https://orcid.org/0000-0001-8711-9263
Accessible Data
Abstract
The Lower Columbia River and Estuary is critical rearing habitat for juvenile Pacific salmon. Extending from the river mouth to Bonneville Dam 235 river kilometers upstream, the Estuary has been altered by dams, dikes, and habitat loss due to deforestation and wetland removal. Since 2008, five sites have been monitored to identify the long-term status and trends in Lower Columbia River and Estuary juvenile salmon rearing habitat. Here, we address predominantly natural origin juvenile Chinook Oncorhynchus tshawytscha diets and identify spatial and temporal trends in prey consumption, stomach fullness, energy consumption, and the metabolic costs associated with fish size and water temperatures. Juvenile Chinook diets consisted mainly of corophiid amphipods, chironomid dipterans, and cladocerans, with other insects filling in most of the remainder of their diets. Juvenile Chinook salmon diets were stable over time, and stomach fullness and caloric intake was comparable among the sites where most salmon were collected. Juvenile Chinook salmon were frequently in water temperatures above fitness thresholds and higher water temperatures and metabolic rates will require increased foraging as water temperatures rise. Reduced growth, earlier migration, and mismatches between prey production and foraging are near term possibilities. Juvenile salmon rearing resiliency in the estuary can be aided by maintaining sufficient river discharge levels for salmon, and by restoring habitat and habitat connectivity to the mainstem channel.
Figures
Citation: Accola K, Cordell J, Oxborrow B, Toft JD, Suzumura A, Grote J (2025) Subyearling Chinook salmon diets in Lower Columbia River estuarine habitats. PLoS One 20(6): e0325939. https://doi.org/10.1371/journal.pone.0325939
Editor: Jason Magnuson, US Geological Survey Columbia Environmental Research Center, UNITED STATES OF AMERICA
Received: February 6, 2025; Accepted: May 20, 2025; Published: June 12, 2025
Copyright: © 2025 Accola et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All "Subyearling Chinook salmon diets in Lower Columbia River estuarine habitats" files are available from the Dryad database and are publicly accessible via this link: https://doi.org/10.5061/dryad.c59zw3rm2 Data used in this study is also visualized at https://public.tableau.com/app/profile/ecosystem.monitoring.program/viz/MacroinvertebrateCommunities/WelcometotheMacroinvertebrateEcosystemMonitoringDashboard.
Funding: This study was funded under Contract #11-2024 by the Northwest Power and Conservation Council/Bonneville Power Administration, to support data collected by the EMP (implemented by LCEP), and to inform regional habitat restoration efforts and action effectiveness monitoring. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Estuaries provide crucial nursery functions for young salmon, including refuge and growth opportunities, as they undergo osmoregulatory transitions from fresh to saline waters [1–4]. Growth during their first year of life is linked to marine survival and spawning [5,6] and the highest proportional growth for salmonids, as visual predators [7], occurs during the juvenile life stage [8]. Larger fish can better evade predation, are less susceptible to starvation, and are more efficient foragers [9]. Juvenile salmon estuarine residency is varied, but a few species, like Chinook Salmon Oncorhynchus tshawytscha, are estuary-dependent [1], with a broad range of residence times [10]. Varying life history types [11,12] and subyearling migration patterns in salmon populations [13] aid adaptation to environmental variability [14]. However, habitat modifications alter prey availability, habitat connectivity, refuge, residency [3,15], and impact density-dependent foraging efficiency [16], so restoring estuarine habitats in conjunction with harvest and hatchery management can aid resiliency in the face of changing climate regimes [15].
The Columbia River (Oregon and Washington, USA) has historically been one of the highest global producers of Pacific Salmon Oncorhynchus spp. [17]. Consequentially, the Lower Columbia River and Estuary (LCRE) is recognized as important juvenile salmon rearing habitat [18–21]. Smaller sizes of subyearling Chinook salmon are the most prevalent species in shallow water habitats in the lower estuary [19,20,22]. Regionally, Chinook salmon are an economically, culturally, and ecologically important species, and the Columbia River is a key source of the species. Historically, spring flooding transported nutrients (wetland-derived prey), cold water, and sediments downstream to nourish migrating juvenile salmon [23], aiding foraging opportunities for their required growth enroute to the ocean [3,8]. However, anthropogenic changes beginning in the 1860s including agriculture, dike building, and dam construction for the hydropower system, altered water quality and habitat conditions [23–25], resulting in loss of 68–70% of vegetated tidal wetlands [26]. Habitat loss undermines the estuary’s ability to support healthy juvenile salmon populations, and the need to protect and recover habitat for threatened and endangered evolutionarily significant units of salmon was recognized [27,28]. In 1995, the LCRE, extending from the river mouth to Bonneville Dam (235 river kilometers (RKm)), was designated an “Estuary of National Significance” [23].
Juvenile salmon need to grow as much as possible before entering the ocean in all estuaries [15], but in the LCRE, where protective marine embayments are lacking compared to places like Puget Sound or the Strait of Georgia, this growth is especially important [22]. Although loss of shallow water habitat has reduced macrodetrital contributions in the LCRE [29], food webs supporting juvenile salmon in the LCRE are still primarily detritus based [30]. Tidal wetlands are a primary source of salmon prey in the wetlands and in the estuary, as salmon disproportionately favor these allochthonous prey sources [30]. The most important prey in LCRE subyearling Chinook diets have been adult dipterans (mostly Chironomidae; non-biting midges) and amphipods (Americorophium salmonis and the more saline-tolerant A. spinicorne) [31,32], which are generally more prevalent in diets at tidally inundated sites [33–35]. Cladocerans (water fleas) have become more prevalent in mid-estuary diets [35]. The large biomass of chironomid larvae make them energetically important [36], and although cladocerans are less nutritious, they are prolific, large bodied, and slow moving, making them more accessible than other prey [37,38].
