q 2003 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or editing@geosociety.org.Geology; April 2003; v. 31; no. 4; p. 343–346; 3 figures; Data Repository item 2003041. 343

Quaternary low-angle slip on detachment faults inDeath Valley, CaliforniaNicholas W. Hayman* Department of Earth and Space Sciences, Box 351310, University of Washington, Seattle, Washington

98195, USAJeffrey R. Knott Department of Geological Sciences, Box 6850, California State University, Fullerton, California 92834, USADarrel S. CowanEliza Nemser*

Department of Earth and Space Sciences, Box 351310, University of Washington, Seattle, Washington98195, USA

Andrei M. Sarna-Wojcicki U.S. Geological Survey, Earth Surface Processes Team, Western Region, 345 Middlefield Road,MS 975, Menlo Park, California 94025, USA

ABSTRACTDetachment faults on the west flank of the Black Mountains (Nevada and California)

dip 298–368 and cut subhorizontal layers of the 0.77 Ma Bishop ash. Steeply dippingnormal faults confined to the hanging walls of the detachments offset layers of the 0.64Ma Lava Creek B tephra and the base of 0.12–0.18 Ma Lake Manly gravel. These faultssole into and do not cut the low-angle detachments. Therefore the detachments accruedany measurable slip across the kinematically linked hanging-wall faults. An analysis ofthe orientations of hundreds of the hanging-wall faults shows that extension occurred atmodest slip rates (,1 mm/yr) under a steep to vertically oriented maximum principalstress. The Black Mountain detachments are appropriately described as the basal detach-ments of near-critical Coulomb wedges. We infer that the formation of late Pleistoceneand Holocene range-front fault scarps accompanied seismogenic slip on the detachments.

Keywords: detachment, Death Valley, Quaternary, extension, tephrochronology.

INTRODUCTIONLow-angle normal, or detachment, faults

are recognized in extended regions worldwide.Detachment faulting is a controversial processthat seemingly defies a mechanical explana-tion despite its prevalence and its possible rolein accommodating large-magnitude extensionand denudation (Wernicke, 1985). Two gen-eral questions are currently debated: (1) canslip occur on normal faults that dip 308 or less(Buck, 1988; Wills and Buck, 1997)?, and (2)is slip on detachments seismogenic (Wernicke,1995; Jackson and White, 1989)? One reasonthat these questions remain is that there arefew documented examples of detachmentfaults that have been active in the QuaternaryPeriod (Wernicke, 1995).

The literature contains arguments for andagainst low-angle, normal-sense slip on de-tachments. For example, the rolling-hingemodel for extension hypothesizes that detach-ments originated as steeply dipping normalfaults that were tilted and became deactivatedas the footwalls isostatically responded to ex-humation (Buck, 1988; Wernicke and Axen,1988). In contrast, geologic and geophysicalevidence from some extended regions indi-cates that slip occurred on normal faults dip-ping #308 (Axen, 1999; Axen et al., 1999;Abbott et al., 2001). We report geologic evi-

*E-mail: nickh@u.washington.edu. Current ad-dress: Nemser—Seismic Hazards Group, URS Cor-poration, 500 12th Street, Suite 200, Oakland, Cal-ifornia 94607-4014, USA.

dence from Death Valley that strongly sup-ports the argument that seismogenic slip oc-curred on low-angle normal faults inQuaternary time.

BLACK MOUNTAIN DETACHMENTSDeath Valley is in the southwestern part of

the Basin and Range Province, an exemplar ofintracontinental extensional tectonics. Sincelate Miocene–early Pliocene time, the DeathValley pull-apart basin between the BlackMountains and Panamint Mountains extendedbetween the Northern Death Valley and South-ern Death Valley dextral strike-slip fault sys-tems (Fig. 1) (Burchfiel and Stewart, 1966).The extension has been accommodated on theeastern margin of Death Valley along theBlack Mountains fault zone, which comprisesgently (range 198–368, average 258) and steep-ly (;608) dipping normal faults with localoblique-to-strike-slip components (Keener etal., 1993). We studied three localities in theBlack Mountains fault zone where the detach-ments and hanging walls are exposed abovethe basin floor: Badwater, Mormon Point, andSouth Mormon Point (Fig. 1). We investigatedthe stratigraphy and structure of the hangingwalls and the relationship of faults in thehanging walls to the detachments.

