the continuing challenge - caltrans · the continuing challenge ... and dcvelopments in seismic...

88
THE CONTINUING CHALLENGE THE NORTHRIDGE EARTHQUAKE OF]A UARY 17, 1994 Report to the Director, California Department of Transportation by the Seismic Advisory Board George W. Housner, Chairman October 1994

Upload: vokien

Post on 04-Sep-2018

220 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

THE CONTINUING CHALLENGETHE NORTHRIDGE EARTHQUAKE OF]A UARY 17, 1994

Report to the Director,California Department ofTransportation

by the

Seismic Advisory Board

George W. Housner, Chairman

October 1994

Page 2: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

THE CONTINUING CHALLENGE

Report on

the Northridge Earthquake

of January 17,1994

Submitted to the

Director, Department of Transportation

State of California

by the

Seismic Advisory Board

George W. Hammer, ChainnanJoseph Penzien, Vice Chairman

Bruce A. BoltNicholas F. Forell

1. M.ldriss] oseph P. Nicoletti

Alexander C. Scorde1isFrieder Seible

Charles C. Thieljr., Editor

October, 1994

Page 3: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

n

First printingJune, 1994

Copyright © 1994 by the State of California, Department of Transportation. Excerpts fromthis report, excepting materials copyrighted by others, may be reproduced for non-commercialuse with attribution to the Departlnent of Transportation's Seismic Advisory Board ReportThe Continuing Chollmge: The Nmhridge Em"thquake ofJontllfry /7, 1994.

Additional copies of this report and information on current prices may be obtained bycontacting the Office of the Chief Engineer, Department ofTransportation (916) 654-6490 orby contacting:

Departmcnt of Genernl ScrvicesPublications SectionP.O. Box 1015North I-lighlands, California 95660, (916) 973-3700

Page 4: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Acknowledgments

The Seismic Advisory Board extends its thanks to all those who assisted it in preparing thisreport, including the following indi,riduals: Mr. James Roberts, Mr. James Gates, Mr. BrianMaroney. Dr. Lalliana Mualchin, and Mr. Ray Zelinski of Cah:rans for provision ofker .infonnation on bridge performance and ground motions; Dr. Nigel Priestley of the Universityof California at San Diego for the collapse assessment of concrete bridges, Dr. Abolhass:mAstanch-Asl of the University of Califomia at Berkeley for observations on steel bridgeperformance, and Dr. Anthony Shakal of the California Strong Motion InstrumentationProgram for btt"ound motion records.

The Board thanks Ms. Gail I-I. Shea, AJbany, CA, for her expert and timely editing ofthe text, and Ms. Laura Moger, Moorpark, CA for her expert layout and design of the repon.

Finally, Dr. Charles C. Thiel Jr. undenook the Clsk of organizing and editing thecontributions of the Board members into a coherent fonn, and contributed portions of the text.l-lis effortS are greatly appreciated.

George W. HousnerChairman

'"

Page 5: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

,v

Three essential challenges must be addressed by the citizens of California, if theyexpect a safe future from earthquakes:

• Ensure that earthquake risks posed by new construction are acceptable.• Identify and correct unacceptable seismic safety conditions in existing structures.• Develop and implement actions that foster the rapid, effective, and economic

response to and recovery from damaging earthquakes.

-Comptting Againrt TImtGovernor's Board of Inquiry on the 1989 Lorna Prieta Earthquake

It is the policy of the State of California that seismic safety shall be given priorityconsideration in the allocation of resources for transportation construction projectS,and in the design and construction ofall state structures, including transportationstructures and public buildings.

-Governor George DeukmejianExecutive Order D-86-90,June 2,1990

The safety of every Californian, as well as the economy of our state, dictates thalour highway system be seismically sound. That is why I have assigned tOp priority toseismic retrofit projects ahead of all other highway spending.

-Governor Pete WilsonOpening of the repaired Santa Monica Freeway, April 11, 1994

Page 6: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Table of Contents

Acknowledgments iii

Section 1

Section 2

Section 3

Section 4

Overview : IDamage to highway bridges in the Northridge earthquake I

Rctrofincd bridges perfonned adequately 3Cal trans seismic design criteria and retrofit program 3

1...:lck of progress on toll bridges 4

Public concerns and questions 4Conclusions 5

Findings and Recommendations 7Bridge performance in the Northridge earthquake 7Retrofit progranl 8))esign 9Caltrans InanagClllent actions 10

State actions 11

Conclusions 12

The Northridge Earthquake and Bridge Perfonnance 13l··hc Northridge earthquake 13Seismological characteristics and ground motion 13Cal trans seismic risk map 15Summary of damagc to highway bridges in theNorthridge earthquake 16Impact on traffic now in Los Angeles County 18

Seismic Advisory Board field trip to damaged structures 18

Similarity of Northridge earthquake to Lorna Prieta andSan Fernando earthquakes 22Performance of new and retrofitted bridges in theNorthridge earthquake 22

Performance ofsteel strucrures 23

Why Did Highway Bridges CoUapse in theNorthridge Earthquake? 25

Bridges designed and built before the1971 San Fernando earthquake 25Bridges designed before 1971 but constructioncompleted after 1971 27Bridges designed and built after 1971, but before basic desi!,1'Jlconcepts were changed in 1987 29

Summary 30v

Page 7: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Section 5

Section 6

Caltrans Bridge Design and Retrofit Program 33Evolution of Calrrans bridge design practices 33Governor's Board of Inquiry investigation 34Cal trans response to Board of Inquiry recommendations 35Peer rcvicw 36Peer review and construction of the Northridge earthquakereplacclncnt bridges 37Cal trans retrofit and design response to LOllla Prieta earthquake 38

Prioritization and screening practices 38Current StaniS of State highway bridge retrofit probrram 39Current status of city and county bridge retrofit program ..42Summary 42

Improving the Caltrans Seismic Program 43Retrofit prioritization procedures 43Site-specific snldies 44Use of earthquake prediction for priority setting 44Research should be enhanced 45

v,

Appendix A Caltrans Response to Governor's Board of InquiryRecommendations and Executive Order ofJune 2,1990 49

Caltrans Report,January 26, 1994 49Appendix B Seismological and Strong MObon Description of the

January 17. 1994 Northridge Earthquake 57Seismological characteristics 57

Strong motion recordings 57Strong motion records from bridges 63

Appendix C San Femando Earthquake, 1971 65Seismic characteristics of the San Fernando earthquake 65Damage to freeway structures 65Design improvemcms 69

Appendix D Lorna Prieta Earthquake, 1989 71Seismic characteristics of the Loma Prieta earthquake 71Impact of the Loma Prieta earthquake on bridges 71

References 75Authors' Affiliations 77

Page 8: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

TablesI-I

3-1

3-2

3-1

5-2

5-35-4

5-5

8-1

D-I

FiguresI -I

3-I

3-2

3-3

3-4

3-5

3-6

Caltrans seismic perfonnance criteria and definitions for the designand evaluation of bridges : 6

Summary of highway bridge collapses in the onhridge eanhquake 17Comparison of the average daily traffic volumes on the damagedhighwllYS before the Northridge eanhquake with the correspondingdllily traffic volumes on February 4, 1994 18

Weights used in the prioritization process 38

Status of the retrofit review of State highway bridges 38

The Caltr:ms bridge retrofit program status as ofJune I, 1994 39

SUtus of toll bridge seismic studies 41

StatuS of the seismic retrofit of city and county bridgesthroughout the State 42

Locations of bridges for which strong motion records were obtainedin the Northridge earthquake from the CSMIP program 63

Calmns bridges sustaining damage gre:ner than $100,000 during theLotnll Prieta earthquake 74

The Joc:nion of bridges that collapsed in the Northridge e:mhquakc ofJanuary 17, 1994 2Isoseismalmap of the Modified Mercalli Intensities for theNorthridge eanh(luake 14

Caltr:ms peak rock acceleration map showing the seismicmean-acceleration values used for bridge design 16

Bridge piers with completely disintegrated core, ruptured hoopreinforcement, and buckled longitudinal reinforcement 18

Steeply inclined shear cracks and cover spalling in one bell[ atFairfux & Washington on 1-10 19

Flexural cracks at the column ends in the adjacent bent atFairfax & Washington on 1·10 19

Columns retrofitted with circular steel jackets at the Cadillac Avenueofframp directly adjacent to Washington and Fairfax 19

V"

Page 9: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

5-1

4-3

4-74-8

4-4

4-5

4-6

3-15

4-14-2

Figures (continued)3-7 Flexure/shear failure in the bottom of the flared columns still visible

in the westbound bridge, even though demolition was in fullprogress at the SR-118 Mission & Gothic Undercrossing 20

3-8 Completed demolition of the southbound connector at theAntelope Valley T-S/SR-14 Interchange five days after the earthquake........ 20

3-9 Abutment joint damage in the separation structures at theAntelope Valley I-S/SR-14 Interchange 21

3-10 Expansion joint damage in the separation struCtures at theAntelope Valley I-S/SR-14 Interchange 21

3-11 The highly skewed geomeuy and short expansion joint seat widthof the still-standing center frame of the 1-5 Gavin Canyon bridge 21

3-12 Typical cracking associated with x-bracing observed in some steelgirder bridges, Northridge earthquake 22

3-13 Cracks in bottom region of plate girder web, northem bridge ofPico-Lyons Overcrossing 23

3-14 Roller and seismic restrainer damage, southern bridge ofPico-Lyons Overcrossing 23

Support damage at top of Pier No.6, Santa Clara River Bridge 23

Gavin Canyon Undercrossing, Northridge earthquake 26

Shear distress pattern at 1-10, Fail"fux & Washington,Northridge earthquake 26

Brittle column failure at 1-10, La Cienega and Venice,Northridge earthquake 27

SR-14/I-S North Connector Overcrossing, Northridge earthquake 17

SR-14/I-S Soum Connector Overhead, Northridge earthquake 28

Short column failure SR-14/1-S South Connector Overhead,Northrid!,TC earthquake 28

SR-118 Mission & Gothic Undercrossing, Northridge earthquake 29

Bull Creek Canyon Channel Undcrcrossing, SR-118,Northridge earthquake 29

Pic charts show the status of the single-column, multiple-columnand bridge retrofit programs for state highway bridgcs 40

S-2 Total numbers of seismic safety retrofit projects for state highwaybridges in different stages ofcompletion since March 1989 40

Vlll

Page 10: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figures (continued)B-1 Location of hypocemers of the Northridge earthquake and aftershocks 58

B-2 Cross-sectional projections of the locations ofaftershocks for therectangles shown in Figure B-1 58

B-3 Time histories of the b'Tound motions recorded at the Newhall andSanta Monica stations of the California Strong MotionInstrumentation Program 59

B-4 Comparison of recorded peak hori7.0ntal accelerations with thosecalculated using the attenuation relationship of Idriss 11991 J forrock and stiff soil sites 60

B-5 Variations of the ratio of peak vertical accelerations to the averagepe:lk horizontal at,'(,.·e1eration versus the closest distance to thefault rupture surface 61

8-6 Normalized sland:lrd response spectra for selected soil sites aslllC:lsured in the Northridge earthquake 62

C-I LOc:ltion of the 1971 San Fernando earthquake 66

C-2 The most dramatic of all bridge damage caused by the San Fernandoearthqu:lkc occurred at the Golden State Freeway (1-5) andFoothill Freeway (1-210) Interchange 67

C-3 Colulllns of the 5:11\ Fernando RO:ld Overhead suffered heavy damageduring the San Fernando earthquake 68

0-1 Isoseismalmap of the d:llllage impacts of the Lom,l Prieta earthquake 72

0-2 'rhe Cypress Viaduct showing the extent of the collapse during the1989 Loma Prieta earthquake 72

0-3 The west sections of the San Francisco-Oakland Bay Bridge fell duringthe Loilla Prieta earthquake 73

0-4 LocatiOIl of the San Frnncisco freeway viaducts damaged in the1989 Lama Prieta earthquake 73

Page 11: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

This page has been left intentionally blank.

Page 12: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Section 1

Overview

C:lltrans appointed the Seismic AdvisoryBoard in September, 1990, as directed byGovernor George Deukmcjian in ExecutiveOrder 0-86-90, dated June 2, 1990, and inresponse to recommendations contained inCompeting Agni1lJ1 Tim~: Gov~,."or's Bonrd of11Iqll;''Y Rt'ptm 01J the !..omn Pr;ml Eorthquolu.The charge to the Seismic Advisory Boardwas to prO\·jdc cominucd, focused evalua­tions ofCaltr:lns seismic policy and technicalprocedures. Since that time, the cight­member Board has regularly reviewedCaltrans seismic design, retrofit, and hazardmitig:ation activities. The Board also hadnumerous discussions on the Cal transprogf"'JIll with senior staff cll&rinccrs andmade numerous recommendations.

The Northridge earthquake ofJanuary17, 1994, in the Los Angeles rcbrion ofCalifornia, provided an opportunity for theSeismic Advisory Board to cvahmte theperformance of Caltrans bridgcs, retrofitprograms, peer review programs, andtechnical procedures. In rt:sponse to theNorthridgc earthquake this report of theSeismic Advisory Board:

• Evaluates the past four years of changesand dcvelopments in seismic designcriteria and the highway bridge retrofitprogram.

• Summarizes Board findinb"S on thepcrfomlancc of highway bridges in theNorthridge earthquake.

• Recommends improvements to Caltransbridge seismic design and retrofitprograms and procedures.

This section provides an overview of thereport. Section 2 gives the detailcd findingsand recommendations of the Board. Theremaining sections and appendices providethe details that led to thc Board's findingsand recommendations.

Damage to Highway Bridgesin the Northridge Earthquake

Caltrans has approximately 12,000 statehighway bridges in California and is respon­sible for a total of 2,52 3 state and interst<ltchighway bridges in Los Angeles County.Additionally, about 1,500 bridges are main­tained by Los Angeles County and 800 bythe City of Los Angeles, and most of theselatter bridges are small, single span-bridgesand most were remotc from thc area ofStrong ground motion. Only a few of the cit}'and county bridges were significantlydamaged. Sections 3 and 4 provide additionalinfonnation on bridge performance.

The Northridge earthquake ofJanuary17, 1994 (Mw:6. 7; Mw is the momentmagnimde) caused the collapse ofsevenhighway bridge stnJcmres and the conse­quent disruption of a large portion of thenorthwest Los Angeles freeway system.Figure I-I shows the locations of thesebridges in relationship to the earthqmlkesource. Of the seven bridges that collapsed inthe earthquake, five had been scheduled asrequiring retrofit. Two bridges, the Mission& Gothic Undercrossing and Bull CreekCanyon Channel on State Route 118, hadbeen identified as not requiring retrofit. Thecollapsed structures can be classified by\'inrage into three groups: three bridgesdesigned and built before the 1971 SanFernando e:lrthquake (Mw:6.6); two bridgesdl.'Signed before 1971, but constructioncompleted after 1971; and, two bridgesdesigned and built a few years after SanFernando, hut not to current standards.

Many othcr bridges in the stronglyshaken region sust':lined damage, but did notcollapse. The damage ranged from minor

Overview 1

Page 13: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

,,

Los Angeles. CA, ,~,

Pll$adena

i·- ~.1 ..I ." ,,-

11 •. 1! i.

Nl'

Epicenter of theJanuary 17, 1994Northridgeearthquake

PACIFIC OCEAN

S.,I/II MotIk:a Bay

'0 l Ii <-- ,1....... j iJ '" ".-;;...

, D, , ,

'-....n

/ ..'P..:iIic eo.t

*

Flgur. 1-1. Thelocation 0/ bridges thai

collapsed in theNorthridge earthquake

o/January 17. 1994;

two bridges. the northand south connector

overcrossing.collapsed at the

f-5/$R-114Interchange.

2 The Continuing Challenge

Page 14: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

cracking and spalling ofconcrete to mores~ere damage that necessitated closing somebndges to mffic while repairs were made.

The bridges in the regions of shakingthat were constructed or retrofitted tocurrent Caltrans criteria had, at most, minordamage. All remained in service and noneposed a significant safety hazard.

Bridge damage was predictable given theground motion recorded during theNorthridge earthquake. 'rhe older bridgeswerc designed for only a small fraction of theground motion they wcrc subjected to in theNorthridge earthquake, and their damage orcollapse could be expected. The types ofdamage observed in the Northridge earth·quake are, in the main, consistent with thoseobserved on older bridges in the 1989 LomaPrieta (Mw:7.0), 1987 VYhittier Narrows(Mw:5.9), and 1971 San Fernando(Mw:6.6), earthquakes. Appendices C ;llld Dcontain a discussion of the irnpaet~ of the SanFernando and LOllla Prieta cartlH]uakes onbridges.

Retrofitted BridgesPerformed Adequately

/}II structures in the reb';on of strongshaking that were retrofitted since 1989performed adequately, thus demonstratingthe validity of the Caltrans retrofit proce·dures; there were 14 retrofitted bridges inthe region of "cry strong shaking and a totalof60 in the rebrion having l>cak accelerationsofO.15g or greater. The retrofitted structuresresisted the earthquake motions much benerthan the unretrofitted structures. TheBoard's conclusion is that if the sevencollapsed bridges had been retrofitted, theywould have survived the earthquake withlittle damage.

Caltrans Seismic DesignCriteria and Retrofit Program

Calmns has seismic desib'" perfomlancecriteria that set st:J.ndards for twO categoriesof bridge structures-important and com­mon. Table I-I (page 6) reproduces theCalmos seismic performance re(luirements.Important strucrures are those that do nothave convenient alternative routes, whoseeconomic consequences of f..ilure arc large,or that provide secondary life safety or aredesignated as illljX>rtant by local clllergCIl{:Yofficials. Technical evaluations arc made foreach typc for twO levels of earthquakeground motions-the functional and safetylevels.

For the safety level eV:lluation, the Boardintcrprets the performance statement asexplicitly containing the goal that collapse beavoided in earthquakes for ;111 state bridges,whether new or retrofitted. For the function­ality level cV:lluation, Cal trans has adoptedperformance criteria that will allow POSt­c;lrthquake damage inspection and repairwith minimal traffic interruptions.

Since the 1971 San Fernando earth·quake, Cal trans has been engaged in a IlHllti­phase bridge retrofit program. To date mostexpansion joint's have been provided withrestrainers or scat extensions and mostcritical single-column-bent bridges havebeen retrofitted. Prompted by the 1987VYhittier Narrows carthquake and amplifiedby the 1989 Lama Prieta earthquake,Caltrans has accelerated the bridge retrofitpr?gram and initiated significant changes inbndge design criteria.

Sections 5 and 6 provide more detail onBoard suggestions for how the Caltransseismic program can be imllrovcd.

Overview 3

Page 15: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Lack of Progress onToll Bridges

\¥hile there is no doubt that Caltr:mshas made steady progress in implementingthe retrofit program, the Board would like tostreSs that there are no construction projectsunderway in the Spring of 1994 for tollbridges, either in southern or northernCalifornia. The size and complexity of tollbridges makes progress slower, but theirimport:lnce puts a premium on completionbefore they afC damaged in an earthquake.Hazard analyses are complete for all 11 tollbridges, and vulnerability analyses have beencompleted for a few. Preparation of retrofitdesigns has not yet started for most, and noconstruction is expected to start for solllerime. The toll bridge projects need greateremphasis to ensure their timely completion.

Public Concerns andQuestions

The Seismic Advisory Board has idellli·fled four impon:mt questions about theperformance ofbri<lges in the Northridgee,lrthquake from those raised by legisl:ltors,newspaper reporters, and the public. Thequestions and Seismic Advisory Boardan~...vers are:

1Question: Do the results of theNorthridge earthquake indicate that theCalmms seismic retrofit ProbTfam has beeneffective and appropriate?

Answer: Yes, the technical standardsappear sound. All 24 of the retrofittedbridges in the region of intense !.'Toumlmotion (PGiL>O.5g) perfonned well. It is ofsome concern to the Board that two of the

4 The Continuing Challenge

sc\'en collapsed bridges had not been selectedfor retrofitting; the other five already hadbeen scheduled for retrofit. The screeningprbt."CSS used to identify retrofit priorities isevolving and generally sound, but needsimprovement. The damage to older bridgeswas essentially of the same type as observedin California earthquakes during the past 25years. "Inere was no way to know that theNorthridge earthquake could happen beforeother possible earthquakes that could haveoccurred at other seismically active sites.

2 Question: If the bridgl."S that collapsedhad been retrofitted before the earthquake,would they have been protected?

Answer: Yes. Observed performanceindicates that the collapse and major damagcsuffered by highway bridges in this e:mh­quake would have bccn prevented if thebridges had been built or retrofitted tocurrent C:lltrans criteria.

3 Question: Is the retrofit program forState of California bridges proceeding at theright p:lce?

Answer: In part. For single-column­bent bridges, nearly all projects arc cithcrcompleted or under construction. Formultiple-column-bent bridges, retrofitconstruction has been complcted for only:lboUf 7% of the projects. For toll bridgcs,fcw (Iesib'll effons have begun. The numberof bridges that have been retrofitted has beencontrolled primarily by thc availability ofresources. The paee could be ()uickened byresoh'ing the administrative, budgetary,contractual, legal, and personnel constraintsthat slow progress.

Page 16: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

4 Question: Do the results of theNorthridge eanhquake indic;lte thatCaltrans seismic design procedures for newstructures need to be modified?

Answer: No. The Board believes thatthe current design procedures are appropri­atc. This earthquake was not a complete test,since longer duration ground motions can beexpected in future California eanhquakes.

evenheless, the Board believes that if thestrucrures had becn designed to currentstandards they would have sustained littlc orno damagc. The Caltrans procedures areexpected to continuc to be improved as newinfomlation and observcd perfonuancebecome availahle.

ConclusionsThe Seismic Advisory Board concludes

that the overall performance of Calrnmsstrucnlres is consistent with the 1990directives of the Governor and the Legishl­nlre on seismic safety of bridges. TheSeismic Advisory Board has witnessedfundamental changes in Calmllls policy sincethe 1989 Loma Prieta earthquake. Immedi·ate1y following the Loma Prieta earthquake,the Cal trans approach to the replacementand retrofit designs of bridges could becharaeteri7..ed as uncertain at the design level,while management was pushing forw:udrapidly. In the Spring of 1994, following theNorthridge earthquake, both managementand design groups seem to be well-synchro·nized and acting with confidence.

Observations of bridge performance inthe Northridge earthquake lead the Board toconclude that the Cal trans seismic designprocedures for new bridges and its retrofit

procedures for existing hazardous bridges aretechnically sound. The Board finds th;lf theretrofit program is proceeding fairly well, blltthat the screening methods used to identifyhazardous bridges could be improved. Themajor issue where subst:mtial improvementscan be made is in the pace of retrofitting theexisting deficient bridge... and particularly thetoll bridges.

Retrofitting of toll bridges is proceedingat a slow pace,limited by budgetary con·straints, even though their vulnerabilities archigh. It is imperative thai retrofitting of tollbridges be funded and implemented withoutdelay. These bridges are too important to theeconomy of California to be left at risk toeanhquake destruction by their currentvulnerable states.

