 think I had problems with some of the samples being too () 25 finely ground to yield a good signal. We're limited, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
47 l I 1 basically, to what bore holes are available and I don't ; I ,s 2 believe there's more than one or two available. 1 (J 3 MEMBER HINZE: Well, we have a huge area 4 between the Solitario and the ESF that is planned to be a 5 major segment of the repository. And I would think that 6 one would be interested in what kind of water is moving j 7 into that area. And, wondering how you can obtain some 8 samples that might help you to get a feeling. I'm kind of 9 concerned that you see no rapid movement along any of the 10 changes in permeability of the units. And this is perhaps 11 good news in the sense that you may not be getting that 12 horizontal movement. But I think that's a very important 13 factor and one that should be worked on. Seems to me that fm., (/ 14 that would be of primary importance. 15 DR. FABRYKA-MARTIN: Yes, I agree. And I 16 should mention that although we didn't see any change -- 17 or, changes in the signal that were unambiguous for the 18 PTn, that was only about 100 meters exposure in the 19 tunnel. And we have a lot more hope for going up the 20 south ramp of being able to evaluate that more fully. 21 But as far as the bore hole question, I'd have 22 to look -- 23 MEMBER HINZE: Have any of your samples come 24 from bore holes? A (,) 25 DR. FABRYKA-MARTIN: Oh, sure. Sure. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
48 1 MEMBER HINZE: So there -- 7_s 2 DR. FABRYKA-MARTIN: And for the non-welded i U) ( 3 samples, yes -- The non-welded samples are consistent with 1 1 the modeling that Bruce showed. 4 I mean, the modeling is 1 5 compared to the data from those bore holes as well as from 1 6 the ESF. And it's consistent. 7 MEMBER HORNBERGER: June, I have a couple of 8 questions. i 9 First, one of the -- it seems to me one of 10 your main conclusions that appeared on your conclusion 11 slide was that your results suggest a 1 to 10 millimeter 12 rate. Is there any other evidence that would contradict l 13 that range or is this the best estimate, most direct l
's / 14 estimate, we have of flux through the repository?
15 DR. FABRYKA-MARTIN: No, it certainly isn't 16 the most direct evidence. There are multiple lines of l 17 evidence that would be consistent with that range of flux. 18 Which is not quite answering the question the way you 19 worded it which was, what evidence is there against that. 20 So, let me answer it. The evidence for is, Allen Flint's 21 infiltration margin program indicates an average flux 22 across the site that, correct me if I'm wrong, is about 5 23 millimeters per year. Is that right? 24 Bo Bodvarsson will be talking, I believe, t' (%) 25 about evidence looking from the thermal gradients at the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
49 1 mountain. Also of that order of magnitude. The chloride 2 mass balance data that we have, I have a slide if anyone O 3 wants to see it. But from looking at the chlcride 4 concentrations in the perched water, in the non-weided 5 units, also indicates fluxes 1 to 10, roughly. I 6 Evidence against that, anyone from the 7 audience? 8 MEMBER HORNBERGER: The second question, then. 9 Clearly, when Bruce presented the modeling, and to really 10 explain your results, to get the mixing right to explain 11 the results, as you've indicated, there had to be ; 12 infiltration into fault zones. A combination of 13 topographic influences so you could concentrate l O 14 infiltration into fault zones that penetrated through. To r 15 what extend have you looked, then, at what fraction of the t 16 total flux through, say, the repository level would occur 17 in these fracture zones? 1 18 Do you see what I mean? There are two -- it's 19 one thing to talk about the average flux through the 20 repository being 5 millimeters per year. 21 DR. FABRYKA-MARTIN: Right. Right. Right. 22 MEMBER HORNBERGER: But it's -- your 23 interpretation is that it's going to be very spatially 24 discreet, even at the repository level. 25 DR. FABRYKA-MARTIN: Right. Right. But the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISt.AND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
50 1 input that we used for the modeling was from Allen Flint's 2 spatial variability map where he -- 0 3 Bruce, what was the range of infiltration 4 rates? From zero on up to a maximum of -- 5 MR. ROBINSON: Of 20. 2,000. Actually, over 6 the -- it averages more like 8 millimeters a year right 7 above -- at the high elevations, higher elevations. It's, 8 in part, elevation dependent. 9 I think the distribution of flow is what 10 you're driving at as opposed to an average. I think s 11 that's going to be a really hard one to tackle except s 12 perhaps through certain stochastic analyses that have been 13 going on, that have gone on to, say, look at permeability 14 distributions and fracture distributions through the 15 mountain. 16 I think when June gets further in this study, 17 it will be one way of allowing us to say, how discreet are 18 these locations and can we even predict where they're 19 going to be. And, the project has undergone some level of 20 prediction to try to do that as well. 21 It's a really tough question. I don't think I 22 have a very good answer for you. 23 MEMBER HORNBERGER: Thank you. 24 CHAIRMAN POMEROY: Bruce, before you go away, () 25 could I just ask you. In your slide that you discussed NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
51 1 the comparison between measured and predicted correlations ,
-s 2 between carbon-14 and chlorine-36 activity, I'm always I ) ~'
3 interested in outlying points. And there is one obvious , l 4 outlying point on that slide. Does that represent a 5 multip14. city of samples or is that a single sample? 6 MR. ROBINSON: That was the baled sample and 7 I'm going to let June explain that. 8 DR. FABRYKA-MARTIN: Yes. As soon as they 9 finish reaming out the hole, they bale the sample and the 10 baled sample, of course, represented a lot of chloride 11 leached from the crushed rock. It would not affect the 12 carbon-14 but dilutes the chlorine-36 to chloride ratio 13 way down. And so, that gives you -- it was like a factor
/~N k- 14 of three dilution.
15 You can see it in the chloride concentration, 16 the chloride-bromide ratio as well. 17 CHAIRMAN POMEROY: Fine. Thank you. 18 Could I interrupt here, also, George just one 19 second to ask that the members, when we are speaking and 20 asking questions, it's been poirded out to me that the 21 people in the back cannot here unless we talk directly 22 into the microphones. So, I'd appreciate it if everybody 23 did that when we ask -- when we comment. 24 MEMBER HINZE: Could I ask, June? Could I I (~T (_) 25 have look at the color simulated diagram. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
52 1 DR. FABRYKA-MARTIN: Oh, you have a copy?
,_ 2 MEMBER HINZE: No, we have black and whites
, ] 3 and I would like to take a --
4 MR. ROBINSON: We've got color versions of i 1 5 that. 6 DR. FABRYKA-MARTIN: Yes. We brought 15 7 colored versions. Oops. We brought 15 colored versions. 8 MR. ROBINSON: Apparently they didn't make it 9 to the front here. I don't know. But maybe we could find 10 those. 11 MEMBER HINZE: All I need is a look at it. 12 MR. ROBINSON: Do you want to put it up on the 13 screen. i \ \/ % 14 MEMBER HINZE: There were a couple of things I 15 wanted to review. 16 Randy, go ahead and finish your -- 17 DR. BASSETT: No, I'm finished. 18 DR. FABRYKA-MARTIN: Do you want a closer 19 look? I could bring the overhead to you? 20 MEMBER HINZE: Yes, that's fine. Please. 21 DR. BASSETT: June, before you leave, I have 22 to lean here. Question for you and Bruce. 23 Because you are speculating the obvious 24 correlation between C-14 and chlorine-36 here, and also (s 25 you speculated that perhaps other isotopes will be useful. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
53 l 1 You've done the preliminary testing for technetium, for 1
,s 2 example. You didn't talk much about your plans for next '
l ! \ l ~ 3 year. Is there a plan to sample in detail, away from a 4 fracture zone, say systematically, and in those samples l l l 5 look for C", perhaps from squeezed samples, or from vacuum 6 extracted, or some other methodology? But C", chlorine-7 36, technetium in a series of samples across -- l 1 8 DR. FABRYKA-MARTIN: I don't know of any plans l l 9 to look for C" in ESF beyond the bore hole studies. And 10 the reason for that is, it's difficult, if not impossible 11 as I understand it, to extract enough pore water from the i l 12 welded units to do a C" analysis. That's one problem. 13 Plus, getting in back far enough from the tunnel walls l
/~N l 'E _/)
s 14 after they've been exposed now for well over a year in 15 some cases that one doesn't have to worry about exchange. 1 16 It's a similar problem, actually, with tritium as far as 17 that exchange issue and the effects of drying out. 18 DR. BASSETT: We've done some vacuum l 19 extraction at Apache Leap that work. I'll show a slide on 20 that. What about the technetium -- 21 DR. FABRYKA-MARTIN: With welded, highly 22 welded samples? Moderately to densely welded samples? 23 DR. BASSETT: Yes, porvaity vf about 6, 7 24 percent.
/T 25 What about technetium or other isotopes?
l i~-) s NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
54 1 DR. FABRYKA-MARTIN: Yes, those we will. What 2 we'll probably do is select -- well, what I will do is (-) 7 select samples after we have the chlorine-36 results back, 4 then select a sweep that are maybe five samples where it 5 was elevated, five samples where it wasn't elevated, then 6 send those and analyze those, then. And do the same thing 7 for iodine-129 on the same sample sweep. 8 And I only plan to do it once. This is just - 9 - the only purpose of doing those analyses is simpic to 10 corroborate the interpretation, not really to look at the 11 distribution of those isotopes in the mountain. 12 MEMBER HORNBERGER: Thank you very much, June 13 and Bruce. ("% 14 By my clock, we're almost exactly on time. We 15 need the microphone. 16 MR. HAYES: Thank you, George. 17 I would just like to emphasize two points that 18 have been made but I really don't want them forgotten. 19 The surficial infiltration range that was 20 discussed, we should not jump to the conclusion that deep 21 percolation equals surficial infiltration, mainly because 22 of the PTn. We still have evidence which suggests the 23 PTn, at least for infiltration, rates under 5 millimeter 24 per year really acts to prevent a lot of that water from () 25 moving into the repository. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
55 1 The other point that's been made but, 'in, I 7s 2 think it's very important, George, and you brought ut we ( ) 3 can talk about average percolation rates but that doesn't 4 really answer our question, is most of this fast pathway 5 operating through localized areas? And I think that's 6 really a very important question for the repository. 7 So, I just want to suggest that we be careful 8 in jumping to, perhaps, conc 3usions about what higher 9 percolation rates or higher infiltration rates may really 10 mean. 11 Thatth you. 12 MEMBER HORNBERGER: Would you identify 13 yourself for the record?
/~s x- 14 MR. HAYES: I'm sorry, George. Larry Hayes 15 with M&O, TRW.
16 MEMBER HORNBERGi'R1 All right. Our next l 1 17 speaker is Bo Bodvarsson. 18 DR. BODVARSSON: Does all of the board have 19 color copies? , l 20 CHAIRMAN POMEROY: I think we all do. 21 DR. BODVARSSON: I have a few more here. 22 My name is Bo Bodvarsson from Lawrence-23 Berkeley National Lab, and I'm here to talk about 24 percolation flux, again. I heard a lot about it already. r' ( ,)\ 25 And talk about it in terms of mostly the UZ NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
56 , 1 model that we are developing at Lawrence-Berkeley Lab in i 2 terms of various parameters like fast flow and -- can 7-i 3 everybody see this all right? Focus it a little bit 4 better? Is that focused? Can you guys see it okay? Can l 5 you turn the lights off close to here or do you have to 6 turn all of the lights off? Or, can you just -- that will 7 be great. Perfect. 8 MEMBER HORNBERGER: Yes, that's better. 9 DR. BODVARSSON: I have a lot of black and 10 white copies in the back if you're interested in it. l l 11 Unfortunately, my secretary has this slick thing in the 12 background that you don't see on the picture which is the 13 Earth. But, when you copy it in black and white, that I A k) ~ 14 comes through more than anything else. And I told her I i l l 15 would prefer Yucca Mountain or something else if it's 16 going to come through like that. But she says that's not 17 your j ob. And you do what you're supposed to do, whatever 18 that is, she says. So, that's too bad. 19 A lot of the information that you get from 20 this talk comes from a recent milestone. We call reports 21 here in Yucca Mountain milestones, as you know. And it's 22 on oe.r UZ three dimensional model. And if you are 23 interested in copies of this milestone, we can provide 24 copies to you. I hope it has been accepted by Russ. Has n ( ,) 25 it? Good. That's good news. NEAL. R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
57
- 1 The outline on my talk is as follows. There l
,~ 2 have been a lot of questions about percolation flux and \)'
3 what data can be used to go after percolation flux. And l 4 Bruce and June described one of the approaches which is l l 5 the chloride-36 approach to looking at the percolation 6 flux. We are going to be looking at it from various other 7 angles, including temperatures, including moisture, 8 tension, and saturations, and all the angles that we can 9 think of. 10 The problem with percolation flux, as you 11 probably know, is that there is no direct way to measure 12 it. We can't put a cup down there and see how much goes 13 into the cup. We have to use all kinds of inverse methods
/
i i k/ 14 to get at it. And that's what we are trying to do. You 15 also have to bear in mind that all of the methods I'm 16 going to present and have been presented are very 17 uncertain. They have a lot of assumptions associated with 18 it so I don't think that we can rule out at this time a 19 lot of differ 1nt flux values. 20 So, basically I'm going to briefly describe a 21 little bit about the model, very briefly. Go through some 22 of the data that we are going to use. Go through some of 23 the model calibration but in light of percolation flux 24 issues. Talk, then, about three dimensional percolation O (_,/ 25 . flux studies that we have conducted. And then talk a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 132* RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
58 1 little bit about the implication of these flux values to 2 important things like seepage into drifts or thermal O 3 loading issues. And then talk about the testing program 4 that is planned and DOE has in the books for the next two 5 years to try to reduce uncertainties, and finally, 6 conclude. 7 So, firstly, briefly about the model. This is 8 a three dimensional model of the unsaturated zone at Yucca 9 Mountain with the boundary condition then being at the top , 10 of the ground surface where we prescribe infiltration. 11 And the boundary condition at the bottom being the water 12 table. In between them, we have a complex mixture of grid 13 blocks, some of them being really fine close to the ! () 14 repository because we want to resolve gradients in that 15 region. And some of them much coarser away where things 16 don't matter so much. And again, some fine, close to the 17 ESF because ESF provides a lot of information as well as 18 ponding conditions. 19 We, then, use this model and I didn't show you 20 the vertical gridding. We, then, use this model to try to 21 match all of the important information that we have from 22 the mountain, including moisture, tension, and saturation, 23 including gas flow and gas pressures because we see gas 24 pressures hundreds of meters into the mountain just when t ( 25 a storm comes by Yucca Mountain. Because the measurements NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
59
-1 that the survey and the other results are very, very 2 accurate, and that gives us an indication of O 3 permeabilities, fault characteristics, and things of that 4 sort.
5 And we also try to match temperature data 6 because the water table temperatures are 33 degrees and at 7 the surface is 18 degrees. And we tried to match the 8 environmental isotopes like tritium, chlorides, whatever 9 we can, to have as good a model as we can possibly make 10 it. And then we try to infer percolation flux from all of 11 these different studies and data. 12 Let me put this one aside. This room'is very 13 small so'I'll put it right there. 14 Now, I want to tell you that the main goal of 15 this work is basically percolation flux, as you all know. 16 Spatial and temporal evaluation of percolation flux. And 17 why is it so important? 18 You all have heard the waste isolation 19 strategy many, mv.ny times and the five major components of 20 the waste isolation stri:stegy. It so happens that the 21 percolation flux really effects four of the five. The 22 ones that are in black. Of course, the percolation flux 23 effects how mach water goes into our drift. It effects 24 the environment to the waste packages, espacially the () 25 humidity and corrosive environment. It effects the waste NEAL R. GROSS COURT REPORTERS AND TRA%CRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 200C0 3701 (202) 234-4433
60 1 mobilization in terms of how much water goes through
- 2 there. And it effects the radionuclide transport, both V
3 through the engineering barrier and as well as through 4 radionuclide transport to the water table. Effects all of 5 those so it's a very important parameter to get a handle 6 on. 7 The only one it doesn't effect greatly, or at 8 all, is dilution in the saturate zone, of course. 9 So, this is the moderation for this model and 10 all of the work we have planned to do. Now I want to go, 11 move away from the models. That's all I'm going to say 12 about the model and move into the data set, what data do 13 we have. rN t )
\_/ 14 And, I first want to give credit to a lot of 15 the collaborators that provide data to our model. A lot 16 of them are from the Survey but from, also, Los Alamos and 17 Sandia, and other places. We, of course, need the 18 geological frame work. We need the matrix properties. We 19 need the fracture properties and fracture matrix 20 interaction. We need the infiltrations. We need the in 21 situ conditions. And here I mean with moisture, tension,
)
i 22 saturations, temperatures, and gas pressures. We need the 23 environmental isotopes that June and Bruce just talked 24 about. The pneumatic data frcm the ground surf ace and f} (_j 25 ESF. And the moisture balance stuff from ESF to try to NEAL R. GROSS COURT REPORTERS AND TRANSORIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
61 l 1 see are we actually losing a lot of water from the ESF l 2 into the formation or are we actually taking water from f-) (G l 3 the formation to the ventilation. 4 And all of that fits into the model which then 5 feeds other parts of the program. And the main 6 developmental model is in this box. 7 Now, during the last one year, there has been 8 a lot of new data collected and a lot of old data 9 analyzed. So, we are seeing a lot of new results and 10 important results that we haven't thought about before. 11 And I want to summarize this a little bit in the slides. 12 Certainly, the gas pressure data that has been collected 13 in about ten wells gives us a huge amount of information t'%
- 14 about fracture permeabilities and fault permeabilities in 15 the mountain.
16 The moisture, tension, and saturation data 17 gives us an indication of the water flow through the 18 mountain. The bore hole temperature profiles give us an 19 indication about the heat flow that may be a combination 20 of heat conduction -- will give us a lot of information 21 about if heat conduction dominates or if it is also due to 22 moisture flow down the mountain. Then we have the 23 chloride-36. The fracture coatings, again, give l l 24 indication about where the water moved in the past and , /~N l (_) 25 where it's moving now. And, you have the perched water i l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS , 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C 20005-3701 (202) 234-4433 l
i l 62 : 1 issues that not only give you where perched water f i 2 accumulates, but also the isotopes give you the ages. And O 3 the well testing on these perched water bodies gives you l 4 how big are these bodies. [ 5 So, all of this tremendous information has to l 6 be incorporated into the model and tried to make a l 7 consistent pictures regarding moisture, gas, and heat flow 8 through the mountain. I 9 I also want to point out that this figure 10 shows kind of the conceptual model that we have been 11 working with in the past that shows the valuable 12 infiltration on the top. Shows the lateral flow component 13 that was always believed to exist in the PTn. The major 14 influx of faults as drainages through the mountain so that l 15 mostly we have very dry region in the repository horizon. l l 16 And as already you know, we haven't seen a lot of water in l l l 17 the ESF. That seems to verify some of this conceptual l l l 18 model. 19 Going, now, more -- I'm finished with the data l. 20 and going to the model calibration issues. And I want to l 21 address model calibration in terms of percolation and what 22 it tells us about the percolation, because that's what the 23 ACNW board requested from this' presentation. l 24 This is more for you to have in your package 25 than for me to describe in details. This shows the NEAL R. GROSS , COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (2'J2) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
63 1 approach we are using to calibrate the UZ model. It has
,s 2 the three components, the gas, the water, and the energy.
( ) 3 They are coupled together but they're also independent to 4 some extent. And we think the key is the perched water 5 location and volumes, the ages, saturation, moisture, 6 tension, increm;ntal isotopes. And all of this gives us 7 some idea about percolation flux coupled with the gas flow 8 and the pneumatic stuff as well as the temperature data 9 here. So, this is for you to have in your packets so that 10 you know what the approach is. 11 But now I want to move gears a little bit into 12 the actual calibration. This is a little diagram that 13 says these are some of the indicators. And there is a O 14 scale. And we want to try to use all of these indicators 15 to give us points on that scale, to tell us percolation 16 flux. There is no single one of this that is so accurate , l i 17 that it's going to give us the volume, we don't have to l l 18 worry about the others. That's the problem. There has to 19 be some agreement among these in order for us to move from 20 one direction to the other. 21 Also, if one of the e says the flux is high or 22 the flux is. low, we shouldn't jump our height in gold, or 23 whatever, because maybe next year we'll find another data 24 that tells us maybe that's not true. So, we need to look
,3
(_,/ 25 at all of these and mark them on there. And that's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
64 1 exactly what I plan to do. 73 2 And I still want to emphasize the , a (~) t < 3 uncertainties in all of this information. We are just 1 4 starting to move through some of the studies to get at the 5 most reliable values that we can think of. I want to 6 start with infiltration which is the top one. And there 7 was some talk before about what is the infiltration rate 8 now that Allen Flint has come up with, and this is the map 9 that I think he's very comfortable with at this time. And 10 the map shows about 40 millimeters per year over the 11 crest, much lower to the eastern part, close to the Bow 12 Ridge Fault and lower also to the western part. 13 If you take an average of the repository
/~h <
t 1
\< 14 horizon, you'd get about 4.5 millimeters per year. But if 15 you do a model calculation like one I'm going to show you, 16 these high values due to lateral flow are going to move 17 somewhat into the repository region. So, at depth, you 18 will get a percolation flux from this about 7 to 8 19 millimeters per year.
20 So, infiltration from Allen Flint suggests 21 that we have -- and I'm going to draw an X here if I can, 22 which you can't see very well. Does anybody have a pen 23 that works on these things? And then I want to put 24 roughly an 8 here. Now let's make a circle. So, what I ( () 25 want to do there so that you remember, first of all there NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASH!NGToN, D.C. 20005 3701 (202) 234-4433
l l 65 l l 1 is variability here. And secondly, which one gives you
,_ 2 what because you're not going to remember all of that.
! \ 3 Now, I want to go in to saturation, moisture, 4 tension data. And here shows just one diagram. We have 5 saturation, moisture -- This shows moisture, tension, and 6 saturation data from one bore hole. I think this may be 7 SD7. I don't remember which one. We have reasonable data 8 from about ten bore holes. And what we tried to do is to 9 match simultaneously for all of the bore holes all the 10 saturation values from cores as well as the moisture, 11 tension or capillary pressure from cores. And then try to 12 see what does this data tell us in terms of percolation 13 flux as well as matrix permeabilities and fracture
,m, i \- 14 permeabilities in about 30 different units going down l 15 through the mountain.
16 And to cut a long story short, what we find by 17 matching all of these welds is that we don't find this 18 data to be extremely sensitive to the saturation -- to the 19 net infiltration values. But they tend to favor in 20 statistics somewhat lower percolation fluxes. 21 So, I'm going to put on this thing here on an 22 X and here you're going to see a X close to about 1 23 millimeter per year based on the moisture, tension, and 24 saturation data. And then I'm going to move on to the O) (_, 25 pneumatics. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005 3701 (202) 234-4433
66 1 Pneumatics data, and I could talk for a whole ,_ 2 hour about pneumatics data because it's extremely ( ) 3 interesting that you can actually see storms with a tiny 4 bit of pressure changes at the surface and you can see 5 that all the way down into the mountain hundreds of 6 meters. This happens to be UZ-7A. And what you see here 7 is the signal and the signal is in the dark gray. And 1 8 then the model simulation is in red. And we did the 9 calibration period and then a prediction. And we matched 10 this data very well, as well as most of the other welds. 11 What does this data tell us? This data tells l 12 us two things. First of all, it gives us very good 13 information about the permeability of the PTn. Because,
/N ,
! 1 !
'- 14 by far, more sensitive to the PTn because the pneumatic 15 diffusivity of the PTn is the largest because that's where 16 the porosity is high and the permeability is lower. It 17 also gives us tremendous information about faults. And 18 that's where I want to show you this graph here which 19 shows the results of our studies on the faults.
20 And, again, the faults information from the 21 pneumatics is somewhat varied. But overall, if you look 22 in the TSw, in the Topopah Springs, the fault permeability 23 to gas is on the order of 1,000 darceys. That's large. 24 Very large. And why do we know that? We know that
/D 25 because when the years that went through here, we had a
( ,/ NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
67 l 1 very quick breakthrough of UZ-4 and 5 just as it hit the p,_ 2 fault that was mapped at the surface. Also, in NRG-6 sees ( l
'~'
3 a signal when the ESF hit the Drillhole Wash, went down 1 4 here, through here instantaneously and into the weld. 5 Very high permeability of the fault to gas. On the order l 6 of 1,000 darceys. 7 The permeabilities of the faults to the 8 surface, in many cases, is less. Surprisingly, the 9 vertical permeabilities. And why do we know that? We 1 1 10 know that because if the vertical permeability in that ' 11 fault close to this weld or this fault close to this weld, 12 to the surface was very high, that weld would not -- a 13 bore hole would not care about when the ESF went through O k- / 14 the fault because it would already get the signal l 15 vertically through the fault. l 16 And so there are many things. Another thing I 17 can -- there's al to of information on the pneumatics from la UZ-7A because UZ-7A is in the Ghostdance Fault and shows 19 the high permeabi ity of the fault in some sections. So, 20 a lot of good information from the pneumatics. But, in 21 terms of flux, it doesn't tell you anything. Because it 22 just tells you gas flow. 23 That is extremely important because it can 24 help us know something about the fault and perhaps we can l') 25 learn if flow is going laterally into these faults are not
- i. y NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
_ _ . ~ . _ _ _ . _ . _ __ _._ _ _._ _ _...-___ _ _-_ _ . _ ___.-_._._ __.__ _. 68 1 from the pneumatics. That's why I included it there 2 because I think it's an extremely important data set that () 3 has been collected. 4 Moving right along. Environmental isotopes, 5 and I don't have to do anything there because Bruce 6 already did it for me. And June did it for me. So, I 7 just mark the number. 8 And they say from the bomb pulse chlorine-36 < 9 we cannot get a global flux. Absolutely corr et. You 10 only get a local flux, if you can get that. You only know 11 a little bit went there. You don't know how much and you 12 know it's focused through the PTn somehow. But from the l 13 chloride-36 ratios on the non-bomb pulse, I think June and j i 14 Bruce concluded something like 1 to 5 millimeters per 15 year. And we will make that a square and put it right 16 here. Environmental isotopes. 17 We also don't have to do anything about l 18 fracture coatings because I don't know anything about 19 fracture coatings and I couldn't do it anyway. But, Sal 20 Peterman has done a lot about fracture coatings aid he 21 told me a year ago, or , told us a year ago, that probably 22 the percolation flux was very low due to the fracture 23 coatings because of this deposition model, continuous 24 depositional model. But now he has some new techniques () 25 that he has used to estimate a global flux over the last NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
i l 69 l 1 10 million years that says the flux is about 1 to 2 1 i ,
- 2 millimeters per year. So, I'm running out of signals but \'~') 3 let's make that a plus. That's certainly a plus, so let's l
l 4 make it 1 to 2 right the r.e . 5 Temperature data, let's go and look at 6 temperature data. What does that tell us? Temperature 7 data, actually, in your viewglaphs you have perched water 8 first so I'll take perched watei- first. So you won't have 9 to reorder your viewgraphs. So, let's look at perched 10 water first. 11 The perched water is extremely important 12 because it tells us somehow the permeability of the 13 underlying layer is so low that whatever water wants to go r '\ ( l
\_/ 14 down vertically, or flowing horizontally can't go down 15 very easily. So, that constrains two things, or one of 16 two things. One is the infiltration into the mountain 17 around here. And the second one would be the permeability 18 of the underlying basal, zeolites, whatever is there. 1 i
19 Then you have the residence time from the isotopes that l l 20 tells us this is 5,000 to 10,000 years old. And then you 21 have the chloride balances in this. 22 If you go through all of this activity and all 23 of these estimates, and you do what Ed Quickless and the 24 others all did in the north ramp report, and you update l {p) , 25 the parameters according to the newest parameters that we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
70 1 have, you find that the perched water sums gives you p- s 2 percolation estimates on the minimum, on the lower side, i / 3 about 2 millimeters per year. That's basically what you 4 get from this. 5 Let's see. What do you think? A diamond? A 6 diamond. So let's make a diamond right here, and it could 7 be larger than that because this analysis assumes the 8 permeability of the underlying zeolites, then assumes a 9 residence time, and it doesn't really take into account 10 the effect of falls. The falls will are short-cutting 11 some of this fluid down into the water table. 12 Finally, temperature data. Let's look at that 13 a little bit. The temperature data has been collected O (~/ 14 from many -- at Yucca Mountain. This is work by John Sass 15 and many others that started in the late 1980s, as well as 16 Joe Rousseau and many others. This shows the bore holes 17 and it also shows the elevation because one thing we have 18 to be aware of, that most of the bore holes are drilled in 19 washes. They're not drilled in the ridge tops, and that 20 may bias some of our analysis for this work. 21 Basically, the temperature of the water table 22 is about 30o and about 18o at the ground surface. What 23 the data teJ1 us basically, based on John Sass's work and 24 Joe Rouseau's work, is that conduction alone does not q_j/ 25 explain the average heat flew that is estimated under NEAL R. GROSS COURT h? PORTERS AND TRANSCRIBERS 1323 MHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.. - _. ....-~ - . ... . ..-_..-.-.-.-.-.- - .. . ... . -_- - . -.
71 1 Yucca Mountain which is something like 40 to 50 milliwatts 2 per meter squared. O 3 If you take the temperature gradient in 4 Topopah and you measure that gradient and you use the 5 thermal conductivity that we have measured and you , ! i 6 multiply the gradient times the thermal conductivity to 7 get the heat flux, that is about 25 milliwatts per meter 1 i 8 squared which is not much considerably less than the 40 to l 9 50 that has estimated to go into Yucca Mountain. All of 10 this is, of cource, uncertain but these are the latest-11 values we have. 12 So there are two explanations for this, and 13 John Sass brought them both up in his report. One is that 14 there is water percolating through the mountain and some 15 of the ene.rgy is transferred just by conduction and the , 16 rest of it is just by heating that water up as it goes 17 down because we have to heat it from 18' to 30o at the 18 bottom. That takes energy. 19 The other one is that no, this is not due to 20 percolating water and the other explanation is that this 21 is due to gas convections, that gas is coming in here and 22 is dry. It picks up water here because at the higher 23 temperature the solubility of water and air is like three 24 percent. Then it moves up, but then it has to give up () 25 water here because the solubility of water at 20* is only NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
1 72 1 one and a half percent. So you have condensation and you 2 have evaporation and since the latent heat for water is 3 very large, it takes a lot of energy to convert water to 4 steam, as we all know when we put the pot on the stove. l 5 You get a lot of energy from a very small flux. It l 1 l
- 6 actually only takes like .2 millimeters per year in i I I d
7 effective water flux to get this deficit in the heat flux. ; i ! j 8 Now, that sounds like that's a very reasonable . l 9 thing because it's maybe much less than we need for water j 10 flow. But one thing is that since the solubility of water , i 11 in the gas phase is only about one to three percent, you ! i 12 need tremendous amount of gas flow to get the required l 13 amount of water to condense. And Sass estimates like 50 , O l l 14 meters per year. That means you cover this whole distance , l l 15 in about 40 years. That sounds a little high. ! 1 i 16 So we haven't looked at both of these cases in 17 details but I'll give you little examples about what 18 happens if you assume this model and what percolation flux 19 would you need to explain the temperatures at Yucca 20 Mountain. That gets to this viewgraph here. This is a 21 study that Ed Kicklis and Joe Rousseau did at Yucca 22 Mountain looking at very shallow bore holes. They were . l 23 interested to know if the infiltration and percolation l 1 24 close to the surface is close to the rich tops or close to () 25 the middle of the washes. So they started to look at with NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1J23 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433 , i
1 l 73 1 a fixed heat flow different infiltration rates. You see . 1 l fs s 2 the higher the infiltration rate, the less the gradient in \ ! \ l
\' '/
3 the Topopah, as you see here. 4 When you take a look at the temperature l 5 gradient in the Topopah, which is this one here that we 6 looked at globally, you find to your surprise that all of 7 the bore holes in the center have about the same gradient 8 which is roughly 16 to 18 degrees C. per kilometer, and 9 that corresponds to some five to 10 millimeters per year 10 flux. So it's very consistent. I was very surprised when 11 I saw that. I expected some bore holes to be high, some l 12 bore holes to be low, but the fact that they are so l l l 13 consistent gives some credibility to this in some sense, j e~s I s i kl 14 shape or form. 15 So the other thing that makes this a little 16 appealing that this may be in the right ball park is when 17 you compare this map that we had here, look at this map 18 here in the blue, and you compare this to the calculated 19 percolation flux based on Allen Flint's map using our UZ 20 model at the repository horizon. And you have here 21 roughly eight millimeters per year in the same area where 22 we estimate from temperatures roughly five to 10 23 millimeters per year. So it's certainly on the same order l 24 of magnitude. r^N l () , 25 So I would add here then to this one here and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
74 1 let me see, let's make it a circle. That's easy because 2 it doesn't do anything. Let's make it one that has a hole 3 in it and put that right here. Now, there are some things l
- 4. here that we have left out. All of us have opinions about l 5 percolation flux. Some thinks it's very low, some thinks 6 it's higher, 7 This seems to send a message that -- the 8 viewgraph is crooked -- seems to send a message that a lot 9 of this information is starting to converge on something 10 in the range of one to five, one to 10 millimeters per l
11 year. Then you have to say is this consistent with: a) j 12 there's no flow into the drifts that we see in ESF, b) the l l 13 depositional models of flacta coatings and other 1 l > l -/ 14 evidence. I remember you asked that question. What l 15 supports this and what doesn't support this? So we need 16 to first look at the implications and then look at the ! 4 17 testing program. I'm going to do that in just a little 18 bit. 19 So this is the calibration and this seems like 20 where we're heading but I can't emphasize enough there are 21 uncertainties in all of these methods. We are moving 22 toward something that seems to give us something that 23 looks a little bit more consistent than it had in the 24 past, but our work is not done. 25 MEMBER HINZE: Bo, may I ask you a question? . NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
75 1 MR. BODVARSSON: Sure. ,-, 2 MEMBER HINZE: There are percolation fluxes i I 3 and there are percolation fluxes. l 4 MR. BODVARSSON: Yes. l 5 MEMBER HINZE: How much are you mixing 6 different types of percolation fluxes here? What is 7 anisotropy? How is anisotropy entering into this, 1 8 etcetera? 9 MR. BODVARSSON: This percolation flux here is 10 the total flow of water that goes vertically through the 1 l 11 repository horizon. 12 MEMBER HINZE: This is the bulk? l l 13 MR. BODVARSSON: Bulk average global flow of Os s' 14 water, and this is a very good question because when you l 15 look at this graph here, this tells you a lot more than I 16 have said already. It tells you other things, too. It 17 says if you believe this and you have all here close to 18 Ghost Dance Fault that show the same thing as outside the 19 Ghost Dance Fault, so it indicates maybe you won't have 20 the lateral flow. Maybe the faults don't act like drains. 21 Actually, the UZ-7A which has a very good temperature 22 record from Joe Rousseau doesn't have any smaller 23 temperature gradient than the rest of it. So it says a 24 lot about that.
/~'N
( ,) 25 But to answer your question, all of these NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
1 l l 76 1 methods give you kind of a global vertical flow and that 2 takes into account the anisotropy because we are not 73 ( ) 3 calculating anything. We are inferring a mass flow. The 4 temperatures, for example, don't care if the flow is in 5 the fractures on the matrix because you just have to heat 6 the water. Does that answer your question? 7 MEMBER HINZE: It does. I'm wondering. From 8 the summaries that you have made, would you reach any 9 conclusions about setback distances of the repository from 10 faults? 11 MR. BODVARSSON: I'm going to get to that in 12 just a little bit. 13 MEMBER HINZE: Great. I3 s# 14 MR. BODVARSSON: Let me eee if I do and if I 15 don't, ask it. 16 I want to now talk briefly about percolation 17 flux studies, what does this mean? I don't want to throw 18 away the low fluxes yet or the high fluxes yet because I 19 still think we need to keep both and do the testing we are 20 planning to do to try to see which ones are the correct 21 ones. So I'm going to start these studies showing you a 22 little bit what happens in the mountain if you have low 23 fluxes. This shows you basically the model and now I'm 24 going to put only like a .1 millimeter per year on top of
/i
(_,/ 25 the mountain and look at the lateral flow and the effect NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
77 1 of faults and all of these things and let's see what we 2 see and then I'm going to put the higher flux like we O 3 inferred and then we are going to look and see what we 4 see. 5 So as you know, this is a three dimensional 6 model and it calculates saturation, moisture fluxes, gas 7 fluxes, heat fluxes and all of those kind of things. This 8 is just a cut through the mountain close to the Ghost l l 9 Dance Fault and shows the effect of the faults being dryer 10 and then some wetter areas in the Topopah and the -- l 11 units. l 12 You also can look at water saturation at the 13 repository. If you have a uniform .1 millimeter per year 14 at the surface and you don't get the uniform .1 millimeter 15 per year at depth, that means you have lateral flow, and 16 you see that close to the fault. You have high l 17 saturations close to the faults. You can see this even 18 better when the water is entering the water table because 19 that's the maximum diversion that you can have. You may 20 also have diversion in the Calico Hills and you see this 21 very distinctly. You he."e areas with practically no flow 22 entering the water table. You have areas with rather 23 large flow entering the water table, if you call this 24 large. This is our low flux case. This might be .4 25 millimetera per year. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBER 3 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
I 78 1 One thing to remember here. The geology is 2 very complex. The permeability contrast is extremely O) \ 3 complex so that this may show you an area or correspond to i 4 an area where the permeable regions -- for example, the 1 5 vitric parcel, the Calico Hills or the Prow Pass -- enters 6 the water table because most of the water will go through 7 the more permeable regions. So this shows very much 8 lateral diversion. If you look at this in a little bit 9 more detail, in a cross section, to see where the water 10 goes, to see if we have lateral diversion or not, let's 11 look at cross section like around UZ-14. East-west cross 12 section I think I have here. I'll only give you one. We 13 see the following. O k- 14 We calibrated all the rock value properties 15 for a low flux now, .1 millimeter per year, and if we have 16 a high flux we recalibrated all the values for a high 17 flux. But this shows you there's a potential for a 18 tremendous amount of lateral flow in the PTn under low 19 flux conditions and tremendous potential for flow into 20 faults laterally like has been hypothesized for low 21 fluxes. 22 There's also, we saw in the simulation, a lot 23 of lateral flow on the top of the zeolites in the Calico 24 Hills that suggests that the flow patterns below the O (_j 25 repository are extremely complex because you have low NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 37C1 (202) 234-4433
79 1 permeability zeolites, high permeability vitric rocks and 2 then you have faults and fractures moving through those. 3 Now, jumping right back to the high flux, and 4 I'm not going to show you a lot of details in that because 5 we don't have time. I'm going to-just show you one l l 6 viewgraph there and this is now just assuming Allen l 7 Flint's infiltration map that averages roughly four and a 8 half millimeters per year at the surface in the repository 9 horizon and something like eight millimeters per year 10 after you had redistributed all of this at the repository l 11 horizon. I'll take a cross section, this one here, to 12 show you a little bit different geology with different l 13 faults. This cross section shows you the following i 14 vertical flow. 15 It shows you for the rock properties that we 16 have estimated for the model results that we are getting, l l 17 if the flux is somewhat higher than one millimeter per 18 year, there doesn't seem to be a lot of lateral flow in i 19 the PTn. There is some redistribution of flow in the PTn 20 because of the variable infiltration on top, but that's 21 what adjusts the infiltration between zero and 14 l 22 millimeters per year to roughly a uniform seven or eight 23 millimeters per year in the repository horizon. 24 What does this mean? If there is no lateral ( k 25 flow here, there is no drainage in faults above the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE.. N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
80 1 repository regions. May not be any more water going 2 through faults than there is going through the repository b v 3 region. The temperature data from UZ-7A, if you believe 4 it and you may choose not to, suggests that's the case, 5 too, because the temperature gradient in the Ghost Dance 6 Fault for UZ-7A is not any different from that of the 7 surrounding wells. 8 We will get a lot more data on this, of 9 course, when the Ghost Dance Fault -- we are now 10 approaching the Ghost Dance Fault in the northern alcove 11 and when we are having a temperature hole going towards 12 the fault, that should give us tremendous temperature data 13 to say if the temperatures has anything to do with 14 percolation flux and to say if the temperature analysis we 15 just presented is any good at all. 16 The other interesting thing -- 17 MEMBER HORNBERGER: Can I just ask a quick 18 question? 19 MR. BODVARSSON: Certainly. Ask any question 20 you want. 21 MEMBER HORNBERGER: Under this scenario, you 22 said that there may be no difference in flux in the faults 23 as in the matrix but you still then have to explain the 24 presence of bomb pulse, so you're saying that the rates () 25 are faster, the movement rates are faster even though the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
'(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
81 i 1 fluxes are the same roughly? 4 I 2 MR. BODVARSSON: Yes. Exactly. That's -O 1 3 exactly the point. The defaults just allow you to go , 4 through the PTn and that's what they allow you to do. 5 Since the saturations are smaller in the faults, you know i I 7 6 the velocity depends on the fault velocity and saturation, 7 you can get it very, very quickly down there. } 8 MEMBER HINZE: Bo, while you're stopped there, 9 if I may ask a question in terms of these hydrogeological i I
- 10 units. Do you assume the consistent properties within j 11 that unit or are these changed with each one of these 12 cells?
i l 13 MR. BODVARSSON: We have two approaches for 14 this. Mostly what we use is a uniform property for each , 15 layer. The reason for that is we get our rock property i 16 data by getting pure measurements from Lori Flint on the l 17 matrix properties. She tells me she has done 20 samples
- 18 in this unit. The average is this. The distribution is 19 this. And the same for all the other units. Then what we
- 20 do, we go through an inverse modeling process to match l
i 21 simultaneously all of the wells with moisture tension and 22 saturation data and we do them simultaneously because we l 23 want to see how consistent they are. 24 From that, we get a single value of () 25 hydrological prope2 ties from all of the layers including NEAL R. GROSS 5 COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE,, N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 4
_ _ _ _ . . _ _ _ _ _ . _ . . . _ _ _ . ~ . . _ _ . _ . _ . _ . _ . _ _ _ _ _ _ _ _ . _ _ _ _ _ 82 1 fracture and matrix properties. This is for an assumed 2 infiltration rate. Then we go and match individual wells O 3 just alone with no effects from another well and then we 4 get the variability of the hydrological parameters for 5 each of the layers. I know this is a little complicated. 6 MEMBER HINZE: I'm following you. 7 MR. BODVARSSON: Are you following it? 8 MEMBER HINZE: Yes. What would this look like 9 if you did a sensitivity study looking at the bound, for 10 example, of Lori Flint's measurements? 11 MR. BODVARSSON: Bounds of? Excuse me? Oh, 12 bounds of the hydrological parameters. 13 MEMBER HINZE: Yes. What would it look 'like? 14 MR. BODVARSSON: That's a good question. In 15 our report we have only one case with this latest 16 infiltration map because we just got it in July 17 unfortunately. We have since then done two other cases. . 18 If I can summarize, what we find is that the hydrological 19 property variations, even though you vary them by a couple 20 of orders of magnitude, the permeabilities are still high 21 enough to allow flux to go through most of the region of 22 interest so that the redistribution in the PTn, you still 23 won't have any lateral flow in the PTn. You will have 24 mostly vertical flow. 25 But the difference may be that in some cases NEAL R. GROSS COURT REPORTERS AND TRANSCRIBEP,3 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
83 l I
. where you have higher matrix permeabilities -- for 2 example, in a repository horizon -- you will have higher O 3 matrix flow there rather than fractured flow. The ratio 4 might be 90 percent matrix flow and 10 percent fracture 5 flow or if you use other parameters, 90 percent fracture 6 flow and 10 perc ent matrix flow. But the global 7 percolation flu:t is going to be the same. But this has 8 major impact on flow in the drifts, as you know.
9 MERBER HINZE: Yes. 10 MR, BODVARSSON: So I'll get to that a little l 11 bit. It's a good question. 12 fic what does all of this mean if we summarize 13 this? Now, ;his means that we have an alternative 14 conceptual nodel. We have the Montazer and Wilson model 15 that we have been using for many years with the lateral 16 flow and th:3 effect of faults and all of these kind of 17 things and nobody is saying this model is no good. We are 18 saying we r.eed to investigate other models. 19 The new model says that we have a lot of 20 vertical flow. We have higher infiltration than we have 21 in the past, we thought we did. We don't see a lot of 22 lateral flow above the repository horizon and, therefore, i 23 we don't expect a lot of focus flowing faults. The j 24 localized flow through the PTn will give us the chlorine-
) 25 36 signal in most of these regions.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234 6433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
84 1 Below the repository you have a different 2 story. You have a very complex flow regime. You have the
- 7. S e' <
/
3 very tight zeolites where you may have lateral flow going 4 all the way to the water table if they are tight enough, 5 and this is where we see pursed water in many cases. You 6 have then in some cases holes in the zeolites where there 7 is a high permeability vitric zone that may allow the e water to go rapidly through those to the water table, and 9 then you may have intersection of faults that again allows 10 quick flow there or you may have flow through the 11 zeolites. So we don't have a lot of data from deep, but 12 it's fascinating and it's interesting. 13 Now, what are we going to do about these? !O
~- 14 What are the implications? The implications are going 15 back to our resaturation strategy. The important thing 16 here is seepage into drift, waste environment, waste 17 mobilization, radionuclide transport, all of these.
18 Looking into those two briefly. 19 Seepage into drifts. There's a recent study 20 that chem Fu and others did at LBL looking at seepage in 21 the drift through PA, through Abe's work and Bob Andrews 22 and some others. They took a permeability pattern from 23 the heater test area and they construed a highly variable 24 permeability pattern with heterogeneities and all of these ('m ( ,) 25 things that people do that know how to do this. I don't NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 85 1 1 know how to do this, but somehow they come up with blue !
- 2 and they come up with red different places. This is ~'
3 great. 4 Then they impose fluxes on this thing and they 5 want to see does this thing go into the drift or not? 6 That's what the impcrtant part is. They started this year 7 with PA, this work, but I think it has great relevance. 8 They use a fairly large pulse of 28 millimeters per year 9 pulse. They let the water go through this heterogeneous 10 rock. They have 100 percent humidity in this thing and, 11 in this case, it doesn't want to go. It wants to go 12 around. 13 Let's take another case and do it 10 times I e~si \/ 14 larger. If you do it 10 times larger, it goes in. You 15 have dripping into there, but that's a large, large pulse, 16 and also the reason they say this happened is because of 17 heterogeneities. You get this low permeability here that 18 kind of chokes it. Now, this looks promising but we 19 haven't look at discreet fractures. We haven't looked at 20 the thermal effects here and the thermal effects on that. 21 But I think this is something that needs to be looked at 22 and may help us if the flux is high. 23 Finally, what are the implications for 24 thermal? I'll give you one example. This is again from ,s (_) 25 the model. If you take a cross section north to south NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
86 l 1 through the repository horizon and you look at thermal
,7 2 loading in this cross section and your repository, this is S
N,_/] , 3 the geological layering in this area. This is your ' 4 repository right here. This is the Calico Hills. You see 5 towards the north closer to Calico Hills. The water table 6 is here and we model a little bit below the water table to 7 allow for the right bonding condition for heat there. IE 8 you cut it off here, you have a constant temperature as 9 Bushek has shown many, many times. It may not be a proper 10 one. 11 We do this for five millimeters per year and 12 what do we get? We get this picture here. That shows 13 temperatures reaching 87 or 88 after 1,000 years and that
\- / 14 is below the boiling temperature. That's below the 97*
15 required to dry it up. That means for this simulation 16 that there's a fairly core simulation, you don't have the 17 dry out so you may have some problems with humidity 18 conditions for the flux. But now we need to look at in 19 more details into individual drifts perhaps will dry out 20 for some time. Tom Bushek has done that I think in the 21 NEPA report. I think he concludes that the flux can be 22 higher than five to get dry out in the drifts, but it can 23 not be a lot higher than that so we need to be concerned 24 to some extent with this. (r~3 ,/ 25 This last viewgraph before the testing thing. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
87 1 This is a dual permeability model and it's with the 83 2 kilowatt per acre that I think is the baseline right now. l O 3 This just shows that the matrix saturation is not drying 4 out. it hrscomes like 40 percent in the repository 5 horizon. Again, emphasizing this is a coarse model and 6 the drift scale detailed model will give us more reliable 7 results. 8 Now, I'm almost finished. How are we-going to 9 test for this? We now have indication flux is higher and 10 we hope in to be lower and there may be implications of 11 flow in the drifts and thermal and all of those kind of i 12 things ard there are tests in the books that are going to l 13 addresa this for us and hopefully help us discriminate 14 thece. 15 You asked Bruce and June before, what are you 16 going to do in chlorine-36? I think DOE has planned a lot 17 of work next year in chlorine-36, in fracture coatings, in 18 modeling, in other areas that we are doing things now. 19 We're certainly continue to look at the temperature 20 analysis to make sure that we haven't screwed up too badly 21 and pe2 haps look at the other alternatives in all of those 22 as weil as in the chloride balances. But these are the 23 tests that are on the books right now. The Ghost Dance 24 Fault testing is going to start I think end of the summer. 25 Is that right, Russ? Maybe early fall. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
88 1 The percolation flux test at the repository 2 horizon is in the book for planning this year and LJ 3 hopefully carried out as soon as we can next year. This 4 test is -- and I'll just show it very briefly -- is trying 5 to get a handle on the percolation flux for various things 6 and we haven't scoped out tota]1 7 what that's going t be. 7 Basically, the idea is percolation flux is an area 8 concept. It's an area beneath it. We need the global 9 value, we need the spatial variations, and we need the 10 temporal variation. We may not be able to get a. temporal 11 variation because of all kind of different issues. It 12 takes a long time to equilibrate. 13 But we are hoping perhaps to look at a case (D 'I '- 14 vith horizontal bore holes away from the ESF over 15 significant enough region like hundreds of mete:rs to look 1Chatthevariabilityandallofthepropertiesand d l '. parameters that we just discussed. So it's going to be 18 looking at this grid and looking at all these 19 fingerprints. That's the plan. Looking at the 20 fingerprints of fracture saturations, ages, matrix 21 saturation, chlorine-36, fracture coatings, temperatures, 22 gas permeabilities, mineralization, all of those. 23 There's a lot of things we are considering. 24 For example, freeze drilling might be one thing. It's not (<_,x) 25 very expensive. We never looked inside the fracture to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
89l 1 see if there's water there and with this freeze drilling, 2 which is not very expensive, we can maybe get a core that O 3 has a fracture and look at the water inside it. And 4 certainly we're also looking at alcove tests and things of 5 that sort. 6 Back to the testing before I conclude. The 7 moisture balance work in ESF has shown promise. It seems 8 like the conclusion from that is that the evaporation due 9 to the ventilation is about the same order of magnitude as 10 the construction water usage. So there's not an order of 11 magnitude loss one way or anocher. There seems to be some 12 drainage from the rock from the evaporation. 13 Rock matrix property measurements. I think 14 these are key. If the flux is high, we need to know in 15 details the matrix properties in the repository horizon to d 16 decide is the fracture flow 90 percent, 50 percent, or 17 lower to go into this flow into drift issue which is a 18 critical issue. That's on the book next year, too. 19 And then some lateral flow tests in the PTn 20 are planned. Limited ones and perhaps that will be looked 21 at some more to see if the assumptions of vertical flow or 22 lateral flow is the right one. 23 Let me just conclude before you can ask all 24 ji.'e r questions you want. The conclusions are various new 25 data and analysis suggests an alternative conceptual model NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
l 90 i i that results in higher percolation flux. The alternative l 2 conceptual model -- and I emphasize again alternative -- O 3 it de-emphasizes the importance of lateral flow in the PTn , 4 and the role of faults as drains above the repository 5 horizon. Faults may certainly be drains below the 6 repository horizon. 7 Flow paths below the repository horizon are l 8 very complicated, as we talked about before. The 9 implication of this higher flux includes perhaps enhanced 10 seepage into drifts and perhaps limited extent of dryout
.11 during thermal loading and, therefore, increased relative 12 humidity, hence, corrosion rates.
13 Finally, DOE has a testing program in place to ; O 14 start to look and hopind to discriminate against low or 15 high fluxes. 16 MEMBER HORNBERGER: Thank you, Bo. Do we have i 17 questions? 18 MEMBER HINZE: Bo, where do you stand in 19 combining your flow modeling with the transport modeling? 20 MR. BODVARSSON: The way we are doing this is 21 that the role of the UZ model is to calibrate against all 22 of the hydrological parameters including gas, moisture, 23 and temperatures and chlorine-36. Then we give the model 24 and all the parameters and results to Los Alamos, to Bruce 25 Robinson actually, that then uses that and does more NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
91 1 detailed calculations on the chlorine-36 issues as well as 2 all the transport studies. So basically, we don't do any O_ 3 transport studies. They do the transport studies. 4 MEMBER HINZE: Are there any problems in 5 meshing these programs? j 6 MR. BODVARSSON: That's a tough one. 7 MEMBER HINZE: There have got to be problems 8 certainly. 9 MR. BODVARSSON: No. I don't think there are 10 problems for the following reasons. There are problems
- 11 sometimes with putting in place the necessary milestone to I
l 12 do that and having the right data at the right time to l l 13 hand over. There's always this question, there's more l 14 data. When should we stop and when should we hand it l l 15 over? But they are gridding their own grids for transport 16 because you need different grids for different processes. 17 We have many different grids. One of the grids is for gas 18 flow where we really have a detailed grid among the ESF 19 because ESF pro rides the signal into the mountain and we ; l 20 must have a very detailed grid there. I 21 Another one is for the temperature where you I 22 don't need a detailed grid because temperatures are so 23 diffusive. The third one then is for the chlorine-36 and 24 the moisture balance. In the transport case, you need l 25 much finer grids to resolve fronts and much different l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.. - _ _ . _ _ . _ . _ . _ _ _ _ _ . _ . - _ _ _ _ _ - m-.__._ . _ - _ _ _ _ _ . _ . _ _ _
l 92 1 grids to resolve fronts. So they have to grid it anyway. l - 2 What we give them is the geological framework that we get l O 3 from the geological model and input on our model. We give ! 4 them all the hydrological properties from both the matrix l 5 and the fracture that they use directly in their model. 6 Then they do a benchmarking against our ; 1 7 results for one case to see that it gets basically the 8 same saturation profiles and the same moisturu tension l I 9 profiles and then they go ahead with their transport 10 studies looking at neptunium or technetium and all of that 11 stuff. l I 12 MEMBER HINZE: What are the major deficiencies 13 that you have in terms of moving your modeling to the , O-14 saturated zone? 15 MR. BODVARSSON: I don't personally believe i 16 that an unsaturated model should extend i0to the saturated 17 zone for the following reason. The unsaturated zone and 18 the saturated zone are very poorly coupled. They're only 1 19 coupled through those areas where you actually have flow 20 from the unsaturated zone into the saturated zone. We are 21 running these models with tens of thousands and probably 22 end up with hundreds of thousands of grid block for the 1 23 unsaturated zone to lock that together with another model 24 that has hundreds and thousands of grid blocks with a very () 25 weak coupling to me doesn't make much sense. NEAL R. GFK1SS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AV ~., N.W. (202) 234 4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
93 1 So, in the saturated zone model, you prescribe 2 a boundary on this at the water table that is basically l 73 b 3 the flux and the radionuclide source terms and then you 4 run with your saturated zone model. You can do that l 5 several times but to couple them you are carrying this 6 huge burden on the unsaturated zone that is for no reason. 7 The other answer to the question is saturated 8 zone modeling is not our responsibility. It's the l l 9 responsibility of USGS and they are doing regional and 10 local saturated zone studies. 11 MEMBER HINZE: Thank you. 12 MR. BODVARSSON: Okay. 13 MEMBER HORNBERGER: Bo, just one quick
- ' 1
's / 14 question. I assume that because of the time scales 15 involved that specifying a constant surface condition is 16 okay. That is, the pulses of infiltration, the temporal 17 pulses of infiltration aren't going to affect simulations 18 at depth. Is that correct?
