Catalog of Chinook Salmon Spawning Areas in Yukon River Basin in Canada and United States

Chinook Salmon Oncorhynchus tshawytscha return to the Yukon River in northwestern North America each summer, migrating to spawning destinations from the lower river to more than 3,000 km upstream from the Bering Sea (Evenson et al. 2009; Eiler et al. 2014). These returns are comprised of stream-type fish with most individuals rearing in freshwater for a year before migrating to sea (Healey 1991). Most returning adults range from 4 to 7 y of age (Healey 1991; Schumann and DuBois 2011; JTC 2016). Almost all of the younger fish (age 4 y) are males, whereas most of older fish (ages 6 and 7 y) are females. Escapements are usually composed of more males than females (Hyer and Schleusner 2005; Eaton 2016; JTC 2016), in part because of additive marine mortality of females that remain in the sea for 1–2 y longer than most males (Healey 1991; Bugaev and Shevlyakov 2007; Myers et al. 2009) and in part because of size-selective fishing mortality on larger individuals during years when substantial large-mesh gillnet fisheries are permitted in the river (Evenson et al. 2009; Hard et al. 2009; Bromaghin et al. 2011). After spawning, Chinook Salmon die, contributing nutrients from the sea to their natal streams (Cederhom et al. 1999).

Chinook Salmon have supported commercial, subsistence, recreational, and aboriginal fisheries within the Yukon River basin for decades but experienced notable declines in abundance beginning in the late 1990s (Evensen et al. 2009; Schindler et al. 2013; JTC 2016). These declines have led to a series of fishery restrictions, including basin-wide closures in 2014 and 2015, and several failures to achieve the internationally agreed upon passage goal into Canada (JTC 2016). Declines in average age and length-at-age of annual returns have also been reported (Bigler et al. 1996; Hyer and Schleusner 2005; Lewis et al. 2015), suggesting that Chinook Salmon stocks in the Yukon River are exhibiting signs of fishing-induced evolutionary changes (Hard et al. 2008; Bromaghin et al. 2011). Rural fishing communities are concerned the bountiful Chinook Salmon runs they depended on previously may not return (Loring and Gerlach 2010). In response to the observed declines in abundance, demographic changes, and subsequent fishery restrictions, there has been a tremendous increase in Chinook Salmon research throughout the basin, guided in part by expert panel recommendations (ADFG Chinook Salmon Research Team 2013; Schindler et al. 2013; JTC 2016). It is our perspective that a thorough accounting of spawning areas in the basin would be useful for many of these research efforts.

Despite the long history of Chinook Salmon research and management within the Yukon River basin (Pennoyer et al. 1965; Evenson et al. 2009; JTC 2016), there is no comprehensive account of spawning areas in tributary rivers and main-stem reaches. Our primary objective in this study was therefore to document and map Chinook Salmon spawning areas throughout the Yukon River basin by using a wide range of data sources. Our secondary objective was to highlight the largest populations by classifying spawning areas as either major or minor producers based on three indicators of abundance: 1) quantitative estimates of escapement (major producer if ≥500 fish, minor producer if <500 fish), 2) radiotelemetry-based proportions of annual production (major producer if ≥1% of the run, minor producer if <1% of the run), and 3) aerial survey index counts (major producer if ≥165 fish observed, minor producer if <165 fish observed).

The Yukon River basin is the largest in Alaska and the fifth largest by drainage area in North America (Revenga et al. 1998). It drains an area of more than 850,000 km², approximately 500,000 km² of which is in Alaska (Brabets et al. 2000). The Yukon River flows more than 3,000 km from its headwaters in northern British Columbia, Canada, to its mouth at the Bering Sea. Average annual flow near the Yukon River mouth is approximately 6,400 m³/s, although peak flow in early summer averages about 20,000 m³/s and extreme flow during flood conditions could exceed 25,000 m³/s (Curran et al. 2003). There are six major tributaries in the Yukon River basin (tributaries that contribute 5% or more to the total area drained and 5% or more to the total flow) including the Pelly, White, and Stewart rivers in Canada; the Tanana and Koyukuk rivers in the United States; and the Porcupine River that transects both countries (Brabets et al. 2000). The White and Tanana rivers originate in heavily glaciated mountains in the Wrangell, St. Elias, and Alaska ranges, and they are the primary sources of suspended sediment in the Yukon River. The Tanana and Porcupine rivers are the two largest tributaries in the basin, with drainage areas of more than 114,000 and 116,000 km², respectively. Despite the similarity in drainage areas, the Tanana River contributes approximately 20% of the total flow in the Yukon River basin, whereas the Porcupine River contributes less than 10%. Hundreds of small tributaries also flow into the Yukon River main stem including tundra-stained streams that meander across soft silty substrates, clear-water streams flowing over gravel, and turbid rivers that seasonally cascade from glaciers in surrounding mountains.

We subdivided the basin into nine geographic regions to illustrate the distribution of spawning areas within the river system. Regions in which all or nearly all of the drainage area are in the U.S. section of the basin included the lower Yukon River downstream from the Koyukuk River mouth (R1), middle Yukon River from the Koyukuk River mouth to the Tanana River mouth including the Koyukuk River drainage (R2), Tanana River drainage (R3), and upper Yukon River from the Tanana River mouth to the U.S.–Canada border (R4). The Porcupine River drainage (R5) flows through both U.S. and Canadian reaches of the basin. Regions in which all or nearly all of the drainage area are in the Canadian section of the basin included the northern Yukon River from the U.S.–Canada border to the White River mouth including the Stewart River (R6), the Yukon River main stem from the White River mouth to the Teslin River mouth including the White River and numerous smaller tributaries (R7), Pelly River drainage (R8), and the upper headwaters from the Teslin River mouth upstream including the Teslin River and other headwater rivers (R9).

