Presumed seed specialists rely on fungi as their primary food source at the sub-Arctic treeline

Presumed seed specialists rely on fungi as their primary food source at the sub-Arctic treeline | 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 Presumed seed specialists rely on fungi as their primary food source at the sub-Arctic treeline Alexandra Windsor, John Markham, James Roth This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6624068/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 Range boundaries limit local populations, which may experience pronounced fluctuations in resource availability, particularly at higher latitudes, often seen as resource pulses. In boreal forests, conifers undergo pulses of seed production followed by intervals of low seed production, profoundly affecting consumers dependent on these resources. Red squirrels ( Tamiasciurus hudsonicus ) are considered seed specialists across the boreal forest. We evaluated how annual changes in white spruce ( Picea glauca ) cone production at the sub-Arctic treeline near Churchill, MB, Canada, influenced squirrels’ use of alternative food sources, predicting that low cone production would increase reliance on alternate foods. Cone crops varied from 2020–2023, with a mast year in 2022 of 471 cones per tree, approximately 70–80% lower than mast years elsewhere, and lower crops in other years (6-115 cones per tree). Furthermore, the number of filled seeds (containing an embryo) per cone was low, ranging from 0.6 ± 0.03 (mean ± SE) in 2022 to 3.6 ± 2.6 in 2023. Using stable isotope ratios of hair and Bayesian mixing models, we found that squirrels primarily consumed fungi (~ 70% of diet), even in mast years, with other food sources varying with cone production. The dominance of fungi in squirrel diet even in mast years, highlights the dietary plasticity of red squirrels beyond seed specialization challenging the seed specialization paradigm. Flexible foraging strategies likely allow populations to persist in resource-limited environments and may facilitate range expansion as climate change reshapes habitats. Population Biology American red squirrel diet facultative specialist stable isotopes range boundaries Tamiasciurus hudsonicus Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Species with extensive geographical distributions often have smaller, fragmented populations near their range limits, where habitats are harsher than at the core of their distributions (Gaston 2003 , Lynch et al. 2014 ). Range boundaries can be shaped by physical barriers directly preventing species’ dispersal (Gross & Price 2000 ) or gradients in climatic variables, which create physiological limits. In the northern hemisphere, it is commonly thought that northern range boundaries are frequently governed by abiotic factors like extreme climate, while biotic factors, such as competition, and food availability, impose additional constraints at southern range limits (Brown et al. 1995 , Sirén & Morelli 2020 ). However, combinations of factors likely work in concert, limiting the expansion of species beyond their range. Compared to core environments, range boundary habitats often exhibit greater climatic variability (Rehm et al. 2015 ), which imposes physiological limits and impact population growth (Sexton et al. 2009 ). Although range boundary habitats can be expansive, they are often low quality or exist as fragmented patches of high-quality habitat. Climatic factors and the corresponding influences on habitat suitability, population density, and reproductive fitness strongly reinforce range limits and can be further exacerbated by inter and intraspecific interactions and resource availability. At northern range boundaries, extreme climates can reduce the availability and accessibility of food sources, particularly during winter. The variation in seasonal climates at higher latitudes dictates large annual and seasonal fluctuations in resource availability (Humphries et al. 2005 ), as shorter growing seasons and cooler seasonal temperatures restrict primary and secondary productivity (Gross & Price 2000 , Callaghan et al. 2004 ). Species respond to fluctuations in food sources in multiple ways. Some species, such as the boreal woodland caribou ( Rangifer tarandus caribou ), migrate seasonally following food sources (Rettie & Messier 2000 , Ferguson & Elkie 2004 ). Other species, such as wolverines ( Gulo gulo) , may change activity patterns or cache food to buffer against fluctuating food availability in an overall low-productivity environment (van der Veen et al. 2020 ). When food is spatially and temporally variable, species may rely on pulses of food sources. Pulsed resources are episodic events of superabundant resource production with long inter-pulse periods of normal or subnormal production. While resource pulses are short-lived, they drive the dynamics of many plant and animal communities. The bottom-up effect of pulsed resources can be seen at the level of the individual, the population, and indirectly at the community level (Yang et al. 2008 , Yang et al. 2010 ). At the individual level, consumer diets shift in response to pulsed resources, which induces numerical responses in consumer populations. These responses then propagate across trophic levels as increased densities in consumers become secondary resource pulses for higher trophic levels (Yang et al. 2010 ). The boreal forest, one of the world’s largest terrestrial biomes, represents the northern range limit for many species (Kayes & Mallik 2020 , Krebs et al. 2023 ). Conifers in the boreal forest experience periodic masting, with years of high seed production followed by long intervals of low or absent seed production (Kelly & Sork 2002 , Lamontagne & Boutin 2007 , Archibald et al. 2012 ). Mast years are often synchronous across large geographical areas (Lamontagne & Boutin 2007 , Krebs et al. 2012 ) and overwhelm seed predators, enhancing seed survival and establishment (Kelly & Sork 2002 ). White spruce ( Picea glauca ) is the dominant tree species in dry habitats at the north limit of the boreal forest (Bonan and Shugart 1989) and exhibits periodic masting (Archibald et al. 2012 , Krebs et al. 2012 , Krebs et al. 2023 ) approximately every 3–5 years (Lamontagne & Boutin 2007 ). Many small mammals rely heavily on spruce seeds, but may switch to alternative food sources, such as fungi, which can be highly abundant but nutrient-poor, during non-mast years (Ostfeld & Keesing 2000 , Fletcher et al. 2010 ). The dietary strategies of species can be defined based on the breadth of their dietary niches. Facultative generalists typically have a very broad dietary niche compared to facultative specialists that exhibit a narrower dietary niche (Shipley et al. 2009 , Pagani-Núñez et al. 2016 ). However, both groups exhibit behavioural plasticity. Facultative generalists optimize resources by switching to more abundant or accessible food sources, while facultative specialists can expand their diet when primary resources are scarce, enabling both groups to exploit fluctuating resources and maintain broad geographic distributions, making them more successful in habitats where food availability is low and heterogeneous (Shipley et al. 2009 , Newbury & Hodges 2018 , Szumski et al. 2023 ). In contrast, obligate generalists, though able to use diverse food sources, have limited specialization and may be less efficient in optimizing their diet in such environments, and obligate specialists are vulnerable due to their reliance on specific resources, which could be scarce (Shipley et al. 2009 , Dehling et al. 2021 ). Due to the low predictability of food abundance, generalist species are common at northern range boundaries (Callaghan et al. 2004 ). Facultative specialists living in these habitats may exhibit behaviours more like facultative generalists when primary food sources are unpredictable and fluctuate dramatically. Individuals may also employ behavioural methods such as food hoarding and increasing litter and clutch sizes to take advantage of years of superabundant food (Callaghan et al. 2004 , Boutin et al. 2006 , McAdam et al. 2019 ). American red squirrels ( Tamiasciurus hudsonicus ) are primarily granivorous, specialising in conifer seeds across most of their distribution (Steele 1998 , Boutin et al. 2006 ). In the northern parts of their range, red squirrels appear to feed almost exclusively on white spruce seeds (McAdam & Boutin 2003). While seeds constitute the majority of squirrel diet, red squirrels can opportunistically consume a large variety of foods including fungi (Layne 1954 , Steele 1998 , Fletcher et al. 2010 ), berries (Benhamou 1996 , Currah et al. 2000 ), and animal prey, including songbird eggs (Bayne & Hobson 2002 , Willson et al. 2003 ) and young lagomorphs (Layne 1954 , Sullivan & Sullivan 1982 , Peers et al. 2020 ). These alternative food sources, however, typically make up a small proportion of annual diet (Layne 1954 ), and are relied on only when spruce cone availability is low. Despite observations of consumption of alternative food sources, detailed quantifications of non-seed food sources are often lacking (Currah et al. 2000 , Fletcher et al. 2010 ). Many life history parameters of red squirrels, including reproduction (Boutin et al. 2006 , Lane et al. 2015 ), growth (McAdam & Boutin 2003), and overwinter survival (Steury & Murray 2003 , LaMontagne et al. 2013 ), are intimately linked to spruce cone production (Humphries & Boutin 2000 , McAdam & Boutin 2003, Réale et al. 2003 , Boutin et al. 2006 ). Assuming seed productivity is consistent, seed availability can be estimated through spruce cone availability by counting the number of cones produced within the study area (Lamontagne & Boutin 2007 , Kucheravy et al. 2021 ). However, the quality and quantity of seeds produced each year can also vary (Waldron 1965 , Zasada 1988 ), further influencing overall food availability. We reconstructed the diet of red squirrels living at their sub-Arctic range boundary using stable isotope analysis and explored how spruce cone production affected squirrel diet. We hypothesised that squirrels would adopt a highly supplemented diet in low cone years and predicted that 1) overall cone production would be low, 2) squirrels would frequently consume alternate food sources, depending on the yearly abundance of cones, and 3) their dietary niche would be broad. We further predicted that 4) fall diet would be broader than spring diet, as alternate food sources such as fungi and berries are more abundant in fall. Methods Study area Our study area near Churchill, MB (58°45’ N, 94°04’ W) on the western edge of Hudson Bay, is on the transition between boreal forest and Arctic tundra. The forest is primarily composed of white spruce, black spruce ( Picea mariana ), and tamarack ( Larix laricina ) (Mamet & Kershaw 2011 , Mamet & Kershaw 2013 ), and is heavily fragmented by patches of wetlands and open tundra (Harper et al. 2011 , Harper et al. 2018 ). The forest has a low density of mature trees (~ 600 stems ha − 1 ) and a sparse canopy of ca. 25% (Lafleur 1999 ). The proximity of the sub-Arctic treeline to Hudson Bay, which is frozen eight months of the year, exposes the region to cool summer temperatures and extreme winters (Mamet & Kershaw 2011 ). In our study area, mean monthly temperatures ranged from − 28.4 to -15.1 ⁰C in winter (November-April) and 1.2 ⁰C to 20.6 ⁰C in summer (June-August) from 2019–2023 (Environment Canada). Fieldwork was conducted in boreal woodland consisting of patchy forest, transitioning to a denser forest and ending in an open-canopy forest near a large fen (Fig. 1 ). We established sites at 250 m intervals along a 7 km transect running north to south, looking for signs of squirrel activity such as vocalisations, cone debris, and middens. Middens are large food caches containing hundreds to tens of thousands of cones (Haines et al. 2022 , Wishart 2023 ), and are easily identified by the concentrated accumulation of cone debris, often reaching surfaces areas up to a few hundred m 2 (Steury & Murray 2003 ). Very few true middens, characterized by large fields of cone debris, were observed in our study area. Here, we define a ‘midden’ as any form of food cache, including true middens and burrows with cone debris that have the