Along-strike upper-plate deformation in response to

Tectonophysics 488 (2010) 162–173

Contents lists available at ScienceDirect

Tectonophysics

j ourna l homepage: www.e lsev ie r.com/ locate / tecto

Along-strike upper-plate deformation in response to metamorphic core complexemplacement, SE Arizona

Frank H. (Trey) Wagner III 1, Roy A. Johnson ⁎Department of Geosciences, University of Arizona, Tucson, AZ 85721, United States

⁎ Corresponding author.E-mail addresses: [email protected]

[email protected] (R.A. Johnson).1 Now at: ConocoPhillips, Upstream Technology, 600

77079-1175, USA.

0040-1951/$ – see front matter © 2009 Published by Edoi:10.1016/j.tecto.2009.07.017

a b s t r a c t
a r t i c l e i n f o
Article history:Received 6 November 2008Received in revised form 8 June 2009Accepted 21 July 2009Available online 29 July 2009

Keywords:Catalina–Rincon metamorphic core complexMetamorphic core complexDetachment faultFinite-element modelingStress rotationSanta Rita fault

Metamorphic core complexes represent concentrated zones of crustal extension; universally, the hangingwalls of these systems are extensively fractured, attesting to the significant horizontal extension above theevolving detachment zone. Recent analysis of a large grid of 2-D seismic reflection lines within the TucsonBasin of southeastern Arizona has facilitated a nearly three-dimensional interpretation of subsurface featuresrelated to Cenozoic crustal extension. Within the northern Tucson Basin, the Catalina detachment fault dips23–35° to the southwest from the western flank of the Catalina/Rincon Metamorphic Core Complex. In thesouthern portion of the basin, the NE-trending Santa Rita normal fault dips 15–20° to the northwest from thewestern flank of the Santa Rita Mountains and is cut by the Catalina detachment beneath the central TucsonBasin. The orientation of the Santa Rita fault is problematical in that its orientation is nearly perpendicular toextension directions in the region, while geologic and seismic reflection evidence indicates that the Catalinadetachment and Santa Rita fault were active synchronously. One possible explanation is that the Santa Ritafault is accommodating along-strike upper-plate deformation in response to core complex emplacement. Totest this, we employed a finite-element modeling approach. A two-dimensional model consisting of ahomogenous elastic material undergoing uniform extension is used to study changes in stresses anddisplacements in proximity to a zone of weakness representing a detachment fault in the upper crust. Awayfrom the detachment fault, primary principal stresses are oriented parallel to the regional extension directionas expected. Near the detachment, extension in the system produces a rotation of the primary principalstresses of nearly 60° with respect to the regional stress field around the end of the fault. Additionally, themodel predicts mechanical failure of the upper-plate of the detachment system to the south of the Catalinacore complex. This model supports our interpretation that the orientation and early displacement noted onthe Santa Rita fault is the result of a perturbation in the regional stress field caused by the Catalinadetachment and the associated brittle failure of the upper-plate from the extreme crustal extensionassociated with core complex emplacement.

© 2009 Published by Elsevier B.V.

1. Introduction

Metamorphic core complexes represent concentrated zones ofextreme crustal extension. Within the Basin and Range province ofwestern North America, metamorphic core complexes can be found ina discontinuous belt from southern Canada to northwestern Mexico(Crittenden et al., 1980; Coney, 1980). These ranges, primarily mid-Tertiary in age, expose metamorphosed mid-crustal rock in thefootwall of low-angle detachments systems (Davis, 1980; Coney andHarms, 1984). In these systems the hanging wall rocks are comprisedof upper crustal rocks (at the time of extension) that bear littleresemblance to the metamorphic rocks within the lower plate of the

m (F.H.(T.) Wagner),

N. Dairy Ashford, Houston, TX

lsevier B.V.

core complex (Davis, 1980). Universally, the hanging wall of thesesystems is extensively fractured, attesting to the significant horizontalextension above the evolving detachment zone (Lister and Davis,1989). These upper-plate units are variably tilted, but predominatelydip toward the core complex with strikes perpendicular to the mainextension direction (Spencer and Reynolds, 1989). High-angle normalfaults break the units into numerous fault-bounded blocks.

A number ofmodels have beenproposed to describe the evolution ofmetamorphic core complexes (Davis, 1980; Crittenden et al., 1980;Wernicke, 1985; Wernicke and Axen, 1988; Lister and Davis, 1989;Spencer and Chase,1989; Block and Royden,1990; Buck,1991; Scott andLister, 1992; Kruger and Johnson, 1994; Nourse et al., 1994; Lavier et al.,1999; Davis et al., 2004; Rosenbaum et al., 2005). Many of the modelsmake predictions for the types of hanging wall deformation expected.Typically, these models have been described in a purely two-dimen-sional fashion, that is, parallel to the direction of extension (orthogonalto strike) as a cross-section through the crust. However, upper-platedeformation must terminate along strike as well. This can logically be

mailto:[email protected]

mailto:[email protected]

http://dx.doi.org/10.1016/j.tecto.2009.07.017

http://www.sciencedirect.com/science/journal/00401951

163F.H.(T.) Wagner III, R.A. Johnson / Tectonophysics 488 (2010) 162–173

accomplished by transferring extension to adjacent structural features,via accommodation zones (Faulds andVarga,1998), or at the terminus ofthe system through distributed deformation.

Fig. 1. Simplified geologic map of the Tucson Basin, Santa Catalina Mts. and Santa Rita Mtslocations and Cenozoic dominant extension directions.

