Freshwater feeding by adult chum salmon Oncorhynchus keta in the Mackenzie River, Northwest Territories

As distributional shifts of marine species into the Arctic become more frequent, these range-expanding individuals may exhibit unique traits that facilitate survival and establishment in new places. In the western Canadian Arctic, increasing incidental harvest of range-expanding Pacific salmon Oncorhynchus spp. is monitored in a collaborative effort with harvesters, Indigenous leadership, and Fisheries and Oceans Canada. As harvesters have observed both feeding behavior and prey in these salmon, it is important to investigate the hypothesis that adult salmon cease feeding in the Mackenzie River, Northwest Territories. Here, we focused on adult chum salmon Oncorhynchus keta harvested during 2017 and 2019 to explore incidences of freshwater feeding in the Delta, Mainstem, and in Great Bear and Great Slave lakes. Stomach content analyses of 150 chum salmon revealed that there were nearly equal numbers of chum with stomach contents vs. empty stomachs (70 vs. 80). When there were contents, stomach fullness remained low (fullness index range: 0.00071%–0.71%) but was the highest in the Delta. Identifiable freshwater prey was present in three chum salmon (2%). Foraging on freshwater fish was only recorded in Great Slave Lake, where two chum salmon each ate one forage fish (Coregonus spp.). Foraging on insects was only recorded in Great Bear Lake, where one chum salmon ate two corixids. Fourteen additional chum salmon stomachs (9.3%) contained prey remnants. Chum salmon stomach contents in all areas included plant materials, which were more prevalent in the Mainstem. Stomachs in all areas also included synthetic fibers, more of which were found in 2017 than in 2019. Although rare and likely opportunistic, the clearest evidence of recent feeding was in chum salmon that swam the furthest upstream. This deviation from the established life-history trait associated with a cessation of feeding in adult Pacific salmon in fresh water represents variation that may be important for these range-edge chum salmon, especially considering the substantial migration distance associated with accessing and then ascending the Mackenzie River.

Introduction

Northward range expansions of subarctic marine species are becoming increasingly common as ocean temperatures warm (Huntington et al., 2020; Stafford et al., 2022; Alabia et al., 2023). Predicting the establishment and proliferation of these opportunistic individuals in new places, however, remains difficult (Dunmall et al., 2016; Bilous and Dunmall, 2020). Habitat requirements and thermal tolerances can shift with life-history stage (Dunmall et al., 2016), and phenotypic traits that influence range expansions, such as those that increase dispersal or reproductive ability, can also appear at the range edge at all life-history stages (Chuang and Peterson, 2016). Similar to invasive species, range-expanding species that arrive in new environments are exposed to selective pressures, such as new climatic conditions and novel species interactions, which may promote rapid adaptation of traits to match local habitats and native species (reviewed in Moran and Alexander, 2014). These range-expanding species may thus possess genetic, morphological, physiological, life-history, or behavioral traits that enhance their survival, functioning, dispersal, and impact, or can rapidly evolve such traits soon after colonizing (Chown and McGeoch, 2023; Pyper et al., 2024). Developing a better understanding of these trait changes in range-expanding species would inform predictions of their potential survival and establishment in new areas, which would help in guiding management decisions regarding biodiversity change (Carbonell et al., 2021).

Range-expanding Pacific salmon Oncorhynchus spp. are following corridors of thermally suitable marine habitats and are being increasingly harvested in the western Canadian Arctic (Dunmall et al., 2024). Chum salmon O. keta are the most commonly harvested Pacific salmon species in the Canadian Arctic, and adults have been caught for generations by subsistence harvesters targeting Arctic species (Stephenson, 2006; Dunmall et al., 2013, Dunmall et al., 2021). The occurrence of these incidental harvests has been monitored using a community-led initiative called Arctic Salmon since 2000 (Dunmall and Reist, 2018). Together with harvesters from across what is now known as the Northwest Territories, Canada, we are monitoring the freshwater harvest of chum salmon in the Mackenzie River system, an area spanning over 1,700 km from the Delta to Great Slave Lake (Dunmall et al., 2013, 2021). While most chum salmon are caught incidentally in subsistence gillnet fisheries targeting Arctic fishes, the occasional chum salmon are caught in the recreational fishery using a rod and reel (e.g., Gibbons, 2016). The appearance of these salmon is concerning to local harvesters, as increasing occurrences provide a tangible example of the impacts of climate change on cultures and ecosystems (Dunmall et al., 2024) and because they may interact with key subsistence fish species, such as Arctic char Salvelinus alpinus and Dolly Varden char S. malma (Chila et al., 2022).

The diet of Pacific salmon can be a useful indicator for understanding prey biodiversity, potential niche overlap, and environmental conditions (Davis et al., 2009; Qin and Kaeriyama, 2016; Graham et al., 2020). In the marine environment, Pacific salmon feed opportunistically, with Chinook O. tshawytscha and coho O. kisutch salmon focusing on fish as prey, while sockeye O. nerka, pink O. gorbuscha, and chum salmon have a more variable diet (Satterfield and Finney, 2002; Brodeur et al., 2007; Urawa et al., 2018). In fresh water, Pacific salmon are known to cease feeding entirely during their spawning migration, with the exception of occasional egg consumption by spawning coho O. kisutch salmon chum salmon, and Chinook salmon O. tshawytscha (Garner et al., 2009), at levels insufficient to obtain biologically significant energy gains (Armstrong, 2010). Harvesters in the western Canadian Arctic, however, have observed food in the non-atrophied esophagi and stomachs of salmon caught in fresh water, as well as salmon striking readily at fishing lures. These observations highlight the need to test the hypothesis that chum salmon cease feeding after entering the Mackenzie River system.

Our goal here is to explore the occurrence of freshwater feeding in adult chum salmon caught in the Mackenzie River system, including Great Bear and Great Slave lakes. Using the stomachs of chum salmon provided to the Arctic Salmon community-led monitoring program (Dunmall and Reist, 2018), our objectives were to categorize stomach contents and identify prey items to the lowest taxonomic level possible and to assess stomach fullness and explore potential differences in contents related to harvest location. We continued the collaborative approach of the Arctic Salmon program (described in Dunmall et al., 2024) to address these community-driven research questions, thereby supporting existing Indigenous and local knowledge of freshwater feeding in chum salmon caught in the Mackenzie River and assessing the potential for diet overlap among range-expanding and key endemic fishes. The information derived here will help to inform decision-making regarding the management of range-expanding species, and also the conservation of Arctic fish species enduring rapid environmental change.

