Volume 5, Number 1 Southern California Earthquake...

Quarterly NewsletterVolume 5, Number 1

SS CC EE CC

2

Exciting Initiative:

EarthScope

3

Interview: John Anderson

9

Hough: Best-Laid Plans

10

Fault: Rose Canyon

16

Evaluating Liquefaction

Hazard

18

Profiles: SCEC Interns

24

Benthien: A Funny Thing

26

Post-Earthquake

Planning Seminar

29

Teachers Sacrifice for

Science

Southern California

Earthquake Center

Southern California

Earthquake Center

Vol. 5, No. 1, 19992 Southern California Earthquake Center Quarterly Newsletter

The Earth Sciences Division(EAR) at the National ScienceFoundation (NSF) is workingwith the scientific communityto orchestrate a new and poten-tially exciting scientificinitiative called EarthScope.EarthScope would consist oflarge arrays of state-of-the-art

seismic and geodetic instru-mentation, much of it focusedon active tectonics of the Pa-cific/Juan de Fuca-NorthAmerican plate boundary of thewestern U.S. A deep earthobservatory along the SanAndreas fault at Parkfield alsowould be part of EarthScope.

SCEC is one of several majorNSF-funded programs helpingEAR shepherd the initiativethrough the Foundation’s ad-ministrative structure. If fundedover the next several years,these new facilities would bean important resource for afuture earthquake center andfor achieving the goal of im-

proving our understanding ofthe earthquake process.

Particularly relevant to earth-quake physics is the geodetic(strain) component of the initia-tive, referred to as the PlateBoundary Observatory (PBO).The PBO is expected to consistof a mix of GPS instrumenta-tion, strainmeters, and InSARimages that can be used toaddress such questions as:

• How is deformation accom-modated within a plateboundary zone?

• What controls the spatialcharacteristics of plateboundary deformation?

• What controls temporalvariations in a plate bound-ary?

• Are there deformationtransients that propagatewithin the plate boundaryzone?

• What is the relationshipbetween vertical andhorizontal tectonics?

• How does plate motionultimately produce anearthquake?

• How do faults and earth-quakes interact with oneanother in time and space?

The PBO would measure defor-mation over a broad spectrumof spatial and temporal scalesand provide sufficient resolu-tion to constrain any transientsassociated with short-wave-length phenomena such asearthquakes and magmaticactivity. A close integration of

seismometers, GPS,strainmeters, and InSAR isnecessary to provide uniformstrain-rate sensitivity at plate-motion strain rates and acrossthe temporal band from severalHertz to a decade.

Thus, the establishment of afully capable PBO will requireprogress in four areas: 1) amore effective integration ofstrainmeters and GPS for a trulybroadband observatory; 2) the

densification of geodetic andseismic instrumentation alongthe northern San Andreas, andperhaps other faults, for in-creased spatial resolution; 3)the linking of the major earth-quake-producing zones tocover the seismogenic part ofthe plate boundary; and 4)increasing the access to InSARdata.

While the main focus of thePBO would be to gain a basicunderstanding of plate bound-ary processes, it also wouldprovide information of practicalvalue. In particular, we wouldbe in a position to detect pre-cursory strain transients, shouldthey occur, that may provepractical for forecasting earth-quakes and volcanic eruptions.Let’s hope EarthScope becomesa reality—it’s our opportunity toadd significant new resourcesto the entire earth sciencescommunity.

EarthScope—a Newand Exciting InitiativeBy Tom Henyey

From the Center Directors

Science DirectorCenter Director

The plate boundaryobservatory is expectedto consist of a mix ofGPS instrumentation,strainmeters, and InSARimages.

We would be in a posi-tion to detect precursorystrain transients that mayprove practical forforecasting earthquakesand volcanic eruptions.

Editor’s Note: As this issue wasprepared, Bernard Minster wasannounced as the new ActingScience Director for SCEC.Please see page 35 for theadministrative structure as ofAugust 1, 1999.

Vol. 5, No. 1, 1999 3Southern California Earthquake Center Quarterly Newsletter

John AndersonInterviewed by Jill Andrews

Interview with SCEC Scientist

John Anderson, director of theSeismological Laboratory at theUniversity of Nevada at Renoand member of the SCEC boardof directors, is interviewed hereabout his work in strong motionseismology, especially in thecontext of the SCEC mission.

SCEC: You and your groupappear to be in a “transitionzone” between basic researchand tangible products—in otherwords, you are trying to findresults that are immediatelyuseful. What have you found sofar, and who will use the resultsof your work?

JA: First, I think it is not as farfrom basic seismic research as

it may appear. I’m thinking of avision for strong motion seis-mology given in 1980 by KeiitiAki (SCEC director, 1991–1996)in an address when he was theoutgoing president of the Seis-mological Society of America.He said: “Our goal is to com-pute seismic motion expectedat a specific site of an engi-

neered structure when the faultmapped by geologists breaks.”So “synthetic” seismograms arethe goal. We want them to beso realistic that engineers canuse them. But we also wantthem to be based on soundunderstanding of the physics ofthe earthquake process so thatthey will be reliable.

SCEC: How can this beachieved?

JA: We have to bring in sourcephysics, wave propagation, andscattering caused by the com-plexity of the Earth. At the least,we have to take what othershave learned and incorporate itinto our models. On the other

hand, one can hope that ourresearch; i.e., the constraints onthe physical processes neededto successfully predict strongmotions, will lead to someincreased understanding of thephysics.

SCEC: Let’s focus on your workas it relates to the Southern

California Earthquake Center,specifically, your work on thePhase III Report (need officialname for this report). Howwould you define the term“strong motion seismologist” tosomeone who doesn’t have abackground in geophysics?

JA: I’d define strong motion asearthquake motions that arestrong enough to feel. Strongmotion seismologists might alsobe recognized by one type ofinstrument that is essential fortheir research. They must haveaccelerographs, which aredistinguished from the instru-ments used by most otherseismologists in that they stayon scale in even the strongestshaking but can’t detect weakearthquakes that are easilyobserved using higher gaininstruments. Twenty years ago,before the digital revolution ininstrumentation, the weakestsignals these instruments couldrecord were just about at thethreshold of human detection.Now, the digital accelerographscan record local earthquakesbelow magnitude 3, which isway below what a person can

normally feel. So accelerographdata is getting more interestingto network seismology, and thedistinction between strongmotion seismologists and othersis getting blurred. Still, theoriginal focus of strong motionseismology is to understandground motions that are strongenough to cause damage.That’s still true.

SCEC: How completely has thedigital accelerograph replacedthe old analog instrument? Doyou still use both?

JA: The most common analogstrong motion accelerograph inthe world, the KinemetricsSMA-1, was discontinued manyyears ago, but there are stilllarge numbers of them out inthe field. The data they produceis very good data, but it’s lessconvenient to use because theyrecord strong motions on film,which has to be retrieved andthen digitized. This is very timeconsuming. The advantage ofdigital instruments is that thedata are available immediately,and the resolution is muchbetter. On analog, one can

Nearly every task SCEC identified for southernCalifornia ought to be done in Nevada, in Utah,near New Madrid and every other seismic zone inthe U.S.


ProfessionalHighlights

John AndersonEducationPh.D., geophysics, Columbia UniversityB. S., physics, Michigan State University

ProfessionalSCEC Board of DirectorsUniversity of Nevada, Reno–Mackay School of MinesProfessor of Geophysics, Department of Geological SciencesSeismological Laboratory Director

Research InterestsJohn Anderson’s research in strong motion seismology has taken him toMexico, Japan, Turkey—and yes, even exotic southern California. He’shosted visiting scholars from India, Iran, China, and Korea, and looksforward to traveling to those places. His research interests span all as-pects of engineering seismology, including applications of geologicaland seismological information to estimate seismicity and seismic haz-ards; recording strong ground motions; understanding the physics ofnear-source ground motions; and applications to engineering problems.

Selected PublicationsAnderson, J. G., and S. Hough, A model for the shape of the Fourieramplitude spectrum of acceleration at high frequencies: Bull. Seism. Soc.Am. 74:1969-1994, 1984.

Anderson, J. G., P. Bodin, J. Brune, J. Prince, S. Singh, R. Quaas, M.Onate, and E. Mena, Strong ground motion and source mechanism ofthe Mexico earthquake of Sept. 19, 1985, Science 233:1043-1049,1986.

Zeng, Y., J. G. Anderson, and G. Yu, A composite source model forcomputing realistic synthetic strong ground motions, Geophysical Re-search Letters 21:725-728, 1994.

Zeng, Y., J. G. Anderson, and F. Su, Subevent rake and random scatter-ing effects in realistic strong ground motion simulation, GeophysicalResearch Letters 22:17-20, 1995.

Zeng, Y., and J. G. Anderson, A composite source model of the 1994Northridge earthquake using genetic algorithms, Bull. Seism. Soc. Am.86:S71-S83, 1996.

Anderson, J. G., and G. Yu, Predictability of strong motions from theNorthridge, California, earthquake, Bull. Seism. Soc. Am. 86:S100-S114, 1996.

Anderson, J. G., Seismic energy and stress drop parameters for a com-posite source model, Bull. Seism. Soc. Am. 87:85-96, 1997.

Ni, S.-D., R. Siddharthan, and J. G. Anderson, Characteristics of nonlin-ear response of deep saturated soil deposits, Bull. Seism. Soc. Am.87:342-355, 1997.

Lee, Y., Y. Zeng, and J. G. Anderson, A simple strategy to examine thesources of errors in attenuation relations, Bull. Seism. Soc. Am. 88:291-296, 1998.

Su, Feng, J. G. Anderson, and Y. Zeng, Study of weak and strong groundmotion including nonlinearity from the Northridge, California, earth-quake sequence, Bull. Seism. Soc. Am. 88:1411-1425, 1998.

Anderson, J. G., and J. N. Brune, Probabilistic seismic hazard analysiswithout the ergodic assumption, Seismological Research Letters 70:19-28, 1999.

resolve differences in accelera-tion of maybe 1,000th of theacceleration of gravity. Withnew digital instruments, there’sabout a 100-fold improvement.

SCEC: How does that affectyour work?

JA: With a really strong earth-quake, you’re well above the“noise” level over a certainfrequency band; with digitalinstruments, you are well abovethe noise level for a widerrange of frequencies. So youhave much more accurate datafor both lower and higherfrequencies than with the old

instruments. Also, the sameinstruments record smallerearthquakes, which can beused as empirical Green’sfunctions for some of our re-search.

SCEC: Let’s shift back to thefield of strong motion seismol-ogy. According to yourcolleague and coordinator ofthe SCEC Phase III report (NedField), logic dictates that strongmotion seismology should be awell-funded branch of seismol-ogy. Is there enough supportprovided by government agen-cies, or could there be more?

JA: More support could beused to accelerate both scien-tific discovery and thedevelopment of practical re-sults. In this regard, I’m not sosure strong motion seismology

has either better or worsefunding than many other fieldsof seismology. For instance,most of our regional seismicnetworks are seriously under-funded. Research on datacollected by those networks isalso under-funded.

SCEC: Why?

JA: There is the usual clichéanswer: that without morefrequent disasters, the publicand Congress tend to forgetabout earthquake hazards, andseismology drops to a lowerpriority on the political agenda.That may contribute to theproblem, but I don’t think it isthe complete answer. Anotherfactor may be an apparentweakness in our case for rel-evance. The building code isthe only place where mostpeople have to spend moneyon seismic hazards. Thesecodes give “cookbook” answersto seismic design for moststructures, and our engineersare probably correct in theirassertion that most structuresare designed well for life safety.With that apparent weakness,how should we convince thegeneral public and Congressthat understanding more aboutthe earthquake hazard is essen-tial for making better decisionson a daily basis, thus reducingthe future risk? Of course, thereare existing programs thatdemonstrate the relevance ofthis kind of research—TriNet isone example. TriNet [Ed. note:the USGS-Caltech-CaliforniaDivision of Mines & Geologyconsortium to upgradeCalifornia’s seismic networks todigital] gives immediate infor-mation for emergency responseand to satisfy the broad curios-ity about earthquakes; those are

Major earthquakes occurrarely, and every one iseither an opportunity or amissed opportunity tolearn more.


More About the SCECPhase III Report

Several years ago SCEC embarked on a multidisciplinary ef-fort to determine how and whether seismic hazard analysiscan be improved by accounting for site effects (the propensityfor certain sites to shake harder than others).

As of 1997, the results of these studies had been combinedinto an engineering-style report (called the SCEC Phase III Re-port), which was reviewed by an independent team of experts.They agreed that the report contained many important findingsbut that it needed some streamlining.

After some delay it became apparent this would not happenwithout a full-time editor. One now exists—Edward (Ned) Field—and the report is well on its way to completion.

Rather than producing a document primarily aimed at engi-neers, Phase III will be published as a collection of scientificpapers in a special issue of the Bulletin of the SeismologicalSociety of America (pending acceptance). Issues that wereaddressed in the earlier version relating to the characterizationof seismic sources (updating Phase II) and to computing syn-thetic seismograms have been omitted. The report will now focusexclusively on how, and whether, probabilistic seismic hazardanalysis (PSHA) in southern Californiacan be improved by ac-counting for site effects.

Field and colleagues made a diligent search for any character-istics that systematically predispose a site to greater (or lower)levels of ground shaking. They found some significant attributes,such as surface geology and depth to bedrock in sedimentarybasins that indeed correlate with observed amplification fac-tors. They also tested which of several attenuation relations aremost consistent with southern California data and with theoreti-cal considerations where data are lacking. They are currentlyquantifying the influence of these factors on PSHA.

A preliminary conclusion is that except in some specific circum-stances, only modest improvements can be made by accountingfor site effects. However, even modest improvements may besignificant in terms of the implied seismic hazard.

strong selling points for thegeneral public. Another ex-ample is SCEC’s outreachprogram, which is excellent. Itdoes a great job of sharing ourexcitement about our scientificdiscoveries. If we as a seismo-logical community did as wellin outreach in the rest of thenation, maybe the national

support for seismology wouldbe much better.

SCEC: How do you resolve this“apparent weakness” in thecase for relevance?

JA: First, I think there are reallylarge uncertainties in the inputsto even the newest code guide-lines and/or requirements, such

as the new probabilistic seismichazard maps produced by theUSGS and the California StateDepartment of Conservation’sDivision of Mines and Geology.

SCEC: Can these uncertaintiesbe resolved over time? Whatcan scientists do to reduce thelarge uncertainties? Or can welearn to work with what wehave and concentrate on miti-gation?

JA: My opinion is that SCEC isdoing exactly what needs to bedone to reduce uncertainties.We are conducting researchthat ultimately will benefitsociety, because in the long runreducing the uncertainties willbe more economical. The latestSCEC RFP is a masterpiece. [Ed.note: every year the SCECboard of directors develops aresearch plan that is circulatedto SCEC researchers as a re-quest for proposals, invitingthem to submit a proposaldescribing what they wouldlike to do to help achieve thatplan in the coming year.]Nearly every task SCEC identi-fied for southern Californiaought to be done in Nevada, inUtah, near New Madrid andevery other seismic zone in theU.S. Every task would make asignificant contribution tounderstanding the earthquakethreat in those areas, also

SCEC: Let’s get back to yourgroup’s contributions.

JA: For the past few years,primarily with SCEC support,we’ve been developing whatwe call a “composite sourcemodel” to predict strong mo-tions. The approach is entirelysynthetic. We don’t use empiri-cal Green’s functions or

empirical earthquake sourcefunctions. Our syntheticGreen’s functions include wavepropagation in an attenuatingEarth, and we’ve added inscattering calibrated from smallearthquakes. The compositesource seems to have the rightamount of complexity, and theparameters can all be corre-lated with macroscopicvariables like seismic momentand energy. We’ve tested itagainst the observations for themajor earthquakes in southernCalifornia, and it seems toperform reasonably. The modelhas a random part to it, butwe’ve shown that the source of

several earthquakes, includingthe Northridge earthquake, canbe described with a specificrealization of the model. We’restill working on improving it.And it is being used in someapplications in private industry.One of the latest phenomenathat we’ve found necessary toinclude in the model isnonlinearity in the soil re-sponse. When we make amodel that is successful for allthe distant stations it signifi-cantly overpredicts the groundmotions at the closest sites.Then when we apply a more-or-less standard correction tothose nearby sites for expectednonlinear stress-strain relation-ship in the alluvium belowthose sites, the predictionscome in much closer to theobservations.

SCEC has given theseismological communitya transportable model forattacking the scientificand social issues associ-ated with earthquakes.


SCEC: To improve seismicdesign and probabilistic seismichazard assessment, you andyour colleagues conduct regres-sion analyses to developempirical ground motion mod-els, which predict groundmotion characteristics as afunction of magnitude anddistance. Can you explain what

is involved with doing regres-sion analyses, define in simpleterms what an empirical groundmotion model is, and explainwhy prediction of groundmotion characteristics as afunction of magnitude anddistance is important to designprofessionals?

JA: There are several ways topredict ground motions. Thestandard engineering approachis with a ground motion predic-tion equation. This approachseeks to predict some groundmotion parameter, such as peakacceleration, as a function ofseveral variables, which haveto include magnitude anddistance. Generally theseequations add additional termsthat seek to characterize the

site conditions in some wayalso.

The first step of the processincludes making up an ad-hocform for the equation—some-thing that you hope willcapture the average trends. Thesecond step uses regressionanalysis to estimate the un-

known coefficients in theseequations. This approachgenerally predicts groundmotions to within a multiplica-tive factor of about two. In oneof our Phase III contributions, Iexamined the physics to try tomake some general statementsabout the expected characteris-tics of these predictionequations, and we also evalu-ated some of the existingequations for use in southernCalifornia.

