Great Valley Nevada Sierra Peninsular Ranges

Jacobson et al.

PSP 121°W Faults Pelona-Orocopia-Rand Schists Forearc sampling areas Ef Elsinore BR Blue Ridge AT Atascadero Diablo DP EHf East Huasna CD Castle Dome Mountains BL Ben Lomond Mountain Great Gf Garlock CH SE Chocolate Mountains CA Cambria PP BL Nf Nacimiento EF East Fork CL Coalinga 37°N Rf Rinconada MP Mount Pinos CP Cajon Pass SAf San Andreas NR Neversweat Ridge DP Del Puerto Canyon Range SGf San Gabriel OR Orocopia Mountains IR Indians Ranch SGHf San Gregorio-Hosgri PR Portal Ridge LP La Panza Range PL SS SJf San Jacinto RA Rand Mountains NB Northern Baja California SL PS SYf Santa Ynez SC Santa Catalina Island OR Orocopia Mountains CL SE San Emigdio Mountains PL Point Lobos IR Sierra 36°N Valley SP Sierra Pelona PM Pine Mountain Block Rf Nf SS Sierra de Salinas PP Pigeon Point TR Trigo Mountains PS Point Sur Nevada PSP Point San Pedro CA AT SA Northern Santa Ana Mts. SGHf LP RA SC Santa Catalina Island SAf SD San Diego Gf SG San Gabriel block 35°N Salinian block EHf SAf SMM SGHf Nf SE SL NW Santa Lucia Range MP SM Santa Monica Mts-Simi Hills SR Nacimiento blk PM SG PR SMI San Miguel Island SYf SR SMM Sierra Madre Mountains Nf BR SY SY SP SG CP SR San Rafael Mountains CTR WTR EF SY Santa Ynez Mountains SMI SM 34°N SGf ETR McCoy Mts. Fm. WTR SA Peninsular OR Accretionary complex SC N SJ Forearc basin f SAf TR CD Ranges NR Magmatic arc and wall rocks Ef CH CA Pelona-Orocopia-Rand Schists 100 km SD MEX AZ McCoy Mountains Formation 118°WNB 115°W Jacobson et al., GSA Bull, 2010 Figure 1. Generalized late Mesozoic–early Cenozoic geology of central to southern California and adjoining areas, based largely on Jennings (1977). Some outcrops of arc rocks and older wall rocks were omitted from southeasternmost California and southwesternmost Arizona to emphasize outcrops of Pelona-Orocopia-Rand schist and McCoy Mountains Formation. Inset delineates the Salinian and Nacimiento blocks. The Nacimiento fault is indicated by a heavier line weight than the other faults. We follow the interpretation of Hall et al. (1995) and Dickinson et al. (2005) that the Nacimiento fault has been offset ~45 km in a dextral sense by slip on the combined Rinconada–East Huasna fault. Sample localities for both Pelona-Orocopia-Rand schists and forearc units are indicated. Repeated labels (SG, SY, and SR) indicate the grouping of samples collected over relatively large areas. Abbreviations not deﬁ ned in the ﬁ gure: AZ—Arizona; CA—California; CTR—central Transverse Ranges; ETR—eastern Transverse Ranges; MEX—Mexico; WTR—western Transverse Ranges.

involved remain unclear. Previous interpreta- 1978; Ehlig, 1981; Jacobson et al., 1988). These Orocopia-Rand schists has long been debated tions include thrusting of 150–200 km (Silver, rocks broadly resemble the Franciscan Com- (Haxel et al., 2002; Grove et al., 2003), although 1983; Hall, 1991; Barth and Schneiderman, plex, but instead of being positioned outboard most workers now favor a model involving low- 1996; Saleeby, 1997; Ducea et al., 2009), dex- of the forearc basin, they sit in structural contact angle subduction of the Farallon plate during the tral strike slip of 2000 km or more (Page, 1982; directly beneath the Cretaceous marginal batho- Laramide orogeny (Crowell, 1968, 1981; Yeats, Champion et al., 1984), and sinistral strike slip lith and adjoining cratonal areas of southeastern 1968; Burchﬁ el and Davis, 1981; Dickinson, of 500–600 km (Dickinson, 1983; Seiders and California and southwestern Ari zona. Poten- 1981; Hamilton, 1988; May, 1989; Malin et al., Blome, 1988; Dickinson et al., 2005). tially equivalent rocks, recovered as xenoliths, 1995; Jacobson et al., 1996, 2002, 2007; Wood A second key feature of southern and coastal occur still farther east in the Four Corners region and Saleeby, 1997; Yin, 2002; Grove et al., central California is represented by the Pelona- of Arizona, New Mexico, Utah, and Colorado 2003; Saleeby, 2003; Kidder and Ducea, 2006; Orocopia-Rand schists (Fig. 1; Haxel and Dillon , (Usui et al., 2006). The origin of the Pelona- Saleeby et al., 2007).

486 Geological Society of America Bulletin, March/April 2011 Exhumation of the Orocopia Schist and associated rocks of southeastern California 5

Figure 3. Geology of the Orocopia Mountains and vicinity (after Crowell, 1975). Contacts within the northwestern half of the upper plate of the Orocopia Mountains detachment fault were modiﬁ ed based on this study. Those in the southeastern half were modiﬁ ed based on the work of Ebert (2004). Lines A–A′ to C–C′ indicate locations of cross sections illustrated in Figure 4. PC—Painted Canyon. Jacobson et al., GSA SP 419, 2007

3. The Orocopia Mountains detachment fault is folded by the schist and upper plate (Orocopia Mountains detachment fault) A reminder of typical geology. Schists are meta-turbiditesthe Chocolate Mountainswith some anticlinorium. metabasite, Consequently, chert and serpentinite. dating the andInverse placed mmic a greater gradients emphasis often onfound relating but thermomodern chronologic exposure is typically under extensional fault systems. “The Pelona-Orocopia-Rand schists consist of 90% or more quartzo-feldspathic schist presumably derived from turbidite sandstone (Ehlig, 1958, 1981; Haxel and Dillon, 1978; Jacobson et al., 1988, 1996; Haxel et al., 1987, 2002). The schists also include up to 10% metabasite, along with minor amounts of Fe-Mn metachert, marble, serpentinite, and talc-actinolite rock. Metamorphism occurred under conditions of moderately high pressure relative to temperature. Mineral assemblages lie mostly within the greenschist and albite-epidote amphibolite faciesfault but locally provides extend into the epidote-blue-a maximum schist and ageupper amphibolite for the facies anticlinorium, (Ehlig, 1981; Haxel and Dillon,which 1978; hasGraham andresults Powell, 1984; to Jacobson fabric and Sorensen, elements 1986; Jacobson within et al., 1988; the Kidder schist. and Ducea, In2006). contrast, Inverted metamorphic Ebert zonations are typical. Initial emplacement of the schists is attributed to a Late Cretaceous–early Cenozoic low- angle fault system referred to as the Vincent–Chocolate Moun- tains thrust (Haxel and Dillon, 1978). However, most present-day contacts between the schists and North American upper plate appear to be low-angle normal faults related to exhumation of the schists, either shortly after underthrusting or during middle Cenozoic extension (Frost et al., 1982; Hamilton, 1988; Jacobson et al., 1988, 2002, 2007; Malin et al., 1995; Wood and Saleeby, 1997; Haxel et al., 2002).”-Jacobson et al., 2010 implications for the passive-roof thrust model of Yin (2002). et al. (2003), Ebert (2004), and Ebert and Yin (2004) focused To investigate the above issues, we performed 40Ar/39Ar on detailed mapping within the upper plate and analysis of the step-heating experiments with hornblende, muscovite, biotite, C l e m e n s W e l l f a u l t . and K-feldspar to constrain the temperature-time histories for Whereas the vast majority of the new data obtained in this both the schist and upper plate in the Orocopia Mountains. study come from the Orocopia Mountains, we also report 40Ar/39Ar Additional control is provided by several apatite ﬁ ssion track step-heating results for three new samples of K-feldspar from the ages, ion microprobe U-Pb dating of zircon from a Late Creta- Orocopia Schist of the Gavilan Hills. These new data were col- ceous leucogranite within the upper plate, and previous U-Pb lected to facilitate comparisons of thermal history between the dating of detrital zircon from the schist (Jacobson et al., 2000; two regions. Grove et al., 2003). The thermochronologic results were inte- grated with preexisting and new observations on the structural REGIONAL BACKGROUND and metamorphic evolution of the schist and upper plate. As detailed below, these data reveal a two-stage exhumation his- The Orocopia Schist of the Orocopia Mountains is exposed tory for the schist in the Orocopia Mountains very much like in a northwest-trending, postmetamorphic antiform that deﬁ nes that found in the Gavilan Hills. the northernmost culmination of the Chocolate Mountains anti- The present study concentrated on the western to northwest- clinorium (Figs. 3 and 4). Proterozoic to Cretaceous igneous and ern part of the Orocopia Mountains (Fig. 3). A similar investiga- metamorphic rocks of the upper plate crop out in a narrow belt tion has recently been conducted by Ebert et al. (2003), Ebert that sits above the schist along the Orocopia Mountains detach- (2004), and Ebert and Yin (2004) in the southeastern part of the ment fault. The structure, metamorphism, and thermochronology range. Our work focused more heavily on the contact between of the schist and upper plate are described in detail below. Downloaded from specialpapers.gsapubs.org on March 8, 2010

384 M. Grove et al.

depositional age for the graywacke protolith for all of the schists and reveal systematic regional variations in the depositional age and provenance of the protolith. When combined with 40Ar/39Ar cooling ages, our detrital zircon U-Pb data can be used to constrain the “cycling interval” during which the schist’s graywacke protolith was eroded from its source region, deposited, underthrust, accreted, and metamorphosed beneath crystalline rocks of southwestern North America. Combined with previously obtained data, new muscovite and biotite 40Ar/39Ar cooling ages from our detrital zircon samples clearly indicate that the cycling interval for erosion, deposition, underthrusting, accretion, and metamorphism varied systematically across the schist terrane. Schist located outboard of the craton in the southern Coast Ranges and northwestern Mojave Des- ert was emplaced well before schists were underplated beneath more cratonal (and progressively more southeasterly) positions in the Transverse Ranges and southeastern California and southwestern Arizona. This spatial-temporal relationship and the overall Laramide depositional age for the schist’s graywacke protolith must be explained by any model that seeks to describe the underplating of the schists beneath southwest North America.

CORRELATION OF THE SCHISTS

Early attempts to correlate the Pelona, Orocopia, Rand, Por- tal Ridge, and Sierra de Salinas Schists (Fig. 1) were based on their ﬁ eld and petrographic characteristics (Ehlig, 1958, 1968). A key observation is that relative proportions of the major rock types are broadly similar in all areas (~90–99% metagraywacke, 1–10% mid-ocean ridge basalt (MORB) metabasite, and trace amounts of Fe-Mn metachert, marble, and serpentinite; Haxel and Dillon, 1978; Haxel et al., 1987, 2002; Dawson and Jacob- son, 1989). Moreover, manyGrove of the etschist al., GSA bodies SP exhibit374, 2003 inverted metamorphic zonation; i.e., peak temperature of metamorphism Figure 2 Tectonic models for origin of Pelona and related schists. A: In decreases structurally downward (Ehlig, 1958; Graham and Eng- subduction model, schist’s graywacke protolith is deposited in trench, land, 1976; Jacobson, 1983b, 1995). In addition, the schists share All of these are compressional so not concernedsubducted, and with accreted how beneath these tectonically are eroded exhumed cratonal rocks. but a number environment of diagnostic mineralogic where featuresthey (Haxelwere and created. Dillon, Lithospheric mantle has been completely removed by this process. B: 1978): (1) poikiloblasts of sodic plagioclase in metagraywacke In forearc model, sediments in forearc basin are overthrust by magmatic that appear gray to black due to inclusions of graphite; (2) local arc; C: In collision model, schist’s protolith is overthrust by a northeast- directed allochthon. D: In backarc model, protolith of schist originates in centimeter-scale clots of fuchsite in metagraywacke; (3) centi- a backarc basin that is subsequently overthrust and closed. meter- to meter-scale lenses of talc-actinolite rock in metagraywacke; (4) sodic-calcic amphibole (winchite to barroisite) in low-grade metabasites; and (5) spessartine garnet, piemontite, stipnomelane, and sodic amphibole in Fe-Mn metachert. multigrain U-Pb analysis of zircons from the schist performed The approximate early Tertiary positions of schist outcrops by Silver et al. (1984) and James and Mattinson (1988) were are indicated in Figure 3. Correlation among the various schists equivocal with respect to the depositional age of the graywacke has been most readily accepted for the Pelona and Orocopia protolith. In contrast, our initial ion microprobe results clearly Schists, which were in close proximity prior to offset on the San indicated the presence of Late Cretaceous detritus (Jacobson et Andreas system (Fig. 3; James and Mattinson, 1988; Dillon et al., 2000; Barth et al., 2003a). In this paper, we present a much al., 1990). Less well established is the relationship of the Pelona– more comprehensive data set involving 40 samples that represent Orocopia schists to the Rand Schist, schist of Portal Ridge, and virtually all of the major schist exposures. This compilation, schist of Sierra de Salinas. which includes six samples from Jacobson et al. (2000) and Barth While the Rand Schist exhibits all the key lithologic attri- et al. (2003a), conﬁ rms a Late Cretaceous to earliest Tertiary butes of the Pelona and Orocopia Schists (Jacobson, 1995), it Jacobson et al.

Sierra Gf NV AZ PSP Nevada RA BL CA

SE Mojave Desert PR SGHf SAf SS M o g o l l o n H i g h l a n d s CP PL Rf SL Nf IR ? Nf LP PP SMM AT SG 100 km EHf MP CA SR SGHf SP BR OR SR Igneous Rocks Nf CA PS PM SGf EF AZ 55–85 Ma TR CD NR CH 85–100 Ma ? 100–135 Ma MEX Cretaceous undifferentiated SY 135–300 Ma N SM Proterozoic SA Peninsular SY Sedimentary (±Volcanic) Sequences SC and Metamorphosed Equivalents SJf Cretaceous-Eocene forearc basin Ranges Accretionary complex Pelona-Orocopia-Rand-Catalina Schists Figure 6 Jurassic-Cretaceous McCoy Mountains Fm. SD SMI Early Cretaceous Alisitos arc NB Paleozoic-Mesozoic

Figure 2. Pre–San Andreas palinspastic reconstruction of southern California, southwestern Arizona, andJacobson northwestern et al., GSA Mexico. Bull, Se2010e text and GSA Data Repository item for explanation (see text footnote 1). Abbreviations for faults and sampling localities are as in Figure 1. Inferred extension of Nacimiento fault west of the San Gregorio–Hosgri fault is based on ﬁ gure 10 of Dickinson et al. (2005). Truncated restored for late Cz deformation rectangular box indicates the approximate area of coverage of the panels in Figure 6.

from >90 Ma at the northwest end of the belt to more nearly parallel to the strike of the mar- Complex. The unit is most widely exposed on <60 Ma in the southeast. Signifi cant variations gin). Alternatively, A. Chapman and J. Saleeby Santa Catalina Island, where it forms a series in provenance are also observed from one end (2009, personal commun.) argue that rotation of fault-bounded slices of metasedimentary, of the belt to the other. of the upper-plate, batholithic rocks did not af- metavolcanic, and ultramafi c rocks (Woodford, In contrast to the northwest-southeast align- fect the structurally underlying schists and that 1960; Platt, 1975, 1976; Sorensen, 1988; Grove ment of schist exposures as a whole, the north- the current locations of the northwestern schists and Bebout, 1995; Grove et al., 2008). Meta- western bodies defi ne a southwest-northeast closely refl ect their relative distributions at the morphism ranges from upper amphibolite facies subbelt extending from the Sierra de Salinas time of underthrusting. This is a further compli- in the structurally highest unit to lawsonite- to the Rand Mountains. One possibility is that cation for reconstructing the Late Cretaceous– blueschist, lawsonite-albite, and albite-actinolite this deviation is an artifact of our failure to cor- early Cenozoic paleogeography of southern facies in the deepest parts of the section (Grove rect for oroclinal bending of the Sierran tail California and vicinity. et al., 2008). The Catalina Schist also occurs as and adjacent parts of the Mojave Desert and widespread submarine outcrops within the inner Salinian block (Kanter and McWilliams, 1982; Catalina Schist continental borderland, as minor outcrops on the McWilliams and Li, 1985). In this case, the belt The Catalina Schist is a moderately high- Palos Verdes Peninsula, and in the subsurface of northwestern schists would originally have pressure metamorphic terrane typically viewed of the Los Angeles basin (references in Grove been oriented approximately north-south (i.e., as the southern extension of the Franciscan et al., 2008).

488 Geological Society of America Bulletin, March/April 2011 Late Cretaceous–early Cenozoic tectonic evolution of southern California

RESULTS 10 Mio 20 Pelona-Orocopia-Rand and Metamorphic bio Catalina Schists 30 Olig mus hbd Aside from the fact that we did not previ- Detrital 40 ously consider the Catalina Schist, our results zircon Eoc are little changed from those obtained by Grove 50 erv al et al. (2003). We thus emphasize only those int points most relevant for comparison with the 60 Age (Ma) Pal forearc data. Cy cling Maast Slip on Nac flt 70 Protolith Age Camp One of the most critical parameters to de- 80 termine for any given schist body is the depo- Sant Con sitional age of the sandstone protolith, which 90 must lie within a time window bounded by the Tur Cen youngest reliable detrital zircon age and oldest 100 reliable metamorphic age (Fig. 3). Grove et al. Age/ SC SE PRRA SS MP SP BR EF OR CH TR CD NR (2003) referred to this time span as the “cycling Epoch Cat-San Emig Rand Pelona Orocopia interval,” because it encompasses erosion in North South 40 39 the source area, transport of that material to the Figure 3. Comparison of Ar/ Ar cooling ages and zircon U-Pb ages ≤100 Ma for the site of deposition, underthrusting beneath the Pelona-Orocopia-Rand-Catalina schists plotted by area (see Figs. 1, 2, and 4 for locations). arc, accretion to the overriding plate, and initial Zircon ages are from Barth et al. (2003), Grove et al. (2003, 2008), andJacobson this et al.,study. GSA Bull, Argon 2010 stages of exhumation. As observed by Grove ages are from Jacobson (1990), Jacobson et al. (2002, 2007), Barth et al. (2003), and Grove etConstraints al. (2003), the oncycling emplacement. interval of the Pelona- mmic ageset al. are (2003, cooling, 2008). detrital The Pelona-Orocopia-Rand zircons are deposition. schists are organized by the conventional Orocopia-Rand schists becomes younger from northwest-southeast grouping, i.e., Rand Schists in the northwest Mojave Desert and envi- northwest to southeast (Fig. 3). Also notable, rons, Pelona Schists in the central Transverse Ranges, and Orocopia Schists in southeast- the duration of the cycling interval increases ern California–southwestern Arizona. The exception is the Rand Schist of the San Emigdio to the southeast, although this may be at least Mountains, which we group with the Catalina Schist. Within the Pelona and Orocopia partly an artifact in the data. For example, we Schist groups, individual areas are ordered from northwest on the left to southeast on the obtained no hornblende 40Ar/39Ar ages from right. Northwest-southeast position for the Rand Schists is not clear owing to potential ro- the easternmost schist bodies (Fig. 3) and tation of the Sierran tail; these bodies were instead ordered from left to right by deposi- thus may not have captured the oldest cool- tional and metamorphic age. Diagonally ruled box indicates time of most likely movement ing ages in this region. In addition, igneous on the Nacimiento fault. Yellow band indicates the cycling interval, which, as discussed in 40 39 bodies younger than ca. 70 Ma are not com- the text, is not well constrained in the southeast. Also note that Ar/ Ar and zircon ages mon in the inferred provenance areas (Barth overlap for the Catalina Schist (the two types of ages have been separated slightly along et al., 2008a), making it diffi cult to constrain the x-axis to clarify this relation). The low grade of metamorphism for this unit (Grove 40 39 the maximum depositional age of sediments et al., 2008) suggests that the Ar/ Ar ages may be infl uenced by excess radiogenic argon younger than latest Cretaceous. For example, or lack of complete recrystallization of detrital mica during metamorphism. Abbreviations in delineating the cycling interval, we ignored for schist localities are same as in Figure 1. Other abbreviations are: bio—biotite; Camp— three early Cenozoic ages from the Orocopia Campanian; Cen—Cenomanian; Eoc—Eocene; hbd—hornblende; Maast—Maastrichtian; Schist due to their status as outliers (Fig. 3). Mio—Miocene; mus—muscovite; Olig—Oligocene; Pal—Paleocene; Sant—Santonian; However, it should be kept in mind that any or Tur—Turonian. all of these ages could be signifi cant. Despite the uncertainties, the schist protoliths clearly range from Turonian to at least as young as Variation in Detrital Zircon Populations as a taceous zircons, with lesser, but still signifi cant, middle Paleocene, representing a time span of Function of Depositional Age and Location proportions of Proterozoic and Jurassic ages. 30 m.y. or greater. Pie diagrams of detrital zircon ages for indi- In contrast, the central to southeastern schists, For purposes of comparison to the forearc vidual ranges and/or adjacent ranges are plot- which have depositional ages of Maastrichtian units, we divided the schists into four groups ted in Figure 4 using the palinspastic base of and younger, are characterized by progressively based on depositional age and geographic Figure 2. Probability density functions for the decreasing abundances of Early to middle Cre- location (Figs. 3 and 4). This contrasts with four regional groupings are illustrated in Figures taceous detrital zircons and increasing num- a threefold division utilized by Grove et al. 5B–5E. As pointed out by Grove et al. (2003), bers of Proterozoic and latest Cretaceous–early (2003). Our fourth category includes the detrital zircon patterns in the Pelona-Orocopia- Cenozoic ages. As concluded by Grove et al. Catalina Schist, which was not considered by Rand schists vary systematically from north- (2003), the patterns imply a shift from outboard Grove et al. (2003), and the Pelona-Orocopia- west to southeast, and thus with depositional to inboard sources with time (e.g., Fig. 2 and Rand schist of the San Emigdio Mountains, age of the protolith. The northwestern schists, following discussion). which Grove et al. (2003) grouped with the which have protolith ages of Turonian to Cam- Compared to the Pelona-Orocopia-Rand Rand schists. panian, are dominated by Early to middle Cre- schists, the zircon age distribution within

Geological Society of America Bulletin, March/April 2011 491 Jacobson et al.

