Detrital K-feldspar 40Ar/39Ar thermochronology was conducted on clastic sedimentary rock samples collected from northern exposures of the Upper Cretaceous Nanaimo Group on Vancouver Island and adjacent Gulf Islands of British Columbia to constrain the denudation history of the local Coast Mountains batholith source region and determine the origin of extraregional sediment supplied to the basin. Strata of the northern Nanaimo Group deposited between 86 and 83 Ma (Comox and Extension formations) exhibit a 130–85 Ma age distribution of detrital K-feldspar 40Ar/39Ar ages that lack age maxima. These are interpreted to have been sourced from the southwestern Coast Mountains batholith. Younger strata deposited between 83 and 72 Ma (Cedar District and De Courcy formations) yield a broader age range (150–85 Ma) with an age maximum near the depositional age. These results indicate focused denudation of deeper-seated rocks east of the Harrison Lake fault. The youngest units deposited after 72 Ma (Geoffrey, Spray, and Gabriola formations) primarily yield younger than 75 Ma detrital K-feldspar ages with pronounced age maxima near the depositional age. This sediment was sourced extraregionally relative to the Coast Mountains batholith.

We sought to constrain the origin of the extraregional sediment by measuring the thermal histories of 74 samples of basement rocks from throughout the Pacific Northwest, and by compiling a database of over 2400 biotite 40Ar/39Ar and K/Ar cooling ages from predominantly Cretaceous batholiths along the western North American margin. This analysis focused upon two previously proposed source regions: the Idaho batholith and the Mojave-Salina margin of southern California. The Nanaimo detrital K-feldspar 40Ar/39Ar age distributions favor the peraluminous Late Cretaceous Idaho batholith and its Proterozoic Belt-Purcell Supergroup sedimentary wall rock as the more likely source of the extraregional sediment and disfavor the Baja–British Columbia hypothesis for 2000–4000-km-scale translation of rocks along the margin during the Late Cretaceous.

The Cordilleran margin of North America is a classic region in which to observe ancient convergent margin processes (Fig. 1). Efforts to reconstruct the histories of continental convergent margins invariably feature geologic investigations of preserved sedimentary sequences deposited within associated sedimentary basins (Dickinson and Seely, 1979). These rich stratigraphic records contain time-series data that represent important and unique repositories of detritus produced by denudation and erosion of the volcanic arc (Ingersoll, 1978). Forearc basin studies require integration of stratigraphic, lithologic, paleontologic, structural, geochemical, and geochronologic data to yield robust constraints bearing on the geologic evolution of the margin (Seiders, 1989; Haggart et al., 2006; Garzanti and Ando, 2007; Surpless, 2015).

Over the past two decades, sedimentary provenance analysis based upon measurement of detrital zircon U-Pb crystallization age distributions (e.g., Gehrels and Pecha, 2014) has become a widely employed method for solving a wide range of geologic problems (e.g., Compston et al., 1984; Gehrels, 2000; Jacobson et al., 2011). While the detrital zircon approach to provenance investigations has become common, provenance analysis with detrital thermochronometers has also been demonstrated to yield useful and complementary information (e.g., Brandon and Vance, 1992; Garver et al., 1999; Hodges et al., 2005; Rahl et al., 2007; Bernet et al., 2009; Peyton and Carrapa, 2013).

Detrital K-feldspar is capable of recording thermal history information from long-eroded rocks, and analysis of these data can provide otherwise unattainable information regarding the thermal evolution of the source region (Harrison and Be, 1983; Harrison and Burke, 1988; Copeland and Harrison, 1990). In particular, detrital K-feldspar eroded from continental margin magmatic arcs has proven to be effective for elucidating their denudation history and the resetting effects of overprinting pluton emplacement (e.g., Lovera et al., 1999; Grove et al., 2003). This is because K-feldspar is an important modal phase in granitoids and has sufficiently high retentivity (350–150 °C) to be capable of retaining Ar during sedimentary burial to depths equivalent to <∼150 °C (e.g., Mahon et al., 1998).

Previous detrital zircon studies indicated that the Late Cretaceous–Paleocene Nanaimo basin of southwestern British Columbia (Fig. 1) experienced a major local-to-extraregional shift in sediment composition from a primitive Jurassic to mid-Cretaceous continental margin arc provenance signature to a more cratonal Late Cretaceous arc provenance signature in a region where Paleoproterozoic and Mesoproterozoic rocks were present in abundance (Mahoney et al., 1999, 2021; Matthews et al., 2017; Coutts et al., 2020). The detrital zircon U-Pb age distribution of the extraregional sediment within the Nanaimo basin of southern British Columbia has been interpreted in terms of either comparatively nearby or more distant source regions. Such evidence interpreted to suggest more distant sources has been used as support for the Baja–British Columbia (Baja BC) model of late Mesozoic Cordilleran assembly in western Canada. The Baja BC model proposes, on the basis of paleomagnetic data, that a large crustal block of composite geologic terranes was tectonically displaced by thousands of kilometers along the North American Cordilleran margin during the Late Cretaceous to its present location in western British Columbia (e.g., Beck, 1980; Housen and Beck, 1999; Matthews et al., 2017). However, the large-scale translation of western Cordillera terranes mandated by the Baja BC model contradicts a wide variety of geologic data, which argue for a Middle Jurassic age of amalgamation of the terranes with western British Columbia (see discussions in Mahoney et al., 1999, 2021; Haggart et al., 2006). The persisting controversy indicates the need for new analytical methods generating new data sets to help resolve the contradictions.

This study had two primary objectives: (1) to use detrital K-feldspar 40Ar/39Ar thermochronology data from locally derived sediment from the Coast Mountains batholith to constrain its denudation history; and (2) to apply the same approach to extraregional sediment for use as a provenance tool capable of selecting between possible extraregional source regions based upon commonality of thermal history. This approach was intended to complement previous detrital zircon provenance analysis through combined consideration of crystallization history and thermal history data.

Toward these goals, we measured 40Ar/39Ar age distributions from detrital K-feldspar from Santonian to Paleocene sandstone samples from the northern Nanaimo basin. In addition, we analyzed a suite of representative basement 40Ar/39Ar samples from throughout the Pacific Northwest (the Coast Mountains batholith, North Cascades, Omineca crystalline belt, Idaho batholith, and Boulder batholith) in an effort to build a comparison data set. We also compiled over 2400 biotite K-Ar and 40Ar/39Ar ages from Mesozoic batholiths throughout western North America to characterize the regional distribution of Ar isotopic cooling ages along the continental margin.

Tectonic Setting

Long-lived arc-trench systems were established along the North American Cordilleran margin during the Permian–Triassic following the late Paleozoic Antler and Sonoma orogenies (Dickinson, 2004). Western British Columbia features two major belts of amalgamated rocks, the Intermontane and Insular belts, which were accreted to western North America sometime during the later Mesozoic (Coney et al., 1980; Monger et al., 1982; Armstrong, 1988; Wheeler et al., 1991; Mahoney et al., 2000; Colpron et al., 2007; Monger, 2014). Collectively, these belts extend ∼2000 km from northern Washington State to southern Alaska, and they are up to 350 km wide when measured perpendicular to the margin (Fig. 1).

