Development of the Whitehorse trough as a strike-slip basin during Early to Middle Jurassic arc-continent collision in the Canadian Cordillera

Synorogenic strata are the depositional records of mountain building and provide opportunities to investigate the timing of plate-tectonic processes along convergent margins. Immature terrestrial and marine siliciclastic strata in orogen-proximal basins have provenance signatures that faithfully record regional geology (Dickinson and Suczek, 1979; Dickinson et al., 1983; Ingersoll et al., 1993; Johnsson, 1993; Link et al., 2005) and are generated in response to discrete crustal thickening, exhumation, and other convergent margin events (e.g., Dickinson, 1974; Cawood et al., 2009, 2012; Horton, 2018; Enkelmann et al., 2019). Detrital zircon U-Pb isotope studies have proven to be useful for constraining the age, provenance, and paleogeography of ancient synorogenic rock units along Cordilleran-type margins, which are characterized by multiphase deformation, metamorphism, and magmatism (e.g., Willner et al., 2008; Amato and Pavlis, 2010; Horton et al., 2010; Wu et al., 2010; Anfinson et al., 2012; Laskowski et al., 2013; Beranek et al., 2016; Romero et al., 2020). Detrital zircon Hf isotope geochemistry has emerged as a complementary tool for identifying the crust-mantle contributions of igneous source rocks, and, as a result, it provides evidence for regional crustal evolution in the stratigraphic record (e.g., Hawkesworth and Kemp, 2006; Flowerdew et al., 2007; Iizuka et al., 2010; Liu et al., 2017). Detrital zircon U-Pb-Hf isotope investigations of synorogenic strata give opportunities to monitor increased crustal contributions during non-steady-state subduction or collisional events, which are typically defined by excursions toward subchondritic compositions or negative (evolved) εHf_(t) values, known as isotopic pull-downs (DeCelles et al., 2009; DeCelles and Graham, 2015; Pecha et al., 2016).

The Intermontane terranes comprise the core of the Canadian Cordillera and are flanked to the east by parautochthonous units of the North American margin and to the west by the more exotic Insular and Alaskan terranes (Fig. 1; e.g., Mortensen, 1992; Monger and Price, 2002; Colpron et al., 2007a; Nelson et al., 2013). The Late Triassic to Middle Jurassic (all stratigraphic and numerical ages follow the time scale of Cohen et al., 2013) tectonic evolution of the northern Canadian Cordillera was marked by waning arc magmatism, arc collision, syn-to postcollisional magmatism, and syntectonic filling of the Whitehorse trough, which overlaps Upper Triassic and older rocks of the Stikinia and Cache Creek terranes from central Yukon to northern British Columbia (Figs. 1 and 2; e.g., Mihalynuk et al., 1994; English and Johnston, 2005; Colpron et al., 2015, 2022). Late Triassic to Middle Jurassic arc collision was accommodated by the enclosure of the Cache Creek terrane and resulted in crustal thickening and burial of Yukon-Tanana terrane rocks to 5–9 kbar (~16–30 km) in Yukon (e.g., Berman et al., 2007; Clark, 2017; Gaidies et al., 2021). Late Triassic to Early Jurassic (205–194 Ma) syncollisional plutons were emplaced at mid-to lower-crustal levels (5–7 kbar, ~16–23 km) into metamorphic basement along the flanks of the Whitehorse trough and subsequently cooled and exhumed at rates of 0.7–7.5 mm/yr (e.g., Johnston and Erdmer, 1995; Johnston et al., 1996; Knight et al., 2013; Colpron et al., 2022). Exhumation processes in central Yukon drove Early to Middle Jurassic erosion and resulted in the syntectonic deposition of the Laberge Group, a >3000-m-thick succession of immature marginal-marine to deep-marine strata that are typically interpreted to have filled a forearc basin (e.g., Dickie and Hein, 1995; English and Johnston, 2005) or a syncollisional basin that overlapped the northern Intermontane terranes (Colpron et al., 2015). Recently, Colpron et al. (2022) proposed that latest Triassic to Early Jurassic end-on arc collision in the northern Intermontane terranes was followed by southward migration of the Stikinia arc (Hazelton Group; Nelson et al., 2022) and that the collision zone in the north was linked to the retreating subduction zone by a sinistral transform fault system, possibly including the Llewellyn and Teslin faults (Currie and Parrish, 1993; Mihalynuk et al., 1999; de Keijzer et al., 2000). This new model predicts that sinistral transtension accommodated extensional exhumation of the orogen and subsidence of the Whitehorse trough as a strike-slip basin. Basal units of the Laberge Group are lithologically diverse and overlie different sedimentary and igneous rock substrates in Yukon (Lowey, 2004; Colpron et al., 2007b), which together imply block-faulted basement topography and multiple erosion-deposition events during Whitehorse trough subsidence. Laberge Group deposition was coeval with the onset of Jurassic foreland basin subsidence in Alberta (Fernie Formation in Fig. 1; Cant and Stockmal, 1989; Price, 1994; Pan et al., 2018) and suggests that the development of the Whitehorse trough accompanied the rise of the Cordilleran hinterland-retroarc thrust system (Colpron et al., 2015, 2022).

The timing of Laberge Group deposition and relative sediment contributions from Intermontane terrane sources during regional exhumation are required to understand the tectonic setting and filling history of the Whitehorse trough. In this article, we build on field studies (van Drecht et al., 2017; van Drecht and Beranek, 2018; Bordet et al., 2019) and report new detrital zircon U-Pb-Hf results from 12 Laberge Group sandstone samples, including six basal samples that sit unconformably on Triassic sedimentary and intrusive rocks of Stikinia. We used this data set to test the relationships between the timing of exhumation and Whitehorse trough deposition. The laser-ablation split-stream technique, which allows for the simultaneous collection of U-Pb and Hf isotope data in zircon (e.g., Fisher et al., 2014; Beranek et al., 2020), was used to identify juvenile to evolved source rocks and characterize an isotopic pull-down recorded by Laberge Group strata. This isotopic pattern matches the one recognized in syncollisional plutonic rocks that intruded the Intermontane terranes during an Early Jurassic episode of crustal thickening in the northern Canadian Cordillera (Colpron et al., 2022). We combined our new detrital zircon U-Pb-Hf isotope data with published fossil and U-Pb results (Colpron, 2011; Colpron et al., 2015; Bordet et al., 2019) to constrain Sinemurian to Toarcian maximum depositional ages for Laberge Group units and basin-filling patterns. These data suggest that the pattern of subsidence in the Whitehorse trough is consistent with development as a strike-slip basin within a sinistral transtensional regime during the early stages of development of the northern Cordilleran orogen (Colpron et al., 2022).

The Intermontane realm of the Canadian Cordillera includes the Yukon-Tanana, Stikinia, and Quesnellia arc terranes and oceanic and accretionary complexes of the Slide Mountain and Cache Creek terranes (Fig. 1; e.g., Nelson et al., 2013). Late Devonian to late Permian arc successions of the Yukon-Tanana terrane intrude and cover the Snowcap assemblage, a metasedimentary basement unit that yields Precambrian detrital zircon age populations consistent with northwest Laurentian provenance (Mortensen, 1992; Colpron et al., 2006, 2007a; Piercey and Colpron, 2009). Development of the Yukon-Tanana arc occurred concurrently with rifting and opening of the Slide Mountain back-arc ocean between the Late Devonian and Permian (e.g., Nelson et al., 2006). The Yukon-Tanana terrane locally forms the depositional basement to the early Mesozoic Quesnellia arc in Yukon and British Columbia (e.g., Simard et al., 2003; Nelson and Friedman, 2004; Roots et al., 2006; Nixon et al., 2020; Colpron et al., 2022). Upper Paleozoic arc successions also form the basement to Triassic arc volcanic and sedimentary rocks of northern Stikinia, but their correlation with the Yukon-Tanana terrane is tenuous (e.g., George et al., 2021); they include the Stikine assemblage in northern British Columbia (e.g., Gunning et al., 2006) and the Takhini and Boswell assemblages in south-central Yukon (Fig. 2; Tempelman-Kluit, 1984; Hart, 1997; Colpron et al., 2006). Overlying Upper Triassic volcanic and sedimentary rocks of Stikinia and Quesnellia have similar stratigraphic and petrogenetic histories and are generally interpreted to have formed part of a single, continuous arc system that now occurs on either side of an accretionary complex and related oceanic rock units of the Cache Creek terrane (Fig. 1; Nelson and Mihalynuk, 1993; Mihalynuk et al., 1994).