San Emigdio Gf matching divisions for the schists. The overall Mts. (3/82) Sierra NV similarity between the two sample sets is im- Nevada AZ pressive. By analogy with the schists, the varia- CA tion from oldest to youngest parts of the forearc Mojave Rand Mts. section is considered to refl ect a transition from SGHf Desert SAf (5/166) outboard to inboard source areas. Portal Ridge Orocopia Pie diagrams of the zircon age distributions (1/37) Blue Ridge (2/70) plotted on a geographic base (Fig. 6) confi rm Rf Orocopia Mts. the temporal progression from marginal to Mt. Pinos (10/253) inboard sources and reveal signifi cant along- Sierra de Nf (5/125) Salinas (5/138) Nf Trigo Mts. Neversweat strike variations in provenance (see also Figs. (4/92) Ridge (2/66) EHf DR2–DR4 [see footnote 1]). Material derived Rand from east of the main Sierran–Peninsular SGHf SGf Ranges arc fi rst appeared at the continental 100 km margin in a few localities in the central part of N Pelona f the study area during Campanian–early Maas- ? trichtian time (Fig. 6B; Cambria and Pfeiffer Sierra Pelona Detrital zircon ages (7/199) slabs, Santa Ynez Mountains, Santa Monica Mountains, Simi Hills). With time, detritus 55–70 Ma East Fork 70–85 Ma (4/110) SE Chocolate Castle Dome originating from inboard sources became the 85–100 Ma Mts. (3/79) Mts. (4/77) dominant component in the central part of 100–135 Ma Catalina the area and was also delivered in signifi cant 135–300 Ma N Schist Peninsular (21/305) quantities both to the northwest and southeast >300 Ma Ranges (Figs. 6C and 6D). It is important to keep in mind that Figure 6 does not take into account displacement on the Figure 4. Pie diagrams of detrital zircon ages from the Pelona-Orocopia-Rand-CatalinaJacobson et al., GSA Bull, 2010 Nacimiento fault. Because movement appears schists plotted on the same palinspastic base as in Figure 2. The Nacimiento fault is indicated to have been over by the late Paleocene, the So where did this stuf come from? After all,by aa bunchheavier lineof reallyweight oldthan zircons the other sofaults. obviously Present-day not outcrops just a pileof Pelona- of ocean Orocopia- floor Eocenematerial reconstruction (Fig. 6D) should be in- Rand schists are shown in black. The Catalina Schist is inferred to restore beneath the dependent of this event. However, the preceding northern Peninsular Ranges (Crouch and Suppe, 1993). Mesozoic magmatic rocks and asso- time frames presumably require varying degrees ciated wall rocks of the Sierra Nevada and Peninsular Ranges batholiths are shown by the of correction, depending on the exact nature and plus pattern. Arc and wall rocks of the Mojave Desert and Salinian block are omitted for age of the Nacimiento fault. If this structure is clarity. Dashed blue outlines delineate three of the four schist groupings used for plotting dominantly a thrust (Page, 1970, 1981; Hall zircon results in Figure 5 (see also Fig. 3). The fourth group includes the spatially sepa- 1991; Saleeby, 2003), then displacement should rated Catalina Schist and schist of the San Emigdio Mountains (see text). Paired numbers have been largely normal to the length of the in paren theses indicate number of samples and number of zircon ages. margin. In this case, the along-strike variations evident in Figure 6 would not be signifi cantly altered. On the other hand, for 500–600 km of the Catalina Schist is geographically distinct pository item [see footnote 1]) and correspond sinistral slip (Dickinson, 1983; Dickinson et al., (Fig. 4); i.e., despite its southerly location, the closely in age to the four schist groups, particu- 2005), the Nacimiento block would restore ap- Catalina Schist is dominated by detrital zircons larly with respect to the three older age divi- proximately outboard of the Diablo Range of of Early Cretaceous age and, as noted previously, sions (Fig. 3). However, the youngest forearc central California, without the Salinian block appears most similar to the Pelona-Orocopia- age group (early to middle Eocene) may have intervening. Further implications of the sinistral Rand schist of the San Emigdio Mountains. As an average age somewhat younger than that of case are considered later herein. we will discuss later, the origin of this pattern is the youngest schist group (Orocopia Schist); Figure 6 reveals an impressive similarity in ambiguous owing to the uncertain nature of slip i.e., as discussed already, the minimum age of the Cenomanian-Turonian to Eocene patterns on the Nacimiento fault. the Orocopia Schists is not well defi ned and from San Miguel Island with those of equivalent conceivably could be no younger than middle age from the vicinity of San Diego and northern Forearc Sedimentary Units Paleocene (Fig. 3). Despite this uncertainty, it is Baja California. This is consistent with mod- clear that the schists and forearc units analyzed els involving large-scale rotation of the west- Age Divisions of Forearc Units here represent at least 30 m.y. of overlapping ern Transverse Ranges from an initial position Samples from the forearc units were divided depositional history during a critically impor- alongside the Peninsular Ranges. into four age groups: (1) Cenomanian-Turonian, tant period in the tectonic evolution of southern One striking anomaly in our results pertains (2) Campanian–early Maastrichtian, (3) middle California. to a single sample of Paleocene age from north- Maastrichtian–Paleocene, and (4) early to ern Baja California (Fig. 6C). Work by one of middle Eocene. These subdivisions were guided Variation in Detrital Zircon Populations as a us (M. Grove) suggests that the distinctive char- by structural and stratigraphic breaks in the Function of Depositional Age and Location acter of this sample refl ects a highly restricted sedimentary record (see the descriptions of in- Probability plots for the forearc age groups source region in the vicinity of the Sierra El dividual sampling areas in the GSA Data Re- are illustrated in Figures 5F–5J alongside the Mayor of eastern Baja California.

492 Geological Society of America Bulletin, March/April 2011 Jacobson et al.

N Orocopia Mts. (2/129) Mts. (3/39) Eocene Sierra Madre

SGf (3/65) Early-Mid Early-Mid Mts.-Simi Hills Nf Santa Monica SAf ?

EHf (4/212) San Diego (2/50) Indians Ranch (3/146)

Rf Nf Northern Baja SGHf Pine Mt. Block (2/26) Santa Ana Santa Mts. (2/184) Mt. (1/28) Ben Lomond (1/53) Mts. (5/74)

SGHf Ynez Santa S. Miguel Is. Jacobson et al., GSA Bull, 2010 GSABull, al., et Jacobson Pass Cajon (1/13) (4/66) San Gabriel Block (5/73) Mts-Simi Hills SGf Santa Monica (3/35)

NW Santa ? Lucia Range

Nf (2/93) Paleocene Mid Maast- Mid SAf San Diego EHf Rf (1/39)

Nf Northern Baja SGHf Mts.(2/82) Santa Ana Santa La Panza Range (4/50) Pt. San Pedro (4/62) (2/29) (2/40) Ranch Indians Pt. Lobos (4/119) (4/119) Tur-Camp Barth et al. Figure 6. Figure Upper McCoy (4/55) (4/85) Atascadero SGf Camp- Santa Monica Mts.-Simi Hills Nf ? (4/169) San Diego

Early Maast Early Rf EHf 300 km 300 SAf (2/158) (3/39) Cyn (1/36) Del Puerto Northern Baja Mts. (3/55)

Rf Nf San Rafael Pigeon Point Mts. (2/56) Santa Ana Santa (2/64) Mts. (2/65) Santa Ynez Santa Slab SGHf (3/64) (3/43) 120 km 120 Cambria S. Miguel Is. Pfeiffer Pfeiffer (2/112) Coalinga Point Sur - DeGraaff- Surpless et al. Surpless

ABC D

Nevada

Ranges

Sierra Sierra SGf Peninsular Peninsular Mojave Desert (1/26) Nf ? Atascadero Cen-Tur (1/58) Mts. (2/61) Santa Ana Santa San Diego (2/88) SAf Rf Northern Baja Mts. (2/31) San Rafael Nf SGHf 55–70 Ma 70–85 Ma 85–100 Ma 100–135 Ma 135–300 Ma >300 Ma 100 km (2/69)

120 km 120 SGHf (3/166) Coalinga DeGraaff- S. Miguel Is. Surpless et al. Surpless Detrital zircon ages 494 Geological Society of America Bulletin, March/April 2011 Forearc in the relative changes show contributionssediments of zirconstime (note through that in ()the numbers are sites/zircons). Late Cretaceous–early Cenozoic tectonic evolution of southern California

POR-Catalina Schists Forearc Sedimentary Units 50 100 150 200 250 300 600 900 1200 1500 1800 2100 50 100 150 200 250 300 600 900 1200 1500 1800 2100 100 100 A Kuu F 80 80 PE 60 60 OR Eoc Orocopia (OR) 40 40 Eocene (Eoc) RA Pelona (PE) KP Mid Maast-Pal (KP) Rand (RA) 20 SE 20 Klu Camp-early Maast (Kuu) San Emig.-Catalina (SE) Cen-Tur (Klu) Cumulative prob. Cumulative 0 0 Orocopia (23/478) B Eocene (28/827) G

Pelona (18/504) C Mid Maast-Pal (30/582) H

Rand (11/341) D Camp-early Maast (32/853) I Relative probability Relative

S. Emigdio-Catalina (24/387) E Cen-Tur (10/333) Vert exag = 2 J Vert exag = 2 for ages>300 Ma for ages>300 Ma

50 100 150 200 250 300 600 900 1200 1500 1800 2100 50 100 150 200 250 300 600 900 1200 1500 1800 2100 Detrital zircon age (Ma) Detrital zircon age (Ma)

Figure 5. Detrital zircon age distributions for the Pelona-Orocopia-Rand-Catalina schists (left column) and forearc sedimentary units (right column). Zircons older than 2.1 Ga comprise <0.5% of the total and are not included. Sample groupingsJacobson are explained et al., GSA in theBull, text 2010. Forearc and schist groups of similar depositional age are plotted next to each other and indicated by the same ﬁ ll pattern. Note x-axis scale break at 300 Ma. Vertical scales also differ to the left and right of the x-axis break such that equal areas represent equal probability throughout the graph, except for plots E and J, for which ages older than 300 Ma are exaggerated by a factor of 2. Black lines in plots G–J indicate schist From the previous sets of info suggestsprobabilities that the from schists plots B–E, respectively.are tightly They aretied raised to above the the forearc x-axis so as notsequences to obscure the plots for the forearc units. Abbrevia- tions for geologic time periods are as in Figure 3. Paired numbers in parentheses indicate number of samples and number of zircon ages. Note that some analyses were excluded from these plots in order to avoid overweighting samples with a large number of analyses (see text and explanation for Fig. 6).

Sandstone Petrology of lithic grains (Table 1). Volcanic frag- COMPARATIVE PROVENANCE Summary modal data for the 59 sandstones ments comprise less than 30% of total lithics. analyzed petrographically are presented in Plagio clase is more abundant than K-feldspar, Detrital Zircons Table 1. More complete results, including sta- although generally not by much. Total lithic tistical values, are provided in the GSA Data contents and P/F values show a weak ten- Sources of Individual Zircon Populations Repository item (Tables DR5 and DR6 [see dency to decrease with decreasing deposi- The most distinctive ﬁ nding of this study is footnote 1]). The Data Repository item also tional age (Table 1; see also Figs. DR5–DR8 the remarkable parallelism in evolution of de- includes QtFL and QmKP ternary diagrams of and Table DR6 [see footnote 1]). Other than trital zircon suites within the Pelona-Orocopia- the results, along with plots of various compo- this, the samples exhibit no systematic trends Rand-Catalina schists and spatially associated sitional parameters shown in geographic coordi- in composition with time. forearc basin. As already noted, the evolving nates using the same format as Figure 6 (Figs. In contrast to the Campanian to Eocene patterns suggest a transition from outboard to DR5–DR9 [see footnote 1]). samples, three of Cenomanian-Turonian age inboard source areas with time, compatible with Most (56) of the samples are Cam panian to from the San Rafael Mountains and Atasca- previous interpretations based on sandstone middle Eocene. This group is rich in quartz dero area are marked by high total lithics, petrol ogy and conglomerate clasts (Gilbert and and feldspar, with low to modest contents Lv/Lt, and P/F. Dickinson, 1970; Lee-Wong and Howell, 1977;

Geological Society of America Bulletin, March/April 2011 493 Jacobson et al.

50 100 150 200 250 300 600 900 1200 1500 1800 2100 2400 2700 3000 100 creasing average age of the Cretaceous–early Cenozoic population and increasing abundance Late Cretaceous–early Cenozoic tectonic evolution of southern California F-POR-C 80 A of Proterozoic grains with decreasing depositional age is somewhat more systematic within POR-Catalina Schists Forearc Sedimentary Units 60 the forearc sequence than the schists (Fig. 5). 50 100 150 200 250 300 600 900 1200 1500 1800 2100 50 100 150 200 250 300 600 900 1200 1500 1800 2100 MC 100 100 This is probably due to the fact that the forearc 40 A Kuu F samples represent a greater geographic range, 80 80 array of depositional environments, and total PE 20 ERG 60 OR 60 PR number of samples and analyses than the Eoc prob. Cumulative Orocopia (OR) 40 40 Eocene (Eoc) 0 schists; i.e., the forearc groups likely reﬂ ect a RA Pelona (PE) KP Mid Maast-Pal (KP) Rand (RA) more complete averaging of the margin than 20 SE 20 Klu Camp-early Maast (Kuu) Forearc-POR-Catalina (F-POR-C) (172/4305) B San Emig.-Catalina (SE) Cen-Tur (Klu)

Cumulative prob. Cumulative the schists. 0 0 Orocopia (23/478) B Eocene (28/827) G Sandstone Petrology

Lower Mesozoic wall rocks - C Results from the sandstone modal analysis Pelona (18/504) C Mid Maast-Pal (30/582) H are consistent with the detrital zircon data but Peninsular Ranges (PR) (4/433) not as diagnostic. For example, the high total lithics, Lv/Lt, and P/F for the Cenomanian- Turonian forearc units (Table 1; see also fi g. 13 Rand (11/341) D Camp-early Maast (32/853) I Upper McCoy Mountains D in Dickinson, 1995) confi rm a source in the Formation (MC) (4/119) western side of the Sierran–Peninsular Ranges arc (cf. Ingersoll, 1983), as concluded based Relative probability Relative on the zircon data. For the younger samples, S. Emigdio-Catalina (24/387) Cen-Tur (10/333) probability Relative E Vert exag = 2 J however, detrital zircon and sand composi- Vert exag = 2 for ages>300 Ma tions are not strongly coupled. Specifi cally, the for ages>300 Ma Jurassic ergs (ERG) (10/890) E Cam panian to Eocene units exhibit a dramatic

50 100 150 200 250 300 600 900 1200 1500 1800 2100 50 100 150 200 250 300 600 900 1200 1500 1800 2100 shift in detrital zircon populations with deposi- Detrital zircon age (Ma) Detrital zircon age (Ma) tional age reﬂ ecting progressively more inboard

