Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 105, NO. B7, PAGES 16,229-16,236, JULY 10, 2000 Strain Zone accumulation at latitude across the Eastern California Shear 36ø30'N WeijunGan,• J. L. Svarc,J. C. Savage, andW. H. Prescott U.S. GeologicalSurvey,Menlo Park, California Abstract. The motionof a linear arrayof monumentsextendingacrossthe EasternCalifornia ShearZone (ECSZ) hasbeenmeasuredfrom 1994 to 1999 with the Global PositioningSystem. The lineararrayis orientedN54øE,perpendicular to the tangentto the localsmallcircledrawn aboutthe Pacific-NorthAmerica pole of rotation,and the observedmotionacrossthe ECSZ is approximatedby differentialrotationaboutthatpole. The observations suggestuniform deformation withintheECSZ (strikeN23øW)(26 nstrainyr-• extension normalto thezoneand39 nstrainyr-• simpleright-lateral shearacross it) withnosignificant deformation in thetwoblocks (theSierraNevadamountainsandsouthernNevada)on eitherside. The deformationmay be imposedby right-lateralslip at depthon the individualmajorfault systemswithinthe zoneif the slipratesare:DeathValley-Furnace Creekfault 3.2_+0.9 mmyr-•, HunterMountain-Panamint Valleyfault 3.3_+1.6 mm yr-1,andOwensValley fault6.9_+1.6 mmyr-1. However,thisestimate of the slip rate on the OwensValley fault is 3 timesgreaterthanthe geologicestimate. 1. Introduction The Eastern California Shear Zone (ECSZ) [Dokka and Travis, 1990a,b] is a -100-km-wide zone of deformation trendingnorth-northwestinto Nevada from the east end of the "big bend" in the San Andreasfault (Figure 1). North of the Garlock fault the ECSZ spansthe Owens Valley-Little Lake, Hunter Mountain-Panamint Valley, and Death Valley-Furnace Creek-SouthernDeath Valley fault systems(Figure 2). The 1992 M=7.5 Landersearthquakeoccurrednear the southernend of the ECSZ and the 1872 M=7.6 Owens Valley earthquake occurred near its northern end (Figure 1). The shear zone accommodates -24% of the Pacific-North American relative plate motion (M.M. Miller et al., Refined kinematics of the Eastern Californua Shear Zone from GPS observations, 1983- 1998, submittedto Journal of Geophysical Research, 2000; hereinafter referred to as Miller et al., submittedmanuscript). As is the case for other zones of distributed deformation [Savageet al., 1999a], it is not clear whetherthe deformation in the zone is simply a manifestationof strain accumulation due to slip at depthon closely spacedfaults within the zone or is imposed by drag from a continuous distribution of deformationin the viscousupper mantle [Bourne et al., 1998]. The former explanation implies that the strength of•the lithosphereresides principally in the upper crust, whereasthe latter explanationimplies that it residesin the upper mantle. To study the distribution of deformationacrossthe ECSZ the U.S. GeologicalSurveyin late 1994 installeda linear array of geodeticmonuments(shearzone array in Figure 2) along an azimuth of about N54øE extending from the Sierra Nevada mountainsto near Beatty, Nevada. The array was surveyedin 1994, 1996, 1997, and 1999 using the Global Positioning System (GPS). That array of monumentsjoined another geodetic strain array (Yucca Mountain array in Figure 2) centeredon Yucca Mountain,Nevada,whichhad beensurveyed in 1993 and 1998 using GPS [Savageet al., 1999b]. Before 1995, codeless Ashtech LM-XII GPS receivers were used. Resolution of phase ambiguities was difficult in those surveys. By 1995 the receivershad been upgradedto Ashtech Z-12's, which receive full-wavelength data and allowed better resolution of phase ambiguities. In the 1994 survey of the shear zone array the individual monumentsin the array were occupied for at least 6 hours in each survey; in the other surveys the occupationswere with few exceptions for 6 or more hourson each of two or more consecutivedays (see the Web locationscited below for occupationhistories). The data were reducedusing point positioning[Zurnbergeet al., 1997] and GIPSY/OASIS-II software,release5 [Webband Zurnberge, 1995]. The solutions are referred to the International Terrestrial Reference Frame 1996 (ITRF96) [Sillard et al., 1998] rotated so that North America is nominally fixed [Savage et al., 2000, appendix] determined so as to minimize (i.e., the rotation is the velocities at the fiducial stations Algo (Ontario, Canada), Brmu (Bermuda), Drao (British Columbia, Canada), Fair (Alaska), Nlib (Iowa), Pie l (New Mexico), and Yell (Northwest Territories, Canada)). Because it is not certain that all of the fiducial stations are in fact fixed with respect to North America, the velocities relative to fixed North America assignedto the monumentsin the array discussed here have a common (systematic) uncertaintyof perhapsone or two millimeters per year. The relative velocities of the monuments within the array, however, are free of that systematicerror. 