Late Paleozoic to early Mesozoic collapse of the marginal ocean basin system along western Laurentia, the vestiges of which comprise the Slide Mountain terrane (Fig. 1), resulted in the docking of Yukon-Tanana and Quesnellia against the North American margin and deposition of Triassic overlap assemblages in eastern Alaska, Yukon, and British Columbia (Unterschutz et al., 2002; Beranek et al., 2010; Beranek and Mortensen, 2011; Golding et al., 2016). Stikinia was likely part of a west-facing arc festoon that protruded into Panthalassa, similar to the modern Aleutian arc, and flourished during the Middle to Late Triassic along western Pangea (e.g., Mihalynuk et al., 1994; George et al., 2021; Colpron et al., 2022). In south-central Yukon, Middle Triassic rocks of the Joe Mountain Formation include Ladinian basalt with tholeiitic arc to back-arc geochemical affinities, siliciclastic and calcareous strata, and mafic intrusions (Hart, 1997; Bordet et al., 2019; Bickerton et al., 2020). The Joe Mountain Formation has tentatively been assigned to Stikinia (e.g., Hart, 1997), but recent studies suggest it may be correlative with primitive arc successions assigned to the Cache Creek terrane in British Columbia (e.g., Kutcho-Nakina assemblages; Zagorevski et al., 2018; Bordet et al., 2019; Bickerton et al., 2020). Zagorevski et al. (2021) assigned these rocks to the newly defined Atlin terrane.

Upper Triassic strata of Stikinia in Yukon comprise the Lewes River Group (Fig. 3; Wheeler, 1961) and include: (1) Carnian volcaniclastic rocks and augite-phyric basalt of the Povoas formation (informal; Tempelman-Kluit, 1984; Hart, 1997); and (2) Carnian to Rhaetian sedimentary rocks of the Aksala formation (informal; Tempelman-Kluit, 1984), which consists of the Casca, Hancock, and Mandanna members (Fig. 3). The Casca member contains Carnian to Norian lithic sandstone, argillite, and conglomerate units that are overlain by Norian to Rhaetian reefal limestone of the Hancock member (Hart, 1997; Bordet et al., 2019). Rhaetian sandstone, mudstone, and conglomerate units of the terrestrial to marginal-marine Mandanna member interfinger with and overlie the Hancock member (Tempelman-Kluit, 1984; Dickie and Hein, 1995; Hart, 1997; Long, 2005). The Aksala formation locally overlaps volcanic strata of the Joe Mountain and Povoas formations (Bordet et al., 2019). Augite-phyric basalt, volcaniclastic, and carbonate rocks of the Upper Triassic Semenof formation (informal; Tempelman-Kluit, 1984; Simard and Devine, 2003) are correlatives of the Lewes River Group to the east of the Teslin fault in central Yukon (Fig. 2).

Norian to Rhaetian arc plutons intrude Triassic volcanic assemblages of Stikinia (Stikine suite, 217–214 Ma; Headless plugs, 208–207 Ma) and Paleozoic basement of the Yukon-Tanana terrane (Pyroxene Mountain suite, 218–214 Ma; Fig. 2; Sack et al., 2020; Colpron et al., 2022). These arc plutons have juvenile whole-rock Nd-Sr and zircon Hf (εHf_[t] = +9.7 to +11.5) isotope compositions that indicate depleted mantle sources with little to no crustal contamination.

The Whitehorse trough is located at the center of an inverted V-shaped map pattern shown by the northern Intermontane terranes in Yukon and British Columbia (Fig. 1), which has typically been explained by Late Jurassic to Early Cretaceous strike-slip duplication of the Stikinia-Quesnellia arc (Wernicke and Klepacki, 1988) or Late Triassic to Middle Jurassic oroclinal bending of the Stikinia-Quesnellia arc and entrapment of the oceanic Cache Creek terrane (Mihalynuk et al., 1994; Colpron et al., 2015). Colpron et al. (2022) proposed an alternative model where entrapment of the Cache Creek terrane resulted from development of a sinistral transform fault system linked to the southward retreat of an Early Jurassic arc (Hazelton Group; Nelson et al., 2022) following initial, latest Triassic collision. In this model, the Whitehorse trough developed atop of the northern Intermontane terranes within a sinistral transtensional setting in the collisional zone to the north (Colpron et al., 2015, 2022). The Cache Creek ocean was consumed by Middle Jurassic time (e.g., Cordey, 2020), and its oceanic assemblages were emplaced onto Stikinia along thrust faults, marking the accretion of the Intermontane terranes against the continental margin (Mihalynuk et al., 1994, 2004).

Hettangian to Sinemurian (ca. 200–190 Ma) prograde metamorphism and burial of Yukon-Tanana terrane basement rocks to midcrustal levels (5–9 kbar, amphibolite facies) in western Yukon constrain the onset of collisional tectonism in the northern Intermontane terranes (e.g., Johnston et al., 1996; Berman et al., 2007; Clark, 2017; Gaidies et al., 2021). This was accompanied by intrusion of latest Triassic to Early Jurassic syncollisional plutons (Minto suite, 205–194 Ma; Sack et al., 2020) in Stikinia and the Yukon-Tanana terrane of central Yukon (Fig. 2). The Minto suite includes large batholiths with metaluminous to peraluminous geochemical signatures and whole-rock Nd (εNd_[t] = −3.6 to +1.3) and zircon Hf (εHf_[t] = 0 to +5) isotope compositions that show a shift toward subchondritic values relative to Late Triassic arc successions (Colpron et al., 2022). Al-in-hornblende pressures of 5–7 kbar suggest emplacement at mid-to lower-crustal depths (Sack et al., 2020). Exhumation of the Yukon-Tanana terrane and Minto suite plutons began by the Pliensbachian and probably was accommodated by crustal-scale normal faults such as the Willow Lake fault in central Yukon (Fig. 2; Knight et al., 2013). The ⁴⁰Ar/³⁹Ar mica cooling dates from Minto plutons suggest exhumation rates of 1.3–7.5 mm/yr (assuming a geothermal gradient of 25–30 °C/km; Sack et al., 2020).

Plutons of the Long Lake (Pliensbachian, 188–183 Ma) and Bennett suites (Toarcian, 178–175 Ma) intruded the Yukon-Tanana–Stikinia boundary during regional exhumation of the western flank of the Whitehorse trough. Long Lake and Bennett suite plutons were emplaced at shallower crustal depths (4–6 kbar) compared to Minto plutons and have peraluminous geochemical signatures and more-evolved whole-rock Nd (εNd_[t] = −5.9 to −1.5) and zircon Hf (εHf_[t] = −6 to +3) isotope compositions (Sack et al., 2020; Colpron et al., 2022). Late Triassic to Early Jurassic plutons that intrude Stikinia and Yukon-Tanana west of the Whitehorse trough show a progressive trend toward more-evolved isotopic values, or an isotopic pull-down, with time (Colpron et al., 2022). The ⁴⁰Ar/³⁹Ar mica cooling dates from the Aishihik batholith (Fig. 2; Long Lake suite) are consistent with exhumation rates of 0.5–2.8 mm/yr (Clark, 2017; Sack et al., 2020).

East of the Big Salmon–Teslin fault system, mafic to felsic plutons of the Lokken suite (Sinemurian to Pliensbachian, 195–184 Ma) that intrude the Yukon-Tanana terrane (Fig. 2) represent the northern extension of Quesnellia arc magmatism in Yukon. Lokken suite plutons have metaluminous geochemical signatures and whole-rock εNd_(t) values of −4.3 to −0.6 and zircon εHf_(t) values of −2.9 to +9.3 (Sack et al., 2020; Colpron et al., 2022). The Lokken suite plutons were emplaced at relatively shallow crustal depths (~3 kbar) and cooled rapidly (≤2 m.y.) compared to plutons west of the Whitehorse trough. Plutons of the Early Jurassic Minto, Long Lake, and Lokken suites contain inherited Late Triassic–Early Jurassic, Permian, Mississippian, Devonian, and Precambrian zircon cores derived from local Intermontane terrane country rocks (Colpron et al., 2022).

Posttectonic plutons of the Bryde and Fourth of July suites (Aalenian to Bajocian, 174–168 Ma) intruded Stikinia, the Cache Creek terrane, and overlapping strata of the Whitehorse trough after their imbrication in the Middle Jurassic (Fig. 2; Mihalynuk et al., 2004; Sack et al., 2020). These intrusive units have mafic to felsic alkaline compositions and generally yield whole-rock Nd (εNd_[t] = +0.5 to +3.5) and zircon Hf (mostly εHf_[t] = +4 to +8) isotope values that show a return to juvenile or superchondritic compositions. This ca. 174 Ma isotopic shift was coincident with the end of marine deposition in the Whitehorse trough and beginning of surface uplift recorded by overlying Upper Jurassic to Lower Cretaceous terrestrial strata of the Tantalus Formation, which filled confined intermontane valleys (Long, 2015). Colpron et al. (2022) suggested that the combined evidence for surface uplift and rapid shift to juvenile isotopic compositions reflects lithospheric delamination after imbrication of the Intermontane terranes.