Figure 5. Detrital zircon age distributions for the Pelona-Orocopia-Rand-Catalina schists (left column) and forearc sedimentary units (right source regions (Figs. 5G–5I). In contrast, sand 50 100 150 200 250 300 600 900 1200 1500 1800 2100 2400 2700 3000 column). Zircons older than 2.1 Ga comprise <0.5% of the total and are not included. Sample groupings are explained in the text. Forearc compositions for these same units show only and schist groups of similar depositional age are plotted next to each other and indicated by the same fi ll pattern. Note x-axis scale break at Detrital zircon age (Ma) subtle variations with age (Table 1). The detrital 300 Ma. Vertical scales also differ to the left and right of the x-axis break such that equal areas represent equal probability throughout the zircons thus provide a reliable indication of graph, except for plots E and J, for which ages older than 300 Ma are exaggerated by a factor of 2. Black lines in plots G–J indicate schist probabilities from plots B–E, respectively. They are raised above the x-axis so as not to obscure the plots for the forearc units. Abbrevia-Figure 7. Detrital zircon age distributions for variousJacobson sedimentary et al., GSA sequences Bull, 2010 in Califor- ages of rocks within the source areas, whereas tions for geologic time periods are as in Figure 3. Paired numbers in parentheses indicate number of samples and number of zircon ages.nia and adjacent areas of the southwestern United States. (A) Cumulative probabilities for modal mineralogy refl ects only their lithology. Note that some analyses were excluded from these plots in order to avoid overweighting samples with a large number of analyses (see textsample groups plotted in B–E. (B) Combined data for the Pelona-Orocopia-Rand-Catalina The latter is apparently relatively uniform both and explanation for Fig. 6). in contrast, foreland sequences are quite diferent. schists and forearc units (this paper). (C) Lower Mesozoic sedimentary framework rocks of spatially and temporally within the eastern side the western belt of the Peninsular Ranges batholith (Morgan et al., 2005; Grove et al., 2008). of the arc and southwestern part of the craton. (D) Upper McCoy Mountains Formation in the McCoy Mountains (Barth et al., 2004). The sandstones analyzed here are broadly Sandstone Petrology of lithic grains (Table 1). Volcanic frag- COMPARATIVE PROVENANCE Summary modal data for the 59 sandstones ments comprise less than 30% of total lithics. (E) Colorado Plateau eolianites (Dickinson and Gehrels, 2003, 2009). See Figure 5 for expla- similar in detrital modes to those from the type analyzed petrographically are presented in Plagio clase is more abundant than K-feldspar, Detrital Zircons nation of scale break at 300 Ma. Paired numbers in parentheses indicate number of samples Great Valley Group of central California east Table 1. More complete results, including sta- although generally not by much. Total lithic and number of zircon ages. of the San Andreas fault (Table 1; Ingersoll, tistical values, are provided in the GSA Data contents and P/F values show a weak ten- Sources of Individual Zircon Populations Repository item (Tables DR5 and DR6 [see dency to decrease with decreasing deposi- The most distinctive fi nding of this study is 1983). Overall, our samples tend to be some- footnote 1]). The Data Repository item also tional age (Table 1; see also Figs. DR5–DR8 the remarkable parallelism in evolution of de- what lower in lithics and P/F than the Great includes QtFL and QmKP ternary diagrams of and Table DR6 [see footnote 1]). Other than trital zircon suites within the Pelona-Orocopia-within the Mojave-Yavapai-Mazatzal basement orogen, are common in Neoproterozoic to Cam- Valley rocks, suggesting a more inboard and the results, along with plots of various compo- this, the samples exhibit no systematic trends Rand-Catalina schists and spatially associatedof the Southwest United States (Karlstrom et al., brian cratonal and miogeoclinal sections of this deeply denuded source area (cf. Dickinson, sitional parameters shown in geographic coordi- in composition with time. forearc basin. As already noted, the evolving nates using the same format as Figure 6 (Figs. In contrast to the Campanian to Eocene patterns suggest a transition from outboard to1987; Anderson and Bender, 1989; Wooden and region (Gehrels, 2000; Stewart et al., 2001; 1985; Marsaglia and Ingersoll, 1992; Ingersoll DR5–DR9 [see footnote 1]). samples, three of Cenomanian-Turonian age inboard source areas with time, compatible withMiller, 1990; Barth et al., 2000, 2001a, 2001b; Gehrels et al., 2002; Barth et al., 2009). This im- and Eastmond, 2007). This probably refl ects Most (56) of the samples are Cam panian to from the San Rafael Mountains and Atasca- previous interpretations based on sandstoneAnderson and Silver, 2005; Farmer et al., 2005) plies that zircons of Grenville age in the schists the younger average age of our samples and middle Eocene. This group is rich in quartz dero area are marked by high total lithics, petrol ogy and conglomerate clasts (Gilbert and and feldspar, with low to modest contents Lv/Lt, and P/F. Dickinson, 1970; Lee-Wong and Howell, 1977;and with predicted zircon “fertility” of the in- and forearc units are recycled. The cratonal and their proximity to the strongly disrupted seg- ferred source rocks (Dickinson, 2008). In con- miogeoclinal sequences also include substantial ment of the arc at the latitude of the Mojave trast, whereas zircons of Grenville age are not populations of anorogenic and Mojave-Yavapai- Desert (see also Dickinson, 1995). Geological Society of America Bulletin, March/April 2011 493abundant in either the schists or forearc units, Mazatzal ages; consequently, some of these they nonetheless appear to be overrepresented grains may be recycled, as well. Comparison to Other Depositional Systems compared to the known distribution of Grenville rocks in the southwestern United States and Contrasts between the Schists and Upper McCoy Mountains Formation northwestern Sonora (Barth et al., 1995, 2001b; Forearc Units The McCoy Mountains Formation of south- Anderson and Silver, 2005; Farmer et al., 2005). Despite the striking similarities between the eastern California and southwestern Arizona On the other hand, detrital zircons of Grenville schists and forearc units, there are some differ- (Fig. 1) is a weakly metamorphosed Upper Ju- age, apparently derived from the Appalachian ences. For example, the overall pattern of de- rassic to Upper Cretaceous fl uvial sequence with

496 Geological Society of America Bulletin, March/April 2011 714 J. Wakabayashi Franciscan Schists POR Schists

2.8 Coe 2.8 Coe Qtz Qtz

2.4 2.4 INTERNATIONAL GEOLOGY REVIEW 673

Begin 2nd cycle for blueschist 2.0 clasts (hypothetical) 2.0 Begin 2nd cycle: subduction subduction between between ca. 90 and 132 and 114 Ma 130 Ma Begin 1st cycle: Begin1st cycle: high-grade block high-grade block 1.6 Jd + AbQtz 1.6 Jd + AbQtz Begin 3rd subduction between subduction between cycle: ca. 155 and Begin 3rd ca. 155 and subduction 170 Ma cycle: 170 Ma of blocks subduction 1.2 1.2 and matrix of blocks between ca. and matrix 72 and ca. 105 Ma 83 Ma 0.8 Ar 0.8 Ar Cc Cc P P (GPa) (GPa)

0.4 0.4 End 1st Cycle: End 1st cycle: exhumation and deposition exhumation and deposition in lower Great Valley Group in lower Great Valley Group (ca. 130-146 Ma) (ca. 130-146 Ma) 400 800 T°C 400 800 T°C

Figure 25. Diagrams showing hypothetical PT paths associated with multiple burial exhumation/exposure cycles for blocks/clasts at the Panoche Road and Blind Beach localities. PT paths are estimated in comparison with similar assemblagesFigure 4. inP-T rocksdiagram examinedshowing metamorphic by detailed facies fields of Spear (1993) and data for northern and southern schists and Catalina petrologic study (Figure 7). The Panoche Road locality refers to high-gradeWakabayashi, blocks Int. withinGeol. Rev., the2015 serpentiniteschist (references mélange given in lensTable 1 that). Numbers is itself along each a path correspondChapman, to theInt. timeGeol. (Ma)Rev., at2017 which specific P–T conditions were experienced. Note that northern schists experienced peak temperatures ~50 to 100°C higher than southern schists (the range of block in siliciclastic mélange. The Blind Beach locality includes the lawsonite-bearing clast in Figurescalculated 19(c) peakand conditions24(a) isand represented24(b) byand ellipses). the Also note ‘hairpin’ to slightly counterclockwise P–T paths. Depth conversion blueschist-overprinted amphibolite clast(s) of Figure 19(d). The multiple PT paths shown for the first cyclefrom pressure reflect assumes the an variety average crustal of PT density paths of 2900 kg/m3. And: andalusite; Ky: kyanite; Sil: sillimanite; AEA: albite–epidote POR Schists are hotter than most Franciscan schists (blue curvesamphibolite; at left). BS; blueschist,Also usually EC; eclogite; GS:have greenschist; an GR:inverted granulite; H: metamorphic hornfels; OA: oligoclase amphibolite; gradient. PP: prehnite– for high-grade blocks. For the Panoche Road locality the serpentinite mélange has blocks of garnetpumpellyite;–amphibolite/eclogite Z: zeolite. Catalina schist subunitas well abbreviations: as A, amphibolite; EA, epidote amphibolite; EB, epidote blueschist; LB, garnet-free amphibolites. The Blind Beach lawsonite-rich clast in Figures 19(c) and 24(a) and 24(b)lawsonitedoes not blueschist. have direct evidence of a high-grade first cycle, although the lawsonites of that grain size are found only in high-grade rocks. The Blind Beach (Figure 19(c)) first cycle PT paths presume a high-grade block first cycle of uncertain nature for the lawsonite-bearing clast.of shearing In contrast, in mylonitic the fabricsfirst along cycle the Rand path fault in deposited after ca. 85 Ma (Figure 3). Following deposi- for the amphibolite clast in Figure 19(d) would most likely be one of the dashed green pathsthe (blueschist-overprinted northern and southern Rand garnet-free Mountains. Evidence tion, the Rand schist was underthrust and experienced amphibolites). for an early upper plate-to-the-SW tectonic event, attrib- peak metamorphism at ~556°C and 8.9–9.4 kbar before uted to subduction of the schist beneath upper plate cooling through 40Ar/39Ar hornblende and muscovite assemblages, is locally preserved along the shear zone closure (i.e., ascending from the deep to middle crust) and extensively overprinted by top-to-the-NE, most by ca. 70 Ma (Jacobson 1990;Groveet al. 2003a; in likely exhumation-related, structures. The press). During underthrusting, the Rand schist devel- The stack of nappes including the Willow Spring slab, neoblastic blueschist faciesJohannesburg minerals gneiss to is, grow in turn, across overlain bythe eastern oped an inverted metamorphic gradient ranging from Sierra Nevada batholith-affinity granitic rocks along transitional greenschist–blueschist facies at deep levels 7 km ENE of the serpentinite mélange locality (Figures 6 various clast boundaries of‘Fault this II’ unit.of Nourse (1989), a member of the southern (~3 km beneath the Rand fault) to transitional greens- and 10) may provide additional constraints on the deposi- A quartz diorite blockSierra from detachment mélange system. The in‘Cinco Sonoma’ window exposes chist–amphibolite facies at relatively shallow levels the same Rand schist–Johannesburg gneiss–granitic (Jacobson 1995). Both Cinco and Rand Mountains tion and accretion of the Panoche Road serpentinite County apparently subductedsequence twice along to the conditions Neogene Garlock of fault ca. (Nourse schist windows are intruded by Miocene dikes and mélange, although a correlative serpentinite mélange hor- 0.4 GPa sometime after its1989 igneous; Silver et al formation. 1995). As a consequence, at 165 Ma structures stocks (Nourse 1989;Silveret al. 1995;Groveet al. in within the Cinco window are highly overprinted by press). izon in the nappe stack is lacking in the Willow Spring (Erickson et al. 2004) butNeogene prior to deformation. its second exhumation, area. There is no equivalent of the coherent high-grade exposure, and redeposition betweenLike San Emigdio ca. 72 and and Tehachapi 83 Ma, schist and windows, 2.3.3. Sierra de Salinas–Gabilan Range–Portal detrital zircon age spectra from Rand schist metasand- Ridge–Quartz Hill Willow Spring slab or the Skaggs Spring schist in the subsequent third burial to prehnitestone are consistent–pumpellyite with a Sierra conditions Nevada-Peninsular The Sierra de Salinas is the type locality of the schist of vicinity of the serpentinite mélange. The blueschist in (Erickson 2011; Section 4.5Ranges). Based arc source. on The youngestthe likely population upper of detrital Sierra de Salinas (Ross 1972, 1976; Barth et al. 2003; zircon grains indicates that the Rand schist was Kidder and Ducea 2006). The schist of Sierra de Salinas contact with the serpentinite mélange is very fine-grained plate source for the quartz diorite block and its degree of lawsonite blueschist, unlike the coarse-grained mafic metamorphism, it is most likely to have been subducted schists of the Willow Spring slab. A reasonable structural for the first time after about 130 Ma, based on the chron- level for the serpentinite mélange body is below the ology of metamorphism and the ages of other mélanges Skaggs Spring schist. Such a structural correlation sug- (serpentinite and siliciclastic) in the Franciscan (see also gests the serpentinite mélange along Panoche Road was Section 5.8.2). deposited after 132 Ma (metamorphic age of Skaggs The blueschist clast from Blind Beach (Figures 19(c) Springs schist) and before 114 Ma. This age estimate and 24(a) and 24(b)) has no direct age constraints. The will be applied for the original deposition of the serpenti- depositional age of the sandstone matrix mélange, prior to nite lens with included high-grade blocks/clasts. the third burial episode, is considered to be between ca. 72 The serpentinite mélange was subducted deeply enough and 83 Ma based on correlation to the KRRM. Lawsonites to recrystallize as antigorite–talc–tremolite–chlorite schist, as coarse as the detrital grains in the clast are restricted to then exhumed and deposited as a large block within mostly the high-grade rocks of the Franciscan, so the first sub- siliclastic and felsic volcanic detritus at ca. 105 Ma, before duction burial event recorded by the clast may have been once again being subducted deeply enough for the ca. 155–169 Ma, as would be true of the nearby Jacobson et al.

on the Nacimiento fault must have occurred A ca. 100 Ma Future Upper McCoy Colorado during emplacement of the Pelona-Orocopia- Franciscan Forearc basin Mountains Fm. Plateau Rand schists (Fig. 3). The missing parts of the arc and forearc basin represent a substantial Pre-Cret. crust volume of material. Whereas much of this mass could have been completely subducted, it seems Top of Farallon Continental reasonable to expect that at least some fraction lithosphere would have been interleaved with the schists 100 km slab (Fig. 8C). However, as pointed out by Dickin- son et al. (2005, p. 14), the Pelona- Orocopia- Rand schists bear little resemblance to the omitted units. The schists consist predomi- B ca. 80 Ma Upper McCoy Future nantly of coherent meta-sandstone with small Mountains Fm. Franciscan Forearc crust: Jur. and Colorado amounts of meta-basalt, meta-chert, marble, E. Cret? Plateau and ultramafi c rock. In contrast, the inferred missing units include parts of the Franciscan Pre-Cret. crust mélange, Coast Range ophiolite, uppermost Ju- rassic through Cretaceous Great Valley Group, Top of Farallon slab Continental and western belt of the Sierran arc, including lithosphere its pre–Late Jurassic metamorphic framework Rand/Catalina Schist rocks. The only component of this assemblage that reasonably matches the Pelona-Orocopia- Jacobson et al., GSA Bull, 2010 Rand-Catalina schists is the Upper Cretaceous C ca. 50 Ma—thrust option Mogollon part of the Great Valley Group. However, it Future Here’s the puzzle to me: WHY would seds go down when the slab FLATTENS? (SlabUpper flattening McCoy isHighlands itself attested to byis the not clearremoval how this of partthe of lower the missing crust, sec- which Forearc basin Mountains Fm. Colorado we haven’t discussed) Rand/Catalina Franciscan tion could have been incorporated within the Schist NF Plateau schists while completely excluding the remain- Pre-Cret. crust der. Mafi c rocks of the Coast Range ophiolite bear superfi cial compositional resemblance to Peraluminous Top meta basites within the schists, but they differ in of crustal melts detail. Specifi cally, the Coast Range ophiolite Farallon Pelona Schist exhibits an arc signature (Shervais and Kim- Orocopia Schist brough, 1985; Giaramita et al., 1998; Sher- slab vais et al., 2005), whereas mafi c rocks in the D ca. 50 Ma—strike-slip option Pelona-Orocopia-Rand-Catalina schists were Mogollon derived overwhelmingly from ocean-fl oor ba- Future Upper McCoy Highlands salts (Haxel et al., 1987, 2002; Dawson and Forearc basin Mountains Fm. Colorado Rand/Catalina Franciscan Plateau Jacobson, 1989; Moran, 1993). Consequently, Schist NF we conclude that the Pelona-Orocopia-Rand- Pre-Cret. crust Catalina schists can be explained entirely as subducted trench materials and off-scraped Peraluminous fragments of oceanic crust from the Farallon Top of crustal melts Farallon plate (Figs. 8B and 8D) without any admixture Pelona Schist of materials excised from along the Nacimiento Orocopia Schist fault. Nonetheless, we cannot rule out the possi- slab SW NE bility that fragments of the missing terranes are hidden within unexposed parts of the Pelona- Figure 8. Tectonic model for underplating of the Pelona-Orocopia-Rand-Catalina schists Orocopia-Rand-Catalina schists (Fig. 8C). and development of the Nacimiento fault. (A) Geometry prior to the onset of fl at subduction In contrast to thrust models, strike-slip models and emplacement of the Pelona-Orocopia-Rand-Catalina schists. (B) Early phase of fl at for the Nacimiento fault imply that the excised subduction preceding initiation of slip on the Nacimiento fault. (C–D) Relations following parts of the arc and forearc basin remained at the cessation of slip along the Nacimiento fault (NF), assuming thrusting and sinistral strike surface but were translated out of the plane of slip, respectively. For simplicity, slip on the Nacimiento fault is shown in both C and D as margin-normal cross sections at the latitude of postdating emplacement of the Pelona Schist but predating emplacement of the Orocopia the Mojave Desert (Fig. 8D). Omission of units Schist. In actuality, movement on the Nacimiento fault is likely to have overlapped with the by strike-slip motion along the Nacimiento fault underthrusting of either or both of those schist groups. Inset in C shows an enlargement of can be reconciled with either dextral or sinistral our interpretation of relations adjacent to the Nacimiento fault system in the case of thrust- movement, as long as the fault cuts in the ap- ing. The North American craton is held fi xed in all panels. Fill patterns for the igneous rocks propriate sense across the strike of the margin. correspond to the color scheme of Figure 2 with the addition that yellow represents ages of Nonetheless, for reasons discussed by Dickin- ca. 70–50 Ma. son et al. (2005), we favor the sinistral option.

498 Geological Society of America Bulletin, March/April 2011 Jacobson et al.

on the Nacimiento fault must have occurred A ca. 100 Ma Future Upper McCoy Colorado during emplacement of the Pelona-Orocopia- Franciscan Forearc basin Mountains Fm. Plateau Rand schists (Fig. 3). The missing parts of the arc and forearc basin represent a substantial Pre-Cret. crust volume of material. Whereas much of this mass could have been completely subducted, it seems Top of Farallon Continental reasonable to expect that at least some fraction lithosphere would have been interleaved with the schists 100 km slab (Fig. 8C). However, as pointed out by Dickin- son et al. (2005, p. 14), the Pelona- Orocopia- Rand schists bear little resemblance to the omitted units. The schists consist predomi- B ca. 80 Ma Upper McCoy Future nantly of coherent meta-sandstone with small Mountains Fm. Franciscan Forearc crust: Jur. and Colorado amounts of meta-basalt, meta-chert, marble, E. Cret? Plateau and ultramafi c rock. In contrast, the inferred missing units include parts of the Franciscan Pre-Cret. crust mélange, Coast Range ophiolite, uppermost Ju- rassic through Cretaceous Great Valley Group, Top of Farallon slab Continental and western belt of the Sierran arc, including lithosphere its pre–Late Jurassic metamorphic framework Rand/Catalina Schist rocks. The only component of this assemblage that reasonably matches the Pelona-Orocopia- Rand-Catalina schists is the Upper Cretaceous C ca. 50 Ma—thrust option Mogollon part of the Great Valley Group. However, it Future Upper McCoy Highlands is not clear how this part of the missing sec- Forearc basin Mountains Fm. Colorado Rand/Catalina Franciscan tion could have been incorporated within the Schist NF Plateau schists while completely excluding the remain- Pre-Cret. crust der. Mafi c rocks of the Coast Range ophiolite bear superfi cial compositional resemblance to Peraluminous Top meta basites within the schists, but they differ in of crustal melts detail. Specifi cally, the Coast Range ophiolite Farallon Pelona Schist exhibits an arc signature (Shervais and Kim- Orocopia Schist brough, 1985; Giaramita et al., 1998; Sher- slab vais et al., 2005), whereas mafi c rocks in the D ca. 50 Ma—strike-slip option Pelona-Orocopia-Rand-Catalina schists were Mogollon derived overwhelmingly from ocean-fl oor ba- Future Upper McCoy Highlands salts (Haxel et al., 1987, 2002; Dawson and Forearc basin Mountains Fm. Colorado Rand/Catalina Franciscan Plateau Jacobson, 1989; Moran, 1993). Consequently, Schist NF we conclude that the Pelona-Orocopia-Rand- Pre-Cret. crust Catalina schists can be explained entirely as subducted trench materials and off-scraped Peraluminous fragments of oceanic crust from the Farallon Top of crustal melts Farallon plate (Figs. 8B and 8D) without any admixture Pelona Schist of materials excised from along the Nacimiento Orocopia Schist fault. Nonetheless, we cannot rule out the possi- slab SW NE bility that fragments of the missing terranes are hidden within unexposed parts of the Pelona- Figure 8. Tectonic model for underplating of the Pelona-Orocopia-Rand-Catalina schists Orocopia-Rand-Catalina schists (Fig. 8C). and development of the Nacimiento fault. (A) Geometry prior to the onsetJacobson of fl at etsubduction al., GSA Bull, 2010 In contrast to thrust models, strike-slip models and emplacement of the Pelona-Orocopia-Rand-Catalina schists. (B) Early phase of fl at for the Nacimiento fault imply that the excised subduction preceding initiation of slip on the Nacimiento fault. (C–D) Relations following Second puzzle: Why would these rocks be cooling at this time? parts of the arc and forearc basin remained at the cessation of slip along the Nacimiento fault (NF), assuming thrusting and sinistral strike surface but were translated out of the plane of slip, respectively. For simplicity, slip on the Nacimiento fault is shown in both C and D as margin-normal cross sections at the latitude of postdating emplacement of the Pelona Schist but predating emplacement of the Orocopia the Mojave Desert (Fig. 8D). Omission of units Schist. In actuality, movement on the Nacimiento fault is likely to have overlapped with the by strike-slip motion along the Nacimiento fault underthrusting of either or both of those schist groups. Inset in C shows an enlargement of can be reconciled with either dextral or sinistral our interpretation of relations adjacent to the Nacimiento fault system in the case of thrust- movement, as long as the fault cuts in the ap- ing. The North American craton is held fi xed in all panels. Fill patterns for the igneous rocks propriate sense across the strike of the margin. correspond to the color scheme of Figure 2 with the addition that yellow represents ages of Nonetheless, for reasons discussed by Dickin- ca. 70–50 Ma. son et al. (2005), we favor the sinistral option.