19 MR. BODVARSSON: It's correct to some extent. 20 Another extent, not. It depends on how frequent those 21 are. We did, for example, a climate variation where you 22 have changes over, say, 10,000 years. That's shown here. 23 This is where we actually increased the infiltration by a 24 factor of three. That was considered reasonable by () 25 Cheryl, for example. And we watch a pulse moving through NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
94 1 the system and see how it affects things like lateral 2 flow, flow into faults, the drainage, the pursed water O 3 issues, and all of those issues. 4 In terms of climate change, of course you know 5 that's what you have to do, and we are doing that to the 6 dual permeability models because for pulses you have to 7 use a dual permeability model. For steady state ECM, it's 8 identical to do a permeability so you don't need to do 9 that. But if you look at smaller variation like if you 10 take 100 year floods and things of that sort, that's all
]
11 going to average out just like you suggested. ! 12 Lynn, aren't you going to ask a question? 13 MS. DEERING: Actually, I don't have a 14 microphone. 15 I think George asked my question. I had two , 16 questions. One was the implication of climate change on i 17 lateral flow and faults as drains with increased 18 infiltration and the second was to what extent would daily 19 infiltration affect your modeling versus averaging it out? 20 MR. BODVARSSON: That's kind of the same 21 thing. Daily variation. I think didn't Andy look at that 22 a little bit? Andy from Los Alamos looked at the daily 23 variation using Allen Flint's and he looked at it for 24 several years and it was no effect, if I remember () 25 correctly. It damps out. I think even 100 years NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
l 95 1 variations are damped out. When you get into thousands of l 73 2 years, it's not damped out. This simulation on the l i t ) i C/ 3 climate change, which is just our first try at it because 4 we have to do this before the June '97 milestone for TSPA. 5 You don't get again tremendously more important effects on 6 the fault except in the very tight units at Topopah I 7 Springs because then you start to get pursed water issues 8 there and the pursed water, because of the leaning of this i 9 thing, go into the fault and drain. I l 10 So you have those in some cases in the bottom l l 11 of the Tiva which is very tight. You may have lateral 12 flow occurring there as well as in some of the tight 13 regions of the Topopah. So the faults become more
/~N. ,
(\ -) 14 important. l l l 15 MEMBER HORNBERGER: We can infer from this 16 though that this may be quite important as you go to 17 interpret, say, the chlorine-36 where you're looking at 18 50,000 year time scales and, therefore, climate 19 variations. 20 MR. BODVARSSON: Absolutely. That's exactly 21 right. I think one model for the chlorine-36 may simply 22 be that you have the 10 year pulses that go into the 23 washes and they actually saturate part of the washers 24 where the alluvium is fairly thin and that gives you the (,
/^)x 25 chlorine-36 in one boom, in one pulse, and you may never NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
96 l 1 see it over the next 10 years, 20 years. That's a very l ,-~ 2 good point. 3 CRAIRMAN POMEROY: Lo, just as a point of 4 information, I think I missed your comment on EC models 5 versus dual porosity models at the very end there and I 6 wondered if you could just repeat it and expand on it a l 7 little bit, particularly applicability of EC models at 8 high rates. 9 MR. BODVARSSON: We did a study on the EC l 10 versus dual K to see where should we use dual K and where 11 should we use ECM for all of our studies. A lot of our l 12 studies are steady state moisture flow studies because we l 13 don' t have a transient. to compare to. We don't know if we (% / \~- ' 6 i 14 should run it for a million years or 10 million years so 15 we run it at steady state for convenience. The reasons 16 are the same anyway. So ECM and dual K are identical for l 17 steady state. There's no difference whatsoever. You'll 18 just take 10 times lcnger to run the dual K so why would 19 you do that? You shouldn't do that. 20 For trarsient moisture flow when you introduce 21 a pulse, then the fracture matrix interaction becomes more 22 important because you're following a pulse through the 23 system and then you need a dual permeability. That's why 24 we used that for the climate change model, the 3-D climate /^)' (_ 25 change model. When you have a transient gas flow, the gas NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
97 1 equilibrium times are only weeks so over distances of 100 fs 2 meters the gas equilibrates with the matrix in just like a V 3 week or two weeks so, therefore, the ECM is the 4 appropriate one but you have to use the total porosity. 5 You can not use just the fracture porosity. You have to 6 use the gas porosity. The matrix plus the fracture. And 7 then you get identical results. 8 For tracer transport you have to use a dual 9 continuum. For any transport you have to use a dual 10 continuum in most cases and you may even have to have more 11 than a single matrix blocks to resolve the gradients for 12 diffusion into the matrix. If you just have a single 13 block, you may not do it properly. 14 And then finally for thermal loading issues, 15 most of the ECM and dual K show similar results for 16 thermal loading except for the percolation flux which is 17 the condensate going down and that becomes a very critical 18 issue, especially late in the temperature rise if you have 19 large pulses of water. So that's why we suggest dual K 20 there in conducting our simulations. 21 MEMBER HORNBERGER: Any other questions? No. 22 Thank you very much, Bo. 23 MR. BODVARSSON: Thanks. 24 MEMBER HORNBERGER: We're very close to our 25 schedule here and we're now scheduled for a 15 minute NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005 3701 (202) 234-4433
98 1 break, so we will do so. 2 (Off the record for a 24 minute recess at O 3 10:48 a.m.) 4 MEMBER HORNBERGE'R: Okay. We are reconvened. 5 Our next presenter is Randy 3assett from the University of 6 Arizona. Randy. 7 MR. BASSETT: Thank you very much. 8 The title of my talk today as presumptuously 9 stated and actually assigned to me, Insights for Yucca 10 Mountain from Fracture Flow Studies at the Apache Leap 11 Research Site, Superior, Arizona. I have chosen this 12 particular display because it represents a sunset and, 13 based on our funding prognostications for the next year, 14 it's an appropriate figure. Hopefully yours is brighter. 15 Actually, ours is not that dim, but we'll see. 16 Here's what I would like to accomplish in the 17 time that I have this morning. I'd like to cover four 18 topics. Fracture dominated episodic valley floor 19 recharge, and this is work that's being conducted by two 20 Master's candidates at University of Arizona. Mechanisms 21 for perched water formation in fractured tuff. That's 22 Betsy Woodhouse, a Ph.D. candidate at University of 23 Arizona. Radiocarbon and tritium as indicators of 24 fracture flow. That's Greg Davidson's dissertation. And () 25 '"U/2"U systematics at the site. That's Ernie Hardin's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
. 99 1 dissertation.
2 obviously, in 45 minutes all five of these 3 individuals will be terribly embarrassed if I said I 4 covered their material in total, but it's simply to let - 5 you know that quite a bit of other material is available l l 6 if you have questions. i l 7 Many of you do not know about the Apache Leap 8 Research Site. Perhaps you've heard some of the talks in 9 the past or seen some of the literature. Our purpose at l 10 Apache Leap is ;o investigate generic, hydrologic, 11 geochemical and transport issues in support of NRC , 12 licensing. It's a field site that is supported basically 13 by the University of Arizona and the Nuclear Regulatory O 14 Commission. 15 We have three locations, three field sites. ; 16 We call them respectively the Deep Slant Borehole Site, ; i 17 which we'll talk about today, the Queen Creek / Haulage 18 Tunnel Site, which we'll also discuss briefly, and then 19 the Covered Borehole Site which is an area where low 20 permeability studies are being conducted and we will not 21 cover that material today. Personnel are listed there. 22 Students, faculty, of course. You may know Slomon Neuman 23 and Pete Wierenga, also faculty members at the University 24 of Arizona. Doctor Neuman is covering the air 25 permeability studies at the site. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
, , - - , .. - - , , - - . - - - - - - n
I 100 l l 1 Obviously, we have some products that we could l 2 refer to. Our project manager / project officer is Tom O 3 Nicholson. 4 Licensing issues. One of the reasons that I'm 5 here today is to simply point out some of the data that 6 have been collected at a site which is much like, I guess, 7 some components of the Yucca Mountain site under wetter 8 conditions. The obvious advantage of our site is that we 9 actually do see fluid moving through fractures in tuff, 10 and so we have the opportunity to do detailed studies and 11 develop methodologies and test and compare models at this 12 particular site. 13 So if, in fact, focus flow does occur at Yucca 14 Mountain and there are faults that you suspect as being 15 the primary pathway, what would you look for? What are 16 scme things that have been found that help us define this 17 particular pathway? Of course, we have a location where 18 we can test some of those concepts. 19 Licensing issues, of course. All this can be i 20 discussed at great length, but most of our contractual 21 responsibilities can be related to a KTI. We are bound, 22 of course, by what the licensing group needs to understand 23 so we are interested, of course, in uncertainty in 24 modeling unsaturated flow through fractured rock. We
) 25 would like to examine some of the assumptions. What's the NEAL R. GFH)SS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
101 1 appropriateness of some of the assumptions that we use 2 when we try to simulate flow through fractured rock? What O 3 about the data collection and if the data collection can 4 be done, can we then do a comparison of models? And then 5 last of all, what kind of isotopic and geochemical 6 modeling and data collection can you do to constrain these 7 results? 8 Let me give you a location map. This is in 9 the handout that's available. Hopefully you have a copy. 10 I can't show both of these at once, so I'll show the 11 location map first and then I'll show you the IR l 12 photograph. We're located east of Phoenix just outside of l 13 the city of Superior, a small town, mining community. So O 14 the city of Superior is located here. This is Highway 60. ; 15 If you follcw Highway 60 all the way across to the east, 16 you'll end up in Globe which is also a very famous Globe 17 Miami mining district. Copper mining is the primary mode la of activity in this particular region. 19 The Magma Copper Company, now BHP, has the 20 Superior Mine located here with their head frame at the 21 top of the escarpment. Very visible for many miles. The 22 head frame here is the terminus basically of what is 23 lovingly referred to as the Never Sweat Tunnel. It's an 24 ore haulage tunnel that allows them to haul ore from the () 25 elevator, the shaft, all the way down into the city of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
102 1 Superior. 2 Now, what's unique about this particular site 4 O 3 is that the tunnel passes underneath an intermittent j 4 stream called Queen Creek. So what we have discovered is 1 5 that because this is located in the rock that we're most I I 6 interested in, tuff, because we have a fracture network 7 which is episodically recharged and because we have access 8 into the tunnel, it's a really obvious place to conduct 9 tracer test, either natural or artificially stimulated 10 tracer test. 11 The length of the tunnel is approximately two 12 miles. It's about the distance from the entrance of the 13 ESF to the thermal testing alcove, so you think about that 14 distance, that's about the distance of the tunnel that we 15 have to work with. 16 We also have this particular watershed 17 instrument and we call this the deep slant borehole 18 watershed because we drilled a borehole in the watershed 19 at a 45* angle, and we'll talk about this in a few 20 minutes. The advantage to the 45o angle is that it I 21 transects all the vertical fractures which was an issue 22 that we talked about earlier this morning. What is a 23 chlorine-36 or carbon-14 or other isotopic variations 24 across a fault zone? If we drill vertical boreholes and () 25 the faults are vertical, then we very frequently have a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
103 1 1 difficult time characterizing the fractures. So that was I 2 the modus of operation at that particular site. ,i i 3 There's also an air permeability site which I 4 will not talk about today. 5 Now let me just show you the IR photograph 6 just to get you oriented. Here's the mine road located I l 7 there and you can see the mine road. So here's Queen l 8 Creek, here's the escarpment. It's about 400 or 500 9 meters above the city of Superior. Here's the mining I 10 activities. There's shaft 9. This is our watershed here. 11 We have an instrumented watershed at this location and ' 12 then the Queen Creek sampling is done at a variety of 13 places here in the creek where fractures transect the 73 ( ) N/ 14 creek and, of course, I'll show you a cross section in 15 just a few moments that identify how the fractures 16 correlate with the tunnel. 17 But let me just tell you first of all why we 18 chose this site a number of years ago. It's because in 19 the region if you talk to the miners, if you talk to the 20 mining companies,.if you talk to the water company in the 21 region, seeps have been historically observed in the Magma 22 Copper Company haulage tunnel. In fact, in several of the 23 drifts they see water periodically entering the tunnel. 24 The water level, the regional water table is extremely (3 s_f 25 deep. They've been dewatering the mine for many years, so NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
104 1 it's almost 600 - 700 meters below sea level now. So it
~ 2 is a very deep regional water table. So there should be O 3 basically no water in the mine except residual water from 4 the drainage or new water entering through fracture 5 systems.
6 So the regional water table is very deep at 7 this point and the points that we'll be sampling in the 8 tunnel are approximately 120 meters below the base of the 9 creek. So that's the 100 meter scale fracture transport 10 issue. So if you think about the distances from the Bull 11 Ridge Fault, for example, to the ESF or if you move 12 further down the ESF, we're looking at transport distances 13 in the range of, say, 40 meters to maybe 200 meters in a O s 1 A/ 14 fractured network, and this fits basically in between 15 those. About 120 to 150 meter transport distance. 16 If you rely on nistorical data, there seems to 17 be a correlation. The correlation is between rainfall j 18 intensity, rainfall duration, flow in the creek, and 19 seepage in the tunnel. The difficulty is that some 20 fractures flow. If you walk down the tunnel -- I hope 21 some day you can do this. We offer a workshop every 22 couple of years. But if you walk down the tunnel, you'll 23 see that there are many fractures. In fact, I was in the 24 tunnel yesterday, in the Esr yesterday, and I was just O 25 ( ,/ struck by the similarity. As you go through Topopah NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE,, N.W. l (202) 234-4433 W/.SH!NGToN. D.C. 20005-3701 (202) 234-4433
l 105 4 1 Springs, you see lots of fractures. You see a very dense, i
,3 2 welded material. If you look at the Apache Leap tuff, i n 3 many fractures, very dense. Porosities range from, say, a 4 low of about four percent to a high of about 15 percent.
5 Many fractures are seen circumscribing the tunnel. 6 But the obvious thing is that some flow, some 7 do not, some flow continuously, some flow episodically. l 8 So the question was: how can we unravel the chemistry of i 9 these fractures? Obviously, to do that would require I 10 continuous monitoring. j 11 Now let's look at this in three dimensions. 12 This diagram should be in your handout. It's a diagram 13 we've been using for quite a while, but actually we have a (D (/ 14 three dicensional model now that we'll be showing in the 15 future as soon as our funding comes through for next year. 1F Here is the valley. The Clean Creek Valley. And then 17 here is a representation of the fracture plane. If you 1 18 walk the creek, you can see multiple fractures. I 19 So what do we mean by fracture plane? That's 20 always the open question. Is it one fracture? Is it a 21 fracture nest? Is it a series of parallel fractures? Is 22 it a fracture that moves horizontally intersecting a flat 23 fracture that then maybe intersects another fracture 24 network? So we really don't know how to define that at f-( ,3) 25 the present time, but we do know that there are a series NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE lALAND AVE., N.W. (202) 234-4433 WASHINGTON, D.c. 20005-3701 (202) 234-4433
106 1 of fracture networks and when you see the tunnel, there 2 are places in the tunnel where water is entering the 3 tunnel that seem to correlate with the dip and strike of 4 the fractures that we can map on the land surface. 5 So first principles would be we can observe a 6 correlation prior to modeling. Now here's the tunnel. As 7 you move back this direction, you move out of the tuff 8 into paleozoic carbonate, so there's really not much of an 9 option to do work in this direction. As you move this 10 way, notice a very interesting feature. This is a very 11 steep escarpment. Here up on top of the escarpment, of l 12 course, we have shaft 9. We do find fractures in this end 13 of the tunnel, and I was really intrigued by some of the i 14 comments that Bo made because I think what we're seeing in 15 our tunnel supports many of his conclusions. 16 That is that there are basically three kinds 17 of flow that we're seeing in the tunnel. The first kind 18 of flow would be flow that is continuous, never changes, i 19 constant discharge. In fact, the chemistry is very 20 constant. The second kind of flow would be episodic where 21 you have flow for a while and then it stops, flow again 22 and then it stops, and we can correlate with the surface. 23 And then there seems to be horizontal flow that is perhaps 24 related to ' perched system so that the chemistry over a 25 large distance horizontally does not seem to change. The NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
107 l 1 manifestation of those kinds of flow is really obvious in 2 the tunnel. 3 Let me show you now some of the parameters 4 that we've monitored. The obvious question is going to i 1 5 be: have you done chlorine-36? And the answer is no. I 6 The reason we haven't done chlorine-36 is-that we have not i 7 had the control over the drilling of the tunnel like DOE 8 has had over the construction of the ESF. The tunnel was 9 drilled for other reasons than nuclear regulatory i 10 research. Consequently, they didn't use a tracer. They ! l 11 didn't consult us on the water chemistry and they have not 1 12 been terribly careful about the kind of gases that they've { 13 used in the tunnel and there's a train that moves in and ) O 14 out of the tunnel about six or seven times a day. So we i i 15 can't find a place in the tunnel that we would feel is not { 16 contaminated or compromised, that it's unambiguously 17 correct. 18 So what we have to do is rely on other 19 isotopes that we think are not affected by what's happened ' i 20 in the tunnel. Now, I would like to do a study of 1 21 chlorine-36, and we'll talk about that in just a.few 22 moments, but prior to that particular study, we looked at i L 23 a variety of other isotopes and water chemistry and let me 24 show you how the results have worked out. () 25 Here I'm plotting distance along the tunnel, NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
108 1 so this is distance from the tunnel entrance in meters so 2 we're starting at about 1,400. Thi's is where the tuff is O 3 first see to the end of the tunnel, about 3,000 meters. 4 I'm plotting on the ordinate on the left hand side 5 bicarbonate concentrations. Across the bottom I've also 6 designated obvious fractures that we can see that have at 7 one time or the other actually demonstrated flow, either a 8 seeping or a weeping of water into the tunnel. So we have 9 quite a number of measurements that we've made over the 10 last couple of years. 11 One of the really interesting things -- and I 12 won't take the time to go into the isotopic composition of 13 rainfall. We have all that data. The composition of the 14 creek. But one of the things that was striking at first is is that all of the fractures on this end of the tunnel 16 from the very end of the tunnel where it dead ends up to 17 the zone that we think is the primary transmissive zone, 18 the water chemistry seems to be very constant. In fact, 19 the mean for the water chemistry is really pretty 20 constant. The bicarbonate values don't seem to vary. 21 Flow is in some places continuous but others it does seem 22 to be episodic. 23 The deuterium values and oxygen values seem to 24 be relatively constant until you get to the fracture zone
) 25 and there we have two or three fractures that seem to flow NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
-. . . - . . - , . - . . . .- .~ . - . . . . --
109 1 large volumes. Perhaps as much as a couple of lit ers a 2 minute to the point where in other periods of time it's O 3 absolutely dry and there's no discharEe whatsoever. But 4 the interesting thing about the data is that if you look 5 at bicarbonate values, they vary tremendously from a low 6 of, say, 50 milligrams per liter to values up as high as, 7 say, maybe 150 - 180 milligrams per liter whereas you , 8 never see that variation in fractures further into the 9 tunnel. 10 Also, the environmental isotopes, the 11 deuterium and the oxygen. We see large variations in i 12 these values. If you go to the surface and look at the 13 concentrations, chemical concentrations in the creek or 14 the isotopic values in the creek in the rainfall, you see j 15 this very same kind of variation. But eery place that we 16 have measured perched water composition or compositions , I 17 from the end of the tunnel where we have continuous 18 discharge, these values are always constant and they seem 19 to be attenuating the isotopic signature that comes from 20 the surface. 21 So preliminary conclu sion is in the fracture 22 zone where we have rapid transmission of fluid, we reflect 23 in the subsurface the changing isotopic and chemical 24 values that you see on the surface in a matter of a rather
) 25 short time period. Other isotopes that we've looked at.
i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
110 1 For example, stable isotopes. How could stable isotopes : 1 2 help us if we're not interested in just the radioactive f~s V 3 suite? 4 Here's the plot of Boron w-11 values. We're 5 using Boron because there is an atmospheric signature of 6 Boron which is probably related to marine Boron. The w-11 7 for marine water is about 39 per mil. We find in surface 8 water that we have relatively heavy values of Boron and so 1 9 we see from the fracture network back into the tunnel, we 10 have values that are relatively heavy in the fracture 11 network and then they seem to move to values of here's 15, 12 seven, and nine. So these values toward the end of the 13 tunnel are closer and are certainly not anywhere in the 14 range that you would expect for rainfall. 15 Now, I won't spend a lot of time on each one 16 of these isotopes. The take home message and the key 17 piece of information here is that we see a difference in 18 the isotopic signature in the fracture network related to 19 what we consider to be rapid transmission of fluid. Rapid 20 flow paths, a focus flow, and I will show you in just a 21 moment, related principally to valley floor recharge. So 22- we have valley floor recharge. There's a difference in 23 the isotopic chemistry, stabilized isotopic chemistry, a 24 difference in the water chemistry. () 25 The other thing is I just want to show sulphur NEAL P GROSS COURT REPoRThrb AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-443's WASHINGTON, D.C. 20005 3701 (202) 234-4433
1 111 ' 1 isotopes. We've really not done much more with this than ,3 2 what I talked about before at a previous conference. But 3 clearly in the southwest the sulp:lur isotopic signature 4 has c hanged over time because of mining activities. l 5 Thers's smelting activities in Arizona. And so the value 6 that you see for sulphur isotopes is always, in this case, 7 more negative than you would find for the values that you 8 see in rainfall. 9 So, rainfall in the Tucson area and its region 10 has a value of about plus eight. Dryfall also in that l l 11 res; ion, plus six to plus eight. Very old waters, have a 12 valte of around plus six. And then waters that are 13 related to rapid runoff, rapid infiltration, drainage of k' 14 t1:e soil zone, anything that's happened in, say, the last 15 50 years, carries with it a much more negative isotopic 16 s::.gnature . 17 So we have without looking even at the 1 18 radioisotopes yet, we clearly see that there's a l 19 dif ference in the fluid in fractures that have been 20 1.ransmitted from the surface to the subsurface rapidly. 21 How do I know it's rapidly? Well, that's the advantage of 22 this site because we actually have liquid water to , 23 examine. 24 I just want to make one point about the rN (_) 25 changing climate. There was a question asked this morning NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
112 1 about climate and what difference does it make if you have f- 2 a lot of water available versus a drought year. Here V 3 we're plotting conductivity which, of course, is the 4 easiest thing to measure in the tunnel as well as 5 measuring it in the surface. Here I'm measuring rainfall. 6 This is rainfall values as a function of days since 7 January 1st. This is the very first year that we 8 collected data. It was a rather wet year. 9 One of the things that we noticed was that 10 whenever we had a rainfall event, there was a change in 11 the composition of the water coming into the tunnel at the 12 zone where we considered to be rapid transmission of fluid 13 but we saw no change, none whatsoever, in fractures that O)
\
N- 14 we considered to be draining perch zones. 15 Furthermore, the interesting thing was that 16 the fractures don't just drain and quit flowing. Even 17 though we think there is a direct connection between the 18 land surface and the tunnel -- in fact, I'll show you in 19 the next slide a very rapid transmission -- there is a 20 fracture network that seems to wet up, fill up, and then 21 drain very slowly. So you get a rapid transmission of the 22 primary pathway and then a very slow draining of the 23 network. So water gets there quickly, but it takes a long i 24 time to dissipate even though it may not be a large volume ( ,s) 25 of water. 1 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
113 l 1 So the only point I want to make here is that 2 it takes a long time. Over this entire period, the water - O 3 continued to flow into the tunnel. Our objective was , 4 first of all to mnduct what we considered to be a natural 5 tracer test and we looked forward to a time when the creek 6 would be dry and the tunnel would be dry so that then we 7 could initiate flow in the fracture network and evaluate 8 the travel time through a fracture network on the order of 9 100 meter scale. 10 So here's precipitation data for two different 11 years. This is March '94 through '95 and March '95 12 through '96 of this year. I just want to show you the 13 difference in rainfall just in this region. You see we O 14 have very frequent storms. Notice this is precipitation 15 in millimeters so this is not North Carolina. This is 16 Arizona. So we have a major. storm that came through and 17 just inundated us with 30 millimeters of rain a couple of 18 times. The rest of the time the rainfall is in the range i 19 of 15 millimeters. That was a wet year, j 20 The dry year, the current year, has been 21 really an interesting c3rcumstance. This storm that 22 happened 250 days into the year, so this is almost i i 1 23 December, I think it was, initiated flow in the creek but 24 it didn't last very long. Beyond that, the creek dried () 25 up. So the creek is dry now from about this point in time NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
114 1 all the way to this point which is approximately February l 73 2 of this year. Now this is important because I'm s O 3 interested in the break through or initial transport of 4 the initial arrival of water wetting up these fracture 5 networks. 6 I know you have bene part of the evolution of 7 thinking from uniform infiltration to focus flow over the 8 last few years. But when we started our work here, the 9 continual resistance that we received was fracture 10 networks will not transmit liquid water for 120 meters. 11 It won't happen, and that was because we were dominated by 12 the soil physics community and the University of Arizona 1 13 which says it's going to imbibe, it'll make it a short
,O
(_ / 14 distance, but it will imbibe and you will not be able to 15 transmit water that long distance. We were anxious to l 16 test the circumstance when the network was completely dry. I 17 This is what we have here. j 18 So this is in February and we're plotting 19 conductivity which is the easiest thing to me7sure at the i 20 creek so this is the influent water, water moving down the l 21 creek at a point where we think the fracture network is l l 22 expressed, and the water coming into the tunnel which l 23 would be the terminus of that fracture network. Remember, 1 l l 1 ! 24 it's been dry for six months so there's no flow in the 7m () 25 tunnel, no flow in the creek. Then you have the rainfall NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
115 l 1 ' event. Happens basically on February 1st about noon. We
-s 2 immediately have -- I mean within minutes you get flow in
() 3 the creek. The conductivity in the creek remains rather 4 constant for several days, so this is from the 1st to the I i 5 9th. And then it stops. So it was not a long event. 6i What happened in the tunnel? Basically on the 7 4th, which is two and a half days later, boom. Water 8 breaks through and the conductivity is a little bit higher 3 so it's apparently picked up some solute along the way, 10 and continues to flow for a long period of time. As a 11 matter of fact, it goes for several months. So you have a 12 short pulse of water in the creek, transmission of fluid 13 for over 120 meter plus network, water into the tunnel,
,/ \ \
K -] 14 and then the fracture network drains for a long period of 15 time. 15 Now, the question is could we initiate a 17 tracer test at that very same site under dry conditions? 18 The question would be what are the effects of antecedent 19 conditions? So we tried that. You have this diagram, but 20 we tried to locate the fractures that we thought were most 21 promising fractures for initiating the tracer test and 22 because we have lots of graduate students who have 23 unlimited time, we tried quite a number of fractures and 24 most of them failed. We never saw anything in the tunnel. (~h () 1 I 25 But we knew that this particular fracture was our highest NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
116 1 probability fracture. This particular one here. 2 I'll just give you the quick answer. If the O 3 antecedent conditions are dry -- and that was the case 4 here. It quit flowing, I think, some time in March and 5 dried out. No water in the creek, no water in the tunnel. 6 We initiated a tracer test and got absolutely nothing. 7 Wait a minute. I'm doing the sequence wrong. We did 8 another tracer test. We did about four or five tracer 9 tests. The tracer test I want to refer to is the one 10 that's shown here. Before the water in the tunnel or the 11 fractures ceased to seep or ceased to flow, we initiated a 12 tracer test in the creek so the creek is dry but the 13 tunnel is still flowing. So the. circumstances would be 14 this. 15 The network still has some water in it. We 16 don't know how much. We don't know the extent of the 17 network. Surfaces in some places are probably still close 18 to saturation. You've had water imbibition, of course, 19 from the fractures so the fracture surfaces are somewhat 20 wet. We added a tracer, something on the order of 2,000 21 gallons of water, traced water, and we monitored what 22 happened in the tunnel. Now, of course, water was still 23 coming through in the tunnel from the drainage of the 24 previous storm, the one that we saw before. And here's 25 what we saw. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE. N W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
-... - ~ - . - - .. - - . . - - .-. - . - - - - - . - .. .. . . .- 117 1 Clearly a break through. Here we see the 2 dates, approximately again a two day break through, at 1 O 3 this particular fracture, which is the one we were most 4 interested in. So we have two conditions now. We have , 5 dry creek bed, no drainage in the tunnel, rain storm wets 6 up the fractures, two day travel time, creek dries up, 7 still have drainage in the tunnel. We piggyback a tracer 8 test on the fracture and we get break through in about two 9 days in a quasi-wetted up fracture environment. , 10 Then we waited a couple of months and tried 11 another fracture test tracer test with twice the amount of 12 water in the exact same fracture and nothing came through. 13 We're still waiting. Waiting for another storm. Hasn't l l 14 happened yet. So you can't graduate. But we will, in l 15 fact, at some point in time allow this young man to 16 graduate as soon as he answers our questions. 17 So a couple of points I would like to make. 18 We'll come back to this fracture flow situation several 19 times. But the issue is valley floor recharge. Does it 20 happen? Yes. Does it happen over long distances? Yes. 21 Do antecedent conditions matter? Yes. Can you model it? 22 That's the phase we're moving into now. And I'm going to j 23 make some comments about that later this afternoon. 24 The answer to the question is you can model it () 25 just about any model you want because -- and this is the i NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 I t _ . _ , ., ,_.
118 1 point I'll make this afternoon -- is that in a tracer test 7s 2 you know the boundary conditions, you know the length, and U) ( 3 you know the break through time. So it doesn't matter 4 whether it's an equivalent continuum, whether it's dual 5 porosity, or whether it's dual permeability. We've 6 modeled it with a dual permeability model. Just fina, It 7 was no problem whatsoever. Used DCM 3-D. We've used 8 HydroGeoChem, George Ye's model. We used hydroflow to 9 model it and then coupled that to HydroGeoChem. It's not 10 difficult to model the tracer test once it's completed. 11 The problem is could we have predicted this a 12 priori? And the answer is no. We didn't know the 13 fracture characteristics. We do know the matrix
\/ 14 characteristics but we didn't know the saturations, we 15 didn't know the aperture or the density or the length or 16 the flow path or any of those other things. But in 17 retrospect -- and I can show you our successful modeling 18 which didn't impress me when he gave it to me and it won't 19 impress you when I show it to you. So I mean we can l 20 simulate break through curves. It's not hard really to 21 simulate two day break through curves in a fracture 22 network with a variety of different models.
23 In fact, you have in your handbook another 24 thing that I was intending to show you but I'm backing off l (_g) 25 on making a big deal out of this. We were going to show l l NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433 l
119 l 1 how fast the matrix wets up when the fracture is saturated
~
2 because the fracture saturation takes two days. So what lO 3 would the moisture distribution in the matrix look like? 4 Well, it doesn't matter really because the infiltration 5 rates or the imbibition rates are so slow the moisture 6 front just incrementally moves away from the fracture ' 7 surface and the point that Bo made this morning, why worry 8 about coupling unsaturated flow models to simulate a 9 saturated case, and this is basically a saturated case and 10 so we don't really accomplish anything by looking at the 11 moisture distribution which, of course, can be done in a 12 variety of models. 13 So in this very first series of experiments, 14 of course, this work is ongoing. Let me just give you 15 some conclusions that we have. First arrival of solute P 16 from intermittent flow in Queen Creek is detected in 17 tunnel fractures within two days. That's 120 meters. We 18 can see this in a tracer test. We can see this in natural 19 circumstances. Hydraulic connection. We know it's the 20 fracture. It's the right one because we can verify that 21 by isotopic data as well as chemical data and the 22 responses of the chemical data. 23 The thing I haven't really talked about much 24 and this will have to be for another time, the discharge () 25 we see in the fractures in most cases is mixed water. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-443'i
120 1 It's a mixed water from two sources. One would b the
,s 2 flow from valley floor recharge and perching which tends
('-) 3 to occur in association with these kinds of phenomenon and l 4 the perched water has a different water composition 1 5 because the perched water is fed not only from the 6 fracture but it's collecting water from the adjacent l l 7 escarpment which has a much longer travel time and a l I 8 different aater composition. So we can look at mixing 9 curves as we move down the tunnel fracture. 10 I have a lot of things here I could mention. 11 We've looked at del 18, del 11, del S34. They're all 12 consistent with the interpretation. We've looked at C-14. 13 It's post-bomb in the fracture. It's 3,000 years old in (h
\- # 14 the perch zone. Obviously different. But what's 15 interesting is that in the perch zone we do detect some l
l 16 tritium, and we'll get into that in just a moment. 17 okay. The last thing though, the last point I 18 want to make on this page which is I think a dilemma for 19 those who want to do predictive modeling in unsaturated 20 environments, and that is that how do you know if a 21 fracture will flow because a priori, I don't believe you 22 can tell that. I don't care how many parameters you 23 measure. It seems to me a difficult task to determine 24 which fracture is going to be the one that's going to (. (_) 25 flow. I mean I really don't think June would argue with NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
121 , 1 me. She's going to have to sample as many fractures as j f 2 she can to see what the distribution of chlorine is around O 3 those fractures to determine which ones actually flow. We 4 all have preconceived notions, but I'm afraid that i 5 geochemical data will be required to unravel which ; 6 fractures actually flow. 7 One final comment here on this particular ; i 8 section. Bo made this statement earlier about perch zones 9 and why he does not think that you're seeing perching at 10 the PTn. I don't want to misinterpret what you're saving , 11 but if there were perching there, you had significant 12 lateral flow, then I would think that there would be a , 13 manifestation of that. There should be many fractures 14 below that perching environment that have chemistry that 15 supports the perching environment. Wherever we have the 16 perch zone, we have fractures below that perch zone that 17 are draining because you have a constant head environment 18 and you have plenty of time. 19 This is no longer an unsaturated environment. 20 You have plenty of time for water to work its way down 21 through the microfractures and you see water seeping at 22 that end of the tunnel where we have the perch zone 23 continuously and you monitor that chemistry and it is 24 absolutely constant. But any place you have water that is () 25 dominated by episodic flow, then it tends to dry up and it l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
. _. _ _ _ _ _ _ . , l
122 1 has a different water chemistry. ! l 1 2 Okay. The bell doesn't ring here so I never 1 O 3 know when to stop. Fifteen minutes. Is that what I have 4 basically? ; 5 MEMBER HORNBERGER: I'll give you 20. 6 MR. BASSETT: I have obviously three Ph.D. i i 7 theses to talk about now. Mechanisms for perched water. 8 We'll go through this quickly. You have the material and, 9 of course, there are references that you can follow up on. 10 The question is perching has become very important. Let ! l 11 me just make a comment about perching. When we began our ; l 12 study at Apache Leap, the conventional wisdom was that l l 1 13 there will not be mid-tuff perching. If you do find l l () 14 perched water, it should be at the transition between l 15 strongly different lithologic boundaries. So when we 16 found perched water we were a little bit surprised and we 17 thought it might be a coincidence. So the question 18 becomes is there a correlation between the properties in 19 the tuff and the mechanisms of perching? 20 So I sent Betsy on a research project to 21 identify mechanisms of perching. Really, there are quite 22 a number. Let's look at this just for a moment. The 23 literature is clear that you can have perching with a 24 simple permeability contrast. I mean this happens in ( 25 limestones, it happens in shales. Happens all kinds of l NEAL R. GROSS ! l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433 l ,
123 1 places where you have a permeability contrast and the 2 ratios of the permeability are such and such. Some people ! lO I 3 say greater than 10, some people say larger. l 4 We also know that there will be perching where 5 you have impermeable zones. Yucca Mountain I think will ; l 6 bear this out. There should be perching above the base of 7 the vitrophere and we see perching or evidence of perching l
. I' 8 above our vitrophere which we have at the base of our flow 9 sheet. There also can be perching that occurs if you have 10 fracture transmission down to a zone and then there is a 1 11 weathering phenomenon which changes the permeability and 12 perhaps even some fracture filling.
13 So, where the weathering environment is i l O 14 extensive, you can have perching. You can have perching 15 where fractures terminate and then there tends to be a 16 zone where there's almost no fracturing of significance 17 and then fracturing begins again. Now, when could that I 18 happen? 19 Well, these could be separate cooling units. 20 In fact, the flow unit that we see at Apache Leap is 21 probably about 2,000 feet thick and we estimate there are 22 at least 20 different flows, but it cooled as one unit. 23 But you could have cooling of different units at different 24 times and different fracture scenarios. But there are () 25 more. I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 , l l
1 124 1 For example, you could have a fracture network ,- 2 like this where at some point deep in the section you b 3 begin to fracture fill. That's a geochemistry problem. 4 As water moves down the fracture, reaction occurs. 5 There's a reaction pathway. Finally at some point you 6 become super saturated and precipitate some minerals. So 7 the fractures terminate because there is an occlusion in 8 the fracture. 9 You also can have horizontal intensity of a 10 fracture. We see this actually at our site. We see major 11 fractures down to a certain zone. Then we see an increase 12 in fracture intensity and then we see a decrease in 13 fracturing below that. So this could be horizontal ) r 7"N 4 ( , k) 14 network, lots of fractures, intersecting other fractures ; 15 for some reason, regional versus cooling, changing ) 1 16 porosity in the matrix rather than fracture dominated. I 17 won't go through the details of these any more than this. 18 I think we've made our point. Then you can have a 19 residual perch zone from water that drains to another 20 location, drains and then leaves a residual perch zone 21 behind. 22 Let me show you one of the concepts at our 23 site. Why am I bringing this up? I'm bringing this up 24 because this is a generic research project and the r, 5.,_) 25 question is, am I going to find perching because already NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
125 1 we've identified perching as important, and where should I l ,-s 2 look? If we are myopic and lack only for perching at t i
~
3 stratigraphic boundaries, I think we may miss some 4 perching. This is mid-tuff perching. In fact, this is i 5 very high up in the tuff section. If you look at the . 1 i 6 properties, I have some thin fin sections there and you l 7 can look at the properties, there's very little difference i 8 aa you move down through the tuff in terms of the porosity 9 distribution or the degree of welding. l 10 But in our particular site, we think this may 1 11 be happening where you have fractures that intersect a 12 zone of higher fracture intensity and then we have a well 13 that's been drilled off of our property by Magma Copper (~) 1
\/ 14 Company. Their intention was to drill down through the l 15 tuff sheet and intersect expiration wells. But in the 16 process of doing this, they encountered a fracture network 17 below the perch zone -- by the way, they did find a perch 18 zone not too far from our well -- and the fractures all 19 below this zone down to a pretty significant depth are 20 filled with silica. So we do know then that at our site 21 we have a combination of perching mechanisms.
22 What's interesting is if you drilled a 23 vertical borehole and were looking at the changes of the 24 matrix properties, you wouldn't see it. It is only seen 1 r\ I () 25 here because we drilled the borehole, left it open, and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 l
126 1 1 there was standing water in the borehole and then we began J p~s 2 to do some studies to unravel that. So this zone has a V 3 pretty significant transmissivity because it's a fracture 1 4 network but it's supported by decreasing matrix porosity 5 and occlusions in the fractures. l 6 Let me just get right to the conclusions on 7 this section then. She'll be terribly disappointed that I i 8 didn't talk about all this. That's the life of a student, l l 9 isn't it? Disappointment. Get used to it. What I wanted l l l 10 to point out is that she did a pumping test, a very good 11 pumping test, and also a slug test, and she evaluated the 12 data using a variety of different analytical methods and 13 came up with some transmissivities which were absolutely 14 what you would be expecting for a fracture network. 15 So I'll just quickly go through some of these 16 conclusions. Potential mechanisms for perching have been 17 identified and we've eliminated several of them from our 18 site. We have eliminated, in fact, the most obvious one 1 19 which is perching between formations. We have not seen 20 evidence of perching between the tuff and the limestone, 1 21 but we see mid-tuff perching. We've looked at the 22 geophysical properties. We've eliminated some of the 23 models. She's done aquifer test analysis. We have seen, 24 if you look at the matrix properties, a very significant ( 25 change in porosity but it does not seem to be correlating NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W. (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
127 1 with the location of the perch zone. It's not 2i interconnected porosity.
'(/ 3 One final comment here. She did do some study 4 on the actual values of if you estimate some values for 5 hydraulic conductivity and then we have some plug flow 6 studies for conductivity of the matrix and she was i 7 estimating that we see nine orders of magnitude difference 8 between the conductivity of the fractures and the 9 conductivity of the matrix. So obviously, why should I 10 use an unsaturated flow model in those circumstances?
11 George, let me just sort of get to the 12 conclusions basically of the other two sections because I 13 think they're germane. 14 MEMBER HORNBERGER: Absolutely. 15 MR. BASSETT: Doctor Pomeroy, if that's all 16 right. 17 I'll just give you one or twc slides here on 18 location. Over in the Deep Slant Borehole site away from 19 Queen Creek now, we drilled this slant barehole to 20 intersect the fractures and our goal war to then map the 21 properties. We drilled it and cored it.. And what did we 22 see? Well, several things. If you lock at borehole 23 geophysical logs, since we know where the fractures are 24 and we know in some cases where water is entering the 25 borehole because some of the fractures do flow NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
,. . - . ~ . - - - - . - - . - . - - - - - - - - . . - - . . _ - - - . - - , - . . - 128 1 periodically, we saw really excellent correlations in a 2 variety of the logs. This will take too long for me to go O 3 through the details of that. Let me just summarize. 4 We saw correlations in the logs with the 5 locations of the fractures. We took the core material and 6 we looked at water content in a very detailed fashion down 7 to the core material and the water content variations 8 correlated beautifully with the geophysical logs. One of 9 the things that we observed was that because these 10 fractures are transmitting water in the recent past, the 11 distribution of moisture has not evened out so moisture 12 close to fractures is still present. So that is an 13 indicator of recent travel time. We see higher moisture 14 content in the core close to fractures. We see higher : 15 moisture content when measured with geophysical logs. 16 That correlation was seen. 17 But one of the most interesting correlations 18 that we saw was with the carbon-14 data. If you look at 19 this, here's a cross section of our borehole and this is 20 the orientation of the fractures. So we think there are 21 principal travel paths. If you looked at this as sort of 22 a map view, this is the direction of the orientation of 23 the borehole. This would be also the direction of flow if 24 it came down. This would be, think of it as valley floor (O
,j 25 recharge. These are just little ephemeral creeks that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
129 1 move down across the escarpment. Fracture orientation is 2 like this. O 3 So as the water rushes down through this 4 valley, there is some obvious recharge through this 5 fracture network. Flow is intermittent and it's very 6 brief. You might only have flow for a few days. But 7 there is apparently sufficient time for water to make its 8 way down these fractures and here you look at the depth. 9 Again, at this site, we're looking at a perched aquifer at 10 a depth of about 160 meters which is similar to the depth 11 of the tunnel. We have fractures that reach the aquifer l 12 and we have a borehole, of course, that penetrates the l l 13 entire section. 14 If you look at the C-14 data, the quick answer 15 is we have taken samples from several locations by a 16 variety of methods. June, this is what I was talking 17 about this morning. In the upper part of the hole where 18 we have significant weathering, we can squeeze the rocks 19 and get enough liquid water that we can do a C-14 analysis 20 on it. But we also in those zones have taken sections of 21 core and put them in a device for doing vacuum extraction. 22 I have a picture of it -- I think I've included that in 23 your handout -- which allows us to extract the C-14 from 24 the sample and the calibration, of course, was to () 25 calibrate this methodology in core in regions where we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
. _ _ -~ _ - _-_
. __ _ _ _ _ . . _ . . . _ _ . _ _ _ _ . _ . _ __.___...._.___.m . . . . . _ . _ .
130 1 also can measure C-14 by other methods and we get the same 7-s 2 number. U 3 So we know that even though fractionation can 4 occur with the stable carbon isotopes, regular carbon does 5 not fractionate significantly by this process. What's the l 1 6 advantage of this process? The advantage is that you can 7 do this carbon-14 analysis on core that has such low 8 porosity that you can not squeeze it. So you can not get I 9 liquid water from squeezing but you can do the vacuum j l 10 extraction and get a C-14 analysis. So by using ' 11 squeezing, free water, and vacuum extraction 12 methodologies, we compared all the methodologies up in the 13 zone where we had all three opportunities. O N/ 14 What we find, the short answer is every place l 15 that you find a fracture there tends to be a change in the 16 distribution of the age of the C-14 as a function of 17 depth. You would think this. In fact, you've seen data 18 like this from Yucca Mountain. Nice smooth curves for C-19 14 as a function of depth. We find disruptions, and 20 there's a whole dissertation written about it so I won't 21 spend a lot of time. Disruptions in the flow of the C-14 22 as a function of depth. 23 Now, one point I want to make here. Here at 24 this point which is right above the perch zone but it's at (~h ( ,) 25 a zone where a major fracture intersects the borehole, we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005 3701 (202) 234-4433
- - - - - - - . - _ - . . --- - . . - . . . - ~ . - . . . .~ -
131 1 get modern carbon. So modern carbon. That zone is i
- 2 corroborated by borehole geophysics, by changing water
'O l 4 3 chemistry, and by the presence of C-14. It would be great 4 for us to be able -- because this is a diagonal borehole, I i 5 hopefully next year we can do some detailed chlorine-36 ] 6 analyses maybe down the borehole and try to see if the 1 7 chlorine-36 values correlate with C-14. 1 i 8 I have another section but actually I think 1 4 4 9 I'm out of time. Let me summarize the last section by 10 just saying a few sentences. In fact, I didn't summarize . I 1 11 this. The C-14 section, Greg Davidson's work. We drill i a 12 the slant borehole. We have now drilled a second borehole 2 13 which is vertical and we saw the same thing in the 14 vertical borehole in terms of the property distribution i 15 and the presence of the first water table. We've 16 characterized the core. All that data are available if 17 you're interested in it. We've done numerous borehole ! 18 geophysical surveys. We've correlated all of the data i 19 with visible fractures and chemical and isotopic 1 20 compositions.
- 21 One other tool that we have used is we've used i
! 22 this COLOG tool to do a borehole geophysical. This is a j 23 video digital image of a fracture in our saturated zone l 24 which you can actually see and then reconstruct using a 25 Japanese sy_".em, REAX, which is a beautiful tool for NEAL R. GROSS , COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. I (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
132 1 identifying fractures.
-s 2 The last section which you can read and just %) 3 basically maybe we can correspond later, but the reason I 4 included the last section is that we also examined the 5 uranium-234, 238 ratio. The reason for that is that at 6 Yucca Mountain an observation has been made -- I think Zel 7 Peterman has done this -- that activity ratios of 234, 238 8 are high. They're above secular equilibrium which would 9 be one. So the question is, why is it that way and is 10 that an indicator of something related to travel time?
11 Well, we began to look at 234, 238 at our site 12 as well and we found the very same thing. We find in our 13 perch zone and at depth, we find 234, 238 activity ratios t'"s
-- 14 as high was six. The correlation is this. In zones where 15 you have rapid transmission of fluid in the shallow part 16 of the aquifer, the activity ratios are always two, 1.4 to 17 two. They're always very low, and that's because of 18 selective leaching, congruent dissolution and the removal 19 of 234 as well as the 238 parent. But deeper in the l
j 20 aquifer there is longer residence time and from about 20 1 21 meters down all the way to the perch zone, we see very 1 22 high ratios in the matrix. So we can correlate high l l 23 permeability matrix with low permeability matrix by 1 1 24 looking at the activity ratios. r-(_,8) 25 The final comment though is why is it that the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS i 1323 RHoDE ISLAND AVE., N.W. l (202) 234-4433 WASHINGTON, D.C, 20005-3701 (202) 234-4433 1 I
133 1 perch zone has high activity ratios? If you look at all fs 2 the other chemistry, the reason is matrix fracture ( )
\_d 3 interaction because the perch zone is driven by fracture 4 compositions but the isotopic ratio indicates that 5 significant uranium has been transported from the matrix 6 into the perch zone. So you have mixed isotopic 7 signatures which, June, we have to deal with all the time 8 where you have certain ages that indicate rapid travel 9 time in chemistry but then you have isotopic ratios that 10 tend to indicate low transmissivity. But the issue is 11 there we're seeing matrix fracture. It's a long story but 12 I'll stop at that point.
13 I just wanteo to cover four topics. One was (^h
\/ 14 rapid transmission over long distance, Queen Creek. The )
15 second one was perched mechanisms, where are they, where . l 16 should we look for them, how significant are they. The j l 17 third is can we use radiocarbon to help us unravel rapid l l 18 transmission of fluids through fractured environments? 19 The fourth, can uranium isotopes also help us? Yes, they 20 can. So I'll stop at that point. 21 MEMBER HORNBERGER: Thank you, Randy. Do we 22 have questions for Randy? 23 Randy, I'll start then while people get their 24 thoughts together. Of course, it's absolutely, totally em i ,) 25 unfair to ask you any question related to Yucca Mountain, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
134 1 but that's what we're interested in. ',-s 2 MR. BASSETT: That's right. l ' )
~'
3 MEMBER HORNBERGER: So that's what I'm going 4 to do. , l 1 5 MR. BASSETT: Okay. 6 MEMBER HORNBERGER: I'm curious about a couple 7 of things. First of all, you indicated that Apache Leap - 8 - I don't want to put words in your mouth but it's clearly I 9 wetter is what you said. 10 MR. BASSETT: Right. l 11 MEMBER HORNBERGER: Now, my question is is 12 this in your mind actually a fairly close analog for Yucca 13 Mountain perhaps under pluvial climate? (>s- 14 MR. BASSETT: I think that's the mind set we 15 have now. I think what we would say is if fluid actually 16 was transmitted, for example, down the Bow Ridge, when did 17 this happen? What was different about that study and 18 today? Did this happen in the past when it was wetter or 19 could it happen again in the future? If it were wet and 20 you had episodic recharge, here's the kind of things that 21 you could look for. 22 On the other hand, does it really have to be 23 wetter to observe this? You notice that we had recharge 24 through our fracture network under relatively drought (3 ( ,) 25 conditions. The only requirement is that you have ponded NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
135 1 water, in our case for about two days. If you can have <- s 2 constant head environment for about two to three days, 3 then you can wet up the fracture. of course, you need 4 sufficient volume to wet up the fracture. Then water 5 makes it through the tunnel. So I'm not so sure it has to 6 be a wetter climate. It has to be perhaps an intense 7 rainfall which allows for saturated conditions at the 8 surface. 9 MEMBER HORNBERGER: Okay. So actually that's 10 a nice seque into the second part of my question. I think 11 Russ told us yesterday that last year they had one to two 12 meters of water in 40 Mile Wash. Do you see, looking at 13 the site-- perhaps not 40 Mile Wash but Solitario Canyon, . r~N I 14 do you see the possibility for occurrences at Yucca 15 Mountain similar to what you've observed at Apache Leap? 16 MR. BASSETT: Boy, I'd hate to get out on a l 17 limb there. I think maybe it would be useful for the ! l 18 modelers in the DOE project to maybe try to simulate 19 what's happening at our site. What is the minimum i 20 requirement to see water move that far and are there 21 similarities between what we're seeing and what's in 22 Solitario Canyon? I don't really know what the parameters 23 are for Solitario nor do I know what the parameters are 24 for 40 Mile. O () 25 But what's interesting is that our NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i
l l 136 1 preconceived notion of the parameters doesn't really l
,f -
2 matter because water made it that far. So then if you do 3 the inverse method to try to unravel it, what is the 4 fracture permeability, what is the fracture aperture, 5 there's no unique solution. It's an infinite number of i 6 set. So I guess what I would say is if you looked at all l 7 probabilities of those combinations of parameters, would 8 Solitario or 40 Mile fit into that domain of 9 possibilities? 10 MEMBER HORNBERGER: Other questions? 11 MEMBER HINZE: Let me just try a brief one if 12 I may, Randy. This goes back to some of the work that 13 Greg Davidson has been doing and trying to decompose the
- 14 perched water into matrix versus fracture sources.
15 Question. How robust is that technique and how applicable 16 is it to the problems of determining magnitude at Yucca 17 Mountain? 18 MR. BASSETT: The magnitude of C-14? 19 MEMBER HINZE: The magnitude of fracture flow 20 versus matrix. 21 MR. BASSETT: Well, I think we've actually had 22 some conversations with Al Yang about this under official 23 sanctions environment at NRC/ DOE exchange. At technical 24 meetings we've talked about this and Al would like to use (~%x_) 25 our equipment. We would like to do some work on Yucca NEAL R. GROSS court REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
137 1 Mountain cores but we are actually prevented from doing 2 so. (g)
~
3 So to actually test the utility of this 4 device, we'd have to loan it, I guess, to DOE but I think 5 that would be useful or either have some waiver given to 6 us that we could take a look at some of the core where 7 perhaps Al has sufficiently squeezed it to get a C-14 8 analysis and then perhaps we could get segments of the 9 core above it and below it and see if this particular 10 methodology is applicable to Yucca Mountain core. I know 11 the red tape in that would be enormous. 12 I really don't know the answer to your 13 question. We've never been able to analyze any Yucca O k- 14 Mountain core and to my knowledge nobody on the DOE 15 project has attempted this particular methodology. 16 MEMBER HINZE: How about the other question? 17 How robust is this technique? There are a lot of 18 assumptions that you have to make. What kind of error 19 functions develop out of it? 20 MR. BASSETT: Greg spent a lot of time looking 21 at different solutions, looking at what would happen if 22 calcium carbonate precipitates. He's looked carefully at 23 the del C-13 values and we're pretty confident that -- I 24 think the value he uses in his dissertation is like is es (_) 25 percent modern carbon or something like this would be the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
138 1 upper limit of error. Even though the del C-13 can be 9 2 large, the C-14 value seems to be pretty accurate. O 3 MEMBER HINZE: Has this been used at any other 4 area other than Apache Leap? 5 MR. BASSETT: Not that I know of. We built 6 the system. 7 MEMBER HINZE: I would encourage whatever it 8 takes to get the cooperation to look at this. It seems to 9 me that this wou]d be an important parameter in terms of 10 the magnitude problem. 11 MR. BASSETT: The paper describing this is in 12 Radiocarbon. I think it comes out in a couple of months. 13 MEMBER HINZE: If we could get a copy of that, 14 pre-publication copy. 15 MR. BASSETT: Yes. I can give you a copy of 16 that. Sure can. 17 MEMBER HORNBERGER: Any other questions? 18 Okay. We're going to break for lunch. We'll take a one 19 hour break. We're 10 minutes late, so we'll reconvene 20 here at five minutes to one. 21 (Whereupon, the meeting was recessed at 11:58 22 p.m. to reconvene at 12:55 p.m. this same day.) 23 24 () 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i _o
139 l 1 A-F-T-E-R-N-O-O-N S-E-S-S-I-O-N i 2 (1:03 p.m.) [\ _h 3 MEMBER HORNBERGER: I think our PA system has l 4 been adjusted. We are ready to start this afternoon's 5 ser~ ion. This morning you will notice that we 1 i 6 co..centrated quite heavily on the hydrology aspects of 7 Yucca Mountain. 8 I should -- I just wanted to take 30 seconds 1 9 or a minute to basically remind us all of the objectives l l 10 of the ACNW. We are obviously mostly concerned about l l 11 regulatory issues, issues that may come to the Commission l l 12 relative to licensing. 13 Thus, we are very interested, obviously, in (
\_/ 14 scientific results, but obviously we are interested in the I
15 context of how, in fact, they may relate to a performance 16 assessment of the repository in the regulatory framework.