In a typical exposure of a detachment faultat these localities, a well-defined principal slipplane separates the hanging wall from a 0–7-m-thick zone of gouge and breccia (Miller,1996). Above the detachments, hundreds of

steeply dipping and listric faults cut the hanging-wall sediments (Fig. 2). In exposures, thesefaults sole into or are cut by the principal slipplane, and hanging-wall sediment is locallypresent within the pervasively deformed faultrocks. These observations collectively indi-cate that slip on the detachments and on thehanging-wall faults occurred during the sameinterval of geologic time.

HANGING-WALLTEPHROCHRONOLOGY

Sediments constituting the hanging walls atBadwater and Mormon Point were depositedin playa lake and alluvial fan environmentsthat are similar to those of today. The mostcontinuous exposures of hanging-wall sedi-ments are at Mormon Point, where the sectioncontains several pronounced tephra layers(Fig. 2). The central tephra is 100–200 cmthick and is correlated with 0.77 Ma Bishopash on the basis of glass composition1 and thenature of biotite phenocrysts. The Bishop ashis also intercalated with alluvial-fan brecciasat Badwater.

There are prominent tephra layers strati-graphically above and below the Bishop ashexposed at Mormon Point. The younger tephracorrelates with the 0.64 Ma Lava Creek Btephra (Lanphere et al., 2002) and the olderwith 0.8–1.2 Ma upper Glass Mountain tephra(Knott et al., 1999). These lower to middlePleistocene sediments are overlain by 0.18–0.12 Ma (oxygen isotope stage [OIS] 6) LakeManly gravel (Ku et al., 1998). At a few lo-calities the Bishop ash is cut by gently dippingdetachment faults. Most of the section aboveand below the Bishop ash is only slightly tilt-ed from its depositional orientation. This ev-idence demonstrates that some slip on the de-tachments is younger than 0.77 Ma.

1GSA Data Repository item 2003041, electron mi-croprobe analysis of volcanic glass shards from mid-dle Pleistocene tephra layers, Death Valley, Califor-nia, and comparative compositions, is availableon request from Documents Secretary, GSA, P.O.Box 9140, Boulder, CO 80301-9140, USA, editing@geosociety.org, or at www.geosociety.org/pubs/ft2003.htm.

344 GEOLOGY, April 2003

Figure 1. Death Valley, pull-apart basin between Northern and Southern Death Valley faultzones (NDVFZ and SDVFZ, respectively). Black Mountains (shown in compiled 30 m U.S.Geological Survey digital elevation model [DEM]) are bound by fault system that consistsof scarps on valley floor (traces from Brogan et al., 1991) and low-angle detachments (tracesfrom Drewes, 1963; Miller, 1991). Hanging walls, to west of and above traces of detachments,consist of Pliocene to Holocene sediments. These sediments are cut by steeply dippingfaults (histogram) that do not cut detachments. Stereographic projections (plotted with Ster-eonet by R. Allmendinger) of 2s contours of poles to hanging-wall fault planes at eachlocality define modes that are synthetic (Mormon Point and South Mormon Point) and an-tithetic (Badwater) to detachment plane (continuous great circles). Striae on detachments(arrows) are parallel to slip direction on hanging-wall fault populations and overall slip di-rection of two fault systems is normal to range front. Orientation of maximum principalstress, s1, is inferred to be steep to vertical and to bisect acute angle between modes ofhanging-wall fault population (histogram and square symbols on stereonet). Cross sectionsA–A9 and B–B9 are shown in Figure 2.

HANGING-WALL FAULTSWe measured the strikes and dips of 267

hanging-wall faults above the Badwater, Mor-mon Point, and South Mormon Point detach-ments. Fault dips have a bimodal distribution(Fig. 1). West-dipping faults that are syntheticto the detachments have a mode of 558 andmean of 658, and the east-dipping, antitheticset has a less-well-defined mode near 708 anda mean of 658. Coulomb materials with an an-gle of internal friction of 308 fail along con-jugate pairs of normal faults wherein the max-imum principal compressive stress, s1, bisectsan acute angle of intersection of 608. Themodes of the fault-population data are com-patible with a steep to vertically oriented s1.This conclusion is consistent with Anderson-ian mechanics, which predicts that s1 is nor-mal to Earth’s surface (Anderson, 1951).Above the detachments, the surfaces of thehanging walls have a 08–68 slope (Fig. 2). Weargue that the near-vertical s1 indicates thatneither the strata in the hanging walls nor thepopulations of normal faults have been tiltedmore than a few degrees.