Shortly after the LCRE was deemed an estuary of national significance, the Lower Columbia Estuary Partnership (LCEP) was established. The LCEP works to protect and restore the LCRE, including by managing the Ecosystem Monitoring Program (EMP), a multi-agency status and trends program that studies juvenile salmon rearing conditions at five sites in the LCRE [23]. Since 2008, EMP scientists have collected and analyzed key biotic and abiotic data related to ecological conditions at tidally influenced estuarine and freshwater emergent habitats used by migrating juvenile Chinook salmon. Data collection includes primary production, zooplankton, fish diets, prey resources, habitat and hydrology, and other abiotic factors. EMP estuarine sites are minimally disturbed and are a subset of the eight hydrogeomorphic reaches across the LCRE [23,39,40]. The sites are currently located at Ilwaco Slough (RKm 6), Welch Island (RKm 53), Whites Island (RKm 72), Campbell Slough (RKm 149), and Franz Lake (RKm 221) (Fig 1).
Map includes site photos, showing river kilometer distance (Rkm) from mouth of the estuary. This Tableau Public map was created by the Wetland Ecosystem Team at the University of Washington, in conjunction with the Lower Columbia Estuary Partnership, in 2025. Basemap © Mapbox, © OpenStreetMap. https://public.tableau.com/app/profile/ecosystem.monitoring.program/viz/MacroinvertebrateCommunities/WelcometotheMacroinvertebrateEcosystemMonitoringDashboard.
Here, we focus on subyearling Chinook salmon diets (78% wild, 22% hatchery) at the five EMP sites to better understand growth in the estuary [41,42]. Diet data has been collected specifically for juvenile Chinook salmon because of their Endangered Species Act listing [27], estuary dependence, and prevalence in the lower estuary [6], manifested by the various life history patterns present throughout the year [4,21,32]. We identified juvenile Chinook foraging trends, including top prey, among sites and through time. We quantified the influence of site and month on salmon fitness via their stomach fullness, energy consumption, and metabolic costs, by assessing their diets and the energetic benefits the salmon derive from them. By identifying trends in this data, we can better understand the role of estuary prey in Chinook salmon fitness.
Methods
Sites
Juvenile Chinook salmon diets and temperature data were analyzed from five EMP sites in the LCRE. Ilwaco Slough (RKm 6) is marsh habitat near the mouth of the Columbia River with low vegetative species richness [35]. It is the most tidally influenced site, with annual average salinities of <10 practical salinity units [23] (Fig 1). Most salmon caught here are chum salmon O. keta [43]. Upstream, Welch Island (RKm 53) is tidally influenced, high marsh habitat with small tidal channels. Whites Island (RKm 72), created from dredge material, is also high marsh habitat influenced by tides and prolonged freshets. Welch and Whites Islands have high vegetative species richness, and most sampled fish are juvenile Chinook salmon [35]. Campbell Slough (RKm 149) monitoring occurs at an emergent marsh that is 1.5 km from the mainstem channel, is frequently grazed by cattle, and has low vegetative species richness. The site is influenced by Bonneville Dam discharge, but pH levels frequently exceed water quality benchmarks [44], and water temperatures regularly exceed salmon fitness standards [35]. The most common salmon caught at Campbell Slough is juvenile Chinook salmon. The Franz Lake (RKm 221) sampling location, located 350 m from the mouth of the channel, is high marsh habitat impacted by beaver dams. Water levels are controlled by Bonneville Dam discharge and water quality is considered to be suitable [35]. Juvenile Chinook and coho O. kisutch salmon are the most frequently caught salmonids. Campbell Slough and Franz Lake have the highest fish diversity, buoyed by warm-water tolerant introduced species, some of which have life stages that prey on juvenile salmon [35,40].
Field and laboratory
Since 2008, subyearling Chinook salmon have been collected for diet and other analyses using boat or raft-deployed 38 x 3-m variable mesh bag seines (10.0 mm and 6.3 mm wings, 4.8 mm bag). The fish were collected under Washington State Collection Permits RCW 77.15.660 and WAC 220-200-04, and under NMFS Northwest Fisheries Science Center Scientific Research Permit 22944. Sampled fish were maintained in buckets with oxygenated, in situ water, with ice added, as necessary, to maintain water temperatures at levels conducive to fish comfort. Trained biologists quickly euthanized fish with a lethal dose of MS-222, placed them on ice to slow digestion for additional analyses not described here. The stomachs were preserved in 10% formalin. In the laboratory, entire stomachs were analyzed, and each prey taxon was counted, blotted dry, weighed to the nearest 0.0001 g, and identified to the finest possible taxonomic unit (crustaceans to genus or species and insects to family or order). Water quality has been continuously measured at EMP sites using Yellow Springs Instruments (YSI) equipped with water temperature, specific conductance, pH, and dissolved oxygen probes.
Analyses
Important components of juvenile Chinook diet composition were determined by calculating an Index of Relative Importance (IRI), which accounts for the numeric composition (counts) of each prey taxa, the gravimetric composition (weight), and the frequency of occurrence (F; ratio of an individual prey taxa occurrence and sample size) [45]. Together, the prey abundance, weight, and frequency of occurrence in the diet were calculated as:
Using IRI results, we used nonmetric multidimensional scaling (NMDS) on a Bray-Curtis dissimilarity matrix (appropriate for percentage data) to visualize dissimilarities of top diet taxa among sites. Our ordination observations led to analysis of similarities to explain statistical differences between groups, and we calculated the cumulative contribution of each species, listing the taxa that account for more than 70% of the differences between sites, i.e., those taxa that were most important in driving differences in the diets among sites.
We visually categorized, then calculated, fish foraging performance, or stomach fullness, by instantaneous ration (IR), which was calculated as the ratio of total prey weight (sum of weights of all individual taxa in the diet) to total fish mass. To understand how prey in diets nutritionally benefit salmon, we quantified energy ration (ER), that is, energy consumption, in which individual prey taxon masses for each fish were multiplied by the energy density (kJ g−1 wet mass) of each prey taxon, summed, and divided by the individual fish mass. Energy counts of prey taxa were compiled and acquired from [46] and [16]. Metabolic costs for fish were measured as maintenance metabolism [47], in which jm is the mass specific maintenance costs at 0ºC (0.003), d is the temperature coefficient for biomass assimilation (0.68), t is water temperature in ºC, and W is fish mass.
