Stress Orientation Determined from Fault Slip Data in Hampel Wash Area, Nevada, and Its Relation to Contemporary Regional Stress Field

University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

USGS Staff -- Published Research US Geological Survey

March 2011

STRESS ORIENTATION DETERMINED FROM FAULT SLIP DATA IN HAMPEL WASH AREA, NEVADA, AND ITS RELATION TO CONTEMPORARY REGIONAL STRESS FIELD

Virgil A. Frizzell Jr. U.S. Geological Survey

Mark D. Zoback U.S. Geological Survey, [email protected]

Follow this and additional works at: https://digitalcommons.unl.edu/usgsstaffpub

Part of the Earth Sciences Commons

Frizzell, Virgil A. Jr. and Zoback, Mark D., "STRESS ORIENTATION DETERMINED FROM FAULT SLIP DATA IN HAMPEL WASH AREA, NEVADA, AND ITS RELATION TO CONTEMPORARY REGIONAL STRESS FIELD" (2011). USGS Staff -- Published Research. 470. https://digitalcommons.unl.edu/usgsstaffpub/470

This Article is brought to you for free and open access by the US Geological Survey at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in USGS Staff -- Published Research by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. TECTONICS, VOL. 6, NO. 2, PAGES 89-98, APRIL 1987

STRESS ORIENTATION DETERMINED FROM FAULT SLIP DATA IN HAMPEL WASH AREA, NEVADA, AND ITS RELATION TO CONTEMPORARY REGIONAL STRESS FIELD

Virgil A. Frizzell, Jr.

U.S. GeologicalSurvey, Flagstaff,Arizona

Mary Lou Zoback

U.S. GeologicalSurvey, Menlo Park, California

Abstract. Fault-slip data were collectedfrom an area of Test Site. This solution fits all the data well, except for relatively young faulting in a seismicallyactive part of the a subsetof strike-slipfaults that strikeN30ø-45øE, subparal- Nevada Test Site 12 km NW of Mercury, Nevada. The data lel to the normal faults of the data set. Nearly pure dip-slip come primarily from intensely faulted Miocene tuffaceous and pure strike-slipmovement on similarly orientedfaults, sedimentaryrocks in Hampel Wash, which is boundedon however, cannotbe accommodatedin a single stressregime. the northby the QuaternaryENE trendingRock Valley fault Superposedsets of striae observedon some faults suggest and on the south by a parallel unnamedfault. Data from temporalrotations of the regional stressfield or local rota- faults with known senseof displacementexhibit a bimodal tions within the regionof the fault zone. distributionof slip angles (rakes). Faults exhibiting steep rakes(typically 75 ø to 90ø) clusterabout a N30ø-35øEstrike; INTRODUCTION most dip 65ø to 80ø . Faultshaving shallowrakes (generally less than 20ø) exhibit a wide range of strikes(from N6øW Much of the actively extendingnorthern Basin and Range to N80øE) and mostlydip between80 ø and 90ø. The predo- province appearscharacterized by a WNW least horizontal minant N30ø-35øE strike of the steep-rakefaults and the principalstress direction [Zoback and Zoback, 1980]. Wright quasi-conjugatenature of a consistentsubset of the shallow- [1976] subdividedthe Basin and Range into two deforma rake faults suggesta maximumhorizontal stress orientation tional fields having similar least horizontal principal stress of about N30ø-35øE and a least horizontal principal stress directions:a northernfield dominatedby steeplydipping nor- directionof N55ø-60øW. Analysis of the data using a least mal faults and a southernfield featuring normal and strike- squaresiterative inversionto determinea mean deviatoric slip faults that formed coevally. Although Anderson and principal stresstensor indicates a normal-faultingstress re- Ekren [1977] disputed Wright's sampling of late Cenozoic gime (S• vertical) with principalstress axes in approximately trends on the basis of more detailed fault relationshipsat horizontal and vertical directions (S•, trend = N19øE and several localities, they suggesteda temporal clockwise plunge= 82øN; S2, N30øE and 8øS; and S3, N60øW and change in the least horizontal stress orientation. Focal 2øE). The maximum horizontal stress, S2, was found to be mechanism data [e.g. Vetter and Ryall, 1983] support nearly intermediatein magnitudebetween S• and S3. The Wright's relatively uniform orientationof the leasthorizontal N60øW least horizontalprincipal stressorientation obtained principal stressdirection as well as his suggestionthat S• from the fault-slipinversion agrees with our geometricanaly- and S2 (maximum and intermediateprincipal stressaxes) sis of the data and is consistent with a modem least horizon- may locally interchange. tal principalstress orientation of N50ø-70øWinferred from In the vicinity of the Nevada Test Site, which is located earthquakefocal mechanisms,well bore breakouts,and hy- in Wright's southerndeformational field, a variety of struc- draulicfracturing measurements in the vicinity of the Nevada tural and geomorphicdata led Carr [1974, p. 10] to infer a N50øW leasthorizontal principal stress direction for defor- This paperis not subjectto U.S. copyright. Publishedin 1987 mationthat beganas early as 10 Ma and possibly'as recently by the AmericanGeophysical Union. as 4 Ma. In a more recent structuralanalysis, Ander [1984] concurredwith the presentN50øW orientation.A paucityof Papernumber 6T0752. dip-slip displacementson NE trendingfaults in the vicinity 90 Frizzell and Zoback: StressOrientation, Hampel Wash Area, Nevada

