Association for Information and Image Management 11 30 Wayne Avenue, Suite 11 00 Silver Spring, Maryland 20910

301 t587-8202

Centimeter

Inches

MRNUFQCTURED TO R I I M STQNDFIRDS

BY RPPLIED IMFlGEi INC.

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsi- bility for the amuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Refer- ence herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recom- mendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

U m L n

E %

m

U

, --,

I a - U

I I , .P , I l l , I 1 1 , , , I / ,

I

ABSTRACT

This proceedings, Dynamic Analysis and Design Con- siderations for High- Level Nuclear Waste Repositories, con- sists of papers presented at the Symposium sponsored by the Nuclear Dynamic Analysis Committee of the Structural Division of the American Society of Civil Engineers held in San Francisco, California on Auaust 19-20, 1992. It brings together the ideas of industry experts, outstanding researchers. program managers. policy makers and practicing engineers in the field of seismic and dynamic analysis and design. These ideas are arranged by six broad caiegories: 1) General overview of analy- sis and design; 2) characterization of faulting; 3) characteriza- tion of design ground vibratory round motion; 4) considera- tions for underground facilities; 8) considerations for surface facilities; and 6) guidelines for instrumentation and monitoring. This proceedings provide discussions on the relative merits and inadequacies of statwf-the-art designlanalysis practices and methodologies in the seismic and dynamic analysis and design field in relation to high-level nudear waste repositories.

Library of Congress Cataloging-in-Publication Data

Dynamic analysis and design considerations for high-level nudear waste repositories : proceedings of the symposium sponsored by the Nuclear Dynamic Analvsis Committee of the Structural Division of the American Sbciety of Civil Engineers and co- sponsored by the U.S. Department of Ener y! Office of Civilian Radioactive Waste Management, 8oIiday Inn Golden GatFay. San Francisco, California. August 19-20. 1992 / Quai A. Hossain, editor.

RZX. IS 8 N 0-87262-94 5 7

1. Radioactive waste sites - Design and construction - Congresses. 2. Earthquake engineerin -Congresses. 3. Structural dynamics - Congresses. I. dossain. Quazi A. II. American Society of Civil Engineers. Nuclear Dynamic Analysis Committee. 111. United States. Dept. of Energy. Office of Civilian Radioactive Waste Management. TD898.155.047D95 1993 621.4838-dc20 93-1 831 4

CIP

The Society is not responsibie for any statements made or opinions expressed in its publications.

Authorization to photmp material for internal or personal use under circumstances not Lllin within the fair us8 provisions of the Copyright Act is granted%y ASCE to libraries and other users registered with tho Copyright Clearance Center (CCC) Transactional Reporting Service, provided that the base fee of $1 .OO per article plus $.15 per page is paid directly-to CCC, 27 Con ress Street, Salem, MA 01970. The identificat!on for ASCE Books is 0-87262/93..$1 i.15. Requests for special per- mission or bulk copying should be addressed to ReprintincJPermissions Department.

Cop ri ht 0 1993 by the American Society of Civil Engineers, All dghs Reserved. Library of Congress Catalog Card No: 93-1 831 4

Manufactured in the United States of America. ISBN 0-87262-945-7

SYMPOSIUM STEERING COMMITTEE

QUA21 HOSSAIN. C h i r m a n Lawrence Livermore National Laboratory

Livermore. CA

MIKE HARDY J.F.T. Agapito & Associates

Grand Junction, CO

KENNETH MARK Bechtei National, Inc.

San Francisco. CA

JACK B'ROUKKE Woodward-Clyde Consultants

Oakland. CA

ASHOK PATWARDHAN Woodward-Clyde Consultants

Oakland. CA

ROBERT SHAW Electric Power Research Institute

Paio Alto. CA

WALTER SILVA Pacific Engineering & Analysis

El Cerrito. CA

CARL STEYP Electric Power Research Institute

Palo Alto, CA

FRANK SWAN, Il l Geornatrix Consultants

San Francisco. CA

MARIE MCGUJNN ESS. Senior Monoger. Technical Services American Society of Civil Engjneers

New York, N Y

me G;overment reserves for itself and others acting on its Malf a royalty free, mmxclusive, irrevoable, world-wide license for govemnental purpses to publish, ~ s t r i b t e , translate, duplicate, exhibit, and perform any such data amrighted by the amtractor.

... ill

I

8 CURRENT PLANS TO CHARKTERIZE THE DESIGN BASIS

GROUND MOTION AT THE YUCCA MOUNTAIN, NEVADA SITE

W.B. Simeckaf , T.A. Grant2, M.D . VoegeleZ , K.M. Cline3

Abstract

A s i t e a t Yucca Mountain Nevada is curren t ly being studied t o assess i t s s u i t a b i l i t y as a potentlal host s i t e for t he nation's f i r s t commercial high level waste repository. The DO€ has proposed a new methodology fo r determining design-basis ground notions t h a t uses both d e t e m n i s t i c and probabi l i s t ic methods. The role of the deterministic approach i s primary. needed by design engineers in the characterization of ground motions. The probabi l i s t ic approach provides a log ica l s t ruc tured procedure for in teqra t ing the range of possible earthquakes t h a t contribute t o t h e ground motion hazard a t the s i t e . I n addition, p robab i l i s t i c methods w i l l be used as needed t o provide input fo r the assessment of long-term repository performance. T h i s paper discusses the loca l tec tonic environment, po ten t ia l seismic sources and t h e i r associated displacements and ground motions. It also discusses the approach t o assessing the design basis earthquake for t he surface and underground f a c i l i t i e s , as well a s selected examples of the use of t h i s type of information i n design a c t i v i t i e s .