Finally, we used generalized linear mixed models to compare spatial and temporal IR, ER, and maintenance metabolism values to understand spatiotemporal cost-benefits for juvenile Chinook. We used gamma distributed (log link) mixed models for the zero-bounded, continuous, skewed data. The models were parameterized in terms of mean and a gamma dispersion parameter
with the structure:
We performed diagnostic tests on data and residuals and tested interactive and random effects. Potential variables included sites, time (month and year), fork length bin (30–65 mm and 66–99 mm; several 100 + mm salmon were excluded from analyses), mark status to distinguish marked hatchery fish from wild or unmarked fish (marked or unmarked; indicated by presence/absence of an adipose fin) the best estimate of fish origin, and water temperature at time of fish sampling.
Our best models computed IR, ER, and maintenance metabolism values among sites and months, and by fish length bin (for IR, ER models) and mark status (for maintenance metabolism models), while accounting for temporal variation that occurs among and within years, that is, the random effects of months nested within years. Mark status, rather than fork length, was used for maintenance metabolism models, because maintenance metabolism is a function of fork length. Because sample sizes were unequally dispersed among mark/unmarked salmon and fish fork lengths (most smaller study fish were wild fry, most hatchery fish were larger), interactive effects of the two would be biased towards smaller and unmarked fish. Rank deficiencies, as each site was not sampled every month, rendered the interactive effects among sites and months nonviable.
Our statistical analyses were calculated using R Statistical Software [48; version 4.2.2.] and R Studio [49]. Final models were developed using the package ‘glmmTMB’ [50; version 1.1.7] and were chosen considering corrected Akaike information criterion and by considering the ecological questions we wanted to answer. Multivariate analyses on prey taxa were conducted using the package ‘vegan’ [51; version 2.6.4]. We organized data using “dplyr” [52; version 1.1.2]. Post-hoc analyses of mixed models included pairwise comparisons of estimated marginal means and calculating the effect size of the comparison, using “emmeans” [53; version 1.8.8]. We used “ggplot2” [54; version 3.4.4] for data visualization. Our study protocols were reviewed and granted written approval through our federal permit (NOAA Section 10 #22944-2R), which serves as the ethics board following Institutional Animal Care and Use Committee guidelines.
Results
Water temperatures averages ranged from 4.4ºC – 10.9ºC in February to 18.9ºC – 23.3ºC in July. In June, the water temperature average ranged from 14.9ºC – 20.8ºC. The maximum temperature averages at Franz Lake and Ilwaco Slough were 14.8 ºC. Most high temperatures occurred at Welch Island, Whites Island, and Campbell Slough, where the majority of juvenile Chinook were sampled (Fig 2).
Top sites and months had the highest juvenile Chinook densities. Black dotted line represents 17.5ºC, which [44] recommends as the limit (seven day average of maximum water temperatures) for Chinook rearing.
Spatial and temporal distribution
A total of 1365 juvenile Chinook salmon diets were analyzed from 2008–2021 EMP fish samples. 18 fish with empty stomachs were removed from analyses. Fish were selected for diet analyses to maximize representation from sites, months, and fish lengths. Most Chinook salmon (91%) used for diet analyses were netted at the middle three EMP sites (Campbell Slough (n = 343), Welch Island (n = 444), and Whites Island (n = 455), with small overall sample sizes at Franz Lake (n = 83) and Ilwaco Slough (n = 40). 78% (n = 1065) of sampled juvenile Chinook were unmarked and considered to be natural origin (wild) fish. Juvenile Chinook counts were highest in May (n = 593), and most salmon were in the smallest length category (30–65 mm fork length; n = 775) (Table 1; see also [20]). In general, most smaller study Chinook were wild fry, most hatchery Chinook were larger fingerlings due to release protocols, yet larger fish were a comparable mixture of wild and hatchery origin.
Diet composition
Prey taxa with the highest IRI in juvenile Chinook diets, for all fish fork lengths (30 mm – 99 mm), were amphipods, cladocerans, and dipterans (Fig 3). Juvenile Chinook primarily consumed chironomids and other dipterans and insects at the most upstream site (Franz Lake), included cladocerans at Welch Island, Whites Island, and Campbell Slough (although predominantly at Campbell Slough), and consumed a mixture of mostly corophiid amphipods and chironomid dipterans at Whites Island, Welch Island, and Ilwaco Slough [35; see also [31]. Ilwaco Slough diets were a mixture of amphipods (Americorophium spp.) and primarily Chironomidae and Psychodidae dipterans.
Prey are sorted by year, fork length bin, and site. Top prey are color sorted by taxa and plot the Index of Relative Importance (IRI) values for taxa each year. IRI accounts for the counts, weights, and frequency of occurrence of prey in the diet.
Americorophium spp. (mostly A. salmonis, then A. spinicorne) comprised 75% of the amphipod counts and weights and were predominant in Welch and Whites Islands’ salmon diets. The remaining amphipods, in order of count and weight importance, were unidentified amphipods, Eogammarus confervicolus, and Hyallela azteca. Daphnia spp. (water fleas) comprised 67% of the counts and weights of cladocerans. Within dipterans, chironomids accounted for 47% of the counts and 39% of the weights. Overall, chironomids and unidentified dipterans, which may be chironomids, comprised 96% of the dipterans present in our juvenile Chinook diets. Adult/emergent chironomids were the most numerous chironomid life stage (44%), and pupa made up the most weight (60%) in the diets.
Diet patterns were portrayed by an NMDS ordination using IRI values to show salmon diet differences among sites, by year, and between size length bins (Fig 4). The fourth most important diet taxon was Hemiptera, followed by the remaining taxa labelled in Fig 3. Fish netted at Whites Island, Welch Island, and Ilwaco Slough had more similar diets, and Campbell Slough and Franz Lake diets had dipterans in common, but differed based on the presence or absence of cladocerans. Analysis of similarities (ANOSIM) and pairwise similarity percentage (SIMPER) site comparisons corroborate the ordinations through statistical analyses, by demonstrating that Campbell Slough and Franz Lake diets were dissimilar from the remaining three sites by the prevalence of amphipods downstream and dipterans and cladocerans upstream (Table 2). Pairwise comparisons show no diet prey differences among different fish length bins. There were no long-term diet differences among years, except that cladocerans did not become prevalent in juvenile Chinook diets until 2016. A few years had temporal diet differences that were likely due to sampling anomalies (e.g., attenuated sampling in 2020 due to COVID-19).