The Santa Catalina/Rincon metamorphic core complex lies adjacentto theTucsonBasinof southernArizona(Fig.1). Recent analysis of a largegrid of 2-D seismic reflection lines within the Tucson Basin facilitated a

. Modified from Hirschberg and Pitts (2000). Included are seismic line locations, well

164 F.H.(T.) Wagner III, R.A. Johnson / Tectonophysics 488 (2010) 162–173

nearly three-dimensional interpretation of the subsurface relationshipsof the Catalina detachment fault and other structures within the basin.Along the southeastern margin of the basin is the Santa Rita fault, arange-bounding low-angle normal fault orientated parallel to thepredominant extension direction. Seismic profiles reveal that theSanta Rita fault merges with the Catalina detachment beneath thecentral Tucson Basin and has accommodated oblique-slip in the upper-plate of the detachment system. Finite-element modeling undertakenfor this work suggests this is the result of stress rotations around theedges of the Catalina core complex with the Santa Rita faultaccommodating distributed deformation between zones with differingmagnitudes of crustal extension.

2. Geological and geophysical background

Data from various sources, including surface geology, seismicreflection profiles, and subsurface well control, are presented here toconstrain the geologic and structural characteristics of the Catalinadetachment system, the Catalina metamorphic core complex and theSanta Rita fault. The data are used to provide the groundwork for finite-element modeling of this system.

2.1. Geology

The Tucson Basin lies within the zone of highly extended terranesof the southern Basin and Range province of North America. TheTucson Basin is bounded to the northeast by the Santa Catalina/Rincon metamorphic core complex (locally referred to as the“Catalina core complex” after Rehrig and Reynolds, 1980), to thewest by the Tucson Mountains, to the south and southeast by theSanta Rita Mountains and to the southwest by the Sierrita Mountains(Fig. 1). Two distinct assemblages of Tertiary sedimentary sequencesare deposited in the basin (Wagner and Johnson, 2006). Units of theoldest sequence range in age from mid-Oligocene to mid-Mioceneand overlie a thick package of mid-Oligocene volcanic units (Eberlyand Stanley, 1978). These older sedimentary sequences are exposedalong the flanks of the core complex in a series of mini-basins formedin the hanging wall of the detachment system during structuralevolution of the Catalina core complex and are cut by numerousantithetic normal faults that merge downward into the masterdetachment (Dickinson, 1991). Capping these units is a regionallyextensive unconformity characterized by erosional unconformitiesand areas of paleosol formation. Stratigraphically above are up to2000 m of relatively undeformed Miocene to Pleistocene alluvial andfluvial sedimentary basin fill (Eberly and Stanley, 1978; Houser andGettings, 2000).

The Catalina core complex forms a broad dome with a peakelevation nearly 2 km above the Tucson Basin and displays aspectssimilar to other Cordilleran metamorphic core complexes in the Basinand Range (for example Davis, 1980; Rehrig and Reynolds, 1980;Crittenden et al., 1980; Coney, 1980; Coney and Harms, 1984; Listerand Davis, 1989; Nourse et al., 1994; Kruger and Johnson, 1994; Krugeret al., 1995). It is composed mainly of Tertiary mylonitic gneiss,Tertiary granite and Precambrian granite. Mylonitic fabrics occur in abelt about 10 km wide in map view along the southwestern flank ofthe core complex (Fig. 1). Metamorphic fabrics indicate top-to-WSWdisplacement (Davis, 1980; Dickinson, 1991) and grade into unde-formed granites structurally downward. Precambrian granites arefound on the northeastern flank of the Catalina core complex in faultcontact with Paleozoic, Mesozoic and Tertiary units.

The Catalina detachment fault bounds the extent of myloniticgneiss exposures to the southwest and separates metamorphosedmid-crustal footwall rock from shallow brittlely-deformed hangingwall rocks and sedimentary deposits. The Catalina detachment isinferred to have a minimum net displacement of 20 to 30 km basedon the distribution of pre-extensional sedimentary and igneous rock

units (Dickinson, 1991). Fission-track dating of footwall mylonitesindicates a period of rapid denudation between approximately 30 and20 Ma (Fayon et al., 2000). The detachment fault dips 25–35° towardsthe southwest from the southwestern flank of the Catalinacore complex and projects beneath the Tucson Basin. The detachmentfault is exposed on the southern end of the Santa Catalina Mountainsand is marked by a 1-meter thick resistant layer of cataclasite directlyoverlying a thick (10 s of meters) chloritic breccia (Rehrig andReynolds, 1980; Davis, 1980; Dickinson, 1991).

On the northwestern and southeastern flanks of the Catalina corecomplex are a series of younger high-angle normal faults including thePirate fault and the Martinez Ranch fault respectively (Fig. 1). Dis-placement on these faults is inferred to have occurred between 12 and6 Ma based on ages of associated sedimentary-rock units (Davis et al.,2004). These faults are interpreted to have controlled the late-stagetopographic uplift of the Catalina core complex, allowing flexuralisostatic uplift of the core complex (Davis et al., 2004; Wagner andJohnson, 2006).

The Santa Rita Mountains, bounding the southern portion of theTucson Basin, are composed largely of Paleozoic and Mesozoicsedimentary and volcanic units (Drewes, 1972; Hardy, 1997). Theseunits are heavily folded and faulted, resulting from northeast-directedcompressional deformation during the Laramide orogeny from LateCretaceous to Eocene in this area. These Laramide reverse faults havereactivated Jurassic to Early Cretaceous normal faults. Middle Tertiaryextension in the Santa Rita Mountains resulted in a southeastwardtilting of the Paleozoic through Tertiary strata. This extension waslikely accommodated on the Santa Rita fault or related structuresbounding the northwestern flank of the Santa Rita. The Santa Rita faultwas first identified through a series of fault scarps, offsetting lateQuaternary alluvial fan deposits, with offsets that range from 1 to 7 m(Pearthree, 1983; Pearthree and Calvo, 1987). Shallow refraction dataindicate fault offsets of 15–20 m in the near subsurface (Rutledge,1984).

Hardy (1997) identified a low-angle detachment fault, which hetermed the Santa Rita fault, in the adjacent foothills of the Santa RitaMountains. Hardy (1997) assigned a mid-Tertiary age to this faultsurface based on “the presence of chloritic breccia and protomylonite,rocks characteristic of mid-Tertiary low-angle normal faults.” Kine-matic indicators on the fault surface indicate top to the westmovement, with a minimum of 4 km of extension based on structuralreconstructions.