)
17 So this morning the -- just to briefly 18 summarize, we heard about percolation rates through the 19 repository area. Both the experimental results, the 20 measurement results, and the modeling results seem to be 21 pointing toward a convergence on bulk overall flow rates, 22 but it is clear that there are still uncertainties 23 remaining relative to the amounts of water and spatial 24 patterns of flow rates through repository, that obviously
/ 4
( j/ 25 will have to be resolved. NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. ?0005-3701 (202) 234-4433
140 1 These, obviously, become important with 2 respect to licensing, as they affect the movement or O 3 potential movement of radionuclides through the tuff and l 4 down to the groundwater. 5 And so this afternoon we are shifting gears a . 6 little bit. Next we are going to hear something about 7 topics in geochemistry, relating to the transport of the , 8 nuclides in the unsaturated zone, and later this 9 afternoon, we will hear about integrated modeling. : 10 Our first speaker this afternoon is Dr. Bill I 11 Glassley from LLNL, and Bill is going to talk to us about 12 geochemical effects in the near field. 13 DR. GLASSLEY: Thank you. Once again, I'm 14 Bill Glassley, with the Lawrence Livermore National 15 Laboratory. I'll be talking about the near field 16 environment and geochemistry program. 17 The intent of that effort is to define what l 18 the chemical environment will be that waste packages will 19 experience, as well as define what the chemical conditions 20 will be that the waste form, itself, will experience, and l 21 how those things, coupled together, ultimately translate 22 into developing a source term in the near field ; 23 environment. l 24 Most of our work focuses on understanding 25 processes and defining processes in a detailed way, and NEAL R. GROSS ; COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
____..__._..____.__-...-_..________.._____m -
.._.m _ _ _
141 1 then providing means for rolling that up into the PA L 2 effort, and feeding that information to the PA program. O 3 If my voice falters, it is because I'm going i' l 4 through a testosterone patch therapy, and it is affecting 5 my voice. No, actually I have a cold and it has been with 6 me for a while. 7 (General laughter.) 8 DR. GLASSLEY: Boy, you were slow, there. 9 What I will be describing first will be the physical 10 environments or regimes that we have to be concerned with, 11 which influence, ultimately, the chemistry. l l 12 There are four of those, one is the region in l 13 che immediate vicinity of waste packages, or the 14 emplacement drifts where the temperatures could be above l 15 boiling. The second regime is the boiling region, itself. 16 Third is the zone of condensation outside of that, and the 17 fourth regime, which is critically important, is the 18 engineered barrier system, because that ultimately is the 19 last physical environment water will see before it gets to 20 the waste form. 21 And then it is the first one it sees as it l 22 leaves waste packages to transport radionuclides out i 23 before its packages do leak. ! 24 Once I've described those regimes briefly, I 25 will talk a little bit about the chemical characteristics NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE,. N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
l 142 1 l 1 of each of those regimes. . l l
,- 2 This is a schematic representation of the i )
i's/ 3 kinds of regimes that may evolve, the three that I was 4 talking about before, minus -- and not indicating EBS. 5 Provided t the waste loading is high enough, 6 and/or the flux through the mountain is low enough, what l l 7 one would expect to develop within the near vicinity of 8 the repository, is a single phase above boiling region, 9 the temperatures have exceeded the nominal boiling point, 10 and we have a single phase, presumably hydrous phase 11 existing in the immediate vicinity of the repository. 12 This, by the way, is a cross-sectional 13 representation, down to about 600 meters below the ground ,
/^~% l
-- 14 surface, and out to about 3,000 meters away from the 15 repository center.
l 16 The distribution of that regime will, 17 obviously, depend upon a range of features. But what we 18 are interested in, primarily, is what will be the 19 characteristics of this thing, this environment that 20 control the chemistry. 21 Similarly, outside of that, there may develop 22 a two-phase boiling region, two-phase water vapor and 23 liquid water, essentially at whatever the nominal boiling 24 temperature is at that point, given the water chemistry.
,O
( ,) 25 And then, finally, outside of that region NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
143 4 l l 1 where condensation can form, as a result of whatever water l l l l 2 vapor transports away from this high temperature region, 7-i l
%J 3 into regions where the temperatures are sufficient to l 4 allow condensation.
5 This schematic representation is meant to show 1 l 6 how these two regimes can evolve through time, or these 7 various regimes can evolve through time, nominally 100 8 years and 1,000 years. The size of these, obviously, will j l 9 depend upon many of the characteristics of how the l 10 repository is constructed, how waste has been placed. 11 The above boiling regime is the region that 12 will probably evolve the most rapidly, as a result of 13 simply the high deposition rate of the energy in the rm i I
\/ 14 environment. It will be a complex mix, presumably, of 15 equilibrium and non-equilibrium domains, and we need to 16 understand the distribution of those.
l 17 Ultimately, though, the chemical l 18 characteristics of that environment are going to depend 19 mainly on evaporation and boiling effects. l 20 Now, the issue of equilibrium versus non- j l 21 equilibrium, really has a profound influence on how we do 1 1 22 any geochemical modeling in this environment. Clearly, if I l 23 it is a relatively slowly evolving condition and 24 equilibrium is achieved, then we can use equilibrium fm is ) 25 thermodynamics to model the system. But if not, then we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
_ _ _ _ _ - . ~... - _ _ ..--.-_ .. . _ _ - l ! 144 I l ' ! I have to deal with kinetics, and things become profoundly l 2 much more complex. I O 3 So one of the things we've attempted to do is l l 4 understand or determine the extent to which equilibrium l 5 may be achieved within that nearfield above boiling i 6 region. l 7 One way -- there are many ways to do it, but i 8 one of the approaches we've taken is to try to define, or l
\
9 take an approach from chemical engineering where one uses 10 what is called Damkoler group number to define the 11 relationship between the rate of a reaction, and advective- ! 12 transport. ! i 13 Now, the Damkoler numbers are dimentionalist O 14 numbers. In this case what this simple equation relates, 15 is the rate of a reaction, which is represented by these 16 various terms, to the distance over which the fluid is i 1 17 moving, equilibrium concentration of whatever species we 18 are interested in, and the velocity of the fluid that is 19 moving through the system. 20 These various parameters represent things such 21 as the -- I should back up a second. This relates 22 specifically to an individual reaction. It is necessary
'23 to define Damkoler numbers for every reaction one is 24 interested in, in order to determine whether or not 25 equilibrium will be achieved, or approached.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005 3701 (202) 234-4433
145 1 For very large Damkoler numbers, things on the 2 order of 1,000 or more, one can assume equilibrium is 7-N' 3 very, very closely approached, if not actually achieved. 4 For s:aall Damkoler numbers, less than ten or one or 5 something in that vicinity, it is questionable whether or 6 not equilibrium will be significantly approached. And for 7 much smaller numbers, clearly, advective transport will 8 dominate. 9 What we have done for one of the important 10 reactions that we are concerned with, in the system, that 11 being the dissolution and/or precipitation of silica in 12 the near vicinity of the repository, we have taken the 13 reaction rates that have been measured for both gS \~- 14 precipitation and dissolution of the various silica 15 polymorphs, considered surface area, which is what this 16 term is about, reaction stoichiometry, reaction kinetics 17 expressions, dependence on some chemical properties of the 18 reaction, in this case hydrogen ion activity. 19 And made the assumption that if we are going 20 to have equilibrium achieved, Damkoler number will 21 probably be close to 1,000, we can solve for, if we know 22 the velocity in the vicinity of where -- in the vicinity 23 of the repository. 24 We essentially know everything else except the n ( ,) 25 length scale over which the fluid would have to move, in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
l 146 l l l 1 order to achieve equilibrium. 73 2 So we've mapped that out in vicinities, in l
\v) 3 regions around an emplacement drift, for various waste
\
I l l 4 loadings. And that is what this result attempts to show, ' 1 1 5 this figure. What the results of these calculations are. 6 What is shown here are two different cross-7 sections through an emplacement drift, which is i 1 8 represented here, vertical scale is in meters above and 9 below the center of the waste package, and meters away 10 from the waste package. i l 11 For 25 years after emplacement and closure, it i 12 is the same closure of the repository that takes place i 13 immediately after emplacement, and 50 years afterwards. 7_. N/ 14 The intensity of these colors gives you an 15 indication of what the length is, over which the fluid ! 16 would have to move, to achieve equilibrium. For the most ) l 17 intense colors, for the particular scenario we are dealing 18 with, here, the fluid would have to move, and this is 19 considering fracture flow, now. Fluid would have to move 20 on the order of 10,000 meters in order to achieve 21 equilibrium. 22 Clearly, that is not going to happen. This is 23 a regime that is going to be dominated by kinetics 24 processes, non-equilibrium processes. We aren't dealing l'3 ( ,/ 25 with equilibrium thermodynamics, here. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 147 1 At 50 years after emplacement, you can see
,3 2 that the envelope has moved out, substantially, but we are l ]
3 still dealing with very large distances in order for fluid j 4 to achieve equilibrium. 5 The fact that -- so what is this telling us? 6 It is that probably within the near vicinity of a waste l l 7 emplacement drift, as far as geochemistry is concerned, 8 equilibrium is not going to be approached, or certainly 9 not achieved, for fracture flow. 10 So what we have, given that, we have decided, l 11 or taken the approach that in order to define the 12 chemistry of the system, what we need to understand, 13 primarily, is how the chemistry of the fluid will evolve (' 14 as a result of evaporation and boiling. That is going to 15 be the dominant control of what we have, what the water 16 chemistry will be in the near vicinity of a waste package, 17 as long as liquid water is present. 18 Water-rock interaction, for this, under these 19 conditions, will not be significant. What is shown here 20 is the result of one set of series of simulations we've 21 done, looking at water chemistry evolution, primarily, 22 simply looking at evaporation of water for two different 23 cases. 24 One case where the fluid that is evaporating (- (_j 25 is in equilibrium with the atmosphere, so we have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
148 1 atmospheric CO2 and O2 present, and that is represented by 7- 2 the box, the points that we've referred to as CO2 and O2 N-] 3 fixed, and those are the squares that, in this figure, 4 approach lower and lower EH values. 5 And the other case where we assume that the 6 system was not ventilated. In essence, the water, as it 7 evolved, would control the gas phase chemistry. And in 8 this case, and that is the case where CO2 and O2 are not 9 fixed. 10 We start out with water that is similar to J-11 13 water, which is water that has interacted with tuffs, 12 something similar to some of the pore water that has been 13 looked at, out at the repository. There are various
^- ,
t i \/ 14 samples from core collected in that vicinity. 15 For the case where the atmosphere is allowed 16 to evolve as heating and evaporation occur, there is 17 essentially no change in the redox state of the system, it 18 remains more or less oxidizing, or rather substantially 19 oxidizing. 20 As is also the case for the situation where we 21 have the water evaporating in the presence of atmospheric 22 gases. In other words, there is no condition we generate 23 reducing environment, where we are evaporating water. 24 As far as PH is concerned, the situation is (3 is '/ somewhat different, primarily because of the role that CO2 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
1 149 1 1 plays in generating carbonate species and bicarbonate j 2 species within the solution.
) l 3 In this case, where the solution evaporates in ,
l 4 the presence of atmospheric gases, the PH starts to 5 increase, and eventually reaches value on the order of l 6 nine and a half, so it is moderately alkaline, whereas in l l 7 the situation where the gas phase is not controlled, PH l 8 remains more or less around neutral conditions. ) 9 I should point out that what we are talking 10 about on the horizontal axis, here, is evaporation of 11 about a kilogram of water, of 55.5 moles of H2O in a 12 kilogram. So we took the calculations to about 97, 98 13 percent dryness.
/' $ b 'L / 14 As far as the aqueous species are concerned, 15 most of them don't have much of an influence on either 16 waste package lifetime, or radionuclide transport, but 17 some do.
18 In this case, looking at chloride as it would 19 evolve during evaporation, and the reason it is important 20 is that it has an influence on waste package corrosion. 21 During evaporation, in this case, chloride 22 increases substantially by about an order of magnitude a 23 little bit more than that from its initial about 7 ppm 24 value. This results from the fact that chloride is (_,/ 25 essentially a conservative ion. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
- - ~ - .-.-.--.. -------.
150 l 1 Until you reach saturation in salt or some 2 other salt species, chloride does not precipitate (::) 3 significantly. So, essentially what we do is concentrate l 1 l 4 chloride in solution to rather high values. l
\
l 5 Now, what we have -- that, essentially, !
)
6 defines the chemistry of water that would exist in 7 constricted pores or regions where water may not be able 1 8 to migrate significantly or is trapped during the heating L 9 phase, and gives us, essentially, an extreme point on the 10 water chemistry evolution curve for the region near the l l 11 waste packages within the rock. l 12 Exterior to that is the boiling condensation 13 region. It is a region where there is a complex interplay l l 14 between water and rock, and one of the things that -- and 15 what we have been looking at and focusing our attention l 16 on, is how water-rock intelaction will influence the water l 17 chemistry. 18 It is impornant to understand that in this 19 environment, the water that you are dealing with, 20 initially, that is generated by the process of the 21 repository evolving through time, essentially is distilled 22 water that condenses from steam phase, interacts with the 23 rock. That is a very reactive system that will result in 24 substantial rock-water interaction effects. ( 25 And I will show you some of the consequences NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
}
151 1 of that interaction in a moment. Nevertheless, for most 7-s, 2 of the environment, the water chemistry is dominated by i ) 3 the rock system, simply because the volume relationship, 4 the amount of water that is available is relatively small 5 compared to the total volume of rock that is there. 6 However, if there are high volume preferential 7 flow pathways, or other things that can concentrate flow, 8 then it is conceivable that the water will end up being I 9 the dominant element in the system. 10 The geometry of the regimes that we are , l 11 talking about is represented here, and the cartoon on the i i 12 left-hand side shows a cross section through the ) l 13 repository block, repository would be represented i r~s) '
\/ 14 approximately there, by that dark band.
15 And outside of it is this two-phase region 16 where boiling condensation would be occurring. On the l 17 right-hand side are shown some of the processes, i 1 18 schematically, that would be occurring within these i 1 19 various environments. 20 Above the repository it is possible what will 21 happen is that water that is vaporized will migrate 22 through fractures, in this band, through the middle of ; 1 23 this that is meant to represent a fracture matrix on l 24 either side of that. t (,,f 25 Vapor migrates up, condenses, is imbibed into NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. j (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
152 1 the matrix and flows down along fracture surfaces. This
,s 2 is distilled water. As it reacts with, or as it moves
/ ) \ / 3 along the fracture surface, it will react with the rock, 4 picking up solutes until it reaches the boiling front, 5 either evaporates or boils away, precipitates what it has 6 dissolved, and the cycle continues ad infinitum. 7 Below the repository it is essentially, it is 8 a similar process, in that vapor is generated and 9 condenses, but it is more or less a single pass system. 10 What I want to show you now are the results of 11 some reactive transport simulations that we've done, 12 looking primarily at the single pass system. 13 The simulator we've been using, we've used a /~% k_ l 14 number of them, but the one that we are currently working 15 most heavily is a code called GIMRT. It is a code that 16 was authored by Carl Steiffel and Sieb Yobassati, and has 17 its progenitors back in MPATH and 1DREACT, and a variety 18 of other reactive transport codes. It is essentially the 19 latest incarnation of that. 20 The code uses the E236 data base, essentially, 21 and models solution chemistry in the same way EQ36 does, 22 but in this case it also includes a flow component, so 23 that one have a homo -- heterogeneous physical framework 24 through which flow is occurring, a flow field is defined, /\ (,) 25 a temperature field is defined and transport through that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
153 1 is then modeled. 2 The simulations we've been doing have been 2-O 3 D, and the results I want to show you represent the 4 condition or a scenario in which water, primarily l 1 5 distilled water, condenses in a temperature regime of ) 1 6 about 95 degrees C, then flows for approximately five 7 meters at that temperature, essentially isothermal flow at 8 95 degrees C, reacting with the tuff, and then proceeds to 9 flow down a thermal gradient of one degree per meter, for 10 another five meters. 11 We simply wanted to understand what the ; 12 evolution of this kind of system would be. It represents 13 something that would potentially evolve in the near 14 vicinity of a waste emplacement drift. 15 We were interested in both the change in the 16 physical framework, and in the water chemistry. 17 Now, what is shown here is the horizontal axis 18 log time, out to 10,000 years, starting at more or less ! 19 zero. The volume -- this should say percent, not fraction 20 -- of various mineral phases that were either present 21 initially and which are cristobalite, feldspar, albite, 22 quartz, and anorthite, that is the initial system, 23 proportions of secondary phases, and the porosity 24 represented in the squares.
) 25 This is at the inlet or quarter of a meter in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4W WASHINGTON, D.C. 20005-3701 (202) 234 4433
154 1 from where the inlet actually is in this system. The s 2 important point that we -- there are a couple of important t ('-) 3 things that take place in here that influence chemistry, 4 significantly. I l 5 One is the fact that cristobalite makes up 30 l l l 6 to 40 percr.nt of the rock in a repository horizon. 7 Cristobalite is the metastable silica polymorph. j i 8 Distilled water interacting with that dissolves it very l 9 quickly and thoroughly, and that evolution is shown here. 10 So within about 100 years, a few hundred years i 1 11 more or less, cristobalite is essentially dissolved out of l l 12 the system, for the flow rate we are talking about here, l 13 which is one millimeter per year, j
\/ 14 The quartz, however, which is another silica 15 polymorph, it occurs originally in the rock, but also as 16 cristobalite dissolves, begins to precipitate, and you see 17 it increase in abundance.
18 Its molar volume is less than that of 19 cristobalite, therefore, what we also see is an increase 20 in the porosity of the system. 21 Eventually, though, we reach a point where 22 because of water-rock interaction, the quartz itself 23 begins to dissolve away, and at that point the porosity 24 increases dramatically, in this case, reaching values of
) 25 about 70 percent by the time 10,000 years is reached.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
155 1 Remember, this is the inlet of the system 2 where distilled water has been interacting with the rock 73 (
'~') 3 for a long period of time.
4 Five meters along that flow pathway, the 5 evolution is somewhat different. Notice, first of all, 6 the vertical axis only goes up to 50 percent instead of 70 7 percent. 8 In this case, again, we see complete loss of 9 cristobalite within a few hundred to 1,000 years, in this 10 case -- in this situation, though, the porosity increases 11 only slightly. It goes from about 11 percent to about 17 12 percent, mainly because quartz precipitates, but we never 13 reach a point where quartz is no longer saturated in the - 14 system. 15 So we are seeing an evolution in which we 1 1 16 essentially will have a small change in porosity, a ' 17 dramatic change in the mineralogy, and that will be the 18 primary evolution of this part of the flow pathway. 19 Water chemistry, in this environment, shows an 20 evolution that is controlled by, or influenced by the 21 mineralogy. In this case, what we are looking at, 22 horizontal axis time, again, same scale, except on even to 23 more zero time, PH on the vertical axis. 24 And although you see a lot of fluctuation in /~' (_)\ 25 here, if you actually look at the scale, there is not much NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005 3701 (202) 234-4433
156 1 change, it remains more or less around seven to eight. It 4 2 is a rock dominated system that is telling us that, at 3 least in this particular situation for the inlet that we 4 just looked at, which is essentially the squares and the 5 five meter situation which is represented here, water t 6 chemistry is not going to evolve a heck of a lot, it will 7 remain a relatively dilute, ear neutral PH condition. 8 I'll talk about the outlet later on. That is 9 the rock-water system, and what it is telling us, I think, ! 10 and we've seen this over and over again, not only in our 11 simulations but in experiments, that the water chemistry 12 changes that will take place as a result of water-rock i 13 interaction, are not very great. O 14 They will span the range of chemistries, at 15 least based on most of the simulations we've done and the . 16 experiments we've done, they will span the range of 17 natural waters that are seen out at the site, right now. 18 So there are,.as far as we can tell, no 19 significant surprises, there. l l 20 What I want to talk about now is the regime 21 within which the waste packages will be completely i 22 encased. That is the infinite barrier system. And it is l j 23 here where changes that could be significant are much more l 24 profound. O) (, 25 The chemical characteristics are controlled, NEAL R. GFH)SS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
157 c l 1 primarily, by the high temperature modification of the l i 2 emplaced materials, whatever they may be. Presumably O 3 cement liners, invert material, pedestals, and a variety 4 of other cementitious materials will be present. 5 Their high temperature behavior, over long 6 periods of time, meaning hundreds to thousands of years, 7 at low water activities, is not known, it simply isn't. ! 8 There is a little bit of data available'for l l 9 some high temperature cements, specially engineered for lo high temperature environments. But it is -- that 11 information represents a short time period and generally l 12 is in a water saturated condition. 13 This remains, for us, a big unknown, and it is 14 where a substantial amount of attention is focused. What j 15 I want -- there is interaction that will also take place 16 with the container material, and what I want to talk 17 about, and show you the results of some of the simulations 18 we've done looking at water interacting with the 19 container. 20 And these simulations try to take into 21 account, in an approximate way, the possible effects of 22 having cement liners present. 23 Schematically, the environment we are talking 24 about looks something like this, where we have -- the (/ 25 green is rock, the tuff pink, possible cement liner, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
158 1 pedestal upon which waste containers would be resting, and I 2 an invert material. () 3 And what we are concerned with is a flow 4 pathway that would be from the rock, into the cement, or i 5 cementitious material, contacting the waste container. 6 In the simulations that I will show you, the 7 assumption was made that we are dealing with the mild 8 steel container that is twelve centimeters thick. l \ ! 9 In doing the simulations, one of the things we l l 10 assumed, because there is not an awful lot of information ! i 11 currently available about cement degradation, at high l l 12 temperatures over 100, is that the water chemistry that a ! 13 waste container is going'to see is the water chemistry you j 14 would expect from a low temperature, fro,n a cement that 15 has not experienced high temperature alteration. 16 The reason we assume that is that the 17 information that is available, thus far, suggests that as 18 you increase temperature -- well, let me back up a second. 19 Water interacting with cement, under normal 20 atmospheric conditions, generally produces waters that 21 have high calcium concentrations, have PH's on the order 22 of 12 to 13, they are very alkaline, and they may or may 23 not significant amounts of sulfate present, depending upon 24 whether or not it is a sulfa-based cement. ( 25 At higher temperatures, the resulting NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
159 l l 1 alteration tends to generate solids in the cement, that 2 would lead to lower PH's. Lower PH cement is less 7S 1 O 3 aggressive, as far as the waste containers are concerned, 4 therefore, in our simulations we wanted to be somewhat 5 conservative, and so we've taken chemistries that have 6 been reported from cements at low temperature, interact 7 those with the mild fuel container. 8 The important thing that I want you to come 9 away with, or I want to emphasize in this, is that it 10 turns out that the consequences for the waste -- what the 11 waste sees, are strongly dependent upon the reaction rate 12 at which the metal, itself, dissolves. 13 Now, again, we've used the reactive transport
,m I i \-) 14 simulator GIMRD. In this case what we did was, assume a 15 cement conditioned water passing through 12 centimeters of 16 mild steel, generating a wide range of hydroxolin and 1
17 oxide phases of a variety of secondary phases associated 18 as a result of the water chemistry in the cement itself. 19 But we were particularly interested in what 20 the PH was of the water that came through the container, 21 and would be the first stuff to contact waste. That is I 22 what this figure is an attempt to show, what the evolution 23 of that PH is. 24 The vertical axis is allowed time, so the ("Nq) 25 simulations go from, again, about zero time to about ! NEAL R. GROSS ! COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. f (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 i l
160 1 10,000 years. The horizontal axis, in this case, is the 73 2 dissolution rate of the metal in moles per square I ) 3 centimeters per second times ten to the minus ten. 4 Now, remember the -- so what we are doing is 5 tracking PH, say we assumed that the rate of dissolution 6 is eight times ten and minus ten, simply track the 7 evolution of the PH at that rate, as a function of time. 8 And what we see is that even though the PH 9 starts out around 13, the instant it hits the container, 10 because the chemistries are so radically different and the 11 generation of hydroxyl secondary reaction products takes 12 place so quickly, we rapidly reduce the PH of the solution 13 to about six. f~h. i\ -) 14 But as time goes on, after about a year to ten 15 years, more or less, the PH starts approaching closer to 16 the values of the original solution. 17 So what we are seeing is essentially 18 conditioning of the reaction pathway as this solution is 19 caming in, and eventually we have a condition where the 20 solution that goes through the container is similar to, 21 but reduced in PH from the input solution. 22 At low rates, and this continues out to 23 whatever high rate you want to choose, but as the rates 24 slow down for reaction between the metal and the incoming (O_/ 25 solution, it takes much, much longer for the solution ever NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 161 I l 1 to reach high PH's. l l I l 2 In other words, this particular system never 7s l N_,h 3 achieves equilibrium, even over that 10,000 year period, 4 because the reactions are so slow. 5 Now, the problem, or the big uncertainty we 6 are left with is, in this situation, how does one model 7 what goes on in the interior of a waste package, unless l l 8 one has very precise numbers for the dissolution rate of a l l i 9 container. 10 And container dissolution rate remains 11 something that needs to be more precisely characterized. 12 Currently, what we -- the ran. of values that are 13 measured for a mild steel container fall within an n/ N- 14 envelope that would be something like this. 15 MR. STEINDLER: Excuse me? 16 DR. GLASSLEY: Sure. 17 MR. STEINDLER: What is the reason why your PH 18 rises, again? 19 DR. GLASSLEY: The -- 20 MR. STEINDLER: I mean, are you assuming the 21 depletion of iron? 22 DR. GLASSLEY: Eventually -- essentially what 23 you are doing is generating, during this time period, a 24 population of hydroxyl bearing iron species, that are more r (_%) 25 or less in equilibrium with the solution that is coming NEAL. R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
i 162 T l 1 in. 2 And by the time you do that, the PH's are l ( 3 beginning to approach what they originally were for that 4 solution that came in, which it started out at around 13. I l 5 MR. STEINDLER: That is exactly my question. 1 6 Why does it do that? Are you assuming that you are 7 depleting the metallic iron content? l 8 DR. GLASSLEY: You are generating a variety of l 9 phases that are no longer reactive with the solution that 10 is coming in. But the iron metal, to start with, is l 11 highly reactive, far from chemical equilibrium with the l l 12 water that is coming in. l 13 And what you generate are a series of mineral 1 14 phases that are closer to equilibrium with that solution. 15 By the time you get to the point where it is closer to 16 equilibrium with the solution, the solution chemistry 17 won't be modified as much by the solids that are there, as 18 it originally was. 19 Does that make sense? 20 MR. STEINDLER: No. 21 MEMBER HORNBERGER: You are making iron 22 minerals, is that what you are saying? 23 DR. GLASSLEY: Yes, making iron minerals 24 during this time period. From this point, what we are 25 concerned with is transport of whatever was dissolved in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
163 1 this period, in this condition, out of the engineer l ? l 2 barrier system. l \)'-' 3 And in this case we are looking, again, at 4 rock-water interaction. And what I want to show you, 5 essentially, is the outlet of that long flow pathway that 6 we had generated earlier, the ten meter one, because it 7 represents, at least to some extent, the kind of 8 modification that one would see along the flow pathway, 9 that materials exiting the EBS would see. 10 In this case, again, it is the same mineralogy 11 as we started with before; it's been through an evolution 12 in which water has moved through it, it is conditioned by l l 13 rock, but now the temperature that the fluid is seeing is I i
\/ 14 decreasing, because it is exiting from the waste packages.
15 In this case behavior is similar to what we 16 described before, in that we dissolve out the 17 cristobalite, we precipitate quartz, porosity, in this 18 case, increases a little bit, but also begins to drop. 19 So we go through a peak of porosity evolution, 20 and then as we start precipitating more and more silica, 21 the porosity starts closing -- decreasing one more time. 22 So there is a much more complex e = ution that 23 is taking place below the repository as a result of the 24 solutions that are coming through. And that evolution i r^N ( ,) , 25 will become important for defining flow pathways out of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
164 1 the EBS.
,- 2 What we have attempted to do, thus far, is l !,j l
3 lock at the processes that take place in individual 4 regimes. The regime near the emplacement drifts, outside 5 of it, and within the EBS. 1 1 6 Our tasks, in the coming years, are focused 7 primarily on finding ways of coupling the flow pathways i l l l 8 between the various regimes so that we can see what the l l 9 overall effect is of solutions moving from the 10 condensations on through the boiling region, into the 11 region that has been affected by the high temperature 12 evaporation process, and then move that solution out of 13 the EBS. l s
/~N l
~' 14 We need to understand and better constrain the l
15 models of the water waste package interaction itself. The 16 fact that there is this large uncertainty in what solution l 17 PH can be of solutions that enter a waste package, based 18 on the dissolution rate of the iron itself, has to be i l 19 resolved, or else we need to find ways cf dealing with the l 20 large range of uncertainty as far as solution entering -- 21 solution chemistry entering the waste packages is ( 22 concerned. 23 And, finally, we need to refine our models of 24 fluid rock interactions for fluids that are exiting the
/"~
(_T ) 25 EBS, particularly understanding better the role of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l l
_ ~ _ . _ _ _ . . _ _ _ _ . _ . _ _ . _ . _ _ _ . _ _ _ _ . _ _ _ _ . _ . _ . _ _ _ . _ . . _ 165 1 development of the secondary mineral assemblages in 2 controlling the evolution of flow pathways. 3 In conclusion, where we stand right now, what 4 we can say is that the water waste interaction is 5 dominated by the most reactive material that occurs along 6 the flow pathway. That is elementary geochemistry. 1 7 But it is clearly important, as far as what we 8 are talking about in the waste package environment is 9 concerned. What will dominate the water chemistry, that 10 the waste packages will see, will be the engineered t 11 barrier materials. They have the biggest potential for l I 12 modifying the ambient water chemistry. 13 As far as the water-rock interactions are i 14 concerned, they are clearly dominated by the rock itself, l 15 when we are out of the EBS. And it will remain that way, 16 unless we deal with regions where there is high fluid k 17 flow, and in that case the evolution of the system becomes 18 much more extreme, and large changes in the flow pathway , 19 characteristics can occur. , 20 Any questions? J'1 VICE CHAIRMAN GARRICK: Let me ask an . 22 engineering question. Based on what you've learned so 23 far, and if I were the designer of the waste packages and 24 the engineered barriers, what additional advice could you 25 give me on the basis of this work? NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
l 166 1 Could you give me some advice on the choice of ,f _ 2 materials, dimensions, configurations? \ l
'~'
3 DR. GLASSLEY: There are several things that 4 are important as far as ultimately the transport of the 5 waste out of the repository. One is the redox condition, 6 the other is PH. 7 What we need to be able to do is have as low 8 an uncertainty on those properties, as possible. What I 9 would ask or recommend or focus on, in discussions with l l' 10 the designers, is identifying those materials that would l 11 give us the lowest uncertainty in those parameters, and 12 that would undergo, to the extent possible, based on 13 historical information and industrial information, the (% l ( N/ 14 smallest degree of change as a result of temperature l 15 effects. 16 If we have that information, then we are in a 17 position to constrain relatively well what the chemistry 18 will be that the waste will see. And that, ultimately, is 19 going to be the primary control on radionuclide transport 20 out of the EBS. 21 So those are the key things that I think we 22 would want to talk about. 23 Once materials were identified, then we could 24 start talking about specifics of how individual phases p) (_ 25 within those could influence the specifics of such things NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
l 167 l 1 as colloid generation, speciation, complexing and ! 2 solution, things like that. 7-s 3 But those are tne key parameters, at this 4 point. I I 5 VICE CHAIRMAN GARRi'/h. : Are those kind of 6 discussions taking place? l 7 DR. GLASSLEY: Yes, they are. Most of the i 8 simulations that we do in the modeling and experimental i 9 work we do is done in collaboration with the people who l 10 are doing the waste package design, as well as those who 11 are doing the waste package performance experimental 12 studies. 13 So there is a lot of collaboration and O wl 14 interaction. 15 H Do I understand correctly, Bill, that what 16 you presented, from what you presented here, that you do 17 not anticipate that deposition will create as an aureole 18 or realm of the EBS any low nermeability, or no i 19 permeability zones? 20 DR. GLASSLEY: No, I wouldn't go that far. I 21 think -- 22 MEMBER HINZE: Well, I'm looking at your 1 23 porosity. l 24 DR. GLASSLEY: We have taken the simulations 25 to -- we were initially interested in trying to determine NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
i 168 I 1 two things. Would flow barriers form as a result of
)
,g 2 deposition of material, and would we generate high b 3 permeability regions that would influence flow pathways? 4 The first effort we undertook was the latter. 5 Do we generate high permeability regions as a result of 6 the dissolution, and to what extent could those influence 7 flow? 8 In FY97 our focus is on carrying those ! 9 simulations to the point where we see, when we dissolve 10 that material, where does it deposit, and t what extent 11 will it potentially close up porosity. And either focus l 12 flow, or prevent flow resulting in the generation of 13 bathtubs, resulting in the generation of preferential flow () i kl 14 pathways. 15 That is what our effort is focusing on, in 16 this coming year. 17 MEMBER HINZE: Both in space and time? 18 DR. GLASSLEY: Yes, in space and time. And 19 depending upon -- also depending upon, by implication, 20 whatever the waste loading strategy becomes. 21 MEMBER HINZE: Another question, I think it is 22 a follow-up to John's question. Is someone concerned 23 abo- modifytrg the chemistry with back-fill, and where is 24 that hea6ed, at this time? 25 DR. GLASSLEY: We were involved in a system N' R. GROSS COURT Rt. ERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
169 1 study that-took place this last year, to examine the 2 chemical consequences of putting in back-fill that was 3 designed to maintain a high enough PH to keep containers 4 from corroding, and have a reducing environment to result, 5 to prevent oxidation of materials. f 6 And what we were able to show was that the i 7 materials that were being discussed, would not achieve the i 8 engineered goals. PH's would be elevated, but not 9 elevated high enough. l 10 PH's could be controlled but probably not for 11 the time period of concern. So that kind of interaction 12 between those involved with designing the back-fill, and 13 those of us doing geochemistry, that interaction is taking 14 place. 15 Does that answer your question? 16 MEMBER HINZE: I gather there is a lot more to 17 be done. 18 DR. GLASSLEY: Yes, as things stand now, 19 currently back-fill is not something that is part of tue 20 reference design. And as various formulations are 21 proposed, we enter into the process of evaluating the 22 performance of those different materials to see if they 23 will achieve the performance goals that are laid out, at 24 least from a geochemical perspective. 25 MEMBER HINZE: On what kind of basis do you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
1 170 1 have input to the TSPA, is this material collected at 2 specific times, or do you interact with them? 'i
\-
3 DR. GLASSLEY: The interaction is in a variety 4 of forms, some is formal, some is informal. We have G specific milestones that we have laid out, that are 6 intended to feed directly, formally, to TSPA activities, 7 and we have a couple of milestones coming up this year 8 that are part of that plug-in. 9 But there is an awful lot of informal 10 interas ' ion that takes place, as well as something new 11 comes up and we get involved in doing simulations for. 12 MEMBER HINZE: Great, thank you. 13 DR. GLASSLEY: Sure.
/~N \ i \/ 14 MEMBER HORNBERGER: Thank you, Bill. I'm 15 sorry, Ginny?
16 MS. COLTON-BRADLEY: I just have a question. 17 What kind of temperatures were you using for the modeling l l 18 of the iron dissolution? 19 DR. GLASSLEY: In that case it was -- we used 20 temperaturec. 25 degrees C P.nd 62 degrees C. We are going 21 to be doing higi-er temperature simulations. We were 22 concerned, at the time, s'2en we were doing them, whether 23 or not the high temperature data for the phases we were 24 using was adequate. O (_,/ 25 And we've gone back, looked at that, and it NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS i 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
1 171 1 1 appears to be so, so we are going to be continuing those, 1 l l s 2 finishing those off this coming year. 3 MS. COLTON-BRADLEY: Okay, now, if you drop 4 the temperature, say down to 55, would you see muc; 5 reaction from those phases? 6 DR. GLASSLEY: Well, the simulation that I l 7 showed you was for 25 degrees C, and the 62 showed some 8 change, but it wasn't significant. 9 MS. COLTON-BRADLEY: Okay. 10 MEMBER HORNBERGER: Thank you, Bill. 11 Our next presentation is by Dr. Inez Triay. 12 It doesn't look like it, I'm wrong. 13 MR. ROBINSON: I'm going to be giving Inez' 14 talk. For those of you who have seen Inez, I apologize 15 that you don't get the treat of hearing her talk about 16 this, because she is not able to travel. She has an ear 17 situation that prevents her from getting on airplanes. 18 What I'm going to do is to present the work 19 that Inez and her group have performed, specifically on 20 radionuclide transport through fractures. 21 This type of work is a subset of the total 22 task that Inez and her colleagues are doing at Los Alamos, 23 to look at the transport characteristics, both the 24 sorption and the solubility of key radionuclides.
- 25 This aspect of the work is to look at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
l (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
3 172
- I, !
j 1 radionuclide transport, specifically through fractures. ! 2 And those fractures present at Yucca Mountain, with and 3 t 4 3 without fracture coatings. ; t l 4 This is a synopsis of what is, in effect, a l 5 hypothesis that is being tested by these experiments. One ; ! 6 of the main concerns that we might have about radionuclide k 7 migration from the repository to the water table, is the i j 8 ability of the natural barrier to retard radionuclides, a 9 and fast paths, as we've seen today, are present at Yucca 1 l 10 Mountain. 4 j 11 Unfortunately the fast paths that we are L i 12 dealing with, here, are not necessarily those that we've i i i 13 been able to study, so far through ESF studies, because i l' 14 what we are talking about in radionuclide migration is I i 15 from the potential repository to the water table. i j 16 So the approach, here, is to try to get an 1 4 17 experimental look at what -- how radionuclides might 1 j 18 transport through fractures at Yucca Mountain experimental ! 19 being in the laboratory. i 20 If water is flowing through fractures, then j 21 perhaps coatings on the fractures prevent radionuclides 22 from diffusing into the matrix, then one would have a l 23 scenario of a fast path, not just for water, but for i 1 24 radionuclides. 25 However, if diffusion were to occur, of the
- NEAL R. GROSS I COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
; (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
173 1 radionuclides, in spite of fracture coatings, then we 2 could have an important retardation mechanism, both for 3 V 3 species and radionuclides that tend to sorp to the rock 4 surface, like a neptunium, or say a plutonium, but even 5 for non-sorbing radionuclides, or weakly sorping 6 radionuclides, such as technetium and selenium, matrix 7 diffusion would tend to have an inhibiting effect on the 8 transport of even a radionuclide that doesn't sorp. 9 So the experimental studies that I'm going to 10 present use natural rock fractures. All the samples that 11 I'll discuss today, these fractures were lined with 12 minerals, natural minerals, there are natural fractures 13 extracted from cores at Yucca Mountain, with a variety of
; h \M 14 different fracture coating mineralc.
15 What we are going to look at are two types of 16 experiments that have been carried out, fractured tuff 17 column experiments, which I'll talk about first, and 18 diffusion cells, which are going to be used to look at the 19 diffusive characteristics, if one had transport through a 20 fracture diffusion, into the plane of the surrounding rock 21 matrix, and how that is impacted by fracture coatings. 22 This just outlines the experimental procedures that 1 23 are used in a fracture tuff column, fractured cores are l 24 run under saturated conditions, one establishes a steady A l () 25 state flow rate through the fracture system, and I'll show NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 l
{ 174 j 1 you a schematic, in a moment, of the experimental 2 apparatus. 3 But after that flow system is established, you 4 inject various radionuclides and among other things, 5 tritiated water and technetium, and a more strongly 6 sorping radionuclide neptunium. 7 Fracture elutions collected at the outlet, in 8 other words, a concentration versus time is then 9 collected, and we can look at various things like the 10 nature of the breakthrough curve, and also more simple 11 means of accepting the experiment, such as the percentage 12 of the radionuclide that is recovered at the outlet. 13 Moving on to the diffusion cell experiments, U 14 which I will get to a little bit later in the talk, 15 radionuclides are placed in contact with the coded -- with 16 a coded fracture in a diffusion cell chamber. i 17 So that in order for the radionuclides to 18 diffuse to the other side of the tuff sample, it will have 19 to have diffused through the fractured coating. 20 So then the appearance of the radionuclide is 21 monitored in the other cell chamber, and that is an l 22 affirmative measurement that in fact the fracture coating 23 did not inhibit the movement, by molecular diffusion, into 24 the tuff matrix. 25 So this is a flow experiment, this is a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W, (202) 234-4433 WASHINGTON. D.c. 20005-3701 (202) 235-4433
175 1 diffusion experiment in the absence of fluid flow. We do f-s 2 both of these things to try to collect data that is V 3 important for radionuclide migration predictions. 4 First the fractured tuff column, you are 5 looking at a schematic of the apparatus. This is looking 6 down on the core sample from above, basically we create a 7 seal around the core, and inject fluid within a region 8 where the fracture is isolated in the core, the fracture 9 is thru-going, as you can see here.
]- And if you isolate the inlet and outlet of 11 this apparatus, one can perform what is in effect a column 12 transport experiment, although in this case, it is not a 13 crushed material, like you typically see, it is a natural t \-- 14 fracture core.
15 MEMBER HORNBERGER: What is the rough scale, 16 Bruce? 17 MR. ROBINSON: Roughly, they vary. But we are 18 talking about on the order of ten centimeters in this 1 19 direction, and a diameter of -- I've got it in the next ) 20 slide, and you will get some hard numbers. l 21 Blank scale five to ten centimeters, roughly, ' 22 and a diameter of, in all cases, about six centimeters. 23 This is a summary of the types of experiments, 24 the types of rock fractures that were studied, and the
/'N
(_,) 25 experimental results. I'm going to walk you through this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l l
176 1 synopsis of the results, before presenting a couple of 7- 2 breakthrough curves in more detail. ( L 3 MR. STEINDLER: Was that a natural fracture or 4 an artificial fracture? 5 MR. ROBINSON: They are, all the cases that 6 I'll be discussing here are natural fractures. They have 7 fracture coating that are typical of those present in 8 Yucca Mountain. 9 The fracture coating minerals range from 10 zeolites and iron oxides, in this case to manganese oxides 11 for the three fracture columns, these three in the middle 12 of the chart. 13 Typical flow rates in milliliters per hour are A/ 14 a half to one milliliter per hour, perhaps more meaningful 15 is the flow velocity through the fracture, and in these 16 experiments they range from about a half to greater than 17 five centimeters per hour. 18 One centimeter per hour is roughly 100 meters 19 per year, so we are talking about, as is typical in i 20 laboratory experiments, rates that are perhaps larger than 21 one would see in the field, but considering the evidence 22 for bomb pulse having traveled a considerable distance in 23 the unsaturated zone, maybe not. 24 The rates perhaps would be indicative of maybe b
\s,- 25 the most rapid moving fluid that one could envision in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
177 l 1 unsaturated zone. 7- 2 Experimental results are summarized in the i t i d 3 bottom, in terms of the percent recovery of these l l 4 materials, the different rated nucleides, tritiated water 5 and technetium, two species we don't expect adsorb to l 6 either the bulk rock or to the fractured minerals, l 7 typically give you recovery percentages on the order of 90 8 to 100 percent. I 9 Experimental difficulties in this sort of l 10 thing sometimes give you recoveries that are predicted to l 11 be greater than 100 percent, but this is just like in the 12 experiment business. 13 When you go to neptunium 237 recoveries, they l0 2 14 are very strongly correlated with the type of mineral that l 15 is coating the fractures. For the manganese oxide 16 minerals, we get recoveries on the order of 10 to 20 17 percent, over the time scale of these experiments. 18 These experiments would typically have at 19 least 10 core volumes and even more. I'll show you a 20 couple of breakthrough curves, in a minute. 1 21 By contrast, for less sorping fracture coating 22 minerals, we get much higher recoveries of neptunium. 23 MR. BASSETT: Say, Bruce? 24 MR. ROBINSON: Yes? () 25 MR. BASSETT: Is the core unsaturated but the (,) l NEAL R. GROSS ! COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
I 178 1 fracture is saturated?
,g 2 MR. ROBINSON: No, the whole system, in this (x -) case, is saturated.
l 3 So what one would be looking at, in I 4 terms of diffusion, would be probably an upper bound on 5 the diffusion coefficient in the medium, because diffusion l 6 coefficient is predicted to go down with -- as the system 7 becomes less saturated. But these are saturated l 8 experiments. l l 9 This is some example breakthrough curves. j 10 This is the -- actually the recovered fraction, so it is a l 11 cumulative curve. We actually inject the pulse but you l 12 are looking at the integral of those results. And when 1 13 they get to one, that means you've got complete recovery. ( )
\/ 14 This is for a case in which the fracture 15 coating materials do not exhibit significant sorption for 16 neptunium, and therefore neptunium breakthroughs are, 1
1 17 although somewhat reduced from the conservative species, 18 tritiated water and technetium, the time of the 19 breakthrough, and the relative amounts of recovery are l l 20 quite similar for all three, with neptunium somewhat 1 21 reduced. l l l 22 This is in contrast to a system in which l 23 significant sorption of neptunium occurs during its flow 1 24 through the column. m (_,I 25 Technetium and tritiated water have similar NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. l WASHINGTON, D.C. 20005-3701 (202) 234-4433 (202) 234-4433
179 1 sorts of breakthroughs, despite the different fracture ; 2 minerals. Neptunium has an extreme delay in the 3 breakthrough, and also a recovery that is obviously much 4 less-than 100 percent, in this case in the order of only 5 about 10 percent. , 6 This is for manganese oxide coated fracture. 7 I took a stab at modeling this sort of a system. The i l 8 work, the experimental work is on-going for the modeling 9 that is going to trail behind that. So we don't have a 10 comprehensive set of modeling results to present here. 11 But if one assumes a one-dimensional axial I 12 dispersion type model in the fracture, with molecular 13 diffusion into an infinite medium, which in this case 14 would be a very good assumption. 15 I think the infinite medium assumption, 16 anyway, would be a very good assumption, given the time 17 and line scales involved here. 18 And, finally, linear reversible sorption with 19 different retardation factors for the fractured surface in ! 20 the rock matrix. Putting in certain parameter values that 21 we estimated for the fracture, we could obtain good fits 22 to technetium and tritiated water elution curves. 23 There is a different -- for the experiment 24 that is being discussed here, there is a different 25 diffusion coefficient for these two species. But that NEAL R. GROSS court REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 _ _ _ . -- _ .i
180 1 varies from experiment to experiment, because basically 73 2 what is going to be controlling the fit to this diffusion 3 coefficient is the percent recovery. 4 And in some cases, these were quite similar 5 for the two radionuclides, and hence one would get a 6 diffusion coefficient that was about the same for the two. 7 If anything, technetium will tend to result in 8 lower predicted diffusion coefficients in this sort of an 9 experiment. 10 Neptunium breakthrough curves can be fitted in 11 a variety of ways, using either matrix sorption or both 12 fracture and matrix sorption. And we are in the process 13 of trying to sort out which would be a more appropriate (D 14 model. 15 Probably the fracture sorption case is more 16 reasonable for an experiment of this sort of a line scale, 17 because we are talking about fracture coatings that are 18 not planes of zero thickness. 19 They have thicknesses in the order of a 20 millimeter, and if the amount of diffusion into that 21 material over the time scale at which it travels through 22 the system is of order of millimeter, then it is going to 23 be experiencing sorption on essentially the fractured 24 coating and nothing else. I \ (_) 25 And so that will guide us -- those sorts of NEAL R. GROSS COURT REPORTERS AND TRANSCRlBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
181 1 considerations will guide us in the future modeling of I 2 these results. lO
~'
3 MR. STEINDLER: What sort of speciation did 4 you have for neptunium and technetium? 5 MR. ROBINSON: In the case -- 6 MR. STEINDLER: In an oxidizing -- 7 MR. ROBINSON: Yes, that is right. The 8 experimental procedure for these types of experiments is 9 one that is typically used for, you know, all sorts of 10 sorption and transport experiments. We carry it out in 11 the fluid, in the fluid with an atmosphere -- the fluid 12 reservoir contains an atmosphere that essentially keeps 13 the chemical conditions in the system a constant over the
,m N- ) 14 time scale of the experiment.
15 I can't really speak too strongly to that, 16 because I'm not the experimentalist, but perhaps somebody, 17 you know, can help you out with that. I really can't give 18 you much more insight than that. 19 Just to compare the sort of results that we 20 see in a transport situation with those of batch sorption 21 on the individual minerals, this is an example of the 22 distribution coefficient that one obtains for different 23 minerals that were present at the fracture coatings. 24 These two minerals, the ones that resulted in (3 (_) 25 a significant uptake of the neptunium, have very high KD NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE,, N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
182 1 values, compared to magnetite and stellerite, which also 2 in the transport experiments, showed very little influence 73 t i
~
3 in terms of retarding neptunium transport. 4 So this is sort of outside confirmation, using 5 single mineral sorption studies that we understand those 6 results fairly well. 7 I'm going to move on to diffusion cell 8 experiment, which -- 9 MEMBER HINZE: Before you leave that, could I 1 10 ask a question? 11 MR. ROBINSON: Sure. 12 MEMBER HINZE: It would seem to me the 13 aperture of these natural fractures, as well as the nature 'A >' 14 of the fracture itself, would be very important in this. 15 Is that true, and if so, how does that enter into this, 16 and into the meaningfulness of what you are coming up l 17 with? l l 18 MR. ROBINSON: These experiments are not l l 19 performed under a load, but the fracture surfaces 20 themselves are in contact with one another. You are 21 absolutely right. If you have gaping fractures through 22 which transport is occurring, you are going to have a lot 1 23 less influence, a lot more channeling right through the j 24 fracture, and a lot less influence of whatever mineral is
?x
(_.) 25 on that fracture surface. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHlNGTON, D.C. 20005-3701 (202) 234-4433
183 1 These are put together, these are natural r~s 2 fractures that are put together to try to somehow
)
3 duplicate what you might have seen when they were in situ, 4 which is a closed fracture with a gap width on the order 5 of a millimeter or less. 6 So you try, in this sort of experiment, you 7 try as much as you can to try and duplicate a condition 8 that would be there in situ. 9 Realizing, of course, that we don't have very 10 good information on what the apertures in situ would 11 really be. It is an experimental uncertainty, for sure. 12 MEMBER HINZE: Okay, all right. 13 MR. ROBINSON: This is a diffusion cell r~xi \# 14 apparatus. I'm going to move over to the diffusion 15 experiments that are being carried out, as well. One 16 places a wafer of tuff material that was extracted from a 17 coated fracture into a chamber like this, closes it up to 18 see -- you close it up to seal the system. 19 The system is saturated with water, and at 20 times zero we put radionuclides into the big cell and 21 monitor the diffusion through the system, by looking at 22 the concentration versus time at the other side of the 23 chamber. 24 So barring experimental difficulties like, you O) (, 25 know, a crack right through the thing or something, it has NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
184 1 to have diffused through the fracture coating, in order to
,3 2 have made it to the other side of the chamber.
i i G 3 This is a picture of a fracture face 4 containing the coating. The coating is the dark material. 5 I'm not the experimentalist, but I'm told this is an 6 example of one of the more heterogenous samples, and that 7 a lot of the samples, the coating appears to cover the 8 entire fracture surface, as opposed to something that has 9 some heterogeneous character to adhere. l l 10 So we are t.alking about fairly ubiquitous ) l 11 fracture coatings on these materials, and they are on the i 12 order of millimeter in thickness, the coatings themselves. 13 This is in contrast to the other side of that h (~I \- 14 same sample, which is essentially the rock matrix material 15 itself, because you've done a cut here, and you are 1 16 exposing the, on the other side of the cell, the rock 17 matrix material itself. 18 Here is an example of the measured results I 19 that you get from this sort of an experiment. This is for i 20 technetium, this is the counts per minute, equivalent to 21 concentration in our -- for the purposes of this 22 discussion, we are talking about what is, in effect, 23 concentration versus time in the monitoring chamber. 24 And so after a period in which the technetium l r'N l (_) 25 is diffusing through the system, it appears, as the other NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
185 1 side of the diffusion cell, and if one is not too far away f3 2 from the zero concentration, one can predict a linear \~/) 3 portion of the curve, in order to analyze the experiment 4 and back out parameters, namely the diffusion coefficient 5 from an experiment like this. 6 This is a summary of three of the experiments. 7 Down below we have, again tritiated water, technetium and 8 neptunium. The summary here is the, in this case, the 9 fraction of the radionuclide that is diffused to the other 10 side of the chamber, but perhaps more informative is the 11 diffusion coefficients that one obtains from this sort of 12 experiment. 13 They are of the same order of magnitude as (h t N ') 14 values that one gets in the absence of these fracture 15 coatings. So the bottom line is that the fracture 16 coatings, themselves, don't seem to significantly inhibit 17 molecular diffusion into the rock matrix. 18 And the reason for this is that the fracture 19 coatings tend to be geologic materials that are not 20 impermeable barriers, but are rather porous, themselves. 21 So to summarize this presentation, diffusion 22 from the fracture into the matrix in the flow experiments, 23 we show that that can take place even in relatively rapid 24 flow rates, and that can have an impact on radionuclide ('~h (_/ 25 migration. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
186 1 Neptunium can be significantly retarded, even fs 2 during fracture flow. And this would be on, specifically, 3 the manganese oxide minerals that were tested in the 4 fracture experiments. 5 We still need to sort out whether both 6 diffusion into the matrix and sorption onto the minerals 7 in the matrix is important, versus that on the fracture 8 walls themselves, and that is going to be a function of 9 flow rates, and that sort of thing. ' 10 But we've got a series of experiments now to 11 look at, that are on-going, but we are developing a data i l 12 base that is going to allow us to sort out some of those 13 issues. I ,o ix' ) 14 This conclusion is sort of a debunking of the 15 idea that is out there, that perhaps fracture coatings 16 might inhibit the movement of radionuclides into the 17 matrix. And the experiments that I've presented here show 18 that that is not the case. 19 In fact, the evidence is that the fracture 20 coatings do not prevent diffusion of the radionuclides 21 from the fracture, into the matrix, and that is the 22 important consideration that we take into account in our 23 transport modeling for the radionuclides. 24 Any further questions? 25 MEMBER HORNBERGER: Thank you, Bruce. Randy? NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
_ _ _ . _ _ . _ . . . . . . _ . _ _ . _ . _ _ . ~ . . _ _ _ _ _ _ _ _ _ _ _ _ . . _ . . _ . . _ . . . _ . . _ . _ - . _ . 187 1 MR. BASSETT: How would your results differ, 2 though, if the matrix were unsaturated? ; O 3 MR. ROBINSON: The -- yes, I've thought about l 4 that question a lot. If one just simply assumes that 5 under an unsaturated condition the diffusion coefficient 6 goes way down, then, I mean, that would be in line with 7 theory. 8 However, the sort of scenario that we are 9 talking about, is a fracture flow situation, in which 10 inhibition has probably taken place, at least to a limited 11 extent, so the skin of rock next to a fracture, is likely 12 to be much more saturated than the bulk measurement of 13 saturation that one would make in the bulk rock. 14 And so I believe, under the conditions in 15 which this is important, ie fracture flow, the saturate -- , 16 assuming complete saturation is probably not as bad as it 17 would have, at first, appeared. 18 Am I making sense? Okay. 19 MEMBER HORNBERGER: Other questions? 20 (No response.) 21 MEMBER HORNBERGER: Bruce, I'm curious how do 22 you see these results, let's say being translated into 23 information valuable for assessing whether or not 24 radionuclides transport to the water table will be 25 important at Yucca Mountain, or the time scale? NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
188 1 MR. ROBINSON: I'd say we are always going to ,~. 2 have uncertainty when one asks the question, is fracture \ls 3 transport of radionuclides important? 4 And so given that uncertainty, this sort of 5 information that -- ie that fracture transport is not 6 necessarily, you know, the end of it, the end of the 7 system; given that sort of information, that gives you a 8 back-up, a sort of experimental evidence that maybe can be 9 used, if we determine with more certainty, for example, 10 that fracture flow is prevalent through all of the layers 11 between the repository and the water table. 12 It would have much less significance, under a 13 scenario in which we were fairly confident that fracture e 4 \~/ 14 flow is not important, okay? So it does depend critically 15 on the nature of the flow system beneath the potential l 16 repository. 17 I mean, we've gemmed up these experiments so 18 that fracture flow is, by definition, achieved. We don't 19 know that in reality. We are probably not going to know l 20 it with certainty one way or the other. So we bring in l l l 21 other lines of experimental evidence to try to, say, cover 22 a case in which significant fracture transport occurs, we 23 then ask the next question, what about sorption on the 24 fracture materials? 25 So for those cases in TSPA analysis, of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
189 1 performance analysis, in which the system in the absence
,3 2 of these effects was dominated by fracture flow, bringing ; 6 k.)