I 116o04' r•] Alluvium(Holocene !•I I EXPLANATIONtoPliocene) •,.,• Tuff(Miocene)

'-"•'•':'-", Tuffeceoustary rocks sedimen-(Miocene toOligocene) -- Contact Fault-Previously mapped.Doffed whereconcealed. Bar andball ondown- thrownside. Arrows -36044 ' sl•owrelative move- menf. Fault-Previouslyun- mapped.Barand ball ondownthrown side, Arrowsshow relative movmenf

ß Senseof faulToffset known © Senseof fault offset unknown ,,,,,JOStrike anddip of beds

NEX/AD•• i -36043 '

0 Ikr, t

Fig. 1. Generalized geologic map of Hampel Wash area, Nevada Test Site, Nevada (after Hinrichs, 1968), showinglocations from which fault slip data were obtained.Most data were collectedfrom previouslyunmapped small faults that are depictedschematically. (a) All data collectedalong or near Rock Valley fault, (b) all data collectedalong southernbounding fault, (c) lower hemisphericequal area projectionfor all Hampel Wash data (n= 160), (d) data collectedfrom faultshaving known senseof displacement(n=50). of Yucca Flat led Ander (p. 54) also to infer a "very recent" been undertaken. The ENE trending Rock Valley fault and clockwisechange in orientationof the least horizontalprinci- a parallel unnamedfault 2 km to the southindicate probable pal stressorientation from N78øW to the modem orientation left-lateral oblique displacementof bedrockand cut Quater- of about N50øW. nary materials [Hinrichs, 1968; Barnes et al., 1982; Ander Several other studies generally support Carr's inferred et al., 1984, pp. 16-17; Carr, 1984, p. 61; Yount et al., N50øW least horizontal principal stressorientation. These 1987]. These faults cut tuffaceoussedimentary rocks in the studiesinclude earthquakefocal mechanismsfrom the vicin- Hampel Wash area where outcrops eroded by ephemeral ity of the Nevada Test Site [Rogers et al., 1983], well bore streamsoffer unique opportunitiesto examine exposuresof breakouts from Pahute Mesa and Yucca Flat [Springer et the ENE trending presumedleft-lateral fault zones as well al., 1984] and well bore breakoutsand in situ stress(hydrau- as faults in the block boundedby them. This report presents lic fracturing) measurements[Stock et al., 1985] in the the results of an analysisof fault slip data collectedin the Yucca Mountain area, all of which indicate either normal Hampel Wash area and modifies preliminary results sum- or strike-slipfaulting that has consistentleast horizontalprin- marized elsewhere [Frizzell and Zoback, 1985]. cipal stressorientations of N50ø-70øW. East-northeaststriking fault zones in a seismicallyactive LOCAL GEOLOGIC SETTING part of the Nevada Test Site [Rogers et al., 1983, Figure 1] have been designatedas left-lateral strike-slipfaults [Carr, Significant stratigraphic and structural changes occur 1974, Figure 3; 1984, p. 61]. Most faults having similar acrossthe Hampel Wash area (Figure 1). These changesare trends generally run along the axes of basins and are thus even apparent on the state geologic map [Stewart and poorly exposed;therefore a thorough examinationof their Carlson, 1978]. The wash itself is underlain by one of the bedrock exposuresand actual sense of movement has not older Tertiary units, the rocks of Pavits Springsof Hinrichs Frizzelland Zoback: Stress Orientation, Hampel Wash Area, Nevada 91

N N

x Normal, left-lateral ß Reverse, left-lateral

Fig. 1. (continued)