I have to admit, though, thatI’m not particularly enthusiasticabout developing new groundmotion prediction equations.The first problem with these isthat there is little to be learnedabout the physics of earth-quakes or wave propagation inthe process. The second is that

the process is endlessly compli-cated. Strong ground motionsare extremely complicated timeseries. Anyone who hopes tocharacterize them adequatelywith a few parameters isdoomed to failure. The groundmotion prediction equationsmay be developed for a dozenor so characteristics of a suiteof seismograms. But that is notenough. Also, every time an-other earthquake happens youhave to carry out a new regres-sion, and every time you usethe prediction equation youhave to worry about whether itis appropriate for the region.That’s why the type of “groundmotion prediction” that we aremainly working on is syntheticseismograms.

There is far more information insynthetic seismograms than inany prediction equation. If wecan generate a reliable syn-thetic seismogram, then theengineer can determine anyparameter he wants from thesynthetic seismogram. We don’tneed to worry about thewhether the prediction equa-tions that we used are correct.Instead, we worry aboutwhether the source model andthe earth structure model that

we used are correct. Ultimately,this will be much simpler thanground motion predictionequations.

SCEC: What can scientists do tohelp design professionals incor-porate research results?

JA: That’s a difficult questionbecause “design professionals”includes a wide range of activi-ties—an architect planning aone-story house, on one hand,or, at the other extreme, struc-tural engineers who designskyscrapers; the latter arepeople who are usually talkingto strong motion seismologistsin the first place. They wouldbe the people more likely toask for seismograms to test theirdesigns.

SCEC: Let’s talk about theprogress of strong motionseismology in relation to betterengineering design on a world-wide basis. Are researchers inyour field making less rapidprogress than if there weremore funds for the field?

JA: The answer is yes. Majorearthquakes occur rarely, andevery one is either an opportu-nity or a missed opportunity tolearn more. There are a lot ofquestions about the earthquakesource that will only be finallyanswered with strong-motionobservations. An examplewould be the question of howlong the San Andreas fault takesto slip in a major earthquake.One thing we know is that

strong motions are highlyvariable over space, and themotion field is seriouslyundersampled. To get a descrip-tion of the strong motion fieldwithout spatial aliasing, weneed many more strong motioninstruments than there are at

An observed strong motion accelerogram compared to a synthetic one.(Prepared by Y. Zeng, 1999, based on work in progress.)

Supporting more strong motion instruments atmajor faults around the world is expensive, but westand to learn the answers to some fundamentalquestions sooner.


Glossary of EngineeringSeismology Terms

Source physics: how one side of the fault slips past the other,and the physical principles that drive the slip

Wave propagation: how the seismic waves get from the faultto wherever they are going.

Scattering: an important part of wave propagation. Scatteringmeans some of the energy goes off in different directions ratherthan traveling in a straight line. Think of baseball: a strongcenterfielder trying to throw a ball to home plate might be ableto throw it in a straight line all the way, but he might also throwit to shortstop, who throws it to third base, who throws it tohome. It’s an indirect route, but the ball still gets there. Like-wise, if there are obstacles at different locations in the Earth,the seismic waves might take an indirect route. Complexitycauses “packets” of energy to arrive at different times.

Empirical Green’s functions: recordings of waves from very smallearthquakes, which are more abundant than large ones, thatcan be used to predict strong ground motions. This is done by“scaling up” the small seismograms.

present. Yokohama is the onlycity in the world that is cur-rently operating a strongmotion network approaching asufficient density to thoroughlydocument the strong motionwavefield at the frequenciesthat affect structures. They have150 accelerographs in an areaof 433 km2, or about one in-strument every 3 km2. Tokyo isworking on an even densernetwork. Mexico City’s networkis also very good. Its density ismuch lower, but the dominantmotions are low frequencywaves, so you can get start toget a good picture of what ishappening. With the complex-ity of seismic waves in basins,this empirical approach may bethe best way to understand andprepare for local variations inthe motions.

Is it fundamental science?Conceptually, if you have agood enough description of thevelocity model and a bigenough, fast enough computer,maybe the spatial variation inthe motions could be predicted.I think that density of stationswould be approaching a levelwhere a new challenge couldbe tackled, though, namely toinvert the complete observedwavefield to derive the struc-ture of the basin.

SCEC: You mentioned Mexico.Describe some of your researchthere.

JA: I’ve been working inMexico since 1980. Jim Brune(UNR) and I installed a strongmotion network in Guerrero[Ed. note: the GuerreroAccelerograph Array is nearAcapulco, which is above thesubduction zone along thePacific coast]. In this zone, we

believe that there are majorearthquake occurrences (M 7.8-8 or greater) quite frequently,about once every 50 years orless. We installed 30 strongmotion instruments on rocksites in the vicinity of GuerreroGap. Shortly after we startedthe project, the September 19,1985, earthquake occurred(M 8.1). The records we got onthe Pacific coast are still almostunique due to their locationdirectly above the fault in aM 8+ earthquake. It turned outthat the Guerrero data wascritical for understanding theground motions in Mexico City,which is 400 km from the fault.The really interesting result wefound was that Mexico Cityexperienced about the samepeak accelerations as the areadirectly above the fault thatmoved, due to the type of soilin Mexico City. An importantlesson from this also is thevalue of international collabo-ration.

I know that supporting morestrong motion instruments atmajor faults around the world isexpensive. The advantage isthat we stand to learn theanswers to some of our funda-mental questions sooner thanwe would if we wait for thesame type of earthquake inCalifornia. That could reduceuncertainties for design ofstructures here; over time, thegreater economic benefit wouldfar outweigh the cost of theinternational experiment.

SCEC: Back to the question ofsupport. What other avenues ofsupport exist for groups such asyours (i.e., private industry) andwhat trade-offs may exist whenyou conduct research withindustry support?

JA: There’s occasionally theopportunity to get involved in adesign project through anengineering consulting firm, butI don’t make any effort topursue those projects, so I don’tdo that very much. A significantsource of support for our grouphas been from Pacific Gas andElectric through PEER (PacificEarthquake Engineering Re-search Center). That is morelike private industry than likeSCEC. We’ve also had supportfrom the grants programs of theNational Science Foundation,the US Geological Survey, andthe California Strong MotionInstrumentation Program. Thetrade-offs question is easy.Support from private industry isvery much more focused—driven by strict deadlines. Youmay not have much time tothink through and try to solve

all the difficulties in detail. Youhave to come up with your bestanswer given the time con-straints that are available. Butthe interaction is very valuable.On those projects, you find outwhat exactly are the problemsthat need to be solved. It’s quitecommon to realize that theirproblems are actually pointingtowards fundamental researchquestions.

SCEC: For instance?

JA: For instance, right nowwe’re dealing with “kappa” atYucca Mountain. That’s aparameter that Sue Hough(USGS Pasadena) and I definedseveral years ago to describethe spectral shape of highfrequency accelerograms.There are some fundamentalquestions about its physics that


we’re being forced to deal withas a result of the way we areusing it for the Yucca Mountainproject. We don’t understandwhy there is so much variabilityfrom one earthquake to thenext.

SCEC: Can you explain thisfurther? Why is this importantto the Yucca Mountain project?

JA: Yucca Mountain has beenusing synthetic seismograms,among other techniques, tocharacterize what the groundmotions might be in the (un-likely) event of an earthquakethere. It’s not very seismicallyactive and there’s not muchdata to work with, so syntheticseismograms play a moresignificant role. One constrainton the synthetic seismograms isthat they should have the samevalue of kappa as observationsfrom small earthquakes. So it’snot a comfortable situationwhen kappa is more variableamong small earthquakes thanyou think it ought to be.

SCEC: A recent paper youauthored that was published inSeismological Research Letters(January-February 1999) ad-dresses uncertainties associatedwith using a catalog of re-corded earthquakes as a proxywhen calculating expectedground motions for one particu-lar site and one particular fault.Can you explain how the“ergodic assumption,” as this iscalled, is used in hazard calcu-lations, and whether the issueof uncertainties associated withit have an important impact onresults?

JA: This is a challenge to ex-plain, since some readers maybe baffled by terminology like

“ergodic,” “aleatory uncer-tainty,” and “epistemicuncertainty.” But our resultspotentially contribute to solve asignificant puzzle. The puzzleis that there are old, precari-ously balanced rocks withinsight of the San Andreas fault.The SCEC Quarterly Newsletter,

Vol. 4, No. 1, featured one onthe cover.

The San Andreas fault is in thebackground, 15 km away.PSHA (Probabilistic SeismicHazard Analysis), as that previ-ous article pointed out, predictsthat ground accelerationssufficient to topple those rocksshould have occurred manytimes in the lifetime of thoserocks. That forced us to thinkmore carefully about the PSHAprocess. We suggested that theproblem arises with how weenter the ground motions andtheir uncertainties into a PSHA.

On average, our observationsof ground motions vary aboutprediction equations within arange of about plus or minus amultiplicative factor of two.This is a standard deviation of aprobability distribution, so withenough cases there are obser-vations that vary by even more.That is predominantly a spatialvariability of ground motions—typically our data aredominated by a few largeearthquakes that have given alot of observations at a lot ofdifferent places. The basicPSHA takes that uncertainty inthe predictions and treats it as ifit is the variability over time.That is what we call the “er-godic assumption.” In thismodel, if we could recordground motions at the same sitefrom a dozen characteristicearthquakes on the San Andreasfault north of Los Angeles, therewould be a dozen differentground motions, and theywould vary over a range fromone half of the mean predictionto twice of the mean predic-tion, and even more. With thismodel, if you wait long enoughat any one place next to theSan Andreas fault, you getreally extreme accelerations.This is probably why the PSHApredicts accelerations thatwould topple precarious rocks.A solution had already beenformulated before we wrotethat paper. The uncertaintyshould be divided into twoparts. Epistemic uncertaintycovers what we don’t knowabout the path and site effects.Aleatory uncertainty coversrandomness that comes in fromthe differences in the sourcefrom one San Andreas earth-quake to the next. Wedemonstrated that if most of theuncertainty is epistemic, then

the contradiction would goaway. A big challenge thatremains is to figure out howmuch of the total uncertaintyfalls into each category. If thegreat earthquakes on the SanAndreas ruptured in an identi-cal way every time, all theuncertainty would beepistemic. The ground motionat the site would be identical inevery earthquake. If it doesn’tknock down a precarious rockthe first time, then it never willunless geological processeskeep making the rock moreprecarious. At some of theprecarious rock sites, Jim Brunemakes a convincing case thatthe geological processes arevery slow now.

SCEC: From your perspective asdirector of the Nevada Seismo-logical Laboratory, what is theimpact of SCEC?

JA: SCEC has given the seismo-logical community atransportable model for attack-ing the scientific and socialissues associated with earth-quakes. We need to continue acenter in California to studyearthquake physics. However,the rest of the country alsoneeds research of the type thatSCEC carried out in southernCalifornia. We should have asimilar center for the GreatBasin, one for the eastern U.S,and one for the parts of thecountry threatened by subduc-tion zone earthquakes. Thesefour centers would each befocused on a different tectonicstyle. This would be an ex-tremely effective way topromote a greater interaction ofearth scientists, scientific dis-covery, and outreach—everything that SCEC does—inthe rest of the country.

Fourier amplitude spectrum of theN85°E component of strong motionacceleration recorded at Cucapahduring the Mexicali Valleyearthquake of June 9, 1980 (M 6.2).Accelerograph was a digitalrecorder that samples at a rate of200/sec. (A) Log-log axes. (B)Linear-log axes, illustrating thedefinition of the parameter kappa(k). This figure is from the paper byAnderson and Hough (1984) thatfirst defined kappa.


I don’t remember what mylong-term plans were when Itook my dog for a walk on onerainy October night in NewYork back in 1989, but I doremember my short-term plan. Iwas going to get home, dry off,and watch a World Series gameon television. The baseballgame didn’t happen, of course;I spent a long night at Lamontinstead. A mere 24 hours afterthat, I was eating a late dinnerat a Thai restaurant in Berkeley,feeling like I must’ve somehowwandered into a Star Trekmovie and been whisked awayby the transporter beam.

A recent SCEC-sponsoredworkshop on earthquake re-sponse seemed like anappropriate occasion to pestera few of my colleagues for theirstories. How had an earthquakechanged their plans and lives inrecent years? Oddly, the firstthree people I asked all hadstories about the same earth-quake.

CDMG geologist Jerry Treimanfirst identified the M 7.2 1992Landers earthquake as havingthrown a monkey wrench intohis plans for a family campingtrip to, of all places, Big Bear.His family did eventually make

Best-Laid Plans

Tales from the Front

the trip, and Jerry was able tobreak away from his fieldworkat times to join them. Did theyfeel a lot of aftershocks? On thecrystalline terrain of the SanBernardino Mountains, report-edly not (providing evidencethat there really is something tothis business of site response).

USC geologist Jim Dolan identi-fied Landers as having been asingularly ill-timed event in hislife. He had carefully plannedto take that Sunday off, his firstday away from work in nearlytwo months. He’d even boughtthe early edition of the SundayNew York Times, in preparationfor his day of leisure. At 10o’clock the next morning,instead of sitting at his kitchentable with the paper and a cupof coffee, Jim was in theMojave Desert making his firstreconnaisance of the surfacerupture. Did he ever get back tothe newspaper? Nope.

The Landers earthquake wasalso memorable for USC seis-mologist Ned Field, who, in1992, was at Lamont. Nedreports having driven into thelab early that day with fellowgraduate student VeganAharonian. On their way in,Ned joked that, with their luck,

a major earthquake wouldhappen, and they’d be the onlytwo people around to deal withit. Not too many minutes later,Vegan found Ned in the com-puter room and let him knowthat they’d gotten their earth-quake. Two days after that, Ned

was in the Coachella Valleywith fellow Lamont researchersPaul Friberg and Noel Barstow.They spent a week in the field,chasing basin edge effects andhoping their unreinforcedmasonry hotel wouldn’t col-lapse if the earth had yetanother major temblor up itssleeve.

Another Lamont scientist, NanoSeeber (who was not at theearthquake response workshop)recalled the Landers earthquakein a different light. Although hisprior plans were also scrubbedfor a Monday morning trip tosouthern California, he recalledhis impromptu fieldwork ad-venture as having been moreenjoyable and productive than

Interested in sharing a field-work story of your [email protected]/HOUGH/

the daily, computer-orientedgrind he left behind.

The SCEC workshop was asuccess. Attendees left with asense of optimism regardingour ability to lay the ground-work to optimize our chancesof a successful earthquakeresponse the next time one iscalled for. One is left wonder-ing, though—is “next time” 10years, or 10 minutes away? 24hours from now, are you goingto be at the appointment onyour calendar, or some placeelse? It’s like the Microsoftslogan with a twist—not“Where do you want to gotoday,” but “Where are yougoing to be tomorrow?”

And if, underneath it all, wedidn’t love it at least a little, wewouldn’t be in the earthquake-chasing business in the firstplace.

By Susan Hough


The northwest striking RoseCanyon fault zone, whichbisects the city of San Diego,comprises a complex set ofanastomosing and en echelonfault strands that include theRose Canyon, Mount Soledad,

Country Club, Mission Bay, OldTown, Spanish Bight,Coronado, and Silver Strandfaults, in addition to manyother secondary faults (Figure1) (Kennedy, 1975; Kennedyand Welday, 1980). Offshoremapping of faults in the South-ern California Borderland usingseismic stratigraphic techniques(Legg, 1985; Legg and Kennedy,1991; Fischer and Mills, 1991)has shown that the Rose Can-yon fault zone extendsnorthwest and joins the New-port-Inglewood fault zone(Figure 2). The continuity offaulting along this zone ledFischer and Mills (1991) toconclude that the Newport-Inglewood and Rose Canyonfaults represent a single struc-tural element.

Holocene Activity of theRose Canyon Fault ZoneBy Scott C. Lindvall and Thomas K. Rockwell

Although motion on the RoseCanyon fault zone is generallyconsidered to be right-lateralstrike-slip (Kennedy, 1975),individual strands within thefault zone display variouscombinations of dip slip andstrike slip. The variable sense ofdip-slip motion along this faulthas locally resulted in the upliftof Mount Soledad, the depres-sion of San Diego Bay, andother physiographic featuresthat provide San Diego withmuch of its aesthetic beauty.

In 1989, two-dimensionaltrenching of the fault on thelowest terrace of Rose Creekdemonstrated that the faultdisplaces Holocene depositsand showed that considerablestrike slip was required toexplain stratigraphic mis-matches across individualstrands of the fault (Lindvall etal., 1990; Rockwell andLindvall, 1990). Subsequentthree-dimensional trenching atthe Rose Creek site was able todetermine a slip rate and recog-nize at least threepaleo-earthquakes (Lindvalland Rockwell, 1995). Morerecent studies have betterconstrained the timing of themost recent surface-rupturingearthquake (Rockwell andMurbach, 1998).

This article presents informa-tion from a three-dimensionalpaleoseismic trenching studyand other investigations andsummarizes our current under-standing of the Holocenebehavior of the Rose Canyonfault zone.

Rose Creek Study SiteBased on our tectonic geomor-phic analysis, we selected a sitewhere the southeastern con-tinuation of the Mount Soledadfault crosses the lowest terraceof Rose Creek and formed alow, east-facing scarp. TheRose Creek site, which was

graded in 1960 and is nowoccupied by an asphalt-cov-ered parking lot of the SanDiego Gas & Electric BeachCities Operating Center, is oneof the few accessible locationsin this urban environmentwhere young, surficial sedi-ments were deposited acrossthe fault.