Intermontane Belt

In southeastern British Columbia, the Intermontane belt (Fig. 1) consists of the Quesnellia, Stikinia, and Cache Creek terranes (Wheeler et al., 1991). Quesnellia and Stikinia are Late Triassic and Early Jurassic intraoceanic arcs, while the intervening Cache Creek terrane is a late Paleozoic–Early Jurassic accretionary complex (Armstrong, 1988; Wheeler et al., 1991). All three terranes were stitched together by Early Jurassic time (Armstrong, 1988), and accretion of the Intermontane belt to North America was complete by the early Middle Jurassic (Ghosh, 1995; Monger and Brown, 2016).

Insular Belt

The Middle Devonian–Early Jurassic portion of the Wrangellia terrane underlies nearly all of the southern Insular belt (Fig. 1; Jones et al., 1977; Monger and Gibson, 2019). On Vancouver Island, Wrangellia consists of Upper Triassic basalt (Karmutsen Formation) and underlying mid-Paleozoic arc and carbonate successions that were affected by later Jurassic magmatism (Monger and Journeay, 1994; Gehrels et al., 2009). Also included in Wrangellia on Vancouver Island are overlying Lower to Middle Jurassic arc volcanic and associated sedimentary rocks, as well as Lower and Upper Cretaceous forearc basin depositional successions. The Cadwallader and Bridge River terranes represent Permian–Middle Jurassic island-arc and accretionary complexes between the Insular and Intermontane belts (Wheeler et al., 1991). The Insular belt was intruded by the Coast Mountains batholith starting in Middle Jurassic time. It is generally agreed that the Insular and Intermontane belts were amalgamated together in Middle to Late Jurassic time (Monger, 2014).

Coast Mountains Batholith

Plutons within the southern Coast Mountains batholith range in age from 180 Ma to 45 Ma (Gehrels et al., 2009; Rusmore et al., 2013; Gibson and Monger, 2014; Cecil et al., 2018). Compositions range from diorite to granite, with tonalite being most abundant (Monger and Brown, 2016). Gibson and Monger (2014) divided the Coast Mountains batholith in the Vancouver area into southwestern and southeastern regions separated by the Harrison Lake fault (Figs. 2 and 3). Plutons southwest of this fault were generally intruded between 167 Ma and 91 Ma and exhibit sub-greenschist- to greenschist-facies, and locally amphibolite-facies, metamorphism. Plutons northeast of the Harrison Lake fault are dominantly 96–84 Ma (Gibson and Monger, 2014) and were affected by upper-amphibolite-facies metamorphism at depths equivalent to 0.5–0.10 GPa near the Harrison Lake fault to sub-greenschist-facies conditions along the Fraser/Straight Creek fault (Fig. 3; Journeay and Friedman, 1993; Brown et al., 2000; Gibson and Monger, 2014). Rapid denudation of rocks east of the Harrison Lake fault occurred as the youngest 86–84 Ma plutons were being emplaced (Journeay and Friedman, 1993; Brown et al., 2000; Gibson and Monger, 2014).

Nanaimo Group Stratigraphy

Upper Cretaceous–Paleocene deposits of the Nanaimo Group unconformably overlie both Wrangellia basement and the Coast Mountains batholith (Fig. 3). This relationship requires a shared Late Cretaceous history across the intervening Strait of Georgia (Muller and Jeletzky, 1970; Mustard, 1994; Katnick and Mustard, 2003). The Nanaimo Group succession comprises Santonian to Paleocene (Danian) siliciclastic strata that dip eastward along eastern Vancouver Island (Fig. 3; Muller and Jeletzky, 1970; England, 1990; Mustard, 1994; Haggart et al., 2005; Ward et al., 2012). Haggart et al. (2005) identified older (Turonian) strata present in the southern area of the Nanaimo basin, but the stratigraphic complexity of these strata precludes their ready assignment to the standard Nanaimo Group succession. Nanaimo Group strata on Vancouver Island and the associated Gulf Islands are locally disrupted by northwest-trending faults and other deformation associated with the Eocene Cowichan fold-and-thrust belt (England and Calon, 1991).

The northern Nanaimo basin is the focus of this paper (Figs. 3 and 4). These strata are represented primarily by exposures on Hornby and Denman islands and along the east coast of Vancouver Island (Fig. 3). The southern Nanaimo basin is exposed in the Gulf Islands along the eastern edge of Vancouver Island (Fig. 3). The northern and southern outcrop belts are remnants of a much larger single basin, with the oldest strata having been deposited in two smaller depocenters separated by a central topographic high, the Nanoose Arch (Fig. 5A). By early Campanian time, deposition of strata was laterally contiguous across the entire basin (Fig. 5B; Mustard, 1994; England and Bustin, 1998).

Deposition of Locally versus Extraregionally Derived Sediment

Santonian to early Campanian strata (ca. 86–83 Ma; absolute ages for stratigraphic data herein follow those of Cohen et al., 2020) of the Nanaimo Group in the northern Nanaimo basin accumulated in a shallow-marine to nonmarine basin (Fig. 5A; Mustard, 1994). These strata include the Comox Formation (Santonian), the Haslam Formation (Santonian–early Campanian), and the Extension Formation (early Campanian) (Figs. 3, 4A, and 4B). Sandstones found within these units contain detrital zircons that yield U-Pb age distributions defined by mid-Cretaceous and Jurassic zircons with maxima occurring at 155, 125, 95, and 85 Ma (Matthews et al., 2017; Mahoney et al., 2021). These age distributions are similar to the values measured within the Coast Mountains batholith in the Vancouver region of mainland British Columbia (Cecil et al., 2018). Both the Jurassic and mid-Cretaceous zircons exhibit primitive initial epsilon Hf isotopic values (εHfi) between +10 and +15 (Mahoney et al., 2021), similar to those measured for the Coast Mountains batholith (Cecil et al., 2011).

After 83 Ma, the Nanaimo basin began to deepen (Fig. 5B; Coutts et al., 2020). Units deposited during this time include the Pender, Protection, Cedar District, De Courcy, and Northumberland formations (Figs. 3, 4C, and 4D). In the northern Nanaimo basin, sediment remained locally derived, consisting of mid-Cretaceous and Jurassic zircons with εHfi values between +10 and +15 (Matthews et al., 2017; Mahoney et al., 2021). The biggest change was a significant increase in Late Cretaceous (100–85 Ma) zircon. The southern Nanaimo basin, however, experienced a completely different source of sediment. Extraregionally derived, muscovite-bearing sediment containing increased amounts of zircon younger than 80 Ma, which is scarce in the Coast Mountains batholith of the Vancouver region (Cecil et al., 2018), began to be delivered to the basin (Fig. 5B; Coutts et al., 2020; Mahoney et al., 2021). This younger zircon had significantly lower εHfi (+5 to −30) values and was accompanied by abundant (20%–30% or more) Paleoproterozoic (ca. 1.7 Ga) and Mesoproterozoic (ca. 1.4 Ga) zircon (Mahoney et al., 2021).

Subsidence within the Nanaimo basin further accelerated after ca. 72 Ma (Fig. 5C; Mustard, 1994; Coutts et al., 2020). Within the northern Nanaimo basin, the latest Campanian–early Maastrichtian Geoffrey Formation, the Maastrichtian Spray Formation, and the Maastrichtian–earliest Paleocene Gabriola Formation were deposited (Figs. 3, 4E, and 4F). Deepening of the basin led to deep-marine conditions and basinwide development of a series of gravel-filled, submarine channel-system deposits (Mustard, 1994; Coutts et al., 2020). At this time, the entire Nanaimo basin received extraregionally derived sediment (Matthews et al., 2017; Coutts et al., 2020; Mahoney et al., 2021). The high proportions of zircons with ages younger than 80 Ma (20%–30%) and with ages of 1.4 and 1.7 Ga in these strata indicate that the sediment is extraregional with respect to the local Coast Mountains batholith and Wrangellian basement.

