Structure and Late Quaternary Activity of the Northern Owens Valley Fault Zone, Owens Valley, California

Home , Owens Valley, Sierra Nevada Fault

Engineering Geology, 27 (1989) 489-507 489 Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands

STRUCTURE AND LATE QUATERNARY ACTIVITY OF THE NORTHERN OWENS VALLEY FAULT ZONE, OWENS VALLEY, CALIFORNIA

STEPHEN J. MARTEL*’ Bureau of Economic Geology, The University of Texas at Austin, University Station, Box X, Austin, TX 78713-7508 (U.S.A.)

(Accepted for publication November 4, 1988)

ABSTRACT

Martel, S.J., 1989. Structure and late Quaternary activity of the northern Owens Valley fault zone, Owens Valley, California. In: A.M. Johnson, C.W. Burnham, C.R. Allen and W. Muehlberger (Editors), Richard H. Jahns Memorial Volume. Eng. Geol., 27: 4899507.

The Fish Springs fault is a primary strand in the northern end of the Owens Valley fault zone (OVFZ). The Fish Springs fault is the northwest strand in a g-km-wide left echelon step of the OVFZ which bounds the Poverty Hills bedrock high. The Fish Springs fault strikes approximately north-south, dips steeply to the east, and is marked by a prominent east-facing scarp. No other faults in the OVFZ have prominent east-facing scarps at the latitude of Fish Springs, which indicates that the Fish Springs fault has accommodated virtually all of the local late Quaternary vertical displacement on the OVFZ. The Fish Springs fault exhibits normal dip slip with no measurable lateral slip. Vertical displacements of a Late Pleistocene (0.314 kO.036 Ma, 20) cinder cone and of an overlying Tahoe- age (0.065-0.195 m.y.) alluvial fan are 76k8 m and 31+3 m, respectively. The maximum vertical displacement of ancient stream channels in a Tioga-age (0.011 to 0.026 Ma) alluvial fan averages 3.3 m. Two nearly equal vertical displacements of the active stream channel in the Tioga-age fan total 2.2 m. Vertical displacement of a stream terrace incised into the cinder cone is 1.2kO.3 m. The minute amount of incision into that terrace indicates that uplift of the terrace probably occurred during the 1872 Owens Valley earthquake. Three displacements of l.lkO.2 m each apparently have occurred at the Tioga-age fan since the midpoint of the Tioga interval, allowing an average recurrence interval of 3500 to 9090 years. Based on the age and displacement of the cinder cone, the average late Quaternary vertical displacement rate is 0.24f0.04 mm/yr (20). At this rate, and assuming an average vertical displacement of 1.1 kO.2 m per event, the average recurrence interval would be 46OOk 1100 years (2~). The recurrence interval for the Fish Springs fault is similar to that for a strand in the southern part of the OVFZ which also ruptured in 1872. Right-lateral, normal oblique slip characterizes the OVFZ. The location of the Poverty Hills bedrock high at a left step in the north-northwest-striking fault zone is consistent with the style of slip of the zone. The pure normal slip on the north-striking Fish Springs fault and the alignment of local cinder cones along north-striking normal faults indicate that the late Quaternary maximum horizontal compression has been oriented north-south at the north end of the OVFZ. Data from southern Owens Valley indicate a similar stress regime there. Late Quaternary slip on the OVFZ is consistent with north-south maximum horizontal compression.

*‘Present address: Division of Earth Sciences, Bldg. 5OE, Lawrence Berkeley Laboratory, University of California, Berkeley, Calif., U.S.A.

0013-7952/89/$03.50 0 1989 Elsevier Science Publishers B.V. 490

INTRODUCTION

The Owens Valley fault zone of east-central California (Fig.l) generated one of California's greatest historical earthquakes (Townley and Allen, 1939; Richter, 1958) and thus poses a significant seismic hazard. However, the structure and behavior of the fault zone have received relatively little attention. The purpose of this study is to augment our understanding of the Owens Valley fault zone by a detailed investigation of the structure and late Quaternary activity near its northern end. The Owens Valley fault zone was first mapped by Knopf (1918). It has been mapped more recently by Bateman (1965), Ross (1965), Nelson (1966), and D.B. Slemmons (Hill, 1972). The displacement along the fault zone and its structure were studied by geophysical means by Pakiser et al. (1964). Most of the geologic studies on the fault zone have focused on effects of the

-Jo

OWENS VALLEY / FAULT ZONE

:.. /" S I E RRA.:~'NEVAOA FAULT ~i....~iZONE

I00 g IO0 KILOMETERS

Fig.1. Index map of the Owens Valley fault zone. 491 great Owens Valley earthquake of 1872. The earliest studies were by Gilbert (1884), Whitney (1888) and W.D. Johnson (Hobbs, 1910). More recent work on 1872 displacement includes that of Bateman (1961) and Bonilla (1968). The prehis- toric behavior of the fault zone has been investigated by Carver (1970), Richardson (1970), Slemmons (Oakeshott et al., 1972) and Lubetkin (1980). Except for Whitney, these workers concentrated on the southern Owens Valley. Along the little-studied northern end of the fault zone, where the oldest exposed Quaternary features are offset, the Quaternary displacement history can be analyzed in detail. The strand of the fault zone examined herein is referred to as the Fish Springs fault. The Fish Springs study area, centered around a prominent basaltic cinder cone, is about 11 km south of Big Pine, California.