3 this into the mix would tend to mitigate that sort of 4 behavior, somewhat, and that is how it would filter its 5 way through a performance analysis. 6 MEMBER HORNBERGER: So you would see, say 7 using retardation coefficients and diffusion coefficients 8 from these experimental results in large scale models? 9 MR. ROBINSON: Yes, I can foresee doing that, 10 we have done that in certain instances, and I think it 11 would be an important factor to consider in certain cases, 12 yes. 13 MEMBER HORNBERGER: Other questions? (m' -) 14 (No response.) 15 MEMBER HORNBERGER: Thank you, Bruce. 16 MR. ROBINSON: Thank you. 17 MEMBER HORNBERGER: Our next topic is the 18 effect of colloids and Dr. John Kessler is here. . 19 DR. KESSLER: Thank you. I'd like to talk 20 about what I -- my personal perspective is on colloid-21 aided transport and fractures, and talk about a few 22 potential reasonable options for Yucca Mountain, and how 23 to deal with them. 24 My motivation is what I think the goal is, and eT (x_,/d 25 that is provide NRC with reasonable assurance regarding NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
190 1 colloid-aided transport. That doesn't mean answer every 2 question there is about colloids, but answering the ones O 3 that need to be answered to provide NRC reasonable j 4 assurance. 5 And how might we do that? Is show the 6 colloid-aided transport has been dealt with,
- 7 conservatively, as we need to, and certainly defensively.
8 And so that sort of guided what I had in mind 9 for my talk for today. 10 I'll do an extremely quick colloid property 11 review. I'm not going to go through a lot of colloid , 12 behavior here, I just don't have time, but I'll try to hit 13 what I think are the highlights of what it is all about. 14 Colloids size ranges roughly in the order of 15 ten nanometers to ten microns, so they are pretty small i 16 little buggers. That means their surface physical 17 chemistry is very important. All the action is what 18 occurs on the surface. 19 Most colloids in solids and water have some i 20 sort of net charge that is just due to substitutions of 21 certain kinds of elements for others, that cause them to 22 have a net charge. 23 So what happens if your colloids and your 24 solids, your fracture walls, or whatever, are oppositely () 25 charged, and they are likely to attract, and the colloids NEAL R. GROSS court REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
l 191 1 are going to stick, they are going to coagulate, they are
. 2 not going to move.
3 on the other hand, if they have the same 4 charge, which is often found in nature, where they tend to 5 be negatively charge, at least for the mineral colloids 6 and the mineral fractures, then they may be stable, that 7 is that there is going to be some sort of net repulsion 8 that is going to keep them suspended. 9 The stability, that is, whether they stay 10 suspended or not, is dependent on a lot of factors, the 11 colloids solid compositions, the PH, ionic strength, there i 12 is many, many more features here that we could get into, 13 but I won't. (m 14 Why do we care for Yucca Mountain? And that j 15 is that some important radionuclides are associated with , l 16 colloids. There's two general classes of colloids that 17 have taken on certain definitions, one is called the true 18 colloids, that is certain radionuclides directly 19 precipitate into solids that are of colloidal size. 20 There is a couple of important actinides that 21 have exhibited behavior like this, that could be 1 1 22 classified as true colloids, plutonium, uranium, thorium, 23 neptunium. Then there is another whole class, which are 24 the pseudo-colloids. r's () 25 That is, you've got natural colloids floating NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON O C. 20005-3701 (202) 234-4433
i 132 l 1 about the system that radionuclides may tend to sorp onto. l 2 I mean, we just saw Bruce giving us examples of neptunium ; O 3 sorption on certain minerals. And if those minerals 4 happen to be colloidal, that would be a case of a pseudo-5 colloid that we need to worry about. 6 Natural colloids that I think about, in terms, 7 are generally the secondary minerals, which we just heard i 8 about, for example clays, there may be organic mineral 9 matter present, there may be a lot of metal oxyhydroxides, ; la especially if we are worried about degradation of the 11 container materials, or whatever. i 12 A lot of nucleides tend to sorp onto colloids. 13 That is just a short list of the kinds of possibilities, l 14 there. 15 so special colloid behavior requires different ; l 16 treatment than for solutes. For instance, we may not be ! 17 able to assume the usual solubility limits, anymore. 18 These things are going -- precipitating into colloidal 19 form, we can't use the pure solution, solubility limits 20 anymore, we don't really know what their solubility limit 21 is. 22 We can't assume colloids sorp and desorp like 23 solutes. There is a whole literature on how colloids, I 24 prefer to talk about attach and detach from other
) 25 surfaces.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.- -,~ - ..-
193 1 Although you will see, in models of colloid 7g 2 transport the use of sorption type isotherms to treat i 1 (' / 3 colloid attachment and detachment. l' 4 Colloid transport is often very different from 5 solute transport, in the sense that the relatively large 6 size affects their ability to entbr small pores. In 7 general, we don't see colloids entering small pores. l 8 I would be very surprieed to find colloid i 9 transport in the porous matrix at Yucca Mountain. I would 10 thin that if it is going to happen, it is going to happen . 11 in the fractures. 12 They also may or may not stay suspended. If 13 they are big enough and dense enough, they are going to
/"'N ,
1 ( ' 6 x' 14 just fall out. They also may not sample all flow lines in l 15 the sense that if they can't get near a wall, they are not l 16 going to sample those low flows, and they may actually l 17 flow along a little faster than your average solute speed. 18 Surface forces can either quickly removed them
, 19 from suspension, as I talked about earlier, or keep them 20 suspended and away from contacting other solids, and they 21 can float along their merry way.
22 So do we have to worry about colloids at Yucca 23 Mountain? Yes, if we just look at those basic properties 24 and don't think about exactly what is happening at Yucca (_s t 25 Mountain. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
.. .-.._ - . . _ . - ~. ..- . . - . - - -.~_--- - . _ _ -
, 194 1 For instance, we may not be able to assume ! 2 that radionuclides and colloids are solubility limited, or i } 3 that colloids are always immobile, because I mentioned
- f 4 that it seems like a lot of the natural colloids, and-a
! 5 lot of the natural surfaces are both negatively charged. 6 Colloids, we can't assume that they are 1 ! 7 retarded, like solutes, for again, for some of these i I 8 repulsive generic reasons. l l 9 So if we were going to start out really i j 10 conservative about the whole thing, we will say, well, a ! 11 let's just say that high solubilities for colloid forming i 12 radionuclides, don't worry about the low solubilities for 13 neptunium. If it forms a colloidal species, let's just w
\ 14 assume it is not-solubility limited, it is being released ,
1 15 congruently with the dissolution rate of the waste form 1 16 itself, and proceeds on down. 17 Next one might be, well, we'll say that they 1 18 migrate in the fastest pathway, because we really don't ; 19 know how they migrate, yet. So that would be the most 20 conservative assumption to start with. 21 And if there is no sort of retardation, there 22 is no sort of attachr, Ms at the walls of the fractures, 23 and they are just going co fly right on through. 24 Well, if you were to conduct a TSPA that way, 25 I think you would quickly find that for important NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
195 1 radionuclides, like neptunium, you would quickly realize gg 2 you can't be that conservative, and that you are going to ( )
~'
3 have to go back to something that is a little bit closer 4 to reality. 5 And that is that most colloids are probably 6 immobile. Arguments are made, based on colloid and solute 7 chemistry, under saturated conditions, generally, but they 8 are still pretty good ones. 9 Los Alamos has been doing that for the 10 project, but it has also been done world-wide, both in the I 11 context of high level waste disposal and other contexts. 12 But we generally find that colloids are fairly immobile. 13 But can this be demonstrated with reasonable (D 14 assurance? After all, that is what we are after, here, in 15 this particular application. 16 Well, there is a lot of lab and field work 17 that suggest they are immobile; there's fairly robust 18 arguments for the range of PH and ionic strengths at Yucca 19 Mountain. 20 Now, the range of PH and ionic strength, 21 again, in the near field, as we heard from Bill and from 22 Bruce just recently, may be a bit large. But you have to 23 worry about them getting transported all the way out, and 24 maybe further downstream, for instance, in the saturated (D zone, there may be a more narrow range of PH and chemistry (_) 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON. D,C. 20005-3701 (202) 234-4433 l
l 196 1 that we could deal with, that we could make a stronger 2 argument for colloids about. O 3 And that is these few loose ends, perturbed 4 chemistry to the EBS thermal output, is that a local 5 effect? I don't know. But maybe that is a way around, is 6 if you look at the entire transport pathway for colloids, 1 7 and not just the near field. 8 Special colloids due to degradation at the 9 container, waste form, other EBS materials, you just heard 10 about that, that is certainly an issue. We need to know l 11 what kind of colloids we are going to have. 12 As we know, the unsaturated zone has a third 13 phase air, and boy does that make life interesting for 14 colloids. Evidence is that colloids prefer the air / water ! 15 interface. There were some fascinating flow visualization 16 experiments done by Wan and Wilson a few years ago, where 17 they gemmed up a system, and they put in different kinds 18 of colloids, and pushed them through a system where there 19 was air. 20 And what they found was that almost all the 21 hydrophobic colloids, and to be very generic about it, 22 those would be more the organic ones, attached to these 23 bubbles that were sitting in their system, at the air / water interface. 24
) 25 However the most, I think, interesting result NEAL R. GROSS Co'lRT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
l 197 1 of their experiments was that most of the hydrophilic l l 7 ,s 2 colloids, and hydrophilic ones are more representative as l i )
'~'
3 a class of your mineral based colloids, which 1 tend to 4 think would be more in existence at Yucca Mountain, are 5 also attached to this air / water interface. 6 And, in addition, their experiments showed 7 that if the bubbles moved, they can pick up even more 8 colloids that happen to be attached to the walls. As 9 those bubbles slipped through one of the pore throats, 10 they yanked some of the colloids that were attached 11 elsewhere on them, and held onto them very strongly, on 12 those bubble, you know, in the air / water interfaces. 13 So what is the conclusion? The air / water ( kY 14 interface has to move to make the colloids move with it. 15 That may be an over-generalization, but I feel pretty 16 strongly that for the UZ, that is where the action is, on 17 the air / water interface. 18 Well, do we have air / water interface movement 19 in the Yucca Mountain and the UZ? What might that be? 20 One is some sort of film flow, which I could envision in 21 the larger aperture fractures, there the colloids just 22 sort of surf down on this low film flow that is running 23 down a fracture. 24 Another might be where you got bubbles that O) is 25 are forced *.hrough some macropore throat or through some NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
198 1 constrictions in fractures, where you could get some sort
,es 2 of conditions to actually move a bubble downstream, i \ >;
f 3 somehow, and bringing the colloids with it. 4 Those would be two general classes of kinds of 5 air / water interface movements, that I would think would be 6 needed to get colloids to move in the UZ. 7 So what kind of conditions do we need for 8 significant film flow transport? Well, obviously some 9 very long, very continuous air / water interface. Some -- 10 maybe perhaps some large through-going fracture is what l 11 you are going to need for that. l l 12 We will have to have, in addition, some sort 13 of significant film flow velocity. Well, the velocity is x-) 14 going to be dependent on the thickness, where the thicker 15 that film of water is, the faster flow is going to be. 16 And the film thickness is dependent on the 17 flow rate, and the matrix properties, and can get pretty 18 bloody complicated. 19 Just as a very quick sketch of what I had in 20 mind, here, let's look at the top one, first. Here is an 21 example of what I could see is a discontinuous air / water 22 interface. I'm trying to visualize, this is an edge-on of 23 a fracture, joining some other fracture. 21 Here is my fracture wall, I maybe have some (~'T l (s ,/ 25 air bubble that sits right down in here, because fractures l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
199 1 do tend to stay wetted, so you've got water right here,
- 2 and you may have all these colloids attached to the
\ t 3 bubble, here, but it's got nowhere to go, because it is 4 intercepted -- it is broken off, really, by come fracture 5 that at least has got water right here at the 6 intersection. 7 Well, that is not good enough to get colloids 8 down. But if you look in the other direction, now I'm 9 looking in the plane of the fracture, one may have, you 10 know, some sort of fracture wall contact, here, but I've 11 still got some air bubble that may go around those 12 contacts, and continue on down. 13 That will be some sort of continuous / ( N- 14 interface, where I've got the little colloids attached to 15 that film, that can continue moving on around 16 constrictions, or whatever in the fracture. 17 But if this was long and continuous, could i 18 conceivably, at least, move colloids down with them, in 19 the UZ. 20 Are fractures likely to be continuous? I was 21 going to say, up until this morning, evidence suggests 22 net, I don't know. But it seems like when I hear Bo talk, 23 I hear about air permeability measurements that there is 1 24 very quick response to barometric pressure changes, which 1 (~) (_, 25 means to suggest that maybe there are. NEAL R. GROSS : COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISI AND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
l 200 1 On the other hand, I hear other evidence that 7~ 2 suggests that they are not completely continuous all the N-]3 3 way through, and I believe that is the general opinion of 4 the project. 5 However, the fractures will be -- or the 6 air / water interface will be more continuous in fractures, 7 if the fractures are drier. But the advantage there is j l 8 the drier f ractures don' t have Au.+ film flow, it is I 9 rather slow. But the bottom A:na .i s , you can't rule it 10 out. Perhaps continuous film flow does exist. l 1 11 What about air bubbles? Can we move air l i i 12 bubbles through some of these constrictions? What do you 13 need? You need a large driving force. The force is a (~h 'N- 14 function of the fracture aperture or pore size, but l l 15 perhaps there is enough force during these episodic #10w l l 4 16 events, that we've heard about today, and on other days, 17 to get air bubbles to move on down further toward the 18 saturated zone. 19 So, once again, perhaps you cannot rule out 20 moving bubbles. 21 So the colloid transport in the unsaturated 22 zone, the UZ, what is the conclusion at this point in , l 23 time? Although unlikely, we can't rule it out entirely, 24 yet. l /"N (m,/ 25 So for NRC's benefit, we have to say where NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON. D C. 20005-3701 (202) 234-4433
201 1 does that leave as? It leaves us in the saturated zone. f- 2 All right, let's go down and look in there. (' 3 Colloid behavior in the saturated zone is a 4 lot better understood. The air / water interface 5 complication is removed, there is a huge literature on 6 colloid filtration theory and the like that we can draw on 7 to make arguments about colloid movement or lack of 8 movement. 9 Lab and field work, to date, suggests that 10 Yucca Mountain colloids are going to coagulate. Again, 11 that has been led by Los Alamos work, but there is a lot 12 of work that is similar to that, that is going on, world-13 wide, to suggest that colloids of similar kinds, under (D \-) 14 similar conditions, are going to coagulate, especially 15 down in the saturated zone, where your chemistries are 16 known better than in the disturbed EBS area. 17 So the conclusion is, and this is a very 18 tentative conclusion, it might be easier, in terms of 19 demonstrating reasonable assurance, to assume the colloids 20 pass through the unsaturated zone. 21 You don't have to concern yourselves with the l l 22 extra UZ complications that way, but of course, you are 23 now forced to make arguments that they are not going to 24 move, at least not enough, in the saturated zone. f) (_,) 25 So what are some scenarios we could think of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
202 1 for the saturated zone? Let's think of one for true 7-~ 2 colloids. We'll assume that these colloid-forming i
~'
3 actinides are not solubility limited, that the colloids 1 4 are transported rapidly through the UZ, but that all the 5 colloids are stopped at the UZ, as the interface. ! 6 Again, I'm assuming that we've got a l 7 coagulating environment once we hit the saturated zone. 8 The actinites now will dissolve in the 9 saturated zone, based on their usual solubility limited 10 constraints, and basically you are starting those 1 1 1 11 actinides at the top of the saturated zone. 12 The advantages and disadvantages of this 13 scenario is easily modeled with the current suite of TSPA f^h k- 14 models. It is certainly conservative in the UZ, but it 15 may not be conservative to assume no transport in the 16 saturated zone. 17 Another scenario might be for pseudo-colloids, 18 same sort of initial conservative assumptions, all 19 colloids are transported through the UZ and large fast 20 flowing fractures. Again, I'm assuming for now all the 21 colloids are stopped at the UZ/SZ interface, and I'll 22 assume instantaneous reversible sorption and desorption of 23 the radionuclides on and off of these pseudo-colloids at 24 that interface. 25 The advantages and disadvantages, it's got the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D C 20005-3701 (202) 234-4433
.. . . ~ . . - - - - - - - _ _ . . . - - - - - . - - - - . . . . . - _ . . . . . . . -
203 1 same ones as scenario one, it is readily modeled, it is 2 conservative in the UZ, but not necessarily conservative O 3 in the saturated zone, but now we need to know two new ! 4 things. 5 We need to know the colloid concentrations, 6 and the -- actually the natural colloids, or the pseudo-7 colloid concentrations, and their sorption 8 characteristics. 9 It is important to know how much of all these 10 other radionuclides the colloids are likely to carry. A 11 lot of work, both at Los Alamos and Paul Sher Institute, , 12 that I can think off, of the top of my head, have 13 concluded that it is important to know whether you have 14 enough colloids to make a dent in the amount of j 15 radionuclides you've got, there are not. i 16 A lot of times, just the colloid concentration 17 is so low, in combination with a relatively low sorption 18 for these, that they are just not going to carry rauch, and 19 you really don't care about them. 20 But you need to know those things, so those 21 are two new things we need to know. 22 Third scenario would be, let's allow colloid 23 migration in the saturated zone. I'm going to assume no 24 colloid movement in the porous matrix. I think, given how 25 -- what low porosity and small pore structure generally NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. i l (202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
204 1 has, it is a pretty safe assumption for colloids, but 7-s 2 migration in the larger water filled fractures is V 3 possible. 4 Again, the standard filtration theory 5 mechanisms are still invoked, I'll quickly mention those, 6 in a second. 7 Advantages and disadvantages, some lab work 8 and considerable modeling of this scenario has been 9 undertaken, to date. And I think a lot of that could be 10 drawn on, in a very straight-forward way, here. 11 Some field experiments are possible to confirm 12 or corroborate some of the modeling that could be done. 13 But there is many more parameters to bound, f'~ 'w- 14 on the other hand, I don't feel that should be 1 15 impossible to do, either. l l 16 Real quickly, here is one particular way that 17 colloid transport in fractures has been modeled, to assume l l 18 the fractures are parallel plate here, where you have some 19 sort of parabolic velocity profile, and what van Der Lee 20 et al did, they split the zone up into two regions, they 21 got one region where they got a lot of these colloid 22 interactions occurring, and basically just sort of an 23 advective region in the center. 24 What are some of these properties? In another x { ) (_/ 25 model, here, this is again, a colloid in a fracture, this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 205 1 particular model looked at external forces where there is 7g 2 a whole bunch of physicocheraical ones there, that could 3 result in a capture. 4 There is drag forces, there is diffusion, and 5 some of these external forces are just gravity. 6 All of these have been specifically modeled. 7 What you wind up needing to know is you are going to turn 8 these into attachment rate and a detachment rate, based on 9 some of those -- all the properties of those. They can 10 get modeled into ultimately how far you are going to get 11 your colloids to move. 12 Another one that I worked on a couple of years 13 ago, was colloid deposition and erosion, where I'm ) I'-'] 14 assuming the saturated zone flow is high enough te shear 15 off the previously deposited colloids. 16 If you get flow very fast, the colloids only 1 17 have a certain amount of strength that they are attached 18 to che walls, you can shear them right off. l l 19 In the end, what this does in fractures, is it 20 causes channeling, you've got regions where you completely , 1 21 clogged the fracture, other regions where you stay pretty 22 wide open. 23 It happens to be a dynamic process, but that 24 is what happens in the end, if you have these right 25 conditions. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C, 20005-3701 (202) 234-4433
206 l 1 Well, in the end I decided that it wasn't I
- g. 2 viable for natural flow conditions at Yucca Mountain, you i
O 3 just don't have enough shear force. The flow is way too 4 slow. This mechanism just doesn't look like it is going 5 to work. 6 It could occur, I suppose, due to strong 7 seismic events, which I would label short-term effects, 8 some sort of pulsing of your saturated zone, due to 9 seismic activity that would very quickly go away, and you 10 might get a bit of additional colloid migration that way. 11 So in the end, what must we understand about 12 colloids at Yucca Mountain? In terms of source term, we 13 might want to know the true colloid generation rate. And Y- 14 I say this is optional, because remember I said one 1 15 approach was just to assume that they are congruently 16 released with the waste form. 17 And, again, it depends on whether you have to l 18 do that. You have to trade off how much effort is going ' 19 to take to define this whether you have to do it or not, 20 when you pull it into TSPA. 21 You are definitely going to have to know your 22 pseudo-colloid concentrations, ad KD's, their sorptions, 23 to be able to definitively say, yes or no, I do or don't 24 have enough colloids with enough sorption capability to () s 25 move enough radionuclides to worry about, here. NEAL R. GROSS COURT REPORTERS AND TRANSCR;BERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-37'11 (202) 234-4433
207 1 You are going to have to know colloid
.f-2 stability. That is, are they going to stay suspended, or 1 ~~J 3 are they going to likely coagulate out? You may need to 4 know colloid size distribution.
5 If you have to worry about how they are going 6 to transport, you are going to need to know this. 7 Otherwise, if you assume they are stable, you probably 8 won't. 9 What we need to know about colloids in the 10 unsaturated zone, may be entirely optional, and I 11 underline may. This may not be the approach taken, but it 12 is certainly an option. 13 We could establish the presence or absence of
\- 14 continuous film flows, that may or may not be easy to do.
15 We may need to know the film flow rate distribution, if we 16 feel that it is really slow, and that is an important part 17 that we will need to invoke for retardation, that we may 18 not need to go after that. 19 Or we could just assume that there is rapid 20 transfer to the saturated zone. So, again, that is why 21 that is optional. And establish whether conditions to 22 move bubbles exist. Again, that -- I don't know whether 23 that is easy to do or not, but if you do decide they 24 exist, you are going to need to know how far and how fast. 1 / ~~ (_ 25 In the saturated zone, again, we may want to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 208 l 1 know the true colloid dissolution rate. If we want to ( s 2 invoke slow dissolution kinetics of these colloids, once l l ' -] 3 they've migrated to the saturated zone, we may want to go 4 for this information. 5 Alternatively, I say that is optional, you can 6 just assume that they are instantly at their solubility l 7 limit. 8 You are going to need to know a colloid 9 attachment and detachment kinetics. There is no way 10 around it, if you are going to invoke colloid migration in i l 11 fractures. l 12 And I would venture to say, there, that 13 there's a lot of work that has been done to date, that . t
/^x< l \- / 14 could be drawn on, to get this work down, or get these 15 numbers down.
16 Solute desorption kinetics from the pseudo-17 colloids, again I say that is optional, you could also 18 assume instantaneous equilibrium. 19 The feeling I'm trying to leave you with is 20 that the amount of work you have to do depends on how 21 conservative you have to be, or how conservative you feel 22 you can get away with, I guess is the best way to put it, 23 there. 24 So in the end we may need to know only three A ( ,) 25 things. The pseudo-colloid concentrations in KD's, the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
209 1 colloid stability. And, again, I say we may only need to 7- 2 know that in the saturated zone. ('"" l 3 The work of the Center, very correctly pointed 4 out that there is a lot more that is going on in the EBS 5 that makes what we think we know about colloid stability 6 less certain. 7 But if we go down in the saturated zone and 8 look there, then life gets a lot easier. Same with 9 colloid attachment and detachment kinetics in the 10 saturated zone. We need to know about that. 11 And, again, I'd like to reiterate, the 12 knowledge about colloid behavior in the UZ, might be 13 unnecessary, if you can demonstrate, with reasonable N- 14 assurance, you've got colloid stopped in the saturated 15 zone. 16 And that is all I had, for today. Any I 37 questions? 18 MEMBER HORNBERGER: Thank you, John. 19 DR. KESSLER: Great, glad that was so clear. l 20 MEMBER HORNBERGER: Just to try to make it 21 clear to me, anyway, in your conclusion, the three things 22 to know, you say you want to know pseudo-colloid 23 concentrations in KD's. 24 I presume this would be through measurements? 25 DR. KESSLER: Yes. NEAL R. GROSS COURT REPORTERS AND TRANSCR:BERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
210 1 MEMBER HORNBERGER: And I also presume that 3 2 you are really talking about the saturated zone, because b 3 you can't collect pseudo-colloids, unless you can collect 4 water? 5 DR. KESSLER: Yes, you are going to talk about 6 saturated zone in one sense, and you are going to have to 7 talk about the EBS in another sense. If you've go 8 colloids that are being generated, at the EBS, under odd ; 9 chemistry, as we've -- well, not natural chemistry, as 10 Bruce and others talked about, just a few minutes ago, ; 11 then we have to know these pseudo-colloids that are 12 generated there, that we may, in our models assume jump 13 straight down into the saturated zone. f% \'}
- 14 So I'm talking a. Jut both saturated zones, 15 pseudo-colloids, and those that may be generated in the 16 EBS.
37 MEMBER HORNBERGER: And you say the same thing 18 for colloid stability? 19 DR. KESSLER: Well, for colloid stability, 20 again, you may need to know what colloids you've got, but 21 arguments have been made, well, gee, you know, they may be 22 stable under this particular set of chemistry, but in the 23 near field, it is a lot wider. 24 Well, if you've go them -- you have to worry
\
[/ (_ 25 about colloids migrating all the way through to the end, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
211 l 1 and there maybe all you need to worry about is the ; I
,_ 2 chemistry that is in the saturated zone, for stability 3 arguments. !
4 But I still have to know what colloids I have, l l 1 5 and there I may have to go back up to the EBS to decide 6 that. 7 MEMBER HORNBERGER: Then I have a question I 8 want to ask Randy, who just lassed the microphone, 9 unfortunately. 10 Randy, in your measurements, in your fracture, l l 11 have you seen any colloids? ' 12 MR. BASSETT: We haven't seen colloids, we do l l 13 see uranium, but we don't see uranium colloids. In fact, 5/ # 14 my question to you was going to be, if you were to design 15 an experiment for a field environment, what surrogate l 16 would you use for an actinide colloid, so that I could get 17 permission? 18 DR. KESSLER: Well, I'd look at the ones where 19 you have some minerals that very strongly sorb and act 20 that is of interest to you, and just go with that mineral, 21 if you can measure that mineral. 22 It is tough. Generally people wind up using 23 some sort of latex microspheres, because you've got to 24 mark then., you've got to be able to track them. And you 7~ (_) 25 are then substituting some other kind of colloid that has NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
212 1 the generic properties you are interested in, for the one ! ,_s 2 you are really interested in.
- \
\ ,)
~
l 3 I think that is okay. In the end you might l 4 want to do some tracers where you fix something and let go 1 5 of maybe one of your natural colloids, for instance, the 6 clays which have, you know, edges that have different 7 properties than the faces, so you really feel comfortable. 8 But I think latex microspheres are something 9 that you can mark and trace, as a perfectly good place to l 10 start. 11 MR. BASSETT: We do see humic substances 12 measurement. I don't know if you are going to consider 13 that a colloid, of course it depends on the particles. (~\
'%- 14 DR. KESSLER: Right. Well, humic I would l 15 consider it a colloid in the sense that there is a lot of l 16 radionuclides also, that are associated with humic 17 substances, and if they have different migration behavior, 18 and you can track humics, humics are'~a good thing to be 1
19 doing, in your instance. I don't know for Yucca Mountain. 20 VICE CHAIRMAN GARRICK: John, is this 21 presentation based on your expertise in the review of the 22 literature, or is that from actually doing some research 23 in this area? 24 DR. KESSLER: The former. I have been looking (^) (_j 25 for a while to find somebody that I thought could do NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
213 1 something in the UZ. Really, what has led me to these
- 2 conclusions, is that I haven't seen my way clear to see
(/ 3 somebody who can do good colloid transport in the UZ, that 4 I thought was directly transferable. 5 I'm getting close, at least that is what some 6 people that I've been talking to have told me. But at 7 this point it is based on my own experience in reading the 8 literature. 9 VICE CHAIRMAN GARRICK: Thank you. 10 MEMBER HORNBERGER: Ginny? 11 MS. COLTON-BRADLEY: I set you up. I'm going 12 to repeat to you what Don Langer said at the colloid 13 workshop at Los Alamos, if you will recall. That is r (
'/ 14 actually where I heard you talk the first time. !
l 15 And he made the comment -- 1 16 DR. KESSLER: I wasn't there. 17 MS. COLTON-BRADLEY: Right, but your -- 18 DR. KESSLER: It was my advisor who gave me 19 the talk. I think he gave me more than two minutes. 20 MS. COLTON-BRADLEY: But his comment, at that 21 point was, if we are going to get the colloids down to the 22 saturated zone, through the UZ, how are you going to stop 23 them? 24 DR. KESSLER: You are getting them through , ew I ks)s 25 because they are surfing their way down on this air / water NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 132'3 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
!! 214 1 interface. How do you get them off of that interface, 2 miraculously, at the saturated zone? That is a good
[~s)
\ /
3 question. I'm not sure that is what you are asking me, 4 though. 5 MS. COLTON-BRADLEY: No. I mean, if you can 6 get them through the UZ, then how are you going to impede 7 them, in the saturated zone, with reasonable assurance? 8 DR. KESSLER: What I assumed was that the 9 attachment on the air / water interface is very strong. And 10 if the air / water interface moves, the colloids are going 11 to move with them. 12 The attachment at the -- if I don't have an 13 air / water interface now, the rules are different. And, O
'\ ') 14 therefore, I would say just because I've assumed --
15 because I've made the assumption that attachment is strong 16 in the air / water interface, and that I can't rule out the 17 air / water interface moving, that is my starting 18 assumptions. 19 That doesn't mean that I can't say they don't 20 move in the saturated zone, because I know a lot more 21 about colloid behavior and how they deposit out in the 22 saturated zone. 23 MS. COLTON-BRADLEY: Are you going to suggest 24 that we clog up the fractures in the saturated zone? A 4 f N s/ 25 DR. KESSLER: Yes. I'm going to suggest that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234 4433 WAF HINGTON. D.C. 20005-3701 (202) 234-4433
215 1 you are going to get colloid deposition. I'm not es 2 suggesting you are going to clog them, because I doubt you (V ! 3 have enough colloids around to clog them. 4 But I would suggest that the environment is 5 depositional, in the sensa that you will be getting 6 colloids to coagulate out in the saturated zone. 7 MS. COLTON-BRADLEY: Could that also happen in 8 the unsaturated zone? 9 DR. KESSLER: Certainly. It is the question 10 of what is the easiest to demonstrate for reasonable l 11 assurance for Yucca Mountain licensing. If you make the ; 12 assumption that .re air / water interface is really l 13 important, and Scv attacn like crazy, Fat you can't ( \ : \2 14 really say whether you've got an air; water interface that 15 is moving or not, that is where I'm starting from, Ginny. l 16 I'm trying to say, what is it that is easiest 17 to demonstrate here? And my personal feeling is that they 18 are less likely to move in the UZ than they are in the 19 saturated zone. But demonstrating it is a different game. l 20 MEMBER HORNBERGER: Let me follow up with one 21 more devil's advocate question that was suggested to me, 22 by Ginny. 23 Isn't the water table an air / water interface? 24 DR. KESSLER: Sure, the top of the water table p), (_ 25 is an air / water interface, and they can sit there forever, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234 4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
l l 216 1 for all'I care. ! [ 2 MEMBER HORNBERGER: So it doesn't move? O 3 DR. KESSLER: The air / water interface is going i 4 to move up and down a little bit, sure. That might get 5 them into the saturated zone. 6 It is one of these things where you have to l 7 think about -- ! 8 MEMBER HORNBERGER: You don't think that the I 9 surface water at the water table moves laterally? l 10 DR. KESSLER: Do I think the air / water . 11 interface at the surface moves laterally? 12 MEMBER HORNBERGER: Well, if the water moves, 13 doesn't the air / water interface move? l i 14 DR. KESSLER: Not necessarily. I don't see l 15 why it would, necessarily. i 16 MR. STEINDLER: Our experience with colloids 17 indicated that they don't, in fact, follow the assumption i 18 that you make, if your assumption about the air / water 19 interface goes to pot, then where are you? 20 We transferred colloids containing americium, 21 for example, essentially suspended in solution, where lots 22 of opportunities for air / water interface are played out, 23 essentially. And it wasn't there. 24 My question to you is, supposing you don't 25 have a system that follows your major assumption, then NEAL R. GROSS CoVRT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
217 1 where are you? e 2 DR. KESSLER: I think you are really in the (_) 3 same place, in the sense that if you are saying, by your 4 question, no, John, you've made it even too simple for the 5 unsaturated zone, and it is even more complicated, then I 6 would say that is even all the more reason why you just go 7 look in the saturated zone, where you know things better. 8 What is driving what I say here, a lot of it 9 is, what is it you can demonstrate? If life is even more 10 complicated in the unsaturated zone, that may make it all 11 the more difficult to demonstrate that they are immobile 12 in the unsaturated zone. 13 That is what is leading me to say look in the
'2 - 14 saturated zone.
15 MR. STEINDLER: If in the saturated zone, 16 air / water interface, you see no colloids, and in fact i 17 movement of those colloids, mineral based actinide 18 containing colloids, is entirely a suspension in the 19 saturated zone, then you have to go back to Ginny's 20 question, how are you going to stop that? 21 DR. KESSLER: You are going to do the modeling 22 and do the experimental work to show that they don't 23 migrate that far. A lot of work has already been done in 24 that area. That is what you are left doing. /~ ( ,T / 25 l'm not saying you are going to be able to do NEAL R. GROSS COURT REPORTERS AND TRANSCRIE,ERS 1323 RHoDE ISLAND AVE., N W. l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 l
218 l 1 it, I'm talking about approaches to dealing with colloids, 3 2 that is what you are left with. I mean, if you can't get 3 around that problem, that is what you have to do, is look, 4 develop an experimental and modeling program to satisfy 5 yourself that colloids don't migrate far enough or fast 6 enough, or carry enough radionuclides for you to worry 7 about. 8 I think what started me here was this whole 9 idea of, well gee, we have this whole unsaturated zone, 10 and do we have to worry about what is going on there, in 11 addition to saturated zone? 12 And if you can make assumptions, conservative 13 assumptions like they go through the unsaturated zone, and r's4 i
\/ 14 we are going to focus on the saturated zone because we 15 understand the processes better, there, then that is what 1
16 you may need to do to -- for reasonable assurance. ! l 17 MEMBER HORNBERGER: Andy? 18 MR. CAMPBELL: Yes, I'd like to follow up a 19 little bit. If you have colloid particles moving, 20 essentially attached to bubbles, in an air / water 21 interface, and that is your assumption in your model, just 22 before your talk we had Bruce Robinson presenting a talk 23 about diffusion into the matrix. 24 And a question that then occurs is, does the l') (_,/ 25 possible movement of colloids on a bubble, if you will, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLANC A\ E. N W. (202) 234-4433 WASHINGTON, D.C 20005-3701 (202) 234-4433
219 1 the surface of a bubble then inhibit its ability, the 7 .s 2 ability of radionuclides to essentially diffuse into the 3 matrix? 4 DR. KESSLER: If we are talking about colloids 5 diffusing into the matrix, I don't think that is an issue, 6 anyway. If we are talking about the bubble being in the 7 way, and acting as some sort of barrier for solute 8 diffusion into the matrix, that could be an issue. 9 In a sense I can't see that it would be, 10 because you already have, in the unsaturated zone, at 11 least in my mind for fracture flow, you've got your 12 bubbles in the center, the solute flow is already going to 13 be right along the fracture walls, anyway, it is not going ('-) 14 to make any difference. 15 MR. CAMPBELL: So you would have to know the 16 partitioning between the actinides attached to colloids on 17 bubble surfaces, and actinides in a soluble phase to do 18 that? l 19 DR. KESSLER: Yes. 20 MR. CAMPBELL: So that is what I'm getting at, 21 is you would have to know what that partitioning is to be 22 able to carry through the modeling, otherwise you end up 23 with a disconnect, you've got one model that says you are 24 sorbing from a soluble phase. f~)h (_ 25 DR. KESSLER: If you wanted your model NEAL R. GROSS OoVRT REPORTERS AND TRANSCRlBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005 3701 (202) 234-4433
220 1 transport -- if you wanted to model actinite transport in g- 2 the UZ, assuming colloida are attached to the air / water b 3 interface, that can also sorb and desorb actinides, yes. , 4 MEMBER HORNBERGER: Any other questions? 5 (No response.) ; I 6 MEMBER HORNBERGER: Thanks very much, John. l l 7 We will now take a 15 minute break, we will 8 reconvene at five minutes past three. 9 (Whereupon, the foregoing matter went off the 10 record at 2:54 p.m. and went back on the 11 record at 3:16 p.m.) 12 MEMBER HORNBERGER: Okay, we are reconvened, l 13 officially, here. The last segment of today's /9
-' 14 presentation, series of presentations, will be on 15 integrated transport modeling.
16 Again, the question is how all of the 17 information we've heard about, previously, in some way or 18 other, can get built into integrated transport modeling, l 19 so that assessment of the safety of the proposed 20 repository can be devised. l 21 We are not laying all that on Bruce's 22 shoulders. Bruce is going to tell us something about 23 coupled flow and transport modeling. 24 MR. ROBINSON: Thanks. The hydrologic D) (_ 25 characteristics of the system that we are studying, the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
. _ _ . _ ._ _ _ .._.. _ . _. _ _ __ _ _ . _ _ .. _ ~._ _ _ . _ ____ .. .. _ _. .._ _ _ ___.. _ .
221 1 unsaturated zone and saturated zone at Yucca Mountain, are 2 clearly important to the transport of radionuclides, but O 3 they are not everything. 4 What I'm going to try to do is to pull f 5 together the hydrologic information into numerical models 6 that also incorporate -- I'm getting some feedback, j 7 That also incorporate some of the chemical l 8 information that we have about the site, and about 9 particular radionuclides in Yucca Mountain fluids. 10 This information also needs to be folded into 11 transport models, and I will show you how we do some of 12 that in this talk. 13 This is a summary of the topics that I'm going 14 to be discussing. The first one, near field radionuclide 15 release calculations, I'm not going to cover in any 16 detail, at all. We have a report that covered that 17 aspect. 18 Basically we need a submodel within our 19 simulations that is used to predict the rate of 20 radionuclide release from the near field environment. 21 The focus of what I'm going to be dealing 22 with, today, is the transport in the unsaturated zone at 23 the site scale, from the potential repository to the water 24 table. () 25 But, keep in mind that a radionuclide release NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
l l 222 1 term, in this sort of a model, is necessary in order to
,- 2 make the calculations and make predictions.
b 3 After briefly covering that, I'm going to run l 4 through a couple of base case conparison simulations for 5 key radionuclides, and what we are going to be looking at 6 is the rate at which radionuclides are predicted to reach 7 the water table. 8 We'll lcok at several key radionuclides, and 9 use that as a screaning mechanism. This has been done in ! 10 the past, we now have new information about the sort of l 11 characteristics that may or may not revise certain 1 12 conclusions that have been made. 13 And so what I want to do is run through some [,,h
\/ 14 calculations for different radionuclides.
15 Then I'd like to '.alk about the impact of 16 fractures, and more generally, the hydrologic properties 17 of both the fracture and the matrix system, and how those 1 18 might impact radionuclide migration. l l 19 I'm going to be focusing specifically on what 20 differences you see, depending on whether you assume i 1 21 fracture, some portion of the flow is fracture dominated l l 22 versus a purely matrix dominated flow. l 23 And here I'm talking about below the 24 repository. That is where radionuclides are migrating. l () (_/ 25 All this study that has been so important in getting a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
l 223 l 1 1 1 handle on infiltration rate at the repository horizon, is l l
.3 2 only half the story. l lN]
l - 3 The carrier fluid flow rate is important, but l i
- 4 so is the nature of the flow system below the repository.
1 5 And I will show you how important the fracture flow 6 scenarios are to radionuclide migration. 7 Then the last two topics are, in a sense, 8 deviations from sort of a base case scenario, in which for 9 transport of radionuclides, we typically assume isothermal 10 or near isothermal conditions, how the repository waste l 11 heat may have an impact on the results of radionuclide 12 migration predictions. 13 And finally for the key radionuclide . /)
\2~
14 neptunium-237, I'll show our efforts at developing a 15 reactive transport model, that tries to capture, in 16 greater detail, some of the processes, chemical and 17 transport processes for that radionuclide. 18 Just one slide on some results from a near 19 field radionuclide release submodel that we use in the 20 larger scale models. This is as a function of 21 infiltration rate, assumed infiltration rate, ranging from l 22 .3 t 3 millimeters per year. l l 23 The rate of radionuclide release versus time 24 since canister failure. I will not be considering any i- m (_,) 25 aspects of the near field system in terms of the integrity NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. , (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
224 1 of the waste package. i 2 My role is to tell the project how I think 73 l\ ) ) 3 radionuclides will transport in the event that a canister 4 does fail. 5 So that is the caveat to place on my -- 6 essentially my entire talk. It is a different subject to 7 say, well, how quickly will these canisters fail, and what 8 will be the result of that process? In the far field, 9 that is where my talk will focus. 10 If the canister does fail, these near field 11 models, typically, predict that within 1,000 years, one 12 will have reached a rate of release that is more or less 13 constant. rN i \ k# 14 And that constant rate is a function of 15 infiltration rate, higher infiltration rates, not 16 aurprisingly, give you higher rates of release from the 17 near field system, l 18 And this would be expected to continue until 19 the radionuclide inventory, at a given location, is 20 depleted. And then the rate would drop to zero over a 21 fairly constrained or short time period. 22 So the results of this type of modeling lead 23 us to basically allow us to set a source term for 24 radionuclides in the near field. And what we assume is a
) 25 constant rate, ignoring kind of the processes that I've NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433 I l 1 \
I 1 l 225 l l 1 blown up on a log scale, here, ignoring those processes l l l,s 2 that will be occurring over a 10 to 1,000 year period, and .( ) l\_/ 3 taking a constant rate of radionuclide release, until that 4 time when the radionuclides will be depleted, and then a 5 zero rate of release, thereafter. 6 What I want to do is move on to the base case 7 comparisons for key radionuclides. To do that, I want to 8 say a few words about the system from the potential 9 repository to the water table. 10 We've got, within the repository horizon 11 itself, the Topopah Springs Tuff fractured, welded. One 12 would expect perhaps seme or even perhaps most of the flow 13 to be controlled within fractures under kind of a worse
/ 14 case infiltration scenario, but perhaps a more even split 15 under the lower infiltration rate scenarios.
16 If one goes down to say .1 millimeters a year, 17 then one would start to predict from models that it would 18 be matrix dominated flow in 1.2e Topopah. But for the 19 range of infiltration rates that I'll present today, 20 transport is fairly rapid down to the Calico hills 21 geologic unit. 1 22 Then you start to encounter non-welded tuffs l 23 that are less fractured, but they do have zones of 24 alteration, which are very important to capture properly n (,))
/
25 in a model or radionuclide transport. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433 l
226 1 This shaded regions show the characterization
,~s 2 in this particular cross-section through the mountain of %) 3 the zeolitic zones. It will become apparent that these 4 zones are very important to radionuclide migration for 5 their retardation capabilities for key radionuclides, like l 6 uranium and neptunium.
7 There's uncertainty in that data. We try to 8 capture it in different models by re-generating a solid l 9 geometry model, and a numerical grid to try to capture the 10 differences, here. 11 I'll show you results from the nominal 12 zeolites gases -- so called, in which the zeolites are 13 predominantly in the Calico Hills and the underlying O M 14 zeolitized portion of the Prow Pass are show in this 15 diagram. 16 Before I move from this, there was some talk l 17 about whether or not dual permeability or equivalent 18 continuum models were appropriate. What I find, in 19 transport calculations, is that it is obviously much more 20 important to consider the role of fractures in rapid l 21 transport through fractures, and for that dual l 22 permeability is a good way to go. l l l 23 If you wanted to use a continuum approach, you l 24 could do that with an assumed porosity that was equivalent 1 (~% (_) 25 to that of the fractures. But in a system like this, we l l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
227 1 predict some units to be matrix dominated, some to be 2 fracture dominated. ifb
\/
3 So unless I want to go through these units and 4 guess, beforehand, I think this is going to be matrix 5 dominated, therefore I'll put in a .3 porosity to 6 represent the matrix, let's say, and then this .0001, 7 something. 8 If I don't want to do that, then a good way, a 9 robust way to do thet is through a dual permeability model 10 that captures the behavior, in either case, within the 11 same simulation. 12 So that is why the choice of conceptual model 13 for transport, typically, dual permeability model, in this i .\2 - 14 syctem, is superior. 15 I show the zeolites. This is an example of 16 batch KD measurements performed by Inez Triay and her 17 group. These are batch studies unlike the transport and 18 diffusion experiments I presented earlier. l 19 But this is the KD of neptunium on zeolitic 20 tuffs and it shows the range that was measured in the 21 experiments. The average is about 2.5, and there is a 22 standard deviation, a pretty standard deviation in these 23 samples. 24 We al. .a that the chemistry of the fluid ,e, (._ ,/ 25 is important to these studies. This is a subset of the NEAL R. GROSS COURT REPORTEqS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005 3701 (202) 234-4433
1 1 228 j 1 entire data set, in which the fluid was J-13 water, from 1 2 well water J-13 controlled to a PH of 7. 3 So that is the constraints that I'm placing on j 4 this subset of the entire data set. We know that as we 5 increase PH, KD values go to about zero. If we increase 6 carbonate concentrations by using a different fluid from i l 7 the carbonate aquifer, that KD's also go to zero. 8 So another aspect of modeling the transport in l 9 this system is try to capture that detail in our model 10 predictions. I'll show you that in a while. 11 MEMBER HORNBERGER: Are these from crushed 12 rock? 13 MR. ROBINSON: Yes, they are.
% 14 MEMBER HORNBERGER: Do you think this is 15 defensible, then, to use this in transport models?
16 MR. ROBINSON: Yes, there's been many, many l l 17 studies that have examined that exact question, and I 18 think it is a definitive -- I think there is a definitive l 19 aitswar that it is defensible, provided you are talking 20 about flow through the matrix rock. 21 In other words, we don't see significant 22 differences due to the crushing of the rock, versus the l 23 intact rock. 24 Now, if we are talking about flow ripping [~h 25 through fractures, that is a different issue. But the KD
'(,,/
NEAL R. GROSS CoVR' AEPoRTERS AND TRANSCRIBERS IJE RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 200r5-3701 (202) 234-4433
229 1 to apply to a matrix dominated flow part of the system, it 2 is defensible to use these measurements, as has been 3 proven, experimentally. 4 Here is a numerical grid that I'll just show 5 to -- the transport calculations, and all the calculations 6 that I'll show today, were carried out in this grid, this 7 is a finite element grid, that captures the stratigraphy, 8 including the zeolitic horizons within the unsaturated 9 zone. 10 The bottom boundary is the water table. We 11 have a strip of very fine discretization near the 12 potential repository, which is expanded in these two 13 photos. We are down to about a two-meter spacing between ( ')
'- 14 nodes in this case, so that we can look at things like 15 when we look at the impact of heat, we will apply heat to 16 individual points that represent the drifts.
17 So we are at the drift scale in a somewhat -- 18 it is still somewhat of a coarse scale, but it is down to 19 the drift scale that we are able to release radionuclides 20 and apply heat to the calculations, yet we still capture 21 the transport through the unsaturated zone at larger 22 scales, as well. 23 Having told you how important dual 24 permeability is, here is an equivalent continuum b s/ 25 simulation. Now, what I'm going to do is to demonstrate NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-7701 (202) 234-4433
230 1 the equivalent continuum -- the model results that one f- 2 gets from an equivalent continuum simulation, then later V 3 on, asses the role of fractures in the dual permeability 4 case. 5 What we used the equivalent continuum modeling 6 for was to screen different radionuclides. This is the 7 simplest type of calculation to do, and we feel that it is 8 appropriate to screen radionuclides at this level, with an 9 equivalent continuum model, recognizing, in the back of 10 our heads, that we are going to have to go back and talk 11 about fractures, as well, 12 This is an example of the rate of neptunium 13 arrival at the water table, versus time. Taking into
/'7 ' 14 account the sorption that we believe will occur in the 15 zeolitic horizons.
16 The arrival times for one millimeter a year 17 infiltration rate, and I'll vary that parameter as well. 18 But at one millimeter a year for different solubilities s 19 assumed in the near field, one gets the following results 20 for the breakthrough. 21 The peak arrival at one millimeter a year is
?? about a half a million years. You can divide that up by 23 about lu. if you assume no sorption, for say a technetium 24 transport, w>ich we think is a non-sorbing radionuclide O l 25 under Yucca Motatain conditions.
(_f NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (200 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
231 l 1 Other thing to look at is that the solubility 2 in these simulations didn't really have much of an effect, O 3 but what we have coming on line now, are some estimates of 4 solubility that go quite a bit lower than this value, and 5 it will start to impact the results of these simulations 6 when we go to, say, 10 the minus 7 solubility that is just 7 starting to become -- that data are starting to become 8 available. 9 So I'm showing you a little bit of a dated 10 result already. But what it shows is the -- basically the 11 arrlval of neptunium at the water table, at one millimeter 12 a year, of about half a million years. 13 Now, this is the prediction that you get for (
's ) 14 the sorption that we assumed of 2.5 for KD.
15 What I'm going to do before moving on to the 16 fracture flow in transport scenarios, is to simply show 17 you a base case -- for this base case comparison for 18 certain key radionuclides. 19 Once the radionuclide reaches the saturated 20 zone, it becomes diluted and transports through the 21 saturated zone as well. If you assume that all of these 22 radionuclides are diluted to the saine extent in the 23 saturated zone, which is probably a pretty good 24 assumption, then what we can do is compare different O). (m, 25 , radionuclides to each other on the basis of a peak dose. NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE.. N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
232 1 So what I've done is, instead of talking about 2 doses themselves, because of the uncertainties in, say, 7-V 3 regulations and also whether to use a drinking water 4 standard or some other standard, and even the time scale 5 over which we should be doing this, what I'm simply trying 6 to do here is to asses the relative importance of various 7 radionuclides. 8 Neptunium, even though it sorbs, still comes 9 out to be the most important in our calculations of these 10 radionuclides, by -- in order of magnitude. l 11 These things can change quite a bit, depending 1 12 on things like dose conversion factors, and that sort of 13 thing. But it gives you a picture of using the drinking i \/ 14 water exposure dose conversion factors that have been used 15 in performance analyses by the project. 16 It gives you a picture for the relative 17 importance of these radionuclides. On the basis of this, 18 what I'd like to do is present results for fracture flow 19 scenarios, for neptunium, and focus on neptunium for the 20 rest of the talk. 21 Now, sometimes when you boil down complex 22 calculations you lose sight of what is actually happening 23 in some of the simulations. This is an example of the 24 type of variability that we predict in transport pathways,
~
k_)3 25 depending on the hydrologic property values you assume. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
233 l 1 This is a simulation which simply shows where l 73 2 radionuclide went. It is not meant to say what is the L \ \
\_)
3 concentration. We are just trying to get a visual about 4 the pathways taken, and specifically the pathways taken in 5 the first 50,000 years of a simulation. 6 So it is the early arriving material, and it 7 is, in this case, in a dual permeability mode that we ran 8 these simulations. 9 Changing the permeability of the zeolitic 10 horizons has a great effect on not only the pathways, but ; 11 also the transport of predictions for, say, neptunium, in l 12 two ways. One, when the permeability of the zeolitle 1 13 layers are low enough, we predict significant fraction of )
\- 14 lateral transport. ;
i 15 It is not a complete barrier, even for the l 16 property values that we are using in this simulation, but 1 17 it is a barrier such that the rapid -- the most rapid 18 moving neptunium is predicted to be that which travels 19 laterally, atop the zeolite layers. 20 I don't want to say that this simulation is 21 more likely than this one, at the moment. But what we are 22 doing is assessing, for various values that are within a 23 range, that is probably an appropriate range, we get 24 differences in the type of transport behavior. l [h l (_./ 25 Not only do you get neptunium transport of a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE. N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
i 234 ) I fraction above the zeolites, but you don't get the benefit l i 1 , 2 of that zeolite sorption. 3 So if this were the case, the neptunium 'ald 4 be predicted to be reaching the water table much more 5 fastly than if it travels through the zeolitic layers. 6 And we are in the process of sorting through 7 the simulations in order to try to sort out which one we 8 think is more appropriate. But both of the permeability 9 values assumed are within the range of measured values. 10 The actual system is a heterogenous system, so 11 maybe both answers are right, to a certain extent, and it 12 would matter greatly the permeability distribution within 13 those zeolites. And that would be something that we would i
' 14 have to look at much more closely.
15 Before moving on to the fracture transport 16 part of the presentation, I just want to summarize how we 17 are studying, in a modeling sense, how we are studying i 18 fracture flow. 19 Finite element reactive transport solutions of 1 20 the sort that I showed in the equivalent continuum models I l l 21 are an appropriate platform to study reactive transport of 22 multiple species, and they are valid for systems with 23 large dispersivities, such that numerical dispersion is l 24 not as much of an issue. [(_-~) 25 However, to really look at what the fraction ' NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
235 1 of material that might be traveling through the matrix, s 2 we've developed a particle tracking module for our dual ( 3 permeability simulation. 4 We can model low dispersivities with this 5 thing, because of particle tracking approach, which 6 minimizes numerical dispersion, and we've developed a 7 transfer function approach that allows matrix diffusion 8 from the fracture, into the matrix, to be captured in a 9 dual permeability model. 10 When you do that, you get a much more 11 realistic picture, if the flow fractions the in the 12 fractures and matrix are correct, this model is going to 13 more correctly capture how that plays out in terms of l I- \' ) 14 radionuclide migration. 15 So we use both of those types of models to 16 look at our results. 17 This is an example, again, of the rate of 18 neptunium arrival at the water table, versus tine. Now 19 I'm varying several things, and I want to walk through 20 this in some detail. 21 I'm varying infiltration rate from one to four 22 millimeters a year, and I'm also employing, for two of the 23 simulations, a particle tracking model which captures the 24 fracture and the matrix portions of the flow more <^ k_N) 25 acceptably. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE l$ LAND AVE., N W. (202) 234-4433 WAbHINGTON, D.C. 20005-3701 (202) 234-4433
236 1 So first the two, one and four millimeter a 2 year simulations, which are basically, even though they 7,) ( v 3 are dual permeability, the results are more like that of a 4 continuum model, because you are assuming rapid transport ) 5 between the two continua. 6 The one millimeter a year curve is here, and l 7 the four is here. For the bulk of the material traveling i 1 8 through the system, infiltration rate will have the 9 predicted effect that we show here, in terms of, let's ! l 10 say, a peak dose. 11 If you go to four millimeters a year the peak 12 dose rises accordingly. And the reason, of course, is you 13 got the same mass of radionuclide if you transport it to o
- 14 the water table more quickly, when it gets to the water 15 table, and then mixes with the saturated zone fluid, it l 16 will be at a higher concentration than if you dribble that 17 radionuclide more slowly to the water table.