Paleomagnetic Evidence for Large Displacements along the Late Cretaceous Margin

Paleomagnetic evidence suggests that elements of the Insular and Intermontane belts were translated northwards on the order of 3000 ± 1000 km during the Jurassic and Late Cretaceous (Beck and Noson, 1972; Coney et al., 1980; Beck, 1980; Irving et al., 1985, 1996; Ward et al., 1997). This interpretation of the paleomagnetic data led to the Baja BC hypothesis (e.g., Umhoefer, 1987).

Paleomagnetic analyses of strata of the Nanaimo basin have provided data bearing on the Baja BC hypothesis (Ward et al., 1997; Enkin et al., 2001; Kim and Kodama, 2004; Kodama, 2012), and these data have also implied that significant (>1500 km) offsets of the basin relative to western North America must have occurred in Late Cretaceous time. However, paleogeographic reconstructions of terrane interactions along readily recognized dextral strike-slip faults of the northern Cordillera can account for hundreds, but not the required thousands, of kilometers of offset required by the Baja BC model (Price and Carmichael, 1986; Umhoefer and Miller, 1996; Umhoefer and Schiarizza, 1996; Wyld et al., 2006).

Relationship to the Source of the Extraregional Sediment in the Nanaimo Basin

Several previous studies have demonstrated that the extraregional young (<80 Ma) zircon within the Nanaimo basin has a Late Cretaceous arc provenance that is isotopically evolved due to assimilation of Paleoproterozoic and Mesoproterozoic crust (Mahoney et al., 1999, 2021; Matthews et al., 2017). These studies have identified two potential source regions with these characteristics: (1) the Mojave-Salinia margin of southern California, and (2) the Idaho and Boulder batholith within the Pacific Northwest (Fig. 1).

The first of these two potential source regions for Nanaimo basin sediment, the Mojave-Salina margin of southern California, is separated from the present-day Nanaimo basin by ∼2000 km, a distance compatible with that proposed for offset of Wrangellia by the Baja BC hypothesis. The Mojave province of southern California features Paleoproterozoic (1.7 Ga) and Mesoproterozoic basement (1.4 Ga and 1.2–1.1 Ga), as well as Permian–Triassic, Jurassic, and Late Cretaceous (80–68 Ma) magmatism (Anderson and Bender, 1989; Wooden and Miller, 1990; Gleason et al., 1994; Barth et al., 2001; Coleman et al., 2002; Barth and Wooden, 2006; Wooden et al., 2013; Cecil et al., 2019). Mojave-Salinian granitoids are isotopically evolved relative to the western domains of the Peninsular Range, Sierra Nevada, and Coast Mountains batholiths (Kistler and Peterman, 1978; Kistler and Champion, 2001; Cecil et al., 2011; Wooden et al., 2013).

Based upon detrital zircon U-Pb age distributions identified in Nanaimo basin strata, Matthews et al. (2017) favored the Mojave-Salinian margin as the extraregional source for these sediments. These workers specifically considered subduction-related schists related to Laramide shallow subduction as correlative with the extraregional sediment that was transported from source to the Nanaimo basin. In the Mojave-Salinia region, these schists, referred to as the Pelona-Orocopia-Rand-Salinas schists (Grove et al., 2003), have the same time-dependent provenance signature as coeval sediment that was supplied to the southern California forearc margin (Jacobson et al., 2011). Sauer et al. (2019) measured the detrital zircon U-Pb and Hf isotopic properties of a suite of Pelona-Orocopia-Rand-Salinas schist samples from the Mojave-Salinian margin and concluded that the schists' detrital zircon U-Pb ages and Hf isotopic properties were indistinguishable from equivalent schists (Swakane Gneiss) that were underthrust beneath the North Cascades arc during the Late Cretaceous. Sauer et al. (2019) further correlated sediment entrained within the western mélange belt beneath the western Cascades thrust system (Sauer et al., 2017) with both the Swakane Gneiss and the extraregional sediment contained within the Nanaimo basin. Garver and Davidson (2015) independently linked sediment from the Mojave-Salinian margin to the extraregional sediment within the Nanaimo basin and further proposed that equivalent sediment was transported from the latitude of the Nanaimo basin to as far north as the Chugach–Prince William region of Alaska via northwest transport of the Yakutat terrane.

The second potential source of extraregional sediment for the Nanaimo basin is the Idaho batholith region (Mahoney et al., 1999, 2021). Prior to the ∼250 km of dextral Eocene slip along the Yalakom and Fraser/Straight Creek faults, the southern Coast Mountains batholith was situated outboard of the North Cascades, ∼400 km northwest of the Idaho batholith (Umhoefer and Miller, 1996; Umhoefer and Schiarizza, 1996). Restoration of subsequent dextral slip along other major faults in the Canadian Cordillera would further decrease this distance (Price and Carmichael, 1986). For example, restoration of fault displacement performed by Wyld et al. (2006) translated the Nanaimo basin and southern end of the Coast Mountains batholith ∼900 km to the south at 100 Ma, in a position outboard of the Klamath Mountains and 750 km southwest of the Idaho batholith.

The Idaho batholith is also isotopically evolved (Fleck and Criss, 2007), yields predominately 80–65 Ma zircon crystallization ages (Gaschnig et al., 2011, 2017), and is intruded into metasedimentary rocks of the Mesoproterozoic Belt-Purcell Supergroup that contain 1.7 Ga and 1.4 Ga zircon (Link et al., 2006; Lewis et al., 2010). Several river systems are envisioned to have flowed from the Idaho region to the Pacific margin and/or the continental interior during Late Cretaceous time, readily transporting material eroded from the batholith to the coastal basins (Heller et al., 1987; Renne et al., 1990; Mahoney et al., 1999, 2021; Janecke et al., 2000; Dumitru et al., 2013, 2016; Dumitru, 2019).

The occurrence of 1.38 Ga zircon within the Lemhi subbasin of the Belt-Purcell Supergroup has also been considered to be relevant to models of sediment source and transport (e.g., Dumitru, 2019). The Lemhi subbasin of the Belt-Purcell Supergroup was intruded by distinctive 1.38 Ga plutons and sills during the Mesoproterozoic (Doughty and Chamberlain, 1996), and again during the Late Cretaceous by the Idaho and Boulder batholiths (Price and Sears, 2001). Dumitru (2019) has argued that sediment eroded from the Lemhi subbasin is the source of the distinctive 1.38 Ga zircon found in the Nanaimo basin, the western mélange belt, and the Swakane Gneiss. A comprehensive study of Mesoproterozoic anorogenic granites by du Bray et al. (2015) indicated that the Lemhi area is the only known source of 1.38 Ga zircon in the Pacific Northwest. Conversely, in case of the Mojave region, plutons older than 1.39 Ga are far more abundant than plutons younger than 1.39 Ga (du Bray et al., 2015).

