Tracking paleodeformation fields in the Mesozoic central Sierra Nevada arc: Implications for intra-arc cyclic deformation and arc tempos

Episodic magmatism in continental arcs was first recognized by the pioneering work done by Armstrong and Ward (1993). They observed that magmatism in North American Cordilleran arcs waned and waxed from Late Triassic to early Eocene time. Recent studies of continental arcs in both the North and South American Cordilleras have confirmed and quantified such episodic magmatism: Periodically reoccurring magmatic flare-up events have occurred, separated by magmatic lulls in the Mesozoic and Cenozoic (Ducea, 2001; Ducea and Barton, 2007; DeCelles et al., 2009, 2015; Paterson et al., 2011; Paterson, 2012). The causes of such cyclicity remain unsolved. Temporal correlation of cyclic arc magmatism and subduction kinematics is tenuous (Ducea, 2001). More attention has been paid to a potentially self-regulating continental arc system (DeCelles et al., 2009, 2015). Unifying models incorporating feedbacks among propagation of retroarc and forearc wedges, foundering of eclogitic arc roots, and underthrusting of retroarc craton are proposed to explain observed cyclicity in magmatism and isotopic variations (DeCelles et al., 2009, 2015). Temporal patterns of these linked cyclic processes, some observed and others hypothetical, are referred to as “arc tempos” (Cao and Paterson, 2013; Paterson et al., 2014; Paterson and Ducea, 2015). However, intra-arc processes, such as the ascent and emplacement of plutons, intra-arc deformation, and crustal thickening, are largely neglected and not integrated into present models of arc tempos. Thus, intra-arc deformation and its relation to arc magmatism and arc tempos deserve better study.

One way to understand relationships between intra-arc deformation and other arc processes is to temporally compare magmatism to intra-arc paleodeformation fields. The Sierra Nevada arc, here used as a case study, is a segment of the North American Cordilleran continental arcs, now exhumed and exposing both the volcanic carapace and the deeper plutons. The latter is referred to as the Sierra Nevada Batholith, a collective term for plutons of various Mesozoic ages (Figs. 1A and 1B). Cyclicity of magmatism in the Mesozoic Sierra Nevada arc is well recognized, consisting of three magmatic flare-ups separated by two magmatic lulls during the Mesozoic (Ducea, 2001; DeCelles et al., 2009; Paterson et al., 2011, 2014). Host rocks of the Sierra Nevada Batholith in the central Sierra Nevada form an ∼200-km-long succession of pendants herein called the Eastern Sierra pendants, a collective term for Ritter Range, Northern Ritter Range, and Saddlebag Lake pendants (Fig. 1C). Though a number of studies have addressed deformation in the Eastern Sierra pendants (e.g., Brook, 1977; Tobisch et al., 1977; Tobisch and Fiske, 1982; Greene and Schweickert, 1995; Schweickert and Lahren, 2006; Nadin and Saleeby, 2008), potential episodic phases of deformation were difficult to determine due to poor temporal constraints. Other studies addressing shear zones mainly focused on the ones developed in the Late Cretaceous (Tikoff and Saint-Blanquat, 1997; Tikoff et al., 2005; Tikoff and Greene, 1997; Greene and Schweickert, 1995; Tobisch et al., 1995; Nadin and Saleeby, 2008). In order to establish an intra-arc deformational history, two challenges have to be addressed: (1) Restore pre–Late Cretaceous intra-arc paleodeformation fields with better timing constraints, and (2) update and a reexamine patterns of intra-arc deformation with respect to arc tempos.

In this paper, we present the geology of the Virginia Canyon area, central Sierra Nevada, where plutons of Triassic, Jurassic, and Cretaceous ages intruded host rocks of Paleozoic and Mesozoic ages. Structures in both plutons and host rocks are constrained using a new geologic map, U/Pb zircon ages, and ⁴⁰Ar/³⁹Ar mineral cooling ages. By combining information from both plutons and host rocks, we examine paleodeformation fields in the central Sierra Nevada. Finally, we discuss how paleodeformation fields can be related to intra-arc deformation and arc tempos. We conclude that, to some extent, intra-arc deformation is cyclic and displays temporal patterns comparable to cyclic magmatism: Feedbacks likely exist between arc processes.

The Eastern Sierra pendants, consisting of Paleozoic and Mesozoic strata, are regionally deformed and intruded by Mesozoic plutons (Fig. 1C). Paleozoic components of the Eastern Sierra pendants represent pre-Mesozoic arc basement, mainly consisting of autochthonous or allochthonous North American passive-margin and deep-marine sediments with minor mafic, submarine igneous rocks (Brook, 1974, 1977; Keith and Seitz, 1981; Bateman et al., 1983; Schweickert and Lahren, 2006). West of these Paleozoic rocks, built on the pre-arc basement and separated by a regional unconformity, the Mesozoic component of the Eastern Sierra pendants represents an upper-crustal section of the Sierra Nevada arc. The Eastern Sierra pendants have experienced multiple phases of Paleozoic to Late Cretaceous deformation resulting in steeply dipping bedding, foliation, subvertical lineation, and multiple shear zones and faults. The Eastern Sierra pendants have also been regionally metamorphosed under greenschist to local amphibolite facies in pluton aureoles. In the following descriptions, the prefix meta- is omitted when protoliths are still recognizable. Jurassic and Triassic plutons intruding the Eastern Sierra pendants are spatially scattered and of small size, while some of the Late Cretaceous plutons formed large intrusive complexes.

The study area, located on the eastern margin of the Tuolumne intrusive complex (Fig. 2), was previously mapped by Chesterman (1975) and Schweickert and Lahren (1993, 2006). We have remapped it at 1:10,000 scale. Figure 3 is a simplified stratigraphic section for the northern part of the study area. Table 1 summarizes new U/Pb zircon ages of host rocks and plutons. Analytical methods, data tables, and plots associated with these ages are presented in the GSA Data Repository.¹

Bedding in the study area is steeply NE dipping. The base of the Mesozoic strata, exposed just east of the study area, is defined by an unconformity that separates Mesozoic strata from the Paleozoic rocks underneath (Brook, 1974, 1977). The lowest stratigraphic unit, a 100–300-m-thick unit consisting of conglomerate interlayered with sandstone (Tcl; Fig. 4A), marks the base of the Triassic Koip sequence, which includes this conglomerate and the volcanic packages described next. Clasts in the conglomerates are poorly sorted, rounded to subrounded, with a wide variety of clast types, including quartzite, chert, argillite, silicic volcanics, and granitoids. The weighted mean of youngest detrital ages of this unit is 221.4 ± 2.7 Ma (sample K-57). Earlier workers have interpreted this conglomerate as a regional marker unit, the Cooney Lake conglomerate, which extends from the northern Ritter Range pendant to the Twin Lakes area (Schweickert and Lahren, 2006). However, our recent mapping indicates that several similar conglomerates occur in different stratigraphic positions throughout the Eastern Sierra pendants, sometimes found near the base of the Koip sequence, but elsewhere mapped with volcanic rocks both above and below.

On top of the Cooney Lake conglomerate, there are volcanic rocks belonging to the Koip sequence. The first volcanic unit is an ∼100-m-thick, white-gray, rhyolitic ash-flow tuff with broken and embayed quartz phenocrysts and pumice fragments called the Rhyolitic Tuff of Saddlebag Lake (Tsl; Schweickert and Lahren, 1993). It has a U/Pb zircon age of 222 ± 5 Ma and is interpreted to relate to the ca. 224 Ma Tioga Pass caldera (Schweickert and Lahren, 2006). On top of the tuff unit, there is a thin layer of phyllitic argillite (Tph). Immediately overlying the phyllite, there is an ∼1-km-thick massive andesite to dacite unit (Tmda; Fig. 4B). A whole-rock Rb-Sr isochron age from an andesite in the same stratigraphic position in the Saddlebag Lake pendant was dated to 224 ± 14 Ma (Keith and Seitz, 1981). West of the massive bedded andesite-dacite unit, there is a package of rhyo-dacitic ash-flow tuff (Trd), a dacite-andesite unit (Tda), and a package of volcaniclastic rocks (Tvc). An open SE-plunging syncline-anticline pair exists in the volcaniclastic unit, which gradationally merges into the dacite-andesite unit (Tda) along its strike. Schweickert and Lahren (2006) and Schweickert et al. (1994) reported 211 Ma and 201 Ma (U/Pb zircon) ages from the dacite-andesite unit (Tda). We obtained an age of 220.3 ± 2.1 Ma from the western part of this unit (sample S-20).