2.2. Seismic data

The seismic datasets in the Tucson Basin consist of grids of two-dimensional reflection profiles originally acquired as separate surveysby Exxon and Phillips Petroleum in the 1970s and early 1980s; thesedatasets were reprocessed by the University of Arizona ReflectionSeismology group following common 2-D techniques. Details of thedata acquisition and processing are presented in Wagner and Johnson(2006). The generalized processing sequence is shown in Table 1.

Interpretations of the seismic data were constrained through wellcontrol available from the Exxon State 32-1 deep exploration welllocated in the central Tucson Basin and the Phillips Petroleum Co.Mountain View State A-1 wildcat well located adjacent the Santa RitaMountains. Synthetic seismograms, based on sonic and density logs,were created for these wells and correlated to depositional-systeminterpretations by Eberly and Stanley (1978) and Houser and Gettings(2000).

Seismic reflectionprofiles image the Catalina detachmentas a zone ofreflectivity extending from the fault trace near the surface dipping downtowards the southwest to a depth of at least 9 km(Figs. 2 and3). Shallowreflectivity results from the juxtaposition of mylonitic lower-plate rocksagainst brittlely-deformed upper-plate Paleozoic and Mesozoic unitsand syntectonic sedimentary deposits. Deeper reflections from the fault

Table 1Generalized seismic processing flow for Tucson Basin 2D seismic data.

1. SEGY in2. Vibroseis correlation (pre-whitened with 1000 ms windowed AGC)a

3. Apply geometry4. Trace edit5. Mute by offset6. Datum statics averaged over 50 CDPs for velocity analysis7. Velocity analysis (every 25 CMPs)8. NMO9. Surface-consistent statics (max power)10. DMO and velocity re-pick11. Stack12. Finite difference migration13. Bandpass filter (10–40 Hz)14. Datum adjustment15. AGC (1500 ms)

a Exxon data only.


zone likely occur due to juxtaposition of upper-plate unmetamorphosedrocks above mylonitized lower-plate rocks. Broad corrugations on thedetachment surface (Davis,1980), parallel to the extension direction, canbe traced down into the subsurface to several kilometers depth (seeWagner and Johnson (2006) for details of subsurface fault geometries).Imaged in Line 2 (Fig. 2) is an incisement nappe on the detachmentsurface, which represents a fault-bounded sliver of upper or lower-platematerial. Towards the eastern end of the basin, immediately above thedetachment surface in the upper-plate, is a large hanging wall rollover-block (Schlische, 1995) that dips into the detachment surface. Seismicprofiles reveal that the upper-plate of the detachment system isextensively faulted. Numerous synthetic and antithetic faults dissectthehangingwall, creating a series of basins that served as catchments forsyntectonic sedimentation.

The Santa Rita fault is imaged in seismic profiles dipping towards thenorthwest at 15–20° (Johnson and Loy, 1992) striking nearly perpendi-cular to the Catalina detachment (Fig. 4). Up-dip projections of the faultsurface in the seismic data are coincident at the surface with theQuaternary fault scarps noted by Pearthree (1983) and Pearthree andCalvo (1987). The low-angle detachment noted by Hardy (1997) in the

Fig. 2. Seismic depth profile Line 2 with interpretations. The Catalina detachment dips towsedimentation are evident: synextensional sediments were deposited during active detaSeparating the two packages is the mid-Miocene unconformity (MMU) surface. See Fig. 1 for

foothills of the Santa Rita Mountains is along strike with the Santa Ritafault, but is 5–7 km south of seismic line 3. Aminimum down-dip offsettowards the northwest of several kilometers is evident by the offset of alarge slide-block in the hanging wall (Fig. 4). This fault block also tilts aseries of parallel reflections towards the east in a rollover structure.Wellcontrol reveals that these units are a thick sequence (N1200 m) ofinterbedded volcanic and sedimentary units in fault contact withgranitic rocks. The Santa Rita fault is clearly imaged down toapproximately 6 km beneath the southern Tucson Basin (Johnson andLoy,1992) (Fig. 3). Seismic reflectionprofiles indicate that the Santa Ritafault intersects the Catalina detachment obliquely along awest plungingline from a depth of approximately 2 km in the east to approximately5 km on the west (Figs. 3 and 5). These profiles indicate, however, thatthe Santa Rita fault does not extend through or offset the Catalinadetachment at depth. North–south-oriented reflection profiles suggestthat the Santa Rita fault merges with the Catalina detachment beneaththe central basin.

Seismic sequence analysis of the Tucson Basin data reveals twomain phases of sedimentation within the Tucson Basin. Synexten-sional sedimentation records faulted, tilted units in small-scale proto-basins found in the deepest part of the Tucson Basin (Figs. 2 and 3),underlain by voluminous volcanic units. K–Ar dates from a tuff unitnear the top of the volcanic sequences reveal an age of ~27 Ma (Eberlyand Stanley, 1978). Synextensional units show fanning dips, fromsteeply dipping at the base to subhorizontal at the top. Additionally,these units are truncated by high-angle normal faults along theiredges. These units are interpreted as Oligocene to Miocene synexten-sional sedimentary rocks deposited in the hangingwall of the evolvingCatalina detachment. Well control from Exxon State 32-1 in thecentral basin indicates clast compositions to be mostly of locallyderived volcanics and Mesozoic sedimentary units. Bounding the topof these units is a major regional unconformity that is evidentthroughout the basin in the reflection profiles. This unconformityseparates the highly disturbed units below from relatively undis-turbed units above. It is likely that this unconformity represents aperiod of time from 20 Ma, the cessation of active detachment faultingon the Catalina detachment, to 10–14 Ma based on K–Ar radiometricages associated with a volcanic sequence in close proximity to this

ards the west beneath the basin and is dashed where inferred. Two phases of basinchment faulting, Miocene basin fill was deposited after core complex emplacement.line location. Intersections with crossing seismic profiles are noted along top of section.