potential to contain cones. To account for juvenile dispersal and the possibility of squirrels taking over previously unoccupied sites, we checked all sites for signs of squirrel activity each year in early June. Food availability To assess annual spruce cone production, we conducted cone counts of individually marked trees each August (five trees per site; 9 sites in 2020, 26 sites in 2021–2023). Each tree was at least 5 m from the nearest sampled tree and had a minimum of 5 cm diameter at breast height (DBH), representing the approximate minimum age at which white spruce trees can bear cones (Lamontagne & Boutin 2007 ). At squirrel-occupied sites, we sampled the five closest trees to the midden. At unoccupied sites, we selected the five closest trees from a flagged reference point, following a bearing determined using a random number generator. Each August, we took 3–5 photographs of each tree on opposing sides. Using ImageJ image processing software, we counted the cones visible in each picture (Nygren et al. 2017 ). The photographs for each tree were matched to avoid double-counting cones. We calculated food availability at each site as the mean number of cones per tree (Kucheravy et al. 2021 ), multiplied by tree density to estimate cones per hectare. To calculate tree density at each site, we counted the number of white spruce trees (> 5 cm DBH) within four 15 m radius plots (two on each side of the flagged reference point) spaced 20 m apart. To estimate seed production, we used data from the control sites, which encompassed our sampling area, from Benjamin et al. ( 2024 ). We collected cones from the upper part of the crown using a pole pruner. The number of cones collected per tree depended on their availability but averaged 13 ± 1 cones. The cones were air-dried for at least 72 h prior to dissecting and removing the seeds. We placed the seeds in 95% ethanol to separate the filled seeds containing an embryo (sinking seeds) from empty seeds (Ho 1984 , Sirois 2000 ). To estimate availability of fungi, in 2022 we collected all epigeous fungal sporocarps within two randomly selected 15 m radius plots at each site. In 2023, we collected all epigeous fungi within four 1-m 2 quadrats placed 5 m from each site’s centre point in the four cardinal directions. At occupied sites, the midden served as the centre, and at unoccupied sites, we used the cone count centre point. All samples were oven-dried for 48–72 hrs, and dry fungal biomass (kg ha − 1 ) was used to reflect fungus availability at each site (Derbridge & Koprowski 2019 ). We monitored squirrel hoarding activity using motion-activated Browning trail cameras (model: Strike Force Extreme and Strike Force HD Pro X) installed at 17 occupied sites (one camera per site). We chose sites with previously observed middens and installed each camera facing the largest midden if a site had multiple middens. Cameras captured bursts of three photos when activated within a three-metre detection range, with a 0.5 second recovery between bursts. We recorded the date, time, location, and type of food items hoarded from August 1 to September 30, reflecting the typical harvest period for squirrels in central areas of the boreal forest (Fletcher et al. 2010 , Archibald et al. 2013 ) and to account for the shorter growing seasons and earlier winters characteristic of the northern boreal treeline. Squirrels entering middens with food or crossing the camera’s field of view carrying food, were classified as hoarding events. Observations may have underestimated activity if a midden had multiple entrances not covered by cameras. Stable isotope sample collection and preparation Red squirrel body hair moults in spring and fall, while tail hair moults only in the fall (Nelson 1945 ). We collected body hair (~ 3–4 clippings, ~ 0.5 cm behind the shoulder) and tail hair (~ 2–3 cm) each June, from a total of 103 squirrels (90 unique individuals, 9 recaptures, and 4 donated by local trappers) to compare seasonal diets. We captured squirrels using Tomahawk live traps (model #202). We recorded the sex and marked each squirrel with a Passive Integrated Transponder (PIT) tag injected under the skin for permanent identification and metal ear tags threaded with coloured pipe cleaners for visual identification We collected samples of potential food sources from forested and forest-wetland regions within our study area in August 2021 (Table S1). Samples included white (n = 9) and black spruce seeds (n = 9), tamarack seeds (n = 6) and buds (n = 15), and the most common species of berries (n = 97), lichen (n = 22), and fungi (n = 110). Hair samples were washed twice with mild, soapy water and rinsed thoroughly before drying in a drying oven for 48 hours. We removed seeds from spruce cones and dried the seeds, fungi, berry and lichen samples in a drying oven for a minimum of 48 hours, and ground all food samples to a powder using a mortar and pestle, washing the tools between samples with 70% ethanol. All samples were analyzed for carbon and nitrogen stable isotope ratios on a continuous-flow isotope ratio mass spectrometer (Delta V Advantage) coupled to a Costech 4010 Elemental Combustion system and a ConFlo IV universal interface. Data analysis We used stable isotope analysis to examine annual and seasonal variation in squirrel diets. The isotopic composition of a consumer’s tissues reflects dietary sources, adjusted by trophic discrimination factors (TDFs) to account for isotopic shifts during assimilation (Phillips 2012 , Parnell et al. 2013 ). We applied rodent-specific TDFs and standard deviations of Δ 13 C = 3.3 ± 2.2‰ and Δ 15 N = 1.9 ± 0.2‰ (Hobbie et al. 2017 , Pauli et al. 2019 ) to source isotope values. Using a K nearest-neighbor (KNN) randomization test (k = 15) we classified 231 samples into five isotopically distinct food groups: berries (species n = 7), lichen (n = 2), conifer seeds and buds (not isotopically distinct: ANOVA: δ 13 C: F 3,35 = 0.65, p = 0.59; δ 15 N: F 3,35 = 1.57, p = 0.38), and fungi divided into two groups (genera n = 5) due to δ 15 N variation across genera and habitats (forest vs forest-wetland) (Table S1, Figure S1). We estimated the dietary contribution of each food group to squirrel diet using Bayesian isotopic mixing models with the MixSIAR and rjags packages (Parnell et al. 2013 ) in R. All models incorporated uniform prior distributions and concentration dependence using the mean elemental concentrations for each prey group. We used Markov-chain Monte Carlo (MCMC) methods to estimate the parameters of the mixing models and ran three parallel MCMC chains with a burn-in of 50,000 iterations. We generated posterior samples using 150,000 iterations of the models and a thinning rate of 50 and checked model convergence using the Gelman and Geweke diagnostic tests produced by MixSIAR (Geweke 1991 , Jackson et al. 2011 , Gelman et al. 2014 ). Stable isotope ratios were also used to estimate dietary niche breadth, where narrower isotopic niches indicate more specialised diets (Bearhop et al. 2004 , Jackson et al. 2011 ). For opportunistic species, niche breadth can shift seasonally and annually with food availability. Isotopic niche breadth was modelled using an ellipse-based approach with the Stable Isotope Bayesian Ellipses in R (SIBER) package in R. We compared isotopic signatures from body (spring diet) and tail (fall diet) hair samples. Previous studies suggest the fall moult occurs between late August and into September (Kranowski 1969 ) and the spring moult occurs between April and May (Layne 1954 ), although populations at higher latitudes can moult between March and June (Lepage & Parker 1988). However, we did not observe signs of moulting during our trapping efforts throughout June and into early July. We therefore assumed that all body hair collected in June was recently grown. We generated metrics for isotopic niche breadths using standard ellipse areas corrected for small sample size (SEA c ) and Bayesian Standard Ellipse Areas (SEA b ) (150,000 MCMC iterations, 95% credible interval). We compared seasonal and annual niche breadths using the credible intervals to estimating the probability (0–1) that one group’s ellipses were smaller than another (Jackson et al. 2011 ). We also examined the dietary niche overlap between ellipses by calculating the proportion of ellipse overlap using 1,000 draws of the posterior estimates of each ellipse. All statistical analyses were conducted in R (version 4.0.5). We used generalised linear mixed effects models (GLMM) to determine if mean cone production varied annually, using ‘site ID’ as a random effect. Since the cone count data was overdispersed, we used a negative binomial link function. For fungal biomass, we used a paired t-test to determine if dry fungal biomass varied from 2022 to 2023. We examined annual variation in filled seed counts per cone using a linear model (LM) with cone production (total cones per tree) and seed production (total number of seeds per cone, comprising both filled and empty seeds) as additional explanatory variables. To examine the effect of cone abundance on hoarded food items, we used trail camera observations to determine the daily mean cone and fungi hoarded per camera. We used a GLMM for each food source with a Poisson link and accounted for spatial variation using ‘site ID’ as a random effect. We tested for annual and seasonal variation in squirrel diet, using a multivariate analysis of variance (MANOVA) to compare δ 13 C and δ 15 N values between years and body and tail hair. We further examined annual diet variations in MixSIAR, with ‘year’ (4 levels) as a fixed effect and ‘site ID’ (26 levels) as a random effect. Squirrels were only trapped at 26 of the 29 sites in our study area. We ran a second mixing model to estimate individual diets using ‘squirrel ID’ as a fixed effect. We incorporated the mean individual dietary proportions from MixSIAR into three beta regression models (berries, fungi, and lichen) with a canonical link function using the betareg package in R to determine if the consumption of alternate food sources fluctuated with cone abundance, using the mean cone abundance at the sites where the individuals were trapped. We used beta regression models since dietary proportions range from zero to one (McAulay et al. 2020 ). We paired all squirrel diets with cone production estimates from the previous fall, as tail hair reflects the previous fall’s diet and body hair collected in summer reflects the spring diet when most food items such as fungi, cones, and berries were not yet available. Thus, squirrels likely relied on food cached the previous fall. Results Tree density across sites averaged 483.7 stems ha − 1 (± 54.2 SE; range 56 to 579). From 2020 to 2023, annual spruce cone production varied significantly (Table S2), with the greatest number of cones per tree in the mast year of 2022 (471.2 ± 33.1), low crops in 2021 (7.9 ± 0.8) and 2023 (6 .2 ± 1.2), and a moderate cone year in 2020 (11.5 ± 6.0). The number of filled seeds per cone was low overall, but also varied annually (Table S2), with the greatest number of filled seeds per cone in 2023 (4.9 ± 1.6), the fewest number in the mast year of 2022 (0.6 ± 0.0), and a low production of filled seeds in 2021 (2.4 ± 0.6). Fungus production across sites ranged from 0.0 to 8.5 kg ha − 1 , with greater abundance in 2022 (3.5 ± 0.5) compared to 2023 (2.5 ± 0.6) (t 25 = 5.51, p = 0.014). Variability of fungal abundance between sites was high but lower in 2022 (coefficient of variation = 67%) compared to 2023 (coefficient of variation = 114%). Mean daily observations of spruce cone and fungus hoarding in August and September fluctuated annually in response to cone production (GLMM cones: z 1,46 = 9.1, p < 0.001; fungi: z 1,46 = -2.3 p = 0.004). Cone hoarding was highest in 2022 (8.1 ± 0.2 cones day − 1 ), lowest in 2021 (0.9 ± 0.3), and relatively low in 2020 (2.3 ± 0.3) and 2023 (1.4 ± 0.6). Fungus hoarding was highest in 2021 (1.0 ± 0.2 sporocarp day − 1 ), lowest in 2020 (0.1 ± 0.0) and 2022 (0.1 ± 0.0), and relatively low in 2023 (0.7 ± 0.1). Male and female squirrels did not differ in δ 13 C (t = -0.44, df = 51.43, p = 0.661) or δ 15 N (t = -0.97, df = 51.43, p = 0.543) (Table S3 and Figure S1), and neither δ 13 C nor δ 15 N values varied seasonally (MANOVA F 1,87 = 2.9, p = 