[1968]; theseMiocene stratawere depositedon the slightly The youngest volcanic unit that is directly offset, the older Horse Spring Formation.Paleozoic carbonate and clas- Wahmonie Formation, yields potassium-argonages of 12 and tic rocks [Burchfiel, 1964] unconformablyunderlie the Horse 13 Ma [Kistler, .1968].Clasts of the younger[8 and 10 Ma, Spring Formation and occupy the ranges to the south and R. J. Fleck, written communication, 1980] basalt of Skull east; silicic ash fall tuffs and flows of Miocene and Pliocene Mountain reside on the surfaces of beheaded alluvial fans age predominatein the ranges to the north [Byers et al., of probable Pliocene age south of the physiographicRock 1976]. Faults in the Paleozoicrocks south of the Hampel Valley [southof Figure 1; Hinrichs, 1968; J. C. Yount, oral Wash area strikeN60ø-80øE, subparallel to the Rock Valley commun., 1985]. These clasts, derived from the Skull Moun- fault, but north of the wash, faults in the upper Tertiary tain area (north of Figure 1), must have traveledsouth across rocks exhibit NNE to N trends. the present position of Rock Valley before the valley was The faults boundingthe block that containsHampel Wash formed. Thus, althoughthe Rock Valley fault may be older displacenumerous Neogene units. Although the amountof than the fans, its topographicexpression, and perhapsthat offseteludes exact measurement,lateral displacementproba- of the southernboundary fault, apparentlydeveloped after bly is not greaterthan a few kilometers[Barnes et al., 1982]. about5 Ma. Multiple eventshave occurredon the Rock Val- 92 Frizzelland Zoback: Stress Orientation, Hampel Wash Area, Nevada

Dip Distribution o

rake<45 ø

r-28

Dip Distribution

rake<45 ø

r-7 rake_>45 ø

r-ll

9O /

•' Rake Distribution

r-26

rake>45 ø -- ...<-

r=10 90

Fig. 2. Distributionsof strike, dip, and rake for entire data set (n= 160) of striae on faults in Hampel Wash area, Nevada, representedon Figure l c. Frizzell and Zoback:Stress Orientation, Hampel Wash Area, Nevada 93 ley fault over the past severalhundred thousand years, and striking (left-lateral) faults suggesta maximum horizontal upper Pleistocenematerials have been clearly offset [Yount principalstress orientation of N30ø-35øEand a leasthorizon- et al., in press]. Although the fault creates prominent tal principal stressorientation of N55ø-60øW. Clearly, the brushlinesin Holocenedeposits, actual breakage of Holocene NE striking strike-slipfaults (both right- and left-lateral) are materials has not been documented. inconsistentwith this simple stressregime.