In the north portion of the study area, volcanic Koip sequence rocks are separated from a package of marine sedimentary rocks by the newly mapped Virginia Canyon shear zone. Based on our mapping in the Twin Lakes, Saddlebag Lake, and Waugh Lake areas, we interpret the contact between the two sequences as an angular unconformity later faulted during the Jurassic and Cretaceous and not a conformable contact as interpreted by Schweickert and Lahren (2006).

The marine sedimentary unit (Jsc, Jms) to the west of the Virginia Canyon shear zone consists of cross-bedded quartzite, phyllite, calc-silicate with local limestone, conglomerate, volcaniclastic rocks, and schists (Figs. 4C and 4D). The weighted means of youngest detrital ages of two siltstones from this unit are 186.1 ± 2.0 Ma and 187.3 ± 2.9 Ma (samples S-33 and S-40). We attribute this marine sedimentary unit to the Jurassic Sawmill Canyon sequence (Schweickert and Lahren, 2006), which was previously called the Horse Canyon sequence by Keith and Seitz (1981).

Just west of and juxtaposed with the Sawmill Canyon sequence, there is a tectonic sliver of well-bedded volcanic, volcaniclastic, and sedimentary rocks (Pzg) exposed in Spiller Canyon (Fig. 4E). The volcanics are andesitic to locally rhyolitic in composition, and commonly clastic with some clasts ranging in size up to 50 cm. Sediments range from siltstones to conglomerates and are interbedded with cross-bedded sandstones (Fig. 4F). Schweickert and Lahren (2006) also found slices of serpentinized ultramafic rocks within this unit in upper Spiller Canyon. One sandstone from this unit shows two youngest detrital zircon ages of ca. 350 Ma and 380 Ma, with older Lower Paleozoic and Precambrian peaks. Detrital zircon ages and the lithology of this unit indicate that it is part of the Golconda allochthon, thrust over the younger Sawmill sequence. A package of rhyo-dacite dated at 227.4 ± 2.2 Ma occurs west of the Golconda allochthon (sample H-07). We interpret this package to be part of the Koip sequence, which was tectonically juxtaposed with other units by Mesozoic faults.

Sandwiched between the Cathedral Peak pluton and the Soldier Lake pluton, there is a belt of volcanogenic sedimentary rocks (Kvs) with a weighted mean of youngest detrital ages of 117.4 ± 2.1 Ma (sample K-34). Belts of similar Cretaceous rocks are also found in the Sawmill Canyon area and in the western Ritter Range pendant south of the study area. We interpret the contact between the Cretaceous rocks and the Koip sequence as an angular unconformity, probably faulted in the Late Cretaceous (Tobisch et al., 2000; Cao and Paterson, 2013).

Host rocks in the study area are intruded by four Mesozoic plutons: the Late Triassic Spiller Canyon pluton, Middle Jurassic Green Lake pluton, and two Late Cretaceous plutons, the Soldier Lake pluton and Cathedral Peak pluton. Contacts between both of these plutons and their host rocks are usually sharp (Fig. 5A). The Spiller Canyon pluton is a fine-grained quartz monzodiorite. The Green Lake pluton and Soldier Lake pluton are medium-grained granodiorite with mafic enclaves (Figs. 5B and 5C). The Cathedral Peak pluton is a granodiorite with large K-feldspar phenocrysts up to 12 cm long (Fig. 5D), locally grading into to a fine-grained granite (Bateman and Chappell, 1979; Memeti et al., 2010b).

The Spiller Canyon pluton is dated at 232.9 ± 1.4 Ma (sample KS-08–019). The Green Lake pluton, Soldier Lake pluton, and Cathedral Peak pluton are dated to 165.2 ± 0.3 Ma, 97.4 ± 0.4 Ma, and ca. 86–87 Ma, respectively (chemical abrasion–isotope dilution–thermal ionization mass spectrometry [CA-ID-TIMS] U/Pb zircon ages; Nomade et al., 2003; Mundil et al., 2004; Matzel et al., 2006; Memeti et al., 2010b).

Series of samples were also collected for ⁴⁰Ar/³⁹Ar hornblende and biotite thermochronology along a NE-SW transect from the eastern margin of the Cathedral Peak pluton to the western margin of the Green Lake pluton (Mundil et al., 2004). Hornblende and two different grain-size fractions of biotite with different closure temperatures were used to track thermal history in study area (Fig. 6). The closure temperatures estimated for hornblende, coarse-grained biotite (800–900 µm), and fine-grained biotite (150–180 µm) are 530 °C, 380 °C, and 330 °C, respectively (Mundil et al., 2004).

Hornblende from the eastern margin of the Cathedral Peak pluton is dated at 85.5 ± 0.9 Ma, and biotite yields ages of 82.0 ± 0.6 Ma (coarse grained) and 80.4 ± 0.7 Ma (fine grained). Hornblende at the eastern margin of the Soldier Lake pluton is dated at 85.2 ± 0.7 Ma, and biotite ages are 83.1 ± 0.7 Ma (coarse grained) and 81.1 ± 0.9 Ma (fine grained). At the eastern margin of the Green Lake pluton, hornblende yields age of 166.2 ± 3.2 Ma, with biotite ages of 85.6 ± 0.7 Ma (coarse grained) and 82.4 ± 0.4 Ma (fine grained; Mundil et al., 2004; Memeti et al., 2010b).

Thermochronology results suggest that the emplacement of the Cathedral Peak pluton had thermally affected the Soldier Lake pluton and Green Lake pluton at ca. 86 Ma. Cooling ages of hornblende and biotite in the Soldier Lake pluton were fully reset to a post–86 Ma age. In contrast, in the Green Lake pluton, only the biotite was reset to the post–86 Ma age, while the hornblende was not affected. These data also reveal a spatial thermal gradient across the area at ca. 86 Ma, implying that the host-rock temperature was above 530 °C up to ∼2 km away from the Cathedral Peak pluton margin and above 380 °C but below 530 °C up to ∼6 km away from the Cathedral Peak pluton margin.

Temperature-corrected, aluminum-in-hornblende barometry coupled with the plagioclase-hornblende thermometry (Johnson and Rutherford, 1989; Schmidt, 1992; Hammarstrom and Zen, 1986; Anderson and Smith, 1995; Holland and Blundy, 1994; Anderson, et al., 2008) were conducted to determine emplacement conditions of the Green Lake pluton, Soldier Lake pluton, and Cathedral Peak pluton. The Green Lake pluton was emplaced at 3.2 ± 0.3 kbar and 731 ± 15 °C. The Soldier Lake pluton was emplaced at 2.3 ± 0.2 kbar and 696 ± 17 °C. The Cathedral Peak pluton was emplaced at 2.4 ± 0.2 kbar and 696 ± 16 °C (Anderson et al., 2010; J.L. Anderson, 2014, personal commun.).

Next, we describe structures in host rocks from oldest to youngest and then describe the time-transgressive foliations and associated lineations in the host rocks.

Triassic thrusts are recognized in the Twin Lakes area immediately north of the study area and in the Tioga Crest area, south of the study area. They are constrained to 218–219 Ma and 218–222 Ma, respectively, by crosscutting relationship with Triassic plutons (Schweickert and Lahren, 2006; Barth et al., 2011). Thrusts are NW striking, have NE or SW vergence, and are now subvertical due to subsequent deformation (Tobisch et al., 2000). Although we did not find direct evidence of these Triassic thrusts in the mapped area, indirect evidence supports their existence. Actual thrust surfaces are cryptic due to the overprinting by the younger phases of deformation. Triassic thrusts are thought to exist in the Koip sequence and are contemporaneous with formation of the Triassic volcanic rocks (Schweickert and Lahren, 2006). They have similar strike and vergence and were probably active around 219 Ma, comparable to other Triassic thrusts in nearby areas. Detailed field mapping and stratigraphic studies are needed to better understand these structures.

Glenberry Lake thrust. The Glenberry Lake thrust is a NW-striking, steeply dipping, NE-vergent thrust fault that placed the western part of the Koip sequence over the eastern Koip sequence. Schweickert and Lahren (2006) mapped this thrust from the southern Twin Lakes area south to Lundy Canyon. The often cryptic fault trace is shown on our map as an inferred fault. The Glenberry Lake thrust cuts the Triassic Koip sequence and thus is at least post–220 Ma. Schweickert and Lahren (2006) interpreted the Glenberry Lake thrust to be Middle Jurassic in age.

