Abstract

New and previously published detrital zircon U-Pb ages from sediment in major rivers of south-central Alaska archive several major episodes of magmatism associated with the tectonic growth of this convergent margin. Analysis of detrital zircons from major trunk rivers of the Tanana, Matanuska-Susitna, and Copper River watersheds (N = 40, n = 4870) documents major <250 Ma age populations that are characteristic of the main phases of Mesozoic and Paleogene magmatism in the region as documented from limited U-Pb ages of igneous rocks. Key points from our detrital record include: (1) Major magmatic episodes occurred at 170, 150, 118, 95, 72, 58, and 36 Ma. The overall pattern of these ages suggests that felsic magmatism was episodic with periodicity ranging between ~14 and 32 m.y. with an average of ~22 m.y. (2) Magmatism in south-central Alaska shows similar age trends with both the Coast Mountains batholith and the along-strike Alaska Peninsula forearc basin strata, demonstrating a spatial and temporal relationship of felsic magmatism along the entire northern Cordilleran margin. (3) Topography and zircon fertility appear to influence the presence and/or absence of detrital zircon populations in individual watersheds. Results from this study indicate that regionally integrated detrital zircon populations from modern trunk rivers are faithful recorders of Mesozoic and Paleogene magmatic events along a convergent margin, but there appears to be a lag time for major rivers to record Neogene and ongoing magmatic events.

INTRODUCTION

The convergent margin of southern Alaska includes several of the highest mountain ranges in North America, two active volcanic arc systems, and several ancient and active sedimentary basins, making it an ideal location to study the distribution of detrital zircon age signals in modern rivers across an active plate boundary. The upper plate of this convergent margin also has a punctuated Phanerozoic history of magmatism that is related to terrane accretion, spreading ridge subduction, and flat-slab subduction (Moll-Stalcup, 1994; Plafker and Berg, 1994; Bradley et al., 2003; Cole et al., 2006, 2007; Trop and Ridgway, 2007; Finzel et al., 2011; Berkelhammer et al., 2019; Brueseke et al., 2019). In this study, we utilize detrital zircons from modern trunk rivers as recorders of the history of Mesozoic and Cenozoic felsic magmatism that occurred during the tectonic development of this margin. We focus on the three main trunk rivers of the watersheds of south-central Alaska: the Tanana River, Matanuska-Susitna River, and Copper River watersheds (Fig. 1). Each of these watersheds drains source regions that include a wide range of plutonic, volcanic, and sedimentary rocks (Fig. 2). In many parts of these watersheds, especially in the heavily glaciated and roadless mountainous areas, there are only limited data on the age of igneous rocks, and many of the published ages are not from modern methods. In general, south-central Alaska does not have an extensive U-Pb geochronologic database for igneous rocks, especially when compared to well-dated regions along the North American Cordilleran margin, such as the Coast Mountains batholith in British Columbia (e.g., Armstrong, 1988; McClelland and Mattinson, 2000; Gehrels et al., 2009; Cecil et al., 2018) and the Sierra Nevada batholith in California (e.g., Ducea, 2001; Paterson et al., 2011; Ducea and Barton, 2007; Paterson and Ducea, 2015). Our study combines new and previously published U-Pb detrital zircon ages (N = 40, n = 4870) from the major rivers of south-central Alaska to: (1) document the provenance record in each of the watersheds, (2) characterize the major Mesozoic–Cenozoic magmatic events recorded by the detrital zircon populations, and (3) examine the regional spatial and temporal trends of the felsic magmatic record along the entire northern Cordillera.

This study is one of a growing number that utilize detrital zircon geochronology of modern river sediment for provenance studies (e.g., Cawood et al., 2003; Saylor et al., 2013; Bradley et al., 2015; Lease et al., 2016; Pepper et al., 2016; Capaldi et al., 2017; Ducea et al., 2018; Jackson et al., 2019). The application of detrital zircon ages in modern river sediment to characterize major magmatic and provenance trends along a convergent margin is useful in that (1) the potential source areas and tectonic setting of a given watershed are better defined relative to ancient basinal settings, allowing for more efficient characterization of zircon age distributions; and (2) in regions that have an extensive and long-lived history of magmatism, a temporal felsic magmatic record of the tectonic development of the margin may be documented. In this contribution, we also discuss how factors such as topography and zircon fertility may have impacted our results and interpretations of the detrital zircon populations of modern trunk rivers. We focus on large trunk rivers because they are considered to be a component of fluvial-deltaic systems with high preservation potential within ancient sedimentary basin systems (Potter, 1978), and thus they should provide a rich archive of regional magmatic and related tectonic events as recorded in the detrital zircon record. One goal of this study is to use our results from the sediment of modern trunk rivers along an active convergent margin to serve as a guide for future studies of the usefulness and limitations of detrital zircon records from ancient sedimentary basins that formed in similar tectonic settings.

REGIONAL GEOLOGIC FRAMEWORK

The southern margin of Alaska records subduction-related magmatism since at least the Early Jurassic, documented by a widespread suite of Mesozoic and Cenozoic igneous rocks (Fig. 2). Jurassic volcanic and plutonic rocks and related basinal strata are exposed throughout the southern part of our study area (dark-purple pattern in Fig. 2; DeBari and Coleman, 1989; Plafker et al., 1989; Trop et al., 2005; LePain et al., 2013; Stevens Goddard et al., 2018). These Jurassic rock assemblages occur sporadically along strike for >1000 km and have been interpreted to represent a magmatic arc built on the Wrangellia composite terrane above a north-dipping subduction zone at some distance from the Laurentian continental margin (present-day coordinates; DeBari and Sleep, 1991; Trop et al., 2002; Clift et al., 2005; Rioux et al., 2007; Hacker et al., 2008; Pavlis et al., 2019). Collision of the Wrangellia composite terrane was likely time-transgressive, starting in southeastern Alaska and progressing northward to south-central Alaska (Pavlis, 1982; McClelland et al., 1992; Ridgway et al., 2002; Box et al., 2019; Trop et al., 2020). Whereas initial collision of the south-central Alaska segment of the composite terrane took place sometime during Late Jurassic–Cretaceous time, geologic relations in southeastern Alaska and western Canada indicate close proximity of the composite terrane with the outboard margin of the Yukon-Tanana and Stikine terranes by Middle Jurassic time (e.g., McClelland et al., 1992; Kapp and Gehrels, 1998; Monger, 2014; Yokelson et al., 2015; Beranek et al., 2017). After collision and incorporation of the Wrangellia composite terrane into the upper plate of the convergent margin, Cretaceous and Early Paleocene postcollisional subduction occurred along the outboard (ocean-ward) margin of this terrane and is marked by the widespread plutonic rocks across both the Wrangellia composite and Yukon composite terranes (red and pink patterns in Fig. 2; Moll-Stalcup, 1994; Plafker and Berg, 1994; Cole et al., 2007; Cole and Chung, 2013). This belt of plutons ranges in age from ca. 110–55 Ma and in the study area extends in a northeast-southwest–trending belt from the Yukon Tanana Uplands to the western Cook Inlet (Fig. 2; Reed and Lanphere, 1972, 1973; Wallace and Engebretson, 1984; Lanphere and Reed, 1985; Moll-Stalcup, 1994). The Paleocene component of this suite of igneous rocks has age ranges of ca. 61–57 Ma, and these rocks are well exposed in the central Alaska Range where they have been referred to as the McKinley plutonic suite (Reed and Lanphere, 1969, 1973; Lanphere and Reed, 1985; Hung, 2008). Subduction along the Pacific margin of the Wrangellia composite terrane was also marked by the development of an extensive Mesozoic and Cenozoic accretionary prism exposed in the Chugach Mountains in the southern part of our study area and shown on Figure 2 (e.g., Plafker et al., 1994; Sample and Reid, 2003; Haeussler et al., 2006; Amato et al., 2013).