STRUCTURAL GEOLOGY OF WESTERN OWENS VALLEY

Owens Valley is a complexly faulted block down-dropped between the Sierra Nevada and the White-Inyo Mountains (Knopf, 1918; Pakiser et al., 1964). The three major structures of western Owens Valley are the Owens Valley and Sierra Nevada fault zones, and the Coyote Warp (Fig.2). The Owens Valley fault zone branches to the east off the Sierra Nevada fault zone about 3 km south of the Alabama Hills. Both fault zones are well defined between the latitudes of Lone Pine and Big Pine. The Coyote Warp, a large flexure, is between Big Pine and Bishop, and lacks major through-going faults (Bateman, 1965). The Warp is bounded on the southeast by the Owens Valley fault zone and on the northwest by the Round Valley fault.

The Owens Valley fault zone

The Owens Valley fault zone is approximately 120 km long and extends from the west side of Owens Lake to a few kilometers north of Big Pine. The maximum width of the fault zone is approximately 3 km. The zone has an overall strike of N17°W and typically dips 60 ° to 70 ° to the east (Tocher et al., 1963). West-side-up, dip-slip displacement along the fault zone is indicated by historical accounts, by east-facing scarps in Quaternary alluvium and lavas, and by displacement of pre-Cenozoic bedrock. From gravity and seismic data, Pakiser et al. (1964) infer vertical displacements of basement of 2400 _ 600 m at Lone Pine, 300 m at Independence, and 1800 ± 450 m at Big Pine. Assuming that vertical displacement rate at Big Pine and Lone Pine would be approximately the data of Giovanetti (1979) and Bacon et al. (1982), the average late Cenozoic vertical displacement rate of Big Pine and Lone Pine would be approximately 0.3 mm/yr and 0.4 mm/yr, respectively. Although left-lateral displacement might have occurred locally during the 1872 earthquake (Whitney, 1888; Gianella, 1959; Pakiser et al., 1964), the evidence for overall right-lateral slip across the fault zone is compelling. The only p,ublished photographs of 1872 strike-slip offset show right-lateral offset (Bateman, 1961) as do the only maps (Bateman, 1961; Lubetkin, 1980). Geodetic 492

I ll8 ° "•'•,.ROUND VALLEY '~'-S- u~ ;FAULT ZONE ~.~

~ / I~ .... ~\~Bishop ~,

\ 0", COYOTE ~. \\ "~ WARP/ / ~J~ i / I\ /MAIN TRACE OF OWENS FAULT ZONE

I Big Pine CRATER MTN t--~,,,~ F IS H SPRINGS FAULT / " TLr.~FISH SPRINGS CINDER CONE POVERTY HILLS // '(~:: RED MTN~ "~-,~ i .~. o_

/ MAIN TRACE OF ( OWENS VALLEY FAULT ZONE Independence # ~,

EXPLANATION z ~,, ALABAMA FAULT ZONE t HILL ~---,~( ,,.n Dotted where interred. -~ I

.... " \ Lone Pine CONTACT 0 ~ Dashed where approximately located. --- -~- ----> ~, ~\ , COYOTE WARP (ANTIFORM) SIERRA "", Approximately located ; NEVADA ~ LAKE ', showing direction of plunge. FAULT ZONE ~ ,," I CINDER CONE OF THE = . ..'" BIG PINE VOLCANIC FIELD ~ ,) i

20 0 20 KILOMETERS 118° I Fig.2. Major geologic structures of the western part of Owens Valley.

surveys across southern Owens Valley have indicated right-lateral deformation at rates of 3.6 + 1.2 mm/yr (Savage et al., 1975) and 2.4 + 0.9 mm/yr (Savage and Lisowski, 1980); the uncertainties are one standard deviation. Finally, right-lateral solutions have been obtained for microearthquakes in northern Owens Valley on northwest-trending faults (Pitt and Steeples, 1975). 493

Structure of the northern end of the Owens Valley fault zone

North of the Fish Springs fault, a series of conspicuous scarps in Crater Mountain clearly marks the main trace of the Owens Valley fault zone. This northernmost section of the zone strikes N19°W. However, no fault traces of that orientation are obvious between Poverty Hills and Crater Mountain. On the basis of gravity data, Pakiser et al. (1964) interpreted the fault zone as extending beneath Poverty Hills on a linear N19°W trend (Fig.3). They

I t

t MAIN TRACEOF'"~'l, Big Pine N OWENS VALLEY ~W 1 FAULT ZONE ~&~- I

; POVERTYHILLS FISH ~~,.,,.~

EXPLANATION

[_~ BASALT MAIN TRACE OF ~/OWENS VALLEY [] GRANITICROCK ~FAULT ZONE

FAULTWITH SCARP '* Dashedwhere approximately located, \ dotted where inferred. Ballon downthrown side.