The Whitehorse trough extends >600 km from Carmacks in central Yukon to Dease Lake in northern British Columbia (Fig. 1). Lower to Middle Jurassic rocks of the Tanglefoot and Richthofen formations (Laberge Group) define the Whitehorse trough in Yukon and comprise ~3000 m of siliciclastic strata that unconformably overlie the Stikinia and Cache Creek terranes (Figs. 2 and 3; Wheeler, 1961; Tempelman-Kluit, 1984, 2009; Dickie and Hein, 1995; Lowey, 2008; Colpron et al., 2015). The Whitehorse trough was deformed by southwest-verging folds and thrust faults during the Late Jurassic and Cretaceous, most likely in association with dextral transpression along the Teslin-Braeburn fault system (White et al., 2012).

The Tanglefoot formation consists of sandstone, mudstone, and conglomerate units with coal, terrestrial plant material, and vertebrate fossils that are indicative of marginal-marine environments, whereas the Richthofen formation contains graded siltstone to sandstone and mudstone couplets and conglomerate units with ammonites, belemnites, planktonic fossils, and trace fossils (Helminthopsis, Phycosiphon, Planolites) characteristic of deep-marine environments (Wheeler, 1961; Tempelman-Kluit, 1984, 2009; Long, 1986; Hart, 1997; Lowey, 2004; van Drecht et al., 2017; van Drecht and Beranek, 2018). Tanglefoot and Richthofen formation strata are restricted to the northern and central regions of the Whitehorse trough, respectively, suggesting a south-directed deepening of the basin (Fig. 2; Tempelman-Kluit, 1984, 2009; Dickie and Hein, 1995; Lowey, 2004, 2008; Hutchison, 2017). Felsic tuff and epiclastic rock units assigned to the Nordenskiöld member occur in both formations (Figs. 2 and 3) and record Pliensbachian (188–186 Ma) eruptive events coeval with intrusions of the Long Lake suite to the west (Colpron and Friedman, 2008).

Clast-to matrix-supported, polymictic, cobble to boulder conglomerate units occur in the Tanglefoot and Richthofen formations and represent debris-flow, sheetflood, and bar deposits of a fan-delta system (Dickie and Hein, 1995; Hart et al., 1995; Lowey, 2004; van Drecht et al., 2017). Paleoflow indicators suggest fan lobe migration and predominately show east-to northeast- and southwest-directed sediment transport along the western and eastern edges of the Whitehorse trough, respectively. Sediment pathways for the conglomerate units were typically transverse to the longitudinal axis of the Whitehorse trough (e.g., Wheeler, 1961; Dickie and Hein, 1995; Hart et al., 1995; Lowey, 2004). In northern British Columbia, Inklin Formation strata (equivalent to the Richthofen formation) show that longitudinal (north-directed) Sinemurian paleoflow was replaced by transverse Pliensbachian paleoflow (Johannson et al., 1997). Conglomerate clast types include augite-phyric basalt, andesite, tuff, granite, granodiorite, diorite, sandstone, volcanogenic sandstone, and limestone (Hart et al., 1995). Basal conglomerate units are dominated by volcanic and sedimentary clasts, whereas younger strata contain a larger proportion of plutonic clasts, which imply unroofing of the adjacent arc (Dickie and Hein, 1995; Hart et al., 1995; Johannson et al., 1997; Shirmohammad et al., 2011). Pliensbachian–Toarcian conglomerate units in northern British Columbia locally contain eclogite clasts that indicate rapid Early Jurassic exhumation at rates of ~4.1 mm/yr (Canil et al., 2006; Kellett et al., 2018). U-Pb dates of plutonic clasts and detrital zircon grains from the Laberge Group have yielded Late Triassic to Early Jurassic and minor Paleozoic age populations that indicate sources from local Intermontane terrane basement and collision-related plutons (Hart et al., 1995; Gordey et al., 1998; Colpron et al., 2015; Kellett and Iraheta Muniz, 2019; Kellett and Zagorevski, 2021).

Twelve rock samples were analyzed for detrital zircon U-Pb geochronology and Hf isotope geochemistry at Memorial University of Newfoundland (Table 1; Fig. 2). Laboratory analyses were conducted using the laser-ablation split-stream technique using two inductively coupled plasma–mass spectrometers (e.g., Fisher et al., 2014). Laser-ablation techniques and data reduction protocols are outlined in the Appendix. Detrital zircon U-Pb-Hf isotope results and reference material values are reported in the Supplemental Material (Table S1¹). U-Pb dates are presented in probability density plots (PDPs) made with a Microsoft Excel macro developed at the Arizona LaserChron Center (https://sites.google.com/a/laserchron.org/laserchron/). Initial ¹⁷⁶Hf/¹⁷⁷Hf ratios are reported as εHf_(t) and represent isotopic compositions at the time of crystallization relative to the chondritic uniform reservoir (CHUR). The εHf_(t) versus U-Pb age plots show the depleted mantle array (Vervoort and Blichert-Toft, 1999) and crustal evolution lines that use ¹⁷⁶Lu/¹⁷⁷Hf = 0.015 (Goodge and Vervoort, 2006). Statistical comparisons (cross-correlation, likeness, and similarity coefficients of PDPs, Kolmogorov-Smirnov and Kuiper tests) were conducted with the DZstats MATLAB program of Saylor and Sundell (2016), and results are reported in the Supplemental Material (Table S2). We followed the statistical assessments of Saylor and Sundell (2016) and favor cross-correlation coefficients (R² values range from 0 to 1, with a cross-plot value of 1 indicating identical age peaks) for interpretation purposes because they are sensitive to the presence or absence, relative magnitude, and shape of age peaks in PDPs. Multidimensional scaling (MDS) plots were constructed with the DZmds MATLAB program of Saylor et al. (2018) to compare U-Pb age results with potential sources and stratigraphic correlations.

The youngest detrital zircon crystals in each sample were used to estimate the maximum depositional ages of Laberge Group strata. Three techniques were used: (1) the youngest statistical peak (YSP) method, which takes a weighted mean of the youngest population of two or more grains that yield a mean square of weighted deviates (MSWD) ≈ 1 (Coutts et al., 2019; Herriott et al., 2019); (2) the youngest cluster at two sigma (YSC) method, which takes the weighted mean of the youngest three or more grains that overlap at 2σ (Dickinson and Gehrels, 2009); and (3) the youngest graphical peak (YPP) method, which is calculated by the youngest peak age in a probability density plot (Dickinson and Gehrels, 2009), in this case determined by the AgePick Excel macro program developed at the Arizona LaserChron Center.

Buff-green, cross-bedded feldspathic lithic arenite that overlies Aksala formation limestone (<20 m above contact) is the stratigraphically lowest part of the Tanglefoot formation sampled along the Robert Campbell Highway, ~25 km east of Carmacks (sample T2 in Fig. 4; Table 1). The sample mostly yielded 252–191 Ma (98%) detrital zircon grains with age peaks of 201 Ma and 228 Ma (Fig. 5A). Detrital zircon grains that ranged 213–191 Ma and 242–217 Ma gave εHf_(t) values that clustered between −0.2 to +5 and +10 to +12, respectively (Fig. 5C). A Pennsylvanian zircon grain (308 ± 3 Ma) yielded an εHf_(t) value of +2.4.

Buff-colored, medium-to very coarse-grained, feldspathic lithic arenite was collected 15 m above sample T2 along the Robert Campbell Highway (sample T3 in Fig. 4; Table 1), and it gave a 219–185 Ma population with an age peak of 195 Ma (Fig. 5B). The Late Triassic to Early Jurassic population had εHf_(t) values that ranged from −4.7 to +9.4 with a dominant cluster between −4.7 and −2.6 (Fig. 5C). Two Mississippian zircon grains of 342 ± 7 Ma and 353 ± 11 Ma had εHf_(t) values of −3.8 and +4.1, respectively.

Maroon, medium-to coarse-grained, feldspathic lithic arenite that crops out at Conglomerate Mountain, along the Klondike Highway ~40 km south of Carmacks (sample T4 in Fig. 2; Table 1), yielded a bimodal distribution of Late Triassic to Early Jurassic (48%) and Late Devonian to Early Triassic (52%) zircon grains (Fig. 5D). The 221–180 Ma population yielded age peaks of 187 Ma and 205 Ma and εHf_(t) values of +0.9 to +11.5 (Fig. 5F). Late Devonian to Early Triassic zircon grains produced a 313 Ma age peak and εHf_(t) values that ranged from +5.7 to +12.7.

