Fatty acids and lipid levels in Atlantic salmon (Salmo salar) change with sea-age and differ from other pelagic fish in the Norwegian sea

Fatty acids and lipid levels in Atlantic salmon (Salmo salar) change with sea-age and differ from other pelagic fish in the Norwegian sea | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Fatty acids and lipid levels in Atlantic salmon (Salmo salar) change with sea-age and differ from other pelagic fish in the Norwegian sea Kjell Rong Utne, Sonnich Meier, Josef Daniel Rasinger, Nikolaos Nikolioudakis, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7827046/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Analysis of the fatty acid composition in body tissue, which reflects the accumulated dietary intake over the last months, is a well-established method for studying trophic interactions in marine food webs. Here we present the fatty acid composition of salmon, herring, mackerel, and their prey sampled in May-Aug at marine feeding areas in the Norwegian Sea. A large proportion of the post-smolts sampled early in the summer had a high proportion of FA associated with age-0 fish. Later in the summer, when they have reached the northern Norwegian Sea, post-smolts have a higher proportion of FAs associated with calanoid copepods. The FA composition indicates of post-smolts indicate a wide feeding niche with a diet changing rapidly in time and space, and somatic growth was prioritized before lipid accumulation until the end of the first summer. Post-smolts have very low lipids levels compared to sub-adult salmon or other pelagic fish feeding in the same geographic region of the northeast Atlantic. Furthermore, the FA composition of post-smolts, which are hypothesized to compete for prey with other pelagic fish species, is partly different from the FA composition of herring and mackerel caught within the same geographic area, suggesting important diet differences between the species. feeding interaction zooplankton lipids diet pelagic Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Most Atlantic salmon ( Salmo salar ) displays an anadromous life-cycle and juvenile salmon (smolts) leave the rivers and typically spend 1–3 year at sea before returning to rivers to spawn (Klemetsen et al. 2003 ). Salmon have rapid growth during the marine phase, but also a high natural mortality often exceeding 95% year − 1 (ICES, 2024 ). The survival at sea for salmon, hereon referred to as “marine survival”, originating from large parts of the Atlantic Ocean has decreased over several decades (Olmos et al. 2019 ). The reasons for this decline in marine survival are complex and multiple factors, operating on varying spatial and temporal scales, may have contributed to a lower marine survival. The Norwegian Sea is an important feeding area for post-smolts (salmon during their first summer and autumn at sea) and sub-adult salmon (after their first winter at sea) originating from several European countries (Jacobsen and Hansen 2000 ; Jacobsen and Hansen 2001 ; Gilbey et al. 2021 ; O'Sullivan et al. 2022 ). During the period 2004–2012, post-smolts feeding in the Norwegian Sea had gradually lower stomach fullness and condition factor, a result that clearly indicate reduced prey availability for salmon within this geographic region (Utne et al. 2021a ). Several studies have suggested decreased prey abundance in the Northeast Atlantic attributed to reduced primary production due to altered water circulation and increased water temperatures (Beaugrand and Reid 2003 ; Beaugrand and Reid 2012 ; Olmos et al. 2020 ; Utne et al. 2022 ; Tyldesley et al. 2024 ). Another hypothesis for the reduced growth and survival of salmon at sea is increased interspecific competition with marine pelagic fish (Potter 2000 ). Especially during the post-smolt phase, the salmon diet can overlap with the diets of mackerel ( Scomber scombrus ) and herring ( Clupea harengus ) (Utne et al. 2021b ). Although competition with pelagic fish does not seem to reduce marine survival of salmon (Utne et al. 2021b ), more research is needed to better understand how feeding interactions affect the marine growth of salmon. Atlantic salmon perform long-distance migrations (Gilbey et al. 2021 ; Rikardsen et al. 2021 ) and must during their marine phase adapt to changing prey composition and abundance. As salmon grows during the marine phase, the size range of the prey they can target changes (Jacobson et al. 2018 ). Fast growth enables salmon to feed on larger prey, but at some stage, salmon need to prioritize storage of energy in the form of lipids instead of allocating energy to somatic growth. This shift in energy allocation allows for sufficient reserves to survive the winter, promote gonadal maturation and allow the long migrating back to the rivers. The link between salmon size, accumulation of lipid reserves, and their feeding strategy is poorly understood for wild Atlantic salmon. Analysis of specific FAs (fatty acid trophic markers: FATMs) is a well-established method for studying trophic interactions in marine food webs as there is limited modifications of FA structures when transferred to a higher trophic level (Dalsgaard et al. 2003 ). While analyses of stomach content only provide a snapshot of the feeding activity up to a few hours before capture, FATM reflects dietary patterns over the last weeks or months (Budge et al. 2011 ). Laboratory experiments with Atlantic salmon given different feed has proven that FATMs deposited in body tissue well reflects their past diet (Budge et al. 2011 ; 2012 ). Some FAs are utilized quickly to cover metabolic costs (Iverson et al. 2004 ), but salmon does also have the active enzymatic capacity to modified FA by elongation and desaturation (Tocher, 2010 ), both of which affects the relative proportion of FAs deposited in the fish. A correction factor is therefore needed to accurately quantify their past diet composition (Iverson et al. 2004 ). However, most FAs mainly are deposited directly in the body tissue, and the FA composition without correction still reflects the trophic niche of the predator (e.g. Thomas et al. 2019 , Schäfer et al. 2024 ). Furthermore, FAs from prey can either be accumulated as membrane lipids (phospholipids, PL), which is a requirement for constructing new cells and therefore needed for somatic growth, or as storage's lipid (triacylglycerols, TAG) available as energy reserve for later use. Total lipid (TL) is the sum of PL and TAG and is normally dominated by TAG in fat fish such as salmon. However, for a lean fish without much stored lipids the proportion of PL relative to TAG is higher than for a fat fish. While TAG largely mirrors the FAs composition of the diet, the PL are under metabolic control and thereby a result of selective allocation of FAs for somatic growth (Olsen et al. 1991 ). To study the dietary history of fish, Olsen et al. ( 2013 ) have recommended to analyze FA profiles in TAG instead of in TL. In the pelagic food web, phytoplankton are the main primary producers responsible for de novo synthesis of especially polyunsaturated FAs (PUFAs), like the long-chain PUFAs 20:5n-3 and 22:6n-3 (Zhukova and Aizdaicher 1995 ). Early life stages of fish larvae feeding heavily on herbivorous copepod nauplii have high levels of phytoplankton FATMs, especially 20:5n-3 and 22:6n-3 (Klungsøyr et al. 1989 ; Folkvord et al. 1996 ). Long chain monounsaturated FAs (MUFAs) such as gondoic acid (20:1n-9), erucic acid (22:1n-9), and cetoleic acid (22:1n-11) are synthesized by copepods such as Calanus spp. and accumulate at high levels in copepod-eating zooplankton and fish, like sandeel ( Ammodytes spp.), capelin ( Mallotus villosus ), herring, and mackerel. Thus, high levels of MUFAs indicate feeding on the Calanus food chain although it does not directly reveal if the diet consisted of Calanus spp., or a predator of Calanus (Petursdottir et al. 2008 ). High levels of PUFAs indicate feeding on fish larvae or fish in their post-larva stage (hereafter referred to as 0-age fish) (Litz et al. 2017a ). The FA profiles from fish sampled at sea can therefore reveal which prey groups dominated the diet the last months prior to capture, and potentially enhance our understanding of the link between the individual’s feeding history and its growth and lipid accumulation. In this study we present for the first time data on the FA composition of Atlantic salmon, herring, mackerel and various species of zooplankton sampled in marine areas commonly used for feeding by pelagic fish in the Northeast Atlantic Ocean. The objective of the study is to provide new knowledge about marine prey groups important for salmon in the Northeast Atlantic Ocean, the width of the trophic niche of post-smolts and sub-adult salmon, and the overlap with niches of other pelagic fishes in the same marine feeding areas. We determine which FAs dominate in Atlantic salmon and analyze how the FA composition varies among individual fish, between post-smolts and sub-adult salmon, and between Atlantic salmon, herring and mackerel. For post-smolts we test whether the FA composition varies among years, geographic area, or the size of the individual. We also test whether post-smolt TL and TAG have different FA compositions, which would indicate that certain prey groups are utilized directly for somatic growth while other prey groups for storing energy reserves. Furthermore, the trophic position and the feeding niche of Atlantic salmon, herring and mackerel are analyzed by estimating how their FA composition resembles the FA composition of zooplankton and age-0 fish. Materials and methods Sampling Sampling of pelagic fish Northeast Atlantic mackerel ( Scomber scombrus ), Norwegian Spring-Spawning herring ( Clupea harengus ) and Atlantic salmon were sampled with pelagic trawls at 13 different scientific surveys targeting mackerel or herring. All post-smolts, herring and mackerel were sampled in July-Aug (International Ecosystem Summer Survey in the Nordic Seas, IESSNS), while sub-adult salmon were sampled both during May-June (N=14, International Ecosystem Survey in the Nordic Seas, IESNS) and during July-Aug (N=72, IESSNS). Post-smolts were sampled during 2012-2020, sub-adult salmon during 2013-2020, and herring and mackerel during 2015-2017 (for more details about survey sampling see supplement Tab. S1). The surveys covered the Norwegian Sea and, in some cases, also the northern North Sea. In total 56 mackerel, 54 herring, 294 post-smolts and 87 sub-adult salmon were sampled within the area 60˚13’-77˚19’N and 12˚28’W-23˚24’E (Fig. 1). Most sub-adult salmon and all post-smolts, mackerel and herring were sampled with a Multpelt 832 pelagic trawl [for trawl gear information see (ICES 2013)] towed at the surface at 4-5 knots (1 knot = 1.852 km·h –1 ), but two sub-adults were sampled from trawl hauls within the range 20-410 m depth in May. Trawling was carried out at predetermined locations with spacing of 50-70 nautical miles in July-August and at random locations in May-June. Following the approach of Gilbey et al. (2021), individuals shorter than 35 cm were classified as post-smolts while larger individuals (43.5-79 cm) were classified as sub-adult salmon. None of the sampled salmon had a body-length within the range 35-43 cm. Once the fish were on deck, they were individually weighed to the nearest gram, and the length (fork length for salmon, total length for herring and mackerel) was measured to a 5 mm resolution. The condition factor (CF) was calculated according to Fulton’s K (K = body weight / fork length 3 ). Sampling of zooplankton and herring larvae. Samples of copepods ( Calanus finmarchicus and Calanus hyperboreus ), amphipods ( Themisto abyssorum and Themisto libellula ), and euphausiids ( Meganyctiphanes norvegica , Thysanoessa inermis and Thysanoessa longicaudata ) were collected from IESNS and IESSNS surveys in 2015-2017 (Tab. S1, sampling positions are shown in Fig. 1). The zooplankton was sampled using a WP2 net (Skjoldal et al. 2004). Species were identified immediately after sample collection (prior to fixation) under a dissecting stereomicroscope and 5-10 individuals were transferred to a 1.5 ml glass tubes, frozen quickly and stored at –25 °C (Tiedemann et al. 2021). Herring larvae were sampled along the Norwegian coast during the Norwegian Spring-Spawning herring larvae survey in April 2015 and 2016 (Tiedeman et al. 2021). Samples of 10 herring larvae from each sampling location were frozen in 16 ml glass tubes. Standardization of tissue sampling of salmon Atlantic salmon has a heterogeneous distribution of fat (Brix et al. 2009). To measure the total amount of lipids in salmon, one must homogenize the whole fish, which is a very time-consuming procedure for large fish. In the present study, the frozen salmon was slightly thawed, and a muscle sample was removed using scalpel. We standardized the sample location of muscle to be taken in front of the dorsal fin, where a 1cm deep, 1 cm wide and 1-2 cm long muscle samples were removed, placed into glass tubes and stored at – 80 ⁰C until the lipid extraction (Fig. S1). After muscle sample removal, each fish was homogenized, and a subsample was collected for lipid extraction. Linear regression was used to identify the degree of correlation and establish the relationship between lipid levels in muscle sample vs homogenized whole fish (n=19; 8 post-smolt and 11 sub-adult salmon, Fig. S2). There was a strong linear relationship between the lipid levels in muscle samples and lipid levels in whole fishes and the estimated formula was: Whole fish lipid (%) = 1.404+0.380*Muscle lipid (%) (R 2 = 0.97, p-value < 0.001, n = 19; Fig. S2, Tab. S2). Total lipids for salmon were calculated from standardized muscle sample in all following analyses. The FA profiles are similar in muscle tissue and whole fish sample, although a few significant differences are found for some FAs only present in low quantities (Tab. S3). Lipid and FA analysis Extraction of FA and lipids The post-smolt were extremely lean, with lipid levels mostly found to be below 2 % (average value of 1.98 ± 0.44 %) and the amount of TAG (storage lipid) in these fish contributed with less than 10 % of the total lipids (TL). The composition of lipid classes is highly dependent on the TL amount in the fish, exemplified by a lean post-smolt sample with 1.3 % TL contains 84 % PL, 7 % TAG and 9 % free FA/cholesterol, while a fat sub-adult salmon sample with 37 % TL contains 10 % PL, 88 % TAG and 2 % free Fas/Cholesterol (Tab. S4). We have analyzed the FA composition in both TL and TAG for salmon, while the only the FA composition in TAG is available for herring and mackerel. Hence, only data from TAG is used when comparing FA composition among fish species. The TAG analysis was done by lipid extracted using hexane:isopropanol (3.2) according to the Hara method (Hara and Radin 1978) and HPLC separation methods described in Madhun et al. (2017). The zooplankton and herring larvae samples were extracted using a modified Folch extraction (Meier et al. 2006). The amount and composition of Fas in TL and TAG were determined by direct methanolysis and gas chromatography following Meier et al. (2006). The TL and TAG fractions were methylated, the resulting FA methyl esters (FAME) were extracted with hexane before being analysed on a HP-7890A gas chromatograph (Agilent, USA) with a flame ionization detector (GC-FID), as previously described in Meier et al. (2006). 67 Fas that contribute with more than 0.1 % of the total Fas were included in our data analysis. There is no data of TAG for 11 of the 294 sampled post-smolts as the samples were compromised. Likewise, no data of TL were generated for 11 sub-adult salmon due to a mistake in the lab. Data are presented as FA profiles (% of total FAs). Four sub-adult salmon were identified as escaped farmed fish and excluded from the initial dataset (see supplement section S3 for more information). As several of the zooplankton (copepods, amphipods and euphausiids) samples contain large amounts of wax esters, the FAME and the fatty alcohols (FAOH) were separated on solid phase column and analyzed individually on GC-FID to avoid coelution, as described in Haug et al. (2021). The FAOHs or wax esters are not found in the lipids of the fish as the FAOHs oxidize very rapidly into the corresponding FAs during prey digestion. The FAOHs, therefore, also contribute to the predator’s FA composition, and when looking at FATM, both the FA and the FAOH from the prey should be considered (Budge and Iverson 2003). In the present work we have therefore added the sum of the quantitative amount of the different corresponding FAs and FAOHs (especially 22:1n-11 FA + 22:1n-11 FAOH in copepods and amphipods and 14:0, 16:0 and 16:1n-7 FAOH in Thysanoessa sp. ) before normalizing to 100 %. The FA and FAOH compositions of the samples are given in the supplementary information, Tables S6-S19. FA biomarkers The copepod Calanus finmarchicus is an important prey for pelagic fish (Dalpadado et al. 2000; Bachiller et al. 2016; Utne et al. 2021a) and carnivorous zooplankton such as Themisto spp. and Euphausiids in the Norwegian Sea ecosystem. Furthermore, several studies have identified the importance of age-0 fish (e.g Clupea harengus, Sprattus sprattus , Gadus spp, Ammodytes spp, Sebastes spp) for post-smolts during their first summer at sea (e.g. Salminen et al. 2001; Utne et al. 2021a; Utne et al. 2022). Hence, we explored further the importance of Calanus spp. and age-0 fish FAs in extracted lipids from salmon, herring and mackerel. For Calanus spp. the following FAs were aggregated in a FATM biomarker; Σ 20:1n-9, 22:1n-11, 22:1n-9, while 22:6n-3 was used as a biomarker for age-0 fish ( Klungsøyr et al. 1989; Folkvord et al. 1996). Statistical analysis Lipid levels A Mann-Whitney U-test was used to test for difference in lipid levels between post-smolts and sub-adult salmon. Principal coordinate analysis To estimate how FA profiles varied among species and life-stages, principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarities was applied. The first PCoA axis (PC1) represents the main axis of variation in the FA composition (Pond and Ward 2011). PCoA was applied to test for any difference in FA composition for 1) post-smolts and sub-adult salmon, 2) all pelagic fish, 3) salmon TAG and TL (separately for post-smolts and sub-adult salmon). PCoA was carried out with the R-package FactoMineR (Lê et al. 2008) and the results visualized by using the R-package factoextra (Kassambara and Mundt 2020). The FA values were log-transformed, mean centered, and standardized by dividing by standard deviation before performing the PCoA. This transformation method levels out the quantitative differences among FAs and ensures that variation in minor FAs also impacts the results. Differences in dispersion of FA profiles among post-smolts and sub-adult salmon, and among post-smolts from different geographic regions and years was tested using a multivariate extension of Levene’s test (Anderson 2006) for homogeneity of variance. A seasonal effect was not tested for post-smolts as 99% of the individuals were sampled within