GEOMETRIC ANALYSIS OF FAULT PATI•ERNS QUANTITATIVE ESTIMATION OF PRINCIPAL STRESS ORIENTATION Exposures in Miocene tuff and tuffaceous sedimentary rocksyielded 160 orientationsof striaeon faults having measured displacementsranging from 4 cm to 1.7 m. Probable Becausethe 50 faults having a known senseof offset are senseof displacementwas determinedfor 50 of these fea- representativeof the completedata set (compareFigures l c, tures by using stratigraphicoffset and macrostructuresand ld, 2, and 3) and constitute the best data collected, they microstructuresdiscussed by Angelier et al. [1985, p. 351- were analyzedby using a least squaresiterative inversion 353]. outlinedby Angelier [1984] to determinethe meandeviatoric Most of the data were collected from small faults between principal stresstensor. This methodof analysisdetermines the two ENE trending faults boundingthe Hampel Wash the parameterswhich define the directionof slip on a fault area. Althoughfault slip data from the vicinity of the bound- plane, i.e., the orientationof the principal stressaxes and ing faultsrepresent a broaddistribution of attitudes(see Fig- a ratio of relative stress magnitudes•= (S2- S3)/(S1-S3), ures la and lb), most faults have shallowlyplunging rakes. by minimizingthe meanangular deviation between the com- True senseof offset was not determinedon any of the ENE putedand observedslip vectoron eachfault. trendingfaults along the boundingfaults, however, apparent Becausethe iterative inversionprocess requires a "starting offset of rock units suggeststhat thesebounding faults have point" stressregime to computeslip directionsfor compari- a left-lateral oblique offset. If the displacementis consis- son with the observedfault slip data, the data were subjected tently left lateral, then the slip data suggestboth minor nor- to two inversions:one with an initial "normal dip-slip fault- mal and reverse componentsof displacementon these near- ing" stress regime (S• vertical) and one with an initial verticalENE trendingfaults. "strike-slipfaulting" stress regime (S2 vertical).Both regimes The entire populationof measurementsfrom the Hampel were assigneda consistentleast horizontalprincipal stress Wash area is representedon a lower hemisphereequal-area (S3) orientationof N60øW. Starting with a normal-faulting projectionin Figure l c. Despite a broad distributionof fault stressregime, inversionof the data set havinga known sense attitudes, strike directionsin the northeastquadrant domi- of motion convergedrapidly on the "final" solution (S•, nate. Distributionsof strike, dip, and slip angles(rakes) are N19øE trend and 82øN plunge; S2, N30øE and 8øS; S3, shownin Figure 2 in two subsetsof the data distinguished N60øW and 2øE) and a •= 0.47. Starting with a strike-slip by rakes. Note that strike-slipfaults (used herein for faults faulting stressregime (S: vertical), the inversionyielded a having rakes less than 45ø) exhibit an apparentbimodal dis- N58øW orientation for the S3 axis, essentiallyidentical to tribution of strike (N35ø-45øE and N70ø-80øE), whereasnor- that derived from the normal-faulting inversion. However, mal faults (faults with rakes greater than or equal to 45ø ) the best fitting • value was 1.0, implying that the maximum exhibit a unimodal distribution(N30ø-35øE). horizontalprincipal stress magnitude equals the verticalstress About a third of the measured faults, 50 of 160, exhibit (S•=S2), or a stressregime transitionalbetween strike-slip characteristicsthat indicatetrue senseof displacement.Faults and normal faulting. Hence, the iterative solutionfor an in- of this subset(Figure l d) exhibit attitudessimilar to those itial strike-slip stressregime moved as close as allowed to of the entire data set. Distributionof dip and rake (Figure a "normal-faulting"solution. Significantly, the "best"strike- 3) appears similar to that for the entire data set, with the slip faulting solutionhad a higher deviation angle between exceptionthat the shallowly dipping dip-slip faults are not observedand computedrakes (34ø ) than did the normal-fault- included in this subsetbecause their senseof displacement ing solution(31ø), indicatingthat the best fit to the overall was not determined in the field. data set was the normal-faultingregime with •= 0.5. Shallow-rake(strike-slip) faults having a known senseof Additional inversionof the data set weightedby increasing offset exhibit a wide range in strike (Figure 3), although amount of displacement[Angelier et al., 1985, p. 351] they can be divided into two nearly distinctfields (Figures yielded similar • values and principal stressaxes that dif- 3 and 4). Faults having right-lateral offsets strike N to NE fered by less than 1ø. Thus the S3 axis determinedfrom (N6øW to N36øE), whereas left-lateral faults strike NE to the fault data approximatesa least horizontalprincipal stress ENE (N36øE-80øE). Two left-lateral faults, numbered on direction of about N60øW, which comparesfavorably with Figure 4 with their slip directionscircled, are exceptionsto other estimatesof the contemporarystress field based on this general pattern and may representmeasurement errors. focal mechanisms,well bore breakouts,and hydraulic frac- Roughly half (11 of 25) of the measuredstrike-slip faults turing. strikeN28ø-45øE, subparallel with the predominantstrike di- The goodnessof fit of the normal-faultingsolution with rection of the dip-slip faults in the data set. These faults a •= 0.5 can be evaluatedfrom the distributionof deviation are discussed in detail in the section on discordant data. anglesbetween the observedand computedrakes (Figure 5). A simplegeometric analysis of the data permitsa qualita- As noted above, the mean deviation angle for this solution tive estimateof the principalstress orientations. The predo- was 31ø; by ignoring the two inconsistentNNW trending minantN30ø-35øE strike of the dip-slipfaults and the quasi- left-lateral strike-slipfaults discussedearlier which may rep- conjugate nature of the northerly (right-lateral) and ENE resent measurementerrors, this mean angle reducesto 27ø . 94 Frizzelland Zoback: Stress Orientation, Hampel Wash Area, Nevada

N Dip Distribution

rake<45 ø

r-9

9 D•pß D•str•but' ' ion rake<45 ø

r=4 rake_>45 ø

r-8

E Rake Distribution

r -10

rake >_45 ø

r=7 90

Fig. 3. Distributionsof strike, dip, and rake for subset(n=50) of faults having known senseof displacement in Hampel Wash area, Nevada, representedon Figure ld. Frizzell and Zoback:Stress Orientation, Hampel Wash Area, Nevada 95

354ø I•T R-L 17 ø

N36øE /

ø.o

/ / /I / / '"

L-L R-L Normal, L-L ß Normal, R-L Reverse, L-L B Reverse, R-L

Fig. 4. Lower hemisphericequal-area projection of fault straie data from strike-slip (R-L right-lateral, L-L left-lateral) faults in Hampel Wash area, Nevada. See Figure 3 for rosediagrams of the samedata.