The findinb"S and recommendations ofthc next section, as well as suggestions inSection 6, provide details on how theCaltrans program can be improved.

Although much has been accomplished,much remains to be done. Earthquakes ofsimilar size to the Northridge earthquake,and even larger, will continue to occur inCalifornia. \'Vith some improvements, theCaltrans program should be continued withdispatch and detennination. The majorforeseeable impediments to a successfulprogram are inadequate or fluctuatingfunding.

The Seismic Advisory Board has confi­dence that the California highway system isprogressing in an orderly fashion to one thatis significantly more seismically safe. TheNorthridge earthquake demonstrates thalCaltrans retrofit effom 10 date have beenresponsive to the seismic haz.1rd and theengineering approach of the Departmcnt ofTransportation is fundamentally sound.

Overview 5

Page 17: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Table 1-1. Caltrans seismic performance critena and definitions for the design and evaluation ofbridges. In the text of this report the term common is used in place of minimum for aI/thosestructures that are designated not importanl by the definition given below. (Department 01Transportation 1994)

Ground Motion at the site

Functional evaluation

Safety evaluation

Minimumperformance level

Immediate service level;repairable damageUmited service level;significant damage

Important bridgeperformance level

Immediate service level;minimal damageImmediate service level;repairable damage

Definitions:Important bridge: (one of more 01 the lollowing items present):

• Bridge required to provide secondary life safety (example: access to an emergencylacility.)

• Time lor restoration 01 functionality after closure creates a major economic impact.• Bridge formally designated as critical by a local emergency ptan.

Functional evaluation ground motion: Probabilistically assessed ground motions thathave a 40% probability of occurring during the usefulliletlme of the bridge. Thedetermination of this event shall be reviewed by a Callrans approved consensus group. Aseparate Functionality Evatuation is required only for Important Bridges. All other bridgesare only required to meet specified design requirements to assure Minimum FunctionalityPerformance level compliance.

Safety evaluation ground motion: Up 10 two methods of defining ground motion maybe used:

• Deterministically assessed ground motions from the maximum earthquake as definedby the Division of Mines and Geology Open·File Report 92·1 (1992).

• ProbabilisticaHy assessed ground motions with a tong relurn period(approximately 1000-2000 years).

For important bridges both melhods shall be given consideration, however, the probabilisticevaluation shall be reviewed by Caltrans approved consensus group. For all other bridges,the motions shalt be based only on the deterministic evaluation. In the future, the role of thetwo methods for other bridges shall be reviewed by a Caltrans approved consensus group.

Immediate service level: Full access to normal traffic available almost immediately.

Repairable damage: Damage that can be repaired with a minimum risk of losingfunctionality.

limited service level: Limited access (reduced lanes, light emergency traffic) possiblewithin days. Full service restoration within months.

Significant damage: A minimum risk of collapse, but damage that would require closurefor repairs.

6 The Continuing Challenge

Page 18: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Section 2

Findings andRecommendations

The Seismic Advisory Board bases thefollowing findings and rccollllllCIHlations onits analysis and review of the NorthridgeearthqU:lkc's impacts on trnnsporr:.ltionstrucmrcs, the Caltrnns retrofit progrnrn,Caltrans response to the rccOlllmcnd;nionscontained in Camper;l1g Agail1st 'Time TheGovtnlor's Bonn! of l11qU;ry Report 011 tbr UnIUIPt-;nn EArtbqllokr, M.II] J990; directionsgiven by Govemor Dcukmejian's ExecutiveOrder D-86-90, dated June Z, 1990; :md tothe requirements arSenate Bills 36X and1104. The basis for these findings andrecommendations t.-an lx: found in thebalance of the report.

The Seismic Advisory Board reCOTll­mends that the imlicatcd actions he under­taken on :I prioriry basis.

Bridge Performance in theNorthridge Earthquake

1 Finding: Caltralls has 12,176 statcbridges :md of thcse 9,206 were designcdprior to the enbrineering impact of the 1971San Fernando earthquake. At this time,knowledge of destructive earthquakes andthe seismic performance of structures was inan undeveloped State so that bridges de­signed prior 10 the San Fernando earth{lUakewcrc not up to current standards of seismicdesign and it W;IS known since the SanFernando, \.vhinier, and Lom:J Prietaeanhqu:Jkes that some of these StruCturescould not survi\·c imense ground shaking.EX;II11ples arc the Nimitz Freew:Jy double­deck viaduct that Coll:lpsed in the 1989 LOlllaPrieta earthquake :llld the bridges thatcollapsed in the Northridge c:Jrthl]H:lke,

2 Finding: Damages observed in theNorthridge e:Jrtlu]uake are, ill the lll;lln,consisrellt with those observed in the. 1989Lama Priem, 1987 Vlhittier Narrows, :lnd1971 San Fernando carthlluakc....

3 Finding: The Northridge earthquakeprovided a v:Jhmble test for Caltrans {k-signprocedures in high-illlensity, modcf'Jtcmagnirude earthquakes, but did not consti­rute a tCSt of their beha\'ior in the larger,long-duration earthquakes that are expectedto occur in the future.

4 .'indillg: Of the seven bridges thatcollapsed, five had heen identified andscheduled for seismic retrofit. '1\\'0, theNlission & Cothic Undererossing and rheBull Creek Canyon Channel Undcrcrossing011 St.ne Route 118, had been evaluated asnot high-risk :111(1 were not scheduled forretrofit.

Recommendation: Caltnlns shoulde\'aIU;He those bridges thaI' were nor includedin the first retrofit group to detenninc if theyrCt]uire retrofining. The evaluation shouldbe performed with thc essential objeeti\·c ofcollapse a\'oid:Jnce in all L':1nhquakcs.

5 Finding: The pcrfonnance of recentlyrctrofitted bridges in the Northridge earth­quake appe~r to be acccpr-Jble. The e\'oh'ingposl'-Loma Prieta earthquake design ;lI1dretrofitting pr;,lCtices used by Clltr:lns appearto be sound, No signific~ntdamage has beenreported to the 60 bridges retrofitted byCtltr,ll1s in the region of strong sh,lking sincethe sl'an of the post-I 987 retrofit progr'llli.Prior ro 19R7, rhc retrofit approach waS 10

Findings and Recommendations 7

Page 19: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

use expansion joint restrainers only. Perfor­mance of joint resrrainers in the orthridgeearthquake was mixed. \¥hile retrofittedbridge performance in this event was accept­able, evaluation of the expected perfonnanceof these bridges in other earthquakes withgreater durations may reveal opportunitiesfor improvement.

RecOlllmendation: A thorough study ofthe performance of bridges in theNorthridge earthquake should be conductedto detennine if changes in Cal trans dcsibtnpractices and priority setting procedures arcneeded. This should be completed throughin-house and independent, external studies,as appropriate. Bridges of both concrete andsteel should be studied.

6 Finding: The public can have confidencein the seismic safety of the Northridgeearthquilke replacement structures becausethey arc being well designed and peerreviewed.

Retrofit Program

7 Finding: Calrrans has made acceptableprogress in implementing the retrofitprogram of single~column-bentbridges, withconstruction either begun or completed on100% of the identified bridges. in addition toretrofitting the single-column-bents, theprogram includes retrofitting the abutmentsand footings as needed. For the multi plc­column-bent bridges, the retrofit programhas been completed for only about 7% of theprojects. It has made slower progress on toll

8 The Continuing Challenge

bridges, where vulnerability smdies arc onlynow being initiated on some, and construc­tion is not underway on any.

. Recommendation: Calrrans shouldidentify the most hn.1rdous highway bridgesin the State and fully retrofit them as quicklyas practical, instead of approaching theretrofit progr-JlllS hy category of structures.

Recommendation: More emphasislllust be given to st;lrting toll bridge retrofitconstruction project... on as rapid a scheduleas practical.

S Finding: The priority setting processused by Caltrans, and as re\'iewed by theSeismic Advisory Board, involves dassil}~ng

StruCtures by vulnerability, seismic hazard,and impact on the community. Each categoryhas scver<ll elements, some of which do notnow appear to be weighted appropriately (forexample, soil conditions at the site and thesystem response of interconnected bridges,such as the sequence of bridges on the SantaMonica Freeway). The present process yieldspriority lists determined by calculations thatdo not take into account all important factorsaffecting seismic safety.

Recommendation: The Calu'3nsprioritizing procedure should be reviewedand modified based on current understand­ing. Attention should be given to the qualityof information uscd in the process, includingthe presence of nonductile columns, variablesoil conditions, and the effect that;l series ofbridges has on the Vlilnerability of a freewayas an interconnected system. Other charac·teristics and their weightings should also bere-examined.

Page 20: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Design

9 Finding: Caltrans design procedureshave twO pcrfonnance categories: importantand common. The perfonnancc objective forimportant bridges is to have full access3V3ilable to norlllal traffic almost immedi­ately following a major earthquake. Theperformance criteria for all common bridgesin a major earthquake arc to avoid collapse,but to allow significant damage and limitedservice. \¥hilc any of three characteristics­secondary safety, economic impact oremergency tlse-can lead to classification as"important," there is some ambiguity in thespecific characteristics that make a bridgeimportant. The public's response to theNorthridge earthquake suggests that morcbridges should be classified as important thanthe current procedure yields.

Rcconuncndation: Caltrans shouldreconsider and broadcn the dcfinition of animponant structurc and the appropriateperfonnancc objectivcs for both importantand common bridge categories. Concur­rently, the acceptance criteria, or limit states,leading to each performance objective shouldalso be defined.

10 Finding: The 'orthridge earthquakeoccurred on a previously unidentified blindthrust fault, a type of fnuh that docs not havea surface tmce. The possibility of blind·thrust earthquakes was well recognized byboth the technical community and Cal trans.The Northridge earthquake producedground motions that were high, but withinthe range considered possible. With few

exceptions, vertical accelerations were notunusually high compared to horizontalaccelerations.

Recommendation: Future seismichazard assessments should consider thelikelihood of blind thrust faults.

11 Finding: The duration of the strongvelocity pulse observed in near field timehistory recordings during the Northridgeearthquake once again affinns its importanceto design. It occurs at sites near faull rup·tures and above thrust faults. The possibilityof a velocity pulsc at a site should be givenconsidcrntion for near field sites in lhedesign of bridges, especially when assessingnon-linear response.

Recommendation: Thc Caltrans bridgedcsign procedures should be assessed, andrevised as required, to detcrmine if theyadequately reflect the structural demandscaused by velocity pulses.

12 Finding: The seismic hazard used inthe dcsibttl of common bridges is based onlyon deterministic evaluations for the maxi­mum earthquakes that can occur throughoutthe state as prepared by the CaliforniaDivision of Mines and Geology (CDMG).There is some debate as to how thcseearthquakes and the faults on which theyoccur should be selected and what attenua­tion relationship should be used to dctcnninethe best estimate of ground Illations at a site.The current map only reflects mean peakground motion estimates; it does not includedumtion effects or velocity pulses, both ofwhich may be important for common bridgedesign.

Findings and Recommendations 9

Page 21: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Recommendation: Caltrans shouldreconsider the technical assumptions leadingto the detenllinistic map and prepare a newone to reflect current understanding of bothseismic ha7..3rd and the way in which thesevalues are used in bridge design.

13 Finding: Caltrans has several hun­dred steel girder bridges in California. Anumber of these in the San Fernando Valley

.area were subjected to strong shaking andsustained severe damage to the end bearingsand to the bearing suppons. None of thesebridges collapsed but at the end of theearthquake they were in a potentiallyhazardous condition.

Recommendation: Caltrnns shouldinvestigate the support systems for steelbrirder bridges and strengthen them asrequired.

14 Finding: Unusual damage wasreported to some steel girder bridges. At thiswriting, twO skew bridges have been identi­fied in the rebrion of strong shaking as havingcracking in girder webs near welded stiffenerplates.

Recommendation: Caltrnns should verycarefully check all steel bridges and clementsin the region of strong shaking to detennineif there has been damage. Bridges omsidethis area throughout the state should bechecked for the possibility of having crackscaused by fatigue.

10 The Conrinuing Challenge

Caltrans ManagementActions

15 Finding: Cal trans has followed thedirections of the Governor based onCompeting Against Time and the directions ofthe Governor's Executive Order. Administra­tively, and in practice, Caltrans is committedto producing seisllliclilly safe rnmsportationstnlcrures.

Recommendation: Cal trans shouldcontinue its commitment to improving theseismic safety of the state's highway bridges.

16 Finding: Peer review of the design ofnew and retrofit bridges has been imple­mented for complex stnlctures. Peer reviewis nor being conducted for the more preva­lent common types.

Recommendation: The scope ofproject'i that are peer reviewed should bcextended to include a few representativeprojects for the more common, prevalenttypes of stnlcntres to validate the design and!or retrofit approach.

17 Finding: There is considerablevariation in how peer review has beenimplemented for different structures.

Recommendation: Peer review shouldbe standardized in tenns of: I) which bridgesare to be scrutini7.edj 2) the scheduling of thereview to allow designcrs time to modify thedesign in response to reviewer commcnts;and, 3) how complete the peer review shouldbc, ranging from the initial stratcb'Y and typeselection to the final seismic design detailing.The specific terms of content and fonnatshould not be standardized-they must beproject-specific.

Page 22: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

18 Finding: Strong motion recordswere obtained from only six bridges located14 to 115 Iniles from the epicenter. None ofthe bridges that collapsed or had substantialdamage were instrumented, thus denying thcopportunity to evaluate the effectiveness ofdesign and analysis procedures by compari­son with actual response.

Recommendation: Both Caltrans andthe California Strong Motion Instrumenta­tion Prof,rram must make a greater commit­ment to installing instruments on bridges,especially toll bridges. Engineers must haverecordinf,rs from bridges and their sitessubjected to high-level ground motions roadvance lhe state-of-the-art in bridge designand analysis.

State Actions

19 Finding: The basic and appliedresearch findings and knowledge that haveallowed the development of improvedseismic design procedures and practices forbridges have come from research on all typesof structure and conditions. The continueddevelopment ofeffective seismic desibrrl andretrofit procedures for bridges will dependon knowledge generated in many areas ofearthquake enbrineering.

Recommendation: Calrrans shouldcontinue its vigorous program of researchand devclollment for bridges.

20 Finding: Budgetary, administrative,legal, and personnel constraints are theprimary reasons why the Caltrans hazardous­bridge retrofit program had not accoll1-

plished as much ;IS desirable prior to theNorthridge earthquake. Tn the past, limita­tions on budget ;Ind personnel were theprincipal drawbacks. Now the issues are: 1)the number of people assigned and their skilllevels; 2) the ability of management toconu!ct with qualified engineers to developdcsigl's; and, 3) the ability to initiatc con­struClion contractS. Calrrans is working nearthe limit of what can be realistically donewith their current personnel levels andprocurement limitations.

Recommendation: If the public wantssafer bridgcs faster than at the current pace,then it will ha\'e to provide greater resources,including both administrative and personnelneeds, and resolve the legislative, legal, andadministrative impediments to implementingretrofit projects quickly.

21 Finding: The twO collapscd bridgeson the Santa Monica Frceway (1-10) wereremoved, new spans were constnlcted, andnormal traffic flow was established byMay 20. This rapid repla.ccment of thedamaged bridges was accomplished by meansof special contractual arrangements thatprovided incentives for completion ahead ofschedule and disincentives for completionbehind schedule. Similar contractualarrangements were made for the completionof all seven collapsed bridges, and Caltr:msplans to have them replaced by the end of1994. This rapid recovery of the heavilytraveled Los Angeles freeway system sets avaluable precedent.

Findings and Recommendations 11

Page 23: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Conclusions

The Board concludes that the publicshould have confidence that new and re­cently retrofitted strucrures arc beingdesigned and constructed based on soundengineering principles that incorporate thelatest research findings and techniaal knowl­edge. The public should have confidence thatCaltr:ms is working diligendy and withdeliberate speed to retrofit hazardousbridges. Retrofit prioritization proceduresneed to be reexamined, possibly leading togreater emphasis on retrofitting importanttoll bridges. On the basis of observations inthe Northridge earthquake, the seismicvulnerability of California's highway struc·tures is significantly decreasing with time.

12 The Continuing Challenge

. Retrofitting of toll bridges is proceedingat a slow pace, limited by budgetary con­straints, even though their vulnerabilities arehigh. It is imperative that retrofitting of tollbridges be funded and implemented withoutdelay. These bridges are too important to theeconomy of California to be left at risk toearthquake destruction by their currentvulnerable St2tes.

The balance of this repon provides thetechnical observations and evidence that theBoard used to reach its findings and fornm­late its recommendations.

Page 24: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Section 3

The NorthridgeEarthquake andBridge Performance

The Northridge EarthquakeThe Northridge earthquake ofJ:mu3'1'

17, 199-J., strongly affcl.:tcd the northern partSof Los Angeles and the Sail Fernando Valley:mel surrounding areas in SOUlhcrn Califor­nia. h was the mOSt cosily single nann"a]disaster in the hislOTV of the United Stales.This m3b'nirudc 6.7 ~:lrthquakc occurred:u4:3 I am local time on a Monda}', and resultedin ahmll 65 l.lc:nhs and over 5,000 injuries.Prclimin:lry damage estimates 3rc ;n themnge 0($1 5-30 billion.

The c:lnh'!Il:lkc occurred in :I highly­populated, urban :lrea. Most affected srmc­rures were buih in rhis century.. The carrh­qU:lkc caused seriolls (!am:tgc and failures incommercial :lnd residential buildings,destruction of the contents of m:my struc­tures, d:llll:1ge 1'(1 critiC:11 ttansp(Irt':1tiOIlsystems, and widespread disruption ofutilities ;llld other lifelincs. Of great puhlicconcern W:1S the coll:lpsc or pani;ll COll;l]JSCof seven bridges of the freew:ly system(Figute I-I). In p:lrt hecause of the time ofoccurrence, nnly one life W:IS lost from theschridgc collapse.;.

The 199-1 J orthridgc earthquake was inan urban :lre:l conmining structures of lllanytypes. It provided :I tirst test for 1ll:ll1ymodern seismic design pt:lcrices. Many ofthese ;lppe:lr to h:n'e heell very successful,hut seJTne now appear 10 be questionable."rhe dalll<lge to steel bridges ,lnd recentlycompleted steel·hr:lced <lllli welded momentframe buildillb'S \\'3S unexpected. Recently­constructed bridges :md post-1987 retrofit­ted reinforced concrete bridges, on the otherh,md appeared to I>crfonn re:lsonably \\"ell.

SeismologicalCharacteristics andGround Motion

The January 17, 199-1, main shock of theNonhri(lge earthqu3kc was generatedbeneath !.he San Fen13nclo Valley ne3rNorthridge at a focal depth uf about 18 kIll.It oc-curred on a blind thrust fault and thusthe principal nlptlltC did not break thesurface. Fib'1lre 3-1 shows the distribution ofNlodified Merolli Intensities (MMI), brivinga sense for the shaken 3re3S. It is of interestthat within intensity VII zone, pockets ofintensity VIIl :lrc lll,lpped south of the Sallt:lMonica Mountains.

Since the 1987 \Vhirrier Narrowse:lrth{IUake, the occurrencc of \)1 ind-t hrllstEmIt earthquakes through the Los Angeles,111<1 San Fernando V:JUe}'s has been widelyaccepted, with the likelihood of earthqu:lkesof magnitudes :lbollt fl.5 gener:lred liy slip onthem. In this sense, the type of faultingwhich produced the Nonhridge e:1I"thquakewas not unexpected. The eX,lCt position ofthe causative Emit, however, W:IS not pre­dicted. The imcnsit)' of shaking :lppe3red to

be system:ltically somewhat higher thanexpected, b:lsed 011 a\'er:lge ;Ittenuatiollcurves for past California earthquakes.

cvertheless, the majority of b"t'Olllldmotions fell within the 8-1% expectationlevels (mean plus one st3ndard devi3tion) andwould thus be :lecommodated b}' prcselllprobabilistic methods of seismic motionassessment. Ap:lrt from a few an01ll3lo11Ssites, contrnl')' to some public impressions,the measured IlC3k vcrtk"al aCI.:derations (3Scompared to the observed hori7.011ta1 \":llues)were also in the expected range of \'31m.'S.

The Northridge Earthquake and Bridge Performance 13

Page 25: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

• Fillmore

-----+-(118

VII

VI

o ro1 ---'1

KM

VIII

Chatsworth •

Tarzana.

Santa Monica

14

VII

VIII

VII

Figure 3·1. lsoseismal map of the Modified Mercalli Intensities for the Northridge earthquake. The locations of principalhighways is indicated for the region of strong shaking, MMI VI is termed moderate and is described by the types of effectsobserved: felt by everyone: many people are frightened, some run outdoors; small objects fall off shelves: pictures fall off walls;plaster cracks: weak masonry buildings Crack. MMI VII is termed strong: weak unreinforced bUildings damaged; unreinforcedmasonry chimneys broken at roof lines; disruption of building contents; plaster cracked. MMI VIII is lermed very strong shaking:damage to nonearthquake-resistant slructures can be significant, with some collapses, particularly those in poor condition;damage to nonSlructuraJ elements in modern, seismically resistant buildings; and substantial disruption of building contents andtoppling aJ.i;.nanchored equipment. MMf IX is termed viOlent: general panic; damage to well-built structures: much interiordamage; frame structures are racked and, if not bolted down, shift off foundations: poor quality unreinforced masonry destroyed;well constructed reinforced masonry is seriously damaged: damage to foundations. The assigned intensity values are based onqualitative observations of damage, not recorded ground motions. (USGS, 1994)

Numerous strong motion instrumentshad been placed by the California StrongMotion Lnstrumentation Program (CSMIP)and the U.S. Gcological Survey. One­hundred and thirty-two instruments within aIOO-mile radius of me fault rupture arearccorded the free-field, strong groundIllations [Shakal et aI, 1994; USGS, 1994J.These records show that:

• Duration of Strong motion was abom9 seconds.

14 The Continuing Challenge

• With few exceptions, peak groundIllotions recorded were within thestatistical ranges expected for such anearthquake.

• Ratios of vertical to horizontal peakground accelerations were typical of pastearthquakes, averaging about 2/3.

Strong motion records wcre obtainedfrom six bridges at distances ranbring from14 to 115 miles. The most significant of

Page 26: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

these was the record from the 1-10/1-405Interchange, a curved concrete box girderstrucmre, 1,037 feet long having nine single­column bents and twO open-seated abut­menrs. The bridge was retrofitted in 1991with steel jackers on some columns. Installa­tion of instmmenrs was completed, fundedby Caltrans, JUSt before the earthquake. Apeak acceleration of 1.83g was recorded atthe box girder near the west abuol1ent. Thisbridge is located about 4 miles west of thesection of the 1-10 Freeway that collapsed.

Appendix B provides more deuil on theseismological and strong motion characteris­tics of the Northridge earthquake.

Calt.ans Seismic Risk MapThe Caltrans seismic criteria use two

different methods for defining dle SafetyEvaluation ground motion (see -r:1.ble 1-1):

• Deterministically assessed groundlIlotions for the maximum earthquake asdefined by the Division of Mines andGeology Open-File Report 92-1(CDMG 1992).