1OnleavefromtheInstitute of Geology, ChinaSeismological The overall accuracy in locating one geodetic monument Bureau,Beijing. with respectto anotherout to 400 km was expectedto be -4 mm (one standard deviation) in each horizontal component Thispaperis notsubject to U.S.copyright.Published in 2000 [Zhang et al., 1997] for a daily solution. However, in this by the AmericanGeophysicalUnion. paper the error estimates, determined by scaling upward the Papernumber2000JB900105. formal errors from the solutionby a factor found appropriate 16,229 16,230 GAN ET AL.: EASTERNCALIFORNIA SHEARZO• ....... :',i'-....... '........... ........... .......... •........... ! .......... occupiedin more than two surveysand thus are the only monumentsfrom the Yucca Mountainarray shownin Figure 3.) In Figure 3 the displacementsfrom arbitrary positions : ............ fixed in our reference frame have been resolved into N36øW i i: "i'• i i 1872 [•• . ...... : O : ' •' ' ....... :Y-,x •?•1857} ' ' ..........*6.... ' .... ,x>- ••._ '"'•3 • • • (direction of the tangentto the local small circle drawn about the Pacific-North American pole of motion [DeMets et al., 1990]) and N54øE components.A straightline, the slopeof which represents the monument velocity relative to the referenceframe,hasbeenfit to the observeddisplacements at each station. The data in Figure 3 are reasonablywell explainedby the linear fits. Thus we assumethat the motion of the monuments has been uniform in time. On the basis of that assumptionwe have usedthe adjustmentprogramQOCA [Dong et al, 1998] (see also the Web site http://sideshow.jpl. nasa.gov:80/~dong/qoca/)in the global mode to find the velocity at eachmonument(Figure 2). The velocity at monument Go05 is clearly anomalous (Figure 2). That monumentis closeto the Coso geothermal .....• .... :...................... :......... • _• .... J....... •__ J...... t_-•q:- -', Figure 1. Map showingthe principalhults in southern field and alsoto the epicenters(Figure2) of threemagnitude5 California and the location of the EasternCalifornia Shear earthquakeswhich occurredin 1996 and 1998 [Bhattacharyya Zone (shaded). The locations of the monuments in the et al., 1999], within the interval of observation.R.W. King geodeticarrayare shownby soliddots. Epicenters of the (personalcommunication,2000), who has analyzed detailed 1857Fo• TQon,1872OwensValley,1952KernCounty, and GPS surveysin the vicinity of the geothermalarea, doesnot 1992LanderseaChquakes am shownby stars. believe that the earthquakesnear Coso had any significant effecton GPS stationsin our array. Wickset al. [1999] report 1992-1997subsidence at therateof 30 mm yr-1centeredon from the linear fits to displacement versus time plots of individual stations [Savage et al., 2000, p. 3102], are somewhatgreater (see error bars in Figure 3), particularly in the early surveys. The additional uncertaintymay be due to monument instability but more likely simply reflects the uncertainty in locating monumentswith respect to the more distant fixed interior of North America. This estimated uncertaintydoes not include the systematicerror in fixing the reference frame with respect to stable North America. An independent estimate of the standard deviation of the velocities for the Yucca Mountain array (abscissa>20 km in Figure 4) basedon the assumptionthat there is no significant relative motion between monumentsin that array is consistent with the standarddeviations used here (J.C. Savage et al., Strain accumulation near Yucca Mountain, Nevada, !993- 1998, submittedto Journal of GeophysicalResearch,2000). In this paper we adopt the following conventions: All uncertainties quoted in the text and tables are standard deviations, but the error bars in the figures extend 2 standard deviationson either side of the plotted points. In discussing strain, extension is reckoned positive. 