0.596). The Bayesian ellipses were also similar in size (female SEA b = 3.7; male SEA b = 3.2) and the proportional overlap between male and female ellipses was also high (0.862), further suggesting a similar diet. However, δ 15 N values varied annually within each season (spring: F 2,86 = 18.53, p < 0.001; fall: F 2,86 = 14.19, p < 0.001), but δ 13 C did not (spring: F 2,86 = 1.72, p = 0.153; fall: F 2,86 = 0.99, p = 0.135). The areas of the Bayesian ellipses reflecting dietary niche breadths in fall (SEA b = 2.9) and spring (SEA b =3.3 2.8) were similar (Fig. 2 ), as the proportion of posterior draws where the ellipses were smaller in fall than spring was 0.514. The proportional overlap between spring and fall ellipses was 0.815, further suggesting a high degree of similarity between seasonal diets (Table S3). The areas of the ellipses were largest in 2022 (spring: SEA b = 2.5; fall: SEA b = 2.7), following a low crop in 2021, and the smallest in 2023 (spring: SEA b = 1.0; fall: SEA b = 1.1), following the mast year in 2022 (Fig. 2 , Fig. 3 ). Similarly, the dietary niche breadth in 2023, following the mast year, was narrower than 2022 (reflecting a low crop in 2021) for both spring and fall (Table S3). Squirrel diet consisted predominantly of fungi (Fig. 4 ) (> 57% in spring and > 62% in fall), with a small conifer cone contribution (~ 30% in spring and fall) (Table S4 and Figure S2). However, consumption of different food groups varied annually (Fig. 4 ). Filled seed production significantly affected the consumption of most alternate food sources (fungi: z = -4.68, p < 0.001, pseudo r 2 = 0.661; berry: z = -4.22, p < 0.001, pseudo r 2 = 0.442). However, lichen consumption was not affected by changes in cone production (z = -1.31, p = 0.054, pseudo r 2 = 0.118). Discussion This study demonstrated that red squirrels, often considered seed specialists, can exhibit a diet predominantly composed of fungi rather than seeds, which to our knowledge has not been found elsewhere. The sporadic and low cone production in our study area, along with a low number of filled seeds within cones, is reflected in this predominantly fungal diet. In interior boreal forest, many studies have recorded extensive white spruce cone production in mast years over thousands of cones per tree (Krebs et al. 2012 , McAdam et al. 2019 , Leeper & LaMontagne 2021 , Krebs et al. 2023 , Wishart 2023 ). While cone production quadrupled in our mast year (2022), the total number of cones per tree rarely exceeded 600 cones per tree, with a maximum of 1288 cones. Given the low tree density and small tree size in our study area compared to interior areas of squirrel habitat (Young et al. 2002 , Sharma & Parton 2007 , Boonstra et al. 2008 ), food availability per hectare remains limited. Although we experienced two years of low crops in our 4-year study period, low cone crops succeeding a mast year are not uncommon due to the immense energy requirements of a mast year (Krebs et al. 2012 ). The single masting event we observed, as well as the 2018 mast year from our study area (Kucheravy et al. 2021 ), did not achieve crop yields to the same extent as other studies (Table 2). Unfavourable environmental and climatic conditions at distributional edges can affect tree growth, resulting low cone production, and a reduction or failure to produce viable seeds. In particular, the dynamics of tree populations at northern latitudinal limits can be linked to the limiting effect of low seasonal temperatures and short growing seasons on pollen and seed production and the subsequent germination and establishment of seeds and seedlings (Henttonen et al. 1986 , Zasada 1988 , Sirois 2000 ). In conifers, while long-term cone production is frequently linked to tree size, for species such as white spruce, climatic limitations in the past and current years rather than endogenous factors often dictate cone crop size (Messaoud et al. 2007 , Krebs et al. 2012 , LaMontagne 2020 ). Seasonal temperature changes in our study region typically exhibit a delay of approximately one month compared to boreal forest regions farther south of the northern treeline (Environment Canada data from Haines Junction Airport, Yukon, and Churchill Airport, Manitoba). Additionally, snow cover can persist for up to two months longer (Scott et al. 1993 ). The variable nature of the climate at the northern boreal forest treelines reduces the likelihood of having consecutively abundant years for spruce cones (Messaoud et al. 2007 ). Additionally, life history theory predicts individuals will prioritize limited energy budgets toward growth, survival, and reproduction to maximize fitness. In the harsh growing conditions presented at the northern treeline, trees may allocate energy toward growth instead of reproduction. In addition to fluctuation in spruce cone production, the quality and quantity of seeds per cone can vary with cone size and year (Waldron 1965 , Zasada 1988 ), potentially making cone numbers a poor predictor of food availability. Without the embryonic tissues, empty seeds have little nutritional value for granivores (Verdú & García-Fayos 1998 ). As the proportion of empty seeds increases within cones, the handling cost of finding highly nutritious seeds also increases (Verdú & García-Fayos 1998 , 2001 ). In our study area, the percentage of filled seeds per cone fell from 8.5% (2021) to 1% in the mast year (2022). Such observations are lower than those observed in other white spruce forests. In northern Ontario, O’Connell (2005) reported that the percentage of filled seed per cone ranged from 26–42% in a non-mast year. Furthermore, in contrast to our observations, Waldron et al. (1965) noted an increase in filled seed in mast years compared to non-mast years, from 12–58% in southern Manitoba. Reduced pollination is commonly associated with empty seeds in conifers (O'Connell et al. 2006 ). In many spruce species, unpollinated ovules can develop into seed coats that contain degenerated gametophytes (Owens 1995 ). Factors such as small stand sizes, low tree density, and environmental stresses such as temperature, water availability, and nutrients, can restrict pollination, increase inbreeding, and raise seed abortion rates (Owens 1995 , O'Connell et al. 2006 , Benjamin et al. 2024 ). For plants residing at the northern limit of their distribution, shorter and cooler growing seasons can result in low seed quality and quantity (Zasada 1992 , Lavoie & Payette 1994 ). Specifically, studies have suggested low seasonal temperatures during spring pollination can lead to seed abortion and cone damage (Zasada 1992 , Owens 1995 ) and inhibit seed maturation (Sirois 2000 ). In white spruce, multi-year reproductive cycles increase susceptibility to such factors affecting seed production. In our study, while the number of filled seeds per cone was low, the total number of filled seeds per tree exceeded those in low cone crop years due to the abundance of cones. However, extracting filled seeds from many empty cones likely increases predator handling costs, influencing seed predator abundance and foraging efficiency (Perea et al. 2013 ). The low seed quality in our study area may further explain why squirrels in this area predominantly consumed fungi. In our study area, fungi appeared from late August to early September with an annual variation typical for boreal forest fungal crops (Luoma et al. 2003 , Krebs et al. 2008 ). Mean dry biomass was comparable with other studies in coniferous forests, including interior boreal forest (2.5 kg ha − 1 ) (Krebs et al. 2008 ) and Douglas-fir forests in the Pacific Northwest (2–5 kg ha − 1 ) (Luoma et al. 2003 ). Fungal reproduction can also experience widespread synchrony (Mehus 1986 ). Our estimates of spruce cone and fungal abundance could be biased by squirrels sampling and caching food prior to our surveys. Despite conducting surveys at consistent times annually, fluctuations in fungal emergence due to rainfall could mean peak abundance was missed. At the northern treeline, red squirrels must constantly adjust to fluctuating seasonal temperatures and food availability. Optimal foraging theory states that species prioritize high-energy foods to maximize fitness (Stephens & Krebs 1986 ), exhibiting specialized diets when preferred resources are abundant and broader diets when resources are scarce. Our results support foraging theory, with a narrower dietary niche breadth following the mast year, although broader than expected if spruce seeds were the dominant food source, likely due to the low quality of seeds at the treeline. In non-mast years, dietary niche breadths generally expanded, with broadest niche observed during the low cone crop in 2021. Fungi were the primary contributor to squirrel diet (~ 70%) across all seasons and years, although other food sources, including berries, were more prominent during the low cone crop year (2021 but reflected in the 2022 diet). Although not a major component, berries increased during low cone years. Like fungi, berries are typically abundant in late summer and early fall and can act as an important source of carbohydrates when spruce seeds are scarce (Stephens et al. 2019 ). Lichen was negligible and showed no variation with cone abundance, consistent with red squirrel’s limited use of this resource (Currah et al. 2000 , Dubay et al. 2008 ). Red squirrels are also known nest predator of understory-nesting birds, such as passerines, in northern coniferous forests (Reitsma et al. 1990 , Sieving & Willson 1998 , Willson et al. 2003 ). However, passerines were excluded from our stable isotopes models because they arrive for the mating season in early June, after the spring moult has occurred. Snowshoe hares were also excluded as a potential food source since predation by red squirrels is rare and mainly involves young hares (< 2 weeks old) in spring and summer (O'Donoghue 1994 ). In more interior northern regions of red squirrel habitat, snowshoe hares typically have three litters (late May, late June - July after the spring molt, and early August before the fall moult) (O'Donoghue 1994 , Oli et al. 2020 ). But delayed spring conditions in our study area shift reproduction into late August, overlapping with fall molt hair growth. However, during August and September, red squirrels are intensely focused on harvesting spruce cones and fungi, reducing their likelihood of consuming alternative food sources. Further, other studies reporting hare consumption by red squirrels have observed squirrels scavenging on hare carcasses in winter (Peers et al. 2020 ), which was not covered by the timeline of hair growth in our study. We observed little seasonal variation in diet during our study period. In early summer, red squirrels will feed on spruce buds and new growth as their winter food caches become depleted, shifting to primary resources, such as spruce seeds and fungi, as they become more abundant in late summer and early fall (Ren et al. 2017 ). As spruce buds were isotopically similar to seeds, we were unable to differentiate between the two food items in spring diet. However, squirrels were observed eating spruce buds in early summer during our study. By late summer and fall, as primary resources become available, squirrels likely consume the same food items they are caching. Throughout their distribution, red squirrels primarily consume conifer seeds and are considered conifer specialists (Smith 1968 , Fletcher et al. 2010 , McAdam et al. 2019 , Wishart 2023 ). In northern interior areas of boreal forest, red squirrels rely exclusively on white spruce seeds (McAdam & Boutin 2003, Boutin et al. 2006 ) with individuals hoarding up to 20,000 cones for winter (Smith 1968 , Hurly & Lourie 1997 ). While spruce cones may not be abundant at the treeline, our results suggest squirrels still prefer conifer seeds, with seed availability influencing the consumption of alternate foods like fungi and berries. Fungi are a known alternate food source for red squirrels, often comprising over half of squirrel’s diet in low cone years (Steele 1998 , Currah et al. 2000 , Koprowski 2005 , Teron & Hutchison 2013 , Derbridge & Koprowski 2019 , Pauli et al. 2019 ). Fungal hoarding by red squirrels fluctuated with cone production, peaking during the low cone crop year (2021) and declining in the mast