We reviewed a large selection of published literature, agency reports, and databases and also other less formal documents for evidence of Chinook Salmon spawning in specific streams or main-stem reaches within the Yukon River basin. Locations were classified as spawning areas if there was consistent and compelling evidence presented, such as repeated observations of prespawning or spawning fish, the presence of redds or postspawning carcasses, or other similar observations. A spawning area could be a 10-km reach downstream from a lake outlet or a 100-km reach along the main-stem or tributary river. A stream might be classified as a single spawning area if the distribution of spawning Chinook Salmon seemed to be relatively continuous even when spawning was occurring along the main stem as well as in one or more tributaries. Alternatively, a stream with discrete aggregations of spawning fish (e.g., spawning in two or more separate forks) would be classified as more than one spawning area. Strong evidence of a spawning area included streams with a history of abundance monitoring projects such as weirs (e.g., Collin et al. 2002; Mears 2015; Wilson 2017), counting towers (e.g., Brase and Doxey 2006; Savereide and Huang 2014), mark–recapture projects (Skaugstad 1993), and visual aerial surveys (e.g., Barton 1984a; Nacho Nyak Dun First Nation 1998; Snow et al. 2012; State of Alaska 2016b). Genetics publications with lists of streams where baseline samples have been collected (e.g., Smith et al. 2005; Beacham et al. 2008; Flannery et al. 2012) as well as agency reports related to supplemental sampling (e.g., Conitz et al. 2012; Snow et al. 2012; MacDonald 2014) were similarly useful. The basin-wide telemetry study (Eiler et al. 2014), along with its associated localized aerial tracking surveys (e.g., Anderton 2003; Osborne et al. 2003; Mercer 2005) provided reliable information on spawning destinations, including several areas that had not been previously reported. We also reviewed information from the Alaska Anadromous Waters Catalog (Johnson and Litchfield 2016a, 2016b; State of Alaska 2016a) that consolidates a wide range of reports and observations in the U.S. portion of the basin, as well as historical and traditional knowledge accounts from terminal fisheries (e.g., Cox 1999; Anderton 2005b; Tobler and Marjanovic 2011) and less formal agency and Yukon River Panel reports (e.g., Walker 1976; Rost 1986; Besharah 2002; Klugie et al. 2003). In many cases, information from telemetry studies and other short-term projects helped strengthen the support for many of these other less definitive observations.

We classified Chinook Salmon spawning areas as major or minor producers by using three indices of abundance: 1) quantitative escapement estimates from weirs, counting towers, or sonar programs; 2) radiotelemetry distribution data; and 3) aerial or stream survey counts (Table 1). We focused on patterns of annual abundance as productivity indicators reasoning that high-quality spawning and rearing habitat will consistently produce more fish than low-quality habitat. Healey (1982) used similar abundance indicators (although telemetry data were not available) to classify 326 British Columbia spawning streams into four abundance categories ranging from less than 200 to more than 5,000 spawning fish. Data from some spawning areas in the Yukon River basin would have supported this level of analysis, but most would not.

Because we considered three indicators of abundance, classification criteria were specific to each indicator. Our highest quality abundance indicators in the Yukon River basin were the quantitative escapement projects and the radiotelemetry distribution data. Spawning streams with a series of annual escapement estimates from weirs or other high-accuracy quantitative methodologies were classified as major producers if they averaged 500 or more Chinook Salmon per year and minor producers if they averaged less than 500 per year. In deciding on this threshold value, we consulted some of the reviews on minimum viable population levels (e.g., Flather et al. 2011; Jamieson and Allendorf 2012) with the intent of using a biologically based value in common use, but we found very little guidance on the matter. In fact, Flather et al. (2011) concluded that there was no population size that would guarantee population persistence. We examined additional literature monitoring genetic health of Chinook Salmon populations based on the effective number of breeders per generation (e.g., Allendorf et al. 1997; Shrimpton and Heath 2003; Olsen et al. 2009), and although general guidelines were discussed, estimates of the relationship between effective number of breeders and annual census values were highly variable and provided little guidance for our purposes. Essentially, our thresholds are arbitrary values along a continuum.

Eiler et al. (2014) estimated the proportional contributions of numerous spawning streams to basin-wide Chinook Salmon production during the 3 y of project operation, 2002–2004. Proportional contributions were based on the distribution of radio-tagged fish weighted by run timing and progressively increased harvest exposure as fish moved upriver. We classified spawning areas as major producers if they averaged 1% or more of the total return and minor producers if they averaged less than 1%. This threshold value probably represents an annual escapement value greater than 500 spawning fish in most cases, but the relationship between spawning fish to number of radio tags varied among years (Mercer and Eiler 2004; Mercer 2005; Spencer et al. 2009) and becomes increasingly volatile with low numbers of tags. The 1% criterion for classification as a major producer standardized distribution across years of the project and avoided the high variation inherent in low values, even though this level represents a somewhat larger escapement than our quantitative classification criteria.