Seven sandstone samples from Nanaimo Group strata were collected from the northern Nanaimo basin (Table 1; Fig. 3). A detailed description of these seven Nanaimo Group samples is presented in Table S11. The Comox Formation sample (179JBM04) and the Extension Formation sample (180JBM04) were both collected on Vancouver Island, while the samples representing the Cedar District (186JBM04) and De Courcy (187JBM04) formations were collected from the southwest corner and west-central part of Denman Island, respectively. Samples representing the Geoffrey Formation (181JBM04), the Spray Formation (183JBM04), and the Gabriola Formation (185JBM04) were all collected from Hornby Island, on the northwestern shore, near Dunlop Point, and on the north side of Tribune Bay, respectively (Fig. 3; Table 1). Note that the Spray Formation sample was collected at the contact with the Geoffrey Formation. Samples were processed using standard crushing, sieving, magnetic, and density separation methods (see Table S2). The 40Ar/39Ar CO2 laser fusion and diode laser incremental-heating methods are summarized in Table S3.

In total, 847 detrital K-feldspars were analyzed by CO2 laser fusion 40Ar/39Ar methods (Table 1; Table S4). To help facilitate interpretation of the detrital K-Ar feldspar results, a comparative suite of 74 representative basement samples was collected from the adjacent Coast Mountains batholith and other Jurassic–Cretaceous batholithic basement exposures throughout the Pacific Northwest (Fig. 2), including the North Cascades, Intermontane terrane, Omineca crystalline belt, Idaho batholith, and Boulder batholith (Table 2; Tables S5 and S6; Fig. S1). K-feldspar, biotite, and muscovite were extracted from these samples, where each mineral was available.

Detrital K-Feldspar 40Ar/39Ar Age Distributions

Figures 6 and 7 show the detrital K-feldspar 40Ar/39Ar cooling age distributions of our samples. We have included detrital zircon U-Pb crystallization age distributions from stratigraphically equivalent samples from Coutts et al. (2020) and Mahoney et al. (2021) on Figure 7 for comparison. The stratigraphic subdivision of Coutts et al. (2020) agrees well with important differences in the nature of the age distributions exhibited by our samples.

Comox and Extension Formations

The Comox Formation and Extension Formation samples were deposited during Santonian to earliest Campanian time, between 86 and 83 Ma. Probability density functions of the detrital K-feldspar 40Ar/39Ar age distributions acquired from the Comox Formation and Extension Formation samples are shown in Figures 6G and 6F, respectively. Cumulative age distributions for the two samples are provided in Figures 7G and 7F, respectively. Note that the cumulative age plots in Figure 7 include Mahoney et al.'s (2021) detrital zircon U-Pb age distributions measured from the same Comox Formation and Extension Formation samples analyzed in this study. Also included are Matthews et al.'s (2017) results for Comox Formation and Extension Formation samples from the northern Nanaimo basin (see also comment by Haggart et al., 2018).

The Comox Formation and Extension Formation samples both yielded relatively uniform distributions of detrital K-feldspar 40Ar/39Ar ages with no age maxima (Figs. 6 and 7; Table 1). In the case of the Comox Formation, 69% (83 of 120) of the measurements passed filters for plagioclase contamination and problematic artifacts. Grains ranged from 260 to 613 µm in diameter with a mean diameter of 414 µm. The age distribution of Comox Formation detrital K-feldspars defines a uniformly varying distribution of ages between 147 Ma and 85 Ma; all but two of the results fall between 124 and 85 Ma. The youngest detrital K-feldspar 40Ar/39Ar age is 84.6 ± 0.1 Ma, within the Santonian (Table S4). Detrital zircon U-Pb ages acquired by Mahoney et al. (2021) for the same sample defined a much broader span (218–86 Ma). The maximum bound on the depositional age (MDA) calculated from the detrital zircon age distribution (86.8 ± 1.6 Ma) can be statistically resolved from the youngest detrital K-feldspar age (84.6 ± 0.1 Ma) at the 95% confidence level.

The Extension Formation sample was quite fine grained and heavily contaminated with albite-rich plagioclase. Only 51 of 119 or 43% of the detrital feldspar analyses passed filter tests for plagioclase contamination and problematic artifacts (Table 1). The accepted grains ranged from 113 to 395 µm in diameter with a mean diameter of 273 µm. A significant fraction of the rejected grains was composed of grains less than 100 µm in diameter, and all but one of the 51 accepted grains defined an age range between 128 Ma and 87 Ma. No meaningful age maxima were evident (Figs. 6D and 7D). The youngest detrital K-feldspar 40Ar/39Ar age for the Extension Formation (87.4 ± 0.3 Ma; Table S4) is consistent with an early Campanian depositional age. Note that detrital zircon U-Pb ages acquired by Mahoney et al. (2021) for the same sample defined a much broader range of ages (193–87 Ma). The youngest detrital K-feldspar age (87.4 ± 0.3 Ma; Table S4) and the detrital zircon MDA (88.1 ± 2.0) ages agree within 95% confidence.

Cedar District and De Courcy Formations

The detrital K-feldspar age distribution of the Cedar District Formation sample is based upon 65% (77 of 119) of the grains analyzed (Table 1). Analyses that passed the filtering process ranged from 378 to 752 µm in diameter with a mean diameter of 589 µm. Most of the excluded analyses were from albite-rich plagioclase. The Cedar District Formation sample yielded a somewhat broader span of detrital K-feldspar ages (150–84 Ma) than either the Comox Formation or Extension Formation samples (Figs. 6E and 7E). However, it differs from the older samples in that 64% of the results in the distribution defined an age maximum between 97 and 84 Ma, near the depositional age. This age maximum was also expressed in the detrital zircon U-Pb age distribution measured from the Cedar District Formation sample by Mahoney et al. (2021). Another significant issue pertaining to the Cedar District Formation sample is that the youngest detrital K-feldspar 40Ar/39Ar age (84.4 ± 0.1 Ma; Table S4) is actually older and statistically resolved from the 77.7 ± 1.4 Ma MDA calculated from the detrital zircon results.

After filtering the data for plagioclase contamination and problematic artifacts, we accepted 78% (93 of 120) of the 40Ar/39Ar analyses of the De Courcy Formation's detrital K-feldspar (Table 1). Analyses that passed the filtering process ranged from 130 to 643 µm in diameter with a mean diameter of 495 µm. The De Courcy Formation yielded a broader age (150–82 Ma) than samples deposited prior to 83 Ma (Figs. 6D and 7D). It also exhibited an age maximum between 94 and 82 Ma that included 51% of the analyses. The overall form of the detrital K-feldspar age distribution mirrors that of the detrital zircon age distribution measured by Mahoney et al. (2021) for the same sample (Fig. 7D). Like the Cedar District Formation sample, the youngest detrital K-feldspar 40Ar/39Ar age (81.7 ± 0.1 Ma; Table S4) is actually older and statistically resolved from the 75.1 ± 1.6 Ma MDA calculated from the detrital zircon results.

Geoffrey, Spray, and Gabriola Formations

The Geoffrey Formation, Spray Formation, and Gabriola Formation samples were all deposited in latest Campanian time or later, approximately younger than 72 Ma. The Geoffrey Formation sample was coarse grained and rich in K-feldspar. Most (110 of 120, or 92%) of the 40Ar/39Ar measurements performed passed the filtering process (Table 1). These grains ranged from 320 to 720 µm in diameter with a mean diameter of 543 µm. All of the measured 40Ar/39Ar ages were Late Cretaceous (Figs. 6C and 7C): Over 95% of the analysis clustered in a range between 82 and 72 Ma. This age distribution is distinct from samples deposited prior to 72 Ma in that most of the 40Ar/39Ar ages in the older samples were older than 82 Ma. The youngest detrital K-feldspar age measured here (71.6 ± 0.1 Ma; Table S4) nearly overlaps at 95% confidence with the MDA calculated from Mahoney et al.'s (2021) detrital zircon U-Pb age results from the same Geoffrey Formation sample (69.3 ± 1.4 Ma).