year (2022). In the northern boreal forest, spruce cone hoarding can be highly variable, and red squirrels will switch from hoarding cones to fungi in years of low production, hanging mushrooms in trees to dry prior to caching (Currah et al. 2000 , Krebs et al. 2008 ). Fletcher et al. ( 2010 ) noted that fungi comprised over 60% of observed caching events when spruce cone production was low, but only 1% of cached items during abundant cone years. The availability and minimal foraging effort needed make fungi a valuable secondary food source for squirrels, and squirrels play a key role in fungal spore dispersal (Krebs et al. 2008 ). Conclusion In summary, the production of white spruce cones varies greatly, and red squirrels have had to adjust their diet according to the availability of primary resources. Compared to similar studies conducted elsewhere in the northern boreal forest (LaMontagne et al. 2005 , LaMontagne 2007 , Lamontagne & Boutin 2007 , Krebs et al. 2012 , Wishart 2023 ), spruce cone production was generally lower at the sub-Arctic treeline. As a consequence, red squirrels living at the sub-Arctic edge of their range rely on fungi as their primary food source, even during mast years. While the caching and consumption of fungi by red squirrels is not a new phenomenon (Currah et al. 2000 , Fletcher et al. 2010 , Pauli et al. 2019 ), fungi have not previously been documented as the primary food source for red squirrels, which emphasizes how marginal treeline is for squirrels. Climate change is altering vegetation patterns in the sub-Arctic and within the treeline, leading to increased tree density and recruitment (Payette et al. 2001 , Mamet & Kershaw 2011 , Mamet & Kershaw 2013 ). In recent decades, changes in masting dynamics have been largely attributed to climate change. In the northern boreal forest, the intervals between mast years have shortened from 5–7 to 2–4 years since 2006 (Krebs et al. 2023 ). As a result, the restrictions placed on squirrels by limited food availability may lessen in the future, increasing the success of future populations and the likelihood of range expansion. However, given the low proportion of filled seed within cones, it is unclear if these shorten intervals between mast years represent increased seed availability. Understanding how factors operate within range boundaries to restrict a species' success and how species conform to these limitations can provide insight into adaptations along latitudinal gradients (Gaston 2009 ) and species’ potential for future range expansion. Declarations Acknowledgements This study was funded by the Natural Sciences and Engineering Research Council of Canada, the Northern Scientific Training Program (NSTP), the University of Manitoba Fieldwork Support Program, and the Churchill Northern Studies Centre (CNSC) Northern Research Fund. We thank the CNSC for their logistical support and all our field assistants for their invaluable assistance in data collection. We are also thankful to Drs. Collin Garroway and Jane Waterman for their advice and loan of equipment. Funding This study was funded by the Natural Sciences and Engineering Council of Canada, the NSTP, the University of Manitoba Fieldwork Support Program, the Oakes Riewe Environmental Studies research award from the University of Manitoba, and the Northern Research Fund from the CNSC. Conflicts of interest/competing interests The authors declare they have no conflict of interest. Ethics approval All data and samples collected were in accordance with the ethical standards of the University of Manitoba (animal care protocol F20-010) and approved by Government of Manitoba Wildlife and Fisheries branch (Permit no: WB2433). Consent to participate Not applicable Consent for publication Availability of data and material The datasets used for this study are available from the corresponding author upon request. Code availability The code used for this study is available from the corresponding author upon request. References Archibald, D. W., Q. E. Fletcher, S. Boutin, A. G. McAdam, J. R. Speakman, and M. M. Humphries. 2013. Sex-specific hoarding behavior in North American red squirrels ( Tamiasciurus hudsonicus ). Journal of Mammalogy 94:761-770. Archibald, D. W., A. G. McAdam, S. Boutin, Q. E. Fletcher, and M. M. Humphries. 2012. Within-season synchrony of a masting conifer enhances seed escape. The American Naturalist 179:536-544. Bayne, E. M. and K. A. Hobson. 2002. Effects of red squirrel ( Tamiasciurus hudsonicus ) removal on survival of artificial songbird nests in boreal forest fragments. The American Midland Naturalist 147:72-79. Bearhop, S., C. E. Adams, S. Waldron, R. A. Fuller, and H. Macleod. 2004. Determining trophic niche width: a novel approach using stable isotope analysis: Stable isotopes as measures of niche width. Journal of Animal Ecology 73:1007-1012. Benhamou, S. 1996. Space use and foraging movements in the American red squirrel (T amiasciurus hudsonicus ). Behavioural processes 37:89-102. Benjamin, J. S., J. D. Roth, and J. H. Markham. 2024. Red foxes increase white spruce seed production at its northern range limit. Basic and Applied Ecology. Boonstra, R., L. Desantis, C. J. Krebs, and D. S. Hik. 2008. Climate and nutrient influences on the growth of white spruce trees in the boreal forests of the Yukon. Climate Research 36:123-130. Boutin, S., L. A. Wauters, A. G. McAdam, M. M. Humphries, G. Tosi, and A. A. Dhondt. 2006. Anticipatory reproduction and population growth in seed predators. Science 314:1928-1930. Brown, J. H., D. W. Mehlman, and G. C. Stevens. 1995. Spatial variation in abundance. Ecology 76:2028-2043. Callaghan, T. V., et al. 2004. Biodiversity, distributions and adaptations of Arctic species in the context of environmental change. AMBIO: A Journal of the Human Environment 33:404-417. Corkery, C. A., E. Nol, and L. Mckinnon. 2019. No effects of asynchrony between hatching and peak food availability on chick growth in Semipalmated Plovers ( Charadrius semipalmatus ) near Churchill, Manitoba. Polar Biology 42:593-601. Currah, R., E. Smreciu, T. Lehesvirta, M. Niemi, and K. Larsen. 2000. Fungi in the winter diets of northern flying squirrels and red squirrels in the boreal mixedwood forest of northeastern Alberta. Canadian Journal of Botany 78:1514-1520. Dehling, D. M., et al. 2021. Specialists and generalists fulfil important and complementary functional roles in ecological processes. Functional Ecology 35:1810-1821. Derbridge, J. J. and J. L. Koprowski. 2019. Experimental removals reveal dietary niche partitioning facilitates coexistence between native and introduced species. Ecology and Evolution 9:4065-4077. Dubay, S., G. Hayward, and C. Martinez del Rio. 2008. Nutritional value and diet preference of arboreal lichens and hypogeous fungi for small mammals in the Rocky Mountains. Canadian Journal of Zoology 86:851-862. Ferguson, S. H. and P. C. Elkie. 2004. Seasonal movement patterns of woodland caribou ( Rangifer tarandus caribou ). Journal of zoology 262:125-134. Fletcher, Q. E., et al. 2010. The functional response of a hoarding seed predator to mast seeding. Ecology 91:2673-2683. Gaston, K. J. 2003. The structure and dynamics of geographic ranges. Oxford University Press on Demand. Gaston, K. J. 2009. Geographic range limits: achieving synthesis. Proceedings of the Royal Society B: Biological Sciences 276:1395-1406. Gelman, A., J. Hwang, and A. Vehtari. 2014. Understanding predictive information criteria for Bayesian models. Statistics and computing 24:997-1016. Geweke, J. 1991. Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments. Federal Reserve Bank of Minneapolis. Gross, S. J. and T. D. Price. 2000. Determinants of the northern and southern range limits of a warbler. Journal of Biogeography 27:869-878. Haines, J. A., et al. 2022. Sex-specific effects of capital resources on reproductive timing and success in red squirrels. Behavioral Ecology and Sociobiology 76:142. Harper, K. A., et al. 2011. Tree spatial pattern within the forest–tundra ecotone: a comparison of sites across Canada. Canadian Journal of Forest Research 41:479-489. Harper, K. A., A. A. Lavallee, and P. Dodonov. 2018. Patterns of shrub abundance and relationships with other plant types within the forest–tundra ecotone in northern Canada. Arctic Science 4:691-709. Henttonen, H., M. Kanninen, M. Nygren, and R. Ojansuu. 1986. The maturation of Pinus sylvestris seeds in relation to temperature climate in northern Finland. Scandinavian Journal of Forest Research 1:243-249. Ho, R. H. 1984. Seed-cone receptivity and seed production potential in white spruce. Forest ecology and management 9:161-171. Hobbie, E. A., et al. 2017. Stable isotopes and radiocarbon assess variable importance of plants and fungi in diets of Arctic ground squirrels. Arctic, Antarctic, and Alpine Research 49:487-500. Humphries, M. M. and S. Boutin. 2000. The determinants of optimal litter size in free‐ranging red squirrels. Ecology 81:2867-2877. Humphries, M. M., et al. 2005. Expenditure freeze: the metabolic response of small mammals to cold environments. Ecology Letters 8:1326-1333. Hurly, T. A. and S. A. Lourie. 1997. Scatterhoarding and larderhoarding by red squirrels: size, dispersion, and allocation of hoards. Journal of Mammalogy 78:529-537. Jackson, A. L., R. Inger, A. C. Parnell, and S. Bearhop. 2011. Comparing isotopic niche widths among and within communities: SIBER - Stable Isotope Bayesian Ellipses in R: Bayesian isotopic niche metrics. Journal of Animal Ecology 80:595-602. Kayes, I. and A. Mallik. 2020. Boreal forests: distributions, biodiversity, and management. Pp. 1-12, Springer International Publishing Cham. Kelly, D. and V. L. Sork. 2002. Mast seeding in perennial plants: why, how, where? Annual review of ecology and systematics 33:427-447. Koprowski, J. L. 2005. The response of tree squirrels to fragmentation: a review and synthesis. Animal Conservation 8:369-376. Kranowski, P. V. 1969. Aspects of red squirrel ( Tamiasciurus hudsonicus ) population ecology in winter in interior Alaska. University of Alaska Fairbanks. Krebs, C., P. Carrier, S. Boutin, R. Boonstra, and E. Hofer. 2008. Mushroom crops in relation to weather in the southwestern Yukon. Botany 86:1497-1502. Krebs, C., J. LaMontagne, A. Kenney, and S. Boutin. 2012. Climatic determinants of white spruce cone crops in the boreal forest of southwestern Yukon. Botany 90:113-119. Krebs, C. J., et al. 2023. Long-term monitoring in the boreal forest reveals high spatio-temporal variability among primary ecosystem constituents. Front Ecol Evol 11:1187222. Kucheravy, C. E., J. D. Roth, and J. H. Markham. 2021. Red foxes increase reproductive output of white spruce in a non-mast year. Basic and Applied Ecology 51:11-19. Lafleur, P. M. 1999. Growing season energy and CO 2 exchange at a subarctic boreal woodland. Journal of Geophysical Research: Atmospheres 104:9571-9580. LaMontagne, J. M. 2007. Spatial and temporal variability in white spruce ( Picea glauca ) cone production: individual and population responses of North American red squirrels ( Tamiasciurus hudsonicus ). Doctor of Philosophy, University of Alberta Edmonton, Canada. LaMontagne, J. M. 2020. Terrestrial ecology: natural selection for mast seeding. Current Biology 30:996-998. Lamontagne, J. M. and S. Boutin. 2007. Local‐scale synchrony and variability in mast seed production patterns of Picea glauca . Journal of Ecology 95:991-1000. LaMontagne, J. M., S. Peters, and S. Boutin. 2005. A visual index for estimating cone production for individual white spruce trees. Canadian Journal of Forest Research 35:3020-3026. LaMontagne, J. M., C. T. Williams, J. L. Donald, M. M. Humphries, A. G. McAdam, and S. Boutin. 2013. Linking intraspecific variation in territory size, cone supply, and survival of North American red squirrels. Journal of Mammalogy 94:1048-1058. Lane, J., et al. 2015. Post‐weaning parental care increases fitness but is not heritable in North American red squirrels. Journal of Evolutionary Biology 28:1203-1212. Lavoie, C. and S. Payette. 1994. Recent fluctuations of the lichen-spruce forest limit in subarctic Quebec. Journal of Ecology:725-734. Layne, J. N. 1954. The biology of the red squirrel, Tamiasciurus hudsonicus loquax (Bangs), in central New York. Ecological Monographs 24:228-267. Leeper, A. C. and J. M. LaMontagne. 