498 Geological Society of America Bulletin, March/April 2011 800 (A) GAVILAN HILLS underthrusting 700 Orocopia Schist Upper Plate Rocks of schist undiff. gneiss accretion of schist 600

hornblende 40Ar/39 Ar ? 500

muscovite 40Ar/39 Ar 400 biotite 40Ar/39 Ar 300 K-feldspar MDD 200 Slip on Slip on Gatuna Chocolate deposition 100 apatite FT fault Mtns. of schist fault protolith congl. of Bear Cyn. 800 (B) leucogranite OROCOPIA MTNS. intrusion

700 l Orocopia Schist Upper Plate Rocks u

t underthrusting o mylonite an-sy unit n accretion of schist ic lg-gn unit (shallow) h

Temperature (°C) chl. zone 600 e lg-gn unit (deep) of schist a bio. zone t

g hornblende 40Ar/39 Ar 500 Slip on CMF(?) muscovite 40Ar/39 Ar 400 ? biotite 40Ar/39 Ar p 300 dee K-feldspar MDD

200 Slip on OMDF apatite FT deposition 100 shallow of schist Maniobra Fm. ? protolith Diligencia Fm. 0 01020304050607080 Time (Ma) Jacobson et al., GSA SP 419, 2007

An important component is the cooling history, indicating episodic normal faulting INTERNATIONAL GEOLOGY REVIEW 671

800 a low-angle detachment fault system that probably corre- (A) GAVILAN HILLS underthrusting lates with the type Rand fault exposed in the Rand 700 Orocopia Schist Upper Plate Rocks of schist undiff. gneiss accretion of schist Mountains (Postlethwaite and Jacobson 1987; 600

hornblende 40Ar/39 Ar ? Chapman et al. 2010). After correcting for regional 500

muscovite 40Ar/39 Ar Neogene clockwise rotations, ductile kinematic indica- 400 biotite 40Ar/39 Ar tors along the Rand fault in the San Emigdio and 300 K-feldspar MDD Tehachapi Mountains yield top-to-the-NE kinematics. 200 Slip on Slip on The Tehachapi–San Emigdio complex structurally Gatuna Chocolate deposition 100 apatite FT fault Mtns. of schist fault protolith underlies granitic and metamorphic framework rocks congl. of Bear Cyn. 800 of eastern Sierra Nevada affinity (the Pastoria plate) (B) leucogranite OROCOPIA MTNS. intrusion

700 l Orocopia Schist Upper Plate Rocks u along the Late Cretaceous southern Sierra detachment

t underthrusting o mylonite an-sy unit n accretion of schist ic lg-gn unit (shallow) h

Temperature (°C) chl. zone 600 e (Crowell 1952; Wood and Saleeby 1997; Chapman et al. lg-gn unit (deep) of schist a bio. zone t

g hornblende 40Ar/39 Ar 500 2012). The southern Sierra detachment is extensively Slip on CMF(?) muscovite 40Ar/39 Ar remobilized as a thrust in the San Emigdio Mountains 400 ? biotite 40Ar/39 Ar p and as the south branch of the Garlock fault in the 300 dee K-feldspar MDD b Tehachapi Mountains. 200 Slip on OMDF The San Emigdio and Techachapi schist windows apatite FT deposition 100 shallow of schist Maniobra Fm. ? protolith preserve inverted metamorphic field gradients that Diligencia Fm. 0 01020304050607080 range from transitional greenschist to lower amphi- Time (Ma) bolite to upper amphibolite facies from the base to the top of the exposed sections. The uppermost portion of the San Emigdio schist locality shows evidence for: (1) partial melting in metasandstone and prograde conversion of hornblende to clinopyr- oxene in mafic schist and (2) fracturing of garnet Jacobson et al., GSA SP 419, 2007 porphyroblasts near peak metamorphic conditions (Chapman et al. 2011). These structural–petrologic Chapman, Int. Geol. Rev., 2017 relations suggest that the San Emigdio schist expo- Figure 2. (a) Compilation of geo-/thermochronologic data from sure is one of two schist exposures (Sierra de Salinas) Chapman suggests this is common to all the POR schists…which seems a bitthe schist,curious. arranged Note by pre-Neogene that this distance is the from opposite the north- ofthat isothermal reached transitional decompression—this granulite facies conditions is westernmost schist locality (the San Emigdio window). See and contains a record of palaeoseismicity along the rapid cooling. Figure 1 for locations and Table 1 for data sources. ‘Low subduction megathrust. Some zircon dates being pushed back due to overgrowths. temperature thermochonometers’ consist of K–feldspar 40Ar/39Ar, zircon (U-Th)/He, and apatite fission track methods. Widespread abundance of garnet, a relatively high Timing of shallow subduction from Liu et al.(2010) and SW U.S. proportion of mafic and ultramafic schist (~15% and 5%, metamorphic core complex formation from Foster and John respectively), and the thermal history of schist in the (1999). The time between the youngest zircon populations San Emigdio and Tehachapi Mountains are unique fea- and 40Ar/39Ar muscovite ages represents the interval in which tures of these exposures. The San Emigdio and peak metamorphism and the onset of cooling occurred, and is highlighted as such. (b) Temperature–time arrays showing two Tehachapi exposures are the oldest and second oldest phases of pronounced cooling (~100°C/Ma) separated by a schist windows, with maximum ages of deposition (cal- period of slow cooling (~1°C/Ma) in Pelona, Orocopia, and culated as a weighted mean from the three youngest Rand schists. Black rectangle represents the depositional age analysed grains in each exposure; Dickinson and Gehrels of the oldest sedimentary rocks that contain schist detritus, 2009) of ca. 94 and 89 Ma, respectively (Grove et al. implying schist exposure at this time (Ehlig et al. 1975; Sylvester and Smith 1975; Dillon 1976; Goodman 1989; 2003a; Jacobson et al. 2011; Chapman et al. 2016; Goodman and Malin 1992; Hughes 1993; Malin et al. 1995; Table 1; Figure 3). However, it should be noted that Ehlert 2003). the maximum depositional age of the San Emigdio exposure may be as old as ca. 101 Ma when metamorphic domains are excluded (Chapman et al. 2013). The schist in the San Emigdio and Tehachapi All analysed samples from these areas contain abundant Mountains resides below a series of Early to mid- Cretaceous detrital zircon grains, with diminishing pro- Cretaceous deep-level plutonic rocks and associated portions of Jurassic, Triassic, and Proterozoic grains, pre-Cretaceous framework metamorphic rocks (the consistent with a Sierra Nevada-Peninsular Ranges arc ‘Tehachapi–San Emigdio complex;’ Saleeby et al. 2007; source (e.g. Grove et al. 2003a; Jacobson et al. 2011; Chapman et al. 2010) along a locally ductile to brittle Figure 3). The lack of significant proportions of M.N. Ducea, A.D. Chapman

Fig. 6. Simplified geologic map of coastal central California south of Monterey Bay, showing the basement and cover units of the Salinian arc, the schist of the Sierra de Salinas, and recent sedimentary cover (modified after Ducea et al. 2007). (top right) Map showing the distribution of arc-related rocks in California. (bottom left) Cross-section A–A' depicts the structural relationships between the upper plate Salinian arc and the underlying schist. X-X' is a line along which the schematic cross sections in Fig. 7 (one for the late Cretaceous configuration and one modern) were drawn.

Salinian Schist timing Fig. 7. Schematic W-E cross sections through the Future “0" age surface of the Coast Ranges A. 74-68 Ma Salinia-Sierra de Salinas region: A: above, depicted Maastrichtian extensional immediately after underplating and during the early basins 82 Ma pluton (Barth et al, 2003) extension that exposed the section in the latest Cretaceous. The panel below, B, shows the geologic 68-69 Ma sediments relationships at present day. The schist of the Sierra E W de Salinas is geochemically equivalent to the Franciscan formation. The approximate location of Salinia upper crust 70 Ma end of shear zone these cross sections are shown in Fig. 6 as line X-X'. (Kidder & Ducea, 2006)

Schist of the Sierra de Salinas 80 (77?) Ma protolith Salinia lower crust (dashed lines (Barth et al, 2003; mark foliation patterns) Kidder & Ducea, 2006) Detachment faults

Ducea and Chapman, Earth Sci. Rev., 2018, after Ducea et al., 2007 B. Modern cross-section W through the Coast Ranges Maastrichtian seds Re-activated Chapman et al., Tectonics,E 2010 Sur fault Sierra de Salinas Santa Lucia Mts. Salinas Gabilan Ranges This is very rapid, perhaps reaching the surface in under 2 m.y. shear zone Salinia (lower crust) Salinia (upper crust) Schist of the Sierra de Salinas

Franciscan

12 M.N. Ducea et al.

4 M.N. Ducea et al. Downloaded By: [University of Colorado Libraries] At: 22:28 19 May 2009

Figure 2. Simplified geologic map of the central Monterey terrane, showing the outcrop pattern of the three basement blocks discussed in the paper (Salinia, Sierra de Salinas and Nacimiento), as well as the location of major Late Cretaceous and Cenozoic faults, and the Salinas shear zone.

(Almgren and Reay 1977; Howell et al. 1977, 1978; Poore et al. 1977; Taliaferro 1944). The Coast Ranges are segmented by numerous Late Cenozoic strike-slip and high- angle reverse faults associated with the San Andreas system (Page et al. 1998). Formation of the modern Coast Ranges ended with an episode of latest Miocene- Quaternary transpressional uplift (Compton 1966; Ducea et al. 2003a). In this paper, we will refer to the region bounded by the Pacific Ocean to the west Figure 3. Time–temperature plot for the Monterey terrane, based on published data from and the San Andreas fault to the east between the Transverse Ranges to the south the Santa Lucia and Sierra de Salinas regions. ProtolithDucea age constraints et al., Int’l for Geol. the schist Rev., detrital2009 and Cape Mendocino to the north, as the Monterey terrane. The lateral zircon U–Pb ages and schist cooling Ar–Ar ages) are from Barth et al. (2003) and Grove et al. (2003). Salinian block thermochronologic constraints are from Mattinson (1978), Kistler and Champion (2001), Ducea et al. (2003b), Kidder et al. (2003) and Kidder and Ducea (2006). Downloaded By: [University of Colorado Libraries] At: 22:28 19 May 2009 The inferred ages of marine clastic overlap strata are from a variety of sources summarized in Consider for instance the Sierra de Salinas schist (part of Rand schistBarbeau groupet al. (2005).in previous analysis). Other half of the story has to be rapid unrooﬁng.

basement) does not have an equivalent in the Sierra de Salinas block, but we interpret that to be a result of limited exposures in the latter. The composition and detrital age patterns in the two blocks are remarkably similar. What differentiates the schist of the Sierra de Salinas from typical Franciscan metamorphic rocks is the higher temperature ‘Barrovian’ metamorphic conditions it exhibits. Indeed, the Sierra de Salinas basement rocks are mica schists (as in their name) with minor inclusions of amphibolite-facies meta-mafic rocks (Ross 1976b), not blueschist- facies rocks. A high pressure/medium temperature path is demonstrated by the schist, with peak temperatures in excess of 700uC close to the Salinas shear zone. High temperatures are not thought to be typical of subduction metamorphism (Davies 1999; Peacock 2003). However, they may be quite common, and commonly overlooked in ultra-shallow subducting systems lacking intervening mantle TC5006 HPA TA. OEO XRSO FTERN N IRAD AIA SCHISTS SALINAS DE SIERRA AND RAND THE OF EXTRUSION OF ROLE AL.: ET CHAPMAN f21 of 3 TC5006

Figure 1 Chapman et al., Tectonics, 2010

Part of that story is the fact that the schists are emplaced against the middle crust MALIN ETProﬁle AL ß REFLECTIONS across TehachapiBENEATH MtnsTILTED CRUST 2073

SL0 - 2.6 - 3.9 J as I ' / /

6.2 -6.5 Velocityin km/s ..... Id *

20- 6.8-7.0 (?)

+50. NW Distance (km) SE

Figure 3. (continued) Malin et al., JGR 1995 continuesto uplift the TehachapiMountains. Using these results the easternMojave are moresuggestive of suboceanicconditions to reconstructthe positionsof the gneissand schist in the past,it [Montana et al., 1991; Leventhal et al., 1992]. wouldseem that the wedgeof lowercrustal reflectors is relatedto The Tehachapigneisses were rapidly uplifted to 10-15 km the subcretion of the schist and subsequenttilting of the depthsometime between 100 and 85 Ma [Pickettand Saleeby, TehachapiMountains in latestCretaceous/early Tertiary time. 1993]. In the Rand Mountains to the east, age and structural relationsindicate emplacement of theRand schist beneath similar crystallinerocks in this sametime interval[Silver and Nourse, GeologicalBackground 1986]. In the TehachapiMountains, uplift anderosion continued Basementrocks north of the Gatlock fault are exposedin into the early Tertiary, so that Eocenemarine rocks of the Tejon scatteredoutcrops and consistof the 100-115 Ma Tehachapi Formation were deposited directly on the gneisses [e.g., gneisscomplex and Rand schist (Figures 1 and2) [Pickettand Goodmanand Malin, 1992]. Based on the presenceof schist Saleeby,1993]. Theserocks once occupied the lowercrust and clastsin the "unnamedconglomerate," which overliesa major weremetamorphosed at depthsof 25 to 30 km (7-10 kbar and550 angular unconformity, the Rand schist in the Tehachapi to 760øC) some 100 m.y. ago [Saleeby, 1990; Jacobson,1990; Mountainsfirst appearedat the surfacein middleMiocene time Jacobson et al., 1988; Saleeby et al., 1987; Sharry, 1981]. [Goodman, 1989]. Postmagmaticductile fabricsin the gneissesare similar to those Geological and geophysicalevidence exists for three other seenin the upper platesof the Vincent and Rand thrusts. These Oligoceneto Miocenetectonic events in the Tehachapiregion faultsplace intrusive rocks over the schistin the San Gabrieland [e.g., Crowell, 1974, 1987; Goodman and Malin, 1988; Rand Mountains respectively[Ehlig, 1981; Silver et al., 1984; Goodman, 1989; Goodman and Malin, 1992]. The earliest event Silver and Nourse, 1986]. Evidently,this event and its associated is a late Oligocene/earlyMiocene period of extension,involving metamorphismtook place in Late Cretaceoustime [Silverand low- and high-angle normal faulting, volcanism, and the Nourse, 1986; Hamilton, 1988; Jacobson et al., 1988]. On its depositionof coarsefanglomerates. Uplift in the mid-Miocene north side, the Rand schist in the Tehachapi Mountains is exposedthe schistand appearsrelated to initial slip on the San boundedby a fault of variable exposureand dip that has been Andreasfault (SAF). Extensionand transtensionresumed after identified both as the "north branch" of the Gatlock fault and as thisperiod, resulting in obliqueslip faulting and rapid subsidence the "Rand thrust" [e.g., Buwalda, 1954; Davis and Burchfiel, of the SSJB to bathyal depths. During late Miocene time, 1973; Burchfieland Davis, 1981]. transtensionalternated with transpression,reflected by cyclesof Sharry [1981] has suggestedthat as in the case of the San subsidenceand uplift in the SSJB. Sincethat time, the SSJBhas Gabriel and Rand Mountains, the schist extends beyond its evolved from submarine, to sublacustrine, to subaerial, to outcrop, perhaps even under the Tejon embayment. This intermontane. Since the Pliocene, the entire region has been possibilitywas discountedto somedegree by Ehlig [ 1968], who dominatedby the transpressionalregime of the modern San suggestedthat only the gneissmay extend to the north. Even Andreasand Garlock faults, with local shorteningtaking place on farther to the north, however, the basement consists of intact, featuressuch as the WWF and Pleito thrust(Figures 1 and 2). shallow level batholithic crust, beneathwhich geophysicaland The left-lateral Garlock fault is the most significantand active deep-crustal xenolith data demonstratethat no significant crustaldiscontinuity exposed in the TehachapiMountains. Most underlyingschist terrane exists [Saleeby, 1986; Dodge et al., of its postulated64 km of displacementappears to have taken 1986, 1988]. Basedon industryand COCORP seismicreflection place sincethe late Miocene [Carter, 1987], althoughthis fault data, Cheadle et al. [1986] and Lawson [1989] have suggested likely originatedin the mid-Tertiary[Goodman and Malin, 1992]. that the schist terrane is present under much of the western Restoration of this slip brings the Tehachapi Mountains in Mojave Desert. In termsof their uppermantles, xenolith studies proximity to the western Rand Mountains (Figure 2) [e.g., from the Sierra Nevada reveal subcontinental geochemical Crowell, 1979]. Thus, if the Garlock fault is taken to be a deep, signatures[Mukhopadhyay et al., 1988], while similar studiesin throughgoingcrustal discontinuity, any model of the crust A. (90-85 Ma) "Great Valley" forearcbasin sed,ments Cretaceousbatholith -o Felsic botholithic crust -12 --.- Metamorphic pendants -24 E

-36 (•:of the Tehachopi Mtns. sub-bat oh lithic-"'--,•.•• _----•-

- 48 -.-- Sub-continental mantle A pproximate intersection B. (85-80Ma) of SW-NE sections SW •, NE NW ,, SE ..,_,•=, ,-,,,<• ,-_-'•,,,,•; 0 , i/•/ ,'..-,,-',-,.,;,- ;• '..',•'•,;),':' • .-, 12 , ,,, ..... )' ;5."K•.-"/ •'I',: ' J '---/; "-",: ' ";;'/;'[" • ; •' u-,.•--'•'.'-%• '?-- .'0',5 ;z'. -•,:,:...•r' ', ' ; '"'-, /•-•; / :,; ,! ; ?; •,, ".' • "•.",: •2436 ';;.2, •:T (RandThrust) 48

hment) •:• ...... oflower mantle.'/• ?/?' Sub-oceanic'"••'"" '"' RSDI Subcretion/ •u•b•ofL•CIregional •(• I•Y•e• g•c t bathollthlcremoval Sub-continentol(/•: )' mantle schistlerrone Approximateintersection crust of NW-SE sections •' tear?