2. the geothermal area at Coso; the subsidenceis reasonably well modeledby the collapseof a sphericalsource(crossin Figure 2) at a depth of 3.5 km. Because the subsidence probably affected Go05, we will exclude that monumentfrom further discussionin this paper. MonumentsFork and V511 may also be affected by the deformation in the Coso geothermal field, but we retain those monuments in this discussion. The velocitiesdeterminedfrom the QOCA adjustmentare plotted in Figure 4 as functionsof the distancein the N54øE direction state line. from monument The velocities Mo93 on the California-Nevada have been resolved into N54øE and N36øW components,perpendicularand parallel, respectively, to the tangentto the local small circle drawn about the PacificNorth American pole of rotation. The N54øE velocity component(Figure 4b) indicateslittle systematicmotion in that direction. The profile for the N36øW velocitycomponent (Figure 4a) suggests shear acrossthe tangent to the small circle, but that shearis confinedentirely to the monumentsin California (negative abscissain Figure 4). The CaliforniaNevada state line seemsroughly to coincide with the eastern edgeof the shearzone at this latitude. Data 3. Strain Analysis Detailed data (latitude and longitude of the monuments, positionsof monumentsas a function of time, and velocities Figure 4 suggeststhat deformationin California (negative inferredfor the monuments) and plotsfor the two arraysare abscissa in Figure 4) is relatively uniform, whereas availableon the Web at http://quake.wr.usgs.gov:80/QUAKdeformation in Nevada (positive abscissain Figure 4) is ES/geodetic/gps/Yucca.qoca/. That Web site will be updated negligible. To examine this further we have solved for the as new data become available. uniform strain that best approximates the observed The observedchangesin monumentpositionsin a reference deformationin California and Nevadaseparately.The QOCA framenominallyfixed with respectto the interiorof the North adjustmentprogram outputsthe uniform strain rate and rigid Americanplate are shownin Figure 3 as a functionof time for body motion most consistentwith the velocity solution for thosemonuments occupiedin morethantwo surveys.(Seethe any subsetof the adjustedarray. We have chosento examine Web site referencedabove for plots of all data. Only two two subarrays,one composedof the monuments in California monuments,l pdi and Shos,in the YuccaMountainarraywere and the other composed of the monuments in Nevada GAN ET AL.' EASTERN CALIFORNIA SHEAR ZONE 16,231 19W iz GPS Station z ß Yucca Mtn. Array 3187 , . \\ ........ \\ . . ß ß > 10mm/¾r , . . 30 Km . & Shear Zone Array , : Fault , •/•, Holoce Quaternary Figure 2. Map of the GPS arraysacrossthe EasternCalifornia ShearZone (solid triangles)and aroundYucca Mountain (solid circles). The velocity relative to the fixed interior of North America for each monument is shown by an arrow. The 95% confidenceellipse is shownat the tip of the arrow. The locationsof the towns Beatty, Nevada, and Lone Pine, California, are shownas shadeddiamonds. The principal faults are shownby sinuousblack lines, and the California-Nevadastateboundaryis shownby the diagonaldashedline acrossthe figure. The diagonalhachuredline trendsN54øE, perpendicular to the tangentto the local small circle drawn about the Pacific-North American pole of rotation. The epicentersof the three magnitude5 earthquakesin 1996 and 1998 that occurred15 km eastof the Cosogeothermalfield are shownby the starsand the centerof the sphericalsourceusedby Wickset al. [1999] to modelthe subsidence in the CosoGeothermalareais shown by the nearby cross. MonumentMo93, locatedon the stateline, is assignedto the Nevada subarray. The principal strain rates and the rotation rates found for those two subarraysare shown in Table 1. From independentGeodolitemeasurements of the 30-km-wide Owens Valley trilateration network, which extends between monumentsCerr and Ovro (Figure 2), Savage and Lisowski [1995] found that the principal strain rates for the western shearacrossvertical planesparallel to the tangentto the local small circle drawn about the North American-Pacific pole of rotation. In a coordinate system with the 2 axis directed the 2-st.d.level with the strainrate for the Californiasubarray in Table 1, the Owens Valley strain rates do nevertheless suggestthat strain accumulationin the western part of the California subarray(Owens Valley) may be higher than the parallel to the tangent to the local small circle. That is, the motion across the shear zone is approximated by rotation aboutthe Pacific-NorthAmerican pole of rotationwith angular velocity increasingwith distancefrom the pole. An interpretation of the California subarray strain rates somewhatmore consistentwith the simple shearis suggested by a resolutionof the strain field in a coordinatesystemwith N36øW parallel to that tangent the strainratesare 