Competition for shared resources is commonly assumed to restrict population-level niche width of coexisting species. However, the identity and abundance of coexisting species, the prevailing environmental conditions, and the individual body size may shape the effects of interspecific interactions on species’ niche width.

Here we study the effects of inter- and intraspecific interactions, lake area and altitude, and fish body size on the trophic niche width and resource use of a generalist predator, the littoral-dwelling large, sparsely-rakered morph of European whitefish (Coregonus lavaretus; hereafter LSR whitefish). We use stable isotope, diet and survey fishing data from 14 subarctic lakes along an environmental gradient in northern Norway.

The isotopic niche width of LSR whitefish showed a humped-shaped relationship with increasing relative abundance of sympatric competitors, suggesting widest population niche at intermediate intensity of interspecific interactions. The isotopic niche width of LSR whitefish tended to decrease with increasing altitude, suggesting reduced niche in colder, less productive lakes.

LSR whitefish typically shifted to a higher trophic position and increased reliance on littoral food resources with increasing body size, although between-lake differences in ontogenetic niche shifts were evident. In most lakes, LSR whitefish relied less on littoral food resources than coexisting fishes and the niche overlap between sympatric competitors was most evident among relatively large individuals (>250 mm). Individual niche variation was highest among >200 mm long LSR whitefish, which likely have escaped the predation window of sympatric predators.

We demonstrate that intermediate intensity of interspecific interactions may broaden species’ niche width, whereas strong competition for limited resources and high predation risk may suppress niche width in less productive environments. Acknowledging potential humped-shaped relationships between population niche width and interspecific interactions can help us understand species’ responses to environmental disturbance (e.g., climate change and species invasions) as well as the driving forces of niche specialization.

The 14 study lakes are relatively small (0.19–23.7 km²), dimictic and meso- or oligotrophic lakes situated along an altitude gradient (275–540 m a.s.l.) in the Finnmark region in northern Norway. The study area has very limited human activity and the catchment areas consist mainly of mountain birch forest and bogs. No major fishery or any stocking activity occur in the study lakes. Besides whitefish, perch, grayling and charr, some lakes have relatively scarce populations of burbot (Lota lota), brown trout (Salmo trutta), pike (Esox lucius), European minnow (Phoxinus phoxinus) and/or nine-spined stickleback (Pungitius pungitius). LSR whitefish is present in all study lakes, whereas six of the lakes host polymorphic whitefish populations having the additional presence of a small pelagic, densely-rakered (DR) and/or a small profundal, sparsely-rakered (SSR) whitefish morph (Siwertsson et al., 2013; Häkli et al., 2018).

Survey fishing was conducted between 8^th August and 17^th September in 2005­–2009 using the same multi-mesh gillnets set overnight in the littoral, profundal and pelagic habitats of each study lake. The used multi-mesh gillnets were 40 m long and composed of eight randomly distributed 5 m panels of 10, 12.5, 15, 18.5, 22, 26, 35 and 45 mm bar mesh sizes (knot to knot). In the littoral (0–8 m depth) and profundal (>12 m depth) habitats, 1.5 m deep bottom nets were used, whereas 6 m deep floating nets were used in the pelagic zone in each lake. Supplementary material for stable isotope (hereafter SIA) and stomach contents (hereafter SCA) analyses were sampled using 30 m long and 1.5 m deep single-mesh (20, 22, 26, 32, 35 mm) gillnets. However, the supplementary samples were excluded from catch statistics. All study lakes were sampled in a single year between 2007–2009, except Stuorajavri with main fish sampling in 2005 supplemented by profundal sampling in 2008, following the same sampling protocols and mostly by the same personnel.

After removal from the gillnets, fish were killed by a sharp blow on the head, kept cool and brought to the field laboratory, where they were identified, measured (fork length, mm) and weighed (g). From a subsample of fish, a small piece of dorsal muscle tissue was dissected and stored at –20°C until subsequent preparation for SIA. The stomachs were removed, preserved in 96% ethanol, and later opened to visually estimate the total fullness on a percentage scale ranging from empty (0%) to full (100%) (Amundsen & Sánchez-Hernández, 2019). The prey items were identified to species, genus or family level and their relative contribution to the total fullness was estimated (Amundsen et al., 1996; Amundsen & Sánchez-Hernández, 2019). All observed 26 prey taxa were included in the calculation of proportional diet similarity index (PSi) for each individual fish (with stomach fullness exceeding 10%), as well as the degree of among-individual diet variation (V = 1 ­– mean PSi) in each whitefish population following the equations described in Bolnick et al. (2002) and Svanbäck et al. (2015).