NMDS plot shows top prey in juvenile Chinook diets between 2008 and 2021. Each point represents fish diets binned by length (30−59 mm, 60−79 mm, 80−99 mm) for each year at each site (color-coded). Black vectors depict significant species loadings (p = 0.05), with longer arrows indicating a stronger association between the species and IRI. The strongest weights are for Amphipoda, Diptera, and Cladocera, signifying they are the most important salmon prey. Inset: energy density values (kJ g-1) for top prey taxa.
Diet benefits and costs
Fish fullness.
Fish fullness, measured by instantaneous ration, was highest at Ilwaco Slough and lowest at Franz Lake (Fig 5). However, sample sizes were small at these sites due to sampling challenges (Franz Lake) and site location (Ilwaco Slough). 2008–2021 Ilwaco Slough diets consisted entirely of unmarked fry. Stomach fullness was comparable among Welch Island, Whites Island, and Campbell Slough, and there was no statistical difference in fish fullness among months. Smaller Chinook had higher fullness marginal mean values than larger fish, so accounting for their body weight, they were filling their stomachs more than larger fish (effect size = 0.39; Table 2). Overall, study fish do not appear to be prey-limited (Fig 6).
Dot plots (in blue) depict the EMMs of generalized linear mixed models’ response variables (Stomach Fullness, Caloric Intake, and Metabolic Costs) plotted by Site and Month. Overlaying boxplots are raw data values that are color-sorted by fish length (30-65 mm = gray, 66-99 mm = black).
Salmon are sorted by site and color-coded by categorization of stomach fullness. Blank means stomach fullness was not categorized. 2 indicates an emptier stomach and 6 indicates a full stomach.
Caloric intake.
Ilwaco Slough Chinook stomachs, dominated by amphipods, produced the highest energy ration values. Franz Lake diets, consisting mainly of dipterans and other wetland insects, had the lowest energy ration values, likely because of lower fullness levels compared to other sites (Fig 5, Table 2). Energy intake was comparable in diets among Welch Island, Whites Island, and Campbell Slough. Accounting for weight, smaller fish had higher caloric intake than larger fish (effect size = 0.39), demonstrating that they derived more nutritional benefit from the prey they were consuming. July was the only month with higher energy ration marginal mean values than other months (Table 2). Higher July caloric intake was likely due to highly nutritious prey including Trichoptera, Hemiptera, and Hymenoptera replacing some amphipods in the Welch and Whites Island’s diets, and fish and insects replacing cladocerans in Campbell Slough diets (see also [55]). Larger wild and hatchery fish fared similarly in diets.
Metabolic costs.
Metabolic costs increased month to month because larger, growing fish have higher metabolic requirements and water temperatures, which increase gastric evacuation rates [56], rise as the year progresses. Higher water temperatures increase metabolic costs more than other factors [56,57] and fish must compensate by increasing their energy intake [56]. Campbell Slough had the highest metabolic costs among the study sites and regularly has water temperatures that exceed salmon fitness thresholds [35]. Most Campbell Slough salmon were larger fish, in large part due to hatchery fish prevalence, while Welch and Whites Island’s salmon had greater size diversity. Marked (hatchery) fish had higher metabolic costs, but this was influenced by the marked fish category having more large fish.
Discussion
Since 2008, we monitored trends of prey quantity, quality, and diet benefits for subyearling Chinook salmon at five minimally disturbed sites in the heavily modified Lower Columbia Estuary [58]. Overall, juvenile Chinook diets at EMP sites appear to be stable, even in current sub-optimal temperature regimes, with salmon experiencing foraging opportunities in which they balance their diets depending on prey availability.
Prey trends
All size classes of sampled Chinook had similar interannual prey trends, in which three taxa (corophiid amphipods, cladocerans, and chironomid dipterans) comprised most of the diets. Rather than an even distribution among the top three prey, there was frequent interannual variation in which one taxon (typically amphipods or dipterans) dominated the year, with other top prey taxa making smaller contributions. Interannual fluctuation in top prey indicates prey availability drives diet patterns, and prey tradeoffs indicate an estuary’s capacity for salmon diet adaptability [59,60]. A possible exception to prey availability driving diet patterns is that larger Chinook appeared to select for amphipods at Welch and Whites Islands even though dipterans were more prevalent than amphipods in neuston samples taken at those sites [35]. Annual fluctuations may be due to differing annual flow regimes, specifically high flow years, e.g., Americorophium spp. play a larger role in wetland diets when they are more readily available from high river flows scouring benthic habitats [43,61].
Chironomids and Americorophium amphipods are also important in main channel yearling salmonid diets, with other insects complementing their diets [62]. Wetland prey presence in main channel diets highlights the importance of tidal wetlands as prey sources to the varying Chinook life history patterns in the LCRE [20], including fast-moving main channel yearling salmon that mostly originate from interior stocks [61,62], and smaller LCEP EMP subyearling Chinook salmon that originate from Lower Columbia stocks [35].
Beginning in 2016, cladocerans became more prominent in salmon diets at Campbell Slough, with amphipods and dipterans remaining important prey throughout the study period. Cladocerans, followed by cyclopoid copepods (the latter were not important in Chinook diets), were also prominent in neuston samples at Welch Island, Whites Island, and Campbell Slough [35]. Amphipods and dipterans (predominantly chironomids) were present in lower quantities in neuston and benthic core samples [35], but their large size and nutrition likely makes them preferable to visual predators like Chinook salmon. Cladocerans may share diet prominence with dipterans at Campbell Slough because amphipods were less available, based on neuston tow and benthic core data [35]. Cladocerans are slow-moving prey that are easily captured but are less nutritious than other top prey, and they may represent as an alternative when there is a paucity of more nutritious prey. Their accessibility likely contributed to their high biomass in diets, enabling higher salmon growth rates in Campbell Slough and similar habitats, where flooding supplies detritus and nutrients, and elevated water temperatures increase phytoplankton and zooplankton productivity [31,60,63,64].