Aerial survey counts have been shown to be variable proportions of annual escapements (Jones et al. 1998; Hilborn et al. 1999). We developed an adjustment factor between aerial survey counts and accurate escapement estimates obtained from weirs, counting towers, or sonar programs by using paired data from four spawning streams in the Yukon River basin in which both types of data were available: East Fork Andreafsky, Gisasa, Chena, and Salcha rivers (Brase and Doxey 2006; JTC 2016). These four spawning streams provided 67 pairs in total of aerial survey counts and escapement estimates. A least-squares linear regression of these data revealed that there was a significant positive linear relationship between the two values (P < 0.001, R² = 42%). The average aerial survey to escapement proportion was 0.33 (SE = 0.025, range 0.013–0.961). Based on these results, spawning streams with a record of aerial survey counts were classified as major producers if they averaged 165 or more fish observed per year (500 × 0.33 = 165) and minor producers if they averaged less than 165 fish observed per year. Spawning areas without any indications of abundance were classified as minor producers.

Historical accounts of traditional fisheries in upriver reaches of the basin (Cox 1999; Anderton 2005b; Tobler and Marjanovic 2011) provided evidence of substantial Chinook Salmon fisheries that would not have occurred unless fish were present in great abundance. Because many of these historical accounts were from many decades ago, they were not used by themselves to classify streams as major or minor producers, but they verified the temporal persistence of certain highly productive spawning areas.

Although these three indices of production level were not directly comparable, the results for spawning areas that were classified using two or more methods were usually classified the same. In a few instances, different abundance indices for a particular spawning area were not in agreement. We then used a relative confidence hierarchy for classification decisions based on the type of information being considered, high-accuracy escapement projects (highest confidence), radiotelemetry proportional estimates (high confidence), and aerial survey counts (less confidence; Table 1).

We tabulated and mapped geographic information on Chinook Salmon spawning areas. Locations of the mouths of spawning streams, or the start points of main-stem spawning reaches, were recorded as latitude and longitude in decimal degrees by using WGS84 datum. We used Google Earth (https://www.google.com/earth/) to measure approximate distances in river kilometers (rkm) following a path fish could swim up the largest channels from the south mouth of the Yukon River (N latitude 62.57484, W longitude −165.01919), which is the largest distributary (McDowell et al. 1987), up the Yukon River main stem and various tributaries to each stream mouth and start point of main-stem reach. In addition, we identified coordinates of upstream locations within each spawning stream or main-stem reach for mapping clarity because identifying a stream of interest with a position at its confluence can be ambiguous. These upstream locations are for mapping clarity only and should not be interpreted as delineating a specific spawning site within a stream. Summarized and detailed geographic data, along with reference source information, are presented in tabular form and in a series of maps illustrating the distribution of Chinook Salmon spawning areas within the Yukon River basin.

We documented 183 Chinook Salmon spawning areas in the Yukon River basin, with 79 in the United States and 104 in Canada (Figure 1; Table 2). Evidence was available to classify 32 spawning areas as major producers, with 18 in the United States and 14 in Canada. The remaining 151 were classified as minor producers, with 61 in the United States and 90 in Canada. Spawning areas were identified from just upstream of the Yukon River delta (rkm 135), to the upper reaches of the Teslin River (rkm 3,204). Details on the distribution and classification of Chinook Salmon spawning areas are discussed below for the nine different regions of the basin. Table 3 summarizes these data among regions and some of the larger tributary rivers within regions. Table 2 provides detailed information for all documented spawning areas. A supplementary data file (Table S1, Supplemental Material) with this information is available in spreadsheet format as well.

Twenty-three spawning areas were identified in the lower Yukon River region in Alaska (Figure 2; Tables 2 and 3). Seven of these spawning areas were classified as major producers (Table 4) including the Andreafsky River (west fork), based on a 48-y record of aerial survey counts that averaged 1,130 fish (Barton 1984a; Estensen et al. 2012); the East Fork Andreafsky River, based on a 25-y record of weir-based escapements that averaged 3,748 fish (Estensen et al. 2012; Mears 2015); the Atchuelinguk River, based on a 13-y record of aerial survey counts that averaged 456 fish (State of Alaska 2016b); the Anvik River, based on 3 y of telemetry proportion data that averaged 3.2% of the run (Eiler et al. 2014); the Rodo River, based on 13 y of aerial survey counts that averaged 314 fish (State of Alaska 2016b); and the Nulato River (north fork) and South Fork Nulato River, based on aerial survey counts that averaged 789 fish (n = 33) and 580 fish (n = 35), respectively (Estensen et al. 2012). Nine years of counting tower and weir escapement estimates that averaged 1,978 fish into the combined Nulato River system support the major production level classifications of the two forks as well (Crawford and Lingnau 2004). Spawning Chinook Salmon were located in five tributaries of the Innoko River drainage during the basin-wide telemetry study (Eiler et al. 2014). However, the total estimated return to the Innoko River averaged <0.5% of the run and did not exceed 0.7% of the run annually, so all of the individual spawning areas in the drainage were classified as minor producers. All seven of the major producing spawning areas in the lower Yukon River region were in streams that flowed from headwaters in the Nulato Hills, a mountainous area northwest of the Yukon River.

Twenty-four spawning areas were identified in the middle Yukon River region in Alaska (Figure 3; Tables 2 and 3). Four of these spawning areas were classified as major producers (Table 4) based on weir escapement estimates including the Gisasa River, based on a 22-y record that averaged 2,289 fish (Carlson 2015); Henshaw Creek, based on a 14-y record that averaged 966 fish (McKenna 2014); South Fork Koyukuk River, based on a 2-y record that averaged 1,438 fish (Wiswar 1998); and Tozitna River, based on a 9-y record that averaged 1,381 fish (Beaudreault et al. 2010). This classification was corroborated for the Gisasa, Henshaw, and South Fork Koyukuk rivers based on aerial survey counts (Estensen et al. 2012; State of Alaska 2016b) and for the Tozitna River based on the basin-wide telemetry proportion data (Eiler et al. 2014). Other documented spawning areas in the Koyukuk, Melozitna, and Nowitna river drainages, as well as smaller main-stem tributaries, were classified as minor producers.