In 1989 we excavated a singletrench across the fault to deter-

mine the style and width offaulting and the type of sedi-ment at the Rose Creek site(Figure 3). The 22-m-longtrench exposed a 1.5 to 2.0-m-thick section of artificial fillimmediately below the pavedparking lot surface. Underlyingthe fill we exposed a section offaulted clayey to silty sand onwhich the surficial soils werepartially intact. The fill wasemplaced in 1960 withoutdisturbing the scarp or theorganic-rich A horizon north-east of the fault (Figures 3 and4). However, southwest of thefault, grading removed much ofthe A and Bt soil horizons.Consequently, the originalground surface, including thefault scarp, was effectivelyburied and preserved or slightlymodified from that seen in the1941 air photos. The fault wasexpressed in the trench expo-

Featured Fault

The dip-slip motion alongthis fault has providedSan Diego with much ofits aesthetic beauty.

The most recent largesurface rupture on theRose Canyon faultoccurred during the pastfew hundred years.

Figure 1. Generalized map showingmajor strands of the Rose Canyonfault zone and major geographicfeatures of the San Diego area.Individual faults within the zoneare the Country Club (CC), MountSoledad (MS), Rose Canyon (RC),Mission Bay (MB), Old Town (OT),Spanish Bight (SB), Coronado (C),and Silver Strand (SS). Faults aredashed where approximatelylocated, dotted where concealed.

All f

igur

es: L

indva

ll and

Roc

kwell


sure as a flower-type structurewith several strands splayingoutward toward the surface.The individual fault strandsoffset Holocene strata across azone about 2 m wide. Somestrands crosscut others andexhibit a complex micromor-phology, suggestive of multipleslip events. Nearly all faultstrands in the trench wallexhibit an apparent west-side-up component of slip, which isconsistent with the east-facingscarp seen in the 1941 photo-graphs. The most recentlyactive fault strands completelyoffset the A, E, and Bt soilhorizons that marked the natu-ral ground surface prior tograding and development in1960 (Figure 3).

Radiocarbon dating of detritalcharcoal samples revealed thatthe stratigraphic section ex-posed in the trench was early tomiddle Holocene in age. Thisinitial phase of trenching at theRose Creek site demonstratedthat the Rose Canyon fault zonewas Holocene active. In addi-tion, the sharp, distinct contactsof the faults and surficial soil

horizons suggest that the mostrecent earthquake likely oc-curred only a few hundredyears ago.

Three-Dimensional TrenchingTwo-dimensional fault expo-sures on a trench wall orientedperpendicular to the faultcannot resolve horizontal slip.Previous studies of strike-slipfaults (Sharp, 1981; Sieh, 1984;Rockwell et al., 1986) havedemonstrated the effectivenessof three-dimensional trenchingin resolving both horizontaland vertical deformation, andwe employed these methods atthe Rose Creek site. Based onthe results of our first trench (T-1), we reoccupied this site withthe intent of exposing the faultzone in three dimensions andresolving the slip by mappingdisplaced fluvial features.

Initially, we excavated twofault-parallel trenches thatstraddle the fault (Figure 6)immediately northwest oftrench T-1. These trencheswere used to establish the site’sstratigraphy and to identifyfluvial features or sedimentarystructures that could be used aspiercing points for quantifyingthe amount of horizontal offset.

In the 20-m-long trench T-2(Figure 4), we found only onesuch feature, a gravel-filledchannel oriented normal to thefault. The channel’s crosssection was exposed in boththe northeast and southwestwalls of the trench and was theonly gravelly channel depositin trenches T-1, T-2, or T-3. Thecoarse-grained channel fill was

distinctive amidst the relativelyfine-grained, weakly stratified,silty to clayey sand depositsthat characterize most of theexposures. The gravel channelwas not present in trench T-3west of the fault zone, becausethe section containing thegravel had been eroded orstripped when the northwest

corner of the site was graded in1960. Therefore we had toexpose the channel in hand-dug excavations across the faultzone beginning with trench T-2.

After logging trenches T-2 andT-3, we excavated the artificialfill at the site down to the top ofthe natural deposits in order totrace the gravel channel to andacross the several strands of thefault. This resulted in a ~12 x18 m rectangular excavationthat was 2-2.5 m deep with 1:1slopes along the sides (Figure7). From the floor of the pit, wefirst excavated a short trench (T-4) across the fault to re-exposethe fault zone (Figure 6). Wealso re-excavated part of trenchT-2 (trench T-2A) to re-exposethe channel. A series of con-necting trenches were thenexcavated by hand to exposethe channel to and acrossdifferent strands of the fault(trenches T-5 to T-11), leavingthe channel preserved in intactblocks of sediment (Figures 6and 7).

StratigraphyThe faulted sediments underly-ing the artificial fill consistpredominantly of a sequence ofclayey to silty sand depositsthat are locally weakly tomoderately stratified (unit C).Also within unit C are several

The evidence suggests the last large surface rupturemay have involved the entire onshore portion of thefault zone.

Figure 2. Regional map showing major northwest-striking dextral-slip faultzones including the Newport-Inglewood (N-I)–Rose Canyon (RC) fault zonethat extends from San Diego to Los Angeles. Other faults are Sierra Madre(SMa), Whittier, (W), Elsinore (E), Palos Verdes (PV), THUMS-HuntingtonBeach (T), Coronado Bank (CB), Descanso (D), Vallecitos (V), San Miguel(SM), Agua Blanca (AB), and Bahia Soledad (BS) faults. Labeled cities areLos Angeles (LA), San Diego (SD), and Ensenada (E). The source zone of the1933 Long Beach earthquake is denoted by the hatched pattern, and thedark shaded area offshore is Lausson Knoll (LK). Numbers refer to fault sliprates in millimeters per year.

Figure 3. Generalized log of trench T-1. Note the upward floweringstructure, mismatch of some stratigraphic units, and location of 14C samples(solid circles). Mapped units are: A, modern A soil horizon; Bt, Bt (argillic)soil horizon; C, stratified overbank alluvium; D, gravelly colluvium. Muchof the A and Bt horizons have been stripped from the southwest side of thefault.


small lenses and sheets of fine,moderately to well-sorted sandand millimeter- to centimeter-thick layers of clay thatgenerally extend laterally for asmuch as several meters. Thesestratified sediments overliemassively bedded, gravelly siltyclay deposits (unit D) that wereexposed at the base of most ofour trenches west of the fault.

Except where deformed withinthe fault zone, the stratigraphyis virtually horizontal through-out our trench exposures(Figures 3 and 4).

The sediments comprising unitC (Figures 3 and 4) are inter-preted to be fluvial overbanksediments from Rose Creekbecause they are fine grainedand thinly bedded, maintainfairly uniform thickness, andare laterally continuous overseveral meters. Three weakburied soil A horizons withinunit C allow for its division intofour subunits, C1 through C4

(Figure 4). Not all buried Ahorizons were recognized inevery exposure, but the lowertwo are commonly preservedand were the most useful forcorrelating between exposuresand across faults. These buriedsoils, which represent shorthiatuses in deposition, arebioturbated, have a darkercolor than the surroundingsediments, and obliterate strati-fication within the overbankdeposits. The best preservedand most laterally continuousstringers of sand and clay are

generally those that directlyoverlie the buried A horizonsand have had the least post-depositional disturbance.

Unit D is a gravelly silty claycapped by a buried A horizon.This distinct unit is consider-ably darker, more massive, andcontains more clay than unit C.Furthermore, clasts rangingfrom pebbles to large cobblesare common throughout thisunit. In trench T-8, severalcobbles were grouped togethernear the top of unit D1 andwere associated with culturalartifacts. Based on the sorting,color, and massive character ofunit D, we interpret this to beprimarily a colluvial or debris-flow deposit.

The modern soil, with a dark,organic-rich A (topsoil) horizon,a discontinuous E (albic) hori-zon, and a weakly tomoderately developed Bt

(argillic) horizon, is developedin the upper 60-80 cm of thesection below the artificial fill.The argillic horizon indicatesthat this soil was exposed at thesurface substantially longerthan at any of the buried Ahorizons.

Gravel-Filled ChannelTrench T-2 exposed a singlegravel-filled channel trendingnearly perpendicular to thefault that we used as the pierc-ing line. The channel isstratigraphically part of unit C2

in all exposures east of the fault(Figure 4). Within the faultzone, the channel fills the baseof a narrow rill or gully thatincised through unit C2 and intounit C3. West of thewesternmost fault strand, thechannel was removed prior to

or during grading of the site in1960.

The channel contains coarsesandy gravel to gravelly sand inthe thalweg. These coarsedeposits fine upward into, andin some places become indis-tinguishable with, the overlyingfine-grained overbank deposits.The sandy gravel at the base ofthe channel is 30-50 cm wideand 10-25 cm deep. Thesubangular to rounded clastswere generally smaller than 5cm in this linear deposit thatflowed east to west across theRose Creek terrace.

Five samples of detrital char-coal from units C2, C3, C4, and

D1 were dated by acceleratormass spectrometry techniquesand yielded dendrochronologi-cally calibrated ages rangingfrom 8.1 to 9.4 kyr. The maxi-mum age of unit C2 is bestconstrained by the youngestdate from the underlying unitC3, or about 8.1 kyr. Becausethe channel is incised into unit

C2 this date also represents themaximum age of the channel.The actual age of unit C2 andthe channel, however, is prob-ably not significantly youngerthan 8.1 kyr.

Determination of Slip and SlipRateThe amount of slip was deter-mined by excavating thedistinctive gravel-filled channelinto and across the severalstrands of the fault zone. Theposition of the channel, indi-vidual fault strands, and eachtrench wall was accuratelymapped using a surveyinginstrument (Figure 6). Based onthis map, we reconstructed thechannel such that adjacent

channel pieces were realigned(Figure 8). We backslipped thewesternmost strand to thelocation where the stratigraphicunit containing the channelwas cut out during the gradingof 1960; because the channelmay have been present evenfarther to the northwest, theamount of slip in the recon-

The most recent event onthe Rose Canyon faultzone was likely on theorder of M 7.

Dating of shells from fissures filled by an overlyingfaulted Indian midden deposit constrains the timing ofthe latest rupture to after A.D.1523.

Figure 4. Log of southeastern part of southwest wall of fault-parallel trenchT-2 showing flat-lying stratigraphy of the site and buried channel that wasused as a piercing line. Units A, E, and Bt represent the topsoil, albic, andargillic soil horizons that formed in silty to clayey sands of unit C. Unit Cwas subdivided into four subunits (C1-C4). In this exposure, units C1 and C2are undifferentiated. The subunits are separated by weak buried A horizons(short vertical wavy lines). Minor discontinuous lenses of sand (stipple) andclay (solid) are present throughout unit C.


struction is a minimum value.This reconstruction results in aminimum of 8.7 m of right-lateral strike-slip that postdatesdeposition of the gravel chan-nel. Vertical displacement ofthe channel across all but the

westernmost strand was consis-tently west-side-up and totaled0.7 m. Although no piercingpoints were available to mea-sure the vertical slip across thewesternmost strand, the 11-cmvertical separation of the top ofUnit D in trench T-9 impliesthat this fault produced onlyminor vertical slip. We estimatea horizontal to vertical slip ratioof about 10:1 for the MountSoledad fault (zone) at the RoseCreek site.

To establish a conservativeminimum slip rate, we use the

minimum of 8.7 m of brittleslip, assuming no rotationaldeformation, and a maximumage of the channel of 8.1 kya.This yields a minimum earlyHolocene to present slip rate of1.1 mm/yr. This rate is a mini-mum from the perspective that(1) the slip determination is aminimum value, (2) the age ofthe piercing channel is a maxi-mum, and (3) other strands ofthe Rose Canyon fault zonemay also be active. If the chan-nel is as much as 2 kyr youngerthan the underlying datedstrata, then the slip rate is about1.5 mm/yr. If the change inchannel trend through the faultzone is the result of horizontaldrag, the lateral displacement isat least 10 m, which increasesthe slip rate to nearly 1.7 mm/yr.

Based on the above uncertain-ties and allowing for theprobability of some slip onother strands of the fault zone,we suggest that the overall sliprate is between 1.1 and 2 mm/

yr, with a best estimate of about1.5 mm/yr. The lower bound of1.1 mm/yr is determined by ourtrenching results, whereas theupper bound of 2 mm/yr isestimated from a comparison ofthe geomorphology associatedwith the three principal strandsof the Rose Canyon fault zone.Based on this comparison, itappears likely that the MountSoledad strand carries themajority of the slip.

Timing of Past EventsFrom the trench exposures, wehave evidence for at least threeevents since about 8.1 kyr. Theearliest event is suggested by alow scarp into which thegravel-filled channel incised.East of the fault, the channel iswithin unit C2 (Figure 4). As thechannel crosses the fault zoneto the west, it incises intoprogressively older and deeperstrata so that the channel cutsinto units C3 and the top of C4.This indicates the presence of ascarp that formed betweendeposition of unit C3 and theformation of the channel.Furthermore, units C1 and C2

thin to the west across thisscarp (Figure 5) indicating thatscarp formation occurred soonafter the deposition of unit C3.Therefore the event that pro-duced the scarp must haveoccurred soon after about 8.1kyr, the maximum age of unitC3. The minimum of 8.7 m ofslip on the channel occurredafter this ~8.1 kyr event.

A second event is indicated bya group of fault splays thatdisplace unit C1 but do notdisplace the base of the Bt soilhorizon (Figure 5). The time ofthis event is unconstrained, butthere must have been sufficienttime for soil development toobliterate the fault traces in thesoil.

The most recent event rupturedthe ground surface and abruptlydisplaced all soil horizons onthe eastern strands of the faultzone in trenches T-1 and T-4(Figures 3 and 5). The absenceof significant bioturbationwhere the Bt horizon is juxta-posed over the A horizonindicates that this event is very

We suggest the overallslip rate is between 1.1and 2 mm/yr.

Figure 5. Detailed map of the northwest wall of trench T-4 showing theprimary structural elements and stratigraphic units. Unit C was subdividedon the basis of weakly developed buried A horizons (shown as shortvertical wavy lines). Unit C contains thin beds and stringers of clay (solid),silt (open), and sand (stippled). Evidence of bioturbation is shown by thenumerous krotovina (k). Note that the northeastern strands displace the A,E, and Bt soil horizons, which is the result of the most recent surfacerupture, whereas the southwestern strands could not be traced to the topof unit C1.

Figure 6. Plan view map of trenches excavated to determine displacementof the buried channel across strands of the fault. Channel is shown bystipple pattern. Faults are heavy lines, dashed where approximatelylocated, queried where uncertain, and are shown at the stratigraphic levelof the channel. Trenches shown with a dashed line were excavated fromthe parking lot surface, all others excavated by hand from the pit floor.Portions of trench margins shown in bold indicate location of other figures.


young, probably less thanseveral hundred years. Thisinterpretation is supported byother studies, which documentthe latest earthquake as occur-ring on the order of ~300 yearsago, as discussed further below.

A minimum of three events inthe last 8.1 kyr implies a maxi-mum recurrence interval on theorder of 4,000 years. This alsorequires that the last two eventsproduced an average of morethan 4 m of slip per event. Theaverage recurrence intervalcould be substantially less thanthe maximum interval of 4,000years. These data indicate thatthe recurrence interval for largesurface-rupturing earthquakes ison the order of a few thousandyears.

Most Recent EventMultiple studies indicate thatthe most recent large surfacerupture on the Rose Canyonfault occurred during the pastfew hundred years. Near Mis-sion Bay at the Rose Creek site,the abrupt, distinct faultedcontacts within the surficial soil

horizons (Figures 3 and 5)imply that very little time haselapsed since the latest event.The absence of significantbioturbation where the Bt

horizon is juxtaposed over theA horizon indicates that thisevent is very young, probablyoccurring less than 500 yearsago (Lindvall and Rockwell,1995).

This estimate agrees with theresults of trenching studies inthe downtown San Diego areathat indicate a surface-ruptur-ing event occurred on strandsof the Rose Canyon fault zonepost-A.D.1420 The timing ofthis event is based on radiocar-bon ages of detrital charcoal infissure fillings from the SportsArena site (Woodward-ClydeConsultants, 1994). The lack ofhistorical accounts of largeearthquakes in the San Diegoarea constrains the timing ofthis event between A.D. 1420and 1769, the year the Spanishestablished the first mission inSan Diego.

Rockwell and Murbach (1998)demonstrated that the mostrecent rupture in the La Jollaarea is consistent with thetiming of the latest earthquakeon portions of the fault zonethat traverse the Mission Bayand downtown areas. Dating ofshells from fissures filled by anoverlying faulted Indianmidden deposit constrains thetiming of the latest rupture topost-A.D.1523 Using the cali-brated lower bound of A.D.1523 and an upper one ofA.D.1769 yields a best estimateof A.D. 1650 ± 125 for the ageof this event (Rockwell andMurbach, 1998). Because thisdate is indistinguishable fromthe dates obtained from thedowntown study (Woodward-Clyde Consultants, 1994), it isplausible, if not likely, that theyrepresent the same earthquake.

It is possible, however, thatthese data could representdifferent earthquake rupturesthat are closely separated intime and space, rather than asingle event.

The evidence for a near-histori-cal event at all three sitessuggests that the last largesurface rupture may haveinvolved the entire onshoreportion of the fault zone. Slip inthis most recent event has beenestimated at both the RoseCreek site and in La Jolla nearthe coast. Based on re-evalua-tion of the three-dimensionaldata from Lindvall andRockwell (1995), it appears thatthe most recent event produced~3 m of slip on the eastern faultstrands at the Rose Creek site(Rockwell and Murbach, 1998).This is consistent with a mini-

Figure 7. Southward view of hand-excavated trenches dug into base of thelarge rectangular excavation, which approximately coincides with the baseof artificial fill. Faults strike from upper left to lower right corner ofphotograph. Note cars in upper left corner for scale.