a month in July-early August. Significant differences in FA profiles between species or sub-groups of a species were estimated with analysis of similarity (ANOSIM) if the dispersion was different between groups and permutational multivariate analysis of variance (PERMANOVA) otherwise (Oksanen et al., 2024). The FAs contributing most to the difference among groups were identified using similarity percentages (SIMPER, Oksanen et al., 2024). To test if the FA composition for post-smolts changed geographically or over time, a PCoA was carried out with individuals grouped according to sampling year. Thereafter a PCoA was carried out with the individuals grouped according to geographic position in the Norwegian Sea (southern = south of 67˚30’N, northwestern = north of 67˚30’N and west of 5 ˚E and northeastern = north of 67˚30’N and east of 5 ˚E, Fig. 1A). Generalized additive mixed-effect models Generalized additive mixed-effect models (GAMMs) were applied to test if the observed variation in total lipids in post-smolts and sub-adult salmon were linked to individual size or geographic position. GAMMs were also applied to test if the first principal component (PC1) of the PCoA for FA composition in post-smolt, and the FATM biomarkers for Calanus spp. and age-0 fish in post-smolts, were related to their geographic position or individual size. The full set of tested models and specifications of model parametrization are given in Table S5. The GAMMs were run separately for post-smolts and sub-adult salmon. The data were inspected for outliers, normality, collinearity, and independence prior to modelling. Some observations were excluded to avoid bias effects of large outliers (see Supplement Tab. S5). A preliminary data inspection prior to the modelling revealed some highly correlated variables (r>0.6), and in these cases one variable was selected to represent a trait in the modelling. This included post-smolt length and body weight (cor = 0.93, p<0.001), and sub-adult salmon body weight and condition factor (cor = 0.64, p<0.001), while sub-adult length and condition factor were moderately correlated (cor = 0.42, p<0.001) but included as covariates in the same model. Post-smolt condition factor was not strongly correlated to length (cor = 0.16, p = 0.009) nor weight (cor = -0.17, p = 0.004). The parsimony principle was used to select the model with the lowest Akaike Information Criterion (AIC) (Burnham 2002). GAMMs were fitted using the R statistical programming v.4.2.2 environment and the package mgcv v. 1.8-41(Wood 2011). All model assumptions were visually assessed using Q–Q plots and residual variation vs. fitted values and leverage. To avoid overfitting the smooth functions, the maximum number of basis dimensions (“knots”) was restricted to 5 for univariate smoothers and 20 for the two-dimensional smoother of geographical position. Results were visualized by using the R-packages ggplot2 (Wickham, 2016), patchwork (Pedersen 2024) and itsadug (van Rij et al. 2022). K-mean clustering Cluster analysis was done using partitioning methods (k-means). The optimum number of clusters for the k-mean clustering was determined to be four using the R-package NbClust (vers. 3.0.1) (Charrad et al. 2014). Data was visualised using the R-package ComplexHeatmap (ver. 2.6.2) (Gu et al. 2016). Results Lipids in post-smolts and sub-adult salmon Post-smolts in the Norwegian Sea had low levels of total lipids with an average value of 1.98 ± 0.45% (min=1.44 %, max=5.16 %), which was significantly lower (Mann-Whitney U test, p<0.001) than the average total lipid levels of 6.88 ± 4.06 % (min=1.69%, max=16.54%) for sub-adult salmon (Fig. 2a). The lipid levels increased with the condition factor for both post-smolts and sub-adult salmon, although the relationship was not linear (Fig. 2b, c). Furthermore, the lipid levels in post-smolts increased with body weight (Fig. 2d). The dependance of lipid levels on body weight for sub-adult salmon was not tested, as body weight and condition factor were highly correlated. Neither geographical position nor sampling month explained variation in lipid levels for sub-adult salmon, but including geographic position slightly improved the model for post-smolt lipid levels. FA composition in post-smolts and sub-adult salmon The variation in FA composition in TAG among individual post-smolts was larger than among individual sub-adult salmon (PERMDIST, F 1,366 = 50.13, p<0.001, Fig. 3a). The following FAs were the five most abundant for both salmon life-stages; 18:1n-9, 22:6n-3, 16:0, 22:1n-11, 20:1n-9. Among the 20 most abundant FAs (Fig. 4a, Tab. S6-S9 show all 67 FAs), 22:6n-3 was more abundant in post-smolts (SIMPER, p-value < 0.001) while 16:1n-7, 18:1n-7 and 22:1n-9 were more abundant in sub-adult salmon (SIMPER, p<0.05) (Fig. 3b). There was also a significant difference for several of the less common FAs. A PCoA was carried out for post-smolts only to further explore the variation in FA composition in TAG. Sampling year had a significant effect on the FA composition (ANOSIM, R= 0.1997, p=0.001) and the associated variance (PERMDIST, F 8,273 = 14.167, p<0.001). A change in the FA composition was noticed in the years after 2016 (Fig. 3c). To identify the FAs explaining most of the temporal change in FA composition, a SIMPER was run with years aggregated into two periods, 2012-2016 and 2017-2020. The relative abundance of 61 of the 67 FAs changed from the first to the second period (SIMPER p<0.05), and the most noticeable change was the increased abundance of FAs associated with calanoid copepods such as 22:1n-11 and 20:1n-9 and decreased abundance of 16:0 and 18:0 in the last period. The results from the K-mean cluster analysis (results presented further below) further support a change in post-smolt FA composition after year 2016. Post-smolts sampled in different geographic regions (south, northeast and northwest) of the Norwegian Sea had significantly different FA compositions in TAG (ANOSIM, R=0.1768, p=0.001) and associated variance (PERMDIST, F 2,279 = 5.376, p=0.005) (Fig. 3d). The relative abundance of most FAs was significantly different for post-smolts sampled in the northern (both northeastern and northwestern regions) and those sampled in the southern Norwegian Sea (SIMPER, p<0.05). The only FAs that had a significantly different relative abundance in post-smolts between the northeastern and northwestern Norwegian Sea was 21:5n-3 (SIMPER, p=0.018). The PC1 for post-smolt mainly represents the relative levels of FAs associated with Calanus spp. (20:1n-9, 22:1n-9, 22:1n-11) and FAs associated with age-0 fish (22:6n-3) (Fig. 3e) and is related to post-smolt sampling location (Fig. 5a). The relative level of Calanus biomarkers is relatively more abundant in the regions north of 67˚N (Fig. 5b) while the age-0 fish biomarker was higher in post-smolt sampled south of 67˚N in the Norwegian Sea and further northeast close to the Norwegian coast (Fig. 5c). The PC1 decreases with increasing lipid levels (Fig. 5d). Hence, post-smolts with relatively high levels of FAs associated with Calanus (20:1n-9, 22:1n-9, 22:1n-11) had higher lipid levels than post-smolts with relatively high levels of FAs associated with age-0 fish (22:6n-3). The PC1 is not associated with post-smolt condition factor, as including condition factor as covariate did not improve the model fit (Tab. S5). TAG and TL for post-smolts and sub-adult salmon The FA composition was significantly different in TAG and TL for post-smolts (muscle samples) (Fig. S4a, PERMANOVA, F 1,556 = 30.7, p<0.001) but with a similar dispersion (PERMDIST, F 1,556 = 0.0096, p=0.9219). For sub-adult salmon the FA composition was not significantly different in TAG and TL (Fig. S4b, PERMANOVA, F 1,161 = 1.576, p=0.21) and the dispersion was similar (PERMDIST, F 1,161 = 0.655, p=0.45). Due to the significant differences in FA profiles for post-smolt TAG and TL, the different FA compositions in TAG and TL was further explored. The difference was mainly seen in the FAs 22:6n-3 and 20:5n-3, which were higher in TL than in TAG, and 16:1n-7, 18:1n-9, 20:1n-9, 20:1n-11 and 22:1n-9 which were higher in TAG than in TL (Fig. 4a, Tab. S6-S9). FAs in pelagic fish and its prey The FA composition of 20 most abundant FAs of salmon, herring, mackerel and prey organisms (calanoid copepods, amphipods, euphausiids and herring larvae) are given in Fig. 4 (Tab. S6-S19 show all the 67 FAs). Difference in FA composition for salmon and herring/mackerel The first two principal components (PC1 and PC2) from a PCoA for TAG in all fish and zooplankton species showed that the FA compositions of post-smolts and sub-adult salmon differ from those of herring and mackerel, with the two latter species having a similar FA composition (Fig. 6a). PC1 mainly represents the variation in saturated FA, C18 MUFAs and long-chain PUFAs, all of which are associated with feeding on carnivorous zooplankton and age-0 fish (positive PC1 score), and C16-PUFA and other FAs associated with feeding on herbivores zooplankton (negative PC1 score) (Fig. 6b, c). Of the 67 tested FAs, the number of FAs with significantly different relative levels were 23 for post-smolts and herring, 14 for post-smolts and mackerel, 18 for sub-adult salmon and herring and 13 for sub-adult salmon and mackerel. Many of the FAs with significantly different relative levels in salmon compared to herring and mackerel are only present at relatively low levels. The FAs in post-smolts or sub-adult salmon, and herring or mackerel, that had significantly different relative levels were FAs associated with feeding on carnivorous zooplankton and age-0 fish (higher in salmon) or FAs associated with feeding on herbivores zooplankton (lower in salmon). For details regarding statistical comparison of all specific FAs among species the reader is invited to see Tables S20-S26. K-mean cluster analysis suggested four clusters in the dataset (Fig. 7). Cluster 1 contains a mix of mackerel and herring (50 % and 49 %), while Cluster 2 contains a mix of sub-adult salmon (41%) and post-smolts (59%). The last two clusters (Cluster 3 and 4) are dominated by post-smolts (98% and 95 %). These two clusters (3 and 4) have higher relative levels of long-chain PUFAs (≥C20), while the clusters containing herring, mackerel and sub-adult salmon have higher relative levels of long-chain MUFAs (≥C20) (Fig. 7). An important difference between Cluster 1 and 2 are the higher levels of short-chain PUFAs (C16 and C18) and saturated FAs (14:0) in Cluster 1 (containing mackerel and herring) than in Cluster 2 (sub-adult salmon and post-smolts). The separation of post-smolts into three different clusters (1, 2, 3) is partly explained by difference in post-smolt size, lipid levels and sampling year. These results mimic the output from the PCoA analyses (Fig. 3 and 4) and are therefore not presented further in detail (see supp. information section S7 for more information). The variable (FAs) along the y-axes is colored to be relative higher (yellow) or lower (blue) in the different clusters. Difference in FA composition for salmon and selected prey organisms The relative levels of most FAs in post-smolts and sub-adult salmon TAG were significantly different from the levels in zooplankton and herring larvae (SIMPER, p<0.05, Fig. 4, see example Tab. S25-S26). The PC1 and PC2 from the PCoAs, including all fish and prey species, show how the FA composition in post-smolts and sub-adult salmon relates to the selected prey organisms (Fig. 6). Along the PC1, post-smolts are clustered in the same space as herring larvae, amphipods and euphausiids (Fig. 6). Compared to the other prey organisms, herring larvae had high levels (≈30%) of 22:6n-3 (Fig. 4b) and were clustered with positive values on the PC1-axis in the two-dimensional space (Fig. 6b). Herring larvae had also high relative levels of 16:0 and 22:5n-3, and so do amphipods and euphausiids (Fig. 4b, 6b). C. finmarchicus and C. hyperboreus have low relative levels of 16:0, 22:6n-3 and 22:5n-3 and high relative levels of 18:4n-3, 20:1n-9 and 22:1n-11 (Fig. 4b), and these two species are clustered in the two-dimensional space with negative values on the PC1-axis (Fig. 6b). Along the PC1-axis there was partly an overlap between sub-adult salmon and the two Calanus species, but not between post-smolts and Calanus spp. Herring and mackerel had high relative levels of 18:4n-3, 20:1n-9 and 22:1n-11 and overlapped with Calanus along the PC1-axis (Fig. 6a, b). Post-smolts and sub-adult salmon differed from Calanus spp. by having significantly lower levels of 14:0 and short-chains PUFAs C16 and C18 (SIMPER, p<0.05). Interestingly, FAs known as products of endogenous metabolism, 18:1n-11 and 22:5n-3, were relatively higher in post-smolts and sub-adult salmon than in all prey groups (Fig. 4). For post-smolts, the relative proportion of 22:6n-3 and 22:5n-3 was higher in TL than in TAG, while the relative proportion of 18:4n-3, 20:1n-9 and 22:1n-11 were higher in TAG than in TL (Fig. 4). Hence, post-smolts have a higher proportion of FAs associated with age-0 fish in the membrane lipids than in TAG, while TAG has a higher proportion of FAs associated with Calanus spp. than the membrane lipids (see also Fig S5). Discussion The present study has identified huge variation among post-smolt FA composition, reflecting their diverse diet and opportunistic feeding behavior. A large proportion of the post-smolts sampled early in the summer had a high proportion of FA associated with age-0 fish, and these FAs are mainly allocated for somatic growth. Later in the summer, when they have reached the northern Norwegian Sea, post-smolts have a higher proportion of FAs associated with calanoid copepods. Post-smolts have very low lipids levels compared to sub-adult salmon or other pelagic fish feeding in the same geographic region of the northeast Atlantic. Furthermore, the FA composition of post-smolts, which are hypothesized to compete for prey with other pelagic fish species, is partly different from the FA composition of herring and mackerel caught within the same geographic area, suggesting important diet differences between the species. Temporal and geographic variation in FAs for salmon The FA composition varied substantially among post-smolts. This is, most likely, due to geographic differences in prey abundance and the number of days post-smolts have spent at sea. Post-smolts sampled in the Norwegian Sea originate from rivers over a large geographic area (Gilbey et al. 2021), where individuals from several countries commence their marine feeding period in completely different regions of the Northeast Atlantic. Therefore, these individuals do not encounter the same prey species during their first period at sea. In any case, the large variation in FA profiles also reflects that salmon are opportunistic feeders, able to target a wide range of prey (Jacobsen and Hansen 2001; Haugland et al. 2006). The variation in FA among post-smolts was especially large for FATMs associated with calanoid copepods and with age-0 fish. The herring larvae analysed for FAs in this study had high levels of long-chain PUFAs such as 22:6n-3, which is a FA also abundant in age-0 fish of other species (Klungsøyr et al. 1989). The proportion of 22:6n-3 remained high in herring sampled from the yolk-sac phase and the following months during the spring and early summer (Fraser et al. 1989), but the relative abundance of long-chain PUFAs is much lower in adult herring and mackerel, as shown in the present study. The Euphausiids and amphipods analyzed in this study also had high levels of 22:6n-3 (~15%), although substantially lower than the observed levels in herring larvae and post-smolts. Hence, we conclude that the high level of 22:6n-3 in post-smolts most probably originates from feeding on age-0 fish. Post-smolts sampled in the northern Norwegian Sea had higher relative levels of FAs associated with calanoid copepods, while fish from the southern Norwegian Sea had higher relative levels of FAs associated with age-0 fish. A north-south gradient for the post-smolt diet has previously been documented from stomach samples, with the biomass of age-0 fish decreasing and the biomass of amphipods in sampled post-smolt stomachs increasing when post-smolts migrate northwards (Utne et al. 2022). The study presented here found a positive correlation between post-smolt lipid levels and FAs associated with calanoid copepods, and high levels of these FAs as well as high lipid levels in sub-adult salmon. Although post-smolts sometimes feed directly on calanoid copepods, it is apparent from stomach content analyses that this prey group is not an important part of the post-smolt diet (Rikardsen and Dempson 2010; Sheehan et al. 2012). We suggest that FAs associated with calanoid copepods in salmon mostly originate from feeding on predators of calanoid copepods, such as amphipods and euphausiids, cephalopods and small fish. The post-smolt FA composition changed in the period 2012-2020, with the main shift occurring around 2016-2017. Vollset et al. (2022) documented how changes in oceanographic conditions and abundance of plankton in the Norwegian Sea in the period 2004/2005 occurred simultaneously as a drop in marine growth during the first year at sea for Norwegian salmon. From 2017 and onwards the oceanographic conditions shifted back towards the pre-2004 conditions, and the zooplankton abundance increased (Skagseth et al. 2022). The change in post-smolt FA composition occurring around 2017 presented in this study supports previous findings of a change in diet for post-smolts in the Norwegian Sea occurring during the same period. The abundance of zooplankton in post-smolt stomachs increased (Utne et al. 2022) and the FA composition presented in this study had a higher relative abundance of FAs associated with calanoid copepods in the period 2017-2020 compared to 2012-2016. The post-smolts were on average sampled ~1 week later and ~1˚ further north in 2017-2020 than in 2012-2016, but with variation among years (Tab. S1), in contrast to the consistent shift in FA composition for post-smolts when comparing the years 2012-2016 and 2017-2020. The FA composition of sub-adult salmon was not significantly different from those of post-smolts, although the FA composition among sub-adult salmon had lower variation than among post-smolts. The most important FAs for sub-adult salmon sampled in the Norwegian Sea was the same as for salmon returning to Canadian rivers after feeding in the Northwest Atlantic Ocean (Bøe et al. 2019). The single FA that differed most between salmon and post-smolts was higher proportion of 22:6n-3 in post-smolts. The PCoA showed that sub-adult salmon FA composition overlap with both euphausiids and amphipods on the PC1, and with C. hyperboreus on the PC2. Unlike for post-smolts, the FA composition of sub-adult salmon does not overlap with C. finmarchicus on the first two PCs. While amphipods and C. hyperboreus primarilyinhabits cold Arctic water masses in the northern Norwegian Sea and further west towards Greenland (Skjoldal 2004), C. finmarchicus primarily inhabits Atlantic water masses in the Norwegian Sea (Strand et al. 2020). Hence, the FA composition of sub-adult salmon is more similar to calanoid copepods associated with cold Arctic water than the warm Atlantic water. We therefore suggest, based on the results presented here, that preferred feeding areas for sub-adult salmon are western parts of the Norwegian Sea, the Arctic front and areas further west towards Greenland. The link between Selective use of FAs, lipid levels and growth of Atlantic salmon The results showed that most post-smolts have low levels of total body lipids (≈2 %). Furthermore, the results presented here showed that post-smolt lipids were dominated by membrane lipids (PL), while sub-adult salmon have most of their stored FAs as storage lipids (TAG), as previously demonstrated from feeding experiments for salmon in tanks (Jobling and Johansen 2003). During smoltification, Atlantic salmon mobilize energy reserves and deplete storage lipids, which are used as catabolic energy for the adaptation to seawater (Sheridan 1989). Hence, the protein content increases in the period around smoltification (Stefansson et al. 2003). For post-smolts, but not for sub-adult salmon, the FA composition in storage lipids (TAG) was significantly different to the composition in the total lipids (TL). Post-smolts have a much higher proportion of FAs associated with age-0 fish in their membrane lipids than in their TAG, suggesting that they quickly utilize long-chain PUFAs for somatic growth rather than for lipid storage. This agrees with salmon using some FAs directly for metabolic costs while other FAs are used for membrane lipids (Litz et al. 2017b), in which case the FA composition of prey is partly modified before allocated to membrane lipids, and that salmon have some ability to perform endogenous synthesis of long-chain PUFAs (Mock et al. 2019). In addition, salmon encountering poor feeding conditions may deplete their FAs originating from calanoid copepods before they deplete FAs originating from age-0 fish (Litz et al. 2017b). It can be beneficial for post-smolts to feed on age-0 fish during the first months at sea, as age-0 fish may be low on lipids but contain a nutrient composition with high levels of long-chain PUFAs in the right proportion for post-smolts to utilize directly for body growth. A high proportion of FAs associated with age-0 fish have also been documented in coho ( Oncorhynchus kisutch ) and chinook ( Oncorhynchus tshawytscha ) post-smolts sampled in the Pacific Ocean within their first months at sea (Daly et al. 2010). Juvenile Chinook salmon is known to shift their energy allocation from somatic growth to lipid storage towards the end of the first ocean summer at sea (MacFarlane 2010; Litz et al. 2017a). Among the sampled post-smolt presented here, the lipid levels were highest when the FA composition had a high proportion of FAs associated with calanoid copepods. These FAs were most abundant in post-smolts sampled in the northern Norwegian Sea. However, there is a correlation between geographic area, time and size of post-smolts sampled in the Norwegian Sea, as they grow while migrating northwards during the summer (Gilbey et al. 2021, Utne et al. 2021a). Farmed Atlantic post-smolts also has low lipid accumulation in the first period after transfer to seawater, even when the fish are fed with highly lipid-rich feed, and lipid accumulation in the muscle only starts at the end of summer (August/September) (Alne et al. 2011; Dessen et al. 2017). This could suggest that accumulation of lipids is related primarily to season and/or body size rather than prey availability. Interactions between salmon and other pelagic fish The observed lipid levels in post-smolts (≈2 %) during the summer are lower than the lipid levels in mackerel and herring (at the lowest around 8 % during winter and early spring) (Slotte 1999; Jansen et al. 2021), but similar to the observed lipid levels in post-smolt of Coho and Chinook salmon in the Pacific Ocean (Daly et al. 2010). Furthermore, the lipid levels in mackerel and herring increase quickly once they commence feeding in late-spring, and their lipid levels peak (≈25 %) around August-September. The low lipid levels in post-smolts caught in July and August suggest that Atlantic salmon have a different energy allocation strategy than mature herring and mackerel. Prioritizing somatic growth over lipid accumulation is reasonable for salmon, given their documented life-history strategy that emphasizes rapid growth during the first year at sea (Klemetsen et al. 2003), as opposed to, for instance, the life-history strategy of herring and mackerel that exhibit relatively slow growth, longevity and multiple years of spawning (Trenkel et al. 2014). Alternatively, low lipid levels in post-smolts may reflect poor feeding conditions the years of post-smolts sampling in this study. The condition factor of the post-smolts sampled in the Norwegian Sea during the years 2012-2019 was significantly lower than during the years 1995-2003 (Utne et al. 2021a). A similar temporal reduction in body-length of salmon returning to river after one year at sea has also been reported for Norwegian, British and French salmon (Vollset et al. 2022, Trehin et al. 2023). Interspecific competition with mackerel and herring can potentially have a negative effect on post-smolt growth and survival (Potter and Corzier 2000), and the diet of the three species partly overlaps when feeding in the Norwegian Sea (Utne et al. 2021b). As presented here, FA profiles varies among post-smolt but in average only partly overlap with herring and mackerel, while the FA profiles of the two latter species only have relatively small differences. We interpret this as an indication for the post-smolt diet niche at least partly deviating from the diet niche of mackerel and herring. However, anadromous salmonids have faster lipid synthesis than most other marine fish and are able to modify the dietary FAs allocated to storage lipids (Tocher 2010). The relatively higher proportion of minor C18 PUFA in the n-7 and n-4 family in the post smolt/salmon compared with herring and mackerel are likely to be elongation of the corresponding C16 PUFA in the n-7 and n-4 family, which were more abundant in herring and mackerel than in salmon. For example, the relative high proportion of 20:4n-3 and 22:5n-3 found in salmon may originate from elongation of 18:4n-3 and 20:5n-3, respectively. A high degree of modification of FAs stored as muscle tissue in salmon, but not in herring and mackerel, would overestimate the dietary differences between salmon and other pelagic fish. However, the FAs contributing to the largest difference between post-smolts and herring/mackerel, namely 22:6n-3 is not modified and thus originate directly from the diet. Any differences in FA modifications among these species is unlikely to fully explain the relatively large niche differences seen between post-smolts and the two other pelagic species. Herring and mackerel were only sampled in three years at less than 10 geographic locations. Sampling herring and mackerel in a larger geographic area and for a longer time period, could potentially reveal areas or periods where their diet niche is more similar to that of post-smolt than shown with the data available for this study. Future perspectives These results presented in this study have provided novel data on FA composition of salmon and other pelagic fish sampled at the marine feeding areas. As salmon are not observed or regularly caught during most of their marine phase, there is still a general knowledge gap for salmon at sea during the autumn, winter and early spring. Hence, future sampling during these seasons could improve our understanding of the link between diet and growth throughout the year, which could be valuable knowledge to understand salmon life-history strategies and factors affecting mortality during the marine phase. Declarations Acknowledgments The work was funded by the Research Council of Norway as part of project 280308 SeaSalar and 243895 EcoNorSe. The authors would also like to thank Therese Smith-Jahnsen Aase and Arve Fossen for technical assistance with analyzing samples in the lab. We thank the many people who assisted with gathering the data presented here, including the skippers and crew onboard the vessels collecting samples. Data availability Data generated or analyzed during this study are available as open-access and can be downloaded from the following link: https://doi.org/10.6084/m9.figshare.28505816.v1 Competing interests The authors have no competing interests to declare. References Alne H, Oehme M, Thomassen M, Terjesen B, Rørvik KA (2011) Reduced growth, condition factor and body energy levels in Atlantic salmon Salmo salar L. during their first spring in the sea. Aquac Res 42(2):248-259. https://doi.org/10.1111/j.1365-2109.2010.02618.x Anderson MJ (2006) Distance-based tests for homogeneity of multivariate dispersions. Biometrics 62(1):245-253. doi:10.1111/j.1541-0420.2005.00440.x. Bachiller E, Skaret G, Nøttestad L, Slotte A (2016) Feeding Ecology of Northeast Atlantic Mackerel, Norwegian Spring-Spawning Herring and Blue Whiting in the Norwegian Sea. Plos One 11(2): 25. doi:10.1371/journal.pone.0149238. Beaugrand G, Reid PC (2003) Long-term changes in phytoplankton, zooplankton and salmon related to climate. Glob Change Biol 9(6):801-817. doi:https://doi.org/10.1046/j.1365-2486.2003.00632.x. Beaugrand G, Reid PC (2012) Relationships between North Atlantic salmon, plankton, and hydroclimatic change in the Northeast Atlantic. ICES J Mar Sci 69(9):1549-1562. https://doi.org/10.1093/icesjms/fss153 Bøe K, Power M, Robertson MJ, Morris CJ, Dempson JB, Pennell CJ, Fleming IA (2019) The influence of temperature and life stage in shaping migratory patterns during the early marine phase of two Newfoundland (Canada) Atlantic salmon ( Salmo salar ) populations. Can J Fish Aquat Sci 76(12):2364-2376. https://doi.org/10.1139/cjfas-2018-0320 Brix O, Gruner R, Ronnestad I, Gemballa S (2009) Whether depositing fat or losing weight, fish maintain a balance. Proc R Soc Biol Sci Ser B 276(1674):3777-3782. https://doi.org/10.1098/rspb.2009.1079 Budge SM, Iverson SJ (2003) Quantitative analysis of fatty acid precursors in marine samples: direct conversion of wax ester alcohols and dimethylacetals to FAMEs. J Lipid Res 44(9):1802-1807. https://doi.org/10.1194/jlr.D300009-JLR200 Budge SM, Penney SN, Lall SP (2011) Response of tissue lipids to diet variation in Atlantic salmon ( Salmo salar ): Implications for estimating diets with fatty acid analysis. J Exp Mar Biol Ecol 409(1-2):267-274. doi:10.1016/j.jembe.2011.09.002. Budge SM, Penney SN, Lall SP (2012) Estimating diets of Atlantic salmon (Salmo salar) using fatty acid signature analyses; validation with controlled feeding studies. Can J Fish Aquat Sci 69(6):1033-1046. doi:10.1139/f2012-039. Burnham, KP, Anderson DR (2002) Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer-Verlag, New York. Charrad M, Ghazzali N, Boiteau V, Niknafs A (2014) NbClust: An R Package for Determining the Relevant Number of Clusters in a Data Set. J Stat Softw 61(6):1-36. doi:10.18637/jss.v061.i06. Dalpadado P, Ellertsen B, Melle W, Dommasnes A (2000) Food and feeding conditions of Norwegian spring-spawning herring ( Clupea harengus ) through its feeding migrations. ICES J Mar Sci 57(4):843-857. https://doi.org/10.1006/jmsc.2000.0573. Dalsgaard J, St John M, Kattner G, Muller-Navarra D, Hagen W (2003) Fatty acid trophic markers in the pelagic marine environment. Adv Mar Biol 46:225-340. DOI:10.1016/S0065-2881(03)46005-7 Daly EA, Benkwitt CE, Brodeur RD, Litz MNC, Copeman LA (2010) Fatty acid profiles of juvenile salmon indicate prey selection strategies in coastal marine waters. Mar Biol 157:1975–1987. https://doi.org/10.1007/s00227-010-1466-9. Dessen JE, Weihe RN, Hatlen B, Thomassen MS, Rørvik KA (2017) Different growth performance, lipid deposition, and nutrient utilization in in-season (S1) Atlantic salmon post-smolt fed isoenergetic diets differing in protein-to-lipid ratio. Aquaculture 473:345-354. doi:10.1016/j.aquaculture.2017.02.006. Folkvord A, Haaland IM, Grahl-Nielsen O, Johannessen A (1996) Fatty acid composition of herring ( Clupea harengus L ) larvae off western Norway. SARSIA 80(4):299-305. https://doi.org/10.1080/00364827.1996.10413604 Fraser AJ, Sargent JR, Gamble JC, Seaton DD (1989) Formation and transfer of fatty-acids in an enclosed marine food-chain comprising phytoplankton, zooplankton and herring ( clupea-harengus l ) larvae. Mar Chem 27(1-2):1-18. https://doi.org/10.1016/0304-4203(89)90024-8 Gilbey J, Utne KR, Wennevik V, Beck AC, Kausrud K, Hindar K, de Leaniz CG, Cherbonnel C, Coughlan J, Cross TF, Dillane E, Ensing D, Garcia-Vazquez E, Hole LR, Holm M, Holst JC, Jacobsen JA, Jensen AJ, Karlsson S, Maoileidigh NO, Mork KA, Nielsen EE, Nøttestad L, Primmer CR, Prodohl P, Prusov S, Stevens JR, Thomas K, Whelan K, McGinnity P, and Verspoor E (2021) The early marine distribution of Atlantic salmon in the North-east Atlantic: A genetically informed stock-specific synthesis. Fish Fish 22(6):1274-1306. doi:10.1111/faf.12587. Gu Z, Eils R, Schlesner M (2016) Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32(18):2847-2849. doi:10.1093/bioinformatics/btw313. Hara A., Radin NS (1978) Lipid extraction of tissues with a low-toxicity solvent. Anal Biochem 90:420-426. Haug T, Biuw M, Gjøsæter H, Knutsen T, Lindstrom U, MacKenzie KM, Meier S, Nilssen KT (2021) Harp seal body condition and trophic interactions with prey in Norwegian high Arctic waters in early autumn. Prog Oceanogr 191. doi:10.1016/j.pocean.2020.102498. Haugland M, Holst JC, Holm M, Hansen LP (2006) Feeding of Atlantic salmon ( Salmo salar L.) post-smolts in the Northeast Atlantic. ICES J Mar Sci 63(8):1488-1500. doi:10.1016/j.icesjms.2006.06.004. ICES (2013) Report of the Workshop on Northeast Atlantic Mackerel monitoring and methodologies including science and industry involvement (WKNAMMM), 25–28 February 2013, ICES Headquarters, Copenhagen and Hirtshals, Denmark. ICES CM 2013/SSGESST:18. 33 pp. ICES (2024) Working group on North Atlantic Salmon (WGNAS). ICES Scientific Reports. 6:36. 415 pp. https://doi.org/10.17895/ices.pub.25730247 Iverson SJ, Field C, Bowen WD, Blanchard W (2004) Quantitative fatty acid signature analysis: A new method of estimating predator diets. Ecol Monogr 74(2):211-235. Jacobsen JA, Hansen LP (2000) Feeding habitats of Atlantic salmon at different life stages at sea. In The ocean life of Atlantic salmon — environmental and biological factors influencing survival. Fishing News Books, Oxford. Jacobsen JA, Hansen LP (2001) Feeding habits of wild and escaped farmed Atlantic salmon, Salmo salar L ., in the Northeast Atlantic. ICES J Mar Sci 58(4):916-933. https://doi.org/10.1006/jmsc.2001.1084 Jacobson P, Gårdmark A, Östergren J, Casini M, Huss M (2018) Size-dependent prey availability affects diet and performance of predatory fish at sea: a case study of Atlantic salmon. Ecosphere 9(1): e02081. doi:https://doi.org/10.1002/ecs2.2081. Jansen T, Slotte A, Schmidt TCD, Sparrevohn CR, Jacobsen JA, Kjesbu OS (2021) Bioenergetics of egg production in Northeast Atlantic mackerel changes the perception of fecundity type and annual trends in spawning stock biomass. Prog Oceanogr 198. doi:10.1016/j.pocean.2021.102658. Jobling M, Johansen SJS (2003) Fat distribution in Atlantic salmon Salmo salar L . in relation to body size and feeding regime. Aquac Res 34(4):311-316. https://doi.org/10.1046/j.1365-2109.2003.00820.x Kassambara A, Mundt F (2020). Factoextra: Extract and Visualize the Results of Multivariate Data Analyses. R package version 1.0.7, https://CRAN.R project.org/package=factoextra Klemetsen A, Amundsen PA, Dempson JB, Jonsson B, Jonsson N, O'Connell MF, Mortensen E (2003) Atlantic salmon Salmo salar L., brown trout Salmo trutta L. and Arctic charr Salvelinus alpinus (L.): a review of aspects of their life histories. Ecol Freshw Fish 12(1):1-59. doi:https://doi.org/10.1034/j.1600-0633.2003.00010.x. Klungsøyr J, Tilseth S, Wilhelmsen S, Falk-Petersen S, Sargent J.R (1989) Fatty-acid composition as an indicator of food-intake in cod larvae gadus-morhua from Lofoten, northern Norway. Mar Biol 102(2):183-188. https://doi.org/10.1007/BF00428278 Lê S, Josse J, Husson F (2008) FactoMineR: An R Package for Multivariate Analysis. Journal of Statistical Software 25(1):1 - 18. doi:10.18637/jss.v025.i01. Litz MNC, Miller JA, Copeman LA, Hurst TP (2017a) Effects of dietary fatty acids on juvenile salmon growth, biochemistry, and aerobic performance: A laboratory rearing experiment. J Exp Mar Biol Ecol 494:20-31. https://doi.org/10.1016/j.jembe.2017.04.007. Litz MNC, Miller JA, Copeman LA, Teel DJ, Weitkamp LA, Daly EA, Claiborne AM (2017b) Ontogenetic shifts in the diets of juvenile Chinook Salmon: new insight from stable isotopes and fatty acids. Environ Biol Fishes 100(4):337-360. DOI: 10.1007/s10641-016-0542-5 MacFarlane RB (2010) 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 67(10):1549-1565. doi:10.1139/F10-080. Madhun AS, Wennevik V, Skilbrei OT, Karlsbakk E, Skaala O, Fiksdal IU, Meier S, Tang YK, Glover KA (2017) The ecological profile of Atlantic salmon escapees entering a river throughout an entire season: diverse in escape history and genetic background, but frequently virus-infected. ICES J Mar Sci 74(5):1371-1381. https://doi.org/10.1093/icesjms/fsw243 Meier S, Mjos SA, Joensen H, Grahl-Nielsen O (2006) Validation of a one-step extraction/methylation method for determination of fatty acids and cholesterol in marine tissues. J Chromatogr A 1104(1-2):291-298. doi:10.1016/j.chroma.2005.11.045. Mock TS, Francis DS, Jago MK, Glencross BD, Smullen RP, Turchini GM (2019) Endogenous biosynthesis of n-3 long-chain PUFA in Atlantic salmon. Br J Nutr 121(10):1108-1123. doi:10.1017/s0007114519000473. Oksanen J, Simpson GL, Blanchet FB, Kindt R, Legendre P, Minchin PR, O'Hara RB et al. 2024. Ordination methods, diversity analysis and other functions for community and vegetation ecologists. 