Methods

Arctic salmon

In collaboration with Fisheries and Oceans Canada (DFO), communities across the western Canadian Arctic have been monitoring generally increasing salmon incidentally caught in subsistence harvests since 2000 in a program called Arctic Salmon (Dunmall and Reist, 2018). In this program, harvesters have the opportunity to provide a whole salmon or salmon head, with harvest date and location information, in exchange for a grocery gift card, and the salmon are used to then address community-driven questions about changing biodiversity. The research goal to assess freshwater feeding in chum salmon was derived from discussion among harvesters, community offices, co-management boards, and Arctic Salmon program DFO researchers as part of the collaborative process (Dunmall et al., 2024), and we received eight letters of support from land claim co-management boards, community leadership, and territorial management agencies during project development. To support the collaborative nature of this project, preliminary results were presented to the Fisheries Joint Management Committee (FJMC) and the Gwich'in Renewable Resources Board (GRRB), which are co-management boards, a total of five times from January 2023 to October 2024. In addition, results were communicated in a newsletter (April 2024) and a Research Bulletin (January 2025; NWT CIMP, 2025). A letter (Figure S1) and the Research Bulletin were sent to leadership in 21 communities throughout the Mackenzie River system in July 2025, providing specific results, seeking feedback, offering a virtual meeting to facilitate discussions, and inviting co-authorship to recognize their valuable contributions throughout the process. While none expressed interest in authorship, they remain instrumental to the development of these outcomes and are included in the acknowledgments. A completed full draft of the manuscript was provided to the FJMC, GRRB, and Ɂehdzo Got’ı̨nę Gots’é Nákedı (Sahtú renewable Resources Board) co-management boards. co-management boards for review in July 2025. Suggestions received from the FJMC were incorporated; the GRRB expressed their appreciation for the opportunity to review and for keeping them updated on the research.

Study area

The Mackenzie River Basin is the largest in Canada and the second largest in North America, with the length of its rivers extending over 4,000 km from its headwaters to the Arctic Ocean (Camsell and Malcolm, 1919). Called Kuukpak (in Inuvialuktun), Dehcho (in Dene Zhatıé), Grande Rivière (in Michif), Deho (in Dene Kǝdǝ́), and Nagwichoonjik (in Dinjii Zhu’ Ginjik), which all translate as “big (or great) river” (Natural Resources Canada, 2025), the Mackenzie River has been essential to Indigenous Peoples for thousands of years (Mackenzie River Basin Board, 2025). Here, we focused on the Mackenzie River within the Northwest Territories, from Great Slave Lake downstream to the Mackenzie River Delta, and including Great Bear Lake.

Whole chum salmon caught near 13 communities on the Mackenzie River and provided to the Arctic Salmon program in 2017 and 2019 were used (Figure 1). As salmon harvest is variable year to year (Dunmall et al., 2024), we focused on whole chum salmon caught in 2017 and 2019, which represent the highest harvest years on record. The harvest location provided for each chum salmon (Figure 1) was categorized into four regions to facilitate assessments of stomach contents relative to distance upstream: the Mackenzie River Delta (including Aklavik, Inuvik, Ft. McPherson, and Tsiigehtchic), the Mainstem of the Mackenzie River (including Ft. Good Hope, Norman Wells, Wrigley, Ft. Simpson, and Ft. Liard), Great Bear Lake (including Dé l̨ ın̨e and Great Slave Lake (including Ft. Providence, Yellowknife, Hay River, and Ft. Resolution). While the chum salmon used in these analyses were harvested in these listed communities, we also recognize and appreciate that chum salmon have also been harvested in other communities, including those outside our study region.

Figure 1

Figure 1. Map of chum salmon Oncorhynchus keta harvest locations in the Mackenzie River system, Northwest Territories, Canada for this study (data in Table 1).

Fish processing

Harvested chum salmon were frozen whole and shipped to the Fisheries and Oceans Canada Freshwater Institute in Winnipeg, Manitoba. All fish were thawed and weighed, sexed by identification of male or female gonads, and assessed as displaying silver or spawning morphology. Stomachs were obtained by cutting the fish along the sagittal plane from behind the mouth to the vent and removing the entire gastrointestinal tract, including the esophagus and intestine, which was frozen to −20 °C for later analyses. Frozen stomachs were sent to North/South Consultants (NSC) in Winnipeg, Manitoba, in 2023. The stomachs were thawed, categorized as empty, nearly empty, or partially full; the stomach and contents [whole stomach (WS)] were weighed. The chum salmon were assessed for differences in sex or size by harvest year and location.

Stomach content analysis

Stomach contents were removed by gentle scraping and rinsing under cold water into a 250-µm sieve, making certain all contents were collected. Sieved contents were placed in glass Petri dishes, examined under a dissecting microscope, and classified into categories of identifiable prey or prey remnants, digested material, plant material, and synthetic fibers, which were then weighed. The empty stomach (ES) was also weighed. The weight of stomach contents (SC) was determined by SC = WS − ES. All identifiable prey were sorted to the lowest taxonomic groups possible, given their state of digestion, enumerated, weighed ( ± 0.001 g), stored separately in labeled Whirl-Pak bags, and refrozen. The taxonomic identification of contents was conducted using all available up-to-date information, relevant keys, original literature descriptions, other materials, and detailed, high-resolution photographs from previous dietary analyses conducted by NSC and compiled into taxon-specific reference binders and photo libraries. Similar to previous diet studies (e.g., Faulkner et al., 2025), hard prey remnants (e.g., fish scales and bones) and other categories of stomach contents (digested material, plant material, and synthetic fibers) were recorded but were excluded from analyses of freshwater prey items, as these items may accumulate in stomachs over time (Santos et al., 2001). As identifiable prey items were scarce, categorizing prey into taxonomic groups and using metrics to quantify prey types [e.g., index of relative importance (IRI)] were not applicable. Instead, identifiable prey and prey remnants were tabulated per fish by region and year.