Figure 8. Generalized map and reconstruction of the faulted channel. Thereconstruction provides a minimum value of lateral offset because thestratigraphic units containing the channel are missing west of thewesternmost fault strand northwest of the shaded triangle.


mum displacement of >1 mestimated at the La Jolla site(Rockwell and Murbach, 1998),which is located over 5 km tothe north. Based on the amountof slip and a minimum rupturelength of at least the onshoreportion of the fault zone, themost recent event that occurredon Rose Canyon fault zone waslikely on the order of ~M 7.

ReferencesFischer, P. J., and G. I. Mills, The

offshore Newport-Inglewood-RoseCanyon fault zone, California:Structure, segmentation, andtectonics, in Environmental Perils,San Diego Region, edited by P. L.Abbott and W. J. Elliot, pp. 17-36,San Diego Assoc. of Geol. for theGSA meeting, San Diego,California, 1991.

Kennedy, M. P., Geology of thewestern San Diego metropolitanarea, California; Del Mar, La Jollaand Point Loma quadrangles in“Geology of the San DiegoMetropolitan Area, California,”Bull. Calif. Div. Mines Geol. 200,9-39, 1975.

Kennedy, M. P., and E. E. Welday,Recency and character of faultingoffshore metropolitan San Diego,California, Map Sheet 40, Calif.Div. Mines and Geol., Sacra-mento, 1980.

Legg, M. R., Geologic structure andtectonics of the inner continentalborderland offshore northern BajaCalifornia, Mexico, Ph.D.dissertation, Univ. of Calif., SantaBarbara, 1985.

Legg, M. R., and M. P. Kennedy,Oblique divergence andconvergence in the Californiacontinental borderland, inEnvironmental Perils, San DiegoRegion, edited by P. L. Abbott andW. J. Elliot, pp. 1-16, San DiegoAssoc. of Geol. for the GSA Mtg.,San Diego, California, 1991.

Lindvall, S. C., and T. K. Rockwell,Holocene activity of the RoseCanyon fault zone, San Diego,California, J. Geophys. Res., v.100, p. 24,121-24,132, 1995.

Lindvall, S. C., T. K. Rockwell, andC. E. Lindvall, The seismic hazardof San Diego revised: Newevidence for magnitude 6+Holocene earthquakes on theRose Canyon fault zone, Proc.Fourth U.S. Natl. Conf. Earth-quake Eng., 11, 679-688, 1990.

Rockwell, T. K., and S. C. Lindvall,Holocene activity of the RoseCanyon fault in San Diego,California, based on trenchexposures and tectonic geomor-phology: GSA, Abstracts withPrograms, v. 22, p. 78, 1990.

Rockwell, T. K., and M. Murbach,Holocene earthquake history ofthe Rose Canyon fault zone,NEHRP Final Technical Report,USGS Grant No. 1434-95-G-2613, p. 36, 1998.

Rockwell, T. K., R. S. McElwain, D.E. Millman, and D. L. Lamar,Recurrent late Holocene faultingon the Glen Ivy North strand ofthe Elsinore fault at Glen IvyMarch, in “Neotectonics andFaulting in Southern California,”Geol. Soc. Am. Guidebook, 167-175, 1986.

Sharp, R. V., Variable rates of lateQuaternary strike-slip on the SanJacinto fault zone, southernCalifornia, J. Geophys. Res., 86,1754-1762, 1981.

Sieh, K. E., Lateral offsets andrevised dates of large prehistoricearthquakes at Pallet Creek,southern California, J. Geophys.Res., 89, 7641-7670, 1984.

Woodward-Clyde Consultants,Fault hazard investigation for theentertainment and sports center,San Diego, California, unpub-lished consulting report preparedfor Centre City DevelopmentCorporation, p.19, 1994

Museum FeaturesWork of SCEC ScientistSCEC’s Thomas Rockwell (San Diego State University) was inter-viewed about his research on the paleoseismicity of the San Andreasfault by the American Museum of Natural History for the Hall ofPlanet Earth, a new permanent exhibition devoted to earth sciences.A portion of that filmed interview will be shown in the new hall.

NISEE Releases Softwareon CD-ROM"NISEE Software Library CD-ROM," a comprehensive collection of112 engineering research software programs, is available from theNational Information Service for Earthquake Engineering (NISEE),University of California, Berkeley. Applications range from strongmotion data processing programs to geotechnical and structuralanalysis tools; user documentation for all programs on the CD-ROMis available through NISEE. Cost is $139, including shipping in theU.S. Order from NISEE, University of California, Berkeley, PEERBuilding 451 RFS, 1301 South 46th Street, Richmond, CA 94804-4698; (510) 231-9403; fax: (510) 231-9461;[email protected]. Web site: WWW.EERC.BERKELEY.EDU.

Geologist PlacementServiceOnsite Environmental Staffing, a placement firm for professionals inthe environmental and engineering fields, has positions availableranging from entry level to registered geologists with leading main-stream geology firms in southern California. Email or fax resumes, orcall to set up an interview. Erik Applebee: (714) 245-4725; (800)338-4199; fax (714) 954-0726; [email protected]

Strong Motion ConsortiumSuccessfully LaunchedThe Consortium of Organizations for Strong-Motion ObservationSystems has submitted its bylaws as a public-interest nonprofitcorporation for earthquake safety to the Secretary of State in Sacra-mento and now has an office and staff in the PEER Building inRichmond, California (COSMOS, c/o PEER, 1301 S. 46th Street,Richmond, CA 94804-4698; Tel: (510) 231-9436; fax: (510) 231-9471; email: [email protected]). The launch of this umbrellaconsortium has met with widespread interest among earthquakeengineers, emergency planners, and seismologists. COSMOS is nowin a position to influence management of the new National SeismicSystem recently approved by Congress. A subcommittee of COSMOShas already produced recommended standards for a national reposi-tory of strong motion records.

A regular board and working committees will be elected at a generalmeeting of all interested persons and eligible members on Septem-ber 16 following the SMIP 99 annual meeting near either Oakland orSan Francisco airport. At that meeting membership fees and COS-MOS operating policies will be discussed. A descriptive brochure isavailable on request from the above address.


A new document to help engi-neers, geologists, and buildingofficials evaluate and takeprotective measures againstliquefaction hazards in south-ern California is now available.

Recommended Procedures forImplementation of DMG Spe-cial Publication117—Guidelines for Analyzingand Mitigating LiquefactionHazards in California wasproduced by a committee ofengineers and geologists withacademic, practicing, andregulatory backgrounds. Thebook was published by theSouthern California EarthquakeCenter.

Liquefaction happens whenloose sandy soils below the

State Guide Tells How to EvaluateLiquefaction DangersBy Mark Benthien

ground water lose strengthbecause of strong groundshaking. This liquefactioncould result in settling andsliding, leading to damage tobuildings on such soils. Thenew report asserts that lique-faction is sufficientlyunderstood now so that soilbehavior during earthquakescan be predicted. Therefore,the consequences of thatbehavior can be planned forand damage possiblyavoided.

The report contains recom-mendations for whatconstitutes an adequateevaluation of the liquefactionpotential of an area, alongwith the hazards that arisefrom liquefaction, including

settling, lateral spreading, flowslides, and loss of bearingcapacity. The report presents anoverview of the recommendedmethods of analysis and ad-dresses the commontechnologies available to miti-gate the effects of liquefaction.Also discussed are the merits ofsoil improvement, structuraldesign for the effects of lique-faction, and the inherent risksassociated with such decisions.

The report is a summary of 18months of study and delibera-tion in response to theCalifornia Department ofConservation’s Division ofMines and Geology (CDMG)Special Publication 117, whichpresents general guidelines forevaluating and mitigating

seismically induced soil lique-faction and landslides.

SP 117 is an outcome of theCalifornia Seismic HazardsMapping Act (SHMA), whichrequires that seismic hazards,including liquefaction, beevaluated and mitigated, ifneeded, for all new construc-tion in the state. The state isnow publishing maps delineat-ing areas that are believed to besusceptible to both liquefactionand landslide hazard.

With the release of SP 117,building officials in the Depart-ment of Building and Safety ofthe City of Los Angeles and theDepartment of Public Works ofthe County of Los Angelesrequested assistance withdeveloping standard proce-dures to implement the new

Using the New Hazard Maps

Thomas Henyey, director of SCEC, responds to a question.

Charles Real describes the Seismic Hazard Mapping Act to assembledreporters, as Thomas Blake looks on.


Liquefaction Implementation Committee

Geoffrey R. Martin (co-chair),Department of Civil Engineering, USC

Marshall Lew (co-chair), Law/Crandall, a Division of LawEngineering and Environmental Services, Los Angeles

K. Arulmoli, Earth Mechanics, Inc., Fountain Valley

Juan I. Baez, Hayward Baker, Santa Paula

Thomas F. Blake, Fugro West, Inc., Ventura

Ebrahim Simantob, R. T. Frankian & Associates, Burbank

T. Leslie Youd, Department of Civil and EnvironmentalEngineering, Brigham Young University

Local and state government liaisons to the committee:

Johnnie Earnest, Orange County

Fred Gharib, Los Angeles County

J. Goldhammer, San Diego County

David Hsu, City of Los Angeles

Steve Kupferman, Riverside County

Jim O’Tousa, Ventura County

Charles R. Real, DCMG

Wes Reeder, San Bernardino County

Workshop Will BeAvailable on CD

A workshop on use of the report was conducted by its authorsfor city and county officials and consulting engineers on June17, 1999, at the Davidson Conference Center at USC. The full-day workshop included a chapter-by-chapter overview of thematerial and afternoon panel sessions for discussion and feed-back. The chapter-by-chapter overview was given using anine-part Microsoft PowerPoint presentation and wasaudiotaped. The presentation will be placed onto CD-ROMalong with the audio for a narrative run-through of the over-view. The afternoon panel discussion sessions were videotaped.Highlights will be placed on the CD-ROM. Questions and theiranswers will be included. The final product will be availablefor sale from SCEC. Watch for an announcement of the CD-ROM’s release at www.scec.org or in this newsletter.

requirements for projects re-quiring their review. Theysought cooperation from otheragencies in southern California.Officials from the counties of

Riverside, San Bernardino, SanDiego, Orange, and Venturaagreed to participate; CDMGand FEMA also lent support tothis effort.

The agencies’ request for assis-tance was made through theGeotechnical EngineeringGroup of the Los AngelesSection of the American Societyof Civil Engineers (ASCE) in1997. A group of practicinggeotechnical engineers andengineering geologists wasassembled to develop imple-mentation procedures. Theliquefaction implementationcommittee was organizedthrough the auspices of SCECand convened at SCEC’s admin-istrative headquarters at USC.Its task was to aid the severalcounties and many cities insouthern California with imple-menting SHMA for liquefactionhazards and to provide anoverview of analysis techniquesand methods of mitigation.

The committee’s report wasreleased on April 19, 1999, anda press briefing was held atUSC to announce its availabil-

Authors of the report take the stage for the afternoon discussion session.(l to r) Ted Smith, Arul Arulmoli, Abe Simantob, Geoff Martin, MarshallLew, Charles Real, Thomas Blake, Juan Baez, and Les Youd.

ity. Members of the implemen-tation committee were presentto answer questions from themedia. Three television newsstations, two radio stations, andtwo newspapers were present.

Copies of the report can beobtained by using the orderform included in this newsletteror by calling 213-740-5843.

Copies of SP117 can be ob-tained by sending a check for$15 payable to Division ofMines and Geology to P.O. Box2980, Sacramento, CA 95814.On the memo line of the check,write “SP117.” Do not add taxor shipping.


Internship program coordinator Mark Benthien recently announcedthe participants in the sixth annual SCEC Summer InternshipProgram. Here is a list of the students, their research mentors,projects and goals, and personal goals.

Natanya BlackGeology, UC Santa BarbaraTom Rockwell, CSU San Diego

Reconstruction of Trench Logs as a Test forInterpreted Paleoseismic Events

Personal goals: after obtaining my under-graduate degree, I intend to continue working

toward my M.S. and Ph.D. My goal is to work at an academicinstitution doing research and teaching. I hope to do research as ageologist in seismic studies and petrology. Teaching is another veryimportant aspiration for me. I want to work toward a broaderunderstanding of geology and earth mechanics in our society. Iwill endeavor to discover new geologic perspectives in my futureresearch, and I hope to be able to convey that knowledge to oth-ers.

Project goals: the goals of this project are very relevant to scienceand society, because earthquakes not only have the potential tocause millions of dollars in damage and terrible loss of life, but ina more insidious way, they can produce an underlying panic orfear in the society. One potential solution to this fear is through thestudy and explanation of the earthquake phenomenon. One ele-ment of this is to study the past occurrences of earthquakes inSouthern California to better understand how often, and when,future earthquakes are likely to occur on Southern California’smany earthquake faults. Ultimately, this information can serve tolessen the anxiety that people experience as they better compre-hend that earthquakes are not random processes (acts of God), butrather occur in a systematic way due to loading on the SouthernCalifornia faults. Developing methods to correctly interpret pastearthquakes is critical in the evolving field of paleoseismology andfor the correct future assessment of seismic activity of SouthernCalifornia’s faults. This knowledge can bring a sense of security,through better education and preparedness, to the society as awhole.

Thumbnail Sketchesof This Year’s Interns

SCEC’s Summer Crop

Debra EinsteinEnvironmental Analysis & Design, UC IrvineMark Legg, Legg Geophysical

Compile Updated Fault Maps of the SouthernCalifornia Continental Borderland (OffshoreRegion) for the Master Model

Personal goals: I would like to work in the field of environmentalassessment and get hands-on experience. I want to learn moreabout faults/fault zones so that I may use that knowledge in assess-ing different areas. This becomes especially important inCalifornia.

Project goals: a project of this magnitude is important, especiallyfor SCEC. To be able to read on a map faults/fault zones and seis-micity is especially significant for assessing types of hazardsaccompanied by building in these areas.

Marie Herrera AdsettsGeophysics, UC Santa BarbaraBruce P. Luyendyk, UC Santa Barbara

Mapping Small-scale Crustal Deformationusing Paleomagnetic Vectors and GIS.

Personal goals: I’d like to understand theresearch processes involved because I understand it’s a difficultand lengthy process. I’d like to become familiar with the geologichistory of this particular area and its relevant significance. I wouldalso like to be able to communicate this information successfullyin both speaking and writing. Academically, I’d like to pursue amaster’s degree in the geological sciences but I must first obtainmy bachelor’s degree. My plan is to enter graduate school by thefall of 2001.

My career goal is rather unstructured at the moment. After obtain-ing my B.S. degree, I hope to work for the USGS as an intern togain some experience and attend graduate school. After complet-ing my master’s degree, I’d like to obtain a position involvingseismology. I am also entertaining the idea of teaching eithergeology or mathematics at a community college.

Project goals: mapping details of the crustal rotations indicatessizes of blocks where strain has occurred. The detailed mappinghistory of deformation is relevant because it suggests how southernCalifornia has deformed, as well as how it may be deformingpresently. Thoroughly understanding the patterns in paleomagneticdata and faults for the western transverse ranges allows for moreprecise predictions involving crustal rotations and fault and basindevelopment for the southern California region.


Grant KierGeology, Sociology/Philosophy, University ofColoradoKarl Mueller, University of Colorado

Determining Activity of the Oceanside de-tachment Offshore Southern California

Personal goals: my long-term goal is to teachearth science in either a high school or college setting. I wish tointegrate my undergraduate studies in sociology, philosophy, andgeology toward studies of natural hazards that have the potentialto affect society. Upon completing my undergraduate work, I willapply to master’s programs in geology at research-oriented univer-sities. After completing a master’s, I hope to work in aneducational setting for a geotechnical company.

Project goals: this project is relevant to the scientific effort tocreate a master model of seismic hazards for southern California.This project may also contribute to an understanding of upliftedterraces along the California coast. Should this project provideevidence for yet unknown activity along the Oceanside detach-ment, this project will have enormous societalimpact—influencing urban planning and development and otherefforts to reduce seismic hazard risk.

Christopher LynchComputer Science/Geology, CSU San DiegoSteven M. Day, CSU San Diego

Inversion of Teleseismic Receiver Functionsvia Evolutionary Programming

Personal goals: besides my interest in thecomputational aspect of this project, I am intensely interested inlocal structure and tectonics, so I’m working to develop the skillsnecessary to enable me to contribute to this area of research. I willgraduate from SDSU in August and hope to continue my studiesthrough a master’s degree program in computational sciencecombining computer and geological/geophysical science.

Project goals: I think this work fits into the SCEC goal of generatingthe “master model,” which includes as a main point the need todetermine the seismic wave velocities to use in computing theo-retical seismograms. This code is being used to explore methods ofdetermining the velocity structure of the crust through inversion ofteleseismic receiver functions. By developing a range of tech-niques and comparing their results we can better verify andconstrain these results, producing a more accurate model of theEarth’s crust. From this we can learn to better understand thetectonic forces that affect us and develop the strategies to existsafely with seismic hazards.

Nathan RobisonGeological Engineering, University of NevadaJohn Anderson, UNR Seismology Laboratory

Comparison and Refinement of SouthernCalifornia Seismic Hazard Analysis

Personal goals: along with my wife anddaughter, I intend to grow as a family that promotes science,bicycle riding, and common decency. I will graduate cum laudewith a B.S. in geological engineering and expect to begin graduatestudy towards a doctorate in mining and geological engineeringwithin the following year. Graduate work is also expected toinclude seismology and mechanics research. A career for the nextseveral years in the environmental and geologic consulting indus-try will give way to a university position after attainment of anadvanced degree.