Beginning in the Late Paleocene and extending into the Early Eocene, the convergent margin was further modified by subduction of a spreading ridge that is interpreted to have migrated from west to east (Moore et al., 1983; Bradley et al., 1993, 2003; Haeussler et al., 2003). Subduction of this spreading ridge is interpreted to have halted subduction-related magmatism on the upper plate, and resulted in the emplacement of slab-window igneous rocks in the arc, forearc, and accretionary prism regions (Cole et al., 2006; Ridgway et al., 2012; Kortyna et al., 2014). In the study area, these igneous rocks are well exposed in the Talkeetna Mountains (light-purple pattern on Fig. 2). Following spreading ridge subduction, normal subduction resumed along the Pacific margin until at least the Late Eocene (Plafker et al., 1994; Finzel et al., 2011). This period of subduction-related magmatism is recorded in a limited number of small plutons ranging in age from ca. 43–37 Ma. These plutons are best documented in the central Alaska Range in our study area (Lanphere and Reed, 1985; Moll-Stalcup, 1994; Hung, 2008; Regan et al., 2019) and are coeval with a major pulse of magmatism in the Aleutian-Alaska Peninsula volcanic arc system located west of our study area.

After this short period of Eocene subduction-related magmatism, the Yakutat microplate, an ~11–30-km-thick oceanic plateau, began to subduct beneath south-central Alaska beginning in the Oligocene (Fig. 2; Plafker, 1987; Ferris et al., 2003; Eberhart-Phillips et al., 2006; Christeson et al., 2010; Finzel et al., 2011; Worthington et al., 2012). Subduction of the thick, buoyant Yakutat microplate prompted the transition from normal subduction to flat-slab subduction in south-central Alaska; this process continues to drive orogenic processes to the present day (Finzel et al., 2011; Arkle et al., 2013; Terhune et al., 2019). Effects of flat-slab subduction of the Yakutat microplate on the upper plate of this convergent margin include increased exhumation and surface uplift (Arkle et al., 2013; Terhune et al., 2019); basin inversion and enhanced sediment accumulation rates to adjacent depocenters (Finzel et al., 2011, 2016; Brennan and Ridgway, 2015; Finzel and Enkelmann, 2017); cessation of magmatism directly above the zone of flat-slab subduction (Finzel et al., 2011; Brueseke et al., 2019); and rerouting of major drainage systems (Benowitz et al., 2019). Along strike, away from the area of flat-slab subduction, active volcanism continues in the Aleutian-Alaska Peninsula volcanic arc in the western part of our study area (AAP in Fig. 2) and in the Wrangell arc in the eastern part of our study area (WVB in Fig. 2; e.g., Richter et al., 1990; Bradley et al., 2003; Brueseke et al., 2019).

METHODOLOGY

Sampling

Modern river samples were analyzed from 17 locations along the main trunk rivers as well as from, in some cases, large tributaries of the main trunk rivers within each of the watersheds. Each sample represents several kilograms of sediment that was collected by shovel from mid-channel longitudinal and transverse bars within each river. The geographic location of each sample is denoted by the yellow circles on Figure 2, and individual sample information can be found in Table 1.

Eight samples were analyzed from the Tanana River watershed. These samples include two from the Nenana River (1 and 2 on Fig. 2), two from the Tanana River (3 and 4), and two from the Delta River (5 and 6). One sample each was also analyzed from the Gerstle and Johnson Rivers (7 and 8, respectively). Four samples were analyzed from the Matanuska-Susitna River watershed (Fig. 2). These include one sample each from the Chickaloon (9), Matanuska (10), Chulitna (11), and Susitna (12) rivers. Finally, five samples were analyzed from the Copper River watershed (Fig. 2). These include a sample from the Chistochina River (13), two samples from the Copper River (14 and 17), and one sample each from the Tazlina (15) and Tonsina (16) rivers.

U-Pb Detrital Zircon Geochronology

U-Pb analysis of detrital zircons from modern river samples were conducted using laser ablation–inductively coupled plasma mass spectrometry (LA-ICP MS) at the University of Arizona LaserChron Center using a Thermo Element2 single-collector ICP-MS following the analytical procedures outlined by Gehrels et al. (2008). Analyses were conducted using a 20 µm spot diameter for laser ablation analysis. Analyses per sample varied between n = 53 and n = 100 depending on the number of zircons in each sample. For analyses younger than 1.0 Ga, the 206Pb/238U ages are reported, whereas for analyses older than 1.0 Ga, 206Pb/207Pb ages are reported. Analyses that show >30% discordance or >5% reverse discordance are not included in the results and interpretations. The results of the analyses are available in Table S1 (see Supplemental Material1) and are displayed on normalized age distribution (relative age probability) plots shown in Figures 3, 4, and 5 for each watershed. These plots were generated using the Age-Pick program in the Isoplot Excel macro (Ludwig, 2003). Peak ages were calculated using the Age-Pick program where each individual peak age was calculated using clusters of three or more detrital zircon grains overlapping in age within 2σ error. These peak ages were used to identify the major detrital zircon age populations that are reported for each sample in the text. A compilation of new and previously published detrital zircon ages for the three watersheds is shown in Figure 6.

RESULTS

Tanana River Watershed

Two modern river samples were collected from the upper parts of the Tanana River watershed; these samples are from the Johnson and Gerstle rivers, which drain the eastern Alaska Range (sample localities 7 and 8 on Fig. 2). These samples are dominated by Paleozoic and Precambrian ages (Fig. 3). The Johnson River yields a Paleozoic population between 378 and 358 Ma, as well as Precambrian age populations between 2.8 and 2.5 Ga, 2.1 and 1.8 Ga, 1.6 and 1.4 Ga, and 1.2 and 1.0 Ga (Fig. 3). The Gerstle River yields almost entirely Precambrian age populations between 2.8 and 2.5 Ga, 2.1 and 1.8 Ga, and 1.4 and 1.0 Ga (Fig. 3).

Two samples were collected from the Delta River, a major tributary of the Tanana River. One sample from the Upper Delta River (sample locality 6 in Fig. 2) yields two major age populations between 340 and 290 Ma and 132 and 114 Ma (Fig. 3). A sample collected from the lower portions of the Delta River (sample locality 5 in Fig. 2) yields major Mesozoic and Cenozoic age populations of 111–92 Ma and 39–34 Ma, one Paleozoic age population of 378–360 Ma, and Precambrian age populations of 2.1–1.8 Ga and 1.4–1.0 Ga (Fig. 3).

Two samples were collected from the Tanana River (sample localities 3 and 4 in Fig. 2). The eastern Tanana River sample yields subordinate Mesozoic and Paleozoic ages between 121 and 107 Ma and 396 and 302 Ma, as well as Precambrian ages between 2.8 and 2.5 Ga, 2.1 and 1.8 Ga, and 1.4 and 1.0 Ga (Fig. 3). The western Tanana River sample yields major Mesozoic and Cenozoic ages between 102 and 89 Ma, 58 and 53 Ma, and 41 and 35 Ma, and Precambrian ages between 2.8 and 2.5 Ga, 2.1 and 1.8 Ga, and 1.4 and 1.0 Ga (Fig. 3).