Like the underlying Geoffrey Formation sample, the Spray Formation sample, collected at its contact with the Geoffrey Formation, was coarse grained and rich in K-feldspar. Only 8 of the 120 40Ar/39Ar measurements (7%) were rejected by the filtering process (Table 1). Accepted grains ranged from 280 to 726 µm in diameter with a mean diameter of 515 µm. Two distinct clusters of 40Ar/39Ar ages occur in the Spray Formation sample distribution (Figs. 6B and 7B). The older 82–72 Ma cluster accounts for 68% of the results and corresponds to a similar cluster in the Geoffrey Formation sample; the younger subordinate cluster (68–65 Ma) accounts for 15% of the distribution. The remaining K-feldspar ages define a tail that consists of older Late Cretaceous and a smattering of Early Cretaceous 40Ar/39Ar ages (Fig. 6B). The youngest detrital K-feldspar age (64.7 ± 0.1 Ma; Table S4) is considerably younger than the 70.3 ± 1.3 Ma MDA calculated from detrital zircon results from the same sample (Mahoney et al., 2021).

Nearly all (107 of 119, 90%) detrital K-feldspar 40Ar/39Ar analyses from the Gabriola Formation sample passed the screening process (Table 1). Grains ranged in diameter from 363 to 805 µm with a mean diameter of 588 µm. Like the Spray Formation sample, the Gabriola Formation sample exhibits two clusters of ages within its distribution (Figs. 6A and 7A). The older 80–69 Ma cluster accounts for 54% of the detrital K-feldspar, while the younger 68–63 Ma cluster represents 32% percent of the 40Ar/39Ar analyses. Most of the remaining grains yielded Late Cretaceous ages. The youngest (63.3 ± 0.1 Ma; Table S4) grain is also considerably younger than the MDA (72.3 ± 1.5 Ma) reported for the sample by Mahoney et al. (2021).

Regionally Distributed Basement Samples

K-feldspar, biotite, and muscovite 40Ar/39Ar thermochronologic data measured from various Jurassic and Cretaceous batholiths of the Pacific Northwest (Fig. 2) are summarized in Table 2. The ages reported in Table 2 are total gas ages calculated from 40Ar/39Ar diode laser incremental-heating experiments consisting of 15–25 steps. Because the results were acquired over a broad region, a detailed assessment of these new basement results is beyond the scope of this paper. Additional information regarding these samples and plots of their 40Ar/39Ar age spectra are provide in Table S8 and Figure S1 of the Supplemental Material (footnote 1).

The most direct approach for evaluating potential source regions involves direct comparison of detrital K-feldspar 40Ar/39Ar age distributions with basement K-feldspar 40Ar/39Ar ages measured from source regions. Unfortunately, K-feldspar 40Ar/39Ar results are not widely available for many basement source regions relevant to the Nanaimo basin. This is primarily due the focus of early studies upon hornblende and biotite as well as the more intensive labor required for K-feldspar thermochronology. In addition, common plutonic rocks such as tonalite and gabbro contain little to no K-feldspar.

Because potential source regions for the Nanaimo basin extend well beyond the local region that we sampled for K-feldspar, we produced a database of biotite K-Ar and 40Ar/39Ar ages from the Cretaceous magmatic arc rocks distributed along western North America (Table S7). Figure 8 shows the variation of zircon U-Pb crystallization ages and Ar isotopic ages from biotite and K-feldspar from strike-perpendicular traverses across the southern Coast Mountains batholith (Figs. 8A and 8D), Sierra Nevada (Figs. 8B and E), and Peninsular Ranges batholith (Figs. 8C and 8F).

As indicated in Figures 8E and 8F, the use of biotite K-Ar and 40Ar/39Ar ages as a proxy for K-feldspar 40Ar/39Ar ages is well justified for the Sierra Nevada and Peninsular Ranges. This is because: (1) biotite (375–300 °C; Harrison et al., 1985; Grove and Harrison, 1996) and K-feldspar (350–150 °C; Lovera et al., 1997, 2002) have overlapping 40Ar closure temperature ranges; and (2) reasonably fast monotonic cooling governed Ar closure of biotite and K-feldspar after intrusion by 85 Ma in both the Sierra Nevada (Nadin et al., 2016) and Peninsular Ranges (Ortega-Rivera, 2003; Grove et al., 2003) (Figs. 8E and 8F).

The situation for the Pacific Northwest is more complicated than it is for the California batholiths due to overprinting of older Jurassic and Cretaceous plutons by Eocene intrusions (Fig. 8D; Armstrong, 1988; Haugerud et al., 1991; Paterson et al., 2004; Gehrels et al., 2009; Miller et al., 2016; Eddy et al., 2016; Cecil et al., 2018). The impact of contact heating produced by the Eocene plutons is exemplified by 40Ar/39Ar results obtained from the Cloudburst pluton within the Coast Mountains batholith (Fig. 9; see location in Fig. 8A). The Cloudburst pluton was emplaced at ca. 147 Ma but yielded 40Ar/39Ar ages for biotite and K-feldspar of 94 Ma and 57 Ma, respectively (Fig. 9). Given the intrusive history of adjacent plutons, it is probable that both biotite and K-feldspar 40Ar/39Ar ages have been reset by repeated pluton emplacement, first during the Late Cretaceous and again in the Eocene.

With the caveat that K-feldspar is more prone to resetting of its 40Ar systematics than is biotite during pluton-related reheating events, we present a contour plot of biotite K-Ar and 40Ar/39Ar ages from Cretaceous batholiths distributed along the western margin of North America as a tool to evaluate potential source regions (Fig. 10). The data and sources used to construct Figure 10 are provided in Tables S7 and S8 (see footnote 1), respectively. Because we were particularly interested in the spatial variation of Late Cretaceous cooling ages in both the Pacific Northwest and in southern California, we also provide enlarged plots of these regions (Figs. 11 and 12).

Areas within Figures 1012 that are colored with hues of blue and light green represent biotite K/Ar and 40Ar/39Ar ages older than 75 Ma. These are diagnostic of detrital K-feldspar age distributions yielded by northern Nanaimo basin sediments deposited prior to 72 Ma. Alternatively, the orange and yellow shaded regions indicate biotite K/Ar and 40Ar/39Ar ages of 75–63 Ma. These match the detrital K-feldspar 40Ar/39Ar ages of northern Nanaimo basin strata that were deposited subsequent to 72 Ma. All regions shaded in red yielded biotite cooling ages that postdate deposition of the Nanaimo Group and generally represent areas affected by Eocene pluton emplacement.

Interpretation of detrital K-feldspar 40Ar/39Ar ages in terms of the thermal history of the basement source region (e.g., Lovera et al., 1999) requires that postdepositional 40Ar loss during sedimentary burial was minimal. We evaluated the possibility that postdepositional 40Ar loss adversely affected the detrital K-feldspar 40Ar/39Ar ages measured in this study.