Besides fish muscle tissue, qualitative samples of putative littoral and pelagic food resources were collected for SIA from each lake in August/September 2007–2009 to study the overall food-web structures (Figure 2) and particularly to estimate the relative littoral reliance (LR), trophic position (TP) and isotopic population niche width (cf. Petta et al., 2020) of LSR whitefish in the study lakes. The LR and TP estimates were calculated using the linear isotopic mixing models described in Post (2002). The LR and TP estimates estimates were used in calculation of population niche width, measured as the total convex hull area (TA; see Jackson et al., 2011 for details) encompassing the LR and TP values of all whitefish individuals in each lake. To validate the robustness of our results, we also present the sample-size corrected SEA_c and the Bayesian SEAb estimates of population niche width (Jackson et al., 2011) along with the TA estimates. Finally, we also calculated the mean nearest neighbour distances (MNND), based on Euclidean distances between individual data points in the LR – TP space, to measure trophic diversity among individuals (Jackson et al., 2011).

References:

Amundsen, P.-A., & Sánchez‐Hernández J. (2019). Feeding studies take guts – critical review and recommendations of methods for stomach contents analysis in fish. Journal of Fish Biology, 95, 1364–1373. https://doi.org/10.1111/jfb.14151

Amundsen, P.-A., Gabler, H.-M., & Staldvik, F.J. (1996). A new approach to graphical analysis of feeding strategy from stomach contents data –modification of the Costello (1990) method. Journal of Fish Biology, 48, 607–614. https://doi.org/10.1111/j.1095-8649.1996.tb01455.x

Häkli, K., Østbye, K., Kahilainen, K.K., Amundsen, P.-A., & Præbel, K. (2018). Diversifying selection drives parallel evolution of gill raker number and body size along the speciation continuum of European whitefish. Ecology and Evolution, 8, 2317–2631. https://doi.org/10.1002/ece3.3876

Jackson, A.L., Inger, R., Parnell, A.C., & Bearhop, S. (2011). Comparing isotopic niche widths among and within communities: SIBER – Stable Isotope Bayesian Ellipses in R. Journal of Animal Ecology, 80, 595–602. https://doi.org/10.1111/j.1365-2656.2011.01806.x

Post, D.M. (2002). Using stable isotopes to estimate trophic position: models, methods, and assumptions. Ecology, 83, 703–718. https://doi.org/10.1890/0012-9658(2002)083[0703:USITET]2.0.CO;2

Siwertsson, A., Knudsen, R., Præbel, K., Adams, C.E., Newton, J., & Amundsen, P.-A. (2013). Discrete foraging niches promote ecological, phenotypic, and genetic divergence in sympatric whitefish (Coregonus lavaretus). Evolutionary Ecology, 27, 547–564. https://doi.org/10.1007/s10682-012-9607-x

Data files

Lake_modelling_data.csv : lake-specific data used in linear models predicting isotopic niche width (TA and SEAc), trophic diversity (MNND) and among-individual diet variation (V) of LSR whitefish as a function of interspecific (Inter) and intraspecific (Intra) interactions and lake altitude and surface area.

Individual_SIA_data.txt : individual stable isotope data of LSR whitefish collected from the 14 study lakes.

Individual_SCA_data.txt : individual stomach contents (diet) data of LSR whitefish collected from the 14 study lakes.

All_fishes_SIA_data.txt : individual stable isotope data of the most abundant fish species collected from the 14 study lakes used to plot size-dependent niche shifts in LR (Figure 4) and TP (Figure 5) estimates.

Data in columns

"waterBody" = lake name

"waterBodyID" = unique lake number in Norwegian lake databases

"Altitude" = altitude aka elevation in meters above the sea level

"Area" = lake surface area in square kilometers (km²)

"MaximumDepth" = lake maximum depth in meters (m).

 "Inter" = proxy for intensity of interspecific interactions measured as the proportion of fish species other than whitefish in the total multi-mesh survey gillnet catches (i.e., total fish biomass in grams including fish from all habitats) in each lake.

 "Intra" = proxy for intensity of intraspecific resource competition. This proxy variable was standardized for different sampling efforts in each study lake (Table S1) by calculating total catch per unit of effort (CPUE; measured as the number of whitefish individuals caught per 100 m² gillnet area per night), including all present whitefish morphs in each lake.

"TA" = total convex hull area (TA; see Jackson et al., 2011 for details) encompassing the LR and TP values of all LSR whitefish individuals in each lake used as a proxy for population (or "isotopic") niche width.

"SEAc" = sample-size corrected standard ellipse area (SEA_c; Jackson et al., 2011) encompassing the core set of LR and TP values of LSR whitefish individuals in each lake, used as a proxy for population (or "isotopic") niche width.