• Probabili~lly assessed ground mo­tions with a long return period (approxi­mately 1000-2000 years).

For import:l.nt bridges both methods arcgiven consideration. However, the probabi­listic evaluation is expected to be reviewed bya Dltrans approved consensus group.

For common bridges, most bridges inthe st:lte system, the motions are based onlyon detenninistic evaluations. Figure 3-2shows the Los Angeles region portion of thisdetennininic CDMG prepared map. Bothpeak f,Tfound acceleration and locations ofspecific faulrs and their maximum credibleearthquakes are indicated. The map shows

that the epicentral area for the Northridgeearthquake had bridge design values of theorder of .6g and .5g, higher than the Agvalue used in the building codes for theextended Los Angcles region. Comparison ofthese numbers for bridges and buildingsshould be done carefully. Compared to typicalbuildings, bridges have relatively small safetyfactors. Buildings have substantial redundan­cies in load paths, so that if one load path isseriously damaged, others can carry theseismic and gravity load. They also have manynonstruetural systems, that, while not in­cluded as part of the seismic resistance system,do participate in the seismic response provid­ing added capacity. Bridges tend to have verylow redundancies, since there are a limitednumber of columns (often only one) and nonon-structural elements to participate in theseismic reslx>nsc. Therefore, even thoughbridges may use a highcr seismic groundmotion they may not yield better behaviorthan a building dcsif,'1lcd to a lesser value.

The CDMG peak rock accelerationmap, Figure 3-2, reflects the mean (bestestimate) ground motions from a selectedgroup of known faulrs. There is some debateas to how these faults should be selected andwhat attenuation relationship should be usedto detennine the best estimate rock motionsat a site. The map only reflects mean peakrock motion estimates; it does not includeduration effectS or velocity pulses, both ofwhich may be important for common bridgedesign. The Board urges that: I) special carebe given in the use of this map; and2) Cal trans should reconsider the technicalassumptions leading to the detenninistic mapand prepare a new one to reflect currentunderstanding of both seismic hazard undthe way in which these values are used inbridge design.

The Northridge Earthquake and Bridge Performance 15

Page 27: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figure 3·2. Caltrans peak rock accelera/ion map showing the seismic mean-acceleration valuesused for bridge design. Both peak rock accelera/ion and locations of specific faults and theirmaximum credible earthquakes are mdicated. (CDMG and Caltrans. 1992)

·50

\ .49, , -

~ .5g,,,

-, ,-,-

---

,

- - 50---

405

*--

Malibu Coast-santa Monica-Raymond FZ (7.5)

\ 118fZ.~1J)1

------- - ... - - - - -"- .-·60

* Epica'ller of NoJ1hridge earthquake

Faull with maximum credible eanhquake magnitude

---------- - - - - ------------·40

, ---------------------.0..4g .... --- ...

, ---

Highway

Peak rock acceleration contour

---

.. -------­- - -"

san ,C3)'e\3l'lO ,H7.S) I,

OOwtdge f {7.5) sa: \- - -.~'Tfta.~~

6g_ - - - - "'... -~-nna TIv.- --- ",so"- - - '·60 r"Jf'~........J

...

Summary of Damage toHighway Bridges in theNorthridge Earthquake

Calrnms is responsible for:l total of2,523 state or interstate highway bridges inLos Angeles County. Addition;llly, about1,500 bridges are m;lintained by Los AngelesCounty and 800 by the City of Los Angeles.Only a few of the City and Count)' bridgeswere significantly damaged. Most of these

latter bridges arc small, single span bridgesand most were remote from the area ofstrong ground motion.

The Northridge earthquake ofjal1\l:lry17, 1994 caused the collapse of scvcn high~

way bridge structurcs (Figure I-I) and thedisruption of:1 large portion of the northwestLos Angeles freeway system. The collapsedstructures CAll he classified by vintage illlothree groups: brid,b'CS designed and builtbefore the 1971, San Fernando cartht]uakc;

16 The Continuing Challenge

Page 28: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Table 3·1. Summary 01 highway bridge collapses in the Northridge earthquake.

Bridge Construction Restrainer Probable causeBridge Route number Design completion retrofit of collapse

Gavin Canyon '·5 53-1797 PIL 1964 1965 1974 Skew !;jeometry andUndercrossing unseating of expansion

joints

N. Connector SR-14/1-5 53-1964 F 1968 1974 1974 Short column brittleOvercrossing shear failure

S. Connector SA-14/1-5 53-1960 F 1968 1974 1974 Short column shearOvercrosslng failure

Mission & Gothic SA-118 53-2205 1973 1976 Flexure/shear failure inUndercrossing architectural flared

columns at bottomof flare

Bull Creek SA-118 53-2206 1973 1976 Flexure/shear failure inCanyon Channel shortened columns byUndercrossing channel wall and low

transverse reinforcementratio

Fairfax & Washington 1-10 53--1580 1962 1964 1974 Flexure/shear failure ofUndercrossing short and stiff columns

La Cienega & Venice 1-10 53--1609 1962 1964 1978 Brittle shear failure of stiffUndercrossing columns

bridges designed before 1971 but con­structed shortly aftcr 1971; and, bridb'CSdesigned and built a few years after SanFernando, but not to current standards.

Bridge damage was predictable given theground motions recorded during thcNorthridge earthquake. The oldcr bridgcswere designed for only ,I small fraction of thcground motions thcy were subjcctcd to inthis earthquake, and their damage or collapsewas incvitablc. The l11any hri(lges in theregions of strong shaking that were con­stmcted or retrofitted to current C:lltr,lIlscriteri,\ had, at most, minor damagc; and allremained in service and none posed ~lll

increased safety thrcat during the earth­quake.

Information published by Caltransidcntifies earthquake damage to StateHighway bridges in the Los Angeles Countyas follows:

I. Initial Assessment DatedJanuary 21,1994: Sibrnific:mt bridge damageoccurred within an area of about 270square miles, as shown in Figure 3-1. Atotal of 506 Cal trans bridges are locatedwithin this area. The reported damagewas:

Collartsed or partl}' collapsed 7Major damagc 4Moderate damage 2Minor damage 18

No damage was rCl>orted to post 1987retrofit or new construction. Damageoccurred primarily in older strucnJresdesigned prior to the 1971 San Fernandoearthquake that were not retrofitted orhad onl}' partial or inadcquate retrofits.An exception to the above is thc partialcollapse of two 1976-vintage bridges Oil

State Highway 118 in the epicentr:llarea.

2. Detailed Assessment Dated Febn'ary9, 1994: This later, and more det:liled,assessment of State Highway uridgedam:lge by Caltnllls lists the following:

Collapsed or partly coll:lpsed 7Major damage 39Other damage requiring repair 194Hinges requiring repair

or rcplacement 46

Table 3-1 lists the seven major bridgesthat collapsed during the Northridgeearthquakc, along with the date of theirdesign and construction and the probablecause of failure. All seven were constructedto design standards that were much lessstringent than those Caltrans currently uses.Section 4 discusses in detail how the bridgesperfonned and current thinking as to whythey failed.

The Northridge Earthquake and Bridge Performance 17

Page 29: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Table 3·2. Comparison of the average daily traffic volumes on the damaged highwaysbefore the Northridge earthquake with the corresponding daily traffic volumes onFebruary 4, 1994. The percentage is the ratio of post-earthquake dally traffic volume topre-earthquake traffic volume.

No. of Pre-EORoute location Lanes Normal Jan. Avg. 2/4/94 Percent

5 South of Ale 170 8 151,000 156,880 149,663 95%

10 East of Rte 405 8 267,000 267,273 113,029 42%

101 West of Ate 405 10 275,000 309,049 267,371 87%

105 East of Rte 405 8 NfA 171,135 186,234 109%

118 West of Rte 405 8 139,000 125,279 48,532 39%

134 East of 101/170 Ie 8 200,000 197,973 264,909 134%

170 North of Rte 101-SE only 4 177,000 78,058 76,143 98%

405 North of Rte 10 10 274,000 271,940 234,834 88%

405 South of Rte 10 10 316,000 321,694 298,851 93%

Many Olher bridges in thc stronglyshaken region susf;.lined dam:lge. but did notcollapsc and rcmained in service, either flll1or limited. The damage r:mged from minorcf<lcking :md spalling of concrete t:o moresevere dam:lge that necessitated dosing thebridge to trnffic while rep:lirs were made.

Impact on Traffic Flow inLos Angeles County

InUllcdiatcly following the orthridgcearthquake, Caltrans moved quickly tomobili7.e construction equipment and

Figure 3-3. Bridgepiers with completely

disintegrated core,ruptured hoop

reinforcement, andbuckled longitudinal

reinforcement.(photo: F. Seible)

18 The Continuing Challenge

personnclto relllO\'e debris and restore orrerOute traffic wherc d:lInagc had occurred tothe highway systcm.

Thc failure ofthc hridges listc(l in'Elhle3·1 caused substantial reroming of tT:lffic.'Elblc 3-2 identifies tile &llll:lged st:ltChighways in the county and their aVel"<lgc(!<lily traffic volumes before the c:lrthllu:lkeand on Fehnlary 4, IBd:lyS aftcr the c:lrth­quake. Tr-dffic records for cach of 10 dayspreceding February .... show that the an:r:lgcdelay on each route decreased as altemateroutes werc opened and drivers beClllleaccustomed to changcd highway conditions.As of February 4, 1994, the delays rangc(!(rom 2 to 25 minutes. many times less th:1Ilthe initial delay times, which had heen asmuch as 2 hours. These tr:lVel time reduc­tions indicate th:lt, while it lll:ly rC(luil'cconsiderable time before the culhpsedbridges are repbccd, C:llmms has estahlishedeffective detours. and, except for StateRoutes 10 and 118. fI";lffic flow was essen­tially restored to normal \'olumcs within afew weeks.

Seismic Advisory Board FieldTrip to Damaged Structures

On January 22. 1994, the SeismicAdvisory Board visited the four m:ljor bridgecollapse sites: the 1-10 Santa Monic:l Free­W:ly, the SR-118 Simi V:lllcy-S:lll FernandoFreeway, the 1-5/SR-14 Antelope V:lllcyInterchange. :md lhe 1-5 Gavin Can)'onUnden.:rossing. Numerous other trips weremade by individual Board members.

Page 30: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figure 3-5. Flexural cracks at lhe columnends in the adjacent bent al Fairfax &Washington on 1-10. (photo: F. Selble)

Figure 3-4. Sleeply inclined shear cracksand cover spalling m one bent al Fairfax &Washmgton on /·10 (phOlO F. Selble)

At all bridgc colbpsc SiICS, (i\·c (!:Irs aCrerthe c;lrtI1l1u:lke. rCI110\':l1 of thc coll:lpscdbridge struCllIrcs W:1S either in progn;ss or:llre:J(ly complelc{l where rO:1dway access wasncccssary. Ilowcver, (1:J111:lge Jl:lrterns andcritiC:11 6illire 1110des were sritl visihle inadjacent llridge .~eclions.

At the 1-10 Ll Cieneg;1 am! Vcnice site,bridge piers \\·ith completely disinfCgr:lte(1core, f1111tllrC(1 hoop reinforcemcnt, andbuckled long-itudiml reinforcelllellt suggestthe explosive ur hrinle f:lilurc mode of someohllese columns (Figurc 3-3). Section 4providcs more inlilrlll:uion on the collapse ofthis :lnd olher m:ljor hridgcs discu~sc(! helcm'.

At F:lirf:lx & ""ashinglOll on 1-10,stccply inclined shear cracks :111(\ cO\'erconcretc sp:ll1ing (Figure 3-4), :mcl flexur.l!cr.lcks:1t thc column ends in the adj:lcctllhClll (Figure 3-5). dearly show thc influcnceof column height ami hounclary conditions,with the shorter l..'olunms fh:ed :H both endsshowing the critical shear fililure patterns.Columns retrolitted \\ ith circular steeljackets (Figure 3-ti), al the C:l<!illac Avcnueoff r.llllp din.'Ctly :lCljacelll In \\'ashinglon andFairfax sho\\ cd onlr minor cracking and

Figu,e 3-6. Columnsretrofifled with circularsteel jackets at theCadillac Avenue offrampdirectly adjacent toWashington and Fairfax.These columns survivedwithoul problems thesame groundacceleratIons thatcaused the Washingtonand Fairfax coJfapse(Figures 3-4 and 3-5).(photo: F. Seible)

The Northridge Earthquake and Bridge Performance 19

Page 31: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figure 3·7. Flexure/shear failure in the

bot/om of the flaredcolumns sll/l visible in

the westbound bridge.even though

demol,,1OfI was in lullprogress at the

SR· I 18 Mission &Go/hie Undercrossmg.

(photo: F Seible)

-.--,

'--"'"

Figure 3·8.Completed demohtton

of lhe soulhboundconnector al IheAntelope Valley

/·5!SR· 14 Interchangefive days afler Ihe

earrhquake.(photo: F, Selble)

spalling in the joint between the steel jacketand the bott'ol1l soffit. These retrollue(lcolumns sU,",'i\·cd, without signs of d<llll:lgC,

the S:lllle ground al'celcrations that causedthe \ V:lshington and Fairh\" columns to

coll:lpse.At the Simi Valley-San Fern,lndo

SR-II fl Freeway, the ne.\"IIre/she:lr failuremode in the flared columns at the bottom ofthe flares W:IS still visible ill the wlostboundhridge, e\'en though demolition was in fullprogn:ss at the SR-118 Mission & GothicUndcrcrossing for the eastbound strul-ture,see Fib'1lfC 3-7.

20 The Contmumg Challenge

At the Antelope V,llley 1-5/SR-14Interchange, dcmolition of the colhlpsedportions of the south ;Ind north conncctorsW:IS alrcady completed (Figure 3-8), butahurmenr :lI1d expansion joint d:1I11:1ge in thesep'II-.ltion structures provided a graphicreminder of the cncountered lorce :lnd(lispbccmcnr Icwls (Figures 3-9 :lllU 3-10).

Finally.:Jt 1-5 Ga\·in Cmyon, thecolbpsed section of the bridge structufC hadhcen rClllm·cd, hur rhe highly skewedgcolllelry :lIld narrow joint Sl':lt width wascvidcnt from the still-st,mding ccntcr fr:lI11e(Figure 3-11)_

Page 32: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figure 3-9. Abu/menfjoint damage in fhesoparation structuresat the Antelope Valley1-5ISR-14 Interchange,(photO: F Seible)

Figure 3·10. Expansion Join/ damage in theseparaflon structures a/tile Antelope Valfey/·5!SR·14 /Ilferchange. (photO: F. Seiblej

Figure 3·11. The highly skewed geometryand short expansion joint seat width of the still­standing center frame of the /-5 Gavin Canyonbridge. (photo: F Seible)

The Northridge Earthquake and Bridge Performance 21

Page 33: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

~ Weld

Steel Girder Bridge

Figure 3-12. Typicalcracking associated

with x-bracingobserved in some steel

girder bridges.Norlhridge eanhquake.

SliltenerPlale

Similarity of NorthridgeEarthquake to Loma Prietaand San FemandoEarthquakes

Immediately after each damagingCalifornia earthquake since 1971 Cal transhas SCnt Post-Earthquake InvestigationTeams (PEQIJ) to survey, investigate, anddocument the damage and collapse of allCaltrans structures in the affected areas. Adetailed documentation of bridge damagecan be found in the PEQJT report publishedafter each of these earthquakes, as well asmany other publications.

There is great similarity in the typcs ofdamage produced by this earthquake andthose seen following the 1971 San Fernandoand the 1989 Loma Prieta earthquakes.Bridges that were damaged in all three ofthese events were similar in age and technicalcharacteristics of design. An examinationof the PEQIT reportS highlights thesesimibrities.

The desibTfl profession has known forsome time of the dcficiencies of the olderbridges that havc been damaged in earth­quakes since 1971. Indeed, the currentCaltrans retrofit program is directed at fixingall of these deficient types of strucrures.Thespecifics of bridge response in the S,mFernando and Lorna Prieta earthquakesare discussed in Appendices C and 0 forcomparison.

Performance of New andRetrofitted Bridges in theNorthridge Earthquake

No significant damage has been re!)()rtedto bridges constructed or retrofitted sincethe 1987 retrofit program beb....n. Two

bridges on State Highway 118 (l\1.ission­Gothic and Bull Creek Canyon) that sufferedpartial collapse during the Northridgeearthquake were designed in 1973, but withcriteria that did not reflect current perfor­mance criteria. These bridges were on thethrust block over the fault slip and wcresubjected to very strong ground motion.

Most of the remaining bridges thatexperienced total or partial collapse, as wellas other bridb'CS with major damage, wereconstructed before 1971 and had nor beenretrofitted. Several of the bridges thatsuffered major damage (i.e., State Route 14/1-5 Interchange) had been previouslydamaged but were completed with essentiallythe same desibTfl and using the original picrsand footinb"S; only one replacement picr andone new pier were designed and constructedwith spiral reinforcement.

Loss of support at superstructurc hingeswas a prevalent cause of distress in the 1971earthquake. The hinge restrainers and seatextenders installed after that earthquakeapparently functioned satisfactorily duringthe Northridge earthquake. Althoughdamage to hinge restrainers, diaphrabTfl1anchorages, or bearing seats occurred at 46separate locations, partial superstructurecollapse due to failure of the rcstrainersoccurred only on Route 1-5 over GavinCanyon, wh~re the highly skewed joints andineffective placement of the rcstrainerscontributed to the loss of support and partialcollapse.

The above observations, as well asdetailed performance assessments (Priestley,Seible and Vang, 1994) indicate that collapseand major damage to bridges by this earth­quake could have been precluded if thebridges had been built or retrofitted tocurrent Caltrans criteria.

22 The Continuing ChafJenge

Page 34: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figure 3-13. Cracks In bot/om region of plategirder web, northern bridge of Pico-LyonsOvercrossmg (photo: A Astaneh-Asl)

Performance of SteelStructures

<::.Iltr:.lI1:. ha~ :.c\'cr..11 hundred SIt:e1 girdcrhridg-cs in the SI:l\e Ilighw:t), Sy<;telll. Ap­proximately 125 "'leel girder hridg{.~ :In:Ineltcd in lhc Cre:1ter Lus Angeles arc:! :111<1SlllllC of these II cre in the strlmgly sh:lkenrcgi'ln III Iring the Ntlnhritlgc c:lrthqll:lkc.These hridge:. h:1Il :til been designed prior tnthe S:lIl FCl"I1:1ndo c:H·thtlll:tke :11111 :IS:I

conselillell(:e of the intense gnlund :.hakingth:llthey expericnced, some sust:linCII signifi­CUlt d:llllage to tht: c11<1 hC;lrin!-", :md t'll thestnlCllIre suppol1ing the he;lrill!,~. Nunc ofthe d;ullagcll hrillb'C:' colbpscd bUI some werein :1 potcnti;!lIy h:tl~'ll"(lolls conditioll :11 lhe endof the {'":lrthqu;lke, It i:. dC;lr th;ll Ihe \\ e;lk­nl"S:-.cs \lcrC:l {.'(m:.cllueIKc of unJcrcstim:lIingthe 1II:tf.,'11ilUdc of till: seismic forces that couldbe {le\'ClopctI during:lll t:'1rth1Iuakc,

In ;l{ldition to the d:un;lge to the bc.lr­ill!,"!'> :md lhe supports, scvcrnl :-tecl girdcrhridge., su't:Jined :1 diAi.:rcnt kind uf d:tlllageto the \\ch-phlte uf:l !!irJer near stiffeners(Figure 3-11). '1\\'0 skcw bridges in theregion of \'e~' ... trullg- "haking h:l\'e heellidentified;1\ having- such cr.lcks. [t ;lppC:lrsth:lt ~(Jme cr;lcks wcrc initi:ttcd dllring- thee:lrthqu:!kc, :Jnd ot-hcrs wcrc thcrc before the

Figure 3-14. Roller and seismic restrainer damage, sou/hem bridgeof Pico-Lyons Overcrossing (photo: A Astaneh-Asl)

Figure 3·15. Support damage al top of PierNo.6. santa Clara River Bridge_(photo: A ASlaneh-Asl)

l.';lnhcju:lke hUI \1 en.: extendcd during theshaking. The \IOrst cr':lcking h:ls :tlre:tdy hecnrep:llred by grinding nut rhc cr..Jd~s andrewelding, I);Ul1:1ge to the stcel girderhridges is illustr:.ltcd in Figures 3-13 through3-15 ;lIul described in :1 rcpon to C;lltr..msprcpared hy A. Asr:meh·Asl and others(Astaneh, Ic)<)4).

The Northridge Earthquake and Bridge Performance 23

Page 35: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

This page has been left intentionally blank.

Page 36: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Section 4

Why Did Highway BridgesCollapse in the NorthridgeEarthquake?

Seven highw;ly hridge struCtl.lfCS col­lapsed during the Northridge c;lrtlHluakc(see ~rilble 3-1 in Section 3 for 10C;ltiollS).These seven can he classified by vintage intothree groups:

t. Bridges designed ;Ind built before the1971 $,111 Fcrn:lndo carthqtl;lke (GavinC;\l1yoll Undcrcrossing on 1-5, theF'lirfax & \,V,lshingron Undcrcrossing on1-10, and the La Cicncga :1m] VeniceUndcn:rossing 011 the [-10 S,mt,! IVloniClFreew;lY)·

2. Bridges designed before 1971 hmconstruction completed aft(;[ 1971 (1- 5/SR-14 Anrelope V;llley Interchange).

3. Bridges designed ,lllt! built after 1971,bur before basic design concepts weredl<lllgcd in 1974 (SR- [ 18 Simi Vallcy­San Fcrn<1ndo Freeway;1t I'l'lission &emhic and Bull Creek ClIlYOIl).

'T"his set·tio]] presents prelimin:lryfindings on why these bridges performed asthey did. Later research ;md evaluationeffort's shoul(l eX;lmine hridges th:n did notfail, so that the differences in perfonn:H1cecan be understood :md their implic:nions fordesign coditied.

Bridges Designed and BuiltBefore the 1971San Fernando Earthquake

About 75% afthe Cllifornia's highw;lyhridges were built before 1<)73, the datewhen modern seismic IJridge design practiceswere introduced that incorpor;1fed thelessons of bridge performance in the 5:1nFern:1I1do earthqu:1ke. Of the 12,176 statebridges, 9,206, or 76%, were designedbefore [<)73.

Three bridge structures that weree01npleted hefore, :1I1d survived, the 19715:111 Fernando earrhquake were the GavinC;myon Undercrossing and the F'1irf.ix &\Vashington Undercrossing, and the L:lCienega & Venice Unden:rossing on the 1-105ant:1 Monica Freew:ly. In the 1960s, whenthese hridge struentres wcrc designed andbuilt, earthquakc cngincering in general wasnot very adv:mced. 'T'his is renected inseismic bridge design guidelines of thatperiod th:lt required clastic bteral loaddesigns typically for only 6 percent of gravity,widlOut :lIly provisions or considerations forinchlstic structural response and ductiledesign det:liling. Key characteristics inherelltin these pre-197 1 bridge strucntres arc

I. Low transverse reinforcement ratios incolumns, typiC:11ly #4@ 12" (1/2"&lInercr reinforcing bars at 12-inchcenters) were provided nominally,without consideration of column si ....e orstrength.