A sample of buff-colored lithic arenite was collected from the stratigraphically highest Tanglefoot formation near Tantalus Butte along the Robert Campbell Highway (sample T5 in Fig. 4; Table 1). Tanglefoot strata in this area are overlain by Upper Jurassic to Lower Cretaceous conglomerate of the Tantalus Formation and are projected to overlie a Pliensbachian crystal lithic tuff horizon to the east (187.1 ± 0.2 Ma; Colpron and Friedman, 2008). This sample mostly contained 212–176 Ma (92%) detrital zircon grains with an age peak of 185 Ma (Fig. 5E). Most εHf_(t) values of Late Triassic to Early Jurassic grains ranged from −4.3 to +4.8, and one 212 Ma zircon yielded a εHf_(t) value of +8.8 (Fig. 5F). An early to mid-Paleozoic (8%) population yielded an age peak of 338 Ma and εHf_(t) values between −8.7 and −5.7. Two Middle Mississippian zircon grains had εHf_(t) values of +4 and +8.7.

Parallel laminated, fine-to medium-grained, feldspathic lithic arenite that disconformably overlies Aksala formation limestone (<5 m above contact) along the eastern flank of Mount Laurier, ~40 km northeast of Whitehorse (sample R5 in Fig. 6; Table 1), yielded 290–225 Ma (58%) detrital zircon grains with an age peak of 205 Ma (Fig. 7A) and εHf_(t) values of +3 to +6.3 (Fig. 7I). Mississippian to Permian zircon grains (42%) showed a broad age range from 356 ± 8 Ma to 269 ± 4 Ma with a probability age peak ca. 329 Ma and εHf_(t) values of +2.3 to +10.7 (Fig. 7E).

Matrix-supported polymictic conglomerate on Richthofen Island regionally overlies Aksala formation limestone in the Lake Laberge area, ~40 km north of Whitehorse (sample R6 in Fig. 6; Table 1). A sample of green feldspathic lithic arenite matrix yielded 232–186 Ma (97%) zircon grains with an age peak of 210 Ma (Fig. 7B). Late Triassic to Early Jurassic zircon grains had εHf_(t) values that ranged from +3.1 to +13.9 with a main cluster between +4.4 and +7.8 (Fig. 7E). Three Paleozoic zircon grains of 340 ± 7 Ma, 343 ± 10 Ma, and 464 ± 8 Ma yielded εHf_(t) values at +11.8, +12.8, and +8.2, respectively.

Lithic feldspathic arenite in faulted contact with Aksala formation limestone (sample R8 in Fig. 6; Table 1) along the eastern shoreline of Lake Laberge yielded 217–194 Ma (98%) detrital zircon grains with an age peak of 205 Ma (Fig. 7C). Late Triassic to Early Jurassic zircon grains gave εHf_(t) values that clustered between +2.8 and +5.3 and one zircon with a εHf_(t) value of −0.2 (Fig. 7E). Two Middle Mississippian zircon grains at 333 ± 5 Ma and 339 ± 6 Ma yielded εHf_(t) values of +6.3 and +5.4, respectively.

Blue-gray, bioclastic lithic wacke that disconformably overlies Aksala formation limestone (<5 m above contact) north of the Takhini River, ~20 km northwest of Whitehorse (sample R9 in Fig. 6; Table 1), contains Lower to Middle Jurassic Pinna sp. (bivalve) and undetermined coarse-ribbed pectinoid bivalve and ammonite fossils (R. Blodgett, 2018, personal commun.). This sample yielded 216–181 Ma (96%) detrital zircon grains with an age peak of 198 Ma and εHf_(t) values that ranged from +4.9 to +12.1 (Figs. 7D and 7E). Two Pennsylvanian grains at 315 ± 9 Ma and 359 ± 6 Ma were also recognized in this sample.

Lithic wacke with mudstone rip-up clasts was collected along the eastern shoreline of Lake Laberge (sample R10 in Fig. 6; Table 1). This sample mostly yielded 270–185 Ma zircon grains (72%) with age peaks of 185 and 202 Ma (Fig. 7F). Late Triassic to Early Jurassic zircon grains had εHf_(t) values between +0.9 and +2.6 (Fig. 7J). Mid-Permian to Middle Triassic zircon grains at 270 ± 8 Ma, 269 ± 6 Ma, 244 ± 8 Ma, and 244 ± 6 Ma gave εHf_(t) values of −3.9, +5.6, +3.9, and +2.2, respectively. Three Middle Devonian to Early Mississippian (14%) zircon grains yielded 356 ± 4 Ma (εHf_[t] = −5.8), 363 ± 6 Ma (εHf_[t] = +1.2), and 390 ± 10 Ma ages. Three Precambrian (14%) zircon grains were also present at 1474 ± 51 Ma, 2563 ± 26 Ma (εHf_[t] = +0.9), and 2829 ± 23 Ma.

Dark gray-blue, medium-grained, lithic wacke that overlies pebble to cobble limestone conglomerate and Aksala formation limestone (<10 m above contact) near Mount Byng, ~50 km northeast of Whitehorse (sample R12 in Fig. 6; Table 1), yielded 250–175 Ma (67%) detrital zircon grains with an age peak of 193 Ma (Fig. 7G). Late Triassic to Early Jurassic zircon grains gave εHf_(t) values between +1.7 and +12 with two clusters around +4 and +10 (Fig. 7J). Silurian to Permian detrital zircon grains (33%) made up a broad population that ranged from 419 to 257 Ma and yielded age peaks of 251, 254, 289, 301, 316, 328, and 392 Ma (Fig. 7G). Silurian to Permian zircon grains gave εHf_(t) values between −7.6 and +11.8.

Dark-gray to buff, fine-to medium-grained, feldspathic lithic arenite that overlies Aksala formation limestone at Mount Slim (<10 m above contact), ~30 km northeast of Whitehorse (sample R16 in Fig. 6; Table 1), contained 244–178 Ma detrital zircon grains with age peaks of 225, 201, and 185 Ma (Fig. 7H). Mississippian to Permian (59%) zircon grains ranged from 350 ± 4 Ma to 262 ± 5 Ma with probability age peaks at ca. 330 and 310 Ma (Fig. 7H). Middle Triassic to Early Jurassic zircon grains yielded εHf_(t) values of −3.2 to +10.6, and Paleozoic zircon grains gave values that ranged from −2.7 to +10.9 (Fig. 7J). One Paleoproterozoic zircon at 2263 ± 74 Ma yielded a εHf_(t) value of −10.8.

Epiclastic rocks nonconformably overlie 216 Ma gabbro of the Stikine suite (Sack et al., 2020) south of the Alaska Highway, ~20 km northwest of Whitehorse (sample R19 in Fig. 6; Table 1). This basal sample (<1 m above contact) was dominantly composed of 203–172 Ma (96%) detrital zircon grains with age peaks at 190 Ma and 173 Ma (Fig. 7I). Late Triassic to Middle Jurassic detrital zircon grains gave εHf_(t) values that ranged from −4.1 to +5 (Fig. 7J).

Tanglefoot formation strata yielded Hettangian to Toarcian maximum depositional ages (Table 1) that are consistent with Early to Middle Jurassic benthic macrofossils in the Laberge Group (Table S3, see ^{footnote 1}; e.g., Colpron et al., 2007b; Lowey, 2008; Colpron, 2011). Using the YSC and YSP methods, the latter of which was shown by Herriott et al. (2019) to correlate best with corresponding chemical abrasion–isotope dilution thermal ionization mass spectrometry (CA-TIMS) analyses, new (this study) and published (recalculated from Colpron et al., 2015) samples of the basal Tanglefoot formation confirm that subsidence in the northern reaches of the Whitehorse trough began by the Sinemurian (Table 1). Tanglefoot formation strata above the basal Sinemurian succession consistently yielded Pliensbachian depositional ages using the YSP method (Table 1). In the case of sample T3, the YSP estimated age (188.0 ± 0.4 Ma) is most consistent with a stratigraphic position below a Pliensbachian tuff (187.1 ± 0.2 Ma; Colpron and Friedman, 2008) ~17 km to the west along the Robert Campbell Highway (Fig. 4), whereas the YSC estimation (185.9 ± 0.3 Ma) would suggest a position above the dated tuff horizon.

Richthofen formation strata overlie the Upper Triassic Lewes River Group and older rocks at several locations in central Yukon (Lowey, 2008; Colpron et al., 2015). Nineteen new (this study) and published (recalculated from Colpron et al., 2015; Bordet et al., 2019) samples, including nine basal sandstones that overlie Lewes River Group rocks, yielded Late Triassic to Middle Jurassic maximum depositional ages for the Richthofen formation (Table 1). The maximum depositional ages for five of the samples reported in Bordet et al. (2019) were determined using CA-TIMS methods and agree with the YSP estimates in four of our samples (Table 1). The other sample (R4) has a CA-TIMS date that most closely matches the YPP estimate of the maximum depositional age. Three samples with Late Triassic YSP estimates reported in Colpron et al. (2015) and Bordet et al. (2019) have maximum depositional ages that are older than biostratigraphic constraints for the Laberge Group, which suggests recycling of underlying Norian to Rhaetian rocks of the Lewes River Group. Most (11 of 19) Richthofen formation samples have Sinemurian to Pliensbachian age estimates and confirm coeval deposition with Tanglefoot formation units farther north. Four basal sandstone units, including epiclastic rocks west of Whitehorse that sit nonconformably on 216 Ma gabbro, yielded Toarcian to Aalenian YSP estimates, which show that subsidence along the west flank of the Whitehorse trough continued into the Middle Jurassic.