10.32614/CRAN.package.vegan O'Sullivan RJ, Ozerov M, Bolstad, GH, Gilbey J, Jacobsen JA, Erkinaro J, Rikardsen AH, Hindar K, Aykanat T (2022) Genetic stock identification reveals greater use of an oceanic feeding ground around the Faroe Islands by multi-sea winter Atlantic salmon, with variation in use across reporting groups. ICES J Mar Sci 79(9):2442-2452. doi:10.1093/icesjms/fsac182. Olmos M, Massiot-Granier F, Prévost E, Chaput G, Bradbury IR, Nevoux M, Rivot E (2019) Evidence for spatial coherence in time trends of marine life history traits of Atlantic salmon in the North Atlantic. Fish Fish 20(2):322-342. doi:https://doi.org/10.1111/faf.12345. Olmos M, Payne MR, Nevoux M, Prévost E, Chaput G, Du Pontavice H, Guitton J, Sheehan T, Mills K, Rivot E (2020) Spatial synchrony in the response of a long range migratory species ( Salmo salar ) to climate change in the North Atlantic Ocean. Glob Change Biol 26(3):1319-1337. doi:https://doi.org/10.1111/gcb.14913. Olsen RE, Henderson RJ, Ringo E (1991) Lipids of arctic charr, salvelinus-alpinus (l) .1. dietary induced changes in lipid class and fatty-acid composition. Fish Physiol Biochem 9(2):151-164. https://doi.org/10.1007/BF02265131 Olsen RE, Taranger GL, Svasand T, Skilbrei OT (2013) Improved method for triacylglycerol-derived fatty acid profiling by various non-lethal and lethal sampling techniques in Atlantic salmon. Aquacult Environ Interact 4(3):251-261. DOI: 10.3354/aei00087. Pedersen TL (2024) Patchwork: The Composer of Plots. R package version 1.2.0. https://CRAN.R-project.org/package=patchwork Petursdottir H, Gislason A, Falk-Petersen S, Hop H, Svavarsson J (2008) Trophic interactions of the pelagic ecosystem over the Reykjanes Ridge as evaluated by fatty acid and stable isotope analyses. Deep-Sea Res Part II Top Stud Oceanogr 55(1-2):83-93. https://doi.org/10.1016/j.dsr2.2007.09.003 Pond DW, Ward P (2011) Importance of diatoms for Oithona in Antarctic waters. J Plankton Res 33(1): 105-118. doi:10.1093/plankt/fbq089. Potter ECE, Crozier WW (2000) A perspective on the marine survival of Atlantic salmon. . In The ocean life of Atlantic salmon — environmental and biological factors influencing survival. Edited by D. Mills. Fishing News Books, Oxford. pp. 19–36. Rikardsen AH, Dempson JB (2010) Dietary Life-Support: The Food and Feeding of Atlantic Salmon at Sea. In Atlantic Salmon Ecology. pp. 115-143. Rikardsen AH, Righton D, Strom JF, Thorstad EB, Gargan P, Sheehan T, Okland F, Chittenden CM, Hedger RD, Næsje TF, Renkawitz M, Sturlaugsson J, Caballero P, Baktoft H, Davidsen JG., Halttunen E, Wright S, Finstad B, Aarestrup K (2021) Redefining the oceanic distribution of Atlantic salmon. Sci Rep 11 (1). doi:10.1038/s41598-021-91137-y. Salminen M, Erkamo E, Salmi J (2001) Diet of post-smolt and one-sea-winter Atlantic salmon in the Bothnian Sea, Northern Baltic. J Fish Biol 58(1):16-35. doi:https://doi.org/10.1111/j.1095-8649.2001.tb00496.x. Schäfer F, Oesterwind D, Sell AF, Kammann U (2024) Fatty acid analyses reveal differences in feeding ecology of North Sea squids that overlap in time and space. Food Webs 40: e00355. Sheehan TF, Reddin DG, Chaput G, Renkawitz MD (2012) SALSEA North America: a pelagic ecosystem survey targeting Atlantic salmon in the Northwest Atlantic. ICES J Mar Sci 69(9):1580-1588. doi:10.1093/icesjms/fss052. Sheridan MA (1989) Alterations in lipid-metabolism accompanying smoltification and seawater adaptation of salmonid fish. Aquaculture 82(1-4):191-203. Skagseth O, Broms C, Gundersen K, Hatun H, Kristiansen I, Larsen KMH, Mork KA, Petursdottir H, Søiland H (2022) Arctic and Atlantic Waters in the Norwegian Basin, Between Year Variability and Potential Ecosystem Implications. Front Mar Sci 9 . doi:10.3389/fmars.2022.831739. Skjoldal HR (2004) The Norwegian Sea ecosystem. Tapir academic press, Trondheim, Norway. Slotte A (1999) Differential utilization of energy during wintering and spawning migration in Norwegian spring-spawning herring. J Fish Biol 54(2):338-355. https://doi.org/10.1111/j.1095-8649.1999.tb00834.x Stefansson SO, Bjornsson BT, Sundell K, Nyhammer G, McCormick SD (2003) Physiological characteristics of wild Atlantic salmon post-smolts during estuarine and coastal migration. J Fish Biol 63(4):942-955. https://doi.org/10.1046/j.1095-8649.2003.00201.x Strand E, Bagøien E, Edwards M, Broms C, Klevjer T (2020) Spatial distributions and seasonality of four Calanus species in the Northeast Atlantic. Prog Oceanogr 185. doi:10.1016/j.pocean.2020.102344. Thomas SM, Kainz MJ, Amundsen PA, Hayden B, Taipale SJ, Kahilainen KK (2019). Resource polymorphism in European whitefish: Analysis of fatty acid profiles provides more detailed evidence than traditional methods alone. Plos One 14(8). Tiedemann M, Nash RDM, Stenevik EK, Stiasny MH, Slotte A, Kjesbu OS, Robert D (2021) Environmental influences on Norwegian spring-spawning herring ( Clupea harengusL. ) larvae reveal recent constraints in recruitment success. ICES J Mar Sci78(2):640-652. doi:10.1093/icesjms/fsaa072. Tocher DR (2010) Fatty acid requirements in ontogeny of marine and freshwater fish. Aquac Res 41(5):717-732. doi:10.1111/j.1365-2109.2008.02150.x. Tréhin C, Rivot E, Santanbien V, Patin R, Gregory SD, Lamireau L, Marchand F, Beaumont WRC, Scott LJ, Hillman R, Besnard AL, Boisson PY, Meslier L, King AR, Stevens JR, Nevoux M (2023) A multi-population approach supports common patterns in marine growth and maturation decision in Atlantic salmon (Salmo salar L.) from southern Europe. J Fish Biol 104 : 125-138. doi:https://doi.org/10.1111/jfb.15567. Trenkel VM, Huse G, MacKenzie BR, Alvarez P, Arrizabalaga H, Castonguay M, Gani N, Gregoire F, Hatun H, Jansen T, Jacobsen JA, Lehodey P, Lutcavage M, Mariani P, Melvin GD, Neilson JD, Nøttestad L, Oskarsson GJ, Payne MR, Richardson DE, Senina I, Speirs DC (2014) Comparative ecology of widely distributed pelagic fish species in the North Atlantic: Implications for modelling climate and fisheries impacts. Prog Oceanogr 129:219-243. doi:10.1016/j.pocean.2014.04.030. Tyldesley E, Banas NS, Diack G, Kennedy R, Gillson J, Johns DG, Bull C (2024) Patterns of declining zooplankton energy in the northeast Atlantic as an indicator for marine survival of Atlantic salmon. ICES J Mar Sci 81:1164–1184. https://doi.org/10.1093/icesjms/fsae077 Utne KR, Pauli BD, Haugland M, Jacobsen JA, Maoileidigh N, Melle W, Broms CT, Nøttestad L, Holm M, Thomas K, Wennevik V (2021a) Poor feeding opportunities and reduced condition factor for salmon post-smolts in the Northeast Atlantic Ocean. ICES J Mar Sci78(8):2844-2857. doi:10.1093/icesjms/fsab163. Utne KR, Thomas K, Jacobsen JA, Fall J, Maoiléidigh NO, Broms CT, Melle W (2021b) Feeding interactions between Atlantic salmon ( Salmo salar ) postsmolts and other planktivorous fish in the Northeast Atlantic. Can J Fish Aquat Sci 78(3):255-268. doi:10.1139/cjfas-2020-0037. Utne KR, Skagseth O, Wennevik V, Broms CT, Melle W, Thorstad EB (2022) Impacts of a Changing Ecosystem on the Feeding and Feeding Conditions for Atlantic Salmon During the First Months at Sea. Front Mar Sci 9. doi:10.3389/fmars.2022.824614. van Rij J, Wieling M, Baayen RH, van Rijn H (2022) itsadug: Interpreting time series and autocorrelated data using GAMMs. R package version 2.4.1. Vollset KW, Urdal K, Utne KR, Thorstad EB, Sægrov H, Raunsgard A, Skagseth O, Lennox RJ, Østborg GM, Ugedal O, Jensen AJ, Bolstad GH, Fiske P (2022) Ecological regime shift in the Northeast Atlantic Ocean revealed from the unprecedented reduction in marine growth of Atlantic salmon. Sci Adv 8 (9). doi:10.1126/sciadv.abk2542. Wickham H (2016) Ggplot2: elegant graphics for data analysis. Springer-Verlag, New York https://ggplot2.tidyverse.org Wood SN (2011) Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J R Stat Soc Ser B 73(1):3-36. doi:https://doi.org/10.1111/j.1467-9868.2010.00749.x. Zhukova NV, Aizdaicher NA (1995) Fatty-Acid Composition of 15 Species of Marine Microalgae. Phytochemistry 39(2):351-356. https://doi.org/10.1016/0031-9422(94)00913-E Additional Declarations No competing interests reported. Supplementary Files Suppinfosub1.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7827046","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":541949560,"identity":"01893931-5594-4f6f-9cf7-a3f6bfdbed83","order_by":0,"name":"Kjell Rong Utne","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA50lEQVRIie3QsQqCQBjA8U8CXQLXE8Fe4cShguhZToKcnFpqMwSfobFHcKu2Lw6cDFensDewJ6gTC2robAy6/yB35/3g7gBUqh+MAGpYvSw4tNdNANlzhgDeFwTeiZ92HcxacxTk7AzNrW5f95NgZ4Be1RJiQ8YEWXjjTaWTYz4PDzEY7kZCHMipIMxPSxQk4WHKQbf7UlLUr+QW0C5iQw4tKaKGIOskVpSJg1Hm0VKLR6dk5qZcSyzZXUjJL1W9ZA4t+LFcJdOBGGRE9mJAUHxoM/Cjx5IWfdrcZj7/myjfqFKpVP/bHbdVU3J6ZwsnAAAAAElFTkSuQmCC","orcid":"","institution":"Norwegian Institute of Marine Research","correspondingAuthor":true,"prefix":"","firstName":"Kjell","middleName":"Rong","lastName":"Utne","suffix":""},{"id":541949561,"identity":"a08c6801-345f-40d0-bc03-58ef8d9fb8f5","order_by":1,"name":"Sonnich Meier","email":"","orcid":"","institution":"Norwegian Institute of Marine Research","correspondingAuthor":false,"prefix":"","firstName":"Sonnich","middleName":"","lastName":"Meier","suffix":""},{"id":541949562,"identity":"0545a0ef-6aa8-40d0-9358-bb5175f53e47","order_by":2,"name":"Josef Daniel Rasinger","email":"","orcid":"","institution":"Norwegian Institute of Marine Research","correspondingAuthor":false,"prefix":"","firstName":"Josef","middleName":"Daniel","lastName":"Rasinger","suffix":""},{"id":541949563,"identity":"c805cbc4-db66-46d3-af85-07f2df158b34","order_by":3,"name":"Nikolaos Nikolioudakis","email":"","orcid":"","institution":"Norwegian Institute of Marine Research","correspondingAuthor":false,"prefix":"","firstName":"Nikolaos","middleName":"","lastName":"Nikolioudakis","suffix":""},{"id":541949564,"identity":"f02d330e-0a17-4d85-b327-11e58298228d","order_by":4,"name":"Erling Kåre Stenevik","email":"","orcid":"","institution":"Norwegian Institute of Marine Research","correspondingAuthor":false,"prefix":"","firstName":"Erling","middleName":"Kåre","lastName":"Stenevik","suffix":""},{"id":541949565,"identity":"2556acd9-6cae-4248-9175-10872011e8cf","order_by":5,"name":"Aril Slotte","email":"","orcid":"","institution":"Norwegian Institute of Marine Research","correspondingAuthor":false,"prefix":"","firstName":"Aril","middleName":"","lastName":"Slotte","suffix":""},{"id":541949566,"identity":"88cc87a9-bdef-4c38-bef2-835f90821e5e","order_by":6,"name":"Katja Enberg","email":"","orcid":"","institution":"University of Bergen","correspondingAuthor":false,"prefix":"","firstName":"Katja","middleName":"","lastName":"Enberg","suffix":""},{"id":541949567,"identity":"d09a5de9-589a-4633-b32d-5db7b2076a64","order_by":7,"name":"Vidar Wennevik","email":"","orcid":"","institution":"Norwegian Institute of Marine Research","correspondingAuthor":false,"prefix":"","firstName":"Vidar","middleName":"","lastName":"Wennevik","suffix":""}],"badges":[],"createdAt":"2025-10-10 12:40:42","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7827046/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7827046/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95654632,"identity":"0d4aa0ee-e3bd-4624-be6c-337f2e5dfc49","added_by":"auto","created_at":"2025-11-11 16:12:37","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5895596,"visible":true,"origin":"","legend":"","description":"","filename":"SEASALARFettsyremanusRiFBF.docx","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/f8518208c1fdf41e4e10406c.docx"},{"id":95566344,"identity":"62c69124-7b81-4b21-b649-f5bf67047bb4","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9543,"visible":true,"origin":"","legend":"","description":"","filename":"b512949a079546f0b3e70ea33234b30d.json","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/aeaaa1473c2e6cfd8c5a80a0.json"},{"id":95566354,"identity":"cea4c4c0-50ab-4113-8b7c-6d9471b73ccc","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2034310,"visible":true,"origin":"","legend":"","description":"","filename":"Suppinfosub1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/8ab076a50950c101eaab1635.docx"},{"id":95655444,"identity":"d0d07f14-07cf-4455-953f-be9fa77a42bb","added_by":"auto","created_at":"2025-11-11 16:16:09","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":188718,"visible":true,"origin":"","legend":"","description":"","filename":"b512949a079546f0b3e70ea33234b30d1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/57367dccf31602d76cb461d0.xml"},{"id":95655295,"identity":"02b92c7f-9ec4-47b0-a66f-f5642dfaff73","added_by":"auto","created_at":"2025-11-11 16:15:12","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":709282,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/e2b54968aa03417bec60fd09.jpeg"},{"id":95654603,"identity":"2d4c57eb-a663-47c4-b57a-fc165eee4538","added_by":"auto","created_at":"2025-11-11 16:12:32","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":303146,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/a6b4e51a00993216ddc437fd.jpeg"},{"id":95655504,"identity":"8919d387-c7aa-458d-ab96-1c8cb6a9c8d0","added_by":"auto","created_at":"2025-11-11 16:16:21","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":520332,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/bbb0614b8c07b5643c249f89.jpeg"},{"id":95566362,"identity":"86f7b9a5-fe08-47c7-869a-35addbcd1c3b","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":376466,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/e19753642277ce5d338c30d9.jpeg"},{"id":95655723,"identity":"303ac578-ff34-425a-9e14-0f6326da73a7","added_by":"auto","created_at":"2025-11-11 16:16:48","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":397008,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/91a7ae93c418ce857b9dc651.jpeg"},{"id":95655680,"identity":"e10c26cc-96da-40d8-b0c9-440256b06aa1","added_by":"auto","created_at":"2025-11-11 16:16:42","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":368878,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/e10e3fe0f2f20db5fb1158b6.jpeg"},{"id":95655662,"identity":"c744d70b-fecc-4fd1-b8bc-fa4a332ad190","added_by":"auto","created_at":"2025-11-11 16:16:40","extension":"jpeg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3101598,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/26c92517557dd915335c87a7.jpeg"},{"id":95655626,"identity":"40e1733e-5728-4b8a-ac09-c5b6184ee5c2","added_by":"auto","created_at":"2025-11-11 16:16:34","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":103991,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/fde726ee7662ac4072fd975b.png"},{"id":95566360,"identity":"6ecb004d-ac1a-4fc6-b579-cb116b69b70e","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36631,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/1d51128adee0ba8a3dfda0a2.png"},{"id":95566359,"identity":"4da4d96a-4a07-420f-9660-ca8030773feb","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":84789,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/fba8ac9ad26833ff6163d660.png"},{"id":95655522,"identity":"b069b6f8-0fb0-4fa7-83e5-53a7653488ed","added_by":"auto","created_at":"2025-11-11 16:16:22","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41050,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/2469a3fe0e87b17eae130624.png"},{"id":95566364,"identity":"247afa84-4668-401d-9bcd-134712b9dc78","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52691,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/ba2333962ce2216b85db2eee.png"},{"id":95566366,"identity":"619d844b-d72a-4092-bad9-3b94d823815b","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52870,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/ff5aa38ac003eb8c14c9422c.png"},{"id":95655575,"identity":"5fcfafcb-9cd6-46a4-91de-7b76690f1573","added_by":"auto","created_at":"2025-11-11 16:16:30","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":110475,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/92c9688ac01ea85b8ab1f6ec.png"},{"id":95566370,"identity":"d018574a-e22d-4ff4-8d2b-9e5364e1e2b5","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"xml","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":188260,"visible":true,"origin":"","legend":"","description":"","filename":"b512949a079546f0b3e70ea33234b30d1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/2ed670eede7dac4d1e2918c8.xml"},{"id":95566371,"identity":"e0955d46-71e4-4ca6-940f-d5c5547202a5","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"html","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":199311,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/b3cf2303223e42d4bca3dbc8.html"},{"id":95566345,"identity":"d14a6f1b-a252-4c1d-a801-beab9cd75c93","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":244293,"visible":true,"origin":"","legend":"\u003cp\u003eThe geographic distribution of sampled pelagic fish and zooplankton a) post-smolts, b) sub-adult salmon, c) Mackerel, d) Herring (adults), e) Herring (larvae), f) \u003cem\u003eC.\u003c/em\u003e \u003cem\u003efinmarchicus\u003c/em\u003e, g) \u003cem\u003eC.\u003c/em\u003e \u003cem\u003ehyperboreus\u003c/em\u003e, h) \u003cem\u003eThemisto\u003c/em\u003e spp. The red lines in panel A show the geographical division of regions applied in PCoA to test for different FA compositions in post-smolts.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/2cf19c734c3def68043a773a.png"},{"id":95654619,"identity":"31467c3e-529b-4e1e-9023-e13932421762","added_by":"auto","created_at":"2025-11-11 16:12:36","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":84075,"visible":true,"origin":"","legend":"\u003cp\u003ea) Lipid levels in the sampled salmon. Partial effect plots estimated with generalized additive mixed models for the relationship between lipid levels and b) condition factor for post-smolts, c) condition factor for sub-adult salmon, d) body weight for post-smolts.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/5463c6188c3b42ce44e16566.png"},{"id":95655145,"identity":"a1f30609-acf9-4e62-b765-c677ef8847e7","added_by":"auto","created_at":"2025-11-11 16:14:22","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":185057,"visible":true,"origin":"","legend":"\u003cp\u003ePrincipal coordinate analysis (PCoA) of the FA composition (log10 transformed values) score plot: a) sub-adult salmon and post-smolts, b) Loading plot of the 20 FAs with highest contribution in the PCoA presented in panel a, c) Post-smolts separated by sampling year, d) Post-smolts separated by geographical area of the Norwegian Sea (see Fig. 1a), e) Loading plot of the 20 FAs with highest contribution in the PCoA presented in panel c and d.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/b860e1319e62b67ea15bb7ae.png"},{"id":95566356,"identity":"3b8bc6db-35bb-47ec-99bd-dd6b979d487b","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":104399,"visible":true,"origin":"","legend":"\u003cp\u003ea) FA profiles for the 20 most abundant FAs from post smolt (TL and TAG), sub-adult salmon (TAG), herring (TAG), mackerel (TAG), and b) prey organisms (TL) comprising copepods (\u003cem\u003eC. finmarchicus\u003c/em\u003e and \u003cem\u003eC. hyperboreus\u003c/em\u003e), amphipods (\u003cem\u003eThemisto\u003c/em\u003e sp), euphausiids and herring larvae.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/0f3df8b72bf1cf90e2e6c012.png"},{"id":95566350,"identity":"0635d3ed-cc35-480e-82d0-c33a36664b7d","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":123492,"visible":true,"origin":"","legend":"\u003cp\u003ea) Geographic variation in PC1 for post-smolts (loading plot presented in Fig. 3e), b) Geographic variation of the \u003cem\u003eCalanus\u003c/em\u003ebiomarker for post-smolts, c) Geographic variation of the age-0 fish biomarker for post-smolts, d) Partial effect plot estimated with generalized additive mixed models for the relationship between the PC1 for post-smolts and their lipid levels. The geographic variation presented in panel A-C are predictions based on the relationship between the geographic position and the respective response variable as estimated with generalized additive mixed models (see Tab. S5).\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/54131ea86dbef8c040f14db5.png"},{"id":95566347,"identity":"6b84f87f-b8df-4af8-8d13-034926018d77","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":121591,"visible":true,"origin":"","legend":"\u003cp\u003ePrincipal coordinate analysis (PCoA) of the FAs composition (log-transformed values) for fish (TAG) and prey species (TL) combined. a) score plot for the fish, b) score plot for the prey, c) Loading plot of the 20 FAs with highest contribution in the PCoA.