I t provides the l eve l of d e t a i l

Introduction

The occurrence of s ign i f icant h i s to r i ca l earthquakes i n t h e Great Basin and the presence of f a u l t s i n the Yucca Mountain area with Quaternary movement indicates the need fo r a de ta i l ed evaluation of ground motion hazards i n the design of repository f a c i l i t i e s and the evaluation of repository performance a t Yucca Mountain. se lec t ion of a seismic design bas is must include consideration of d i f f e r ing design and performance concerns t h a t a r e applicable during various time periods over the 10,000 year repository l i fe t ime. preclosure time period (C100 yrs) is t he period when surface

The

The

1U.S. Dept. of Energy, P.O. Box 98608, Las Vegas, NV 89193 ZScience hpplications International Corp., 101 Conveniion Center Dr.,

3Woodward-Clyde Federal Services, 650 Maryland Ave., S te . 695, Ste. 407, Las Vegas, NV 89109

Washington, D . C., 20024

i 76

CUKKENI' PLANS

f a c i l i t i e s will be i n operation, waste w i l l be emplaced, and over which r e t r i evab i l i t y must be maintarned. (10,000 yrs) i s the period over which re leases t o the accessible

environment from the reposltory must be controlled and remam within limits prescribed by regulatlon. time period i s the f i r s t 300 t o 1 , 0 0 0 years when the engineered ba r r i e r system i n the repository m u s t provide subs tan t ia l ly complete containment. performance concerns, the repository is a l so composed of surface and underground f a c i l i t i e s t ha t may have d i f fe ren t requirements when considering seismic design.

The postclosure t i m e period

fntluded within the postclosure

In addition t o the d i f f e r ing tune periods and

T i t l e 10, Chapter 1, Part 60 of the Code of Federal Regulations (10 CFR 60) spec i f ies the performance objectives and general design c r i t e r i a for a mined geologic repository. repository f a c i l i t i e s be designed so t h a t protection is afforded against radiation exposures and releases i n accordance w i t h standards established by the U . S. Nuclear Regulatory Commission (NRC) (10 CFR 20) and t h e U . S. Environmental Protection Auency (EPA) ( 4 0 CFR 191) . 10 CFR 60 also requires tha t s t ruc tures , systems, and components t h a t a r e unportant t o sa fe ty be designed s o tha t na tura l phenomena an t ic ipa ted a t the repository w i l l not i n t e r f e re with necessary safe ty functions. Those engineered s t ruc tures , systems, and components essent ia l t o the prevention or mitigation of an accident t ha t could result i n a radiation dose t o the whole body, or any organ, of 0.5 rem or greater a t o r beyond the nearest boundary of t h e controlled area during the preclosuze period a re considered mpor tan t t o sa fe ty . bas i s fo r repository f a c i l i t i e s must s a t i s f y these requirements. While 1 0 CFR 60 imposes performance objectives t o be met i n t h e design of f a c i l i t i e s , i t does not specify a pa r t i cu la r methodology t o be used i n developing a seismic design bas is a s was done f o r nuclear power p lan ts i n 10 CFR 100, Appendlx A . For a repository, it i s the applicant 's respons ib i l i ty t o develop the seismic d e s i p bas i s and then demonstrate t ha t t h e NRC performance objec t ives w i l l be met.

1 0 CFR 60 requires tha t

Any proposed methodology for developing a seismic design

Local Tectonic Environment A t Yucca Mountain

Reaional and loca l tec tonic models. Tectonic models for fau l t ing i n the Yucca Mountain area f a l l i n t o three general categories: detachment, high-angle normal, and s t r ike - s l ip f au l t i ng models. tha t bound Yucca Mountain a re l i s t r i c and merge with a shallou- dipping detachment f a u l t a t depths of about 2 km. Scott (1990) has also suggested tha t t h i s detachment i s par t of a system of regional extent on which the majority of l a t e Ter t ia ry extension i n t h e region has occurred. Malaonado (1990) and Carr and Monsen (1988) ind ica te tha t movement on the detachment f au l t surfaces t h a t a r e found i n t he area ceased about 7 .5 my ago. Therefore, detachment f a u l t s a r e not considered s igni f icant seismic sources i n t he Qua te rnaq .

Scott (1990) has proposed tha t t h e north-trending f a u l t s

Doser and Smith (1989) have reviewed t h e source parameters f o r large h i s t o r i c Basin ana Range earthquakes. They found tha t a l l tbe

I I , I1

17H NUCLEAR WASTE Ht'l'OSITORIES

earthquakes occurred on f a u l t s w i t h dips greater than about 40°, t ha t t h e f a u l t s were generally planar with no evidence of l i s t r i c or low angle faul t ing, and tha t focal depths were generally 8 km o r greater . T h i s would indicate tha t s ignif icant earthquakes i n the region are most l i k e l y t o occur on r e l a t ive ly high-angle f a u l t s that penetrate the c rus t .

A s t r ike - s l ip faul t ing model that predicts t ha t f au l t i ng pat terns would be similar t o the 1932 Cedar Mountain earthquake has been proposed for t he f a u l t s a t Yucca Mountain (Bell, 1988) . Several s tudies indicate tha t s t r ike - s l ip faul t ing may have been increasingly important i n the Walker Lap.e zone during Pliocene and Pleistocene time ie.g., Reheis and McXee, i 9 9 1 ) . Wright (1989) proposed t h a t Crater Flat is part of a larqer s t r ike - s l ip f a u l t zone. t h a t a s e r i e s of negative gravity anomalies extending from Stewart Valley t o Crater Flat represent a se r i e s of pull-apart basins he called the "AMrgosa Desert r i f t zone" t h a t formed a t t he northern end of the Pahrump Valley s t r ike - s l ip f au l t zone (Figure 1). Schweickert (1989) a l so proposed tha t s ignif icant amounts (up t o 26 km) of r igh t - l a t e ra l o f f se t has occurred along a northwest-trending f a u l t zone i n the southern Amargosa Valley and Crater Flat areas on the basis of o f f se t Mesozoic s t ruc tu ra l features .

He concluded

Local Faultinq Rates and Patterns. The Yucca Mountain s i t e area i s located i n series of eastward t i l t e d blocks of Miocene tu f f t ha t a r e bounded by dominantly normal, north-trending f a u l t s . The potent ia l underground repository is located i n one of these fault-bounded blocks. 13-my-old t u f f s is about 300 m on these f au l t s . trenching s tudies conducted ear ly i n the program indicated t h a t most of these f a u l t s experienced some movement during the Quaternary time period (Swadley e t a l . , 1 9 8 4 ) . movement h i s to r i e s have been l imited t o a few l o c a l i t i e s .