18 When I move on to a fracture flow and 19 transport scenario, we get at one millimeter a year. This l 20 curve with the points on it, and the four millimeter a 21 year equivalent to that is the solid curve. 22 The fracture flow part of the curve shows up 23 in terms of a bimodal distribution of arrivals at the 24 water table. And the question would be, what is the /O (_) 25 cr..terion on which we are judging these simulations? NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1373 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASH' NGTON, D.C. 20005-3701
. (202) 234-4433
237 1 If we are saying it is the peak dose or the gm 2 peak concentration or the peak rate of arrival, that is (_) 3 important. Then, in these simulations, we predict that 4 the fracture part of the transport, even though it's only 5 about ten percent of the total, would be more important i 6 than the rest of it. 7 That is not always the case in these 8 simulations, but in general it is true that the fracture 9 part of the breakthrough curve would tend to be the 10 controlling factor if one were using, say, a peak dose 1 1 11 downstream at the accessible environment in the saturated 12 zone. 13 And we've done simulations in which we've (~'h ' - 14 taken this sort of a rate of arrival at the water table, 1 15 and done calculations in the saturated zone, using this as 16 an input to a saturated zone calculation, and essentially, 17 to a first approximation, these arrivals just get 18 translated to whatever your observation point in the 19 accessible environment is assumed to be, without more 20 change. 21 There are, for certain radionuclides, 22 dev2ations from that conclusion, but basically, if it is 23 coming to the water table in this type of a way, it gets 24 diluted, but the shape of the curve at the accessible O \_,/ 25 environment is net changed that much. NEAL R. GROSS COURT REPORTERS AND TRANSCRMRS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
238 1 So this is a good way to look at how things g~ 2 might play out in a -- at the end of an assessment, which N 3 is at the accessible environment in the saturated zone. 4 We focused more closely on the early time, 5 only up to 200,000, only 200,000 years in this case, the 6 rate of neptunium arrival for various infiltration rates, i 7 including a variable infiltration rate case that is 8 consistent with the infiltration maps that are being 9 developed by the USGS. 10 We get these sorts of breakthrough curves for 11 neptunium. The variable infiltration rate, even though it 12 averages about five millimeters a year, similar to the 13 four case, is more concentrated above the potential I N/ 14 repository, and hence gives you a breakthrough predicted 15 to be quite a bit sooner than an average infiltration rate 16 case. 17 Another interesting fact about that, though, 18 is that if you are talking about peak dose over long 19 times, it may not be part of the breakthrough that is 20 important for that, because the peak for the remainder of 21 the curve is also of similar magnitude in the way that it 22 would play out in the saturated zone through mixing and 23 dilutioning and at adsorption, might be different. 24 And so you might be controlled by this peak, (O,) 25 rather than this peak, in assessing the site, if it was a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
239 1 peak dose based standard that you were working under. 7s 2 MR. STEINDLER: Are these all based on the b) i 3 same inventory? 4 MR. ROBINSON: Yes, they are, We do not study 5 variations of that. I would venture to say that there are 6 variations in the hydrologic conditions, or uncertainties, 7 let's put it that way, would be -- would span a greater 8 range than any uncertainties we have about the inventory 9 that is going to be present at Yucca Mountain, l l 10 VICE CHAIRMAN GARRICK: Bruce, can you say 11 something, or are you going to say something about the 12 delay time between the peak arrival rate and the peak l l 13 dose, and what it depends on? I
'ss/ 14 MR. ROBINSON: You mean in a saturated zone?
15 VICE CHAIRMAN GARRICK: Either one, either for 16 fracture or flow or whether infiltration rate is -- 17 influences that, as well. 18 MR. ROBINSON: Let me just backtrack and say a 19 little bit about that, now. In general, what we see is 20 that higher infiltration rate, the higher the infiltration 21 rate, the higher the predicted peak dose. That is one 22 conclusion that you didn't need a fancy model to make. 23 For fracture flow, we do find, although I'm 24 not showing it in this slide, we do find that whether or () () 25 not the fracture portion of the flow controls the peak NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
240 I 1 dose, is dependent on fracture properties that we don't 1 1 ,3 2 know very much about, below the pctential repository. (_) 3 In other words, what I'm saying is that if we i 4 were to do more site characterization below the 5 repository, this peak could go away, and you would be -- 1 6 or it could get bigger, in which case you would be -- but j I 7 if it were to go away, ie, a fracture flow would become 8 less important to the calculation of transport. l 9 You would be talking more about the peak of 10 this very broadly distributed arrival at the water table. 11 Then when we typically take these simulations, 12 as I said, dump them into a saturated zone flow and i 13 transport model, to predict the effects of dilution and / t (' -) 14 dispersion, in order to make a prediction of what a 15 concentration would be at the accessible environment. 16 So that is another aspect of the work. I'm j 17 focusing, today, on the unsaturated zone. 18 VICE CHAIRMAN GARRICK: Yes, take one of your 19 curves, say the four millimeters per year that has a peak 20 rate of arrival at something less -- something around 21 100,000 years, I guess. 22 Have you calculated where and when the peak 23 dose would occur for that arrival date? 24 MR. ROBINSON: Yes, if it was 100,000, it is O k, ,/ 25 probably 100,000 plus 10,000 at the accessible NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
l 241 1 environment. That gives you a feel for the relative -- in g~ 2 other words, the saturated zone transport is predicted to V 3 be quite a bit faster than the percolation through -- of 4 the bulk of the material through the UZ, and so therefore 5 it is a pretty good approximation, at least conceptually, 6 to say this just gets stuffed into the saturated zone, 7 where it dilutes and disperses. 8 But the shape of this doesn't curve, and it is 9 not -- these curves don't really change, and they j 10 essentially arrive at the accessible environment, on this l 11 scale, at about the same time as they arrive at the water 12 table. l 13 It is more dilute, but the transport times are
- 14 pretty short, j l
15 MR. CAMPBELL: Bruce, are you defining the l l l 16 accessible environment as five kilometers from the -- 17 MR. ROBINSON: Since we -- I don't believe 18 have definitive guidance on that, we look at a variety of 19 things, because in models you can observe the 20 concentration wherever you want. 21 We've looked, typically, at five kilometers, 22 and 25 kilometers, but there is -- I'm not making any 23 statement about what the accessible environment is, I'm 24 just examining the ramifications of that in models. (.s ) 25 MR. CAMPBELL: Well, I was trying to get a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 242 1 feel for that extra 10,000 years, it does travel 5 2 kilometers or 25 kilometers? O' 3 MR. POBINSON: That would be more like 25 4 kilometers. i 5 MR. CAMPBELL: So in essance, then, the 6 limitation on transport time is essentially the time it l 7 takes to get to the water table? 8 MR. ROBINSON: That is right. 9 MR. CAMPBELL: Not the water table itself? ) 10 MR. ROBINSON: That is right. And, on top of 11 that, it is not my area of expertise, but there could be 12 significant delays in the escape from the near field 13 environment. We've taken a very conservative approach O xl 14 looking at the model results in the UZ. 15 Our rates of release from the near field tend 16 to be higher than what you might expect, based on the fact 17 that we are trying to be conservative in that. 18 So there's a lot of different aspects, here. I 19 You've got to start somewhere in terms of looking at UZ 20 transport. It is an uncertain aspect of the modeling, as 21 well. 22 MR. CAMPBELL: But there is, at least, based 23 upon the modeling, an order of magnitude difference 24 bet veen the t ransport, longer transport times or more in 25 the unsaturated zone, relative to the saturated zone, l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234 4433
243 1 whether you are talking about 5 kilometers or 25
- 2 kilometers.
\) 3 MR. ROBINSON: That is correct, especially for 1 4 things that sorb in the UZ, like neptunium. If we took a 5 careful look at saturated zone neptunium transport, we 6 probably start to assume sorption, as well. 7 But especially for something like the 8 neptunium which is delayed by an order or magnitude, at 9 least the bulk of it is, delayed by an order of magnitude 10 from that of a technetium, that would certainly be the 11 case. 12 MR. CAMPBELL: One more question. I don't 13 quite understand why you have a double peak, say the four (~ '\ -l 14 nillimeter per year particle tracking. All of those show 15 an initial sharp spike followed by a dip, followed by 16 another rise. Maybe I missed, but what is the cause of 17 that? 18 MR. ROBINSON: In every hydrologic unit, we 19 assume fracture and matrix properties that are, we think, 20 representative of that unit. Now, the fastest moving 21 material represents either one of two things; transport 22 through the Zeolitic Calico Hills through fractures, where 23 it is not able to sorb, or transport laterally on top of 24 that zeolitic layer. O k_,) 25 Both are, in fact, the case, depending on what NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
244 1 property values you use. But that represents, in this
,s , 2 case, about ten percent of the flow and transport in these 3 predictions. The other 90 is that which is percolating 4 through the matrix in the zeolitic horizons, sorbing, and 5 becoming much more delayed.
6 So it is this dual flow system that is giving 7 rise to these curves. 8 CHAIRMAN POMEROY: Bruce, before you go away, 9 I was just curious, I think we will get to this a little 10 bit later, in another discussion, but since you have done 11 some calculations, following through the saturated zone 12 out to 25 kilometers, do you think there is good 13 information, and where do you derive that information from O
\' # 14 on the geologic chemical properties and so forth, in the 15 saturated zone?
16 MR. ROBINSON: The data set is not as -- well, i I 17 this is a truism. It is not as complete as we would like 18 it to be, but you know, everybody would always say that l 19 about whatever system they are looking at. l 20 We've got hydrolic -- basically the 21 potentiometric surface is defined by on the order of 100 22 wells in the study region that we are looking at. And we ( 23 have some geologic information from a subset of those 24 wells.
/O \ ,)
m 25 We are in the process of building a model that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASH;NGTON, D.C. 20005-3701 (202) 234-4433
l 245 l 1 is calibrated to those measurements, and then the next , 1 f- 2 step would be, since these are transport calculations, to l ( 3 look at whatever chemical evidenet that we might be able 1 I 4 to bring to bear. 5 There are carbon 14 age estimates for fluids 6 in the saturated zone that we intend to look at when -- as , 7 the second stage of a process in which we first simulate 8 the hydrolic potentiometric surface and then move on to 9 the chemical indicators, to try to calibrate the system. 10 These last two topics, I won't spend a whole 11 lot of time on, but I'd like to focus on what impact we 12 think repository waste heat might have on radionuclide 13 migration. (
\ 14 What I want to say, on the outset, about that, 15 is that we are looking at a very restricted question that 16 I'm asking, here. Specifically, what I'm going to be 17 doing is examining the impact of the heat in the 18 thermohydrologic system.
19 If you assume that those processes have no 20 impact through, say, dissolution or precipitation, 21 assuming they have no impact in a permanent sense, on the 22 hydrologic properties of the system, so that is the key 23 assumption. 24 We will carry through the radionuclide (^\ () 25 simulations with all the thermalhydrologic behavior, but NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
246 1 1 we will assume that the hydrologic properties, the
- 2 characteristic curves, the permeability is unchanged from C 3 its initial value.
4 It is a critical assumption to remember in 5 presenting this, but what it is getting at is the question 6 of whether the repository heat, itself, and rapid flow-7 through fractures, during a portion of the heat-up and 8 cool-down cycle really has an effect on the long-term 9 behavior in the system. 10 This is a blow-up of a simulation of fluid 11 saturation. Blue is high saturation, red is obviously 12 zero or low saturation, at two different times in this 13 cross section. /'T s 14 These simulations are at one millimeter a 15 year, and you predict dry-out to occur at the individual 16 nodes in this simulation that are selected to be the drift 17 nodes in the calculation. 18 You see dry-out, and movement of steam 19 upwards, and condensation in a two-phase zone above, that 20 is at high fluid saturation, because it is basically a 21 two-phase system that has very -- very good transport of 22 heat characteristics compared to that of thermal 23 conduction, alone. 24 These are the sorts of variations that one /~ (_T) 25 would see from an ambient water percolation case. And, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
247 1 again, the question is, what does that mean for fs 2 radionuclide transport? 3 And I think the answer is derived from looking 4 at the time scales over which these sorts of processes 5 occur. 6 After 1,000 to 10,000 years, the repository 7 horizon is predicted to re-wet, and go back, in this case, 8 to the ambient flow system. Temperatures might remain 9 somewhat elevated, somewhat being in the, you know, 50, 60 10 degree, but it is still liquid flow after about, in this 11 one millimeter a year simulation, about a couple of 12 thousand years. 13 So the effect that that has on something like '- ' 14 neptunium is almost nothing, because of the difference in 15 the transport time scales, and the long transport times in 16 the zeolitic units, we predict a rate of arrival at the 17 water table, that is almost no different, as with the heat 18 effects included in one, in which the heat effects are not 19 included. 20 Once again, we are not considering the 21 possibility of precipitation, and how that might -- 22 dissolution, and how that might affect the hydrologic 23 parameters of the system. 24 That would be something that would need to be t'h (,,) 25 looked at in an individual study. NEAL R. GROSS COURT REPORTERS AND TRANSCR!BERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C 20005-3701 (202) 234-4433
248 1 The final topic is to try to develop a 73 2 chemical model for neptunium transports, since it is 3 showing up as being one of the key radionuclides in the 4 inventory. We'd like to do a better job than simply 5 assuming a KD, a distribution coefficient for neptunium. 6 And this is an example of a calculation in l 7 which we've taken literature values for the equilibrium l 8 constants of different neptunium species in the fluid at I 9 Yucca Mountain, we considered the carbonate system in 10 these calculations, as well, because of the tendency for 11 as PH increases, to form complex -- carbonate complexes of 12 neptunium. l l 13 One needs to consider that if you are in a l 5 14 regime in which these negatively charged carbonate 15 complexes form at the expense of the neptunial cation, 16 that would be a situation in which sorption coefficients 17 would go down. 18 If, in the simplest case, we assume that the 19 KD that we measure in the laboratory under certain 20 conditions is simply due to that of the neptunial cation, 21 anything that drives the system toward the other species 22 at the expense of the neptunial, would lower the sorption 23 coefficient. 24 What we've done is develop a reactive ID (_,/ 25 transport capability that is based on a generalized set of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
249 1 reactions such as aqueous speciation reactions into our 2 flow and transport model, so that we can carry out the 3 detailed chemical transport calculations, in the process 4 of doing an unsaturated zone simulation. 5 The next slide shows the same sort of results 6 that I've been presenting for neptunium, but now I'm 7 examining, with the reactive transport simulator, the 8 effect of the fluid chemistry, based on the aqueous 9 speciation absorption model that I just showed an example 10 of. 11 So it is rate of arrival versus time for 12 different fluid chemistries. For a fluid like well water 13 J-13, controlled at a PH of 7, one predicts the highest 14 amount absorption, the biggest delays, as you increase PH 15 in that fluid, absorption coefficients go down, effective 16 absorption coefficients, which are really an outcome of 17 the chemical calculation, they go down, and arrival times 18 are sooner and the peak is higher. 19 The UE-25 P1 water, at PH of 8.5 predicts 20 almost no sorption for neptunium, and hence arrivals at a 21 much earlier times. So there is no sorption for that 22 particular fluid. 23 Now, how does this -- this sort of a picture 24 can be boiled down in a figure like this. This is the ( 25 model simulations in a batch system. The fraction of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
250 1 neptunial cation versus FH, in the two different fluids. ,g 2 As PH increases for either of them, the 3 fraction goes down, sorption goes down. Furthermore, for 4 a higher carbonate concentration fluid, the position in PH 5 at which this occurs is lower. 6 Now, the obvious question is what does the 7 unsaturated zone fluid resemble? Does it resemble the J-8 13 and the P1 water? 9 And USGS measurements that have been recently 10 published point more towards the J-13 type of carbonate 11 concentrations, and PH's of order eight, seven to eight, 12 really, which would tend to support the assumption that 13 we've been making in most of our calculations, that the l ('-) 14 higher KD values are appropriate for neptunium. I 15 It is this sort of a speciation model that l 16 we'd like to build on, various processes into, such as the 1 l 17 possible impact of, say, a rising PH due to cement 18 materials and that sort of thing. l 19 And we feel like this is a good platform to 20 ask questions like that, through models. 21 I've got two conclusion slides, so don't think 22 I'm done after the fourth one, here. I didn't show much 23 of the results, except for an example of a near field 24 model result, which allowed us to assume, in the far field 25 simulations that a constant release rate source was NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
251 1 appropriate.
- 2 Discussed a ranking, based on a peak dose d 3 criterion that showed the neptunium comes out as being 4 quite important. Technetium non-sorbing radionuclide, E probably close to it in importance, as well.
6 These results qualitatively, and I think maybe 7 even quantitatively, do agree with performance analyses 8 that have been done. So, despite the fact that ws've got 9 a firmer basis for the sorption characteristics, there are 10 no real surprises in this ranking. 11 I show the importance of hydrologic properties 12 in the units below the potential repository, and how that 13 would control the migration patterns of radionuclides. /~T (\~l' 14 It is important to remember that we have much 15 less information below the repository than we do above the 16 repository down to the repository, because of the 17 exploratory studies facility, and how that has allowed us 18 to look at the possibility of fast pathways to that level. 19 We don't have similar information below the l l 20 repository. It is going to require a very careful study 21 of the existing data, and perhaps collection of new data 22 to really pin down property values to a level where we can 1 23 really be confident in them. 24 The fracture transport results, obviously (~) (_) 25 showed an early arrival. What may be more important is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS f 1323 RHODE ISLAND AVE., N.W. l (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 252 1 that that may be the portion of the radionuclide inventory l 1 2 that is responsible for the peak dose, if that were the l( s
)
3 regulatory criterion that we were investigating. l 4 Finally, I show the repository waste heat had 5 very little impact on the long term breakthrough of, say, 1 l 6 neptunium at the water table, with the important caveat j 1 i 7 that hydrologic and transport properties are assumed to 1 8 not change, as a result of that vigorous heating and 9 thermohydrologic behavior. 10 And finally the reactive transport modeling 11 for neptunium, I believe, puts on a very much firmer basis 12 the KD based models that we've been using up till this 13 point.
/,_\
t )
'/ 14 Additional questions?
15 I Thank you, Bruce. Anyone have additional 16 questions for Bruce? Andy? 17 MR. CAMPBELL: In the heating models that we 18 saw earlier today, in both the refluxing above the heat 1 19 source, and below, you were getting the -- the models were l l 20 saying you were getting higher ionic strengths. 21 And you were getting some statual dissolution 22 of mineral phases, so you were getting higher ionic 23 strengths in those solutions than you would start with. 24 In thece models, J-13, as I understand it, has
/N
(_,) 25 a lower ionic strength and a different composition, other NEAL R. GROSS COURT REPORTERS AND TRANSC RIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
253 1 than just PH, with the P-1 water. And to what extent does l l g3 2 ionic strength play a role in those differences that you b 3 are seeing? l 4 And two, if you have water coming off the l l l 5 repository, which is full of a lot more than the natural l
- 6 waters in terms of various ionic species, does that -- how I l 7 would you factor that into your transport models?
l 8 MR. ROBINSON: Let me just, for talking l l l 9 purposes, if I can find it, I can't find it, recall that 10 the sorts of effects, chemical effects that are important , l l 11 for, say, a neptunium migration are only once the 12 neptunium and whatever plume is a result of the heating, 13 reaches the zeolitic horizon. i ,_) \
' 14 And that is over 100 meters below the i
15 potential repository. So that would be a -- in other 16 words, there is a great distance over which these ehemical 17 effects would have to travel, and not be buffered by the 18 existing ground water system, in order for them to become 19 important co, say, absorption of neptunium. 20 So that would argue that these sorts of 21 effects might be too near field, and that by the time the l 22 radionuclides reach the zeolites, hundred or 150 meter 23 below the repository, those effects would have been 24 dampened down in the -- and the natural system parameters 25 would be more appropriate than those that were -- chemical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
254 . 1 effects that were occurring in the near field. 2 MR. CAMPBELL: Have you done the buffer O 3 capacity cap studies and calculations -- , 4 MR. ROBINSON: We've not done those, yet. [ 5 That is an area that we want to look at, both in terms of 6 effect of the heat on the natural system, and also with a 7 system that may probably have considerable cementatious j
\
8 materials that would also have an effect for man-made 9 materials, and the natural system, are both going to l 10 change. 11 The zeolites tend to not get too hot, because 12 of their distance from the repository. Not too hot, 13 meaning they are certainly not in the 200 degree range, 14 like we see near the -- in the very near field. 15 CHAIRMAN POMEROY: Bruce, let me ask you 16 quickly, you indicated that the hydrologic properties in 17 the units below the potential repository control the 18 migration patterns of radionuclides, and that they are not 19 well known, with which I happen to agree. 20 But is there a plan in place to get you -- get 21 that necessary data within a time frame that is going to 22 allow incorporation of that information in the TSPA VA? 23 MR. ROB 7NSON: I think various research 24 organizations in the project are going to take their best 25 stab at looking at the units below the repository. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
255 l 1 Obviously the ideal situation would be to have studies 2 facility shaft that goes into the Calico Hills, given that ) (s) wi 3 that -- assuming that that is not the case, you've got 4 various organizations looking at core measurements, and 5 doing the statistical analyses to get those parameters in 6 the units -- in well bores from well bore samples, 7 calibrating them to, say, saturation, as in Bo 8 Bodvarsson's presentation. 9 Looking at additional chemical sampling, ; 1 10 basically, if you don't have a shaft, you have to do it 11 through well bores, and there are going to be fundamental 12 limitations that are always going to exist when that is 13 the case. (3 ! I
'/ 14 But the idea is to extract as much information 15 from the bore holes as possible.
16 CHAIRMAN POMEROY: Thanks. 17 MEMBER HORNBERGER: Thank you, Bruce. 18 Next presentation is by Dr. Abe Van Luik, who 19 is going to talk about the TSPA process. 20 DR. LUIK: As you can tell, we are shifting 21 gears now. You've heard from principal investigators, and 22 now all of a sudden you are going to hear from a DOE l 23 person. 24 And we are also shifting gears in the (~')N (_ 25 technical depth at which we are going to present things. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
_. . _ . . - . . - . .. . __ - ..- _ ~ - .._ _ 256 1 What I'd basically would like to convey to l l 2 you, is where has TSPA been, what has it been doing while 3 all of these wonderful pieces of work are going on; how is 4 it reacting to the latest developments in process level 5 modeling, particularly Bo's model. 6 And as the question was raised by Dr. Pomeroy, 7 where do we go from here to make sure that we have a l 8 comprehensive and integrated TSPA for the VA. 1 9 So quickly I'll be covering two aspects of the i 10 work that we've done since 1995, basically '<,1ked on in I 11 1996. Quickly cover current performances work, i 12 particularly the implications that we have done some 13 calculations on, just this week on evaluating the role of 14 the alternative and saturated zone flow model, and then 15 bring you into what we are going to do from this point 16 forward. 17 The reason I'm throwing some of these things 18 up, is just to show you that when you got done listening 19 to, basically, a detailed description of two process 20 models, but there is a lot more to a TSPA analysis, TSPA 21 being total system performance assessment. 22 In terms of preparing for the model i 23 abstraction work that is going to take place in 1997, we i , 24 have done a number of sensitivity analyses, looking at 25 alternative models of the sensitive components of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (204) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
1 I 257 i 1 1 total system performance, i l f-- 2 Perhaps I should get into the 20th century, 3 and use that cute pointer. Where is it? 4 I also want to show what we have been doing in 5 terms of TSPA's. One of the questions I was asked 6 recently by someone outside the program, is, if you only 7 do these things every two years, how in the world are you l 8 ever going to catch up to what is going on in science and i 1 9 engineering by 1997 and give them feedback? 10 And I want to, by showing you some examples, l l 11 show you that we have been doing TSPA's since TSPA '95. l 12 The results of the work that I'm going to show you, most 13 of it has been documented, several deliverable documents { \-- 14 are currently in DOE review, and when I checked with Eric 15 Smith's staff yesterday, several have been okayed, so they 16 will be delivered into the public domain very quickly. 17 And I guess I mentioned that so you don't 18 think that I'm snowing you. These things will all be 19 available. 20 As far as model abstraction sensitivity 21 analyses, we have been doing these. They are in a report 22 which is a report that, I believe, is one of the ones that 23 we cleared just recently. 24 But we have looked at UZ flow models, thermal p/ (- 25 hydrology models and waste package degradation models. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
l l 258 1 The sub-bullets under here are just two out of three or ,f ss 2 four that are addressed under each one. V 3 But, for example, we've been looking at the l 4 spatially variable infiltration and hydrologic property's 5 role, the sensitivity of performance to that kind of l 6 variability at the drift scale. 7 As Bo mentioned this morning, he is working on
]
l 8 100 meter by 100 meter scale, and as was mentioned just a j 9 few minutes ago, when you look at the drift scale, you 10 have to worry about heterogeneities as a much smaller 11 scale than how they influence flow. 12 I believe that we would be best served by 13 moving along. We have looked at things like waste form ! 1 (")' \' 14 alteration, radionuclide mobilization, and transport. 15 Looking, again, at alternative models of in drift 16 hydrology, most of the TSPA work is looking at taking the 17 materials being produced at the site scale, and bringing 18 them down to the repository scale. ; 1 19 We have looked at unsaturated and saturated 20 zone transport issues, and in fact, I'll be talking more, 21 in just a minute, about the evaluations that we have very 22 recently made, of the new models for percolation flux 23 distribution in the UZ. 24 We have done some work which has been l r~x 1 _ 25 documented in seismic and tectonic activity scenarios, and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
259 1 we are part of an overall engineering effort to get a fx 2 handle on criticality effects. 1 . %.)\ ! 3 And that is another issue that I should point 4 out, is that PA has been accused, in the past, of working l 5 by itself in a little room by itself. Most all of these 1 6 work is being done in cooperation with the other parts of 7 the program, particularly the science and the engineering l l 8 parts, but also the biosphere patients that come out of i 9 the environmental group. 10 TSPA analysis that we have conducted since , l 11 TSPA 1995, if the system study on enhanced engineered 12 barrier options is -- comes out, you will see some of 13 these results in there.
\~/ 14 We have looked at evaluating alternative 15 performance measures, the integrated release, if you will 16 recall, was the 40 CFR 191 performance measure. We have l 17 looked at peak doses to critical groups at various 18 locations.
19 We are expanding that work, also, to look at 10 peak doses to critical groups of different descriptions. 21 Uncertainties that have been investigated in this work, of 22 course, are the UZ flow and transport models, especially 23 how that translates into drift scale flow models, waste 24 package degradation, waste form alteration and saturated f k,_ 25 zone flow models. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
260 1 All of these things have been the subject of g- 2 uncertainty analyses in this work. V 3 So now we shift gears to current performance 4 assessment work, work that is going on, even as we speak, 5 except, of course, for those people who are in attendance 6 here, instead of doing their work. 7 (Laughter.) 8 DR. LUIK: Currently, sensitivity analyses are 9 being conducted to evaluate the impacts of percolation 10 flux versus seepage flux, advective flux and contact with 11 the waste form, waste package degradation, cladding 12 degradation, and the form of radionuclide release. 13 In other words, is it coming out as a gas, is 14 it coming out as a solute in liquid form. And then the 15 location of the critical group, we have paid some 16 attention to, because of current discussions on what our 17 performance measure is going to be from a regulatory 18 perspective. 19 I'll give you one example. It is a work in 20 progress, and at the time I made up these view graphs, 21 which was yesterday, we had a preliminary example 22 available. I believe that by now we have two available, 23 but it is too late. 24 TSPA 1995 was modified to make it reasonably 1 r~s k ,) 25 conservative, for this particular illustration, and an l NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
260 1 All of these things have been the subject of
,-s 2 uncertainty analyses in this work.
t V 3 So now we shift gears to current performance 4 assessment work, work that is going on, even as we speak, 5 except, of course, for those people who are in attendance 6 here, instead of doing their work. 7 (Laughter.) 8 DR. LUIK: Currently, sensitivity analyses are 9 being conducted to evaluate the impacts of percolation 10 flux versus seepage flux, advective flux and contact with 11 the waste form, waste package degradation, cladding 12 degradation, and the form of radionuclide release. 13 In other words, is it coming out as a gas, is
\- 14 it coming out as a solute in liquid form. And then the 15 location of the critical group, we have paid some 16 attention to, because of current discussions on what our 17 performance measure is going to be from a regulatory 18 perspective.
19 I'll give you one example. It is a work in 20 progress, and at the time I made up these view graphs, 21 which was yesterday, we had a preliminary example 22 available. I believe that by now we have two available, 23 but it is too late. 24 TSPA 1995 was modified to make it reasonably 1 f% l (-) 25 conservative, for this particular illustration, and an NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE IGLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005 3701 (202) 234-4433
~ . _ . . _ _ . _ _ _ . . _ _ _ . . _ _ _..__.___ _ __ _ . . .___ _ . .._ - _ _. _ . _ . _ . . _ . _ . _ . . _ . _
261 1 optimistic case. We used six representative columns from 2 the work that was shown you by Bo Bodvarsson this morning, 3 and the average percolation flux, at depth, was increased 4 from what we had in TSPA '95, to the seven millimeters a 5 year that was indicated in the information that we got 6 from Bo. 7 We used, because we were doing transport, a 8 dual permeability model to define fracture matrix flux and 9 velocity distributions. 10 If you will recall, this morning you saw this 11 diagram and what this illustration is, and we made color l 12 copies for the Board. Did you get color copies? Some of 13 you did, some of you didn't. 14 If you want color copies, you have Bo 15 Bodvarsson's -- yes, okay. It is the same diagram, l 16 basically, but showing the six places where Bo calculated l 17 for us what the flux would be in the center of the six 18 representative locations in the repository for which we 19 did calculations.
. 20 And you can see that most of these locations l 21 are right on the edge of, or in, the higher flux area, and 22 we may look at the average for these six locations, it was l 23 right at seven.
24 Let me describe the two cases that we did 25 model. The conservative case made an assumption that 36 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
.. . _ . _ _ . _ _ _ . . _ . __.__.______.__._-___.___m . _. _ 262 1 percent of the packages see dripping water. This is 2 approximately the same percentage as was assumed in TSPA 3 '95, and it has a basis in the drift scale modeling that 4 we did, where we varied the parame'.ers to look at the role 5 of heterogeneity at that scale. 6 We conservatively assumed one over L fracture 7 matrix partition particle transition rate, where L is the 8 formation thickness, and this is part of the input to the 9 RIP model which uses a Markovian transport module. 10 This will be something that will be addressed, 11 as I speak later on, about the abstraction process. We 12 will abstract, instead of using this type of modeling, the 13 results of the transport modeling being done by Los 14 Alamos, of course. 15 For the conservative case we use no back-fill, 16 for the optimistic case, we assign four percent of the 17 packages to see dripping water, which is of the same 18 magnitude as what Bo shows for the likelihood of fracture 19 flow at the repository horizon for the very large scale. 20 But, again, because Bo gets this result, 21 doesn't mean it is not closer to that in real life, at, 22 you know, at a very smaller scale, much smaller scale. 23 We also threw in galvanic protection of the 24 waste packages, when half of the outer barrier is gone, 25 the inner barrier can start being attacked. Here, when NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
1 263 ) 1 any part of the outer barrier was gone, the inner barrier ) l ,r~3 2 started to be attacked. , 3 We threw in our best guess at the cladding, 1 4 how long it lives, and what the eal liability of it is, 1 5 over time. And then we threw in a more conservative 6 transport parameter, one over point one L. I 7 So that is the difference between the two 8 cases that we modeled. ! 9 Let me also cover the assumptions that are 10 common to both cases, because they are based still on TSPA 11 '95 modeling. We assumed that 83 metric tons of uranium 12 acre thermal, or actually mass loading. We assumed it 13 drips on the waste container release model for those --
~
I )
\/ 14 that percentage of containers seeing drips.
l 15 But we also assumed that diffusion, through 16 corrosion pits, was the avenue for radionuclides to escape 17 before contacting advective flow a meter or so away from 18 the container. . 19 This is -- this was a minor case in TSPA '95. I l 20 In TSPA '95, the reference case was drips on waste i l 21 container, and advective release for many of the packages 22 from the waste container. 23 This is one of the modifications that we've 24 been making, from looking at the drift scale modeling from o _) 25 the TSPA '95 platform. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
264 1 We assumed drinking water doses of two liters g- 2 a day at both five kilometers and 30 kilometers down V 3 gradient from the repository. We used, instead of the 4 flux matrix fracture flux numbers that we used in TSPA 5 '95, which were quite low, many cases about 1.25. 6 We increased this to what I showed you a while
]
7 ago, an average of seven, and we also assumed that iodine, 8 chlorine and carbon migrate through the engineered barrier 9 system as aqueous species. No flying iodine in these 10 particular calculations. l 11 And here are the preliminary results, and 12 these are not numbers that you should, you know, quote to 13 your grandmother over the phone, these are very t
\- 14 preliminary calculations.
15 If we look at TSPA '95 comparative case, this 16 is not the same case that was a reference case in TSPA 17 '95, but this was a case reported in TSPA '95 that is 18 almost exactly the same as the higher flux case. 19 You can see that increasing the flux brings 20 the peak up earlier, makes it just a little bit higher, 21 and that this is the technetium iodine peak, and the later 22 the neptunium peak comes in, and here are the contributing 23 radionuclides here. 24 Once the neptunium peak comes in, it rises a
~x l(,/> 25 little bit higher than the preliminary peak. We did not NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234 4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
265 1 carry the calculations thus far, beyond 100,000 years, we em 2 did these calculations over the last two days, there just ( ) \ '/ 3 wasn't time to add them in. 4 But you can see that the conservative case, 5 and we believe it is a realistically conservative case, 6 not an optimistic case, at all, delivers doses for the 7 drinking water scenario that we modeled, whether there is 8 a person there or not, we didn't care for this thing, of 9 about maybe 30 or 40 milligram per year for the about i 10 10,000 year period. 11 The optimistic case, on the other hand, and it 12 is my personal belief that when we get done with the l l
,_s 13 factoring in the results of the more elaborate process .
/, 3 -'/ 14 level modeling, that we will be somewhere between these l 15 two cases, or maybe even at this case, here. 16 The doses are quite life legible over 100,000 17 year period. 18 If we look at 30 kilometers, which is where 19 the majority of people live near Yucca Mountain, we see 20 that the doses are brought down almost in order of 21 magnitude, but the shapes of the curves are still pretty 22 much the same, which reflects the results that Bruce 23 Robinson showed a while ago, too, that the basic effect of 24 moving from 5 to 30 kilometers is more mixing, more r"Na
\m,/ 25 dilution.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
266 1 But basically the solution that reaches the 2 well, at 30 kilometers, is about 25 times more dilute than O- 3 the solution that reached it five kilometers. 4 These results are being re-worked, even as we 5 speak, and more and more material is being brought into 6 these calculations because, obviously, these new results 7 from Bo's modeling are being tested by Bo, himself, to 8 see, you know, what other interpretations of the mountain 9 can be made. 10 And it is our job in TSPA to immediately feed 11 back what the implications are for system performance. 12 The significance that we see, from a TSPA 13 perspective of the new work by Bo and company, is that 14 they have increased percolation flux, and they have 15 increased the bulk average matrix permeability. 16 It decreases the main UZ advective travel 17 time, and that has been demonstrated both by Bo and by 18 Bruce. The higher flux may increase the percent of 19 packages likely to encounter seepage. I mean, that is 20 almost intuitive. 21 But the high permeability on the other end may 22 decrease the percent of packages likely to encounter 23 seepage. The higher flux is likely to stay in the matrix, 24 because the matrix is so much more capable of conducting 25 water. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
267 l ! 1 The higher flux may, however, decrease time of l 2 reduced humidities. The thermohydrology effects, even as O 3 we speak, are being addressed by Livermore. 4 Higher permeability may increase the time to 5 initial breakthrough of radionuclides, depending on l 6 percent of fi':x and f ractures, and like I said, this ; 7 evaluation is a work in progress, all of these statements i l 8 are being questioned vehemently by the modelers. j 9 Bo showed you, this morning, an illustration 10 of one of the analysis that was done with a drift at 11 ambient temperatures. 100 percent relative humidity in 12 that drift, throwing a 28 millimeter per year pulse into 13 that drift. 00 .4 And this is the kind of result that, you know, 15 could even though the average flux here might be seven 16 millimeters, there is no reason to think that at seven 17 millimeters distributed broadly, across this area, it 18 could be that there is a focus of 28 and some areas of 19 zero, you know? That is the way nature works. 20 But you need to push a heck of a lot of water 21 through this type of system with this wide open 22 permeability, in order to get dripping into the drifts. 23 And this is when we finish this work, it 24 should give us a firmer basis for assigning that 25 percentage of waste packages that may see dripping water. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
1 l 268 I 1 So we are not done with these analysis, by any 1 l l g3 2 means.
\ )
3 Where do we go from here? All of this work, 4 on our part, is preparatory to TSPA VA, the viability 5 assessment is due in 1998, as you heard. And so we have 6 been, this last year, doing a lot of sensitivity analysis, 7 a lot of evaluations of alternative conceptual models, in 8 cooperation with the site investigators, and also with the 9 engineers. 10 And all of this was preparatory to the work 11 that is going to be done for TSPA VA this coming year, 12 which starts next week, and in 1998. 13 The objective of the way that we are (,_).
\' 14 approaching this is to assure that the TSPA VA captures 15 the process leve) of modeling being performed by site 16 engineering, and also the environmental function of the 17 project.
18 We are bringing in, also, and this is not we 19 PA, but the program is bringing in project external 20 experts, and we are going to involve them in two ways, 21 through focused expert elicitations, and through a 22 comprehensive peer review. 23 And in the question and answer discussion 24 tomorrow, we will be prepared to answer some questions as
/ \
(m,) 25 of the status of these two efforts. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
. - . . - . - - . - - - - - ..-~- - - -- .- -_ .
269 l l 1 Just this last week, from the M and O we l 2 received, and are currently reviewing, a TSPA VA plan i 3 which defines the overall approach, the roles of the t l 4 different M and O participants and the GS, and 5 responsibilities, discusses the method of ensuring the 6 most representative process models that are abstracted in 7 the TSPA VA, and then presents a status for each process l 8 model that is to be abstracted. 9 The current status of abccraction, the NRC's i ! 10 treatment in IPA-2 or recent communications, relevant i ! 11 uncertainty sources of information, expected output from I i
- 12 the abstraction process itself, the key personnel within )
13 the project that we think we would like to have involved, l 14 and a schedule for these abstraction exercises. 1 15 In other words, we are planning the details of [ 16 the work for the next two years, down to who is going to i j 17 do what and when. t j 18 The approach that we are taking to doing TSPA i i 19 VA, is to form abstraction and testing teams, composed of I j 20 the process model development, and the performance l 21 assessment staff. What we want to do is assure that the I 5 22 use we in performance assessment make of these models is 23 proper. That the models are tested, and that we have l 24 appropriately bounded the uncertainties in those models. k) 25 To do some of this work, we need to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS l 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
270 1 supplement, as I mentioned, the in-house expertise, with 7s 2 an expert elicitation specific to certain focused areas, 3 to quantify the uncertainties. 4 And we want to focus the TSPA analysis on key 5 attributes, consistent with what we have discovered from 6 our previous TSPA experience, the waste containment and 7 isolation strategy, and also the NRC's KTI's. 8 And, finally, what you've all been waiting 9 for, here is when the larger aspects of this work are 10 going to be done. Abstraction workshops start this fall, , 11 finish in the spring. 12 Sensitivity analyses start this winter and go ) I' 13 on all year long. The documentation c' the abstraction t]
\ 14 process, which is what you are seeing here, is in the fall 15 of 1997 to the Spring of '98.
16 The reference case analyses, we are looking at 17 November '97, just a couple of months, and then the 18 sensitivity analyses for the other cases that will be 19 presented, to be representative, maybe four cases 20 representative of the whole range. 21 And that will be another three months after 22 the reference cases are done. Documentation is going to 23 take about four months, and the peer review process will 24 carry through the delivery of the documentation, and then O (_) 25 we expect them about a few months later, maybe about nine NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISt.AND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
y 271 1 months later, to give us a document telling us what we can
- 2 improve for the TSPA LA, because we shouldn't forget that 7
3 one of the main purposes of the TSPA VA, is to put 4 together a comprehensive case that can be commented on, to 5 tell us what we need to improve for the TSPA, for the 6 license application. 7 That is the ultimate goal, is the license 8 application. 1 9 And since I wasn't interrupted, I see that I 10 have no questions. I'll sit down. You have just redeemed 11 some time. 12 CHAIRMAN POMEROY: Before you do that, what is 13 the date of the viability assessment transmission to the fb \/ 14 President? 15 DR. LUIK: That is a question for which I 16 would have to consult someone else. I didn't -- 9-98, 17 okay. The thing that confused me, I knew the 9-98, but 18 the transmission to the President. Is that an actual l 19 thing that we are going to do. 20 VICE CHAIRMAN GARRICK: I'm saving most of my 21 questions for tomorrow, but I wanted to see if you could 22 elaborate jut a little bit on what is special about 23 viability assessment. 24 What is there to be more of there than just p) (_, 25 another iteration of the performance assessment? NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE IS LAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
1 272 1 DR. LUIK: My view of what is special about c 2 the viability assessment is that it is the ultimate dry 7 I ( 3 run for the license application. The intent is to allow 1 4 the NRC, or beg the NRC to review it, and give us some 5 serious comments on it, just like they are doing on TSPA l 6 '95. l l l 7 This will be the first assessment where we 1 8 actually physically fold in every aspect of the site and 9 the engineering and the biosphere program. So that when 10 we publish the TSPA VA, the whole program will be able to 11 stand behind it and say, yes, my work was probably 12 considered and folded into this TSPA. 13 So, by its very nature and construct, it will 14 be a different animal than what you have seen before, 15 which yes, did have a lot of input from site and 16 engineering, but not to the extent where we will sit down 17 with them, test their models with them, give them comments 18 on, and have them, you know, address our comments. 19 It will be a much more comprehensive and 20 defensible product than you've seen before, yet it will 21 still have some shortcomings. Hopefully, you know, not 22 huge and crucial ones, but ones that we can actually 23 address in the two years between VA and LA. 24 VICE CHAIRMAN GARRICK: Thank you. Ch (__,) 25 MEMBER HINZE: Abe, if I might, your waste NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234 4433
273 l 1 isolation strategy preceded, I believe, the recent l 2 modeling by Bo and the development of the increased bulk l O-s 3 infiltration rates. l l 4 Does this mean that we are going to see some 5 new hypotheses, modifications? What is going to be 6 developing along this line? 7 DR. LUIK: The waste isolation strategy is 8 being reviewed in some detail by the organizations that 9 are bringing us these results, to see if there should be 10 changes at the fundamental level, or if the changes that 11 need to be made are in the, as Bo indicated this morning, 12 in the area of where do we put our resources to build our 13 best case. rr (\ ') 14 For example, Bo talked -- where is Bo? Bo j 15 talked about, you know, making a distinction between his 16 model, his new interpretation of the mountain, and the 17 previous interpretation by doing a series of tests. 18 And I think what the panel of eight people or 19 whatever it is, I think it is eight people, that are re-20 writing the waste isolation strategy are now looking at 21 is, given our new view of the mountain, what testing can 22 we do to see if that should, indeed, be our new view, or 23 if there is some view in between the old and the new that 24 is really more appropriate.
/~h
(. ,) 25 So I think what we are seeing is a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234--4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
274 1 comprehensive review of what we've done to this point, in 2 light of this new information. O 3 And I can't second guess these people, because 4 they are representing each of the organizations, and each 5 of the viewpoints. t 6 MEMBER HINZE: Well, the important point is l 7 that it is coming down the pike. 8 DR. LUIK: Yes, it is. 9 MEMBER HINZE: One of the things that -- I 10 hate to say this. Bo, would you remove yourself? 11 (Laughter.) 12 MEMBER HINZE: One of the better diagrams that 13 I've seen today, in Bo's presentation, that laid out the 14 Yucca Mountain project UZ moisture gas heat and transport 15 modeling, this seemed to integrate things together, for 16 the UZ zone, and if you will, took the place of a long-17 winded strategy, if you will. 18 But I'm wondering what's going on, is there 19 something comparable in the saturated zone we've heard 20 here, today, about the unknowns in the saturated zone? 21 What do you have that is of comparable nature 22 in the saturated zone? 23 DR. LUIK: We were just delivered, a couple of 24 weeks ago, the first version of a saturated zone flow 25 model from the USGS. We are going to comment on that, and NEAL R. GFH3SS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 ,
~-- . .
275 1 I think by the middle of next year have a second version
- 2 of that model, that we will then abstract.
[v~) 3 It will be the basis for the flow portion of 4 the transport model being put together by Los Alamos, and 5 then we will abstract the whole package into our TSPA 6 analysis. 7 The reason that UZ flow is way ahead of SZ 8 flow, of course, is related to the regulatory environment 9 that we lived in, for a long time, which basically said, 10 calculate out to this point, and accumulate for 10,000 11 years. 12 Since most of our performance was in the 13 unsaturated zone, there was little emphasis on the
~N
( ) # 14 saturated zone, we just conservatively assumed parameters 15 for it. 16 And we are trying to address that particular 17 standard. Now that we have a different standard that is 18 going to require us to calculate, at some distance from 19 the repository, what the concentrations are, rather than a 20 cumulative mass, what the concentrations are, we have a 21 different ball game, and so we are doing catch-up on the 22 saturated zone. 23 MEMBER HINZE: So there will be something here 24 that will be released in the next month or so, or -- (- We are a 's_.) 25 DR. LUIK: I'm looking at Russ. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
276 1 team. 7- 2 MR. PATTERSON: This is Russ Patterson, DOE.
~
3 We've had two deliverables, one is the preliminary site 4 scale saturated zone flow model. That was delivered and 5 it was a level three, which was made into a level two with 6 a 5.1Q review, which is a system we go through. 7 That deliverable is available, it is very 8 preliminary in nature, but it is there, it is the site 9 scale, which is still a very large scale flow model. 10 That deliverable was also given to Los Alamos, 11 who has done a transport model on the saturated zone, and 12 that deliverable also is available. And we have those 13 now, and I assume that within the next few weeks, they (~h ( N 'l 14 should be available for people to have copies of. 15 And the next reiteration of those, of course, 16 will be, like Abe said, I believe March to June time frame 17 of next year. And those will be used for the TSPA 18 abstraction. 19 MR. ROBINSON: He wants me to talk about the 20 experiments at the C wells, that is going to be 21 incorporated into the curated zone flow model, and also 22 the saturated zone transport models. l 23 We have reactive tracer experiments, and non-24 reactive tracer experiments, both going on in the C wells. r"N. (_) 25 We are hoping to get those completed by March, so that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 V'iSHINGTON, D.C. 20005-3701 (202) 234-4433
277 1 they can go into the modeling. And there will be -- there g- 2 will also be experiments, h<pefully, going on past March, l v 3 but they probably won't make it into this modeling effort, 4 this year, in '97. 5 Is that good enough. 6 MEMBER HINZE: Can I ask you one more 7 question, Abe? Could you give us any heads up as to what 8 you are doing in terms of scoping of the critical group, 9 the work that you have going on, is there -- you mentioned 10 five and 30 kilometers. l 11 Is there anything that you can add to that at 12 this point? 13 DR. LUIK: What I would add to that, at this () 14 point, is really stepping into the bailiwick of the 15 environmental shop within our program. I attended, two 16 weeks ago, a three-day workshop on defining the biosphere 17 for Yucca Mountain. 18 And I'm pleased to say that we have a very 19 active effort of looking at what other nations have done, 20 and looking at what we have done at the NTS and at other 21 federal sites, like Hanford. 22 And the basic thing is that we have these 23 compartment models that describe the biosphere. What we 24 need to do is see what is appropriate for this particular s
. 25 location in terms of what pathways in the biosphere could NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
278 1 be important to creating the ultimate total dose. 2 Because we have not had a definitive model of Q) 3 _ hat sort, we have just kept calculating the stylized two 4 liters a day drinking water dose, what may be required for 5 the new EPA standard is a total dose, which would require 6 us to look at what are people actually doing in terms of I 7 growing their own food in Amargosa Valley. 8 You know, do they eat their own cow's livers, 9 you know, and that kind of thing. And that is what we 10 have the environmental group looking at. 11 As far as just moving the particular, you 12 know, location of the person, which we have done in some 13 sensitivity studies, that is very good. But you realize, ; . \- [ 14 of course, that until a definitive regional -- not 15 regional, but a larger scale flow model for the saturated 16 zone is complete and defensible, we are somewhat 17 handicapped, there. 18 MEMBER HINZE: Thank you. 19 MEMBER HORNBERGER: Abe, just a quick 20 question. We heard John Kessler give an outline of 21 possible ways to calculate things having to do with 22 colloid transport. How do colloids figure into your TSPA? 23 DR. LUIK: I believe that we have not 24 addressed them, except in a sensitivity case, in which we O k_) 25 -- and Bob Andrews may have to come up and correct me NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D C. 20005-3701 D'Vi 234-4433
279 1 here. In a sensitivity case where we lower the KD to ! i 2 simulate what the possible effect of colloids are. l O 3 But we are basically waiting for the word to
\
l j i 4 come down from Los Alamos, as to how to properly model 5 this. They have that particular assignment. 6 CHAIRMAN POMEROY: Abe, going on from the -- l
)
7 what is going to happen in the next few months, do you l 8 have a longer term plans to develop additional information 9 along the potential pathways, in terms of geological, 10 chemical information between the site and the proposed 11 critical group, wherever that may be? 12 DR. LUIK: Okay. This is an area where, I 13 think, I'm a little bit at a loss, because I believe that 14 we want to look at our ability to asses, you know, the 15 importance of doing that, versus looking at the importance 16 of doing -- further work in the unsaturated zone to get.a 17 firmer fix on Bo's model, for example. 18 That could be much more important than 19 swinging that dose, than looking at, you know, defining ; 20 the pathways. I think our major emphasis, even with the I 21 new standard, will be on near field and unsaturated zone 22 performance, with a more bounding estimate of what could 23 happen in the saturated zone. 24 CHAIRMAN POMEROY: Including the unsaturated 25 zone, below the rS!:vsitory? NEAL R. GROSS COURT REPORTERS AND 1RANSCRIBERS I 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433 j
l 280 i 1 DR. LUIK: Yes, yes. But that may change, you l 2 know, as the TSPA VA is dcne, and we receive comments back krw)
'~'
3 on tnat, we may have some alteration in strategy from that j l 4 point on. So, we are flexible. l 5 CHAIRMAN POMEROY: All right, thank you very ; i 6 much Abe. l 1 7 MEMBER HORNBERGER: The last presentation 8 listed for this afternoon is, again, Randy Bassett, 9 hydrogeochemical transport models. j I 10 MR. BASSETT: George, I'd like to defer my 11 comments to mid-discussion, if that is all right. I just 12 have a few comments, I modified them as the day 13 progressed, and so I hava an opportunity, in discussion
'# 14 time, I think, to summarize.
15 I'd like to do that, just for the sake of ) l 16 time, if that is all right? 1 17 MEMBER HORNBERGER: Sure, that will be fine. l 18 All right, so we are in the period of 19 discussion and meeting summary. I think that, actually, 20 for me, Abe's presentation was a pretty good summary of 21 the kind of things that need to be considered. 1 22 So I will just ask if people have points that l l 23 they want to make, points that they want discussed, 24 questions that need to be asked. l
- O ks -) 25 We might get to leave sooner than you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
281 1 thought, Randy. l 2 MEMBER HINZE: Well, let me throw one thing 3 out, here. And I guess this goes back to a question that 4 I asked of Bo in his presentation, and this is this term, 5 bulk infiltration. 6 We hear that -- one of the things I heard, I 7 think from Bruce, was that 90 percent of the flow through 8 the UZ is in the matrix. And I'm worried about that 9 number, and whether the bulk infiltration may be hiding 10 some very rapid flcws of sotne high magnitude. 11 We know that there is some rapid from the work 12 of June and her colleagues. And I'm wondering if in terms 13 of studies, what is being done to get a handle on the rh ( Nl 14 magnitude of the flow through the UZ zone by virtue of 15 fracture flow? 16 MEMBER HORNBERGER: Who can handle that, Bo? 17 MR. BODVARSSON: Yes, I'll try to handle 18 that. That is a very good question, because the flow and 19 the drift, of course, is very much controlled by how much 20 is in the fracture, and how much is in the matrix. 21 And if you have eight millimeters per year 22 going down the mountain, you would rather see a lot of it 23 in the matrix bypassing the drifts, obviously. It iF a 24 very good point. T'N ( ,) 25 I'll tell you my thinking about that, then NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 13?3 RHoDE ISLAND AVE., N.W. (202) 234 4433 WASHINGTON, D C. 20005-3701 (202) 234-4433
282 1 somebody else can add to that. 7ss 2 The rock property data in the unsaturated zone O 3 comes from various sources. The most important one is the 4 measurements that Lori Foint is doing on matrix 5 properties, where they are actually measuring i 6 permeabilities, porosities, and trying to measure one kind 7 of parameters, and things of that sort. l 8 And we have recently gotten a centrifuge 9 together to look at the relative permeability affects, as 10 well, of those samples. 11 You also know that we have quite a lot of 12 samples from the unsaturated zone. I would estimate a 13 thousand samples or more that have been measured, and give p.
-- 14 us some distribution of these properties.