CONTACT 5 O 5 KILOMETERS

Independence

Fig.3. Two interpretations of the structure of the Owens Valley fault zone near the Poverty Hills. A = this study; B = from Pakiser et al. (1964). 494

contended that Poverty Hills represents a gravity slide mass, derived from the Sierra Nevada, that overlies the Owens Valley fault zone. However, this interpretation seems incompatible with the geology of Poverty Hills. If Poverty Hills were a gravity slide mass, it would have to predate the 1.0 Ma rhyolite dome (Moore and Dodge, 1980) between Poverty Hills and the Sierra Nevada. Given over one million years, a '~buried" Owens Valley fault zone should have had sufficient time to rupture the Poverty Hills mass, yet no through-going faults have been mapped in Poverty Hills (Bateman, 1965). Only if the trace of the fault zone lies east of Poverty Hills could west-side-up displacement on the fault zone bring plutonic rock to the surface (Fig.3). Three short scarps just east of Red Mountain (Fig.3) indicate that the trace indeed does deflect to the east side of the hills. This interpretation is compatible with the geophysical data of Pakiser et al. (1964), the geology of Poverty Hills, and the character of displacement on the Owens Valley fault zone. If the trace of the Owens Valley fault zone lies east of the Poverty Hills, then a geometric discontinuity in the fault zone must occur north of Poverty Hills, for no fault trace links the northeast corner of the hills with the fault zone trace on Crater Mountain. The Fish Springs fault is the only west-side-up fault with a prominent scarp in the alluvium between Poverty Hills and Crater Mountain, and it is most likely the primary element in the Owens valley fault zone between those two topographic highs. The depth to which the discontinuity in the fault zone extends is unknown. Below some depth the fault zone might be a continuous structure. The decrease of 2100 m in bedrock displacement across the fault zone between Lone Pine and Independence might be due to the discontinuity in the fault zone at Poverty Hills. Under this interpretation the Poverty Hills bedrock high is at a left step in a right-lateral, normal oblique shear system. The Coyote Warp is at an analogous, but much larger left step between the Owens Valley fault zone and the Round Valley fault. Theoretical work (Segall and Pollard, 1980) and field studies elsewhere (Clayton, 1966; Sharp and Clark, 1972) demonstrate that areas between left-stepping echelon fault strands in right-lateral shear systems are likely sites for uplift.

LATE QUATERNARY DISPLACEMENT ON THE FISH SPRINGS FAULT

Five major geomorphic features of late Quaternary age dominate the Fish Springs study area (Fig.4). These are: (1) the compound east-facing scarp; (2) the Fish Springs cinder cone; (3) an alluvial fan exposed west of the fault, hereafter referred to as the west fan; (4) an alluvial fan east of the fault and north of the cinder cone, hereafter referred to as the north fan; (5) an alluvial fan east of the fault and south of the cinder cone, hereafter referred to as the south fan. These features were mapped at scales of 1:2000 and 1:500 by plane table and at a scale of about 1:2000 on aerial photographs. The cinder cone is the oldest feature. The fault trace splits the cone asymmetrically, the east segment being larger than the west segment. The 495

/| 1 EXPLANATION Qwf i" I ",..,,,, / ~Ir~ FISH SPRINGS FAULT ~" t Qnf / ./ Qnf I o/ SLOPE WASH AND TALUS ...... _..... NORTH FAN-- NORTH LOBE ;" • ~°°°°

Qwf tI"* "" ...... SMALL ALLUVIAL FAN WEST FAN Qsw / ,~ I~ Qnf "'.. QfsbA~ ' NORTH FAN-- SOUTH LOBE "", Qwf (?) NORTH ALLUVIAL FAN /I i,,k, .... " ~Qaf ~ ~.~~'... SOUTH ALLUVIAL FAN

WEST ALLUVIAL FAN SPRINGS N POTENTIAL EXPOSURES EAST OF CINDER~ FISH SPRINGS FAULT ARE QUERRIED ~,Qf'sbll FISH SPRINGS CINDER CONE

Qsw L.. BASALT OF THE FISH SPRINGS 5 CINDER CONE

Qwf ll]_no. ~ "- I jl BASALT NOT PART OF THE FISH SPRINGS CINDER CONE

MAP SYMBOLS

- - - ~ZSG ...... APPROXIMATELY LOCATED ; ///~ 200 METERS DOTTED WHERE CONCEALED.

,'~1 3af / (sc ...... ) CONTACT DASHED WHERE APPROXIMATELY LOCATED; DOTTED WHERE CONCEALED

SCARP PROBABLY OF LANDSLIDE ORIGIN Fig.4. Geologic map of the Fish Springs study area. Modified from Martel (1984). west segment has been uplifted with respect to the east segment. The west fan partly buries the cone and is also offset by the fault. The north and south fans are forming east of the scarp and bury the down-dropped west fan. The north fan is offset by the fault. A small stream terrace incised into the cinder cone is also offset. The surface of the south fan apparently is not offset.

The Fish Springs fault scarp

This scarp is tallest in the west segment of the cinder cone. The alluvial scarp is tallest immediately north and south of the cinder cone. The height of the alluvial scarp decreases from the cinder cone towards the apices of the north and south fans. The scarp height increases north of the north fan, but decreases south of the south fan. The scarp height variation north of the cinder cone primarily reflects the burial of the scarp by the north fan. The decrease in scarp height south of the cinder cone probably is due to a combination of increasing burial by the south fan and decreasing displacement near the south end of the fault. 496

The fault trace was located on the alluvial scarp at the contact between abundant in-place boulders upslope of the trace and abundant translocated cobbles downslope of the trace (Fig.5). The contact apparently marks the top of a debris slope at the base of the scarp and should coincide with the fault trace (Fig.6). The slope above the boulder/cobble contact is generally a few degrees steeper than the slope below it. Wallace (1977) noted similar effects at the top of debris slopes on fault scarps in Nevada. At several places south of the cinder

Fig.5. View to west of fault trace at the cobble/boulder contact. Field assistant is standing on the contact.