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/63b40b948ab0c0e5a56bc78c.png"},{"id":95566352,"identity":"ede41e16-a7e7-4dda-9ff0-40f6bbdf6a2a","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":329363,"visible":true,"origin":"","legend":"\u003cp\u003eHeat plot from k-mean cluster analysis from the FAs profiles of the\u003cstrong\u003e \u003c/strong\u003eTAG\u003cstrong\u003e \u003c/strong\u003efrom post-smolts, sub-adult salmon, herring and mackerel (n=480). The color on the x-axes shows the species, post-smolt (red), sub-adult salmon (grey), herring (yellow) and mackerel (purple). The variable (FAs) along the y-axes is colored to be relative higher (yellow) or lower (blue) in the different clusters.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/e3255d9423190d292ff49da3.png"},{"id":105033608,"identity":"2f4147a5-2444-4353-abca-b69e89ca0986","added_by":"auto","created_at":"2026-03-20 07:20:34","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1949855,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/4b0d3fe9-efa3-4ae0-a1a0-30d3b688e256.pdf"},{"id":95566348,"identity":"8479ae1c-37dd-4940-8a7b-0e207ca77d08","added_by":"auto","created_at":"2025-11-10 16:18:33","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":2034310,"visible":true,"origin":"","legend":"","description":"","filename":"Suppinfosub1.docx","url":"https://assets-eu.researchsquare.com/files/rs-7827046/v1/ccb51634d85de8ade1504570.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Fatty acids and lipid levels in Atlantic salmon (Salmo salar) change with sea-age and differ from other pelagic fish in the Norwegian sea","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMost Atlantic salmon (\u003cem\u003eSalmo salar\u003c/em\u003e) displays an anadromous life-cycle and juvenile salmon (smolts) leave the rivers and typically spend 1\u0026ndash;3 year at sea before returning to rivers to spawn (Klemetsen et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Salmon have rapid growth during the marine phase, but also a high natural mortality often exceeding 95% year\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (ICES, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The survival at sea for salmon, hereon referred to as \u0026ldquo;marine survival\u0026rdquo;, originating from large parts of the Atlantic Ocean has decreased over several decades (Olmos et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The reasons for this decline in marine survival are complex and multiple factors, operating on varying spatial and temporal scales, may have contributed to a lower marine survival.\u003c/p\u003e\u003cp\u003eThe Norwegian Sea is an important feeding area for post-smolts (salmon during their first summer and autumn at sea) and sub-adult salmon (after their first winter at sea) originating from several European countries (Jacobsen and Hansen \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Jacobsen and Hansen \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Gilbey et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; O'Sullivan et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). During the period 2004\u0026ndash;2012, post-smolts feeding in the Norwegian Sea had gradually lower stomach fullness and condition factor, a result that clearly indicate reduced prey availability for salmon within this geographic region (Utne et al. \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2021a\u003c/span\u003e). Several studies have suggested decreased prey abundance in the Northeast Atlantic attributed to reduced primary production due to altered water circulation and increased water temperatures (Beaugrand and Reid \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Beaugrand and Reid \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Olmos et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Utne et al. \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Tyldesley et al. \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Another hypothesis for the reduced growth and survival of salmon at sea is increased interspecific competition with marine pelagic fish (Potter \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). Especially during the post-smolt phase, the salmon diet can overlap with the diets of mackerel (\u003cem\u003eScomber scombrus\u003c/em\u003e) and herring (\u003cem\u003eClupea harengus\u003c/em\u003e) (Utne et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e). Although competition with pelagic fish does not seem to reduce marine survival of salmon (Utne et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2021b\u003c/span\u003e), more research is needed to better understand how feeding interactions affect the marine growth of salmon.\u003c/p\u003e\u003cp\u003eAtlantic salmon perform long-distance migrations (Gilbey et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Rikardsen et al. \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and must during their marine phase adapt to changing prey composition and abundance. As salmon grows during the marine phase, the size range of the prey they can target changes (Jacobson et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Fast growth enables salmon to feed on larger prey, but at some stage, salmon need to prioritize storage of energy in the form of lipids instead of allocating energy to somatic growth. This shift in energy allocation allows for sufficient reserves to survive the winter, promote gonadal maturation and allow the long migrating back to the rivers. The link between salmon size, accumulation of lipid reserves, and their feeding strategy is poorly understood for wild Atlantic salmon.\u003c/p\u003e\u003cp\u003eAnalysis of specific FAs (fatty acid trophic markers: FATMs) is a well-established method for studying trophic interactions in marine food webs as there is limited modifications of FA structures when transferred to a higher trophic level (Dalsgaard et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). While analyses of stomach content only provide a snapshot of the feeding activity up to a few hours before capture, FATM reflects dietary patterns over the last weeks or months (Budge et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Laboratory experiments with Atlantic salmon given different feed has proven that FATMs deposited in body tissue well reflects their past diet (Budge et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Some FAs are utilized quickly to cover metabolic costs (Iverson et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2004\u003c/span\u003e), but salmon does also have the active enzymatic capacity to modified FA by elongation and desaturation (Tocher, \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), both of which affects the relative proportion of FAs deposited in the fish. A correction factor is therefore needed to accurately quantify their past diet composition (Iverson et al. \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). However, most FAs mainly are deposited directly in the body tissue, and the FA composition without correction still reflects the trophic niche of the predator (e.g. Thomas et al. \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, Sch\u0026auml;fer et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Furthermore, FAs from prey can either be accumulated as membrane lipids (phospholipids, PL), which is a requirement for constructing new cells and therefore needed for somatic growth, or as storage's lipid (triacylglycerols, TAG) available as energy reserve for later use. Total lipid (TL) is the sum of PL and TAG and is normally dominated by TAG in fat fish such as salmon. However, for a lean fish without much stored lipids the proportion of PL relative to TAG is higher than for a fat fish. While TAG largely mirrors the FAs composition of the diet, the PL are under metabolic control and thereby a result of selective allocation of FAs for somatic growth (Olsen et al. \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e1991\u003c/span\u003e). To study the dietary history of fish, Olsen et al. (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) have recommended to analyze FA profiles in TAG instead of in TL.\u003c/p\u003e\u003cp\u003eIn the pelagic food web, phytoplankton are the main primary producers responsible for \u003cem\u003ede novo\u003c/em\u003e synthesis of especially polyunsaturated FAs (PUFAs), like the long-chain PUFAs 20:5n-3 and 22:6n-3 (Zhukova and Aizdaicher \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e1995\u003c/span\u003e). Early life stages of fish larvae feeding heavily on herbivorous copepod nauplii have high levels of phytoplankton FATMs, especially 20:5n-3 and 22:6n-3 (Klungs\u0026oslash;yr et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e1989\u003c/span\u003e; Folkvord et al. \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e1996\u003c/span\u003e). Long chain monounsaturated FAs (MUFAs) such as gondoic acid (20:1n-9), erucic acid (22:1n-9), and cetoleic acid (22:1n-11) are synthesized by copepods such as \u003cem\u003eCalanus\u003c/em\u003e spp. and accumulate at high levels in copepod-eating zooplankton and fish, like sandeel (\u003cem\u003eAmmodytes\u003c/em\u003e spp.), capelin (\u003cem\u003eMallotus villosus\u003c/em\u003e), herring, and mackerel. Thus, high levels of MUFAs indicate feeding on the \u003cem\u003eCalanus\u003c/em\u003e food chain although it does not directly reveal if the diet consisted of \u003cem\u003eCalanus\u003c/em\u003e spp., or a predator of \u003cem\u003eCalanus\u003c/em\u003e (Petursdottir et al. \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). High levels of PUFAs indicate feeding on fish larvae or fish in their post-larva stage (hereafter referred to as 0-age fish) (Litz et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e). The FA profiles from fish sampled at sea can therefore reveal which prey groups dominated the diet the last months prior to capture, and potentially enhance our understanding of the link between the individual\u0026rsquo;s feeding history and its growth and lipid accumulation.\u003c/p\u003e\u003cp\u003eIn this study we present for the first time data on the FA composition of Atlantic salmon, herring, mackerel and various species of zooplankton sampled in marine areas commonly used for feeding by pelagic fish in the Northeast Atlantic Ocean. The objective of the study is to provide new knowledge about marine prey groups important for salmon in the Northeast Atlantic Ocean, the width of the trophic niche of post-smolts and sub-adult salmon, and the overlap with niches of other pelagic fishes in the same marine feeding areas. We determine which FAs dominate in Atlantic salmon and analyze how the FA composition varies among individual fish, between post-smolts and sub-adult salmon, and between Atlantic salmon, herring and mackerel. For post-smolts we test whether the FA composition varies among years, geographic area, or the size of the individual. We also test whether post-smolt TL and TAG have different FA compositions, which would indicate that certain prey groups are utilized directly for somatic growth while other prey groups for storing energy reserves. Furthermore, the trophic position and the feeding niche of Atlantic salmon, herring and mackerel are analyzed by estimating how their FA composition resembles the FA composition of zooplankton and age-0 fish.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003eSampling\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSampling of pelagic fish\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNortheast Atlantic mackerel (\u003cem\u003eScomber scombrus\u003c/em\u003e), Norwegian Spring-Spawning herring (\u003cem\u003eClupea harengus\u003c/em\u003e) and Atlantic salmon were sampled with pelagic trawls at 13 different scientific surveys targeting mackerel or herring. All post-smolts, herring and mackerel were sampled in July-Aug (International Ecosystem Summer Survey in the Nordic Seas, IESSNS), while sub-adult salmon were sampled both during May-June (N=14, International Ecosystem Survey in the Nordic Seas, IESNS) and during July-Aug (N=72, IESSNS). Post-smolts were sampled during 2012-2020, sub-adult salmon during 2013-2020, and herring and mackerel during 2015-2017 (for more details about survey sampling see supplement Tab. S1). The surveys covered the Norwegian Sea and, in some cases, also the northern North Sea. In total 56 mackerel, 54 herring, 294 post-smolts and 87 sub-adult salmon were sampled within the area 60˚13\u0026rsquo;-77˚19\u0026rsquo;N and 12˚28\u0026rsquo;W-23˚24\u0026rsquo;E (Fig. 1). Most sub-adult salmon and all post-smolts, mackerel and herring were sampled with a Multpelt 832 pelagic trawl [for trawl gear information see (ICES 2013)] towed at the surface at 4-5 knots (1 knot = 1.852 km\u0026middot;h\u003csup\u003e\u0026ndash;1\u003c/sup\u003e), but two sub-adults were sampled from trawl hauls within the range 20-410 m depth in May. Trawling was carried out at predetermined locations with spacing of 50-70 nautical miles in July-August and at random locations in May-June. Following the approach of Gilbey et al. (2021), individuals shorter than 35 cm were classified as post-smolts while larger individuals (43.5-79 cm) were classified as sub-adult salmon. None of the sampled salmon had a body-length within the range 35-43 cm. Once the fish were on deck, they were individually weighed to the nearest gram, and the length (fork length for salmon, total length for herring and mackerel) was measured to a 5 mm resolution. The condition factor (CF) was calculated according to Fulton\u0026rsquo;s K (K = body weight / fork length\u003csup\u003e3\u003c/sup\u003e). \u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eSampling of zooplankton and herring larvae.\u003c/h3\u003e\n\u003cp\u003eSamples of copepods (\u003cem\u003eCalanus finmarchicus\u0026nbsp;\u003c/em\u003eand \u003cem\u003eCalanus hyperboreus\u003c/em\u003e), amphipods (\u003cem\u003eThemisto abyssorum\u0026nbsp;\u003c/em\u003eand \u003cem\u003eThemisto libellula\u003c/em\u003e), and euphausiids (\u003cem\u003eMeganyctiphanes norvegica\u003c/em\u003e, \u003cem\u003eThysanoessa inermis\u003c/em\u003e and \u003cem\u003eThysanoessa longicaudata\u003c/em\u003e) were collected from IESNS and IESSNS surveys in 2015-2017 (Tab. S1, sampling positions are shown in Fig. 1). The zooplankton was sampled using a WP2 net (Skjoldal et al. 2004). Species were identified immediately after sample collection (prior to fixation) under a dissecting stereomicroscope and 5-10 individuals were transferred to a 1.5 ml glass tubes, frozen quickly and stored at \u0026ndash;25 \u0026deg;C (Tiedemann et al. 2021). Herring larvae were sampled along the Norwegian coast during the Norwegian Spring-Spawning herring larvae survey in April 2015 and 2016 (Tiedeman et al. 2021). Samples of 10 herring larvae from each sampling location were frozen in 16 ml glass tubes.\u003c/p\u003e\n\u003ch2\u003eStandardization of tissue sampling of salmon\u003c/h2\u003e\n\u003cp\u003eAtlantic salmon has a heterogeneous distribution of fat (Brix et al. 2009). To measure the total amount of lipids in salmon, one must homogenize the whole fish, which is a very time-consuming procedure for large fish. In the present study, the frozen salmon was slightly thawed, and a muscle sample was removed using scalpel. We standardized the sample location of muscle to be taken in front of the dorsal fin, where a 1cm deep, 1 cm wide and 1-2 cm long muscle samples were removed, placed into glass tubes and stored at \u0026nbsp;\u0026ndash; 80 ⁰C until the lipid extraction (Fig. S1). After muscle sample removal, each fish was homogenized, and a subsample was collected for lipid extraction. Linear regression was used to identify the degree of correlation and establish the relationship between lipid levels in muscle sample vs homogenized whole fish (n=19; 8 post-smolt and 11 sub-adult salmon, Fig. S2). There was a strong linear relationship between the lipid levels in muscle samples and lipid levels in whole fishes and the estimated formula was:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhole fish lipid (%) = 1.404+0.380*Muscle lipid (%) (R\u003csup\u003e2\u0026nbsp;\u003c/sup\u003e= 0.97, p-value \u0026lt; 0.001, n = 19; Fig. S2, Tab. S2). Total lipids for salmon were calculated from standardized muscle sample in all following analyses. The FA profiles are similar in muscle tissue and whole fish sample, although a few significant differences are found for some FAs only present in low quantities (Tab. S3).\u003c/p\u003e\n\u003ch2\u003eLipid and FA analysis\u003c/h2\u003e\n\u003ch3\u003eExtraction of FA and lipids \u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe post-smolt were extremely lean, with lipid levels mostly found to be below 2 % (average value of 1.98 \u0026plusmn; 0.44 %) and the amount of TAG (storage lipid) in these fish contributed with less than 10 % of the total lipids (TL). The composition of lipid classes is highly dependent on the TL amount in the fish, exemplified by a lean post-smolt sample with 1.3 % TL contains 84 % PL, 7 % TAG and 9 % free FA/cholesterol, while a fat sub-adult salmon sample with 37 % TL contains 10 % PL, 88 % TAG and 2 % free Fas/Cholesterol (Tab. S4). We have analyzed the FA composition in both TL and TAG for salmon, while the only the FA composition in TAG is available for herring and mackerel. Hence, only data from TAG is used when comparing FA composition among fish species. The TAG analysis was done by lipid extracted using hexane:isopropanol (3.2) according to the Hara method (Hara and Radin 1978) and HPLC separation methods described in Madhun et al. (2017). The zooplankton and herring larvae samples were extracted using a modified Folch extraction (Meier et al. 2006). The amount and composition of Fas in TL and TAG were determined by direct methanolysis and gas chromatography following Meier et al. (2006). The TL and TAG fractions were methylated, the resulting FA methyl esters (FAME) were extracted with hexane before being analysed on a HP-7890A gas chromatograph (Agilent, USA) with a flame ionization detector (GC-FID), as previously described in Meier et al. (2006). 67 Fas that contribute with more than 0.1 % of the total Fas were included in our data analysis. There is no data of TAG for 11 of the 294 sampled post-smolts as the samples were compromised. Likewise, no data of TL were generated for 11 sub-adult salmon due to a mistake in the lab. Data are presented as FA profiles (% of total FAs). Four sub-adult salmon were identified as escaped farmed fish and excluded from the initial dataset (see supplement section S3 for more information). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs several of the zooplankton (copepods, amphipods and euphausiids) samples contain large amounts of wax esters, the FAME and the fatty alcohols (FAOH) were separated on solid phase column and analyzed individually on GC-FID to avoid coelution, as described in Haug et al. (2021). The FAOHs or wax esters are not found in the lipids of the fish as the FAOHs oxidize very rapidly into the corresponding FAs during prey digestion. The FAOHs, therefore, also contribute to the predator\u0026rsquo;s FA composition, and when looking at FATM, both the FA and the FAOH from the prey should be considered (Budge and Iverson 2003). In the present work we have therefore added the sum of the quantitative amount of the different corresponding FAs and FAOHs (especially 22:1n-11 FA + 22:1n-11 FAOH in copepods and amphipods\u003cem\u003e\u0026nbsp;\u003c/em\u003eand 14:0, 16:0 and 16:1n-7 FAOH in\u003cem\u003e\u0026nbsp;Thysanoessa sp.\u003c/em\u003e) before normalizing to 100 %. The FA and FAOH compositions of the samples are given in the supplementary information, Tables S6-S19.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eFA biomarkers\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe copepod \u003cem\u003eCalanus finmarchicus\u003c/em\u003e is an important prey for pelagic fish (Dalpadado et al. 2000; Bachiller et al. 2016; Utne et al. 2021a) and carnivorous zooplankton such as \u003cem\u003eThemisto\u003c/em\u003e spp. and Euphausiids in the Norwegian Sea ecosystem. Furthermore, several studies have identified the importance of age-0 fish (e.g \u003cem\u003eClupea harengus, Sprattus sprattus\u003c/em\u003e, Gadus spp, Ammodytes spp, Sebastes spp) for post-smolts during their first summer at sea (e.g. Salminen et al. 2001; Utne et al. 2021a; Utne et al. 2022). Hence, we explored further the importance of \u003cem\u003eCalanus\u003c/em\u003e spp. and age-0 fish FAs in extracted lipids from salmon, herring and mackerel. For \u003cem\u003eCalanus\u003c/em\u003e spp. the following FAs were aggregated in a FATM biomarker; \u0026Sigma; 20:1n-9, 22:1n-11, 22:1n-9, while 22:6n-3 was used as a biomarker for age-0 fish ( Klungs\u0026oslash;yr et al. 1989; Folkvord et al. 1996).\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eStatistical analysis\u0026nbsp;\u003c/h2\u003e\n\u003ch3\u003eLipid levels\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eA Mann-Whitney U-test was used to test for difference in lipid levels between post-smolts and sub-adult salmon.\u003c/p\u003e\n\u003ch3\u003ePrincipal coordinate analysis\u003c/h3\u003e\n\u003cp\u003eTo estimate how FA profiles varied among species and life-stages, principal coordinate analysis (PCoA) based on Bray\u0026ndash;Curtis dissimilarities was applied. The first PCoA axis (PC1) represents the main axis of variation in the FA composition (Pond and Ward 2011). PCoA was applied to test for any difference in FA composition for 1) post-smolts and sub-adult salmon, 2) all pelagic fish, 3) salmon TAG and TL (separately for post-smolts and sub-adult salmon).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePCoA was carried out with the R-package FactoMineR (L\u0026ecirc; et al. 2008) and the results visualized by using the R-package factoextra (Kassambara and Mundt 2020). The FA values were log-transformed, mean centered, and standardized by dividing by standard deviation before performing the PCoA. This transformation method levels out the quantitative differences among FAs and ensures that variation in minor FAs also impacts the results.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDifferences in dispersion of FA profiles among post-smolts and sub-adult salmon, and among post-smolts from different geographic regions and years was tested using a multivariate extension of Levene\u0026rsquo;s test (Anderson 2006) for homogeneity of variance. A seasonal effect was not tested for post-smolts as 99% of the individuals were sampled within a month in July-early August. Significant differences in FA profiles between species or sub-groups of a species were estimated with analysis of similarity (ANOSIM) if the dispersion was different between groups and permutational multivariate analysis of variance (PERMANOVA) otherwise (Oksanen et al., 2024). The FAs contributing most to the difference among groups were identified using similarity percentages (SIMPER, Oksanen et al., 2024). To test if the FA composition for post-smolts changed geographically or over time, a PCoA was carried out with individuals grouped according to sampling year. Thereafter a PCoA was carried out with the individuals grouped according to geographic position in the Norwegian Sea (southern = south of 67˚30\u0026rsquo;N, northwestern = north of 67˚30\u0026rsquo;N and west of 5 ˚E and northeastern = north of 67˚30\u0026rsquo;N and east of 5 ˚E, Fig. 1A).\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eGeneralized additive mixed-effect models\u003c/h3\u003e\n\u003cp\u003eGeneralized additive mixed-effect models (GAMMs) were applied to test if the observed variation in total lipids in post-smolts and sub-adult salmon were linked to individual size or geographic position. GAMMs were also applied to test if the first principal component (PC1) of the PCoA for FA composition in post-smolt, and the FATM biomarkers for \u003cem\u003eCalanus\u003c/em\u003e spp. and age-0 fish in post-smolts, were related to their geographic position or individual size. The full set of tested models and specifications of model parametrization are given in Table S5. The GAMMs were run separately for post-smolts and sub-adult salmon. The data were inspected for outliers, normality, collinearity, and independence prior to modelling. Some observations were excluded to avoid bias effects of large outliers (see Supplement Tab. S5). A preliminary data inspection prior to the modelling revealed some highly correlated variables (r\u0026gt;0.6), and in these cases one variable was selected to represent a trait in the modelling. This included post-smolt length and body weight (cor = 0.93, p\u0026lt;0.001), and sub-adult salmon body weight and condition factor (cor = 0.64, p\u0026lt;0.001), while sub-adult length and condition factor were moderately correlated (cor = 0.42, p\u0026lt;0.001) but included as covariates in the same model. Post-smolt condition factor was not strongly correlated to length (cor = 0.16, p = 0.009) nor weight (cor = -0.17, p = 0.004). The parsimony principle was used to select the model with the lowest Akaike Information Criterion (AIC) (Burnham 2002). GAMMs were fitted using the R statistical programming v.4.2.2 environment and the package \u003cem\u003emgcv\u003c/em\u003e v. 1.8-41(Wood 2011). All model assumptions were visually assessed using Q\u0026ndash;Q plots and residual variation vs. fitted values and leverage. To avoid overfitting the smooth functions, the maximum number of basis dimensions (\u0026ldquo;knots\u0026rdquo;) was restricted to 5 for univariate smoothers and 20 for the two-dimensional smoother of geographical position. Results were visualized by using the R-packages ggplot2 (Wickham, 2016), patchwork (Pedersen 2024) and itsadug (van Rij et al. 2022).\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eK-mean clustering\u003c/h2\u003e\n\u003cp\u003eCluster analysis was done using partitioning methods (k-means). The optimum number of clusters for the k-mean clustering was determined to be four using the R-package NbClust (vers. 3.0.1) (Charrad et al. 2014). Data was visualised using the R-package ComplexHeatmap (ver. 2.6.2) (Gu et al. 2016).\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eLipids in post-smolts and sub-adult salmon\u003c/h2\u003e\n\u003cp\u003ePost-smolts in the Norwegian Sea had low levels of total lipids with an average value of 1.98 \u0026plusmn; 0.45% (min=1.44 %, max=5.16 %), which was significantly lower (Mann-Whitney U test, p\u0026lt;0.001) than the average total lipid levels of 6.88 \u0026plusmn; 4.06 % (min=1.69%, max=16.54%) for sub-adult salmon (Fig. 2a). The lipid levels increased with the condition factor for both post-smolts and sub-adult salmon, although the relationship was not linear (Fig. 2b, c). Furthermore, the lipid levels in post-smolts increased with body weight (Fig. 2d). The dependance of lipid levels on body weight for sub-adult salmon was not tested, as body weight and condition factor were highly correlated. Neither geographical position nor sampling month explained variation in lipid levels for sub-adult salmon, but including geographic position slightly improved the model for post-smolt lipid levels. \u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eFA composition in post-smolts and sub-adult salmon\u003c/h2\u003e\n\u003cp\u003eThe variation in FA composition in TAG among individual post-smolts was larger than among individual sub-adult salmon (PERMDIST, F\u003csub\u003e1,366\u003c/sub\u003e= 50.13, p\u0026lt;0.001, Fig. 3a). The following FAs were the five most abundant for both salmon life-stages; 18:1n-9, 22:6n-3, 16:0, 22:1n-11, 20:1n-9. Among the 20 most abundant FAs (Fig. 4a, Tab. S6-S9 show all 67 FAs), 22:6n-3 was more abundant in post-smolts (SIMPER, p-value \u0026lt; 0.001) while 16:1n-7, 18:1n-7 and 22:1n-9 were more abundant in sub-adult salmon (SIMPER, p\u0026lt;0.05) (Fig. 3b). There was also a significant difference for several of the less common FAs.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA PCoA was carried out for post-smolts only to further explore the variation in FA composition in TAG. Sampling year had a significant effect on the FA composition (ANOSIM, R= 0.1997, p=0.001) and the associated variance (PERMDIST, F\u003csub\u003e8,273\u003c/sub\u003e= 14.167, p\u0026lt;0.001). A change in the FA composition was noticed in the years after 2016 (Fig. 3c). To identify the FAs explaining most of the temporal change in FA composition, a SIMPER was run with years aggregated into two periods, 2012-2016 and 2017-2020. The relative abundance of 61 of the 67 FAs changed from the first to the second period (SIMPER p\u0026lt;0.05), and the most noticeable change was the increased abundance of FAs associated with calanoid copepods such as 22:1n-11 and 20:1n-9 and decreased abundance of 16:0 and 18:0 in the last period. The results from the K-mean cluster analysis (results presented further below) further support a change in post-smolt FA composition after year 2016. \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePost-smolts sampled in different geographic regions (south, northeast and northwest) of the Norwegian Sea had significantly different FA compositions in TAG (ANOSIM, R=0.1768, p=0.001) and associated variance (PERMDIST, F\u003csub\u003e2,279\u003c/sub\u003e= 5.376, p=0.005) (Fig. 3d). The relative abundance of most FAs was significantly different for post-smolts sampled in the northern (both northeastern and northwestern regions) and those sampled in the southern Norwegian Sea (SIMPER, p\u0026lt;0.05). The only FAs that had a significantly different relative abundance in post-smolts between the northeastern and northwestern Norwegian Sea was 21:5n-3 (SIMPER, p=0.018). The PC1 for post-smolt mainly represents the relative levels of FAs associated with \u003cem\u003eCalanus\u003c/em\u003e spp. (20:1n-9, 22:1n-9, 22:1n-11) and FAs associated with age-0 fish (22:6n-3) (Fig. 3e) and is related to post-smolt sampling location (Fig. 5a). The relative level of \u003cem\u003eCalanus\u003c/em\u003e biomarkers is relatively more abundant in the regions north of 67˚N (Fig. 5b) while the age-0 fish biomarker was higher in post-smolt sampled south of 67˚N in the Norwegian Sea and further northeast close to the Norwegian coast (Fig. 5c).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe PC1 decreases with increasing lipid levels (Fig. 5d). Hence, post-smolts with relatively high levels of FAs associated with \u003cem\u003eCalanus\u003c/em\u003e (20:1n-9, 22:1n-9, 22:1n-11) had higher lipid levels than post-smolts with relatively high levels of FAs associated with age-0 fish (22:6n-3). The PC1 is not associated with post-smolt condition factor, as including condition factor as covariate did not improve the model fit (Tab. S5).\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eTAG and TL for post-smolts and sub-adult salmon\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe FA composition was significantly different in TAG and TL for post-smolts (muscle samples) (Fig. S4a, PERMANOVA, F\u003csub\u003e1,556\u003c/sub\u003e= 30.7, p\u0026lt;0.001) but with a similar dispersion (PERMDIST, F\u003csub\u003e1,556\u003c/sub\u003e= 0.0096, p=0.9219). For sub-adult salmon the FA composition was not significantly different in TAG and TL (Fig. S4b, PERMANOVA, F\u003csub\u003e1,161\u003c/sub\u003e= 1.576, p=0.21) and the dispersion was similar (PERMDIST, F\u003csub\u003e1,161\u003c/sub\u003e= 0.655, p=0.45). Due to the significant differences in FA profiles for post-smolt TAG and TL, the different FA compositions in TAG and TL was further explored. The difference was mainly seen in the FAs 22:6n-3 and 20:5n-3, which were higher in TL than in TAG, and 16:1n-7, 18:1n-9, 20:1n-9, 20:1n-11 and 22:1n-9 which were higher in TAG than in TL (Fig. 4a, Tab. S6-S9).\u003c/p\u003e\n\u003ch2\u003eFAs in pelagic fish and its prey\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eThe FA composition of 20 most abundant FAs of salmon, herring, mackerel and prey organisms (calanoid copepods, amphipods, euphausiids and herring larvae) are given in Fig. 4 (Tab. S6-S19 show all the 67 FAs).\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eDifference in FA composition for salmon and herring/mackerel\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe first two principal components (PC1 and PC2) from a PCoA for TAG in all fish and zooplankton species showed that the FA compositions of post-smolts and sub-adult salmon differ from those of herring and mackerel, with the two latter species having a similar FA composition (Fig. 6a). PC1 mainly represents the variation in saturated FA, C18 MUFAs and long-chain PUFAs, all of which are associated with feeding on carnivorous zooplankton and age-0 fish (positive PC1 score), and C16-PUFA and other FAs associated with feeding on herbivores zooplankton (negative PC1 score) (Fig. 6b, c). Of the 67 tested FAs, the number of FAs with significantly different relative levels were 23 for post-smolts and herring, 14 for post-smolts and mackerel, 18 for sub-adult salmon and herring and 13 for sub-adult salmon and mackerel. Many of the FAs with significantly different relative levels in salmon compared to herring and mackerel are only present at relatively low levels. The FAs in post-smolts or sub-adult salmon, and herring or mackerel, that had significantly different relative levels were FAs associated with feeding on carnivorous zooplankton and age-0 fish (higher in salmon) or FAs associated with feeding on herbivores zooplankton (lower in salmon). For details regarding statistical comparison of all specific FAs among species the reader is invited to see Tables S20-S26. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eK-mean cluster analysis suggested four clusters in the dataset (Fig. 7). Cluster 1 contains a mix of mackerel and herring (50 % and 49 %), while Cluster 2 contains a mix of sub-adult salmon (41%) and post-smolts (59%). The last two clusters (Cluster 3 and 4) are dominated by post-smolts (98% and 95 %). These two clusters (3 and 4) have higher relative levels of long-chain PUFAs (\u0026ge;C20), while the clusters containing herring, mackerel and sub-adult salmon have higher relative levels of long-chain MUFAs (\u0026ge;C20) (Fig. 7). An important difference between Cluster 1 and 2 are the higher levels of short-chain PUFAs (C16 and C18) and saturated FAs (14:0) in Cluster 1 (containing mackerel and herring) than in Cluster 2 (sub-adult salmon and post-smolts). The separation of post-smolts into three different clusters (1, 2, 3) is partly explained by difference in post-smolt size, lipid levels and sampling year. These results mimic the output from the PCoA analyses (Fig. 3 and 4) and are therefore not presented further in detail (see supp. information section S7 for more information). \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe variable (FAs) along the y-axes is colored to be relative higher (yellow) or lower (blue) in the different clusters.\u003c/p\u003e\n\u003ch3\u003eDifference in FA composition for salmon and selected prey organisms\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe relative levels of most FAs in post-smolts and sub-adult salmon TAG were significantly different from the levels in zooplankton and herring larvae (SIMPER, p\u0026lt;0.05, Fig. 4, see example Tab. S25-S26). The PC1 and PC2 from the PCoAs, including all fish and prey species, show how the FA composition in post-smolts and sub-adult salmon relates to the selected prey organisms (Fig. 6). Along the PC1, post-smolts are clustered in the same space as herring larvae, amphipods and euphausiids (Fig. 6). Compared to the other prey organisms, herring larvae had high levels (\u0026asymp;30%) of 22:6n-3 (Fig. 4b) and were clustered with positive values on the PC1-axis in the two-dimensional space (Fig. 6b). Herring larvae had also high relative levels of 16:0 and 22:5n-3, and so do amphipods and euphausiids (Fig. 4b, 6b). \u003cem\u003eC. finmarchicus\u003c/em\u003e and \u003cem\u003eC. hyperboreus\u003c/em\u003e have low relative levels of 16:0, 22:6n-3 and 22:5n-3 and high relative levels of 18:4n-3, 20:1n-9 and 22:1n-11 (Fig. 4b), and these two species are clustered in the two-dimensional space with negative values on the PC1-axis (Fig. 6b). Along the PC1-axis there was partly an overlap between sub-adult salmon and the two \u003cem\u003eCalanus\u003c/em\u003e species, but not between post-smolts and \u003cem\u003eCalanus\u003c/em\u003e spp. Herring and mackerel had high relative levels of 18:4n-3, 20:1n-9 and 22:1n-11 and overlapped with \u003cem\u003eCalanus\u003c/em\u003e along the PC1-axis (Fig. 6a, b). Post-smolts and sub-adult salmon differed from \u003cem\u003eCalanus\u003c/em\u003e spp. by having significantly lower levels of 14:0 and short-chains PUFAs C16 and C18 (SIMPER, p\u0026lt;0.05). Interestingly, FAs known as products of endogenous metabolism, 18:1n-11 and 22:5n-3, were relatively higher in post-smolts and sub-adult salmon than in all prey groups (Fig. 4).\u003c/p\u003e\n\u003cp\u003eFor post-smolts, the relative proportion of 22:6n-3 and 22:5n-3 was higher in TL than in TAG, while the relative proportion of 18:4n-3, 20:1n-9 and 22:1n-11 were higher in TAG than in TL (Fig. 4). Hence, post-smolts have a higher proportion of FAs associated with age-0 fish in the membrane lipids than in TAG, while TAG has a higher proportion of FAs associated with \u003cem\u003eCalanus\u003c/em\u003e spp. than the membrane lipids (see also Fig S5). \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study has identified huge variation among post-smolt FA composition, reflecting their diverse diet and opportunistic feeding behavior. A large proportion of the post-smolts sampled early in the summer had a high proportion of FA associated with age-0 fish, and these FAs are mainly allocated for somatic growth. Later in the summer, when they have reached the northern Norwegian Sea, post-smolts have a higher proportion of FAs associated with calanoid copepods. Post-smolts have very low lipids levels compared to sub-adult salmon or other pelagic fish feeding in the same geographic region of the northeast Atlantic. Furthermore, the FA composition of post-smolts, which are hypothesized to compete for prey with other pelagic fish species, is partly different from the FA composition of herring and mackerel caught within the same geographic area, suggesting important diet differences between the species.