Project goals: justifiable revision of attenuation models has thepotential to save billions of dollars in public works and public andprivate construction costs. Establishment of techniques for reduc-ing the uncertainty of upper-bound acceleration models may be oflong-term use to the science.

Kelly SchmokerGeology, CSU, San BernardinoSally McGill, CSU, San Bernardino

Paleoseismic study of the San Andreas faultnear San Bernardino

Personal goals: I would like to go on to gradu-ate school in the field of geology, focusing on paleoseismicstudies. Eventually, I would like to become a professor and still doresearch.

Project goals: the San Bernardino segment of the San Andreas faultrepresents a major hazard for the Inland Empire area. It is impor-tant to know how frequently this portion of the fault producesearthquakes. Previous estimates of the probability of an earthquakeon the San Bernardino segment of the San Andreas fault wereabout 28 percent in the next 30 years (Working Group on Califor-nia Earthquake Probabilities, 1995). This was based on assuming arecurrence interval of 146 years. Prior work at the Plunge Creeksite suggests that it may have been more than 350 years since thelast earthquake at that site. If this is true, then either the probabilityof an earthquake in the next 30 years is more than 28 percent orthe average recurrence interval is longer than the 146-year esti-mate that was used to calculate this probability. Our work thissummer will help to distinguish between these two possibilities.


Ashley StreigGeology, Occidental CollegeKerry Sieh/Doug Yule, Caltech

Paleoseismic Investigation of the San AndreasFault at Burro Flats

Personal goals: I plan on attending graduateschool upon my graduation from Occidental College in May 2000.I hope to attain both my master’s degree and Ph.D. in a field ofgeology or earth sciences. This internship is a great opportunity forme to gain research experience in a field where I hope to continuemy education. Through this research project, I hope to focus myacademic goals in the field of geology.

Project goals: the seismic history in Burro Flats will help us deter-mine how frequently earthquakes happen in this region and howintense the ground shaking may be. We will determine how thefault behaves in Burro Flats between San Bernardino and Indio.Interestingly, this is the only segment of the San Andreas fault thathas not had an earthquake in the last 200 years, so examination atthis segment is important to the overall understanding of the SanAndreas fault system. Seismic history of this region will help usgauge future activity in this region. As the population of the InlandEmpire continues to grow, this information will help inhabitantsprepare for future earthquakes. The results of this study will showus how large earthquakes are and how frequently they occur alongthis section of the fault, allowing assessment of risk in the area.

Kathryn van RoosendaalGeophysics, CSU, NorthridgeDavid Okaya, USC

3-D analysis of LARSE ’94 data

Personal goals: my long-term goal is to obtainat least a master’s degree in volcanology and

geophysics. I am especially interested in geophysical modeling ofmagma chambers and other plutonic structures.

Project goals: the LARSE project is of great importance to theinhabitants of the Los Angeles basin and the San Fernando Valley.The Northridge earthquake drove home the importance of locatingblind thrust faults and other structures far beneath the surface. Thestudy is also important scientifically for its look into the structureof the lithosphere and the workings of plate tectonics.

Adam WebberGeology, UC Santa BarbaraE. A. Keller, UC Santa Barbara

The Earthquake Hazard for the Rincon CreekAnticline (RCA) and the Associated BuriedReverse Fault, Carpinteria, CA.

Personal goals: as an undergraduate geology major, I am striving toexperience a broad spectrum of fields within this major, becausethis is my chance. These are my academic goals, and what I hopeto achieve is an understanding for what I want to pursue in thefuture. I have a love for understanding the present and past pro-cesses of the earth, and this is what I want to focus on in thefuture. I am set on going to graduate school once I have receivedexperience and an idea of what I want to pursue. Plate tectonicsand geomorphology are two specific areas that I have recentlystudied and have gained a strong interest in. I hope to follow upon these newfound interests by gaining some experience duringthe summer of 1999.

Project goals: the importance of this project with respect to sci-ence and society is to gain a better understanding of theearthquake hazard that the Rincon Creek Anticline (RCA) presentsto the city of Carpinteria, CA. By determining a rate of uplift andwhether it was a constant rate or rapid jumps as a result of earth-quake activity, the threat posed to Carpinteria can be analyzed.The RCA is a typical young fold of the Santa Barbara fold belt, andby determining localized characteristics of the structure, extrapo-lated information can aid in studying the larger picture of the SantaBarbara fold belt.


Many Jurisdictions in L.A. AreaSCEC Outreach Presentations to the EmergencyPreparedness Commission

On June 9, 1999, Jill Andrews and Mark Benthien gave presenta-tions to the Emergency Preparedness Commission for Los AngelesCounty and the cities in the area. Andrews spoke about SCEC andits outreach programs, highlighting the programs that SCEC isconducting with the city and county of Los Angeles, that include:

• A joint task force of engineers, scientists, and city and countyofficials to study vulnerable buildings in Los Angeles—specifi-cally “tuck-under” or “soft story” structures such as theNorthridge Meadows apartment building that collapsed in theNorthridge earthquake—and nonductile concrete structures likeparking garages.

• A partnership with the California Division of Mines andGeology to conduct a series of workshop to educate cityofficials on how to use liquefaction hazard maps. [Ed. note: Seeseparate article in this issue for a description of a workshop.]

Andrews also described “The Real Meaning of Seismic Risk”—aproposed symposium that would feature lively exchanges amongscientists, engineers, building officials, policymakers, insurers,developers, and the media—with the goal of encouraging publicparticipation in and understanding of earthquake science throughinteractivity with SCEC.

The symposium would consist of a panel of experts with differingor opposing views who would make short presentations on urbanseismic risk issues. Topics could include:

• A critique of methods used to interpret the earthquake threat

• Vulnerability of tall buildings and other structures near faults

• Whether the “life safety” design code is the best practice, givenwhat we now know from Northridge

• Cost-benefit analyses of various retrofitting techniques andcodes for new construction

• Socioeconomic impacts of earthquakes and secondary hazardsin California vs. other natural hazards outside the state

A public report, with audio and video tapes of the symposium’sproceedings, would be made available through SCEC.

Andrews concluded her presentation with a request for thecommission’s partnership in this project, which they granted. ChrisWright, president of the Los Angeles chapter of the Business andIndustry Council on Earthquake Planning and Preparedness(BICEPP) also indicated that BICEPP will partner with SCEC on thisproject.

Benthien then gave a presentation on the Los Angeles RegionSeismic Experiment (LARSE II), which will be conducted in Octo-ber 1999. The study will use ultrasound-like images of faults, basindepths, and other features deep in the Earth’s crust to determinewhere earthquakes can occur and how the ground will shake as aresult.

To generate the vibrations needed to form the “seismic image,”more than 100 small buried charges will detonated over a four-to-five-day period along a line from Pacific Palisades to the westernMojave Desert. To record the vibrations, 1,000 seismometers willbe placed along the same line. More than 60 separate landownerswill grant permission for this joint USGS-SCEC project to be lo-cated on their property.

SCEC Will Pay for MembersStress-Triggering and DeformationSoftware Training Workshop

Three pieces of software in current use for studies of earthquakeand fault interaction will be the subject of a two-day hands-onworkshop to be held in Palo Alto on September 8-9. The goal is forthe participants to develop sufficient skill that they will use thesoftware in their own research and teach its use to others.

The course will introduce participants to Coulomb 1.0, 3D-DEF,and VISCO1D. The applications will be presented by their authors,who will walk participants through tutorials and provide simplemanuals. All software and manuals will be available electronically.

To give everyone keyboard/monitor access, the workshop is lim-ited to 45 participants. Housing will be in suites at the SchwabResidential Center of the Stanford Business School. Instruction willtake place at the Mitchell Earth Sciences 20-station training center.

SCEC News Briefs


Participants will arrive on the afternoon of September 7, followedby a welcoming reception that evening. Other events include abarbecue on Wednesday evening and an afternoon run. The work-shop is scheduled to conclude by 6 p.m. on September 9. SCECwill cover tuition, airfare within California, airport van shuttle,meals, and accommodations for its members.

Coulomb 1.0, principally written by Shinji Toda (ERI, Japan)—anevolution of GEN 1.0 by Geof King (IPG, Paris)—is a fast, menu-driven Mac program rich in graphics that performs 3D elasticdislocation and a limited number of 2D boundary element calcu-lations of deformation and stress in an elastic half-space. RossStein will help Shinji Toda teach this session.

3D-DEF, written by Joan Gomberg (USGS, Memphis) and MichaelEllis (CERI, Univ. Memphis), performs elastic dislocation bound-ary-element calculations. The program enables a variety ofboundary conditions to be applied, which makes the model quiteflexible. The program and manual can be obtained viaanonymous FTP to beagle.ceri.memphis.eduin the “/pub/gomberg” directory.

VISCO1D, written by Fred Pollitz (UCD) calculatesthe response of a spherically stratified elastic-viscoelastic medium to the stresses generated by fault slipor dike opening occurring in one of the elastic layers.

To register, or for more information, email Ross Stein [email protected]. Please indicate your choice of software, so ses-sions can be planned accordingly.

Workshop to Be Held October 3-5Plate Boundary Observatory toBe Defined and Planned

The Plate Boundary Observatory (PBO) is a proposed facility forinvestigating active tectonic and magmatic processes of the Pa-cific/Juan de Fuca–North American plate boundary throughmeasurements of crustal deformation. The study of plate boundarydeformation is a research area that deserves increased attentionfrom a broad spectrum of Earth scientists. Toward that end, aworkshop will be convened on October 3–5 in Snowbird, Utah, tohelp develop the plans for such an observatory.

The PBO Steering Committee invites participation from a broadspectrum of earth scientists in a workshop to help define the PBOconcept and plan for its implementation. The workshop will pro-duce a report outlining the scientific basis for the PBO, itsinstrumentation requirements and deployment strategy. It willdescribe the ways the facility can advance earth science researchand contribute to education and outreach.

The workshop will be limited to 100 participants. Each applicantto the workshop is asked to provide a brief statement of interests,including how he or she can contribute to the goals of the work-shop. Partial support (air travel, hotel, and meals) will be providedby workshop funds. To apply for an invitation, see: HTTP://WWW.SCEC.ORG/NEWS/99NEWS/PBO.HTML.

The chief observational requirement of the PBO is a characteriza-tion of the three-dimensional deformation field over the maximumranges of spatial and temporal scales. The PBO should be de-signed to study long-term, regional tectonic processes as well asshorter-term, smaller-scale processes that may be more closelyrelated to natural hazards, such as earthquakes and volcaniceruptions.

It is proposed that the PBO be coordinated and integrated with theproposed U.S. Array and the San Andreas Fault Observatory atDepth (SAFOD) projects as part of the EarthScope initiative being

developed at NSF. In addition to advancing our basic scientificknowledge of active tectonic processes, the facility will

improve seismic and volcanic hazard assessment andcontribute to earth science education and out-

reach in the U.S.

The PBO Steering Committee consists ofYehuda Bock, Andrea Donnellan, Don Helmberger,

Tom Henyey, Ken Hudnut, Gene Humphreys, ChrisMarone, Meghan Miller, Bernard Minster, Barbara Romanowicz,Paul Segall, Paul Silver, Bob Smith, Seth Stein, Wayne Thatcher,George Thompson. The workshop is jointly sponsored by theNSF’s Division of Earth Sciences, UNAVCO, IRIS, NASA, USGS,IGPP, and SCEC.

Moving to Private SectorSteven Ganz Leaves WSSPC

Steven Ganz, first executive director of the Western States SeismicPolicy Council (WSSPC), recently moved to the private sector,joining an Internet start-up based in San Francisco. In that move,the Earthquake Information Providers (EqIP) also loses a member ofits steering committee.

A graduate of the John F. Kennedy School of Government atHarvard, Ganz’s master’s degree is in public policy and urbanplanning. Using that background and management and entrepre-neurial skills during his tenure at WSSPC, he developed severaloutstanding programs, including a state-of-the-art Web service anda series of excellent annual meetings and workshops featuringearthquake-related public policy issues, recent research on seismichazards in the western states, and insurance-related issues andpolicy recommendations.


Following EqIP’s approval of the revised structure, the process ofrefining the site began, under the auspices of the MultidisciplinaryCenter for Earthquake Engineering in Buffalo:

• Refining and filling in the contents of the database

• Developing the Web interface to retain simplicity but add morecapability

• Maintaining the site

• Coordinating the participation in and use of the site by EqIPmembers

• Marketing the site—getting it better known and more widelyused

A new proposal submitted to NSF will ask for support for thesecontinuing efforts at least through the year 2000.

With the Red Cross and AllstateSCEC Helps Add Earthquakes toDisaster Curriculum

SCEC has joined with the American Red Cross and the AllstateFoundation in a project that will help teachers integrate disastersafety concepts into their regular lesson plans.

Announced at a news conference June 1, 1999, the Children’sDisaster Safety Curriculum will provide lessons, activities, anddemonstrations that teachers can use to incorporate a hazard-related topic into other classroom subjects. To address the veryreal threat of earthquakes in our region, basic earth science andearthquake information will be included in the curriculum materi-als.

“Integrated curricula that use real-life or hands-on examples toconvey lessons in science, math, language arts, and social studiesare indeed the best tools for teaching today’s students,” saidSCEC’s outreach director, Jill Andrews. “An issues-based integratedcurriculum such as this one is especially valuable in light of therequirements of the new National Science Education Standards.”

The Children’s Disaster Safety Curriculum will be made availablethrough local Red Cross chapters at a nominal cost. For moreinformation, contact Rocky Lopes, Disaster Services, AmericanRed Cross National Headquarters, at (703) 206-8805 [email protected].

He developed partnerships with other organizations that focus onseismic issues and research and was a technical advisor and leaderamong such partnerships.

Jill Andrews, SCEC outreach director, said, “We in the SCEC com-munity will greatly miss Steve’s professional approach to bringingtogether people in both public and private organizations to reachconsensus on earthquake-related public policy issues.

“We thank Steve for his contributions to society through associa-tion with WSSPC, and wish him all the best in his new endeavor.”

In a recent message to the earthquake research community, Ganzsaid of the move, “I am very sad to leave the organization, but atthe same time very excited about my new opportunities. I havethoroughly enjoy working at WSSPC. I truly believe that WSSPC isan organization with a focused mission and a unique ability tocontinue making a significant impact in improving the means toreduce the threat of earthquake hazards.”

EQNet Developed with SCEC ParticipationOmnibus Earthquake Resource Site Now Online

Chairperson Jill Andrews announced that the Earthquake Informa-tion Providers (EqIP) group has finished a full revision of thegroup’s Web-based catalog of earthquake-related resources. Inaddition, the group is submitting a proposal to NSF to continue thedevelopment of the site.

EQNet (WWW.EQNET.ORG), a project originally launched by EqIP in1996 is intended to be a “one-stop” source for locating Internetsites related to earthquake hazards. Part of its purpose was toconsolidate access to those resources, and part of it was to elimi-nate duplication among earthquake information providerssupported by NSF.

All EqIP members have Web sites and links pages, all of whichrequire maintenance. EQNet allows them to reduce the amount ofmaintenance on their own link pages by centralizing a resourcethey can all link to. Such a shift allows individual sites to focus ontheir areas of expertise.

Ultimately, EQNet may not only be a reference tool for EqIP mem-bers (and others), but also a means of promoting and attractingmore visitors to all EqIP members’ Web sites.

EQNet’s webmaster, Ed Hensley, dismantled the original staticpages, redesigned the structure of the site, and remounted it as anonline database. At the same time, he simplified the interface,making access to the information in the database easier.


A Funny Thing Happened onthe Way to the Fieldwork

Beneath the Science

For this issue of the SCECQuarterly Newsletter, we re-quested anecdotal or humorousstories of summer field re-search—the memorableexperiences aelong the way tothe final scientific result. Hereare a few of the responses.

Shirley Baher, UCLAWhile in Hawaii when I wastaking ESS135 A, B, C, we hada field trip to Kilauea to per-form magnetic and seismicexperiments. While there, weall had to perform an extraproject and write a specialsection about it in our finalreport. I had chosen to provethe Curie temperature. This isthe temperature at whichmagnetization of material goesto zero. To perform the experi-ment, I had a magnetometerand heat probe. I would insertthe heat probe into recentlyformed lava, Paul

Davis would read the tempera-ture, and I would record themagnetometer readings everyminute. The two measurementswould be combined, a graphwould be drawn, and thetheory could be proven (or notproven).

The only problem with thesetup was that as the lavacooled it cemented a $200 heatprobe along with it. To compli-cate matters, there was a steady

stream of lava heading ourdirection, and it was setting fireto the surrounding brush. So Iwas tugging at the heat probewith all my might, surroundedby a small brush fire and anapproaching 3-ft-wide flow oflava. I finally gave up the fightand watched as the heat probewas buried even further bylava.

Sally McGill,CSU San BernardinoI have had many interestingfield assistants in the course ofmy paleoseismic fieldwork insouthern California. I would

like to pay tribute to them notonly for their dedication togetting the work done but alsofor their cheerful spirits whichmade the long months in thefield more enjoyable. AlthoughI mention only two of myassistants here, I have fondmemories of my time with all ofthem. Dawn Grant, who as-sisted with trench logging onthe Garlock and San Andreasfaults, drew beautiful, creativedoodles in the margins of thelogs, including a depiction ofhell, located beneath the base

of our trench, and upside-down people in Chinabeneath that. Her doodles alsocommemorated the scatteredinteresting events that broke upthe monotony of trench log-ging, such as the discovery of ascorpion in the trench one dayand a sidewinder under thetrailer another day.