Apparent ve r t i ca l displacement of the Preliminary

To date, detai led s tudies of f a u l t

The Paintbrush Canyon f au l t (Figure 1) i s located ju s t t o the east of t he proposed surface f a c i l i t i e s and ramp por t a l s along the east s ide of fidway Valley. Whitney and Muhs (1991) give a f a u l t length of 33 km and have found tha t the f a u l t is divided i n t o 5 segments with Quaternary faul t ing extending for distances of 3-4 km i n each of t he southern three segments. Maximum known Quaternary displacement occurs a t Busted Butte where the oldest of four s o i l horizons near t he base of a sand ramp is displaced 4 . 1 m (Whitney and Muhs, 1 9 9 1 ) . over l ies an eolian unit containing the 738-ka-old Bishop ash (Whitney and Muhs, 1991) . a t Busted Butte. Whitney and Muhs (1991) note t h a t sl ickensides indicate a l e f t - l a t e r a l component of movement on the f a u l t r e su l t i ng a t o t a l oblique slip of about 5.8 m over t he past 700 ka or a s l i p r a t e of 0.0083 m/yr .

The s o i l i s estimated t o be about 700 ka old since it

Five episodes of f au l t i ng have o f f se t t he 4 soils

The S o l i t a r i o Canyon f au l t forms the western boundary of t he potential repository block and has a t o t a l f a u l t length of about 30 km, including two main splays a t the south end of the f a u l t (Figure

I79

1) . Swadley e t a l . ( 1 9 8 4 i report Quaternary displacement In trenches across the f a u l t , b u t these trenches do not provide information on slip r a t e s or s ingle event displacements. Bell 11988) used low-sun-angle photography and scarp p ro f i l i ng t o ident i fy a Holocene scarp on the f a u l t w i t h a scarp i n Early t o mddle Pleistocene deposits has a displacement of about 1 m. Bell (1988) found that nearly contlnuous scarps extend for about 12 .5 km along t h i s f au l t b u t t h a t displacements i n thick calcretes were small, indicating tha t large v e r t i c a l o f f se t s (several meters) have not occurred along t h i s f a u l t i n the l a t e Quaternary.

displacement of about 0.2 m. A nearby

The Fatigue Wash and Windy Wash f a u l t s a r e two closely-spaced, pa ra l l e l f a u l t s t ha t l i e west of the S o l i t a r i o Canyon f au l t i n Crater Flat . Analysis of low-sun-angle photographs by Ramelli e t al. (1989) indicates t ha t these f a u l t s form low scarps i n :he alluvium of Crater Flat and t h a t they in t e r sec t toward the south (Figure 1) . The

FIGURE 1. Fbgional model for Quaternary faulting in the Yucca MounUin Region m a from Wright (1989). Solid lines are faults wilh Quaternary movemen& patterned arom

are puU-apan basins formed by movsmenr on major Slrke-8llp fauns. Inset shows bcrtion ol faub mu Yucca MounWn.

I HO NUCLEAR WASTE KEWSITORIES

combined t o t a l length of the f a u l t s i s about 37 km. a scarp 0.7 one of the f a u l t s were investigated by Whitney (1986). Whitney distinguished a t l ea s t 7 episodes of f au l t movement i n t he trenches w i t h 4 episodes occurring i n the l a s t 300 ka; indicating an average recurrence interval of about 75 ka for events a t t h i s location. Maximum cumulative ve r t i ca l offset during these l a s t 4 episodes i s 0 . 4 m (Whitney, 1986) t o 2 m (Shroba e t a l . , 1990) . The youngest event recorded i n the trench i s believed t o be Holocene i n ape w i t h displacement l e s s than 0 . 1 m. consistent w i t h a Quaternary s l i p r a t e of 0.008-0.01 m / y r on t h i s f au l t system.

The Bare Mountain f a u l t l i e s along the western edge of Crater Flat a t t he base of Bare Mountair;. f a u l t is marked by the s teep range front of Bare Mountain w i t h faceted spurs truncating the ridges between drainages (Reheis, 1986). Reheis (1986) found evidence for Holocene or l a t e Pleistocene movement on the f a u l t along two 3-km-long segments of the f a u l t . . Reheis (1986) a l so found evidence tha t t h i s event may have resul ted i n a displacement of 1.75 m on the southern segment.

Trenches across

T h i s information i s

It is about 23 km long. The

Planned Approach fo r Selecting a Ground-Motion Dosiqn Basis 1

Any st rategy for select ing a ground-motion d e s i 9 basis fsi-

repository f a c i l i t i e s must consider the various factozs chat a f f e c t t he decision. In the case of t he Yucca Mountain s i t e , a ground- motion design bas i s must recognize the d i f f e r ing nature of surface and subsurface s t ruc tu res w i t h varying design l i fe t imes covering the preclosure (<lo0 yr) or postclosure (up t o 10 ,000 yr) periods. design basis mus t also consider t h e r e l a t ive hazards posed by the f a c i l i t i e s being designed a s compared t o other c r i t i c a l f a c i l i t i e s i n determining the l eve l of safety tha t i s acceptable. is used mus t demonstrate t o the NRC and the public t ha t t he r e su l t i ng design adequately protects the public health and safety against any credible earthquake ground motion. defensible, robust, and reasonably reproducible by other workers i n order t o withstand the intense review t o which the project w i l l be subjected. King (1990) defines robustness t o mean tha t t he design basis t ha t is provided should not be sensi t ive t o new information other than major changes i n t he perception of the seismic hazard.

The

Any method t h a t

The method used must a l so be

King (1990) a lso notes the need :or a ground-motion design bas i s It i s always possible t o increase the sa fe ty

However, a t some point t he cost of doing t ha t i s cost e f f ec t ive . of a s t ructure or system. so cannot be j u s t i f i e d by t h e amount of increased safety achieved or because of the competing demands for f inancial resources. effectiveness can be determined by evaluating the costs of adopting d i f f e ren t design bases against t h e i r effectiveness i n preventing earthquake damage and maintaining safety functions when subjected t o a range of anticipated seismic events.

of f au l t i ng and seismicity i n the s i t e area.