15 What we do then is to try to calibrate those 16 properties for different infiltration rates with the 17 matrix saturations and capillary pressures. 18 So the latest ones, of course, using the Allen 19 Flint's map, which seems to be kind of getting us together 20 with some of the other measurements, and trying to get a 21 handle of the best property values that we can use in the 22 modeling of the UZ. 23 We have done two cases, recently. One of them 24 is in our Milestone Report, and that one was a preliminary n k_) 25 version of Lori's data. She hadn't finished her Milestone NEAL R. GROSS COURT REPollTERS AND TRANSCRIBERS 1323 RHCDE ISLAND AVE., N.W. (202) 234-4433 WASHING 7oN D.C. 20005-3701 (202) 234-4433
283 1 yet, and didn't have all the uncertainties with the values l 2 of the properties. 1 Q) 3 But what we are seeing in the I tuff 4 inversions, is that the modeling wants to have higher 5 matrix permeabilities in the Topopah region, than what has 1 6 been measured on the average, j 7 One reason might be is that you measure 8 permeabilities on little plots, and that does not take 9 into account, basically, the global permeability of the 10 matrix, that it is more of a conductors and parallel. 11 And the affects of the big one, the ten to 12 minus sixteen meters squared, if you are familiar with 13 that, versus the ten to minus eighteen. You may have m N- 14 hundred of ten to the minus eichteen, but that is one ten 15 to the minus sixteen is going to give you a hell of a lot 16 of flow. 17 MEMBER HINZE: Amen. 18 MR. BODVARSSON: Plus, of course, the little 19 micro factors and things of that sort. So one thing that 20 I've been thinking about, a little bit, is to try to get a 21 larger scale permeability, perhaps, connected to one of 22 these alcove, where we look at maybe a meter zone, and it 23 wouldn't be very expensive to do that kind of a test over 24 a larger region, to try to verify these permeability
) 25 values, because we are TSPA is going to have to rely a lot NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
284 1 on those values. 2 MEMBER HINZE: I go back to Randy's comment 3 about the 45 degree hole is a much better test of the near 4 vertical or vertical, and Lori's tests are missing those 5 kinds of pieces of evidence. j 6 MR. BODVARSSON: Right, right. So that is one 7 thing, I think, I'm sure that DOE is going to consider in 8 the future, in terms of what changes we want to make to 9 the plan. 10 We are all thinking about what else can we do I 11 to learn more, both from the saturated zone and the 12 unsaturated zone, but there is another evidence, also, l 13 that relates to fracture versus matrix flow. 14 That comes very -- at least it comes to my 15 mind every time you mention that. And that is the "= is 16 if you do a chloride balance and an isotol slance ne 17 perched water, you get the majority of fracture flow going 18 through the perched water zone. , 1 l 19 So that is another indication that you have l 20 zones. And we see that, in all our simulations. In the 21 Topopah Springs, for example, there are generally in most 22 of our simulation zones that 90 percent fracture flow and 4 23 other zones, that may be predominantly matrix flow. 24 Because it only takes like microdarcys of 10 25 or microdarcys of permeability in the matrix, to get NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
l 285 l 1 significant matrix flow. 1 g 2 So it is still, I think, an unresolved 1 gY 3 question, and one thing that we are thinking about, is to 4 do more measurement in the ESF, in the middle -- which is 5 the repository zone, to get the better handle on the 6 matrix permeability, as well as the larger scale test. 7 MEMBER H1NZE: Is there an alcove that is in 8 the position that is representative of vertical fractures 9 which you could get a sufficient amount of volume to do a 10 reasonable test? 11 MR. BODVARSSON: There is the thermal area, 12 which is a large area, where you can actually -- 13 MEMBER HINZE: That is not really -- that is ,Q t \J 14 not hardly the broken -- 15 MR. BODVARSSON: No, that is not the broken 16 lip zone. But, you see, you can make fairly cheaply, I 17 understand from those that know those things much better 18 than I know those things, you can make a fairly cheap, 19 what they call a niche or something like that, it is a 20 very short kind of alcove, which I don't think has to be 21 more than a couple of meters from the ESF drift, and then 1 22 we can isolate that from the drift, and then to a larger 23 scale permeability test that wouldn't be so expensive. 1 24 MEMBER HORNBERGER: Just one quick follow-up b) (_ 25 on that. When you say -- what did you say, you needed a NEAL R. GROSS l COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 244-4433 l l l
1 286 1 microdarcy to get a lot of flow in the matrix? s 2 DR. BODVARSSON: No, if you take like an 3 average permeability of the Topopah, it might be one 4 microdarcy, one times ten, to minus 18 meters squared, 5 that can carry through it, when it is fully saturated, 6 like point three millimeters per year of fluid. i 7 Now, if you are off by one -- it doesn't take i 8 much volume to get that, you have right there three 1 9 millimeters going through the matrix. 10 MEMBER HORNBERGER: But in your estimation, to 11 get up to ten microdarcy, you would be talking about micro 12 fractures? 13 MR. BODVARSSON: I'm not sure you may have to. ()T Y- 14 It may be just going from a small scale to a larger scale. 15 You know, this conducts some serious problem. I'm not 16 sure, Lori Flint would be much better qualified to answer j i 17 that question than I would. l 18 MEMBER HINZE: Thanks very much, Bo. While 19 I've got this microphone pushed over here, let me ask, or 20 make an observation, if you will. And that is that the 21 NRC ste.'f is very much interested in coupled processes, as 22 they should be, and are working diligently in this area. 23 And one of the things that I heard, I think 24 several times, today, is that we are left with not a good (D
._/ 25 handle on the thermal changes that -- or the hydrologic NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 287l 1 transport changes that may come about as a result of the
- 2 thermal processes.
N) 3 And I would include, with that, the effect of l 4 1 4 the thermal effects on zeolites. Does anyone have a 5 feeling for when that kind of information is going to be 1 6 available for the modelers to feed in to their studies? l 7 Bruce seems to be wanting to jump at that. 8 MR. ROBINSON: Yes, although I'm obviously not 9 the person that did that work, there is a lot of work that ; 10 has been done to try to characterize those types of 11 processes, and even somewhere to try to build them into 12 models. 13 For example, for the affect of heat on (~'\ \~/ 14 zeolites, there has been a lot of experimental work at Los 15 Alamos, looking at dehydration and rehydration, as one 16 goes through a cycle of heating and cooling. 17 The results of that are that in theery, as one 18 heats the zeolites, they do give off water. That could 19 impact the water budget within the zeolites, okay? 20 The processes at the laboratory scale appear 21 to be reversible in the range that we have studied it. 22 And, furthermore, when you look at the effect in a 23 thermalhydrologic calculation of zeolites and this 24 hydration dehydration type of reaction, they don't have a
/~.
k,,), 25 big effect on the overall performance system, unless there NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005 3701 (202) 234-4433
288 1 were an irreversible process that would have occurred, 2 that would have perhaps changed the sorptive O 3 characteristics. 4 That type of thing has not been observed, yet, 5 experimentally. So there are data, and -- 6 MEMBER HINZE: Are these data in the form of 7 journal articles, or how are they -- l 8 MR. ROBINSON: I'm going to need help with 9 that. There are journal articles. 1 1 10 MS. COLTON-BRADLEY: And I don't even work for ' 11 Los Alamos. Dave Bish has done a lot of that work, and 12 there is a lot of Los Alamos reports on clays and clay l 13 minerals. (~h
\ -) 14 He has published quite a bit of that, and it 15 is running around.
16 MEMBER HINZE: Well, the thermal effect on 17 zeolites are just, of course, one of the things. And we 18 heard about the -- I think the question marks that were 19 showing up in terms of precipitation and dissolution, and 20 that perhaps is more important than the irreversible 21 effects on zeolites. 22 When is this information going to be available 23 for the modeling? Is that already all available, or what 24 is the status? 25 MR. ROBINSON: It is a -- in going from an NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N W. (202) 234-4433 WASHINGTON D C. 20005-3701 (202) 234-4433
289 1 experimental observation like that one, to a numerical 7s 2 model, is I believe, a pretty tricky business. We took a 3 stab, a couple of years ago, Bill Glassley and I, at 4 looking at what, say, a very simple but perhaps one of the 5 more important reactions silica dissolution and 6 precipitation, what effect that might have at the 7 repository scale. l l 8 In a site scale simulation we hypothesized i I 9 that where silica was over-saturated, it would 10 precipitate, and under-saturated, it would not. And then 11 we further said that we thought we knew what effect that 12 would have on the permeability. 13 Now, precipitation in fractures probably most
'N/ 14 likely would decrease the permeability of the fractures.
15 Dissolution, on the other hand, I think the 16 experimental evidence, on a broader scale, is less 17 conclusive about what dissolution reactions do to the j 18 permeability of a medium. 19 I've seen studies where you etch away the pore 20 space, and it is highea permeability. I've seen studies 21 where the fractures collapse, because you have dissolved 22 away the asperities, and the permeability goes down. 23 So I think there are going to be some 24 fundamental problems with building those sorts of data f~) (_/ 25 into the types of models that we are doing here. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
290 l 1 Maybe at a drift scale it is more reasonable. 73 2 In looking at things like what would that do to choke off
)
3 a fracture that was intercepting the drift? 4 As you go to the larger scales, it becomes, to 5 me, more and more problematic. Bill wants to talk to this 6 one, too. 7 DR. GLASSLEY: Thanks. This coming fiscal 8 year we have planned a series of experiments, plugged flow 9 reactor experiments, looking at the rock material in four 10 different units, as a means of testing the coupled 11 reactive transport codes. 12 What we are going to do is, you know, a well 13 defined flow field where the material is already well b \/ 14 characterized, put through solutions of known composition, 15 monitor the evolution of porosity and permeability, 16 chemical changes and mineralogical changes along that flow 17 pathway, to establish the extent to which it is possible, 18 using current code capabilities, to model and predict the 19 behavior of that system. 20 That should give us at least a first order 21 handle on the extent to which our modeling capabilities 22 actually mimic some of the things that we think may be 23 going on in the kind of repository material we are talking 24 about, for a fully saturated system. /N (_) 25 And that information will be, hopefully, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON D.C. 20005-3701 (202) 234-4433
291 1 incorporate'd into milestones that will be feeding the TSPA 2 effort, by putting rolled up synthesized models that 7s !'~') 3 describe this kind of behavior in a form that could be 4 used in the TSPA codes. 5 MEMBER HINZE: Will those include fractured 6 rocks? 7 DR. GLASSLEY: The plan is ultimately to 8 include fractured material, not crushed material, yes. 9 But that will be toward the end of the year. 10 MEMBER HINZE: You will start off with crushed 11 material and then move to -- 12 DR. GLASSLEY: That is right. The reason for 13 that is that we want to have a high enough surface area I 1
'- 14 available.
15 MEMBER HINZE: Thank you very much. 16 MEMBER HORNBERGER: John? 17 VICE CHAIRMAN GARRICK: I just want to make an 18 observation and then a speech. 19 (Laughter.) 20 VICE CHAIRMAN GARRICK: No, I'm just kidding. 21 First let me comment, as a committee member, that I like I 22 the format of the proceedings of today. I like the 23 approach of talking about specific technical issues, and 24 then pushing them into some sort of context relative to b) (_ 25 bottom line. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
292 1 And I think that this Committee is going to
,f 3, 2 try to do a lot more of that. And I think it is obvious
)
\J 3 that we haven't done enough of that. 4 I think that it is very important for us to 5 see more clearly the connection between the scientific 6 work and the engineering design, and be convinced that i l 7 there is a connection. 8 I think it is very important for us to l l 9 appreciate that the TSPA has real purpose, and is a l l 10 process that builds our confidence, that it can become, 11 increasingly, the basis for us to evaluate where we stand i 12 with respect to specific issues. l 13 I hope that this is the beginning of things to l l
,,\ l v )
\/ 14 come. I would like to see a lot more alignment of l 15 supporting evidence for different hypotheses. 16 I think, sooner or later, if we are going to 17 convince the public that this repository can do the job, 18 we are going to have to do it in such a way that we don't 19 hide behind assumptions and a hypothesis. 20 I think, sooner or later, the waste people 21 have got to learn how to do risk assessment. And risk 22 assessment is removing the conditionalities, removing the 23 hypotheses, and putting forth what our best case is for 24 what we think the risk is. ( ,) 25 So I think today was encouraging in that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
293 1 regard, but it is obvious we've really got a long ways to 2 go. But some initial steps would be to begin to see what 3 the supporting case is for one millimeter percolation rate 4 versus ten. 5 Because I don't think that we are going to 6 convince people very effectively if our performance i 7 assessments give us a multiplicity of CCDF's, and you 8 choose the infiltration rate that you want. l 9 That is the way it is now, for the most part. 10 And I think that the kind of activity and the kind of 11 discussion we had today is a movement in a direction to 12 put forth exactly what we believe, and what is the 13 scientific evidence for that belief. 14 So that, I think, is where we are headed. And i 15 I'm hopeful that the TSPA VA will be a giant step forward i 16 in that regard. 17 MEMBER HORNBERGER: Well, I see that Paul was 18 afraid that I wasn't going to give that symbol of power 19 back to him, but I really was. 20 Randy, before Paul takes charge of the meeting 21 back, would you like to have a chance to make some 22 comments? 23 MR. BASSETT: Can I make just a couple of 24 comments? And this is based on this circumstance, and 25 that is that because we have seen both matrix and fracture NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
294 i 1 flow at Apache Leap, I'd like to make some observations 2 about that. ! O 3 And also a concern that because there seems to 4 be a diminishing number of field experiments on the 5 horizon, for example, I don't think the Fran Ridge 6 experiment is going to happen, things like that. 7 So there are some other block experiments I 8- think will be very instructive, that may not happen I 9 don't know. It is still on, that is good to hear. 10 Because here is a problem that we faced at i 11 Apache Leap, which was a little bit discouraging. l l 12 Actually, let me just read through a couple of our items, l l 13 it won't take me very long. ) O 14 only a small number of the needed. parameters l 15 are experimentally determined. If you really think about 16 what we are doing, especially for fracture related 17 environments. 18 We measure a lot of matrix parameters, but for 19 fracture we really don't. Instruments are required for 20 hydraulic parameters, boundary conditions, transfer 21 functions, initial chemical compositions, sorption i 22 constants. 23 And there is no easy option for ground truth 24 of these components So there is an awful lot of 25 estimates, okay? NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
1 295 1 1 Then the second problem that we face, that you g- 2 face, is the spatial distribution of rock properties is 3 not known very well, so consequently we tend to assume 4 uniform properties in these grids, and large grid sizes. ) l 5 Now, what is the problem with that? Well, l 6 when we do our predictive modeling, based on uniform 7 properties, we miss the fine structure. The problem with ) l 8 that is that flow and transport simulations are not ! 9 corroborated, then, by the independently measured 10 geochemical data. 11 So we end up with two data sets. The 12 predictions made from our hydrologic data, and then the l 13 independently measured geochemical data are not the same. 14 Then we have the problem of perching. Now, I 15 think I can say that based on the measurements that we 16 have made, and based on the projected measurements that we 17 would have ever made, I don't think I would have ever 18 predicted mid-tuff perched zones at Apache Leap. 19 So the consequence is, I can't model it, until 20 I find it. And I probably would have never predicted it, 21 yet there it is. So that is a dilemma that we face. 22 Next level. Fracture connectivity, 23 permeability and participation. Perhaps your ability to 24 transmit fluid is really best evaluated with isotopic n s, 25 indicators along key points, along the flow path. NEAL R. GROSS COURT REPor* ERS AND TRANSCRIBERS 1323 RHC'E ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
296 1 The problem with that is that it requires
,s 2 significant drilling and coring, and how do we know waere
( ) \v/ 3 these fast paths are without the chemical data? 4 And the final thing is that even with tracer 5 test results I am left with the dilemma of trying to bring 6 into conformity the isotopic data and the hydrologic 7 coupled chemical model. 8 And I found I had a lot of degrees of freedom, 9 when I really got into this. There are five or six 10 hydrologic flow models, approaches that I could have 11 taken. 12 And we covered this, dual porosity, dual 13 permeability, discrete fracture, uniform -- you can take rx ( I
' / 14 almost any of those models and simulate fracture flow.
15 And then you have the chemical freedom, which 16 is sorption constants, initial chemistry, non-equilibrium 17 circumstances, multi-components speciation. 18 And so when I have all these degrees of 19 freedom, what I found out was that the prediction was non-20 unique, and I basically just adjusted my prediction based 21 on the answer. 22 So I have to know the answer to make the 23 prediction. And the example, I guess, for you guys, might 24 be the c' lorine 36 data. I don't think you would have D. 's j 25 predicted that until June measured it. NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
l 297' 1 I certainly would not have predicted some of I l 5
~s\ 2 the things that we have seen at Apache Leap, until we l l
V 3 measured it. , I 4 But you know what is really interesting about 5 that, is I can model any scenario for which I have data. 6 So I'm a little concerned about the concept of 7 coupled modeling helping us a lot. Thermal, mechanical, 8 hydrologic, chemical coupled modeling really being the 9 answer, because what we end up doing is we build this 10 large model, and then we take it apart, and look at a very ; 11 small subset. , 1 12 And we assume, instead of, as you have already ) 13 pointed out, instead of dual porosity, dual permeability, /~'s 1 2 14 with all the other bells and whistles that we can put on 15 with our models, we can use a 1-D flow model with 1 l l 16 equivalent continuum and models it just fine. l l l 17 So I'm a little discouraged at my small ' 18 system, and how you scale that up to a large system like l 19 Yucca Mountain. 20 Now, I'm sure there are bounding calculations 21 that we can do to narrow down the dispersion a little bit, 22 but it is just kind of an observation that I wanted to 23 make about our site. 24 VICE CHAIRMAN GARRICK: What do you see as an q_) 25 alternative? NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
298 1 MR. BASSETT: I think, for us, the importance 7- 2 of geochemical measurement can't be over-emphasized. I (' 3 really think we need to understand the chemical data. I 4 don't think the hydrologic predictions, without the 5 chemical data, and iterative conceptual modifications, I 6 don't think that will work without the chemical data. 7 And I really would encourage, as much as 8 possible, the understanding of the various sources of 9 chemical data. 10 MEMBER HORNBERGER: In case anybody in the i 11 audience didn't know, Randy is a geochemist. 12 I think that Bo actually commented quite 13 lucidly on the difficulty of the inverse problem, and in l (~~\ l t }
\' 14 fact this is what Randy is referring to. l 15 Also, I think it was stated, again, probably 16 by Bo, that a multi-signal calibration is essential, and I 17 think that by all means the geochemical data play, I 18 thing, a huge role in this. Perhaps even more important 19 than the hydrologic data. Are there any other comments, 20 questions? Paul, I turn it back to you.
21 CHAIRMAN POMEROY: Thank you, George. I have 22 no further comments. It is late in the day, already. So 23 with that we will adjourn until 8:30 tomorrow morning. 24 (Whereupon, the above-entitled matter was rh (,) 25 adjourned at 5:18 p.m.) NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W. (202) 234-4433 WASHINGTON. D.C. 20005-3701 (202) 234-4433
O CERTIFICATE l This is to certify that the attached proceedings before the United States Nuclear Regulatory Commission in the matter of: Name of Proceeding: 86TH ADVISORY COMMITTEE ON NUCLEAR WASTE (ACNW) MEETING Docket Number: N/A l Place of Proceeding: LAS VEGAS, NEVADA were held as herein appears, and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and, thereafter reduced to typewriting by me or under the direction of the court reporting company, and that the transcript is a true and I accurate record of the foregoing proceedings. 1 x, // / JL (( CORBETT RINER official Reporter Neal R. Gross and Co., Inc. O NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHoDE ISLAND AVENUE, NW (202) 234-4433 WASHINGTON, D.C. 20005 (202) 234-4433
*l O
RECENTISOTOPIC (36Cl) EVIDENCE FOR FRACTURE FLOW AT YUCCA MOUNTAIN by JUNE FABRYKA-MARTIN and Q BRUCE ROBINSON LOS ALAMOS NATIONAL LABORATORY LOS ALAMOS, NEW MEXICO (505) 665-2300 (Fabryka-Martin) (505) 667-1910 (Robinson) ADVISORY COMMITTEE ON NUCLEAR WASTE September 26,1996 Las Vegas, NV O
O OUTLINE o Chlorine-36 as a Hydrologic Tracer e Objectives of ESF Sampling Program
. Data, Interpretation, and Uncertainties
- Elevated 36C1/Cl ratios O - L ss-elevated 36C1/Cl ratios e Comparison with Transport Calculations e Conclusions O ACNW 9/26/96
O teed u! uone^ela o o 8 8 O
$ $ 8 8 o
$xNmy - h ., ,!? }~
*h *,
q ;. --
~
s a. a ,- m s I .- e e e. e + Iu., v l
, Y*
m , 5 t 6
$ $tI el
>=
- i:. jf b .-
@ o 8 --
ll
,2
]$g,p J ;;g. _
,!?/ : :
,.y ./- . 8_
o ! n.
- m z
l l h'
~
g HC h' ' 2 , O
% d- . -
e- - C. " g , s T 3 T 8~ n ;, 7* ,7
'** - o__
l a o 8 O 9 8 sJeletN u! uolle^e;3
O CHLORINE-36 AS A HYDROLOGIC TRACER . (Half-life. 301,000 years) Sources Estimated Value Relative , at Yucca Importance at Mountain Yucca
- Mountain 4
(36C1/Cl x 10 5) ATMOSPHERIC SOURCES . Anthropogenic sources Up to 200,000 Dominant in
- Global fallout (peak global young waters i
- Local NTS activities fallout)
Natural atmospheric sources 500 at present- Dominant in
- Reactions of cosmic rays with day, but up to pre-bomb O 4 Ar,36Ar, and 35Cl 1500 over past waters 0.5 My IN-SITU PRODUCTION In Rocks and Minerals Near the Variable. Probably Surface Function of negligible
- Reactions of cosmic rays with exposure age relative to 3'K,4 Ca and 35C1 and elemental atmospheric composition sources In Deep Subsurface Rocks and 20-50 Generally Waters negligible
- Neutron capture by 35Cl ACNW 9/26/96
OBJECTIVES OF ESF STUDY # i . Evaluate frequency and distribution of ! preferential flow paths i l i j e Provide bounding estimates for the travel i l time of water in the matrix of the Topopah l Spring welded (TSw) unit at the potential
- repository horizon j O l
- Evaluate the extent to which the Paintbrush nonwelded (PTn) unit reduces vertical fluxes
- and/or increases groundwater travel times 1
I i 1 , ACNW , 9/26/96
1
- O APPROACH l 1
Comprehensive Sampling of ESF Rocks for Analysis of Chlorine-36, Chloride and Bromide l Sampling Category Sample Inventory, Stations 2 to 57, as of i 8/29/96 Collected Analyzed l
- O Systematic sampling 25 19
- every 200 m
. Feature-based sampling 107 71 Sampling of PTn subunit 22 18 contacts (usually three per contact) Total 154 98 i, O 9/26 9
DISTRIBUTION OF 36Cl/Cl RATIOS MEASURED FOR ESF SAMPLES i 5000 o Feature-based samples (fractures, faults, breccia, unit contacts, etc.) f = Systematic samples 5 4000 o i 5 _ o _ ; 2 -
$ 3000 c
d a o o o B_ O - o
" a -
l 8 2000 a o o y -- ._ g .
. g ~<
5 ..; . . . .' j"t o g.g n Estimated range o,
- - of meteoric signal E 1000 -
__.. 19..aj .,,,,3 o-ok lo"= 7"[ o 1,p,. "o :
.B a
y over past 50 ky
$ p,;,,,g,u 4;,;.;,,,,,,,,. o o,,,go i i i i i , , , , - Present-day ,
45 40 35 30 25 20 15 10 5 0 meteoric : background ESF Station l i ACNW , 9/26/96 ! YM06e. J l
LOCATIONS OF ELEVATED 36C1/Cl SIGNALS l RELATIVE TO FAULTS MAPPED AT THE SURFACE 1 1 , . , -s
' *, , i\ '\
o)\ O Elevated seCl/Cl ratio i.\ \###\\
- .\ss,,'s./ i
'\
.\ \ (ratio > 1500 x 10-15) s N %s \ t \ SCALE (meters) 4 1000 1500
'25' i 0 500 N i20 j zg ,
.\g,\ "5\ ..'s.Q{ N ,,j ,
! ,% 7 !
>= ==;_...p. N* 15 '\ g',es ' '
f 30 - 8'T \3 \ i
!['1 Sundance \ s% ,$ /- !! i i imd.. .,T \.5d1!10 , g',iti si l i Faulti 5 5 /,' \
- g. _. / s.\ g8!
. .i i ' ;;!i i
% o -,/b/:! ,!',! Ii,-b! l! ) Portal I d .
s
..s l :r i ib i,4
- sj!.ixj! '
,.r; l
40e. rym.m.. ; i e. ,; ,is
.. s ; :
i .O _ N
-.~...,'st.!
1 I I i *g 1
/ i
! ' ht. E .,,N ! \ !! l' l re - i 8 !
\. i \ '/
O.i8
\
3 x' i.i .is!\, ._ s' ! i1 s s 45- - s. - s E.
/c 's
! c J..._ . '/>!
11 i s < 8 r; a NN . j7 e is. !E II . t; !jiiij\ !? I i c:f .,% "' 9 ,,- s b . ,b h!
.....s 1 50 -
'! '/
.I Tk
'N ii f 51 %k*! '*%'
e= ':s. ., .!*l. i I.. ,! .11/ f. i. i. l,\, E'
*/T * '
l !aT'
.IE
{h 1/ -
. I!/ f: i n ,,,/
! t. ! : \ f 1, .. \ i t 55
. 'i , \ c~T..,!!!
r"'J d/ l) s!; !'rp. I ;/ I 1 i %~g _.,. . ts 5 f 'j R$, { ; i g. l f*gQ
,.5 iPO
/
.-' 8 s.,Is
.')),if ig; i i -
f. Q3 i \^x [x t i 6f/ ' Q :LN < . i9 ij
; j!
L.4 g
; o!9
!'i *; g i / -s
): \s 1
c 65 O 70)}k,/5 ' jj }
;; ii
** /\ \is; s,3iA4 '
\'. 3;~ill i!
5 l - q. Location of faults based on Day et al. (1996) ACNW 9/26/96 YM082f.ai
LOCATIONS OF ELEVATED 36Cl/Cl SIGNALS RELATIVE TO SELECTED FAULTS MAPPED IN THE ESF TUNNEL
= 36Cl/Cl ratios less than 1500 x 10-15 (77 samples) a 36Cl/Cl ratios more than 1500 x 10-15(21 samples) 1500 1400 o soo m s 1300
- TCw Bow o
;-- 1200
'? PTn \ d98
" Fault -
y ? TSw _ North 2 Portal en o mm ua.t h ,,, 1100 g 3.;c}ama. ) 1000 ? Sundance f} _ Fault Drill Hole Wash 1 il l 900 Fault zone Broken zone i i i i .. i :
-50 -45 -40 -35 -30 -25 .2u -15 -10 -5 0 ESF STATION Projected locations of unit contacts and faults are based on a cross-section in Barr et al. (in prep.)
Note: At least 60% of the samples with elevated 36Cl/Cl ratios are within 100 m of a major fault; at least 81% are within 100 m of any fault. ACNW 6c.ai
- . l 1
O PRELIMINARY ANALYSES OF OTHER BOMB-PULSE NUCLiDES i IN ESF SAMPLES
- e Iodine-129 and technetium-99 were both present in two deep samples that also contained elevated chlorine-36 ESF Station 2 (Bow Ridge Fault gouge),40 m depth UZ-N55, cuttings from depth of 53 m l
e Iodine-129 was also present in a sample from ESF Station 34+71, which contained elevated chlorine-36 iO e Cesium-137 and plutonium were observed in surface soils but not in any of the deeper samples l e These distributions are consistent with our understanding of l the geochemical behavior of these nuclides.
- Tritium was below detection in five samples from ESF tunnel walls, and slightly above background (up to 22 TU) in drillcore from Alcove #3 (Station 7+54, top of PTn)
ACNW 9/26/96
INTERPRETATION OF G l ELEVATED 36Cl/Cl RATIOS 1 AS INDICATORS OF FAST PATHS l i CONTINUING EVALUATION TO REDUCE SOURCE TERM UNCERTAINTIES e Sources of contamination e Cosmogenic production in soil calcite e Reconstruction of past 36 C1/Cl signal in atmosphere: What is maximum upper bound? PLANNED ACQUISITION OF CORROBORATING EVIDENCE REGARDING FAST PATHS e Correlation with topography and structural features e Correlation with surface infiltration estimates e Measurement of other bomb-pulse nuclides in ESF (3H,99Tc, 29}) e Measurement of chloride porewater concentrations in ESF as surrogate indicator ofinfiltration rates
- Solute transport modeling of bomb-pulse signals in ESF l
ACNW 9/26/96 h
O INTERPRETATION OF PRE-BOMB 36C1/Cl RATIOS AS INDICATORS OF GROUNDWATER TRAVEL TIMES (Slide 1 of 2) MULTIPLE APPROACHES e Establish upper limits for ground-water travel times from ground surface to the ESF based on radioactive decay of maximum possible 36C1/Cl signal in the O atmosphere e Calculate travel times by transport simulations using reconstructed 36C1/Cl signal in atmosphere o Establish probable lower limits for travel times by matching peaks in the reconstructed 36C1/Cl signal in the atmosphere ACNW 9/26/96
RECONSTRUCTION OF ATMOSPHERIC asCl/Cl RATIO DUE TO CHANGES IN RATES OF 38CI PRODUCTION AND STABLE CHLORIDE DEPOSITION Ratios are decayed to present-day values - {1500 _ 6 -
; :q :
51000 : m [ f\^v^\ev.
-]'
j 500 T 0 O 100 200 300 400 500 600 700 800 900 1000 Age (ky) COMPARISON OF RECONSTRUCTED RATIO WITH RATIOS MEASURED FOR PACKRAT MIDDEU SAMPLES FROM SOUTHERN NEVADA 1500 = Packrat sample - b : = n V: b1000 = ' ^' M' - e - _A'v . vW%r'[ U 500 .h g _ _
~
, , , , . i i 0 5 10 15 20 25 30 35 40 Age (ky) g l ACNW 9/26/96 YM102.ai j
i f O INTERPRETATION OF PRE-BOMB 36Cl/Cl RATIOS AS INDICATORS OF l GROUNDWATER TRAVEL TIMES i (Slide 2 of 2) i CONTINUING EVALUATION TO REDUCE I SOURCE TERM UNCERTAINTIES l e Ongoing evaluation of sources of contamination ! e Evaluation of surface calcite as additional source e Reconstruction of past 36C1/Cl signal in atmosphere PLANNED ACQUISITION OF CORROBORATING EVIDENCE REGARDING TRAVEL TIME e Correlation with topography, structural features, and surface infiltration estimates e Comparison with U-series dating of fracture minerals e Correlation between "C and 36Cl signals in perched water, borehole and ESF samples e Measurement of chloride porewater concentrations in the ESF as surrogate indicator ofinfiltration rates e Solute transport modeling of reconstructed signal ACNW 9/26/96
i i Percolation Flux - Porewater Age Relation Simulated Breakthrough (Cumulative Sample Age) at Station 35 for DifTerent Infiltration Rates 100 g,-
- _ 1 v I l C 10 - j .
G . i 3 g l t O 6. I i 5 M 1- 3 J e
. ~
k ~ k I b w E EI E 0.1 - S mm/yr / - 9
- j Increased Fracture *lI
.2 Perm. In PTn a' .
g ,___________ 'gl 3 0.01 - l cd l E
'i I o
s l es : 0.001 . . . . : . . ., ........, . . . ....., /, . . ..... . . . . ....i I........ 1e+00 1e+01 1e+02 1e+03 1e+04 1e+05 1e+06
- Time (yrs) .
j e o - -- e
O O O'
"'"""o/ci g". os y$)
in u?se " ' ni,,,,,re oo "%,o," ga,'a'ion
'ent ,,+'/ .
?
bl
~
9 y' j <;A g;.,i,z, /
#c r
n- -
% M~ -
a 1
^,
/ ; ggo;%,g%Q;'
l-l'h^~g
- . . s ~
~ . , 1 y ds,.
. +^ - ,
5' N
/ .;
f*,4 ' m- s
/
O 3
~ 00 GOO 900 I200
%cO6 %' rap G
- a n a j,,,. ,. '" A L"S Ala'hos
RECONSTRUCTED g ATMOSPHERIC 14C AND seCl ACTIVITIES FOR THE PAST 50,000 YEARS l m1oo - 1200 p O v E 80 - asci 1000 S
*[J h -
soo E h 60 '5 Q soo 85 l W 4o 5 y - 400 g o 20 e 14C - 200 M o o 1o 20 ao 40 so Age (ka) O COMPARISON BETWEEN MEASURED AND PREDICTED CORRELATION BETWEEN 14C AND seCl ACTIVITIES
^
100
= Measured values for perched water O a Reconstructed values for past 25,000 yr E 80 a -
2.5 v - y
) 60 Isotope 5 correlation .$ 40
_ =
, y curve ^
g a - 10 d g 20 - 33 m 20 g , , , , , O 200 400 600 800 1000 g 36Cl/Cl Ratio (x 10-15) ACNW 9/26/96 YM107A.ai
\ 'O CONCLESIONS IMPLICATIONS OF 36Cl RESULTS FOR FLOW IN THE UNSATURATED ZONE AT YUCCA MOUNTAIN e The bimodal distribution of 36C1/Cl ratios demonstrates the existence ofisolated fast paths from the surface to the ESF.
- Penetration of recent water into TSw unit is indicated by l bomb-pulse 36Cl in ESF fractures. However, bomb-pulse signals by themselves do not indicate magnitude of fluxes.
Q e Working hypothesis: Fast paths that carry water into the TSw may be associated with major fault zones that cut through the PTn, in conjunction with areas of higher-than-average infiltration e Transport calculations indicate that arrival of bomb-pulse 36 C1 at the ESF is consistent with increased fracture permeability in the PTn, as may be associated with faults, in combination with infiltration rates of 1 mm/yr or more. e Transport simulations with base-case properties indicate that infiltration rates on the order of 1 to 10 mm/yr (rather than 0.1 mm/yr) are required to yield 36C1/Cl signals in the range of 500-1000 x 10-" at the ESF. O ACNW 9/26/96
-a,. --.. - - - .- ..+x- --~ -s-. - -a =s.---.- - . -~a. -.s-- - - - -,--a a. -
y---. u.-- .- --~- - - -. .-.-----a.-a-.. - - - -- .
, D i
l N0 O O O
O O O~ w. y
- u. . . =
Deep Percolation and Flow t -Ph1?. .,deling at Yucca'Muuntain
.f,...
s
. 7, : *
.- y. , s, ,
. El 86th ACNWMeeting .
#~
September 2$,,4996 .
, \,+
1
$ 'W Law
?G.S.
Berkeley V.f 80$,% bit, Yon . 1 onal Labora'$"fh hv' ' tj o/ 4 E
/f .
f
^
A U [ ERNEar ORLANDO LAWRENCE BERKELEY N ATIONAL LABORATORY prar=== = EARTH SCIENCES DIVISION ACNW'96
x' DEVELOPMENT AND CAllBRATION
.\.s i s OF THE THREE-DIMENSIONAL A -
SITE-SCALE UNSATURATED ZONE q
'\ \ MODEL OF YUCCA MOUNTAIN,
~iI*s.
m ._ NEVADA l Cross Section: S-N (ur.in43 s < -- > N [A 1soo -
- "' EDITED B Y:
G.S. B O DVAR SS ON i l noo
-- . .._._._ i,1, s
lll
.. AND g ,,,, ; i . . . s. I j .; .
.2 I ,4 I i i no
. . ;' i . ; E lll
\-_ ..]i ,') . . . :4 . -J-' >1- -
- z. -
.<s 900 I - *
.1-
. ... ~
.o 4,
- h. ~~
g 'l ' ,, .# g v. .e g e
,s
/
l:*,U -
. ,s. l-C e'i
\,\ o ti,-
S
$-[
=
AUGUST 1996
.s
\
fffffIf !!! co rsT c aL ANoo L AWRENCE BCNKELcv NATIONAL L#WORATORT i'<'<"-- 'i EARTH SCIENCES O 1VI S I O N
Flow Modeling and Percolation u UZ Model i e ibration a ion Flux Studie a Te Reduce Un rc Ig
- a Conclu k gs m
4 }4 l f I c ~EBra LAmoD tAwaE~cE BERKELEY N ATIDNAL LABORATDRY scaxc'cv'^- EARTH SCIENCES DIVISION ACNW'96
g i i l gi ia g i e a a . . g 2.380E5 i----------------- --A-----------}---------------- -j---------
- _ j ' '
- : ~
i fMG-2 i j _ - - A' .
-{ 3 l' .'k :
---a ,
s L { e w M :
- 2.360E5 +--------"-~o * -s --i---------
'yx - --
s
- i ~g , . *\ i -
E i o - :,
~
4 ' l V
' , .g dUZ-14 :
- p : f :
] jf E G-1 .
'N i
~
4 : d- '
>c :
22.340E5 +------ --Mf , H-s
- + - - - - - - - -
O ,
. ; s .
SD-9 - oc 7 q . J
.1 O ,i . ',1 ' .
sM:
- - , t i -
O i ' 3 ,,7 i O 4 i ( x .flw)i . O2 4 320E5 1
+-----------: - f}
.k,j. --- -E 1 i', 6 'e
-+-
i
- > El m ; < l
-: - . I SD-7 '
.c :
p; uM/ g- g j 5,,' - r F M x (8
. e/:$! _
i s e i . i 1 ' -< u-3 & a
-i , . }. '
i 02.300E5
$----y-- --i---------
~i , ,J , , i , k? , , i . L , , i ,
1.680E5 1.700E5 1.720E5 1.740E5 EAST NEVADA COORDINATES (m) A fffffff lli ERNEST ORLANDO LAWRENCE
~
BERKELEY N ATIONAL LABORATORY fi 53 g _ . - , ~ _
-,,,-~ yn
.,hi . t, i! ,ll,il l.' l1' c
e ig s s f
- x 6 S 9' I
-O WWN C A y g
, ~
J
/,
4 3..
?
g '. e L
, Q' - , %3M, .4.
n t .. a ? r V'M. - t . S . J n sy.,g : ;. N g. o - i . L. L,s._+-: s, t . a n . l o E n N.. .
- y. ., .
9 = s
' %a*y ..
I M y
/,/
4 - e p ,
~,~p t .
s (7/,n, p .,. a . wf
. /
W
~ -
O A ,
~a
. c :
;n s,
d
},
a- ' O. i o n Ol t
- '- u a -
' i D
x a.L ?
%"3.N'
. i'
?
u i? . l F S
, N
' 7 - ...
- ' ~-
+
+-
n j y2_ Y' s i o n _ ~, ~ ~
/
t ; , _ a
'--(W;:.'y//,
.^ * ,
l > o -
./ -
c r , e - .
~ .,6
'/
P '
' h >-
- N u W', ;%
g y, O
l Yucca Mountain Project UZ Moisture, Gas, Heat and Transport Modeling Fracture Properties infiltration in-Situ Environmental isotopes Fracture / Matrix Int. A. Flint Conditions J. Fabryka-Martin & A. Yang L Anna & R.Zimmerman J. Rousseau I Pneumatic Data Matrix Properties
- l L Flint & C. Rautman UZ Site-Scale Model E. Kwicklis,G. LeCain, G. Patterson, &
Conceptual Models J. Rousseau E. Kwicklis & B. Bodvarsson A ' Geology / Geophysics 3y t M. Chornack, R. Clayton, m m ESF Moisture Balance W. Day, E. Majer, 3D-Numerical Models D. Chesnut, A. Flint, R. Spengler, & W. Zelinski B. Bodvarsson & E. Kwicklis & J. Wang I I , 1f 1 f 1 f Moisture Model Thermal Model Gas Flow Model ; Y.Wu C.Haukwa R. Ahlers & N. Lu I f lf II lf if r Transport Model TSPA Thermal Modeling Gas Transport B. Robinson R. Andrews T. Buscheck & Y.Tsang E. Kwicklis & Y.S. Wu
. _ . - . . . . ~ - - - - . . - - - _ _ - - - . - - . _ -- _ - - - . . - . - _ . - - - . - . . - - - - _ - . - _ . - - -
Borehole
~
Temperature w Profiles Fault
- J ' -
Air-K Tests
% Ghost Dance
._ %g K fault Variable Spatial &
'w( -
Sofitario '
. p- k* -
2.,,,,
< ' .y Temporal Infiltration Canyon -
( s Fault -- - g Bo idge
# 9r Je s
Qww',.
' i rm um
.-{, ~~
- *K Q', "
s ,gg;g
- a s:$i *'. u _ . h 1, .
.. ,[j ..PTnqg_; ;- ,,
y . w m.,. . ._ ~ - . ,r.,, .. .. 4
%. . p _ . . . :
. . . _., w ~~1 N pe<.:. -
- s. ,, w w.n. + --
N-
.g*.-- -
Temp.,l
- ., , a, , ._.a_u , r , . : ' * - n.~a -
4 a :. e e se v -
,, ,. - > * . Rif Data K
# ~ p ., , . .--% u ~ = ', ( h ,; . j
~ ~
_ T ,
,, _ , . - . Q . ..
Perched Water ~ \ ~ b . (Young Age) ' 2l!' J' " _ l .
,.s.
,l// w
..#5
\ ' -
\& /)w;Lr
/
v 1.i. ll
.._ w.. . . c. , .w . . 9.
i
< , . , . .e.m x. .
n p #, . N-- .*-t. . gc g
/ .
. U/ p ~u
/ igp?c d l f f
^
// x;,,
.,-- khi.1.
~'- ) .
A r47 i o Q. g.' Wg .g ] ' gg. ;+
$ re ' pnr -e *= +w r -s w,
.s - 7-- , - r- r WMx">
f sqw:. t ;..w m mey L,Md?%3tg*g.2,y. g.3 g e t 2."s"3g*g l Gas Pressure l
^
A IIfffIf lll ERNEST ORLANDO LAWRENC:" BERKELEY N ATION AL LABORATORY Old X5cc BERKELEY L As EARTH S CIEN C ES D I V I S I C3 N TechintMtg'96 .
UZ Flow Model Gas Balance Water Balance Energy Balance h Conceptual Conceptual Model Conceptual Model Model -Geological Issues
-Heat Transfer Mechanisms
-PTn Microstructure -Hydrologicalissues
-Heat Transfer Functions :
-Fracture -Mathematical Parameters Formulation
-Anisotropy 4
$ Heat Flow Data Perched Water at Surface Atmospheric Location & Volume A A t y Pressure Signal
$ T
, Temp. Data Perched Water Ages l n from Wells N
' Gas Pressure Sat. & Moist. I Wells V Temp. Data Tension from Wells s from ESF e
Gas Pressures Environ. O Due to ESF isotope Ages N Heat Flow at g E Water Table Y P Gas Ages from isotopes i I ESF Moisture Balance T M f Inpact of o i o Major Faults impact of f Fracture Coadng Ages - Major Faults T I L * * '
- Data from Ghost M
- inishe:i Dance & Other Other Major Faults 1P Major Faults Y Finished g CALIBRATED UZ - Moisture / Heat / Gas Model y=
O O O Percolation Flux Indicators Percolation Flux (mm/yr) infiltration 100 Saturation & Moisture Tension Data 10 Pneumatic Data Environmental isotopes I l l Fracture Coatings 0.1 l Temperature Data L j Perched Water Data 0.01 l
^
A ERNEST ORLANDO LAWRENCE l BERKELEY N ATeDNAL LABOR ATORY rnewamen EARTH SCIENCES DIVISION ACNW'96 figs
.,,,,,,,,,,,,,g ,,,,,,,,, ,,,,,,,,, , , ,
,,,,m, m., , , , , , ,
sq;. . 1300 TCW , j **, f,g!, 1300 - l h y., TCW ..
. .. ,; < ,c -
- a. o ,.*41*,
9, g
-----. ... y _
. .t 1200 iTSw *
"E...- -
., .:, . . i 1200 e .e .
.i.'. 6: :: .,*
~
5y
'.', ^ .
TSw -
- USGS Data ,
0 Ave. Data *. *~ v E .
..4*#*
1100 - * .- . _------------ Uncalibrated I g,j'f - 1100 - g
.i* -
g . Calibrated ,. 4 _ *; .8 ,
' I , i Hydro. Unit .#
1000 -
., t .th 1000 - . # ".g - -
i m . .xm.
.e._ -
,._ s.
==* ._ __.
LU - 3 o
-CHn .
..,9 ,s.
900 - CHn: 900g -
.po
{,, . .:* * * *
- o -
i 800 L *h!l.:.*.; 800 L
~
.*..=,*._
.c .l .'.*o
.Se. .**.'#..
700 '''''''''''''''''' _ '""' '""' '""' t 700 ' ' " 0.0 0.2 0.4 0.6 0.8 1.0 10 10' 10' i 10* 10' Liquid Saturation Capillary Pressure (bars) i A L b fffffff lll ERNEST ORLANDO 1.AWRENCE BERKELEY N ATION AL LABOR ATORY (jg3433
. EARTH SCIENCES DI ION in Ug'96 f
USW UZ-7a 91 89 -
.__ . , g gg - -
9 87 - - - g e so I' . se
- n. **
J
~
- t -
i. w es l i , , \ u - 87
.- as n - -
i ! Prediction .. 86
' .i! . . . .
.i...i...i.~,
H -58 -28 o Julian Day 1996 28 58 I J i l l 1 I _ - _ _ _ _ . - _ _ _ _ _ _ _ - _ _ - _ _ _ _ _ _ _ _ _ - _ . - _ _ _ - . _ - \
l 9
. . / '- '
i . j~ M l Q 4 ) 4 .y ^3 ;! x ., f ,
- ' \. I,
j ..
'4 .
! k
'? A h,p? ,*. .: Pneumatic ,
.h[
.42f. j ] l Diffusivity l k ypk g
%, l'i #: ._
. .J (m'/s) l
).
l
'.'@) 1.00x10
39 g,; .. %'% *. i l 5.32x10* 4 I i -- 4.34x10 i {N f, hg ' 2.06x10* l l p l .b 4.69x10 l j 3.95x10
$ g nif U 4
$ f $k 7 g j .
2.75x10 l QlI ,.a - 1.67x10* 1.00x10 4 o i . ?' l 171000
._ butE.
172000 Easting (m) 173000 174000 i O l l l 1
O O O \ _i ..i....,.... _ 238000 - J. st 1.00 I 0.99 k $ 0.98 y 236000 - o,'g g 0.94
$ - i Z-1114 0.92 i
$ _ - 1 3 . 0.90 5 _
) NRG 7/7a 0.85 U 80 234000 -
g , O ~
~
0.70 i (U 0.65 0.60 , 0.55 Z 232000 - - 0 50 _ .- 0.40 l 0.30 !
~ '
O.20
~
O.10 !
~
~
0.00 230000 - - i ' I i . ,,I i,,,I , , , , 168000 170000 172000 174000 A Nevada Coordinate (m) ffIffII lll ERNEST ORLANDO LAWRENCE B ERKELEY N ATION AL LABOR ATORY ' 6 7536 BERKELEY LAs EARTH SCIENCES D I V I S I C3 N TechintMig'96 ,
i Temperature (C) l
~ I
~ g
~
\
Mg /750.0 ; 17 @ O wT7 7, H-8 MRg 1650.0 !
- i. ; I6MO j
E M 155ao l
#. # Q [l M Ismo
" LW) 1 ggs 4
,,'.l ffg0 ,
y imo
~
a< m _q, "
^~
/J@.0
} ' *~~ a g.
a, _y, 1250g e wK4 * '
$ .w,.
1800 n g s 120ao gn iisao
?, - , '
%.. cog E ,
,u ~
y,mg
. g#
* ' di
~
70sao i h*-
\ uo, e l~
y~ Im79
' s 9300
~""
l
\
~ #> AMO I
# 45' .c
%s 830g j
\ .
r 233CCC emo
- , s.
D *-1 - + ~ r '> t. 236000 2370C0 f500 [ 2.15000
..,~7, , .., r 0*** 2$300 m'0 2 4000 700 0 p.
23290 Northung (m) &[
. 3 231000 23GOOO i
l IIII I lll ERNEST ORLANDO LAWRENCE BERKELEY N ATION AL LABOR ATORY {; ggy e u r .-. scie ~ces o ig i a s g*'u'r96
m as abu men.m am_ mm e m e= m ..mem.a. m Am -eu u m N.E asm +hm esm.-em e- m -mm a w pm.ma a e m .=*E.ehh6m a L ea m m a..w emm mewmaema w a mum en r .m gumum e a e amm.em. %.mm, man.-*; m.ai..hai. .hamM.a..% & 4_ . A a he b _W __ p 4,,_ g , A h4mm, b. A MA m&.a R+A69.64 h &h-O O O /
"2k"$:::y., 88) i
~
hp
- del p '%/c *W i g, l
ay. Ih' y Perco,dym "c~i ,,, l
/g
Use of Temperature to Estimate Percolation Flux l
~ !
I 1200 *,~ I* - u . .
- 3 i .
0.1 mnWr 238000 - I
----- 1.0 mmfyr I
10.0 mmfyr Jg, 's, .f;3 p 1100 -
' ' -'~ 20.0 mm/yr , Temperasw ia y s
. m nr sere r
.\ , A em ,, (" cam) i I ,\ '. y 236000 - 2s i
[ s
~
'\ ', dT/&=0.020
- Cart e 2s.00 g \g ,
' w >= ;
s, dT7dz=00208bt .C 2440 a w soo ,$7/g=o,ot37 eCfmk ( , y g 234000 - MM 22.50 2m y' 's o / 22,5o
' dT/dr=0.0188'C/m \ 's, -$
2m soo .
. 's, - %
{ 21.50 2m s e i 20.50 i
.s 's, 2 232000 -
20x0 !
. , , ,\ , 3 , toso ,
e as . . <= ; TEMPEiMTURE,1N DEGREESCELSIUS * , Comparison of measured temperatures at ~
/
~
l' UZ#5 with simulated temperatures using a ' 168000 170000 172000 174000 one-dimensional model and steady-state ! afiltration fluxes of 0.1,1s0,10.0 and 20.0 Nevada coordinate (m) ' mm/yr (Rousseau et al.,1996) Estimated temperature gradient in the
^ TSw based on data from Sass et al.1988 A and Rousseau et al.,1996 (Bodvarsson &
f""" E' concer ont*~oo u4wacuce Bandurraga, eds.,1996) acaxe,_cv w vio~ tt..onaroav l op,, ,,, ,, wings 2 EEEEEE!IEEEIn e EARTH SCIENCbS DIVISION 4Wsay '
0 O O Vertical Mass Flux at Repository Horizon l '\ i - x I ,'i tQ f h i <.
/
r . gm f
/
p!' i fM I% .. i
/ Shj !
bdi
}hY "
4 i thi 1
-as
-lo
/
l
-75 j
-20
-25 ( '
-30 .. ;r-
-35 ;
P *
-40 '
y
-45
~50
-15
-60 ,
i l - 1 f , I
\t "T?
^
j t A , ERNEST ORLANDO LAWRENCE i l B ERKELEY N ATION AL LABOR ATORY
- g rnerern e EARTH SCIENCES DIVISION ACNW96 figs i
i
8 y 2 _1! o o i
>iii..i..,,,, ,,3 , ,o b'-
(t) ,
~
r c _
'< / . ,
s.( -
. .,., ,1 -' /r se ~ ~.
.o~ t -
m O V < s , m- c-
*g:;
,, o ,%,! , A,.(.. g" ~ ..q x)-/4. ,. .u -4 .
. 3,>y '~' x. ,.-
,'- . C Q #
'f ,.).~
' m--- /-j. , : '
w y .
~.%' y).', , / ./ L .~&.\ ; '. ,7 .\< &~,Q: O g [
f' Q ~ _j f'f's.p. y41/, 6:'~ .4:' g g :- o(w@, y{s): jo f 7 <. s/fi/
,N,X / )F ,.
x ps ,.., ~.4-4 2;1
~
m 0
~
?'> $ -
O - E W .& 'PW5 sSi w
? t. %htC -
0 o . ./ -L : e 0 0
./ d:-c:h..>.:., 4,j.wk.z .tu .7m% 'ne&: 9;u+w {~= cm ~t~
y ,- y c ..,, t.,..g,: ; _Lm r;.) s.
.>,-v.+ _
o o ,a o w <,a m y,s :u. gY'rg't,4.~yy-.y+ 1 .~r,:m'
' 7t[ r s Aif.. p5 ^o.
.9 L t-; p g, aa w g 9 a y.1,c,c,7..e
- > . y g.c,
-l ,
gl.. _ p ;' 58 0 R ~ r .;i ;f..:',l 4 '#i ) A g _ uoe.,soon,m - ., . , , X . Z $jZ
<;w 1
, , +...
' N J <
_ e . o
. i , , , , i , , , , i , , , , i , , ,
,' i , -- g az oo g e
o o o o o o o o o e
" iy@
J o o o o o aZ m e v N o 1 n n n n n OF N N N N N p w u a r-Ifl W % ww (m) eleulpaooo speAeN i5 4 wmW
<m 6
6 6 f' % s kbwi
ca . -m m,%,g. __-u
--==--=4=6w======.+====-_-m4.m--m==--eem - - - = - = - - = = - - = . -----=.-womau--*===-%meam o ,;e***
/~// /'/>%m
/ </ ~
t 7~% s
'N 4 >
r ,)y"x~~ .s
~~x,,,,9s
.A , ~
p
/
/
.p g
\ '
1;i\,
\^'
s O.9 s O 1 7 015
\
r 03
-,O
' 0.65 g g6
\
@^
0 55 ,a
.> ~
0.5 { g,45 g2 N, 0.35 O.3 0.25 y.d 07 f.y. 0.15 . 0 05 [.. IIffffI s III ERNEST ORLANDO LAWRENCE BERKELEY N ATIONAL LABORATORY I'",8720 \ EARTH S CIEN C ES D IVI S I O N 3e&\** gqb
.~ - . - _ _ _ _ _
- __ _ . _. 7 ater Saturation at Rep sitory Horizon o - a.
1 i
/ ~ u a. s l
t
!- \
/
/ XYZ. 1 2
1.0 I an \ 1800 , a 1 j D s s a, ars 400 C t 06s ma
.: t s
"l y :M:
Q , G3 0 25 j D sti n9 ai
# '5 00 go (tY at _
aos IIIIIII III ERNEST ORLANDO LAWRENCE f B ERKELEY N ATION AL LABOR ATORY g 9 E' A R F M S CIEN CES OI ION nng'96
Vertical Fracture Mm Flux at Water Table 1 a s
'l
\
I en \
\ ?
W' s 4-g,A
~
\ ., b - 9, ~
j \
/ N*
T- % /'<>. d\ l :s N%,,
! 'N3
< lY ao 0
-0 025 IfN[
-Q 5
-Q ! ' 00 $
-a m
-0 15
-0 175
;hA -02
& mm 0 F) i M'
1 j M sn arts s ;ng o ga . %.2 J -a3 aars I FH B aas an ~ llI ERNEST ORLANDO LAWRENUE [IIIIII DERKELEY N ATION AL LABOR ATORY TechintMtg'96 N EARTH S CIEN C ES DIVISIO N
l Cross Section: W-E (uz-1/14) 1800 2 SL 1.00 0.95 i 0.90 1600 - solitario -
,. 0.8s Canyon - ' O.80 Fault -
_.} 0.75
~ ~
(/) Z-1/14 r 0.70 O
~
\ ~
1400 _
\ _ f 0.ss h
0.60 0.55 C 0.50 1200 __ [4 ,
> A 0.35
.9 ,
0.30 W m 0.25 I 1000 -
,- --- 0.20 s q 0.1s
_ , p 0.10 p 0.Os 800 - - ll 0.00 lll l l -
. . . . i . ... .... ....i c 170000 171000 172000 173000 .
Nevada Coordinate (m) ;
-w
/\ ;
I I Ill ERNEST O RLANDO LAWRENCE DERKELEY N AT,0N AL LABOR ATO RY q e m, - scic~cce o ig i o ~ g, iu.os
O - t Cross Section: W-E (uz-1/14) 1800 2
''''''''''' L st f
1.00 l 1000 yrs o,gs ! 0.90 4 1600 - somario - 0.85 Canyon Fault 0.80 O (O
' k 0.75 cc
\ Z-1/14 s
h o,70 1400 - 3 0.65 t E - 0.60 g.) r 0.55
}
h 0.50 l "@ 1200 .45 l (U 7
> y 0.40
- 0.35 h -
0.30 W 0.25 i 1000 - l- m 0.20 l 0.15 j_ I 0.10 l
. 0.05 l 800 -
1 0.00 8 t iI - Ill I i t
' ' i ,
. . . . i .... i
. . . . . . . i .-
170000 171000 172000 173000 Nevada Coordinate (m) , GA IIIIIII lll ERNEST ORLANDO LAWRENCE ( DERKELEY N ATION AL LABOR ATORY N EARTH S CIEN CES DIVISIC3N TechintMig'96
,s f'w J y /1 s f' s 9 / / / / /,,.- / r# / j.-' ,/ r ,1 l
/ /
"6 m.