Area of erosion _ _~~_ _a_attopof faullscarp

Area of debris accumulation fault scarp

Fig.6. Profile of an idealized compound fault scarp showing the location of the fault trace. 497 cone the boulder/cobble contact corresponds with the base of a band of shrubs• At several places on the scarp north of the cinder cone, the soil is redder below the boulder/cobble contact than above it. The bases of fresh scarps on the compound scarp also coincide with the top of the debris slope• Across the fault trace the cinder cone is obscured. A scarp across the west segment of the cinder cone (Fig.7) probably is not a fault scarp, but rather a landslide scarp, formed as the slope was oversteepened by faulting• If that scarp marked the fault trace, then, based on the solution of a three-point problem, the fault dips 59 ° to the east. This is the minimum dip for the fault, for the cinders

~f \Z~ ~ EXPLANATION

...... ~j~Kf ApproximofelyFAULT located i I Dotted where concealed.

SCARP Probably of landslide origin.

~J I00 0 I00 METERS <-. \~, 'co.To,.,,,,'--~.,;v.L ,O,,,T~,,~' , \/

.tO

I I I I A'

-- ....

I I S j I

Fig.7. Topographic map of a section of the Fish Spring fault scarp showing the locations of the cross-sections of Fig.8. 498

A A' G G' Qwf .--, [370 L13;0 .... ~ oo, :~x-- ...... _[,36o (g) --~-'- 3.5rn

13551 _ \ _ 1-1355 B H H'

1365 -~---~_L ..... 1565 .... % .... ~===:::-4.o 3.5m (b),~5o .,, ,_ ~28m.... - ..... 1350 ,380 ~ ~.~,,- ",360 (h) ~ Ont ,3554 ___~ 2.51'5 1-1355 C C' I ['

1370 --__ ~370 (C) I360 ...... Qfsb x~'~ 30m 1560 ,366 -- -~ .... z.Bm ,388 (i) ...... Z 1362 Onf "~\,~__ -4_ -_ 1362 ~340 1340 ~um 9*f~ .... :J/ 1358 ~ ~<~ 1358 D D d J"

1366 ...... ~ F1366 (d) 1350 ...... 1350 ...... q) 1362 ' 8m " Qnf \ ~.~.-i~ [382 1358 !358 K K

..... ~L---- ,~6o (k) ~358 on~ ~ - - ..... 1358 q ...... - >'-I.0m 1355"L_ ~'~--~IOm .... ~~ Qnf~. -~r 1355 1354 Q m J F F' L l" 1365 ~ --2.3~ 2.5 m 1365

1333 'i~- o.,r '35~ (I) 1335 ~ \ ~355 o°, ~,~ .... ~ \\ [ .... 1350 \\~ ~ 1350

Fig.8. Cross-sections of features offset by the Fish Springs fault. See Fig.7 for explanation of symbols.

west of this scarp show no evidence of being faulted. However, as shown in Fig.8a, if the fault dip is 59 °, the amount of cinder talus immediately east of the fault would be roughly five times the amount available from the source area west of the fault. If the fault is vertical, the amount of cinder talus nearly equals the amount available from the source area. Thus, the fault dips between 59 ° and 90 ° and is probably close to vertical.

The Fish Springs cinder cone

The cinder cone records the greatest displacement on the fault. A cross- section perpendicular to the fault (Fig.7) indicates that the fault has vertically separated the cinder cone by 82_+ 2 m (Fig.8a). The uncertainty represents the range of possible separation values and should be interpreted as approximately two standard deviations. The minimum vertical displacement is 68 m (assuming a 59 ° fault dip), and the maximum vertical displacement is 84 m (assuming a 90 ° fault dip). The vertical displacement of the cinder cone is thus 76_ 8 m. In order to estimate the lateral offset of the cinder cone, the position of the 499 contact between the cinder cone and west fan was compared to the theoretical contact for a cone offset vertically but not laterally. The contact between the cinder cone and the overlying fan is exposed in the fault scarp. Where well- exposed, the contact marks a sharp transition from abundant cinders, covered by sparse granitic boulders, to abundant granitic boulders. In some places, however, gravels sloughed off the west fan have obscured the contact. In these places, the contact was mapped at the highest position on the slope of orange-brown vesicular cinders, but the position of the contact could lie slightly higher. The theoretical contact was found by displacing a model of the original surface of the cinder cone to yield a vertical separation of 82__+ 2 m, and then intersecting that surface with the mapped topography. The model of the original shape of the west half of the cinder cone was made by extrapolat- ing the western surface of the east segment of the cinder cone. Theoretical contacts were constructed for fault dips of 59 °, 70 ° , and 90 ° . The sections of the theoretical contacts that are the most important for assessing the amount of lateral slip, the northernmost and southernmost sections, were identical for the three different dips. The mapped and theoretical contacts between the cinder cone and overlying gravels are compared in Fig.9. The location of the southern part of the theoretical contact relative to the mapped contact indicates that no right- lateral offset has occurred, but does not preclude a small amount of left-lateral offset. Similarly, the northern portions of the cone/fan contact indicate that no left-lateral offset has occurred, although a small amount of right-lateral displacement cannot be ruled out. The simplest explanation is that no lateral offset has occurred, although a few meters of offset might go undetected. Imperfections in the fit of the modeled contact to the mapped contact are attributed to the difficulties in mapping the exact location of the cone/fan contact, to differences in shape between the model and the actual cinder cone, and to variations in vertical displacement along the strike of the fault.