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eTemporal and geographic variation in FAs for salmon \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eThe FA composition varied substantially among post-smolts. This is, most likely, due to geographic differences in prey abundance and the number of days post-smolts have spent at sea. Post-smolts sampled in the Norwegian Sea originate from rivers over a large geographic area (Gilbey et al. 2021), where individuals from several countries commence their marine feeding period in completely different regions of the Northeast Atlantic. Therefore, these individuals do not encounter the same prey species during their first period at sea. In any case, the large variation in FA profiles also reflects that salmon are opportunistic feeders, able to target a wide range of prey (Jacobsen and Hansen 2001; Haugland et al. 2006). The variation in FA among post-smolts was especially large for FATMs associated with calanoid copepods and with age-0 fish. The herring larvae analysed for FAs in this study had high levels of long-chain PUFAs such as 22:6n-3, which is a FA also abundant in age-0 fish of other species (Klungsøyr et al. 1989). The proportion of 22:6n-3 remained high in herring sampled from the yolk-sac phase and the following months during the spring and early summer (Fraser et al. 1989), but the relative abundance of long-chain PUFAs is much lower in adult herring and mackerel, as shown in the present study. The Euphausiids and amphipods analyzed in this study also had high levels of 22:6n-3 (~15%), although substantially lower than the observed levels in herring larvae and post-smolts. Hence, we conclude that the high level of 22:6n-3 in post-smolts most probably originates from feeding on age-0 fish. Post-smolts sampled in the northern Norwegian Sea had higher relative levels of FAs associated with calanoid copepods, while fish from the southern Norwegian Sea had higher relative levels of FAs associated with age-0 fish. A north-south gradient for the post-smolt diet has previously been documented from stomach samples, with the biomass of age-0 fish decreasing and the biomass of amphipods in sampled post-smolt stomachs increasing when post-smolts migrate northwards (Utne et al. 2022). \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe study presented here found a positive correlation between post-smolt lipid levels and FAs associated with calanoid copepods, and high levels of these FAs as well as high lipid levels in sub-adult salmon. Although post-smolts sometimes feed directly on calanoid copepods, it is apparent from stomach content analyses that this prey group is not an important part of the post-smolt diet (Rikardsen and Dempson 2010; Sheehan et al. 2012). We suggest that FAs associated with calanoid copepods in salmon mostly originate from feeding on predators of calanoid copepods, such as amphipods and euphausiids, cephalopods and small fish.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe post-smolt FA composition changed in the period 2012-2020, with the main shift occurring around 2016-2017. Vollset et al. (2022) documented how changes in oceanographic conditions and abundance of plankton in the Norwegian Sea in the period 2004/2005 occurred simultaneously as a drop in marine growth during the first year at sea for Norwegian salmon. From 2017 and onwards the oceanographic conditions shifted back towards the pre-2004 conditions, and the zooplankton abundance increased (Skagseth et al. 2022). The change in post-smolt FA composition occurring around 2017 presented in this study supports previous findings of a change in diet for post-smolts in the Norwegian Sea occurring during the same period. The abundance of zooplankton in post-smolt stomachs increased (Utne et al. 2022) and the FA composition presented in this study had a higher relative abundance of FAs associated with calanoid copepods in the period 2017-2020 compared to 2012-2016. The post-smolts were on average sampled ~1 week later and ~1˚ further north in 2017-2020 than in 2012-2016, but with variation among years (Tab. S1), in contrast to the consistent shift in FA composition for post-smolts when comparing the years 2012-2016 and 2017-2020.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe FA composition of sub-adult salmon was not significantly different from those of post-smolts, although the FA composition among sub-adult salmon had lower variation than among post-smolts. The most important FAs for sub-adult salmon sampled in the Norwegian Sea was the same as for salmon returning to Canadian rivers after feeding in the Northwest Atlantic Ocean (Bøe et al. 2019). The single FA that differed most between salmon and post-smolts was higher proportion of 22:6n-3 in post-smolts. The PCoA showed that sub-adult salmon FA composition overlap with both euphausiids and amphipods on the PC1, and with \u003cem\u003eC. hyperboreus\u003c/em\u003e on the PC2. Unlike for post-smolts, the FA composition of sub-adult salmon does not overlap with \u003cem\u003eC. finmarchicus\u003c/em\u003e on the first two PCs. While amphipods and \u003cem\u003eC. hyperboreus\u0026nbsp;\u003c/em\u003eprimarilyinhabits cold Arctic water masses in the northern Norwegian Sea and further west towards Greenland (Skjoldal 2004), \u003cem\u003eC. finmarchicus\u003c/em\u003e primarily inhabits Atlantic water masses in the Norwegian Sea (Strand et al. 2020). Hence, the FA composition of sub-adult salmon is more similar to calanoid copepods associated with cold Arctic water than the warm Atlantic water. We therefore suggest, based on the results presented here, that preferred feeding areas for sub-adult salmon are western parts of the Norwegian Sea, the Arctic front and areas further west towards Greenland.\u003c/p\u003e\n\u003ch2\u003eThe link between Selective use of\u0026nbsp;FAs, lipid levels and growth of Atlantic salmon\u003c/h2\u003e\n\u003cp\u003eThe results showed that most post-smolts have low levels of total body lipids (≈2 %). Furthermore, the results presented here showed that post-smolt lipids were dominated by membrane lipids (PL), while sub-adult salmon have most of their stored FAs as storage lipids (TAG), as previously demonstrated from feeding experiments for salmon in tanks (Jobling and Johansen 2003). During smoltification, Atlantic salmon mobilize energy reserves and deplete storage lipids, which are used as catabolic energy for the adaptation to seawater (Sheridan 1989). Hence, the protein content increases in the period around smoltification (Stefansson et al. 2003). For post-smolts, but not for sub-adult salmon, the FA composition in storage lipids (TAG) was significantly different to the composition in the total lipids (TL). Post-smolts have a much higher proportion of FAs associated with age-0 fish in their membrane lipids than in their TAG, suggesting that they quickly utilize long-chain PUFAs for somatic growth rather than for lipid storage. This agrees with salmon using some FAs directly for metabolic costs while other FAs are used for membrane lipids (Litz et al. 2017b), in which case the FA composition of prey is partly modified before allocated to membrane lipids, and that salmon have some ability to perform endogenous synthesis of long-chain PUFAs (Mock et al. 2019). In addition, salmon encountering poor feeding conditions may deplete their FAs originating from calanoid copepods before they deplete FAs originating from age-0 fish (Litz et al. 2017b). It can be beneficial for post-smolts to feed on age-0 fish during the first months at sea, as age-0 fish may be low on lipids but contain a nutrient composition with high levels of long-chain PUFAs in the right proportion for post-smolts to utilize directly for body growth. A high proportion of FAs associated with age-0 fish have also been documented in coho\u0026nbsp;(\u003cem\u003eOncorhynchus kisutch\u003c/em\u003e) and chinook (\u003cem\u003eOncorhynchus tshawytscha\u003c/em\u003e)\u0026nbsp;post-smolts sampled in the Pacific Ocean within their first months at sea (Daly et al. 2010). Juvenile Chinook salmon is known to\u0026nbsp;shift their energy allocation from somatic growth to lipid storage towards the end of the first ocean summer at sea\u0026nbsp;(MacFarlane 2010; Litz et al. 2017a). Among the sampled post-smolt presented here, the lipid levels were highest when the FA composition had a high proportion of FAs associated with calanoid copepods. These FAs were most abundant in post-smolts sampled in the northern Norwegian Sea. However, there is a correlation between geographic area, time and size of post-smolts sampled in the Norwegian Sea, as they grow while migrating northwards during the summer (Gilbey et al. 2021, Utne et al. 2021a). Farmed Atlantic post-smolts also has low lipid accumulation in the first period after transfer to seawater, even when the fish are fed with highly lipid-rich feed, and lipid accumulation in the muscle only starts at the end of summer (August/September) (Alne et al. 2011; Dessen et al. 2017). This could suggest that accumulation of lipids is related primarily to season and/or body size rather than prey availability.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eInteractions between salmon and other pelagic fish\u003c/h2\u003e\n\u003cp\u003eThe observed lipid levels in post-smolts (≈2 %) during the summer are lower than the lipid levels in mackerel and herring (at the lowest around 8 % during winter and early spring) (Slotte 1999; Jansen et al. 2021), but similar to the observed lipid levels in post-smolt of Coho and Chinook salmon in the Pacific Ocean (Daly et al. 2010). Furthermore, the lipid levels in mackerel and herring increase quickly once they commence feeding in late-spring, and their lipid levels peak (≈25 %) around August-September. The low lipid levels in post-smolts caught in July and August suggest that Atlantic salmon have a different energy allocation strategy than mature herring and mackerel. Prioritizing somatic growth over lipid accumulation is reasonable for salmon, given their documented life-history strategy that emphasizes rapid growth during the first year at sea (Klemetsen et al. 2003), as opposed to, for instance, the life-history strategy of herring and mackerel that exhibit relatively slow growth, longevity and multiple years of spawning (Trenkel et al. 2014). Alternatively, low lipid levels in post-smolts may reflect poor feeding conditions the years of post-smolts sampling in this study. The condition factor of the post-smolts sampled in the Norwegian Sea during the years 2012-2019 was significantly lower than during the years 1995-2003 (Utne et al. 2021a). A similar temporal reduction in body-length of salmon returning to river after one year at sea has also been reported for Norwegian, British and French salmon (Vollset et al. 2022, Trehin et al. 2023). \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eInterspecific competition with mackerel and herring can potentially have a negative effect on post-smolt growth and survival (Potter and Corzier 2000), and the diet of the three species partly overlaps when feeding in the Norwegian Sea (Utne et al. 2021b). As presented here, FA profiles varies among post-smolt but in average only partly overlap with herring and mackerel, while the FA profiles of the two latter species only have relatively small differences. We interpret this as an indication for the post-smolt diet niche at least partly deviating from the diet niche of mackerel and herring. However, anadromous salmonids have faster lipid synthesis than most other marine fish and are able to modify the dietary FAs allocated to storage lipids (Tocher 2010). The relatively higher proportion of minor C18 PUFA in the n-7 and n-4 family in the post smolt/salmon compared with herring and mackerel are likely to be elongation of the corresponding C16 PUFA in the n-7 and n-4 family, which were more abundant in herring and mackerel than in salmon. For example, the relative high proportion of 20:4n-3 and 22:5n-3 found in salmon may originate from elongation of 18:4n-3 and 20:5n-3, respectively. A high degree of modification of FAs stored as muscle tissue in salmon, but not in herring and mackerel, would overestimate the dietary differences between salmon and other pelagic fish. However, the FAs contributing to the largest difference between post-smolts and herring/mackerel, namely 22:6n-3 is not modified and thus originate directly from the diet. Any differences in FA modifications among these species is unlikely to fully explain the relatively large niche differences seen between post-smolts and the two other pelagic species. Herring and mackerel were only sampled in three years at less than 10 geographic locations. Sampling herring and mackerel in a larger geographic area and for a longer time period, could potentially reveal areas or periods where their diet niche is more similar to that of post-smolt than shown with the data available for this study. \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eFuture perspectives\u003c/h2\u003e\n\u003cp\u003eThese results presented in this study have provided novel data on FA composition of salmon and other pelagic fish sampled at the marine feeding areas. As salmon are not observed or regularly caught during most of their marine phase, there is still a general knowledge gap for salmon at sea during the autumn, winter and early spring. Hence, future sampling during these seasons could improve our understanding of the link between diet and growth throughout the year, which could be valuable knowledge to understand salmon life-history strategies and factors affecting mortality during the marine phase. \u0026nbsp;\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgments\u003c/p\u003e\n\u003cp\u003eThe work was funded by the Research Council of Norway as part of project 280308 SeaSalar and 243895 EcoNorSe. The authors would also like to thank Therese Smith-Jahnsen Aase and Arve Fossen for technical assistance with analyzing samples in the lab. We thank the many people who assisted with gathering the data presented here, including the skippers and crew onboard the vessels collecting samples.\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eData availability\u003c/p\u003e\n\u003cp\u003eData generated or analyzed during this study are available as open-access and can be downloaded from the following link: https://doi.org/10.6084/m9.figshare.28505816.v1\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eThe authors have no competing interests to declare.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAlne H, Oehme M, Thomassen M, Terjesen B, R\u0026oslash;rvik KA (2011) Reduced growth, condition factor and body energy levels in Atlantic salmon Salmo salar L. during their first spring in the sea. Aquac Res 42(2):248-259. https://doi.org/10.1111/j.1365-2109.2010.02618.x \u003c/li\u003e\n\u003cli\u003eAnderson MJ (2006) Distance-based tests for homogeneity of multivariate dispersions. Biometrics 62(1):245-253. doi:10.1111/j.1541-0420.2005.00440.x.\u003c/li\u003e\n\u003cli\u003eBachiller E, Skaret G, N\u0026oslash;ttestad L, Slotte A (2016) Feeding Ecology of Northeast Atlantic Mackerel, Norwegian Spring-Spawning Herring and Blue Whiting in the Norwegian Sea. Plos One 11(2): 25. doi:10.1371/journal.pone.0149238.\u003c/li\u003e\n\u003cli\u003eBeaugrand G, Reid PC (2003) Long-term changes in phytoplankton, zooplankton and salmon related to climate. Glob Change Biol 9(6):801-817. doi:https://doi.org/10.1046/j.1365-2486.2003.00632.x.\u003c/li\u003e\n\u003cli\u003eBeaugrand G, Reid PC (2012) Relationships between North Atlantic salmon, plankton, and hydroclimatic change in the Northeast Atlantic. ICES J Mar Sci 69(9):1549-1562. https://doi.org/10.1093/icesjms/fss153\u003c/li\u003e\n\u003cli\u003eB\u0026oslash;e K, Power M, Robertson MJ, Morris CJ, Dempson JB, Pennell CJ, Fleming IA (2019) The influence of temperature and life stage in shaping migratory patterns during the early marine phase of two Newfoundland (Canada) Atlantic salmon (\u003cem\u003eSalmo salar\u003c/em\u003e) populations. Can J Fish Aquat Sci 76(12):2364-2376. https://doi.org/10.1139/cjfas-2018-0320\u003c/li\u003e\n\u003cli\u003eBrix O, Gruner R, Ronnestad I, Gemballa S (2009) Whether depositing fat or losing weight, fish maintain a balance. Proc R Soc Biol Sci Ser B 276(1674):3777-3782. https://doi.org/10.1098/rspb.2009.1079\u003c/li\u003e\n\u003cli\u003eBudge SM, Iverson SJ (2003) Quantitative analysis of fatty acid precursors in marine samples: direct conversion of wax ester alcohols and dimethylacetals to FAMEs. J Lipid Res 44(9):1802-1807. https://doi.org/10.1194/jlr.D300009-JLR200\u003c/li\u003e\n\u003cli\u003eBudge SM, Penney SN, Lall SP (2011) Response of tissue lipids to diet variation in Atlantic salmon (\u003cem\u003eSalmo salar\u003c/em\u003e): Implications for estimating diets with fatty acid analysis. J Exp Mar Biol Ecol 409(1-2):267-274. doi:10.1016/j.jembe.2011.09.002.\u003c/li\u003e\n\u003cli\u003eBudge SM, Penney SN, Lall SP (2012) Estimating diets of Atlantic salmon (Salmo salar) using fatty acid signature analyses; validation with controlled feeding studies. Can J Fish Aquat Sci 69(6):1033-1046. doi:10.1139/f2012-039.\u003c/li\u003e\n\u003cli\u003eBurnham, KP, Anderson DR (2002) Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer-Verlag, New York.\u003c/li\u003e\n\u003cli\u003eCharrad M, Ghazzali N, Boiteau V, Niknafs A (2014) NbClust: An R Package for Determining the Relevant Number of Clusters in a Data Set. J Stat Softw 61(6):1-36. doi:10.18637/jss.v061.i06.\u003c/li\u003e\n\u003cli\u003eDalpadado P, Ellertsen B, Melle W, Dommasnes A (2000) Food and feeding conditions of Norwegian spring-spawning herring (\u003cem\u003eClupea harengus\u003c/em\u003e) through its feeding migrations. ICES J Mar Sci 57(4):843-857. https://doi.org/10.1006/jmsc.2000.0573.\u003c/li\u003e\n\u003cli\u003eDalsgaard J, St John M, Kattner G, Muller-Navarra D, Hagen W (2003) Fatty acid trophic markers in the pelagic marine environment. Adv Mar Biol 46:225-340. DOI:10.1016/S0065-2881(03)46005-7\u003c/li\u003e\n\u003cli\u003eDaly EA, Benkwitt CE, Brodeur RD, Litz MNC, Copeman LA (2010) Fatty acid profiles of juvenile salmon indicate prey selection strategies in coastal marine\u003c/li\u003e\n\u003cli\u003ewaters. Mar Biol 157:1975\u0026ndash;1987. https://doi.org/10.1007/s00227-010-1466-9.\u003c/li\u003e\n\u003cli\u003eDessen JE, Weihe RN, Hatlen B, Thomassen MS, R\u0026oslash;rvik KA (2017) Different growth performance, lipid deposition, and nutrient utilization in in-season (S1) Atlantic salmon post-smolt fed isoenergetic diets differing in protein-to-lipid ratio. Aquaculture 473:345-354. doi:10.1016/j.aquaculture.2017.02.006.