Two samples were collected from the Nenana River, a major tributary of the Tanana River, which drains the central Alaska Range (sample localities 1 and 2 in Fig. 2). The sample from the upper Nenana River (locality 2) yields predominantly Mesozoic and Cenozoic ages between 101 and 92 Ma, 67–53 Ma, and 41–35 Ma (Fig. 3). The sample from the lower Nenana River (locality 1) contains Mesozoic and Cenozoic ages between 103 and 89 Ma and 39 and 37 Ma, as well as Precambrian ages between 2.7 and 2.5 Ga, 2.1 and 1.8 Ga, and 1.4 and 1.0 Ga (Fig. 3).

Our new modern river detrital zircon ages combined with ages from samples of modern river sediment from previously reported in Trop et al. (2019, 2020) yield a total of 1012 detrital zircon ages (Fig. 6A). Cenozoic and Mesozoic detrital zircon ages account for 14% and 34% of the total age distribution, respectively, with major populations of Miocene–Pliocene (9–5 Ma), Late Eocene (41–34 Ma), Late Cretaceous–Early Eocene (70–53 Ma), and Early to mid-Cretaceous (126–107 and 103–89 Ma) ages (Fig. 6A). Paleozoic detrital zircons account for 13% of the total distribution and yield a major Mississippian–Permian (333–290 Ma) age population (Fig. 6A). Finally, Precambrian detrital zircons account for the largest percentage (39%) of detrital zircon ages and yield three major Archean–Proterozoic populations between 2.8 and 2.5 Ga, 2.1 and 1.8 Ga, and 1.4 and 1.0 Ga (Fig. 6A).

Matanuska-Susitna River Watershed

Two samples were collected from trunk rivers draining the central Alaska Range and northern Talkeetna Mountains. The Susitna River sample (sample 12 in Fig. 2) yields an assortment of Precambrian through Cenozoic ages but yields two major Mesozoic and Cenozoic age populations between 105 and 91 Ma and 61 and 52 Ma (Fig. 4). The Chulitna River sample (sample locality 11 in Fig. 2) yields a similar age distribution as the Susitna River sample with two Mesozoic and Cenozoic age populations between 106 and 89 Ma and 68 and 49 Ma (Fig. 4).

Two samples were collected from rivers draining the southern Talkeetna Mountains in the Matanuska Valley. The sample from the Chickaloon River (sample locality 9 on Fig. 2) yields only Mesozoic and Cenozoic ages, with a major age population between 177 and 144 Ma (Fig. 4). The sample from the Matanuska River (sample locality 10 in Fig. 2) yields predominantly Mesozoic and Cenozoic ages that are between 184 and 148 Ma, 110 and 105 Ma, and 74 and 60 Ma (Fig. 4).

Detrital zircon ages from modern trunk rivers in the Matanuska-Susitna River watershed, both from this study as well as from previously published samples of modern river sediment from Finzel et al. (2016), Lease et al. (2016), Finzel and Enkelmann (2017), and Trop et al. (2019), combine for a total of 3322 detrital zircon ages (Fig. 6B). The age distribution of the Matanuska-Susitna River watershed is dominated by Cenozoic (34%) and Mesozoic (61%) ages that yield major Middle to Late Eocene (46–34 Ma), Late Cretaceous–Early Eocene (80–50 Ma), mid- to Late Cretaceous (115–81 Ma), and Late Triassic–Early Cretaceous (203–140 Ma) age populations (Fig. 6B). Paleozoic ages are sparse, accounting for only 2% of the total age distribution (Fig. 6B). These Paleozoic ages are primarily Silurian (443–419 Ma), Devonian (416–359 Ma), Mississippian (356–327 Ma), and Permian (292–260 Ma) in age. Precambrian ages are also relatively sparse (2% of the total distribution), but generally occur between 2.8 and 2.5 Ga, 2.1 and 1.7 Ga, and 1.4 and 1.0 Ga (Fig. 6B).

Copper River Watershed

One sample was collected from the Chistochina River; this sample is from the upper part of the watershed that drains the eastern Alaska Range (sample locality 13 in Fig. 2). This sample yields major Mesozoic and Cenozoic ages between 127 and 119 Ma, 110 and 93 Ma, and 64 and 46 Ma, as well as abundant Precambrian ages between 2.7 and 2.6 Ga and 1.8 and 1.6 Ga (Fig. 5). Two samples were collected from the upper and lower portions of the Copper River (sample localities 14 and 17 in Fig. 2). The upper Copper River sample contains major Mesozoic, Paleozoic, and Precambrian ages between 106 and 94 Ma, 311 and 292 Ma, and at ca. 2.6 Ga and 1.8 Ga (Fig. 5). The lower Copper River sample contains abundant Cenozoic, Mesozoic, and Paleozoic ages between 76 and 48 Ma, 98 and 93 Ma, 132 and 106 Ma, 163 and 140 Ma, and 320 and 297 Ma, and a subordinate cluster of Precambrian ages between 2.0 and 1.8 Ga (Fig. 5). Two samples were collected from trunk rivers that drain the Chugach Mountains (sample localities 15 and 16 in Fig. 2). One sample from the Tazlina River displays primarily Mesozoic and Paleozoic ages between 78 and 71 Ma, 104 and 91 Ma, 160 and 127 Ma, and 315 and 294 Ma (Fig. 5). One sample from the Tonsina River contains a range of Mesozoic to Precambrian ages that cluster between 80 and 70 Ma, 111 and 98 Ma, 163 and 155 Ma, 199 and 185 Ma, 370 and 355 Ma, and 2.0 and 1.8 Ga (Fig. 5).

Detrital zircon ages from modern trunk rivers in the Copper River watershed, both from this study and ages from one sample of modern river sediment reported in Day et al. (2016) combine for a total of 536 detrital zircon ages (Fig. 6C). The Copper River watershed displays a predominance of Mesozoic detrital zircon ages that account for 62% of the total population. These Mesozoic detrital zircons yield major Late Cretaceous (80–68 Ma), mid-Cretaceous (111–90 Ma), Early to mid-Cretaceous (132–117 Ma), and Late Jurassic–Early Cretaceous (161–140 Ma) age populations (Fig. 6C). Paleozoic detrital zircons account for 15% of the total age distribution and yield a major Pennsylvanian–Permian (309–293 Ma) age population (Fig. 6C). Finally, Precambrian detrital zircons constitute 19% of the total population, with major age populations between 2.8 and 2.5 Ga, 2.2 and 1.7 Ga, and 1.4 and 1.0 Ga (Fig. 6C).