The entirety of the stomach contents (all categories) was explored to assess differences in contents among regions and years. Relative stomach mass (RSM) was calculated by dividing the weight of the empty stomach (ES) by the weight of the fish (W) and then converting to a percentage. To determine if RSM varied by year, location, or sex, an analysis of variance (ANOVA) was carried out (F-statistic). Stomach fullness was assessed using the fullness index derived by Hureau (1970):

$\text{FI}=\text{FW}/\text{W}\times 100$, where the fullness index (FI) is calculated by the weight of contents (FW) divided by the weight of the fish (W) and then converted into a percentage. To determine if FI varied by year, sex, or location, an ANOVA was carried out. Differences among the stomach content categories (identifiable prey, digested remains, plant material, and synthetic fibers) were assessed using a multiple logistic regression (Wald statistics), with year, sex, location, and presence/absence. For chum with stomach contents, an analysis of covariance (ANCOVA) was carried out on the specific materials (covariate: fish mass; note, when contents were present but <0.001 g, 0.001 g was used for statistical testing). Grubbs’ test was used to identify outliers. For all analyses of variances, data were 1) tested for normality using a Shapiro–Wilk test and then 2) checked for homogeneity of variances using a Brown–Forsythe test. All values passed except relative stomach mass, which was log-transformed. SigmaPlot 15 (Systat; CA, USA) was used for statistical analyses and graphing. R (R Core Team, 2021) was used to produce violin plots. Adobe Illustrator (Systat; CA, USA) was used for graphics.

Results

A total of 150 chum salmon harvested in the Mackenzie River in 2017 and 2019 were assessed for stomach contents (Table 1). The chum salmon sampled were caught from August to October in 2017 and from June to October in 2019 (Table 1). While the catch month was not recorded for 19 other salmon in 2019, they were also harvested within this timeframe, as that is the fishing season. All had transitioned into displaying morphological characteristics consistent with spawning except for seven salmon (five in the Delta and two in the Mainstem) that were still ocean silver in color. None were spawned out. More males than females were harvested across all locations and years (ANOVA; mean difference: 6.75 more males; F_1,15 = 22.1, p < 0.001) (Figure 2A), and the proportion of males to females remained similar among locations and years (average 2.6 to 1; 2017, X²³ = 1.132, p = 0.769; 2019, X²³ = 3.94, p = 0.268). Chum salmon from 2017 were heavier than those from 2019 (ANOVA; mean difference: 283 g; F_1,149 = 6.03, p = 0.015), and males were larger than females (ANOVA; mean difference: 643 g; F_1,149 = 31.2, p < 0.001) (Figure 2B). Chum salmon mass also differed by sampling location (ANOVA; F_3,149 = 7.20, p < 0.001), as chum from Great Bear Lake (GBL) were smaller than both Great Slave Lake (GSL) and Mainstem chum (mean differences: 750 and 675 g, respectively; t-values 3.84 and 3.95, respectively, both p < 0.001). However, within each year, there were no sampling location or sex-specific differences in chum mass (p-values of 0.893 and 0.304, respectively).

Table 1

Table 1. Number of harvested chum salmon Oncorhynchus keta in 2017 and 2019, by month, year, and harvest region in the Mackenzie River, assessed for stomach contents.

Figure 2

Figure 2. The number of females and males, their mass, relative stomach mass, and stomach fullness index of adult chum salmon harvested in the Mackenzie River in 2017 and 2019. GBL, Great Bear Lake; GSL, Great Slave Lake; Main, Mainstem; Delta, Mackenzie River delta. Locations provided in Figure 1 and Table 1. Asterisk indicates difference across years; dissimilar uppercase letters indicate difference within years; lowercase letters indicate difference across years. Similar numbers indicate differences within a location across years.

Stomach contents were examined to determine the type and occurrence of prey. Combined across 2017 and 2019, there were roughly equal numbers of chum with stomach contents vs. empty stomachs (70 vs. 80). This differed within years, however, as 65.1% of the chum sampled in 2017 had stomach contents vs. 32.1% of the chum in 2019 (Wald, 16.8, p < 0.001). Sampling location and sex were not factors (p-values 0.128 and 0.766, respectively). Identifiable prey or prey remnants were present in only 11% of all salmon assessed (n = 12 salmon in 2017 and n = 5 salmon in 2019; Table 2). Excluding those with hard prey remnants (e.g., fish scales and bones) and other categories of stomach contents (digested material, plant material, and synthetic fibers), only three males had recently consumed freshwater prey items (Table 2). This included two chum salmon from 2019 (corixid; unidentified small fish) and one from 2017 (Coregonus sardinella) (Table 2). While the forage fish eaten in 2019 was unidentifiable to species, it was likely either another least cisco C. sardinella or a lake herring Coregonus artedi, as those are the two most prevalent forage fish in Great Slave Lake (Zhu et al., 2017). Interestingly, all of these chum were harvested in either GBL (in 2019) or GSL (in 2017); no chum salmon harvested in the Delta or Mainstem had identifiable freshwater prey in their stomach contents. Only chum salmon caught in GSL had entire fish in their stomachs. Therefore, the chum salmon that swam the furthest in fresh water provided the clearest evidence of freshwater feeding, which was on aquatic insects and fish.

Table 2

Table 2. Identifiable freshwater prey (bolded) or prey remnants in stomachs of chum salmon caught in the Mackenzie River Delta (Delta), the Mainstem upstream from Tsiigehtchic to the mouth of Great Slave Lake (Mainstem), Great Bear Lake (GBL), or Great Slave Lake (GSL). Each line represents a different salmon.