Eighteen spawning areas were identified in the Tanana River drainage in Alaska (Figure 4; Tables 2 and 3). Five of these spawning areas were classified as major producers (Table 4) including Barton Creek, based on a 9-y record of aerial survey counts that averaged 232 fish (State of Alaska 2016b); Chatanika River, based on a 7-y record of tower counts that averaged 997 fish (Brase and Doxie 2006); Chena River, based on a 28-y record of escapement that averaged 6,450 fish; Salcha River, based on a 28-y record of escapement that averaged 9,051 fish; and the Goodpaster River, based on a 9-y record of escapement that averaged 2,034 fish (Estensen et al. 2012; Savereide and Huang 2014). Telemetry proportion data suggest that the Salcha and Chena rivers were two of the largest spawning populations in the Yukon River basin, averaging 9.1 and 4.8% of the Chinook Salmon run, respectively (Eiler et al. 2014). Four of the five major producing spawning areas in the Tanana River region flow southwest from headwaters in the Yukon–Tanana uplands. All but one of the minor producing spawning areas are on the south side of the Tanana River, including a cluster of small tributaries of the glacial Kantishna River.

Eleven spawning areas were identified in the upper Yukon River region in Alaska (Figure 5; Tables 2 and 3). The Teedriinjik River (known as Chandalar River until its name was changed in 2015) was the only spawning area in the region classified as a major producer (Table 4). This river drains a portion of the eastern Brooks Range and has long been known as a major fall Chum Salmon Oncorhynchus keta spawning area (Barton 1984a; Daum and Osborne 1998). Aerial surveys of the Teedriinjik River were typically conducted in the fall (September and October) to coincide with the return of fall Chum Salmon (Barton 1984a), but too late to observe Chinook Salmon during their late July–August spawning period. Chinook Salmon have been known to spawn in the Teedriinjik River since the mid-1980s (Rost 1986; Daum 1989); however, Eiler et al. (2014) established the spawning area as a major producer based on 3 y of telemetry proportion data, averaging just over 3% of the entire run. A weir was operated for 4 y on Beaver Creek (Collin and Kostohrys 1998; Collin et al. 2002), which drains a portion of the north side of the Yukon–Tanana uplands. The average annual escapement was 187 Chinook Salmon, ranging from 114 to 315 fish, leading to a classification of the stream as a minor producer. Nine other spawning areas in the region were classified as minor producers.

Nine spawning areas were identified in the Porcupine River drainage in the United States and Canada (Figure 6; Tables 2 and 3). The Sheenjek River was the only spawning area in the region classified as a major producer (Table 4). Similar to the Teedriinjik River, the Sheenjek River drains a portion of the south slope of the eastern Brooks Range, known to be a major producer of fall Chum Salmon (Barton 1984a; Dunbar 2013); and most of the aerial surveys in the drainage reported by Barton (1984a) took place in September and October, too late for observing spawning Chinook Salmon. Rost (1986) conducted an aerial survey of the Sheenjek River in August 1985 and counted 45 spawning Chinook Salmon, establishing it as a spawning area. Eiler et al. (2014) subsequently reported an average production of almost 2% of the run in the Sheenjek River during the 3 y of the basin-wide telemetry study, establishing the stream as a major producer. Two other U.S. tributaries, the Coleen and Salmon Fork Black rivers, and six Canadian tributaries, including two streams in the Old Crow River drainage and four streams in the upper Porcupine River, were all classified as minor producers. The distribution of radio-tagged fish during the telemetry project (Anderton 2003, 2005a) and aerial survey counts of spawning fish and redds (Snow et al. 2012) indicate that the largest spawning area in the Canadian portion of the drainage is the Miner River. Future assessment projects may provide sufficient evidence to reclassify this stream as a major producer.

Twenty spawning areas were identified in the northern Yukon River region in Canada (Figure 7; Tables 2 and 3). Two spawning areas in the region were classified as major producers (Table 4), the Klondike and McQuesten rivers. The Klondike River was classified as a major producer based on an average escapement of 2,377 fish per year during three seasons of sonar operations (2009–2011; Mercer 2012). This classification was further supported by the telemetry proportion study, which suggested that the Klondike River produced an average of about 2% of the run during the 3 y of operation (Eiler et al. 2014). The main-stem McQuesten River was classified as a major producer based primarily on telemetry proportion data. Eiler et al. (2014) estimated that an average of 5.4% of the entire run returned to the Stewart River, but they did not estimate the contributions of individual spawning areas within the drainage. However, radio tracking surveys during the 3 y of this study showed that most of the radio-tagged fish returning to the Stewart River migrated to the McQuesten River (Osborne et al. 2003; Mercer and Eiler 2004; Mercer 2005). We therefore inferred that approximately 2.6% of the run was produced in the McQuesten River. Historical accounts of traditional fisheries and more recent survey data (Cox et al. 1997; Nacho Nyak Dun First Nation 1998; Tobler 2003a) support this classification as well. The other 18 spawning areas in the region were classified as minor producers. Historical accounts, however, suggest that Chinook Salmon returned to the Mayo River in large enough numbers to support a substantial aboriginal fishery before a dam was constructed in 1952 with no provision for fish passage (Cox 1999; Tobler and Miles 2004). The dam effectively blocked access to major spawning areas extending upstream approximately 90 km to the outlet of Mayo Lake. Chinook Salmon are still observed in the lower Mayo River (Osborne et al. 2003; Mercer 2005), but there is no indication that the numbers are large. A weir was operated on the Chandindu River for 4 y between 1998 and 2003, with annual escapements averaging 146 Chinook Salmon (range 85–239; Duncan 2000; McCready 2004), which led us to classify it as a minor producer. No other quantitative escapement monitoring projects have been conducted in this region.