DISCUSSION

Tanana River Watershed Provenance

The Tanana River watershed shows <250 Ma age populations of 126–107 Ma, 103–89 Ma, 70–53 Ma, 41–34 Ma, and 9–5 Ma (Figs. 3 and 6A). The Early to mid-Cretaceous (126–107 Ma) age population matches well with 126–113 Ma plutons, which are associated with the Early to mid-Cretaceous Chisana arc in the Wrangellia terrane (e.g., Snyder and Hart, 2007; Graham et al., 2016; Manselle et al., 2020). These plutons are best exposed in the eastern Alaska Range in the headwaters of the watershed (Fig. 2). The mid-Cretaceous (103–89 Ma) detrital zircons were likely derived from an extensive belt of similar-aged plutonic rocks that are part of the Yukon composite terrane (Fig. 2; 110–85 Ma; mid-Cretaceous igneous belt of Plafker and Berg, 1994; Hudson, 1994; Hart et al., 2004; Dusel-Bacon et al., 2015). The Late Cretaceous–Early Eocene detrital zircon population was most likely derived from a suite of Upper Cretaceous to Lower Eocene plutonic rocks exposed in the central Alaska Range (Fig. 2; 75–56 Ma; the Alaska Range–Talkeetna Mountains magmatic belt of Moll-Stalcup, 1994; Cole et al., 2007; Cole and Chung, 2013). Another possible source for these zircons is recycling of Cenozoic sedimentary strata of the Tanana basin exposed north of the central Alaska Range (TB in Fig. 2); strata of this basin have detrital populations with dominant peaks between 75 and 50 Ma (Brennan and Ridgway, 2015). Late Eocene (41–34 Ma) detrital zircons were likely derived from a suite of small Eocene plutons that are exposed in the central and eastern Alaska Range within the watershed (Fig. 2; Moll-Stalcup, 1994). The late Miocene (9–5 Ma) detrital zircon age population was derived from igneous rocks associated with Wrangell arc magmatism (30–0 Ma); these rocks are exposed in the Wrangell Mountains at the headwaters of the watershed (Fig. 2; Richter et al., 1990; Preece and Hart, 2004; Brueseke et al., 2019). Overall, <250 Ma detrital zircon ages from modern river sediment in the Tanana River watershed are interpreted to reflect derivation from predominantly primary plutonic and volcanic sources, with a smaller contribution from recycled sedimentary sources.

Detrital zircon ages >250 Ma constitute a significant percentage (~52%) of the detrital zircon age distribution of the Tanana River watershed (Fig. 3). The Tanana River watershed contains a significant age population between 330 and 290 Ma (Fig. 2) found primarily in the most upstream sample. We attribute this population as being derived from plutonic rocks of the Skolai arc (320–290 Ma) in the Wrangellia composite terrane and/or metasedimentary strata that were derived from the arc, all of which are exposed in the eastern Alaska Range (e.g., Nokleberg et. al., 1994; Wilson et al., 2015). Precambrian detrital zircon grains constitute the largest percentage of the Tanana River watershed age distribution (39%; Fig. 3) and yield major age populations between 2.8 and 2.5 Ga, 2.1 and 1.8 Ga, and 1.4 and 1.0 Ga. The source of these Precambrian grains is likely from Neoproterozoic–early Paleozoic metasedimentary basinal strata associated with the parautochthonous continental margin of North America. These metasedimentary strata are extensively exposed in the Tanana River watershed and contain similar detrital zircon age populations between 2.7 and 2.6 Ga, 1.9 and 1.8 Ga, and 1.6 and 0.9 Ga (e.g., Dusel-Bacon et al., 2017; Romero et al., 2020), making them likely sources of these detrital zircons in the modern river sediment samples.

Matanuska-Susitna River Watershed Provenance

The Matanuska-Susitna River watershed shows predominantly <250 Ma age populations (94% of the total age distribution) between 203 and 140 Ma, 115 and 81 Ma, 80 and 50 Ma, and 46 and 34 Ma (Fig. 6B). The Late Triassic to Early Cretaceous (203–140 Ma) detrital zircon population overlaps in age with Late Triassic to Late Jurassic plutons of the Talkeetna arc (201–153 Ma) and Middle Jurassic–Early Cretaceous (175–135 Ma) plutons of the Chitina arc in the Wrangellia composite terrane (Fig. 2; Plafker et al., 1989; Plafker and Berg, 1994; Rioux et al., 2007). Plutons of the Talkeetna arc are exposed extensively throughout the Talkeetna Mountains and are likely the primary sources of the detrital zircons in this watershed. Plutons of the Chitina arc, however, are mostly exposed outside of the Matanuska-Susitna River watershed in the southern Wrangell Mountains (Fig. 2). The prevalence of detrital zircons (153–140 Ma) that overlap in age with the Chitina arc, therefore, likely reflect erosion of Mesozoic sedimentary strata exposed throughout this watershed. These secondary sources include (1) the Kahiltna assemblage in the Clearwater Mountains (KB in Fig. 2); (2) forearc basin strata of the Matanuska basin exposed in the southern Talkeetna Mountains (MB in Fig. 2); and (3) accretionary prism strata exposed in the Chugach Mountains (Fig. 2). Sedimentary strata in all three of these potential sources contain detrital zircon ages that were derived from Chitina arc plutons (Hampton et al., 2010; Amato et al., 2013; Stevens Goddard et al., 2018). Mid- to Late Cretaceous (115–81 Ma) detrital zircons overlap in age with mid- to Late Cretaceous plutons that are exposed in the Talkeetna Mountains and central Alaska Range (Fig. 2; Wilson et al., 2015) and indicate a likely primary igneous source for these zircons. An extensive belt of mid-Cretaceous (110–85 Ma) plutons are also exposed in the Yukon Tanana Uplands, which lie well outside of the watershed (Fig. 2; Dusel-Bacon et al., 2015). Zircon from these sources may be related to erosion and recycling of Cretaceous strata of the Kahiltna and Matanuska basins as well as the Chugach accretionary prism (Fig. 2). All these strata contain detrital zircons that are interpreted as being derived from this mid-Cretaceous plutonic suite (Hampton et al., 2010; Stevens Goddard et al., 2018; Romero et al., 2020). The Late Cretaceous to Early Eocene (80–50 Ma) detrital zircon age population can be attributed to an extensive suite of similar-aged plutons of the Alaska Range–Talkeetna Mountains magmatic belt (75–56 Ma); these plutons are exposed throughout the watershed (Fig. 2; Cole et al., 2007; Cole and Chung, 2013). Another potential source may include coeval lavas associated with the Colorado Creek volcanics located in the central Alaska Range; these lavas yield 40Ar/39Ar ages between 71 and 68 Ma (Trop et al., 2019). Finally, Middle to Late Eocene (46–34 Ma) detrital zircons overlap in age with spatially limited Eocene to Oligocene (40–30 Ma) plutons that are exposed in the central Alaska Range (Fig. 2; Moll-Stalcup, 1994; Wilson et al., 2015). In summary, detrital zircons from the Matanuska-Susitna River watershed likely reflect derivation from both primary plutonic and volcanic sources, as well as secondary sedimentary sources.