2. Short se;lt width at abutments :lndS\lperstrllewre expansion joints due to

underesril11ated displaccment demalHls,as ;1 result of the low 1:1teral design loads.

The collapse :It the 1-5 Gavin CanyonUndercrossing (Figure 4-1) e;ln be :lttributedto geumctrie complexities arising from the66 degree skew orient:llion of ahutlllents andin-span cxpansion joints, ,1S well ;1$ the 8-inchse;lt width. Even though retrofitted with firstgeneration restrainers in 1974 (:IS .1 directresult" of the 1971 San Fern:l1ldo e'1rth<IIl'lkc),restrainer design and detailing were notsuflieiellt to ptevent L111se:lting :Ind subse­quent sllperstructllre failure in the acutecorners of the bridge at the in-sp:mexpansion joints.

Why Did Highway Bridges Collapse in the Northridge Earthquake? 25

Page 37: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figure 4-1. /·5. Gavin Canyon Undercrossmg. Northridge earthquake.(photo: F. Seible)

Since the restr.lIner retrofit in 1974,conrinued rcsc,lrch and devc!ollinent ofn:str.liner units funded by Clltr.I1lS [Sdn:l,Nhil\';1r, :\Ild Zelinski, I989j, ;\S well ,IS:I benerundefSt<lllding of the skcw geometry problcmsunder seismic 1o.lds [Pricsdcy, Seible andChai. 19921. would resulr now in significantlybrger 5(::\1 width extensions ;lnt! restrainercapaciries. Ilowc..-er, qumnit:lti\'e f:lilurc3SSCSSlllenL'i illllnedi:trdy following theNorthridge e:lnhquake [Priestley, Scihlc andU:lIlg. 19941 :11S<) indicate lhat shl.';.tr Elilurc inthe shoneI' benL":1t [fle Grwin ClllyonUndercrossing W:IS imminent, h,l(l unsc:Hing,llll! superstructure Elilme not occurred.

Pre-1971 tr:\llsverse rcinforccmentdetailing in columns W,IS :llso :\ prohlem (lueto the low tr:IIlS\'crsc reinforcement (nominal#-I-@ 12") for Ihe two hridge wlbpses:1I Ihe1-10 Sanr:1 1\lonicJ Frecway. As st:llcd :lbO\'e,the st:lte-of-the-:lrt in bridge dcsi~n :11 therime pro\'idetl tr:ms\'ersc column reinforce­ment only nomin,llIy, :md nOl :IS a rL"Sult ofengineered L':1pacit}' requirettlenL,> that wouldtoday result in tr.ms\'crse reinforcementr.\tios exceeding the nOlnin:III~'-llnl\'idedonesby a f:lclOr of R 10 10 or morc. AS;1 consc­quence of the low tr:lllsverse reinforcementalld underest'im,ltcd /lcxlIl":11 over-strength,the 1-10 Slructures ,It F,lir(;lx &. \.V'lshinglOn(Figure 4-2), ;1I\d at L:J Ciencg:.1 :lnd Venice(Figtlrc 4--3), coll:Jpscd with column sheMf:lilures either hefore or shorrly after theirinitial /lcxur:11 yielding. QU:lntilati\,e failure

26 The Continuing Challenge

Figure 4-2. Shear distress pat/ern a/I-I0.Fairfax & Washington. Nor/hridge earthquake.(photo: F. Seible)

a5.<;t:ssments of these structures (Priestley,Sci hIe and Uan!!, 19941 showe(1 th:1t thefailure modes encountered could have l>cellpre\'ented with :I\'ailallle culumn retrofitj:leketing- techI10101,')'.

Rctrofit designs tnr hoth of the coll:lpscdl~ 1() stTucturcs were CIHllp1cte ,II Ihe lillIe 1)1'

Ihc Northri(lgc carthquake. Ilowe\'er, retrofitirnplemcnt,ltion rq)(lrtedly was cornpoulHledlIy 1eg:t1 problems conccrning the leased:lirsp:lCc under the I,a Cieneg'.l :l11d \ cniL"(.'stnJeturcs. recmphasizing the faci th:1t time isofessence in the seismic rctrofit progr:nn.

Page 38: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Bridges Designed Before1971 but ConstructionCompleted After 1971

The second group of bridge structllrcs,the 1-5/SR-14Antciopc V;lllcy inrcrch,\l1ge,were designed to prc-1971 design sl";mdards,hur wen: cornplcn.:d in 1974. -rhis suggeststhat lessons !c,lrncd from the 1lJ7! 5,111Fernando eanh(lUake should h,1\'C beenirnplclTlcnrccl in rile redesign.

A common misconception following theNorthridge earthquake W,15 rh,][ the salTle

bridge stnu.:rurcs tll,]t collapsed during the1971 San FCrn,\1Hlo carthqu'lkc colbpscd,lg,lin this time. Pre- and post-c,lrrhqu,lkcacti;]] photographs lIcnnings, 1971], dearlyshow that:

I. "I"he 1971 bridge colhlpse was in adifferent sCp:lI",ltion strllCf\lrc, namelythe 1-5/SR-14 south scpaf,uioll :mdovcrhc:ld.

2. During the 1971 San Fern:lndo event,the two bridge sections th~H collapsed III

the Northridge e;lrt'hquake were underconstnll.:tion (S R- 14/1-5 south connecttlr

Figure 4-4. SR-14/1-5 North Connector Overcrossing, Northridgeearthquake, (photo: F. Seible)

overcrossing Sp,IIlS 1 :lI1d 2 had rhebottom soffit, webs and cap beams cast),or wcre almost completc (rhe SR-141l-5North Conneetllr Overerossing wascomplete, with the exception of OIlCcolumn ~111(1 the last rwo Sp,lIlS;H thenorth end).

.L All columns in the Antelope ValleyInterchange, except for the onc mcn­tioned above in 2, wcre completed ;1Ilt!

featured the nominal pre-I 971 #4 @ 12"transverse colllmn reinforcement.

Bec,l\Ise unseating ;It exp;lIlsion joints intall bridges with single-colulllll bents wasidentified as the prim,lry reason for collapseof bridge structures in the 1971 SanFernando earthquake, the decision;1t thetime must h,\ve been to cOllllllete the int"er­change with the already built substructures,hur with added c.xpansion joint restrainers to

tic thc superstructure together and prevent"unscating ,It exp:ltlsion joints.

During the Northridge eartIHI\l:lke, thetwo hridge f,\ilures <1lthe 1-5/SR-14 lnter­dl.mgc (Figures 4-4 through 4-0) cm be:mrihuted to brittle shear f,lilure of short orstiff eolullins. These columns, proportiol1~llly

to their stiffnesses, atrract"ed more seismicforce than their more flexible adjacent bentsami did 110t h,\ve the neceSS<lry defonnat"ioncap,lcity due to rhe pre-1971 tr:msvC\"screinforcell\ent det,liling IPriestley, Seihle ;lndU;llIg, 19941. In addition, the shear failure orthe llonh COlll1cC!'or ofhent #2 w<\s aided hy

Why Did Highway Bridges Collapse in the Northridge Earthquake? 27

Page 39: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

,-~

Figure 4·5. SR-14/1-5 South Connector Overhead. Northridge earthquake (photo: F. Seible)

Figure 4-6. Short column failure SR- 14/1-5 South Connector Overhead, Northridge earthquake.(photo: F. Selble)

28 The Conrmuing Challenge

Page 40: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figure 4-7. SR-l 18 Mission & GothIcUndercrossmg. Northridge earthquake. (phoIO:F Seiblej

an effective shonening of the COIUlIlll by theconstnlctinn uf a rnlck ramp shoulder striparound lhe (:UIUIIlII with l"()Jnpacted aggre­g:Jte alld ;lsplulticollCrCtC o\'crhly. 'I'hus.both structurcS:lI the 1-5/SR- H Intcrchangecullapsed hy hrittle she:!r f;lilllre in the shurtcolumns next to thc ;lbutlllCI1tS with subse­quent superstrucmre unseating:ll the;\l)uttllent ;lml superstructure flexural fallurc;11 the adjaccnl bern ;\s a direct consequenceof the shon eCllullln col1:lpse.

Bridges Designed and BuiltAfter 1971, but Before BasicDesign Concepts WereChanged in 1973

Finall)'. lhe two hridgl-'S thaI bilcd on lheSR-118 Simi \~1Ik'Y-S:m Fernando Freewa}' at1\1iS!>;on & Gothic ami Bull Cn..-ck C~nyon wcredC"Jrly post-197I vllll:lbTC in both design andCOnSmK1.101l. Both hridb'C f.lilllrt'S during theNonhridge l-':Jrthtlu3ke can he amilffitl-xl to:

I. Signifk'":ll1tly increased stiffncss duc todecrease in length or effeeti\'c col Ullln­shortening by heav}' column fl:m.:s at.\Iission & Gothic (Figure -1--7), and by:\channel w:III :11 Hull Creek Canyon dl;ll\\';IS huilt inn.:gral with the columns :11nnga benl Ii lie (Fig-ure 4~8).

-Figure 4-8. Bull Creek Canyon Channel Undercrossing, SR-/18.Northridge earthquake. (photo: T. Saido)

2. Higher shear demands due to flexuralo\'er strength; and,

3. DCbrr:tding shcar C'.lpaeities underindastit· '--relic loading 3t high ductilitydemand.

At the J\Iissioll & Gothic Undercrossing,the column flares. extending o\'er half thecolU111n heighl, wcrc 1110derately reinforced:Jlong rhe fl:ll'e. Thesc moderarely reinforcedHares more th;1I1 doubled the flexural capac­ity oft-he colulllll top in the flared colul11lldirecrion. Tr:lllsvcrse reinforcement, in rheform of;1 smooth #5 spiral with 3.5" pitch,was provided ;llong the eillire circularcolumn core, :md flexural plasric hinging wasforced to the hutrulll of the flare. Theincre:lsed shear dcmand led to she;u failurcand \'ertical(,."Olumn bar buckling in theplastic hinge region (Figure -1--7) at high localcurvature ductilitics. At Bull Creck unyoll,only the (,.'Olul1m ends m'er a dist,lIlcc of onecolumn diamcter werc confined with rightlyspaced spirals. while the columl1 centerportion fean\red :l!,r:lin :1 12~ pitch. In rhebent... wid1 the imcgral ch:lllncl wall (Fi!,'Ure4-8), the channel \\,:111 top was dearl)' in theregion of low rr:.lIlsverse reinforcement r:ltiothat provided little or no ductility to theinel;lstic HcxlIr;11 hinge Ihal forrned on lOp of

Why Did Highway Bridges Collapse in the Northridge Earthquake? 29

Page 41: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

the channel wall and fitiled in shear at lowflexural ductilities (Priestlev, Seible andUang, 1994]. .

Summary

Following the Nonhridge eanhquakethere was speculation that blamed thecollapse of bridge struCtures on:

I. Shon seat width at expansion joints.2. High vertical accelerations.

Unseating of the failed bridge sections atthe abutments and expansion joints as theprimary collapse source would have requiredsignificantly larger displacements at thecollapsed bents than the available displace·mcnt (:apacities at these bents. Therefore,the bents would have failed before unseatingcould have occurred. WIth the exception ofGavin Canyon, the likely failure sequencestarted with column shear or flexure/shearfailures, with unseating as :J. direct cOllse·qucnee of the shortening or collapse of theadjacent bents.

Vertical ground accelerations measuredduring the Northridge eanhquake were notdisproportionately larger than the measuredhoriwntal accelerations, when comparedwith other eanhquakes (Appendix B). All thedescribed bridge failures can be explained byonly the probable horizontal accelerations atthe respective bridge sites, IPriestley, Seible:and Uang, 1994], without contributionsby, or interaction with, vertical groundaccelerations.

Design. The bridge structures thatcollapsed in the Nonhridge earthquake seemto have been designed based on the best:available infonnation at the time of thedesign. Changes during the construction

30 The Continuing Challenge

phase of some of these structures (i.e., theI-5/SR-14 bridges) were not feasible-shon ofco1.tlplete demolition and reconstruroon­since all columns and most superstructureswere already completed at the time of the1971 San Fernando eanhquake and theprimary cause of collapse of tall singlecolumn structures was attributed to unseat­ing and in-span expansion joints. Retrofit ofthese bridge structures with expansion jointrestrainers was implemented immediatelyfollowing the San Femando earthquake.

Retrofit Delays. Subsequent upgradingof these first-generation restrainer retrofits:and column retrofit implementation waspending adequate funding and a lack ofcomprehensive seismic design researchresults until the 1989 Lorna Prieta earth­quake. The increased funding followingLoma Prieta acceler:ated the retrofit programand retrofit designs were completed for theI-10 St'nlctllres, but had not yet been imple­memed at the time of the Northridge earthqu:ake. Legal problems with leased airspace atLa Cienega & Venice, under the bridgestrucmrcs, have been cited as part of thecause of the delay in retrofit implementation.

The 1-5/SR-14 Strucrures were sched­uled for retrofit evaluation and design,somewhat lower on the Cal transprioritization list, due to the post-1971construction completion date (see Section 5,"Caltr-,ms Bridge Design and RetrofitProgram," subsection Retrofit ScreeningCriteria). The retrofit design was alsodelayed by the c1:assification of the 1-5/SR-14Interchange in an Alquist-Priolo SpecialSrudy Zone and appropriate smdies onpossible faulting and maximum expectedvenical and horizontal offSets had beeninitiated but not yet completed.

Page 42: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Retrofit Screening Criteria. The SR­118 bridges were originally on the assessedrisk priority list, but were subsequentlyremoved due to various criteria such as nointernal expansion joints, post-1971 designand construction, and redundancy in themulti-column bents. The reduced effectivelength of bridge columns through flares,channels or barrier walls, and ground surfacemodifications arc apparently not routinelychecked as part of the Caltrans structuralvulnerability assessment.

These stmcUlral vulnerabilities ofeffective column shortening, the effects ofheavy uni-directional skew geometry, and abetter assessment of flexural over strengthand aCUial shear capacities, arc known

problem areas and need to be incorporatedinto the routine vulnerability assessment ofbridge structures as quickly as possible.

Retrofit Would Have PreventedCollapse. A quantitative assessment of howeffective state-of-the-art bridge retrofittechnology [Priestley, Seible and Uang,1994J developed for and by Caltrans over thepast eight years shows that all seven of theNorthridge e:lrthquakc bridge collapsescould have been prevented with currentseismic retrofit tcchnolob'Y' Thus, the criticalelements were llot a lack of technical under­standing or design errors, but rather over­sights in the structural vulnerability assess­ment and, most importantly, the retrofitimplementation time factor.

Why Did Highway Bridges Collapse in the Northridge Earthquake? 31

Page 43: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

This page has been left intentionally blank.

Page 44: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Section 5

Caltrans Bridge Designand Retrofit Program

Evolution of Caltrans BridgeDesign Practices

The d;llll:lgC [0 bridges in theNorthridge c:lrthqu:l.ke had similarities to

that observed in previous earthquakes,p;JrIicularly the 1971 San Fernando and the1989 l,om:l Prict:l earthquakes. E.1ch of theseearthquakes caused major damage and thecollapse of sevcr.l1 buildings, bridges, andother strucmres. Each e\'cnr has had and willcontinue to have a major effect on rheimprovement of me seismic desi!:,"'l andretrofit procedures for Ca]rr:.ms bridges inCalifornia.

In 1943, the C:llifornia State Division ofllighways introduced a specific static seismiclatcralload requirement into its designspccific:lIiolls for the first time. Bridgedesign, at th:1l rime, considered ;\Il equivalelltlateral seismic load as a pereent:agc of thede:ad weight. The percent:age v:aried from2 to 6 depending upon the soil conditions.']\1'0 pcn:elll, four percent, and six percentwerc specified for bridges founded on rock,on soils h:lving 4 tonslft.1 bearing capacity,:and 011 piles, respenivel)'.

In 1963, the Division's Bridge Depart­ment adopted the Stnlctural EngineersAssociation of C:ilifornia (SFAOq codefonnul:ation requiring that the equivalenrlateral seismic load (EQ) be detenninedusing the formula EQ"" KCD in which D isthe dead load of the Stnlcture, C is a seismicb;lse·sh(..-ar coefficient given by o.osrr llJ

(I' being the fundalllent::l.l period of thesmlcrure) with 0.10 specified as ;In upperlimit, ;Inc! K is:a coefficient representingenergy :Ibsorption capacity of the structure.A K-\'aluc of 1.33 w:as specified for bridgeshaving wall supportS with height-to-lellgth

ratios of 2.5 or less, 1.00 W,IS specified forbridges having single-column or pier sup­portS with height-to-length ratios gr.caterthan 2.5, :and 0.67 w:as specified for bridgessupported on continuous frames. The designprovision also specified that the product KCshould never be less th:an 0.02.

In hindsight, and as was demonstrated bythe San Fernando eart!t(luake, using buildingdesign provisions for bridges was notappropriate. Bri(lges and buildings sharesome characteristics, but differ in fundamen­t::I.l ways that make their behavior \'erydifferent. The building prO\,isions expressedthe profession's evalu:uion of the totality of abuilding's characteristics and their expecta­tion of its performance in an earthquake.The provisions should onl)' have been usedfor structures that :ue similar to buildinb'S,which bridges arc not.

The damages lO bridge strtlchlreS duringdle San Fernando eartIH]u:ake made it veryclear th:at the ahove 1963 code provision wasin:adc(luate. Thus, the Californi:a StateDivision of Ilighways immediately institutedchanges to im.:rease the 1963 code force levelby the factor of 2 for all bridge.'i supported onspread footinb"'S and by the factor 2.5 forthose bridges supported on pile foundations.Besides increasing the code force le\"el, manystructural derails were impro\'ed consider­ably. These changes applied only to newdesigns. Cal trans knew that lIlany of thedeficient bridge strucolres in use weredesigned using the pre-19; I design criteria,and, consequently, initiated a seismic retrofitprogralll.

A brief chronologie-dl summary of someof the major de\'clopmenr.'i in these criteria isgiven helow:

Caltrans Bridge Design and Retrofit Program 33

Page 45: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

1971-1986: Research results from the1971 San Fernando eanhquake, as weU asrecommendations developed by the ATC-6project caused Caltrans to implement newbridge design criteria. During this period,ARS (acceleration response spectrum)ground motion curves and response reduc­tion factors were adopted, which, in general,led to higher design force levels and thespecification of robust spiral ties for columnswas implemcnted.

1986-1989: A retrofit program devel­oped by Caltrans identified single-columnbridge hcnts as being potentially the mOStvulnerable to carthquake damage. Rcsearchsponsored by Caltrans at the University ofCalifornia, San Diego, led to a retrofitprocedure that uses stcel jackets to incrcaseflexural ductility and shear capacities.Immcdiatc implementation was begun forthesc hem types.

Post-1989: Following the 1989 LOlllaPricta earthquake, Caltrans sponsoredacceler.\tcd retrofit research primarilyconducted at the University of California atBerkcley (VCR) and Univcrsity of Californiaat San Diego (UCSD) and appointed aSeismic Advisory Board. Peer review panelswere selcctcd for the retrofit or replacementof the damaged San Franciseo viaducts.The Applied Technology Council project(ATC-H) was initiated to review and revisebridge design criteria. \Vhile the draft rcsultsof this project are available and have beenpartially adopted hy Cal trans, the project hasyet to be completed. Administrative issuesh:lve held up the contract for the final period.

Although Cahrans design criteria havenot been fonnally revised, ad hoc criteria :lnddesign memor.mda have been developed andimplemented for replacement, :lS well as

34 The Continuing Challenge

retrofit, of existing bridges. These revised orsupplementary criteria include guidelines fordev~lopment ofsite-specific ground motionestimates, capacity design to preclude brittlefailure modes, rational procedures for jointshear design, and the definition of limit statesfor various perfonnance objectives.

Hinge restrainers had been installed in atotal of about 1,200 bridges in Los AngelesCounty sincc the 1971 San Fernandoeanhquake. At the time of the Northridgeearthquake, Cal trans had identified 716high-riSk bridges for retrofit in Los AngelcsCounty-retrofit had been completed for115 of these bridges, consisting chiefly ofstcel jackcts and footing strengthening atsingle-column hents.

In 1990, a total of 700 city and countybridges were targeted for retrofit. OnlY:lslll:lll number of these were completc at thetime of the eanhquake, due to fundinglimitations.

Governor's Board of InquiryInvestigation

A post-Lama Prieta earthquake reviewwas conducted bY:l Board of Lnquiry ap­poimcd by Governor George Deukmejian onOctober 26, 1989. After extensive hearingsand smdies, the Board submitted its findingsand recommcndations to the Governor in acomprehensive report, C'mllpnillg Agtli1lstTmtt, JVby 31,1990. This report gavespecific recommcndations for action by theGovernor, the Director of the Departmcnt ofTranspomJtion, and transportation agenciesand districts. OnJunc 2, 1990, GovernorDeukmejian issued Executive Order D·86-9O

Page 46: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

to implemcnt lhe Board of Inquiry's recom­mendations, which contained the followingitems of importance to Cal trans programs.

I. It is the policy of the Stlte of Californiathat seismic safety shall be brivcn priorityconsideration in the allocation ofresources for transportation constmctionprojects, and in the design and construc­tion of all state structures, includingtransport:uion structures and publicbuildings.

2. The Director of the Department ofTransportation shall prepare a dctailedaction plan to ensure that all trnnsporta­tion stmctures maintlined by the Stileare safe from collapse in the event of anearthquake and that vital transport'Jtionlinks arc designed to maintain theirfunction following an earthquake. Thephm should include a priority listing oftransport:ltion stmcmres which will bescheduled for seismic retrofit. TheDirecror shall transmit this action planto the Governor by Augtlst 31, 1990.

3. ·rhe Director of the Departmcnt ofTrnnsportation shall eStlblish a fonnalprocess whereby the Department seeksand obtains the advise of external expertSin establishing seismic safety policies,standards, and technical practices; andfor seismic safety reviews of plans forconstmction or retrofit of complexstmCnires. The Director shall transmit asUlllmary of this proccss to the Governorby August 31, 1990.

4. The Director of the Departmelll ofTrJnsportation shall assign a highpriority to development of a program ofhasic and problcm-focused rescarch onearthquake engineering issues, to include

comprehensive earthquake vulnerabilityevaluations of important transportationstructures and a program for placingseismic activity monitoring instrumentson transportation structurcs. TheDirector shall transmit a description ofthe resCflrch program to the Governorby AUbrust 31, 1990.

Caltrans Response to Boardof Inquiry Recommendations

Caltrans has made fundamental changesin its operations in response to the Board oflnquiry recommendations, and has takcnactions to fulfill these requirements on apriority basis.