Cost

A f i n a l consideration i n adopting a design basis i s the nature A t Yucca Mountain, t he

CURRENT PLANS 181

low r a t e of a c t i v i t y on the f a u l t s and the apparent long recurrence intervals for seismic events on t h e f a u l t s (50 t o 100 k a ) indicates that methodologies tha t a r e used i n more act ive areas, such a s coastal California, may not be sui table or yield comparable r e s u l t s when applied i n t h i s case.

After considering the f ac to r s discussed above, the DOE decided t o use the method tha t i s described i n the S i t e Characterization Plan (SCP) (DOE, 19881, King e t a l . ( 1 9 8 9 ) , and King (1990) - The method adopted for determining the ground-motion design basis for preclosure f a c i l i t i e s uses a spec i f i c deterministic approach supplemented by a probab i l i s t i c hazard analysis . A d e t e m n i s t i c approach was selected as the primary method because a reproducible method can be defined and because of the NRC's continued emphasis on the 3etpzzLinistic approach fcr determining ground-motion desi? 3ases. considerations were f e l t t o outweigh +_tie use of a p robab i l i s t i c assessment a s a primary basis Siren though it vas the opinion of most of the technical staff ;flat a p robab i l i s t i c assessment was a technically mcze defensible approach i n t h i s s e t t i ng .

These

Deterministic methodologies usually attempt t o define a .maximum" earthquake t i e d t o a spec i f i c f au l t t h a t i s believed t o be the source of t he strongest motions t o which a s i t e could be subjected. method is t he determination of what const i tutes t he appropriate "maximum" event. Another concern i s tha t using a "mxmum" event approach a t Yucca Mountain would produce a r e su l t t ha t is unrea l i s t i ca l ly conservative given the long recurrence in t e rva l s and low s l i p r a t e s i n t he area and the operational functions of Surface f a c i l i t i e s . The DOE adopted a conservative method that would produce r e s u l t s comparable t o exceedance probabi l i t ies for t he design-basis ground motions used on more hazardous f a c i l i t i e s such a s U.S. nuclear power plants . that the annual exceedance probabili ty f o r the determinis t ical ly derived Safe Shutdown Earthquake a t these plants is generally i n the range of t o lo-' (Reiter and Jackson, 1983).

One of the major areas of controversy i n applying t h i s

Studies of a number of these pouer plants indicate

The method proposed by DOE (19881 uses a 10,000 year Cumulative S l ip Earthquake (CSE) as the design basis fo r preclosure f a c i l i t i e s important t o sa fe ty . produce the observed or estimated average Quaternary s l i p r a t e on a f a u l t p were it t o recur a t equal, 10,000-year-time in t e rva l s . T h i s approach is considered t o produce reproducible r e su l t s because the interval of concern is defined and not subject t o varying in t e rp re t a t ions of what const i tutes an undefined 'maximum" event. This approach i s a l so considered t o be robust is l t ha t application of the CSE methodology is expected t o ident i fy only one o r two faults t ha t w i l l control the design basis . f au l t i s a parameter t ha t i s more readily determined than other parameters, such as the age or magnitude of individual paleoseismic events, and i s l e s s l i ke ly t o be subject t o large var ia t ions. approach also has advantages over techniques tha t use fault-rupture length because of t he uncertaint ies i n f a u l t rupture pat terns noted

The CSE i s a hypothetical earthquake t h a t would

The averaqe s l i p r a t e f o r a

The

I LI u I ,

182

J i l I I ,

N U c' 1 E AH WA S'JI: K E POS I7'OH I ES

by Bell (1988) for the b x i i n and Range Province.

It is concluded tha t the use of a 10 ,000 year CSE meets t he in ten t of es tab l i sh ing a design basis w i t h an annual probabi l i ty of exceedance of t o However, it should be noted tha t such an earthquake, l i k e any design earthquake, i s purely hypothetical and t h a t ac tua l events t ha t occur i n the fu ture may be la rger o r smaller than t h i s event. seismic design bas is tha t ensures miriirnal disruption t o t he operation of f a c i l i t i e s tha t are important t o sa fe ty when subjected t o earthquakes up t o t h i s l eve l . As noted by King (1990), greater-than- CSE events during the preclosure period a re unlikely but possible. Therefore, i t m u s t be demonstrated tha t sa fe ty functions can be maintained for such events, including maximum-magnitude earthquakes. Engineering analyses t o demonstrate la rge safe ty margins t o accommodate earthquake loads which exceed the nominal CSE design basis a re an innerent par t of t he method.

The CSE methodology i s intended t o define the

It is a l s o planned t o conduct a comprehensive p robab i l i s t i c hazard ana lys i s a s part of s i t e charac te r iza t ion . design, t he probabilistic analysis w i l l be used t o confirm tha t t he de te rminis t ica l ly derived design bas is f a l l s within the desired pzobabili ty range. The probabi l i s t ic ana lys i s w i l l a l s o allow a l t e rna te conceptual models of regional and loca l tec tonics and a l t e rna t ive , non-Poisson models of earthquake occurrence t o be analyzed and incorporated, i f appropriate. characterlzation data is collected and the technicai pos i t ion of t h e NRC becomes be t t e r defined, the DOE will per iodica l ly re-evaluate the r e l a t ive weight t ha t should be given t o de te rminis t ic ana p robab i l i s t i c approaches i n se lec t ing a preclosure design bas is .