The hanging-wall faults are kinematicallycoupled with the detachments: the general di-rections and amounts of slip observed on thehanging-wall faults have been equivalently re-solved on the detachments, and the detach-ments and the hanging walls have not inde-pendently tilted. The evidence for kinematiccoupling of the two fault systems includes thefollowing: (1) We note that none of the ob-served hanging-wall faults cuts completelyacross a principal slip plane and related faultrocks (Fig. 3). Moreover, many hanging-wallfaults are listric and curve to join the principalslip plane. (2) The mode of the strikes of thehanging-wall fault populations at each localityis parallel to the local strike of the related de-tachment (Fig. 1). (3) There is a normal senseof separation of strata across exposed faults,and a net down-to-the-west or -northwestthrow across the population of faults. The pres-ence of a conjugate set in the bulk populationis consistent with dip slip on the hanging-wallfault system. (4) Striae on the principal slipplanes of the detachments are parallel to themodal dip direction of the hanging-wall faultsand the direction of net extension across thehanging-wall fault system.

We note that there are striae and other ki-nematic indicators recording lateral andoblique slip on parts of the Black Mountainfault system (Keener et al., 1993). Local de-viations from pure dip slip on the faults thatwe studied are spatially coincident with thehinges of the lower-plate antiforms (Wright etal., 1974; Holm et al., 1994), unmapped irreg-ularities in the detachment surface (Nemser,2001), or the influence of dextral slip on the

Northern and Southern Death Valley faultzones (Keener et al., 1993).

DISCUSSIONAge of Slip and Original Orientation ofthe Detachments

At Mormon Point and Badwater, the Bishoptephra (0.77 Ma) is directly cut by the detach-ment, and the youngest sediment in the over-lying strata is the 0.18–0.1 Ma (OIS 6) gravel.These Quaternary strata are relatively flat ly-ing except in a rollover anticline (Fig. 2),where a significant listric hanging-wall faultplaces the Quaternary section against Pliocenesediment. The high-angle fault system thatcuts the entire Quaternary section is kinemat-ically coupled with the detachment. Moreover,the modes of the fault populations are mostcompatible with a s1 that is normal to the sur-

faces of the hanging walls. Collectively, theseobservations are evidence that the detach-ments have not tilted from an initial dip of198–368. Any kinematic model for detachmenttilt without corresponding hanging-wall tiltwould conflict with these observations andwould require structures that are not observed(e.g., young breakaway basins or extensionalduplexes).

The field relationships only require that thedetachment and hanging-wall fault systemslipped since 0.18–0.1 Ma. However, slip re-stricted to the last 0.18 m.y. would be incom-patible with the observed extension across thefault system. We estimate ;600 m net hori-zontal extension, or ;14% elongation alongthe 5 km NW-SE section across the high-anglefaults in the Quaternary sediments at MormonPoint. This estimate is derived from hundreds

GEOLOGY, April 2003 345

Figure 2. Cross sectionA–A9 across MormonPoint illustrating (1) sed-imentary horizons ofknown or inferred agesand (2) amounts of slipon more significant nor-mal faults. Intersectionsbetween hanging-wallfaults are inferred wherenot exposed, such asdepth trace of frontalfault scarp. Projectedabove section A–A9 is ad-jacent cross section, B–B9, depicting hanging-wall rollover anticline,density of significantfaults, and presence ofPliocene sediments atupdip, easternmost partof section. For locationsof cross sections, seeFigure 1. Photograph issouth-facing view ofhanging-wall fault, inQuaternary sediment,soling into detachment atMormon Point. OIS is ox-ygen isotope stage.

of small-displacement (,1 m) normal faultsand several large-displacement (.10–100 m)faults (Fig. 2) and is a minimum because itdoes not include translation on the detachmentsthat was unaccompanied by hanging-wall de-formation. We adopt a value for slip rates of,1 mm/yr, supported by geodesy and paleo-seismology of the Basin and Range (Wernickeet al., 2000). For example, 600 m of extensionsince the deposition of the 1.2 Ma upper GlassMountain tephra would have proceeded at anaverage slip rate of 0.5 mm/yr.

The Pliocene sediments above the CopperCanyon detachment, and correlative sedimentsat Mormon Point and Badwater, contain a sim-ilar hanging-wall fault system (Holm et al.,1994). The kinematics in these older sectionsare more complicated to reconstruct than inthe Quaternary section yet are compatiblewith coupled slip on low-angle detachmentsand hanging-wall faults in Pliocene time.