Triassic to Jurassic detrital zircon grains (40%–99% of each sample, X = 79%) are interpreted to have provenance from arc and collision-related Mesozoic igneous rocks. Primary sources include mafic to felsic intrusive suites (and eroded volcanic equivalents) in southern Yukon that are well characterized by zircon CA-TIMS U-Pb, zircon Hf isotope, and whole-rock Nd-Sr-Pb isotope geochemical studies (Sack et al., 2020). For example, Sinemurian to Pliensbachian strata of the Tanglefoot formation along the Robert Campbell Highway (samples T2 and T3) contain Late Triassic to Early Jurassic zircon grains with chondritic to superchondritic Hf isotope compositions (Fig. 5) that are consistent with provenance from ca. 205–194 Ma Minto suite plutons near the northern apex of the Whitehorse trough (Fig. 2; Colpron et al., 2022). A Pliensbachian sandstone along the Robert Campbell Highway (sample T5) has subchondritic Early Jurassic zircon grains consistent with provenance from ca. 188–183 Ma Long Lake suite rocks to the west of the Whitehorse trough, which show a range of εHf_(t) values from +6.4 to −5.8 (Sack et al., 2020; Colpron et al., 2022).

Richthofen formation strata contain Triassic to Early Jurassic detrital zircon grains with dominantly chondritic to superchondritic Hf isotope compositions (Fig. 7) that are consistent with provenance from ca. 217–214 Ma Stikine and ca. 205–194 Ma Minto intrusive suite rocks (Sack et al., 2020). Samples with significant late Sinemurian and younger (≤195 Ma) detrital zircon populations also generally have chondritic to superchondritic Hf isotope compositions that may indicate sources in the ca. 195–184 Ma Lokken suite to the east, which typically has more juvenile εHf_(t) values compared to coeval intrusions of the Long Lake suite (Colpron et al., 2022). An exception is sample R19, an epiclastic unit with a Toarcian to Aalenian maximum depositional age that sits nonconformably on 216 Ma gabbro west of Whitehorse, in which late Sinemurian and younger detrital zircons have subchondritic εHf_(t) values, which are more common in plutons of the Long Lake suite to the west.

Paleozoic detrital zircon grains (<1%–59% of each sample, X = 19%) are interpreted to have provenance from Devonian to Permian igneous rocks of the Yukon-Tanana and Stikinia terranes that form the basement to Late Triassic arc assemblages and host Late Triassic to Early Jurassic intrusive suites in southern Yukon. Late Devonian to Mississippian detrital zircon grains, such as those at Conglomerate Mountain (sample T5; Fig. 2) and east of Lake Laberge (sample R10; Fig. 6), show subchondritic Hf isotope compositions (Fig. 7) that are consistent with primary and recycled sources in the Yukon-Tanana terrane (Finlayson assemblage and related intrusive suites; Murphy et al., 2006; Piercey et al., 2006). Superchondritic Late Mississippian to Pennsylvanian detrital zircon grains in the Tanglefoot formation (Conglomerate Mountain) and Richthofen formation (Mount Laurier, Mount Slim, Mount Byng; Fig. 6) yielded ca. 315 Ma and 330 Ma peaks and suggest sources in the Takhini and Boswell assemblages of Stikinia that crop out along the flanks of the Whitehorse trough (Fig. 2; Hart, 1997; Simard et al., 2003; Bordet et al., 2019).

Precambrian detrital zircon grains were observed in only two samples of the Richthofen formation (R10 and R16). They are interpreted to be recycled from local exposures of pre–Late Devonian metasedimentary successions in the Yukon-Tanana terrane (Snowcap assemblage; Piercey and Colpron, 2009).

The Whitehorse trough has been interpreted as a forearc basin that developed in proximity to the Stikinia arc as a result of subduction of the Cache Creek ocean (part of Panthalassa; Mihalynuk et al., 1994; Dickie and Hein, 1995; English and Johnston, 2005). Colpron et al. (2015) showed that deposition of the Laberge Group was coincident with rapid exhumation of metamorphic rocks and their contained intrusions along the flanks of the trough. They suggested that the Whitehorse trough first developed as a forearc basin that evolved into a syncollisional, piggyback basin in the Early Jurassic during early growth of the northern Cordilleran orogen. Our new U-Pb-Hf isotope data and maximum depositional age calculations for Laberge Group strata, in combination with improved constraints on the magmatic and metamorphic evolution of the Intermontane terranes (Clark, 2017; Sack et al., 2020; Gaidies et al., 2021; Colpron et al., 2022), provide new opportunities to evaluate the filling patterns of the Whitehorse trough.

Figure 8 shows the maximum depositional age estimations of Laberge Group strata for two broad transects along and across the Whitehorse trough in southern Yukon. Nineteen samples of the Richthofen formation, including the YSP ages of nine basal sandstones (Table 1), illustrate that submarine fan and fan-delta deposits record subsidence events that likely correspond to late Rhaetian to early Sinemurian, late Sinemurian to Pliensbachian, and Toarcian regional exhumation. Six samples of Tanglefoot formation strata, including the YSP ages of two basal sandstones (Fig. 8; Table 1), also indicate that the oldest marginal-marine deposits in the northern Whitehorse trough could be linked to early Sinemurian exhumation. Colpron et al. (2015) noted, based on a limited number of samples (n = 8) and using YPP estimates, that maximum depositional ages from basal strata (n = 4) appear to be progressively younger toward the west. They suggested that an early uplift event along the western side of the Whitehorse trough was followed by western transgression of Laberge Group strata during the Sinemurian to Toarcian. To test this hypothesis and better resolve initial subsidence within the trough, half of the new samples reported in this study (n = 6/12) were collected from basal strata of the Laberge Group. Our analysis of maximum depositional ages, using a larger number of samples (n = 25; Table 1) and YSP estimates, confirms that the youngest basal Laberge strata (Toarcian–Aalenian) that sit unconformably on Lewes River Group rocks occur mostly along the western edge of Whitehorse trough exposures, but this data set also illustrates a more complex age pattern near the center of the trough (Fig. 8).

Pliensbachian to Toarcian basal strata of the Richthofen formation that cover Upper Triassic reefal limestone at Grey Mountain (sample R15) and Mount Slim (R16), east of Whitehorse (Fig. 6), have maximum depositional ages that are considerably younger than basal and nonbasal strata to the north, near Lake Laberge (Fig. 8). This suggests that reef complexes of the Aksala formation in the Whitehorse area remained as topographic highs for most of the Early Jurassic or that tectonic exhumation processes in the central Whitehorse trough were focused on the Grey Mountain and Mount Slim region during the Pliensbachian–Toarcian.

Richthofen formation strata collected near Mount Byng (sample R12) along the eastern edge of the trough have Pliensbachian to Toarcian maximum depositional ages (Figs. 6 and 8; Table 1) and suggest that the flanks of the Whitehorse trough were highlands early in basin development, perhaps associated with basin-bounding faults, and they only subsided in the Pliensbachian–Toarcian. The pattern of maximum depositional ages along strike of the trough shows generally younger estimated ages in shallow-marine strata of the Tanglefoot formation in the north, and mostly older estimated ages in more distal strata of the Richthofen formation near Lake Laberge, to the south (Fig. 8).

Colpron et al. (2015) proposed that crustal thickening and exhumation of the Yukon-Tanana terrane in Yukon mark the onset of early Mesozoic orogenic activity in the northern Cordillera. In their model, the northern Intermontane terranes and Whitehorse trough were part of the growing orogen that advanced onto western North America during the Early to Middle Jurassic and formed part of the hinterland-retroarc thrust belt system that supplied sediment to the oldest foreland basin successions of western Canada (Poulton, 1989). Colpron et al. (2015) framed the Late Triassic to Jurassic evolution of the northern Cordillera in the context of the oroclinal enclosure model of Mihalynuk et al. (1994). New evidence regarding the metamorphic history of the Yukon-Tanana terrane (Clark, 2017; Dyer, 2020; Gaidies et al., 2021), documentation of syncollisional, Early Jurassic plutons in southern Yukon (Sack et al., 2020; Colpron et al., 2022), and the development of the latest Triassic to Early Jurassic Hazelton arc in British Columbia (George et al., 2021; Nelson et al., 2022) have led to revised models in which end-on collision of the Stikinia arc with the Yukon-Tanana and Quesnellia terranes in the latest Triassic to Early Jurassic initiated the development of the northern Cordilleran orogen. Colpron et al. (2022) provided a model that proposes the collision zone in Yukon was linked by a sinistral transform fault system to the retreating Hazelton arc to the south (Nelson et al., 2022). Detrital zircon U-Pb-Hf isotope results from synorogenic strata of the Laberge Group (Table 1) combined with statistical assessments (Fig. 9; Table S2, see ^{footnote 1}), biostratigraphic constraints (Table S3; Colpron, 2011), and published tuff ages (Colpron and Friedman, 2008) provide the opportunity to reevaluate the position of the Whitehorse trough in this revised Late Triassic to Jurassic tectonic framework for the northern Cordilleran orogen.