Fig. 3. Seismic depth profile Line 4 with interpretations. The Santa Rita fault extends from the southern end of the section dipping towards the north where it terminates against theCatalina detachment beneath the center of the Tucson Basin. See Fig. 2 for explanation of abbreviations. See Fig. 1 for line location. Intersections with crossing seismic profiles arenoted along top of section.


surface in various southern Arizona locales (Eberly and Stanley, 1978;Menges and Pearthree, 1989; Dickinson, 1991; Kruger et al., 1995).

Miocene sedimentation involved generally continuous, relativelyundisturbed sedimentary sequences that symmetrically drape thebasin. These sedimentary deposits signify the end of major extensionassociated with the Catalina detachment and record the gentlesubsidence and filling of the central Tucson Basin beginning after10–14 Ma associatedwith the local onset of Basin and Range extension(Wagner and Johnson, 2006). These sedimentary rocks are comprisedof largely mylonitic granite clasts derived from the adjacent Catalina

Fig. 4. Seismic depth profile Line 3 with interpretations. The Santa Rita fault is evident extenTertiary volcanics are imaged in the hanging wall of the fault system. See Fig. 1 for line loca

core complex. Thickest in the middle of the basin, these units onlapthe basin edges, and are progressively tilted with increasing depthattesting to the overall structural depression of the central basin. Thedeepest units are constrained to the basin's center, with progressivelyshallower units covering an increasingly larger aerial extent.

3. Finite-element modeling

Seismic profiles indicate that the Santa Rita fault merges with theCatalina detachment beneath the central basin. Additionally, the

ding from the eastern side of the profile dipping down towards the west. East dippingtion. Intersections with crossing seismic profiles are noted along top of section.

Fig. 5. Seismic depth profile Line D5 with interpretation. The Santa Rita fault and Catalina detachment merge beneath the central Tucson Basin. The Catalina detachment flattens outand continues towards the south where it begins to dip southward again along the edge of the profile. The flat section imaged on the Catalina detachment is due to this line obliquelycutting part of a broad corrugation feature on the detachment surface. See Fig. 1 for line location. Intersections with crossing seismic profiles are noted along top of section.


distribution of relatively undisturbed mid-Miocene and youngersedimentary sequenceswithin the basin above the Catalina detachmentand Santa Rita fault suggests that major motion on these two systemsoccurredprior to themid-Miocene. These constraints place the two faultsystems active approximately synchronously. The difference in exten-sion directions between the Catalina detachment and Santa Rita fault isproblematical, however, given that the strike of Santa Rita fault is nearlyparallel to major extension directions in the area, strike-slip motionwould be expected on the fault. Of importance, however, is the locationof the Santa Rita fault, in an area that borders zones of differing amountsof crustal extension and strength, with a zone to the north of extremecrustal extension and weakened crust across the Catalina/Rinconmetamorphic core complex, and an area to the immediate south thathas experienced only moderate crustal extension.

In order to understand how the Santa Rita fault could be accom-modating oblique strain from the Catalina detachment system, a finite-element approach was adopted. As described below, this approachprovides constraints on regional stress field orientations associated withdifferential extension. From this, an explanation for the unusualorientation of the Santa Rita fault is that it is the result of a perturbationof the regional stress field in response to large extension and decoupledcrust across the Catalina metamorphic core complex. The Santa Rita faultacts to accommodate the large strain around the southern end of theCatalina metamorphic core complex, diffusing the strain by rotatingmaterial in from the west and south.

3.1. Model construction

Finite-element analysis allows the modeling of the behavior of acomplicated systemby breaking it up into amesh ofmany simple shapeseach assigned realistic physical properties and rules for interaction.Evaluation of the changes in mesh geometry and in associated physicalproperties provides insights to actual deformation processes. Theapproach used here is to model in map view how a two-dimensionalelasticmaterial underextremeextensional stresses deformsandhow thestate of stress is perturbedwhen a broad zone of weakness, representingthe Catalina detachment, is introduced. As this model is constructed inmap view the vertical state of stress is assumed to be constant.

In our analysis we utilized the ABAQUS suite of general-purposenonlinear finite-element analysis and visualization programs (Hibbitand Sorenson, 2003). A two-dimensional model consisting of ahomogenous elastic material undergoing uniform extension was used

to study changes in stresses and displacements in proximity to a zone ofweakness representing a detachment fault in the upper crust. While anoversimplification of actual crustal properties, an elastic rheology issufficient for understanding thefirst-orderdistributionof stresses in thissystem, especially prior tomajor deformation. An elastoplastic rheologyis also employed to study the location and geometry of failure expectedin a brittle material; the material undergoes elastic deformation until adefined elastic limit is reached, at which point failure in the systemoccurs.

For simplicity in model construction, the map view finite-elementgrid was rotated approximately 30° clockwise, with respect to theactual geometry of the Catalina core complex, to allow themodeling ofregional extension as purely left–right (Fig. 6). The finite-element gridextends 560 km left to right (corresponding to NE–SW) and 320 kmtop to bottom (corresponding to NW–SE); the area of primary interestis only 100 km by 60 km. This is sufficiently large to ensure that anystress perturbations around the area of interest have equalized by themodel's edges and also ensures no far-field boundary conditions affectthe center of the model. The grid includes approximately 2800 nodesand 5300 triangular elements, with an increased mesh density nearand within the area of interest. Both triangular and rectangular meshelements were tested, and both produced similar model results.Triangular elements were ultimately used, as they proved moreadaptable to generating varying mesh densities across the model.