Caltrans, and in particular the Divisionof Strucmres under James Roberts, hasresponded positively and quickly to theserecommendations. They have issucd annualstatus reports on "Caltrans Response toGovernor's Board of Inquiry Recommcnda·tions and Executive Order ofJune 2, 1990."(See Appendix A for an abstract of the lateststatus report,January 26, 1994.)

Caltrans in response to the recommen­dations has:

I. Appointed a Seismic Advisory Board t'O

rcview its programs and advise ontechnical and administrative programs,bringing oversight and contributionsfrom an extended community of earth·quake engineering specialists. It has metapproximately quarterly since then toreview and advise Caltrans on proposals,progress, and implementation actions tomeeting the recommendations.

2. Developed an action plan to assureseismic safety of state-owned bridges.

Caltrans Bridge Design and Retrofit Program 35

Page 47: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

3. Perfonned vulnerability assessments ofthe 24,000 st<1te, county, and city bridgesand <1 developed prioritized list toimplement a seismic retrofit program.

4. Developed a bridge seismic performancepolk".

5. Implemented independent technical peerreview of the seismic aspects of impor­tant projects, thus opening their designprocess to influence by a broadertt."chnical community.

6. Developed a priority list for the retrofit­ting of high hazard structures based on arational procedure.

7. Initiated seismic retrofit design andconstruction for approximately 2,000high-hazard struetures to be completedover a 10-year period (1989-1999).

8. Instituted changes to the Caltrans BridgeSeismic Design Specifications andCriteria.

9. Established an Office of EarthquakeEnbrincering and conducted extensivetraining in seismic design for over 200bridge engineers.

10. Increased commitment to researchfunding with an initial invesnnent of$8million, followed by annual expendituresof$5 million on problem-focusedseismic research topics.

The above actions are in various stagesof progress and will require a continuingmanagemcnt (:ommitment to their comple­tion. The Seismic Advisory Board evaluatesCal trans performance as consistent with thedirections of the Executive Order andlegislative directions on seismic safety.Appendix A provides a detailed review of theactions, through January 24, 1994, by theDepanment of Transportation in response tothese recommendations and directions.

36 The Continuing Challenge

Peer ReviewIn response to the Board of Inquiry

report, Caltrans implemented a seismicsafety peer review process for selectedimportant new or retrofit bridge designprojects following the Lama Priem earth­quake. The peer reviews to date havc hadseveral different forms and functions,including:

I. Review of seismic design criteria only.2. Review of criteria and designs.3. Review ofcompleted designs only.

Both Caltrans and outside engineeringconsulmnts have taken the review processvery seriously, resulting not only in structuralseismic safety improvements, but also in alearning experience for all parties involved.

The Seismic Advisory Board recOlll­mends that the peer review process bestandardized to make it more effective. Thescope of review process should be st:llldard­ized in tenns of:

1. Which bridges are to be scrutinized.2. Scheduling the review to allow designers

and reviewers the time necessary toscrutini1..e and/or improve on the seismicperformance of the bridge structurc.

3. Completeness of the peer review,starting with the initial concept andstrategy of design and type of selcctionand continuing through the final seismicdesign detailing.

The specific terms of content or formatshould not be standardized; they must beproject-specific.

Peer review is currently only imple­mented for a few special structures (seeAppendix A). Yet, as the Northridge experi­ence shows, the seismic performancc of

Page 48: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

common strucrures affects the functionalityof the transportation system as a whole. TheBoard believes that peer review also shouldbe implemented for some selected commonbridges. This will help ensure that all newand retrofit designs benefit from the besttechnical knowledge :md experience, not juStthe "important" strucrures. What makes abridge "important" needs review and clarifi­cation in light of the regional transportationsystem so that it will be clearer to the publicand bener defined for engineering designpurposes.

As a very positive nOte, it should bestated that experiences with the peer reviewprocess to date indicate that peer reviewseems to be the vehicle thai integrates thelatest seismological, gL'Ological, geotechnicaland strllcrural findings imo Caltnms seismicdesign for bridges and bridge retrofits.

Peer Review andConstruction of theNorthridge EarthquakeReplacement Bridges

Caltrans set a very fast schedule for theremoval of all collapsed and unsafe bridges,the replacement of nine bridge structures,and the retrofit of one bridge-all to becompleted by Dccember I, 1994. Thisschedule provided new challenges not onlyfor dcmolition and construction, but also fordcsign and seismic safety review. Demolitionand construction contracts were givcn on aninvited, prequalified limilcd bid basis withheavy incentivc and penalty clauscs for earlyor late completion, rcspectively. For thisreconstruction effort dcsign submittals arestaged to just stay ahead of the construction,

and the seismic safety review is based onevolving and continuously changing designconcepts and documents.

Nevertheless, Caltrans is illll>lementingsignificant changes from past desib"1l practicebased on lessons learned from theNorthridge c:trthquake in terms of:

1. Elimination of most in-span expansionjoints through longer jointless super­strucnlres or special hinge bents.

2. Elimination of excessively skewed jointsand abutments.

3. Balancing the stiffness of bridge columnswithin individual frames.

4. Design alternatives in steel and concretefor selected structures.

5. Use of site-specific geological andseismological data for ARS curves andsubstnlcttlre stiffness.

6. Consideration of potenti:tl verticalexcitations for the superstructure.

All these new concepts, and the associ­:Hed changes and deviations from establisheddesign procedures, were implemented andaccomplished in this short period. Thisindicates a commendable flexibility andcapability of Caltrans designers and engi­neering consultants. The Board believes thatthe public can have confidence in the seismicsafety of these replacement stnlcrures.

Caftrans Bridge Design and Retrofit Program 37

Page 49: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Table 5-2. Status of the retrofit review ofState highway bridges.

Stage Number

Total brid!iles on thestate highway system 11,895

Total screened out in firstscreening (priorityindex) stage 4,612

Total screened out insecond screening stage(plan review) 3,595

Total screened out in thirdscreening stage (detaileddynamic analysis) 1,151

Total remaining in the retrofitprogram; some will bedropped as the detailedanalysis continues 2,537

for site specific analyses because existing ARScurves were not felt to be appropriate. Majordeficiencies of the original structures were inthe areas of insufficient joint shear reinforc­ing; inadequ:ne reinforcing steel anchorageand laps; large torsion resulting from freewayconfiguration; and the absence of longitud­inal frame action. Rather than an incremen­tal retrofit program, where the first stepwould permit the opening of the freeways totraffic and the second step would refine theretrofit to provide for post earthquakeoperational capilcity after a future majorearthquake, it was decided to combine thiswork into a single effort.

Prioritization and ScreeningPractices

Cal trans experience with retrofit of thedamaged bridges in the San Francisco Bayarea following the Lama Prieta earthquake in1989 emphasized the need for a "strucnlralsystcms" approach to the retrofit of older,seismically deficient bridges instead of a"strucnmll elements" approach. With thesupport of the Seismic Advisory Board,Caltr;ms initiated a comprehensive retrofitprogram.

The decision on whether a bridge shouldbe considered for retrofitting is based on afour step-process for prioritizing strUCt1!resfor retrofit. At any step in the process, astructure can be assessed as acceptable andnot further considered.

33%

38%

29%

25%

22%

8%

12%

16.5%

16.5%

28%

15%

7%

12%

7%

7%

14%

10%

CharacteristicWeight

Caltrans Retrofit and DesignResponse to Loma PrietaEarthquake

vVithin weeks after the Lorna Prietaearthquake, Caltrans embarked Oil a massiveprogram to replace, remove or retrofitdamaged freeways in the Bay Area. Toexpedite the process, Caltrans retainedoutsi<le consultants to design and detail theretrofits of major freeways. Directions to theconsultant were issued in the form of memos.

As required by Executive Order D-86-90from the Governor, Caltrans appointcd PeerReview Panels consisting of private practitio­ners, academicians and researchers toevaluate the work of the Caltrans and itsconsultants. The peer review panels heldregular meetings to review both retrofitcriteria and mcthods, resulting in substantialchanges in the work underway. in thc case ofthe San Francisco Double Deck Freewayretrofit program, a number of major issucssurfaced. Poor soil conditions led to the need

Categoryfcharacterlstlc

Hazard (H,)

Soil conditions

Peak rock acceleration

Duration

Vulnerability (VI)Year designed

Outriggers or shared columns

Abutment type

Skewness

Drop type failure

Bent redundancy

Impact (I,)

ADT on structure

Leased air space (residential, office)

Leased air space (parking, storage facility)

ADT under/over structure

Facility crossed

Route type on bridge

Detour length

Critical utility

Table 5-1. Weights used in the prioritization process. The priority isdetermined from the formula below where A, is the fault activity (0.25-1.0).H, is the sum of the hazard characteristic values, V, is for vulnerability andI, for Impact. Specific characteristic values range from zero to the maximumvalue given based on the characteristics of an individual bridge.Priority rating index", (A, H,)(O 60 1,+0.40 V,)

38 The Continuing Challenge

Page 50: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Table 5-3. The Calrrans bridge relrofil program status as of June I, 1994. The proportions arebased an the total number af structures affected. M is an abbreviation for miliians. (Caltrans 1994)

Estimated Construction ConstructIonCategory of structure total cost complete underway Remaining

1. Single column retrolit $120 M 87% 13% 0%2. Multiple column retrofit $1,650 M 2% 7% 91%3. Toll bridge retrofits $650M 0"'<' 0% 100%

Screening pnx:cdures for retrofittingstructures arc dC\'c!opcd in four steps:

I. A computerized prioriti7A1tion alb'"Orithmwas dC\'c1opcd to e\·:l!uatc the \~Jrious

attributes of L-Jch bridge and to :lssign aquantified r:mking for retrofit. It cmploysthree major (:ategories for evaluation:I) vulnerability ofstmctures; 2) seismicha7.ard; and 3) impact on the cOllllllunity.Each of these c:ucbrories h:IS a number ofspecific elemcnts (Tahle 5-1).

2. Initial screening of the approxilll:ncly24,000 state, county, and CilY bridges inC:llifornia to determinc their seismicvulnerability. About 7,000 state bridgesand 4,000 county and city bridges wereidentified :lS being plllel1tially h:17...1rdous.

3. Det:liled pl:ln review of :IIJ 11,000potel1tially hazardous bridges.

4. Demiled seismic cV:llllation of theremaining bridges in order of priority toidentify structural deficiencies forrctrofit (sec Tahlc 5-2).

5. Design and prepanltion of the nccessaryconstruction dOClllllents to illlplclIlcntthe retrofit. Unlike prior rctrofit pro­grams, this progr.U11 systclllaticallyaddresses deficiencies ill all thc stnlcl'uralcomponents of each bridge.

Lnitially there were 11,895 srate highwaybridgc strucnlres 10 be ranked. Of this brrouP,1,537 wcre judged to he h:17~1rdous, allhoughsince nUl alllhini-slep studies are complete,this numbcr may be reduced. 'l'3ble 5-2shows the se()uence of reductions in numbersat the three assessment steps of the review,yielding a best estim:uc al this timc ofabout2,000 bridges that will need retrofitting. Ofthese 2,000 high-risk bridges, 716 were inLos Angeles County. Ellbrlnecring design has

been completed for 800 bridges in the Sr:ltcjconstruction has !leen complet'ed for 250;approxim:Hcly 400 are in the process of beingrctrofitted.

Although most state bridges werescreened out of the current retrofit program,this docs nor mean that they satisfy IlltKlemdesign and construction standards. At a laterdate, further consideration must be brl\'en tothe potential for severe damage or collapsc.

Current Status of StateHighway Bridge RetrofitProgram

A totll] of2,537 state bridges rClllain inthe current rctrofit program based 011 thescreening procedure discllSsc(! :lhovc.Further scrcening and preliminary .~tructur:ll

evalll:llion havc rcsulted in identification ofabout 2,000 st,ne bridges that n:lluiredetailing evalu:nion and retrofit. As indic:ltedin Figure 5-2, retrofit plans have beencompleted for llpproxilllately ROO hridges,bids have been opened for 400, :Jnd retrofith;ls been completed on ahout 250.

The retrofil progral1l for stale highwaybridges has been divided intO three parts­single-column, multiple-coluTTln, and mil­bridge programs-with a TOtal cstim:lted COSI

ofS2,420 million (l'3ble 5-3). All projects arcin progress with the indicated l>crcemagesfor strucrures with collstruction complete,and under construction. Figure 5-1 graphi­cally illustrates the proportion of SlnlcturCSimpacted in each of the nine projL'Ct catego­ries from Rendy For Ass;gmllmt to COllm'fll1;O"C(IIlIpletrd.

Calrrans Bridge Design and Rerrof;t Program 39

Page 51: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

3

TollBridgeProgram

Single­ColumnProgram

5

Multlple­ColumnProgram

II Construction completed

II Under construction

a Bids opened

_ Advertisement dale set

Planning Specifications &Engineering

o Structural Plans complete

o Strategy determined

o Strategy being analyzed

~ Ready lor assignment

Figure 5·1. Pie charts show the status o( the single-column. multiple-column and bridge retrofitprograms for state highway bridges. The segment represents the proportion of the structures in theparlicular category of completion compared to the number of structures affected.

I_ Structure plans complete /- --0-- Bids opened

- ---.......- Bid openings scheduled ..--0-- Construction complete

'"V ~

r nrP-"'

/ V """'"~

~ '$

700

800

100

oMar 89 Sep 89 Mar 90 Sep 90 Mar 91 Sep 91 Mar 92 Sep 92 Mar 93 sep 93 Mar 94

Figure 5-2. Total numbers of seismic safety retrofit projects for state highway bridges in differentstages of completion since March 1989, Note that the program accelerated substantially followingtile Lorna Prieta eartllquake on October 17, 1989.

600••2500u,~400..~300E,z200

40 The Continuing Challenge

Page 52: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Tabl. 5-4, Status of toll bridge seismic studies. The dates given are those when the phase ofIhe study was completed or is scheduled to be completed; T indicates target. C indicatescompleted, and' indicates that the target date is based on Caltrans' ability to con/ract workwith consulting engineering firms.

T6196T6196Unscheduled'T 6196'T 6196'T 6/96·T 6/96·

Unscheduled'T 6196·

T 6196'

N/AN/A

T 8/94T 5194T 5194C 11/93T 12/94C 12/93C 8/93

T 5/94C 7/95C 11/91NlANlA

C 12/92T 5194C7193C2I93C6I93C 2/93

C 12/92C 9/93

T 7/94T 7/94T 12/94T 12/94

Dete of completion or scheduhld completion

Htwlrd Vulner8blllty Design PS&E-San Francisco-Oakland Bay Brjdge(12 distinct design projects) 1

West spansEast spans

Dumbarton BridgeSan Mateo-Hayward BridgeRichmond-San Rafael BridgeCarqulnez BridgeBenicia-Martinez BridgeAntioch Bridge2

San Diego-Coronado BridgeTerminal Island (Vincent Thomas) Bridge

Commodore Schuyler Helm Bridge3

Terminal Island (Gerald Desmond) Bridge'

Not.. 1. Some retrotit construction is complete (Pier E-9), having been initiatedimmediately after the loma Prieta earthquake; other construction will be initiatedin the summer of 1994 and continue until all projects are completed.

2. A recentty completed design that may require retrofit work.3. This bridge is on soft ground over an oil field that has been settling tor years. No

amount of retrofitting would guarantee continued operation.4 Bridge built by the Port of long Beach. It will become part of the state highway

system in the near future. Caltrans is negotiating an agreement to have Port bringit up to current maintenance standards before acceptance. Seismic vulnerabilitystudies will be initiated upon acceptance.

Figure 5-2 graphically emphasizes theremaining vulncrability of thc state highw;ly~ystelll. In particular, it should be noted thatretrofit plans have not yet been completedfor any of the II toll bridges in California.Most of these bridges, such as the SanFrancisco Bay Bridge, arc vital to thee<.:onomic welfare of the area.

None of the toll bridge retrofit projectsare yet in construction. Toll bridb'CS arc verycomplex structures, requiring substantiallymore effort than do conventional bridges.The seismic retrofit progrnm for thesestructures has been approached in threeparts:

I. Haz.ard analysis2. Vulner:ability analysis3. DcsibtJ'l planning, specifications ;md

engineering (PS&E)

lable 5-4 gives the status of these studiesfor each of the II roll bridges for whichCaltrans is responsible. j-1a1.-,ud analyses havebeen completed for all the toll bridges, withthe results having been made available forvulnerability assessments. Vulncr:abilityanalyses are completc for somc of thescbridgcs and retrofit designs are underway.

"'hile the rate of preparation of retrofitplans, as indicated in Figure 5-2, app<.-ars to

be accelerating, the debilitating effect of thelawsuit by the PEG group (Caltrans employ­ees) is evidenced by the flat portion of thecurve in May·July of 1993. Additionally, itappears that there is a r:apidly increasing gal>between the number of retrofit plans that arecompleted and the number of comracts thatare advertised and aw:mlcd.

Caftrans Bridge Design and Retrofit Program 41

Page 53: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Tabl. 5-5_ Sratus of the seismic retrofit ofcity and county bridges throughout rhe State.Note that the classifications are somewhatdifferent than those used for stale highwaybridges in Table 5-2.

Stage

Total bridges on thehighway system

Total screened outin first screening(priority index) stage

Total screened out insecond screeningstage (plan review)

Total remaining in the retrofitprogram after secondscreening stage

Totallefl to be assessedin third screening stage(detailed analysis)

Number

12,000

6,607

4,152

1041

648

Current Status of City andCounty Bridge RetrofitProgram

Table 5-5 indiC3tes the status of theretrofit prob'T:Ull for county and city bridgesundemkell by Caltrans as direeted by law,even though the bridges themseh'es arcunder local jurisdiction. Of the 12,000 tomlbridges in these systems only 1,041 (8.7%)remained afTer the second screening com­pared to 31 % for the st~He bridges. Statebridges tend to be larger than local govern­ment bri(lgcs ,md thus tend to be 1110revulnerable and have larger impact in case ofElilure or closure. Construction is completefor 18 and underway for :lJlother 36; 165remain to have the retrofit design processinitiated, ,llld 648 have yet to have had theirthird level investig,ltion.

Summary

In SUllun.lry, the 1971 San Fernando :mdthe 1989 Loma Prieta earthquakes both hada major impact on increasing the awarenessof the seismic risks to bridge structures inCalifornia. Following the Loma Prietaearthquake, Cal trans responded positi\'ely tothe recommendations of the Governor'sBoard of Inquiry for improving the seismicsafety of highway bridges outlined inC011lpnil1g Agail1st Tit/It:.

A/I post-Lama Prieta retrofitted bridgesperfonned well in the Northridge earth­quake. All of the bridges in the rCbrlOn of

42 The Continuing Challenge

strong 1lI0tion are in the "common" class. Ofthe seven bridges that collapsed, five hadbeen identified and scheduled for seismicretrofit. 1".'0, the Mission & GothicUndercrossing and Bull Creek CanyonChannel Undercrossing on State Route 118,had been evaluated as not eurrently re(luiringretrofit. In light of this experience, it isadvisable t'O review the prioriti:r.ing procedureand reexamine the retrofit decision for thoscbridges that were eliminated from theretrofit program t'O determine if they sh(wldhc reconsidered. The seismic safety ofcOl1lmon other bridges should also bee:mmined for possible retrofit.

It is essential that retrofit of the remain­ing deficient bridge structures in Californiabe ~lcceleralCd so that, hopefully, it will becompleted before the next major eartluluakeoccurs. V/h:Hevcr ~lctions and support .1rerequired to accomplish this must be providedby the GO\'ernor ~llld the State Legisl:Hure.

Page 54: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Section 6

Improving the CaltransSeismic Program

Retrofit PrioritizationProcedures

The Nonhridge C:lrtluluakc has shownth:n the ambitious retrofit I>rioriti:t.:luollprogr::un undcrT'".lkcn hy Caltnllls for all ofthe State's H ,000 bridges has bc...-cn cffceti\'c.Ilowc\'cr, it is not flawless and needs colllin­lied scrutiny :lmlupdaring. This sectionpresents some of the Seismic AdvisoryBoard's suggestions for improvement based011 observations of system pcrfonnallcc in theNorthridge c:lrth(lliakc :lnd rc(:onsidcrationof some past decisions Calrr;lIls h;IS made.

It can Ix: ar!>'1.lcI! th:lt the uncertainties inassessing seismic hilzard :Jlld the CUlllplcxiticsof (Ietermlning strllct'ul~ll vu[ncr,lbilitycannot be qu,llltilicil il1m:1 deterministic,llumeriCl1 risk ,lsseSSlllclll and retrofitpriol'itiz'ltion. Calt'l",l1lS is well aW,lre of thisargulIlent. The CaltT;lIlS risk ,lsscssmentalgorithm, which has been cOlltinuollslymodified ,md updated over rhe p,lst threeye,lrs, is only used as a preJfrem;lIg toni. Allthe bridge structures th,lt colhlJlSed in theNorthridge earthql1'lke were initilily identi­fied as potenti,ll retrofit eandid:ltes using this'llgorithm. However, tWO of these wereremoved from the list on the third step ofscreenlllg.

The subsecluellt lll:mu,11 screening of allidentified C':.Hulid:lte structures relies he:wilyon subjective judb1TTlenr.... by individual reviewengineers :md, consequently, is prone tohuman error and omissions. Abrain, Caltr.mstries to minimize t1u:se prohlems by having,It least tWO independent reviews, witharbitration in C:.ISC of differences. The factthat only two bridgL"S of 506 hridges Io<:atedwithin the strong shaking ,lre:l (PGA~O.5g)were misjudged by the Cahr.ms risk assess­ment pnK."cdure is :letu:llly ll"ite remarkable.

After a preliminary assessment ofobservations from the orthridbrc earth­quake, the Seismic Advisol)' Board SU!''"gL"Ststhe following :lctions be taken by C:lhrans:

1. Review of Phase I expansion jointretrofits in terms of restrainer orienta­tion, restrainer capacity, detailing andscat width c..\"tension, and vulnerability ofcolumns.

2. Train re\'iew engineers to look foreffective length of columns as modifiedby flares, walls, or ground surfaL"Cconditions.

3. Train re\'iew engineers to assess poten­tial ductile vs. brittle failure modes incollllllns.

4. Rescreen all bri<lge stnlchlres based onthe latest hazard and vulnerahilit)'findings.

These :lctions should be incorporincd ,ISexpcdiently ,IS possihle into the retrofitprioritiziltion procedure ,lllli tr,liningprograms.

\-Vhile almost all bridge design is basedon lincar ehlstic methods, dillllagc is <;:lusedby nonlinC:lr r~sponse. Linear elastic designmethods arc formulated so that they provideadequate nonlinear response in IIIOSt L":.ISCS.\Vhil~ linear 'lpproaches Ill;ly be quiteserviceable in mOSt L":.\5CS, some applic.":.\tionsrequire non linear analysis to properlyunderstand th~ Stnlct\lres response :ll1d judgethe adc<]UaL'y of a design. Nonlinear analysiscan be :Ipplied both to dynamic time historyand static approaches, in the btter C"Jse push­o\'er analyses arc one commonly uscd:Ipproach. The Board believes it to beimporelllt that Cahrans develop the staffC"Jpablc of pcrfonning nonlinear analyses forboth complete bridge systems and for

Improving the Caftrans Seismic Program 43

Page 55: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

sections of bridges. It further believes that llll

Calmms design engineers should be wellinfonned as to the Limitations of linear elasticanalysis and the circumstances when each isappropriate.