Figure 10 shows a series of maps that illustrate filling patterns and progressive development of the Whitehorse trough during the Early to Middle Jurassic. These maps were constructed using the present distribution of Laberge Group strata in Yukon and depositional age constraints in Table 1 and Table S3. The present structural geometry of the Whitehorse trough is the result of Cretaceous (and younger?) dextral transpression superposed on older Jurassic structures (Colpron, 2011; White et al., 2012; Calvert et al., 2017). Fold and thrust structures in the northern Whitehorse trough are gentle and suggest shortening on the order of 15%–30% (White et al., 2012). However, the amount of transcurrent displacement along intrabasin faults is largely unconstrained, and a rigorous palinspastic restoration of the Whitehorse trough region was not attempted for these maps. Nevertheless, the older histories of some intrabasin faults (e.g., Laurier, Goddard, and Braeburn faults) can still be inferred based on truncation of stratigraphic trends (dashed red lines in Fig. 10; Colpron et al., 2007b; White et al., 2012) and facies changes (Bordet et al., 2019). Figure 11 is a schematic representation of the tectonic evolution of sedimentary basins in the northern Intermontane terranes following the model of Colpron et al. (2022).

The latest Triassic to earliest Jurassic end-on collision of Stikinia (Fig. 11A) was probably triggered by the northwest advance of North America as Pangea began to breakup (Monger and Gibson, 2019; George et al., 2021; Colpron et al., 2015, 2022). Collision-related processes are documented by Early Jurassic tectonic burial and amphibolite-facies metamorphism in the Yukon-Tanana terrane (Johnston et al., 1996; Berman et al., 2007; Clark, 2017; Gaidies et al., 2021). The peraluminous, syncollisional 205–194 Ma plutons of the Minto suite were emplaced at midcrustal depths during deformation (Sack et al., 2020; Colpron et al., 2022); deposition of Laberge Group strata in the Whitehorse trough also began at that time (Figs. 10A and 11B; Colpron et al., 2015).

Crustal thickening was followed by exhumation of the Yukon-Tanana terrane (Figs. 11B and 11C) and its contained syncollisional plutons beginning in the Sinemurian (Clark, 2017; Colpron et al., 2022). The Willow Lake fault (Fig. 2; Colpron and Ryan, 2010) is a northwest-striking extensional fault that provides evidence for crustal-scale, Early Jurassic tectonic exhumation of the Yukon-Tanana terrane during the initial deposition of Laberge Group strata. The hanging wall of the Willow Lake fault consists of undeformed and unmetamorphosed rocks with Mississippian cooling ages (Knight et al., 2013). Polydeformed metamorphic rocks in the footwall of the Willow Lake fault yield ca. 197 Ma and younger ⁴⁰Ar/³⁹Ar biotite and hornblende ages, which, in combination with 205–196 Ma cooling ages for Rhaetian plutons of the region, support rapid Early Jurassic exhumation of basement and collision-related plutons from midcrustal depths (Berman et al., 2007; Knight et al., 2013). For example, early Sinemurian rocks of the Tatchun batholith (Fig. 2) that occur on the northeastern flank of the Whitehorse trough were exhumed during the Early Jurassic at 2.1–7.5 mm/yr (Sack et al., 2020).

Sinemurian subsidence in the Whitehorse trough in Yukon was probably facilitated by sinistral transtension as a transform fault system developed together with the reorganization of subduction after collision (Figs. 10A and 11B; Colpron et al., 2022). The depositional life span (<20 m.y.) and sediment accumulation rates (>0.2 mm/yr, decompacted) recorded by Laberge Group strata support the initiation of the Whitehorse trough as a strike-slip basin fed by drainage systems along rapidly exhuming flanks. In their study of Laberge Group strata in Yukon, Dickie and Hein (1995) proposed that Lower to Middle Jurassic submarine fan deposits are consistent with rapid tectonic subsidence adjacent to basin-bounding, strike-slip faults. Potential candidates for the western boundaries of the Whitehorse trough include the Tally Ho and Llewellyn fault zones near the Yukon–British Columbia border (Fig. 2) where early ductile fabrics are inferred to have developed during Late Triassic to Early Jurassic sinistral strike-slip displacement (Hart and Radloff, 1990; Currie and Parrish, 1993; Mihalynuk et al., 1999). The proto–Big Salmon–Teslin fault system was the eastern boundary of the Whitehorse trough (Colpron et al., 2022) and accommodated early sinistral displacement (de Keijzer et al., 2000). Potential intrabasin faults that may have been involved in the development of the Whitehorse trough include the Braeburn, Goddard, and Laurier faults, which locally bound distinct pre-Jurassic domains (Colpron et al., 2007b; White et al., 2012; Bordet et al., 2019). For example, the Laurier fault juxtaposes different Triassic successions (Joe Mountain formation on the east, Lewes River Group on the west) and is overlapped by Norian to Rhaetian Aksala formation rocks (Fig. 6; Bordet et al., 2019). The oldest maximum depositional age estimates and fossil determinations for Laberge Group strata (Fig. 6; samples R1 to R8 in Fig. 10A) occur mainly in the axial regions of the Whitehorse trough, especially near Lake Laberge, in proximity to the Goddard and Laurier faults.

Early Sinemurian terrestrial to marginal-marine strata of the Tanglefoot formation (samples T1 and T2 in Fig. 10A) yielded 201 Ma age peaks and indicate that Minto suite sources, including the rapidly exhumed Granite Mountain and Tatchun plutons, fed sediment routing systems near the northern apex of the Whitehorse trough. A population of Ladinian to Norian (ca. 242–230 Ma) detrital zircon grains with superchondritic Hf isotope compositions in sample T2 further suggests that point sources in the Joe Mountain and Povoas formations were locally sampled in the Carmacks area. To the south, early Sinemurian submarine fan deposits of the Richthofen formation (e.g., samples R4 to R8 in Fig. 10A) yielded unimodal 210–200 Ma age peaks that are consistent with Minto suite sources to the northwest of the Whitehorse trough. North-to-south, early Sinemurian stratigraphic connections are undetermined in the Whitehorse trough, but sample T1 notably yielded PDP cross-correlation coefficients of 0.71–0.98 (X = 0.88) and 0.66–0.95 (X = 0.70) when compared with Richthofen formation samples R1–R4 (from Colpron et al., 2015; Bordet et al., 2019) and R5–R8 (this study), respectively. Based on the number of grains analyzed (see Saylor and Sundell, 2016), we do not reject the hypothesis that these early Sinemurian or older strata were drawn from the same sources. The MDS plot in Figure 9A further shows that early Sinemurian Tanglefoot formation strata cluster or have similarity with some of the oldest Richthofen formation samples (R3–R5, R7), which further suggests common provenance areas during that time. Most of these Richthofen formation strata crop out around the southern Braeburn, Goddard, and Laurier faults, and these structures may have controlled north-to-south axial drainage patterns (Fig. 10A). Sample T2 is a statistical outlier because of its unique 242–230 Ma population, and it is distanced from Richthofen formation samples in the MDS plot, which implies dissimilarity, but its nearest neighbor (black tie line) is sample R1 (Fig. 9A). Late Devonian to Late Pennsylvanian (360–300 Ma; 329 Ma peak; Fig. 7A) detrital zircon grains in the basal sandstone at Mount Laurier (sample R5 in Fig. 10A) are consistent with Yukon-Tanana provenance (Klinkit assemblage, Piercey et al., 2006), but may also point to contributions from the age-equivalent Boswell and Takhini assemblages of Stikinia near the northeastern and western basin flanks, respectively (Hart, 1997; Colpron, 2011; Bordet et al., 2019; Fig. 2). The detrital zircon U-Pb-Hf isotope results for these Tanglefoot and Richthofen formation strata are interpreted to support the early Sinemurian establishment of longitudinal drainage along the axis of the Whitehorse trough.