Energy-dense insects became more prevalent at the middle three LCEP EMP sites as the summer progressed, especially at Welch and Whites Islands. Although [65] found the highest energy transport from LCRE tidal wetlands was from non-chironomid dipterans and (predominantly) corixid hemipterans, unidentified dipterans (which could have been chironomids) were important in our salmon diets, but hemipterans (‘true bugs’) did not have as high importance in our study diets as would be expected, given their high nutritional value. The exceptions were a few outlying years, when larger fish at Welch and Whites Islands complemented their corophiid amphipod and chironomid dipteran diets with hemipterans, especially during summer months. Most hemipterans in our study were corixids, which tend to have tougher exoskeletons and be less easily digestible than other hemipterans. As aquatic invertebrates, they may be able to escape predation more easily than terrestrial insects, so it is unclear whether smaller Chinook fry have more difficulty capturing them, experience gape limitations, or it is not worth the energy expenditure to pursue and prey on them.
Spatial and temporal trends
Most juvenile Chinook salmon (91%) were netted at the middle sites (Welch Island, Whites Island, and Campbell Slough). Challenges of sampling access at Franz Lake, which was hindered by beaver dams and water levels [35], and the location of Ilwaco Slough near the mouth of the estuary, where salmon may quickly pass through to the ocean, likely contributed to low Chinook densities at these sites. Although few juvenile Chinook were netted at Ilwaco Slough, juvenile chum counts were highest at this site and Chinook were present in low densities in spring months. Stomach fullness results from our study differ from [31], who found upstream diets had the fullest stomachs, while downstream stomachs were emptier. Their downstream subyearling Chinook were offshore and migrating rapidly, while our Ilwaco Slough subyearling fish were smaller and likely utilizing off-channel habitats to increase growth before leaving the estuary.
Salmon counts peaked in May each year, then decreased throughout summer – more quickly for small fish – because some small fish transitioned to a larger size class, while others left the estuary or died. A July increase in caloric intake values indicates that salmon were acquiring nutritious, energy dense prey during this time. Although caloric intake was highest in July, stomach fullness was not. Most nutritious food was found in diets from sites where predominantly larger salmon were found (Welch Island, Whites Island, Campbell Slough). Prey included high energy dipterans, hemipterans, hymenopterans, trichopterans (all insects), and amphipods, which can help mitigate higher metabolic rates due to elevated water temperatures [66,67]. This was especially true at Campbell Slough, where reduced connectivity to the main channel in low water periods may have increased water temperatures, but also facilitated retaining wetland insects more than open sites like Welch and Whites Islands.
Juvenile Chinook salmon have exhibited steady annual presence at Welch Island, Whites Island, and Campbell Slough, even with frequent high temperatures surpassing minimal water temperature thresholds for juvenile salmon/steelhead fitness (80 days exceeding 19ºC in most years) [35]. Salmon densities decline in June, but May has continued to be the peak month of juvenile Chinook abundances sampled throughout our study period. The Washington Department of Ecology recommends a threshold of 17.5ºC as a water temperature standard [44]. [68] found that compared to warmer streams, Chinook grew faster with maximum temperatures of 16ºC. Lower water temperatures also benefit Chinook over their non-salmonid competitors [69]. However, water temperature-related related changing migration patterns for our study are not yet apparent.
The three lower estuary sites (Ilwaco Slough, Welch Island, Whites Island) are tidally driven and inundated several hours each day, while Franz Lake and Campbell Slough are off channel sites with reduced connectivity to the mainstem and are therefore more affected by freshet and dam discharge. River discharge is important in creating ephemeral flooded habitats [66,70] that moderate higher water temperatures, especially at sites with reduced connectivity to the mainstem channel. Franz Lake and Campbell Slough have higher fish diversity and richness, driven by introduced species that are tolerant of warm temperatures and low dissolved oxygen [21]. The six most common fish (five native, one non-native) that have life stages that prey on juvenile salmon are freshwater fish found primarily at Franz Lake and Campbell Slough [35].
Summary
Agriculture, residential development, and dams have modified critical juvenile salmon rearing habitat in the LCRE [23]. As a more estuary-dependent species, juvenile Chinook are especially susceptible to estuarine modifications [1,3]. Their marine survival is positively correlated with how much the estuary is unmodified [6], because natural estuarine habits offer refuge from predation [6] and prey production for growth [15]. However, the status and trends of juvenile Chinook diets at five monitoring sites in the Lower Columbia River and Estuary shows stable diets, dominated by amphipods and dipterans, and cladocerans. Other prey, including other insects, complement the diets. Our results verify the importance of extant tidal wetlands habitats for juvenile salmon diets. One of the top salmon prey for subyearling Chinook, chironomid dipterans, are linked to both aquatic and terrestrial habitats as they progress from larvae to adults. Wetlands are sources of other nutritious vegetation-associated prey, including other insects. [21] found juvenile Chinook use emergent marsh habitat for multiple weeks, growing 0.53 mm/day, re-entering the same habitats after vacating them during ebb tides. Re-entry indicates foraging-related movement patterns, as salmon use secondary channels against outgoing tides during peak emergence times of prey insects [21]. Main channel complexity is also important for growth, as vegetated shoreline habitats can provide insects for salmon and channel complexity can reduce flushing of salmon in high flow periods [70].
While juvenile Chinook migration currently peaks in May, increasing annual water temperatures will lead to higher energy requirements to maintain growth to compensate for increased metabolic costs, or to fish emigrating at smaller sizes than is optimal for first year survival. First year growth includes early ocean entry, so estuarine growth may be able to be postponed in system-specific situations [71], but early marine salmon have increased metabolic requirements in warm ocean regimes [67], and the LCRE lacks protective marine embayments outside the estuary. Improving estuary wide habitat conditions and cold water refugia in the LCRE will benefit juvenile salmon by easing metabolic stressors and providing high quality prey. The current state of climate change, including extreme summer heat events, highlights the need for water temperature mitigation and flooded habitats via dam discharge, channel access to the mainstem to promote and sustain healthy water temperatures, and a high-quality prey field for optimal growing conditions for juvenile salmon.