East Spiller Canyon thrust. The newly mapped East Spiller Canyon thrust placed the Jurassic Sawmill Canyon sedimentary rocks over older Golconda allochthon in the NW of the study area. The East Spiller Canyon thrust is a NW-striking, SW-vergent thrust, and it cuts the eastern margin of the 232.9 Ma Spiller Canyon pluton, resulting in NE-side-up subsolidus shearing structures in the pluton. The East Spiller Canyon thrust is in turn crosscut by the Soldier Lake pluton. We conclude the East Spiller Canyon thrust is younger than 186.1 Ma and older than 97.4 Ma. It is probably contemporaneous with other kinematically similar, Middle Jurassic thrusts.

West Spiller Canyon thrust. We interpret the contact between the Spiller Canyon sliver and the Koip sequence in the NW corner of the map as a thrust (West Spiller Canyon thrust). The West Spiller Canyon thrust is cut by the 97.4 Ma Soldier Lake pluton, which constrains its timing to between 227.4 Ma and 97.4 Ma. Kinematics on the West Spiller Canyon thrust suggest SW-vergent motion, resulting in thrusting of the older Golconda allochthon over younger Koip volcanics. The West Spiller Canyon thrust and East Spiller Canyon thrust thus form the bounding faults of the Golconda allochthon. We propose that the West Spiller Canyon thrust is Middle Jurassic, contemporaneous with the East Spiller Canyon thrust.

Excelsior Mountain thrust.Schweickert and Lahren (2006) mapped one fault (Excelsior Mountain thrust) in the same location we show in Figure 2. It was previously interpreted to separate the Mesozoic volcanic-sedimentary units (on the east) from the Golconda allochthon (on the west). We retain this name, but according to our mapping and geochronologic results, the Excelsior Mountain thrust separates volcanics of the Koip sequence (on the east) from the sedimentary rocks of the Sawmill Canyon sequence (on the west). In the middle of the study area, the Excelsior Mountain thrust is located within the Koip sequence. In its NW extension, the Excelsior Mountain thrust bifurcates into two branches that bound an overturned anticline (Excelsior Mountain thrust–east and Excelsior Mountain thrust–west; Fig. 4G). Based on kinematics in the field and crosscutting relationship, the Excelsior Mountain thrust is a post–186 Ma, NE-vergent thrust that faulted the unconformity between the Koip and Sawmill Canyon sequence. It was probably active during the Middle Jurassic with other thrusts in the study area. The Excelsior Mountain thrust was also reactivated as part of the Virginia Canyon shear zone in the Late Cretaceous, discussed in a later section.

Folds related to Jurassic thrusts. Series of kilometer-scale, fairly open to locally tight folds were mapped in host rocks between the different thrust faults. Axial planes of these folds range from subvertical to moderately dipping with hinges plunging SE or NW (Fig. 7A). Syncline-anticline pairs are developed in the Koip sequence and in the hanging wall of the Glenberry Lake thrust and Excelsior Mountain thrust. A syncline in the northern part of the area has two limbs consisting of local marine sedimentary rocks, which belong to the Jurassic Sawmill Canyon sequence. An overturned anticline was also mapped between the two fault branches of the Excelsior Mountain thrust (Fig. 4G). Timing of these folds is post–186 Ma based on the youngest strata involved. Spatial and kinematic relationships between these folds and the nearby thrusts suggest that they are related to the same phase of Middle Jurassic deformation.

Virginia Canyon shear zone. Our new 1:10,000 scale mapping in the study area and Saddlebag Lake areas led us to the recognition of a major oblique, dextral, transpressional, ductile shear zone, the Virginia Canyon shear zone, that evolved into a brittle fault system with one major strand centered within the ductile shear zone. The ductile phase of the shear zone is wider in the north of the study area and deforms much of the Koip sequence and adjacent units to the west. The Virginia Canyon shear zone spatially matches the location of the Excelsior Mountain thrust, which suggests that the Virginia Canyon shear zone may have taken advantage of the preexisting fault zone during its formation.

The ductile phase of the Virginia Canyon shear zone is defined by a NNW-striking, subvertical foliation more strongly developed than outside the Virginia Canyon shear zone (Fig. 7B), SE-plunged stretching lineations, and the presence of shear sense indicators. In the Virginia Canyon shear zone, the simple shear component is less developed in competent volcanic rocks, evidenced by weak asymmetric volcanic clasts. The simple shear deformation is best developed in sedimentary rocks that show a series of outcrop-scale dextral asymmetric folds, S-C structures, and rotated clasts with asymmetric tails (Figs. 4C and 4D). The pure shear component of the ductile phase of the Virginia Canyon shear zone manifests in the NNW-striking foliation and tight asymmetrical, west-side-up folds with axial planes oriented parallel to the shear plane. Boundaries of the ductile Virginia Canyon shear zone are gradational. Scattered ductile shears with the same kinematics exist out to the western margin of the Green Lake pluton and into the eastern margin of the Cathedral Peak pluton. The ductile phase of the Virginia Canyon shear zone also deformed the eastern margin of the Cathedral Peak pluton under magmatic and high-temperature subsolidus conditions.

The mineral assemblage in the calc-silicate rocks in the Virginia Canyon shear zone contains quartz, garnet, calcite, epidote, hornblende, chlorite, and plagioclase. Quartz exists as oriented, recrystallized aggregates with irregular boundaries indicating subgrain rotation recrystallization in upper-greenschist conditions around 400–500 °C (Hirth and Tullis, 1992; Passchier and Trouw, 2005). The Virginia Canyon shear zone was at least active at ca. 85 Ma when regional temperatures reached 400–500 °C (Fig. 6). Shearing of the Cathedral Peak pluton by the Virginia Canyon shear zone in the south of the study area suggests the Virginia Canyon shear zone was coeval with or predates emplacement of the Cathedral Peak pluton at ca. 86 Ma. The Virginia Canyon shear zone bears geometric, kinematic, and temporal similarities with other shear zone segments of the Sierra Crest shear zone system (Fig. 1B; Greene and Schweickert, 1995; Tikoff and Saint-Blanquat, 1997). To the south, it is continuous with the Cascade Lake shear zone (Tikoff et al., 2005).

A brittle, dextral fault with the same orientation and kinematics was found in the center of the Virginia Canyon shear zone. This fault can be traced into the Saddlebag Lake and Sawmill Canyon areas for ∼10 km to the south of the study area (Hartman and Paterson, 2013; Whitesides et al., 2010) and northward into the Twin Lakes area. It offsets a Cathedral Peak pluton–originated dike in the Saddlebag Lake area, indicating that motion along the brittle fault is post–86 Ma. We suggest that this brittle fault motion reflects continued movement as the shear zone crossed the brittle-ductile transition and was at least active around 79.5 Ma (Fig. 6), when regional temperatures dropped below 300 °C and may have continued to develop to 70 Ma.

Cretaceous Koip sequence fault. We interpret the contact between Cretaceous volcanogenic sedimentary rocks and the Triassic Koip sequence as an angular unconformity based on observations that Cretaceous rocks of similar ages unconformably overlie older rocks in other parts of the Eastern Sierra pendants (Tobisch and Fiske, 1982; Longiaru, 1987; Hartman and Paterson, 2013). The unconformity was likely subsequently faulted by SW- or NE-vergent thrusting between 117.4 Ma and 97.4 Ma. We herein name this fault the Cretaceous Koip sequence fault.

Ductile deformation in the host rocks is represented by a well-developed regional slaty to schistose foliation (Figs. 8 and 9) and associated lineations. Depending on protolith, foliation is defined by alignment of deformed clasts and pebbles, or oriented, recrystallized quartz and biotite. Host-rock foliation statistically strikes NW and dips steeply to the NE, oriented 315/78 (strike/dip; Fig. 7C) and is subparallel to bedding (Fig. 7D). Host-rock lineation is defined by alignment of plagioclase and/or hornblende phenocrysts, stretched quartz aggregates, elongated lapilli and clasts in volcaniclastic rocks, and stretched pebbles in conglomerates. Lineation plunges moderately to steeply SE with an average orientation of 93/70 (trend/plunge; Fig. 7E). Tobisch and Fiske (1982) noted that the regional foliation in the Eastern Sierra pendants was caused by multiple phases of subparallel deformation overprinting each previous fabric in the Paleozoic and Mesozoic. Next, we briefly describe evidence that foliation and associated lineation in our study area are also time-transgressive.

Steeply dipping bedding and bedding-parallel, NW-striking, steeply dipping foliation in the Golconda allochthon are cut by the 232.9 Ma Spiller Canyon pluton. We interpret these structures to have formed initially during the Permian–Triassic Sonoma event when deep-marine sediments and associated volcanics were thrust over the western North American continental margin (e.g., Dickinson, 2004), and then steepened during later Mesozoic deformation.