I C. (80-70Ma)

SW SE

• 24 •4•E

Continued schist subcretion Proﬁle across Tehachapimega thrust) Mtns D. (70-60M0) MALIN ET AL ß REFLECTIONS BENEATH TILTED CRUST 2073 SW • (c) NE WWF NWSP2 GF SL0 - 2.6 - 3.9 J as I ' / / ...... :-.,•, '""'"'''"' ',':'-';"'-" 6.2 -6.5 Velocityin km/s ..... Id *

20- .'.•••'•ß ...... ß':,..ß..'.v:',:':'.::'-:...... ,:..:.:..:.'v.::..:..::.,'::.:.,:.:..:i:..;.•':v.:..:.,::.. . 6.8-7.0 (?)

+50. I•-callyrising'sub-oceanic NW Distance (km) SE mantle NW E. {PRESENTDAY)Figure 3. (continued) SE continuesto uplift the TehachapiMountains. Using these results the easternMojave are moresuggestive of suboceanicconditions Tejon to reconstructthe positionsTehochopiof the gneissand schist in the past,it [Montana et al., 1991;Mojave Leventhal et al., 1992]. wouldseem that the wedgeof lowercrustal reflectors is relatedto The Tehachapigneisses were rapidly uplifted to 10-15SYMBOLS; km Emboymentthe subcretion of the Mtnsschist and subsequent •,Gorlocktilting of theF. depth sometime between Desert100 and 85 Ma [Pickettand Saleeby, TehachapiMountains ---in latest/Cretaceous/early ß Tertiary time. 1993]. In the Rand Mountains0 to the east,--&_ age and structuralMajor active EocenetoRecent strata relationsindicate emplacement of theRand schist beneath similarthrust zone crystallinerocks in this sametime interval[Silver and Nourse, GeologicalBackground 1986]. In the TehachapiMountains, uplift anderosion ^ continued Major inactive Basementrocks north of the Gatlock fault are exposedin into the early Tertiary, so that Eocenemarine rocks of thethrust Tejon scatteredoutcrops and consistof the 100-115 Ma Tehachapi Formation were deposited directly on the gneisses [e.g., detachment and/.__.•z • • gneiss'"'""complex '.'-•""and ...... Rand schist (Figures'"' .... 1•. and.... 2) "•••-"'• [Pickettand Goodman'••'•'••"•••••••-'-•.'.':'-':'-".•and Malin, 1992]. Based on the presence ß of Secondschist order Neogene ..:.-•'.•':Saleeby,...... 1993]. Theserocks once occupied the lowercrust and clastsin the "unnamedconglomerate," 24 which overliesa activemajor thrust orbrittle- ductile ,ransition ,.:.:::::: were...... metamorphosed ¾::.':F?:?:'.. at depths of 25 to 30 km (7-10.•::•i.::;•_•.'.:'.h..:i.i•.'• kbar and550 angular'•'••••••••:.•:•'-"'"'•"'••'''" unconformity,'' ' the Rand schist in the Tehachapifault to 760øC) some 100 m.y. ago [Saleeby, 1990; Jacobson,1990; Mountainsfirst appearedat the surfacein middleMiocene time Jacobson et al., 1988; Saleeby et al., 1987; Sharry, 1981]. [Goodman, 1989]. ,• Second order Moh o ::'::':':":'"'"":"? Postmagmatic' ...... ductile "•'•' fabricsin •' the"' gneisses.... are'"' similar-'"'•-' to those'-' '•• Geological'•" -•"•"'•••••'•••• and geophysical3evidence 6 exists for three inactive other thrust seenin the upper platesl_'owof theser Vincent rctau and Rand I f•nthrusts. trsTheseucture Oligocene to Miocenetectonic J events in the Tehachapiregion faultsplace intrusive rocks over the schistin the San Gabrieland [e.g., Crowell, 1974, 1987; Goodman and Malin, fault1988; 0 25 50 75 IOOkmRand Mountains respectivelyinherited[Ehlig, 1981;from Silverextensional et al., 1984; Goodman, 1989; Goodman and48 Malin, 1992]. •The © earliest Motion event inond out t i , ...-..! i .J Silver and Nourse,collapse 1986]. Evidently,this deformation event and its associatedof is a D.late Oligocene/early Miocene period of extension,involving of cross section metamorphismtook place in Late Cretaceoustime [Silverand low- and high-angle normal faulting, volcanism, and the Nourse, 1986; Hamilton, 1988; Jacobson et al., Malin1988]. On et its al., depositionJGR 1995of coarsefanglomerates. Uplift in the mid-Miocene north side, the Rand schist in the Tehachapi Mountains is exposedthe schistand appearsrelated to initial slip on the San Figure 8. This Figureshows a schematicboundedsummary by a fault of variableof exposure theand subcretion dip that has been Andreasand fault extensional(SAF). Extensionand transtensioncollapse resumed after model for the identified both as the "north branch" of the Gatlock fault and as thisperiod, resulting in obliqueslip faulting and rapid subsidence restoredTehachapi and Rand Mountains regions. the "Rand thrust"(a) [e.g., Buwalda, Generalized 1954; Davis and Burchfiel,starting of the conditionsSSJB to bathyal depths.of During the latebatholithic Miocene time, crust with 1973; Burchfieland Davis, 1981]. transtensionalternated with transpression,reflected by cyclesof theapproximate location of thedeep-level TehachapiSharry [1981] has suggestedrocks that asindicated. in the case of the San(b)-(d) subsidenceand Orthogonal uplift in the SSJB. Sinceviewsthat time, the SSJBduring has the crustal Gabriel and Rand Mountains, the schist extends beyond its evolved from submarine, to sublacustrine, to subaerial, to outcrop, perhaps even under the Tejon embayment. This intermontane. Since the Pliocene, the entire region has been evolutionthat hasexposed the deep-crustalpossibilitymetamorphicwas discountedto somedegree coreby Ehlig of [ 1968], the who dominatedTehachapiby the transpressionaland regimeRand of theMountains. modern San (e) The suggestedthat only the gneissmay extend to the north. Even Andreasand Garlock faults, with local shorteningtaking place on currentconfiguration of the crust. The intersectionsfarther to the north, however,of theviews basement consistsare ofas intact,indicated featuressuch as thein WWF Figure and Pleito thrust9. (Figures Extensional 1 and 2). collapse shallow level batholithic crust, beneathwhich geophysicaland The left-lateral Garlock fault is the most significantand active structureswere favorablyoriented for reactivationdeep-crustal xenolith dataas demonstrate reversethat nofaults significant duringcrustaldiscontinuity either exposed in Tertiary the TehachapiMountains. crustal Most rotations or underlyingschist terrane exists [Saleeby, 1986; Dodge et al., of its postulated64 km of displacementappears to have taken 1986, 1988]. Basedon industryand COCORP seismicreflection place sincethe late Miocene [Carter, 1987], althoughthis fault Plioceneto Recentcontraction. (For additionaldata, Cheadleevidence et al. [1986] andfor Lawson NW-SE [1989] have suggested differences likely originatedin shown the mid-Tertiaryin[Goodman theand crust Malin, 1992].see Ross [ 1985, that the schist terrane is present under much of the western Restoration of this slip brings the Tehachapi Mountains in 1989],Saleeby et al. [1987],Dodge et al. [1986,Mojave Desert. 1988], In termsof theirand upperSaleeby mantles, xenolith studies [1990].) proximity to the western Rand Mountains (Figure 2) [e.g., from the Sierra Nevada reveal subcontinental geochemical Crowell, 1979]. Thus, if the Garlock fault is taken to be a deep, signatures[Mukhopadhyay et al., 1988], while similar studiesin throughgoingcrustal discontinuity, any model of the crust Proﬁle across TehachapiJ. SALEEBY Mtns into Sierra

Figure 3. Generalized cross section from the central Sierra Nevada region southward alongSaleeby, oblique GSA crustal Bull, 2003 section and across Tehachapi- Rand deformation belt into the northernmost Mojave Desert with the Rand Mountains complex restored to its pre-Neogene position relative to the Sierra Nevada. General location of section line shown in Figure 1. This section shows contrasting deep-crust–upper- mantle structures for central and southernmost Sierra Nevada region following Laramide orogeny, revealing a segmentation in the subducted slab with the removal of the subbatholith mantle lithosphere and underplating of the Rand schist to the south. The subduction trajectory of the slab was roughly at right angles to and directed into the cross-section trend. References given in text.

1984; May, 1989; Hall, 1991; Silver and Mat- Yeats (1992), Miller et al. (1992), and Tickoff the San Joaquin volcanic field, offer an in tinson, 1986); 2) transrotational deformation and Maxson (2001). depth picture of the primary lithospheric of the Transverse Ranges (Hornafius et al., Figure 1 also shows the trace of a longitu- structure for the SNB (Ducea and Saleeby, 1986); 3) displacement along the Garlock fault dinal cross section along the southernmost 1996, 1998; Saleeby et al., 2003). The bath- (Malin et al., 1995); and 4) transtensional SNB as it extends southward across the Te- olithic crust is predominately felsic to 35 strain across the Death Valley region (Snow hachapi Range and the restored Rand Moun- km deep. Below lies an 10-km-thick tran- and Wernicke, 2000). The compatibility of tains of the northernmost Mojave Desert (Fig. sition zone containing batholithic rocks and these restorations with the structure of the 3). This cross section displays the critical mafic garnet granulite residues/cumulates that Mojave Desert presents a problem, for which structural features of the Tehachapi-Rand de- in turn grade into an 35-km-thick section the answer is presumed to lie in the disputed formation belt, a complex structural system dominated by garnet clinopyroxenite residues kinematic patterns of Neogene extension and that lies along the southern Sierra-Mojave De- that, like the granulites, were produced during distributed dextral shear along the Mojave sert transition and that is directly related to the batholith generation. The garnet pyroxenites segment of the eastern California shear zone denudation of the southern SNB oblique crust- are interlayered with progressively more spi- (cf. Dokka, 1986; Glazner et al., 1989, 1994; al section. nel and deeper garnet peridotites that extend Dokka and Travis, 1990; Henry and Dokka, to an 125 km depth. This lithospheric sec- 1992). Furthermore, it is asserted here that CONTRASTING DEEP CRUST-UPPER tion remained intact in the central Sierra re- Laramide tectonics left much of the Mojave MANTLE SECTIONS IN LATEST gion at least until the mid-Miocene time of Desert basement in a tectonically layered CRETACEOUS-PALEOGENE TIME xenolith entrainment. It is shown diagram- state, which in conjunction with only scattered matically at the northern end of the Figure 3 basement exposures expressed today, greatly The southern 100 km of the SNB exposes cross section. The garnet pyroxenite and pe- complicates the restoration of superposed a southward-deepening oblique crustal section ridotite levels of the section correspond to the Neogene deformation. The Figure 1 recon- through Cretaceous batholith-generated crust subbatholith mantle lithosphere that represents struction also incorporates generalized pre- from 2 kb to 9 kb conditions, or 7 km the conductively cooled mantle wedge be- Cenozoic reconstructions of the greater Basin to 35 km depths (Ague and Brimhall, 1988; neath the Cretaceous Sierran arc. and Range province region (Coney and Saleeby, 1990; Pickett and Saleeby, 1993; The subbatholith mantle lithosphere struc- Harms, 1984). Major structures of the Lar- Wood and Saleeby, 1998). The oblique crustal ture exhibited for the central SNB contrasts amide and Sevier deformation belts are shown section, in conjunction with seismic data sharply from the deep structure observed after Allmendinger (1992), Christiansen and (Fliedner et al., 2000) and xenolith data from along the Tehachapi-Rand deformation belt,

658 Geological Society of America Bulletin, June 2003 faults per metre. Like Beehive Flat, only sidered to postdate the formation of the been able to identify a dominant fault struc- one dominant set of straight fibers is typi- fault, it is useful to view the data in a fault- ture within the zone. Therefore, we con- cally present. However, five fault surfaces parallel reference frame, with the primitive clude that our measurements at the outcrop have a superimposed set of subhorizontal circle of the stereogram oriented parallel to scale are representative of the regional-scale striae, which we attributed to late Cenozoic the regional attitude of the fault zone. The brittle strain, which appears to be contrac- strike-slip deformation known to occur on data are transformed by rotation around the tional in the plane of the fault zone. the Hayward and Calaveras faults, which regional strike of the fault zone. This refer- flank the Diablo Range uplift (Fig. 2). These ence frame emphasizes the fact that the TECTONIC IMPLICATIONS slip data are not included in our analysis shortening axes now lie subparallel to the As discussed above, the traditional model here. In present coordinates, 79% of the fault zone (Fig. 3C), indicating crustal short- (e.g., Ingersoll, 1979) (Fig. 4A) for the tec- faults have a dip-slip sense of motion di- ening in a northeast (70°) subhorizontal tonic evolution of the Franciscan complex rected generally to the west (38 out of 55) but direction. Furthermore, those faults with does not account for the metamorphic break also to the east (17 out of 55). Most of these normal geometry in present coordinates ap- at the Coast Range fault. The extensional Vol. 128, 1988 Subduction-Channel Model: 1 479 faults (44 out of 55) would again be classed parently originated as reverse faults, dipping model (e.g., Piatt, 1986) postulates that the as normal faults. The remaining 21% are si- mainly to the east. metamorphic break is due to extensional prism by underplating, the requirement for and the depths of upflow of melange, nistral and dextral strike-slip faults, in equa98l At Del Puerto Canyon, the contoureRING AND dBRANDON faulting. Our kinematic data provide no ev- and numerous other details. It supplies a basis for classification of convergent proportions. shortening and extension directions (Fig. 3, idence of fault-parallel extension; instead, margins and provides quantitative constraints on, and detailed insight into, such E and F) show nearly identical relations, al- data indicate that brittle deformation within aspects as the patterns and rates of uplift from the trench slope through the forearc Analysis though the extension axes display a better the fault zone was dominated by fault- region, the mechanisms of blueschist uplift and metamorphic zonation during The contoured shortening directions for developed point maximum. In a fault-paral- parallel contraction. all sampled faults in the Beehive Flat area lel referenccontinuede frame, thsubduction,e data als othe indicat distributione W ande focu magnitudes here on oftw oearthquakes, viable models and, ththee (Fig. 3A) show a well-defined point maxi- fault-parallevolumesl shortenin of gdeeply (Fig. 3Gsubducted) in a north sediment- tectoni that ccan wedg contributee mode l toof arcWentwort magmas.h et al. mum oriented in a subvertical direction, east (55°) subhorizontal direction. Those (1984) (Fig. 4C) and the thin-skinned model subparallel to the present down-dip direc- faults with normal geometry in present co- of Suppe (1979) (Fig. 4D). Both models at- tion of the fault (great circles in Fig. 3 indi- ordinates apparently originated as Themainl Subduction-Channely tribute the structur Model e of the Franciscan- cate fault-zone orientation). The contoured east dippinBasicg revers Assumptionse faults ,and as Newindicate TerminOlogyd Grea t Valley contact to younger out-of- extension axes (Fig. 3B) show a girdle pat- when viewed in a fault-parallel reference sequence thrust faults. Both are compatible tern; a maximum, however, is apparent and frame. A central assumption of the subduction-channelwith fault-paralle modell contractio is thatn theon thsubductinge Coast has a subhorizontal direction trending east- Deformatiosedimentn withi deformsn the Coas approximatelyt Range faul tas aRang (note necessarilyfault, the metamorphiNewtonian) cviscous break fluidat th eas southeast (110°). Because the present steep zone appearit sis t odragged have bee byn thecharacterize descendingd bplatey beneathfault, an thed thoverridinge substantia crystallinel thinnin plateg o fand, th eif dip of the Coast Range fault is generally con- distributed onerevers is present,e faulting the . accretionaryNo one ha prism.s Coas Thet Ranglayere of ophiolite deforming. Fo sedimentr the thin-skinne comprisesd a kind of shear zone, called the subduction channel (or flow channel). The name emphasizes that the kinematic patterns within the channel differ significantly from Present reference frame Fault-parallel referencthosee fram ine typical shear zones (Figure Traditiona6), especiallyl mode wherel (e.g.upflow, Ingersoll occurs. ,Sediment 1979) shortening axes extension axes shortening axes extensiosubductionn axes occurs because the downward-directed shearing overcomes the buoy- BX^TT^X C/"" s. D. ancy and the (normally) adverse pressure gradient that act upon the low-density 7 / / / / // // // /, , d /atfsediment.l \ dThe exact pattern of flow and the thickness of the subduction channel ////// /y / / / / / ' • \ \ ' Subduction Channel model -----TR ENCHW,CR D ,r R C W,4R B--,.- B Extensional model (Harms et al., 1992) .... ~e_!o..a__t_~on ...... Inlet_~ ...... _sL~__~S v__q_...... _._.....--- front ~ Aeeretionary prism A GorvYUrtra~ine . ,~"-...~N~u/-.. ~ (if present) f ~ plate ~ Hanging Wall ")

FIG. 1. Interaction of exhumation mechanisms in various plate-tectonic settings. Figure 3. Contoured maximum-shortening and maximum-extension axes for brittle strain at Beehive Flat and Del Puerto Canyon. Regional strikeCloos, s, and Shreve, PAGEOPH, 1988 ate and down-dip direction, d, of fault zone are shown. DiagramCoulombs are synoptic ; wedge there is no evidence of structural domains. Contours are according to Kamb (1959) and are in multiples of uniform density startindeformationg at 1, with c WedgFigure 6e model (Wentworth et al., 1984) interval of 0.66. Major features of the subduction-channel model, shown schematically for a margin with an accretionary prism. The subduction• channel is the sediment-filled shear zone inV which. ' deformation is concentrated.\ \ \ \ • Figure 4. Four models of tectonic evolution of Franciscan complexFor clarity,. Sym its- thickness is greatly exaggerated.'/CRI The deformation front and the inlet bound the zone bols and patterns as in Figure 2, except for Franciscan, whicofh compression,is stippled which is the region of bulldozer-like action in front of the overriding prism. A control and undivided. A: Traditional model showing original configuratioRingpoint and Brandon,n (CP) of Cre is Intl -located Geol Rev., where 2010 the channel capacity decreases as a result of a sharp increase in the den- taceous convergent margin. Fore-arc massif is thrust onto Franciscasity of then com overlying- rocks, in the slope of the topographic surface, or in the dip of the descending plate, plex. B: Extensional model explaining contact of Franciscan compleor in a combinationx and of the three. A zone of upflow of subduction melange can develop when the amount Great Valley Group by Late Cretaceous-earlDownloaded by [University of Colorado at Boulder Libraries] 21:06 15 October 2013 y Tertiary top-side-east normal of sediment arriving at the control point exceeds its capacity. FIG. 2. Schematic illustration of the three exhumation processes: normalThin-skinne faulting, ductiled mode flow, andl (Suppe erosion. Normal, 1979) faults. C: Coast Range ophiolite as part of Cretaceous tectonic indenter that faultingOut-of-sequence covers both brittle normal faulting in the upper crust and normal-sense ductile shear zones in the deeper crust. wedged eastward between overlying Great Valley sequence and Sierran Ductile thinning refers to wholesale vertical shortening in the wedge. The circle shows an undeformed particle accreted arc. Coast Range fault is interpreted to be top-side-east out-of-sequencthrusts e thrust fault that postdated and obscured originaatl subduction-relatethe base of the wedge,d whichstruc -becomes deformed (indicated by ellipses), along the exhumation path. ture of margin. D: Model of Cenozoic west-vergent, out-of-sequence thrust faults. Ring and Brandon, Geology, 1994 ent geologic settings. Our paper distinguishes itself as an important exhumation process. We deliber- from earlier review articles in the following features: ately do not pay too much attention to normal fault- GEOLOGYSubduction, Augus tchannel 1994 model--stuf(1) goes discussion back of up divergent when tooand convergentmuch was settings; going down...maybeing because the importancebest for things of normal like faulting blueschist 73for7 knockers and the like. Wholesale deformation of the wedge (2) inclusion of a broad range of examples drawn exhuming deep-seated rocks has been stressed in from around the world; and (3) emphasis on erosion numerous papers and reviews (e.g. Platt, 1986, TC5006 HPA TA. OEO XRSO FTERN N IRAD AIA SCHISTS SALINAS DE SIERRA AND RAND THE OF EXTRUSION OF ROLE AL.: ET CHAPMAN f21 of 3 TC5006