Burial History

The total thickness of the Nanaimo Group in the northern Nanaimo basin is ∼3 km (Muller and Jeletzky, 1970; Mustard, 1994; Mahoney et al., 2021). England (1990) constructed a series of reflectance-depth plots throughout the Nanaimo basin that suggest at least 1.3–2.0 km (locally >4 km) of strata that were initially overlying the Nanaimo Group were subsequently removed by postdepositional erosion. We infer that these strata were correlative to Paleocene–Eocene nonmarine strata of the Whatcom basin, which are widely exposed to the southeast in northwest Washington State. These strata unconformably overlie the Nanaimo Group, and we estimate their thickness in the northern Nanaimo basin to have been ∼1 km. Available information, including structural and stratigraphic reconstructions, vitrinite reflectance data, and apatite fission-track data, indicates that maximum Eocene burial conditions of basal Nanaimo Group strata in the northern basin were ∼4 km and 130–140 °C (Kenyon and Bickford, 1990; England et al., 1997). Fission-track and vitrinite reflectance data indicate that the Nanaimo basin remained at near-maximum temperatures from 60 to 40 Ma, until basin inversion began at ca. 40 Ma (England et al., 1997).

Anticipated 40Ar Loss in the Stratigraphically Lowest Nanaimo Group Formations

We approximated the burial history of the Comox and Extension formations within the northern Nanaimo basin with a Gaussian curve centered upon 50 Ma. Maximum temperatures were maintained between ca. 60 and 40 Ma (Fig. 13; England, 1989; England et al., 1997). An ∼25 °C/km geotherm is indicated (England et al., 1997; Farley et al., 2001). The multidomain diffusion model (e.g., Lovera, 1992) was used to calculate the impact of the burial histories shown in Figures 13A and 13B upon detrital K-feldspar 40Ar/39Ar ages. The diffusion properties of a representative K-feldspar were employed (see N13 in Lovera et al., 2002). Similar calculations used plagioclase diffusion properties from Cassata et al. (2009).

Figure 13 shows the extent of postdepositional 40Ar loss expected for detrital K-feldspar due to burial heating. The blue, cyan, and green curves in Figures 13A and 13B are most relevant because they cover the range of likely burial histories experienced by the most deeply buried strata of the Nanaimo Group (i.e., the Comox and Extension formations). Detrital K-feldspar that experienced maximum burial temperatures of 120, 145, and 170 °C (4–6 km depth) yielded fractional loss (f) of 0.2%, 1.0%, and 4.1%, respectively (Figs. 13C13D). Although available data indicate that maximum burial of 4–5 km (125–145 °C) is most probable, the anomalously young 40Ar/39Ar ages we measured from detrital plagioclase contaminating our samples are an indication that the burial temperatures could have exceeded 145 °C. Plagioclase 40Ar retentivity is generally somewhat less than that of K-feldspar (Cassata et al., 2009). Our plagioclase calculations indicate fractional loss of 0.7%, 3.1%, and 11% for maximum burial temperatures of 120, 145, and 170 °C, respectively (Fig. 13D). Because some plagioclase grains yielded 40Ar/39Ar ages that were younger than the depositional age of these samples, it is possible that temperatures were as high as 170 °C. A peak temperature of 170 °C would produce up to ∼4% postdepositional 40Ar loss in K-feldspar (Fig. 13D). Shallower strata would have experienced cooler conditions and minimal 40Ar loss.

Denudation History of the Local Coast Mountains Batholith Source Region

The Nanaimo Group is in depositional contact with the southern Coast Mountains batholith and Wrangellian basement (Fig. 3), and, prior to ca. 72 Ma, the northern Nanaimo basin received locally derived sediment from each of these sources (Figs. 3 and 5A; Mustard, 1994; Monger, 2014; Matthews et al., 2017; Coutts et al., 2020; Mahoney et al., 2021). Given the dominance of Karmutsen Formation basalts in Wrangellia and the subordinate number of intruding Jurassic plutons, most detrital zircon and K-feldspar present within the Nanaimo Group can be reasonably assumed to have been eroded from Coast Mountains batholith granitic basement.

Jurassic plutons are most abundant in the southwestern Coast Mountains batholith (Fig. 8; Monger, 2014; Cecil et al., 2018). Detrital zircon U-Pb age distributions reported by Matthews et al. (2017) and Mahoney et al. (2021) for the Comox and Extension formations in the northern Nanaimo basin yielded a high percentage (43%) of Jurassic U-Pb detrital zircon ages. This indicates that sediment supplied to these units was preferentially derived from the southwestern part of the Coast Mountains batholith (Figs. 8A and 8D; Cecil et al., 2018).

Detrital K-feldspars from the Comox and Extension formations defined a relatively continuous distribution of 40Ar/39Ar ages between 125 and 85 Ma (Figs. 6F, 6G, 7F, and 7G). Only three Jurassic 40Ar/39Ar detrital K-feldspar ages were measured. Given reasonable arc geotherms of ∼40 °C/km (Barton and Hanson, 1989; Rothstein and Manning, 2003), erosion depths of ∼8–9 km (340–380 °C) are required to fully erase radiogenic 40Ar accumulated by the Jurassic plutons and produce a smoothly varying K-feldspar 40Ar/39Ar age distribution from an arc terrane with a protracted emplacement history (Lovera et al., 1999; Grove et al., 2003). These temperatures are comparable to those expected for the observed sub-greenschist- to greenschist-facies metamorphism that affected the Coast Mountains batholith southwest of the Harrison Lake fault (Journeay and Friedman, 1993; Gibson and Monger, 2014).

Subsequent to 83 Ma, the northern Nanaimo basin began to receive a higher proportion of zircon with 100–80 Ma U-Pb crystallization ages (Fig. 7). The Cedar District and De Courcy formations are similarly enriched in 100–80 Ma detrital K-feldspar 40Ar/39Ar ages (Figs. 6D, 6E, 7D, and 7E). The coincidence of 100–80 Ma age maxima from both the zircon U-Pb and K-feldspar 40Ar/39Ar systems indicates that the locus of erosion within the Coast Mountains batholith became localized east of the Harrison Lake fault, where 100–80 Ma plutons are abundant (Figs. 8A and 8D; Cecil et al., 2018). Denudation of the hanging wall east of the Harrison Lake fault was most rapid between 86 and 84 Ma (Brown et al., 2000; Gibson and Monger, 2014). Figure 11 confirms that 105–85 Ma cooling ages predominate west of the Harrison Lake fault, while deeper rocks east of the fault record 85–68 Ma cooling ages. The fact that plutons east of the fault are predominantly tonalites poor in K-feldspar may explain the paucity of 82–76 Ma detrital K-feldspar ages (relative to zircon) in the age distributions of the De Courcy Formation, and to a lesser extent in the Cedar District Formation.

Sediment supplied to the Cedar District and De Courcy formations may also have been transported from locations further east within the Intermontane belt. Although restoration of Eocene displacements along the Fraser/Straight Creek and Yalakom faults needs to be considered, Figure 11 shows that Intermontane belt rocks tend to have older Ar isotopic ages (>105 Ma) than the rocks of the Coast Mountains batholith. Figure 6 shows that the Cedar District and De Courcy formations yielded a higher percentage of detrital K-feldspar cooling ages older than 105 Ma than did the older Comox Formation and Extension Formation samples. The appearance of these older grains in the Cedar District and De Courcy formations indicates that an overall eastern shift in erosional denudation took place post–83 Ma. This is consistent with Late Cretaceous inversion of the Methow basin (Surpless et al., 2003). It is further probable that subsidence along the western margin of the Coast Mountains batholith occurred synchronously with subsidence within the Nanaimo basin.