The Triassic–Cretaceous units all contain fabrics cut by Jurassic to Cretaceous plutons, and thus must have formed in separate strain events during the Mesozoic. The 165.2 Ma Green Lake pluton cuts across fabrics in the Triassic Koip sequence, requiring that these fabrics formed between 221.4 Ma and 165.2 Ma. Fabrics in the Jurassic Sawmill Canyon sedimentary rocks are crosscut by the 97.4 Ma Soldier Lake pluton, suggesting that these fabrics formed between 186.1 Ma and 97.4 Ma.

Fabrics in the 117.4 Ma Cretaceous rocks, crosscut by the 97.4 Ma Soldier Lake pluton, must have formed between 117.4 Ma and 97.4 Ma. Cretaceous strain is also indicated by the change of microstructures in host rocks. Dacitic-rhyolitic tuffs (thin section sample K17; Figs. 2 and 10A) close to the Cathedral Peak pluton margin (<0.5 km) show high-temperature (600–700 °C), subsolidus microstructures indicated by “chessboard” pattern–type subgrains in quartz and static recrystallization of plagioclase (Passchier and Trouw, 2005). A dacitic lapilli (thin section sample L9; Figs. 2 and 10B) collected ∼1.7 km away from the Cathedral Peak pluton margin consists of a metamorphic mineral assemblage of quartz + muscovite + chlorite + clinozoisite + plagioclase + K-feldspar + opaque oxides ± hornblende, suggesting lower-amphibolite-facies conditions (500–600 °C) during fabric formation. Opaque minerals preserved in clinozoisite porphyroblasts define an internal foliation (S_i). The S_i folia are curved into the matrix and become the S_e, parallel to the NW-striking foliation seen in outcrops. Such a relationship is interpreted as syntectonic growth of clinozoisite (Passchier and Trouw, 2005) during formation of foliation. A rhyolitic tuff (thin section sample R10), ∼5 km away from the Cathedral Peak pluton margin, consists of a fine-grained biotite, muscovite, and chlorite matrix. Quartz phenocrysts show bulging recrystallization, suggesting deformation temperatures of ∼300–400 °C (Passchier and Trouw, 2005). The eastward trend of decreasing deformation temperature based on microstructures and mineral assemblages suggests that the host-rock foliation contains Late Cretaceous strain accumulated during the emplacement of the Cathedral Peak pluton at ca. 86 Ma. Similar Late Cretaceous strain accompanying pluton emplacement has also been observed in the Ritter Range pendant by Sharp et al. (2000).

Magmatic fabrics are foliations and associated mineral lineations formed under melt-present conditions above the magma solidus (Paterson et al., 1998; Vernon, 2000). They are defined by the alignment of primary magmatic crystals of euhedral to subhedral shapes such as plagioclase, K-feldspar, biotite, hornblende, and sphene (Paterson et al., 1998; Vernon, 2000). Although magmatic fabrics are not good recorders of finite strain, they preserve the youngest strain increment during the final stage of pluton crystallization (Paterson et al., 1998). Magmatic fabrics present: (1) local incremental strain related to magma chamber flow dynamics and/or (2) superimposed regional incremental strain (Paterson et al., 1998; Callahan and Markley, 2003; Žák et al., 2007, 2009). Magmatic fabrics related to local strain in a magma chamber usually vary their orientations at the meter scale and sometime have a pluton-margin–parallel pattern. Magmatic fabrics related to regional strain display a consistent orientation (Paterson et al., 1998; Žák et al., 2007). The latter type has been used to deduce regional strain fields (Benn et al., 2001; Callahan and Markley, 2003; Žák et al., 2009).

The U/Pb zircon ages obtained from a pluton can be used as an approximate age of magmatic fabric formation. The time span for full crystallization following magma emplacement is relatively short at the emplacement depths encountered in the central Sierra Nevada. For a conductively cooling magma chamber with a 10 km diameter (D), and thermal diffusivity (κ) of 10⁻⁶ m²/s, the characteristic cooling time (t_c = D²/κ) is ∼3 m.y. (Gerya, 2010). For the Green Lake pluton and Soldier Lake pluton with diameters less than 10 km, we estimate full crystallization at 10² k.y. scales. For the Cathedral Peak pluton, Memeti et al. (2010b) and Paterson et al. (2011) calculated hypersolidus conditions of only 600–1000 k.y. for the entire Cathedral Peak pluton using two-dimensional thermal models. Petford et al. (2000) also suggested time scales of 0.1 m.y. for pluton crystallization. Consequently, in comparison with time scales of formation for metamorphic host-rock fabrics, we treat magmatic fabrics in plutons as strain “snapshots” for the U/Pb zircon age determined for the pluton.

We introduce a nomenclature to better categorize the magmatic fabrics based on one first proposed in Miller and Paterson (2001). We use M to denote magmatic foliation. A subscript indicates origin of the foliation and the relative foliation-forming sequence. In keeping with the use of S₀ to refer to layering associated with bedding (nontectonic origin of layers), we use M₀ to denote magmatic foliation caused by internal flow in magma chambers. We use the designation M₁, M₂,…, M₅, to designate the sequential formation of different magmatic foliations associated with regional tectonism. M_L denotes magmatic lineation, and the same subscript numeral designation indicates the sequence of formation for multiple lineations (M_L,1, M_L,2,…).

A superscript numeral varying from 0 to 5 defines magmatic fabric intensity as defined by the following parameters: 0 denotes no visible fabric; 1 is a weak fabric that is very difficult to recognize; 2 means the fabric is defined by moderate crystal alignment that is readily measured; 3 represents strong crystal alignment and may include crude layering; 4 is a very strong mineral alignment with continuous layering; and 5 denotes extremely intense alignment and layering. For example, refers to a magmatic foliation that is second oldest with an intensity of 3.

The 232.9 Ma Spiller Canyon pluton has a weak to moderate magmatic foliation defined by aligned plagioclase, biotite, and hornblende overprinted by a subsolidus mylonitic foliation caused by East Spiller Canyon thrust–related shearing. For the magmatic fabrics, the elongated shape of the Spiller Canyon pluton makes the margin-parallel and difficult to differentiate. These foliations strike 330/90, subparallel to the long axis of the pluton, and are associated with a subvertical magmatic lineation (⁠⁠) defined by alignment of magmatic hornblende and plagioclase. Magmatic fabrics in the Spiller Canyon pluton and other plutons are summarized in Table 2.

The 165.2 Ma Green Lake pluton has two sets of foliations (M₁ and M₂), a single magmatic lineation, and several types of mafic enclaves (Fig. 5B). The dominant magmatic foliation (⁠⁠) strikes NW (324/88), and the overprinting minor one (⁠⁠) strikes SW (245/75; Fig. 7F). The two sets of magmatic foliation share a single subvertical magmatic lineation (⁠⁠; Fig. 7G), defined by aligned magmatic hornblende and plagioclase. Margin-parallel M₀ is only locally found in the Green Lake pluton.

Mafic enclaves ranging from a few centimeters to several decimeters are common in the Green Lake pluton. Paterson et al. (2003) summarized three types of mafic enclaves in the field based on their geometries and relationships to magmatic fabrics. All types of enclaves have x axes parallel to magmatic lineation (x, y, z refer to the long, intermediate, and short axis of an enclave ellipsoid). Type I enclaves have x-y planes parallel to ⁠. Type II enclaves have x-y planes parallel to ⁠. Rare type III enclaves, called quadruple-pronged enclaves, have two sets of prongs that parallel both and ⁠, respectively (Table 2; Fig. 5B).

The 97.4 Ma Soldier Lake pluton has fabrics that are very similar to those in the Green Lake pluton. It has two sets of magmatic foliation, a dominant one (⁠⁠) that is oriented 316/75 and a second weaker one (⁠⁠) that is oriented 92/86 (Fig. 7H). The two sets of magmatic foliation share a single magmatic lineation (⁠⁠) that plunges steeply SE (Fig. 7I). The Soldier Lake pluton also has three types of mafic enclaves with the same relationships to magmatic fabrics as described for the Green Lake pluton. Local small-scale (a few to tens of meters), sinistral, subsolidus shear zones (Fig. 5C) in the Soldier Lake pluton strike approximately E-W and dip steeply N (Fig. 7J). These shear zones in the Soldier Lake pluton offset Cathedral Peak pluton–originated aplitic veins, indicating they formed after 86 Ma. Margin-parallel M₀ is also locally found in the Soldier Lake pluton.