Figure 1 Chapman et al., Tectonics, 2010

Part of that story is the fact that the schists are emplaced against the middle crust TC5006 CHAPMAN ET AL.: ROLE OF EXTRUSION OF THE RAND AND SIERRA DE SALINAS SCHISTS TC5006

zone with a top to the NNE sense of shear. With the exception McWilliams, 1982; McWilliams and Li, 1983, 1985] observed of the San Emigdio Mountains, which experienced significant in the batholithic rocks from the same region, and suggests Pliocene‐Quaternary contractile deformation [Davis, 1983], that the schist escaped major systematic rotation. It could be lower plate transport directions are uniform at S30W ± 10° argued that the Garlock fault may be a transrotation boundary within a ∼10,000 km2 region (Figure 1). The consistency [e.g., Dickinson, 1996] and that any rotation in the southern in schist transport direction stands in marked contrast to Sierra Nevada (e.g., in the Tehachapi schist body) would apparent vertical axis rotations of 45–90° [Kanter and not be expected in the Sierra de Salinas or Rand windows. However, the geometry and kinematics of Neogene‐Quaternary faulting in the southern Sierra Nevada are inconsistent with the proposed boundary [Mahéo et al., 2009]. 40 39 c. 100 Ma [30] In the SNB, circa 86 Ma Ar/ Ar cooling ages indicate that the schist reached its present position relative to the upper plate at about the same time that the upper plate cooled sufficiently to lock in a magnetic orientation (circa 88–80 Ma). These observations leave little time for rotation prior to schist arrival, and rule out major crustal block rotation following circa 86 Ma. We assume that following schist emplacement and cooling, any upper plate rotation that occurred would also involve rotation of the underlying schist. We conclude that a significant fraction of upper plate rotation of the southern SNB occurred in the Late Cretaceous coin- cident with schist ascent and cooling at circa 88–86 Ma. It is important to reiterate here that fabrics at the base of the upper plate and in the schist are generally parallel. We suggest that the trenchward flowing schist exerted traction at the base of the upper plate, driving upper plate extension and rotation. Chapman et al., Tectonics, 2010 We further speculate that this coupling increased toward the Chapman et al. worked with detailed structural info Rand fault (i.e., from high to low structural levels in the upper plate), leading to the observed upper plate clockwise deflec- tion into parallelism with the schist. This interpretation is at odds with models linking clockwise rotation in the southern SNB to post‐Cretaceous dextral transpression [Burchfiel and Davis, 1981; McWilliams and Li, 1985] or Miocene transtension [Ross et al.,1989;Dokka and Ross, 1995]. Although the west directed thrusting model of May

Figure 13. Block diagrams of the southern SNB showing the tectonic context of northward tilting and westward de- flection of the upper plate and the development of lower plate transport directions. (a) Circa 100 Ma. Immediately prior to collision of the LIP. (b) Circa 95 Ma. Segmentation of the Farallon slab into shallow and more deeply subducting portions, with intervening lateral ramp. To the north, the subbatholith mantle lithosphere is preserved, while to the south, the mantle wedge is sheared off by the subducting LIP. Schist protolith is shed into trench and begins to subduct while upper plate uplift and extensional collapse occur above the shallow segment. (c) Circa 85 Ma. Gravitational collapse of the upper plate drives trench‐directed channel- ized extrusion of the schist. Strain coupling between the schist and upper plate leads to continued extension and exhumation in the upper plate, and clockwise rotation in the upper plate. Schist flow is dominated by simple shear near the boundaries of the channel and by pure shear at the center of the wedge. Abbreviations KWF, Late Cretaceous Kern Canyon–White Wolf fault system; LIP, large igneous province; MSL, mean sea level; pKCF, proto‐Kern Canyon fault; SBML, subbatholith mantle lithosphere; SOML, sub- oceanic mantle lithosphere.

16 of 21 TC5006 CHAPMAN ET AL.: ROLE OF EXTRUSION OF THE RAND AND SIERRA DE SALINAS SCHISTS TC5006 zone with a top to the NNE sense of shear. With the exception McWilliams, 1982; McWilliams and Li, 1983, 1985] observed of the San Emigdio Mountains, which experienced significant in the batholithic rocks from the same region, and suggests Pliocene‐Quaternary contractile deformation [Davis, 1983], that the schist escaped major systematic rotation. It could be lower plate transport directions are uniform at S30W ± 10° argued that the Garlock fault may be a transrotation boundary within a ∼10,000 km2 region (Figure 1). The consistency [e.g., Dickinson, 1996] and that any rotation in the southern in schist transport direction stands in marked contrast to Sierra Nevada (e.g., in the Tehachapi schist body) would apparent vertical axis rotations of 45–90° [Kanter and not be expected in the Sierra de Salinas or Rand windows. However, the geometry and kinematics of Neogene‐Quaternary faulting in the southern Sierra Nevada are inconsistent with the proposed boundary [Mahéo et al., 2009]. 40 39 [30] In the SNB, circa 86 Ma Ar/ Ar cooling ages indicate that the schist reached its present position relative to the upper plate at about the same time that the upper plate cooled sufficiently to lock in a magnetic orientation (circa 88–80 Ma). These observations leave little time for rotation prior to schist arrival, and rule out major crustal block rotation following circa 86 Ma. We assume that following schist emplacement and cooling, any upper plate rotation that occurred would also involve rotation of the underlying schist. We conclude that a significant fraction of upper plate rotation of the southern SNB occurred in the Late Cretaceous coin- cident with schist ascent and cooling at circa 88–86 Ma. It is important to reiterate here that fabrics at the base of the upper plate and in the schist are generally parallel. We suggest that the trenchward flowing schist exerted traction at the base of the upper plate, driving upper plate extension and rotation. We further speculate that this coupling increased toward the c. 95 Ma Rand fault (i.e., from high to low structural levels in the upper plate), leading to the observed upper plate clockwise deflec- tion into parallelism with the schist. This interpretation is at odds with models linking clockwise rotation in the southern SNB to post‐Cretaceous dextral transpression [Burchfiel and Davis, 1981; McWilliams and Li, 1985] or Miocene transtension [Ross et al.,1989;Dokka and Ross, 1995]. Although the west directed thrusting model of May

Chapman et al., Tectonics, 2010 ing portions, with intervening lateral ramp. To the north, the subbatholith mantle lithosphere is preserved, while to the south, the mantle wedge is sheared off by the subducting LIP. Schist protolith is shed into trench and begins to subduct while upper plate uplift and extensional collapse occur above the shallow segment. (c) Circa 85 Ma. Gravitational collapse of the upper plate drives trench‐directed channel- ized extrusion of the schist. Strain coupling between the schist and upper plate leads to continued extension and exhumation in the upper plate, and clockwise rotation in the upper plate. Schist flow is dominated by simple shear near the boundaries of the channel and by pure shear at the center of the wedge. Abbreviations KWF, Late Cretaceous Kern Canyon–White Wolf fault system; LIP, large igneous province; MSL, mean sea level; pKCF, proto‐Kern Canyon fault; SBML, subbatholith mantle lithosphere; SOML, sub- oceanic mantle lithosphere.

16 of 21 TC5006 CHAPMAN ET AL.: ROLE OF EXTRUSION OF THE RAND AND SIERRA DE SALINAS SCHISTS TC5006

zone with a top to the NNE sense of shear. With the exception McWilliams, 1982; McWilliams and Li, 1983, 1985] observed of the San Emigdio Mountains, which experienced significant in the batholithic rocks from the same region, and suggests Pliocene‐Quaternary contractile deformation [Davis, 1983], that the schist escaped major systematic rotation. It could be lower plate transport directions are uniform at S30W ± 10° argued that the Garlock fault may be a transrotation boundary within a ∼10,000 km2 region (Figure 1). The consistency [e.g., Dickinson, 1996] and that any rotation in the southern in schist transport direction stands in marked contrast to Sierra Nevada (e.g., in the Tehachapi schist body) would apparent vertical axis rotations of 45–90° [Kanter and not be expected in the Sierra de Salinas or Rand windows. However, the geometry and kinematics of Neogene‐Quaternary faulting in the southern Sierra Nevada are inconsistent with the proposed boundary [Mahéo et al., 2009]. 40 39 [30] In the SNB, circa 86 Ma Ar/ Ar cooling ages indicate that the schist reached its present position relative to the upper plate at about the same time that the upper plate cooled sufficiently to lock in a magnetic orientation (circa 88–80 Ma). These observations leave little time for rotation prior to schist arrival, and rule out major crustal block rotation following circa 86 Ma. We assume that following schist emplacement and cooling, any upper plate rotation that occurred would also involve rotation of the underlying schist. We conclude that a significant fraction of upper plate rotation of the southern SNB occurred in the Late Cretaceous coin- cident with schist ascent and cooling at circa 88–86 Ma. It is important to reiterate here that fabrics at the base of the upper plate and in the schist are generally parallel. We suggest that the trenchward flowing schist exerted traction at the base of the upper plate, driving upper plate extension and rotation. We further speculate that this coupling increased toward the Rand fault (i.e., from high to low structural levels in the upper plate), leading to the observed upper plate clockwise deflec- tion into parallelism with the schist. This interpretation is at odds with models linking clockwise rotation in the southern SNB to post‐Cretaceous dextral transpression [Burchfiel and Davis, 1981; McWilliams and Li, 1985] or Miocene transtension [Ross et al.,1989;Dokka and Ross, 1995]. Although the west directed thrusting model of May

Figure 13. Block diagrams of the southern SNB showing the tectonic context of northward tilting and westward de- flection of the upper plate and the development of lower plate transport directions. (a) Circa 100 Ma. Immediately prior to collision of the LIP. (b) Circa 95 Ma. Segmentation of the Farallon slab into shallow and more deeply subducting portions, with intervening lateral ramp. To the north, the subbatholith mantle lithosphere is preserved, while to the c. 85 Ma south, the mantle wedge is sheared off by the subducting LIP. Schist protolith is shed into trench and begins to subduct while upper plate uplift and extensional collapse occur above the shallow segment. (c) Circa 85 Ma. Gravitational collapse of the upper plate drives trench‐directed channel- ized extrusion of the schist. Strain coupling between the schist and upper plate leads to continued extension and exhumation in the upper plate, and clockwise rotation in the upper plate. Schist flow is dominated by simple shear near the boundaries of the channel and by pure shear at the center of the wedge. Abbreviations KWF, Late Cretaceous Kern Canyon–White Wolf fault system; LIP, large igneous province; MSL, mean sea level; pKCF, proto‐Kern Canyon fault; SBML, subbatholith mantle lithosphere; SOML, sub- oceanic mantle lithosphere.

Chapman et al., Tectonics,16 2010 of 21

Note timing and substantial normal faulting. SEGMENTATION OF THE LARAMIDE SLAB

reduction in material strength (Tullis and Yund, 1980; Blanpied et al., 1995; Kohlstedt et al., 1995; Wintsch et al., 1995; Tullis et al., 1996). Stable isotope and petrologic studies of the Franciscan Catalina schist indicate very high water fluxes during progressive subduction and metamorphism from 5 to 11 kb conditions (Bebout and Barton, 1989). The Cat- alina schist is very similar to the schists that are underplated beneath the MSB. An analo- gous high-water flux migrating from the underplated schist protoliths into the overlying quartzofeldspathic batholithic crust, as evi- denced by widespread retrograding in the upper plate rocks, would promote substantial weakening of the upper plate as well. It is suggested that these extreme tectonic and rheological conditions resulted in the production of a highly weakened orogenic crustal section along the plate edge that was highly suscep- Saleeby, GSA Bull 2003 tible to gravitational collapse. Malin et al. (1995) point this out and imply that this alone can account for the collapse and forearc breachment of the MSB. It is further suggested below that the passing of the shallow slab segment and the resulting steepening of the slab dip promoted regional extension along the plate edge and, in conjunction with the extreme rheological conditions noted above, a prolonged episode of vigorous orogenic collapse characterized medial to late Laramide time for the MSB. Figure 4 is a model that shows the sequence of events that led to the uplift and collapse of the MSB. This model is intended to be ge- neric for the MSB region and reflects the sequence of events affecting the northern- MSB–southern-SNB transition perhaps up to 10 m.y. earlier than for the southern MSB region. The model is based on the concept of the shallow slab segment resulting from the subduction of the Towisangna Ridge (Livac- Figure 4. Series of generalized cross section models of the southwest Cordilleran plate cari et al., 1981; Hendersen et al., 1984; Barth edge in the region of the Mojave-Salinia segment of batholithic belt (MSB) showing how and Schneiderman, 1996). The Figure 4 model the Laramide shallow slab segment deformed the forearc-arc region as subduction flat- also incorporates the work of McNulty and tened; then as subduction steepened, regional extension promoted orogenic collapse and Farber (2002), which documents the early forearc breachment of the disrupted arc segment. Superpositioning of structures is not phases of crustal collapse of the Peruvian An- shown in continental plate for reasons of simplicity. des above its corresponding shallow slab segment. This work suggests that the initiation of orogenic collapse there is linked in time with wallrocks (Jacobson et al., 2000). These data al., 1986; Trehu and Wheeler, 1987; Li et al., the passing of the trailing flank of the sub- suggest rapid uplift and erosion of the MSB 1992; Malin et al., 1995; Hauksson, 2000). ducted Nazca Ridge beneath the forearc re- sediment source, as well as rapid tectonic Exposures of the schists are dominated by me- gion. The corollary added here is that the sub- transport of the corresponding clastic taclastic rocks but also contain significant hy- duction of less buoyant abyssal lithosphere, in wedge(s) into the metamorphic environment. drous metabasalts and lesser serpentinites. The the wake of the subducted Nazca Ridge, in- Seismic data from beneath the Tehachapi- tectonic replacement of the subbatholith man- duces a steepening in the slab that further in- Rand deformation belt, as well as from Salinia tle lithosphere by such an assemblage effec- duces extensional tectonics in the overlying and the western Mojave Desert, suggest un- tively replaces a dry pyroxene ⇥ garnet ⇥ orogenically thickened crust. derplating of perhaps up to an 30-km-thick olivine rheology with a wet mica ⇥ feldspar Figure 4A shows in highly generalized wedge of schist-affinity material (Cheadle et ⇥ quartz rheology, resulting in a tremendous fashion the principal plate edge elements in

Geological Society of America Bulletin, June 2003 663 SEGMENTATION OF THE LARAMIDE SLAB

reduction in material strength (Tullis and Yund, 1980; Blanpied et al., 1995; Kohlstedt et al., 1995; Wintsch et al., 1995; Tullis et al., 1996). Stable isotope and petrologic studies of the Franciscan Catalina schist indicate very high water fluxes during progressive subduction and metamorphism from 5 to 11 kb conditions (Bebout and Barton, 1989). The Cat- alina schist is very similar to the schists that are underplated beneath the MSB. An analo- gous high-water flux migrating from the underplated schist protoliths into the overlying quartzofeldspathic batholithic crust, as evi- denced by widespread retrograding in the upper plate rocks, would promote substantial weakening of the upper plate as well. It is suggested that these extreme tectonic and rheological conditions resulted in the production of a highly weakened orogenic crustal section along the plate edge that was highly suscep- tible to gravitational collapse. Malin et al. (1995) point this out and imply that this alone can account for the collapse and forearc breachment of the MSB. It is further suggested below that the passing of the shallow slab segment and the resulting steepening of the slab dip promoted regional extension along the plate edge and, in conjunction with the extreme rheological conditions noted above, a prolonged episode of vigorous orogenic collapse characterized medial to late Laramide time for the MSB. Figure 4 is a model that shows the sequence of events that led to the uplift and collapse of the MSB. This model is intended to be ge- neric for the MSB region and reflects the sequence of events affecting the northern- MSB–southern-SNB transition perhaps up to 10 m.y. earlier than for the southern MSB region. The model is based on the concept of the shallow slab segment resulting from the subduction of the Towisangna Ridge (Livac- Figure 4. Series of generalized cross section models of the southwest Cordilleran plate cari et al., 1981; Hendersen et al., 1984; Barth edge in the region of the Mojave-Salinia segment of batholithic belt (MSB) showing how and Schneiderman, 1996). The Figure 4 model the Laramide shallow slab segment deformed the forearc-arc regionSaleeby, as subduction GSA Bull 2003 flat- also incorporates the work of McNulty and tened; then as subduction steepened, regional extension promoted orogenic collapse and Farber (2002), which documents the early forearc breachment of the disrupted arc segment. Superpositioning of structures is not phases of crustal collapse of the Peruvian An- shown in continental plate for reasons of simplicity. des above its corresponding shallow slab segment. This work suggests that the initiation of orogenic collapse there is linked in time with wallrocks (Jacobson et al., 2000). These data al., 1986; Trehu and Wheeler, 1987; Li et al., the passing of the trailing flank of the sub- suggest rapid uplift and erosion of the MSB 1992; Malin et al., 1995; Hauksson, 2000). ducted Nazca Ridge beneath the forearc re- sediment source, as well as rapid tectonic Exposures of the schists are dominated by me- gion. The corollary added here is that the sub- transport of the corresponding clastic taclastic rocks but also contain significant hy- duction of less buoyant abyssal lithosphere, in wedge(s) into the metamorphic environment. drous metabasalts and lesser serpentinites. The the wake of the subducted Nazca Ridge, in- Seismic data from beneath the Tehachapi- tectonic replacement of the subbatholith man- duces a steepening in the slab that further in- Rand deformation belt, as well as from Salinia tle lithosphere by such an assemblage effec- duces extensional tectonics in the overlying and the western Mojave Desert, suggest un- tively replaces a dry pyroxene ⇥ garnet ⇥ orogenically thickened crust. derplating of perhaps up to an 30-km-thick olivine rheology with a wet mica ⇥ feldspar Figure 4A shows in highly generalized wedge of schist-affinity material (Cheadle et ⇥ quartz rheology, resulting in a tremendous fashion the principal plate edge elements in

Geological Society of America Bulletin, June 2003 663 Late Cretaceous–early Cenozoic tectonic evolution of southern California