Application of the Extensional CoulombWedge Model

Our observations and inferences concerningQuaternary slip on detachment faults are atodds with most models for detachment fault-ing. For example, a strict application of therolling-hinge model predicts that low-angledetachments are inactive in the shallow crust(Buck, 1988; Wernicke and Axen, 1988). Oth-er models for the initiation of, and slip on,low-angle normal faults involve rapid exten-sion rates, the rotation of the stress field rel-ative to the fault plane, or the elevation offluid pressure (Wills and Buck, 1997). To pro-vide an alternative explanation for active slipon this fault system we propose that the

wedge-shaped sedimentary sections along theBlack Mountains are extensional Coulombwedges (Fig. 3) (Xiao et al., 1991). This mod-el predicts that a critically or subcritically ta-pered wedge is stable and therefore can slideon a detachment without undergoing internaldeformation. Hanging-wall wedges that havetapers greater than critical, however, will failand extend by normal faulting in the hangingwall; slip on each fault is resolved onto thedetachment.

One control on the maintenance of the crit-ical taper is the effective basal friction of thedetachment (meff). When meff is lowered fromtypical values of 0.6–0.85 [f 5 308, m 5tan(f)] the detachment is weakened. Onemechanism for weakening could be the de-velopment of fault rocks with unusually lowstatic friction (m). Alternatively, if the ratio offluid pressure to lithostatic pressure along thedetachment (lb) is higher than that within thewedge (l), then meff 5 m[(1 2 lb)/(1 2 l)]and meff , m. We use the observed taper atMormon Point with a surface slope (a) of 268and a basal dip (b) of 308 (Fig. 2), and weassume that the sediment is cohesionless, thatl does not deviate from hydrostatic through-out the wedge, and that f 5 308 for the wedgematerials. The extensional solution to theCoulomb wedge problem (Fig. 3) (Dahlen,1984; Xiao et al., 1991) predicts that whenmeff . 0.4 or fb . 218, the observed taper issupercritical. If the detachment were to beweakened such that fb # 218, the wedgewould be stable or critically stable. If m 50.6–0.85 for the fault rocks within the detach-ment, then the fluid pressure is the dominantcontrol on detachment weakening, requiring

that lb 5 0.7 for l 5 0.5; the latter is a nom-inal value for the hydrostatic fluid pressure.Such an elevation of fluid pressure within afault zone is within the range of measured flu-id pressures within faults (Xiao et al., 1991).Regardless of the assumptions used to applythe model or the mechanism for the modestweakening of the detachment, the extensionalCoulomb wedge model adequately describesthe behavior on the Black Mountain faultzone.

Seismic Slip on the DetachmentsThe Black Mountains fault zone includes

range-front fault scarps in upper Pleistoceneor Holocene alluvial fan and playa sediments(Fig. 1) (Brogan et al., 1991). The scarps arethe surface expression of steeply dippingfaults that in central Death Valley accrue dom-inantly normal slip (Brogan et al., 1991). Theintersections of the frontal faults and the de-tachments are buried beneath the valley floor(Fig. 2). It has been argued that either thefrontal faults resolve themselves on the de-tachments (Hamilton, 1988; Burchfiel et al.,1995) or, alternatively, cut the detachments atdepth (Miller, 1991; Keener et al., 1993; Chi-chanski, 2000). In the latter case, Holoceneand possibly older slip on the frontal faultswould have superseded any slip on the de-tachments and necessarily deactivated them.

We propose that the steep frontal faults be-long to the larger population of hanging-wallfaults that resulted from extension of thehanging-wall wedges. It is probable that someof the members of the population that are nowsituated above the floor of the valley once de-fined the mountain front, just as the youngest