Heidi Anderson assisted mewith trench logging on theGarlock fault while she was anundergraduate geology major atCaltech. After graduating, shetook a break from geology tostudy fashion design in Paris.This was an outgrowth of hersummer undergraduate re-search project on the history ofhoop skirts. As she learnedmore about the history offashion she wanted to illustratethe ages of the strata in mytrenches with sketches of thefashions that were popular atthose times. Eventually shereturned to assist me on an-other Garlock trench, but thistime she brought a portable,hand-operated sewing machinewith her, and she spent herevenings in the trailer sewing atrench-suit for herself. Heidi isnow a graduate student ingeology at UC Berkeley.

Shawn Larsen, LawrenceLivermore National LabI was on Rob Reilinger’s 1989GPS campaign in eastern Tur-

key. My guide and translatorwas Lieutenant Alp, an armyofficer in the Turkish military.We also had a driver. One sitewas located near a smallKurdish village. We stopped atthe local military command,where Lieutenant Alp discussedsomething with the local com-mander. He later told me thatthey were concerned aboutpossible Kurdish rebel activity.We headed off to the GPS siteat 3 o’clock the next morning.It was pitch black. Suddenly,our driver came to a completestop as we encountered fourKurdish men standing in theroad carrying automatic weap-ons. Worse, they opened up thedoor and piled into our car—acar built for four people. Ohgreat, I’m never going to seemy cat again. Lieutenant Alp,who was purposely wearingcivilian clothes to avoid beingshot, started talking to them inTurkish. Meanwhile, I wassilently praying in English. It’snight and I’m on a desertedroad in the middle of a foreigncountry. I’m crammed into asmall car with six other people,four of them heavily armed. Aswe were driving along the road,I kept using my finger to pushthe barrel of a machine gunaway from my head. Thinkingback, it’s fortunate that themovie “Pulp Fiction” hadn’t

We had a Mongolianguide who was usuallydrunk and therefore notvery useful.

By Mark Benthien


come out before this time. Afterwe got to the site, LieutenantAlp informed me that the armygave these men guns and askedthem to protect us from pos-sible rebel activity. It turns outthat these four Kurdish menwere perhaps the friendliestpeople I have ever met.

I was on Rob Reilinger’s 1990GPS campaign in the ImperialValley. I usually traveled solo tomy assigned sites, but this timeKenji Satake was with me tolearn something about GPS.The station for the night wasKANE, located near the junc-tion of the Elmore Ranch andSuperstition Hills faults. It’s inthe middle of nowhere, onlyaccessible with a 4-wheel drivevehicle along a dilapidated dirtroad. We headed west towardsthe site late in the day. It wasextremely difficult to see, sincewe were blinded by the glarefrom the setting sun. As wewere bouncing back and forthtrying to maintain contact withthe road, we could make outwhat appeared to be a truckand people standing around acampsite 100 yards in front ofus. No problem. It’s alwaysnice to meet strangers in themiddle of nowhere. We movedto within 50 yards. We werestill blinded by the sun butcould definitely distinguish theoutline of two people. Still,something didn’t quite lookright. At about 25 yards Islowed and started to roll downmy window. I wanted to sayhello and tell these peoplewhat we were doing. But ... ohmy gosh. There were a manand a woman standing therewithout any clothes on. Need-less to say, I decided not tostop, and quickly rolled thewindow back up. As we

passed, the woman had thedecency to head for cover. Onthe other hand, the man stoodthere scratching his rear likeHomer Simpson. He lookedlike him, too. It was not a prettysight.

Mark Benthien, USC:During Paul Davis’ Lake BaikalTeleseismic experiment, I wasstationed in Mongolia alongwith a team of Russian col-leagues. Our seismographstations were about 50 kmapart, in areas where all dirtroads and hills look identicaland finding the stations was achallenge. We had a Mongo-lian guide who was usuallydrunk and therefore not veryuseful, and I was a 20-year-oldkid from California—so ourcredibility was about equal.One night we were late inarriving at the next station andhad no landmarks to guide us.Our only help was a handheldGPS receiver that I had pro-grammed with the coordinatesof the station. The instrumentlisted the direction we wereheading and the direction anddistance to the station. As weapproached the station, Ipointed a flashlight ahead ofthe car so the driver (who didnot speak English) could drive“cross country” in the rightdirection by following the spotof light on the ground ahead.This was in 1992, so GPS wasnew to us, and no one else inthe car trusted that I would findthe site. As we got close, weswerved back and forth as Iadjusted our course until thedriver slammed on the brakesto stop less than 2 feet from thestation. I guided the team forthe rest of the summer.

Off-Scaleauthors who are not earth scientists but wish they were

“One of the Three Most InterestingSpectacles I Have Beheld”

After five years aboard the H.M.S. Beagle, Charles Darwin

found himself becalmed off Chile in March of 1835 with time to

write this letter to his sister.

My dear Caroline,

We now are becalmed some leagues off Valparaiso and instead of

growling any longer at our ill fortune, I begin this letter to you. . . .

The voyage has been grievously too long; we shall hardly know

each other again; independent of these consequences, I continue to

suffer so much from sea-sickness, that nothing, not even geology

itself, can make up for the misery and vexation of spirit.

The papers will have told you about the great Earthquake of the

20th of February. I suppose it certainly is the worst ever experi-

enced in Chile. It is no use attempting to describe the ruins—it is the

most awful spectacle I ever beheld. The town of Concepcion is now

nothing more than piles and lines of bricks, tiles and timbers—it is

absolutely true there is not one house left habitable; some little

hovels built of sticks and reeds in the outskirts of the town have not

been shaken down and these now are hired by the richest people.

The force of the shock must have been immense, the ground is

traversed by rents, the solid rocks are shivered, solid buttresses 6-

10 feet thick are broken into fragments like so much biscuit.

How fortunate it happened at the time of day when many are out of

their houses and all active: if the town had been overthrown in the

night, very few would have escaped to tell the tale. We were at

Valdivia at the time. The shock there was considered very violent,

but did no damage owing to the houses being built of wood. I am

very glad we happened to call at Concepcion so shortly after-

wards: it is one of the three most interesting spectacles I have

beheld since leaving England—A Fuegian Savage—Tropical

Vegetation—and the ruins of Concepcion. It is indeed most wonder-

ful to witness such desolation produced in three minutes of time.

—Charles Darwin, 1835


About 50 SCEC-affiliated re-searchers and students led byUSGS Pasadena Scientist-in-Charge Lucy Jones met in lateJune to organize a new planaround the scientific objectivesof a post-earthquake investiga-tion and to discuss how thoseobjectives could be achieved.

Following a brainstormingsession, the participants brokeinto three groups—geology,

geodetics, and seismology—todiscuss the questions raised inthe brainstorming session andformulate post-earthquake fieldexperiments that would addresseach of those questions. At theend of the day, each grouppresented its conclusions, andJones facilitated a lively sessionthat focused on implementationmethods.

The following notes outlinemany of the ideas broughtforward in each of the breakoutsessions. The results of theworkshop will be synthesized

into an official SCEC responseplan. The group agreed itshould meet again at the SCECannual meeting to discuss theresulting plan and logistics.

Geology Group Summary(Tom Rockwell)Following a large earthquake,geologists need to do the fol-lowing:

In the first hours:• Resolve scope of surface

rupture. They would need animmediate response teamthat is ready to go, withnecessary equipment,

clearance, and transporta-tion to resolve extent ofsurface rupture. This requiresaccess to a helicopter, withdirect communication backto a data collection center.Aerial photos from anysource would be extremelyuseful. This also requires

real-time epicentral informa-tion, including size, focalmechanism, etc.

• Install quadrilaterals/alignment arrays/afterslipstudies, GPS, point align-ments.

Planning SCEC’s Role in Clearinghouse

SCEC Scientists Conduct First Post-EarthquakeResponse Planning SeminarBy Jill Andrews and Mark Benthien

CollaboratingWhen It Counts

The California Post EarthquakeInformation Clearinghouse

For two weeks after the January 17,1994, Northridge earthquake, theCalifornia Office of EmergencyServices (OES) Pasadena officeserved as the nerve center of anextensive reconnaissance effort.Every morning, earth scientists, en-gineers, social scientists, andpublic policy experts headed outto the damaged areas. After a dayof looking at and studying theearthquake’s effects, these expertsreturned to a crowded conferenceroom in Pasadena to spend the

evening sharing and reviewingtheir observations.

That was the first extensive opera-tion of the California Post EarthquakeInformation Clearinghouse, aproject of many organizations in-volved in earthquake studies. Theclearinghouse is a unique way inwhich information can be ex-changed among various types ofinvestigators who come from otherstates and counties.

Within one day of the Northridgeearthquake, clearinghouse field in-vestigators were on site, bringingback damage reports that aidedemergency response activities andfocused earth science investiga-tions. In a few days, the dailyinformation became an impressivebody of data for future analysis anduse.

Clearinghouses had been orga-nized following other Californiaquakes. The California Division of


In the first day:• Aerial photography of the

rupture as soon as therupture zone is established,so more detailed work ofdocumentation, includingdetailed mapping and slipmeasurements, can begin.

• Slip measurements. There isa need to establish acommon standard of qualityto be followed by all

participants; to perhapsestablish a field mappingsystem complete with GPSreceivers for more preciselocations; to compile daily asummary of all field data at acommon data collectioncenter or clearinghouse; andto publish the data as soonas possible.

• Detailed mapping; includefocus on structure complexi-ties

There is also a need for a struc-tured response with commonstandards. This should be acollaborative effort between the

USGS, CDMG, SCEC geolo-gists, and others.

Geodesy Group Summary(Ken Hudnut)Geodesists would like to havecontinuous instrumentation outin the field, which wouldeliminate the need to rush outto the field after the event.SCIGN has not built this capa-bility into its program.Geodesists need GPS arraysacross the fault to measuredisplacement.

Topics of interest for suchinvestigations:• Coseismic

• Stress triggering (observablerate changes adjacent toother faults—co-seismicmotion on neighboringfaults, for example.)

• Campaign deployments

• Instrumenting buildings

• Long-period, long-wave-length deformations

• Various models to explain

Geodesists would rely onSCEC/USGS to coordinateefforts. A possible structure:• SCEC: Crustal Deformation

Working Group (Agnew:logistics)

• SCIGN: (Hudnut: chairscientific integration, sourcemodeling, feedback toAgnew’s group)

• USGS Menlo Park: Crustaldeformation mega project(Thatcher)

continued on next page

Mines and Geology (CDMG) andthe USGS worked with OES andthe Earthquake Engineering Re-search Institute (EERI) to run themodest operations out of localschool gymnasiums, firehouses,community colleges, and evenmotels.

The value of sharing informationafter these events was obvious, butthey also realized that the clearing-house was an efficient way toinvolve and track the large num-ber of investigators who came tothe damage scene.

A few months before theNorthridge earthquake, the four

leading organizations were joinedby the California Seismic SafetyCommission (CSSC). They reachedan informal agreement to establishand operate a large clearinghouseafter the next major earthquake.That agreement came just in time.

Hundreds of investigators passedthrough the doors of the Northridgeclearinghouse during its two weeksof operation. Although it as an un-qualified success, it also showedthe need for formal plans and theinvolvement of more organiza-tions.

The Plan

Clearinghouse collaborators havebeen meeting quarterly since

March 1996 to plan for the needsof all the involved organizations.

In summary, the plan calls for theclearinghouse to: 1) be the check-in and check-out point for allinvestigators and officials who ar-rive at the scene; 2) collect andverify perishable reconnaissanceinformation; 3) convey that infor-mation to the planning/intelligencefunction of the OES Regional Emer-gency Operations Center; 4)provide updated damage informa-tion through daily briefings andreports; 5) track investigators in thefield; and 6) perhaps even directtheir movements for maximum

continued on next page


Adapted from an articlepublished in CaliforniaGeology, March/April 1999,by Sarah K. Nathe.

Seismology Group Summary(Ralph Archuleta)The group discussed whatexperiments could be deployedto answer the scientific ques-tions. Timing, priority,organization, and personnelwere taken into account.Archuleta constructed an orga-nizational chart that indicatedcoordination of data analysisand modeling, communication,and field investigations, as wellas considering real-time inven-tory of instruments.

Scientific questions consideredincluded:• Site effects

• Basin effects, basin edge,3D structure

• Geology + correlation tosite response

• Non-linearity (timeurgency)

• Coherency of near-field,scattering array aroundborehole stations

• Ground motion andprecarious rocks

• Deep structure

• Source effects—earthquakephysics (time urgency)

• Near field (source)recordings [time urgent]

• Source properties(temporal changes)

• Rupture dynamics

• Rupture nucleation?(good rock sites)

CLEARINGHOUSE continued from previous page

• Rupture stopping?

• Subsurface structure

• Fault zone trapped waves(fault complexity)

• Where is the fault?

• Stress triggering(modeling)

• Crustal structure? Refraction

Other issues discussed by theentire group:• The need for cell phones

and whether they aredependable in a post-earthquake setting

• Walkie-talkies

• The need to construct a Webpage that could be used as acommunications vehicle andthat is password protected.

• The need for ID cards, lettersof permission to enteraffected areas

• The need to interface withCalifornia’s OES and itstechnical clearinghouse

• The need to invite SCECresearchers to futuretechnical clearinghousemeetings

• The need to recruit volun-teers to continue the scienceplan effort

• The need to conductexercises based on scenarioearthquakes

• The need to identify fundsfor a post-earthquakescenario

coverage with minimal disruptionto residents.

In the few years since Northridge,the technology for capturing datahas been vastly improved, andwhat was a comparatively primi-tive system in 1994 is now anetworked geographic informationsystem capable of tracking the in-vestigators as well as their findings.

Within the clearinghouse manage-ment group, CDMG and USGS areresponsible for conductingseismologic and geologic assess-ments of earthquakes. EERI has acharter from the National ScienceFoundation to investigate the struc-tural and social effects of all majorearthquakes in the U.S. andabroad. The CSSC is the main seis-mic policy body in the state,recommending new legislation andregulations to minimize earth-quake losses. OES coordinates all

the emergency planning and re-sponse activities in the state.

Besides the members of the man-agement group, ten otherorganizations have signed on asparticipants in the clearinghouseplan:

Applied Technology Council

California Institute ofTechnology

California Universities forResearch in Earthquake Engi-neering

Federal Emergency Manage-ment Agency, Region IX

National Aeronautics andSpace Administration

Pacific Earthquake EngineeringResearch Center

Southern California EarthquakeCenter

Structural Engineers Associa-tion of California

Technical Council on LifelineEarthquake Engineering

UC Berkeley SeismologicalLaboratory

Triggering a Clearinghouse

An earthquake in an urban areawill trigger establishing a clearing-house when the quake is damagingand has a magnitude of 6.0 orabove. A clearinghouse may beestablished under other conditionsif recommended by CDMG staffduring initial field surveys follow-ing an earthquake.

A federal disaster declaration is notnecessary to activate the clearing-house, but the clearinghouse will

always be activated when there isa federal disaster declaration.

In the first 24 hours after a seriousquake, the OES region in which theearthquake strikes will provide, orwork with other governmentalunits to arrange for, the clearing-house space.

The duration of the clearinghouseoperation depends on the extent ofthe damage and the length of theresponse period. While there is stilla need for information to supportresponse activities, or perishabledata to be gathered, the field in-vestigators will survey the damagedarea.

COLLABORATING continued from previous page


Uniting Science and Education

Teachers Sacrifice YetAnother WeekendBy Jill Andrews

I used to think I wanted to be aK-12 teacher. I thought it wouldbe an easy career—teachingstraight out of texts to well-behaved students who aremotivated and eager to learn,with lots of time to rest, read,and travel during the summer. Iheld that idyllic but erroneousview for years until I beganspending a weekend eachquarter with a group of ridicu-lously busy, hard-workingteachers and teacher trainerswho are advising our Web

authors as they develop SCECand SCIGN online educationalmodules.

As I sat down to write a reporton our second SCEC Outreach/DESC-Online Advisory TeamWorkshop, held in June, Istarted thinking about what itmeans to give up a weekend.Much of a teacher’s eveningsand weekends are already filledwith planning, grading, train-ing, or researching. Why give up four more weekends to help

with someone else’s project?

Their answer: it’s for a goodcause. Teachers are greatly inneed of well-constructed, user-friendly earth-science curriculaaligned with the national stan-dards. Our DESC-Onlineproject offers an opportunity forteachers to collaborate withscientists to produce a first-rateWeb-based teaching and learn-ing tool for middle school andhigh school students.

Over the course of those twodays in June, we immersedourselves in learning what ittakes to create and flesh out astoryline for a middle schoolversion of our two existingmodules—“Investigating Earth-quakes through Regional Seis-micity” (WWW.SCECDC.SCEC.ORG/MODULE) and “Exploring the Useof Space Technology in Earth-

quake Studies” (SCIGN.JPL.NASA.GOV/LEARN). Here’s how we did it.

Midnight, Friday, June 4: All thepreparations for the workshopwere complete. SCEC staffmembers and Web authors allarrived hours earlier at themotel near UC Santa Barbara,but so far none of the teacherswere there, and I was worried.Did they have the right date?Since it was so late, there wasnothing to do but go to bed andhope they arrived by morning.

Had I known these peoplebetter, I wouldn’t have worried.Having spent all day in a train-ing workshop in OrangeCounty, they started out about10 p.m. and drove from Irvineto Santa Barbara. AlthoughMeridith Osterfeld, PhilLaFontaine, Cindy Anderson,and Don Whisman got only afew hours’ sleep, they were up

DESC-Online workshop participants take a break from the long hours ofcurriculum development: (l to r) Phil LaFontaine, Meridith Osterfeld, JillAndrews, Katrin Hafner, Cindy Anderson.