6.7-+6.5nstrainyr'1, t}12=-40.7-+4.4 nstrainyr'1, and•22= 18.2_+7.2 nstrain yr'1. In thiscaset}•tis notsignificant, t}22 is only marginallysignificant,and t}12differs significantly portionof theECSZ(OwensValley)were82_+15 nstrainyr'1 fromthe rotationrate(0 (Table 1) onlyin sign(i.e., N73ø_+3øW and-39_+15nstrainyr'1 NI7ø_+3øE,extension Thus, to a reasonable approximation the deformation field reckoned positive. Although that strain rate is consistentat representssimple (t}t2=-(0) shear acrossvertical planes averagerate over the entire subarray(Table 1). As suggestedby Figure 4 the strain rates for the California subarray(Table 1) can be approximatedby simple 16,232 GANETAL.:EASTERN CALIFORNIA SHEAR ZONE .......• GOO,• 700 ......• .......... • 3187- 600 • •• Fork 500 E E 400 •-• .........• ....... • T•s, + s,ovj 7 _ T19sT. .........._• Stov .................. .• '•' .......... 4 •............... • _ P166•' • ........... • X137_- X137• ............ •.................. • .......... ....... • .............. • .........•u•r 300 _ L 200 -- M137 - ................ ...................... ,,. .................. _- 200 Oerr i i 13dd Hold 100 •.......... •(• G165 • ;• _ .z. Hold 5................. -•.......................... • ............... • - 100 F23x •............... • ............... • ......... • - ............................... '• • 5 .•. ...... ........ .•. ............................. • •uo• Mo93• 1 _ _ ............................ • .............................. • 1994 ,,,•1•,,,I 1995 .... 0- I,,,•1•1,,,, 1997 1998 1996 • Bm8• 1999 2000 1994 ,,,•1,1,,1•,•,1 1995 1996 Time, Yr 200 .... I .... I .... I .... .... 1997 1998 1999 2000 Time, Yr I''''1''''1'''' _ _ _ Ovro- E E 150 • - Ovro- - • _ -• ................ 3............... '-•........... -•Jctn-!.................................. •........................ '•............... •Jctn! _ E ß 100 --•P16• ._ _ _ ,Sho to 5O z - • -- I 3: IShog I ! I -- _ _ 1pdi- o 1993 '-r,,, I,,, 1994 • I, •,, 1995 I,,,, 1996 I,,,, 1997 I,,,, 1998 I,• 1999 in 2000 Time,Yr x,,,I,,•l•l•,,I 1993 1994 .... 1995 1996 1997 I•l•l•n 1998 1999 2000 Time,Yr Figure3. Plotsof displacement referred to anarbitrary originasa function of timefor all monuments in the geodetic arraysmeasured in morethantwo surveys.The displacements havebeenresolved intoN36øWand N54øE components. Theerrorbarsrepresent twostandard deviations oneither sideoftheplotted point. the 2 axis parallel to the N23øW trend of the ECSZ. In that coordinatesystemthe strain rates are s//= 24.9+6.5 nstrain .! the extension rate •2 parallelto thetrendof theshearzoneis negligible. Our preferredinterpretation of deformationin the yr-], t}[2=-39.2+4.4nstrainyr-], andt•j2= 0.0_+7.2 nstrain ECSZ at this latitude is simple shearacrossthe trend of the yr-•. Noticethattheshear ratet•[2 differssignificantly from zone plus extensionperpendicularto it. the rotationrate to (Table 1) only in sign(i.e., t•/'2=-to ) and The deformation of the Nevadasubarray(Table 1) is at best GAN ET AL.: EASTERN CALIFORNIA SHEAR ZONE 20 16,233 Earthmodelin which the relativemotionbetweentwo rigid (a) plates is accommodated by uniform deformation in the parallel-sidedzone that separatesthem [Savageet al., 1995, appendix]. Thatis, thecurvature of theEarthis neglected and 15 southern Nevada E E • o and the Sierra Nevada mountains are represented by rigid blocksseparated by a parallel-sided zone (ECSZ) subjectto uniformdeformation (Figure5b). We have shownabovethatthe strainratein theNevadasubarray to the 10 eastof the shearzone is not significant,and the SierraNevada blockto thewestof the shearzoneis generallytreatedasrigid [e.g.,Hearn and Humphreys,1998]. Thusapproximating the SierraNevadablockand the southernNevadablockas rigid z 5 plates is reasonable. The strain within the ECSZ (California subarray) conforms to therequirements (•/2=-r0 and t}22=0) • •: •-- 0 ß .o• ..... -' $0 X i * -1 O0 7111 ' ,.... -$0 ß for a zoneof deformation separating two rigid plates[Savage et al., 1995, appendix](noticethat in this referencethe sign I I I •- , i•/ ß ! .... 0 conventionfor r0 is oppositethat employedhere). Thus the ! lOO lO ECSZ canbe interpreted[Savageet al., 1995,appendix]as a zoneof uniformdeformationseparating rigid plates(southern Nevada and the Sierra Nevada mountains). Notice that in this flat-Earthmodel,thereis no rotationof theSierraNevadaplate relativeto southern Nevada. Specifically, thereis no apparent rotation of the Sierra Nevada block relative to fixed interior North America that would permit a fan-like openingof the Basin and Range province to accommodatemore east-west spreadingin the north than in the south [Bogen and Schweickert, 1985]. Nor was such a rotation found in the preferred representation (model M2) of deformation in the sameareaby Hearn and Humphreys[1998]. 