Twenty-one spawning areas were identified in the Yukon River main-stem region in Canada (Figure 8; Tables 2 and 3). Four spawning areas and one main-stem reach in the region were classified as major producers including Tatchun Creek, the Nisling, Little Salmon, and Big Salmon rivers; and the Yukon River main stem between the mouths of the White and Teslin rivers, respectively (Table 4). Tatchun Creek is a small tributary of the Yukon River with a 27-y history of aerial and stream walk counts averaging 201 fish (range 52–643) and weir counts from 1997 to 1999 with an average escapement of 618 (range 252–1,198; Estensen et al. 2012). The Nisling River, a tributary of the White River, was classified as a major producer because the White River produced an average of 2.6% of the run during the 3 y of the telemetry study (Eiler et al. 2014), with approximately 51% of radiotagged fish located in the Nisling River (Mercer and Eiler 2004; Mercer 2005; Wilson 2006). We therefore inferred that approximately 1.3% of the run was produced in the Nisling River. The Little Salmon River was classified as a major producer based on a 27-y record of aerial survey counts that averaged 559 fish (Estensen et al. 2012). The Big Salmon River was classified as a major producers based on an 11-y record of sonar escapements that averaged 5,380 fish per year into the drainage (Mercer and Wilson 2016). Although five spawning areas have been identified in the drainage, aerial survey records over 43 y (Estensen et al. 2012) and aerial tracking surveys during the 3 y of the telemetry project (Osborne et al. 2003; Mercer and Eiler 2004; Mercer 2005) indicated that a large majority of fish spawn in the main-stem Big Salmon River. Further support for this classification was provided by the telemetry study, which estimated an average of 5.2% of the run produced in the Big Salmon River (Eiler et al. 2014), and aerial survey counts, which averaged 911 fish (Estensen et al. 2012). Spawning Chinook Salmon were observed in several reaches along the Yukon River main stem during the 1970s and 1980s (Walker et al. 1974; Walker 1976; Milligan et al. 1985), although in the large river environment it was not possible to know the extent or magnitude of spawning activity in the main stem. Large numbers of radio-tagged fish, however, remained in the middle and upper reaches of the main stem during the 3 y of the telemetry study, averaging 4.9% of the run annually (Eiler et al. 2014), indicating that the main stem was a major producer. The 17 remaining areas, including tributaries of the White River and the Yukon River main stem, were classified as minor producers.

Twenty-six spawning areas were identified in the Pelly River drainage in Canada (Figure 9; Tables 2 and 3). Three of these areas were classified as major producers (Table 4). Blind Creek is the only stream in the Pelly River drainage with long-term escapement data from weir counts (Wilson 2017). Average escapement over a 17-y period (1997–2016 with three seasons missed) was 588 fish (range 157–1,155). Eiler et al. (2014) estimated an average of 9.5% of the entire Yukon River run returned to the Pelly River during the basin-wide telemetry study. Radio-tagged fish were widely distributed throughout the drainage with notable concentrations observed in Blind Creek, Ross River, and South Macmillan River (Mercer and Eiler 2004; Mercer 2005). Of these three streams, Blind Creek received the fewest number of radio-tagged fish during the 2003 and 2004 seasons with escapements of 1,155 and 792 Chinook Salmon, respectively (Wilson 2017). Consequently, we assumed that comparable or greater numbers of fish returned to the Ross and South Macmillan rivers, which were therefore classified as major producers. This classification was further supported by a 15-y record of aerial survey counts in the Ross River (beginning in 1968) that averaged 331 fish (range 102–949; Estensen et al. 2012). Sparling (2003) observed 395 Chinook Salmon in the Earn River during a helicopter survey in 2002, suggesting that this stream might be a major producer. However, Barton (1984a) reported an average of 39 fish annually during four aerial surveys conducted between 1968 and 1983, so we classified the Earn River as a minor producer. Altogether, we classified 23 spawning areas as minor producers in the Pelly River drainage.

Thirty-one spawning areas were identified in the upper headwaters region in Canada (Figure 10; Tables 2 and 3). Four spawning areas were classified as major producers including the Teslin River downstream from Teslin Lake, the Wolf and upper Nisutlin rivers in the upper Teslin River drainage, and the Takhini River downstream from Kusawa Lake (Table 4). Eiler et al. (2014) estimated that an average of 9.8% of the entire run returned to the Teslin River during the basin-wide telemetry study. During all 3 y, the terminal locations of over half of the tagged fish were between the Teslin River mouth and the outlet of Teslin Lake (Osborne et al. 2003; Mercer and Eiler 2004; Mercer 2005). Consequently, this main-stem reach was classified as a major producer. The Wolf River was classified as a major producer based on a 36-y record of aerial surveys averaging 221 fish (Estensen et al. 2012). Similarly, the upper Nisutlin River was classified as a major producer based on a 41-y record of aerial surveys between the late 1960s and 2010, averaging 439 fish. The Takhini River was classified as a major producer based on a 16-y record of aerial survey counts, averaging 260 fish (Barton 1984a). The Morley River may have been a major producer in the past, but recent data do not support this classification. Barton (1984a) reported average counts of 166 fish (range 7–571 fish) during 12 aerial surveys between 1969 and 1983, which by itself would result in the area being classified as a major producer. During the basin-wide telemetry study however, only a single radio-tagged fish was located in the river (Osborne et al. 2003; Mercer and Eiler 2004; Mercer 2005), suggesting that the Morley River currently supports a smaller return. Another river with classification complications is Michie Creek, which is located upstream from the Whitehorse Rapids Dam (Gordon et al. 1960) and is supplemented annually with juveniles produced in the Whitehorse Rapids Fish Hatchery that are marked with coded wire tags and clipped adipose fins (Boyce 2000; Yukon Energy 2005; JTC 2016). During its 41 y of operation, the hatchery has released an average of approximately 150,000 juveniles annually, mostly upstream from the dam and mostly in Michie Creek (JTC 2016). Approximately 50% of the Chinook Salmon enumerated at the Whitehorse Rapids Fishway are thought to return to Michie Creek (Matthews 1999). The 1990–2015 average fishway count was about 1,300 fish, suggesting an average escapement of about 650 fish to Michie Creek (JTC 2016). However, during this same time period an average of 57% of the returns through the fishway were first generation hatchery fish without adipose fins, indicating that the Michie Creek spawning population is supported primarily by hatchery rather than wild production. We therefore have classified Michie Creek as a minor producer. Altogether, 27 spawning areas in the upper headwaters region were classified as minor producers.