The ca. 86 Ma Cathedral Peak pluton has been previously mapped by Žák et al. (2007) and Memeti et al. (2010b). Žák et al. (2007) recognized four types of magmatic foliations in the central part of the Cathedral Peak pluton. In their classification, NW-SE–striking type 3 and E-W–striking type 4 foliations are pluton-wide foliations, which they attributed to regional incremental strain. We view the type 3 and 4 magmatic foliations in the Cathedral Peak pluton as equivalent to M₁ and M₂ foliations, respectively, in the Soldier Lake pluton and Green Lake pluton, and we use the corresponding notation and in the Cathedral Peak pluton to indicate this interpretation. In the Cathedral Peak pluton, is oriented 299/83 (Fig. 7K), slightly more W-E than M₁ in the Soldier Lake pluton and Green Lake pluton. The two magmatic foliations also share a steeply plunging magmatic lineation (⁠⁠). We observed very few E-W–striking and margin-parallel M₀ foliations in parts of the Cathedral Peak pluton in the study area compared to the more developed M₂ and M₀ in other parts of the Cathedral Peak pluton and Tuolumne intrusive complex (Žák et al., 2007).

Magmatic fabrics can be associated with U/Pb zircon ages to represent the instantaneous maximum contraction and extension due to the short period of formation (<1 m.y.) as discussed before. Figure 11A summarizes the orientations of plutonic and host-rock structures in map view. Here, x, y, and z represent the directions of instantaneous principal strain. Magmatic foliation is perpendicular to the direction of the maximum instantaneous shortening (ε₃, z direction), and magmatic lineation parallels the direction of the maximum instantaneous elongation (ε₁, x direction).

The strike of the steeply dipping dominant magmatic foliation (M₁) in the 232.9 Ma Spiller Canyon pluton, 165.2 Ma Green Lake pluton, 97.4 Ma Soldier Lake pluton, and 86 Ma Cathedral Peak pluton is 330°, 324°, 316°, and 299°, respectively (errors related to best-fit calculation are shown in Table 2). The direction of ε₃ rotated counterclockwise with time. The 17° difference in the M₁ between the two Late Cretaceous plutons (Soldier Lake pluton and Cathedral Peak pluton) is notably larger than the 8° difference between the Jurassic Green Lake pluton and Soldier Lake pluton and the 6° difference between the Triassic Spiller Canyon pluton and Green Lake pluton. In contrast, the magmatic lineation in all four plutons is always subvertical. Thus, we infer that ε₃ rotated counterclockwise at variable rates through time while the direction of ε₁ remained constant.

One hypothesis for this rotation of ε₃ is that it reflects the decreasing angle of convergence (α) between the western margin of the North American plate and the vector of plate motion of the subducting Farallon plate (Fig. 11B). During homogeneous transpression, the vector of plate motion does not parallel the direction of ε₃. Angle α has a simple relationship with the angle (θ_p) between the plate margin and the direction of ε₃, which is α = 2θ_p − 90^ο (Fossen and Tikoff, 1993; Teyssier et al., 1995). If the Sierra Nevada arc did not localize strike-slip faults when the four plutons were emplaced, then α decreased from 90° at 232.9 Ma to 28° at 86 Ma, suggesting an increasingly wrench-dominated, transpressional tectonic setting in the Late Cretaceous. According to Tobisch et al. (1995), once α passes a critical value between 20° and 30°, tranpressional shear zones will form to accommodate the strike-slip strain. We infer that the Late Cretaceous Sierra Crest shear zone (Tikoff and Saint-Blanquat, 1997) and the Virginia Canyon shear zone developed once ε₃ rotated to this critical value.

Tikoff and Saint-Blanquat (1997) argued that regional deformation in the Sierra Nevada arc shifted from contractional to dextral transpressional at ca. 90 Ma. Tobisch et al. (1995) suggested such transition occurred between 100 Ma and 90 Ma. We estimate that the angle of convergence α rotated counterclockwise on average 0.20°/m.y. between 232.9 Ma and 97.4 Ma. The rate of rotation increased to 3.1°/m.y. between 97.4 Ma and 86 Ma. Despite our lack of further constraints on the specific timing of this rotation rate increase, a transition from contraction to transpression between 97.4 and 90 Ma fits both our data and previous models by Tikoff and Saint-Blanquat (1997) and Tobisch et al. (1995). If plate motion had changed before 97.4 Ma, we would expect to find pre–97 Ma noncoaxial structures. The geology in the study area and other areas in the Eastern Sierra pendants, however, is consistent with contraction through much of the Mesozoic and into the Late Cretaceous (see summary in Table 3).

Reconstructions of North American and Farallon plate motions have conventionally been based on a hotspot reference frame and seafloor magnetic patterns (Engebretson et al., 1985). Post–84 Ma relative plate motion reconstructions between the North American and Farallon plates indicate dextral strike-slip kinematics with slip rates comparable to, or greater than, those for the modern San Andreas fault (Engebretson et al., 1984, 1985; Doubrovine and Tarduno, 2008). However, plate motions are poorly constrained prior to ca. 84 Ma due to limitations on older hotspot and seafloor paleomagnetic data (Engebretson et al., 1985; Doubrovine and Tarduno, 2008). The kinematics of North American and Farallon plate interactions prior to 84 Ma are thus commonly based on orientation and kinematic data for a series of Cretaceous shear zones in the Sierra Nevada (Tobisch et al., 1995; Tikoff and Saint-Blanquat, 1997). For the first time, we have been able to semiquantify plate motions using magmatic fabrics to track plate convergence angles. This allows us to compare the temporal pattern of plate kinematics with other arc processes in later discussion.

Next, we briefly discuss and evaluate other possible mechanisms for the change in strike of magmatic foliation in Mesozoic Sierran plutons through time.

(1) The change in orientation of the magmatic foliations could be related to the large-scale oroclinal bend in trends of Precambrian to Mesozoic sedimentary strata across the eastern Sierra Nevada and western Great Basin (Albers, 1967).

In the east-central Sierra Nevada, the Jurassic thrust system north of the Saddlebag Lake pendant trends almost E-W as part of a Z-shaped curve that has been attributed to this regional oroclinal bend (Wetterauer, 1974; Albers, 1967; Lahren and Schweickert, 1995). Previous workers (Albers, 1967; Stewart et al., 1968; Wetterauer, 1977; Lahren and Schweickert, 1995) have suggested that this curvature in the thrust system is probably of Cretaceous age and kinematically related to the Mojave–Snow Lake fault, a cryptic dextral strike-slip fault. Memeti et al. (2010a) constrained the location of the Mojave–Snow Lake fault to a narrow, NW-SE corridor across the Tuolumne intrusive complex (Fig. 1C) and the timing of the dextral displacement along the Mojave–Snow Lake fault to 145–102 Ma. Paleomagnetic data from Geissman et al. (1984) also suggest that only minor clockwise regional oroclinal bending took place during the interval 100–70 Ma and younger. Consequently, the change in strike of the magmatic foliation in the post–100 Ma plutons is unlikely to have been associated with the pre–100 Ma regional oroflexural bending.

(2) The change in orientation of the magmatic foliations could be related to the cooling and intrinsic contraction of magma chambers during magma crystallization, which might shift the late incremental shortening direction.

The thermal strain associated with this volume change during the formation of the magmatic foliation can be calculated using the volumetric coefficient of thermal expansion of the magma α = 3 × 10^–5 K^–1 (Turcotte and Schubert, 2001) and assuming a 100 K drop in temperature (ΔT) before reaching the solidus: ε_T = ⅓ α ΔT. If the body is initially isotropic (Turcotte and Schubert, 2001), the thermal strain is ε_T = 1 × 10^–3. This implies that even over length scales of several kilometers, the volume change during pluton cooling would not produce strains greater than 0.1%. Thus, the cooling-related contraction of magma chambers is unlikely to affect the orientation of magmatic fabrics.

(3) The change in orientation of the magmatic foliations in the Late Cretaceous plutons could be related to Late Cretaceous shallow subduction caused by collision of an oceanic plateau with the North American margin.

Shallow subduction began around 90–85 Ma (Saleeby, 2003), which matches the timing of increasing rotation of magmatic foliation. Structures in the Sierra Nevada that relate to shallow subduction include: (a) extension and denudation of Cordilleran crust above the Rand-Pelona-Orocopia Schists (Saleeby, 2003; Chapman et al., 2013) and (b) low-angle thrust emplacement of these schists and the Franciscan complex. The average azimuth of schist transport is 205° with a shallow plunge (6.5°) toward the SW, consistent with the vergence of shear zones that led to the emplacement of these schists (Chapman et al., 2012). The direction of schist transport and shear-zone kinematics are difficult to reconcile with the approximately NE-SW shortening strain fields indicated by the magmatic fabrics. Thus, shallow subduction is unlikely to be the cause of the pattern of magmatic fabrics in the central Sierra Nevada.