Finally, retrofit designs, and particularlyimplementations, need to proceed as quicklyas possible for all bridge types, withoutconsideration of restrainers first, single­column bents second, multi-column bentS

third, and toll bridges separately or fourth.The highest risk (including the greatestconsequcnce of failure) is likely to be foundindependent of the these basic types ofcategories.

For example, the Board could make anargument for retrofitting all major rollbridges first, since their failure or closurewould have severe and extreme economicimpacts. A good argument could also bemade for doing multi-column bent bridgesncxt, since those columns are prone to shearfailure due to their shorter length and thepotential for brittle failure mechanismswithout ductility or energy absorption towithstand the duration of strong groundshaking. Assigning priorities for retrofitbased on categories like these tends toobscure the relevance of a particular bridge'svulnerability compared to others.

The Board believes that the methodsused to assign priority needs rethinking. Thecurrcnt system uses one index to determinethe priority and it may not be robust enoughto order bridges properly. It may be neces­sary to use more complicated approachesthan the current one. Furure priority assign­ment systcms must recognize all of thefactors and s!)Ccific characteristics for eachbridge that contribute to its h:l7.ard, vulner­ability, and importance valuations.

44 The Continuing Challenge

Site-Specific StudiesThe Northridge earthquake indicates

the importance of Calrrans undertakingspecific geotechnical srudics for all newdesigns or retrofits of major bridges ;md forsites that are expected to perfonn poorly.The evidence of the Nonhridgc canhquakcis that motions at several of the collapsedbridges may havc been sib"tlificantly amplifiedby the local soil conditions. Current weight­ing factors in the prioriti7.ation procedureapply too little weight to the geotechnicalconditions of the site and their variationsalong the length of a strucrure.

Use of EarthquakePrediction for PrioritySetting

The Northridge earthquake could nothave been forecast as to specific place, timeand size. Though it occurred on a blind­thrust fault, it was not unexpected, bccause itlies within in a broad region where suchactive faults are known to occur.

In general, quantitative statements onthe probability of £Ururc earthquakes are oflimited value in deciding CaJtrans retrofitpriorities. Nevertheless, when the impor­tance factors for structures arc considered,there are certain lectonic regions of tile Statewhere the likelihood of intense strongIllotions in the lifetime of the strucn-res canbe specified with some confidence. Thisgeolobrical and geotechnical knowlcdge,where available, should guide the design ofnew structures and the prioritization forretrofit of sttucrures.

Page 56: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Research Should beEnhanced

Caltrans has made a commitment tosupport research and development, and hassustained a $5 million pef year researchprogram on bridges since the Lama Prietaearthquake. The evidence suggests that itwill continue this commitment. There is,however, morc to do than this budgetcOlllmitment allows.

The Northridge earthquake confirmedlaboratory test result... on the effectiveness ofbridge column retrofitting to preventstructural collapse, thereby mitigatingdamage and decreasing dismption followingan earthquake. However, it needs to beemphasized th'lt the last two earthquakeswith significant hridge damage-LamaPrieta in 1989 and Northridge in 1994­were of the "moderate" category. The"large" carth<]uakcs, still to occur, willprovide more severe testing of design andretrofit practices. Thus, no field evidence isyet available on how retrofitted bridges, ascurrently designed, will perform under max ­imum expected ground motions. Research isthe only way ro develop a full understandingof what bridges and other structures will besubjected to during longer-duration groundmotions prior fO their occurrence. The"large earthquakes" will probably notproduce significantly higher peak groundmotions at a point, but will produce thesehigher motions over much greater area andfor a longer time, thus affecting a vastlylarger number of strucmres than wereeffected in the Northridge earth{IUake.

The Caltrans retrofit program firstadded restrainers in expansion joints and,next, retrofitted single-columns with steeljackets. 'T'hese weak elements in the bridge

structure provided a strucmral fuse whosefailure protected other elements of thestrucmre that also may be very damage­prone. Now that these retrofitted elementscan be expected fO survive, adjacent struc­rural elements, such as footings, superstruc­ture/column connections and, in particular,abunnents, will be subjected to previouslyunreached force levels and deformationdemands. Thus, significant damage in theseadjacent bridge members can be expected infuture seismic events.

It is essential that the Caltrans compre­hensive seismic bridge retrofit researchprogram be continued and accelerated toaddress the behavior of bridges as complete,interconnected systems, including soileffects. Research should focus not just onsafety or "no collapse" criteria, but also onfunctionality or serviceability criteria thatclearly OlJtline the expected bridge damage,repairability, and bridge or route closureconsequences.

The Seismic Advisory Board recom­mends that the following research and studyagenda be addressed;

I. The cOlllributions of site-specificgeological, seismological, andgeotechnical characteristics.

2. Response of bridges as completesystems llsing nonlinear as well as linearapproaches.

3. Detennination of the types and extent ofdamage to different types of elementsthat corresponds to the thresholds fordifferelll performance levels; for ex­ample, what degree of cracking in areinforced concrete column is acceptablebefore it is dosed for repairs, or must beremoved and replaced.

Improving the Ca/trans Seismic Program 45

Page 57: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

4. Predicting the expected damage statesfor specific bridge systems using differ­ent desibttl approaches and constructiondetails in specified earthquake groundmotions, including those used for bothfunctional and safety evaluations.

5. Functionality criteria for post-earth­quake repairability and serviceability.

6. A comprehensive strategy wherebyresearch results and other st:l.te-of-the­art advances in engineering are incorpo­rated into Caltrans desibttl practice. Thisimplementation prob'J'am should coordi­nare research programs and ensure thatnew knowledge and technology areincorporated into both the structuralvulnerability assessment as well as intoretrofit implementation and new bridgedesign in the shortest time framepossible.

The Roard also recommends thatCalrrans P3Y special attention to re3ssessingand upgrading Phase I rcstminer rctrofittcchnolob'Y, sincc it is now known that larger

46 The Continuing Challenge

forces and displacements can be expectedfrom retrofitte<1 bent systems. This mayrequire replacing sOll1e restrainer systemsand strenb'thening some foundations, andverifY that columns will nOt fuil in she:Jr.

Finally, the design and evalll:Hion criteriafor design of bridge struchlres needs to heregularly reviewed and revised to renect thelatest scismologic:ll, geological, geot'echnic:ll,and structuml research findings.

The focus of rescarch and professionalpractice development to dare has been onreinforced concrete and older steel bridges.Observed damage to modern steel buildingssuggests thar there may be serious seismicperfonnance problems associated withcurrent design and (.'onstruction proceduresfor steel bridges. These issues need to beinvestigared with the objective of ensuringthat Caltrans bridge dcsibtflS are nor subjectto the sallle types of Elilllre problems.

Page 58: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Appendices

Page 59: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

This page has been left intentionally blank.

Page 60: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Appendix A

Caltrans Response toGovernor's Board of InquiryRecommendations andExecutive Order of dune 2, 1990January, 1994 Status Report

This Appendix reproduces, with minordeletions, the DCP:lrt11lCllt of'ThHlspOrl,Hion's J,lIlU,lry 1994 report on theS[,ltllS of actions in response to theGovernor's Board of Inqui ry Rccommend:l­[ions ,Ind Executive OnJcr ofJune 2, 1990,ICaltr:lllS, 19941.

The Seismic Advisory Board finds th,athe Department of'Tr,lllsport,ltion hasresponded very positively to: I) Each of thecleven rccol11111cndations the Governor'sBo,lrd of Inquiry presented in their reportComprfll/g Against Time, (hlted Nby 31, 1990;2) Governor George DCllklllcjian's ExecutiveOrder D-86-90, (btcdJUllC 2,1990; 3) there(]uirerncnts or Senate Bill 36X, signed intobw hy Governor DcuklllCji:lll on November6, 1989; 4) to Senate Bill 2 [04 (Kopp). Eachof these four documents recommended ,1Ildmandated specific :lspects of a continuing.1lId aggressive seismiC safety progr:1tn fortr.lnsport:ltion structures. The followingC,lltr,lns det:liled report, dated J:muary 26,1994, :lddresses e.lch of the cleven itemsrccolllmended ill Competillg Against Timespecific'llly and includes att:lChments con­taining more detail.

Caltrans Report,January 26, 1994

The progress report of the clIrl'enl statusof the Departmcnt's Bridge Seismic SafetyProgr:llll is addressed in "Seismic SafetyRetrofit Progr:lm-Annual Report" to bepresented to the Californi:l TransportationCommission on February 24, 1994. From ,1total of 2,500 Single Column supportedbridges on the State highw,ly system, 259have been identified as needing retrofitupgrading after all screening :m<l :lnalysishave been completed. Of these 259 bridges,258 have been completed or are underconstruction, and the remaining arc beingdesigned .1I1d will go to construction within afew months. The relll:lining 9,500 bridges onthe system arc Illultiple column supportedstructures. initial screening pared that listdown to 4,500 that require dynamic analysisto determine their status, and we estimateth:lt approximately 774 will require retrofit­ting. In this category, 96 :lre under construc­tion or completed, plans arc completed on470, and design is underw:ly on 208. Theremaining bridges arc in various stages of'lll'llysis ,inti screening. \Ve have more than50 consulting firms :lssisting the Departmentst:lff and will complete all pbns on thesebridges hy December 31, 1994. The 8 TollBridges arc in v,lrious St.lgCS of analysis anddesign and will be under construction by theend of 1995.

Caltrans Response 10 Governor's Board of Inquiry Recommendations 49

Page 61: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

1 BOARD OF INQUIRY RECOMMENDATION No. 1 FOR AcnON BY THE GoVER.."mR.

Affinn the policy that seismic safety shall he a paramount concern in thedesign and construction of transportation structures. Specific goals ofthis policy shall be that all transportation structures be seismically safeand that important transportation structures maintain their functionafter earthquakes.

Governor Deukmcjian's ExecutiveOrder D-86-90 ofJune 2, 1990. "It is thepolicy of the State of California that seismicsafety shall be given priority considemtion inthe allocation of resources for transportationconstruction projects, and in the design andconstruction of all state structures, includingtransportation structures and public build­ings" and

"The Director of the Depamncnt ofTmnsportation shall prepare a detailed actionplan to ensure that all transportation struc­tures maintained by the state arc safe fromcollapse in the event of an earthquake andthat vital transportation links are designed to

maintain their function following an earth­quake. The plan should include a prioritylisting of transportation strucrures which willbe scheduled for seismic retrofit. TheDirector shall trnnsmit this action plan to theGovernor by August 3I, 1990."

Stotus on ]nntlllTJ 26, 1994: "Report ToThe Governor On Seismic Safety, August 31,1990" was submitted by the Department inresponse to the Executive Order. The reportdescribes the action plan to assure seismicsafety of state~ownedbridges. Included is the

50 The Continuing Challenge

initial list of 7,000 bridges potentiallyneeding some degrec of seismic analysis,evaluation and possible retrofitting. Alsodescribed in this report arc the makeup ofthc Seismic Advisory Board, the SeismicResearch Advisory Panel, use of SeismicSafcty Peer Review Panels and the SeismicResearch Program.

By July, 1990, Department Enginceringstaff had completed an initial "VulncrabilityAnalysis" of the 23,000 state, county and citybridges and produced a prioritized list fromwhich to completc the seismic safety retrofitprogram. Thc logical strategy was to retrofitor replacc thc most vulncrable bridges on thelist first, when this approach is possiblc. Thatinitial screening list contained approximately7,000 state and 4,500 city and county bridgeswhich required furthcr analysis and c..'\'3Iua­tion before a dctennination could be made asto seismic retrofit needs.

A bridge seismic perfonnance policy wasdeveloped in 1991 by the Department ofEngineering staff for state-owncd bridgesand was approved by thc Department'sSeismic Advisory Board in September, 1992.

Page 62: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

2 BOARD OF INQUIRY RECOMMENDATION No.4 FOR ACTION BY THE DIRECTOROF THE DEPARTMENT OF TRANSPORTATION. Prepare a plan, includingschedule and resource requirements, to meet the transportationseismic perfonnance policy and goals established by the Governor.The plan shall include the timely seismic retrofitting of existingtransportation structures.

Status on January 26, 1994: "BridgeSeismic Retrofit Program,January 1, 1991."A schedule for project completion and costestimates was submitted to the Governor onJanuary 1,1991 in response to his ExecutiveOrder D-86-90 and to the Legislature inresponse to Senate Bill 2104. For the firsttime an attempt to segregate the 7,000 state·owned bridges into priority groups wasincluded in the report. This was for the

purpose of identifying those most criticalbridges that needed immediate repairversus those which would only be repairedfor purposes of reducing future damage. Aform of triage if you think of it that way. Itis estimated by Caltrans structural engi­neers, based on their past experience, thatthe number of state bridges actuallyneeding any work will be pared down to

approximately 4,500.

3 BOARD 0'" INQUIRY RECOMMENDATION No.5 ,,'OR ACTION BY THE DlRECrOR OFTHE DEPARTMENT OF TRANSPORTATION. Fonn a pennanent EarthquakeAdvisory Board of external experts to advise Caltrans on seismic safetypolicies, standards and technical practices.

Governor Deukmejian's Executive Status on January 26, 1994: In responseOrder D-86-90 of]une 2, 1990. "The to the Governor's Executive Order D-86-90Director of the Department ofTransporta- the Seismic Advisory Board was formed andtion shall establish a formal process whereby reported in the "Report To The Governorthe Department seeks and obtains the advice On Seismic Safety, August 31, 1990."Theof external experts in establishing seismic Department has established a Seismicsafety policies, standards, and technical Advisory Board of eight leading experts inpractices; and for seismic safety reviews of this field. Department Bridge Earthquakeplans for construction or retrofit of complex Engineering staff and management meetStructures. The Director shall transmit a with the Board quarterly to obtain theirsummary of this process to the Governor by approval of new criteria and solicit theirAugust 31, 1990." advice on future developments. Four of the

eight Board members were also members ofthe Governor's Board of Inquiry and one is

Ca/trans Response fO Governor's Board of Inquiry Recommendations 51

Page 63: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

the chainnan of the Engineering CriteriaReview Board for BCOC. Another memberis the Chair of the Seismic Research Advi­sory Panel. 10 date they have reviewedCaltrans Division of Structures' designprocedures, the seismic vulnerability analysis

algorithm and screening procedures, theseismic perfonnance criteria, the processutilized by thc Scismic Safety Peer ReviewPanels and many other aspects of ourearthquake enbrineering operations.

4 BOARD OF INQUIRY RF.COMMENDATION No.6 FOR AcnON BY TIlE DIRECf'OR OF

THE DEPARTMENT OF TRANSPORTATION. Ensure that Caltrans seismic designpolicies and construction practices meet the seismic safety policy and goalsestablished by the Governor:

Recommendation A. Review and revisestandards, perfonnance criteria, specifica­tions, and practices to ensure that they meetthe seismic safety goal established by theGovernor and apply them to the design ofnew stnlctures and rehabilitation of existingtransportation structures. These standards,criteria and specifications arc to be updatedand periodically revised with the assistance ofexternal technical expertise.

Status on Jat/uary 26, 1994: The AppliedTechnolo!.'Y Council was awarded a contractin 1990 to review and recommend changes tothe Caltrans Bridge Seismic Design Specifi­cations and criteria. That work is nearingcompletion and the final report with recom­mendations will be available by the end of1994. In the interim, many changes havebeen made to the Seismic Design criteria andprocedures as various research work iscompleted and resultsl recommcndations canbe incorporated into the specific~ltions.

Recommendation B. Institute indepen­dent seismic safety reviews for importantstructures.

52 The Continuing Challenge

Statllson Jl111l1ary 26, /994: We haveeng<1ged several Seismic Safety Peer ReviewPanels to review critical and major projectsand to comment on our bridge seismicdesign criteria and details. These PeerRevicw Panels vary in size depending on thespecific project. These panel members haveparticipated in detailed seismic safety designreviews of several major Caltrans projects inall areas of thc St:l.tc.

Projects that have been Peer Reviewedare selected by the Division ofStrueturesbased on size, complexity or some uniquefeatures and include the following:

I. 1-110 Transirway (Harbor Freeway)elevated viaduct in Los Angeles

2. 1-480 (Embarcadero Freeway Viaduct)double deck viaduct in San Francisco

3. 1-480 (Terminal Separation) MultipleLevel Interchange in San Francisco, atthe west end of the SFO Bay Bridge-forretrofitting

4. 1-280 (China Basin Viaduct) double levelviaduct in San Francisco

5. 1-280 (Southern Freeway Viaduct)double level viaduct in San Francisco

Page 64: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

6. 1-2801US 10 I(Alemany Interchange)multiple level interchange in SF

7. US 101 (Central Freeway) double levelviaduct in San Francisco

8. 1-215/) 0 Interchange in San Bernardino9. 1-5801l-980/SSR24 Interchange in

Oakland10. 1-480 (Terminal Separation) multiple

level interchange in San Francisco, at theWest end of the SFO Bay Bridge-forreplacemcnt structures

II. 1-880 replacement project in Oakland12. San Francisco-Oakland Bay Bridge

Recommendation C. Conduct avigorous program of professional develop­Illent in earthquake engineering disciplines atall levels of the oq,rani7,ation.

Status on January 26, 1994: The Office ofEarthquake Engineering was created in 1990by combining some smaller units in theDivision of StruCtures. The office is staffedwith several engineers with advanced de­grees, including 6 with Ph.D.s (a first inCaltrans Structures Division). We have hireda staff Seismologist to augment the existingstaff of engineering geologists. Wc havecreated a staff seismic design specialists at theSenior Engineer level in each of the 14Bridge Design sections. We have conductedextensive training in seismic design for over200 bridge design engineers. We have heldnumerous workshops between the seismicresearch communiry and our designers andconsultants to exchange the latest in seismictechnolob'Y'

A Bridge Seismology Committee wasauthorized by the Board of Directors of theStructural Engineers Association of Califor­nia (SEAOC) inJuly, 1991 and organizedshortly after that time. The committee ischaired by James H. Gates, Chief of the

Caltrans Office of Earthquake Engineering,and is represented by several Cal trans seismicspecialists and structuf<ll designers a~ well asSEAOC members from each of the four localassociations. Many of the SEAOC membersof this sub-committee are also members ofthe Peer Review Panels and the ATC ProjectEngineering Panel, so we have not beenworlcing in a void.

Recommendation D. Fund a continu­ing program of basic and problem-focusedresearch on earthquake enbrineering issuespertinent to Caltrans responsibilities.

Governor Deukmejian's Executive Order0-86-90 ofJune 2, 1990. "The Director ofthe Department of Transportation shallassign a high priority to development of aprogram of basic and problem-focusedresearch on earthquake engineering issues,to include comprehensive earthquakevulnerability evaluations of importanttransportation structures and a program forplacing seismic activity monitoring instnl­ments on tf<lnsportation stnlctures. ThJ;Director shall transmit a description of theresearch program to the Governor byAugust 31, 1990."

Status on January 26, 1994: This pro­gram was described in the "Report To TheGovernor On Seismic Safety, Au~,'ust 31,1990." The initial investment in bridgeseismic research was $8 million and isoutlined in the report. Subsc<)uently, theDepartment management has agreed to aproblem~focused seismic research program atan annual expenditure level of $5 million(approximately 1% of the Caltrans bridgecapital expenditure program).

Caltrans Response to Governor's Board of Inquiry Recommendations 53

Page 65: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Recommendation C. Perform compre­hensive earthquake vulnernbility analyses andevaluation of important trnnsponationstnlctures throughout the state, includingbridges, viaducts, and interchanges, usingstate-of-the-art methods in earthquakeengineering.

Status on January 26, J994: ByJuly, 1990DOS scaff had completed an initial "Vulner­ability Analysis" of the 23,000 state, countyand city bridges and produced a prioritizedlist from which to complete the program.The logical strategy was to retrofit or replacethe most vulnerable bridges on the list first,when this approach is possible. That listcontained approximately 7,000 state and4,500 city and county bridges which requiredfurther analysis and evaluation. To date 1,033bridges have been identified that requireseismic retrofitting.

Recommendation D. Implement acomprehensive program ofseismic instru­mentation to provide measurement of theexcitation and response of transportationstructUres during earthquakes.

Staws on January 26, 1994: The Depart­ment increased its annual support of theCalifornia Division of Mines and Geology(CDMG) Strong Motion InstrumentationProgram (SMtP) from $40,000 to $100,000.Working with the Seismic Advisory Boardwe have agreed to fund instrumentationfor 20 additional bridges at a one time costof $700,000.

Recommendation B. Develop a long­term strategy and program for the seismicstrengthening of existing substandardstructures, including the San FranciscoFreeway Viaducts, that considers theiroverall behavior, the degree ofseismic risk,and the impomnce of the structure to thetransportation system and to the community.

Status on January 26, 1994: This pro­gram was described in the report to theGovernor "Bridge Seismic Retrofit Program,January I, 1991." The importance factor isnow a part of the Seismic PerformanceCriteria adopted by the Department andapproved by the Seismic Advisory Board.The current scatus of the seismic retrofitstrengthening prOb'Tam is described in the"Bridge Seismic Retrofit Report" datedJanuary 7,1994.

5 BOARD OF L'fQUlRY RECOMMENDATION No. 7 FOR ACTION BY TIlE DiRECfOR OF

THE DEPART,\1LVT OF TRANSPORTATION. Take the foUowing actions for

specific structures:Recommendation A. Continue to

sponsor and utilize the Independent ReviewCommittee's technical reviews of the engi­neering design and construction proposedfor the short-term repair and strengtheningof the San Francisco Freeway Viaducts.

Stlltus on JanUIJry 26, 1994: The Depart­ment has continued to utilize the services ofthe independent Seismic Safety Peer ReviewTeams for over four years and the work isnearly complete. Most reconstruction iseither under contract or scheduled forcontract award this calendar year.

54 The Continuing Challenge

Page 66: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

and has a consultant preparing seismicretrofit plans for the bridge. Caltrans has noinfonnation on seismic instrumentation,however.

Recommendation C. Instinne indepen­dent seismic safety reviews for importantstructures.

StlltUS 011 Jamlllry 26, 1994: It is knownthat the Golden Gate Bridge and Transpor­tation District has conducted a seismic safetyreview of the bridge.

Recommendation D. Conduct avigorous program of professional develop­ment in earthquake engineering disciplines atall levels of their orh"3nizations.

Statusrm]omlo1'Y 26,1994: It is notknown to Caltrans whether any of theagencies has conducted such a program.