\u003c/li\u003e\n\u003cli\u003eFolkvord A, Haaland IM, Grahl-Nielsen O, Johannessen A (1996) Fatty acid composition of herring (\u003cem\u003eClupea harengus L\u003c/em\u003e) larvae off western Norway. SARSIA 80(4):299-305. https://doi.org/10.1080/00364827.1996.10413604\u003c/li\u003e\n\u003cli\u003eFraser AJ, Sargent JR, Gamble JC, Seaton DD (1989) Formation and transfer of fatty-acids in an enclosed marine food-chain comprising phytoplankton, zooplankton and herring (\u003cem\u003eclupea-harengus l\u003c/em\u003e) larvae. Mar Chem 27(1-2):1-18. https://doi.org/10.1016/0304-4203(89)90024-8\u003c/li\u003e\n\u003cli\u003eGilbey J, Utne KR, Wennevik V, Beck AC, Kausrud K, Hindar K, de Leaniz CG, Cherbonnel C, Coughlan J, Cross TF, Dillane E, Ensing D, Garcia-Vazquez E, Hole LR, Holm M, Holst JC, Jacobsen JA, Jensen AJ, Karlsson S, Maoileidigh NO, Mork KA, Nielsen EE, N\u0026oslash;ttestad L, Primmer CR, Prodohl P, Prusov S, Stevens JR, Thomas K, Whelan K, McGinnity P, and Verspoor E (2021) The early marine distribution of Atlantic salmon in the North-east Atlantic: A genetically informed stock-specific synthesis. Fish Fish 22(6):1274-1306. doi:10.1111/faf.12587.\u003c/li\u003e\n\u003cli\u003eGu Z, Eils R, Schlesner M (2016) Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32(18):2847-2849. doi:10.1093/bioinformatics/btw313.\u003c/li\u003e\n\u003cli\u003eHara A., Radin NS (1978) Lipid extraction of tissues with a low-toxicity solvent. Anal Biochem 90:420-426.\u003c/li\u003e\n\u003cli\u003eHaug T, Biuw M, Gj\u0026oslash;s\u0026aelig;ter H, Knutsen T, Lindstrom U, MacKenzie KM, Meier S, Nilssen KT (2021) Harp seal body condition and trophic interactions with prey in Norwegian high Arctic waters in early autumn. Prog Oceanogr 191. doi:10.1016/j.pocean.2020.102498.\u003c/li\u003e\n\u003cli\u003eHaugland M, Holst JC, Holm M, Hansen LP (2006) Feeding of Atlantic salmon (\u003cem\u003eSalmo salar L.)\u003c/em\u003e post-smolts in the Northeast Atlantic. ICES J Mar Sci 63(8):1488-1500. doi:10.1016/j.icesjms.2006.06.004.\u003c/li\u003e\n\u003cli\u003eICES (2013) Report of the Workshop on Northeast Atlantic Mackerel monitoring and methodologies including science and industry involvement (WKNAMMM), 25\u0026ndash;28 February 2013, ICES Headquarters, Copenhagen and Hirtshals, Denmark. ICES CM 2013/SSGESST:18. 33 pp.\u003c/li\u003e\n\u003cli\u003eICES (2024) Working group on North Atlantic Salmon (WGNAS). ICES Scientific Reports. 6:36. 415 pp. https://doi.org/10.17895/ices.pub.25730247\u003c/li\u003e\n\u003cli\u003eIverson SJ, Field C, Bowen WD, Blanchard W (2004) Quantitative fatty acid signature analysis: A new method of estimating predator diets. Ecol Monogr 74(2):211-235.\u003c/li\u003e\n\u003cli\u003eJacobsen JA, Hansen LP (2000) Feeding habitats of Atlantic salmon at different life stages at sea. In The ocean life of Atlantic salmon \u0026mdash; environmental and biological factors influencing survival. Fishing News Books, Oxford.\u003c/li\u003e\n\u003cli\u003eJacobsen JA, Hansen LP (2001) Feeding habits of wild and escaped farmed Atlantic salmon, \u003cem\u003eSalmo salar L\u003c/em\u003e., in the Northeast Atlantic. ICES J Mar Sci 58(4):916-933. https://doi.org/10.1006/jmsc.2001.1084\u003c/li\u003e\n\u003cli\u003eJacobson P, G\u0026aring;rdmark A, \u0026Ouml;stergren J, Casini M, Huss M (2018) Size-dependent prey availability affects diet and performance of predatory fish at sea: a case study of Atlantic salmon. Ecosphere 9(1): e02081. doi:https://doi.org/10.1002/ecs2.2081.\u003c/li\u003e\n\u003cli\u003eJansen T, Slotte A, Schmidt TCD, Sparrevohn CR, Jacobsen JA, Kjesbu OS (2021) Bioenergetics of egg production in Northeast Atlantic mackerel changes the perception of fecundity type and annual trends in spawning stock biomass. Prog Oceanogr 198. doi:10.1016/j.pocean.2021.102658.\u003c/li\u003e\n\u003cli\u003eJobling M, Johansen SJS (2003) Fat distribution in Atlantic salmon \u003cem\u003eSalmo salar L\u003c/em\u003e. in relation to body size and feeding regime. Aquac Res 34(4):311-316. https://doi.org/10.1046/j.1365-2109.2003.00820.x\u003c/li\u003e\n\u003cli\u003eKassambara A, Mundt F (2020). Factoextra: Extract and Visualize the Results of Multivariate Data Analyses. R package version 1.0.7, https://CRAN.R project.org/package=factoextra \u003c/li\u003e\n\u003cli\u003eKlemetsen A, Amundsen PA, Dempson JB, Jonsson B, Jonsson N, O\u0026apos;Connell MF, Mortensen E (2003) Atlantic salmon Salmo salar L., brown trout Salmo trutta L. and Arctic charr \u003cem\u003eSalvelinus alpinus (L.):\u003c/em\u003e a review of aspects of their life histories. Ecol Freshw Fish 12(1):1-59. doi:https://doi.org/10.1034/j.1600-0633.2003.00010.x.\u003c/li\u003e\n\u003cli\u003eKlungs\u0026oslash;yr J, Tilseth S, Wilhelmsen S, Falk-Petersen S, Sargent J.R (1989) Fatty-acid composition as an indicator of food-intake in cod larvae \u003cem\u003egadus-morhua\u003c/em\u003e from Lofoten, northern Norway. Mar Biol 102(2):183-188. https://doi.org/10.1007/BF00428278\u003c/li\u003e\n\u003cli\u003eL\u0026ecirc; S, Josse J, Husson F (2008) FactoMineR: An R Package for Multivariate Analysis. Journal of Statistical Software 25(1):1 - 18. doi:10.18637/jss.v025.i01.\u003c/li\u003e\n\u003cli\u003eLitz MNC, Miller JA, Copeman LA, Hurst TP (2017a) Effects of dietary fatty acids on juvenile salmon growth, biochemistry, and aerobic performance: A laboratory rearing experiment. J Exp Mar Biol Ecol 494:20-31. https://doi.org/10.1016/j.jembe.2017.04.007.\u003c/li\u003e\n\u003cli\u003eLitz MNC, Miller JA, Copeman LA, Teel DJ, Weitkamp LA, Daly EA, Claiborne AM (2017b) Ontogenetic shifts in the diets of juvenile Chinook Salmon: new insight from stable isotopes and fatty acids. Environ Biol Fishes 100(4):337-360. DOI: 10.1007/s10641-016-0542-5\u003c/li\u003e\n\u003cli\u003eMacFarlane RB (2010) Energy dynamics and growth of Chinook salmon (\u003cem\u003eOncorhynchus tshawytscha) \u003c/em\u003efrom the Central Valley of California during the estuarine phase and first ocean year. Can J Fish Aquat Sci 67(10):1549-1565. doi:10.1139/F10-080.\u003c/li\u003e\n\u003cli\u003eMadhun AS, Wennevik V, Skilbrei OT, Karlsbakk E, Skaala O, Fiksdal IU, Meier S, Tang YK, Glover KA (2017) The ecological profile of Atlantic salmon escapees entering a river throughout an entire season: diverse in escape history and genetic background, but frequently virus-infected. ICES J Mar Sci 74(5):1371-1381. https://doi.org/10.1093/icesjms/fsw243\u003c/li\u003e\n\u003cli\u003eMeier S, Mjos SA, Joensen H, Grahl-Nielsen O (2006) Validation of a one-step extraction/methylation method for determination of fatty acids and cholesterol in marine tissues. J Chromatogr A 1104(1-2):291-298. doi:10.1016/j.chroma.2005.11.045.\u003c/li\u003e\n\u003cli\u003eMock TS, Francis DS, Jago MK, Glencross BD, Smullen RP, Turchini GM (2019) Endogenous biosynthesis of n-3 long-chain PUFA in Atlantic salmon. Br J Nutr 121(10):1108-1123. doi:10.1017/s0007114519000473.\u003c/li\u003e\n\u003cli\u003eOksanen J, Simpson GL, Blanchet FB, Kindt R, Legendre P, Minchin PR, O\u0026apos;Hara RB et al. 2024. Ordination methods, diversity analysis and other functions for community and vegetation ecologists. 10.32614/CRAN.package.vegan\u003c/li\u003e\n\u003cli\u003eO\u0026apos;Sullivan RJ, Ozerov M, Bolstad, GH, Gilbey J, Jacobsen JA, Erkinaro J, Rikardsen AH, Hindar K, Aykanat T (2022) Genetic stock identification reveals greater use of an oceanic feeding ground around the Faroe Islands by multi-sea winter Atlantic salmon, with variation in use across reporting groups. ICES J Mar Sci 79(9):2442-2452. doi:10.1093/icesjms/fsac182.\u003c/li\u003e\n\u003cli\u003eOlmos M, Massiot-Granier F, Pr\u0026eacute;vost E, Chaput G, Bradbury IR, Nevoux M, Rivot E (2019) Evidence for spatial coherence in time trends of marine life history traits of Atlantic salmon in the North Atlantic. Fish Fish 20(2):322-342. doi:https://doi.org/10.1111/faf.12345.\u003c/li\u003e\n\u003cli\u003eOlmos M, Payne MR, Nevoux M, Pr\u0026eacute;vost E, Chaput G, Du Pontavice H, Guitton J, Sheehan T, Mills K, Rivot E (2020) Spatial synchrony in the response of a long range migratory species (\u003cem\u003eSalmo salar\u003c/em\u003e) to climate change in the North Atlantic Ocean. Glob Change Biol 26(3):1319-1337. doi:https://doi.org/10.1111/gcb.14913.\u003c/li\u003e\n\u003cli\u003eOlsen RE, Henderson RJ, Ringo E (1991) Lipids of arctic charr, \u003cem\u003esalvelinus-alpinus (l)\u003c/em\u003e .1. dietary induced changes in lipid class and fatty-acid composition. Fish Physiol Biochem 9(2):151-164. https://doi.org/10.1007/BF02265131\u003c/li\u003e\n\u003cli\u003eOlsen RE, Taranger GL, Svasand T, Skilbrei OT (2013) Improved method for triacylglycerol-derived fatty acid profiling by various non-lethal and lethal sampling techniques in Atlantic salmon. Aquacult Environ Interact 4(3):251-261. DOI: 10.3354/aei00087.\u003c/li\u003e\n\u003cli\u003ePedersen TL (2024) Patchwork: The Composer of Plots. R package version 1.2.0. https://CRAN.R-project.org/package=patchwork\u003c/li\u003e\n\u003cli\u003ePetursdottir H, Gislason A, Falk-Petersen S, Hop H, Svavarsson J (2008) Trophic interactions of the pelagic ecosystem over the Reykjanes Ridge as evaluated by fatty acid and stable isotope analyses. Deep-Sea Res Part II Top Stud Oceanogr 55(1-2):83-93. https://doi.org/10.1016/j.dsr2.2007.09.003\u003c/li\u003e\n\u003cli\u003ePond DW, Ward P (2011) Importance of diatoms for Oithona in Antarctic waters. J Plankton Res 33(1): 105-118. doi:10.1093/plankt/fbq089.\u003c/li\u003e\n\u003cli\u003ePotter ECE, Crozier WW (2000) A perspective on the marine survival of Atlantic salmon. . \u003cem\u003eIn \u003c/em\u003eThe ocean life of Atlantic salmon \u0026mdash; environmental and biological factors influencing survival. \u003cem\u003eEdited by\u003c/em\u003e D. Mills. Fishing News Books, Oxford. pp. 19\u0026ndash;36.\u003c/li\u003e\n\u003cli\u003eRikardsen AH, Dempson JB (2010) Dietary Life-Support: The Food and Feeding of Atlantic Salmon at Sea. \u003cem\u003eIn \u003c/em\u003eAtlantic Salmon Ecology. pp. 115-143.\u003c/li\u003e\n\u003cli\u003eRikardsen AH, Righton D, Strom JF, Thorstad EB, Gargan P, Sheehan T, Okland F, Chittenden CM, Hedger RD, N\u0026aelig;sje TF, Renkawitz M, Sturlaugsson J, Caballero P, Baktoft H, Davidsen JG., Halttunen E, Wright S, Finstad B, Aarestrup K (2021) Redefining the oceanic distribution of Atlantic salmon. Sci Rep \u003cstrong\u003e11\u003c/strong\u003e(1). doi:10.1038/s41598-021-91137-y.\u003c/li\u003e\n\u003cli\u003eSalminen M, Erkamo E, Salmi J (2001) Diet of post-smolt and one-sea-winter Atlantic salmon in the Bothnian Sea, Northern Baltic. J Fish Biol 58(1):16-35. doi:https://doi.org/10.1111/j.1095-8649.2001.tb00496.x.\u003c/li\u003e\n\u003cli\u003eSch\u0026auml;fer F, Oesterwind D, Sell AF, Kammann U (2024) Fatty acid analyses reveal differences in feeding ecology of North Sea squids that overlap in time and space. Food Webs 40: e00355.\u003c/li\u003e\n\u003cli\u003eSheehan TF, Reddin DG, Chaput G, Renkawitz MD (2012) SALSEA North America: a pelagic ecosystem survey targeting Atlantic salmon in the Northwest Atlantic. ICES J Mar Sci 69(9):1580-1588. doi:10.1093/icesjms/fss052.\u003c/li\u003e\n\u003cli\u003eSheridan MA (1989) Alterations in lipid-metabolism accompanying smoltification and seawater adaptation of salmonid fish. Aquaculture 82(1-4):191-203.\u003c/li\u003e\n\u003cli\u003eSkagseth O, Broms C, Gundersen K, Hatun H, Kristiansen I, Larsen KMH, Mork KA, Petursdottir H, S\u0026oslash;iland H (2022) Arctic and Atlantic Waters in the Norwegian Basin, Between Year Variability and Potential Ecosystem Implications. Front Mar Sci \u003cstrong\u003e9\u003c/strong\u003e. doi:10.3389/fmars.2022.831739.\u003c/li\u003e\n\u003cli\u003eSkjoldal HR (2004) The Norwegian Sea ecosystem. Tapir academic press, Trondheim, Norway.\u003c/li\u003e\n\u003cli\u003eSlotte A (1999) Differential utilization of energy during wintering and spawning migration in Norwegian spring-spawning herring. J Fish Biol 54(2):338-355. https://doi.org/10.1111/j.1095-8649.1999.tb00834.x\u003c/li\u003e\n\u003cli\u003eStefansson SO, Bjornsson BT, Sundell K, Nyhammer G, McCormick SD (2003) Physiological characteristics of wild Atlantic salmon post-smolts during estuarine and coastal migration. J Fish Biol 63(4):942-955. https://doi.org/10.1046/j.1095-8649.2003.00201.x\u003c/li\u003e\n\u003cli\u003eStrand E, Bag\u0026oslash;ien E, Edwards M, Broms C, Klevjer T (2020) Spatial distributions and seasonality of four Calanus species in the Northeast Atlantic. Prog Oceanogr 185. doi:10.1016/j.pocean.2020.102344.\u003c/li\u003e\n\u003cli\u003eThomas SM, Kainz MJ, Amundsen PA, Hayden B, Taipale SJ, Kahilainen KK (2019). Resource polymorphism in European whitefish: Analysis of fatty acid profiles provides more detailed evidence than traditional methods alone. Plos One 14(8).\u003c/li\u003e\n\u003cli\u003eTiedemann M, Nash RDM, Stenevik EK, Stiasny MH, Slotte A, Kjesbu OS, Robert D (2021) Environmental influences on Norwegian spring-spawning herring (\u003cem\u003eClupea harengusL.\u003c/em\u003e) larvae reveal recent constraints in recruitment success. ICES J Mar Sci78(2):640-652. doi:10.1093/icesjms/fsaa072.\u003c/li\u003e\n\u003cli\u003eTocher DR (2010) Fatty acid requirements in ontogeny of marine and freshwater fish. Aquac Res 41(5):717-732. doi:10.1111/j.1365-2109.2008.02150.x.\u003c/li\u003e\n\u003cli\u003eTr\u0026eacute;hin C, Rivot E, Santanbien V, Patin R, Gregory SD, Lamireau L, Marchand F, Beaumont WRC, Scott LJ, Hillman R, Besnard AL, Boisson PY, Meslier L, King AR, Stevens JR, Nevoux M (2023) A multi-population approach supports common patterns in marine growth and maturation decision in Atlantic salmon (Salmo salar L.) from southern Europe. J Fish Biol 104\u003cstrong\u003e:\u003c/strong\u003e125-138. doi:https://doi.org/10.1111/jfb.15567.\u003c/li\u003e\n\u003cli\u003eTrenkel VM, Huse G, MacKenzie BR, Alvarez P, Arrizabalaga H, Castonguay M, Gani N, Gregoire F, Hatun H, Jansen T, Jacobsen JA, Lehodey P, Lutcavage M, Mariani P, Melvin GD, Neilson JD, N\u0026oslash;ttestad L, Oskarsson GJ, Payne MR, Richardson DE, Senina I, Speirs DC (2014) Comparative ecology of widely distributed pelagic fish species in the North Atlantic: Implications for modelling climate and fisheries impacts. Prog Oceanogr 129:219-243. doi:10.1016/j.pocean.2014.04.030.\u003c/li\u003e\n\u003cli\u003eTyldesley E, Banas NS, Diack G, Kennedy R, Gillson J, Johns DG, Bull C (2024) Patterns of declining zooplankton energy in the northeast Atlantic as an indicator for marine survival of Atlantic salmon. ICES J Mar Sci 81:1164\u0026ndash;1184. https://doi.org/10.1093/icesjms/fsae077\u003c/li\u003e\n\u003cli\u003eUtne KR, Pauli BD, Haugland M, Jacobsen JA, Maoileidigh N, Melle W, Broms CT, N\u0026oslash;ttestad L, Holm M, Thomas K, Wennevik V (2021a) Poor feeding opportunities and reduced condition factor for salmon post-smolts in the Northeast Atlantic Ocean. ICES J Mar Sci78(8):2844-2857. doi:10.1093/icesjms/fsab163.\u003c/li\u003e\n\u003cli\u003eUtne KR, Thomas K, Jacobsen JA, Fall J, Maoil\u0026eacute;idigh NO, Broms CT, Melle W (2021b) Feeding interactions between Atlantic salmon (\u003cem\u003eSalmo salar\u003c/em\u003e) postsmolts and other planktivorous fish in the Northeast Atlantic. Can J Fish Aquat Sci 78(3):255-268. doi:10.1139/cjfas-2020-0037.\u003c/li\u003e\n\u003cli\u003eUtne KR, Skagseth O, Wennevik V, Broms CT, Melle W, Thorstad EB (2022) Impacts of a Changing Ecosystem on the Feeding and Feeding Conditions for Atlantic Salmon During the First Months at Sea. Front Mar Sci 9. doi:10.3389/fmars.2022.824614.\u003c/li\u003e\n\u003cli\u003evan Rij J, Wieling M, Baayen RH, van Rijn H (2022) itsadug: Interpreting time series and autocorrelated data using GAMMs. R package version 2.4.1.\u003c/li\u003e\n\u003cli\u003eVollset KW, Urdal K, Utne KR, Thorstad EB, S\u0026aelig;grov H, Raunsgard A, Skagseth O, Lennox RJ, \u0026Oslash;stborg GM, Ugedal O, Jensen AJ, Bolstad GH, Fiske P (2022) Ecological regime shift in the Northeast Atlantic Ocean revealed from the unprecedented reduction in marine growth of Atlantic salmon. Sci Adv \u003cstrong\u003e8\u003c/strong\u003e(9). doi:10.1126/sciadv.abk2542.\u003c/li\u003e\n\u003cli\u003eWickham H (2016) Ggplot2: elegant graphics for data analysis. Springer-Verlag, New York https://ggplot2.tidyverse.org \u003c/li\u003e\n\u003cli\u003eWood SN (2011) Fast stable restricted maximum likelihood and marginal likelihood estimation of semiparametric generalized linear models. J R Stat Soc Ser B 73(1):3-36. doi:https://doi.org/10.1111/j.1467-9868.2010.00749.x.\u003c/li\u003e\n\u003cli\u003eZhukova NV, Aizdaicher NA (1995) Fatty-Acid Composition of 15 Species of Marine Microalgae. Phytochemistry 39(2):351-356. https://doi.org/10.1016/0031-9422(94)00913-E\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"feeding interaction, zooplankton, lipids, diet, pelagic","lastPublishedDoi":"10.21203/rs.3.rs-7827046/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7827046/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAnalysis of the fatty acid composition in body tissue, which reflects the accumulated dietary intake over the last months, is a well-established method for studying trophic interactions in marine food webs. Here we present the fatty acid composition of salmon, herring, mackerel, and their prey sampled in May-Aug at marine feeding areas in the Norwegian Sea. A large proportion of the post-smolts sampled early in the summer had a high proportion of FA associated with age-0 fish. Later in the summer, when they have reached the northern Norwegian Sea, post-smolts have a higher proportion of FAs associated with calanoid copepods. The FA composition indicates of post-smolts indicate a wide feeding niche with a diet changing rapidly in time and space, and somatic growth was prioritized before lipid accumulation until the end of the first summer. Post-smolts have very low lipids levels compared to sub-adult salmon or other pelagic fish feeding in the same geographic region of the northeast Atlantic. Furthermore, the FA composition of post-smolts, which are hypothesized to compete for prey with other pelagic fish species, is partly different from the FA composition of herring and mackerel caught within the same geographic area, suggesting important diet differences between the species.\u003c/p\u003e","manuscriptTitle":"Fatty acids and lipid levels in Atlantic salmon (Salmo salar) change with sea-age and differ from other pelagic fish in the Norwegian sea","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-10 16:18:28","doi":"10.21203/rs.3.rs-7827046/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e347280a-5b68-40f0-9208-6531e240623f","owner":[],"postedDate":"November 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-15T13:39:31+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-10 16:18:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7827046","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7827046","identity":"rs-7827046","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}