4. Fault Model Here we attemptto explain the deformationobservedacross -lO N54øE Distance from Mog3, km Figure 4. Profiles of the (a) N36øW and (b) N54øE componentsof velocity as a funcition of distanceN54øE from monumentMo93 on the California-Nevadaboundary. The the ECSZ as beinggeneratedby strikeslip at depthon the threeprincipalfault systems (the DeathValley-Furnace Creek, the HunterMountain-Panamint Valley, andthe OwensValleyLittle Lake) that crossthe geodeticarray(fault crossings are indicatedby vertical dashedlines in Figure 4). We have representedthe fault systemsby simple dislocationsin an elastichalf-space[Okada, 1985] and determinedthe Burgers vector for each dislocationsuchthat the observedvelocity error bars represent two standarddeviations on either side of the plotted point. The continuous solid lines indicate the velocitespredictedby the dislocationmodel of Table 2. The field is best explained. The faults are represented by monumentsClai, B191,Mile, 67tj and Whme (Figure 2) from (e.g., slip on the Hunter Mountain fault is the same as on the Panamint Valley fault). Notice that no allowance has been dislocationsegmentsas shownin Figure 5a. The Owens Valley-Little Lake fault systemhas been represented by a dashedvertical lines indicatethe locationof profile crossings single screw dislocation and the Death Valley-Furnace Creekof the principal faults. Southern DeathValleyandHunterMountain-Panamint Valley fault systems haveeachbeenrepresented by segmented screw dislocations.All dislocations areplacedat a depthof 15 km marginally significant [Savageet al., 1999b]. Wernickeet al. (the locking depth), and the slip on each fault systemis to be constantalongthe lengthof the system [1998] have argued that independent GPS surveys of constrained 1991 to 1998 indicate a N65øW strain rate of 50_+8 nstrain yr-1. However,thedataavailable to Wernicke et al. [1998] made for normal slip on the Death Valley-FurnaceCreekSouthern DeathValleyandHunterMountain-Panamint Valley used here, and we are confident that the strain rate in the fault systems noron theSierrafrontal faultsystem,although were much less completethan the U.S. GeologicalSurvey data Nevada subarrayis very low (Table 1). The observedabsence of significantstrain accumulationin the Nevada subarrayis of some importanceas the proposedU.S. high-level radioactive waste disposalrepositoryis located close to monumentMile (Figure 2). The observed deformation of the entire geodetic array (CaliforniaplusNevadasubarrays) canbe represented by a flat- suchslip probablyoccursthere; data on vertical deformation wouldbe requiredto effectivelyconstrainsuchnormalslip. The bestfit valuesfor slip rateson the variousfault systems are shownin Table 2. With the importantexceptionof the 2 mmyr-1geologic sliprateestimate fortheOwens Valleyfault [Beanland and Clark, 1994], the slip ratesin Table 2 are withinthe rangeof previousgeologicandgeodeticestimates 16,234 GAN ET AL.: EASTERN CALIFORNIA SHEAR ZONE Table 1. Uniform Principal Strain and Rotation Rate Approximationsto Observed Deformation in theSubarrays Subarray California Nevada t}l, nstrain yr-l 53.6_+6.9 N77ø_+ 3W 9.9_+5.4 N60ø_+13øE t•2, nstrain yr-l 0), nrad.yr-l -28.7e5.5 N13ø_+ 3øE -11.5_+7.4 N30ø_+13øE 37.1_+4.1 6.0_+4.4 Quoted uncertaintiesare standarddeviations (see Table 1 of Hearn and Humphreys [1998] for a recent compilation of slip rates) and reproduce the observed deformation satisfactorily (continuous lines in Figure 4). Moreover, the slip rate estimatesin Table 2 are completely changingthe locking depth to 10 or 20 km does not change the slip rates significantly and gives almost as good a fit to the observationsas the 15 km locking depth employed. The discrepancybetween the geologicestimateof slip rate consistent with similar estimates of Bennett et al. [1997], (2_+1mm yr-1 [Beanlandand Clark, 1994])on the Owens Valleyfaultandthe estimate (6.9_+1.6 mm yr-l ) in Table2 McClusky et al. [1999], and Miller et al. (submitted manuscript,2000) basedon independentGPS measurements. However, becausethe velocity profiles acrossthe shearzone are relatively smooth, our dislocation models for the individual faults are not closely defined. For example, calls into questionthe validity of representingdeformationin the ECSZ solely by slip