RESULTS 10 Mio 20 Pelona-Orocopia-Rand and Metamorphic bio Catalina Schists 30 Olig mus hbd Aside from the fact that we did not previ- Detrital 40 ously consider the Catalina Schist, our results zircon Eoc are little changed from those obtained by Grove 50 erv al et al. (2003). We thus emphasize only those int points most relevant for comparison with the 60 Age (Ma) Pal forearc data. Cy cling Maast Slip on Nac flt 70 Protolith Age Camp One of the most critical parameters to de- 80 termine for any given schist body is the depo- Sant Con sitional age of the sandstone protolith, which 90 must lie within a time window bounded by the Tur Cen youngest reliable detrital zircon age and oldest 100 reliable metamorphic age (Fig. 3). Grove et al. Age/ SC SE PRRA SS MP SP BR EF OR CH TR CD NR (2003) referred to this time span as the “cycling Epoch Cat-San Emig Rand Pelona Orocopia interval,” because it encompasses erosion in North South 40 39 the source area, transport of that material to the Figure 3. Comparison of Ar/ Ar cooling ages and zircon U-Pb ages ≤100 Ma for the site of deposition, underthrusting beneath the Pelona-Orocopia-Rand-Catalina schists plotted by area (see Figs. 1, 2, and 4 for locations). arc, accretion to the overriding plate, and initial Zircon ages are from Barth et al. (2003), Grove et al. (2003, 2008), andJacobson this et al.,study. GSA Bull, Argon 2010 stages of exhumation. As observed by Grove ages are from Jacobson (1990), Jacobson et al. (2002, 2007), Barth et al. (2003), and Grove etWould al. (2003), seem the cyclingif this intervalis driven of the by Pelona- plateauet subduction al. (2003, 2008). that Thewe havePelona-Orocopia-Rand 40-50 m.y. of suchschists subduction... are organized by the conventional Orocopia-Rand schists becomes younger from northwest-southeast grouping, i.e., Rand Schists in the northwest Mojave Desert and envi- northwest to southeast (Fig. 3). Also notable, rons, Pelona Schists in the central Transverse Ranges, and Orocopia Schists in southeast- the duration of the cycling interval increases ern California–southwestern Arizona. The exception is the Rand Schist of the San Emigdio to the southeast, although this may be at least Mountains, which we group with the Catalina Schist. Within the Pelona and Orocopia partly an artifact in the data. For example, we Schist groups, individual areas are ordered from northwest on the left to southeast on the obtained no hornblende 40Ar/39Ar ages from right. Northwest-southeast position for the Rand Schists is not clear owing to potential ro- the easternmost schist bodies (Fig. 3) and tation of the Sierran tail; these bodies were instead ordered from left to right by deposi- thus may not have captured the oldest cool- tional and metamorphic age. Diagonally ruled box indicates time of most likely movement ing ages in this region. In addition, igneous on the Nacimiento fault. Yellow band indicates the cycling interval, which, as discussed in 40 39 bodies younger than ca. 70 Ma are not com- the text, is not well constrained in the southeast. Also note that Ar/ Ar and zircon ages mon in the inferred provenance areas (Barth overlap for the Catalina Schist (the two types of ages have been separated slightly along et al., 2008a), making it diffi cult to constrain the x-axis to clarify this relation). The low grade of metamorphism for this unit (Grove 40 39 the maximum depositional age of sediments et al., 2008) suggests that the Ar/ Ar ages may be infl uenced by excess radiogenic argon younger than latest Cretaceous. For example, or lack of complete recrystallization of detrital mica during metamorphism. Abbreviations in delineating the cycling interval, we ignored for schist localities are same as in Figure 1. Other abbreviations are: bio—biotite; Camp— three early Cenozoic ages from the Orocopia Campanian; Cen—Cenomanian; Eoc—Eocene; hbd—hornblende; Maast—Maastrichtian; Schist due to their status as outliers (Fig. 3). Mio—Miocene; mus—muscovite; Olig—Oligocene; Pal—Paleocene; Sant—Santonian; However, it should be kept in mind that any or Tur—Turonian. all of these ages could be signifi cant. Despite the uncertainties, the schist protoliths clearly range from Turonian to at least as young as Variation in Detrital Zircon Populations as a taceous zircons, with lesser, but still signifi cant, middle Paleocene, representing a time span of Function of Depositional Age and Location proportions of Proterozoic and Jurassic ages. 30 m.y. or greater. Pie diagrams of detrital zircon ages for indi- In contrast, the central to southeastern schists, For purposes of comparison to the forearc vidual ranges and/or adjacent ranges are plot- which have depositional ages of Maastrichtian units, we divided the schists into four groups ted in Figure 4 using the palinspastic base of and younger, are characterized by progressively based on depositional age and geographic Figure 2. Probability density functions for the decreasing abundances of Early to middle Cre- location (Figs. 3 and 4). This contrasts with four regional groupings are illustrated in Figures taceous detrital zircons and increasing num- a threefold division utilized by Grove et al. 5B–5E. As pointed out by Grove et al. (2003), bers of Proterozoic and latest Cretaceous–early (2003). Our fourth category includes the detrital zircon patterns in the Pelona-Orocopia- Cenozoic ages. As concluded by Grove et al. Catalina Schist, which was not considered by Rand schists vary systematically from north- (2003), the patterns imply a shift from outboard Grove et al. (2003), and the Pelona-Orocopia- west to southeast, and thus with depositional to inboard sources with time (e.g., Fig. 2 and Rand schist of the San Emigdio Mountains, age of the protolith. The northwestern schists, following discussion). which Grove et al. (2003) grouped with the which have protolith ages of Turonian to Cam- Compared to the Pelona-Orocopia-Rand Rand schists. panian, are dominated by Early to middle Cre- schists, the zircon age distribution within

Geological Society of America Bulletin, March/April 2011 491 Late Cretaceous–early Cenozoic tectonic evolution of southern California

RESULTS 10 Jacobson et al. Mio 20 Pelona-Orocopia-Rand and Metamorphic bio Catalina Schists 30 Olig mus forearc sequences would have been deposited would migrate hbd in the same direction, leading to Aside from the fact that we did not previ- Detrital Rand 40 on opposite sides (east and west, respectively) ously consider the Catalina Schist, our results progressively zircon younger packets of accreted mate- 70 50 Ma Eoc are little changed from those obtained by Grove al of the intraoceanic terrane. However, it seems Catalina (NW Ma 50rial away from the trench. We refer to the pre- erv et al. (2003). We thus emphasize only those int points90 most relevant for comparison with the highly unlikely that the detrital zircon age pat- Pelona-Orocopia 60dicted geometry as a “tectonic onlap” based on Age (Ma) Pal terns of the schists and forearc units would be forearcMa data. its resemblance to an upside-down sedimentaryCy cling Maast Slip on Nac flt 70 so strongly correlated over >30 m.y. (Fig. 5) if Protolithv Age onlap (clearest in Fig. 8D). s. One of the most critical parameters to de- Camp their protoliths had been deposited in isolated SE 80 termine for any given schist body is the depo- Continental options) Sant basins. The McCoy data, in particular, imply MigratingCon Aseismic Ridge margin sitional age of the sandstone protolith, which 90 must lie within a time window bounded by the Tur that a backarc basin would be highly enriched in AsCen indicated already, Barth and Schneider- youngest reliable detrital zircon age and oldest 100 cratonal detritus compared to the forearc region reliable metamorphic age (Fig. 3). Grove et al. manAge/ (1996)SC proposedSE PR RAthat SS the Pelona-Orocopia-MP SP BR EF OR CH TR CD NR A Epoch Cat-San Emig Rand Pelona Orocopia (Figs. 6B and 7D). Consequently, we interpret (2003) referred to this time span as the “cycling Rand schists were derived by thrusting of the interval,” because it encompasses erosion in 40 39 the zircon results to indicate that the forearc the source area, transport of that material to the Figurearc 3. over North Comparison the forearc of Ar/ basin.Ar cooling They ages attributed and zircon U-Pb this ages ≤100South Ma for the Ma Pelona-Orocopia-Rand-Catalina schists plotted by area (see Figs. 1, 2, and 4 for locations). units and schists represent the proximal (forearc 90site of deposition, underthrusting beneath the event to subduction of an aseismic ridge (cf. arc,Rand accretion to the overriding plate, and initial Zircon ages are from Barth et al. (2003), Grove et al. (2003, 2008), and this study. Argon basin) and distal (trench) facies, respectively, stages of exhumation. As observed by Grove agesLivaccari are from Jacobson et al., (1990), 1981; Jacobson Henderson et al. (2002, et 2007), al., Barth1984) et al. (2003), and Grove Catalina of a single depositional system on the outboard et al. (2003), the cycling interval of the Pelona- et al.with (2003, its 2008). long The dimension Pelona-Orocopia-Rand oriented schists slightly are organized more by the conventional Orocopia-Rand schistsPelo becomes younger from northwest-southeast grouping, i.e., Rand Schists in the northwest Mojave Desert and envi- side of the Sierran–Peninsular Ranges arc. Be- na-Orocopia rons,northerly Pelona Schists than in thethe central convergence Transverse Ranges,vector and between Orocopia Schists in southeast- ( northwest to southeast (Fig. 3). Also notable, NW 70 Ma the duration of the cycling interval increases ern California–southwestern Arizona. The exception is the Rand Schist of the San Emigdio tween the previously identifi ed weaknesses Mountains,the Farallon which we and group North with theAmerican Catalina Schist.plates. Within In this the Pelona and Orocopia tovs. the southeast, although this may be at least Schist groups, individual areas are ordered from northwest on the left to southeast on the (e.g., Burchfi el and Davis, 1981; Crowell, 1981) partlySE an artifact in the data. For example, we case, the intersection point between ridge and 40 39 right. Northwest-southeast position for the Rand Schists is not clear owing to potential ro- Continental obtainedoptions) no hornblende Ar/ Ar ages from and zircon data, we see little justifi cation for re- tationtrench of the Sierran would tail; have these bodies migrated were instead southward ordered from with left to right by deposi- margin the easternmost schist bodies (Fig. 3) and 50 Ma taining the backarc models. thus may not have captured the oldest cool- tionaltime, and metamorphicpresumably age. causing Diagonally the ruled thrust box indicates between time arc of most likely movement 100 km ing ages in this region. In addition, igneous on the Nacimiento fault. Yellow band indicates the cycling interval, which, as discussed in and forearc basin to propagate in the same direc-40 39 B bodies younger than ca. 70 Ma are not com- the text, is not well constrained in the southeast. Also note that Ar/ Ar and zircon ages Northwest-Southeast Decreasing Age of the mon in the inferred provenance areas (Barth overlaption. for As the discussedCatalina Schist previously, (the two types we of agesconsider have been that separated slightly along et al., 2008a), making it diffi cult to constrain the x-axis to clarify this relation). The low grade of metamorphism for this unit (Grove Pelona-Orocopia-Rand Schists the schists were derived40 39 from trench materials the maximum depositional age of sediments et al., 2008) suggests that the Ar/ Ar ages may be infl uenced by excess radiogenic argon Figure 9. End-member possibilities for age or lackrather of complete than recrystallizationfrom the underthrust of detrital mica forearc during metamorphism.basin. Abbreviations younger than latest Cretaceous. For example, Jacobson et al., GSA Bull, 2010 One of the most distinctive features of the variation withinin delineating a sheet the of cycling Pelona-Orocopia- interval, we ignored for schistNonetheless, localities are thesame geometricas in Figure 1. argumentOther abbreviations of ridge are: bio—biotite; Camp— Rand schistthree underplated early Cenozoic ages beneath from the Orocopia North Campanian; Cen—Cenomanian; Eoc—Eocene; hbd—hornblende; Maast—Maastrichtian; Pelona-Orocopia-Rand schists is the >30 m.y. Schist due to their status as outliers (Fig. 3). Mio—Miocene;migration mus—muscovite;seems applicable Olig—Oligocene; to either Pal—Paleocene;situation. Sant—Santonian; American crust. Northwest and southeast Tur—Turonian. Woulddecrease seem in depositional if this is and driven emplacement by plateau ages subductionHowever, that it should we be kept have in mind that40-50 any or m.y.A similar of such interpretation subduction... was also proposed by from northwest to southeast (Fig. 3). Because of locations of all the of these Catalina ages could Schist be signifi reflcant. ect Despite the Dickinson et al. (1988, p. 1036) to explain north the uncertainties, the schist protoliths clearly the relatively linear distribution of the schist out- cases of 500–600range fromand Turonian 0 km to of at leastsinistral as young slip as Variationto south in Detrital migration Zircon Populations of Laramide as a taceous deformation zircons, with in lesser, but still signifi cant, crops, it is not clear whether this age variation on the Nacimientomiddle Paleocene, fault, representing respectively. a time span The of Functionthe foreland of Depositional region. Age and Location proportions of Proterozoic and Jurassic ages. 30 m.y. or greater. Pie diagrams of detrital zircon ages for indi- In contrast, the central to southeastern schists, relates more to distance inboard from the conti- fi gure is highlyFor schematic purposes of comparison and does to thenot forearc take vidual ranges and/or adjacent ranges are plot- which have depositional ages of Maastrichtian nental edge (Fig. 9A) or to position along strike into accountunits, the we narrowing divided the schists of into the four margin groups ted inSinistral Figure 4 usingSlip the on palinspastic the Nacimiento base of and Fault younger, are characterized by progressively that would havebased resulted on depositional from age displacement and geographic Figure 2. Probability density functions for the decreasing abundances of Early to middle Cre- of the margin (Fig. 9B). Consideration of the location (Figs. 3 and 4). This contrasts with four regionalThe groupings model are of illustrated a migrating in Figures ridge-trenchtaceous detrital zircons in- and increasing num- Catalina Schist provides some additional con- on the Nacimientoa threefold fault, division whether utilized by by Grove thrust- et al. 5B–5E.tersection As pointed out seems by Grove particularly et al. (2003), effectivebers of Proterozoic in ex- and latest Cretaceous–early ing or strike (2003).slip. Our fourth category includes the detrital zircon patterns in the Pelona-Orocopia- Cenozoic ages. As concluded by Grove et al. straints but does not fully solve the problem. As- Catalina Schist, which was not considered by Randplaining schists vary systematicallythe northwest-southeast from north- (2003), thedecrease patterns imply a shift from outboard suming no sinistral slip on Nacimiento fault, the Grove et al. (2003), and the Pelona-Orocopia- westin to agesoutheast, of and the thus schists, with depositional assuming to inboard sinistral sources slip with time (e.g., Fig. 2 and Rand schist of the San Emigdio Mountains, age of the protolith. The northwestern schists, following discussion). Catalina Schist would initially have been located which Grove et al. (2003) grouped with the whichon have the protolith Nacimiento ages of Turonian fault. to Cam- This isCompared illustrated to the in Pelona-Orocopia-Rand outboard of the Pelona and Orocopia Schists older schists toRand be schists. underplated at increasingly panian,Figure are dominated 10, where by Early panel to middle A Cre-representsschists, the the zirconpaleo- age distribution within (southeast option in Figs. 9A and 9B). In view of greater distances inboard (Fig. 8). In this case, geography prior to fault slip. At this time, the the ca. 95–90 Ma age of the Catalina Schist, this age contours would parallel the strike of Geological the northwestern Society of America schist Bulletin, localities March/April 2011would have been 491 reconstruction is consistent with the isochrons of margin (Fig. 9A), and the young age of the situated relatively close to the trench, requir- Figure 9A, but not those of Figure 9B. In other southeastern schists would be a consequence ing only modest fl at subduction to emplace words, the southeast option for the Catalina of their position relatively far inboard. We en- trench materials beneath the former arc (Figs. Schist requires at least some component of de- vision this process involving transfer, due to 8B and 10A). At the same time, the future sites creasing age from outboard to inboard within the buoyancy, of some fraction of trench materi- of the Pelona and Orocopia Schists would have schist terrane. Alternatively, for the case of 500– als from the subducting (Farallon) to overrid- resided far inboard, in a region of arc magma- 600 km of sinistral strike slip on the Nacimiento ing (North American) plate above the hinge at tism (Fig. 10A; Mattinson, 1990; Barth et al., fault, the Catalina Schist would restore to the which the subducting plate descended toward 1995, 1997, 2001a, 2008a; Kidder et al., 2003; northwest end of the Pelona-Orocopia-Rand the mantle. With time, such subcreted materi- Barbeau et al., 2005). With commencement schist belt. This location is consistent with either als could become quite thick, leading to tectonic of Nacimiento slip, the latter regions would decreasing age outboard-inboard (Fig. 9A) or and/or erosional denudation of the overlying have been drawn toward the continental mar- northwest-southeast (Fig. 9B). These uncertain- plate (cf. Platt, 1986; Yin, 2002). In this fashion, gin (Fig. 9B) in the same way that slip on a ties require that we consider several mechanisms underplated materials would be driven upward, low-angle normal fault will translate rocks of to explain the observed variation of age within consistent with thermochronologic evidence the lower plate closer to the surface. As a con- the schist terrane. for exhumation of the Pelona-Orocopia-Rand sequence, progressively more southeastward schists concurrent with subduction (Jacobson (inboard) regions would have been transferred Progressive Subduction Erosion et al., 2002, 2007). This would allow permanent into the zone of underplating near the conti- Grove et al. (2003) proposed that buzz-saw incorporation of some trench materials within nental edge. In this scenario, the schists need removal of the base of North America could the overriding plate. As the fl at slab continued not defi ne more than a relatively narrow band allow younger schists to “leap-frog” beneath to erode inboard, the locus of underplating (becoming younger to the southeast) beneath

500 Geological Society of America Bulletin, March/April 2011 Jacobson et al.

forearc sequences would have been deposited would migrate in the same direction, leading to on opposite sides (east and west, respectively) Rand progressively younger packets of accreted mate- 70 50 Ma

of the intraoceanic terrane. However, it seems Catalina (NW Ma rial away from the trench. We refer to the pre- highly unlikely that the detrital zircon age pat- 90 Pelona-Orocopia dicted geometry as a “tectonic onlap” based on

terns of the schists and forearc units would be Ma its resemblance to an upside-down sedimentary

so strongly correlated over >30 m.y. (Fig. 5) if v onlap (clearest in Fig. 8D). s. their protoliths had been deposited in isolated SE Continental options) basins. The McCoy data, in particular, imply margin Migrating Aseismic Ridge that a backarc basin would be highly enriched in As indicated already, Barth and Schneider- cratonal detritus compared to the forearc region A man (1996) proposed that the Pelona-Orocopia- (Figs. 6B and 7D). Consequently, we interpret Rand schistsLate Cretaceous–early were derived byCenozoic thrusting tectonic of the evolution of southern California the zircon results to indicate that the forearc arc over the forearc basin. They attributed this Ma 90 10 units and schists represent the proximal (forearc Rand RESULTS event to subduction of an aseismic ridge (cf. basin) and distal (trench) facies, respectively, Livaccari et al., 1981; HendersonMio et al., 1984) Catalina 20 of a single depositional system on the outboard Pelona-Orocopia-Rand and with its long dimension oriented slightlyMetamorphic more Pelo bio side of the Sierran–Peninsular Ranges arc. Be- Catalina Schistsna-Orocopia northerly than the convergenceOlig vector between ( mus NW 70 Ma 30 tween the previously identifi ed weaknesses the Farallon and North American plates. Inhbd this vs. Aside from the fact that we did not previ- Detrital (e.g., Burchfi el and Davis, 1981; Crowell, 1981) SE case, the intersection point40 between ridge and ously consider the Catalina Schist, our results zircon and zircon data, we see little justifi cation for re- Continental options) trench would have migratedEoc southward with margin are little changed from those obtained by Grove 50 Ma al taining the backarc models. time, presumably causing50 the thrust between arc nterv et al. (2003). We100 thuskm emphasize only those i and forearc basin to propagate in the same direc- points most relevant for comparison with the B Late Cretaceous–early Cenozoic tectonic evolution60 of southern California Northwest-Southeast Decreasing Age of the tion. As discussed previously,Age (Ma) Pal we consider that forearc data. y cling Pelona-Orocopia-Rand Schists the schists were derived from trench materials C Maast Slip on Nac flt Figure 9. End-member possibilities for age rather than from the underthrust70 forearc basin. variation within a sheetProtolith of Pelona-Orocopia- Age workers have interpreted this “borderland-style” One of the most distinctiveA features ca. 75 of Ma the blk Nacimiento DiabloOne of the most critical B parametersNonetheless, ca. 60 to de-Ma the geometricCamp argument of ridge Pelona-Orocopia-Rand schists is the >30 m.y. Rand schist underplatedrange beneath North migration seems applicable80 to either situation.geometry as direct evidence that strike-slip termine for any given schist body is the depo- Sant decrease in depositional and emplacement ages American crust. Northwest and southeast A similar interpretation wasCon also proposed by sitional age of the sandstone protolith, which 90 faulting, whether in a dextral or sinistral sense,

locationsp l a t e of the Catalina Schist reﬂ ect the Tur from northwest to southeast (Fig.Farallon–North 3). Because of must lie within a time windowDickinson bounded byet al.the (1988, p. 1036) to explain northplayed an important role in emplacement of the the relatively linear distribution of the schist out- cases of 500–600 andyoungest 0 km of reliable sinistral detrital slip zircon to agesouth and migration oldest of LaramideCen deformation in America 100 Salinian block (Nilsen and Clarke, 1975; How- crops, it is not clear whether this age variation on the Nacimiento fault,reliable respectively. metamorphicNevada Theage (Fig. the 3). foreland Grove et region. al. Age/Nevada SC SE PRRA SS MP SP BR EF OR CH TR CD NR relates more to distance inboard from the conti- ﬁ gure is highly schematic(2003) and referred does to not this take time span as the “cycling Epoch Cat-San Emigell andRand Vedder, 1978; Pelona Vedder et al., 1983; Orocopia Dick- into account the narrowinginterval,” of because the margin it encompasses erosion in inson et al., 2005). However, whereas strike-slip

nental edge (Fig. 9A) or to position along strike SinistralF a Slip on the Nacimiento Fault Figure 3. Comparison of 40Ar/39Ar cooling ages and zircon U-Pb ages ≤100 Ma for the of the margin (Fig. 9B). Consideration of the that would haveNf resultedthe source from area,displacement transport of that Thematerial model to the of a migrating ridge-trenchNorth in-faulting can generate local topographic highsSouth r a l l o n p Nf Pelona-Orocopia-Rand-Catalina schists plotted by area (see Figs. 1, 2, and 4 for locations). 100 km on the Nacimiento fault,site ofwhether Sierradeposition, by thrust- underthrusting beneath the Sierra Catalina Schist provides some additional con-F a r a l l o n tersection seems particularlyZircon ages effectiveare from inBarth ex-and et al. lows, (2003), it isGrove not et clear al. (2003, that 2008), this process,and this study. by Argon straints but does not fully solve the problem. As- ing or strikeCS slip. arc, accretion to the overridingplaining plate, and initial the northwest-southeast decrease stages Gf of exhumation. As observed by Grove agesGf are from Jacobson (1990),itself, Jacobson can account et al. (2002, for 2007), the substantial Barth et al. net(2003), sub- and Grove suming no sinistral slip on NacimientoN fault, the in age of the schists, assuming sinistral slip et al. (2003), the cycling interval of the Pelona- et al. (2003, 2008). The Pelona-Orocopia-Rand schists are organized by the conventional Catalina Schist would initially have been located on the Nacimiento fault. This is illustrated insidence implied by the regional transgression Orocopia-Rand schists becomes younger from northwest-southeast grouping, i.e., Rand Schists in the northwest Mojave Desert and envi- Salinian blk Sal of the Salinian block. Saleeby (2003, Fig. 4C)

outboard of the Pelona and Orocopia Schists older schists to be underplated at increasingly Figure l a t e 10, where panel A represents the paleo- northwestMojave to southeast NV (Fig. 3). Also notable, rons, PelonaMojave SchistsNV in the central Transverse Ranges, and Orocopia Schists in southeast- Nacimiento bl inia attributed this event to extension within the up-