Differences in the entirety of the stomach contents (all categories) were explored using RSM. As may be expected, RSM decreased as the fish moved upstream, from the Delta to the Mainstem, and then to either GBL or GSL (across locations, F_3,149 = 49.8, p < 0.001) (Figure 2C). Specifically, the RSM of fish in the Delta (0.573) decreased by approximately 50% when in the Mainstem (0.318) and decreased further when in either GBL (0.206) or GSL (0.114). All the locations differed from each other (p < 0.001), with the exception of GBL and Mainstem (p = 0.076) (t-values: Delta vs. GSL, 12.0; GSL vs. Mainstem, 7.64; Delta vs. GBL, 6.03; Mainstem vs. Delta, 5.59; GSL vs. GBL, 4.63; GBL vs. Mainstem, 1.790). Sex was not a factor (F_1,149 = 2.58, p = 0.110). Across the years, RSM differed (F_3,149 = 6.43, p < 0.001), but no clear pattern was apparent. Specifically, the RSM of fish from the Delta was ~40% larger in 2019 vs. 2017 (t = 2.424, p = 0.017), but this difference was lost or reversed as the fish moved farther upstream (e.g., fish in the GBL region had a 40% smaller RSM; t = 2.258, p = 0.025). Note, the mass of the empty stomach was unrelated to whether it had contents (p = 0.451).

For stomachs that had contents (all categories of materials), the fullness was low (FI range: 0.00071%–0.71%). As with RSM, FI decreased as the chum moved upstream (across the sampling locations, F_3,69 = 4.06, p = 0.011) (Figure 2D). The major difference was that the FI was greater in fish from the Delta, and the lack of identifiable prey in these fish, except for one Neomysis and one Malacostraca (Table 2), suggests that contents were most likely from the marine environment, as expected. Overall, chum from the Delta had 2.5× greater FI than Mainstem (t = 3.086, p = 0.018); within 2019, the FI of Delta chum was 3.37× greater than that of GBL and 2.78× greater than that of the Mainstem (Delta vs. GBL t = 3.51, p = 0.005; vs. Mainstem, t = 2.71, p = 0.043). As with RSM, sex was not a factor for FI (p = 0.464). There was only one within-year difference, and that was in GBL chum, for which the FI was 3.82× greater in 2017 than in 2019 (t = 2.49, p = 0.015); the reason for this difference was not clear. Note, one 2.36kg male with 23 g of stomach contents collected in the Delta in 2017 was excluded from statistical analysis, as it was identified as an outlier; its inclusion would have further enhanced the greater FI observed in the Delta.

Of the chum stomachs that had contents, more digested remains were present in chum from 2017, and more remains were identifiable as prey items in 2017. Specifically, in 2017, 72.1% of chum with stomach contents had digested remains in their stomachs vs. 25.9% in 2019, which led to a significant difference by year (Wald, 6.10, p = 0.014), but not by catch location (p = 0.237) or sex (p = 0.98) (Figure 3A). Eleven chum (7.3%) had material identifiable as fish remnants (i.e., scales, bones, and digested fish); in 2017, four were from the Delta, four were from the Mainstem, one from GSL, one from GBL; in 2019, one was from the Delta; the majority (eight) were female (Table 2; Figure 3B). Because of this, both year and sex were significant (i.e., more evidence of piscivory in females from 2017; Wald values of 4.29 and 8.60, p-values of 0.038 and 0.003, respectively). The overall mass of the contents did not differ by location, year, or sex (Figure 3C).

Figure 3

Figure 3. The proportions of stomachs categorized as empty or with contents, and the frequency and mass of stomach contents of chum salmon harvested in the Mackenzie River delta (Delta), Great Bear Lake (GBL), Great Slave Lake (GSL), and mainstem (Main) during 2017 and 2019. Asterisk indicates significant difference across years.

The stomach contents of chum salmon harvested in the Mackenzie River included items other than prey or prey remnants, such as plant materials, synthetic fibers, feathers, and sand and gravel (Figure 3B). When present, plant material was consistently identified as woody debris. Unlike RSM or FI, there was no clear upstream migration-based pattern to the presence of plant materials, despite the locations differing from each other [Wald, 8.15, p = 0.004; neither year (p = 0.079) nor sex (p = 0.341) was a factor]. Specifically, plant materials were most frequently present in chum from the Mainstem (78.2%), not the Delta (37.5%), and GBL and GSL were similar (58.3% and 54.5%, respectively). The mass of plant materials did not vary by catch location (p = 0.238), year (p = 0.190), or sex (p = 0.443) (Figure 3C). The presence of synthetic fibers in the stomach differed across years (Wald, 3.94, p = 0.047), with nearly 3× more chum having fibers in 2017 vs. 2019 (32.6 vs. 11.1%, respectively). There was no clear pattern in the presence of synthetic fibers, as neither catch location nor sex was a factor (p-values of 0.611 and 0.739, respectively). Fiber mass was unrelated to catch locations (p = 0.723), year (p = 0.735), or sex (p = 0.424); however, numerically, there were fewer fibers in chum salmon from GBL and GSL (Figure 3C). Two chum stomachs, one from the Delta and one from the Mainstem, both from 2017, contained a small amount of feathers (0.12 and 0.77 g). Sand and gravel were also found in the contents of several fish; no location/year/sex differences were apparent.

Discussion

Adult chum salmon in the Canadian Arctic consume prey in the fresh water of the Mackenzie River system, albeit rarely and likely opportunistically rather than for energetic gains. Nearly half of the harvested chum salmon stomachs (70 out of 150 total) had contents, including identifiable prey, digested remains, plant materials, synthetic fibers, feathers, or sand and gravel. However, only 14 chum salmon stomachs contained prey remnants, and three chum salmon stomachs contained identifiable freshwater prey. Importantly, the clearest evidence of recent feeding occurred in the Mackenzie River’s great lakes, including one chum salmon consuming insects in Great Bear Lake and two instances of piscivory in Great Slave Lake. Although stomach fullness was low overall, chum salmon stomachs had contents, and digestive tracts were thus not fully degraded for chum salmon harvested as far upstream as Great Slave Lake. This suggests a deviation from the established life-history trait associated with a cessation of feeding in adult Pacific salmon in fresh water, a deviation that may be more prevalent in chum salmon at the range edge.