Southward migration of the Hazelton arc in the Early Jurassic (Fig. 11B; Nelson et al., 2022) led to continued development of a sinistral transform system that progressively enclosed the northern Cache Creek terrane (Colpron et al., 2022). Sinistral transtension in Yukon promoted continued exhumation of the flanks of the Whitehorse trough and plutons of the Long Lake and Lokken suites that had been intruded at mid-to upper-crustal levels to the west and east of Whitehorse trough, respectively (Fig. 2; Sack et al., 2020). Laberge Group strata accordingly documented subsidence resulting from late Sinemurian to Pliensbachian tectonic exhumation processes and were mostly deposited in the northwestern sector of the basin (Figs. 10B and 11C).

Late Sinemurian to Pliensbachian strata in the Carmacks area yielded 205–185 Ma age peaks from Minto and Long Lake suite sources and generally showed lower εHf_(t) values than those recognized in older strata. For example, early Sinemurian sandstone along the Robert Campbell Highway has Late Triassic to Early Jurassic detrital zircon grains with εHf_(t) values that generally range from 0 to +5, whereas overlying late Sinemurian to Pliensbachian strata (samples T3 and T5 in Fig. 10B) have similar-aged grains with εHf_(t) values of −5 to +5 (Figs. 5C and 5F). These excursions toward subchondritic Hf isotope compositions reflect increasing crustal contributions to Long Lake suite magmatism (Sack et al., 2020; Colpron et al., 2022).

Pliensbachian strata at Conglomerate Mountain in the northern Whitehorse trough (sample T4 in Fig. 10B) crop out near the trace of the Braeburn fault, which may have been part of a western bounding fault system that accommodated the exhumation of Triassic–Jurassic and Carboniferous rock units. Conglomerate Mountain strata have age peaks that correspond to Long Lake suite (187 Ma), Minto suite (205 Ma), and Takhini assemblage (313 Ma) sources from the western flank of the Whitehorse trough (Fig. 5D). Late Sinemurian to Pliensbachian strata of the Richthofen formation along the Braeburn and Goddard faults to the south, such as samples R9 to R14 (Fig. 10B), yield PDP cross-correlation coefficients of 0.32–0.54 (X = 0.40) when compared to Conglomerate Mountain strata, from which we cannot rule out different or similar sources (e.g., Saylor and Sundell, 2016), but these samples do not cluster together in the MDS plots (Fig. 9B). On the other hand, samples R9 to R14 have detrital zircon U-Pb signatures that are like Pliensbachian samples T3 (PDP cross-correlation coefficients of 0.82–0.94, X = 0.89) and T5 (PDP cross-correlation coefficients of 0.54–0.80, X = 0.66) along the Robert Campbell Highway, and these cluster together in the MDS plots (Fig. 9B), which is consistent with longitudinal drainage patterns. The detrital zircon Hf isotope compositions of these Richthofen formation rocks, however, do not show the same subchondritic trends as samples T3 and T5 in the northern Whitehorse trough. For example, basal sandstone units that sit on Lewes River Group strata north of the Takhini River and at Mount Byng (samples R9 and R12 in Fig. 10B) have detrital zircon Hf isotope compositions pointing to age-equivalent, but probably different juvenile sources. Sample R9 on the western flank has superchondritic zircon Hf isotope compositions consistent with sources from the Minto suite (Figs. 7D and 7E), perhaps calling for south-directed drainage along the Braeburn fault system, whereas sample R12 on the east flank has zircon Hf isotope compositions that likely indicate sources in the Lokken suite and Yukon-Tanana terrane to the east (Figs. 7G and 7J). These results taken together support late Sinemurian to Pliensbachian establishment of transverse drainage systems in at least some parts of the Whitehorse trough, including near Conglomerate Mountain in the north and near Mount Byng in the south.

Late Early to Middle Jurassic entrapment of the Cache Creek terrane and imbrication of the Intermontane terranes in the northern Canadian Cordillera (Mihalynuk et al., 2004) probably resulted from continued westward drift of North America as the central Atlantic Ocean began to open (Monger and Gibson, 2019; Colpron et al., 2022). Laberge Group strata that coincided with the emplacement of Bennett suite plutons (Figs. 10C and 11D), prior to thrust imbrication of the Stikinia and Cache Creek terranes, are widespread and indicate subsidence from the basin flanks to central axis during the final stages of Whitehorse trough development. At Grey Mountain near Whitehorse, basal Richthofen formation strata (R15 in Fig. 10C) are probably early Toarcian and sit unconformably on Lewes River Group reefal limestone. These Norian to Rhaetian rocks likely formed block-faulted highlands early in the development of the Whitehorse trough, perhaps inherited from a preexisting Triassic fault system along the central axis or a pop-up block within a sinistral strike-slip system, and only subsided during the late Early Jurassic. Other Tanglefoot and Richthofen formation strata (samples T6 and R16–R19), including a Toarcian basal sandstone unit at Mount Slim along the southern Laurier fault, have Late Triassic to Early Jurassic and Paleozoic age populations that may indicate local provenance rather than well-mixed, longitudinal drainage signatures. Statistical comparisons between samples T6 and R15–R19 yielded PDP cross-correlation coefficients of 0.33–0.85 (X = 0.65), from which we cannot rule out different or similar sources (e.g., Saylor and Sundell, 2016), but the Richthofen formation samples do show sample T6 as a common neighbor in a MDS plot (Fig. 9C). Basal Richthofen formation strata west of Whitehorse nonconformably overlie Stikine suite gabbro (sample R19 in Fig. 10C) and imply that fault-related highlands along the western flank of the Whitehorse trough did not subside until the Toarcian to Aalenian. This hypothesis can be tested by future investigations of Laberge Group strata in the Fish Lake area to the southwest of Whitehorse (Fig. 6), where there is comparatively less fossil and detrital zircon provenance information.

The northern Intermontane terranes and Laberge Group are intruded by stitching 174–168 Ma plutons of the Bryde suite (Sack et al., 2020) and represent the end of Whitehorse trough deposition. This final phase of Laberge Group deposition is recorded by Aalenian to Bajocian strata that are adjacent to the northern Braeburn fault near Carmacks and the ca. 172 Ma Teslin Crossing pluton near the Teslin fault in the east-central Whitehorse trough (Colpron, 2011; Sack et al., 2020). Upper Jurassic to Lower Cretaceous terrestrial strata of the Tantalus Formation unconformably overlie the Laberge Group marine rocks and document surface uplift associated with lithospheric delamination or slab breakoff after final collision of the northern Intermontane terranes (Colpron et al., 2022).

The Triassic to Jurassic tectonic evolution and paleogeography of western North America have long been the subjects of debate, with considerable interest about the timing of Cordilleran terrane accretion and establishment of the hinterland-retroarc thrust belt and foreland basin system (e.g., Murphy et al., 1995; Fuentes et al., 2009; Colpron et al., 2015; LaMaskin et al., 2015; Golding et al., 2016; Beranek et al., 2017; Monger and Gibson, 2019; Pană et al., 2019; Nixon et al., 2020). Crustal thickening and terrane imbrication events in Yukon and British Columbia (e.g., Murphy et al., 1995; Mihalynuk et al., 1999; Nixon et al., 2020) are most consistent with an accreted position for the Intermontane terranes by the Early Jurassic and support the hypothesis that the Whitehorse trough was carried eastward within the nascent Cordilleran orogen (Colpron et al., 2015, 2022). Eastward convergence of the Intermontane terranes with western North America began in the Sinemurian–Pliensbachian (Figs. 10B and 11C; e.g., Nixon et al., 2020). Provenance studies of Fernie Formation strata (see location in Fig. 1), which represent the oldest foreland basin units of western Canada and are coeval with the lower Laberge Group, have recently demonstrated that Sinemurian to Pliensbachian “basal sandstone” rocks yield Late Triassic to Early Jurassic detrital zircon grains (including age peaks of 207 and 192 Ma) sourced from the Intermontane terranes (Pană et al., 2019). Integrated zircon U-Pb and fossil studies have also identified Pliensbachian (188 Ma) bentonitic clay horizons in the lower Fernie Formation and require ash-fall contributions from the Intermontane terranes to the west (Hall et al., 2004), analogous to the ages of the oldest Nordenskiöld member tuffs in the Whitehorse trough (Colpron and Friedman, 2008).