Constant-displacement boundary conditions were applied to theedges of the model. The top and bottom edges of the model areallowed to move in the horizontal direction, parallel to extension, butprevented from moving vertically. This produces symmetry about thetop and bottoms of the model, creating, in effect, a mirror image ofstresses across these boundaries. Constant-velocity displacements areapplied to the left and right edges of the model, directed in thenegative and positive directions, respectively, to simulate regionalextension. Typical continental crust values (Turcotte, 2002.) used formaterial parameters in the model are a Young's modulus of 70 GPaand Poisson's ratio of 0.3. A plastic failure of 4 MPa, similar to that ofgranite, is assigned to selected models to allow material failure inthese systems (Brace and Kohlstedt, 1980; Bott, 1997). As this modeldoes not attempt to account for any gravitational or inertial forces,defining a density for this material was not necessary.

The interpretations and conclusions in this study are based on theorientation andmagnitudes of principal stresses and failure is predictedusing von Mises stresses (Hibbit and Sorenson, 2003). Principal stress

Fig. 6. Complete finite-element grid with the region of interest shown zoomed in on the bottom grid. Note that the entire regionwas rotated approximately 30° clockwise to allow forease of modeling along “natural” axes. Horizontal arrows represent a constant-displacement boundary condition; rollers on the top and bottom indicate the boundary is fixed in thetop-to-bottom direction but allowed to move freely left-to-right. The detachment fault is modeled either as a physical break in the model or as a zone of weaker material.


orientations prove useful in understanding how the regional stress fieldis perturbed in response to the introduction of a feature representing adetachment fault. The von Mises stress, also called the equivalent oreffective stress or the maximum distortion energy criterion, combinesthe modeled deviatoric stresses in two or three dimensions and

expresses them as a scalar. Higher von Mises stresses are predictors oflocalized failure in a model as determined by the set plastic failurecriterion.

Several different model geometries were tested to determine anappropriate method for modeling an extensional system in map view


(Fig. 7). Our goal was to select a model that represents the first-ordertectonic features and stresses in the area without creating instabilitiesin the model. One approach was to model the Catalina detachment asa physical break in the crust. This is commonly done for modelingextensional systems in cross-section (Melosh and Williams, 1989;Bott, 1997; Mohapatra and Johnson,1998). This approach proved to beinappropriate for map view because extension produced a gap in themodel leading to extreme stresses at the tip of the fault.

An alternate approachwas tomodel the detachment as a broad zoneofweakermaterial representingexhumed lower-platematerial. Stressesare still transmitted across the fault zone, but at a much lower mag-nitude than in the surroundingmedium. This alleviated the problems ofgaps being created in themodel and also lessened stress concentrationsat tip points. This is a plausible way to model a detachment, as somestresses would be expected to transmit across a fault plane. Varying thewidth of the zone of weakness did not show any significant qualitativechange inmodel behavior around the area of interest. Indeed, allmodelsinvolving some inherent weakness across an extensional settingproduced similar stress field changes around that weakness. A finalrefinement was to round the sharp corner of the fault zone to alleviatestress concentrations (Fig. 7d).

3.2. Model results

Elastic solutions for the model represent the state of the modelafter 3 km of extension has occurred across the detachment fault. Thisrepresents the early stages of extension in the system. While asignificant displacement, this only represents 10% of the totalextension across the Catalina detachment. Attempts to represent the

Fig. 7. Principal stress plots used in testing four different model geometries for stability. a) Mproduced undesirable concentrated stresses at the tip of the fault. b) Modeling the detachmenc) Modeling the detachment as a zone of weakness extending to the edge of the model. This pmanipulate in ABAQUS. d) Refinement of the previousmodel by rounding over the corner ongiving similar principal stress orientations to all three of the previous models.

total estimated 30 km of extension across this system produce un-stable results.

As the grid undergoes extension, stresses build nearly uniformlyacross the model. The elastic model shows that elastic stresses arehighest around the bend in the weakened area representing thefootwall of the detachment system (Fig. 8). This may be due partly tomodel geometry effects, but is most likely a real feature of the systemas this is in the area of transition fromweak to strong material in linewith the regional stresses. Stresses are also high in the areaimmediately south (to the left and downward in the model) of thefault bend and also in the area east of (below in themodel) the zone ofweakness. The footwall of the system, represented by weakermaterial, readily deforms and therefore has much lower elastic stressbuild-up. Away from the detachment fault, primary principal stressesare oriented parallel to the extension direction as expected, but nearthe detachment, extension in the system produces a rotation of theprimary principal stresses around the end of the fault (Fig. 9).Northeast (to the right side in the model) of the detachment zone, thestress field begins to deflect from the regional stresses. Progressingaround the bend in the fault, extensional stresses are deflected bynearly 60° clockwise. The deflected stress field continues around thefront of the zone of weakness and gradually returns to normalapproximately 50 km from the front of the fault zone.

With plastic failure incorporated into the elastic model, failureunder extension in the system occurs very early (Fig. 10). The plasticfailure criterion is met within the first 50 m of extension in the model.Not unexpectedly, plastic failure occurs first in the area near the bendin the weak zone, where elastic stresses build up most rapidly. Zonesof failure also form east and southeast of (to the right of and below)

odeling the detachment as a physical break in the crust with no initial width. Extensiont as a zone of weakness with finitewidth. This helped to alleviate concentrated stresses.roduced very similar results to the previous model after extension, but proved easier to

the detachment. This helped to lower stress concentrations around the corner, while still

Fig. 8. Von Mises stresses due to extension on the system. Note that highest stresses build up around the bend in the detachment and below the zone of weakness representing thedetachment. The Catalina detachment and Santa Rita fault approximate locations are plotted on this figure for reference.


the zone of weakness. Two lobes of high stress extend out from thezone of failure, one towards the south–southwest (bottom left of themodel) and the other towards the east (bottom right of the model).The high-stress lobe that extends to the south–southwest correspondsto initiation and subsequent slip on the Santa Rita fault as discussedbelow. Association of a dominant structural feature with the high-stress lobe that extends towards the east from the zone of failure is notas clear. Although this area coincides with the Martinez Ranch fault,this fault formed much later (12–6 Ma) than the Santa Rita fault(Davis et al., 2004). A more direct temporal correlation exists withlocalized intense faulting, deformation and deposition of a thicksequence of Oligocene–mid-Miocene conglomerates, sandstonesand interbedded volcanic rocks in the Cienega Basin (Dickinson, 1991;

Fig. 9. Principal stress plots after extension. Primary principal stress orientations should be ordistorted around the zone of weakenedmaterial representing the detachment fault and coreof slip would be possible across the Santa Rita fault zone even given its oblique orientation

Richard andHarris,1996), south of the SantaCatalina–RinconMountains(Fig. 1).