Given its observed rarity, the freshwater feeding was likely opportunistic, or even accidental, rather than active. Salmon are known to be aggressive during spawning runs, but that energy is generally directed toward intruding conspecifics (McVeigh et al., 2007), rather than putative prey items. The instances of feeding on nutritionally substantive prey items, such as fishes, occurred in the upper reaches of the study area, perhaps reflecting the costs associated with a long freshwater migration (Brönmark et al., 2014). However, as the chum salmon ate forage fish (Coregonus spp.), which are common in Great Slave Lake (Zhu et al., 2017), the consumption may also have been opportunistic. Foraging on insects was only recorded in Great Bear Lake, where chum salmon ate corixids, commonly known as water boatmen. As corixids dwell along the shallow (<5 m deep), rocky nearshore areas of the lake (Johnson, 1975), this suggests that chum salmon may use those habitats in Great Bear Lake. Also, digestion may persist throughout the freshwater movements upstream, as chum salmon caught in the Delta had fuller stomachs than other harvest locations, there were no differences among harvest locations in digested materials (number of chum with stomach contents that included digested materials or mass of digested materials), and the mass of the stomach declined with upstream migration distance.

The instances of plant material, synthetic fibers, feathers, and sand or gravel in the stomachs further suggest that some contents may have been ingested accidentally while actively swimming upstream. The Mackenzie River is turbid, with a mud bottom, whereas Great Bear and Great Slave lakes are both deep and oligotrophic, with Great Bear Lake described as strikingly clear (Johnson, 1975). Accordingly, more than any other harvest location, chum salmon caught in the Mainstem had stomachs containing plant material. More than half of the chum salmon stomachs in Great Bear and Great Slave lakes contained plant material, indicating either that harvest may occur soon after the chum salmon enter the lakes or that the digestion rate for plant material may be slow. While two chum salmon stomachs contained feathers, this is also likely the accidental ingestion of debris, as there was no other evidence of feeding on birds present in any stomach (e.g., no beaks or bird bones). The presence of synthetic fibers in chum salmon stomachs is not particularly surprising, given that microplastics are present in the western Arctic marine environment (D’Angelo et al., 2023) and fresh waters (Bourdages et al., 2024) and that plastics have been found in several Arctic marine fish species (reviewed in Kögel et al., 2022) and recently in Arctic char (Hamilton et al., 2024). Indeed, given the ubiquity of plastic pollution in the Arctic presently, what may be more interesting is the variation, which highlights the need to continue to investigate the occurrence of microplastics, especially in ecologically and culturally important northern species and great lakes.

Given that most chum salmon in the fresh waters of the Mackenzie River system only rarely had stomachs that contained prey items, the potential for competition for food among salmon and northern fish species is low. Those that did consume identifiable freshwater prey, however, demonstrated diet overlap with endemic fish, as the chum salmon showed evidence of eating insects and fish, which are consumed by culturally and ecologically important fish species in the Northwest Territories (reviewed in Wight et al., 2023). Chum salmon stomachs also contained fish and invertebrate remains; however, this may indicate feeding in the marine environment due to the longer retention time for hard prey remnants (Santos et al., 2001). Our data showing the greater FI of chum from the Delta vs. other locations also support this. Dietary niche overlap in the marine environment may be limited among chum salmon, Arctic char, and Dolly Varden, as chum salmon forage on marine invertebrates, whereas Arctic char and Dolly Varden prey on marine forage fish during the summer marine foraging season in the Beaufort Sea (McNicholl et al., 2025). It is curious that more chum salmon stomachs contained digested remains and identifiable prey in 2017 than in 2019; this year-to-year variation is relevant to monitor, as it could inform observed variation in environmental conditions or prey availability for range-expanding chum salmon as well as for Arctic endemic fish. While there is limited risk of competition for prey among range-expanding and Arctic fish species currently, it remains important to understand the potential impacts of a rapidly warming Arctic on endemic fish species (Martin et al., 2020), including shifts in prey availability and habitat use (Reist et al., 2006), and interactions among species, especially with potentially increasing occurrences of subarctic species shifting northward (Dunmall et al., 2024).

Chum salmon are not new to the western Canadian Arctic; they have been harvested at low levels in the Mackenzie River system for generations (Chila et al., 2022) and they are the only salmon species with documented overwinter egg development and nearshore juvenile survival in the North American Arctic (Dunmall et al., 2022). However, they are responding to environmental changes rapidly occurring across the Arctic in recent years and are now being caught in more places than in the past, and the years with higher harvests are becoming more frequent (Dunmall et al., 2024). Harvesters are concerned about the increased presence of salmon in the Inuvialuit Settlement Region in the western Canadian Arctic, both because of the potential for interaction with Arctic species and because they are a tangible representation of climate change impacts to subsistence resources and traditional lifeways (Chila et al., 2022). Indeed, the Canadian Arctic has become the epicenter for distributional shifts of salmon, as Pacific salmon are expanding north and east to the western Canadian Arctic (Dunmall et al., 2024), Atlantic salmon are shifting to the eastern Canadian Arctic (Bilous and Dunmall, 2020), and pink salmon are invading across the Atlantic Ocean to eastern Canada from an introduced source in Russia (McNicholl et al., 2021; Dunmall et al., 2025). It remains important to better understand traits that facilitate range expansions in order to inform management and conservation decisions regarding predicting the establishment of new populations and potential interactions between range-expanding or invading fishes and endemic species.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

Ethical approval was not required for the study involving animals in accordance with the local legislation and institutional requirements because the research was completed on already dead salmon obtained from Indigenous harvesters who were targeting Arctic fish and occasionally caught salmon incidentally, which they provided voluntarily to support research on salmon in the Canadian Arctic.

Author contributions

KD: Writing – review & editing, Methodology, Investigation, Writing – original draft, Funding acquisition, Supervision, Data curation, Conceptualization, Formal Analysis, Project administration. KW: Investigation, Methodology, Writing – original draft, Writing – review & editing, Formal Analysis. DM: Methodology, Investigation, Conceptualization, Writing – review & editing. KT: Writing – original draft, Visualization, Supervision, Writing – review & editing, Methodology, Formal Analysis, Investigation.

Funding

The author(s) declared financial support was received for this work and/or its publication. Funding for the Arctic Salmon program was provided by Fisheries and Oceans Canada, the Fisheries Joint Management Committee, the Gwich’in Renewable Resources Board, and the Ɂehdzo Gotınę Gots’e Nákedı (Sahtú Renewable Resources Board). The Government of Northwest Territories Cumulative Impact Monitoring Program (CIMP 221) provided funding specifically to investigate freshwater feeding in chum salmon from already provided samples.