Evaluation of Potential Extraregional Sediment Sources

Post–72 Ma sediment contained within the Geoffrey, Spray, and Gabriola formations is clearly extraregional because it is dominated by 40Ar/39Ar detrital K-feldspar ages younger than 80 Ma that cannot be reconciled with adjacent Coast Mountains batholith basement (Figs. 7A and 7B). In addition, the post–72 Ma samples contain muscovite and cratonally derived Paleoproterozoic and Mesoproterozoic (1.7 and 1.4 Ga) zircons (Fig. 7; Mahoney et al., 1999, 2021; Matthews et al., 2017; Coutts et al., 2020). Although this transition occurred after 72 Ma in the northern Nanaimo basin, extraregional sediment appeared after 83 Ma within the central and southern Nanaimo basin (Coutts et al., 2020; Mahoney et al., 2021).

Figure 14 summarizes the 40Ar/39Ar thermal history results and U-Pb crystallization ages from the major batholithic segments shown in Figure 1. Sediment contained within the Nanaimo basin was derived significantly from erosion of Cretaceous magmatic arc terranes and host rocks enriched in 1.4 Ga and 1.7 Ga detrital zircon (Mustard, 1994; Mahoney et al., 1999, 2021; Monger, 2014; Matthews et al., 2017; Coutts et al., 2020). Inspection of Figure 14 allows us to confidently rule out the arc rocks within the North Cascades and Intermontane belt as the source of these sediments, due to the scarcity of wall rocks capable of supplying large proportions of coarse muscovite and Middle Proterozoic zircon to the Nanaimo basin (e.g., Brown and Gehrels, 2007). Similarly, we can rule out both the Sierra Nevada and Peninsular Ranges arcs based upon their inability to deliver a high proportion of zircon U-Pb ages younger than 80 Ma and K-feldspar 40Ar/39Ar ages younger than 70 Ma (Figs. 8E, 8F, 13, and 14; Irwin and Wooden, 2001; Todd et al., 2003; Grove et al., 2003; Ortega-Rivera, 2003; Chapman et al., 2012; Morton et al., 2014; Premo et al., 2014; Sharman et al., 2015; Jiang and Lee, 2017; Nadin et al., 2016).

Idaho/Boulder Batholiths

The predominantly Late Cretaceous Idaho batholith is located in the central Idaho region (Figs. 1 and 2; Gaschnig et al., 2011). This isotopically evolved batholith (Fleck and Criss, 2007) intrudes the Mesoproterozoic Belt-Purcell Supergroup, a 15–20-km-thick Mesoproterozoic sedimentary basin that contains abundant 1.4 Ga and 1.7 Ga zircons (Fig. 14; Ross and Villeneuve, 2003; Link et al., 2007; Lewis et al., 2010).

The Idaho batholith is a significant source of 75–63 Ma K-feldspar (Fleck and Criss, 2007). The 83–67 Ma peraluminous suite represents the majority of the Atlanta lobe and consists of compositionally restricted biotite granodiorite and muscovite-bearing granite (Gaschnig et al., 2017). The Bitterroot lobe is situated to the northeast of the Atlanta lobe, and it consists of a 76–68 Ma metaluminous outer zone made of quartz diorite, tonalite, and hornblende granodiorite, as well as a 66–53 Ma peraluminous inner zone of biotite granodiorite and muscovite-bearing granite (Fig. 2; Gaschnig et al., 2017). The Boulder batholith in southwestern Montana also intrudes the Mesoproterozoic Belt-Purcell Supergroup (Figs. 1, 2, and 11) and is dominated by the voluminous 77–75 Ma Butte Granite; other intrusions were emplaced between 82 and 74 Ma (du Bray et al., 2012).

As indicated in Figure 14, the Idaho/Boulder batholith region contains all of the ingredients needed to account for the extraregional sediment within the Nanaimo basin, including 75–63 Ma K-feldspar 40Ar/39Ar ages (Table 2; Fig. 11). The Mesoproterozoic Belt-Purcell wall rocks contain 1.4 and 1.7 Ga zircons, including the distinctive 1.38 Ga grains from the Lemhi subbasin. The abundance of muscovite in the peraluminous granites of the Idaho and Boulder batholiths is also significant (Gaschnig et al., 2017; duBray et al., 2012). While muscovite-bearing granitic rocks are uncommon in continental margin batholiths along western North America, they are abundant in an inland belt that stretches from Sonora to British Columbia and includes the Idaho batholith (e.g., Miller and Bradfish, 1980).

Mojave-Salinian Margin

The Mojave magmatic arc is situated between the Sierra Nevada and Peninsular Ranges batholiths of southern California (Figs. 1 and 12). The term “Salinia” refers to Mojave batholithic and associated rocks that were displaced ∼415 km northwest of their original location by Neogene dextral slip along the San Andreas and related faults (e.g., Page, 1970; Dickinson et al., 2005). Prior to the development of the San Andreas fault, the southern Sierra Nevada and northern Peninsular Ranges arcs were separated by >500 km; Figure 12 has been palinspatically restored to illustrate this relationship.

Both the southern California forearc and the Mojave-Salinian arc experienced significant deformation during Late Cretaceous Laramide shallow subduction (Page, 1982; Dumitru et al., 1991; Hall, 1991; Barth et al., 2003; Grove et al., 2003; Saleeby, 2003; Ducea et al., 2009; Jacobson et al., 2011; Chapman et al., 2012, 2016). Upwards of 150 km of the continental margin appear to have been excised during the Late Cretaceous as a result of this deformation (Saleeby, 2003; Jacobson et al., 2011). The intense Laramide deformation of the margin has been attributed to subduction of an oceanic plateau after 80 Ma (Saleeby, 2003; Liu et al., 2010). Extraregional sediment first appeared in this breached segment of the continental margin batholith during latest Maastrichtian–early Paleocene (Sharman et al., 2015). The flux of extraregional sediment was highest during the Eocene (Jacobson et al., 2011; Sharman et al., 2015; Shulaker et al., 2019).

Sufficient 40Ar/39Ar results have been reported from basement K-feldspar from the southern California and Baja California region to permit direct comparisons with the detrital K-feldspar 40Ar/39Ar age distributions measured from the Geoffrey, Spray, and Gabriola formations here (Fig. 15). As indicated in Figure 15B, basement K-feldspar 40Ar/39Ar results are available for the southern Sierra Nevada (Figs. 15B and 15C), the northwestern Mohave area (Figs. 15D and 15E), the southwestern Mojave region (Figs. 15F and 15G), and the northern Peninsular Ranges (Figs. 15H and 15I). The source publications for these results are provided in Table S8 (footnote 1). Application of the Kolmogorov-Smirnov test indicated that basement K-feldspar 40Ar/39Ar age distributions from the Sierra Nevada and Peninsular Ranges are distinct at 95% confidence from the detrital K-feldspar 40Ar/39Ar age distributions measured for the Geoffrey, Spray, and Gabriola formations (Table 3). In the case of the Mojave region, only the comparison between the Gabriola Formation and the northwest Mojave region yielded a P value (0.36) that indicates the two distributions overlap at 95% confidence (Table 3). The clustering of 75–70 Ma K-feldspar 40Ar/39Ar cooling ages in the northwestern Mojave region has been attributed to subduction refrigeration due to Laramide shallow subduction (Shulaker et al., 2019). In spite of this similarity in K-feldspar 40Ar/39Ar ages between the northwestern Mojave region and the age distribution from the Gabriola Formation, the lack of overlap of the Geoffrey and Spray formations argues against the Mojave region as the source of extraregional sediment to the Nanaimo basin.