Based on our review of the information, spawning Chinook Salmon were widely distributed throughout the Yukon River basin, with fish traveling from 135 km to more than 3,200 km to reach their final destinations. More than 180 spawning areas were identified, ranging from lower river tributaries to the upper headwaters. Eiler et al. (2014) observed that fish returning to the Canadian portion of the basin were more uniformly distributed, whereas those returning to U.S. portion were more clumped in their distribution, which is consistent with our findings (Figure 1).

Chinook Salmon spawning populations in the Yukon River basin seem to be consistent with a metapopulation structure as described by Policansky and Magnuson (1998) and Schtickzelle and Quinn (2007) for Pacific salmon and other anadromous fish. Salmon metapopulations are generally defined as groups of local spawning populations that experience a range of habitat qualities and environmental variables, yet are in close enough proximity to enable a small amount of gene flow through straying. Theoretically, this type of population structure is capable of surviving environmental perturbations that might lead to extinction of one or more local populations while providing a source for the colonization of suitable but vacant habitat (Hanski and Gilpin 1991; Harrison 1991; Hanski 1998). Average straying rates in stream-type Chinook Salmon have usually been estimated as <5%, with straying more common between nearby streams than more distant reaches (Quinn 1993; Candy and Beacham 2000; Westley et al. 2013). Investigations of genetic diversity and structure among Chinook Salmon spawning populations in the Yukon River indicate that geographic structuring is most pronounced at the country of origin level (upper reaches of the basin vs. lower reaches), less pronounced among regional spawning aggregations, and weakest among local spawning populations within a region (Smith et al. 2005; Beacham et al. 2008; Flannery et al. 2012). All these lines of evidence are consistent with a metapopulation structure in the Yukon River basin.

Chinook Salmon returning to the Yukon River basin originated in a mixture of major and minor producing spawning areas. We based the assessment on three different indices of spawning abundance. Healey (1982) used similar indices (i.e., escapement estimates and aerial surveys, but no telemetry-based distribution information) to classify 326 Chinook Salmon spawning populations in British Columbia into four size categories (<200, 200–1,000, 1,000–5,000, and >5,000 fish) and concluded that a majority of populations in British Columbia were very small (49% <200 fish, 81% <1,000 fish). For comparative purposes, we reclassified the mean stream escapement estimates presented by Healey (1982; Table 10) by using our numerical size criteria (≥500 = major producer, <500 = minor producer) and found that we would have classified 67% (n = 218) of the spawning areas in British Columbia as minor producers and 33% (n = 108) as major producers. Our data suggest that small populations are even more prevalent in the Yukon River with 83% (n = 151) of spawning areas classified as minor producers and only 17% (n = 32) as major producers.

These 183 Chinook Salmon spawning areas identified here represent the current knowledge for the Yukon River basin, but there is nothing absolute about this number or the relative production levels of these localized populations. It is unlikely that additional major spawning aggregates exist within the basin given the extensive coverage by aerial surveys (Barton 1984a; State of Alaska 2016b), the pipeline assessment work conducted in the upper basin during the 1970s and 1980s (e.g., Walker 1976; Beak Consultants Limited 1977), the contributions of Canadian First Nations conducting fisheries research in familiar drainages (e.g., Nacho Nyak Dun First Nation 1998; Sparling 2003), the basin-wide telemetry study (e.g., Anderton 2005a; Eiler et al. 2014), and the on-going compilation of information provided by the Alaskan Anadromous Waters Catalog (Johnson and Litchfield 2016a, 2016b; State of Alaska 2016a). However, it is likely that some minor spawning aggregations remain undocumented due to turbidity in certain reaches of the basin that prevent viewing during aerial surveys, the remote nature and limited access of many tributaries, and other detection challenges. Other authors may also split or combine spawning areas or classify escapement levels differently and achieve different totals. Similarly, the relative size of these spawning populations may change over time in response to changing habitat conditions or improved assessment methods and abundance estimates. Streams that are currently vacant may eventually be colonized (Schtickzelle and Quinn 2007), discovered, and added to this catalog. Alternatively, the number of fish currently using established spawning areas may decline to the point where they become vacant. Nonetheless, the Chinook Salmon spawning areas cataloged here provides a baseline that makes it possible to assess both spatial and temporal changes within the basin.

Please note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental material. Queries should be directed to the corresponding author for the article.