The west fan

The west fan is characterized by a weathered surface that slopes gently to the east. The surface displays reddish brown soils, moderately- to highly- weathered granitic boulders, and is armored by platy oxidized metamorphic pebbles and cobbles. The west fan is either exposed or buried very shallowly east of the tallest parts of the alluvial scarp. South of the cinder cone is a flat, slightly dissected surface (Qwf(?) on Fig.4) that might be a downdropped remnant of the west fan. This surface is characterized by patches of oxidized metamorphic pebbles, a dark tan soil, and partly weathered boulders, as is the west fan. This surface is distinctly different from the light-colored sandy surface of the adjacent south fan. The flat surface does not resemble a small alluvial fan constructed at the base of the scarp because it lacks the convex-up cross profile of fans and its clasts do not decrease in size downslope. Immediately north of the cinder cone 500

EXPLANATION I00 0 I00 METERS i i Contour intervo//O meters CONTACT Location accurate to within 2 meters.

.... o**o FAULT 1 \ .~DINDERS Approximately located; N PARTIALLY Dotted where concealed. BURIED BY -- IOO-- FANGLOMERATE \ . CONTOUR FANGLOMERATE ) ri FISH / Relative elevation I SPRINGS ~(3 THEORETICAL /'? ~FAULT /'-" CONTACT, if NO LATERAL /; (

,. / / tA

70 [ JT oNE

\ \ -TJI \ 'Y F,N,LOMER,T

Fig.9. Comparison of the cinder cone/west fan contact as mapped with a theoretical contact for the case of no lateral offset. is the southern margin of the north fall. As is discussed below, the north fan probably is less than 26,000 years old. The west fan probably is buried shallowly by the distal margin of such a young fan. Displacement of the west fan surface can be estimated from profiles across the tallest part of the alluvial scarp because the west fan is exposed west of the scarp and is at or near the surface to the east of the scarp. The vertical 501 separation of the west fan south of the cone is 28 m (Fig.8b). North of the cone the separation is slightly more than 30 m (Fig.8c). Assuming that the west fan lies no more than 2 m below the southern margin of the north fan, the vertical separation of the west fan surface is estimated as 30 + 2 m and the correspond- ing vertical displacement as 31± 3 m.

The small terrace in the west segment of the cinder cone

A small uplifted stream terrace occurs in the northern half of the west segment of the cinder cone. The scarp bounding the terrace consists of cinders, has a slope that locally exceeds 80 °, and is not faceted. The lack of facets indicates that the terrace probably was offset during a single event. The vertical separation and displacement on the terrace are 1.2+0.3 m (Fig.8d). Post-uplift incision of the terrace by a 1- to 2-m-wide intermittent stream has resulted in the knickpoint in the stream profile migrating only 5 m upstream from the fault. This erosion conceivably could have occurred in a single intense storm, and the possibility that the terrace uplift is older than 1872 seems very remote. As no significant earthquakes have occurred in the Fish Springs area since 1872, the terrace displacement most likely occurred during the 1872 earthquake.

The north fan

The complex interaction among deposition, stream migration and downcutt- ing, and faulting has resulted in the formation of two lobes on the north fan (Fig.4). The younger, southern lobe lies almost entirely east of the fault. It is characterized by bouldery ridges 1-4 m tall and numerous sandy distributary channels. The intermittent main stream of the southern lobe has been eroding the north slope of the cinder cone. The apex of this lobe is where the main stream crosses the fault. The older, abandoned northern lobe exhibits a distinct convex-up cross profile and covers about twice the area of the southern lobe. The apex of the northern lobe is on the upthrown block, 200 to 300 m west of the fault. The northern lobe is characterized by a subdued topography punctuated by a few partly buried bouldery ridges. The light colored, sandy surface of the north lobe contrasts sharply with the darker surfaces of the west fan. The maximum height of the fault scarp in the northern lobe is approximately 3.5 m. Three prominent terraces on the upthrown block border the main stream through the north fan. They record previous locations of the stream. The terraces are successively higher to the north. These surfaces show that the main stream of the north fan migrated to the south during deposition of the north fan. These terraces are not offset laterally by the fault. The north fan is interpreted to have formed in the following manner. The apex of the north lobe was initially at the fault scarp (Fig.10). Eventually, deposition on the downthrown block reached the top of the scarp, allowing the stream channel on the upthrown block to be backfilled. This accounts for the 502

WEST FAN

BIRTH OF NORTH LOBE ~ FAUU

--~GROWTH OF NORTH LOBE

BACKEILLING OF CHANNEL ON UPTNROWN ~ BLOCK

••'• ABANDONMENT " ,'/ I \ // OF NORTH LOBE / \ ~ y ANO .% CO,BTROCT,ON ~ OF SOUTH LOBE

Fig.10. Inferred development of the north fan.