at depth on the three principal fault systemsthere. Although Beanland and Clark [1994] admit some uncertaintyin the geologic estimateof the slip rate on the Owens Valley fault, their argumentsfor a low slip rate are convincing.For example,the 6.9_+1.6 mm yr-1 slip rate proposedhere would require an earthquakesimilar to the 1872 Owens Valley event roughly every 1000 years, whereas the geologicevidencesuggestsonly three suchearthquakesin the last 10,000 to 20,000 years [Beanland and Clark, 1994]. Dixon et al. [2000] suggestthat the current high slip rate on the Owens Valley fault inferred from GPS might be explained by postseismicrelaxation following the 1872 earthquake. The slip rate in the later part of the earthquakecycle would be expectedto be much less, so that the slip rate averagedover theentireinterseismic intervalcouldapproach 2 mmyr-l. 5. Discussion Hearn and Humphreys [1998] modeled the deformation acrossthe Eastern California Shear Zone using geologic slip ratesandthe sparsegeodeticdatathenavailable. They divided the region into fault blocks using the major fault systemsas boundariesand then, subject to constraintsimposedby the geodeticand geologicobservations,determinedthe slip rates on those faults which minimized the elastic energy accumulation. Their model M2 represents the best approximation to block motion consistent with the data. BecauseHearn and Humphreys[ 1998] model the block motion (i.e., motion averagedover many earthquakecycles) whereas we observethe interseismicdeformation,direct comparisonof of their model with our observations is not possible. However, the observedmotions of the rigid blocks on either sideof the ECSZ are generallyconsistentwith model M2. For example,their model M2 predicts12.7_+1.5mm yr-• Table 2. Slip Rates Estimatedfrom GPS Measurements Figure 5. Two modelsof deformationin the ECSZ. (a) Deformation attributedto slip at depth on individual faults. for FaultSystems withintheEastern CaliforniaShearZone Fault System R.-L. Slip Rate, mmyr-1 The tracesof the dislocationslip planesused to representthe three major fault systemsare shownby the heavy straight lines. (b) Deformation attributed to uniform strain accumulation (heavycrossedarrowswith principalstrainrates in nanostrainper year) in a parallel-sided shearzone(shaded) with no strain accumulation in the blocks on either side of it. Death Valley-FurnaceCr.-S. Death Valley Hunter Mountain-PanamintValley Owens Valley-Little Lake Quoted uncertaintiesare standarddeviations 3.2_+0.9 3.3_+1.6 6.9_+1.6 GAN ET AL.: EASTERN CALIFORNIA SHEAR ZONE N50ø_+5øW for the velocityof theSierraNevadablockand2.5 mmyr-1N87øWfor thevelocityof thesouthern Nevadaplate, whereasour observedvelocity on the Sierra Nevada plate (monument 3187)is 13.9_+1.4 mmyr-1 N39ø_+6øW andthe average velocityfor thestations in theNevadasubarray is 3.0 mmyr-1 N71øW. In distributing slipwithintheECSZmodel M2 requires a 1.5-2.3mmyr-l right-lateral sliprateonthe 16,235 representedin the thin viscous sheet model of deformation [England and Jackson, 1989], where the entire lithosphereis modeled by a thin viscouslayer. The elastic strain imposed upon the uppercrustis eventuallyrelieved by ruptureon faults so that in the long term the upper crust deforms by block motion. Our observationsacrossthe ECSZ suggestthat deformation OwensValley-Little Lake fault system(in good agreement is relatively uniform across the shear zone, not clearly withthegeologic estimate of Beanland andClark[1994])and concentratedupon the principal faults in the zone. This would a 5-6 mmyr-• right-lateralslip rate on the DeathValleyFumaceCreek-SouthemDeathValley fault system,whereaswe infer(Table2) 6.9_+1.6 mmyr-• ontheformerfaultsystem and 3.2_+0.9 mmyr-• onthelatter.Theagreement is betterfor the HunterMountain-Panamint fault system(modelM2 predicts2- appear to favor the hypothesis that the deformation is imposed by a shear zone at depth in the upper mantle. However, if reasonable locking depths are assigned to the principalfaults within the ECSZ and slip ratesfor thosefaults selected to be as consistent with the observed deformation as possible,the deformationat the surfaceof the shear zone is In summaryour observations are consistent with the overall relatively uniform (Figure 4). However, the slip rate assigned pattemof deformation predicted by Hearn and Humphreys to the Owens Valley fault in this model is not in agreement [1998]but differ in the detaileddistribution of slipinsidethe with the secularslip rate inferred from geology. ECSZ where our