2021. Cone characteristics and insect predation levels vary across years in mast seeding white spruce. Canadian Journal of Forest Research 51:1550-1557. Lepage, P. and G. Parker. 1988. Copper, nickel, and iron levels in pelage of red squirrels living near the ore smelters at Sudbury, Ontario, Canada. Canadian journal of zoology 66:1631-1637. Luoma, D. L., J. M. Trappe, A. W. Claridge, K. M. Jacobs, and E. Cazares. 2003. Relationships among fungi and small mammals in forested ecosystems. Mammal Community Dynamics in Western Coniferous Forests: Management and Conservation Cambridge University Press, Cambridge, United Kingdom:343-373. Lynch, H. J., M. Rhainds, J. M. Calabrese, S. Cantrell, C. Cosner, and W. F. Fagan. 2014. How climate extremes—not means—define a species' geographic range boundary via a demographic tipping point. Ecological Monographs 84:131-149. Mamet, S. D. and G. P. Kershaw. 2011. Radial-growth response of forest-tundra trees to climate in the Western Hudson Bay Lowlands. ARCTIC 64:446-458. Mamet, S. D. and G. P. Kershaw. 2013. Multi-scale analysis of environmental conditions and conifer seedling distribution across the treeline ecotone of northern Manitoba, Canada. Ecosystems 16:295-309. McAdam, A. G. and S. Boutin. 2003. Effects of food abundance on genetic and maternal variation in the growth rate of juvenile red squirrels. Journal of evolutionary biology 16:1249-1256. McAdam, A. G., S. Boutin, B. Dantzer, and J. E. Lane. 2019. Seed masting causes fluctuations in optimum litter size and lag load in a seed predator. The American Naturalist 194:574-589. McAulay, J., P. J. Seddon, D. J. Wilson, and J. M. Monks. 2020. Stable isotope analysis reveals variable diets of stoats ( Mustela erminea ) in the alpine zone of New Zealand. New Zealand Journal of Ecology 44:1-13. Mehus, H. 1986. Fruit body production of macrofungi in some North Norwegian forest types. Nordic Journal of Botany 6:679-702. Messaoud, Y., Y. Bergeron, and H. Asselin. 2007. Reproductive potential of balsam fir ( Abies balsamea ), white spruce ( Picea glauca ), and black spruce ( P. mariana ) at the ecotone between mixedwood and coniferous forests in the boreal zone of western Quebec. American Journal of Botany 94:746-754. Nelson, B. A. 1945. The spring molt of the northern red squirrel in Minnesota. Journal of Mammalogy 26:397-400. Newbury, R. K. and K. E. Hodges. 2018. Regional differences in winter diets of bobcats in their northern range. Ecology and evolution 8:11100-11110. Nygren, M., K. Rissanen, K. Eerikäinen, T. Saksa, and S. Valkonen. 2017. Norway spruce cone crops in uneven-aged stands in southern Finland: a case study. Forest Ecology and Management 390:68-72. O'Connell, L., A. Mosseler, and O. Rajora. 2006. Impacts of forest fragmentation on the mating system and genetic diversity of white spruce ( Picea glauca ) at the landscape level. Heredity 97:418-426. O'Donoghue, M. 1994. Early survival of juvenile snowshoe hares. Ecology 75:1582-1592. Oli, M. K., C. J. Krebs, A. J. Kenney, R. Boonstra, S. Boutin, and J. E. Hines. 2020. Demography of snowshoe hare population cycles. Ecology 101:e02969. Ostfeld, R. S. and F. Keesing. 2000. Pulsed resources and community dynamics of consumers in terrestrial ecosystems. Trends in Ecology and Evolution 15:232-237. Owens, J. 1995. Constraints to seed production: temperate and tropical forest trees. Tree Physiology 15:477-484. Pagani‐Núñez, E., C. Barnett, H. Gu, and E. Goodale. 2016. The need for new categorizations of dietary specialism incorporating spatio‐temporal variability of individual diet specialization. Journal of Zoology 300:1-7. Parnell, A. C., et al. 2013. Bayesian stable isotope mixing models. Environmetrics 24:387-399. Pauli, J. N., et al. 2019. Quantifying niche partitioning and multichannel feeding among tree squirrels. Food Webs 21:e00124. Payette, S., M.-J. Fortin, and I. Gamache. 2001. The subarctic forest–tundra: the structure of a biome in a changing climate: the shifting of local subarctic tree lines throughout the forest–tundra biome, which is linked to ecological processes at different spatiotemporal scales, will reflect future global changes in climate. BioScience 51:709-718. Peers, M. J., et al. 2020. Prey availability and ambient temperature influence carrion persistence in the boreal forest. Journal of Animal Ecology 89:2156-2167. Perea, R., M. Venturas, and L. Gil. 2013. Empty seeds are not always bad: simultaneous effect of seed emptiness and masting on animal seed predation. Plos One 8:e65573. Phillips, D. L. 2012. Converting isotope values to diet composition: the use of mixing models. Journal of Mammalogy 93:342-352. Réale, D., D. Berteaux, A. McAdam, and S. Boutin. 2003. Lifetime selection on heritable life‐history traits in a natural population of red squirrels. Evolution 57:2416-2423. Rehm, E. M., P. Olivas, J. Stroud, and K. J. Feeley. 2015. Losing your edge: climate change and the conservation value of range‐edge populations. Ecology and Evolution 5:4315-4326. Reitsma, L. R., R. T. Holmes, and T. W. Sherry. 1990. Effects of removal of red squirrels, Tamiasciurus hudsonicus , and eastern chipmunks, Tamias striatus , on nest predation in a northern hardwood forest: an artificial nest experiment. Oikos:375-380. Ren, T., et al. 2017. Seasonal, spatial, and maternal effects on gut microbiome in wild red squirrels. Microbiome 5:1-14. Rettie, W. J. and F. Messier. 2000. Hierarchical habitat selection by woodland caribou: its relationship to limiting factors. Ecography 23:466-478. Scott, P. A., R. I. Hansell, and W. R. Erickson. 1993. Influences of wind and snow on northern tree-line environments at Churchill, Manitoba, Canada. Arctic:316-323. Sexton, J. P., P. J. McIntyre, A. L. Angert, and K. J. Rice. 2009. Evolution and Ecology of Species Range Limits. Annual Review of Ecology, Evolution, and Systematics 40:415-436. Sharma, M. and J. Parton. 2007. Height–diameter equations for boreal tree species in Ontario using a mixed-effects modeling approach. Forest Ecology and Management 249:187-198. Shipley, L. A., J. S. Forbey, and B. D. Moore. 2009. Revisiting the dietary niche: when is a mammalian herbivore a specialist? Integrative and comparative biology 49:274-290. Sieving, K. E. and M. F. Willson. 1998. Nest predation and avian species diversity in northwestern forest understory. Ecology 79:2391-2402. Sirén, A. P. K. and T. L. Morelli. 2020. Interactive range‐limit theory (iRLT): An extension for predicting range shifts. Journal of Animal Ecology 89:1-15. Sirois, L. 2000. Spatiotemporal variation in black spruce cone and seed crops along a boreal forest-tree line transect. Canadian Journal of Forest Research 30:900-909. Smith, M. C. 1968. Red squirrel responses to spruce cone failure in interior Alaska. The Journal of Wildlife Management:305-317. Steele, M. A. 1998. Tamiasciurus hudsonicus . Mammalian Species:1-9. Stephens, D. W. and J. R. Krebs. 1986. Foraging theory. Princeton University Press. Stephens, R. B., E. A. Hobbie, T. D. Lee, and R. J. Rowe. 2019. Pulsed resource availability changes dietary niche breadth and partitioning between generalist rodent consumers. Ecology and Evolution 9:10681-10693. Steury, T. D. and D. L. Murray. 2003. Causes and consequences of individual variation in territory size in the American red squirrel. Oikos 101:147-156. Sullivan, T. P. and D. S. Sullivan. 1982. Influence of fertilization on feeding attacks to lodgepole pine by snowshoe hares and red squirrels. The Forestry Chronicle 58:263-266. Szumski, C. M., J. D. Roth, and D. L. Murray. 2023. Canada lynx foraging strategies: Facultative specialists become obligate generalists toward the distribution edge. Ecosphere 14:e4629. Teron, J. N. and L. J. Hutchison. 2013. Consumption of truffles and other fungi by the American red squirrel ( Tamiasciurus hudsonicus ) and the eastern chipmunk ( Tamias striatus )( Sciuridae ) in northwestern Ontario. The Canadian Field-Naturalist 127:57-59. van der Veen, B., J. Mattisson, B. Zimmermann, J. Odden, and J. Persson. 2020. Refrigeration or anti-theft? Food-caching behavior of wolverines ( Gulo gulo ) in Scandinavia. Behavioral Ecology and Sociobiology 74:1-13. Verdú, M. and P. García-Fayos. 1998. Ecological causes, function, and evolution of abortion and parthenocarpy in Pistacia lentiscus (Anacardiaceae) . Canadian Journal of Botany 76:134-141. Verdú, M. and P. García-Fayos. 2001. The effect of deceptive fruits on predispersal seed predation by birds in Pistacia lentiscus. Plant Ecology 156:245-248. Waldron, R. 1965. Cone production and seedfall in a mature white spruce stand. The Forestry Chronicle 41:316-329. Willson, M. F., T. L. D. Santo, and K. E. Sieving. 2003. Red squirrels and predation risk to bird nests in northern forests. Canadian Journal of Zoology 81:1202-1208. Wishart, A. E. 2023. Variation in resource acquisition in a food-caching mammal, the North American red squirrel ( Tamiasciurus hudsonicus ). Doctor of Philosophy, University of Saskatchewan. Yang, L. H., J. L. Bastow, K. O. Spence, and A. N. Wright. 2008. What can we learn from resource pulses. Ecology 89:621-634. Yang, L. H., K. F. Edwards, J. E. Byrnes, J. L. Bastow, A. N. Wright, and K. O. Spence. 2010. A meta‐analysis of resource pulse–consumer interactions. Ecological Monographs 80:125-151. Young, P. J., V. L. Greer, and S. K. Six. 2002. Characteristics of Bolus Nests of Red Squirrels in the Pinaleno and White Mountains of Arizona. The Southwestern Naturalist 47:267-275. Zasada, J. 1992. The reproductive process in boreal forest trees. A systems analysis of the global boreal forest:211-233. Zasada, J. C. 1988. Embryo growth in Alaskan white spruce seeds. Canadian Journal of Forest Research 18:64-67. Tables Table 1: Comparison of white spruce cone counts and the percentage of seeds that were filled seeds within spruce cones near Churchill, MB, Canada, from interior forests within red squirrel distribution. Location Total cones/tree Years Reference Mast year Treeline 74 – 863 2022 this study Treeline 0 – 407 2018 Kucheravy et al. 2021 Yukon 1,000 – 2,500 1993 – 2010 Krebs et al. 2012 Yukon 750 – 3,000 2005 – 2022 Krebs et al. 2023 Yukon 1,000 – 3,500 2008 – 2022 Wishart 2023 Michigan 2,400 – 2,700 2012 – 2017 Leeper & LaMontagne 2021 Non-mast year Treeline 0 – 123 2020 - 2023 this study Treeline 0 – 93 2019 Kucheravy et al. 2021 Yukon 0 – 200 1993 – 2010 Krebs et al. 2012 Michigan 216 2012 – 2017 Leeper & LaMontagne 2021 Wisconsin 6 – 186 2012 – 2014 Corona et al. 2022 Filled seeds/cone (%) Mast year Treeline 1.2 ± 0.1 2022 this study Southern Manitoba 58 1954 – 1963 Waldron 1965 Alaska 60.5 1957 – 1959 Zasada & Gregory 1969 Non-mast year Treeline 6.4 ± 1.4 2021 – 2023 this study Southern Manitoba 12 – 48 1954 – 1963 Waldron 1965 Northern Quebec 66 2001 – 2004 Messaoud et al. 2007 Northern Ontario 26 – 42 1994 O’Connell 2006 Alaska 22 – 62 1954 – 1963 Zasada & Gregory 1969 Additional Declarations The authors declare no competing interests. Supplementary Files OecologiaelectronicsupplementaryWindsor.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-6624068","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":454024732,"identity":"1e2c3800-c85f-4f46-a37f-0176ca1f3f04","order_by":0,"name":"Alexandra Windsor","email":"data:image/png;base64,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","orcid":"","institution":"University of Manitoba","correspondingAuthor":true,"prefix":"","firstName":"Alexandra","middleName":"","lastName":"Windsor","suffix":""},{"id":454024733,"identity":"745ca69a-2612-4c96-9f8c-0eb955769ca1","order_by":1,"name":"John Markham","email":"","orcid":"","institution":"University of Manitoba","correspondingAuthor":false,"prefix":"","firstName":"John","middleName":"","lastName":"Markham","suffix":""},{"id":454024734,"identity":"a1df4e3c-2e8a-49ba-9601-ab25f0169a30","order_by":2,"name":"James Roth","email":"","orcid":"","institution":"University of Manitoba","correspondingAuthor":false,"prefix":"","firstName":"James","middleName":"","lastName":"Roth","suffix":""}],"badges":[],"createdAt":"2025-05-09 01:27:52","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-6624068/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6624068/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":83300172,"identity":"787f9fad-73de-4f31-b231-43009782bff6","added_by":"auto","created_at":"2025-05-22 14:53:02","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":170148,"visible":true,"origin":"","legend":"\u003cp\u003eA map of the study area (outlined in red) located near Churchill, MB, Canada. Dashed lines delineate separate habitat types (Corkery et al. 2019).