Next steps should include bioenergetics modeling and pairing diet data with stable isotopes and biophysical indicators at LCEP EMP sites to increase our understanding of estuarine influence on juvenile salmonid growth (e.g., [42,72]. Further analyses may include studying hatchery and wild Chinook diets in isolation to investigate potential diet differences and competition for key resources (see [73]), because wild salmon have higher estuarine growth [74]. Diets of pooled hatchery and wild fish with different lengths did not differ in our study, and diet contents through time were stable. Thus, though smaller fish were predominantly wild and larger fish had higher percentages of hatchery fish, we expect all juvenile Chinook in the LCRE to have similar diets.
Acknowledgments
We thank the many partners involved in the research of the Lower Columbia Estuary Partnership (LCEP) Ecosystem Monitoring Program (EMP), including the current team of EMP research scientists that also includes Curtis Roegner, Tawnya Peterson, Joseph Needoba, and Ian Edgar. We are grateful to the LCEP oversight team, including Catherine Corbett, and formerly Sarah Kidd and Sneha Rao Manohar. We thank the team at NOAA Fisheries, including Regan A. McNatt, Susan A. Hinton, Lyndal L. Johnson, G. Curtis Roegner, Daniel Lomax, and Sean Sol, for collection of neuston and juvenile Chinook salmon samples. Finally, we thank Joe Needoba and the team at OHSU for collecting water temperature data.
References
- 1.
Healey M. Estuarine comparisons. Juvenile Pacific salmon in estuaries: the life support system. 1982. p. 315–41.
- 2. Levy DA, Northcote TG. Juvenile salmon residency in a marsh area of the Fraser River estuary. Can J Fish Aquat Sci. 1982;39(2):270–6.
- 3.
Simenstad CA, Fresh KL, Salo EO. The role of Puget Sound and Washington coastal estuaries in the life history of Pacific salmon: an unappreciated function. In: Kennedy VS, editor. Estuarine Comparisons. Academic Press; 1982. p. 343–64. Available from: https://www.sciencedirect.com/science/article/pii/B9780124040700500260
- 4. Thorpe JE. Salmonid Fishes and the Estuarine Environment. Estuaries. 1994;17(1):76.
- 5.
Reimers PE. The length of residence of juvenile fall Chinook salmon in Sixes River, Oregon. 1971;111.
- 6. Magnusson A, Hilborn R. Estuarine influence on survival rates of coho (Oncorhynchus kisutch) and chinook salmon (Oncorhynchus tshawytscha) released from hatcheries on the U.S. Pacific coast. Estuaries. 2003;26(4):1094–103.
- 7. Schabetsberger R, Morgan CA, Brodeur RD, Potts CL, Peterson WT, Emmett RL. Prey selectivity and diel feeding chronology of juvenile chinook (Oncorhynchus tshawytscha) and coho (O. kisutch) salmon in the Columbia River plume. Fisheries Oceanography. 2003;12(6):523–40.
- 8.
Quinn T. The behavior and ecology of Pacific salmon and trout: 2nd edition. 2018.
- 9. Anderson JT. A Review of Size Dependant Survival During Pre-recruit Stages of Fishes in Relation to Recruitment. J Northw Atl Fish Sci. 1988;8:55–66.
- 10. Reimers PE, Loeffel RE. The length of residence of juvenile fall chinook salmon in selected Columbia River tributaries. Fish Comm Oregon Res Briefs. 1967;13(1):5–19.
- 11.
Rich WH, Holmes H. Experiments in marking young chinook salmon on the Columbia River, 1916 to 1927. (1929), pp. 215- S 264, 85 figs. Washington: Bull, US Bur Fish, vol XLIV, 1928. 1929.
- 12. Norrie CR, Morgan CA, Burke BJ, Weitkamp LA, Miller JA. Freshwater growth can provide a survival advantage to Interior Columbia River spring Chinook salmon after ocean entry. Mar Ecol Prog Ser. 2022;691:131–49.
- 13. Rich WH. Early life history and seaward migration of Chinook salmon in the Columbia and Sacramento Rivers. US Bur Fish Bull. 1920;37:74.
- 14.
Groot C, Margolis L. Pacific salmon life histories. UBC Press; 1991.
- 15.
Bottom D, Simenstad C, Baptista A, Jay D, Burke J, Jones K. Salmon at River’s End: The Role of the Estuary in the Decline and Recovery of Columbia River Salmon. 2005.
- 16. David A, Simenstad C, Cordell J, Toft J, Ellings C, Gray A. Wetland loss, juvenile salmon foraging performance, and density dependence in Pacific Northwest estuaries. Estuaries and Coasts. 2015;39.
- 17.
Netboy A. The Columbia River Salmon and Steelhead Trout, Their Fight for Survival. Seattle: University of Washington Press; 1980.
- 18. Kareiva P, Marvier M, McClure M. Recovery and management options for spring/summer chinook salmon in the Columbia River basin. Science. 2000;290(5493):977–9.
- 19.
Fresh K, Casillas E, Johnson L, Bottom D. Role of the Estuary in the Recovery of Columbia River Basin Salmon and Steelhead: An Evaluation of Selected Factors on Population Viability. 2005.
- 20. Roegner GC, McNatt R, Teel DJ, Bottom DL. Distribution, size, and origin of juvenile chinook salmon in shallow-water habitats of the lower Columbia River and estuary, 2002–2007. Marine and Coastal Fisheries. 2012;4(1):450–72.
- 21. McNatt RA, Bottom DL, Hinton SA. Residency and Movement of Juvenile Chinook Salmon at Multiple Spatial Scales in a Tidal Marsh of the Columbia River Estuary. Trans Am Fish Soc. 2016;145(4):774–85.
- 22.