Table S1. Chinook Salmon Oncorhynchus tshawytscha spawning areas compiled in 2017 for the Yukon River basin ordered by the country in which a spawning area is found, region, and drainage listed sequentially from the Yukon River mouth upstream to the final spawning destinations.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S1 (54 KB XLSX).

Reference S1. Beak Consultants Limited. 1977. A survey of fall spawning fish species in waterbodies within the influence of the proposed Alaska Highway Pipeline in Yukon Territory, 1977. Report prepared for Foothills Pipe Lines (Yukon) Ltd., Calgary, Alberta.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S2 (5156 KB PDF).

Reference S2. Beaudreault S, Post J, Kretsinger C, Karlen B. 2010. Abundance and run timing of adult salmon in the Tozitna River, Alaska, 2001–2009. Bureau of Land Management, Fisheries Resource Monitoring Program Report 07-208, Fairbanks, Alaska.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S3 (895 KB PDF).

Reference S3. Brabets TP, Wang B, Meade RH. 2000. Environmental and hydrologic overview of the Yukon River basin, Alaska and Canada. U.S. Geological Survey, Water-Resources Investigations Report 99-4204.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S4 (6275 KB PDF). Also available at http://pubs.usgs.gov/wri/wri994204/pdf/wri994204.pdf (May 2017).

Reference S4. Brinkman AR, McCall PL, Haddix JA. 2015. Documenting anadromous headwaters in the Tanana Flats of interior Alaska, in response to proposed road development. U.S. Army Garrison, Fort Wainwright Alaska, Center for Environmental Management of Military Lands, Colorado State University, Fort Collins.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S5 (5615 KB PDF).

Reference S5. Carlson JG. 2015. Abundance and run timing of adult salmon in the Gisasa River, Koyukuk National Wildlife Refuge, Alaska, 2014. U.S. Fish and Wildlife Service, Alaska Fisheries Data Series 2015–3, Fairbanks.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S6 (840 KB PDF). Also available at http://www.fws.gov/alaska/fisheries/fish/Data_Series/d_2015-3.pdf (May 2017).

Reference S6. Collin N, Kelly L, Kostohrys J. 2002. Adult salmon runs and streamflow data at a resistance board weir on Beaver Creek, Alaska, 1998–2000. U.S. Department of the Interior, Bureau of Land Management, BLM–Alaska Open File Report 85, Anchorage.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S7 (1044 KB PDF).

Reference S7. Collin N, Kostohrys J. 1998. Enumeration of adult salmon and hydrologic data at a resistance board weir on Beaver Creek, Alaska, 1996–1997, U.S. Department of the Interior, Bureau of Land Management, BLM–Alaska Open File Report 70, Anchorage.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S8 (6393 KB PDF).

Reference S8. Curran JH, Meyer DF, Tasker GD. 2003. Estimating the magnitude and frequency of peak streamflows for ungaged sites on streams in Alaska and conterminous basins in Canada. U.S. Geological Survey Water-Resources Investigations Report 03-4188, Anchorage.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S9 (3534 KB PDF). Also available at http://pubs.usgs.gov/wri/wri034188/pdf/wri034188_v1.10.pdf (May 2017).

Reference S9. Daum D. 1989. 1987 Chandalar River Chinook Salmon telemetry results. U.S. Fish and Wildlife Service, Memorandum Trip Report, Fairbanks, Alaska.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S10 (229 KB PDF).

Reference S10. Esse DA, Kretsinger CF. 2009. Abundance and run timing of adult salmon in Clear Creek, Hogatza River, Alaska, 2000–2005. Bureau of Land Management, Central Yukon Field Office, Program Report DIFR BLM/AK/F03000-6500/FY09/1120/07, Fairbanks.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S11 (1101 KB PDF).

Reference S11. Hyer KE, Schleusner CJ. 2005. Chinook Salmon age, sex, and length analysis from selected escapement projects on the Yukon River. U.S. Fish and Wildlife Service, Alaska Fisheries Technical Report 87, Anchorage.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S12 (1840 KB PDF). Also available at http://www.fws.gov/alaska/fisheries/fish/Technical_Reports/t_2005_87.pdf. (May 2017).

Reference S12. McDowell S, Signorini S, Pace S, Borchardt J. 1987. Yukon Delta processes: physical oceanography. National Oceanic and Atmospheric Administration, Outer Continental Shelf Environmental Assessment Program 57:335–661.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S13 (12329 KB PDF).

Reference S13. McKenna B. 2014. Abundance and run timing of adult salmon in Henshaw Creek, Kanuti National Wildlife Refuge, Alaska, 2014. Tanana Chiefs Conference, Fisheries Resource Monitoring Program Report 14-209, Fairbanks, Alaska.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S14 (2048 KB PDF).

Reference S14. McKenna B, DeCovich N. 2016. Chinook Salmon tissue sample collections for the analysis of Yukon River DNA baseline samples in Alaska, 2015. Yukon River Panel, Report CRE-78-14B, Whitehorse, Yukon.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S15 (1114 KB PDF).

Reference S15. Mears JD. 2015. Abundance and run timing of adult Pacific salmon in the East Fork Andreafsky River, Yukon Delta National Wildlife Refuge, Alaska, 2014. U.S. Fish and Wildlife Service, Alaska Fisheries Data Series 2015–5, Fairbanks.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S16 (949 KB PDF). Also available at https://www.fws.gov/alaska/fisheries/fish/Data_Series/d_2015-5.pdf (May 2017).