After evaluating the previous hypotheses, we conclude that changing subduction kinematics is presently the best model to explain our magmatic fabric data. However, we acknowledge that we still do not fully understand the significance of the more minor set of M₂ foliations, and a more systematic study of fabrics in Triassic to Cretaceous plutons is needed to address these. An additional complication that we acknowledge is that the incremental strain recorded in plutons may also be affected by the partitioning of far-field plate deformation caused by heterogeneity in the arc crust (Žák et al., 2009).

The Sierra Nevada arc displays episodic magmatism (Ducea, 2001; DeCelles et al., 2009; Paterson et al., 2011), and three magmatic flare-ups occurred with peaks at 225 Ma, 161 Ma, and 98 Ma (Paterson and Cao, 2014). We propose that intra-arc deformation in the central Sierra Nevada also displays a temporal pattern that more or less mimics the episodic magmatism.

To describe the intra-arc deformation, we name periods during which more deformation occurs as higher-strain-rate periods and distinguish them from lower-strain-rate periods during which less deformation occurs. A complete intra-arc deformational cycle consists of a higher-strain-rate period and a lower-strain-rate period. Two-and-a-half intra-arc deformational cycles are proposed in the central Sierra during the Mesozoic. In the following discussion, we summarize the field evidence and timing for each cycle.

We propose the first higher-strain-rate period (HSP-1) started as early as subduction between the Farallon and the North American plates initiated around 245 Ma and after the Sonoma event and sinistral California-Coahuila shearing (Dickinson, 2004; Dickinson and Lawton, 2001). The beginning of HSP-1 may overlap the last stages of these two events. The earliest arc magmatic record is the 256–240 Ma plutons in the eastern Sierra–northern Mojave region (Christiansen, 1961; Miller et al., 1995) and 230–220 Ma in the east-central Sierra Nevada and Mono Basin–northern Owens Valley region (Barth et al., 2011). We propose that the HSP-1 might have lasted until ca. 220 Ma. The absence of stratigraphy of arc volcanic rocks prior to ca. 230 Ma in the Eastern Sierra pendants suggests that the central Sierra area was under exhumation when arc magmatism started, and volcanism is not well preserved (Miller and Sutter, 1982; Carr et al., 1992; Miller et al., 1995). Structures associated with NE-SW contraction are found in Paleozoic–Triassic strata during HSP-1 (Table 3). Strain preserved in these strata may include strain caused by pre-arc events that are difficult to separate. These deformational events have similar kinematics, suggesting a more or less arc-perpendicular contractional deformational field.

The first lower-strain-rate period (LSP-1) following HSP-1 started at ca. 220 Ma and ended ca. 180 Ma, corresponding to the first magmatic lull. In terms of the stratigraphic records, the Triassic volcanic Koip sequence found in many of the central Sierran pendants is overlain by the Sawmill Canyon sequence, a package of Jurassic shallow-marine sedimentary rocks. We interpret the contact between the two as a regional angular unconformity, which is bracketed between 187.3 and 220.3 Ma in the study area and 164 Ma and 203 Ma in the Ritter Range pendant (Cao et al., 2012; Tobisch et al., 2000). The significance of the unconformity is enigmatic, but it suggests that the arc was below sea level during the magmatic lull. At the plate-boundary scale, the western North American margin started to be affected by slab rollback caused by the dense Panthalassa abyssal plate from ca. 240 Ma to 170 Ma (Saleeby and Dunne, 2015). Busby-Spera (1988) suggested that slab rollback triggered synextensional volcanism between 235 and 190 Ma in the Mineral King section of the southern Sierra Nevada. In the central Sierra, the Sawmill Canyon sequence in the study area and Saddlebag Lake area suggests a shallow-marine environment back to the Late Triassic to Early Jurassic (this study; Schweickert and Lahren, 2006). However, whether it reflects regional extension related to slab rollback or a local intra-arc basin is difficult to determine due to the lack of widespread extensional structures in the central Sierra.

We propose that a second higher-strain-rate period (HSP-2) started at ca. 180 Ma and ended at ca. 150 Ma and generally corresponds to the Jurassic magmatic flare-up. A major NE-SW contractional event was recorded by the Middle Jurassic contractional structures in the study area and throughout the central Sierra Nevada (Table 3). Schweickert and Lahren (2006) related this Middle Jurassic contraction to the Nevadan orogeny in the east-central Sierra. Dunne and Walker (2004) also described repeated and episodic, broadly coaxial and coaxial-planar NE-SW contraction during a similar time period (>188 Ma to 140 Ma). At the plate-boundary scale, the Insular superterrane intra-oceanic arc was dextrally obliquely colliding with the Cordilleran margin ca. 170–157 Ma (Hillhouse and Gromme, 1984; Saleeby, 2000; Saleeby and Dunne, 2015).

The second lower-strain-rate period (LSP-2) started at ca. 150 Ma and ended at ca. 110 Ma, corresponding to the second magmatic lull. Volcanic rocks formed during this period of time (ca. 140–130 Ma) are found in Ritter Range pendant (Tobisch et al., 2000; Paterson et al., 2014). Volcanic and sedimentary rocks of Late Cretaceous ages ranging from ca. 115 Ma to 90 Ma (e.g., Minarets caldera) are commonly found in multiple host-rock pendants in the central Sierra (Tobisch and Fiske, 1982; Tobisch et al., 2000; Memeti et al., 2010a; Hartman and Paterson, 2013). These Late Cretaceous strata are separated from older strata by a regional angular unconformity. The timing of the unconformity is bracketed between 97.4 Ma and 117.4 Ma in the study area, 113 Ma and 170 Ma in the Saddlebag Lake area, 110 and 148 Ma in Oak Creek pendant, and ca. 100 Ma and 142 Ma in Ritter Range pendant (Hartman and Paterson, 2013; Longiaru, 1987; Tobisch and Fiske, 1982). The unconformity might be related to the renewed volcanism in the following Cretaceous flare-up. At the plate-boundary scale, Saleeby and Dunne (2015) suggested that the SW Cordillera margin underwent an episode of sinistral tangential shearing from ca. 157 Ma to 140 Ma as a marginal effect of the Mojave-Sonora megashear (Anderson and Silver, 1979, 2005). The shearing was accompanied by regional emplacement of the Independence dike swarm at ca. 148 Ma in the eastern Sierra region (e.g., Chen and Moore, 1982; Glazner et al., 1997).

A third higher-strain-rate period (HSP-3; ca. 110–80 Ma) more or less corresponds to the Late Cretaceous magmatic flare-up. The existence of deformed Late Cretaceous plutons and many contemporary shear zones provides a higher temporal resolution of deformational fields during this period of time. Field evidence for NE-SW contraction is present in many host-rock pendants and contemporaneous Late Cretaceous plutons (Table 3). At the plate-boundary scale, oblique subduction began at this time. Corresponding to this event, many studies, including the present study, have documented a major transition of the deformational field from contraction to dextral transpression at ca. 90 Ma (Tikoff and Saint-Blanquat, 1997; Greene and Schweickert, 1995; Tobisch et al., 1995). This transpressional deformational field did not rotate older regional foliation in host rocks. There are two possible explanations for this phenomenon: First, the Late Cretaceous strain was not strong enough to rotate the coaxial strain that had accumulated since the early Mesozoic. Second, the simple shear component of the transpressional strain is likely to have partitioned into the Virginia Canyon shear zone, while the pure shear component was added to the existing foliation coaxially (Tikoff and Greene, 1997).

Other Late Cretaceous structures in the central Sierra Nevada include spatially scattered conjugate kinks and box folds that crenulated the Cretaceous foliation in the host rocks (Tobisch and Fiske, 1976; Paterson, 1989). Those crenulations record NW-SE shortening postdating the NW-striking regional foliation, and they may be related to the elastic recovery of the orogen (Paterson, 1989). The magnitude of this deformation is insignificant and not a main phase of deformation. Magmatism shut down at ca. 84 Ma (Chen and Moore, 1982) and an increase in the exhumation rate of the Sierra Nevada began at about the same time (Cecil et al., 2006). We view the third cycle as incomplete, containing only the record of a higher-strain-rate period (HSP-3) and lacking a lower-strain-rate period.

To quantify the intra-arc deformation, we estimated the bulk intra-arc strain and propose a model describing the temporal history of strain rates of intra-arc deformation.