Acknowledgments

We thank the harvesters for continually teaching us about salmon in the Canadian Arctic. This study was only possible because harvesters shared their knowledge and observations about chum salmon eating in the freshwaters of the Mackenzie River system, as well as their actual salmon catch. For this paper, we are grateful for the letters of support received (indicated by *) and specifically acknowledge the Aklavik Hunters and Trappers Committee (HTC), the Inuvik HTC, the Ehdiitat Gwich’in Council (Aklavik Renewable Resources Council, RRC), Gwichya Gwich’in Council (Tsiigehtchic RRC), the Nihtat Gwich’in Council (Inuvik RRC), Tetł’it Gwich’in Council (TEETŁ’IT ZHEH (Fort McPherson) RRC), the Rádelıh ko (Fort Good Hope) Ɂehdzo Got’ıne (Renewable Resources Council, RRC)*, Norman Wells RRC, Délı̨nę RRC, Pehdzeh Ki First Nation (Wrigley), Łíídlíí Kύé First Nation (Ft. Simpson)*, Ft. Simpson Métis, Acho Dene Koe First Nation (Ft. Liard), Deh Gah Got’ie First Nation (Ft. Providence), Fort Providence Métis Council, West Point First Nation (Hay River), Ka’a’gee Tu First Nation (Hay River), and Deninu K-ue First Nation (Ft. Resolution). We also thank the Aboriginal Aquatic Resource and Oceans Management program in the Dehcho*, as well as staff in the Government of Northwest Territories (GNWT) Environment and Natural Resource Offices in the Dehcho*, Sahtú*, and South Slave regions, and in Fisheries and Oceans Canada offices in Inuvik, Yellowknife, and Hay River. Funding for the Arctic Salmon program was provided by Fisheries and Oceans Canada, the Fisheries Joint Management Committee*, the Gwich’in Renewable Resources Board*, and the Ɂehdzo Got’ı̨nę Gots’é Nákedı (Sahtú Renewable Resources Board)*; the Government of Northwest Territories Cumulative Impact Monitoring Program (CIMP 221) provided funding specifically to investigate freshwater feeding in chum salmon from already provided samples. Thanks to Mike Johnson (North/South Consultants) for the stomach contents taxonomic identifications and to Morgan Steffler (University of Alberta) for graphics work using R. We also appreciate all involved in the Arctic Salmon program, including those outside our study region for this paper, and we recognize the importance of the broader collective effort in monitoring biodiversity change across the Canadian Arctic.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fevo.2025.1686114/full#supplementary-material

References

Alabia I. D., García Molinos J., Hirata T., Mueter F. J., and David C. L. (2023). Pan-Arctic marine biodiversity and species co-occurrence patterns under recent climate. Sci. Rep. 13, 4076. doi: 10.1038/s41598-023-30943-y

PubMed Abstract | Crossref Full Text | Google Scholar

Armstrong J. B. (2010). Comment on ‘‘Egg consumption in mature Pacific salmon (Oncorhynchus spp.)’’. Can. J. Fish. Aquat. Sci. 67, 2052–2054. doi: 10.1139/F10-127

Crossref Full Text | Google Scholar

Bilous M. and Dunmall K. M. (2020). Atlantic salmon in the Canadian Arctic: potential dispersal, establishment, and interaction with Arctic char. Rev. Fish. Biol. Fish. 30, 463–483. doi: 10.1007/s11160-020-09610-2

Crossref Full Text | Google Scholar

Bourdages M. P. T., Provencher J. F., Hurtubise J., Johnson N., and Vermaire J. C. (2024). Microplastics and anthropogenic microparticles in surface waters from Yellowknife Bay, Great Slave Lake, Northwest Territories, Canada. J. Great. Lks. Res. 50, 102348. doi: 10.1016/j.jglr.2024.102348

Crossref Full Text | Google Scholar

Brodeur R. D., Daly E. A., Schabetsberger R. A., and Mier K. L. (2007). Interannual and interdecadal variability in juvenile coho salmon (Oncorhynchus kisutch) diets in relation to environmental changes in the northern California Current. Fshrs. Oceanog. 16, 395–408. doi: 10.1111/j.1365-2419.2007.00438.x

Crossref Full Text | Google Scholar

Brönmark C., Hulthén K., Nilsson P. A., Skov C., Hansson L.-A., Brodersen J., et al. (2014). There and back again: migration in freshwater fishes. Can. J. Zoo. 92, 467–479. doi: 10.1139/cjz-2012-0277

Crossref Full Text | Google Scholar

Camsell C. and Malcolm W. (1919). “The Mackenzie River Basin,” in Memoir 108 Geological series, Geological Survey of Canada, Department of Mines. (Ottawa: J. de Labroquerie). Available online at: https://www.canadiana.ca/view/oocihm.82004/5.

Google Scholar

Carbonell J. A., Wang Y.-J., and Stoks R. (2021). Evolution of cold tolerance and thermal plasticity in life history, behaviour and physiology during a poleward range expansion. J. Anim. Ecol. 90, 1666–1677. doi: 10.1111/1365-2656.13482

PubMed Abstract | Crossref Full Text | Google Scholar

Chila Z., Dunmall K. M., Proverbs T. A., Lantz T. C., Aklavik Hunters, Trappers Committee, et al. (2022). Inuvialuit knowledge of Pacific salmon range expansion in the western Canadian Arctic. Can. J. Fish. Aquat. Sci. 79, 1042–1055. doi: 10.1139/cjfas-2021-0172

Crossref Full Text | Google Scholar

Chown S. L. and McGeoch M. A. (2023). Functional trait variation along animal invasion pathways. Annu. Rev. Ecol. Evol. Syst. 54, 151–170. doi: 10.1146/annurev-ecolsys-102220-013423

Crossref Full Text | Google Scholar

Chuang A. and Peterson C. R. (2016). Expanding population edges: theories, traits, and trade- offs. Glob. Change Biol. 22, 494–512. doi: 10.1111/gcb.13107

PubMed Abstract | Crossref Full Text | Google Scholar

D’Angelo A., Trenholm N., Loose B., Glastra L., Strock J., and Kim J. (2023). Microplastics distribution within western Arctic seawater and sea ice. Toxics 11, 792. doi: 10.3390/toxics11090792

PubMed Abstract | Crossref Full Text | Google Scholar

Davis N. D., Volkov A. V., Efimkin A. Y. A., Kuznetsova N. A., Armstrong J. L., and Sakai O. (2009). Review of BASIS salmon food habits studies. N. Pac. Anadr. Fish. Commun. Bull. 5, 197–208. Available online at: https://www.researchgate.net/publication/268345548_Review_of_BASIS_salmon_food_habits_studies (Accessed October 31, 2025).