O \' 9s k %e , x
^s O 4 7
+ j*
x . e " s 4(e x . d 4: # 4 f $x_ ' ' s . y ft j gg . ' a{if.k? m;1 y
\j-vg}{4 N
'x 4 ' s
+
y ,p,v N q+ =
; g ,,
k \ n \ E N i 4 ~ g @
?,
gtg i
- EARTH SCIENCES DIVISION rrrrrrr h O
Scenario 3: Simulation with """"' " Ventilated Drift drift scale runs with steady-state not infiltration, considering situation afterdrift drilling, without presence of waste canisters 50 % relative humidity in the drift due to ventilation - evaporation at drift wall creates strong capillary gradient presented for XZ_1 Plane Absolute Permeability Field (in log,, m') O "
,,I N :
T
. hE . . 5 r
.r'
=
- '0
-i2.50 li$
2.0
- I . * *
,,, , e 14.00
~~~
-' d 5' "E
N 0.0 . r k'*r*~f, E s 15.00
- 1 :h -15.50 g'
i? " . .
-=a; ;; _ , .
-iu0 16.00
. . .g .. -17.00
.0 4.0 -4.0 2.0 0.0 2.0 4.0 4.0 x[m]
O
m EARTH SCIENCES DIVISION rerrtre air uno ( Scenario 4 Increase in Magnitude of Flux compared to the initial Situation for 28 mm/yr Pulse (in % mmfyr) s
- je 2.00 g -
' 1.50 4.0
- 1.00 0.50 M 9 0.00
-4.0
-4.0 -2.0 0.0 2.0 4.0 6.0 x [m]
i l 9 l l= .. . . _ . - -- - . . . - . . _ . ._. . . _
- EARTH SCIENCES DIVISION .,
n errrrer a !U i """"""" l , Scenario 4 1 l I l Increase in Magnitude of Flux compared to the initial Situation for 280 mm/yr Pulse (in log,, mmfyr) {
~
l J y .; . 6.0 . 2.50 l t .' / - 2.00 1.50 l 4.0 - i 4
/
1.00 lQ e .
; 0.50 0.00 "E 20 ~
-o 50 N -1.00
/
l 0.0 l l ! -2.0 l 1 -
-4.0 -
-4.0 -2.0 0.0 2.0 4.0 6.0 x [m]
1 i !O i a A___
- - .- _ . - - . . . _ _ . - . - . _ . _ . ~ . _ . - . . . . _ . . _ . - . _ . . - - - - . - .
O I f > I . [ ' - r e.n e
~
,, in seei t vai s ' sumum es e e g Ul3 i" i ElllllllI i'llf ; ; - i N .
==
~
ER. .
.v -
- s. p.s as v.
= s- ,s:
.=., /
o i ElEl'I Ei'i E! i E"= = I IEi U -
! ! ^
M :* t". : 1 gs i : E-5 .R= .n ' '; .i i : w B i "n ;i M ;' d ! 1 W' L. I g.: l
-m i -i ':* .. :r it ' i.=".i== *ll> :
1 O . n.sn . T'T' ^ ' " '~f -- : ! ? M. I h ,!,, ==! ! ! ." ' ' ! E f l . i+1/ 4 2; g !E! ! !! -' E !ElI-'i! ! Q:
- a; ! i: illFs .: :: i gi
-5 $
5E ?-h =i 55 i' -l o
= ==
se - 3 = r. _ ;: :l i s, i.g 1. 1 1 o. _ g .... ..... .. 3... .
,3
;._ p .. .
. =g g= i = .i . 1:, .i .s, -
l =. s. .: < o=> g
, : 3 := r: : : o-5
. ==
it .-;;3 :
- ( j i : 4
- = ;
- fs 8 t
E5 - l:: s~ 1 - ,: :
! w s
c: 11 g; em a.8ri! x, i :: = .. l lE l: l g n ;,Eii?i+; _p 3 l ill t Ei ;i! l
- ................;......,,i E c .;
g> p.,,, . 5, t ax : =: - .s-
.g -- .l : .:
e i = ; !! _ 1;' i D i z; E i O n
= ! !: ~ 4: :ll
=: =.
= = =' =1
; ) .. ::
- .i :
?
l =_ =l . l _ l
!o a :;::: ::.w s u g i ;
i
. : := n: cs ta :*a
== y- 4 .:
j... .. l ,8jj l l ,
. i , ... ..
- l , != ,! . ,! l
. . . . i ....i.. . . i. l 1
3 a a (W) NOl1VA373 l Ol d
i , l l l l I - - - . - s , . . . .
~
j TEMPERATURE at 1000 years ! j ceae.eues.k, wn.4.4%, ape,'e3 % l
- : : . =
! i uguld Fhra bi Maestr .
i i i ! - i
- : : a 1500 ---.-..j...................,..l.........................:....,,,,__,_,,,,,,,,,,_j,,,,,
r i g Ej Ii d;
- 87.68 OE -
_, 82.83 77.96
^
E - 73.13 POSITORY e i - 68.28 2 - I 63.42 I' O - sa 57 - 1000 53.72
. 4a.a7 W 44.02 g .
W 39.17
- 34.32 tabis 29 47 ! 24.61 i
500 _ 19.76 STEP eize=100
- ..............:,q..... *
. j j : .**** -
t j , ,
, ........j......
j , , ,
.... .j , , ,
.l 1
230000 232000 234000 236000 NORTH NEVADA COORDINATES (m)
4 l 1 g . . . . . . . g . . . g l MATRIX UQUID SATURATION et 1000 yeere j i Cese:DueR,ind5=4.4mndyr, APD=43 kWlacre : _ i i i ! i e - '
- : : : SL 1500 - - - - -- + - -- --- ---- - , i ----- - - -- -- I :j - - - - -- 4 -.------ -- -j-----
.-- 0 956 a
- E: :
- E ..
- 0.911 m -
.:.'i S "'M h.... _
O.867
,,,,,,,,,,,,n~w~= ' ' ' ' '
O.822 i w E - REPOSITORY - 0.778 ! Z - 0.733 ! i O , 0.689
;- 1000 - -
0.s44 , 4 0.000
> t t
g . , . 0.555 l _J -
- 0.511 l E _
. 0.466 !
0.422 !
~ ~
0.377 500 - - 0333
, . l . . . . ' l '" . . . .I . . -
. 'l' . . . .- >
230000 232000 234000 236000 NORTH NEVADA COORDINATES (m) O O O
Percolation Flux Studies m Percolation Flux Tests at Repository Horizon e Fault Test [ 3 4 m alance in the g i a trix Property h m La w Tests in g , SE w
^
k) . l ERNEST ORLANDO LAWRENCE BERKELEY N ATIONAL LABORATORY scRxEucy tAs EARTH SCIENCES DIVISION acuw* l
Conclusions u Various new data and analysis suggest an alternative conceptual model that results in percolation flux at the repository horizon of
.~ 5mm/97.% -
w ' a The alterhative conceptual model de-emphasizes the importance of lateral flow in the PTn and the role of faults as drains above the repositdr'y horizon. 7 ' *y - N g- 1
/e :
' .e. !
E Flow!p)Ths below the repository harifon are more' complicated due to perclie potential for lateral flow aliove the ' zeolitic tdi ater occurrences 2*pfErfg*eochemical sign L,gr ,,
$i
~~
{ge inversionggyA
' ~ ~ - ;
E implication gher percolatiqn fid!dj i$clude: (a) enhange"<fd , potential page into drifts,(rid'(b) perhaps limited di ! ! " dry out" ring thermal loading, and therefore incre
- hpmidity inhrifts.
A
<<<<<<< m
\ . _ ., . . L . c- . E - x\ BERKELEY N ATIONAL LA50RATORY j m==remen EARTH SCIENCES DIVISION uw
- Q Q
\ i
r 5 Summary of Problems Considered Problem Recommended Numerical Formulation Steady-state moisture flow ECM Transient moisture flow Dual-continua Transient gas flow ECM Tracer transport Dual-continua Thermal Loading Dual-continua 1
5 Summary of Problems Considered Problem Recommended
~
Numerical Formulation Steady-state moisture flow ECM Transient moisture flow Dual-continua Transient gas flow ECM Tracer transport Dual-continua Thermal Loading Dual-continua
m a a.a s.4awam-m+, .,+a--+a,w A w ----m ---------w ----_..-w- ~ O b i s 1 . N I
$, N y1 s sN s Ns
\ s O
s NN\ s SP gN NT. s N s N r
/N A \ N \
i,
% h=: -
^
N%
)
A s c5 y i
' ^
O
- 4 '2.0
! l 1
- O Parameter Contour Maps Areal Percolation Test ,,,c,,,
i at the Repository Horizon sawadaa
$1} waer/Fr.ctwe M4 I
. . c. N#
i '.- 7% a m ,ix s a w an oa { '. . 4
\
- M-M )
Fractwe Coadap/ Ages
~
' ?
l
.,5 ;
M .., Moisture Tensioa g o'~ . , (@) '*-"* b M ! 1 **"
$1 } m As.
- q i
3 ,, ca. -s "sr e ce2 i eH3 ce3 i C 3, i n m
- h
& A matrix Permeeldtity
!!:, ! ; l ljil! ;l ; < i!!!l : l ll1 d
, ' s s se m ,,, e' i m ' e Gm J7 .e o e smAma s.
e e . e. s. e.eeee* m z. i. t.esem*
- ~
1 - - - ; - - - m y 0 _ 0 0 i 0 0 3
)
4 n
. 0 1
7 1 - i
)
_ z u ( m ( E-0 0 t e i 0 2 a _ W . 7 1 i n n dr o . o i t o c . 0 C e , ' i 0 0 a S . 4 1 7 d s s M 1 1 a Z- '. v o e r '. C . l 0 N n ' 0 ot ma 0 - . y n ua i i 0 sc F 7 1 1
- - : - - ~ : : . . . - :-
0 0 0 0 6 1
# 1
= 0 0
1 0 s gW vC ,>&m r
' i
;! ! ll l!!; ,i l l!!lt l! l! ! !! i Ijii t
g 0 o s 1o. ,,e s s t
,,s.'
e' 8 emr e0
- e. 7. o ' "ome s.e e. m em u. s. e s 0 s ,e o e oa.e2e e e. m "
t. e' _
- - ~ ' .
s r 0 y 0 0 i 0 _
) 0 3 4 '
0 . 7 . n i 1 1
)
z-u ( m ( E-0 0 t e i 0 2 a - W ' 7 1 i n n
. dr o '
o i t o c ' 0 C e 0 -0 0 S ' i 1 a e '. 7 d s l
/ '
1 a s ' Z\ 1 v o' r
'. e C ' ,
0 N n - 0 o ea sc y na r i i 0 0 7 1 _1 f - _ - - ' - _ .;_.'_. 0 0 0 0 0 0 0 0 0 0 0 0 8 8 4 2 0 s 1 1 1 1 1
,W a C9.W,>jg 0
j ;
,- ' Jn - l
~..]' 7. ~ ,.;; y W r 1gy. p t a e O. - :cv.sv ,._;,;v. 1,r., 7.;;3.g.
=y .
.': ; p. W :s,K w ,p.; v:%e u., e g: W.*gj .
,x. . ? .a %. s,:u- + 7,w;_ fg.lg K _.yppG_9
.o : m, ; y, , _cq;; km.4..
- . '. L : .., q , p / , I :; .[ h. $\qh. 3
[ . h;:f l"[* .
.l b.
. :l} _ .; .
b lh.f
,j }}. f ;l. . .} } ),}'}'; GQ . .
., 7.
.i
.f ,
( }, .
, y g y;g gggggg g q q q q[gy
} *((-.. .. .
?
,. ,. g . ..y ;;"un:- g-
>...g ,- t.
+~...:M. - . .
.. v .
s.v . g ..., g, s ;;. ?. ,, Q. q 4:' .::.a
,_=
; os < y. ;.... ..:.3 e . .. ,y.n. . y. 2;-Qs.g.y,m s y f..
. ; , ,en j . .. g.m c . '
e : ,. .~
. ,- y.
,yy: ,...,,.y . n e n.'y.x. p.. w3 ;a.
- e. g -
- 'f ;. ,
- p y. ., ., .
. .: , . . .h
,. .,.,; .,.:}j
'. ,;. l_.
O '., .; b. .,;+~ q y;f; x;f:(j};q._y.p Q.)q . ll; :. _ , ,,k _lf.f
;{ ' L 2
) . .; .. . ~ ' .- * '~
A
. .3 :,, f ; ;4 y a:% .
"^ ~ .
. yy;y.: 4 4;;'s 4
_* [-,. c-
.3 9 _f:- ?Q;*~j [ 7L
} j.T.L Q f_ f Q 's * ' ..; ,.
A,nsigh.ts, fo> r, Y_ q1:\:la.u c c a u @{ V ]@s.
+
l3- . ... ,,. .,l }. . ; . .; ' ..
.,. Q;
. .p. l.. f..y ry;. ;f, v .vQ ..Q:. u .v .Q. v ifn,. fR... yg& & py ( i.;,_. f~ .
.L a
- .? +%
- .~..
>_ : e -
.4p. ; , . -
g ;y.x 4..%u - r ]'
.ue 'y:' .y,e.,j,~ ...
.s ..
.'. . h
_[
~
, f., j; .
- v. . . .. ;
- g. ;
,., g .
,,g 7 qf .
..t p-
- i. ;-
F-u C
.? .. % ,.
.- . ?. ' .y.; "
- i. .L ::t ;' - -'" .,:-n. T..a - y q <!:s $.k i '..a '?
5 - , .
.g i ; ?l3,7$.p%* :.c g? *.%.
N, e
~.
... r.
' ' . l.
,e,
(; ; s#
..,,.,,y.. g . y! : [ ';.
.,c ,.g, s..,'.v.Qq..g
, ~
- c. .
.2 .L y
__g c.
...e....; . .u - . :r.v2:, + ::w :.p. r .. a.. r. ; w. >w>t t..-
.p..
. w;w n
.n.
, .... ; u
..~~) . . n ,_ $. .q:w s
. y,,3,r9,,. 3.4 ., .Wm..yd. ,- y g. .l6, .v .
,a
.k .C ~'"+s.,.m es..
~
f L. . .
?.1.;;J.Q{Qg:.$..Q. y' '
~ - %r.Tq',. .a:y J.. , s . . %F,&lVf.. s . ; . . t q,a ~,4 '. .,.3 yyAW.3
..q>.
, : 7 t .,; - - . .. s n
g.
' ,. .+
q-g .. 1.j.;% :: . s.~ s .,,;- . . . , .yn .
- . c..., .:-. # . . . w. x .. w %f.
g, 4.n... ;.p.g.v,m 3.,p,;c t ., g, .j. .s%; gf . ; g .3 .a.
.* m
- :. , ... ....:;. .n.3; 4.. e. e . w r ~. ' ' ,
; ; g. , t '.
> m ;;: .:w;. g.&p.;
. : . ;..Q;.. 4 1, , . . , y. . g,; i, 5' . . h '.. ,.,;. 7[( i
'. .h ;,. [ h.It.h . [ #
~.h S:
s
- e. . . .
.y Q*.
- u. s, . ~ . a. - ~ . m .w mgg ma.
((f , , g . 6: hO .. , 3: m- y,_,. &. . a'.. Ty h._y.;! e" U, n.e . .. Q: a ';
; 17 ;. a g y .
- f , .;<Q s, e k.:
^t - - -
g _ ja e' +
~. ,:
73s-~x g l
$Y. h l .g.~
' dk , ,'- , . ,' '; $ hIWCY.&. . e f l
ex
.w, c:
., i '1[
J : : g ((J jf-_f:;.h..,g.A9MYiN I
}iff
, ~aw==2MS&L.
~
ff'ih!EW'1[0l'$:Y .?SS$FYl%ik[f$hj:{"%)*%Y
- I: M Ql* %
p(b;gf.g V{::y. . O,;ggy%,;; '. p '- ^
^
m ~ %Q, 1 k-
- k$h@kh$$pf9&q h? ^ ^
thy' . l j,$ydeg5;bt ?-
~
# l o 1
.;:- m0, :a - no yy <
a e
- o
_t. o ,) 0 0.0 1 0 un 3,4,1
\;$s'f& o~ _
l
.1;>-
Ji bl l t s l
?
- f. y r,
, a hk$$_hA["MS$M 3 . n.g_"
4 A,+ ghg g,ye
' 'fu. ,w ydA;%
mcz
&l& MrQy;,/g-a t,
+ ,. .
- u r-s.
O
, h: wwg,y 8 * '
'l 4 m?' % % ;'
. - 0. 0b);L'. N 0 00 0 M ki % d 0 '
hurdk:.o3 - e%- h ' s .:%
.s u, - ?
' ' ?,:
1: , b;_ M9my.g4q; ( 'c;. > > ,
*c P sy k.
Ygh, s s y.?g , <
. x ;-:,7 <
pi;$ gg, , t , myY Jp , p
' k5?b;i ,
"D /
3 .: j .., -- l k $ I _I I5 l I - Ull h ei @ x)@N ,m t2@E o ,e Kh , t ,
. % $,. ' M3 - . ayp@,
1, - - . . _ .
~.
_ - , a
l 86TII ACNW MEETING LAS VEGAS, NEVADA l SEPTEMBER 26,1996 l INSIGIITS FOR YUCCA MOUNTAIN FROM FRACTURE FLOW STUDIES AT TIIE APACIIE LEAP RESEARCH SITE, I SUPERIOR, AZ l R.L. Bassett, Pb,D Professor I Department ofIlydrology and Water Resources { The University of Arizona Tucson, AZ 85721 (520) 621-6215 I bassett@hwr. arizona.edu IO l FRACTURE DOMINATED EPISODIC VALLEY FLOOR RECIIARGE.....M.J. Thomasson; D.M. Stephens I
- MECIIA.NISMS FOR PERCIIED WATER FORMATION IN l FRACTURED TUFF....E.G. Woodhouse RADIOCARBON AND TRITIUM AS INDICATORS OF FRACTURE FLOW.....G.R. Davidson, Ph.D 234g/238U SYSTEMATICS OF INFLUENT, FRACTURE, MATRIX AND PERCIIED WATER IN TUFF..... E.L. Hardin
[O Y
\ .O APACHE LEAP RESEARCH SITE G PURPOSE: Investigate generic hydrologic, geochemical and transport issues in support of the NRC licensing staff.
O FACILITIES: THREE RESEARCH LOCATIONS
- Deep Slant Borehole Site (DSBS)
- - Queen Creek / Haulage Tunnel Site (QCS)
- Covered Borehole Site (CBS)
O PERSONNEL
- Staff: D.L. Thompson
- Students: G.R. Davidson, E.L. Hardin, W.A. Illmann, i D.M. Stephens, M. Thomasson, E.J. Woodhouse l0 - Faculty: R.L. Bassett, S.P. Neuman, P.J. Wierenga O PRODUCTS l
- Publications, Presentations
- Evans Workshop (Biennial)
- Instrumentation
- Databases
- Measurement Methods
- Collaborative Research w/Other Agencies O SPONSORSHIP J
- U.S. Nuclear Regulatory Commission.....
Project Officer: T.J. Nicholson
- Department of Hydrology and Water Resources jQ University of Arizona, Tucson, AZ 85721
g FRACTURE DOMINATED EPISODIC VALLEY FLOOR RECHARGE LICENSING ISSUES ;
- Uncertainty in modeling unsaturated flow through fractured unsaturated rock. !
- Appropriateness of assumptions and simplifications m !
mathematical models of flow through fractured unsaturated I rock. - i
- Confirmation of the basic concepts of flow in fractured unsaturated rock by data collection model comparison.
O
- Investigate the use ofisotopic and geochemical modeling studies to determine the rate and pathway of solute. flux in unsaturated fractured rock.
l
- aL -_m a ___s 2 g
g U) y
'd g N ~ W '
}'l f -
[, y i I
# .j o
ax4,m% Q E yp % c f ( g- . [%, ,\<{ ' 2 y
~x$y y + +;,, Q' i I s
~J - f% 4()( :
8 F $ q $ #, , ~- 3u4 2 , i7 2 i- f _.i i
/
i km %- s p ,,
'g !
}
, , 7_ . .. ,-,'
e, j ,pt yys) 6
'.:;, 5 p.ts, a' k (,,wt'j/l, 7
xs .,5 . e\ '3, ".v.s Wh
- ),
m' g .,'..., j:.., - ~ g
,-.:<~..y ;;-% ,.
l 5
,;~, [' r.g--$, ', _- ; x .: \
,< * ' ;;*) , ,y N,Td$ ;' ,;
eE
. E I
I I I I I e I I
9 O
-* y ,,w..v , # '.
, . m
.1 ig},gg<:.
., ?
*$ .- ~ * ;- f . , *,t~ ?{ 1 . *1 ,
* *,y: M . g . '<
l , ' _ .t ,h.:.f ; ,; j.
' l% .? y 4 w $u a t es s e.s 3 w2 w+rA4 ,
xi
- h. .... . d, ?+
N l5.. ' 1
/- ' .
, ,4,t
, , - %. '.,y.',>ft,L.'y.,i
,,f* w) 5 e
.-.. 6 4 ' ,e
'4)h., .;
gy
- + y J. p y, .
h ....,. v..r 3,A
, .- , - * 'g ' . .' . '
*3 p'l i(p . '; .<-
1
- y. , -
4 v..X$
'.,g . ' * ~ . <
; :c . .
O ?l'O,GhJ,y:h.x?!Q. Y g .. I y . ,- Y~ f s o.
*j. ;N
.y e,. '
s
-[. p [ , ~,% .
s
'a
. . .- *O ,
.~.- . c ., . 1,
,g. ,
1 ..
.p., ?'
) .
t
# g
. x. ..._. .
i {
] INITIAL OBSERVATIONS 9 Seeps have been historically observed in the Magma Copper Company ore haulage tunnel bored into the Apache ; Leap Tuff. The tunnel is more than 2000 m above the regional water table and greater than 120 m below land I surface even at the shallowest point. O Relying on historical data from mining companies, water l companies, personal observations it was clear that a correlation exists between rainfall quantity and distrit4Rion, stream discharge and fracture discharge at some locations in the mine. 9 Some fractures are dry, some seep continuously, others 3 periodically. Detecting changes in flow and chemical composition required automatic monitoring instrumentation in the tunnel. O
- l l
- e MAGMA MLNE Preapitation Data ip,) Conecuan romt N p, ,
V x,s rD e < r .y - r!r
}\' ,
e thlh.f:II "'
? 8-
,1l;
; .\ ' '$,' ,.o
- r CREEK QUEEN j I/(!]Ji'l,;/
4j ; /j '-{sj jf fj't j o
-4, 4.60 4
- j. yft f . f. s.ni.. ,y./ ., .! U/ k
.NN . .
s k ,O Pty , .g f . :, ,
~ '
7-o N p MA
% JS M A; To Superior
" / t m'i an k
si N~ K w w - i
'V
-- M=
~ ~
( f,f,$f.~.'{ '
. .. . . i ....
~ ~ ~ ' I -
1400- .M.6: .. 1 a a #6 Shaft- Perched #9 Shaft - i *):1,- c~- zone
~ 4 'i" -
1300- ~~< i5 -
' ". W
- ; Si? m W "
a **'* Udi
^ 1200- - J'WU ' e' e e' ' e' r- r-
.E t r' r' e r' r' # r' c
.9 1100-
'//r e' e' / /
' / / / / / e' /
W,'/#7 / Gray
/ / e' / '/
Unit - e /['
'Ri / Brownj ,.,.
' e' e' //
$ 1000 '
' e' #* e' * * # # # # # 'j E Never Sweat Tunnel * * " 6P5 i i i # # $ h1-
~ , ~ e' / v'e'r'e' -a e, 900 ~
, a - a a .
.f ", ", ", "a ",
Underlying Sedementary Rocks %w NW ,r, 800 sa 700 C) 2000 2500 0 500 1000 1500
'v Distance from Tunnel Entrance (m)
O O O . TRAONG FRACTURE FLOW - BORON ISOTOPES aussicaex
~ - 1 enorosenneuneawaec
/--
TRAONG FRACTURE FLOW- $D AND HCO3
,/ ,/ -
muae I 1
.! ! / i i e i i ; ,=
!H' 1 , . = - -
E I $ *
- M' i.
,en s
I saw I!.*
,, l nm is er . 8 :
nu isu n : W
> I . .oo 8
I
, ,: j'.g .,j ...
Ja a
-e
! ' '. _ _=
' ' ~ ~ . .n J TRACNG FRACTURE FLOW - SULFUR ISOTOPES i O
- -80 a
?=
, T . .so D= -I II III lill l I 1111 l I I
,, i*m teco isoo 200o 2 00 240o 2e'oo 2soo 3o08
- oi - - m
" , Dry Fes ' _
j .
~
1 ned 1sevel
==
E.1munal Poe.ms . 1 U -t 5 k k e a d$ I
(O 400 . . . . 500 c EC at CP2 3 Days
~**
- Precipitaion Data ' 2 Days
( ^ co 350 ;
~' l~ -
400 2 Days ,11 Days l . [ .e 300' r i o
.t .
300 0 c ! D ', i C [ j 250 - s i
+
[ l I, 200 2o ,
*C g 200 -
100 {
. t 5 '
/ -
[ 150
- I 'l - ' ' ' 'd- " -'
700 720 740 760 780 800 820 840 Days Since January 1st,1991 O 500 , , approaching perched ( Co O 400 First arrival == zone composition h i j ; Flow y [ .= V Flow ' y 300' ] [ ]o %u , ""'" ,y n average perched U zone 0 ~ 3 : C0mposition [ .s 3s 0 : t s 'j ., D f.[ - [' O CP3 [G ,
$e,o '
gg ' O CP2 0 700 720 740 760
\
Days Since January 1st,1991
-% b
?
, l i J_
.O l _ i .- e 0 , 0 4 e 00 - 4
^
A _ k 0 l l i
' 5 ' 0 3 I 5I 3
I _. 5 6 9 9 A h h c r b c I a 0 r 0 _ i 0 a ' M 3 M 0 _
- i 3
4 9 5 _ 9 h h c r r a c r _ a e r a _ M Y a e Y M . f n 0 o Oi f o o n 0 h 5y o '
' 5y t
a 2a it 2a it D a D p t ip " i c ic - e r e b r P ^ P ' i
" 0 0
2
' 0 0
2
^
A i 0 0 h 5 '
' 5 1
I 1 A ' E e 0 0
' 0 '
0 1 1 0 5 0 5 0 5 0 0 0 5 0 5 0 5 0 5 O 4 3 3 2 2 1 1 4 3 3 2 21 1
!$ - EE gh.a3n.
O
z - ii!!i"hk i
;.' 'D
~
^
s- ..
~.
h! j \, \ ..
~
9 a ;- . [ ,
.:. __.. 1 b -J.' . , .I .
D
Conductivity Readings in Queen Creek and N.S. Tunnel 400 375 tip 350 o 1 325 b l 300 -w, ___ _275 -r._v - - - j250
'g 225 15'W ~*-
200 Tunnel 175 1 150 Creek 125 100 75 50 25
- 0
^
, -25 l l l l l l l 01/31 02/02 02/04 02m 02me 02/10 02/12 02/14 Date a .; p lN Electrical Conductivity in Never Sweat Tunnel 324
10:00 322 10:45 320 _
n 't0 2 318 e
@316 n 11:15 l l 'iS 4 314 - - - - -
- I' - f1. ,,
, : ', ,j 312 - ,-, - I- '
< og.co a
ty 310 l l l l l l 06/14 06/15 06/16 06/17 06/18 b 06/19 06/20 06/21 06/22 Date tune of usecten : 1I 00
A.. 9 O , I, - e, ,, ;, le 3 3 -
' 3082ths le Il
, 31 il 9 4 19 "b 3 9 I) ll it 8 IS 3 l'
8 is n i. %-== me=== l t 17 m l' ' 1200 4-t LI <k It wg . I 2 3 6 l ll i p
=4- 3+ 13 -> 3 a +
O , , 4 14 24 4 l E 36 p' 3 13 3 3 l l j, il gi 2 12 3 5, I d4" Il "b 21 I b D M QTCSCDtatiOD Of the three-Column grid for the trRCer uphudit.
%=
1 N m u N l, N N
... . 'N I N u
N
* = - - -
Preliminnry simulation of sodium breakthrough at the tracer site. ~~ l ~j Bre.akthr.ough
.imNa i
~
u / [p /
/
/
/
u J
/
s '
' * * * . a ,
, ,e Sodium breakthrough versus time at the tracer site.
9
DIRCHLET BOUNDARY l Finite Element Grid 11 rows x 14 columns __ _____ _____ 154 elements 4
- 2-D Flow and Transpon Coupled to HGC Unsaturated Matrix Saturated Fracture Zone Experimentally Determined Matrix
_____ Properties Tracer Test Constraints E b>k b>k n d W O - ol & 6 9 el o zs za { .. I I I I I I I I I I I I I I I I I i iiiiiiiI i iiI O aaa aaa aa a"a aaa saagggaagggass
conctusions: O First arrival of solute from intermittent flow in Queen Creek is detected in tunnel fractures within two days [120 m transport distance - fractured tuff] e Hydraulic connection is verified by chemical and isotopic composition, and continually monitored electrical conductivity. e Tunnel fracture discharge is a mixed water originating from Queen Creek valley floor recharge and from perched water draining the escarpment. e Distinction between water sources can be made using O chemical composition and environmental isotopic variability (6 80,SD,6"B,634S). o Other isotopic methods identify the fast transport pathway of valley floor recharge ("C,3H). e Numerous fractures are visible and mappable but the actual number that transmit liquid water is a small subset. It is not clear how to predict a priori which fractures would flow. O
g
- Discharge from fractures connected to the perched zone is continuous and the water composition is essentially uniform.
- Simulations of flow and transport are compatible with observed fracture discharge. Models incorporate experimentally determined rock hydraulic properties, boundary conditions, and constraints imposed by tracer test data.
e Antecedent conditions strongly influence valley floor fast path recharge on the 100 m scale:
- under drought conditions, small scale tracer tests are insufficient to initiate fracture flow; under conditions of recent precipitation, small
$ scale tests yield liquid water connections from valley floor to the tunnel; runoff events rapidly recharge the fracture network and generate fracture discharge in the tunnel.
b l
- - . - - - . _ - = - - - . - - .. - - -_ - - . - - .. -
FUTURE WORK )O
- e Controlled tracer experiments tracing valley floor j recharge on the 100 m scale in the fractured tuff
) section between Queen Creek and the Ore Haulage tunnel under wet antecedent conditions with multiple
- tracers (conservative, isotopic, and reactive).
- Simulate transport with multicomponent reactive transport i model comparing fracture / matrix hydrologic modelling
- strategies.
!
- Drill, core and log two monitoring wells to identify fracture network persistence and orientation.
I lO
. sample fracture surfaces and adjacent matrix in new boreholes and in the ore haulage tunnel (after closure) j under ambient conditions for indicator and environmental j isotopes, and residual pore water chemical compositions, to i compare with samples collected at Yucca Mountain.
i j i i O
[ O MECHANISMS FOR PERCHED WATER FORMATION IN FRACTURED TUFF E. G. Woodhouse [ Ph.D. dissertation, University of Arizona (In prep.) [ LICENSING ISSUES ( e Does a correlation exist between the physical properties of [ the tuff and the location of perched water? Can the identification of these properties support a limited selection of perching mechanisms at ALRS? O ( ( ( { ( ( ( (0 4 /
g Potential M~echanisms for Perching Ki/K2 > 10; K:/L < 25 Ag
%, j_
-- a W'
___V_____._I__________ n L _ _ V _ _ __ _ _ _._ ___ _ _ _ _ _ _ _ ___E________________ B ASAL VITROPHYRE, K = 0
~
BASAL TUFF
- a. Individual flow model b. Cooling zone model h
%-[' J h_
e V __ _ __ 4 _ et Q K
" - 7- -
I h% Ki weathenng*g ZODCS ' fhgggg'i l / acture filling Ki > K2 and Ki > L
- c. Faulting model d. Fracture set model Ref. Woodhouse E. O., PhD dissertation. University of Arizona Departnuts of flydroke and Water Resources, in preparation.
O P **"*i"' "**"""*' ' ' "*'*"'"3 (co nt' d.) AA ^AA
' y w-
/~ '
.7
's \
\
/,N/ s s s (ys x/
' s N /
p % L y\<; /y \}/pff
-$9]7,,
l / 'l 1 I
- e. Fracture fill model f. Fracture density model
.O v A
^'
NEE 55@Eh$gppn5$25EI siE
$terconnected porosity [NCD Sist _,w, dis 3 M M R N M@j$ %MM Mi2l5s
. m _m.....s,,....,
P3fy,$ E$ gps!P; decreasing permeability, increased density
/'
- 57 -
perched water j/ former hich water stand
~
X l zone of alteration isolatedporosity'hility negligible permet p regional water table
- g. Matrix properties model h. Weathering / alteration model Ref Woodhouse, E. O., PhD duncrtatson, University of Aruana Depenmern of flydrology and Water Rescurces, in preparation.
O b - ' f%)igj @h y }:f!sl3P%jr$$E@P h 77 35 m y' /// - I
/
/
/f' '
/ ,J /,/&/ , l/ '
// /
/l R 135 m b \ g o _
.{h3,,/ \% i A t5 _p 200 m 1
\ i z \
$ interconnectedporosity \ dominantly unfilled fracture F/]/' decreasingpermeability,
/pjpwA \ dominantly filled fracture increased density isolated porosity negligible permeability Combination Model:
reduced matrix permeability reduction in fracture density increase in fracture filling O ~*---t- ~ - " - - > ' - - - '"' '* " ' ' '" -
O Wl'5WMtwdb!!DR ?_wwwedil0Premm WY W5%A,h ti3 VuraiMifdiFs Mi
- na ta x e g .gg ggy g y jfg ikse jDSB ,d a;32 z_9 D lj fau t
// / '
Zone y 9, /lf+ g y# ' ' ', df N 135 m __ _/_ _ _. __1 - _ L O __ 200 m
-+_9__-p/ j -f
$ interconnectedporosity \ dominantly unfilled fracture 7 rmeability, g decreasing
/5/
mereased ensity
\ dominantly filled fracture isolated porosity negligible permeability Combination Model including the DSB:
reduced matrix permeability
. fault zone intercepts borehole increase in fracture filling Q w w e .co,no % % % < m w auxue w we a_m,.w %
comparison or analyses of ALRS aquifer test data O _ Test type Model Conditions T ,r (m2/ day) Notes pumping Gringarten- discrete fracture 0.161 poor recovery Ramey (19"3) match pumping Warren-Root dual porosity not available no straight line (1963) formed recovery Theis (1935) confmed/uncon- 0.18 homogeneous, fined, porous isotrop. assumpts. slug Cooper et al. confined; 0.167 less close match (1967) unsteady-state for late times slug Bouwer-Rice unconfmed, 0.057 system may not be (1976) steady-state steady-state slug Karasaki linear / radial 0.17 based on values (1988) flow from other models l O
g RESULTS AND CONCLUSIONS Potential mechanisms for perching have been identified (see accompanying figures, a through h). Physical characteristics of the tuff were compared and correlated. Properties included geophysical, isotopic, geochemical, physical, mineralogical, and alteration characteristics ranging from submicroscopic to macroscopic scales. Results allowed some perching models to be eliminated; others were combined to form representation of the perching system at ALRS (see accompanying Combination Model figures). 9 . Aquifer test analyses indicate that the flow system is fracture dominated (see accompanying table). Fracture-controlled flow is supported by isotopic and geochemical data. Matrix analyses indicate that changes in the properties with depth cause severe reduction in porosity and permeability (see accompanying photographs of polished rock and thin sections, which were vacuum-impregnated with blue epoxy; blue color therefore represents pore space). Modeling efforts of the pump test and slug test data indicate an effective aquifer transmissivity on the order of 0.16 to 0.18 m 2/ day.
The Apache Leap system is more accurately portrayed by a O discrete fracture model than by a dual-porosity model. g Estimated fracture hydraulic conductivity is nine orders of magnitude greater than estimated matrix hydraulic conductivity, based on the Gringarten-Ramey (1973) [ model. I = The Apache Leap perched aquifer is a fracture-dominated system. l I I IO I I I I I O m
9 i M 1 3'. k' .t - 1 + .' - P 9o P O i or 7; f>
' N ,
., s E O c-N
,f f - O
. j
% (\
l
. _ _ _ , _ _ . - - - - - - - -_ _-_ _-_ - - - - - - - - - - - - - - - - - - - - - - - ~ ~ - - - - ' - - ~ - - ' - - - ~ - - " ' - - - -
D 9 .
~
$ ') 0
=E due
=e J .'
l
^
+ ^
1 e 'a-
.p.v r
3 . e A* ' , ' y!** R $ R p ?...{* ?
^
N h <, i ,, g . . _s. ~ g - Y. . 19?,*.';.4 k+ *& C 5 7 .ya( U
^
,. ,1 ..
3_ w- .
~'
d( '.
?),_,$ip~,414 9<n. .,1
. . . c' g
- v
- 4. .~e' ,. '
- t. ~,
'q;_
** .. ' A.
O' Mia 2 ti s., ~
' J,N;
' .; . p ar
'Cg: s a ;.
Ih j
4 O - E o-T
.3 O #
4
*fh
.+ w l
l a e u t n k o O
O ggggy.q. d q ..
.'o.
'*s_ .
3
. wE ~
$ E
'ij An ,.
Y' - a .- v ,- .
~.
m,g . o q q} s}Q ,: 4a.. w %; [ . r e e ( a
+
. M
& '.]
o 4e .: e:
.. e ,
g , . w, ,. (-a .[. . . N" ~ T. - 1 i
g RADIOCARBON AND TRITIUM AS INDICATORS OF FRACTURE FLOW G. R. Davidson, Ph.D LICENSING ISLLES e The uncertainty caused by the lack of experimental confirmation of the basic physical concepts of groundwater flow through unsaturated fractured rock, and uncertainty caused by the lack of established data collection and interpretation techniques required to model groundwater flow through unsaturated fractured rock. O P i
' - _
- e
,,3,,
s.
?n-y
- r. .
n a t ym w
, g u.
; ; ,' w-q q
- q. y- . . .
(qc
- - a
; M ,u -
l
, , , M-( ..M ':
- :. ." 4
<*)'i_ .
s-r ....
- J. ,, W; . : , .
z f..>$q
$1%.;.c lsg$o$N$~l5 ;%, + . dl
- c. .e .,
$ f; ; ,- -
j
.$ s u sq, _ g , . .
.i
, f
'v ./
y: -.3 ,
.x. .
m 2e {
+.. . ,
#9 y " ,r? "
I 4 .
$(h (4 % % ' , '
,f J
.g 4Pt
,, i - [
f' i i ., , g- . s 8
, .i f t
- s. g .
t
<' IcQ) Fr:- ' *l
' Q%
+ s y~ i. '
i ,,k. g ~ ' ' d '. -; . ' q '
%. ,Wy
.$k f i. ,, 9l4'fp'.'
, q; < ,7
's
.. .s ,.4.
e ..T , g y.; j, ,.
- e. qq .
.. r -
.'- L q- n'
; y.'
9% { c..' d ,'.f ,
* 'J-i s ,#c M, r ..g' g,? ' De
, ~ . ,. a, I
y ' Y'_ -
..,- g- ,
, ;l' q
;> s -
I c - . i) ls Q.N _ % '"
.+
)" 4 !#
;, f
=;
, . l
{ 4 ,)
.p 4 4 j <J
- x y, > ; . . m,y f,v.
, 1.
1 e m.tsk.,sf$ n ww jgs }f.g f
- :~ . -e
, ; ' T ),8; -
b D v .ra ,. '. :.g .
, .. :.n.
- \ *;3.%f Q[q!'i' i
( '$ - 2 q a. 8
?. -
-jh}:p@c.r, :9
- y. .j;8$$ .&M 3 I . 6-
~ k, t ,% . [g' ' u ik m.. .g7
- g. . . ;,4 -A ,. - ,
, .' ; 1 6' ., , 'm 2. m i c o a' ijfat,g y
A ' .-. ,
, g' .y 'i ., _
4
m_ _ _ _ _ _ _ _ _ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
,t 'a i
1 7- . t 7 1 4 :- 5 4 4g$ \ ag g .' 3'i 7 i r,
-w.-
- et - %
g g j '
\
i3G *
, Vy ^
3" 5 1 ..]} w h'i I I
~
's [ ,
~
Ig((i
, N{;i]
AE 4'u sw' h aa
*" , ' ey , .c
;. 1 e 1
-?
' gvt.t ,"
.h $l hN['hl5kNI$ $ hik.Nhf b fg
%g%Q~m:wd p- t
, , 'i, t gg -
.n .
...z S,
, j g k g$,1;F.An,,t-
. '.ili.v; w
.9
,. e _ _ . _ . . -
' c {h tgh P. 4.;,
' :s. .x ,
3 f.
, te .u , ~ '
2
. [ f'. ' f M M l3 yg g- ~4
..d[m,.WE n
% ; ci' et - ,
=
m3, . 2 2.; a . s fh.
~'
AN, e 't E
.e ..
i i) ' f, . e , . h:K'@ ' n: $
. "~l
^ '
~W re w:y-j w.. ;.;&' y. m* , ::[f*'..-' .sa.
$l->...fTwn. n h*
r . :;
. ;; sq
! e,
$?]ii$$$Nb ' ~?, }N ? ?Y Y F
gpg i[a E g ,,ys I h- , , w: a f. N& a 2R 'i b . 12tch J m ,, t.1.:..
. h-(#h]>, -.
geaxx .
*a r
. w o
2 . ~ . > ,.. 3
- w -n{ 8 x T.
. k" h iI'x %g,, m y.(( n;.t:. ,c
-~ ~ .
- 9. %
t . 5 1 ; 'Yl ~g - l ' f,' 5 ,'
.l
,. n:. .m .., , . r;m .
? g
, w w. pas ., - q .
iv t t 3
,i:<g; g apyyd .. a _ 5, r%a + -
. 7 e. %7;- :* L' '}.
)
hO 'I-f ;h
.hhy[w ne r
s: ;r s y
..y.t , A ,
;- Q .
g
.4.*- - !.,,,/ ., u i @";h -
,,?$P 2rv.My- '" .I _,. !!
N p pw/ ~m -4 ( :
=
p . k; N f , t ? ~' ' : - --A $ '
\
f 'f k
# . .ff.; $ .
.. 9 5 .g. . '). ; \4
,a+ ,
^ ! k ul:..J :
. L .. l , , ., , l
~
O O O . ohm-m (resistivity) 20 40 60 0 l : l l 0 . , , l , a T
-$ e _ - . .
3 l c 4 c v i I
-$ }; _
"- a g____
^
nMw{ l
> f 1 5 .em - - -- ,
1 am m ii; -
-!5 -bn a -
sa. 80 --
- c. ; .
e . .... o Q ...............'1 C
-100 -- ,
'L....................
1 100 -- 1 8y x 6 . 1
=
.y g c
-120 -- c .2 '.=5 -
g - 120 -- .
~-
., _._ w;. , , twme-a
-140 - bn ~ "
3 -140 -- . F3 perched aquifer
-160 l . l :
-160 ;. ' '
O I 2 3 ' 0 10 20 30 40 50 60 70 80 near/far ratio (neutron) & g/cm' (density) Water per Mass ofBulk Rock (mVKg) I . _ .. -s -
0 I lf ' f/ ll
/
f
-O-Formation Air O Comp. Pore-Water
+ Surface & Aquifer DOC
-e-Distilled Pore-Water o Surface & Aquifer DIC
- ,l -
~ l l n - [ ,
/ 0 , , , .0 . O-E O b -
\~
l 8' -20 ll 4L I I' 1
-100 -- i
? 1
-120 --
40 - g
-140 --
l I W'*if"il. l *)
'T}ll ;\
'~
~
-160 -60 -
m 0 20 40 60 80 100 120 140 $ 7 __ Horizontal Distance (m) E -80 - a. m o
-100 -
i ! i i i i' -120 -
<- .m.
\ } -140 -
_ _ . .'i - Sl g _ I - .. _ .'. _ .. _ _...._.. _ ..._ +G perched aquifer
-160 jncn.,,zon, j j 0.6 0.7 0.8 0.9 1.0 1.1 1.2 i
;- Fractures m! '#
. C (fmc) k Regional Flow j
O G Table 2 Carbon-14 and 'H results from the saturated zone. Numbers in parentheses indicate the volume of water pumped before sampling (saturated zone), and the depth core was taken from (pore water). Sample "C (fme) i 12e *H (TU) l 12e urn W ater 0.665 0.013 0.00 0.09 (1.900 L) umW ater 0671 (50,900 L) 0.011 0.00 0.09 Pumped Water
- 0.664 0.013 nd" nd
~t51,700 L)
'* ~
0.809 0.012 0.00 0.08 ($4 7 ) 0.813 0.012 0.14 0.12 150.0 ) 0.820 0.020 nd nd 155A m) .
- Above the current water table
" not determined I
I Table 3 Geochemical and isotopic data collected from surface runoff, from the saturated zone at the DSB, from seeping fractures in the east end of the I mine haulage tunnel, and from the saturated zone at UZP-4. The SO//Cr values are mass ratios.
. Surface "" *#
DSB UZP-4* Runoff Tunner "C (fme) 1.180 0.67 0.69 0.68
'H (TU) 6 <0.1 1.2 1.9 5*'S ( ) -4 to + 1 6-7 2-4 7 60// Cr 6 - 10 0.4 - 0.5 2.3 - 2.5 0.4 - 0.8
- data from Bassett et al. (1994)
O I (Davidson et al.,1996)
o g RESULTS AND CONCLUSIONS 9 Drilled and cored the Deep Slant Borehole (DSB) 45 from horizontal, and drilled the upgradient Vertically Oriented Borehole (VOB) in the instrumented watershed; each located a perched water zone at ~150 m depth. O Core has been characterized by measuring properties such as: grain density, water content, porosity, hydraulic conductivity, pore water chemical & isotopic composition, geophysical properties, mineralogy, petrography. O Conducted numerous borehole geophysical surveys to determine rock properties and fracture locations. O 9 Numerous visible fractures, some observed periodically transmitting water into the borehole. O Radiocarbon profile indicates reversals in fracture zones. O Post-bomb "C was detected in water from a fractured interval near the saturated zone (perched interval) indicating relatively rapid transport. O Chemical and isotopic composition of the perched zone as compared with matrix water composition supports fracture dominated recharge. 9 Developed new sampling methods for "C in soil, formation q v gas and core materials.
l0 2xum8e SYSTEMATICS OF INFLUENT, FRACTURE, MATRIX l AND PERCIIED WATER IN TUFF I E.L. Hardin l (Ph.D. dissertation, University of Arizona) l l LICENSING ISSUES e The uncertainty caused by the lack of experimental [ confirmation of the basic physical concepts of groundwater l flow through unsaturated fractured rock, and uncertailty caused by the lack of established data collection and lO interpretation techniques required to model groundwater flow through unsaturated fractured rock. e Characterizing the chemistry of the groundwater in l partially-saturated hydrologic zone of Yucca Mountain, Nevada. I o Development of a conceptual groundwater flow model that l is representative of the Yucca Mountain groundwater flow system. O )
O Apache Leap Research Site 234 U/238 UAR Welding 1234567 0 Surface Partial Pore 20
@ Waters:
gp Squeezed
%a #g# 40 - & MQ Leached ap Ng 64 v 60
.' .,.c::
- . . n , . r . ' , , ,. .. .. s .;, t ,;
, ,E h
,j ' h '. . ,
YG& % .y
.y v;;uw.uv
~
- '::- S# ;/f,3. . .'
- g. .p,- 0,C.f'.l: ...v:i,[ 100 3 "/ b. ,;:
n.::::y;;
;;e ? Q Ei ;'.G; y; < . e.1
_ : .my( - y MNiJ
;, m .+ .,,;
' .:1 .
120 q '+I-iG g g,y x..2.q~ i:!;5.lNr,,. - ,v. g k
.J ,
./.:..
t.3,' :.s.,....g 1 h g.wan: m m'. 140 nc.. _r~ ,x . .
~
vw -
. Perched 0 4/8 Free waters
+ 4/8 Squeezed O 4/8 MQ Leached O 4/8 Hydroxylamine N
Leached (Hardin,1996) 1
O EM39 Susceptibility Logs - ALRS 0 s-Qir 20
- DSB
/- Casing 40
{f f z 1 y> 60 es ,
- 2 ---?
^
2[ 80 # F G d = O ea 6
$ 100 S
g
>sVOB DSB i '
120 { 6 4 -
"c t
. ?-
f } 140 M'f ' . Perthed Water f - - Table { 0 0.5 1 4 O (Hardin,1996) Susceptibility (10
- cgs /g) s
] J /
Mixing Plot of Surface, Fracture, Vadose Zone and Perched Waters, ALRS Perclied: l D Oak Flat 3 6 ALRS O 4/8 Free waters "e c X 4/8 squeezed pore waters M * '"""*' 5 + 4/8 MQ leached h 7 4 - - g Below AR transition o 3 d Surface & 00 - >6cn Near-surface , hractuhes 1 O Stfeam - - 0 0.05 0.1 0.15 Reciprocal U Concentration (1/ ppt) (Hardin,1996) _ - . - = .. .-
Sampling Locations for 234U/238U Investigation Apache Leap Research Site Ephemeral stream ~2m boreholes, fracture zones U7~20-50 ppt, AR~1.4 Ur~100 ppt, AR~2 A I I
" Squeezed" pore waters U7~100-400 ppt, AR~2 " White" tuff
" Gray" tuff l
MQ flushed pore waters AR~ 4-5 (gray tuff) Deep slant borehole V / ! Perched water table 1 U7= 220 ppt, AR = 6.2 Mine tunnel MQ flushed pore water AR~5 U7= 672 ppt, AR = 5.3 1 (Hardin,1996)
g.- 1 O l Deep Slant Borehole - ALRS I Ksat (mm/yr) I 0.1 1 10 100 g 0 - -
, 7
*gM W I 20
\
l l / \ 40 - fo ' I .so 60 / g 9 v - a.g - 80 o g,,,(mmfy,) Q + GeometricMean Ksat f l 100 I 120 I - l 140 - I O (Hardin,1996) I ;
O Deep Slant Borehole - ALRS Ksat (mm/yr) O.11 10 100 0
* /;~{
20 *
/
@i> o 40 -
O 60 /
~
o p ~ c / j O Lat (mmlyr) A f o Geometric Mean hat Calc. Lat ' l " 120 log 10(Lat)= 140- -1.26+3.93 exp(-0.0332*z) { L J ..I ...l . ..I (Hardin,1996)
e n RESULTS AND COlyCLUSIONS U 9 Fracture-matrix interactions leaves durable 234 U isotopic evidence in the matrix.
> Mechanisms for 234 U fractionation which are likely in Yucca Mountain tuff pore waters:
- Selective leaching in weathering zone near surface
- Auto-oxidative selective leaching by pore water in tuff matrix that contains soluble minerals with trace 238U4+.
> Uranium series systematics of fracture lining minerais reveal the 234 U/238U AR of fracture waters from O which the minerals precipitated.
> Comparison with matrix pore water activity ratios will reveal the extent of fracture-matrix mass transfer, i.e.
where interaction is limited the ratios will diverge. 9 Effects of fracture-matrix interactions have been observed at Apache Leap
> The effects of matrix imbibition and matrix diffusion on site performance are represented by complex models coupling hydraulic and geochemical processes, but properties and compositions of the Apache Leap tuff are quite amenable to model calibration.
O
g Magnetic susceptibility logs show weathering along fractures.
> Matrix hydraulic conductivity shows a decrease from 10 to 1 mm/yr in region ofinterest.
O Retardation of U transport by secondary minerals is strongly rate limited, and effectively limited to adsorption at exposed surfaces. O .g
,: ' , ' f! l ; ; ;! ill i ; l{ l ym m_
t o - r o s p t c s e n m, a t e
- yf T r
W s a y r o
- r t l
a d a a r c n l e c 6 b o a
- i a u9 9 yL m
me e s N1 n6 o2 l el a s n s oi at a o h a e e r e l GNer Cl ee t b mo t i i m a r d R me i l r e 1
- l mt o
p l i i v
- e n CS e WLec i n y
_. Fo r o e r w r ts i s a L v
._ e ac d a A 1,
+
~
Np 1 k h . b i i o d
. h
^ h f g
4 .~j !
$ i ,el
I ! P I! ;: i 1lll l il Il: l _ o s e _ . . l a i m _ i t g _ n e r e u f _.n o l f s _ i f i c e -- o t s n n i r l i o e o t u i t p t c i r a r O c g n a s n o i h n ~. e i l i o t a c o , s ge n l i l b a t as e nd c t a n e ._ __d i h ce oi s .m b o nB yigABC v i l Si o E m h e s e Pe r - - - C sa a _i a e s t o Pnn
- e _$
h" h" n
. ,= 8 a; ln- -l
O O O Repository heating can result in a single-phase above-boiling zone, a two-phase boiling zone, and a condensate zone EE O ' ' ' ' ' ' i i ' '
' Cwl I l i
'TCwl PTn - PTn -
100 - - y TSw1 T@ TSw1
- W B 200 -
]3 A
~
A - B " --- TSw2 TSw2 DN {300 f n;-;; ;-y
~ ~
j TSw3 TSw3 E O
~.
o g CHnv _ CHnv _ CHnz A CHnz 500 - B g . pp ,I
. 1 i l i I i l i . l . ! , i O 500 1000 1509 2000 2500 30004 500 1000 1500 2000 2500 3000 Radial distance from repository centerline (m) t = 100 yr t = 1000 yr E811n twBGe'e43
-. :=___x-_:_-__=___==___--_. __ _ _ _ - _ _ _- _i
l i l;; ; ' ; ; l l' .l:!1!I i i;! ; 11 ! _O - e r u n-t a o r n - ed t cs pn et c
~. ma l
f e t emu rf e ef e s i . h r cg m - gib il i t n i i sl i h usir i _ g t qin te bo _O e aef a c R __ no o mad o r n _ g m i xd aa h n _. lum mlc o
, t i
_ n ' i l oxu v ei r at c a i ' _ i r o e pib mp l l i o _ B , _d i mi u e a oqh v e . m pCeCe a v R o '- bm , u $'~ O ".kw hla tN !t' l
~
O O O , Equilibrium and Kinetics in the Near-Field LL I. Events at high flow rates
- Damk6hler number: rate of reaction relative to the rate
~
t of advective flow D1= [pij Sj Kj (aH+)n L]/[c eq V]
- For the case where DI > 1000, solve for L i - The result maps the regions where equilibrium is less likely to be achieved (i.e., where distances are less than a few millimeter per year) i
Fluid Equilibrium Conditions O For Fracture Flow Around Repository 25 years after 50 years after waste emplacement waste emplacement 25 years 50 years 23.2 11.6 , 0.0
-11.6
-23.2 0.1 9.05 18.2m 0.1 9.05 18.2m Colors indicate travel distance required for water / rock equilibrium.
Darker colors indicate: O for 25 yrs. maximum distances in excess of 10,000 m for 50 yrs. maximum distances of about 1,000 m
M O . l l 1 I 0.80 4 l 1 0.70 CO2 & O2 Not Fixed O O o - , . .q.e u o = l f 0.60 11 O l d O O IkO 0.50 = CO2 & O2 Fixed 0.40 0.00 10.00 20.00 30.00 40.00 50.00 60.00 Moles of Water Evaporated
O l I 9.50 g iP
'# ~
O CO2 & O2 Fixed gi E lE D
" J G 8.00 k
rm 0o 7A0 h 8 e o* cm 6.50 CO2 & O2 Not Fixed 6.00 < 02 10.00 20.00 30.00 40.00 50.00 60.00 Moles of Water Evaporated i
1 . l l O 1 l l l 1 l 1000 m l l e O toom - rt 1 % l 5 C D n E elP- - O o
- =
a t som _ .tr [ T, o- 1 I I 1m ' ' ' ' om tom 20m .wm som som som Moles of Water Evaporated l l l l l i l l l l O . 1 , l 1 l
- ! ! l\; i! llll \l! !!
o s y d s a n - s s s e a k e w r h m e ctclnt o r fe ua t i g a yfe k,p e w n b dn cw oo R m
~
e ei o rf l n _._.ew ct a yl nt a i
- o. t eibi p t t e uidn c eer a
s b l f net y r ae o n a i y pnf e sr e mr l i e _ p t r - d . r o p n _ n e i t mt de o e a ym h wrt u _ i C _ x e cd sl iop o
/ l p r e _
mvl g el l e mk t i eh v n 1 o c at sh ge CroWdChd i i i l i _ m .