Cao et al. (2012) compiled finite strain data of host rocks based on over 650 field measurements from the central Sierra Nevada (for data sources and methods, see GSA Data Repository material [see ^{footnote 1}]). Figure 12 shows the results plotted in a z-shortening versus x-extension diagram (x, y, and z refer to axes of the finite strain ellipsoid caused by cumulative Mesozoic deformation). The results show that the ductile strain associated with host-rock foliation is heterogeneous but averages 49.4% arc-perpendicular shortening with a standard deviation of 16.7%. The deviation of plotted results from the central bold curve (y-extension = 0) indicates that the ductile strain is not perfectly plane strain. Except for the strain results from the slates, which experienced nondeformational flattening during lithification, most of the results plot within the y = 0 ± 15%–20% range, suggesting limited arc-parallel extension or volume change. We use 50% shortening as the intra-arc deformation related to host-rock foliation formation.

The strain related to brittle deformation on thrusts and bedding rotation can be estimated by assuming that little volume loss occurred during brittle deformation, and extension along the strike of thrusts was also negligible. Tobisch et al. (2000) proposed that a maximum of 45° tilting could be achieved by duplex structures associated with thrusting. We estimate that these constraints correspond to ∼30% shortening (Fig. 13).

The combination of strain associated with foliation (∼50% shortening) and thrusting (∼30% shortening) is equal to ∼65% arc-perpendicular shortening. Folding and later shortening may have further rotated the bedding to steeper orientations. Although we do not have good constraints from field observations, we relax the arc-perpendicular shortening to 85% to accommodate the strain required to rotate bedding to its present dip. The actual value for intra-arc finite shortening strain may vary, but we can treat ∼65% and ∼85% shortening as the lower and upper bounds, respectively. Next, we use the upper bound of ∼85% shortening to estimate the strain rates during the Mesozoic.

We propose a model for partitioning the total strain at different periods of time and estimate the strain rates needed to achieve the amount of shortening observed. Figure 14A illustrates the relationship between the finite pure shear strain coaxially accumulated along the maximum shortening direction (z direction) and time. Curves represent natural strain rates of different values ranging from 10⁻¹⁶ s^–1 to 10⁻¹⁴ s^–1. We propose that total ∼85% shortening can be achieved with the following combination: (1) an ∼30-m.y.-long period of bulk shortening strain at a rate of 10⁻¹⁵ s^–1, resulting in ∼61% shortening, (2) an ∼60-m.y.-long period of bulk shortening at a strain rate of 10^–15.5 s^–1, resulting in ∼47% shortening, and (3) an ∼80-m.y.-long period of bulk shortening at a strain rate of 10⁻¹⁶ s^–1, resulting in ∼25% shortening.

In the previous discussion, we proposed that intra-arc deformation could be divided into higher-strain-rate periods and lower-strain-rate periods. To quantify the deformation intensity, as shown in Figure 14B, we further propose that one higher-strain-rate period consists of a 10-m.y.-long peak (HSP-P), during which the bulk shortening strain rate is 10⁻¹⁵ s^–1, followed by a 20-m.y.-long transitional period (HSP-T), during which the bulk shortening strain rate is 10^–15.5 s^–1. One lower-strain-rate period corresponds to a 40-m.y.-long period during which the bulk shortening strain rate is 10⁻¹⁶ s^–1. The shortening strain accumulated during a higher-strain-rate period and half of lower-strain-rate period is 41% and 7%, respectively. Figure 14B shows strain rates before, during, and after a higher-strain-rate period. Panel 3 of Figure 15 illustrates the entire evolution of strain rates.

The length of one HSP-P agrees with the typical time span of a deformational phase in an orogenic system, most of which last less than 30 m.y., with many lasting 5–10 m.y. (Pfiffner and Ramsay, 1982). Pfiffner and Ramsay (1982) also suggested that measurable strain associated with cleavages in orogenic systems is accommodated in less than 10 m.y., or even less than 0.1 m.y., which correlates to strain rates at the order of 10⁻¹⁵ s^–1 to 10⁻¹³ s^–1, respectively. Paterson and Tobisch (1992) also argued that regional cleavage formation in the Foothills terrane could form in 23 m.y. with a strain rate of 10⁻¹⁵ s^–1, or it could form in 2 m.y. at strain rates of 10⁻¹⁴ s^–1. Thus, a 10-m.y.-long HSP-P with strain rates of 10⁻¹⁵ s^–1 fits these conclusions fairly well.

Spatially localized or temporally short-lived structures, such as shear zones, may have much higher strain rates than 10⁻¹⁵ s^–1. However, at the bulk arc scale, the strain rate cannot be as high as those that produced the localized structures. Otherwise, the crust would be unrealistically deformed over tens of millions of years. For example, continuous coaxial pure-shear deformation over 5 m.y. at the strain rate of 10⁻¹⁴ s^–1 will lead to ∼85% shortening. Our model does not exclude the possibility that during the higher-strain-rate periods, intermittent short deformation phases at finer sub–10-m.y. temporal scales could have higher strain rates.

Downward transfer of host rocks is proposed by some researchers to solve the “space problem” during pluton emplacement (e.g., Saleeby, 1990; Tobisch et al., 2000; Paterson and Farris, 2008). In the Sierra Nevada, plutons intruding older volcanic rocks or sedimentary rocks suggest that the latter supracrustal materials must be transported downwards to emplacement levels that range from 2 to 3 kbar in the central Sierra Nevada to 9–10 kbar in the southern Sierra Nevada (Saleeby, 1990; Chapman et al., 2012). Other lines of evidence of downward transfer of host rocks include steep stretching lineations in both plutons and surrounding host rocks and rotation and folding of bedding adjacent to plutons (Paterson and Farris, 2008).

Episodic intra-arc shortening and magma intrusion may lead to episodic crustal thickening. If exhumation of the arc crust is limited (Fig. 15, panel 6), it will lead to episodic transport of supracrustal materials downward and would fertilize the magma source regions in addition to crustal inputs from forearc and retro-arc regions and the downgoing plate (e.g., Kay et al., 2005; Ducea, 2001; DeCelles et al., 2009; Hacker et al., 2011; Chapman et al., 2013, 2014). Oxygen isotopic results suggest that crustal sources have contributed up to 50% to the Sierran magma (Ducea and Barton, 2007), but isotopic data cannot constrain the exact source of supracrustal materials, leaving open the possibility that the intra-arc downward transfer of host rocks is one likely source.

Figure 15 presents our attempt to summarize and compare different geological processes within the same temporal framework. Panel 1 is a comparison between relative magmatic addition rates, higher-strain-rate periods, lower-strain-rate periods, and unconformities in the central Sierra Nevada through time. The magmatic flare-ups in the central Sierra Nevada initiated at ca. 250 Ma, 180 Ma, and 120 Ma, and each typically lasted ∼20–30 m.y. The peak of each flare-up event occurred at ca. 225 Ma, 161 Ma, and 98 Ma (Paterson and Cao, 2014). We propose that the three higher-strain-rate periods match the three flare-ups, and the two lower-strain-rate periods coincided with the two magmatic lulls. The black box indicates the 10 m.y. peak period of each flare-up and HSP-P. Regional angular unconformities in the central Sierra Nevada are also plotted in panel 1 (Fig. 15). Each unconformity formed before a magmatic flare-up. This suggests potential cyclicity of sedimentation/volcanism.

Panel 2 (Fig. 15) plots the bulk intra-arc natural shortening strain rates through time. It suggests that the strain rates mimic the magmatic cycles. During the lower-strain-rate periods, bulk arc strain rate is on the order of magnitude of 10⁻¹⁶ s^–1 and increases to 10⁻¹⁵ s^–1 during the HSP-P. The “DTH” in panel 2 refers to downward transport of host rocks. We propose this process is likely to coincide with the peak period of flare-ups and HSP-P, when the arc is experiencing strong magmatism and deformation.

Panel 3 (Fig. 15) shows the shortening strain through time. Since natural strain is plotted, the slope of strain accumulation decreases when the finite strain is high. Finite shortening strain is labeled at the end of each higher-strain-rate period. Arc-perpendicular shortening, evidenced by the formation of thrusts, strata tilting, and host-rock fabric, occurred during the higher-strain-rate periods (Table 3). For the lower-strain-rate periods, structures and the amount of associated strain during these periods remain less well constrained.