presence of north lobe sediments west of the fault. Deposition on the upthrown block eventually filled the stream channel to near the level of the west fan surface. Then, either avulsion of the stream channel or stream piracy by a fault scarp gully resulted in abandonment of the apical stream channel and the start of southward channel migration. Development of the terraces was in part due to lowering of the local base level as the stream migrated towards its present position and encountered lower points on the surface of the north lobe. The scarp of the north fan probably reflects three major displacements on the Fish Springs fault. Evidence of these displacements comes from: (1) desert varnish rings on a fault scarp boulder a few meters south of the main stream channel through the north fan; (2) knickpoints in the profile of the stream channel; and (3) displacements on sandy surfaces of the north fan. Two distinct desert varnish rings on the fault scarp boulder (Fig.ll) suggest that the north fan has been displaced at least twice. This boulder is immediately upslope from the fault trace. Desert varnish rings on fault scarp boulders coincide with the levels of stable ancient ground surfaces (e.g., Lubetkin, 1980). Seismic displacements can lower the ground level around fault scarp 503

Fig.11. Desert varnish rings on a fault scarp boulder a few meters south of the main channel of the north fan. The rule is set to 30 cm. Weathering lines are near the top and bottom of the rule.

boulders, allowing varnish rings to record displacements. As measured on the downslope side of the boulder, the lower desert varnish ring is approximately 1 m above the present ground surface, suggesting that the most recent displacement of the ground surface at the boulder was about 1 m. This displacement is consistent with the 1.2 _0.3m displacement of the cinder cone's stream terrace. The higher varnish ring is approximately 0.5 m above the lower ring, suggesting that the total displacement at the boulder in the last two events has been about 1.5 m. However, this displacement is substantially less than the maximum displacement of the north fan, indicating that the maximum fan displacement represents more than two events. Knickpoints in the present main channel of the north fan suggest that the channel has been displaced twice. A profile along the channel shows two distinct knickpoints upstream of the fault trace (Fig.8e). The downstream knickpoint is at the top of the fault scarp. The displacement at this scarp is 1.2 m, an amount consistent with the displacement on the cinder cone terrace and at the fault scarp boulder. The upper knickpoint suggests a previous displacement of 1.0 m. The total displacement of 2.2 m is still less than the maximum displacement of the northern lobe, suggesting that the northern lobe has been displaced more than twice. The vertical displacement of the northern lobe can be inferred from profiles F-F through L-L' (Fig.8). On most of the profiles, the slope of the surface east of the fault parallels the slope west of the fault, whereas on a few profiles (G-G', and H-/-F) the east slope is steeper than the west slope. However, any slope angle differences are small, and the displacement can be estimated to within 15%. The profiles reveal two sets of displacements. Displacement of the terrace 504 located immediately north of the modern channel is approximately 2.5 m, similar to the channel displacement of 2.2 m. The mean displacement of sandy terraces further north (profiles G-G', H-H', I-r, J--~, and KK', Fig.8) is 3.3 m, 1½ times the channel displacement. As the modern channel apparently has been displaced twice, three displacements are indicated for the oldest parts of the north fan. The vertical displacement during each of the last three ruptures of the north fan apparently was 1.1+0.2m. The most recent displacement most likely occurred in 1872. This suggests that 1872 displacement might be representative of displacements on the Fish Springs fault.

The south fan

The south fan has been deposited by Birch Creek. It is characterized by a light colored sandy surface with concentrations of unweathered gravel along its distributaries. The surface of the south fan appears to be unbroken by the fault. If the south fan surface was offset recently, as seems likely, then the record of offset may have been erased either by Birch Creek or by the recent land development at the head of the south fan.

Ages of offset features

The cinder cone has been dated radiometrically at 0.314___ 0.036 m.y. (2a) by Martel et al. (1987). The west and north fans are considered to be of Tahoe- and Tioga-age, respectively, based on a comparison (Martel, 1984) of their weathering characteristics with other Owens Valley gravels that have been dated (Lubetkin, 1980; Gillespie, 1982). The age of the Tahoe glaciation is controver- sial and is conservatively estimated as between 0.065 and 0.195 Ma (Martel et al., 1987). The age of the Tioga glaciation is 0.011-0.026 Ma (Birkeland et al., 1971; Smith, 1979; Atwater et al., 1986; Mezger and Burbank, 1986). The three displacements of the north fan apparently occurred after the northern lobe was abandoned, and as the northern lobe forms at least two-thirds of the north fan, the displacements probably occurred during the latter half of the Tioga interval.

DISCUSSION

Sense of movement during the late Quaternary Period

The Fish Springs fault exhibits west-side-up normal slip with no discernible strike-slip displacement. The lack of measurable lateral offset for the oldest feature of the study area, the cinder cone, and some of its youngest features, the north fan stream channels, suggests that the style of displacement has been consistent throughout the latter part of the Quaternary Period. 505

Average recurrence interval

Three 1872-type displacements apparently have occurred since the latter part of the Tioga glaciation (10,600-18,000 B.P.), suggesting an average recurrence interval of 3500 to 6000 years. However, from this data, the maximum recurrence interval would be estimated more conservatively at 9000 years (had the recurrence interval been calculated in 1871 only two events would have occurred in the prior 10,500 to 17,900 years). Based on the 76+8m vertical displacement of the 0.314 ± 0.036 Ma cinder cone, the average late Quaternary displacement rate is 0.24+0.04 mm/yr, similar to the late Cenozoic displacement rate across the Owens Valley fault zone at Big Pine. Assuming that the last three displacements of 1.1 + 0.2 m are average displacements, the long-term average recurrence interval for the Fish Springs fault would be 4600 ± 1100 years (2a).