346 GEOLOGY, April 2003

Figure 3. Solution to extensional criticalCoulomb wedge problem described by Xiaoet al. (1991): a 1 b 5 c1 2 c0, where a issurface slope, b is detachment dip, c0 isstress angle, and c1 is function of friction,f, basal friction, fb, and a. As explained byDahlen (1984), c1 is periodic function withtwo minima and extensional solution is for2p/2 < c0 < 0. Thus, any given a and b havetwo critically stable solutions that we solvedfor numerically and plot here as curvesbounding shaded region. Any a and b thatplot on this curve or within shaded region,are either critically stable or stable, respec-tively, and can translate down detachmentdip without internal faulting. Any a and bthat plot within unshaded region are unsta-ble and hanging-wall faulting is required forwedge to accommodate tectonic extension.Mechanisms for maintaining criticality varyin nature, but we consider fb to be control-ling parameter; wedge at Mormon Point (a 526, b 5 30) is critically stable for fb 5 21.However, if fb > 21, boundary for stabilityfield would shift to right (dashed line for fb5 24) and wedge would become unstable.Alternatively, if fb < 21, boundary wouldshift to left (dashed line for fb 5 18) andwedge would become stable. Mechanism forchanges in friction of detachment could beeither presence of fault rocks with anoma-lously low friction, or elevated fluidpressure.

scarps do today. According to our hypothesis,the frontal faults do not cut the buried down-dip continuations of the detachments; rather,slip on them is kinematically coupled to slipon the buried detachments. If our hypothesisis true, then the seismic ruptures thought to beresponsible for the young scarps in Death Val-ley (Brogan et al., 1991) must have propagat-ed up the detachment faults. Our proposal thatat least some slip on the detachments was seis-mogenic agrees with recent interpretationsmade elsewhere in the Basin and Range Prov-ince (Axen, 1999; Axen et al., 1999; Abbottet al., 2001).

CONCLUSIONSStructural and stratigraphic data from the

hanging wall of the Black Mountain detach-ments require that the detachments accruedlow-angle, normal-sense slip in the Quater-nary, likely since the Pliocene and into Ho-

locene time. The detachment slip was coupledwith slip on a system of high-angle faults inthe hanging wall. A modest slip rate, a steepto vertically oriented principal stress, and theangular relationship between the surface of thehanging walls and the detachments support thebest mechanical explanation for detachmentfaulting in this setting: the sediment overlyingthe detachment forms an extensional Coulombwedge. Our conclusions lead to the interpre-tation that the seismic slip that created thehigh-angle fault scarps on the valley floororiginated on the detachments at depth.

ACKNOWLEDGMENTSHayman, Nemser, and Cowan received support

from the donors to the Petroleum Research Fund,administered by the American Chemical Society.Knott and Sarna-Wojcicki received support fromNational Science Foundation grant EAR 94-06029and the U.S. Geological Survey. We thank JohnBartley, John Dewey, Martin Miller, and Brian Wer-nicke for valuable reviews.

REFERENCES CITEDAbbott, R.E., Louie, J.N., Caskey, S.J., and Pullam-

manappallil, S., 2001, Geophysical confirma-tion of low-angle normal slip on the histori-cally active Dixie Valley fault, Nevada:Journal of Geophysical Research, v. 106,p. 4169–4181.

Anderson, E.M., 1951, The dynamics of faulting:London, Oliver and Boyd, 206 p.

Axen, G.J., 1999, Low angle normal fault earth-quakes and triggering: Geophysical ResearchLetters, v. 26, p. 3693–3696.

Axen, G.J., Fletcher, J.M., Cowgill, E., Murphy, M.,Kapp, P., MacMillan, I., Ramos, V.E., and Ar-anda, G.J., 1999, Range-front fault scarps ofthe Sierra El Mayor, Baja California: Formedabove an active low-angle normal fault?: Ge-ology, v. 27, p. 247–250.

Brogan, G.E., Kellogg, K.S., Slemmons, D.B., andTerhune, C.L., 1991, Late Quaternary faultingalong the Death Valley–Furnace Creek faultsystem, California and Nevada: U.S. Geolog-ical Survey Bulletin 1991, 23 p.

Buck, R.W., 1988, Flexural rotation of normalfaults: Tectonics, v. 7, p. 959–973.

Burchfiel, B.C., and Stewart, J.H., 1966, ‘‘Pull-apart’’ origin of the central segment of DeathValley, California: Geological Society ofAmerica Bulletin, v. 77, p. 439–441.

Burchfiel, B.C., Molnar, P., Zhang, P., Deng, Q., andWang, Y., 1995, Example of a supradetach-ment basin within a pull-apart tectonic setting:Mormon Point, Death Valley, California: Ba-sin Research, v. 7, p. 199–214.

Chichanski, M., 2000, Low-angle, range-flank faultsin the Panamint, Inyo, and Slate Ranges, Cal-ifornia: Implications for recent tectonics of theDeath Valley region: Geological Society ofAmerica Bulletin, v. 112, p. 871–883.