Eighteenpercent of the measuredstriae (nine of 50) exhibit 15-• orientationsdiscordant (deviate by more than 45ø ) with the best solution. Exclusion of the discordant data (discussed below) yieldsa smallermean deviation angle of 22ø, indicat- •10 ing that the solution representsa good fit to most of the I.l.J ' data.

DISCORDANT FAULT DATA

As anticipated,the shallow-rakefaults that strike between N32ø-45øEexhibit slip directionscompletely discordant (predicted slip directionsthat differ by 70ø or more from ob- l serveddirection) with the solutionobtained by the inversion. 0 Io 20 ,50 4O 60 )70 This result was predictablebecause these strike-slipfaults parallelthe dip-slip faults of the data set. ANGLEOF DEVIATION Superposedsets of strike-slipand dip-slip striationswere observedon four steeplydipping fault surfacesin this strike Fig. 5. Distribution of deviation angles between rakes ob- range (Table 1), but the relative age of the striationswas served in the field and theoretical rakes determined from the not determined.Steep dips characterizethe four faults, and computer-derivedsolution. 96 Frizzell and Zoback: StressOrientation, Hampel Wash Area, Nevada

TABLE 1. SuperposedSets of Striae on Fault Surfacesin rent regional state of stressinferred from earthquakefocal Hampel Wash Area, Nevada mechanisms,well bore breakouts, and hydraulic fracturing measurements for the Nevada Test Site area (summarized Fault Strike, Dip, Rake, Deviation by Stock et al., [1985, table 3]). Becausethese indicators NøE deg. deg. angle,deg.* variouslyreflect normal and strike-slipfaulting regimeshav- ing nearly uniform least horizontal principal stressdirection