Implications of 1872 rupture of the Fish Springs fault

Discontinuities along fault zones in some cases act as either barriers or nuclei for fault ruptures (Aki, 1979; Lindh and Boore, 1981). The discontinuity in the fault zone at Poverty Hills apparently was too small to prevent rupture through the Fish Springs area in 1872. If 1872 displacement had terminated at the Poverty Hills discontinuity, the Fish Springs fault could have become "loaded" and posed a near-future seismic hazard. However, in light of evidence for 1872 rupture of the fault and its long average recurrence interval, the fault probably is "unloaded" and a small seismic hazard in the near future. Rupture of the Fish Springs fault may have loaded the area to the northwest though, for the Coyote Warp is the site of several of the largest Owens Valley earthquakes since 1872 (Martel, 1984).

Relationship of the Fish Springs fault to the Owens Valley fault zone

The displacement behaviors of the Fish Springs fault and the Lone Pine fault, which are near opposite ends of the Owens Valley fault zone, are similar in some interesting aspects. The Lone Pine fault ruptured three times within the last 10,000 to 20,000 years (Lubetkin, 1980) and has a 5000-7000-year recurrence interval (L.K.C. Lubetkin, written commun., 1981). The Fish Springs fault apparently has ruptured three times in the last 10,600 to 18,000 years and has a long-term average recurrence interval of 4600+1100 years. Perhaps ruptures of the Fish Springs fault are associated with ruptures of the entire Owens Valley fault zone. The structure and displacement on the Fish Springs fault may provide some insight into the regional stresses acting on the Owens Valley fault zone. The pure normal slip on the north-striking fault suggests that the maximum horizontal compression may be oriented north-south. A different orientation could cause lateral slip on the fault. Additionally, both the Fish Springs cinder cone and Red Mountain (another cinder cone) are linked to Crater Mountain 506 by north-striking normal faults. Such alignments of cinder cones tend to parallel the maximum horizontal compression (Nakamura et al., 1977). Carver (1970) concluded that the late Quaternary maximum horizontal compressive stress at the south end of Owens Valley has been oriented approximately north-south based on the north-south orientation of faults with pure dip-slip displacement and on the north-south orientation of desiccation cracks on the Owens Lake playa. A north south orientation of the maximum horizontal stress is significantly different from the N33°E orientation inferred by Zoback and Zoback (1980). Although right-lateral, normal oblique slip on the Owens Valley fault zone is consistent with both stress regimes, a N33°E maximum horizontal stress orientation is consistent with neither the observations at Fish Springs nor those in southern Owens Valley by Carver. Because the maximum horizontal compressive stress in California west of the Sierra Nevada is approximately north-south (Zoback and Zoback, 1980), perhaps Owens Valley is also in that stress regime.

ACKNOWLEDGEMENTS

The author is most grateful for the guidance of the late R.H. Jahns in the initial stages of this project. Special thanks are also due to A. Gillespie of the Jet Propulsion Laboratory, T. Harrison of the State University of New York at Albany, S. Waichler, and L. Martel. The financial contributions of the Shell fund, the Department of Applied Earth Sciences at Stanford University, and my parents are gratefully acknowledged.

REFERENCES

Aki, K., 1979. Characterization of barriers on an earthquake fault. J. Geophys. Res., 84:6140 6148. Atwater, B.F., Adam, D,P., Bradbury, J.P., Forester, R.M., Mark, R.K., Lettis, W.R., Fisher, G.R., Gobalet, K.W. and Robinson, S.W., 1986. A fan dam for Tulare Lake, California, and implications for the Wisconsin glacial history of the Sierra Nevada. Geol. Soc. Am. Bull., 97:97 109. Bacon, C.R., Giovanetti, D.M., Duffield, W.A., Dalrymple, G.B. and Drake, R.E., 1982. Age of the Coso Formation, Inyo County, California. U.S. Geol. Surv. Bull., 1527, 18 pp. Bateman, P.C., 1961. Willard D. Johnson and the strike-slip component of fault movement in the Owens Valley, California, earthquake of 1872. Bull. Seismol. Soc. Am., 51: 483-493. Bateman, P.C., 1965. Geology and tungsten mineralization of the Bishop district, California. U.S. Geol. Surv. Prof. Pap., 470, 208 pp. Birkeland, P.W., Crandell, D.R. and Richardson, G.M., 1971. Status of correlation of Quaternary stratigraphic units in the western United States. Quat. Res., 1:208 227. Bonilla, M.G., 1968. Evidence for right lateral movement on the Owens Valley, California, fault zone during the earthquake of 1872, and possible subsequent fault creep. In: W.R. Dickinson and A. Grantz (Editors), Proceedings of Conference in Geologic Problems of San~amdreas Fault System. Stanford Univ. Publ. Geol. Sci., pp.4-5. Carver, G.A., 1970. Quaternary Tectonism and Surface Faulting in the Owens Lake Basin, California. Mackay School of Mines Tech. Rep. AT-2, 103 pp. Clayton, L., 1966. Tectonic depressions along the Hope fault, a transcurrent fault in north Canterbury, New Zealand. N.Z. Geol. Geophys., 9:94 104. Gianella, V.P., 1959. Left-lateral faulting in Owens Valley, California. Geol. Soc. Am. Bull., 70: 1721. Gilbert, G.K., 1884. A theory of earthquakes of the Great Basin, with a practical application. Am. J. Sci., 3rd Ser., 127:49 53. Gillespie, A.R., 1982. Quaternary Glaciation and Tectonism in the Southeastern Sierra Nevada, 507