Bottom DL, Anderson G, Baptista A. Salmon Life Histories, Habitat, and Food Webs in the Columbia River Estuary: An Overview of Research Results, 2002-2006. report, Portland, Oregon: 2008.
- 23.
Rao S, Kidd SA, Fuller RN, McNatt R, Poppe K, Peterson TD, et al. Lower Columbia River Ecosystem Monitoring Program Annual Report for Year 15 (October 1, 2018 to September 30, 2019). Prepared by the Lower Columbia Estuary Partnership for the Bonneville Power Administration. Available from the Lower Columbia Estuary Partnership, Portland, OR. 2019.
- 24.
Myers JM, Busack CA, Rawding D, Marshall AR, Teel DJ, Van Doornik DM, et al. Historical population structure of Pacific salmonids in the Willamette River and the lower Columbia River basins. Northwest Fisheries Science Center (U.S.), editor. 2006; Available from: https://repository.library.noaa.gov/view/noaa/3461
- 25. Simenstad CA, Small LF, David McIntire C, Jay DA, Sherwood C. Columbia river estuary studies: An introduction to the estuary, a brief history, and prior studies. Progress in Oceanography. 1990;25(1–4):1–13.
- 26. Marcoe K, Pilson S. Habitat change in the lower Columbia River estuary, 1870–2009. Journal of Coastal Conservation. 2017;21:1–21.
- 27.
Good TP, Waples RS, Adams PB. Updated status of federally listed ESUs of West Coast salmon and steelhead. 2005.
- 28.
National Marine Fisheries Service (NMFS). Columbia River Estuary ESA Recovery Plan Module for Salmon and Steelhead. NMFS Northwest Region. Portland, OR. January. Prepared for NMFS by the Lower Columbia River Estuary Partnership (contractor) and PC Trask & Associates, Inc., subcontractor. 2011.
- 29. Sherwood CR, Jay DA, Bradford Harvey R, Hamilton P, Simenstad CA. Historical changes in the Columbia River Estuary. Progress in Oceanography. 1990;25(1–4):299–352.
- 30.
Bottom D, Jones K, Herring M. Fishes of the Columbia River estuary. Final Report on the Fish Work Unit of the Columbia River Estuary Data Development Program. 1984 Jan 1.
- 31. McCabe GT Jr, Emmett RL, Muir WD, Blahm TH. Utilization of the Columbia River estuary by subyearling chinook salmon. Northwest Science. 1986;60.
- 32.
Bottom D, Baptista A, Burke J, Campbell L, Casillas E, Hinton S, et al. Estuarine Habitat and Juvenile Salmon: Current and Historical Linkages in the Lower Columbia River and Estuary. 2011.
- 33. Bottom DL, Jones KK. Species composition, distribution, and invertebrate prey of fish assemblages in the Columbia River Estuary. Progress in Oceanography. 1990;25(1):243–70.
- 34.
Lott MA. Habitat-specific feeding ecology of ocean-type Chinook salmon in the lower Columbia River estuary. M.Sc. Thesis. Seattle: University of Washington. 2004. p. 124.
- 35.
Kidd SA, Edgar I, Rao S, Fuller RN, McNatt R, Poppe K, Peterson TD, et al. Lower Columbia River Ecosystem Monitoring Program Annual Report for Year 16 (October 1, 2020 to September 30, 2021). Prepared by the Lower Columbia Estuary Partnership for the Bonneville Power Administration. Available from the Lower Columbia Estuary Partnership, Portland, OR. 2022.
- 36. Merritt R, Cummins K, Berg M. An Introduction to The Aquatic Insects of North America: Fourth Edition. The Journal of Animal Ecology. 2008.
- 37. Craddock DR, Blahm TH, Parente WD. Occurrence and Utilization of Zooplankton by Juvenile Chinook Salmon in the Lower Columbia River. Transactions of the American Fisheries Society. 1976;105(1):72–6.
- 38. Drenner RW, Strickler JR, O’Brien WJ. Capture probability: The role of zooplankter escape in the selective feeding of planktivorous fish. J Fish Res Bd Can. 1978;35(10):1370–3.
- 39.
Simenstad C, Burke J, O’Connor J, Cannon C, Heatwole DW, Ramirez M, et al. Columbia River Estuary Ecosystem Classification—Concept and Application. US Geological Survey Open-File Report 2011-1228. 2011 Jan 1.
- 40.
Sagar J, Borde A, Johnson L, Peterson T, Needoba J, Corbett C. Juvenile salmon ecology in tidal freshwater wetlands of the lower Columbia River and estuary: synthesis of the ecosystem monitoring program, 2005-2010. Portland: Lower Columbia Estuary Partnership; 2013.
- 41. Simenstad CA, Cordell JR. Ecological assessment criteria for restoring anadromous salmonid habitat in Pacific Northwest estuaries. Ecological Engineering. 2000;15(3–4):283–302.
- 42. Chittaro P, Johnson L, Teel D, Moran P, Sol S, Macneale K, et al. Variability in the performance of juvenile Chinook salmon is explained primarily by when and where they resided in estuarine habitats. Ecology of Freshwater Fish. 2018;27(3):857–73.
- 43.
Kidd SA, Schwartz MD, Fuller RN, McNatt R, Poppe K, Peterson TD, et al. Lower Columbia River Ecosystem Monitoring Program Annual Report for Year 14 (October 1, 2017 to September 30, 2018). Prepared by the Lower Columbia Estuary Partnership for the Bonneville Power Administration. Available from the Lower Columbia Estuary Partnership, Portland, OR. 2019.
- 44.
Washington State Water Quality Standards. WAC 173-201A-200. Fresh water designated uses and criteria. 2023.
- 45. Liao H, Pierce CL, Larscheid JG. Empirical Assessment of Indices of Prey Importance in the Diets of Predacious Fish. Transactions of the American Fisheries Society. 2001;130(4):583–91.
- 46.
Gray A. The salmon river estuary: restoring tidal inundation and tracking ecosystem response. 2005.
- 47. Fiechter J, Huff DD, Martin BT, Jackson DW, Edwards CA, Rose KA, et al. Environmental conditions impacting juvenile Chinook salmon growth off central California: An ecosystem model analysis. Geophysical Research Letters. 2015;42(8):2910–7.