The p robab i l i s t i c analysis w i l l a l so be the primary tool i n

For preclosure

As more site

postclosure design and perfonnance analysis. tee et al. (1991), a primary postclosure design concern is t he performance of the encjineered ba r r i e r system. the engineered ba r r i e r system (according t o t h e current reference design) a re the waste emplacement borehole, t he a i r gap around the uaste container, and the waste container i t s e l f . The multiple occurrence of strong mocion during t h e postclosure phase may be an important consideration fo r waste container design because of : (1) d x e c t mechanical e f f e c t s caused by movement of container within t h e vas te emplacement hole and subsequent puncture or rupture of the container, espec ia l ly a t times of increased corrosion; and/or (2) stress-induced rock spa11 within the waste emplacement hole t h a t can reduce or eliminate a i r space surrounding the waste package, r e su l t i ng i n container puncture or perroanent uas te container-wall contact , potent ia l ly amplifying cvrrosion r a t e s . Thus, t he re may be a requirement for a design t o address l a t e r a l movement within t h e emplacement hole or a requirement for increased material strength.

for the repository, it is possible tha t t he repository w i l l be subjected t o a nwnber of ground motion events of varying magnitude

As noted by the SCP and

The bas ic elements of

When considering the 1,000- o r 10,000-year performance periods

from both local and regional sources. Repeated or long-term wall contact from an e a r l i e r e v e n t may result i n s l i gh t container damage and increased corrosion a t contdct points or damaoe locations. The container would t h e n be more susceptible t o fur ther damage and corrosion due t o shaking and additional repeated wall contact during subsequent events, po ten t ia l ly reducing cont-arner performance. The cumulative e f f ec t of a nwnber of events producing moderate ground motion at the repository may be more s igni f icant t o long-term uaste package performance than the e f f e c t s of a s ing le la rge event, because (unlike other f a c i l i t i e s s u c h as a nuclear power p lan t ) there w i l l be no provision for repa i rs a t t he reposltory a fcer an event occurs.

In addition, t he d i f f e ren t design purpose of t h e ground motion evaluation f o r t he postclosure period, i n comparison t o an engineered f a c i l i t y such a s a nuclear power plan:, must be considered. nuclear power p lan t , a s i n a l e large event is hypothesized i n order t o design fo r t he a b i l i t y t o s a f e l y s h u t down the reactor i n response t o tha t event. For t he postclosure period of a repository, the concern i s i n evaluating the range of events +Yat a re reasonably l i ke ly t o occur over a long time period i n ordei to assess the performance of a system tha t m y not have ac t ive human monitors or operators. Because of these differences, a determinis t ic methodoiogy using a s ing le postulated design earthquake on a s ingle source s t ruc ture is considered inappropriate. consideration of t he f u l l ranae of events and seismic sources t h a t could contribute t o the ground motion hazard a f fec t ing the repository during the postclosure time period. A probab i l i s t i c model w i l l a l so allow an exp l i c i t consideration of t he uncer ta in t ies present i n this type of analysis.

For a

A p robab i l i s t i c model w i l l allow

Current Ground-Motion Estimates and Data Needs

Deterministic Estimates. As discussed above, implementing the CSE methodology requires estimating the expected displacement associated with the hypothetical CSE from the average s l r p r a t e on the f au l t and t h e n estimating the s i z e of t he earthquake from empirical cor re la t ions using h i s to r i ca l earthquake da ta . For this paper, a corre la t ion between the s e i s m c moment and mximum displacement of h i s t o r i c events i n t he Basin and Range Province is used. basin and Range Province because the parameters of f au l t rupture appear t o vary from province t o province and data fram a s ingle province appears t o be a b e t t e r predictor of events i n t h a t province. Seismic moment was used i n t h i s study because published values for t h i s parameter are avai lab le f o r nearly all Basin and Range earthquakes; moment may be ca lcu la ted using e i t h e r geologic, qeodetic, o r seismologic information; and rupture length and displacement a re more d i r e c t l y re la ted t o t h i s measure than t o o ther measures of earthquake magnitude.

The correlations used i n t h i s study use only data from t he

Figure 2 shows a corre la t ion between log moment and log maxim\nn displacement. f igure is

The formula f o r the regression line shown on the

NUCLEAR WASTE KEYOSITORIES

Log Moment = 25.005 + 1.123 (log D ) + 0.209 (log D ) 2 .

The upper l l m i t f o r s l i p r a t e s on f a u l t s i n the immediate s i t e are currently appears t o be about 0 . 0 1 m / y r . f a u l t s would then be 0 . 1 m for a 10,000 year CSE calculat ion. Multiplying t h i s by 3 t o estimate a maximum displacement (Bonilla, 1982) and using the equation above y i e l a s a moment of 3.0 x l o z 5 dyne-an or moment magnitude of 6.3. Ground Acceleration (PGA) of about 0 . 5 g (Campbell, 19821, i f the closest f au l t (Paintbrush Canyon) i s considered a s the source. The annual probabi l i ty of exceedance fo r t h i s hypothetical design event on the Paintbrush Canyon f a u l t i s then about 1.5 x l o m 5 using Molnar (1979).

The average s l i p on these

T h i s corresponds t o a Peak

- 1 0 1 2 - 2

COG MAXIMUM DISPLACEMENT (meters)

FIGURE 2. Relationship between log seismic mornen1 and log maximum displrcem for hiatwic Baain and Range earthquakes.

Bell 11988) a r p e s that d i s t r ibu t ive f au l t i ng should be considered i n eva1uatir.g po ten t i a l ground motion a t Yucca Mountain and that an event s imilar t o t h e 1932 Cedar Mountain earthquake IrZ, 6.9) could occur i n the area w i t h f a u l t movement occurring on s..eral of the f a u l t s in t he area. Mountain earthquake was about 2 m (Bell, 1988). 2, there i s a good correlat ion between moment and maximum displacement for larger earthquakes. and Range earthquakes have always been accompanied by maximum surface displacements of commensurate s i ze . There is currently no documented evidence t o suggest displacements associated w i t h individual events are larger than about 0 . 5 m indicat ing t h i s scenario may not apply t o Yucca Mountain. However, t h i s and other types of multiple f a u l t or multiple event hypotheses should be evaluated when analyzing f a c i l i t y perfonnance and sa fe ty margins.