\u003c/p\u003e","description":"","filename":"image.png","url":"https://assets-eu.researchsquare.com/files/rs-6624068/v1/5763e0ea9a7a80a4850a9dd7.png"},{"id":83299863,"identity":"e19f4f73-4512-4a23-9b54-6ce79213d413","added_by":"auto","created_at":"2025-05-22 14:45:02","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":83536,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of seasonal dietary niches from 2020 – 2023 and the associated standard ellipse areas (SEA). Bayesian ellipses are generated using the SIBER package in R. The dotted lines indicate convex hulls and solid lines indicate standard ellipse areas. The black dots in the SEA represent the median, the red ‘x’ the mode, and the boxes indicate the 50, 75, and 95% credible intervals.\u003c/p\u003e","description":"","filename":"image.png","url":"https://assets-eu.researchsquare.com/files/rs-6624068/v1/4d7e81cfd71d587de3a9b15d.png"},{"id":83301051,"identity":"091e55ff-7e34-4212-90fc-779834cab0aa","added_by":"auto","created_at":"2025-05-22 15:01:02","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":214993,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of A) spring and B) fall dietary niches from 2020 – 2023 and the associated standard ellipse areas (SEA). Bayesian ellipses were generated using the SIBER package in R. The dotted lines indicate convex hulls and solid lines indicate standard ellipse areas. The black dots in the SEA represent the median, the red ‘x’ the mode, and the boxes indicate the 50, 75, and 95% credible intervals.\u003c/p\u003e","description":"","filename":"Screenshot20250508at7.33.32PM.png","url":"https://assets-eu.researchsquare.com/files/rs-6624068/v1/b1c1daa9bfb36b5d4ad5ec0b.png"},{"id":83299862,"identity":"c1b87e7c-2483-4a4f-80f6-1ce8a0d41ab3","added_by":"auto","created_at":"2025-05-22 14:45:02","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":44205,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of the annual contributions of five food sources to spring squirrel diet from 2020 – 2023 (mean \u003cu\u003e+\u003c/u\u003eSD). Proportions were estimated from Bayesian mixing models using the MixSIAR package in R. Note: 2022 diet (blue) reflects the low cone crops in 2021, and 2023 diet (purple) reflects the 2022 mast year.\u003c/p\u003e","description":"","filename":"image.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6624068/v1/d47e2d91eb3630581007a87c.jpeg"},{"id":83302005,"identity":"aba1355a-2d91-4d5e-bce2-c03d60628c3e","added_by":"auto","created_at":"2025-05-22 15:17:03","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1120157,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6624068/v1/0faed02c-c25f-4375-8566-ff9fafd67574.pdf"},{"id":83300173,"identity":"57f3ff72-1c68-4d7b-9566-5b12d6cef60e","added_by":"auto","created_at":"2025-05-22 14:53:02","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":274641,"visible":true,"origin":"","legend":"","description":"","filename":"OecologiaelectronicsupplementaryWindsor.docx","url":"https://assets-eu.researchsquare.com/files/rs-6624068/v1/fbb54898ec579e7b59f132d7.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003ePresumed seed specialists rely on fungi as their primary food source at the sub-Arctic treeline\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eSpecies with extensive geographical distributions often have smaller, fragmented populations near their range limits, where habitats are harsher than at the core of their distributions (Gaston \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, Lynch et al. \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Range boundaries can be shaped by physical barriers directly preventing species\u0026rsquo; dispersal (Gross \u0026amp; Price \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2000\u003c/span\u003e) or gradients in climatic variables, which create physiological limits. In the northern hemisphere, it is commonly thought that northern range boundaries are frequently governed by abiotic factors like extreme climate, while biotic factors, such as competition, and food availability, impose additional constraints at southern range limits (Brown et al. \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e1995\u003c/span\u003e, Sir\u0026eacute;n \u0026amp; Morelli \u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). However, combinations of factors likely work in concert, limiting the expansion of species beyond their range. Compared to core environments, range boundary habitats often exhibit greater climatic variability (Rehm et al. \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), which imposes physiological limits and impact population growth (Sexton et al. \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). Although range boundary habitats can be expansive, they are often low quality or exist as fragmented patches of high-quality habitat. Climatic factors and the corresponding influences on habitat suitability, population density, and reproductive fitness strongly reinforce range limits and can be further exacerbated by inter and intraspecific interactions and resource availability. At northern range boundaries, extreme climates can reduce the availability and accessibility of food sources, particularly during winter. The variation in seasonal climates at higher latitudes dictates large annual and seasonal fluctuations in resource availability (Humphries et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), as shorter growing seasons and cooler seasonal temperatures restrict primary and secondary productivity (Gross \u0026amp; Price \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, Callaghan et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Species respond to fluctuations in food sources in multiple ways. Some species, such as the boreal woodland caribou (\u003cem\u003eRangifer tarandus caribou\u003c/em\u003e), migrate seasonally following food sources (Rettie \u0026amp; Messier \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, Ferguson \u0026amp; Elkie \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Other species, such as wolverines (\u003cem\u003eGulo gulo)\u003c/em\u003e, may change activity patterns or cache food to buffer against fluctuating food availability in an overall low-productivity environment (van der Veen et al. \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). When food is spatially and temporally variable, species may rely on pulses of food sources.\u003c/p\u003e \u003cp\u003ePulsed resources are episodic events of superabundant resource production with long inter-pulse periods of normal or subnormal production. While resource pulses are short-lived, they drive the dynamics of many plant and animal communities. The bottom-up effect of pulsed resources can be seen at the level of the individual, the population, and indirectly at the community level (Yang et al. \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2008\u003c/span\u003e, Yang et al. \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). At the individual level, consumer diets shift in response to pulsed resources, which induces numerical responses in consumer populations. These responses then propagate across trophic levels as increased densities in consumers become secondary resource pulses for higher trophic levels (Yang et al. \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe boreal forest, one of the world\u0026rsquo;s largest terrestrial biomes, represents the northern range limit for many species (Kayes \u0026amp; Mallik \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2020\u003c/span\u003e, Krebs et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Conifers in the boreal forest experience periodic masting, with years of high seed production followed by long intervals of low or absent seed production (Kelly \u0026amp; Sork \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2002\u003c/span\u003e, Lamontagne \u0026amp; Boutin \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Archibald et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Mast years are often synchronous across large geographical areas (Lamontagne \u0026amp; Boutin \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Krebs et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) and overwhelm seed predators, enhancing seed survival and establishment (Kelly \u0026amp; Sork \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). White spruce (\u003cem\u003ePicea glauca\u003c/em\u003e) is the dominant tree species in dry habitats at the north limit of the boreal forest (Bonan and Shugart 1989) and exhibits periodic masting (Archibald et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2012\u003c/span\u003e, Krebs et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e, Krebs et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) approximately every 3\u0026ndash;5 years (Lamontagne \u0026amp; Boutin \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Many small mammals rely heavily on spruce seeds, but may switch to alternative food sources, such as fungi, which can be highly abundant but nutrient-poor, during non-mast years (Ostfeld \u0026amp; Keesing \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, Fletcher et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe dietary strategies of species can be defined based on the breadth of their dietary niches. Facultative generalists typically have a very broad dietary niche compared to facultative specialists that exhibit a narrower dietary niche (Shipley et al. \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2009\u003c/span\u003e, Pagani-N\u0026uacute;\u0026ntilde;ez et al. \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). However, both groups exhibit behavioural plasticity. Facultative generalists optimize resources by switching to more abundant or accessible food sources, while facultative specialists can expand their diet when primary resources are scarce, enabling both groups to exploit fluctuating resources and maintain broad geographic distributions, making them more successful in habitats where food availability is low and heterogeneous (Shipley et al. \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2009\u003c/span\u003e, Newbury \u0026amp; Hodges \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2018\u003c/span\u003e, Szumski et al. \u003cspan citationid=\"CR94\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In contrast, obligate generalists, though able to use diverse food sources, have limited specialization and may be less efficient in optimizing their diet in such environments, and obligate specialists are vulnerable due to their reliance on specific resources, which could be scarce (Shipley et al. \u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e2009\u003c/span\u003e, Dehling et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Due to the low predictability of food abundance, generalist species are common at northern range boundaries (Callaghan et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Facultative specialists living in these habitats may exhibit behaviours more like facultative generalists when primary food sources are unpredictable and fluctuate dramatically. Individuals may also employ behavioural methods such as food hoarding and increasing litter and clutch sizes to take advantage of years of superabundant food (Callaghan et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2004\u003c/span\u003e, Boutin et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2006\u003c/span\u003e, McAdam et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAmerican red squirrels (\u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e) are primarily granivorous, specialising in conifer seeds across most of their distribution (Steele \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e1998\u003c/span\u003e, Boutin et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). In the northern parts of their range, red squirrels appear to feed almost exclusively on white spruce seeds (McAdam \u0026amp; Boutin 2003). While seeds