- 48.
R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2023.
- 49.
Posit team. RStudio: Integrated Development Environment for R. Boston, MA, Posit Software, PBC. 2024. URL http://www.posit.co/
- 50. Brooks M, Kristensen K, van Benthem K, Magnusson A, Berg CW, Nielsen A. glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. R Journal. 2017;9:378–400.
- 51.
Oksanen J, Simpson G, Blanchet FG, Kindt R, Legendre P, Minchin P. Vegan community ecology package version 2.6-2 April 2022. 2022.
- 52. Wickham H, François R, Henry L, Müller K, Vaughan D. dplyr: A Grammar of Data Manipulation. R package version 114, https://github.com/tidyverse/dplyr [Internet]. 2023
- 53. Lenth RV. emmeans: Estimated Marginal Means, aka Least-Squares Means. R package version 1.10.4.900003. 2024. https://CRAN.R-project.org/package=emmeans
- 54.
Wickham H. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York; 2016. https://ggplot2.tidyverse.org
- 55. Haskell CA, Beauchamp DA, Bollens SM. Trophic Interactions and Consumption Rates of Subyearling Chinook Salmon and Nonnative Juvenile American Shad in Columbia River Reservoirs. Trans Am Fish Soc. 2017;146(2):291–8.
- 56. Benkwitt C, Brodeur R, Hurst T, Daly E. Diel feeding chronology, gastric evacuation, and daily food consumption of juvenile chinook salmon in Oregon coastal waters. Transactions of the American Fisheries Society. 2009;138:111–20.
- 57. Brett JR, Higgs DA. Effect of temperature on the rate of gastric digestion in fingerling sockeye salmon, Oncorhynchus nerka. J Fish Res Bd Can. 1970;27(10):1767–79.
- 58.
Thomas D. Changes in Columbia River Estuary habitat types over the past century. Columbia River Estuary Data Development Program. Columbia River Estuary Study Taskforce; 1983.
- 59. Woo I, Davis MJ, Ellings CS, Hodgson S, Takekawa JY, Nakai G, et al. A Mosaic of Estuarine Habitat Types with Prey Resources from Multiple Environmental Strata Supports a Diversified Foraging Portfolio for Juvenile Chinook Salmon. Estuaries and Coasts. 2019;42(7):1938–54.
- 60. Goertler P, Jones K, Cordell J, Schreier B, Sommer T. Effects of Extreme Hydrologic Regimes on Juvenile Chinook Salmon Prey Resources and Diet Composition in a Large River Floodplain. Trans Am Fish Soc. 2018;147(2):287–99.
- 61.
PNNL (Pacific Northwest National Laboratory) and NMFS (National Marine Fisheries Service). Restoration action effectiveness monitoring and research in the lower Columbia River and estuary, 2016–2017. Richland, Washington: PNNL; 2020.
- 62. Weitkamp LA, Beckman BR, Van Doornik DM, Munguia A, Hunsicker M, Journey M. Life in the Fast Lane: Feeding and Growth of Juvenile Steelhead and Chinook Salmon in Main‐Stem Habitats of the Columbia River Estuary. Trans Am Fish Soc. 2022;151(5):587–610.
- 63. Koehler ME, Fresh KL, Beauchamp DA, Cordell JR, Simenstad CA, Seiler DE. Diet and Bioenergetics of Lake‐Rearing Juvenile Chinook Salmon in Lake Washington. Trans Am Fish Soc. 2006;135(6):1580–91.
- 64. Katz JVE, Jeffres C, Conrad JL, Sommer TR, Martinez J, Brumbaugh S, et al. Floodplain farm fields provide novel rearing habitat for Chinook salmon. PLoS One. 2017;12(6):e0177409. pmid:28591141
- 65. Roegner GC, Johnson GE. Export of macroinvertebrate prey from tidal freshwater wetlands provides a significant energy subsidy for outmigrating juvenile salmon. PLoS One. 2023;18(3):e0282655. pmid:36930681
- 66. Sommer TR, Nobriga ML, Harrell WC, Batham W, Kimmerer WJ. Floodplain rearing of juvenile chinook salmon: evidence of enhanced growth and survival. Can J Fish Aquat Sci. 2001;58(2):325–33.
- 67. Daly EA, Brodeur RD. Warming Ocean Conditions Relate to Increased Trophic Requirements of Threatened and Endangered Salmon. PLoS One. 2015;10(12):e0144066. pmid:26675673
- 68. Bisson PA, Davis GE. Production of juvenile chinook salmon, Oncorhynchus tshawytscha, in a heated model stream. Fishery Bulletin. 1976;74(4):763–74.
- 69.
Hillman TW. The effect of temperature on the spatial interaction of juvenile chinook salmon and the redside shiner and their morphological differences. Doctorate dissertation. Idaho State University; 1991.
- 70. Jeffres CA, Opperman JJ, Moyle PB. Ephemeral floodplain habitats provide best growth conditions for juvenile Chinook salmon in a California river. Environmental Biology of Fishes. 2008;83(4):449–58.
- 71. MacFarlane RB. Energy dynamics and growth of Chinook salmon (Oncorhynchus tshawytscha) from the Central Valley of California during the estuarine phase and first ocean year. Can J Fish Aquat Sci. 2010;67(10).
- 72. Chalifour L, Scott DC, MacDuffee M, Stark S, Dower JF, Beacham TD. Chinook salmon exhibit long-term rearing and early marine growth in the Fraser River, British Columbia, a large urban estuary. Can J Fish Aquat Sci. 2021;78(5):539–50.
- 73. Cordell J, Toft J, Gray A, Ruggerone G, Cooksey M. Functions of restored wetlands for juvenile salmon in an industrialized estuary. Ecological Engineering. 2011;37:343–53.
- 74. Arbeider M, Pemberton-Renaud V, Hodgson E, Moore J. The estuarine growth and residency of juvenile Pacific salmon in North America: a compilation of empirical data. Canadian Journal of Fisheries and Aquatic Sciences. 2023;81.