observations suggest deformation is 3 mmyr-• right-lateral slipwhereas weinfer3.3_+1.6 mmyr-•). concentrated more to the west side of the shear zone (Owens 6. Conclusions Valley-LittleLakefaultsystem)andmodelM2 indicates that the deformationis concentratedon the east side (Death ValleyFumaceCreek-SouthemDeath Valley fault system). Distributed shear such as we have observed in the ECSZ is To a first approximationdeformationacrossthe ECSZ can be attributed to distributed uniform simple shear across vertical planesparallel to the tangentto the local small circle drawn about the North America-Pacific pole of rotation. However, the observations imply a marginally significant generally associated withductileflow. However,because the uppercrustitselfis thought to be brittle,ductileflow thereis rated22=18.2-+7.2 nstrain yr-1parallel to theshear not a viable explanation of the observed distributed extension direction(N36øW) that is not explainedby the model. A better deformation. Rather, the deformation must be attributed to elasticdeformationin the upper crustthat occursin response solution is to resolve the strain rates into a coordinate system with the 2 axis parallel to the trend of the ECSZ (N23øW). In to deformation in the lower crust and upper mantle. Two that coordinatesystemthe deformationcorrespondsto simple explanations are availablefor the observed concentration of strainaccumulation alongplateboundaries.The explanations right-lateralsheare•2 -39.2_+4.4nstrainyr-• acrossa are basedon different assessments of the relative strengthsof vertical plane striking parallel to the shear zone plus an the uppercrustandthe uppermantle. The conventional model (Figure 5a) assumesthat the extension,df•= 24.9_+6.5 nstrainyr-• perpendicular to the trend of the zone, with no significantextensiond2'• parallel strengthof the lithosphereresidesprincipallyin the upper to the zone. This correspondsto the flat-Earth deformation crust;thatis, the uppercrustactsas a stressguide. A shear zone model of Savage et al. [1995, appendix] in which the zone at the surfaceis then interpretedas the consequence of Sierra Nevada block is translatingboth parallel to the Eastern elastic strain accumulationdue to slip at depth on a fault California ShearZone and away from it relative to the southern Nevada block (Figure 5b). In that model the motion is accommodated by uniform extension and shear within the [Savage and Burford, 1973]. For example,considera transform boundary.In the absence of a plateboundary fault, the contactingplates would both be uniformly strained. ECSZ, and the Sierra block does not rotate to accommodate a However,wherea transformfault existsat the plateboundary, fan-like opening of the Basin and Range province. The the relativemotionof the platesis accommodated by slip at deformation can also be attributed to slip at depth on the depthonthefault,withonlytheuppermost 10to 20 kmof the principalfault systemswithin the ECSZ (Figures4 and 5a and fault locked. This flaw (a slippingfault at depth)causesthe Table 2). However,suchan explanationrequiresa slip rate on surfacestrainaccumulation to concentrate alongthe fault with the Owens Valley fault at least 3 times greater than that negligible strainaccumulation beyond a horizontal distance of inferred from geology [Beanland and Clark, 1994], a morethanaboutsix lockingdepthsfrom the fault; hencethe discrepancy which calls into question the validity of the appearance of a shearzoneat thesurface. Alternatively, onemayargue[England andJackson, 1989; model. Acknowledgment.We thankDananDongfor hisassistance in using Bourne et al., 1998] that the strength of the lithosphere his adjustmentprogramQOCA. resides primarilyin theuppermantle.Theuppercrustis then regarded asa badlyfractured layerthatpassively followsthe References deformationof the underlying upper mantle, which is undergoing ductileflow. A shear zone(Figure5b)observed at Beanland, S., and M.M. Clark, The Owens Valley Fault Zone, eastern California,and surfaceruptureassociated with the 1872 earthquake, thesurface is regarded simplyaselasticdeformation drivenby U.S. Geol. Surv. Bull., 1982, 29 pp., 1994. basaltractionsexertedupontheuppercrustby ductileflow in Bennett,R.A., B.P. Wernicke, J.L. Davis, P Elosegui,J.K. Snow, M. J. theuppermantle,presumably flow concentrated at a zoneof Abolins, M.A. House, G.L. Stirewalt, and D.A. Ferrill, Global weakness at theplateboundary (shearzonein uppermantle). PositioningSystemconstraints on fault slipratesin the DeathValley region,CaliforniaandNevada,Geophys. Res.Lett., 