Copper River Watershed Provenance

Detrital zircon ages from modern trunk river sediment of the Copper River watershed are dominated by <250 Ma age populations (67% of the total age distribution) of 161–140 Ma, 132–117 Ma, 111–90 Ma, and 80–68 Ma (Fig. 5). Late Jurassic to Early Cretaceous (161–140 Ma) detrital zircons in the watershed overlap in age with plutons associated with the Chitina arc (175–135 Ma). These plutons are exposed south of the Wrangell Mountains in the eastern part of the watershed (Fig. 2) and were likely the predominant primary sources for these zircons. Other possible sources for this age population include recycled zircons from sedimentary strata of the Wrangell Mountains basin (WMB in Fig. 2) and metasedimentary strata of the Chugach accretionary prism (Fig. 2). Both of these potential sources contain significant populations of Late Jurassic to Early Cretaceous detrital zircons (e.g., Amato et al., 2013; Day et al., 2016; Fasulo et al., 2020). Early Cretaceous plutons that may have contributed 132–117 Ma detrital zircons in the Copper River watershed have not been documented, aside from sparse Cretaceous-aged plutons that have been mapped in the eastern Alaska Range (e.g., Wilson et al., 2015). These plutons may be a possible source of Early to mid-Cretaceous detrital zircons present in our sample from the Chistochina River, which drains the eastern Alaska Range (Fig. 2). The source(s) for Early Cretaceous detrital zircons from samples farther downstream are most likely recycled from metasedimentary and sedimentary strata of the Chugach accretionary prism and Wrangell Mountains basin (Fig. 2); both contain similar aged detrital zircons (e.g., Amato et al., 2013; Fasulo et al., 2020). Similar to Early Cretaceous plutons, mid-Cretaceous plutons are also lacking in the Copper River watershed (Fig. 2); however, the watershed contains abundant mid Cretaceous (111–90 Ma) detrital zircons (Fig. 6C). These detrital zircons appear to be more strongly represented in the upstream samples (Fig. 5). These samples are more proximal to potential sources from the eastern Alaska Range where mid Cretaceous plutons are exposed just outside of the watershed boundary (Fig. 2). Two possible interpretations for the source of sediment of this age range include: (1) mid-Cretaceous igneous rocks that may be mainly buried beneath the extensive volcanic and glacial cover of the Wrangell arc (WVB in Fig. 2) and beneath the extensive Quaternary cover of the Copper River basin (CRB in Fig. 2); and/or (2) mid-Cretaceous detrital zircons that may have been transported from the eastern Alaska Range into the Copper River watershed by glaciers during glacial maxima. Finally, Late Cretaceous (80–68 Ma) detrital zircons are most abundant in the two samples from the Tazlina and Tonsina Rivers, both of which drain the Mesozoic accretionary prism strata in the Chugach Mountains (Fig. 2). Upper Cretaceous accretionary prism strata contain abundant detrital zircons between 80 and 68 Ma (Amato et al., 2013; Gross Almonte et al., 2019) and are interpreted as the predominant source for Late Cretaceous zircons in the modern watershed.

Detrital zircons >250 Ma constitute a significant percentage of the total age distribution in the Copper River watershed (34%; Fig. 6C). The watershed contains Pennsylvanian–Permian (309–293 Ma) detrital zircons that were likely derived from plutons and related metasedimentary strata associated with the Pennsylvanian–Permian Skolai Group. These units are part of the Wrangellia composite terrane and are exposed in the eastern Alaska Range in the northern part of the watershed (Fig. 2). Precambrian detrital zircons are also abundant and display age populations of 2.8–2.5 Ga, 2.2–1.7 Ga, and 1.4–1.0 Ga (Fig. 6C). These populations are similar to those found in the Tanana River watershed. We attributed Tanana River populations to being derived from Precambrian to Paleozoic metasedimentary rocks exposed in the Yukon Tanana Uplands (Fig. 2). Similar rocks are exposed in the eastern Alaska Range in the upper reaches of the watershed (Fig. 2; Wilson et al., 2015), making these rocks a likely source for the Precambrian detrital zircons.

Characterizing the Mesozoic–Cenozoic Magmatic Record along the Southern Alaska Convergent Margin

A composite detrital zircon age distribution of grains <250 Ma from modern rivers of south-central Alaska is shown in Figure 7. In this section, we evaluate and characterize the integrated detrital zircon record of the major Mesozoic and Cenozoic magmatic events along the southern Alaska convergent margin. Arc magmatism associated with Jurassic to Early Cretaceous subduction is recorded by igneous rocks associated with both the Talkeetna (201–153 Ma) and Chitina (175–135 Ma) arcs that are exposed in the Talkeetna Mountains and southern Wrangell Mountains, respectively (Fig. 2; Plafker et al., 1989; Plafker and Berg, 1994; Rioux et al., 2007; Stevens Goddard et al., 2018). The modern river detrital zircon record suggests magmatism in these arcs was continuous between 200 and 140 Ma, with peak magmatism occurring ca. 170 and 150 Ma (Fig. 7). Another pulse of magmatism is recorded in the integrated detrital zircon data set between 125 and 113 Ma (Fig. 7; peak magmatism ca. 118 Ma). This age range matches closely with recently published 126–117 Ma U-Pb zircon ages from plutons of the Chisana arc exposed in the Wrangell Mountains (Fig. 2; Graham et al., 2016; Manselle et al., 2020). The Talkeetna-Chitina-Chisana arcs are all interpreted to have been active prior to and coeval with collision of the Wrangellia composite terrane to the Mesozoic continental margin (Plafker and Berg, 1994; Rioux et al., 2007; Trop and Ridgway, 2007; Box et al., 2019; Trop et al., 2020).

The next felsic magmatic pulse recorded in the integrated detrital record occurs between 105 and 85 Ma, with peak magmatism occurring ca. 95 Ma (Fig. 7). An extensive suite of 110–85 Ma plutonic rocks are exposed throughout both the Yukon composite and Wrangellia composite terranes (Fig. 2). These plutons have been interpreted to have been emplaced following a period of crustal thickening associated with subduction-related magmatism along the southern Alaska convergent margin (Moll-Stalcup, 1994; Hart et al., 2004; Dusel-Bacon et al., 2015; Romero et al., 2020). Following this magmatic phase, the modern river detrital record documents a Late Cretaceous–Eocene magmatic pulse between 80 and 50 Ma, with peak magmatism occurring ca. 72 and 58 Ma (Fig. 7). This period of magmatism coincides with the development of the younger part of the Alaska Range–Talkeetna Mountains magmatic belt (75–56 Ma). This suite of igneous rocks has been interpreted to represent magmatism, metamorphism, and crustal thickening along the southern Alaska convergent margin (Davidson et al., 1992; Cole et al., 1999; Ridgway et al., 2002; Cole et al., 2007; Davidson and McPhillips, 2007). A Middle to Late Eocene magmatic pulse is documented between 45 and 35 Ma with peak magmatism occurring ca. 36 Ma (Fig. 7). This phase of magmatism is reflected by a small number of plutons geographically restricted to mostly the central Alaska Range (40–30 Ma; Fig. 6), as well as recently dated mafic dike swarms in the central Alaska Range that yield ages between ca. 38 and 25 Ma (Trop et al., 2019). These igneous rocks were coeval with initiation of Aleutian arc magmatism in the western Alaska Peninsula (Jicha et al., 2006; Finzel and Ridgway, 2017). The decrease in magmatism after this brief phase coincides with the beginning of flat-slab subduction of the Yakutat microplate along the convergent margin. This stage of subduction is interpreted to have caused both cessation of magmatism directly above the flat-slab region (Finzel et al. 2011) and slab-edge magmatism along the eastern margin of the subducting microplate (Brueseke et al., 2019). The youngest detrital zircon populations recorded from the trunk rivers of south-central Alaska are 10–5 Ma with an age peak ca. 7 Ma (Fig. 7). This age range coincides with only part of the age range of the magmatism of the Wrangell arc exposed in the Wrangell Mountains (Fig. 2); magmatism in this belt is interpreted to have begun ca. 30 Ma and has been relatively continuous since then (Richter et al., 1990; Preece and Hart, 2004; Berkelhammer et al., 2019; Brueseke et al., 2019). Zircons younger than 30 Ma, however, are relatively sparse in our detrital record from modern trunk rivers (Fig. 7). The absence of detrital zircon populations representing the entire age range of the Wrangell arc may be related to several factors. One factor is that there may be some lag time needed to allow deeper exhumation and extensive erosion of plutons that underlie the lavas of the active Wrangell arc before significant zircon-rich detritus can propagate downstream into the main truck rivers of the watershed. Supporting this interpretation are ongoing U-Pb geochronology detrital zircon studies of the proximal, smaller tributaries draining the Wrangell arc; these studies report ages that cover the entire duration of Wrangell arc magmatism (Trop et al., 2019). More sample coverage of the trunk rivers is needed, but significant amounts of this age detritus do not seem to have reached the major rivers and also suggest a possible lag time. Another related possibility is that the 10–5 Ma detrital zircon population in our samples is the product of erosion of airfall tephras associated with the Wrangell arc. Similar age detrital zircon populations have been reported from localized Miocene–Pliocene basins in the eastern Alaska Range that are attributed to reworking of tephras into fluvial strata (Allen, 2016; Allen et al., 2018). In addition, in the Upper Miocene strata of the Tanana basin near the town of Healy, Alaska, two thick tephras have been documented, one 8.4 m thick and the other 5.0 m thick (Ridgway et al., 2007). These tephras have 40Ar/39Ar ages from biotite, plagioclase, and hornblende of 6.4–6.7 Ma (Triplehorn et al., 2000). These findings suggest that extensive airfall tephra may have been deposited across south-central Alaska from 10–5 Ma; this episode of volcanism might be what is recorded in our trunk river sediment samples.