Maxlmuna displacement during the Cedar As shown on Figure

This shows t h a t larger Basin

The bounds on preclosure ground motion for use i n engineering analyses of t h e perfonrance of a pa r t i cu la r design during low- probabili ty events can be examined by considering a number of scenarios. full- length rupture on individual f au l t s , c o s e i h c movement on several f a u l t s (e.g. , Bare Mountain, Windy Wash, S o l i t a r i o Canyon,

These scenarios should consider p o s s i b i l i t i e s including

CURRENT PLANS

and Paintbrush Canyon), and possible longer rupture lengths on f a u l t s than current ly documented C e . g . , Bare Mountain f a u l t extends i n t o Amargosa Desert w i t h s t r i k e - s l i p movement). maximum magnitudes of aboat M,, J to 7 114.

These scenarios y i e ld

These analyses should a l s o include multiple-event scenarios where separate events occur on several f a u l t s w i t h tune in t e rva l s a s short a s a few minutes between events. behavior would be the 1954 Dixie Valley-Fairview Peak earthquake sequence. T h i s sequence consisted of 4 events that produced surface ruptures on d i f f e ren t f a u l t s defining a zone about 40 !un wide. The earthquakes ranged from 4 6 . 1 t o 7 . 2 and the int.ervals between events ranged from about 4 minutes t o 4 months (Thompson, 1984). Coseismic or multiple-event faul t ing i s suggested by t h e occurrence of volcanic ash deposi ts f i l l i n g voids along several f a u l t s and a potent ia l ly synchronous Holocene event on some f a u l t s (Bare Mountain, Windy Wash, Bow Ridge). Such a sequence of events might include an event on t h e Bare Mountain f au l t (4-7) and separate events on t h e Paintbrush Canyon, t h e S o l i t a r i o Canyon, and the Fatigue Wash-Windy Wash f au l t systems of M,,-6-6 1/2. agreement w i t h t h e short rupture seqm-nts and r e l a t ive ly small displacements observed on these f a u l c s .

An example of t h i s type of

T h i s scenario would be i n

P robab i l i s t i c Estimates: The most comprehensive p robab i l i s t i c assessment performed t o date for the Yucca Mountain s i t e is by URS/John A . Blume ( 1 9 8 7 ) . Because of the uncertainty i n estimating the r a t e of seismici ty i n a region with low a c t i v i t y and long recurrence in t e rva l s from the h i s to r i c record, t h i s study included both area and f a u l t sources. Earthquake occurrence on f a u l t s t h a t were considered s ign i f i can i t o the s i t e was described by a truncated Gutenberg-Richter densicy function. Faults were assigned seismici ty by developing a correlat ion between f au l t length and average f a u l t deformation r a t e . Fault a c t i v i t y was constrained by s l i p r a t e , maximum maqnitude, and b-value, w n i l e area sources were constrained by f au l t a c t i v i t y , and pre-underground weapons t e s t i n g h i s t o r i c a l seismicity for t he period 1932-1562. Average a c t i v i t y i s about 0.015 events of M24/1,000 kmzlyr over a 20 ,000 km2 area t h a t excludes seismicity of easterr . Caiifornia, and lower seismicity regions both n o r t h and south of the s i t e . evaluated i n t h i s study. t h a t deformation accumulates equally for both horizontal (striice- s l i p ) and v e r t i c a l components. A b value of 0.9 was used, and ground motion at tenuat ion w i t h source distance and magnitude were estimated t h e relat ionships of Campbell (1982): The preferred p robab i l i s t i c hazard r e s u l t s were s imilar t o results of an e a r l i e r evaluation based on h i s to r i ca l ly averaged seismicity., From t he preferred hazard model (Figure 31, a design bas i s of 0.4 g *as reconrmended using a probabi l i ty of exceedance of 5 x for t h e preclosure surface f a c i l i t i e s .

Several a l t e rna te f au l t i ng models were The preferred mode: of t he study assumed

Ground motion concerns related t o t h e postclosure period have been revieved by Lee e t a l . ( 1 9 9 1 ) . Blume (1987) study a s a basis for for i t s estimates. As discussed

This study used the URS/John A.

I I I I , , , I ,

.

previously, postclosure considerations involve concerns related t o both the s i ze of potent ia l around motion and the number of strong- motion events t ha t might orzur during periods of i n t e r e s t . study found tha t a peak ground motion of 0 . 6 g had an approximate 10% chance of exceedance i n a 1 , 0 0 0 year period ( the substant ia l ly complete containment period for the engineered ba r r i e r system). For the same design acceleration and a 10,000 year period, t he re was a nearly equal probabili ty of occurrence for two or mme everits (0.283) a s f o r one eve:.; ; r ) .366) . Figure 4 shows the cumulative probabi l i ty of t he number c': 2 0 . 4 g events that occur i n 10,003 years equaling or exceeding a given value for the preferred model. estimated t h a t the most probable nunher of occurrences of 0.1 g and greater i n 1,000 years ranges from about 8 t o 12 .

T h i s

It was a l so

Data Needed From S i t e Characterization. The determinis t ic and p robab i l i s t i c e s t w t e s given sbove are presented solely t o i l l u s t r a t e t he models and do not represent f i n a l posi t ions on design bases. seismicity i n t he s i t e area tha t additional information i s needed from si te character izat ion. evaluation, data are needed on the slip r a t e s of local f a u l t s (including any potent ia l s t r i ke - s l ip component), the displacement

There a re su f f i c i en t uncertainties concerning the nature of

For a l l methods of seismic hazard

10

10

0.6

0 0

0.4

0 2

0 0 1 2 3 4 S 0 7 D 0 l O l l l 2 1 3 1 4

NUUDER OF WEMJ

FIGURE 4. Cumulative probability of N

and greater episodes of site accelerations >0.4 g over 10.000 year

period using model in Figure 3.

From Lee el ai. (1991). , lo*

0 1 0 2 0 5

w norwtw Ground Accrlomtlon (g)

FiGURE 3. PGA excoedance rates for the preferred model of URSNohn A. Blume (1987).

and recurrence in t e rva l s of individual events on local f au l t s , t he synchronicity of events (or lack o f ) occurring on different f au l t s , dynamic s o i l properties a t the s i t e , and the 3 D geometry and interconnections of the d i f f e ren t 1c;cal f a u l t s .