constitute the majority of squirrel diet, red squirrels can opportunistically consume a large variety of foods including fungi (Layne \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e1954\u003c/span\u003e, Steele \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e1998\u003c/span\u003e, Fletcher et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), berries (Benhamou \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e1996\u003c/span\u003e, Currah et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2000\u003c/span\u003e), and animal prey, including songbird eggs (Bayne \u0026amp; Hobson \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2002\u003c/span\u003e, Willson et al. \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2003\u003c/span\u003e) and young lagomorphs (Layne \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e1954\u003c/span\u003e, Sullivan \u0026amp; Sullivan \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e1982\u003c/span\u003e, Peers et al. \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). These alternative food sources, however, typically make up a small proportion of annual diet (Layne \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e1954\u003c/span\u003e), and are relied on only when spruce cone availability is low. Despite observations of consumption of alternative food sources, detailed quantifications of non-seed food sources are often lacking (Currah et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, Fletcher et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). Many life history parameters of red squirrels, including reproduction (Boutin et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2006\u003c/span\u003e, Lane et al. \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), growth (McAdam \u0026amp; Boutin 2003), and overwinter survival (Steury \u0026amp; Murray \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, LaMontagne et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), are intimately linked to spruce cone production (Humphries \u0026amp; Boutin \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, McAdam \u0026amp; Boutin 2003, R\u0026eacute;ale et al. \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, Boutin et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). Assuming seed productivity is consistent, seed availability can be estimated through spruce cone availability by counting the number of cones produced within the study area (Lamontagne \u0026amp; Boutin \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Kucheravy et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). However, the quality and quantity of seeds produced each year can also vary (Waldron \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e1965\u003c/span\u003e, Zasada \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e1988\u003c/span\u003e), further influencing overall food availability.\u003c/p\u003e \u003cp\u003eWe reconstructed the diet of red squirrels living at their sub-Arctic range boundary using stable isotope analysis and explored how spruce cone production affected squirrel diet. We hypothesised that squirrels would adopt a highly supplemented diet in low cone years and predicted that 1) overall cone production would be low, 2) squirrels would frequently consume alternate food sources, depending on the yearly abundance of cones, and 3) their dietary niche would be broad. We further predicted that 4) fall diet would be broader than spring diet, as alternate food sources such as fungi and berries are more abundant in fall.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy area\u003c/h2\u003e \u003cp\u003eOur study area near Churchill, MB (58\u0026deg;45\u0026rsquo; N, 94\u0026deg;04\u0026rsquo; W) on the western edge of Hudson Bay, is on the transition between boreal forest and Arctic tundra. The forest is primarily composed of white spruce, black spruce (\u003cem\u003ePicea mariana\u003c/em\u003e), and tamarack (\u003cem\u003eLarix laricina\u003c/em\u003e) (Mamet \u0026amp; Kershaw \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2011\u003c/span\u003e, Mamet \u0026amp; Kershaw \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), and is heavily fragmented by patches of wetlands and open tundra (Harper et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2011\u003c/span\u003e, Harper et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The forest has a low density of mature trees (~\u0026thinsp;600 stems ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) and a sparse canopy of ca. 25% (Lafleur \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). The proximity of the sub-Arctic treeline to Hudson Bay, which is frozen eight months of the year, exposes the region to cool summer temperatures and extreme winters (Mamet \u0026amp; Kershaw \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). In our study area, mean monthly temperatures ranged from \u0026minus;\u0026thinsp;28.4 to -15.1 ⁰C in winter (November-April) and 1.2 ⁰C to 20.6 ⁰C in summer (June-August) from 2019\u0026ndash;2023 (Environment Canada).\u003c/p\u003e \u003cp\u003eFieldwork was conducted in boreal woodland consisting of patchy forest, transitioning to a denser forest and ending in an open-canopy forest near a large fen (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). We established sites at 250 m intervals along a 7 km transect running north to south, looking for signs of squirrel activity such as vocalisations, cone debris, and middens. Middens are large food caches containing hundreds to tens of thousands of cones (Haines et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2022\u003c/span\u003e, Wishart \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), and are easily identified by the concentrated accumulation of cone debris, often reaching surfaces areas up to a few hundred m\u003csup\u003e2\u003c/sup\u003e (Steury \u0026amp; Murray \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Very few true middens, characterized by large fields of cone debris, were observed in our study area. Here, we define a \u0026lsquo;midden\u0026rsquo; as any form of food cache, including true middens and burrows with cone debris that have the potential to contain cones. To account for juvenile dispersal and the possibility of squirrels taking over previously unoccupied sites, we checked all sites for signs of squirrel activity each year in early June.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eFood availability\u003c/h3\u003e\n\u003cp\u003eTo assess annual spruce cone production, we conducted cone counts of individually marked trees each August (five trees per site; 9 sites in 2020, 26 sites in 2021\u0026ndash;2023). Each tree was at least 5 m from the nearest sampled tree and had a minimum of 5 cm diameter at breast height (DBH), representing the approximate minimum age at which white spruce trees can bear cones (Lamontagne \u0026amp; Boutin \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). At squirrel-occupied sites, we sampled the five closest trees to the midden. At unoccupied sites, we selected the five closest trees from a flagged reference point, following a bearing determined using a random number generator. Each August, we took 3\u0026ndash;5 photographs of each tree on opposing sides. Using ImageJ image processing software, we counted the cones visible in each picture (Nygren et al. \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). The photographs for each tree were matched to avoid double-counting cones. We calculated food availability at each site as the mean number of cones per tree (Kucheravy et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), multiplied by tree density to estimate cones per hectare. To calculate tree density at each site, we counted the number of white spruce trees (\u0026gt;\u0026thinsp;5 cm DBH) within four 15 m radius plots (two on each side of the flagged reference point) spaced 20 m apart. To estimate seed production, we used data from the control sites, which encompassed our sampling area, from Benjamin et al. (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). We collected cones from the upper part of the crown using a pole pruner. The number of cones collected per tree depended on their availability but averaged 13\u0026thinsp;\u003cspan type=\"Underline\" class=\"Underline\" name=\"Emphasis\"\u003e\u0026plusmn;\u003c/span\u003e\u0026thinsp;1 cones. The cones were air-dried for at least 72 h prior to dissecting and removing the seeds. We placed the seeds in 95% ethanol to separate the filled seeds containing an embryo (sinking seeds) from empty seeds (Ho \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e1984\u003c/span\u003e, Sirois \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2000\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo estimate availability of fungi, in 2022 we collected all epigeous fungal sporocarps within two randomly selected 15 m radius plots at each site. In 2023, we collected all epigeous fungi within four 1-m\u003csup\u003e2\u003c/sup\u003e quadrats placed 5 m from each site\u0026rsquo;s centre point in the four cardinal directions. At occupied sites, the midden served as the centre, and at unoccupied sites, we used the cone count centre point. All samples were oven-dried for 48\u0026ndash;72 hrs, and dry fungal biomass (kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) was used to reflect fungus availability at each site (Derbridge \u0026amp; Koprowski \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWe monitored squirrel hoarding activity using motion-activated Browning trail cameras (model: Strike Force Extreme and Strike Force HD Pro X) installed at 17 occupied sites (one camera per site). We chose sites with previously observed middens and installed each camera facing the largest midden if a site had multiple middens. Cameras captured bursts of three photos when activated within a three-metre detection range, with a 0.5 second recovery between bursts. We recorded the date, time, location, and type of food items hoarded from August 1 to September 30, reflecting the typical harvest period for squirrels in central areas of the boreal forest (Fletcher et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, Archibald et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) and to account for the shorter growing seasons and earlier winters characteristic of the northern boreal treeline. Squirrels entering middens with food or crossing the camera\u0026rsquo;s field of view carrying food, were classified as hoarding events. Observations may have underestimated activity if a midden had multiple entrances not covered by cameras.\u003c/p\u003e\n\u003ch3\u003eStable isotope sample collection and preparation\u003c/h3\u003e\n\u003cp\u003eRed squirrel body hair moults in spring and fall, while tail hair moults only in the fall (Nelson \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e1945\u003c/span\u003e). We collected body hair (~\u0026thinsp;3\u0026ndash;4 clippings, ~\u0026thinsp;0.5 cm behind the shoulder) and tail hair (~\u0026thinsp;2\u0026ndash;3 cm) each June, from a total of 103 squirrels (90 unique individuals, 9 recaptures, and 4 donated by local trappers) to compare seasonal diets. We captured squirrels using Tomahawk live traps (model #202). We recorded the sex and marked each squirrel with a Passive Integrated Transponder (PIT) tag injected under the skin for permanent identification and metal ear tags threaded with coloured pipe cleaners for visual identification\u003c/p\u003e \u003cp\u003eWe collected samples of potential food sources from forested and forest-wetland regions within our study area in August 2021 (Table S1). Samples included white (n\u0026thinsp;=\u0026thinsp;9) and black spruce seeds (n\u0026thinsp;=\u0026thinsp;9), tamarack seeds (n\u0026thinsp;=\u0026thinsp;6) and buds (n\u0026thinsp;=\u0026thinsp;15), and the most common species of berries (n\u0026thinsp;=\u0026thinsp;97), lichen (n\u0026thinsp;=\u0026thinsp;22), and fungi (n\u0026thinsp;=\u0026thinsp;110).