107B 14 90 17S of extension,they lend credenceto Wright's [1976] model in which the least horizontal stress orientation remains con- 107B' 14 90 48S stant and the maximum horizontal and vertical stress exchangemagnitudes to explain the coeval existenceof normal 174B 36 78W 90 11 and strike-slipfaulting in the region of the southernBasin 174B' 36 78W 30N 71 and Range. Such mixed-mode patternsof faulting have been reported 174D 32 69W 85S 7 by Angelier et al. [1985] at Hoover Dam, Nevada, 140 km 174D' 32 69W 16N 72 SE of Hampel Wash. Fault data at Hoover Dam have been interpretedto indicate two late Cenozoic (about 12 to 5 Ma) deformational events, with a 55ø clockwise rotation of S3 176A 23 90 22S between the two deformational events. Both of their defor- 176A' 23 90 90 mational events involved a combinationof normal dip-slip and strike-slipfaulting as indicatedby a bimodal distribution Senseof offset determinedonly for faults 174B and of rakes. Angelier et al. interpretedthese data as indicating 174D (see text and Figure 6). that the maximum horizontal and the vertical stress alternate *Angle betweenmeasured rakes and thosetheoretically without affectingthe orientationof S3 [Angelier et al., 1985, determined from solution. p. 361]. Our data set also exhibitsa nearly bimodaldistribu- tion of rakes. However, the solution obtained on our data set indicatesthat both the normal dip-slip faulting and slip very steep rakes (about 85ø ) representthe dip-slip motion on three of the four. The slip sensewas determinedon two fault surfaces(Table 1 and Figure 6, 174B and 174D) located along the same N30ø-35øE-strikingfault zone in Hampel Wash. Both exhibit normal right-lateral oblique sensesof displacement.Although the steeporientations of striaemay, in other circumstances,represent local "block settling"rather than a direct responseto regional tectonic stress,this does not appear to be the case here because,given N30ø-35øE the fault orientation, these steep rakes fit the solution discussed above. A second subset of the data (10 of 50) yielded angles of 30ø to 40ø betweenthe observedand predictedslip directions. Thesefaults includeN to NNE and ENE strikingfaults havingsteep rakes (greater than 80ø ) plus additionalshallow- rake faults striking N17ø-44øE. The solutionthat best fits the bulk of the data (S3, N60øW; S•, approximatelyvertical; and •= 0.5) predictsthat faultshaving strikes of N20ø-45øE would be expectedto show primarily dip-slip displacements, whereas steep faults striking approximatelyN-S should be right lateral, and ENE faults shouldbe primarily left lateral. Both the measureddip-slip displacementson NNE and ENE faults and, in particular, the strike-slip displacementson faults striking N20ø-45øE very likely would not have oc- curred under the inferred stressregime and suggesteither temporalvariation in the regional stressfield or local block rotations. Such variations are also indicated by the super- Fig. 6. Representationof two fault surfacesexhibiting super- poseddip-slip and strike-slipstriae described above. posed sets of striae (see Table 1). Striae 174B and 174D fit the solution (least horizontal principal stressorientation DISCUSSION of N60øW), and striae 174B' and 174D' are discordant.For comparison,striation 187 (N34øE strike, 76øW dip, 48øS rake) representsa striation on a third fault surfacethat de- The orientationof the stressaxes computedfrom fault slip viates 38ø from the orientationtheoretically predicted from data collectedin the Hampel Wash area agreeswith the cur- the solution. Frizzell and Zoback:Stress Orientation, Hampel Wash Area, Nevada 97 on the strike-slip faults of the "quasi-conjugate"N-to-NNE dated on smaller right-lateral faults which would rotate, and ENE-to-E setsgenerally agree with the normal-faultsol- along with the blocks they bound, from the orientationin ution with a •= 0.47. The deviationangles between actual which they developed. and predictedslip for thesequasi-conjugate strike-slip faults As shown on Figure 4, most of the right-lateralfaults of varied between 0ø and 40ø and had a mean value of 17ø, the data set strike approximatelyN-S (as expected in the a value somewhat lower than the overall misfit of the entire modem stressregime) or are rotatedclockwise from this pos- data set. The deviation angle for the dip-slip subsetof the ition (to the NE); furthermore, the left-lateral faults appear entire data set ranged from 1ø to 40ø with a mean value to have rotated counterclockwisefrom their expectedENE of 16 ø. direction. Thus, in a field example very similar to the con- Thus the mixed strike-slipand dip-slip faulting in Hampel ceptualmodel of Ron et al. [ 1984], the resultsappear incom- Wash may be explainedsimply in termsof a singledeforma- patible with the model. A detailed paleomagneticstudy has tional event in a normal faulting stress regime with a not been undertaken in this area and would be of considera- •=• 0.5 (the maximum horizontal stress intermediate in ble interest. magnitudebetween S3 and the vertical stress).Our limited The final explanationof the inconsistentright-lateral and data set does not