24, 3073-3076, The weak upper crust above is draggedalong by the 1997. deformationof the upper mantle. This viewpoint is 16,236 GAN ET AL.: EASTERN CALIFORNIA SHEAR ZONE Bhattacharyya,J., S. Gross, J. Lees, and M. Hastings, Recent earthquakesequencesat Coso: Evidencefor conjugatefaulting and stressloading near a geothermalfield, Bull. Seisrnol.Soc.Am., 89, 785-795, 1999. Bogen, N.L., and R.A. Schweickert,Magnitude of crustal extension acrossthe northernBasin and Range province: Constraintsfrom paleomagnetism, Earth Planet.Sci. Lette, 75, 93-100, 1985. Bourne,S.J.,P.C. England,andB. Parsons, The motionof crustalblocks drivenby flow of the lower lithosphere:Implicationsfor slip ratesof faults in the South Island of New Zealand and southern California, Nature, 391,655-659, 1998. Savage,J.C., M. Lisowski,J.L. Svarc,and W.K. Gross,Strain accumulation across the central Nevada seismic zone, 1973-1994, J. Geophys. Res.,100, 20,257-20269,1995. Savage,J.C., J.L. Svarc,and W.H. Prescott,Geodeticestimatesof fault slip ratesin the SanFranciscoBay area,J. Geophys.Res.,104, 49955002, 1999a. Savage,J.C., J.L. Svarc, and W.H. Prescott,Strain accumulationat Yucca Mountain, Nevada, 1983-1998, J. Geophys. Res., 104, 17,627-17,631, 1999b. Savage,J.C., J.L. Svarc,W.H. Prescott,and M.H. Murray, Deformation acrossthe forearcof the Cascadiasubductionzone at Cape Blanco, Oregon,J. Geophys.Res.,105, 3095-3102,2000. DeMets, C., R.G. Gordon, D.F. Argus, and S. Stein, Current plate Sillard, P., Z. Altamimi, and C. Boucher, The ITRF96 realization and its motions,Geophys.J. Int., 101,425-478, 1990. associated velocityfield, Geophys.Res.Lett., 25, 3223-3226, 1998. Dixon, T.H., M. Miller, F. Farina,H. Wang, and D. Johnson,Presentday motionof the Sierrablockand sometectonicimplicationsfor the Webb, F. H., and J.F. Zumberge,An introductionto GIPSY/OASIS-II, Rep. JPL D-11088, Jet Propul. Lab., Calif. Inst. of Technol., Basinand Rangeprovince,North AmericaCordillera,Tectonics,24, 1-24, 2000. Dokka, R.K., and C.J. Travis, Late Cenozoicstrike-slipfaultingin the Mojave Desert,Califonia, Tectonics,9, 311-340, 1990a. Dokka, R.K. and C.J. Travis, Role of the eastern California shear zone in accommodating Pacific-NorthAmericanplate motion, Geophys. Res. Lett., 17, 1323-1326, 1990b. Dong, D., T.A. Herring, and R.W. King, Estimating regional deformationfrom a combinationof spaceand terrestrialgeodetic data, J. Geod., 72, 200-214, 1998. England,P., and J. Jackson,Active deformationof the continents,Anu. Rev. Earth Planet. Sci., 17, 197-226, 1989. Heam, E.H., and E.D. Humphreys,Kinematicsof the southernWalker Lane belt and motion of the Sierra Nevada block, California, J. Geophys.Res., 103, 27,033-27,049, 1998. McClusky,S., S. Bjornstad,B. Hager,E. Heam, R. King, B. Meade,M. Miller, F. Monastero, and B. Souter, Geodetic constraints on deformationin the southernWalker Lane Belt, California (abstract), in Proceedingsand Abstracts,1999 SCEC Annual Meeting,p. 81, South,Calif. EarthquakeCent., Univ. of South.Calif., Los Angeles, Pasadena, 1995. Wernicke, B., J.L. Davis, R.A. Bennett, P. Elosegui,M.J. Abolins, R.J. Brady, M.A. House, N.A. Niemi, and J.KoSnow, Anomalousstrain accumulation in the Yucca Mountain area, Nevada, Science, 279, 2096-2100, 1998. Wicks, C.W., W. Thatcher, F.C. Monastero, and M. Hastings, Deformation of the Coso Range, east-central California, from satellite radar interferometry (abstract), Eos Trans. AGU, 80(46), Fall Meet. Suppl.,F1194, 1999. Zhang, J., Y. Bock, H. Johnson,P. Fang, J. Genrich, S. Williams, S. Wdowinski, and J. Behr, Southern California permanent GPS geodeticarray: Error analysisof daily positionestimatesand site velocities,J. Geophys.Res., 102, 18,035-18,055, 1997. Zumberge,J.F., M.B. Heftin, D.C. Jefferson,M.M. Watkins, and F.H. Webb, Precisepoint positioningfor the efficientand robustanalysis of GPS data from large networks,J. Geophys.Res., 102, 5005-5017, 1997. 1999. Okada, Y., Surface deformation due to shear and tensile faults in a half- space,Bull. Seismol.Soc.Am., 75, 1135-1154, 1985. Savage,J.C., and R.O. Burford,Geodeticdeterminationof relativeplate motionin centralCalifornia,J. Geophys.Res., 78, 832-843, 1973. Savage,J.C., and M. Lisowski, Strainaccumulationin OwensValley, California, 1974-1988, Bull Seismol.Soc. Am., 85, 151-158, 1995. W. Gan, W. H. Prescott, J. C. Savage, and J. L. Svarc, U.S. Geological Survey, MS/977, 345 Middlefield Rd, Menlo Park, CA 94025. ([email protected]) (Received October 18, 1999; revised March 9, 2000; acceptedMarch 24, 2000)