Comparison with Along-Strike Magmatic Records in the Northern Cordillera

From a regional perspective, south-central Alaska forms one part of the northern Cordilleran continental margin that spans from the western Alaska Peninsula to southern British Columbia. Figure 8 shows a comparison between the Mesozoic and Cenozoic magmatic records of south-central Alaska, the western Alaska Peninsula, and the Coast Mountains in British Columbia. Note that in Figure 8, the magmatic record of the western Alaska Peninsula is based on detrital zircon from Mesozoic and Cenozoic forearc basin strata (Finzel and Ridgway, 2017), the magmatic record of south-central Alaska is based on detrital zircon from modern river sediment (this study), and the magmatic record of the Coast Mountains batholith in British Columbia is based on a magmatic flux curve calculated from the age and volume of exposed plutons (Gehrels et al., 2009). In general, all three records show a magmatic flare-up in the Middle to Late Jurassic–Early Cretaceous (ca. 180–140 Ma) that suggests a regional widespread magmatic event. All three data sets show a general lack of magmatic activity between 140 and 120 Ma that coincides with a well-documented magmatic lull throughout the Northern Cordillera (Fig. 8; Armstrong, 1988; Gehrels et al., 2009). This lull has been interpreted to reflect (1) initial collision of the Wrangellia composite terrane with the continental margin (Trop and Ridgway, 2007; Stevens Goddard et al., 2018); (2) a period of oblique sinistral displacement along the margin (Chardon et al., 1999; Butler et al., 2006; Gehrels et al., 2009; Beranek et al., 2017); and/or (3) an episode of ridge subduction along the northern Cordilleran margin (Pavlis et al., 1988; Mahar et al., 2019). Widespread magmatism resumed between 120 and 80 Ma (Fig. 8), with the emplacement of plutons in south-central Alaska and the Coast Mountains batholith. This stage is also marked by regional crustal metamorphism, thickening, and shortening (Rubin et al., 1990; Gehrels et al., 2009; Romero et al., 2020). All three magmatic records generally document significant magmatism between 80 and 50 Ma, with the Coast Mountains batholith indicating a discrete period of extensive magmatism ca. 55–48 Ma (Fig. 8; Gehrels et al., 2009). Magmatism during this period is associated with the widespread emplacement of Late Cretaceous to Eocene plutons in a transpressional regime followed by widespread extension in what is now British Columbia (Andronicos et al., 2003; Hollister and Andronicos, 2006). Late Cretaceous to Eocene plutons are also common in the Alaska Range and Talkeetna Mountains in south-central Alaska (Fig. 2; Moll-Stalcup, 1994; Cole et al., 2007). The magmatic record of the Coast Mountains batholith lacks evidence for significant magmatic activity during the time after ca. 48 Ma (Fig. 8). In British Columbia, this time period is characterized by a transition from a convergent to transform boundary as the Pacific plate changed to a more NW-directed plate motion (Engebretson et al., 1985; Gehrels et al., 2009). The magmatic record of the Alaska Peninsula forearc basin shows only minor events after ca. 48 Ma (Fig. 8). In the Alaska Peninsula, arc magmatism, however, is interpreted to have been active from Late Eocene to the present in the Aleutian arc (Jicha et al., 2006). The absence of a clear detrital record of post–48 Ma magmatism may be due to a lack of extensive exhumation and erosion of post–48 Ma plutons in this active arc system. South-central Alaska shows ca. 45–35 Ma magmatism that reflects a period of normal subduction along the margin (e.g., Plafker et al., 1994). Plutons of this age have been exhumed and are at the surface in the central Alaska Range (Csejtey et al., 1992). Following the 45–35 Ma period of magmatism, flat-slab subduction of the Yakutat microplate commenced, which halted magmatism above the flat-slab region in south-central Alaska (Finzel et al., 2011). Active magmatism in the Wrangell arc along the eastern edge of the flat-slab has been documented since ca. 30 Ma (Richter et al., 1990; Berkelhammer et al., 2019; Brueseke et al., 2019) and is partly recorded in the modern trunk river detrital record with populations between 10 and 5 Ma (Fig. 8). In general, our comparison of the magmatic record of south-central Alaska with the magmatic records of the western Alaska Peninsula and British Columbia shows a general temporal relationship between the major periods of Mesozoic and Paleogene magmatism along the entire northern Cordillera margin that is readily documented by the ages of igneous rocks, the detrital zircons of ancient sedimentary strata, and the detrital zircons of modern river sediment.

Arc Magmatic Episodes in South-Central Alaska

An important avenue of ongoing research is centered on understanding the episodic magmatic record of arc systems (e.g., Ducea, 2001; Ducea and Barton, 2007; DeCelles et al., 2009; Gehrels et al., 2009; Paterson and Ducea, 2015; Beranek et al., 2017; Cecil et al., 2018). These studies have shown that the development of continental arc systems is punctuated by high-volume magmatic “flare-ups” separated by low-volume magmatic “lulls” (e.g., Ducea et al., 2015). Detrital zircon ages from modern rivers in south-central Alaska provide a documentation of the major episodes of felsic magmatism on the upper plate of this convergent margin during the past 200 m.y. Mesozoic and Paleogene peak magmatism along this part of the plate boundary occurred ca. 170, 150, 118, 95, 72, 58, and 36 Ma based on our detrital zircon record from modern trunk rivers (Fig. 7). The overall pattern of these ages suggests that felsic magmatism was highly episodic, with the periodicity of magmatic episodes ranging between ~14 and 32 m.y. with an average of ~22 m.y. The cyclicity displayed in our detrital zircon record shows similarities to cyclical patterns that have been documented in other regions along the Cordilleran margin. In the central Andes, for example, periods of magmatic flare-ups during the past 200 m.y. have an apparent cyclicity of ~25–45 m.y. (Haschke et al., 2006; DeCelles et al., 2015). Along these same lines, studies of North American Cordilleran arcs, such the Sierra Nevada batholith in California and the Coast Mountains batholith in British Columbia, document flare-up events with an apparent cyclicity of ~25–50 m.y. (Ducea, 2001; Ducea and Barton, 2007; DeCelles et al., 2009; Gehrels et al., 2009; Beranek et al., 2017; Cecil et al., 2018). In addition, similar to our study, comparisons of the U-Pb ages from igneous rocks of the Sierra Nevada batholith with U-Pb detrital zircon ages from sedimentary strata in adjacent basins found a general agreement between timing of magmatic flare-ups and lulls recorded in these different components of that settings (de Silva et al., 2015; Paterson and Ducea, 2015). At a first order, this suggests to us that the study of detrital zircons from modern rivers is a powerful approach to delineate the magmatic processes and cyclicity along a long-lived convergent margin. Our study from modern rivers builds on detrital zircon studies that interpreted magmatic arc flare-ups and lulls based on the ancient sedimentary record (e.g., Barth et al., 2013; Laskowski et al., 2013). As pointed out by Paterson and Ducea (2015), however, the detrital zircon record may not be a sensitive recorder of the volume of magma addition rates to the upper plate during a flare-up event. It is also important to note that detrital zircons best reflect episodes of felsic magmatism and, therefore, are not an accurate recorder of major mafic and intermediate magmatic episodes (e.g., Martínez Ardila et al., 2019). Another important consideration in evaluating magmatic episodes in southern Alaska is that well documented magmatic cessation events are associated with processes such as terrane accretion, spreading ridge subduction, and oceanic plateau subduction (Bradley et al., 2003; Finzel et al., 2011; Finzel and Ridgway, 2017; Stevens Goddard et al., 2018; Jones et al., 2021).