Google Scholar

Dunmall K. D., Bean C. W., Berntsen H. H., Ensing D., Erkinaro J., Irvine J. R., et al. (2025). Invading and range-expanding pink salmon inform management actions for marine species on the move. ICES. J. Mar. Sci. 82, fsae199. doi: 10.1093/icesjms/fsae199

Crossref Full Text | Google Scholar

Dunmall K. M., Langan J. A., Cunningham C. J., Reist J. D., Melling H., Aklavik Hunters and Trappers Committee, et al. (2024). Pacific salmon in the Canadian Arctic highlight a range-expansion pathway for sub-Arctic fishes. Glob. Change Biol. 30, 1–14. doi: 10.1111/gcb.17353

PubMed Abstract | Crossref Full Text | Google Scholar

Dunmall K. M., McNicholl D. G., Farley E., and Reist J. D. (2021). Reported occurrences of Pacific salmon in the Canadian Arctic continue to increase whereas reports of Atlantic Salmon sightings remain low. N. Pac. Anadr. Fish. Comm. 17, 88–91. doi: 10.23849/npafctr17/88.91

Crossref Full Text | Google Scholar

Dunmall K. M., McNicholl D. G., Zimmerman C. E., Gilk-Baumer S. E., Burril S., and Von Biela V. R. (2022). First juvenile chum salmon confirms successful reproduction for Pacific salmon in the North American Arctic. Can. J. Fish. Aquat. Sci. 79, 703–707. doi: 10.1139/cjfas-2022-0006

Crossref Full Text | Google Scholar

Dunmall K. M., Mochnacz N. J., Zimmerman C. E., Lean C., and Reist J. D. (2016). Using thermal limits to assess establishment of fish dispersing to high-latitude and high-elevation watersheds. Can. J. Fish. Aquat. Sci. 73, 1–9. doi: 10.1139/cjfas-2016-0051

Crossref Full Text | Google Scholar

Dunmall K. M. and Reist J. D. (2018). “Developing a citizen science framework for the Arctic using the “Arctic Salmon” initiative,” in Impacts of a changing environment on the dynamics of high-latitude fish and fisheries. Proceedings for the 31st Lowell Wakefield Fisheries Symposium. Eds. Mueter F. J., Baker M. R., Dressel S. C., and Hollowed A. B. (Fairbanks, Alaska Sea Grant: University of Alaska). doi: 10.4027/icedhlff.2018.02

Crossref Full Text | Google Scholar

Dunmall K. M., Reist J. D., Carmack E. C., Babaluk J. A., Heidi-Jørensen M. P., and Docker M. F. (2013). “Pacific Salmon in the Arctic: Harbingers of Recent Changes,” in Responses of Arctic Marine Ecosystems to Climate Change. Proceedings for the 28th Lowell Wakefield Fisheries Symposium. Eds. Mueter F. J., Dickson D. M. S., Huntington H. P., Irvine J. R., Logerwell E. A., MacLean S. A., Quakenbush L. T., and Rosa C. (University of Alaska, Fairbanks, Alaska Sea Grant). doi: 10.4027/ramecc.2013.07

Crossref Full Text | Google Scholar

Faulkner C. W., Ittinuar S. J., Tartak C. U., L’Herault V., Harris L. N., Davoren G. K., et al. (2025). Spatiotemporal variation in Arctic char (Salvelinus alpinus) foraging ecology along western Hudson Bay, Nunavut, Canada. Can. J. Fish. Aquat. Sci. 82, 1–13. doi: 10.1139/cjfas-2024-0032

Crossref Full Text | Google Scholar

Garner S. R., Heath J. W., and Neff B. D. (2009). Egg consumption in mature Pacific salmon (Oncorhynchus spp.). Can. J. Fish. Aquat. Sci. 66, 1546–1553. doi: 10.1139/F09-103

Crossref Full Text | Google Scholar

Gibbons M. (2016). Two fishermen catch ‘surprise’ salmon outside Yellowknife (My True North Now). Available online at: https://www.mytruenorthnow.com/16787/news/yellowknife-news/two-fishermen-net-surprise-salmon-outside-yellowknife/ (Accessed September 8, 2024).

Google Scholar

Graham C., Pakhomov E. A., and Hunt B. P. V. (2020). A salmon diet database for the North Pacific Ocean. Sci. Data 7, 332. doi: 10.1038/s41597-020-00676-y

PubMed Abstract | Crossref Full Text | Google Scholar

Hamilton B. M., Harris L. N., Maksagak B., Nero E., Gilbert M. J., Provencher J. F., et al. (2024). Microplastic and other anthropogenic microparticles in Arctic char (Salvelinus alpinus) and their coastal habitat: A first-look at a central Canadian Arctic commercial fishery. Sci. Tot. Environ. 932, 172854. doi: 10.1016/j.scitotenv.2024.172854

PubMed Abstract | Crossref Full Text | Google Scholar

Huntington H. P., Danielson S. L., Wiese F. K., Baker M., Boveng P., Citta J. J., et al. (2020). Evidence suggests potential transformation of the Pacific Arctic ecosystem is underway. Nat. Clim. Chang. 10, 342–348. doi: 10.1038/s41558-020-0695-2

Crossref Full Text | Google Scholar

Hureau J. C. (1970). Biologie compare de quelques poisons antarctiques (Nothotheniidae). Bull. l’Institut. Océano. Fondation Albert Ier de Monaco. 68, 1–44.