The Early to Middle Jurassic sinistral transform fault system that generated the Whitehorse trough was restricted to the collision zone north of the Stikine arch in the northern Intermontane terranes (Figs. 1 and 11; Colpron et al., 2022; Nelson et al., 2022), but it heralded a profound change in plate dynamics that would later affect outboard regions of the North American Cordillera (Monger and Gibson, 2019). By the late Middle Jurassic, transcurrent regimes accommodated the accretion of the Alexander-Wrangellia-Peninsular composite terrane along the western margins of Stikinia and the Yukon-Tanana terrane (e.g., McClelland and Gehrels, 1990; Gehrels, 2001; McClelland et al., 1992). Middle to Late Jurassic tectonism drove tectonic exhumation and coarse-grained deposition in the Talkeetna forearc and arc regions of southern Alaska (e.g., Trop et al., 2005; Trop and Ridgway, 2007), perhaps along the sinistral Bruin Bay fault system (Betka et al., 2017). The Coast Mountains continental arc system was established along parts of the Insular-Intermontane terrane boundary by the Late Jurassic (van der Heyden, 1992; Gehrels et al., 2009; Beranek et al., 2017). Late Jurassic to Early Cretaceous oblique convergence resulted in a major sinistral transcurrent fault system, which in British Columbia may have accommodated >800 km of displacement, and generation of Oxfordian and younger structural basins at releasing bends, which were in a back-arc position relative to the Coast Mountains arc (e.g., Anderson, 2015).

Crustal thickening events can result in lower-to upper-crustal geochemical contributions to arc and collision-related igneous rocks (e.g., Pearce et al., 1984; Ducea et al., 2015). In Andean-type accretionary systems, the underthrusting of melt-fertile lower crust beneath a continental arc during retroarc deformation results in the partial melting of lithosphere with only minor contributions from asthenospheric mantle (Ducea and Barton, 2007; DeCelles et al., 2009). These crustal thickening events are generally identified by upper-plate igneous rocks with subchondritic radiogenic isotope compositions, which contrast with those generated from superchondritic mantle sources during normal, steady-state subduction. It follows that syntectonic siliciclastic strata and their recycled derivatives yield detrital zircon U-Pb ages and Hf isotope compositions capable of constraining increasing crustal contributions to upper-plate magmatic systems (e.g., Laskowski et al., 2013; Pecha et al., 2016; Pepper et al., 2016), including those that transitioned over time from normal, steady-state subduction to arc collision. Late Triassic to Jurassic detrital zircon grains in Laberge Group strata support this hypothesis and generally show increasing crustal contributions or decreasing εHf_(t) values with time, replicating the isotopic compositions of their igneous rock sources in the northern Intermontane terranes during the south-progressing collision and entrapment of the Cache Creek terrane (Sack et al., 2020; Colpron et al., 2022). We interpret that 205–194 Ma detrital zircon grains with chondritic to subchondritic Hf isotope compositions in the Tanglefoot formation are the result of latest Triassic to Early Jurassic tectonic burial, prograde metamorphism, and midcrustal emplacement of syncollisional Minto suite plutons.

Stikinia is generally considered to have juvenile basement or to have been constructed from depleted mantle–derived arc rocks with little to no crustal contamination (e.g., Logan et al., 2000; Dostal et al., 2009; Bordet et al., 2019). For example, late Paleozoic detrital zircon grains with Takhini assemblage provenance at Conglomerate Mountain (sample T4) yielded εHf_(t) values of +7.8 to +12, which approach the depleted mantle array (Figs. 5D and 5F). However, Colpron et al. (2022) concluded that some Early Jurassic plutons emplaced into northern Stikinia, such as peraluminous Long Lake suite rocks with subchondritic zircon Hf isotope compositions and inherited Paleozoic zircon cores, incorporated old crustal material during arc collision. In northern British Columbia, George et al. (2021) recognized that latest Triassic to Jurassic zircon grains in the Hazelton Group, which were part of the arc system to the south of the active collision zone in Yukon (Figs. 11B–11D), have an average εHf_(t) value of +10, but locally these rocks yield early Paleozoic, Proterozoic, and Neoarchean detrital zircon grains derived from basement highs. The crustal affinity or identity of these isotopically evolved materials is uncertain, but the depleted mantle ages of subchondritic Early Jurassic igneous and detrital zircon grains in southern Yukon suggest that Stikinia could have Proterozoic crystalline basement domains or contain sedimentary assemblages that were derived from Precambrian rocks. For example, sample R19 yielded subchondritic 176–172 Ma detrital zircon grains that plot between the 1.0 and 1.5 Ga crustal evolution lines in Figure 7J. A second option for the origin of this crust is that Laurentian cratonal rocks were shoved beneath Stikinia and the Yukon-Tanana terrane as they were carried eastward over the edge of North America during Early Jurassic retroarc thrusting (Fig. 11C). These findings from northern British Columbia and southern Yukon illustrate that high-n or targeted detrital zircon U-Pb-Hf isotope studies of arc-proximal basins provide new opportunities to understand the architecture and crustal evolution of Stikinia and, by extension, arc terranes in ancient accretionary and collisional orogens.

Detrital zircon U-Pb-Hf isotope studies of Laberge Group strata have constrained the timing of syntectonic deposition in the Whitehorse trough and their significance to northern Cordilleran tectonics and paleogeography. Maximum depositional age estimates for Tanglefoot and Richthofen formation strata demonstrate that tectonic subsidence began by the early Sinemurian following latest Triassic–earliest Jurassic end-on arc collision between northern Stikinia and the Yukon-Tanana terrane. The subsequent southward retreat of the Hazelton arc of Stikinia resulted in the development of a sinistral strike-slip fault system that accommodated regional exhumation of the northern Intermontane terranes, syncollisional magmatism, and subsidence of the Whitehorse trough. Early Sinemurian marginal-marine and submarine fan strata have statistically comparable detrital zircon U-Pb-Hf isotope signatures and support the establishment of longitudinal drainage along the axis of the northern Whitehorse trough by ca. 197 Ma. Intrabasin strike-slip faults may have accommodated axial sediment transport from northern source regions near Carmacks to southern depocenters near Whitehorse. Late Sinemurian to Pliensbachian and Toarcian turbidite and debris-fan strata were mostly derived from transverse drainage systems, and they reveal that the flanks of the Whitehorse trough were fault-related highlands that only subsided by the early Middle Jurassic. Early Jurassic detrital zircon grains from the Long Lake plutonic suite show excursions to subchondritic Hf isotope values through time that represent increased crustal contributions to syncollisional magmatism in the collision zone to the north.

This project was supported by the Yukon Geological Survey and Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grants to Luke Beranek. John Hanchar and Core Research Equipment and Instrument Training (CREAIT) staff members Wanda Aylward, Rebecca Lam, and Markus Wälle aided laboratory studies at Memorial University of Newfoundland. Presubmission feedback on this project from Emily Finzel and Steve Piercey was greatly appreciated. Associate Editor Joel Saylor and journal reviewers Ken Ridgway and Jeff Trop provided constructive and helpful reviews that improved this manuscript. This is Yukon Geological Survey contribution 059.

Zircon crystals were separated from rock samples, handpicked onto double-sided tape, and mounted in epoxy. After polishing to expose the interior of the crystals, cathodoluminescence imaging of the mounts using a JEOL JSM 7100F scanning electron microscope was completed at the Memorial University of Newfoundland. The images were used to locate homogeneous regions of the zircons and to avoid complex internal structures, cracks, and zones of potential Pb loss.

Isotope ratios were measured using the laser-ablation split-stream method at Memorial University of Newfoundland following the procedures of Hutter and Beranek (2020). Zircon crystals were ablated with a GeoLas 193 nm excimer laser-ablation system using a 40 µm spot size, laser fluence of 5 J/cm², pulse rate of 10 Hz, and 600 shots with a total analysis time of ~120 s (~30 s background measurement, ~60 s ablation, and ~30 s washout). Ablated material was evacuated from the ablation chamber via He carrier gas and split using a baffled Y-connector. The Hf isotopes were acquired using a Thermo Finnigan Neptune multicollector–inductively coupled plasma–mass spectrometer (ICP-MS), and U-Pb isotopes were measured using a Thermo Finnigan Element XR magnetic sector ICP-MS. Time-integrated U-Pb signals were analyzed offline using Iolite software (Paton et al., 2010). Age calculations were made using the VizualAge data reduction scheme, which includes a correction routine for down-hole fractionation (Petrus and Kamber, 2012). U-Pb ages were calibrated to the 1065 Ma zircon standard 91500 (Wiedenbeck et al., 1995), and ¹⁷⁶Hf/¹⁷⁷Hf ratios were compared to those of the 337 Ma zircon standard Plešovice (Sláma et al., 2008). Zircon U-Pb analyses with high error (>10% uncertainty) or excessive discordance (>10% discordant, >5% reverse discordant) were excluded from plots and interpretation. The reported ages for grains younger and older than 1200 Ma are based on ²⁰⁶Pb/²³⁸U and ²⁰⁷Pb/²⁰⁶Pb ages, respectively. Age-corrected epsilon Hf (εHf_[t]) calculations used the decay constant of Söderlund et al. (2004) and present-day chondritic uniform reservoir values of Bouvier et al. (2008).