The similarity in magmatic cyclicity displayed in our modern river detrital zircon record in south-central Alaska (14–32 m.y.) to the magmatic cyclicity documented in the central Andes (25–45 m.y.) and North American Cordilleran arcs (25–50 m.y.) suggests that the entire Cordilleran margin has been impacted by similar cyclical convergent margin processes. Multiple studies have explored potential causes for cyclicity within continental arc systems (e.g., Haschke et al., 2002, 2006; DeCelles et al., 2009, 2015; Paterson and Ducea, 2015). Proposed explanations range from lower-plate controls, such as plate convergence velocity, geometry and age of the down-going slab, and thermal anomalies at the base of the lithosphere to upper-plate controls, such as retroarc shortening and crustal delamination. Our detrital zircon data from modern rivers in south-central Alaska alone are not sufficient to determine the specific tectonic processes that may have driven magmatic cyclicity. The recognition that our data do exhibit cyclicity that is comparable to along-strike parts of the Cordilleran margin, however, indicates that this is a plate-scale process and as has been pointed out by several studies warrants additional future study.

Comparison of Detrital Zircon Records between Individual Watersheds versus Integrated Watersheds and Application to Studies of Ancient Basins

The regionally integrated data set from watersheds of south-central Alaska demonstrates that detrital zircons from the major rivers contain a comprehensive record of Mesozoic and Paleogene magmatic events that have occurred along this convergent margin (Fig. 7). Data sets from any single individual watershed, however, do not capture the entire magmatic record (Fig. 6). Prominent zircon populations between 200 and 140 Ma, for example, are well developed in the Matanuska-Susitna River and Copper River watersheds but are generally lacking in the Tanana River watershed (Fig. 6). In the Matanuska-Susitna River watershed, major detrital zircon populations are lacking between 130 and 115 Ma; whereas in the Tanana River and Copper River watersheds, detrital zircons of this age are present. It is also interesting to note that the youngest detrital peak age in the Copper River watershed is 46 Ma, in the Matanuska-Susitna River watershed, it is 36 Ma, and in the Tanana River watershed, it is 7 Ma. These differences suggest a high degree of partitioning and less time for homogenization of detrital zircons from these younger magmatic episodes between individual watersheds.

One of the first-order controls on the differences between the detrital zircon populations between the three watersheds appears to be the high-relief topography that characterizes the southern Alaska convergent margin. As shown in Figure 1, the northwest-flowing Tanana River watershed is divided from the south-flowing Matanuska-Susitna River and Copper River watersheds by the high peaks of the central and eastern Alaska Range. The lack of 200–140 Ma detrital zircons in the Tanana River watershed can be accounted for by this major topographic barrier, as most of the Jurassic–Lower Cretaceous plutonic and volcanic assemblages (i.e., Talkeetna and Chitina arc plutons in the Wrangellia composite terrane) are presently exposed south of the Alaska Range. The presence of the 130–115 Ma detrital zircon in the Tanana River and Copper River watersheds but not in the Matanuska-Susitna River watershed is most likely due to restriction of plutons of this age to the eastern Alaska Range (red pattern located between 62°N and 63°N in Fig. 2). Another interesting comparison can be made between the Matanuska-Susitna River and Copper River watersheds. Both watersheds are situated south of the Alaska Range, and both flow to the south (Fig. 1). However, as discussed earlier in this section, there are differences in the detrital zircon age spectra between the two watersheds (Figs. 6B and 6C). The differences in the detrital zircon populations suggest that the topographic divide between the two watersheds (i.e., the eastern Talkeetna Mountains; Fig. 2) exerts significant control on the sources of sediment available in each watershed and, therefore, the presence or absence of some detrital zircon populations.

Another controlling factor in zircon presence in south-central Alaska watersheds appears to be lithologic controls on zircon fertility of potential source rocks. Previous studies have noted the potential impact of zircon fertility (i.e., zircon abundance) in interpreting detrital zircon provenance (e.g., Moecher and Samson, 2006; Dickinson, 2008; Spencer et al., 2018). Our interpretation of the magmatic record of south-central Alaska is certainly biased toward magmatic phases that produced zircon-fertile rocks (i.e., felsic plutonic and volcanic rocks) as opposed to more zircon-poor rocks such as mafic igneous rocks. This is significant in a region such as south-central Alaska, which has a magmatic history involving felsic, intermediate and mafic magmatism (e.g., DeBari and Coleman, 1989; Richter et al., 1990; Moll-Stalcup, 1994; Cole et al., 2006, 2007; Rioux et al., 2007; Berkelhammer et al., 2019; Brueseke et al., 2019). Zircon-rich magmatic phases may be represented in our data set by the large detrital zircon populations of 110–85 and 80–50 Ma present in all three watersheds (Fig. 6). The phases of 110–85 and 80–50 Ma magmatism are documented by the mid-Cretaceous igneous belt best exposed in the Yukon Tanana Uplands and the Late Cretaceous–Eocene Alaska Range–Talkeetna magmatic belt best exposed in the central Alaska Range, respectively (Fig. 7). Geochronologic and geochemical studies of plutons in both belts indicate that these plutons were emplaced due to high degrees of crustal melting during or following phases of crustal thickening (Hudson, 1994; Hart et al., 2004; Cole and Chung, 2013; Dusel-Bacon et al., 2015; Romero et al., 2020). In both instances, melting of thickened crustal rocks would have produced zircon-fertile magmatism on the upper plate of this convergent margin. As a result, the modern river detrital zircon record of south-central Alaska appears to be dominated by zircons representing these two periods of zircon-fertile magmatism. Other factors such as erodibility of different lithologies and areal extent of exposure of different lithologies will need to be better known to fully evaluate the role of zircon fertility in the detrital population from the modern rivers.

The differences in detrital zircon populations that we have documented between individual watersheds of south-central Alaska have implications for the application of detrital zircon geochronology to studies of ancient sedimentary basins. Our results point out the advantage of having detrital zircon samples from a geographically wide distribution within an ancient basin system, if the goal of the study is to establish a comprehensive record of sources of sediment. In some cases when designing a provenance study, it seems prudent to focus detrital zircon sample collection in a specific part of an ancient basin because this “type section” has the most continuously exposed stratigraphic section and the most rigorous age control (e.g., Brennan and Ridgway, 2015). This “type section” approach may be warranted or even necessary in some cases, but this strategy may lead to an incomplete record of all the source areas that supplied sediment to the basin. This concern may be especially critical in provenance studies of ancient collisional and accretionary convergent margins when the absence or presence of a specific source of sediment is used to determine the proximity of an allochthonous terrane to a continental margin or the timing of collision of the “exotic” terrane to a convergent margin. As discussed in an earlier section, our data also indicate that there may by some lag time in response to sediment propagation into the main trunk rivers of a watershed during an episode of Neogene and active magmatism. This finding might be important when analyzing detrital zircon from Neogene sedimentary strata with the goal to determine the initiation of magmatism and related subduction processes along a convergent margin.