Comparison of Results. Both the p robab i l i s t i c estimates and the deterministic 10,000 year CSE estimates given above appear capable of providing a ground motion design basis t ha t meets the goal of having exceedance p robab i l i t i e s i n t he l o e 3 t o range. For example, the CSE estimate of a PGA of about 0 . 5 g corresponds t o an exceedance probabili ty about 2 x IOeb u s i n g the results of t h e URS/John A. Blume (1987) study. While the DOE has s:ated in the SCP t ha t the level of conservatism contamed i n the above goals is adequate for protecting public health and safety, the question remains a s t o what level of conservatism w i l l eventually be found t o be acceptable from a regulatory and public perception viewpoint. backaround on t h i s question, a srudy was conducted by Subramanian e t

;a l . (1989) t o evaluate the cost e f f e a i v e n e s s of adopting various ground motion design bases for preclcsure f a c i l i t i e s on meeting public health and safety requirements. d m x s s e d i n the next section.

I n order t o provide

The r e s u l t s of t h i s study are

Impact of Ground-Motion Design Basis on Safety and S i t e S u i t a b i l i t y

Subramanian e t a l . (1989) have assessed the costs and benefi ts associated uirh varying t h e seismic design basis of the preclosure waste-handling f a c i l i t i e s . expert judgment s o l i c i t e d i n an interdiscipl inary workshop environment. The estimated cos t s for individual a t t r i b u t e s and the assmptions underlying these cost estimates (seismic hazard levels , f r a g i l i t i e s , radioactive-release scenarios, e t c . ) were subject t o large uncertaint ies a t t h i s ea r ly stage of the project . uncer taint ies were generally iden t i f i ed but not t r ea t ed exp l i c i t l y . The assessment considered a range of vibratory ground-motion level3 w i t h PGAs between 0.2 and 1.0 g ( v e r t i c a l PGA - horizontal P a ) . Both accident- and nonaccident-reiated costs and benefi ts ( a t t r i bu te s ) were quant i ta t ively estimated a s a function of the design basis . Tns assessments were based on expert opinion derived from previous analyses and observed performance of heavy indus t r i a l f a c i l i t i e s , rather than a f ac i l i t y - spec i f l c quant i ta t ive analysis .

The assessment was based primarily on

These

Accident-related a t t r i b u t e s considered i n the Subramanian et al. (1989) study included the p robab i l i s t i ca l ly derived costs of public rachological exposure, worker radiological exposure, o f f s i t e property damage and cleanup, ons i t e damage, mission delays. Nonaccident- re la ted a t t r i b u t e s considered included s t ruc tu ra l design and construction costs, equipment procurement and quai i f icat ion costs, l icensing costs, s i t e character izat ion Costs, and the costs of potent ia l progrart-snatic delays. The assessment focused on the main waste-handling building. t h i s building r e su l t i n a s t ruc tu re w i t h 5-foot-thick reinforced concrete shear-wall construction. therefore inherently r e s i s t an t t o ground motion.

Shielding requirements for the hot cells in

The resul t ing s t ruc tu re is

188 NUCLEAR WASTE REPOSITORIES CURRENT I’1,ANS

a

189

Subramanian e t al. (1989) found that , notwithstanding t h e uncertaint ies and approxhat ions of the s tudy , t h e repository surface f a c i l i t i e s a r e low seismic risk f a c i l i t i e s because t h e expected cost and r i s k t o t he p * b l i c ( i .e. , accident-related costs) i s very low a t a l l design levels . The assessment found tha t accident r i s k s associated with any seismic-design level above 0.2 g would be extremely small. least a 95% chance t h a t , i f t he building were designed t o 0.4g ( the basis of the current conceptual design), it would suffer only l i g h t damage w i t h no l o s s of containment i f an event producing 1.0-g ground motion should occur. The study concluded t h a t , from a cost-benefit standpoint, t he optimum design level could vary from 0 .24 t o 0.53 g. I n the view of t h e study, a design basis below 0.3 g was not precluded by safety considerations but would most l i k e l y r e s u l t i n increased l icensing costs . However, i t was found tha t choosing a design bas i s above 0.3 g would have l i t t l e impact ori reducing r i s k t o the public.

For example, the study found there would be a t

The results of t h e Subramanian et a l . (1989) study ind ica t e tha t w h i l e t h e select ion of a ground-motion design basis may be a major l icensing, design, and cost issue, i t does not s ign i f i can t ly a f f e c t site s u i t a b i l i t y determinations or public safety even i f r e l a t ive ly minimal values (0 .2 g o r above) are selected.

References

Bell, J. W., 1988, Quaternary geology and act ive f au l t i ng a t and near Yucca Mountain, in Evaluation of t h e geologic r e l a t ions and seismotectonic s t a b i l i t y of the Yucca Mountain area, Nevada Nuclear Waste S i t e Investigation (NNWSI), Final Report; Center for Neotectonic Studies, Mackay School of Mines, University of Nevada-Reno, p. 1-1 t o 1-45.

Bonilla, M. G., 1982, Evaluation of potent ia l surface f au l t i ng and other t ec ton ic defo,mation; U.S. Nuclear Regulatory Commission, NUREG/CR-2991, 58 p.

Caurpbell, K. W., 1982, A preliminary methodology for t h e regional zonation of peak ground acceleration; Thi rd Internat ional Earthquake Microzonation Conference Proceedings, Sea t t l e , WA, v. 1, p. 365-376.

Carr, M. D. and S. A . Monsen, 1988, A f i e l d guide t o the geology of Bare Mountain; & Weide, D. L., and M. L. Faber (ed.) , This extended land, Geological journeys i n t h e southern Basin and Range: Field t r i p guidebook, Geological Society of America, Cordilleran Section Meeting, Las Vegas, NV, p. 50-57.

DO€ (U.S. Department of Energy), 1988, S i t e Characterization Plan for t h e Yucca Zlountain Si te ; U.S. Dept. of Energy Publication DO€/RW-0199, Washington, DC.