require temporal variationsof the relative left-lateraloffsets on NE strikingfaults, as being due to local magnitude of the maximum horizontal and vertical stress stressreorientations, is always a potential sourceof noise suggestedby Wright [ 1976] and Angelier et al[ 1985]. in fault slip studies[e.g., Angelier, 1979, p. T20]. However, Predictably, the shallow-rake faults which strike about one would expect such noise to produce random rotations N20ø-45øEcompose the largestsubset of data incompatible of the faults rather than the observed clockwise rotation of with the solution becausethey strike within 10ø to 15ø of right-lateral faults and counterclockwiserotation of left-lat- the computed maximum horizontal stress direction and the eral faults. preferred orientationof primarily dip-slip faults. Slip on these shallow rake faults is not only incompatiblewith the SUMMARY normal faulting stressregime having an S3 of N60øW but also incompatiblewith simplepermutations of the maximum Fault slip data collected in a seismicallyactive part of horizontal and the vertical stress because the strike of these southernNevada in an area boundedby relatively youngleft- strike-slipfaults is so closeto the maximumhorizontal stress lateral faults exhibit a bimodal distribution of rakes that indi- direction. catesnearly pure strike-slipor pure normal dip-slip faulting. An inconsistentsense of displacementof the shallow-rake Normal dip-slip faults cluster about a N30ø-35øE strike; faults strikingN20ø-45øE further hampers the understanding whereas shallow-rake faults strike about a dominant node of their significance:both right- and left-lateral offsets are of N30ø-45øE (which includesboth right-lateraland left-lat- observed.These apparentlycontradictory offsets may be at- eral offset) and subordinatequasi-conjugate nodes of N66ø- tributableto (1) rotationof the regionalprincipal stressfield, 80øE (left-lateral offset) and N-to-NNE (right-lateralfaults). (2) tectonicrotation about a verticalaxis of blockscontaining An iterative inversion of the data indicatesa best fitting de- the faults, or (3) second-orderlocal stressreorientations. viatoricstress tensor characterized by a normalfaulting stress As discussed in the introduction, several workers have regime (S• vertical) havingprincipal stress axes in approxi- suggesteda late Cenozoic rotation or changein orientation mately horizontaland vertical planes and a least horizontal of the regional stressfield in the southernNevada area (An- principalstress (S•) orientationof N60øW. An intermediate derson and Ekren [1977]; Ander [1984]; see also Zoback value for •= 0.5 indicates that the magnitude of the et al. [1981]). However, all theseauthors have indicatedonly maximum horizontal stresslies approximatelyhalfway be- a clockwise change in rotation. The observationof both tween the magnitude of S• and S•. The S• orientation is right-lateraland left-lateraloffset on the N30ø-45ø strike-slip consistentwith the inferred modem least horizontalprincipal faults appearsto rule out a simpleclockwise change in stress stressorientation in the vicinity of the Nevada Test Site of orientation. N50ø-70øW previously obtained from earthquake focal Paleomagneticstudies in southernNevada have revealed mechanisms,well bore breakouts, and hydraulic fracturing several examples of the secondexplanation, block rotation measurements.The intermediate value of • indicates that about a vertical axis. At Yucca Mountain 35 km NW of both the quasi-conjugatestrike slip and normalfaulting may HampelWash, Scottand Rosenbaum[1986] reporta 13 m.y. be compatiblewith deformationin a normal faulting stress old tuff unit displayingprogressive clockwise rotation from regime. The maximum horizontal and vertical stressneed north to south with the southern end rotated 30 ø relative to not exchangemagnitudes as suggestedby Wright [ 1976] and the northern end. Within the Lake Mead shear zone near Angelier et al. [1985]. Hoover Dam, Ron et al., [1986] reported 27 degreesof The solution fits all the data well except for strike-slip counterclockwise rotation in 11-12 Ma volcanic rocks. Ron faults that strike N30ø-45øE, subparallelto the normal faults et al. [1984] have proposeda model for local block rotation of the data set. The strike-slipoffsets observed on theseNE resultingfrom strike-slipfaulting; their model predictsthat striking faults, as well as superposedsets of striaeobserved internalrotation (accommodated on a conjugateset of strike- on somefaults, suggesteither rotationsof the regionalstress slip faults) occurs within regionsbounded by major strike- field or local block rotations within fault zones. slip faults. Applying this model to the Hampel Wash area which is boundedon the north and southby left-lateralob- Aknowledgments.Jim Yount shared his understandingof lique faults, one would expect an internal counterclockwise the Rock Valley fault and Mark Gorden helped in data col- rotation. This counterclockwise rotation would be accommo- lection. This article benefited from helpful and sometimes 98 Frizzell and Zoback: StressOrientation, Hampel Wash Area, Nevada