6 Bo.o\RD OF L"QUlRY RECOMMEI''DATION No.8 FOR AcnON BY THE TRA..,....SPQRTKI10N

AGF..NCrr,S AND Disnucrs. Agencies and independent districts that areresponsible for transportation systems, rail systems, highway structures,airports, ports and harbors-should:

Recommendation A. Adopt the sameseismic policy and goals established by theGovernor for State transportation structuresand implement seismic practices to meetthem.

Status on January 26, 1994: It is difficultfor Caltrans to determine whether any ofmese agencies adopted policies and goals orwhether they implemented practices to meetthem. Caltrans has no authority to requirethese agencies to cOlilply with the directive.

Recommendation B. Perform compre­hensive earthquake V\llnerability analysis andevaluation of important transportationstmcrnres (e.g., the BARl' 'nans-Bay Tubeand Golden Gate Bridge) using state-of-the­art methods in earthquake engineering, andinstall seismic instrumentation.

StlltllSonJall11flry 26,1994: It is a knownfact that the Golden Gate Bridge andTransportation District has conducted aseismic vulnerability analysis of the Bridge

Caftrans Response to Governor's Board of Inquiry Recommendations 55

Page 67: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

This page has been left intentionally blank.

Page 68: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Appendix B

Seismological and StrongMotion Description of theJanuary 17, 1994Northridge Earthquake

SeismologicalCharacteristics

'meJalluary 17, 199-t lIIain shock of theNorthridge e-arthquakc SC(lucncc (/\I\\':6.7)had;l hYJloccnrr:.lllo<""arion of H· 13'1\',liS" 31'\\1 with:J flX-:l1 depth of :,!>out 18 kill(sec Figure B-1). 'ne hypocemrallocatiollsofhumlrcds of aftershocks have fixed thefault plane as dipping 42" to the south underthe S:m Fcm:mdo Valle}' :lnd striking about112" (sec Figure IJ-l). The :mal}'sis of thewave forms in this c:lrthl]u;lkc at nC:lr :lTlddistant St:ltiullS. together with the measuredgeodetic disphtcCl11CIll.~, indic:llCS that therupmrc rncch:misl11 011 this f:lult, called theNorthri<lgc fault, was :1 thrusting Illcch,mislllwith Gr:lll:Hla Ilills ,lT1d the S:lll FCI'!l:lIlcloV:lllc}' on the h:lllging w:lllllloving northeast.

The :lrea of rupture is inferred to beapproximately 16 km alnng the strike .,nd 15kill in the direction of slip. Slips across thethrust f;mlt h:Hlm:ui111ul11 values of ,Ibout 3.5Ill, with all average slip of ahollt" 1.5 lI\ overthe whole faull'cd surface. No groundrupture of the thrust f.,ull was observed :llthe surface and the indication is th.,t therupture plane lost its structul';.ll coherence ata depth of ahout 5 kill 011 the north :melabout 10 km Oil the south: tlUlt e1ispbce­ment's al:)()\'e 5 kill in depth were perh'lpstr.lIlsferred til Ilumerous surfici.,1 r.lUitS in theS:lllt.1 Susan:lh Mountains. Because theI:mlting did not appe;lr:1t thc surface, theseismic source is termed a "hlind-thrust." Itthus resembles the \Vhinier Narrows seismicsource that was located undcr east LosAngeles and ruptured on Octoher I, 1987,(AI\\' '" 6.0). Thc seismograms indie..."ate th.,tthe release of energy was not uniform alongthe slip surfuee; there was at least one

secondary release of eneq,'y initiated about 3seconds after the start of the first rupnlre atthe hypocenter.

A number of moderately large after­shocks were produced by additional fuultingin the following weeks, the hugcst on .\Iarch10, 1994 h:H'ing;l mab'llil'Ude of 1\lw '" 5.3,occurring on :1 subsidiary f;mlt with slip inthe upper 5 kill of the cmsl.

An eartlHJuake of this si7-c was notunexpecred in the Los Angcles and SanFernando h:lsins whcre lTlany cap:lble faullSof various types :Irc m:IPllC(\ on the St:lt'efault tll<lp. In addition, it h:ls becn de:lrlydemonstratcd by the 19H7 \·Vhinicr Narrowse:lrthquakc th,l\ suust":llnial carth(l\lakes canbe generated by slip on thrust faults thaI" donot havc de,11" cxpressions:1t the surface. 'rhenetwork of blind-thrust faull'S througholll' rheregion has been established by dcep drilling,llld by geological rccollstnlctiol1s of lheregional tectonic defonnatiolls. Thescmcthods, howevcr, do nOl USU:ll1y lead lOunique locations :llld dimcnsions of suchburied faults. Thc par;1111ctcrs that define thefault rupture pr(){lueing the):m\l:lry 17Northridgc earthlJllakc were not previouslypublished.

Strong Motion RecordingsNumerous strong motion instruments

had heen placed by the Californi'l StrongMotion Instnllllent'ltion rrogram (CSI\ 11 P)and the U.S. Geologie..-al SUl""I'ey. One­hundred and thirty-two of these Instrulllentsplaced in thc free field arc within a 100 mileradius of the fault rupt'ure area, (Shakal ct ai,199-1-; USGS, 1994-1. The main shock yieldedrerordinb"S at 193 sites; of these, 116 werefree+field rerordinb"S, 77 were in engineered

Seismological and Strong Motion Descflptlon of the Northridge Earthquake 57

Page 69: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

SAN GABRIEL MOUNTAINS

0.0+

2.0+

3.0+

loIAGHITUDES

•o <4.0+

5.0+

6.0+

116"".

Angeles

Los ....

".

~"'/.>,..'" '"

;,.: ~ '~. , ' '", oe _ til'"

10", '.,r1t.t

y ___. oil....,.. ."Ii .. ...

'ko 1"1 " -.~

. ,,'---~-....-::- • i\'.- .'

..:~ ..~"•

".

".

MaUbu

--_/---".

10 Klol111""",1

I

".

".

,,-

Figure 8·1. Location of hypocenters of the Northridge earthquake and aftershocks. The edge ofthe aftershock distribution rends to locate the extremes of the fault rupture in the main shock.Figure B-2 shows the depth cross section for the two rectangles A and B. (data provided byCaltech and USGS. Pasadena, CA; figure provided by Egill Hauksson, Caltech)

o.•..

SGF C'

.-

NHF VF SFF

... ! .... ~,"..w..:-

SGF Bt:NHF SSFHrssrNHr

... ' • • • ... I:. ••'0'_.' 0 ',• '. , J• ." • !"'" lo '.

. ',..' '''if''-"-11:'-'.'; .··.N . .. 0

.* ',' ~ .. ",..*~

A

%~~

wc -20

o

--.-10>~-

-'0o 10 20 30 0 10 20 30 0 \0 20 30

DISTANCE (Kt.4) DISTANCE (Kt.4) DISTANCE (Kt.4)

Figure B-2, Cross-sectional projections of the locations of aftershocks for the rectangles shown in Figure 8-1.(data provided by Caltech and USGS, Pasadena, CA; figure provided by Egill Hauksson, Caltech)

58 The Continuing Challenge

Page 70: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Recordings at Santa Monica

2010

,Vartlcal Component

1

, , ,-1.0

o

00

1.0

0.'

.Q.'

Vertical Component

0'1--.-1·-------------1

Recordings at Nawhall1.0,--"""'--,--":"'---r--~~-__,

0.0

·1.0L--~__L'--~--_,L--~--..Jo m ~ ~

.Q,I---'----------------1

-I.O,--...,...---,--.,-----,---~-___, -1.0 ,--...,...--,------r-----__,East-West Component

~ o.s1---.,-j--,----------------1cgf 0.0J!

! .a51---..11----------~----__1

East-West Component

~0.51------1-------------1c.!!f 0.01--......J!~~ .a.S

_, 0 ~--.L---~'=_--'---~'=_--~--:'o ~ ro ~

-1.0!:-----'::---'------::-----_:o ~ ~ ~

1.0,--.,-----,---~-_.----__, 1.0,-------,-----_.------,Nortn-South Component North-South Component

051---+11--,------------1 0.1-----------------1

00 001--"tI

.Q.I--'--'l...--'-----------I .Q51---------------I

-1.0 !:---'---'::----'----::--~--_:o m ~ ~

Tlma - Seconds

-1.0 !:---'---'::----'----::--~--_:o m ~

Time - Seconds

Figure B-3. Time histories of the ground motions recorded during the Northridge earthquake at the Newhall andSanta Monica sralions of the California Strong Motion Instrumentation Program. (data from Shakal el aI, 1994)

structures, including motion recordings fromscvcn bridgcs [Shaknl ct ai, 19941. None ofthe seven bridgcs that were damaged wcrcinstrumented. Fib~lre B-3 shows timehistorics for two ground motion stations,Newhall (near the l-5/SR-14 failure) andSanta Monica. These records show that theduration of strong motion was abolll 9seconds (sec Fib'l1re I-I for :lpproxilll:ltelocations).

There was, initially, an inference in themedia that every where this earthquakeproduced unusually strong ground motions.The record at the "T.1l7.ana site, for example,shows extraordinary duration of strongshaking lasting for about 8 seconds, with anumber of peaks measuring Ig. It is likely,however, that there is a special gt:olobrical ortopographic circumstance that explains thehigh values at this site. Stmcnlres near thesite do not show damab'"C consistent with suchhigh readings. In the 1987 Whittier Narrowse2rthquakc, the Tan.ana site recorded peak

accelerations about 12 times larger thanthe average of a number of nearby sites.Other recordings in the region of strongestshaking suggest maximum values somewhatless than.9g.

Peak horizontal acceleration (PGA) forfree field sites is a comlllonly used measure­ment to characterize the intensity of groundmotion at a site. Figure B-4 shows thevariations of PGA with distance from thefAult rupnlrc surfJcc for rock and soil sites.The lines on this figure arc estimates for aMw=6.7 earthquake using the attenuationrelationship developed by Idriss 119911. Thecentral line is the median estimate, while theupper and lower lines arc the median plusand minus one standard deviation.

Further snldy ofsite charneteristics willbe required to understand these variations.\Nh.jle the recorded values arc somewhathigher than the estimation curves, they arcnot un typical of the scatter observed for

Seismological and Strong Motion Description of the Northridge Earthquake 59

Page 71: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

t

•.-

Median & Medlan:l: a UsingAnenluallon Relatlonshlpfor ROCk and Stiff Soli Sitesby ldrls.. (1991)•••

Motions recorded duringlhe 1994 NorthridgeEarthquake

Average of the twohorizontal components• Rock Sites

Soli Sites

O.Q1'---------::-----------------------=::------''''----------'10 Closest Olslance 10 the Rupture Surface _ km 100

Figure 8·4. Comparison of recorded peak horizontal accelera/ions with those calculated using the atrenuation relationship ofIdriss {1991} for rock and stiff soil siles, The measured values are higher than the median eslimales using commonly acceptedattenuation relationships; the median eslimales are about 20% higher. but well within statistical expectations. Soil and rock sitesuse separate symbols. (data from Shakal et ai, 1994)

other carth<]uakcs. Systematic deviation:lbove :lnd below the medi:lll :lltciluationcurvc has been observed in sevcr.tl othereartll(lUakes.

There was an early suggestion thatvertical accelerations were system:lticallyhigher in the Northridge eatthquake thanwas expected based on earliet recordings ofsimilar-si7.ed earthquakes in California. Thenormal assumption in building code andengineering practice is that the vertical tohori7.0ntal PGA ratio is 213. Figure8·5 shows that the average trend for thisratio for all distances is approximately 213 inaverage for small distances and less forgreater distances.

An importanl method for characterizingthe seismic excitation used in design is theresponse spectrum. Nonnalizcd sl)CC[r,J arc

given in Fib'1lre 13-6 for ~l group of soil sitesso that the general nature of the groundmotions C~lJl be compared. These show aconsiderable SC:llter as is qUite typical forsuch plots. The ARS spcctmm lIsed byCaltnns for design at sites with 10 to 80 feetof alluvium are also shown. Nhllly of the sitesused probably do not fall in this category;when further infonnation is available on siteconditions where recordings were written.For low periods, the recorded spectra aresystem~lticallyhigher than the ARS spectra;at medium to high periods, in excess of 0.5seconds, the ARS spectmm is comparable tothe average of the soil site spectra given.

60 The Continuing Challenge

Page 72: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

'60160

II

II

''0

Mollons recorded during the1994 Northridge Earthquake• Rock Sites• Soli Sites

••

60 00 100 120Closest Oistanee 10 the Rupture Surface - km

• III• Ratio:: 2/3 •

~~.~---------~---------

•II

2.0

18II

c~

1.•e•"uuC 1.4;;c

~0 1.2~

~••'1; 1.0

Iiii ••u 08uc

" ----"< 0.6 - •• ••>~••~.2 0.4

"~0.2 -

0.0a 20

F/gu,.. B·S. Variations of the ratio of peak vertical accelerations /0 the average peak horizontal acceleration versus the closesldistance to the fault rupture surface. (dala from Shakal et ai, 1994)

Seismological and Strong Motion Description of the Northridge Earthquake 61

Page 73: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

s,----------,--------r-----,------,------,--------,

L.egend

Spectral Shapes for MotionsRecorded at Soil Sites During the

Northridge Earthquake

4f--tH-++---j------+-----H

3.."~~•~•E0<~0e~•m

2

Spectral Shapes Used byCaltrans for Soil Sites(10 to 80 It of Alluvium)

spectral damping = 0,05

o~~~~±::;~~0.0 0.5 1,0 1.5 2.0 2.5 30

Period - seconds

Figure B-6. Normalized standard response spectra for selected soil siles as measured in the Northridge earthquake,The ARS spectrum used by Gal/rans is given for comparison,

62 The Continuing Challenge

Page 74: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Table B-1. Locations of bridges for which strong motion records were obtained in the Northridge earthquake from the C$MIPprogram. (data from Shakal et ai, 1994)

Distance from Maximum accelerationlocation Highway epicenter (km) Site type Ground Structure

Los Angeles 1-10/405 Inter-change Bridge 22 deep alluvium NA 1.83g

Los Angeles Vincent Thomas Bridge 57 alluvium .25g .65g

Devore 1-15/215 Inter-change Bridge 104 deep alluvium .08g .24g

San Bernardino 1-10/215 Inter-change Bridge "" deep alluvium .13g .47g

Beaumont 1-10/60 Inter-change Bridge 146 alluvium over granite .049 .09gbedrock

North Palm 1-10/62 Inter-change Bridge 181 deep alluvium .02g .11gSprings

Strong Motion Records fromBridges

Strong motion records were obtainedfrOlll six bridges during the Northridgee;lnhqu;lke ;It dist;lllces mnging from 22 km(14 mi) to 181 kill (115 rni). T:'lble B-1 liststhe bridges, their locations and maximumaccelerations observed. Only Olle instru­mented bridge was in the region of strongshaking (the 1-10/405 Interchange), and onein the region of moderate shaking (VincentThomas).

The most significant record is from the1-1011-405 interchange. It is a 1037 feet long,curved concrete box girder structure. It hasnine single-column bents and two open­seated abutments. The bridge was retrofittedin 1991 with steel jackets on some columns.Installation of instruments was completed,fun<lcd by Caltmns, just before the earth­qU;lke. A peak acceleration of 1.83g wasrecorded at the box girder Ilear the west;lbutment. This bridge is located about 4miles west of the section of the 1- I0 Freew;lythat collapsed. 'rhe bridge did not suffer anysignificant damage.

These arc the instrumented bridges thatrecorded the Northridge earthqu;lke. The1-10/405 record is of great importanc~ sinceit is for a retrofitted bridge near the epicen­ter. The focus of the instrumentationprogram is on complex geometries ;lndretrofitted bridges. Some considerationshould be given to instrumentingllnretrofitted bridges and more commontypes of bridges. Records from these types ofstructures would provide a basis for under­standing how retrofitting is affecting bridgeperformance. C;lltr.ms has steadily increasedits COlllmitment to the CSMJP program forbridge instnll11cntation. The increase from$40,000 in 1988 to $1,000,000 per ye;lrstarting inJuly, 1994 should give the oppor­tlillity for significant improvements. Milc afew additional bridgcs in northcrn ;lndsouthern Californi;l have been instnllllentedto d;lte, much of the instrumentation effort isyct to be completed, particularly for tollbridges.

Seismological and Strong Motion Description of the Northridge Earthquake 63

Page 75: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

This page has been left intentionally blank.

Page 76: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Appendix C

San FernandoEarthquake, 1971

The 197 I San Fernando maill shock wa.~

one of the most important earthquakes in thedevelopment' of modern seismic design andretrofitting pr:lcticcs-for bridges as well asfor buildings. The c:lrthquakc (Mw=6.7) ofFebruary 9, 1971, C311scd 53 de'lths and ancsti11l;1tcd dam:1gc of less than $1 billion in1971 dollars.

By comparison, the Northridge eanh­CjU;l).;C, with magnitude 6.7, occurred in apopulated area, caused 56 deaths, andestimated &Hmgc of $15-30 hillion. Each ofthese earthquakes caused major damage andthe collapse of several buildings, bridges, andother stTIlCl11res. Each event has had and willcontinue to have a major effect on theimprovement of the seismic design andretrofit procedures for Cal trans bridges inCalifornia.

Previous design practiccs wcrc dClllon­sU'ated to be deficient in the San Fernandoearthquake. This section reviews theearthquake's impacts on bridges and howCaln'ans responded to the discovery th:lttheir seismic design practices needed revi­sion. The 1971 S:ln Fern:lndo earthquake isof gre:lt import:mcc in the snl(ly of thebehavior of Caltnms structures because itssize .md fault type is similar to theNorthridge e:lrthqllake, and it occurred on :lneighboring thrust-fault to the llorth of theNorthridge Emir rupt'llre.

Seismic Characteristics ofthe San FernandoEarthquake

The San Fernando earthqll:1ke epicenrer\\,:lS in ;1 lightly populated area at the edge of;1 San Fernando Valley. It occurred at 6:07AM 011 February 9, 1971, with a hypocenter

oflatimde 34· 24'.ON, longinlde liS· 23'.7Wand focal depth of about 13 k1l1 (Figure C-I).Its moment magnitude was calculated atMw = 6.6. Surface faulting occurred j'n theSan Fernando Valley and in the foothills ofthe San Gabriel Mountains, particularly inSylmar. The surface faulting had both thrustand left-lateral motions. The strike variedfrom place to place with a mean value N70·W :ll1d a dip of 45·. The totallel1f,rth ofsurface faulting was approximately 15 kill,with some bteral offsets. [n striking contrastto its 1994 seismogenic neighbor, the slipfault in this casc dipped to lhe north underthe San G:lbriel Mountains.

Although the San Fernando c:lrthqu:lkewas of 1lI0der,lte magnitude, accclcrographmeaS\lrements and obsen>ed damage toengineered structures indicate that theintcnsity of surface ground shaking in theimmediate vicinity of the epicenter wasconsiderably higher than predicted at thattime for such an event. 'rhe highest recordedpeak ground accelerations (PGA) measuredby the San Fernando earthquake, equal to

1.26g horizontally and O.72g vertically, wererecorded on a rock ridge Ilear the abutment.This PGA was significantly greater than anyrreviolls recordings. At the sites of lhe fivecollapsed freeway structnres, the horizontalPGA levels were estimated to be approxi­mately O.6g, which was considered t"O be nearthe upper-bound value at th:lt time.

Damage to FreewayStructures

1-5/1-210 Interchange. The mostdramatic of all bridge damage caused by theSan Fernando earrh{IUake occurred at theGoldcn Statc Frcew;lY (lnterst:ltc 5) and

San Fernando Earthquake. 1971 65

Page 77: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

....-.

VIII·XI

VII

-.-Figure C·1. LOC8tlOO of the 1971 San Fernando earthquake. Roman numerals indicate the Modified MercalliIntensity; see Figure 3-1 for description 01 the intensity levels.

Foothill Freeway (Interstate 210) Inter­change. The highest ovcn:rossing at this site,which l.::Jrricd southbound tr,lffic fromFoothill Freeway onto the Golden StateFrccw:lY collapsed during the earthquake(Figure C-2). The box girder deck of thisovercrossing (approximately 770 feet inlength) was supported on six piers and endabuoncnrs. The twO mOSt southerly pierswere supported on spread footings which, inmm, were supported on driven concretepiles. The four most northerly piers weresupported directly on a single round pile castdirectly in a 6-f001 diallleter drilled hole.The box girder h:1(1 one expansion joint ncarmid-crossing in addition to those at theabutmcnt....

The two principal causes of collapse ofthis particular overcrossing wen~:

66 The Continuing Challenge

I. The large vibratory motions induced inthe superstruchlre by the high intensityground motions.

2. The relative ground displacements thatoccurred between abutment SUPlxlrts.

Considering the locations and orienra­dons of the abutments and expansion joints,and considering the general foml of thedeck curvature in plan view, it is quiteapparent that the deck was highly con­strained against large displacements in alldirections except in the westerly direction.As the vibratory lIlotions of the dcck builtup with considerable bi3s in the that direc­tion, the deck scp3r;ltcd at the celltrallylocated expansion joint, allowing one end ofa span to fall off its support and initiatecollapse of the entire StniCture.

Page 78: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figure C·2. The most dramatic of al/ bridge damage caused by the San Fernando earthquake occurredat the Golden State Freeway (1-5) and Foothill Freeway (1-210) Interchange. The highest overcrossing atthis site, which carried southbound tralfic from Foo/hill Freeway onto the Golden State Freeway,collapsed during the earthquake. Note the similarity 10 the performance of the 1-14/1-5 connectionovercrossing in the Northridge earthquake, Figure 4-2. (photo: National Earthquake EngineeringInformation Service archive)

San Fernando Earthquake, 1971 67

Page 79: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

-Figure C·3. Columns of rhe San Fernando Road Overhead suffered heavy damage during the San Fernando earthquake Notethe Similarity to /he perfOlmance of /he /. f /8 Mission and Go/hie undercrossmg In the Northridge earthquake, Figure 4-6.(photo: Na\lOna! Earthquake Engineering In'ormation Service archive)

1-5lState Highway 14 (mel·change.Another llrall1:l1ic coll~IPSC O<:l.:urrcd duringthe c'lrth{]uakc .It the Golden Stale Frccw<lV(Intcrst':lte 5):md S(;ltC Ilighwa)' 14 Intcr- .change. where .;;c\'cral mcrcrossille'S werestill under constnlction. A long (approxi­mately 4OO-fl) ccntml S<.'Ction (If the highesto\'crcrussing (completed before the c:lrth­

quake) CUIl:lpscd. This IOllg, prestressed(:oncretc section of bridge deck W:lS sup­ported :It each end 011 be:lring p;lds ;11e-xpansion joillL'i and by :J single columnst:mding 160 feet high. The initial Cluse ofl.:ollapsl.: W;IS the brge reblive deck dispbce­ment at one eXI';msion joilll, which :l1Jowedthe hox girder to Eill off its support andinitiate l.:ollapse of rhe enrire central section,Both cantilewred portions of rhe del.-·k, ;lS

they hinged down, broke off;11 the lOp of rhecentr.ll colulIln, which ;1110\\00 them to f;tllalmost straight down frolll their origilla)

68 The Contlnumg Challenge

positions. The cclllral colulnn then fell tothe west. This intereh;mge W;lS d:nnag"edag;lin in the Northridge e:lrtlul'lake (secSection 3).