$pm%
o _ [e ;,
;i
l! t ji! il' ii !:l,! I
):!! l I
- ~
~
O i n o i n o t t a a s s n e g U n n " e
- d i d n l i
l n o i o o o C B B C
-" i - ? -
-^ - MLWM -
?.
1 I
>A i
- w l
i j O I i \ s M *- j.M
- w. _
M# e - - _ 'T- - i m s j g e
- r. )
- 0 s y - - - -
0 0 r e r ^ b 6 t t e s l mm i o r (r f et m \ e ce n n e h a c c t y sr o i Dt o e i s o G 0 p e r 0 0 0 0 0 0 0 0 0 0 0 0 0 0 . O 2 4 6 8 0 2 4 .. 1 1 1 7li eb 8{E E3E2 Ee 5o$ i
T i O GIMRT sim6dg2: vol%(t) at 0.25 m (inlet)
, , , . . , . . . i i
.fo _. ....... -
porosity - cristoballte
- : x-feldspar -
w... . - -
.:. albite --
quartz C anorthite -
- --c xaolinite
.- 50 - ~ - ~ ~
-o- pyrophyllite
# ^
gibbsite g . 7 diaspore - g a_. . . . - O s
+ -
E . i 30 7 -'
~'~
2> -
- = &#;-
.\T . I \ ~
10 -- .
\ _
\
i 0' ~
- .- . -2 i 22 --^^ l I
1 0 1 2 3 4 log (time (yr)) O
\
O GIMRT sim6dg2: vol%(t) at 5 m (center)
,2
, , . . . i . . . i ,
- porosity ,
- cristoballte k-feldspar f 40 - -
- --- - albite ~
7 quartz ;
~
anorthite C i
- --o- kaolinite ,
d '
+ pyrophyllite g# ^
gibbsite I i _ l 7 diaspore O 1l i i E _ _ _ __ y 3- .._.. ...... ...,.....-..... o . .
+ %"
10- ,- --- t-i
. l ,
0 E , , O i E ~., -" ,
-1 0 1 2 3 4 log (time (yr))
O
E
- O T
i i GIMRT sim6dg2: pH(t) f ' ' ' 8.2 L 8.0-
~
- inlet (0.25 m)
- : center (5 m) ~
- outlet (10 m) 7.8--- r
+
7.6 -v <
=-
7.4 - Q - 7.2- . 7.0-6.8-. - - 6.6 i i i i i
-6 -4 -2 0 2 4 log (time (yr))
O
;llll i;:
ll ;1ltI :!I i! i iIf
~
O y b _ d _ l l e f oo r s _ t no n l t a _ oi s t e ca _.eml d sic e p cif idt / __ s o t i r r s ev ceir an t O e r i l m a r ute/ s t a at a ar e e _hr i g e ce mn i l m r e R l cmen a pdt i n S R _ i mt e ca m ?- n e t a p e '- o l eh K h gmC C _C h i e id
- a l
+,_
_; . h~ O
.?
W&m lk c
;i l
hsa wa Amusau m mes- e m--_e m Amsquu u m m.a.m-
-En
-em.
I.w-4 ' t s { ' l
..s .
',.t 4
4 ..% . h
. ?N l *
' (!f.
?:
-.? , :
3 ..
~
. .. ... . : 9 - -
- e. -
.- , y
. 6 6 " '
n h, ,
)[
l 5 .h J.' m1
- y
- 9 . .
' Q }g . .
~
)4Hk , o .
--+ '
. , , , w g . . .;i . -
9
- , g! D
.s.
r
- e . - T ' ~
M
. 1-r.
y- u
. ., f . .
C d c a ,
- .j -
f) e..... e, . .' Y g p , f . .
,e s
O l l E ' 3 t ] 9 y - 4 ')
, w .
1 y y .. . 7 - J :I'., '. ' ..,'a:,
.m , , . . , ,
p ,. ; : . -
\.
.,'.-::..':, . g h2 - . < .i 3 n
". . . .x . '
\. ' . _ _ , , _ . _ _
. .. . . -- - u. --
y e" a i o
1 9 l l 4 ef 1 1 1 i 1 l i ( e st ger - , t j k;.;. w a ..n.
. ,* 'f
,(*
.,.,y;l "
e .. s ( a
' , p{ 1 2., u.i< s .
*n l , .f *
&; ,' y
$f ' ,
&W w .9,.. &
d
. . . i 'P a , . $,
, pIM m
n
,a .Ynce,
':f I[ /,,,*f(k , , Y ' 4
, , s 'b e.
3 l -wy: ~ . . _y , " #p/ . 7gy%gg. i d.". .' ,' e 3y: $hf[ W J' 8 y" h. rl' *pa S
<t
< 4e e w y7:7 n y y"ha '
w .' n* ,'s 'p 64 * [,[m . %r ~'-( , yf ',
.-' i [- *,4 '
s / (., t .- 'S s i ;lt.id h a &,%L, ,' M M..n, < w e ,q.e, m."
<a w n m l:~["
M au ' h.G. n"M v.&,. , , 'v ,[+.J.d :rk.O & t h y sw j , s . .o y; s e ~.
%.g , m e r y-.m -
- y. 7 g 1yf - ) , f ., ,
S .g. 4 ".s,' i i y e r
'w mm f.Si . *t
).$. e
;r -
1 I '
- e. .4,Q ' ? d w,. Ii M5 '
l 5 l v.::O, ,.# ([.I sn .c Jiy @
- y. c .y ? " . Q,.
,l,' 3 ,"
. t
.", % . S ' ~c .i , (J.. . ,2a % - @, s ,. .,.. l }'g y / (3.,
...,:,[ , + ,
,e - man 'E.~ . . - ' ' b [.ia, a ne f
$t 5,. p , I
[f b t { .' J' , hy-.< ( y + l1 r , s p e-V.
'?,g i, '
Y p y;.
d'
- 1
~ m \.
.; c m;7"%
3 f _(q
,4..
, t A ;. i ,'> 6 5r- 5
' - flf
' , .LJ
. g'f 1. w, 'm"t' . ~ ;,g. . 3 ..
- s I I' k [ :M'4 r y r,, , p ,g:r%:
44 g).,,g
+
' -' pw,t ..%, @[%, , i
.$.g.*>+' 61 e M;;l , ., n s!b O j% q M"
, , . = , e .g NJ A , A , m .4 ';(# , f p g;L W6 '
c ir
+- ,,w"
, , , ms
.mp Wp o swqo, E,. ,.
- p;., N e w n-<u~ .ta;
- 4. . wy 2'0Jm.~ M, m c , .o 2 wN > c x
> W+
ii e
~ .f.k M s
),.p N 6,1.- nr ' *yw g [~'J' g*cp l- ,- Jt / sJ ip - yM 5
%. Af's '>;.3 di f
Q f}'.,(.W. 'M u q >.; m:'. ,;
,.r,,Q v . , . . r^ a >: p -
n' i#D , i' j > ][ , (/ ' ', f .4 - f: .[ ;g i' ( , h Sf , p ys Y'*
...+
s kg
- s. on M ,. s j[ j%[. 6 2.s., .
iy ,
.x
, [y[,[ _;p ;u -. s - :cg e
- f. n';: 7h,y .<@s! Y (n.tM% 'Y ,
a e l 5y 1
.',n s **T.+.j '. + ,u s >
~ '?
if ,, , . , /d
, 4
-h,, <
xc, n/ %y. M! M ,'d 'N;;$jd '~ . g%> 8
+
l
; g. [.
'3
%.s
..y.
, ;:. -. e, 1
, h,
- , ..:{ G7m +y.nn
. - m p h % b. e
, - s . .
WBC' w
..g
., j.,' i
. ~ .
i ,( *
,,g<.2'm' a.'M .y '. '9. f.Dl i
%:{ {l *
, &. -y, ,'i q *
>*)e,Y Y '. ~j 'r '
4 'h ? j '.
. , . - 3., i 4 , , +x s < ,ie r t A % 5 [(,:a.y[ *, )
4 OI $k ' hP. IO M p' # [ ' ' 'A : 3 r \ it 91 gm. . ; ;q,. ,s
> q:,
w., y+... .
,M a
j - n q/ mpvs
<;. , gn ny:m c y +,
\+
g w>-
- c. ~u <
;i s, t
;' ; .'.-"^3.p. f, A M( tfnf 3 . .g , b - 4 x - * *
- m. ; m s y
C+ag op',QQ />y . ,, ~ u g o ' g.A~ b
'p 5 "y q. ;+
e u
, , ,.. %, . . - Q.,( ' Q# s; .'3,, r
.*],, > .' j ;
- i. .
- e. .
n 4 i l; py sK,; 9 AA r g AV p
., r
,y. 1
.s ;,'v.?i V v'+
'*'{- -%'
(j '-
' 7%8. n d'- I "
Wv
+
- s .g pV' , s
~r ,
.- p :3. -'. a , $
- u. .q ak -n <
.. ' W e f :. *
,** [ , , ll? a
- j,f.
f a r fW : ;\ > , as : .,{s% %. ~. ,1 % :
-. ,A v, ,,.'
n ,, '.Yl . y8 4, J
...f, $ G, ,h n l'- r (I ' ('
c , , y o M,,j . 'g. s,.'" -< s < r - ; uf ' * - p Rj? , e y"fp, g 4 fr. 14a q
~. i
I
.s
' a
(_ f O >' c- . Y:. < m w S,.
- x. .
4 +. E,..A o d
,,r
- ., g h 4 k ,-
- e
-p I ,,i'(;
i i P a mi ess -weev'r w w wwe wv* - '-v- 4 me-w- m-------r --a---rw---r-e-w -- - - - - - ' ' - - - " - " ' * - ' ' ' ' - - - ' " - - ' " ' ' ' - " " " ' " - " " ' * " " ' " ' " ' ' ' ' " ' ' ' ' ' ' ' ~ ' ' " " ' ' - ' - " ' ' - " - ' ' " " " - ' ' ' ' ~" ~
!! 1 :; l ; i ; !ll ! ! l O
l n t a o e n i t c nmo e t a r t wnad ne ed e e br e n t _ n st a p e I i e _ nwd r e i od y O n t c eln g i a a r o n t eiti o r s n v t nd t set o n e. i v. w ea e yo C s u r l c r r a w e t- a n r k t e i l t ntso e ct c i a m. x smfe o ef a W 4
+
McEre a yQ !
,g" O '
w e n" i i,![9 i
4 i O pH of Cement-Conditioned Water, After Passing Through 12 cm hon Container,2 C r l 2.0 4.0 6.0 8.0
;l:
l : l l l 3.75 -- ' -
- 3.75 l
- 10.0 ..
l 2.50 -- -
- 2.50 9.8 ~
-9.
~
1.25 - -
,r_
;: 10.0 1.25
-6 C 7.C 0.00 - -
6.8 0.00 g .
-6. ~^
6.0 . g . ..
-1.25 -- -
1.25 2.50 -- -
-2.50 3.75 -- -
3.75
-5.00 -- -
5.00 l : : l l l . : l l 2.0 4.0 6.0 8.0 Iron Rate (mol/m^2/sec) x 10^( 10) O
' I
' , > ' !! t f'l! ! I i I
~ .
O r s e n - t i o a t a w m r s c o e mkro s f s s n ha S y a r p - R b t e y r - E d s a e ad n - . o s w. a m f i d i h po d c m. r - t n u a a o n e a fs m. O m O ? m n s s t r o t ua e y ot pn e r i. m a s o u z - w n n m p m. or a p h - r. s r t ait o n t 2 l pca e - a r
=
wr aic ie l v T oteSD Fi n l - a 4g,
,1M O w ' - ne -
l! l ai
O GIMRT sim6dg2: vol%(t) at 10 m (outlet) 50 - ' ' ' ' ' ' ' ' ' ' ' '
- porosity
- cristoballte -
40 - :- : k-feldspar r albite
- I
-+ quartz -
C anorthite -
--o-- kaolinite -
c 30 --+ pyrophyllite - -- O a v
-+- 9ibesite .
0 . . W . E -;- - ; -- -- -- ; - 22^ o 20 -.- O . .
. . . . . . . . . -i
+ w - - -
10- - 4 0, E ,
, E , 5- & E---.... r
-1 0 1 2 3 4 log (time (yr))
O
.! ! l ll ltllli O bg y yns n bo do i
- el a dg t al teer na a r _ it n io me om m,s - o n d i n& o t n do sig n e c e t or _ am i t cde e _ r e e aim n (c O t r r edu - S _ a _ i t sp h et a t inno v kt cpw i l N _. w w )? oecl o O oyr xf - i S l f r r wr eh a - l e e g _ l t at eh t k, i l t ach C H WawpWrofo a bia a 0s a O lmr Ei k ll'
1i !l il l\l!! ! !!! !!f s e f m no ot i _ g i t c e n e e a r r ms _ k n s e a e w t t epe nl og a _ r w-re i el _ o t kvb c em et W b a s w ode r
- - s e d s r
yf siSaBl u . a onlu t w siftof Era _ u ht l ec o ge F ada po r sni e eim l i t n fo wmt dx o ni n oey r s u lf e tag i r e msadd c l psk a nl6 ei u n o u n cf i f F o o a er c e CCpRfo s Wj; m Q - dg g!' '
O o o p Radionuclide Transport through Fractures I. Triay, D. Ware, and B. Robinson 1
O O O
~
One of the main concerns of the Yucca Mountain Project regarding the ability of the natural barrier to retard radionuclides is the existence of fast paths (from the , repository to the water table) In these transport paths, water flows through fractures and the coatings on the fractures prevent radionuclides from diffusing into the matrix If diffusion were to occur, despite the fracture coatings, this would be the most ; important retardation mechanism since it applies to strongly sorbing ' radionuclides, poorly sorbing radionuclides (such as Np), and non-sorbing radionuclides (such as Tc and Se) Experimental Studies used natural rock fractures lined with stellerite, magnetite, hollandite, and romanechite from Yucca Mountain , fractured tuff column experiments diffusion cells using fractured tuff
. _ . . _ _ _ _ _ . . ._.. - . . .. _ _ ..- . _ _ - _ _ ~._. . _ _ _ _ _ . - _ _.. _. _.-___ _ _ O O O
~
Experimental Procedure for Radionuclide Transport through Fractured Tuff Columns t
- i Fractured cores were saturated and a flow of groundwater was initiated Various radionuclides were injected (H-3, Tc-95m, Np-237)
Fracture elutions were collected as a function of time and analyzed for the percentage of radionuclide recovered ' I Experimental Procedure for Radionuclide Transport through Diffusion Cells j Radionuclides are in contact with a coated fracture in one of the cell's chambers l i The appearance of the radionuclides (that have diffused through the fracture I and the matrix) is monitored in the other cell chamber (which is in contact with the tuff matrix itself)
Fracture Column Assembly O Top view
/
NottoScale
%' j ~
/,
i , r, a ' r,
', a
%% g D-spacing,s g FlowFitt ing yTrough 4" _-- End Cap O
\ M a t'o'/ pp %,s's'a'a'e's'o','%;;;, e'opp a'p'pp,p,
' 'a p',,py,p' TR t 4 # # # # # # # # 4 4 4 9'///////// /,w //////,' s
?/////////'/,W //////,'
?//,W ////,//////////,' < SilasticCoatin9 9 / # # # # # 4 4 $ # y p p p p p p 4 fg W :% : %X w ,'4 X 144FC*tdrM :
w, ,w% W,%%%% W%W%W%%%%%,%%v, ,%v%%w, , EpoxyFilling
'gW%v, 4 % % %% % % % g
# # # # # d 4 4 %9%,%,% %*%,% # # e/d g 4% # g g ,
% % % % % % % %% % % % % % % % % % g % ,
% % % % % % % % % %%% 4 % % 4 ,
ta w ,w ,,. ,%:::::::::%::::::::%:::::::::: a ,,,p a ,,,, w , 4 LeuciteTubing warsag% , %a, t ! , e# # # # # # 4 - y , 3,! ,,,,,4 3,3y, 9 # # # / ## //4 4 # # # , f P /,/// g % %% % % g %% % % % g g g g g % g g i "aW%%%va%%%W/, W%%%w/,a.a%.,w%%w/,, w, ,,,
, , , aw, , .
W ,g %, t,a a s yw%,s a s % assa, v,s,% , ,W ,y, % %,,%,,w , OC d wwww'tvaww%w,%v,
% % % % %%g%
W%%
% %% % %' s %'% ,
y een a o a a a%va saf 'Ja,%, ,%, ,%, Yw, =.r age _=__ig a= NottoScale aa
-- E -
o DonutSpacer
/n v O Fractured Columns i Tinif-~ s'UE-35---'- 6S\V C~4-- IGM C-T- IGfW G-4 -
- DE 2T0ZFf'6 TDEr2T02NfG
' UZ#16 919 82980 82981 -
'2954 '595 81336 8 IhTa'trix ---'--" lkaIIFel5pd iA -
- EIEdiTEll ir-IKIEdilEldspir-IKIEaIilEldspai IXfkilIFeIdi[Gr--I I alTFeTdipir" k 8 Mi iand 8 ii}c _n_er_a_ls_ _ _ _ y,.a.r_tz__ _ y_a_nd O_p-_ _ _ _8 a.n.d _ -_ ___8 an 8 an 8 Iand O_ gal _i _ _be_detemune e' To_d a_Qu_a_rt_z_._E. be temune i . To_d_
racture i nte i fioTlandite is lio
. Tlan ite isTollandite i Coating i Magnetite i Romanechite i Romanechite i Romanechite
,-Minerals i l i iI (Oxide minerals) 8 i (Oxide minerals) ii
- i_ _ --- t_ --
e TisTf"T_~~-~~~~;ilfeviIri_T,iFJ--
;i tie _iiTrifief-- i --
DeTiTrified i Dev i&TfieI-- 3_DivTtiffreJ--- i 1-i DevTtHffeil----_, i
, M a_ter ype _ _ _ _ _,'h,yntlietic ---
, , SyntEc_lic.---- r, SyntTietic J-lI ,r fyntTwtic J-II r, fy- nthet'ic J-13--- ,r Eynihetic J-13-- i - - - -
i
- i UE-25_p#1
- UE-25 #1 : i i F 7 9.1 al i e_gB.t_E_(
ililameter _ _ _.i i 6.1 cm_ _) i- '_5_2_ 1 ______t _ _ _ _5 i C0 _ _ _'n_[9 ______L f_ _- _p~ ~ -- '_-iE.1f_ _-----~ i C3 _ _ _ _ _ _i 6.1
' - ~'Tf_
~ . ~ . ' _--
_______a_ ___s ' i}(cm_) t Tow _liin_te_ _ _ a _i~_0.5 i ~0.5 i ~0.5 L
~0.5 u _l.0
~
+i~_1.0
_a i 8(m i'Flo_l/_h w 8 r_)_. ________y _ _ _ s 3_.2 8 i 0.9 s iOJ r s _+I.3________y 8 i i 0.4 i 5.7 i i i Velocity i i 8 8 ' ' ' l (cm/hr) _ j _ _ _ _ _ _ _ _ _ j_ _ _ _ _ _ _ _ _ _ l_ _ _ _ _ _ _ _ _ _ [ _ _ _ _ _ _ _ _ _ [ _ _ _ _ _ _ _ _ _ _1_ _ _ _ _ _ _ _ _ _ j sil W i 97 i108 82 i 96 90 i 93 e i i i I I I ,"Recov_ert___a_________a ,_________8g _ _ _ _ _ _ _ _ _ ,_ _ _ _ _ _ _ _ _ _ ,'_ _ _ _ _ _ _ _ _ _ _ t _ _ _ _ _ _ _ _ _ _ _, Tc % i 91 8 96 92 i 95 i 91 i 91 i lR i 'TS'97 ecovert___ jill _______ji9 If _ _ _ _ _ _ _ _ yIS_ _ _ _ _ _ _ _ sEl _1 -________j i i E9 e l_R_ecovert___j ' j ,
,'_4.6_ ________j 1.4 ____j iR 2
,_1_"_N p i 1 .3 ______a_________---_-_-----
, 30.8 _ _'_7T1 If9 i i i
_ _ . . - - - - - - ^- _ _ . . . _ _ _ 8 g i 8 s 9 k 0 8 8 8 ~~k
@ o o
=
- E
$. o i o -
8] gw O. 3 8 " I St o
- 3 R A
o y 13 e o A o 15 A 2 e o 3 o s
--g' Z 0 l:l 0
,-J m
o 9 o o
$l' o h.
o ll' 8 g u
' o ~~s o
l 0 o i o X My o; W ^ ; X%-X-X-X4X-X>il>N b O ! sss! ! ! 5 5 5 d uolpeJd ptJeA00*H
O O O Natural Fracture USW G4-2981 1.00 333x Tc 95tn 0.90 0.80
- 3)33)3ppnnDD>m)D)DDDDDDDDD2DDDDDDDD22DDDDDD23)DDDDo)DDD)X H-3 0.70 y ,,[ '
x .' . ' c xt j 0.60 x' S _N,, 2 0.50 f( !o X Y W 0.40 x " X 0.30 -
- x 0.20 - .
0.10 -W eX+:-:-:-:-:9:-x-:4:-:+:-:-:-:-x+:. ,g7 0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 Volume Eluted, ml
O O O
~
Simulation of Fractured Tuff Column i
- Model -
- one-dimensional axial dispersion in the fracture
- molecular diffusion occurs into an infinite medium (the rock matrix)
- linear reversible sorption is assumed, with different retardation factors for the fracture surface and the rock matrix :
- Parameters
- fracture volume = 1 ml - fracture aperture = 0.32 mm , - dispersivity = 0.3 cm - matrix porosity = 0.1 Good fits were obtained for '>5"Tc and 3H elutions using a difference in the diffusion coefficient of a factor of five
. The 237Np elution can be fitted assuming Matrix sorption only Fracture and matrix sorption . i I
(
) .
i Batch Results for Np Sorption in J-13 Well Water l Major Minerals in' Kd (ml/g) l Solid Phase Composition l l Solid Phase l t __ ___ ___________,._ ___________ _ _ _ _ _,l _(d e termined b XRD)______________Y______________.l, I t 3 l Stellerite ,_ _ _ _ _ _ _ _ _ _ _ _ _ _l _ _ _ ~0 _ ,._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _,l N / A I , ______________________________, l i l Hollandite , _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _7_x,_10
; 2
_ _ _ _ _ _ _ _100% _____ Hollan _ _ _ _,l_ _ _ _ _ _ _ _ _ _ _ _ l_d i t e I I I 2 I l Romanechite l 6 x 10 N/A ,__________________,__________________l______________________________,l I I I l Magnetite i l 7 l 85% Magnetite,12% Hematite, l !__________ _______ _[__ _______________!_and 3 % q_oethite,_______________j
O O O
~
Schematic X-Section of Diffusion Cell Assembly Reservoir Access < 1/4 X 28 tapped hole 1/2" Top of Sample llole dia. 21/2" deep, remainder 1/16"(typ) hole depth I" genemlly. 2 la,, Cured Plexiglas Epoxy 0-ring groove N ' O-ring ! f saneas Pan A
.J ' ,
. g;a ve ,-
t r,r n Reservoir Access
,, -- _0 1/4 x 28 tapped f '
' ,~ sane as Pan A
, , q., , , Plexiglas
\ q ,
#,;n.-{>_, ;~~.' /
- 21/2" ti ,i
~~D g 1 1
,y__- ,s'l2 I/2",8 s
v ._3 l,l '3..
% Part A ,7 , g_ ,_ 4 (77 g
1 3/8" Q D Oft 3 All comer holes I/4" clearance, all 3 pieces
% X Pari B II' -
must fit together k- __' 63/8" 13/8" Part C y
aA - E m a. 4 4, g a. ,, A __L O O O
~
Figure of Coated Fractured Tuff in Diffusion Cell
O O O Figure of Rock Matrix in Diffusion Cell Tuff I
O O O Diffusion Cell with Fractured Tuff UE-25b1h 2025 in J-13 1.6E+05 1.4 E+05 + 1.2E+ 05 + U
. 1.0 E + 05 1
6 8.0E+04 + Tc-95m Fit e 6 .0E+04 - S - + 4.0E+04 - , 2.0E+04 - -
+
0.0E+00 0 0 O f t : t4***: *: : : : 0.0E+00 5.0E+05 1.0E+06 1.5E+06 2.0E+06 2.5E+06 3.0E+06 3.5E+06 Time, s _______________.__m_ _ . _ _ _ _ - _ . . _ . . _ _ _ _ _ - _ _ _ _ _ _ _ _ . _ _ - _ _ _ . _ . _ _
0 O O
.i Diffusion Cells Diffusion Cell Tuff USW G4-2954 I USW G4-2954 UE-25b1h 2025 Matrix Quanz Quartz Quartz Minerals Feldspar Feldspar Feldspar Mica Mica Smectite IIematite Hematite Fracture Coating Manganese Oxides Non-Coated Stellerite "
Tuff Type Devitrified Devitrified Devitrified Volume of Untraced Chamber, ml 85 85 85 Volume of Traced Chamber, ml 800 800 800 Sample Diameter, cm 5 5 5 Sample Width, cm 1.5 1.5 1.5 Water Type J-13 J-13 J-13 H Fraction Diffused at 0.02 0.03 0.007 t-3 E 6 secs '5"Tc Fraction Diffused at 0.01 0.02 0.002 t - 3 E 6 secs 237Np Fraction Diffused at 0.004 0.01 0.002 t - 3 E 6 secs 3H Fitted Parameters D = 3 x 10 6cm2 /s D = 4 x 10 6cm2 /s D = 4 x 10 7cm2 /s '5"Tc Fitted Parameters D = 1 x 10 6cm2 /s D = 1 x 10 6cm2 /s D = 2 x 10 7cm2 /s 2 'Np Fitted Parameters D = 1 x 10 6cm2 /s D = 1 x 10 6cm2 /s D = 2 x 10-7cm2 /s Kd = 3 ml/g Kd = 0 Kd = 0
O O O Results of Radionuclide Transport Experiments through Fractured Tuff Diffusion from the fracture into the matrix can take place even at relatively fast flow rates Np can be significantly retarded, even during a fracture-flow scenario Neptunium retardation in fractures could be due to both diffusion into the matrix and sorption onto the minerals lining the fracture walls Contrary to previous assumptions about the role of fractures in radionuclide retardation, preliminary results from these experiment indicate that fracture flow ' does not necessarily result in a fast pathway for actinide migration through fractures There is no evidence that fracture coatings (in the diffusion cells studied) prevent i the diffusion of the radionuclides from the fracture into the matnx j i L i
.n
- . _ _. . . - - - - . - - - . - - - - - - - - - ~ ~ - - - - - -
- ~ - - - - - - - ~ ~ - - - - - ~ ~ - - - ~ ~ ~ ' ' ~ ~ ~ -
- !EPRI i I
- i 1
- O :
i
!, f i !
i l l , I l i i l ! l
- bw
% , - ,j , Colloid-Aided Transport nm ,
in Fractures 4 i l 1 i s l ! 1 Reasonable Approaches for
,i Yucca Mountain !
lQ j! Presented to { l Advisory Committee on Nuclear Waste i l Las Vegas, Nevada l l l September 26,1996 i i J
- ! i I
i
; John Kessler
! l
- i
, i ! { f i !O 4 $
l s A. l i O Goal
- i Provide NRC with reasonable assurance regarding l
colloid-aided transport l Show that colloid-aided transport has been dealt with: l
- conservatively (as necessary)
- defensibly ACNW 9/2646 (t) EPRI O
Quick Colloid Property Review Size Range: about 10 nm to 10 pm Surface Physicochemistry Very Important Most colloids and solids in water have a net charge
- If oppositely charged, they attract and colloids stick (coagulate)
- If the same charge, colloids may be " stable"(stay suspended)
- Stability dependent on many factors: colloid / solid composition, pH, ionic strength, many more...
acuwm m Era O
.s ,
O Some Important R.adionuclides are Associated with Colloids
)
+ "True" colloids (direct radionuclide precipitation)
- Several imponant actinides: Pu, U, Th, Np
" Pseudo" colloids (radionuclide sorption onto natural colloids)
- Natural colloids: secondary minerals (e.g., clays);
organic matter, metal oxyhydroxides
- Many nuclides sorb onto colloids: Th, Am, Np, Cs, Co, )
Eu, U, more...
] -- a mu O
Special Colloid Ethavior Requires Different Treatmen:. (than for solutes)
. Can't assume usual solubility limits
. Can't assume colloids " sorb" and " desorb" like solutes Colloid transport often very different from solute transport
+ Relatively large size affects their ability to:
- enter small pores
- stay suspended
- sample all flow lines a
surface forces can either
- quickly remove them from suspension
- keep them suspended and away from contacting other solids m ,a.= ic r.m O 1 l
1 I
O Do We Have to Worry About Colloids at Yucca Mountain? YES
- Can't assume:
) - radionuclides on colloids are solubility limited
- colloids are always immobile
- colloids are retarded like solutes
- Most conservative starting assumptions:
- High solubilities for colloid-forming radionuclides
- Migration in the fastest, pathways
- No retardation 1
ACNW OfJeMe (5) EPRI O Back to Reality: Most Colloids are Probably Immobile
- Arguments are made based on colloid and solute chemistry under saturated conditions a
Can this be demonstrated with reasonable assurance? Much lab and some field work suggests this
- Fairly robust arguments for the range of pH and ionic strengths at YM
+ Still a few loose ends:
- Perturbed chemistry due to EBS, thermal output (local effect only?)
- Special colloids due to degradation of container, waste form, other EBS materials
= = . ,
rm O
O The Unsaturated Zone (UZ) Has a Third Phase: Air Evidence that colloids prefer the air / water interface (e.g., Wan and Wilson,1992) ,
- Almost all hydrophobic colloids attached to bubbles ( air / water interface)
I
- Most " hydrophilic" colloids attached there as well
- If the bubbles move, they can pick up more colloids attached to the pore walls l
- Attachment on the bubbles seemed strong
=
Conclusion:
the air / water interface has to move to make the colloids move with it 1 ACWW 6%As m E71u AirAVater Interface Movement in the i Yucca Mountain UZ? I l l Film flow in larger aperture fractures (colloids
" surf")
Bubbles forced through macropore throats, I fracture constrictions i ACNW Wh90 @ EMG O
1 - n v 1 i Conditions Needed for Significant i Film Flow Transport ! Long, continuous air / water inteiface l 4
- (e.g.,large through-going fracture)
Significant film flow velocity j - Dependent on film thickness (thicker = faster)
- Film thickness dependent on flow rate and matrix properties ACNW M646 <es EPRI
- O
- Are Fractures Likely to be Continuous?
- Evidence suggests not
, More continuous if the fractures are drier
- But drier fractures don't have much film flow But, perhaps continuous film flow caimot be ruled out f
ACNW W2eMe (le EPRI O
% kew -
1 13CONTINuouj
\ ll J \
k W / A!K/WATetz
/yr6gpAc,e O: l
' I '
A j/
,\
- L v
/ l e\
ev . >
/
eN Qy 7 3 ;(g, g5 / (Paricues cue <r 6 kdT6/L NTIHlAola AlR/WA rtK INtcKFKf 9, m
?
ggy ' WALL-CN TRtT
\
l AllL ,
\
- Ath ,
l
~ !
o N l
' t WAT6k flL Pls Meu q V
3/ \
O l l Conditions Needed to Move Air Bubbles Through Constrictions
- Large driving force
- Force is function of fracture aperture or pore size
- Perhaps enough force during " episodic" flow events?
Therefore, moving bubbles may not be easily ruled out ACNW 9/2M6 (II) EPf(I O Colloid Transport in the UZ: Conclusion
- Although unlikely, it cannot be ruled out entirely Where does that leave us?
- Answer: in the Saturated Zone (SZ)
ACNW 9/2M6 (12) EPRI O
c- , O Colloid Behavior in the Saturated Zone Better Understood Air / water interface complications removed
. Large literature on colloid filtration theory 1 Lab / field work to date suggests YM colloids will coagulate j
Conclusion:
It might be easier to assume the colloids pass through the UZ
- Don't have to concern ourselves with the extra UZ complications that way l
ACNWW2Wh H3 EPR! O 1 SZ Scenario #1 (for "True" Colloids): Assumptions:
- Colloid-forming actinides are not solubility limited - All colloids are transported rapidly through the UZ - All colloids are stopped at the UZ/SZ interface - Actinide colloids dissolve in the SZ (solubility limited)
- Advantages / disadvantages of this scenario:
- Can be easily modeled with the current suite of TSPA models - Conservative in the UZ - May be non conservative to assume no transpon in the SZ e, - o., Em O
l O l SZ Scenario #2 (for " Pseudo" Colloids): 1 Assumptions:
- All colloids are transported through the UZ in large, fast flowing fractures 1
- All colloids are stopped at the UZ/SZ interface
, - Instantaneous, reversible sorption / desorption Advantages / disadvantages:
- Same as Scenario #1 (readily modeled, conservative in !
UZ, non conservative in SZ) )
- Need to know two new things: colloid concentrations, I colloid sorption characteristics
_ _ m, m, ! n (.) l i SZ Scenario #3: 1 Colloid Migration Allowed l
- Assumptions:
- No colloid movement in the porous matrix
- But migration in the larger, water-filled fractures possible
- Standard filtration theory mechanisms still invoked
,
- Advantages / disadvantages:
- Some lab work and considerable modeling of this '
scer.ario has been done
- Field experiments are possible
- Many more parameters to bound (shouldn't be i
impossible to do) l ACNW F2646 06) EMd (%
O
\
~
'N l
'1o u < Velocity ZA -
Profile i
> A
>X y
j g! gVelocity
/ Profile d 7 Region i 3
~
Colloid Z- --
----------Z=6
+ Region 11 Z=0 s
Figure 5-3. Schematic representation of convection (Region I) and attachment (Region II) regions that influence determination of avera colloid-front velocity (adapted from van der Lee et al.,1993) O
I - l
, . l
[ l 1 l 1 l
.ici.::
- ..;.. u
.::. :- u. : . .. .. . ..
.o e
.?:.;;i?:.;. :
u
- g. e :
.. .: .. m<
oo
,m
.: m
; e ..
m e u
.: . .. g b*.*.* m .. :.'. . '
- f.. .
~
u ;
.. : 'q ;
e . ::.: u. :.. .; I e::. ::...
- .o_ :.:. .
E
.... . . . c i- ~
i E . ;:n.. . .:..
..:ciW.:r....
- .t:
- j. \
? }. $ ,*. i.:?'
j o Q :...; u
)
ee e . .
.~
u g
..... u
.. .:........ . o L
; .. . . y. : ..:9 .: . :*-
eo
. s i , L .
< L a
... e ,
u
.. . . e..e ;
- f. &ff '.:..,,**,**..
2r. .... . ;..: tr) U
- s
..to I
O SZ Scenario #4: Colloid Deposition / Erosion Assumption: SZ flow high enough to shear off previously deposited colloids
- Causes channeling Doesn't seem viable for natural flow conditions at YM '
- SZ flow velocity not high enough Could occur due to strong seismic events (very short term effect) I ACNW W26m6 (17) EPRI O
What We Must Understand About Colloids at Yucca Mountain-1 Source Term:
"True" colloid generation rate (optional)
+ Or assume congruent release with waste form
- " Pseudo" colloid concentrations and Kd's
- Colloidstability
- Colloid size distribution (optional)
+ If colloids are stable ACNW 9/k96118) EPRI O
O l l What We Must Understand About Colloids at Yucca Mountain (continued): Unsaturated Zone (optional):
- Establish presence of continuous film flows
- Film flow rate distribution (optional) )
- Could assume rapid transfer to SZ
- Establish whether conditions to move bubbles exist l
- If yes, how far and how often?
ACNW 9f&96 (Ivi EPRI O What We Must Understand About Colloids at Yucca Mountain (continued): Saturated Zone:
- True" colloid dissolution rate (optional)
- Or assume instantly at solubility limit
- Colloid attachment / detachment kinetics
- Solute desorption kinetics from " pseudo" colloids (optional)
- Or assume instantaneous equilibrium ACNW 9fh4 (20s EPRI O
l
1 0' 1
Conclusion:
We May Need to Know Only Three Things
" Pseudo" colloid concentrations and Kd's Colloid stability (perhaps only in the SZ)
Colloid attachment / detachment kinetics in the SZ Knowledge about colloid behavior in the UZ might be unnecessary ACNW 9/J6N6 (21) EPRI O Colloid-Aided Transport in Fractures Reasonable Approaches for Yucca Mountain John Kessler, Electric Power Research Institute Presented to the Advisory Committee on Nuclear Waste,26 september 1996 ACNW W2MI6 (22) EPitj O
a_,g- a __ s m,.4i-. 4Jm.2 m. d.d4 .a.4 _ a :m s 44aa-ma. -.-**m.-- E. - --*--m. --- e
~
a p ' Ha 4 6 O i I i , 1 i j 1 i 1
)
l i l 1 O I 4 l l O
o o o "
=
Coupled Flow and Transport Modeling For Yucca Mountain Bruce A. Robinson Earth and Environmental Sciences Division Los Alamos National Laboratory (505) 667-1910, robinson @lanl. gov Earth and EnvironmentalSciences Los Alamos
O O O - Topics of Discussion s Near-field radionuclide release calculations a Base-case comparison for key radionuclides a Impact of fractures and matrix hydrologic properties on radionuclide migration a Impact of repository waste heat on radionuclide migration m Reactive transport modeling for 237Np Earth and EnvironmentalSciences Los Alamos
O O O -
~
Near-Field Release Rate, Effect ofInfiltration Rate 1e-08 _ Infiltration Rate = 3 mm/y O E 1e I I _
.E 0.3 e -
E - S
=
E - 8 5 y 1e E b e lii g - 1e-11 . . . . . . . . , .
.......i .
1 10 100 1000 10000 100000 Time Since Canister Failure, years uz__viewgraphs-57 Geoanalysis Group, Los Alamos
O O O - Stratigraphy and Zeolites Ach %
* ~
~~
Nominal Zeolites Potential Repository
% /
~ r_ Y Maximal Zeolites ut_viewgraphs-56 Geoanalysis Group, Los Alamos
O O O ' l 237Np K g Distribution on Zeolitic Tuffs 20 e - n o Kd, cc/g
. .s : minw ty ---
u;Hipraimp.wmmwwwseew. o us 15 L. -Q 'A t QQ
.o e 10
%4-niMi:i!$Wli!!!bll74 h;S his skiihiidHidHi!!f9fiWfr ili f!!Hii !!fhiliG&i cC . v i d .
E g Ed i O
) i!I ';'Niiii'TN! Ih! [g '!hh!f .fhfb!hb!!bi!!hlNdNb!ffhN!!h!!dd!!!d!Nlk!!h j .%
.5 0 1 2 3 4 5 6 to to to to to to to 1 2 3 4 5 6 7 Ko on Zeolitic Tuff f
Earth and EnvironmentalSciences Los Alamos i
O O O - Two-Dimensional Grid - Cross Section at Antler Ridge Fun grid gSMFnWGWr5fr]Emr-m gjgy,in4L u n y;' W f n U W T 'f f i f w itr + w y T i ' F M f f
,wm r
Lis m,a , 3, n,y y zy7?jt+m' y w a r. w g (s E q q g 2 2 5p u nuan
# % 2 2 T- i et y !:
w ~3~.ggy, , ; w.~ . .., Yf-*117fffg47fgg_j 1999
'eu;mm$ P ..
v?aQc.s-Qg+ - RHfHi,h.- g ~ ,..
.a4r a%;pww& #;n et g. g 3 4+v+p tth 4' 1
~~s_ _ sw . . ,
Expanded Views at the Potential Repository
/#%Q ') i
\
\ /
y,vMQil ?i$ud'i%* ' i
, t/<';Q<p N h
{' f s
~ -
&m .
_b b ? k I'k N b i'$ h 52$ $ $ $ , VWA&M4i%h$fA 6
'WYW WVT-- /
A h h9g iv utviewgraphs-52 Geoanalysis Group, Los Alamos
O O O
*I 237 Np Flux at the Water Table,1 mm/y ECM 5.0e-12
- - - - 10~3 moles /l
~
4 S <n 10 moles /l .
.-.... - - - - - - - 5 ,,,,3fg o 4.0e .. ..
E _e .
.o . .- . ;
a : .,
>- i .
u : . E 3.0e '
/ .
3: ,' / '.,-
@ y ,'
- 5
~
i i l '.. ca . i E 2.Ce '
/
c
- 1 v
4 l l
- a. . , ;
Z l / Es i
/
N /
\
o 1.0e ~ , y - ' i
! I tc : i
/ ,. o 0.0e+00 ~~.' i . , .
i 0 200000 400000 600000 800000 1000000 Time Since Release of Radionuclide, years i utviewgraphs-17 Geoanalysis Group, Los Alamos :
O O O - Normalized Peak Doses Relative to 237Np Drinking Water Exposure Assumed Neptunium 1 Technetium 0.04 Uranium 0.003 - 0.09 Selenium 0.006 Earth and EnvironmentalSciences _ Los Aiamos
,---w-, ---------,_,w--- -
-==h-wmm- wau e-%aw 4m haa_s.-e4 ...ma,.--g- mM4 d as.
6 d I O v O Z 'l /w a Z O m m J o
*=
o. Q 6 O E C y C L L .% C o V 6 6 to w L L ~~m
~ A T _b
*. lll"
==
=
0
" e M c M. .j c n
Q b A N V
. = '
#J Y.. l C o
.E I W %
=
L i M E ,- l C -t
==
+
. .3. ' n.' k
' * {'_,
= : s.le.
i G
~ ,Ih.h f .'
ef
' llt". :.*
*W
>q i 1 Sp.
~,.. . :;&.-
4 y p {.r ..) &e, b-- g
.. 3T. e _:-
, . : e, 4 N M.
i 5 i s k o O
O O O ~, Transport Model Options a Finite element reactive transport solution
- Multiple, reactive species
- Valid for systems with large dispersivities a Particle Tracking Module :
- Valid for advection-dominated systems (low i dispersivity)
- Transfer function approach allows matrix diffusion model to be employed in dual permeability -
Earth and EnvironmentalSciences Los Alamos l l \ _ . _ _ _ . _ _ _ _ _ _ _ _ _ . . _ _ _ . _ _ _ . _ . . _ . . . . . _ . _ _ _ _ _ _ _ . _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ . _ _ _ _
O O O - 237 Np Flux at the Water Table for Different Fracture Transport Models 3.0e-11 4 mm/y, particle tracking y - 1 mmly, particle tracking g _
- - - 1 mm/y 3 ....... 4 mm/y E
ai Ii s 2.0e Ni 3: / . a> : . 5 ~ i .
% \ i3 '.
E !i '. y : i '., h I .
< 1.0e i i c.
4 i z ' . y j ,' ! . o i T -
'.___R~~
, : 9 , .
g . qs -. - -~ - D' ;'" '
, ,,A -
j' -
.....""'~
0.0e+00 . i . i 0 200000 400000 600000 800000 1000000 Time Since Release of Radionuclide, years ut_viewgraphs-31 Geoanalysis Group, Los Alamos t - - - _ _ _ _
O O O - 237 - Np Flux at the Water Table Due to Fracture Transport g,,, , Effect ofInfiltration Rate Scenario ij :. 1 mm/y a ~ !! /: i'i 4 m m/y m i: i l. ! :. -------- variable infiltration
$ j! l :,/ i E :i $ : > :.
g 3.0e ! : ,, T:i m i:::::
. :f
.i .
.! n,s.
u : .
- 'e
.a . .
a : .
!!ll : i ! '.
e i i : '. 5 2.0e : a : a : .
.2: -
- : I- .
- u. . .
m . . . '
< : : ; 9 --
O. i \ .- ' \ . 2 : : : ' ' S; 1.0e : : ' - a j :,
. s
. . t ,
o : .: et m -
- A cc . o
- i *o-e.e.e-seee**-**** * **~
2 .*'. 0.0e+00 .- .- , . . . . 0 50000 100000 150000 200000 Time Since Release of Radionuclide, years i uz_viewgraphs-34 Geoanalysis Group, Los Alamos x _ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - _ _ _ - - _ _ - _ - - - - - - _ _ _ - _ _ _ - - _ _ - - - -
-un.w_w. --wwaw.-- - ,-- - -m--- - - -ana-se ese.aes m hh
'-'a. -.' - - s-w -e-'-wm.e-m e- maw-- -
6 ma meen e rm.w mu.e h.m.m m.m m m. J m maam, , e
- 1 A
1 0 , l Z O l wc: I i s N e Z n i m V em w O Cu. 858 i
~
O 1 W N \ b. U m b O - h M C G '
=
M ~ man h A v EM Q M ,.
.m v
U w A A w 5.h. E L1 O r O dn \ E .n
.=., t w v' am.
Om w
. .Z.
==
E I
.O.
w 6
.W
- 6 6 M FT M Z U U m e T + -
m m C g
.I C
~
~ w e
1 w x I M }, 2 e N 3 a
- O a
J
O O O - 237 Np Flux at the Water Table,1 mm/y Effect of Potential Repository IIcat 5.0e-12 Isothermal
------- Heat Effects included
'm y 4.0e ~
ai
.O .
A lii lii 3.0e 3: e Q . lii E 2.0e 12
~
z S ce g 1.0e 2 E - y 0.0e+00 . 0 200000 400000 600000 800000 1000000 Time Since Release of Radionuclide, years uuiewgran ns.58 Geoanalysis Group, Los Alamos
O O O Neptunium Speciation, J-13 Water 1e-01 ; 1e : 1e-03 , 7 ! "2003 HCO3
.b. Mf ~
1e NpO + 2
^
Ea y 1e e - y NpO2(OH) o ie { NpO2(OH) 1e-07 = NpO2 I 3I 1e-08 - i - i - i - i - i - i 3 4 5 6 7 8 9 pH High Sorption Low Sorption oo.ie ,gs.u Geoanalysis Group, Los Alamos
O O O - 237 Np Flux at the Water Table,1 mm/y, Reactive Transport Model
~
Effect of Fluid Chemistry
- 3. M i
\
i J13, pH 7 j i
\ -------- J13, pH 8.5 .
- - - UE-25 P#1, pH 8.5 o t E I s
I 2 \ F 2.0e-11 - I t g S ' \ N I y ' 1 e \ 5 - I \ lii
~
I \ n I \ E I \ I
< 1.De \
O. I \ z , . k en I 3 t ...- ....___ y I ...- i 0.0e+00 . i - i . 0 200000 400000 600000 800000 1000000 Time Since Release of Radionuclide, years uz_viewgraphs-59 Geoanalysis Group, Los Alamos
O O O Synthesis of Neptunium Speciation Calculations 1.0 0.8 - UE-25 P#1 J13
' O.6 -
O
- lt c -
i e a it 0.4 - 0.2 - 0.0 . i . 3 4 5 6 7 8 9 10 pH i i uz_viewgraphs4 Geoanalysis Group, Los Alamos I
~
O O O - E Conclusions i a Near-field model results imply that a co astant release rate source term is appropriate m Using a peak-dose criterion, the ranking of ' radionuclides is: Np > Tc > U > Se s Hydrologic properties in the units below the potential repository control the migration patterns of radionuclides a Fracture transport results in earlier arrival at the water table and may control peak dose Earth and EnvironmentalSciences Los Alamos
3
- ~
O O O - l Conclusions (continued) m Repository waste heat has little impact on long-term breakthrough at the water table Note: changes in hydrologic and transport properties are not yet considered m Reactive transport model for 237Np confirms the results of simple Ka-based models 1 i i Earth and EnvironmentalSciences LOS Alamos i
E W E E .
" J U'.S. DEPARTMENT OF ENERGY "
OFFICE OF CIVILIAN RADIOACTIVE WASTE MANAGEMENT ADVISORY COMMITTEE ON NUCLEAR WASTE
'1
SUBJECT:
TSPA Insights on Significance of Alternative Conceptual Models PRESENTER.: Dr. Abe VanLuik PRESENTER'S TITpE Technical Synthesis Team Leader AND ORGANIZATigN: U.S. Department of Energy Yucca Mountain Site Characterization Project Office Las Vegas, Nevada TELEPHONE NUMEPER: (702) 794-1424 September 26,1996 l f "! I i -
O O O . Outline Performance Assessment Work Since TSPA 1995
- Model Apstraction Sensitivity Analyses
- TSPA Sedsitivity Analyses a
Current Performance Assessment Work
- TSPA implications of alternative unsaturated zone flow and transport conceptual models
- Planning for TSPA-VA VANLUIK/ACNW%.2
o o o . Performance AssessmentWork Since TSPA 1995 o Model Abstrpction Sensitivity Analyses
- Objective:. Evaluate alternative models of the sensitive componertts to total system performance o TSPA Analysies Conducted Post TSPA-1995
- Objective: Evaluate alternative measures of total system performance and compare performance of barriers and components for alternative conceptual models o
Results of Thls Work Have Been Documented (several j deliverable documents are in DOE review) i i VANLUIK/ACNW%.3 l
O . O O . Example Model Abstraction Sensitivity Analyses ; o UZ Flow Models <
- spatially variable infiltration and hydrologic properties
- alternative conceptual models of fracture - matrix coupling o '
Thermal Hyqrology Models ,
- analyses at drift and site scale in 2- and 3-D o
Waste Package Degradation Model
- alternative corrosion allowance and corrosion resistant materials aehavior models-i I
VANLUIK/ACNW%.4 : L
o o o .: Example Model Abstraction Sensitivity Analyses (continued) o Waste Form Alteration
- models of cladding degradation o Radionuclidq Mobilization and EBS Transport
- alternative models of in-drift hydrology o Unsaturated and Saturated Zone Transport
- alternative models of percolation flux distribution in UZ
- alternative advective flux distributions in the SZ o Disruptive F atures, Events and Processes 9
- seismic /tqctonic activity scenarios
- potential (scoping) criticality effects VANLUIK/ACNW96.5
^
- O O O ..
TSPA Analyses Conducted Since TSPA-1995 o Evaluating Alternative Performance Measures
- integrated' release (compared to time varying inventory)
- peak dose to critical group at various locations j o
Uncertainties investigated
- UZ flow and transport models
- Drift-scale flow models :
- Waste package degradation models
- Waste form alteration models i
- - SZ flow models I
r VANLUIK/ACNW%.6
O O O Current Performance Assessment Work
. l o Sensitivity Apalyses are Being Conducted to evaluate impact of:
. percolation flux versus seepage flux
- advective flux in contact with waste form
- waste package degradation model
- cladding degradation
- form of radionuclide release (i.e., gaseous vs aqueous)
- location of critical group VANLUIK/ACNW%.7
o o o . Sensitivity Analyses to Evaluate Alternative Conceptual Models of UZ Flow
=
This Work ig In Progress; A Preliminary Example is Available at This Time ( TSPA-1995 was Modified to Make a Reasonably Conservative Case and an Optimistic Case
- representative columns from the USGS/LBNL 3-D flow model wqre used with spatially variable infiltration average" percolation flux at depth was increased to 7 mm/yr.
- dual permeability model was used to define fracture -
matrix fhyx and velocity distributions VANLUIK/ACNW%.8 i
O l erti al Mass Flux at Repository Horizon X,
\:
i
. xy. '
s rea N D 4 n oM
/ , ,
g" O g
/ b 4 1 w p - 4 A g, 2 s s $ o
/ -
eg as
- f:
/
/ -
'o
*n%
o
/ fD 50 e :
M (p
! (D i ao
-<s
%m (D
- I CO ll ,
E ff
-ro ;;
I YO D
!4 '-
m ll ) l
@ 3 I
8 7<. (D f 9
-me.n,_,,-,-.--.wr----we ,v a-,,..,--.,v,-,_--a,-_,,---,,,_,a_e_aw- -----a,----,---- - - - - - - , - - - - -n------,---a-ww- - - - - - - - - - , , - - , - - - - - - , - , , -
4 ! O i l Vertical Mass Flux at Repository Horizon l
\l ,
'N/ ~
l j 1 ~2 <
~ 4 2 -
!O _ s i 0.0
- -05 '
i -i.0 6 l .
- 1.5
{ 20 l -2 s
- -30 1
-35 l 40 i -45 l -10
-55
, 40 4,
-1.0
, / \ !
l ' ) - 1.5
-20 3 }
l I' O l 1
)
O O O .! Description of Two Cases Modeled a Conservativg-case Assumptions:
- 36% of packages see dripping water
- 1/L fractgre/ matrix particle transition rate (conservatively low, whi4h gives more transport through fractures; L= formation thickness) j
- No backfill Optimistic-cpse Assumptions:
- 4% of paqkages see dripping water
- 50% galvgnic protection of waste packages
- Fuel-rod cladding reduces release rate
- 1/(0.1L) particle transition rate between fractures and matrix (less fracture transport than conservative case) backfill VANLUIK/ACNW%.10 i
o o o . Assumptions Common to Both Sensitivity Analysis Cases o Based on TSPA-1995 Model o 83 MTU/ acre Thermal Loading o " Drips on W!aste Container" Release Model: Diffusion , Through Corrosion Pits Before Contacting Advective Flow o Drinking Wa,ter Doses (2L/ day) at Both 5 km and 30 km Downgradiept. o Differences Fl rom TSPA-1995:
- Used repqsitory-horizon matrix / fracture Darcy fluxes and pore velodities from LBNL-1996 DKM model for parameter set #4 (ITC) UGH fit) and infiltration map #2 (Flint, Hevesi,
, and Flint,1996) 1 29J,36CI, and 14C migrate through EBS as aqueous species ; VANLUIK/ACNW96.11
O O O . 100,000-yr Total Drinking Water Dose History 83/ drips on WP/at 5 km LBNL-1996 DKM fluxes: parameter set #4. infiltration mad #2 1 04 :
=
i i i i
.: =
Conservativ,e - 1Q3
! -- Optimistic !
- . ----- TSPA-1995 : :
m 1 02 ,. . . . . . . . . , . . u , s K : ,... 3.......-----.... . ...,,,,,, ; s -
...... -----v ...........
E 10' - --=---:- ------+-- -- -- -
--+- --
- :- a q) : f--
$ 10 ;J r f n - -- - -
+ .
v : . G) . . :. . . (gontributing radionuclides: **Tc,jf*l,237Np) ,; 10-1 _. . . . m o : : Q 10-2 j ..
. . . g .
. -i ~ ~ " .. . ..q
} Q-3 . . . . .
- . . ./ : i
"' 'y' ' ' ' ' ' ' ' ' '
10-4 i l 0 20000 40000 60000 80000 100000 ; Time (yrs) ,
-_______________m.__ . _ _ _ _ _ _ __
O O O 100,000-yr-Total Drinking Water Dose History 83/ drips on WP/at 30 km LBNL-1996 DKM fluxes: parameter set #4. infiltration map #2 10 8
~
8 5 Conservative 5
~
ptimistic .._; 10 5
------ TSPA-1995 n .
" +
> 10 =
g
' = . . . . . . . . . . . . . .+ "
- ~ .- ..- ..
g E
.57-a) 10-' e
+
.- +- - - = = +- , -"
km. :: ** , E E f~ E
~
10-2 ,
~
[ .
+
(Contributing radionuclides: Tc, ' 'l, ' 7" Np)
.+ . , ,
O
- ~
a 10-3 , 5
+. ---
_. = ..- . _ _ . . . , g
. f *.._,...
e -
.}}