Dcser, D . I . , and R . B. Smith, 1989, An assessment of source parameters of earthquakes i n t he Cordillera of the western United States; Bul le t in of t h e Seismological Society of America, v. 19 , n. 5, p . 1383-1404.

King, J. L., G . A. Frazier , and T. k. Grant, 1989, Assessment of faul t ing and seismic hazards a t Yucca Mountain: Radioactive Waste Management and the Nuclear Fuel Cycle, v . 13, n. 1-4, p. 155-170.

Ring, J. L. , 1990, Approach t o developing a ground-motion design bas i s f o r f a c i l i t i e s important t o safety a t Yucca Mountain; High Level Radioactive Waste Management, Proceedings of the Internat ional Topical Meeting, American Nuclear Society, La Grange Park, I L and American Society of Civi l Engineers, New York, NY, v. 1, p. 103-158.

Lee, R. C. , J. L. King, and T. A. Grant, 1991, Multiple event considerations fo r postclosure seismic hazard evaluations a t Yucca Mountain, Nevada: High Level Radioactive Waste Management, Proceedings of t h e Second Annual Internat ional Conference, American Nuclear Society, La Grange Park, I L and American Society of C iv i l Engineers, New York, NY, v. 1, p . 76-82.

Maldonado, r., 1990, Structural geology of t h e upper p l a t e of t h e Bullfrog Hills detachment f a u l t system, southern Nevada; Geological Society of America Bulletin, v . 102, p - 992-1006.

Molnar, P., 1979, Earthquake recurrence in t e rva l s and p l a t e tectonics; B u l l . Seismolcgical SOC. of America, v . 69, n . 1, p. 115-133.

Ramelli, A. R., T. L. Sawyer, J. W. Bell, C. M. dePolo, F. F. Peterson, R. I. Dorn, 1389, P r e l h n a r y analysis of f a u l t and f r ac tu re pa t t e rns a t Yucca Mountain, southern Nevada; Proceedings of Nuclear Waste I so l a t ion i n the Unsaturated Zone Focus 89, American Nuclear Society, La Grange Park, IL, p. 336-343.

Reheis, PI., 1986, Preliminary study of Quaternary f au l t i ng on t h e east s ide of Bare Mountain, Nye County, Nevada; U. S. Geol. Survey, Open F i l e Report OFR-86-576, 14 p .

Reheis, M. C. and E. H. McKee, 1991, Late Cenozoic his tory of s l i p on the Fish Lake Valley f au l t zone, Nevada and California: Pacif ic C e l l , Friends of t h e Pleistocene guidebook for f i e l d t r i p t o Fish Lake Valley, California-Nevada, May 31-June 2, 1991, p. 26-45.

Reiter, L., and R. E. Jackson, 1983, Seismic hazard review f o r t h e systematic evaluation program - a use of probabi l i ty i n decision making; U . S. Nuclear Regulatory Conmission RJREG-0967, 55 p.

190 NUCLEAR WASTE KEPOSITORI ES

ScQtt , R. B . , 1990, Tectonic Setting of Yucca Mountain, Southwest Nevada, Wernicke, B. P. (ed.), Basin and Range Extensional Tectonics near t h e Latitude of Las Vegas, Nevada; Geological Society of America Memoir 1 7 6 , p. 251-282.

s t r i k e - s l i p f a u l t beneath Crater F la t , Nevada (abs . ) ; Geological Society of America, Abstracts w i t h Program, 1989 Annual fleeting,

Shroba, R. R., J. L. Whitney, E. M. Taylor, and K. F. Fox Jr . , 1990,

Schweickert, R. A. , 1989, Evidence for a concealed dext ra l

p. A90.

Quaternary movement on north-trending f a u l t s a t Yucca Mountain, Nevada: preliminary r e s u l t s (abs . ) ; Geological Society of America, Abstracts with Programs, v. 22, n. 3, p. 83.

Subramanian, C. V., N . Abrahamson, A. H. Hadjian, L. J. Jardine, J. B. Kemp, 0. K. Kiciman, C. W. Ma, J. King, W. Andrews, and R. P. Kennedy, 1989, Preliminary seismic design cost-benefit assessment of t h e t u f f repository waste-handling f a c i l i t i e s ; Sandia National Laboratories Report SAND88-1600, Albuquerque, NM.

Swadley, W. C., D. L. Hoover, J. N. Rosholt, 1984, Preliminary report on l a t e Cenozoic fau l t ing and s t ra t igraphy i n t h e v i c i n i t y of Yucca Mountain, Nye County, Nevada; U.S. Geol. Survey Open File Report 84-788, 4 1 p.

Valley earthquakes of 1954; U.S. Geol. Survey Open-File Report 85-290, p. 27-42.

Thanpson, G. A., 1984, Perspective from the Fairview Peak-Dixie

URS/John A. Blme L Associates, 1987, Technical bas i s and parmfetric study of ground motion and surface rupture hazard evaluations a t Yucca Mountain, Nevada: Sandia National Laboratories Report SAND86-7013, 99 p.

Whitney, J. W., R. R. Shroba, F. W. Simonds, S. T. Harding, 1986, Recurrent Quaternary movement on t h e Windy Wash f a u l t , Nye County, Nevada (abs.): Geological Society of America Abstracts w i t h Programs, v. 18, p. 787 .

nLitney, J . W., and D. R. Muhs, 1991, Quaternary movement on t h e Paintbrush Canyon-Stagecoach Road f a u l t system, Yucca Nountain, Nevada (abs.); Geological Society of Azerica, Abstracts u i t h Programs, v. 23, n. 5, p. A119.

Wright, L. A., 1989, Overview of the ro le of s t r ike-s l ip and normal fau l t ing i n t h e Neogene h is tory of the region northeast of Death Valley, California-Nevada; E l l i s , M. A. , (ed. ) , Late Cenozoic evolution of t h e southern Great Basin; Nevada Bureau of Mines and Geology Open F i l e 89-1, p. 1-11.

I