"V" = the degree of among-individual diet variation (V = 1 ­– mean PSi). See below the description for proportional diet similarity index PSi.

"MNND" = mean nearest neighbour distance (MNND) based on Euclidean distances between individual data points in the LR – TP space and measuring trophic diversity among individuals (Jackson et al., 2011).

"scientificName" = Latin (scientific) name of the fish species

"Length" = total length of fish in millimeters (mm).

"PSi" = proportional diet similarity index calculated for each individual fish with stomach fullness exceeding 10%. PSi indices were further used to calcualte the degree of among-individual diet variation (V = 1 ­– mean PSi) in each whitefish population following the equations described in Bolnick et al. (2002) and Svanbäck et al. (2015). The PSi index compares each individual’s diet to that of the entire population, with values ranging between 0 and 1. In populations where individuals specialize on different kinds of prey, the PSi values tend to be low and the resulting population-level V values tend to be high, i.e. approaching 1 (Bolnick et al., 2002; Svanbäck et al., 2015).

"d13C" = stable carbon isotope value (δ¹³C) of fish muscle tissue used in calculation of littoral reliance estimates (LR, see descriptions below).

"d15N" = stable nitrogen isotope value (δ¹⁵N) of fish muscle tissue used in calculation of trophic position estimates (TP, see description below).

"LR" = estimate of the relative reliance of fish on littoral (versus pelagic) carbon sources calcualted using the two-source isotopic mixing model described in Post (2002). See more detailed description of LR calculation below.

"TP" = estimate of the trophic position of fish in the lake food web calcualted using the two-source isotopic mixing model described in Post (2002). See more detailed description of TP calculation below.

The LR (eqn 1) and TP (eqn 2) estimates were calculated using the linear isotopic mixing models described in Post (2002):

LR = (δ¹³C_fish – δ¹³C_pel) / (δ¹³C_lit – δ¹³C_pel)                                                              eqn (1)

TP = λ + (δ¹⁵N_fish – [δ¹⁵N_lit x LR + δ¹⁵N_pel x (1 – LR)] / Δ_n                                     eqn (2)

where δ¹³C_fish and δ¹⁵N_fish refer to isotope values of individual fish; δ¹³C_lit, δ¹⁵N_lit, δ¹³C_pel and δ¹⁵N_pel refer to the lake-specific δ¹³C and δ¹⁵N values for the littoral and pelagic isotopic end-members; λ is the trophic position of the organisms used to estimate δ¹⁵N_lit and δ¹⁵N_pel, (here λ = 2 for primary consumers); and Δ_n is the mean trophic fractionation of muscle tissue δ¹⁵N (i.e., 2.9‰; McCutchan et al., 2003). For LR calculation, δ¹³C_fish were corrected for trophic fractionation by subtracting 1.3‰ from the original δ¹³C value (McCutchan et al., 2003). The littoral and pelagic isotopic end-members were defined as the mean isotope values of algae-grazing littoral benthic invertebrates (i.e., snails, Gammarus lacustris amphipods and chironomid larvae; δ¹³C ≥ -25‰) and pelagic zooplankton (δ¹³C ≤ -28‰), respectively.

References

Bolnick, D.I., Yang, L.H., Fordyce, J.A., Davis, J.M., & Svanbäck, R. (2002). Measuring individual-level resource specialization. Ecology, 83, 2936–2941. https://doi.org/10.1890/0012-9658(2002)083[2936:MILRS]2.0.CO;2

Jackson, A.L., Inger, R., Parnell, A.C., & Bearhop, S. (2011). Comparing isotopic niche widths among and within communities: SIBER – Stable Isotope Bayesian Ellipses in R. Journal of Animal Ecology, 80, 595–602. https://doi.org/10.1111/j.1365-2656.2011.01806.x

McCutchan, J.H., Jr, Lewis, W.M., Jr, Kendall, C., & McGrath, C.C. (2003). Variation in trophic shift for stable isotope ratios of carbon, nitrogen, and sulfur. Oikos, 102, 378–390. https://doi.org/10.1034/j.1600-0706.2003.12098.x

Post, D.M. (2002). Using stable isotopes to estimate trophic position: models, methods, and assumptions. Ecology, 83, 703–718. https://doi.org/10.1890/0012-9658(2002)083[0703:USITET]2.0.CO;2

Svanbäck, R., & Persson, L. (2004). Individual diet specialization, niche width and population dynamics: implications for trophic polymorphism. Journal of Animal Ecology, 73, 973–982. https://doi.org/10.1111/j.0021-8790.2004.00868.x