4. Interpretation and discussion

By combining seismic reflection data, well control, surface geologicinformation and integrating these with finite-element model resultswe can describe the tectonic evolution of the Catalina core complexand also the along-strike deformation of the upper-plate and the roleof the Santa Rita fault system in accommodating the transfer oflocalized strain into a lesser deformed region (Fig. 11).

From geologic control, extension in the area began around 30 Ma,with a middle-to-late Oligocene pulse of arc magmatism preceding

iented nearly horizontal (left-to-right) due to stretching but become progressivelymorecomplex. Extensional stresses become rotated such that a significant normal componentto the regional stress field.

Fig. 10. Von Mises stress map with plastic failure criterion applied. Plastic failure occurs first in area around corner in detachment system. Two lobes of high initial failure extend outfrom the corner, one down and towards the right and a smaller lobe down and to the left. The orientation of this smaller lobe suggests that failure on the Santa Rita fault may havebeen the result of this perturbed stress field.


detachment faulting, which blanketed the region in lava flows andash-fall units (Eberly and Stanley, 1978; Spencer and Reynolds, 1989;Dickinson, 1991). At around this time, crustal extension becamefocused around the evolving Catalina detachment system. Synexten-sional sediments record the brittle deformation of the hanging wall

Fig. 11. Schematic model of fault interactions overlaid with stress vectors derived fromfinite-element models. Beneath the central Tucson Basin, the Santa Rita fault mergeswith the Catalina detachment. Normal motion occurs on the Santa Rita fault because ofthe deviation of the regional stress field around the Catalina detachment.

above the detachment system, dissecting the newly depositedvolcanics along a series of high-angle normal faults and creating aseries of small proto-basins. Concurrent with extension was the upliftand emplacement of the Catalina core complex into the upper crust asindicated by thermochronological analysis.

South of the Catalina detachment and core complex, along strike ofthe system, upper-plate deformation was accommodated by thenewly active Santa Rita fault. Although there are no absolute ageconstraints on the timing of motion on the Santa Rita fault, severalfactors indicate that activity began around this time: (1) themaximum age of faulting can be constrained by the faulted andsubsequently tilted volcanic sequences imaged in the seismic andsampled through well control. These volcanics are consistent withvoluminous Tertiary volcanic sequences noted beneath the TucsonBasin and in various locales around southern Arizona that immedi-ately precede detachment faulting (Eberly and Stanley, 1978; Spencerand Reynolds, 1989; Dickinson, 1991). (2) Although evidence suggeststhe Santa Rita fault may still be active, major motion on the Santa Ritafault is likely not younger than that of the Catalina detachment, giventhat seismic profiles indicate the Santa Rita fault merges with theCatalina detachment and does not appear to offset it beneath thebasin. (3) Finite-element modeling predicts that extension on theCatalina detachment and associated core complex emplacementwould have created increased stresses in the upper-plate of thedetachment system to the south (Fig. 11) and modeled plastic failuresuggests that early failure in the hanging wall block was likely.

Extension on the Catalina detachmentwas directedmainly NE–SW(Dickinson, 1991; Davis et al., 2004), consistent with extensiondirections in the area. Motion on the Santa Rita fault was directedobliquely to the Catalina detachment. From seismic profiles the strikeof the Santa Rita fault is nearly perpendicular to that of the Catalinadetachment. The exact direction of motion on the Santa Rita faultcannot be determined from the seismic profiles (i.e., purely dip-slip orstrike-slip or some component of the two), however the presence of alarge rollover anticline in the hanging wall suggests that a significantcomponent of movement was down-dip, or towards the WNW. Hardy(1997) noted top-to-the-west movement on the mid-Tertiary detach-ment he mapped just to the south of the main seismic profiles. Finite-


elementmodels confirm that the stressfieldwould have beenperturbedsuch that a large componentof normal slipwould have been favorable inthe region of the Santa Rita fault, even given the almost parallelorientation of the Santa Rita fault to the regional stress field. Modeledprimary principal stress orientations in the area of the Santa Rita faultmatch closely with kinematic indicators on Santa Rita fault exposures.

Upper-plate deformation continued on the Catalina detachmentsystem until ~20 Ma andmotion on the Santa Rita fault had also ceasedby this time. Miocene sedimentation begins, after a significant tectonichiatus, at approximately 10–14 Ma. This sedimentation attests to asignificant change in tectonic style, with low-angle, large-magnitudedetachment faulting giving way to high-angle, low-magnitude Basin-and-Range-style faulting (Wagner and Johnson, 2006). These sedimentsrecord that themainTucson Basin began to subside symmetrically alonga N–S axis, withmaximum subsidence concentrated west of the surfacetrace of the Catalina detachment and northwest of the Santa Rita fault.

5. Conclusions

In this study, analysis of a grid of 2-D seismic reflection profiles inthe Tucson Basin, coupled with well control and geological informa-tion from the Santa Catalina–Rincon Mountains and the Santa RitaMountains helped to constrain the along-strike deformation andtiming of the upper-plate of a low-angle detachment system andassociated metamorphic core complex. Early extension in the area,beginning in the middle-to-late Oligocene, focused on the Catalinadetachment. Brittle deformation of the upper-plate of the detachmentsystem was concurrent with core complex uplift and emplacement.