DESC-Online Background

In early 1998 SCEC launched a formal review and testing ofits educational materials under development, with a commit-ment to bringing the Web-based products online as quickly aspossible. The scientific review, now complete, focused on sci-entific accuracy and pedagogical effectiveness. At the sametime, we assembled a group of educators to work with ourWeb authors to test the existing modules and help us create anew module for middle and high school students. The group,called the Development of Earth Science Curricula (DESC)Online Advisory Group, has met four times to date (with an-other three weekends scheduled through the remainder ofSCEC’s fiscal year) and has completed a design and storylinefor the new module. Between workshops, group members workand communicate via the Internet to further develop portions ofthe modules to present at subsequent workshops. The Web au-thors use input from the advisors and add graphics, illustrations,animated text, etc. to make the material dynamic and interest-ing for both teachers and students.


Meridith Osterfeld—Regional Director, K-12Alliance

Phil LaFontaine—Regional Director, K-12Alliance

Cindy Anderson—Regional Director, K-12Alliance & 6th gradeteacher at Vista UnifiedSchool District

Don Whisman—Staff Developer, K-12Alliance & 8th gradeteacher at Tierra Del SolMiddle School

Ursula Sexton—the first elementaryteacher to win the Presi-dential Award

early and ready for work onSaturday. Ursula Sexton, theother member of the group,arrived later that morning.

The ABCs of the 4As9 a.m., Saturday, June 5: Afterintroductions, Phil gave thegroup a clear explanation of theall-important storyline, thebasis for constructing a usablecurriculum. The standards, heexplained, include what shouldbe taught but not how to teachit. Their “4A” model was devel-oped, he explained, to aidteachers in constructing astoryline that addresses the“how” and the “what.”

In the 4-A model, the first “A”stands for “Alignment”—identi-fying what teachers think

students should know, called“prethinking” the learningprocess. Using earthquakes asan example, we came up withthe following ideas:

Earthquakes happen in differ-ent areas for different reasons.

Earthquakes have conse-quences, but with mitigation,we can change these.

Earthquakes reshape the earth.

Earthquakes are the result ofwhat the earth is doing—constantly changing. (Thisconcept is known as a “BigIdea.”)

Earthquakes are measurable.

This exercise leads to state-ments of what we thinkstudents should know.

The second “A”—“Augmenta-tion”: This step asks theteacher to decide what makessense to students. What toolsdo students need, what“hooks” should exist in apackage that will make themremember what they learn.Research shows that moststudents retain information ifit is conveyed in the format ofa story—especially if it’s astory that is relevant to thestudent’s life.

During the process of aug-mentation, a teacher developsa storyline, starting with the“Big Idea” and movingthrough general concepts,appropriate levels of subcon-cepts. Each component leadsthe student to the next. If thestory makes sense and is amemorable one, it is easilyrecalled, along with the facts

related to the topic. A teacher’sgoal is to construct a storylinethat students can learn from,that is rigorous, and that meetsthe requirements of the stan-dards.

The third “A” stands for “Acces-sibility.” All students inCalifornia, for example, shouldhave access to knowledge. Thisrequirement gives rise to “Deci-sion Point Assessment”—i.e., ifsome students don’t “get it,” theteacher has to figure out whatneeds to be done to help thosestudents learn.

The fourth “A” is “Assessment,”which is conducted before,during, and after lessons. Alesson that covers waves, forexample, should include teach-ers’ background information onthe topic—called “front load-ing” the concepts—so teacherscan anticipate students’ ques-tions. Assessment during andafter is equally important toensure that students are under-standing and retaining theknowledge.

A Portion of theMiddle School Module

Unifying Concept: Earth processes create changes that areobservable and measurable over time.

Module Concept: The Earth changes through a process called“plate tectonics.”

Subconcept #1: Observations provide evidence of changesin the Earth.

• Observations are made using a variety of tools.• Observations of local land forms and/or local structures/

features and formation of a hypothesis of how it came to be.• A compilation of detailed, local observations leads to a

larger/regional/global picture.• The present is a key to the past. Patterns of observations

have helped to build a unifying theory called “plate tecton-ics.”

Subconcept #2: Plate tectonics explains important featuresof the Earth’s surface and major geologic events.

• The Earth is layered (clues on the surface provide clues towhat’s below).

• Lithospheric plates move at the rate of centimeters per yearin response to movement in the mantle.

• Major geologic events result from these plate interactions andmotions.

Subconcept #3: Major geologic events impact society.


August 19995-8 SCEC Intern Colloquium

9-12 Ninth InternationalConference on Soil Dynamics andEarthquake Engineering (SDEE ’99).Sponsors: Institute of Solid EarthPhysics, University of Bergen,Norwegian Society for EarthquakeEngineering. Bergen, Norway.Contact: K. Atakan, email:[email protected]; WWW.IFJF.UIB.NO/SEISMO/SDEE99.HTML.

15-30 Field trip to western Turkey:Extensional tectonics, modern andhistorical earthquake surface breaksand archaeoseismology. 90 (212)285 6299; 90 (212) 285 6210,fax; [email protected] [email protected]

31–Sept 2 Autonomous Data-Gathering Systems in ExtremeEnvironments, Jet PropulsionLaboratory, Pasadena, CA. Abstractand reg. deadline: July 15. Contact:Andrea Donnellan, JPL, 818-354-4737, [email protected];HTTP://GEODYNAMICS.JPL.NASA.GOV/ANTARCTICA.

September 19996-9 Western States SeismicPolicy Council 21st AnnualConference. Santa Fe, NewMexico. Contact: WSSPC, (415)974-6435; fax: (415) 974-1747;email: [email protected].

7-9 Stress Triggering andDeformation Software TrainingWorkshop. For information, seeWWW.SCEC.ORG/NEWS/99NEWS/STRESS.HTML.

20 USC—Davidson Confer-ence Center: Science Seminar:HAZUS Demonstration andDiscussion; Active faulting andearthquake hazards in the LosAngeles metropolitan area.

27-29 SCEC Annual Meeting,Palm Springs, CA. Contact SCEC,213/740-5843 or seeWWW.SCEC.ORG.

26-30 California EmergencyServices Association AnnualConference and Training: “Defining21st Century Emergency Manage-ment, Managing Reality vs.Perceptions in a Media World.”Palm Springs, CA. Contact: WendyMilligan, (805) 644-0899;fax: (850) 642-2883; email:[email protected].

October 19991-Nov 15 Los Angeles—LARSE IIExperiment. Interested parties andvolunteers needed. Contact MarkBenthien, SCEC Outreach.

3-5 Snowbird, Utah—PlateBoundary Observatory (PBO)Workshop: to apply, see HTTP://WWW.SCEC.ORG/NEWS/99NEWS/PBO.HTML.

13 California Post-EarthquakeInformation Clearinghouse Meeting,Oakland, CA. For more informa-tion, contact T. Toppozada, CDMGSacramento.

27-28 Sixth Annual Congress onNatural Hazard Loss Prevention.Sponsor: Institute for Business andHome Safety. Memphis, TN.Contact: (617) 292-2003; fax:(617) 292-2022; email:[email protected]; WWW.IBHS.ORG.

27-29 Second Meeting onSeismology and Seismic Engineer-ing of MediterraneanCountries—Sismica99. Faro,Portugal. Tel/fax: +351 (0)89803561 (ext. 6545); fax: +351(0)89 823539; email:[email protected]; WWW.UALG.PT/EST/ADEC/SISMICA99/SISMICA99GB/INDEX.HTM.

November 199918 USC—Science Seminar.Topic: To be announced

December 199913-17 Fall American GeophysicalUnion meeting, San Francisco, CA.See HTTP://EARTH.AGU.ORG/MEETINGS/SM99TOP.HTML.

16 USC—Science Seminar.Topic: To be announced

January 200020 USC—Science Seminar.Topic: To be announced

30-Feb 4 12th World Conference onEarthquake Engineering. Auckland,New Zealand, P.O. Box 2009,Auckland, New Zealand;tel: 64-9-529 4414;fax: 64-9-520 0718;email: [email protected];WWW.CMSL.CO.NZ/12WCEE;also see WWW.EERI.ORG/MEETINGS/12WCEE.HTML.

Phone 213/740-1560 for SCEC OutreachPhone 213/740-5843 for General InformationFax 213/740-0011 Email: [email protected]

CalendarGetting Down to Sixth-GradeLevelNoon: Following the 4A over-view, Meridith and Cindy ledthe group through a “Star Lab”activity to demonstrate howsome middle school studentslearn about tectonics and theEarth’s interior. This meant thatwe first had to imagine our-selves as sixth graders (not hardfor some of us), crawl inside thegiant air-filled sphere they hadbrought with them, and view adepiction of the Earth’s tectonicplates with the aid of a laserdisc that projected an image onthe inner skin of the sphere.

After viewing the plates inmotion, we moved to a labora-tory where we performed anexperiment that illustrated theprocesses at work in the Earth’sinterior. As “students,” we wereasked to explain what wethought was occurring in ourliquid-filled beakers as wedropped dye into the liquid andmoved a burner underneath toheat the liquid. Most of usfigured out that we were learn-ing about the process involvedwith the Earth’s core heatingthe magma, causing convectionand thus tectonic activity. Wewere told that when studentsare in a lab setting, learning the“hands-on” way as we weredoing, they become completelyengrossed in the process andbehavior problems are nearlynonexistent.

After our experiment, weviewed a video that used ani-mated graphics and filmfootage to help the teachersexplain what we had just seen,touched, and learned. By thetime we were finished (thewhole process took about twohours), we were ready to think

on the level of a sixth grader—a most effective way to createa science-based storyline.

Putting the Pieces Together2:00 p.m.: Facilitated byMeridith, the group began thetime-consuming but rewardingprocess of creating an earthscience storyline, Big Idea,concepts, and subconcepts fora new middle school module.The product will be an 8- to-12-week segment of 3250-minute lessons.

Finishing the JobSunday, June 6: After dinnerand a late-evening Saturdayfilled with talk about our work,homes and hobbies, we weretired when Sunday morningrolled around, but were readyto finish what we’d begun. Ourgoal was to provide the Webauthors with enough informa-tion to start the process ofreworking our community-college level modules into aproduct appropriate for middleschool. Jim Russell, vice presi-dent for education andoutreach for the Institute ofBusiness and Home Safety wasa welcome observer who gaveus the societal impact point ofview. The group continued toflesh out the outline and fin-ished in time for lunch togetherbefore parting company.

Driving home late that after-noon, I was pleased with whatwe had accomplished andgrateful to those already over-committed teachers forspending yet another weekendimproving the nation’s scienceeducation curricula. All of us—scientists, parents, and studentsalike—can be thankful for thatkind of dedication.


Publications

277. McGill, S. F. and C. M. Rubin, Surficial Slip Distribution on the CentralEmerson Fault During the 28 June 1992 Landers Earthquake, Journal ofGeophysical Research, 104, no. B3, pp. 4811-4833, 1998.

384. Madariaga, R., K. Olsen, and R. Archuleta, Modeling dynamic rupturein a 3D earthquake fault model, Bulletin of the Seismological Society ofAmerica, 88, pp. 1182-1197, 1998.

391. Day, S. M., Efficient Simulation of Constant Q using Coarse-GrainedMemory Variables, Bulletin of the Seismological Society of America,88,1051-1062,1997.

396. Jackson, D. D. and Y. Y. Kagan, VAN method lacks validity, EOS,Transactions of the American Geophysical Union, 79, no. 47, pp. 573 and579, 1998.

Varotsos and colleagues (the VAN group) claim to have successfullypredicted many earthquakes in Greece. Several authors have refuted theseclaims, as reported in the May 27, 1996, special issue of GeophysicalResearch Letters and a recent book, A Critical Review of VAN [edited byLighthill, 1996]. Nevertheless, the myth persists. In the letter we summarizewhy the VAN group’s claims lack validity.

407. Bazzurro, P., and C. A. Cornell, Disaggregation of Seismic Hazard,Bulletin of the Seismological Society of America, 89, no. 2, pp. 501-520,1999.

408. Hearn, E. H., and E. D. Humphreys, Kinematics of the Southern WalkerLane Belt and Motion of the Sierra Nevada, California, Journal of Geo-physical Research, 103, pp. 27033-27049, 1998.

Deformation in the southern Walker Lane Belt region of the southwesternGreat Basin accommodates significant portions of both Pacific-NorthAmerica transform motion and Basin and Range extension. However,apparent kinematic inconsistencies between geodetic and fault slip data inthis region make it difficult to understand the nature of the interactionbetween these processes. Kinematic modeling of the region in applied tothis region in a manner that enforces kinematic consistency and simulta-neously includes fault geometry and slip rate data, GPS survey data, andVLBI/VLBA site velocity data. A model is found that is consistent with theset of available data, and this model differs significantly from prior modelsfor the region. Our model has the Sierra Nevada block, which bounds theGreat Basin to the west, moving N49W +-1 at a rate of 12.7 +-0.5 mm/yr,with only minor amounts of counterclockwise rotation. Garlock fault slip isa consequence of the relatively great westerly motion of the Sierra Nevada,which lies immediately north of the Garlock Fault. We also find that thesouthern Great Basin adjacent to our study area moves westward at a rateof about 2.5 mm/yr, which is slower than the velocity in the central GreatBasin.

412. Ward, S. N., On the Consistency of Earthquake Moment Rates,Geological Fault Data, and Space Geodetic Strain: The United States,Geophysical Journal International, 134, pp. 172-186, 1998.

New and dense space geodetic data can now map strain rates overcontinental-wide areas with a useful degree of precision. Stable strainindicators open the door for space geodesy to join with geology andseismology in formulating improved estimates of global earthquakerecurrence. In this paper, 174 GPS/VLBI velocities map United States’ strainrates of <0.03 to >30.0 x10-8/y with regional uncertainties of 5 to 50%.Kostrov’s formula translates these strain values into regional geodeticmoment rates. Two other moment rates, M_seismic and M_geologicextracted from historical earthquake and geological fault catalogs, contrastthe geodetic rate. Because M_geologic, M_seismic and M_geodetic derivefrom different views of the earthquake engine, each illuminates differentfeatures. In California, ratios of M_geodetic to M_geologic is 1.20. The nearconsistency points to the completeness of the region’s geological fault dataand to the reliability of geodetic measurements there. In the Basin andRange, Northwest and Central United States, both geodetic and seismicgreatly exceed geologic. Of possible causes, high incidences of understatedand unrecognized faults most likely drive the inconsistency. The ratio ofM_seismic to M_geodetic is everywhere less than one. The ratio runssystematically from 70-80% in the fastest straining regions to 2% in theslowest. Although aseismic deformation may contribute to this shortfall, Iargue that the existing seismic catalogs fail to reflect the long-term situation.Impelled by the systematic variation of seismic to geodetic moment ratesand by the uniform strain drop observed in all earthquakes regardless ofmagnitude, I propose that the completeness of any seismic catalog hingeson the product of observation duration and regional strain rate. Slowlystraining regions require a proportionally longer period of observation.Characterized by this product, gamma distributions model statisticalproperties of catalog completeness as proxied by the ratio of observedseismic moment to geodetic moment. I find that adequate levels ofcompleteness should exist in median catalogs of 200 to 300 year durationin regions training 10-7/y (comparable to southern California). Similarlevels of completeness will take more than 20,000 years of earthquake datain regions straining 10-9/y (comparable to southeastern United States).Predictions from this completeness statistic closely mimic the observedM_seismic to M_geodetic ratios and allow quantitative responses topreviously unanswerable questions such as: “What is the likelihood that theseismic moment extracted from a earthquake catalog of X years falls withinY% of the true long term rate?” The combination of historical seismicity,fault geology and space geodesy offers a powerful tripartite attack onearthquake hazard. Few obstacles block similar analyses in any region ofthe world.

413. Trecker, M. A., L. D. Gurrola and E. A. Keller, Oxygen IsotopeCorrelation of Marine Terraces and Uplift of the Mesa Hills, Santa Barbara,California, U.S.A., Late Quaternary Coastal Tectonics (probably) Spec. Publ.Geol. Soc. London, 146, pp. 57-59, 1998.

Resolving marine terrace chronologies is critical for determining uplift ratesalong tectonically active coastlines. Unfortunately, lack of suitable datingmaterials often makes it difficult to establish the chronology. We presenthere oxygen isotopic data from 21 shells of Olivella biplicata from fourmarine terraces in the Santa Barbara and Ventura area located in southernCalifornia, U.S.A. Uranium series analysis of two fossil corals Balanophylliaelegans from two Santa Barbara area marine terraces yielded ages of 47 ka+/- 500 yrs and 70 +/- 2 ka (oxygen isotope stage 3a and 5a, respectively)Gurrola et al. 1996 and 1997). Olivella shells from these dated calibrationpoints yield average values of 1.117" and 0.627" respectively. Mollusksfrom undated terraces hypothesized to be of the same age yielded averagevalues of 1.010" and 0.751" respectively. The data indicate that stableoxygen isotopic signatures preserved in marine terrace mollusks canprovide a useful tool for correlating undated terraces to those of knownage, as well as for correlating terraces across faults and folds. Using isotopicdata coupled with a U-series dated wave-cut platform we calculate a rate

The following is a list of recent publications based on SCEC-funded research.SCEC authors must obtain SCEC contribution numbers for all papers to acknowl-edge SCEC funding. In return, their papers are added to the SCEC PublicationDatabase. This database is reported to the NSF during each SCEC evaluation. Toreceive a SCEC contribution number, complete the online form at WWW.SCEC.ORG/RESEARCH/SCECNUMBER.HTML, which requires authors, title, publication name, abstract(very important), and any other bibliographic information available. The SCECnumber will be returned via email along with the proper NSF/USGS/SCECacknowledgement statement.