San Fernando Road Q,'crhe:ul. TheSan Fern:mdo Road O\erhe;ld suffered hea",d;llllag~during the San Fern:mdu earth· .qU:lke. One Sp:lll crossing the SouthernP;lt:ific IbilfOad fell frulil its bc:lring SUPIHlrt;n one end, c:lUsing it to co1h11:tSe (FigureC-3). The deck ofother Sp;IIlS remained inpLlCe; though their supporting l.:o)unms wereIXldly d::unaged. M;my of lhe st iffer l'ollllllllSsuffered shear failures, while the lIlorellexilJlc coluillns suffere(l hC;l\'v tlexllr,lldamagcs:1( their tops. Thc lIla'in reinforcing1I:lrs :It these )oc;niolls hucklcd dlle to thehigh ("()Illpressi\'e forces produced hr lheearthquake. Once the concrete CO\er.lg..·~p:Jlled ofT the baTh. the ric'i were in:1dellu:nt'

Page 80: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

to provide their needed lateral constraint andto provide confinement to the core concrete.Other forms of damage to the San FernandoRoad Overhead included the dislodging ofsteel rocker bars from their support assem·blics. These failures were caused by the largerelative displacements produced betweendeck and support.

Other Damage. Other freeW'ay struc­tures that suffered light to heavy damagesduring the San Fernando earthquake were:

• Foothill Boulevard Undcrcrossing at theFoothill Freeway

• Roxford Street Undercrossing at Foot·hill Freeway

• Polk Street Undercrossing at FoothillFreeway

• Hubbard Street Undercrossing atFoothill Freeway

• B1edsol Street Undercrossing at FoothillFreeway

• Tyler Street Pedestrian Overcrossing atFoothill Freeway

• Culvcrt undcr Foothill Freeway• Via Princessa Undercrossing on State

Highway 14• Santa Clara Overhead Crossing on State

Highway 14

The types of damages to these structuresincluded:

I. Flexural yielding and crushing ofconcrete at tops ofcolumns.

2. Shear fr:Jeturing of columns followed bycrushing ofconcrete causing mainreinforcing bars to buckle outward.

3. Fracturing of piles supportingabutments.

4. Wing walls broken away fromabutments.

S. Differential settlement of soils bchindabutments.

6. Flexural cracking in diaphragmabutments.

7. Breakage of concrete due to pounding atexpansion joints.

8. Downward slippage of concrete apronsat abuonents.

Design ImprovementsThe 1971 San Fernando earthquake

experience, made it clear that the freewaystructures then existing had serious deficien­cies. The damages caused by the earthquakepointed out the need to improve designdetails as follows:

I. Expansion Joints: Collapse of highovercrossings was initiated by bridgespans falling off their supports atabutments and expansion joints due to

excessive displacements of the spansrelative to their bearing supports atexpansion joints and at abutments.These needed to be widened to providemore effective ties across expansionjoints, and eliminate expansion jointswherever feasible.

2. Columns: Inadequate ties, both in sizeand spacing, contributed to shear andflcxure-type failures in the columns.Design derails, particularly the size andplacement of reinforcing bars and tieshad to be improved. Such changes werecritical to s:Jtisfuctory perfonnance undermaximum seismic loading conditions.

San Fernando Earthquake, 1971 69

Page 81: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

3. Column Caps: Damages indicated alack of teinforcing hars tying columncaps to their respective box-girder bridgedecks. Desi!,1'J1 details, again, had to beimproved.

4. Column Foundations: Failures at thebases ofcolumns using single cast-in­place piles or spread footings with drivenpiles showed inadequate anchorage ofthe main reinforcing bars. Correctivemeasures had to be taken so that suffi­cient anchorage was provided to developthe full strength of the main reinforcingbars.

5. Abutments and Wrng Walls: Abut­ments and wing wall failures caused byexcessive dynamic forces transmitted bybackfill earth pressures and seismic deckforces showed the need to strengthenthese elements so that they wouldperform satisfactorily under maximumexpected seismic conditions.

70 The Continuing Challenge

Page 82: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Appendix 0

Loma PrietaEarthquake, 1989

In the C":lIWltion ofrhc dalllage (0structlJrcs, p;lrticularly freeway overpassesand bridges, the second carth(]llake forcomparison is the Loma Prieta earthquake ofOctober 17. 1<)89, which O<.x:urrcd in centralC:llifornia, 60 miles south of San Francisco.The Lama Prieta earthqwlkc (1\'1\\1:7.0)occurred in a lightly popul:ncd :lrca in theSam:! Cruz. lIlount:lins, caused 62 deaths, andestimated d:lImge of$6-7 billion in 1989dollars.

The Loma Prieta canh(IUakc l.":luscd themost damage of :lIly California carthqu:lkcsince the 1906 t.:arthquake. Northridge nowhas this distinction. Lama Pricf:1<.:ausedextensive damage to seven double-decker\,j:lducrs, including the C)'press Vi;\duCI,which f:lilcd. l):Il11:1gC dosed the San Fr;lll­cisco-Oakl;lIld Hay Brid!,'"C for onc 1I10mh.

Seismic Characteristics ofthe Loma Prieta Earthquake

The epicenter was located approxim:ltely16 km northe;lst of the city of S;lIlt;1 CnJz.The f()(,:;11 depth was :Ipproxim;ncly 18 klllbelow the surface, with a f.lUlt pl;lI1~ dippingabout 10· frolll the vertie.-Ollw the west.''''hi Ie it was origin.llly thought to h:l\'eoccurred hy slip of the San Andreas f:lUh,sollle geologists and seismologists nowbelieve th:1t it ot:curred by rupture of a f.lultto the west of the San Andre.--::as. An:llysis ofseismograms showed that the ruprure spread;lhout 20 km to the north and 20 km to thesouth, with seismic w:wcs f':uliating for8 seconds from the moving slip fronts. Likethe Nurthridge earthquake source, the LamaPriet;l rupture :llso propag>ated toward theearth's surface but stopped ;It .1 depth ofapproxilll'ltcly 5 k111. The direction of slip

was of oblique nature with a right Literaloffsct of :lbout 1m and a vcrtical offset ofabout 1.5 Ill. [t C:lll he termed a blind obli{]lleslip earthquake source.

Impact of the Loma PrietaEarthquake on Bridges

Only a Slll:lll percema!,'C of the hridges inthe area susClined any earthquake damage at:Ill. Moreover, IIlOSt of the hridf,'l..'S damaf,'C(!in this l.-::arthljuake were conStructed before1971, before constnlction sr:md;lrds \\ crestiffened to reflcct lessons learned in rhe1971 S;}n FeTll;lndo eJrthquake. The b'Tl.':ltestdamage during the 1.,0111a Prieta earthquakeoccurred to older structures on soft b'TOund.

Caltrans DistriCt 4, whose jurisdictionapproximates the ;lrea of grcatest Lorn:1Pricta e'lnhqll:lkc damage, is rcsponsihle for\ ,896 st'lte bridges, of which C) I (4.8 percent)incurrcd sOllie degree of d;11l1age (mostlyminor) during lhc c'll'lIu\llake. Structurald.ullage or the potentbl rhre;1f to Iluhlit's;lfery was suOicicmly serious in the (;:lse of13 St1.tC bridges lh:n they wcre dosed totr.:affic for some rime. rElble D-I lists rheCaltralls bridges th;}t Sllsra.ined majordamage.

The le\'c1 of ground sh:lking in the LomaPrieta earthquake was (for most bridges inthe nay Area) smaller in Ixlth duration andimensiry than would be expeCted in larb'Crand closer c:lrthquakcs. Morcover, theduration W;}S not sufficiclll to excite all of thediffcrent llI(){lcs of the Bay and Goldcn GalcBridges th:ll would be cxcited in a longer­duration cvent, nor W;}S the level of sh;lkingsuflil'icntly closc to th;lt expected in Ill;ljor

Loma Prieta Earthquake, 1989 71

Page 83: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Figure D-1. Isoseismal map of the damageimpacts of the Lama Prieta earthquake. See

Figure 3-1 for the definitions of MMllevels. oI

10I

20I

30 MilesI

6........_~..

6 6Monterey

6

6

6

VI

6

_ Collapsed Portion

iJ ~~.f. j~~y~. ~S'" 241h 28th 32nd 34th ~ _"",e

20th Coi'l1 :--.....-:::J~

96 & 97Bent 27 Bents 45. Bent 63

46 & 47

"h

Figure D-2. The Cypress Viaduct showmg the extent of the collapse durmg the 1989 Lama Prieta earthquake. (Hausner 1990)

earthquakes. Lama Prieta was, then, alimited test of the strengths of bridgeclements.

The most tragic impact of the earth­quake was the life loss caused by the collapseof the Cypress Viaduct (Fibrure D-2), whilethe most disruption was caused by theclosure of the Bay Bridge (Figure 0-3) for amonth, leading to costly commute alterna-

tives and probable economic losses. Inaddition, some of the steel rocker bearingssupporting the navigator spans of thc SanMateo-Hayward Bridgc failed. Any onc or allof these could have led to catastrophicdamage if shaking had been longer or morcintense.

72 The Continuing Chalfenge

Page 84: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

_ San Francisco

1-·--50---

FIgure D-4. Location of the SanFrancisco freeway viaducts damaged

in the 1989 Loma Prieta earthquake.The shaded areas are fitled portions

of the bay within the originalshoreline. (Hausner 1990)

-)1\e Loma Prieta earthquake was, forthe San Francisco freeway viaducts, a minor­ro-moderatc earthquake in ground lIlotiuns.The via(luets, however, suffered majordamage. These viaducts (Emhan.-aderoFreeway, 1erminal Separation Viaduct,Celltl'"JI Viaduct, China Basin Viachlct,Southern Freew:1Y Viaduct and AlelllanyViaduct) in San Francisco (Figure D-4) wereall buill with the same technology used forthe Cypress Viaduct and arc the only struc­tures in the state of this design. All thefreeway structures, with the exception ofthe Alemany Viaduct, were damaged duringthe earth<luake and subsequently closed rotr"Jffic.

Bridges maintained by loc"3l govern­ments also incurred damage, though none ascatastrophic as that sustained by some of theCalrrans stmctures. A partial survey byBoard of Inquiry mff found thai at leasl43 locally maintained strucrures in theearthquake area were damaged, of which atleast 5 were closed to traffic for some time.

one, however, collapsed. Reports frompost-earthquake reconnaissance te:llnsindicated thal1Tlost local bridges perfonnedrem:lrbbly well.

FIgure D-3. The west sections of theSan Francisco--OakJand Bay Bridge fellduring the Lorna Prieta eanhquake.The upper and lower closure spans felJwhen the bolts attaching the east(right) truss-span were severed and itmoved to the east. pulling the linkspans off their western supports.(Hausner 1990)

~""~_... CentralV"duct

China Basin....,.,,k-J I ""_-- Viaduct

Southern FreewayViaduct

AlemanyV~uct

Lorna Prieta Earthquake, 1989 73

Page 85: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Table D-1. Cal/rans bridges sustaining damage greater than $100.000 during the Loma Prieta earthquake, Conditions are as ofMay I. 1990. (Hausner 1990)

Santa Cruz CountyStruve Slough Bridge(SR 1)

Cypress Street Viaduct(1-880)

Name of Bridge Location Description of Damage

Bridges Closed to Public Traffic after the Earthquake:

San Francisco-Oakland San Francisco Bay Upper and lower closure spans at Pier E9 fell; spans at Pier E23Bay Bridge (1-80) (Alameda Co.) were near failure; concrete pedestal base of Pier E17 cracked;

connection bolts at Piers E-17 through E-23 damaged; opened fortraffic after one month; 1 death and 12 injuries.

Oakland (Alameda Co.) Collapse of 48 bents, causing the upper roadway to collapseonto the lower roadway; 41 deaths and 108 injuries with 1subsequent death; demolished, reconstruction uncertain.

Extensive collapse of the "twin~ bridges; opened on January 25,1990 after reconstruction.

West Grand AvenueViaduct (1-80)

Southbound connectorover-crossing (1-980)

Mora DriYeoyer-crossing (1-280)

Central Freeway Viaduct(US10l)

Southern Freeway Viaduct(1-280)

China Basin Viaduct(1-280)

Terminal SeparationViaduct (1-480)

Embarcadero Viaduct(1-480)

Route 92/101 Interchange(US 101)

San Mateo-Hayward Bridge(SR 92)

Port of Oakland(Alameda Co.)

West Oakland(Alameda Co.)

Santa Clara County

San Francisco

San Francisco

San Francisco

San Francisco

San Francisco

San Mateo County

Between San Mateoand Alameda Counties

Damage to bents, columns and earthquake restrainers;open to traffic after seyeral days.

Damage to two outrigger bents; opened on October 23, 1989.

Damaged column requiring reconstruction; opened to traffic aftera few hours.

Damage to bents and columns; retrofit required; portions are stillclosed.

Damage to bents; retrofit required; still closed to traffic.

Damage to bents; retrofit required; opened to traffic after 6weeks.

Damage to steel span bearings; retrofit required; still partiallyclosed to traffic.

Damage to bents and columns; retrofit required; still closed totraffic.

Damage to bearings, expansion joints, footings, and columns;opened to traffic after 2 weeks.

Failure of steel rocker bearings; opened to traffic after a fewhours.

Other Bridges Requiring Major Repairs after the Earthquake:

Temescal Creek (1-80) Alameda County Several large cracks in concrete box culYert walls and ceiling.

Distribution structure Alameda County Damage to bent caps and columns.(1-580)

Alameda County

Alameda County

Contra Costa County

Distribution structure(1-580)

Fifth Ayenue over-crossing(1-880)

Aoute 242/680 separation(SA 242)

West connector oyer-crossing Contra Costa County(SA 242)

Damage to bent caps and columns.

Damage to columns, bent caps, bearings, and substructure.

Damage to bearing system at Bent 4.

Cracks and spalls; damage to bearings and joint seals.

Benicia-Martinez Bridge(1-680)

Richardson Bay Bridge(US 101)

Pajaro River Bridge(US 101)

Alemany Viaduct (1-280)

Napa Aiver Bridge (SA 37)

Contra Costa County

Marin County

Santa Clara County

San Francisco

Solano County

Damage to open deck expansion joints.

Damage to bearings, caps, columns, and earthquake restrainers.

Anchor bolt and expansion joint damage; cracks and spalls.

Spalling and column damage; retrofit required.

Superstructure shifted 4" longitudinally; earthquake restrainersdamaged.

74 The Continuing Challenge

Page 86: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

References

Asr:l1lcIH\sl, A., B. Bolt, K. McMillin, O. Modjrahcdi, S. eha and R. Donikian, 1994. "SeismicPerformance of Steel Bridges during the 1994 Northridge E:lrthqtlake, n Report NumberUCB/CE-Stccl-94/01, Department of Civil Ellbrinccring, Universiry of California atBerkeley, April 22, 1994.

C.1lifornia Division of Mines and GCO]ob'Y, 1992. Peak rock accelcr.ltion seismic risk map, Open­file Report 92-01, Sacramento, C:llifornia.

DcpartmcnrofTransponaoon, 1994. "The orthridgc Earthquake," Post-Earthquake lmcso·gation RCIK)rt, Calrr:ms, Division ofStrucrures. February 8,1994.

Department ofTransponation, 1972. Rrport COllun1ing Enrthqllakr-Rr1;rtfl1lt Duig" AdvRlICU ftrStnu Higbv:IIY fwd Bridgr Constmetion. Third Biennial Report, requested by: AssemblyConcurrent Resolution No. 49, Resolution Chaptcr I o. 1H, 1972 Regular Session, Statc ofCalifomia, Business and Transportation Agency. October, 1976.

Dcpartmcnt ofTransportation, 11)91. Repon to tbt GOVt17lQr 011 St;sm;c Safety. , Robert E. Best,Dircctor. August 31, 1990.

Departmcnt ofTransportation, 1991. s.n. Z104 Rrp011 to the Goveli/(n' al/(I Ltgisllfhll'r, 8ridgrSeismic RmYJjif Progm7ll. Roben E. Best, Director. January 1, 1991.

Department ofTransportation, 1994. Ctl1N'1I11S Respome 70 GOVCI'IIOI''J !JonI'd OJ IIIIJlth) Rt'fQ1II7Il1'II­dlf!iolls And F.xrClltive OHler 0llllllC Z, 1990; 1994: SUIfUS RI'pOI1, Roberts, J. E., Dcpartment"of'lhlllsportaUoll, Sacr:ulH.:nto, C:llifornia. January, 26, 1994.

Eanhqtl:lke Enginecring Resc;uch 1nstitutc, 1994. N011bridgr Barrbili/lfte, JIIIIIIII1) 17, 1994,Prl'limilJluy ReroI1lJtl;sst/Il(( Rrp011. EERI, Oakland, Californi;1. March, 1994.

EQE Intcrnational, 1994. Tbl' ]nlll/llry 17, 1994 NOl1hridgt, C"lifonl;1I r'.OnhqflaKe.: All EQESU11I11IfllY Report. EQE. San Francisco, CA. March, 1994.

GO\'cmor's Office, 1990. Smte of Californi;l, Execuu\'c Ordcr 1)·86-90.Junc, 1990.

Fung, G.G., R.). LeBeau, E. D. Klein,). Belvedere, and A. E Goldschmidt 1971. The SailFenlll11do EiI11hlfflilke, Field Im;trt;gfll;oll of8r;dgt Domflgt. State of Califomia, Business andTr:msport:nion Agency, Departmcnt of Public V.'orks, Division of Highway Bridges, 1971.

Housner, George, 1990. CU1IIptt;lIg AglI;lIS1 Ti1llr: Rtpurt to Govml0" Grorgt Dmhmjioll from TbrGovenl0r's Bonrd of l11ql/;ry Oil tht 1989 !.htlfo Prieto Eitrthlfllflke, George \0\', I-Iousner, Chair­man, c.c. Thiel, EdilOr. Statc of Califomia, Office of Planning and Research, Sacralllcnto,California. A'l:ly, 1990.

Idriss, I.M., 1991. Prottdllrcs[01' Se/rcr;l1g Eitrrh'llIlIkt Ground N!ol;OllS, :t Repon, by the Cemer forGcotcchnical Modeling, Dcpal·tlllcnt of Civil & Envirollmcnral Engineering, University ofCalifornia, Davis, to the Nation:lllnstitutc of Standards & 'lcchnolob'Y, US Department ofCommerce, Gaithersburg, MD, Scptcmber, 1991. The rcport was revised in r\Jbrch 1993 to

include cstim,ltion of spectr:J1 orc!in,ltes at v;lrious d,ll1lping ratios and was re-issucd byNIS'r as Rcport No NlST GCR 93-625,

References 75

Page 87: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Iwasaki, 'r ,J. Penzien, and R. Clough, 1972. LiIl:mwrr SU1"vey-Sris1Jlic Ejjects all /-ligllWilYBri,lges. Report No. EERC 72-11, Earthquakc Enginecring Rescarch Centcr, University ofCalifornia at Bcrkeley. Novcmber, 1972.

Jennings, PC, Editor, 1971. c"lIgil1eering Featl/res oftbe Silll Fe17/(wtlo EtwtbqllHke Felmli111 9,1971. California Institute of"'echnoI06'Y, Report EERL 71-02. June, 1971.

Moehle, J. P., Editor, 1994. Prdilllininy Report 011 fbI' SeislI/%giat/ (/I/(I ElIgilluril/g Aspects oftheJIIIII/(I1Y 17, 1994 NOI1b"itlge Earthquilke. Report No. UCB/EERC-94/01, EarthquakeEngineering Research Center, University of California at Berkeley. January, 1994.

N.ltional Oceanic and AtIllospheric Administration, 1973. Sail [<'el'l1(/IIt1o, Cil/ijomi" Em'tbqllilke ofPebrllmy 9, 1971. U. S. Departmcnt of COlllmerce, Leonard M. Murphy, Scientific Coordi­nator (thrcc volulllcs), Washington, D.C. 1973.

Priestley, M.J.N, F. Seible, V.H. Chai 1992. Design Guidelinesfm' IIssessment Retrofit Hnd Repair 0/Bridges ftr Seismic Pnj"017lf(lIIce, University of California, San Diego, Structural SystemsResearch Project Report No SSRP 21/01. August, 1992.

Priestlcy, MJ .N., F Scible and C.M. U<lng, [994. TI)(' Nortbridgl' Etmbqllilkl' o/JilllI/my /7, 1994-Dln/Mge 1I'11it/ysis ofSdectrd Frel'Wi!J B1'idges. University of California at San Diego,Structur:ll Systcms Research Report No. SSRP-94/06. February, 1994.

Rep011 o/fbl' Los Angeles Coullty EtI1'tbqllilke C01llIllissio1/, Sail FC17IIIIulo E(/I'tbqllake, Febrllll1Y 9,197/. Novembcr, 1971.

Selna, L.G., L.J. Malvarand RJ. Zelinski, 1989. "Bridge RctrolltTesting; Hinge Cable Re­strainers," in Jou1"!ltli o/5,'trlfetllnli ElIginm·illg. ASCE, Vol. 114, No.4, pp. 920-934, 1989.

Sh'lkal, A .ct al 1994. CSMlP SfnJllg Motion Records}oml the No1'tbridge, C((liforllia F'iI11bqunke 0/Janunry 17, /994. Report OSl\11S 94-07, California Strong Motion Instrumentation Pro­gram, California Division of Mines and Geolo6'Y, California Department of Conscrv,ltion,Sacramento, California. Fcbruary 18, 1994.

Thiel, c.c., G.vV. I-Iousncr, and LT. l(lbin, 1991. "State Rcsponse to the LOllla Prieta Earth­quake; Competing Ag'linst Time," RIIJ/eti" o/tbt' SI'islllologim/ Society 0/1I111eri((/. Volumc 81,No.5. Octoher, 1991.

Thiel, ec., 1994. "Pcrfonmnce of Steel and Concrete Buildings in the NorthridgeEarthquake." Testimony belore the Seismic Safety COlTlmission, Burbank, California.March, 1994.

76 The Continuing Challenge

Page 88: THE CONTINUING CHALLENGE - Caltrans · THE CONTINUING CHALLENGE ... and dcvelopments in seismic design criteria and the highway bridge retrofit ... bridge seismic design and retrofit

Authors' Affiliations

Bruce A. BoltUniversity of California, Berkeley

Nicholas F. ForcllForellJElsesser Engineers, San Francisco

George W. HausnerCalifornia Institute of Technology

I. M. IdrissUniversity of California, Davis

Joseph 1~ NicoleniURS Consuh::ams, San Francisco

Joseph PenzienInternational Civil Engineering Consultants, Berkeley

AlCJt:andcr C. ScordclisUniversity of Califomi:l, Berkeley

Frieder SeiblcUniversity of California, San Diego

Charles C. Thiel Jr.

Telesis Consultants, Piedmont

Authors' Affiliations 77