(southeast option in Figs. 9A and 9B). In view of greater distances inboard (Fig.SAf block 8). In this case, geography prior to ernfaultSA California–southwestern blockslip. At this time, the Arizona. The exception is the Rand Schist of the San Emigdio Peninsular the duration of the cycling interval increases the ca. 95–90 Ma age of the Catalina Schist, this age contours would parallel the strike of the northwestern schistMountains, localities would which have we been group with the Catalina Schist. Within the Pelona and Orocopia to the southeast, although this may be at least n blk f per plate of the Vincent–Chocolate Mountains reconstruction is consistent with the isochrons of margin (Fig. 9A), and the young ageCA of the situated relatively closeSchist togroups, the trench,CA individual requir- areas are ordered from northwest on the left to southeast on the partly an artifact in theAZ data. For example, we AZ thrust system. Alternatively, or in addition, we Figure 9A, but not those of FigureAseismic 9B. In other southeastern schists wouldobtained be no a consequence hornblende 40 Ar/ing39Ar only ages modest from flright. at subduction Northwest-southeast to emplace position for the Rand Schists is not clear owing to potential rotation of the Sierran tail; theseobserve bodies that were forearc instead subsidence ordered from is left a commonto right by deposi- words, the southeast option for the Catalina of their position relativelythe easternmostfar inboard. schistWe en- bodiestrench (Fig. materials 3) andk beneath the former arc (Figs. tional and metamorphic age.feature Diagonally in regions ruled boxof inferred indicates subduction time of most erosion likely movement Schist requires at least some componentridge of de- vision this processCA involvingthus may transfer, not have due captured to 8B the and oldest 10A). cool- At the same time, the future sites creasing age from outboard to inboard within the buoyancy, ofRanges someMEX fractioning ages of in trench this region. materi- In addition,of the Pelona igneous and Orocopiaon the Nacimiento Schists would fault. have Yellowas an band isostatic indicates response the cycling to removal interval, which,of material as discussed in Nf 40 39 schist terrane. Alternatively, for the case of 500– als from the subductingbodies (Farallon) younger to than overrid- ca. 70 Maresided are not far com-inboard,the in text,a region is not of wellarc magma- constrainedfrom inthe the base southeast. of the Also overriding note that plate Ar/ (CliftAr and and zircon ages overlapNf for the Catalina Schist (the two types of ages have been separated slightly along 600 km of sinistral strike slip on the Nacimiento ing (North American) monplate inabove the inferredthe hinge provenance at tism areas (Fig. (Barth10A; Mattinson, 1990; Barth et al.,Vannucchi, 2004). By analogy, we propose that et al., 2008a), making it diffi cult to constrain the x-axis to clarify this relation). The low grade of metamorphism for this unit (Grove fault, the Catalina Schist would restore to the which the subducting plate descended toward 1995, 1997, 2001a, 2008a; Kidder et al., 2003; 40 39 the maximum depositional age of sediments et al., 2008) suggests that themarine Ar/ transgressionAr ages may be of infl the uenced Salinian by excess block radiogenic may argon northwest end of the Pelona-Orocopia-RandPOR-Catalina the mantle. Schists With time, such subcreted materi- Barbeau et al., 2005). With commencement younger than latest Cretaceous. For example, CSor lack of complete recrystallizationhave been of detrital related, mica at during least inmetamorphism. part, to tectonic Abbreviations schist belt. This location is consistent with either als could become quite thick, leading to tectonic of Nacimiento slip,for the schist latter localities regions are would same as in Figure 1. Other abbreviations are: bio—biotite; Camp— decreasing age outboard-inboard (Fig. 9A)Prot. or basementand/or erosional w/55–85 denudation Main plutonsdelineating of the the overlying cycling interval,have webeen ignored drawn toward the continental mar-erosion of North American mantle lithosphere three early CenozoicAseismic ages from the Orocopia Campanian;Peninsular Cen—Cenomanian; Eoc—Eocene; hbd—hornblende; Maast—Maastrichtian; northwest-southeast (Fig. 9B). These uncertain-Late Cret.plate batholith (cf. Platt, (85–100 1986; Yin, Ma) 2002). In this fashion, gin (Fig. 9B) in the same way that slip on aand lowermost crust during underplating of the Schist due to their status as outliers (Fig. 3). Mio—Miocene; mus—muscovite; Olig—Oligocene; Pal—Paleocene; Sant—Santonian; ties require that we consider several mechanisms underplated materials would be driven upward, low-angle normal fault will translate rocks of Early Cret. batholith (100–135However, Ma) it shouldridge be kept in mind that any or Tur—Turonian. Pelona-Orocopia-Rand-Catalina schists. to explain the observed variation of age within consistent with thermochronologic evidence the lower plate closer to the surface. As a con- all of these ages could be signifi cant. Despite This interpretation implies that the age of the the schist terrane. Foothillsfor belt exhumation of the Pelona-Orocopia-Rand sequence, progressively more southeastward the uncertainties, the schist protoliths clearly CA transgression in any given area should correlate schists concurrent with subduction (Jacobson (inboard) regions would have Rangebeen transferred Forearc basin range from Turonian to at leastN as young as Variation in DetritalMEX Zircon Populationswith the timeas a of taceous local zircons, schist with underplating. lesser, but still At signifi cant, Progressive Subduction Erosion et al., 2002, 2007). Thismiddle would Paleocene, allow permanent representing into a time the zonespan of of underplatingFunction of Depositional near the conti- Age and Location proportions of Proterozoic and Jurassic ages. Accretionary complex 100 km least to a fi rst approximation, this appears to be Grove et al. (2003) proposed that buzz-saw incorporation of some30 trench m.y. or materials greater. within nental edge. In this scenario,Pie diagrams thes ofschists detrital need zircon ages for indi- JacobsonIn contrast, etthe centralal., GSA to southeastern Bull, 2010 schists, removal of the base of North America could the overriding plate. As Forthe flpurposes at slab continued of comparison not to defi the ne forearc more thanvidual a relatively ranges and/or narrow adjacent bandthe ranges case arefor plot-the Salinianwhich have block depositional and central ages Trans- of Maastrichtian allow younger schists to “leap-frog”Figure 10. beneath Origin to of erode Nacimiento inboard, the faultunits, locus we by of dividedsinistral underplating the schists strike (becominginto slip four (modifi groups younger edted to from in the Figure southeast) Dickinson 4 using beneath the palinspasticverse Ranges. base of The and correlation younger, are does characterized not hold by true progressively [1983] using more recent estimates basedfor age on depositionalof faulting). age Faulting and geographic is assumed Figure to2. Probabilityhave been density in functions the San for Emigdio the decreasing Mountains, abundances where of theEarly schist to middle Cre- location (Figs. 3 and 4). This contrasts with four regional groupings are illustrated in Figures taceous detrital zircons and increasing num- driven by subduction of an aseismic ridge. Ridge scale is based on analogy to the Yakutat is early Late Cretaceous in age, but the base of Jacobson imagine a long ridge500 hitting obliquely. Geological Society of Americaa threefold Bulletin, division March/April utilized 2011 by Grove et al. 5B–5E. As pointed out by Grove et al. (2003), bers of Proterozoic and latest Cretaceous–early block (Fuis et al., 2008) and ridges (2003).on the OurCocos fourth and category Nazca includesplates (Gutscher the detrital zirconet al., patterns 2000). in thethe Pelona-Orocopia- marine sectionCenozoic in the upper ages. As plate concluded is Eocene by Grove et al. Present-day outcrops of Pelona-Orocopia-Rand-CatalinaCatalina Schist, which was not schists considered are by shown Rand in schists black. vary Only systematically (Nilsen, from 1987b). north- However,(2003), the patterns because imply this a shift region from outboard the northwesternmost schist bodiesGrove would et al. have (2003), been and thein Pelona-Orocopia-place at the timewest represented to southeast, and by thus is with at the depositional northwesternmost to inboard end sources of withthe schist time (e.g., belt, Fig. 2 and Rand schist of the San Emigdio Mountains, age of the protolith. The northwestern schists, following discussion). panel A. Outlined area labeled “Nacimientowhich Grove Blk” et al. is (2003) based grouped on present-day with the which outcrop have protolith of Fran- ages of itTuronian is conceivable to Cam- thatCompared subsidence to thewas Pelona-Orocopia-Rand limited by ciscan Complex between the NacimientoRand schists. and San Gregorio–Hosgri faults panian,southeast are dominated of Point by Earlythe to fl middleexural Cre- rigidityschists, of a thefull zirconthickness age distributionof litho- within Sur (Fig. 1). Panel A indicates inferred location of this body prior to sinistral slip on the sphere beneath the adjacent Sierra Nevada. Nacimiento fault. Panel B shows its location outboard of the Salinian block immediately following Nacimiento fault slip in the early Cenozoic, which closely matches Geological the Society present-day of America Bulletin,CONCLUSIONS March/April 2011 491 spatial relationship (Fig. 1). Note that the reconstruction of panel B is a simplifi ed equivalent of that in Figure 2. Heavy arrow shows Farallon–North America relative plate motion. Similar detrital zircon patterns exhibited by CS—Catalina Schist; Nf—future (panel A) and actual (panel B) traces of Nacimiento fault; the Pelona-Orocopia-Rand-Catalina schists and Gf—future trace of Garlock fault; SAf—future trace of San Andreas fault; other abbrevia- coeval sedimentary units of the forearc basin tions are as in Figure 1. provide evidence that these two sequences comprise parts of a single depositional system on the outboard side of the Sierran–Peninsular Ranges the outer edge of the continent, in contrast to posite the Diablo Range to the current position arc. The >30 m.y. time frame for emplacement the sheet-like geometry typically envisioned adjacent to the Salinian block (Fig. 10; see pre- of the schists and the short cycling interval are (e.g., Fig. 9). This could explain the relatively vious discussion). most consistent with a subduction origin. The linear nature of the schist exposures, which sandstone protoliths of the schists are best in- seems fortuitous if the distribution at depth is Marine Sequences of the Salinian Block terpreted as trench sediments complementary to more laterally persistent. In the sinistral slip the forearc basin sequence. model, extraregional detritus is supplied to the The Maastrichtian to Paleogene marine trans- Emplacement of the schists directly beneath coast via a gradually enlarging “window” be- gression of the Salinian block and adjacent areas the Cordilleran magmatic arc requires the re- tween the southern Sierra Nevada and northern has long been viewed as an important tectonic moval of North American mantle lithosphere Peninsular Ranges (Fig. 10B). Sinistral slip on signal. Clast distributions in conglomerates indi- and lowermost crust. The initial stages of this the Nacimiento fault would also translate the cate a number of localized basins, at least during underplating may have occurred at the end of a Nacimiento block from an initial location op- early phases of deposition (Grove, 1993). Some period of modest subduction erosion associated

Geological Society of America Bulletin, March/April 2011 501 Late Cretaceous–early Cenozoic tectonic evolution of southern California

RESULTS 10 Mio 20 Pelona-Orocopia-Rand and Metamorphic bio Catalina Schists 30 Olig mus hbd Aside from the fact that we did not previ- Detrital 40 ously consider the Catalina Schist, our results zircon Eoc are little changed from those obtained by Grove 50 erv al et al. (2003). We thus emphasize only those int points most relevant for comparison with the 60 Age (Ma) Pal forearc data. Cy cling Maast Slip on Nac flt 70 Protolith Age Camp One of the most critical parameters to de- 80 U-Pb plutonictermine for any given ages, schist body isuncertainty<5 the depo- Sant Ma sitional age of the sandstone protolith, which Con 65 90 must lie within a time window bounded by the Tur Cen youngest reliable detrital zircon age and oldest 100 reliable metamorphic age (Fig. 3). Grove et al. Age/ SC SE PRRA SS MP SP BR EF OR CH TR CD NR POR Schist(2003) referred to this time span as the “cycling Epoch Cat-San Emig Rand Pelona Orocopia 70 interval,” because it encompasses erosion in 40 39 emplacementthe source area, transport of that material to the Figure 3. ComparisonNorth of Ar/ Ar cooling ages and zircon U-Pb ages ≤100South Ma for the site of deposition, underthrusting beneath the Pelona-Orocopia-Rand-Catalina schists plotted by area (see Figs. 1, 2, and 4 for locations). estimatesarc, accretion to the overriding plate, and initial Zircon ages are from Barth et al. (2003), Grove et al. (2003, 2008), and this study. Argon ages are from Jacobson (1990), Jacobson et al. (2002, 2007), Barth et al. (2003), and Grove 75 (mostly Chapman, 2017)stages of exhumation. As observed by Grove et al. (2003), the cycling interval of the Pelona- et al. (2003, 2008). The Pelona-Orocopia-Rand schists are organized by the conventional Orocopia-Rand schists becomes younger from northwest-southeast grouping, i.e., Rand Schists in the northwest Mojave Desert and envi- northwest to southeast (Fig. 3). Also notable, rons, Pelona Schists in the central Transverse Ranges, and Orocopia Schists in southeast- the duration of the cycling interval increases ern California–southwestern Arizona. The exception is the Rand Schist of the San Emigdio 80 to the southeast, although this may be at least Mountains, which we group with the Catalina Schist. Within the Pelona and Orocopia partly an artifact in the data. For example, we Schist groups, individual areas are ordered from northwest on the left to southeast on the obtained no hornblende 40Ar/39Ar ages from right. Northwest-southeast position for the Rand Schists is not clear owing to potential ro- Age (Ma) Age the easternmost schist bodies (Fig. 3) and tation of the Sierran tail; these bodies were instead ordered from left to right by depositional and metamorphic age. Diagonally ruled box indicates time of most likely movement 85 thus may not have captured the oldest cooling ages in this region. In addition, igneous on the Nacimiento fault. Yellow band indicates the cycling interval, which, as discussed in 40 39 bodies younger than ca. 70 Ma are not com- the text, is not well constrained in the southeast. Also note that Ar/ Ar and zircon ages mon in the inferred provenance areas (Barth overlap for the Catalina Schist (the two types of ages have been separated slightly along et al., 2008a), making it diffi cult to constrain the x-axis to clarify this relation). The low grade of metamorphism for this unit (Grove 40 39 90 the maximum depositional age of sediments et al., 2008) suggests that the Ar/ Ar ages may be infl uenced by excess radiogenic argon younger than latest Cretaceous. For example, or lack of complete recrystallization of detrital mica during metamorphism. Abbreviations in delineating the cycling interval, we ignored for schist localities are same as in Figure 1. Other abbreviations are: bio—biotite; Camp— three early Cenozoic ages from the Orocopia Campanian; Cen—Cenomanian; Eoc—Eocene; hbd—hornblende; Maast—Maastrichtian; Schist due to their status as outliers (Fig. 3). Mio—Miocene; mus—muscovite; Olig—Oligocene; Pal—Paleocene; Sant—Santonian; 95 Tur—Turonian. -800 -600 -400However, it should-200 be kept in mind that any0 or 200 400 600 all of these ages could be signifi cant. Despite S35Ethe uncertainties, Palinspastic the schist protoliths (trench clearly parallel, km) range from Turonian to at least as young as Variation in Detrital Zircon Populations as a taceous zircons, with lesser, but still signifi cant, NAVDAT + numerous other papers middle Paleocene, representing a time span of Function of Depositional Age and Location proportions of Proterozoic and Jurassic ages. 30 m.y. or greater. Pie diagrams of detrital zircon ages for indi- JacobsonIn contrast, etthe centralal., GSA to southeastern Bull, 2010 schists, For purposes of comparison to the forearc vidual ranges and/or adjacent ranges are plot- which have depositional ages of Maastrichtian units, we divided the schists into four groups ted in Figure 4 using the palinspastic base of and younger, are characterized by progressively based on depositional age and geographic Figure 2. Probability density functions for the decreasing abundances of Early to middle Cre- location (Figs. 3 and 4). This contrasts with four regional groupings are illustrated in Figures taceous detrital zircons and increasing num- From Jones, GSA talk, 2019. Heavy red dots are Plomosas (73a threefold Ma, division Seymour utilized by Grove etet al. al.,5B–5E. As2018) pointed out by and Grove et al.Cemetery (2003), bers of Proterozoic Ridge and latest (65Cretaceous–early Ma, Jacobson et al., 2017). Note P-O (2003). Our fourth category includes the detrital zircon patterns in the Pelona-Orocopia- Cenozoic ages. As concluded by Grove et al. schists come in right as magmatism dies while Rand schistsCatalina seem Schist, which to was potentially not considered by Rand overlap schists vary systematically with from arc. north- (2003), the patterns imply a shift from outboard Grove et al. (2003), and the Pelona-Orocopia- west to southeast, and thus with depositional to inboard sources with time (e.g., Fig. 2 and Rand schist of the San Emigdio Mountains, age of the protolith. The northwestern schists, following discussion). which Grove et al. (2003) grouped with the which have protolith ages of Turonian to Cam- Compared to the Pelona-Orocopia-Rand Rand schists. panian, are dominated by Early to middle Cre- schists, the zircon age distribution within

Geological Society of America Bulletin, March/April 2011 491 How to Emplace a Schist?