Inyo County, California. PhD. thesis, California Institute of Technology, Pasadena, Calif., 695 pp. Giovannetti, D.M., 1979. Volcanism and sedimentation associated with the formation of southern Owens Valley, California. Geol. Sot. Am. Abstr. Prog., 11: 79. Hill, M.R., 1972. A centennial anniversary of the great Owens Valley earthquake of 1872. Calif. Geol., 25: 51-54. Hobbs, W.H., 1910. The earthquake of 1872 in the Owens Valley, California. Beitr. Geophys., 10: 3522385. Knopf, A., 1918. A geologic reconnaissance of the Inyo range and eastern slope of the southern Sierra Nevada, California, with a section on the stratigraphy of the Inyo Range by Edwin Kirk. U.S. Geol. Surv. Prof. Pap., 110, 130 pp. Lindh, A.G. and Boore, D.M., 1981. Control of rupture by fault geometry during the 1966 Parkfield earthquake. Bull. Seismol. Sot. Am., 71: 6041-6049. Lubetkin, L.K.C., 1980. Late Quaternary Activity along the Lone Pine Fault, Owens Valley, California. M.S. thesis, Stanford University, Stanford, Calif., 85 pp. Martel, S.J., 1984. Late Quaternary Activity on the Fish Springs Fault, Owens Valley Fault Zone, California. MS. thesis, Stanford University, Stanford. Calif., 100 pp. Martel, S.J., Harrison, T.M. and Gillespie, A.R., 1987. Late Quaternary displacement rate across the Fish Springs fault, Owens Valley fault zone, California. Quat. Res., 27: 1133129. Mezger, L. and Burbank, D., 1986. The glacial history of the Cottonwood Lakes area, southeastern Sierra Nevada. Geol. Sot. Am. Abstr. Progr., 18: 157. Moore, J.G. and Dodge, F.C.W., 1980. Late Cenozoic volcanic rocks of the southern Sierra Nevada, California, I. Geology and petrology. Geol. Sot. Am. Bull., 91: 1995-2038. Nakamura, K., Jacob, K.H. and Davies, J.N., 1977. Volcanoes as possible indicators of tectonic stress orientation: Aleutians and Alaska. Pure Appl. Geophys., 115: 877112. Nelson, C.A., 1966. Geologic Map of the Waucoba Mountain Quadrangle, Inyo County, California. U.S. Geol. Surv. Geologic Quadrangle Map, GQ-528. Oakeshott, G.B., Greensfelder, R.W. and Kahle, J.E., 1972. 187221972 one hundred years later. Calif. Geol., 25: 55-61. Pakiser, L.C., Kane, M.F. and Jackson, W.H., 1964. Structural geology and volcanism of Owens Valley region, California - a geophysical study. U.S. Geol. Surv. Prof. Pap., 438, 68 pp. Pitt, A.M. and Steeples, D.W., 1975. Microearthquakes in the Mono Lake-northern Owens Valley, California, region from September 28 to October 18, 1970. Bull. Seismol. Sot. Am., 65: 8355844. Richardson, L., 1972. Left-lateral, east-trending faults in the Alabama Hills. Calif. Geol., 25: 63-64. Richter, C.F., 1958. Elementary Seismology. W.H. Freeman, San Francisco, Calif., 768 pp. Ross, D.C., 1965. Geology of the Independence quadrangle, Inyo County, California. U.S. Geol. Surv. Bull., 1181-0, 64 pp. Savage, J.C. and Lisowski, M., 1980. Deformation in Owens Valley, California. Bull. Seismol. Sot. Am., 70: 122551232. Savage, J.C., Church, J.P. and Prescott, W.H., 1975. Geodetic measurement of deformation in Owens Valley, California. Bull. Seismol. Sot. Am., 65: 865-874. Segall, P. and Pollard, D.D., 1980. Mechanics of discontinuous faults. J. Geophys. Res., 85: 4337-4350. Sharp, R.V. and Clark, M.M., 1972. Geological evidence of previous faulting near the 1968 rupture on the Coyote Creek fault. In: The Borrego Mountain Earthquake of April 9, 1968. U.S. Geol. Surv. Prof. Pap., 787: 131-140. Smith, G.I., 1979. Subsurface stratigraphy and geochemistry of Late Quaternary evaporites, Searles Lake, California. U.S. Geol. Surv. Prof. Pap., 1043, 130 pp. Tocher, D., St. Amand, P., Engel, R., Von Huene, R., Bateman, P.C. and Ross, D.C., 1963. Guidebook for Seismological Study Tour. International Union of Geodesy and Geophysics, 110 pp. Townley, S.D. and Allen, M.W., 1939. Descriptive catalogue of earthquakes of the Pacific coast of the United States 176991928. Bull. Seismol. Sot. Am., 29: l-297. Wallace, R.E., 1977. Profiles and ages of young fault scarps, north-central Nevada. Geol. Sot. Am. Bull., 88: 1271-1281. Whitney, J.D., 1888. The Owens Valley earthquake in two papers. Annual Report of the State Mineralogist, California State Mining Bureau, 8: 2888309. Zoback, M.L. and Zoback, M., 1980. State of stress in the conterminous United States. J. Geophys. Res., 85: 611336156.