Google Scholar

Johnson L. (1975). Distribution of fish species in Great Bear Lake, Northwest Territories, with reference to zooplankton, benthic invertebrates, and environmental conditions. J. Fshrs. Bd. CA. 32, 1989–2004. doi: 10.1139/f75-235

Crossref Full Text | Google Scholar

Kögel T., Hamilton B. M., Granberg M. E., Provencher J., Hammer S., Gomiero A., et al. (2022). Current efforts on microplastic monitoring in Arctic fish and how to proceed. Arc. Sci. 9, 266–283. doi: 10.1139/as-2021-0057

Crossref Full Text | Google Scholar

Mackenzie River Basin Board (2025). State of the Aquatic Ecosystem Report 2021. Available online at: https://soaer.ca/about/#introduction (Accessed November 21, 2025).

Google Scholar

Martin C., Parlee B., and Neyelle M. (2020). Fishing livelihoods in the mackenzie river basin: stories of the délıne got’ine. Sustainability 12, 7888. doi: 10.3390/su12197888

Crossref Full Text | Google Scholar

McNicholl D. G., Gallagher C., Tuktoyaktuk Hunters and Trappers Committee, and Dunmall K. M. (2025). Dietary niches of endemic and range-expanding salmonids in the western Canadian Arctic. Environ. Biol. Fish. 108, 2071–2085. doi: 10.1007/s10641-025-01733-4

Crossref Full Text | Google Scholar

McNicholl D. G., Harris L. N., Loewen T., May P., Tran L., Akeeagok R., et al. (2021). Noteworthy occurrences among six marine species documented with community engagement in the Canadian Arctic. Anim. Migr. 8, 74–83. doi: 10.1515/ami-2020-0113

Crossref Full Text | Google Scholar

McVeigh B. R., Healey M. C., and Wolfe F. (2007). Energy expenditures during spawning by chum salmon Oncorhynchus keta (Walbaum) in British Columbia. J. Fish. Biol. 71, 1696–1713. doi: 10.1111/j.1095-8649.2007.01635.x

Crossref Full Text | Google Scholar

Moran E. V. and Alexander J. M. (2014). Evolutionary responses to global change: lessons from invasive species. Ecol. Lett. 17, 637–649. doi: 10.1111/ele.12262

PubMed Abstract | Crossref Full Text | Google Scholar

Natural Resources Canada (2025). Kuukpak/Dehcho/Grande Rivière/Deho/Nagwichoonjik (Geographic names in Canada. Government of Canada). Available online at: https://natural-resources.Canada.ca/maps-tools-publications/maps/geographical-names-Canada/kuukpak-dehcho-grande-riviere-deho-nagwichoonjik (Accessed November 21, 2025).

Google Scholar

Northwest Territories Cumulative Impact Monitoring Program (NWT CIMP) (2025). Research Bulletin. Assessing diet and interaction among salmon and Arctic fishes in the Mackenzie River. Available online at: https://www.gov.nt.ca/ecc/sites/ecc/files/resources/128-cimp_bulletin_75_en_web.pdf (Accessed November 4, 2025).

Google Scholar

Pyper N. R., Painting C. J., and McGaughran A. (2024). Home and away: the role of intraspecific behavioural variation in biological invasion. New Z. J. Zoo. 51, 151–174. doi: 10.1080/03014223.2024.2336035

Crossref Full Text | Google Scholar

Qin Y. and Kaeriyama M. (2016). Feeding habits and trophic levels of Pacific salmon (Oncorhynchus spp.) in the North Pacific Ocean. N. Pac. Anadr. Fish. Commun. Bull. 6, 469–481. doi: 10.23849/npafcb6/469.481

Crossref Full Text | Google Scholar

R Core Team (2021). R: A language and environment for statistical computing (Vienna, Austria: R Foundation for Statistical Computing). Available online at: https://www.R-project.org/ (Accessed June 15, 2025).

Google Scholar

Reist J. D., Wrona F. J., Prowse T. D., Power M., Dempson J. B., King J. R., et al. (2006). An overview of effects of climate change on selected Arctic freshwater and anadromous fishes. Ambio 35, 381–387. doi: 10.1579/0044-7447(2006)35[381:aooeoc]2.0.co;2

PubMed Abstract | Crossref Full Text | Google Scholar

Santos M. B., Clarke M. R., and Pierce G. J. (2001). Assessing the importance of cephalopods in the diets of marine mammals and other top predators: problems and solutions. Fish. Res. 52, 121–139. doi: 10.1016/S0165-7836(01)00236-3

Crossref Full Text | Google Scholar

Satterfield F. R. and Finney B. P. (2002). Stable isotope analysis of Pacific salmon: insight into trophic status and oceanographic conditions over the last 30 years. Prog. Oceanogr. 53, 231–246. doi: 10.1016/S0079-6611(02)00032-0

Crossref Full Text | Google Scholar

Stafford K. M., Farley E. V., Ferguson M., Kuletz K. J., and Levine R. (2022). Northward range expansion of subarctic upper trophic level animals into the Pacific Arctic region. Oceanography 35, 158–166. doi: 10.5670/oceanog.2022.101

Crossref Full Text | Google Scholar

Stephenson S. A. (2006). A review of the occurrence of Pacific salmon (Oncorhynchus spp.) in the Canadian Western Arctic. Arctic 59, 37–46. Available online at: https://www.jstor.org/stable/40512766 (Accessed March 26, 2012).

Google Scholar

Urawa S., Beacham T. D., Masa-Aki F., and Kaeriyama M. (2018). “Ocean ecology of chum salmon,” in The ocean ecology of Pacific salmon and trout. Beamish R. J. editor. (Bethesda, Maryland: American Fisheries Society), 161–317.

Google Scholar

Wight K. J., McNicholl D. G., and Dunmall K. M. (2023). A systematic review of the trophic ecology of eight ecologically and culturally important fish species in the North American Arctic. Polar. Biol. 46, 409–425. doi: 10.1007/s00300-023-03133-9

Crossref Full Text | Google Scholar

Zhu X., Chapelsky A., Carmichael T. J., Leonard D. L., Lea E., Tallman R. F., et al. (2017). Establishment of ecological baseline metrics for integrated ecomonitoring and assessment of cumulative impacts on Great Slave Lake fisheries ecosystems. Can. Tech. Rpt. Fish. Aquat. Sci. 3223, x + 58. Available online at: https://publications.gc.ca/collections/collection_2017/mpo-dfo/Fs97-6-3223-eng.pdf (Accessed November 4, 2025).

Google Scholar