Immediately south of the Catalina detachment, along strike of thesystem and in the upper-plate, is the Santa Rita fault. Seismic profilesreveal that the Santa Rita fault is linked to Catalina detachmentbeneath the central Tucson Basin and basin sedimentological evidencerequires nearly synchronous movement on these faults. Finite-element models predict early mechanical failure of the upper-plateof the detachment system to the south of the Catalina core complex.Additionally, modeled principal stresses around the detachment showa nearly 60-degree rotation of the stress field from the regionalstresses. This is consistent with the oblique orientation of the SantaRita fault to the regional stress field.

These models suggest that the oblique motion on the Santa Ritafault is the result of a perturbation in the regional stress field by theCatalina detachment and the associated failure of the upper-platefrom the extreme crustal extension resulting from core complexemplacement. Additionally, the Santa Rita fault is acting to partitionstrain between zones of differing crustal extension, with a zone to thenorth of extreme crustal extension, on the order of 100% or moreacross the Catalina/Rincon metamorphic core complex, transitioningto a zone immediately south that has experienced only moderatecrustal extension. This along-strike upper-plate deformation wouldseem to be a necessary component of concentrated crustal extension.

Acknowledgments

We thank Randy Richardson, Bill Dickinson, Charles Ferguson, JonSpencer and Robert Krantz for helpful discussions and insights. Twoanonymous reviews also helped us in making improvements to thepaper. We are grateful to ConocoPhillips and ExxonMobil for releasingthe seismic data used in this study to the University of Arizona.Seismic processing and interpretation software were provided byLandmark Graphics Corporation through the Landmark UniversityGrant Program. Funding for this project was provided in part by theNational Science Foundation from grants EAR-8804667 and EAR-8817241. Financial support for F. Wagner was provided in part byConocoPhillips, BP, ChevronTexaco, the University of Arizona GalileoCircle and from the H. Wesley Peirce Memorial Scholarship.

References

Block, L., Royden, L.H., 1990. Core complex geometries and regional scale flow in thelower crust. Tectonics 9 (4), 557–567.

Bott, M.H.P., 1997. Modeling the formation of a half graben using realistic upper crustalrheology. Journal of Geophysical Research-Solid Earth 102 (B11), 24605–24617.

Brace, W.F., Kohlstedt, D.L., 1980. Limits on lithospheric stress imposed by laboratoryexperiments. Journal of Geophysical Research 85 (NB11), 6248–6252.

Buck, W.R., 1991. Modes of continental lithospheric extension. Journal of GeophysicalResearch-Solid Earth 96 (B12), 20161–20178.

Coney, P.J., 1980. Cordilleran metamorphic core complexes; an overview. In: Max Jr., D.Crittenden, Peter, J. Coney, George, H.Davis (Eds.), Cordilleran Metamorphic CoreComplexes. Memoir— Geological Society of America. Geological Society of America(GSA), Boulder CO, United States, pp. 7–31.

Coney, P.J., Harms, T.A., 1984. Cordilleran metamorphic core complexes — Cenozoicextensional relics of Mesozoic compression. Geology 12 (9), 550–554.

Crittenden Jr., M.D., Coney, P.J., Davis, G.H., 1980. Cordilleran metamorphic corecomplexes. Memoir — Geological Society of America, vol. 153. Geological Society ofAmerica (GSA), Boulder, CO, United States. 490 pp.

Davis, G.H., 1980. Structural characteristics of metamorphic core complexes, southernArizona. In: Crittenden Max Jr., D., Coney Peter, J., Davis George, H. (Eds.),Cordilleran Metamorphic Core Complexes. Memoir — Geological Society ofAmerica. Geological Society of America (GSA), Boulder, CO, United States, pp. 35–77.

Davis, G.H., Constenius, K.N., Dickinson, W.R., Rodriguez, E.P., Cox, L.J., 2004. Fault andfault-rock characteristics associated with Cenozoic extension and core-complexevolution in the Catalina–Rincon region, southeastern Arizona. Geological Societyof America Bulletin 116 (1–2), 128–141.

Dickinson,W.R.,1991. Tectonic settingof faultedTertiary strata associatedwith theCatalinacore complex in southern Arizona. Special Paper— Geological Society of America, vol.264. Geological Society of America (GSA), Boulder, CO, United States. 106 pp.

Drewes, H.D., 1972. Structural geology of the Santa Rita Mountains, southeast of Tucson,Arizona. U. S. Geological Survey Professional Paper. U. S. Geological Survey, Reston,VA, United States. 35 pp.

Eberly, L.D., Stanley, T.B., 1978. Cenozoic stratigraphy and geologic history ofsouthwestern Arizona. Geological Society of America Bulletin 89 (6), 921–940.

Faulds, J.E., Varga, R.J., 1998. The role of accommodation zones and transfer zones in theregional segmentation of extended terranes. In: Faulds James, E., Stewart John, H.(Eds.), Accommodation Zones and Transfer Zones; The Regional Segmentation ofthe Basin and Range Province. Special Paper — Geological Society of America.Geological Society of America (GSA), Boulder, CO, United States, pp. 1–45.

Fayon, A.K., Peacock, S.M., Stump, E., Reynolds, S.J., 2000. Fission track analysis of thefootwall of the Catalina detachment fault, Arizona; tectonic denudation, magma-tism, and erosion. Journal of Geophysical Research, B, Solid Earth and Planets 105 (5),11047–11062.

Hardy Jr., J.J., 1997. Superimposed Laramide and middle Tertiary deformations in thenorthern Santa Rita Mountains. Pima County, Arizona.

Hibbit, K., Sorenson, 2003. ABAQUS, version 6.4, Pawtucket, R.I.Hirschberg, D.M., Pitts, G.S., 2000. Digital geologic map of Arizona; a digital database

derived from the 1983 printing of the Wilson, Moore, and Cooper 1:500,000-scalemap. Open-File Report — U. S. Geological Survey. 2000. U. S. Geological Survey.Reston, VA, United States. Pages: 67. 2000.