Reference S16. Melegari JL, Osborne BM. 2008. Evaluation of sonar for enumeration of Chinook Salmon in the Chandalar River, Yukon Flats National Wildlife Refuge, Alaska, 2004–2006. U.S. Fish and Wildlife Service, Alaska Fisheries Data Series 2008–1, Fairbanks.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S17 (441 KB PDF). Also available at http://www.fws.gov/alaska/fisheries/fish/Data_Series/d_2008_1.pdf (May 2017).

Reference S17. Rost PJ. 1986. Aerial surveys for summer and fall salmon in the upper Yukon River drainage, 1985. U.S. Fish and Wildlife Service, Fairbanks Fishery Resources Progress Report FY86-9, Fairbanks, Alaska.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S18 (553 KB PDF). Also available at http://www.fws.gov/alaska/fisheries/fish/Progress_Reports/p_1986_09.pdf (May 2017).

Reference S18. Smith P. 1996. Chinook Salmon (Oncorhynchus tshawytscha) spawning and juvenile rearing habitat in Janet Creek: inventory, utilization and restoration. Report prepared for the Stewart Valley Salmon for the Future Society and the Yukon River Salmon Restoration and Enhancement Fund, Whitehorse, Yukon.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S19 (977 KB PDF).

Reference S19. VanHatten GK. 2005. Abundance and run timing of adult salmon in the Kateel River, Koyukuk National Wildlife Refuge, Alaska, 2001–2003. U.S. Fish and Wildlife Service, Alaska Fisheries Data Series 2005–2, Fairbanks.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S20 (229 KB PDF). Also available at http://www.fws.gov/alaska/fisheries/fish/Data_Series/d_2005_2.pdf (May 2017).

Reference S20. Wiswar DW. 1998. Abundance and run timing of adult salmon in the South Fork Koyukuk River, Kanuti National Wildlife Refuge, Alaska, 1997. U.S. Fish and Wildlife Service, Alaska Fisheries Data Series 98-1, Fairbanks.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S21 (587 KB PDF). Also available at http://www.fws.gov/alaska/fisheries/fish/Data_Series/d_1998_1.pdf (May 2017).

Reference S21. Wynne-Edwards VC. 1947. The Yukon Territory. Chapter II, pages 6–20 in Northwest Canadian fisheries surveys in 1944–45. Fisheries Research Board of Canada, Bulletin 72.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-045.S22 (954 KB PDF).

We appreciate the dedication of all the people who have conducted research on Chinook Salmon in the Yukon River basin and documented their findings. These documents are the basis of this paper, and it would not have been possible without them. Online access to agency reports and other fisheries data provided by the Alaska Department of Fish and Game, U.S. Fish and Wildlife Service, Department of Fisheries and Oceans Canada, Yukon River Panel, and Alaska Resources Library and Information Services were invaluable. The tedious task of determining upstream distances to spawning stream mouths throughout the basin was performed primarily by B. Carter and O. Edwards (U.S. Fish and Wildlife Service). Reviews by J. Adams (U.S. Fish and Wildlife Service); J. Conitz, B. Borba, and A. Padilla (Alaska Department of Fish and Game); M. Bradford (Fisheries and Oceans Canada); the Associate Editor of the JFWM; and two anonymous reviewers improved the focus, quality, and clarity of the manuscript.

Any use of trade, product, website, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Sulukna River: Gerken J, Esse D. 2016. In July 2007, Jon Gerken (U.S. Fish and Wildlife Service) and Dave Esse (Bureau of Land Management) conducted a boat survey of river conditions on the Sulukna River in the upper Nowitna River drainage in preparation for an upcoming Inconnu Stenodus leucichthys spawning habitat project. They observed between 50 and 100 prespawning Chinook Salmon along the course of the river and captured and photographed several of them.

Sethkokna River: Kretsinger C, Karlan B. 2016. Following a July 30, 2014, aerial helicopter fish survey of the Sethkokna River in the upper Nowitna River drainage, Carl Kretsinger and Bob Karlan (Bureau of Land Management) reported observing 98 Chinook Salmon along the course of the river, all associated with spawning redds. They observed approximately 40 redds.

Selwyn River: Department of Indian Affairs and Northern Development, Department of Fisheries and Oceans, Department of the Environment, and Yukon Territorial Government. 1985. An Environmental Review of Big Creek, Yukon as related to Placer Mining, Prepared for Placer Research and Development Committee. According to Al von Finster (Fisheries and Oceans Canada), who participated in the review, spawning Chinook Salmon were observed in the Selwyn River.

Kirkland Creek: In 1995, Al von Finster (Fisheries and Oceans Canada) filed a 5-page memo with Fisheries and Oceans Canada describing observations of Chinook Salmon spawning in Kirkland Creek during an overflight of Division Mt. Coal and potentially affected water bodies and stream courses.

Rose River: On August 15, 1996, Al von Finster (Fisheries and Oceans Canada) filed a 1-page memorandum with the Habitat and Enhancement Branch, Yukon and Transboundary Rivers Division, Department of Fisheries and Oceans, documenting his observation of Chinook Salmon spawning in the Rose River.

Klinkit Creek: In the late 1980s, helicopter pilots supporting mineral exploration in the Klinkit Lake area reported observing Chinook Salmon spawning immediately downstream of the lake outlet to habitat biologist Al von Finster (Fisheries and Oceans Canada).

Mendenhall River: On August 24, 2003, Paul Sparling and Mark Connor filed a preliminary report with the Yukon Department of Highways and Public Works on a fisheries utilization assessment conducted at the Alaska Highway crossing on the Mendenhall River, August 19th & 20th, 2003. They reported that they captured several fish species including a juvenile and an adult Chinook Salmon.