challengingcomments by Ernie Anderson,Doc Bonilla, Ben Nye and Esmeralda Counties, Nevada, Mem. Geol. Soc. Page, Ze'ev Reches, Hagai Ron, JoanneStock, and George Am. 10, 251-262, 1968. Thompson. Rogers, A.M., S. C. Harmsen, W. J. Carr, and W. Spence, Southern Great Basin seismologicaldata report for 1981 REFERENCES and preliminarydata analysis,U.S. Geol. Surv. Open-File Rep. 83-669, 1-240, 1983. Ander, H. D., Rotation of late Cenozoic extensional stresses, Ron, H., R. Freund, Z. Gaffunkel, and A. Nur, Block rota- Yucca Flat region, Nevada Test Site, Nevada, Ph.D. tion by strike-slip faulting: Structuraland paleomagnetic thesis,77 pp., Rice Univ., Houston,Tex., 1984. evidence:J. Geophys.Res., 89, 6256-6270, 1984. Ander, H. D., F. M. Byers, Jr., and P. P. Orkild, Geology Ron, H., A. Aydin, and A. Nur, Strike-slip faulting and of the Nevada Test Site (Field Trip 10), in Western block rotation in the Lake Mead fault system, Geology, Geological Excursions, vol. 2, edited by Joseph Lintz, 4, 1020-1023, 1986. Jr., pp. 1-35, Departmentof GeologicalSciences, Mackay Scott, R. B. and J. G. Rosenbaum, Evidence of rotation School of Mines, Reno, Nev., 1984. about a vertical axis during extensionat Yucca Mountain, Anderson, R. E., and E. B. Ekren, Late Cenozoicfault pat- Southern Nevada, Eos, Trans. AGU, 67, 358, 1986. terns and stress fields in the Great Basin and westward Springer, J. E., R. K. Thorpe, and H. L. McKague, displacementof the Sierra Nevada block: Comment, Geol- Borehole elongationand its relation to tectonic stressat ogy, 5,388-389, 1977. the Nevada Test Site, Lawrence Livermore Lab. Rep., Angelier, J., Determinationof the mean principaldirections UCRL-53528, 43 pp., 1984. of stressfor a given population,Tectonophysics, 56, T17- Stewart, J. H., and J. E. Carlson, Geologic map of Nevada, T26, 1979. scale 1:500,000, U.S. Geol. Surv., Reston, Va., 1978. Angelier, J., Tectonic analysis of fault slip data sets, J. Stock, J. M., J. H. Healy, S. H. Hickman, and M.D. Geophys.Res., 89, 5835-5848, 1984. Zoback, Hydraulic fracturing stress measurementsat Angelier, J., B. Colletta, and R. E. Anderson, Neogene Yucca Mountain, Nevada, and relationshipto the regional paleostresschanges in the Basin and Range: A case study stressfield: J. Geophys.Res., 90, 8691-8706, 1985. at Hoover Dam, Nevada-Arizona, Geol. Soc. Am. Bull, Vetter, U. R., and A. S. Ryall, Systematicchange of focal 96, 347-361, 1985. mechanism with depth in the western Great Basin, J. Barnes, H., E. B. Ekren, C. L. Rodgers, and D.C. Hed- Geophys.Res., 88, 8237-8250, 1983. lund, Geologic and tectonic maps of the Mercury quad- Wright, L., Late Cenozoic fault patternsand stressfields rangle, Nye and Clark Counties, Nevada, scale 1:24,000, in the Great Basin and westward displacementof the U.S. Geol. Surv. Misc. Invest. Map Ser., Map 1-1197, Sierra Nevada block, Geology, 4, 489-494, 1976. 1982. Yount, J. C., R. R. Shroba, C. R. McMasters, H. E. Huc- Burchfiel, B.C., Precambrianand Paleozoicstratigraphy of kins, and E. A. Rodriguez, Trench logs from a strand Spector Range quadrangle, Nye County, Nevada, Am. of the Rock Valley fault, Nevada Test Site, Nye County, Assoc. Pet. Geol. Bull., 48, 40-56, 1964. Nevada, scale 1:10, U.S. Geol. Surv. Misc. Field Studies Byers, F. M., Jr., W. J. Carr, P. P. Orkild, W. D., Quinli- Map MF-1824, 1987. van, and K. A. Sargent, Volcanic suites and related caul- Zoback, M. L., and M.D. Zoback, State of stress in the drons of Timber Mountain-OasisValley Caldera complex, conterminousUnited States:J. Geophys.Res., 85, 6113- southernNevada: U.S. Geol. Surv. Prof. Pap. 919, 1-70, 6156, 1980. 1976. Zoback, M. L., R. E. Anderson, and G. A. Thompson, Carr, W. J., Summary of tectonic and structuralevidence Cainozoic evolution of the state of stressand style of tec- for stress orientation at the Nevada Test Site: U.S. Geol. tonism of the Basin and Range province of the western Surv. Open-File Rep. 74-176, 1-53, 1974. United States: Philos. Trans. R. Soc. London, Ser. A, Carr, W. J., Regional structuralsetting of Yucca Mountain, 300, 407-434, 1981. southwestern Nevada, and late Cenozoic rates of tectonic activity in part of the southwesternGreat Basin, Nevada and California: U.S. Geol. Surv. Open-File Rep., 84-854, V. A. Frizzell, Jr., Western Regional Geology, U.S. 1-109, 1984. GeologicalSurvey, 2255 North Gemini Drive, Flagstaff,AZ Frizzell, V. A., and M. L. Zoback, Striae on faults in Ham- 86001. pel Wash, southernNevada, indicate least horizontalprin- M. L. Zoback, Seismology, U.S. Geological Survey, cipal stressdirection of N60øW: Eos Trans. AGU, 66, Menlo Park, 94025. 1056, 1985. Hinrichs, E. N., Geologic map of the Camp Desert Rock quadrangle, Nye County, Nevada, 1:24,000 scale, U.S. (ReceivedMay 22, 1986; Geol. Surv. Geol. Quadrangle Map, GQ-726, 1968. revised December 8, 1986; Kistler, R. W., Potassium-argonages of volcanic rocks in acceptedDecember 12, 1986)