It is also important to consider when extraregional sediment from the Mojave region reached the margin. Detrital K-feldspar 40Ar/39Ar age distributions are available for Late Cretaceous through Eocene strata deposited in the Santa Ana Mountains at the northern limit of the Peninsular Ranges batholith (Fig. 16; see location in Fig. 12; Lovera et al., 1999; Grove et al., 2003). This location is significant because it records a local to extraregional transition from derivation of sediment from local Peninsular Ranges arc basement to that from the southeastern Mojave region. Distinctive rhyolite metatuffs present in Upper Paleocene and Eocene strata within the Santa Ana Mountains have been correlated with the Jurassic volcanic arc in the southeastern Mojave region (Kies and Abbott, 1983).

Detrital K-feldspar 40Ar/39Ar age distributions from the Santa Ana Mountains are shown in Figure 16. The Cenomanian through Maastrichtian (100–66 Ma) forearc strata (Figs. 16C16F) contain 110–75 Ma detrital K-feldspar 40Ar/39Ar and 120–80 Ma zircon U-Pb ages that are characteristic of the northern Peninsular Ranges batholith (Figs. 8 and 15; Lovera et al., 1999; Grove et al., 2003). A regional unconformity exists between uppermost Cretaceous and Cenozoic strata (McCulluh et al., 2000). The palynology of the overlying Silverado Formation indicates that the depositional age of its basal strata is Thanetian (59–56 Ma; Gaponoff, 1984). The basal Silverado formation also has a Peninsular Ranges age distribution (cf. Fig. 16C with Figs. 16E16F; Sharman et al., 2015). However, much younger detrital K-feldspar 40Ar/39Ar and detrital zircon U-Pb ages (75–65 Ma) present in the upper Silverado formation document the arrival of extraregional sediment (Fig. 16B). The overlying Santiago Formation contains almost pure extraregional sediment with detrital zircon that includes Late Cretaceous (80–69 Ma), Jurassic, Permian, and Proterozoic (1.7 and 1.4 Ga) U-Pb ages characteristic of the Mojave province (Jacobson et al., 2011; Sharman et al., 2015). The detrital K-feldspar 40Ar/39Ar age distribution has age maxima at 70–55 Ma and a tail of older ages up to 1 Ga (Fig. 16A). The latter appears to reflect thermal overprinting of K-feldspar from Proterozoic gneisses caused by Permian–Triassic, Jurassic, and Late Cretaceous magmatism in the Mojave region.

Implications for the Baja–British Columbia Hypothesis

Previous studies of key features of the extraregional sediment within the Nanaimo basin have suggested that these sediments could have been derived from either the Idaho batholith region or the Mojave-Salinia segment of the southern California margin (e.g., Mahoney et al., 1999, 2021; Matthews et al., 2017; Sauer et al., 2019). Results from this study revealed significant problems associated with a Mojave-Salinian source for the extraregional sediment found within the Nanaimo basin. Basement K-feldspar 40Ar/39Ar ages from the Mojave-Salinian margin are (with the exception of the Gabriola sample) distinguishable from the detrital K-feldspar age distributions of the Nanaimo basin samples (Fig. 15; Table 3).

Detrital zircon evidence indicates that sediment containing an appreciable population of Proterozoic zircon eroded from the Mojave region first arrived at the San Gabriel–La Panza, Santa Lucia, and Santa Ynez basins along the southern California margin during the late Maastrichtian–early Paleocene (ca. 70–62 Ma; see fig. 3 of Sharman et al., 2015). The same data sets show, however, that the full extraregional character of the sediment, including high proportions of Jurassic zircon, was established later, during the late Paleocene and early Eocene (ca. 62–48 Ma), along the southern California margin (Fig. 15; Jacobson et al., 2011; Sharman et al., 2015). This postdated the appearance of extraregional sediment within the Campanian (80–74 Ma) De Courcy Formation in the southern and central Nanaimo basin (Coutts et al., 2020; Mahoney et al., 2021) by at least 4 m.y. and potentially by more than 12 m.y. (Fig. 16).

In summary, analysis of detrital K-feldspar 40Ar/39Ar age distributions of Nanaimo Group strata in the northern Nanaimo basin, and comparison with potential source regions in the Mojave-Salinia region and the Idaho batholith favor the latter over the former as the source for the extraregional detritus present within the northern basin. The data presented in this paper are best interpreted as disfavoring the Baja BC hypothesis. Ongoing research that is comparing Pb isotopic data from the Nanaimo basin with data from the Idaho batholith and Mojave-Salina regions will further test this provenance linkage. Regardless, the results in this paper demonstrate that detrital K-feldspar 40Ar/39Ar age distributions can be an important provenance tool that is capable of addressing a wide range of geologic problems.

Detrital K-feldspar 40Ar/39Ar age distributions from samples of Nanaimo Group strata in the northern Nanaimo basin of western British Columbia suggest three distinct types of age distributions. Strata deposited prior to 83 Ma yielded cooling ages that are similar to those exhibited by mid- to Late Cretaceous plutons within the southwestern Coast Mountains batholith. These sediments were deposited in a shallow-marine to nonmarine environment and are interpreted to have received sediment primarily from the southwestern Coast Mountains batholith. Strata deposited between 83 and 72 Ma yielded age maxima that are interpreted to reflect denudation localized east of the Harrison Lake fault of south-central British Columbia. Samples deposited after 72 Ma yielded distinctly younger age distributions inconsistent with local Coast Mountains batholith basement, and they are interpreted to reflect a shift from local to extraregionally derived sediment. These analyses revealed maxima in the detrital K-feldspar 40Ar/39Ar age distribution at 68–63 Ma. This shift is correlated with a significant increase in the grain size and abundance of detrital K-feldspar and the appearance of accompanying muscovite. Our preferred source for this sediment influx is the Idaho batholith, arguing against the Baja BC hypothesis.

This research was supported by grants from Stanford University's Enhancing Diversity in Graduate Education (EDGE) doctoral fellowship program. The 40Ar/39Ar analyses were conducted at the Noble Gas Laboratory at Stanford University. Ziva Shulaker and Carl Jacobson contributed to the compilation of biotite data for this publication's contour map, which was started with data from the Western North American Volcanic and Intrusive Rock Database (NAVDAT), the British Columbia Geological Survey's BCAge database, and Nadin et al. (2016). We are grateful to Jim Monger, Dave Kimborough, and Tim England for discussions in the field, and to Dan Gibson for providing samples from the Breakenridge pluton. Careful reviews of the manuscript provided by Andrew Zuza, Mariah Romero, and Jeff Amato greatly improved its organization and clarity.

1Supplemental Material. Table S1: Descriptions of each Nanaimo Group sample. Table S2: Description of K-feldspar separation method. Table S3: 40Ar/39Ar data collection methods. Table S4: Nanaimo detrital K-feldspar 40Ar/39Ar data. Table S5: K-feldspar 40Ar/39Ar incremental-heating data from the Pacific Northwest. Table S6: Biotite and muscovite 40Ar/39Ar incremental-heating data from samples from the Pacific Northwest. Table S7: Biotite K-Ar and 40Ar/39Ar age database used to construct Figures 10, 11, and 12 for selected Cretaceous batholiths of western North America, including references. Table S8: References used to construct Figures 8, 10, 11, 12, and 13. Figure S1: 40Ar/39Ar step-heating plots for K-feldspar, biotite, and muscovite from basement samples from the Pacific Northwest. Please visit https://doi.org/10.1130/GEOS.S.16556340 to access the supplemental material, and contact editing@geosociety.org with any questions.
Science Editor: Andrea Hampel
Associate Editor: Christopher J. Spencer
Gold Open Access: This paper is published under the terms of the CC-BY-NC license.