The SCEC Quarterly Newsletter now publishes the references only for publishedarticles, no longer listing ones that are submitted, in review, in press, etc. Thecomplete list (both searchable and sortable) is available at WWW.SCEC.ORG/RESEARCH

PAPERS.HTML and will no longer be printed in the newsletter in its entirety each year.A hardcopy version of the list can be obtained by calling 213-740-5843 oremailing [email protected].


of uplift ranging from 0.62 +/- 0.03 mm/yr (where the elevation of the firstemergent terrace is 41 m) to 0.54 +/- 0.05 mm/yr (where the first emergentterrace is at 36 m) for marine terrace flights preserved on the Mesa hillsanticline located in the city of Santa Barbara, California.

414. Vidale, J. E., D. C. Agnew, M. J. S. Johnston and D. H. Oppenheimer,Absence of earthquake correlation with Earth tides; an indication of highpreseismic fault stress rate, Journal of Geophysical Research, 103, pp.24567-24572, 1998.

417. Bowman, D. D., G. Ouillon, C. G. Sammis, A. Sornette and D.Sornette, An Observational Test of the Critical Earthquake Concept, Journalof Geophysical Research, 103, no. 24, pp. 359-372, 1998.

We test the concept that seismicity prior to a large earthquake can beunderstood in terms of the statistical physics of a critical phase transition. Inthis model, the cumulative seismic strain release increases as a power-lawtime-to-failure before the final event. Furthermore, the region of correlatedseismicity predicted by this model is much greater than would be predictedfrom simple elasto-dynamic interactions. We present a systematicprocedure to test for the accelerating seismicity predicted by the criticalpoint model and to identify the region approaching criticality, based on acomparison between the observed cumulative energy (Benioff strain)release and the power-law behavior predicted by theory. This method isused to find the critical region before all earthquakes along the San Andreassystem since 1950 with M 6.5. The statistical significance of our results isassessed by performing the same procedure on a large number of randomlygenerated synthetic catalogs. The null hypothesis, that the observedacceleration in all these earthquakes could result from spurious patternsgenerated by our procedure in purely random catalogs, is rejected with99.5% confidence. An empirical relation between the logarithm of thecritical region radius (R) and the magnitude of the final event (M) is found,such that log R µ 0.5 M, suggesting that the largest probable event in agiven region scales with the size of the regional fault network.

419. Harris, R. A. and R. W. Simpson, Suppression of Large Earthquakes byStress Shadows—A Comparison of Coulomb and Rate-and-State Failure,Journal of Geophysical Research, 103, pp. 24439-24451, 1997.

420. Harris, R. A., Stress Triggers, Stress Shadows, and Implications forSeismic Hazard, Journal of Geophysical Research, 103, pp. 24347-24358,1997.

429. Ward, S. N., On the Consistency of Earthquake Moment Release andSpace Geodetic Strain Rates: Europe, Geophysical Journal International,135, pp. 1011-1018, 1998.

In this article, 141 VLBI/SLR/GPS velocities map European strain rates from<0.09x10^-8/y to >7.0x10^-8/y with regional uncertainties of 20 to 50%.Kostrov’s formula translates these strain values into regional geodeticmoment rates M_geodetic. Two other moment rates, M_seismic extractedfrom a 100-year historical catalog, and M_plate taken from plate-tectonicmodels, contrast the geodetic rates. In Mediterranean Europe, the ratios ofM_seismic to M_geodetic stand between 0.56 and 0.68. In Turkey the ratiofalls to 0.18. Although aseismic deformation may contribute to this deficit,the magnitudes of the shortfall coincide with the variations expected in100-year catalogs.

431. Xu, H. S. M. Day, and J. B. Minster, Model for Nonlinear WavePropagation Derived from Rock Hysteresis Measurements, Journal ofGeophysical Research, 103, pp. 29915-29929, 1998.

We develop a method for modeling nonlinear wave propagation in rock atintermediate strain levels, i.e., strain levels great enough that nonlinearitycannot be neglected, but low enough that the rock does not incurmacroscopic damage. The constitutive model is formulated using asingular-kernel endochronic formalism, and this formulation is shown tosatisfy a number of general observational constraints, including producinga power law dependence of attenuation (Q-1) on strain amplitude. Oncethe elastic modulus is determined, and a second parameter fixed to give Q-1 linear in the strain amplitude, the model has 2 remaining free parameters.

One of these represents cubic anharmonicity, and we set in to agree withlaboratory observations of harmonic distortion. The other parametercontrols the amount of hysteresis, and it is set to approximate stress-straincurves measured in laboratory uniaxial stress experiments on Bereasandstone by Boitnott and Haupt. The constitutive equations, thoughfundamentally nonlinear and rate-independent, have a superficial, formalresemblance to viscoelasticity, which we exploit to produce an efficient,stable numerical algorithm. We solve 1D wave propagation problems forthis constitutive model using both finite difference and pseudospectralmethods. These methods are shown to reproduce, to high precision,analytical results for quasi-harmonic wave propagation in a nonlinearlyelastic medium.

Application of the Berea sandstone model to quasi-harmonic wavepropagation shows several departures from results obtained with nonlinearelasticity. The Berea model shows more rapid decay with distance of thefundamental frequency component, due to nonlinear, amplitude-dependent attenuation, than does nonlinear elasticity. The Berea modelalso shows enhanced excitation of the order 3 harmonic, in agreementwith laboratory observations. In addition, the growth with propagationdistance of the harmonics of the source excitation shows a saturation,relative to the nonlinear elasticity results. This behavior reflects thecompeting effects of amplitude growth via energy transfer from the sourcefrequency, and energy dissipation due to hysteresis, the dissipationincreasing as the harmonic amplitude grows. In additional numericalexperiments, we find that a two-frequency source function generatesharmonics with frequencies which can be expressed as linear combina-tions of integer multiples of the two source frequencies, in agreement withpublished laboratory results for other solids.

434. Sato, T. and H. Kanamori, Beginning of earthquakes modeled with theGriffith’s fracture criterion, Bulletin of the Seismological Society ofAmerica, 89, no. 1, pp. 80-93, 1998.

We present a source model for the beginning of earthquakes based on theGriffith’s fracture criterion. The initial state we choose for this model is acritical state of pre-existing circular fault, which is on the verge ofinstability. After the onset of instability, the fault grows with a progressivelyincreasing rupture speed, satisfying the condition of fracture energybalance at the crack tip. We investigate the difference in rupture growthpatterns in two classes of models, which are considered to represent end-member cases. In the first model (Spontaneous Model), we assume that thesurface energy varies smoothly as a function of position in the crust. In thismodel, faults with small initial dimensions grow in the medium with smallsurface energy, and those with large initial dimensions, in large surfaceenergy. The rupture velocity increases progressively until it reaches itslimiting velocity. The synthetic velocity seismogram at far-field shows aweak initial phase during the transitional stage. The time taken to reach thelimiting velocity is proportional to the initial length of pre-existing fault.Therefore the duration of the weak initial phase scales with the initiallength of fault. In the second model (Trigger Model), we envisage that thereare many pre-existing faults in the crust with various length. These faultsare stable because they encounter some obstacle at their ends (e.g. faultsegmentation, strong asperity etc). This situation is modeled with a localincrease in the surface energy near the ends of fault. An earthquake istriggered when the obstacle is suddenly removed (i.e., sudden weakening)or the stress is suddenly increased locally to overcome the obstacle. Oncean earthquake is triggered then the fault growth is governed by the ambientsurface energy. In this model, the rupture speed attains its limiting velocityalmost instantly. The synthetic velocity seismogram at far-field shows anabrupt, linear increase in amplitude without the weak initial phase thatappears in the Spontaneous Model. The Spontaneous Model is character-ized by a small trigger factor and the Trigger Model by a large trigger factor,where the trigger factor is defined as a fractional perturbation of the surfaceenergy at the ends of fault relative to the ambient surface energy. Thus, theseismic initiation phase with and without the slow initial phase can bothoccur depending on the trigger factor. The variability in the observedseismic initiation phase may represent a variation of surface energy(strength) distribution surrounding the pre-existing cracks. No simplemodel can explain the scaling relation between the nucleation momentand the eventual size of earthquake.


ITEM PRICE QUANTITY TOTAL Putting Down Roots in Earthquake Country (10 minimum) $1.00*

Non-profit organization reduced rate (10 minimum) $0.50*Shipping: $3.00 + $0.05 for each copy over 10 copies <------

Future Seismic Hazards... (Phase I) $15.00 Seismic Hazards... 1994 to 2024 (Phase II) $5.00 SCEC Quarterly Newsletter: One year (4 issues) $25.00

Two years (8 issues) $40.00 Field Trip Guides

Palos Verdes Peninsula $5.00Newport-Inglewood and Whittier-Elsinore $5.00Palos Verdes Fault Field Trip $5.00

C-Cubed Task Reports: Complete Set $250.00Individual reports: Indicate quantity after each itemH-1 $100.00__ H-2 $25.00__ H-3 $10.00__H-4 $25.00__ H-5 $100.00__ H-6 $20.00__H-7 $40.00__ H-8 $50.00__ H-9 $25.00__

Printed Workshop Proceedings:Earthquakes and Insurance I $15.00

Recommended Procedures for Implementation of DMG SP 117 (Analyzing and Mitigating Liquefaction in Calif.) $10.00

Credit card: Number:_______________________, Exp. date: __________Fax this form to 213-740-0011. Signature:________________________

Please make checks payable to USC : Your check number: _______________Mail this form to the address above.

Name: ___________________________________________________________________________________________________

Organization: ___________________________________________________________________________________________________

Address: ___________________________________________________________________________________________________

___________________________________________________________________________________________________

Telephone: __________________________________________ Fax: __________________________________________________

E-mail: ___________________________________________________________________________________________________

SS CC EE CC

VISA MC DISC

Southern California Earthquake CenterUniversity of Southern California

Los Angeles, CA 90089-0742Phone: 213-740-1560 Fax: 213-740-0011

Tax included in all prices. Shipping & handling included except for RootsOur Tax I.D. # is 95-1642394.

* Price is reduced for quantities of 1000 or more. Call 213-740-0323 for details.

TOTAL:

Your card will be billed by USC

Products and Publications Order Form

Putting Down Roots in Earthquake Country—The U.S. GeologicalSurvey and SCEC produced two million copies of this illustrated 32-pagecolor publication. Its message is that earthquakes are inevitable butunderstandable and that damage and serious injury are preventable.

Future Seismic Hazards in Southern California, Phase I: Implications ofthe 1992 Landers Earthquake Sequence—Primarily a study of the impli-cations of the Landers earthquake, this report discusses the recentincrease in the frequency of earthquakes in southern California.

Seismic Hazards in Southern California: Probable Earthquakes, 1994 to2024 (Phase II)—This report represents a major advance in our knowl-edge of how often shaking from earthquakes in specific areas of southernCalifornia will be strong enough to cause moderate damage.

SCEC Quarterly Newsletter—Features include contributions by SCECscientists and working group participants; a compilation of currentlyavailable resources, published materials, and databases; a “Fault of theQuarter,” showcasing a southern California fault; and an interview witha prominent SCEC scientist in each issue.

Publication Descriptions

Field Trip GuidesPalos Verdes Peninsula—Written for teachers and students as well as thegeneral public, this guide offers a lively narrative on a number of sites atwhich to observe fossils, rock structures, and faults. Two foldout mapsare included.

Palos Verdes Fault—This field trip guide is designed for engineers,geotechnical professionals, and earth scientists. Unlike the broad infor-mation provided in the Palos Verdes Peninsula field trip guide, this guidefocuses on the Palos Verdes fault. Included is a discussion of the fault asa whole as well as information pertaining to the many sites along theroute.

Newport-Inglewood and Whittier-Elsinore Fault Zones—Rather thanoffering a route to follow for a field trip, this guide discusses the two faultzones, allowing the reader to design his or her own trip. Emphasis is onthe methods scientists use to learn about faults, such as trenching.

Caltrans/City of Los Angeles/County of Los Angeles Task ReportsThe Task H-1 report describes improved empirical models for scaling theamplitudes of earthquake response spectra for the southern Californiaregion. (3 volumes)

The Task H-2 report focuses on potential destructive earthquakes alongthe Hollywood and Santa Monica faults.

The Task H-3 report examines the use of weak motion amplificationfactors for microzonation, and the relationship between weak and strongmotion amplification.

The Task H-4 report describes the development of empirical models forscaling duration of strong ground motion by utilizing regression analysesof recorded data.

The Task H-5 report describes the compilation of a GIS basedgeotechnical database of the Los Angeles Basin for use in strong groundmotion site characterization. (3 Volumes)

The Task H-6 report documents the earthquake performance and lique-faction-related damage to bridges in the magnitude 7.8 Luzon,Philippine earthquake (July 1990), and the magnitude 7.5 Costa Ricanearthquakes (April 1991).

The Task H-7 report demonstrates seismic hazard analysis methodologywith respect to selected sites in the Los Angeles basin, and illustratesseveral methods for generating acceleration time histories.

The Task H-8 report describes the use of geotechnical data to reassessand improve the Los Angeles geological database used to developliquefaction potential maps. This report complements Task H-5.

The Task H-9 report focuses on the cataloging of available strong motionrecords for vertical ground acceleration time histories, together with thecomputed acceleration response spectra.

Recommended Procedures for Implementation of DMG Special Publi-cation 117: Guidelines for Analyzing and Mitigating LiquefactionHazards in California is intended to help engineers, geologists, andbuilding officials evaluate and take protective measures against thepotential liquefaction hazard in many areas of southern California.


A component of the Center’s Outreach ProgramFor more information on

the SCEC Outreach Program,see the Outreach Web page at

WWW.SCEC.ORG

OR contact:

Jill Andrews, Outreach Director213/740-3459 or [email protected]

Mark Benthien, Outreach Specialist213/740-0323 or [email protected]

Ed Hensley, Newsletter [email protected]

A National Science Foundation National Scienceand Technology Center in partnership with the

United States Geological Survey

Administration

Thomas HenyeyCenter Director

Bernard MinsterActing Science Director

John McRaneyDirector of Administration

Jill AndrewsOutreach Director

Board of Directors

Bernard Minster, ChairmanScripps Institution of OceanographyUniversity of California, San Diego

John AndersonUniversity of Nevada, Reno

Ralph ArchuletaUniversity of California, Santa Barbara

Robert ClaytonCalifornia Institute of Technology

James DolanUniversity of Southern California

Ken HudnutUnited States Geological Survey

David JacksonUniversity of California, Los Angeles

Thomas RockwellSan Diego State University

Nano SeeberLamont-Doherty Earth Observatory of Columbia University

Southern CaliforniaEarthquake Center

SCEC Quarterly Newsletter

To Subscribe: One year’s subscription costs$25.00. Please make payment by check,money order, or purchase order payable to“University of Southern California/SCEC.” Donot send currency. Price includes postagewithin the U.S. Overseas airmail costs or spe-cial courier services will be billed. SCECscientists, students, and affiliated agenciesreceive this newsletter free.

Mail your name, mailing address, phonenumber, email, and check for $25 to:

Southern California Earthquake CenterUniversity of Southern CaliforniaLos Angeles, CA 90089-0742

Subscribers: Check your mailing label for thelast issue of your current subscription. If “free”is listed, you will continue to receive the news-letter indefinitely. If any changes need to bemade to your address, please call 213-740-5843 or email [email protected]

Newsletter on the Web:WWW.SCEC.ORG/NEWS

Have questions? Call, fax, or email:Tel: 213/740-1560Fax: 213/740-0011Email: [email protected]

bscr

iptio

n In

fo

SCEC Quarterly Newsletter is published quarterly by the SouthernCalifornia Earthquake Center, University of Southern California, LosAngeles, CA 90089-0742, USA, telephone 213/740-1560 or 213/740-5843, fax 213/740-0011, email: [email protected].

EditorEd Hensley

Feature WriterMichael Forrest

FeaturedInterview: John Anderson .................................................... 3

Rose Canyon Fault ........................................................... 10

Report: Evaluating Liquefaction Hazard .............................. 16

SCEC’s Summer Interns ..................................................... 18

Planning SCEC’s Role in Post Earthquake Clearinghouse ...... 26

SCEC and Teachers Design Instructional Modules ................ 29

DepartmentsFrom the Directors: EarthScope ............................................ 2

Tales from the Front by Susan Hough .................................... 9

News Briefs ..................................................................... 21

Beneath the Science by Mark Benthien ............................... 24

Off-Scale ......................................................................... 25

Calendar ......................................................................... 31

Publications ..................................................................... 32

SCEC Publication Order Form ........................................... 34

Subscription Info .............................................................. 35

InsideThis Issue

SCEC NewsletterContributors

ADDRESS CORRECTION REQUESTED

Writing:

Jill Andrews

Mark Benthien

Thomas Henyey

Susan Hough

Scott Lindvall

Thomas Rockwell

Sara Tekula

Photography:

Mark Benthien

Michael Forrest

Sara Tekula

Assistant Editors:

Barbara Anderson

Linda Townsdin

Column and Incidental Illustrations:Jim Hensley

Cover:Computer enhanced trench log of Rose Canyon fault (see page 10)

Southern California Earthquake CenterUniversity of Southern CaliforniaLos Angeles, CA 90089-0742

SS CC EE CC

Volume 5, Number 1 Southern California Earthquake...

Documents

Transcript of Volume 5, Number 1 Southern California Earthquake...