Panel 4 (Fig. 15) shows data sets that constrain the exhumation of midcrustal rocks in the central Sierra Nevada during the Cenozoic (Cecil et al., 2006; House et al., 1997; Dumitru, 1990) and Mesozoic. The data set for Mesozoic exhumation is based on the aluminum-in-hornblende barometer coupled with U/Pb zircon ages of four Mesozoic plutons (Cathedral Peak pluton, Green Lake pluton, and Soldier Lake pluton are from study area; Eagle Peak pluton [EPG] is from the nearby Twin Lakes areas; Paterson et al., 2014). Panel 4 suggests that (1) from the Late Triassic to Late Cretaceous, the central Sierra Nevada was exhumed ∼5–7 km; (2) the exhumation was generally steady and slow during the majority of the Mesozoic, and there is no cyclic pattern based on the available data; and (3) the sudden increase in the amount of exhumation at ca. 85–90 Ma coincides with the shutdown of the Cretaceous magmatic flare-up. A similar sudden increase of exhumation is not observed for the Triassic and Jurassic flare-ups.

Panel 5 (Fig. 15) shows data sets that constrain the convergence angle between the Farallon and North American plates during the Mesozoic (Tobisch et al., 1995; Page and Engebretson, 1985; Ducea, 2001). The central dashed line separates regimes of dextral and sinistral kinematics of plate subduction. Convergence angles fluctuated in the Triassic and Jurassic and switched to dominant dextral transpression after 97–90 Ma in the Late Cretaceous, and they show no cyclic pattern. Gray bars in panel 5 mark the possible kinematic transitions.

Panel 6 (Fig. 15) summarizes regional extra-arc events that occurred during arc development (see previous discussion for citations). Tectonism along the western margin of North America shows no direct temporal correspondence with the cyclic pattern of magmatism and intra-arc deformation (panel 1 and panel 2, Fig. 15). The kinematic transition matches the switch of regional tectonic events. For example, the kinematic transition in the Early Triassic may be related to the transition from California-Coahuila sinistral shearing to head-on convergence when the arc initiated. The transition in the Middle Jurassic may be related to transition from dextral collision of the Insular superterrane to Mojave-Sonora sinistral shearing. The transition in the Late Cretaceous is related to the dextral transpression discussed earlier.

Panel 7 (Fig. 15) shows the inferred intra-arc deformation fields. It suggests that the arc-perpendicular contraction occurred from the Triassic to Early Cretaceous, and dextral transpression took place in the Late Cretaceous. We do not exclude any possibility that short-period, neutral extension and dextral/sinistral transtension/transpression may have occurred, but field evidence for these structures is scarce in the central Sierra Nevada.

The temporal match between arc magmatism and deformation is likely to reflect the feedbacks between the two processes. The intrusion of voluminous plutons increased the thermal profile and decreased the bulk viscosity of the arc crust (Barton and Hanson, 1989; Babeyko et al., 2002), which promoted ductile deformation to develop in middle and lower crust. In the upper crust, thrust faults developed at the same time to accommodate the crustal shortening. Elevated thermal profiles induced by magmatism have also been argued for by Saint-Blanquat et al. (1998) and Tobisch et al. (1995) to facilitate the development of ductile shear zones and the migration of the thermal axis in the Cretaceous Sierra Nevada (Chen and Moore, 1982).

Intra-arc deformation also influences magmatism. For example, Tobisch et al. (2000) suggested that steeply dipping bedding in host rocks reduces the surface that resists sinking and facilitates the downward transfer of host rocks and the ascent of magma. Steep bedding and thrust faults may provide subvertical anisotropic zones for guiding transport in magmatic dikes. Dikes will thermally and rheologically soften the host rocks and facilitate the diapiric ascent of larger, subsequent magmatic pulses. Miller and Paterson (2001) proposed a model of magma emplacement evolving from sheeted dikes to blobby diapirs in the Cascades core, Washington. This model may be applicable to the Sierra Nevada as well. The downward transfer of host rocks may fertilize the magma source region with supracrustal materials and promote the generation of melt and a residual arc root. Such a process is likely to be most efficient in hot arc crust when ductile flows of host rocks are active. The possible consequence of such feedbacks between magmatism and deformation is the establishment of an intra-arc convection system (Babeyko et al., 2002), which recycles arc crustal materials.

Chin et al. (2012) suggested that arc lithosphere thickened by magmatism and tectonic deformation may pinch out the asthenospheric mantle and interact with the subducting slab, cooling the arc lithosphere and terminating arc magmatism. Similarly, Karlstrom et al. (2014) suggested that thickening of the lithosphere could cause the migration of the arc front and truncation of mantle melting. Whether these processes are related to the proposed arc tempos deserves further study.

We do not fully understand the exhumation path shown in panel 4, Figure 15. The apparent noncyclic pattern may be due to the low temporal resolution of the data. Surface erosion, exhumation, and rock uplift depend on both intra-arc (e.g., crustal thickening) and extra-arc factors (e.g., climate and annual precipitation rate), either cyclic or noncyclic (C.-T. Lee, 2014, personal commun.).

Plate convergence angle fluctuated throughout the Mesozoic and has little correlation to the cyclic magmatism, although subduction is the ultimate cause of arc magmatism. This implies that the Sierra Nevada arc may be a self-regulating systems, as proposed by DeCelles et al. (2009, 2015).

The decoupling between regional tectonic events (panel 6, Fig. 15) and intra-arc deformation (panel 7, Fig. 15) is likely to be controlled by strain partitioning. In the Sierra Nevada arc, later strike-slip motion may have taken advantage of preexisting strike-slip structures in the forearc region. This would protect the arc from the transcurrent deformation caused by regional tectonic events. However, magmatic fabrics may still be sensitive to plate-kinematic changes, as long as forearc structures were not active when the fabric-recording pluton was emplaced.

We end this paper with two points we want to emphasize. First, a unifying model explaining the arc tempos in a continental arc (DeCelles et al., 2009) should include intra-arc processes: Their links to other processes merit further study. Second, the concept of “arc tempos” should be treated carefully because not all arc processes show temporal cyclic patterns. The evolution of continental arcs is likely influenced by interrelated cyclic processes, which are overprinted by noncyclic processes controlled by events external to the arcs.

Structures in plutons and host rocks in the Virginia Canyon area are coupled with geochronologic data to establish a temporal history of paleodeformation fields in the Mesozoic central Sierra Nevada.

(1) Deformation in the host rocks of Paleozoic–Mesozoic age in the central Sierra Nevada includes (a) Triassic, Jurassic, and Cretaceous NE- or SW-vergent thrust faults and associated folds; (b) time-transgressive, regional NW-striking foliation and subvertical lineation resulting from overprinting by coaxial strain due to contractional events throughout the Mesozoic; and (c) Late Cretaceous dextral, transpressional shear zones.

(2) Magmatic fabrics (M₁) in Mesozoic plutons tracked subtle changes of paleodeformation fields. The strikes of steeply dipping magmatic foliation in several plutons display a counterclockwise rotation over time: NW-SE–oriented magmatic foliation in the older plutons evolves to more E-W–oriented magmatic foliation in the younger plutons. Such a change of magmatic foliation was likely linked to the change of subduction kinematics from arc-perpendicular to oblique convergence in the Late Cretaceous.

(3) We propose that, in some ways, similar to magma flare-ups, intra-arc deformation in the central Sierra Nevada exhibits cyclic patterns. Magmatism promotes intra-arc deformation, while deformation causes crustal thickening, downward transport of crustal materials to magma source regions, and fertilization of the magma source regions with supracrustal materials. Models explaining the arc tempos in continental arcs should include the intra-arc processes. The evolution of continental arcs may be influence by intra-arc cyclic process, accompanied by noncyclic processes controlled by events external to the arcs.

We acknowledge support from National Science Foundation grants EAR-0537892 and EAR-0073943 awarded to Scott Paterson, U.S. Geological Survey EDMAP award G12AC20178, Geological Society of America Graduate Student Research Grants, and the Graduate Student Research Fund from University of Southern California (USC) Department of Earth Sciences. We thank undergraduate students Tiffany Ikeda, Andrew Whitesides, Phillip Ehret, Ian Cox, Brittany Gelbach, Emily Van Guilder, Jill Hardy, Kristan Culbert, and Sean Zalunardo, who were associated with USC’s Undergraduate Team Research program, for their assistance in constructing excellent maps in the central Sierra Nevada and for helping to obtain laser-ablation–inductively coupled plasma–mass spectrometry zircon ages. We thank Mark Pecha, George Gehrels, and other laboratory scientists from Arizona LaserChron Center for helping in the dating laboratory. Greg Davis, Sean Hartman, Babsi Ratschbacher, Snir Attia, and Katie Ardill are gratefully thanked for their discussions and comments on the manuscript. We also thank Editor John Goodge, reviewers Allen Chapman, Doug Walker, Jean-Louis Vigneresse, and an anonymous reviewer for their helpful comments that greatly improved the quality of the manuscript. Finally, Cao would like to thank his fiancée Jingjing for her support and patience throughout the past 6 yr.