Dahlen, F.A., 1984, Noncohesive critical Coulombwedges: An exact solution: Journal of Geo-physical Research, v. 89, p. 10,125–10,133.

Drewes, H., 1963, Geology of the Funeral PeakQuadrangle, California, on the eastern flank ofDeath Valley: U.S. Geological Survey Profes-sional Paper 413, 78 p.

Hamilton, W.B., 1988, Detachment faulting in theDeath Valley region, California and Nevada:U.S. Geological Survey Bulletin 1790, p. 51–85.

Holm, D.K., Fleck, R.J., and Lux, D.R., 1994, The

Death Valley Turtlebacks reinterpreted asMiocene–Pliocene folds of a major detach-ment surface: Journal of Geology, v. 102,p. 718–727.

Jackson, J.A., and White, N.J., 1989, Normal fault-ing in the upper continental crust: Observa-tions from regions of active extension: Journalof Structural Geology, v. 11, p. 15–36.

Keener, C., Serpa, L.F., and Pavlis, T.L., 1993,Faulting at Mormon Point, Death Valley, Cal-ifornia: A low-angle normal fault cut by high-angle faults: Geology, v. 21, p. 327–330.

Knott, J.R., Sarna-Wojcicki, A.M., Meyer, C.E.,Tinsley, J.C., III, Wells, S.G., and Wan, E.,1999, Late Cenozoic stratigraphy and tephro-chronology of the western Black Mountainspiedmont, Death Valley, California: Implica-tions for the tectonic development of DeathValley, in Wright, L.A., and Troxel, B.W., eds.,Cenozoic basins of the Death Valley region:Geological Society of America Special Paper333, p. 345–366.

Ku, T.H., Luo, S., Lowenstein, T.K., Li, J., andSpencer, R.J., 1998, U-series chronology of la-custrine deposits in Death Valley, California:Quaternary Research, v. 50, p. 261–275.

Lanphere, M.A., Champion, D.E., Christiansen,R.L., Izett, G.A., and Obradovick, J.D., 2002,Revised ages for tuffs of the Yellowstone Pla-teau volcanic field: Assignment of the Huck-leberry Ridge Tuff to a new geomagnetic po-larity event: Geological Society of AmericaBulletin, v. 114, p. 559–568.

Miller, M.G., 1991, High-angle origin of the cur-rently low-angle Badwater Turtleback fault,Death Valley, California: Geology, v. 19,p. 372–375.

Miller, M.G., 1996, Ductility in fault gouge from anormal fault system, Death Valley, California:A mechanism for fault-zone strengthening andrelevance to paleoseismicity: Geology, v. 24,p. 603–606.

Nemser, E.S., 2001, Kinematic development of up-per plate faults above low-angle normal faultsin Death Valley, California [M.S. thesis]: Seat-tle, University of Washington, 102 p.

Wernicke, B., 1985, Uniform-sense normal simpleshear of the continental lithosphere: CanadianJournal of Earth Science, v. 22, p. 108–125.

Wernicke, B., 1995, Low-angle normal faults andseismicity: A review: Journal of GeophysicalResearch, v. 100, p. 20,159–20,164.

Wernicke, B., and Axen, G.J., 1988, On the role ofisostacy in the evolution of normal fault sys-tems: Geology, v. 16, p. 848–851.

Wernicke, B., Friedrich, A.M., Niemi, N.A., Ben-nett, R.A., and Davis, J.L., 2000, Dynamics ofplate boundary fault systems from Basin andRange Geodetic Network (BARGEN) andgeologic data: GSA Today, v. 10, no. 11,p. 1–6.

Wills, S., and Buck, R.W., 1997, Stress-field rota-tion and rooted detachment faults: A Coulombfailure analysis: Journal of Geophysical Re-search, v. 102, p. 20,503–20,514.

Wright, L.A., Otton, J.K., and Troxel, B.W., 1974,Turtleback surfaces of Death Valley viewed asphenomena of extensional tectonics: Geology,v. 2, p. 53–54.

Xiao, H.B., Dahlen, F.A., and Suppe, J., 1991, Me-chanics of extensional wedges: Journal of Geo-physical Research, v. 96, p. 10,301–10,318.

Manuscript received 6 August 2002Revised manuscript received 21 November 2002Manuscript accepted 3 December 2002

Printed in USA