Houser, B.B., Gettings, M.E., 2000. Stratigraphy and tectonic history of the Tucson Basin,Arizona, based on re-examination of cuttings and geophysical logs of the ExxonState (32)-1 well. Open-File Report— U. S. Geological Survey. 2000. U. S. GeologicalSurvey. Reston, VA, United States. Pages: 38, 2 sheets. 2000.

Johnson, R.A., Loy, K.L., 1992. Seismic-reflection evidence for seismogenic low-anglefaulting in southeastern Arizona. Geology 20 (7), 597–600.

Kruger, J.M., Johnson, R.A., 1994. Raft model of crustal extension; evidence from seismicreflection data in Southeast Arizona. Geology (Boulder) 22 (4), 351–354.

Kruger, J.M., Johnson, R.A., Houser, B.B., 1995. Miocene–Pliocene half-graben evolution,detachment faulting and late-stage core complex uplift from reflection seismic datain south-east Arizona. Basin Research 7 (2), 129–149.

Lavier, L.L., Buck, W.R., Poliakov, A.N.B., 1999. Self-consistent rolling-hinge model for theevolution of large-offset low-angle normal faults. Geology 27 (12), 1127–1130.

Lister, G.S., Davis, G.A., 1989. The origin of metamorphic core complexes anddetachment faults formed during Tertiary continental extension in the northernColorado River region, U.S.A. Journal of Structural Geology 11 (1/2), 65–94.

Melosh, H.J., Williams, C.A., 1989. Mechanics of graben formation in crustal rocks — afinite-element analysis. Journal of Geophysical Research-Solid Earth and Planets 94(B10), 13961–13973.

Menges, C.M., Pearthree, P.A., 1989. Late Cenozoic tectonism in Arizona and its impacton regional landscape evolution. In: Jenney, J.P., Reynolds Stephen, J. (Eds.),Geologic Evolution of Arizona. Arizona Geological Society Digest. Arizona GeologicalSociety, Tucson, AZ, United States, pp. 649–680.

Mohapatra, G.K., Johnson, R.A., 1998. Localization of listric faults at thrust fault rampsbeneath the Great Salt Lake Basin, Utah: evidence from seismic imaging and finiteelementmodeling. Journal of Geophysical Research-Solid Earth 103 (B5),10047–10063.

Nourse, J.A., Anderson, T.H., Silver, L.T., 1994. Tertiary metamorphic core complexes inSonora, northwestern Mexico. Tectonics 13 (5), 1161–1182.

Pearthree, P.A., 1983. Distribution, recurrence, and possible tectonic implications of lateQuaternary faulting in Arizona / by Philip A. Pearthree, Christopher M. Menges,LarryMayer. In: C.M.Menges and L. Mayer,1951— (Editors), Neotectonic analysis ofArizona. Arizona Bureau of Geology and Mineral Technology, Tucson.

Pearthree, P.A., Calvo, S.S., 1987. The Santa Rita fault zone; evidence for large magnitudeearthquakes with very long recurrence intervals, Basin and Range Province ofsoutheastern Arizona. Bulletin of the Seismological Society of America 77 (1), 97–116.


Rehrig, W.A., Reynolds, S.J., 1980. Geologic and geochronologic reconnaissance of anorthwest-trending zone of metamorphic core complexes in southern and westernArizona. In: Crittenden Max Jr., D., Coney Peter, J., Davis George, H. (Eds.), CordilleranMetamorphic Core Complexes. Memoir — Geological Society of America. GeologicalSociety of America (GSA), Boulder, CO, United States, pp. 131–157.

Richard, S.M., Harris, R.C., 1996. Geology and geophysics of the Cienega Basin area, Pimaand Cochise Counties, Arizona. Open-File Reoprt 96-21. InArizona Geological Survey.37 pp.

Rosenbaum, G., Regenauer-Lieb, K., Weinberg, R., 2005. Continental extension: fromcore complexes to rigid block faulting. Geology 33 (7), 609–612.

Rutledge, J.T., 1984. A shallow seismic refraction survey over a late Quaternary faultscarp west of the Santa Rita Mountains, Arizona. M.S. Thesis, University of Arizona,Tucson, 93 pp.

Schlische, R.W., 1995. Geometry and origin of fault-related folds in extensional settings.Aapg Bulletin-American Association of Petroleum Geologists 79 (11), 1661–1678.

Scott, R.J., Lister, G.S., 1992. Detachment faults — evidence for a low-angle origin.Geology 20 (9), 833–836.

Spencer, J.E., Chase, C.G., 1989. Role of crustal flexure in initiation of low-angle normalfaults and implications for structural evolution of the Basin and Range Province.Journal of Geophysical Research-Solid Earth and Planets 94 (B2), 1765–1775.

Spencer, J.E., Reynolds, S.J., 1989. Middle Tertiary tectonics of Arizona and adjacentareas. In: Jenney, J.P., Reynolds Stephen, J. (Eds.), Geologic Evolution of Arizona.Arizona Geological Society Digest. Arizona Geological Society, Tucson, AZ, UnitedStates, pp. 539–547.

Turcotte, D.L., 2002. Geodynamics/Donald L. Turcotte, Gerald Schubert. CambridgeUniversity Press, Cambridge. New York.

Wagner, F.H., Johnson, R.A., 2006. Coupled basin evolution and late-stage metamorphiccore complex exhumation in the southern Basin and Range Province, southeasternArizona. Tectonophysics 420, 141–160.

Wernicke, B.P., 1985. Uniform-sense normal simple shear of the continental lithosphere.Canadian Journal of Earth Sciences= Journal Canadien des Sciences de la Terre 22 (1),108–125.

Wernicke, B.P., Axen, G.J., 1988. On the role of isostasy in the evolution of normal faultsystems. Geology (Boulder) 16 (9), 848–851.

Documents

Along-strike upper-plate deformation in response to