Limitations of the Study

This study was designed to investigate the Mesozoic and Cenozoic magmatic record of the southern Alaska convergent margin as recorded in the detrital zircon populations of modern river sediment. Our analysis explores this record but also has several limitations. Our sampling density and small-n data sets, for example, may have influenced our interpretation of the presence of detrital zircon populations in each of the three watersheds and in our comparison between watersheds. Given the larger number, for example, of new (this study) and previously published detrital zircon samples from the Matanuska-Susitna River watershed, the total number of samples (N) and therefore analyses (n) for the Matanuska-Susitna River watershed (N = 23 and n = 3322) is significantly larger than the number for the Tanana River watershed (N = 11 and n = 1012) and Copper River watershed (N = 6 and n = 536). This results in our comparisons between the age distributions in each watershed being skewed toward the greater number of samples represented by the Matanuska-Susitna River watershed. The greater number of samples, and therefore number of analyses, also favors source regions in the Matanuska-Susitna River watershed to exhibit a stronger signal in the integrated detrital zircon age spectra shown in Figure 7.

Another limitation of our study is that it utilizes small-n data sets for new samples reported herein; these data sets range between 53 and 100 constituent analyses (average n = 91). Many of our samples were analyzed over the past decade, a time period when n = 100 was a standard approach for detrital zircon studies (e.g., Gehrels et al., 2006, 2008; Gehrels, 2012). Improvements in instrumentation in recent years have allowed for a greater number of measurements in a more efficient and cost-effective way for detrital zircon studies (Pullen et al., 2014). Results from large-n studies have demonstrated that in some cases small-n data sets (n = 100) may not be sufficient in accurately characterizing all the minor detrital zircon populations within and between samples (Saylor and Sundell, 2016). Detrital zircon studies of Mesozoic–Cenozoic strata in the Cook Inlet basin in south-central Alaska, however, reveal that an expanded large-n (n = 300) data set does not show a significant difference from a more limited (n = 100) data set in the presence of both major and minor age populations, given the small number of age groups present within the distribution of that specific data set (Finzel et al., 2016; Finzel, 2019). Given that our modern river samples have a variety of age groups within the overall age distribution and cover multiple regional watersheds, a large-n data set might improve our results and interpretations, especially for minor sources of sediment.

With the acquisition of large-n data sets, the application of new statistical approaches to more accurately model the relative contributions of different sources in detrital zircon data sets may prove useful for better characterizing potential sources of sediment for south-central Alaska watersheds in future studies (e.g., Saylor et al., 2013; Sundell and Saylor, 2017). Lastly, the coupling of U-Pb geochronology with Hf isotope geochemistry on detrital zircons from modern river sediment in our study area can help reconstruct both the magmatic history as well as the crustal evolution of the southern Alaska convergent margin. Pepper et al. (2016) demonstrated the power of this coupled approach for reconstructing the magmatic and crustal evolution record along the Andean convergent margin in South America using modern river sediment. A similar approach along the southern Alaska convergent margin could be similarly insightful, given the rich and long history of magmatism, terrane accretion, and crustal orogenesis that has occurred.

CONCLUSIONS

In this study, we examine a regional U-Pb geochronology detrital zircon data set from sediment of the main trunk rivers of watersheds in the south-central part of the southern Alaska convergent margin. Key interpretations and insights include:

  • (1) Our data set demonstrates that the detrital zircon record from these modern trunk rivers is a faithful recorder of the major episodes of Mesozoic and Paleogene felsic magmatism along this convergent margin. Our integrated data set also tracks the magmatic record documented in the along-strike, plutonic rocks of the Coast Mountains batholith in British Columbia as well as the magmatic record documented in the along-strike, ancient forearc basin strata of the western Alaska Peninsula.

  • (2) Our data set does not, however, capture the complete record of the ca. 30 Ma to present-day magmatism represented by the Wrangell arc in the eastern part of south-central Alaska. This absence suggests that there may be a lag between the start of magmatism and the propagation of zircon-rich detritus into the main trunk rivers of a watershed. This finding may be important in studies of Neogene sedimentary strata and modern river sediment when utilizing U-Pb detrital zircon geochronology to identify the initiation of magmatism and related subduction processes along a convergent margin.

  • (3) Our findings show that detrital zircon data sets from any single individual watershed do not capture the entire record of Mesozoic and Paleogene magmatism along this convergent margin. We interpret this to reflect the distinct partitioning of watersheds by the high topography of south-central Alaska.

  • (4) An insight from the comparison between the integrated regional and individual watersheds is that for detrital zircon provenance studies of ancient sedimentary basins, a strategic approach is to sample several different geographic areas throughout the basin exposures to better reflect the different watersheds that may have supplied sediment to the basin. In some basin analysis studies, it is often attractive to focus on the one “type” stratigraphic section, especially if that section has independent and previously published age control. Results from our study suggest that such an approach may not always capture the complete record of sediment sources for a large ancient basin. In studies based on the one “type” section approach, the absence of a potential source of sediment may be more of a function of the individual ancient watershed extent rather than the absence of the source of sediment at a regional scale. This insight may be particularly important for sedimentary basins that formed in ancient collisional and accretionary convergent margin settings where the absence or presence of a specific source of sediment may be interpreted as evidence for the timing of terrane collision.

  • (5) Results from our study point out that in highly inaccessible areas, such as roadless, glacier-covered mountain ranges, with limited bedrock U-Pb igneous ages, modern river sediment can provide a record of the major magmatic events for that region. In south-central Alaska, Mesozoic and Paleogene felsic magmatic episodes occurred at 170, 150, 118, 95, 72, 58, and 36 Ma. The overall pattern of these ages suggests that felsic magmatism along the southern Alaska convergent margin was highly episodic, with the periodicity of these magmatic events ranging between ~14 and 32 m.y., with an average periodicity of ~22 m.y.

  • (6) Our study has several limitations and may be improved, especially with a larger number of samples analyzed, having the same number of detrital zircon age determinations from each individual watershed for more robust comparisons, and more complete sample coverage along the trunk rivers. At a first order, however, the study shows that detrital zircon studies from modern rivers are a powerful approach to delineate the magmatic and related tectonic processes along a long-lived convergent margin.

ACKNOWLEDGMENTS

This research was partially funded from National Science Foundation grants (EAR-1550034 and EAR-0910945) to Ridgway and from the Alaska Geological Society to Fasulo. We benefited from discussions with Chris Andronicos, Brandon Keough, and Nat Lifton. We appreciate constructive feedback from Chris Andronicos on an earlier version of the manuscript. Patrick Brennan, Emily Finzel, Andrea Stevens Goddard, and John Witmer helped with analyzing the samples. George Parker assisted in the collection of the river samples. We thank the staff scientists at the University of Arizona LaserChron Center for their assistance with geochronological analyses. This contribution was much improved by the constructive reviews of Matt Brueseke, Stephen Box, and the Geosphere Associate Editor.

1Supplemental Material. Table S1: U-Pb analytical results from detrital zircons from modern rivers in south-central Alaska. Please visit https://doi.org/10.1130/GEOS.S.14346974 to access the supplemental material, and contact editing@geosociety.org with any questions.
Science Editor: Andrea Hampel
Associate Editor: G. Lang Farmer
Gold Open Access: This paper is published under the terms of the CC-BY-NC license.