\u003c/p\u003e \u003cp\u003eHair samples were washed twice with mild, soapy water and rinsed thoroughly before drying in a drying oven for 48 hours. We removed seeds from spruce cones and dried the seeds, fungi, berry and lichen samples in a drying oven for a minimum of 48 hours, and ground all food samples to a powder using a mortar and pestle, washing the tools between samples with 70% ethanol. All samples were analyzed for carbon and nitrogen stable isotope ratios on a continuous-flow isotope ratio mass spectrometer (Delta V Advantage) coupled to a Costech 4010 Elemental Combustion system and a ConFlo IV universal interface.\u003c/p\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eData analysis\u003c/h2\u003e \u003cp\u003eWe used stable isotope analysis to examine annual and seasonal variation in squirrel diets. The isotopic composition of a consumer\u0026rsquo;s tissues reflects dietary sources, adjusted by trophic discrimination factors (TDFs) to account for isotopic shifts during assimilation (Phillips \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2012\u003c/span\u003e, Parnell et al. \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). We applied rodent-specific TDFs and standard deviations of Δ\u003csup\u003e13\u003c/sup\u003eC = 3.3\u0026thinsp;\u0026plusmn;\u0026thinsp;2.2\u0026permil; and Δ\u003csup\u003e15\u003c/sup\u003eN = 1.9\u0026thinsp;\u0026plusmn;\u0026thinsp;0.2\u0026permil; (Hobbie et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Pauli et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) to source isotope values. Using a K nearest-neighbor (KNN) randomization test (k\u0026thinsp;=\u0026thinsp;15) we classified 231 samples into five isotopically distinct food groups: berries (species n\u0026thinsp;=\u0026thinsp;7), lichen (n\u0026thinsp;=\u0026thinsp;2), conifer seeds and buds (not isotopically distinct: ANOVA: δ\u003csup\u003e13\u003c/sup\u003eC: F\u003csub\u003e3,35\u003c/sub\u003e = 0.65, p\u0026thinsp;=\u0026thinsp;0.59; δ\u003csup\u003e15\u003c/sup\u003eN: F\u003csub\u003e3,35\u003c/sub\u003e = 1.57, p\u0026thinsp;=\u0026thinsp;0.38), and fungi divided into two groups (genera n\u0026thinsp;=\u0026thinsp;5) due to δ\u003csup\u003e15\u003c/sup\u003eN variation across genera and habitats (forest vs forest-wetland) (Table S1, Figure S1).\u003c/p\u003e \u003cp\u003eWe estimated the dietary contribution of each food group to squirrel diet using Bayesian isotopic mixing models with the MixSIAR and \u003cem\u003erjags\u003c/em\u003e packages (Parnell et al. \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) in R. All models incorporated uniform prior distributions and concentration dependence using the mean elemental concentrations for each prey group. We used Markov-chain Monte Carlo (MCMC) methods to estimate the parameters of the mixing models and ran three parallel MCMC chains with a burn-in of 50,000 iterations. We generated posterior samples using 150,000 iterations of the models and a thinning rate of 50 and checked model convergence using the Gelman and Geweke diagnostic tests produced by MixSIAR (Geweke \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e1991\u003c/span\u003e, Jackson et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2011\u003c/span\u003e, Gelman et al. \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eStable isotope ratios were also used to estimate dietary niche breadth, where narrower isotopic niches indicate more specialised diets (Bearhop et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2004\u003c/span\u003e, Jackson et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). For opportunistic species, niche breadth can shift seasonally and annually with food availability. Isotopic niche breadth was modelled using an ellipse-based approach with the Stable Isotope Bayesian Ellipses in R (SIBER) package in R. We compared isotopic signatures from body (spring diet) and tail (fall diet) hair samples. Previous studies suggest the fall moult occurs between late August and into September (Kranowski \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e1969\u003c/span\u003e) and the spring moult occurs between April and May (Layne \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e1954\u003c/span\u003e), although populations at higher latitudes can moult between March and June (Lepage \u0026amp; Parker 1988). However, we did not observe signs of moulting during our trapping efforts throughout June and into early July. We therefore assumed that all body hair collected in June was recently grown. We generated metrics for isotopic niche breadths using standard ellipse areas corrected for small sample size (SEA\u003csub\u003ec\u003c/sub\u003e) and Bayesian Standard Ellipse Areas (SEA\u003csub\u003eb\u003c/sub\u003e) (150,000 MCMC iterations, 95% credible interval). We compared seasonal and annual niche breadths using the credible intervals to estimating the probability (0\u0026ndash;1) that one group\u0026rsquo;s ellipses were smaller than another (Jackson et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). We also examined the dietary niche overlap between ellipses by calculating the proportion of ellipse overlap using 1,000 draws of the posterior estimates of each ellipse.\u003c/p\u003e \u003cp\u003eAll statistical analyses were conducted in R (version 4.0.5). We used generalised linear mixed effects models (GLMM) to determine if mean cone production varied annually, using \u0026lsquo;site ID\u0026rsquo; as a random effect. Since the cone count data was overdispersed, we used a negative binomial link function. For fungal biomass, we used a paired t-test to determine if dry fungal biomass varied from 2022 to 2023.\u003c/p\u003e \u003cp\u003eWe examined annual variation in filled seed counts per cone using a linear model (LM) with cone production (total cones per tree) and seed production (total number of seeds per cone, comprising both filled and empty seeds) as additional explanatory variables. To examine the effect of cone abundance on hoarded food items, we used trail camera observations to determine the daily mean cone and fungi hoarded per camera. We used a GLMM for each food source with a Poisson link and accounted for spatial variation using \u0026lsquo;site ID\u0026rsquo; as a random effect.\u003c/p\u003e \u003cp\u003eWe tested for annual and seasonal variation in squirrel diet, using a multivariate analysis of variance (MANOVA) to compare δ\u003csup\u003e13\u003c/sup\u003eC and δ\u003csup\u003e15\u003c/sup\u003eN values between years and body and tail hair. We further examined annual diet variations in MixSIAR, with \u0026lsquo;year\u0026rsquo; (4 levels) as a fixed effect and \u0026lsquo;site ID\u0026rsquo; (26 levels) as a random effect. Squirrels were only trapped at 26 of the 29 sites in our study area. We ran a second mixing model to estimate individual diets using \u0026lsquo;squirrel ID\u0026rsquo; as a fixed effect. We incorporated the mean individual dietary proportions from MixSIAR into three beta regression models (berries, fungi, and lichen) with a canonical link function using the \u003cem\u003ebetareg\u003c/em\u003e package in R to determine if the consumption of alternate food sources fluctuated with cone abundance, using the mean cone abundance at the sites where the individuals were trapped. We used beta regression models since dietary proportions range from zero to one (McAulay et al. \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). We paired all squirrel diets with cone production estimates from the previous fall, as tail hair reflects the previous fall\u0026rsquo;s diet and body hair collected in summer reflects the spring diet when most food items such as fungi, cones, and berries were not yet available. Thus, squirrels likely relied on food cached the previous fall.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eTree density across sites averaged 483.7 stems ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (\u0026plusmn; 54.2 SE; range 56 to 579). From 2020 to 2023, annual spruce cone production varied significantly (Table S2), with the greatest number of cones per tree in the mast year of 2022 (471.2 \u0026plusmn; 33.1), low crops in 2021 (7.9 \u0026plusmn; 0.8) and 2023 (6\u003cem\u003e.2\u003c/em\u003e \u0026plusmn; 1.2), and a moderate cone year in 2020 (11.5 \u0026plusmn; 6.0). The number of filled seeds per cone was low overall, but also varied annually (Table S2), with the greatest number of filled seeds per cone in 2023 (4.9 \u0026plusmn; 1.6), the fewest number in the mast year of 2022 (0.6 \u0026plusmn; 0.0), and a low production of filled seeds in 2021 (2.4 \u0026plusmn; 0.6). Fungus production across sites ranged from 0.0 to 8.5 kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, with greater abundance in 2022 (3.5\u0026thinsp;\u0026plusmn;\u0026thinsp;0.5) compared to 2023 (2.5\u0026thinsp;\u0026plusmn;\u0026thinsp;0.6) (t\u003csub\u003e25\u003c/sub\u003e\u0026thinsp;=\u0026thinsp;5.51, p\u0026thinsp;=\u0026thinsp;0.014). Variability of fungal abundance between sites was high but lower in 2022 (coefficient of variation\u0026thinsp;=\u0026thinsp;67%) compared to 2023 (coefficient of variation\u0026thinsp;=\u0026thinsp;114%).\u003c/p\u003e \u003cp\u003eMean daily observations of spruce cone and fungus hoarding in August and September fluctuated annually in response to cone production (GLMM cones: z\u003csub\u003e1,46\u003c/sub\u003e = 9.1, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; fungi: z\u003csub\u003e1,46\u003c/sub\u003e = -2.3 p\u0026thinsp;=\u0026thinsp;0.004). Cone hoarding was highest in 2022 (8.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.2 cones day\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), lowest in 2021 (0.9\u0026thinsp;\u0026plusmn;\u0026thinsp;0.3), and relatively low in 2020 (2.3\u0026thinsp;\u0026plusmn;\u0026thinsp;0.3) and 2023 (1.4\u0026thinsp;\u0026plusmn;\u0026thinsp;0.6). Fungus hoarding was highest in 2021 (1.0\u0026thinsp;\u0026plusmn;\u0026thinsp;0.2 sporocarp day\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), lowest in 2020 (0.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.0) and 2022 (0.1\u0026thinsp;\u0026plusmn;\u0026thinsp;0.0), and relatively low in 2023 (0.7\u0026thinsp;\u0026plusmn;\u0026thinsp;0.1).\u003c/p\u003e \u003cp\u003eMale and female squirrels did not differ in δ\u003csup\u003e13\u003c/sup\u003eC (t = -0.44, df\u0026thinsp;=\u0026thinsp;51.43, p\u0026thinsp;=\u0026thinsp;0.661) or δ\u003csup\u003e15\u003c/sup\u003eN (t = -0.97, df\u0026thinsp;=\u0026thinsp;51.43, p\u0026thinsp;=\u0026thinsp;0.543) (Table S3 and Figure S1), and neither δ\u003csup\u003e13\u003c/sup\u003eC nor δ\u003csup\u003e15\u003c/sup\u003eN values varied seasonally (MANOVA F\u003csub\u003e1,87\u003c/sub\u003e = 2.9, p\u0026thinsp;=\u0026thinsp;0.596). The Bayesian ellipses were also similar in size (female SEA\u003csub\u003eb\u003c/sub\u003e = 3.7; male SEA\u003csub\u003eb\u003c/sub\u003e = 3.2) and the proportional overlap between male and female ellipses was also high (0.862), further suggesting a similar diet. However, δ\u003csup\u003e15\u003c/sup\u003eN values varied annually within each season (spring: F\u003csub\u003e2,86\u003c/sub\u003e = 18.53, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; fall: F\u003csub\u003e2,86\u003c/sub\u003e = 14.19, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), but δ\u003csup\u003e13\u003c/sup\u003eC did not (spring: F\u003csub\u003e2,86\u003c/sub\u003e = 1.72, p\u0026thinsp;=\u0026thinsp;0.153; fall: F\u003csub\u003e2,86\u003c/sub\u003e = 0.99, p\u0026thinsp;=\u0026thinsp;0.135). The areas of the Bayesian ellipses reflecting dietary niche breadths in fall (SEA\u003csub\u003eb\u003c/sub\u003e = 2.9) and spring (SEA\u003csub\u003eb\u003c/sub\u003e =3.3 2.8) were similar (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), as the proportion of posterior draws where the ellipses were smaller in fall than spring was 0.514. The proportional overlap between spring and fall ellipses was 0.815, further suggesting a high degree of similarity between seasonal diets (Table S3). The areas of the ellipses were largest in 2022 (spring: SEA\u003csub\u003eb\u003c/sub\u003e = 2.5; fall: SEA\u003csub\u003eb\u003c/sub\u003e = 2.7), following a low crop in 2021, and the smallest in 2023 (spring: SEA\u003csub\u003eb\u003c/sub\u003e = 1.0; fall: SEA\u003csub\u003eb\u003c/sub\u003e = 1.1), following the mast year in 2022 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Similarly, the dietary niche breadth in 2023, following the mast year, was narrower than 2022 (reflecting a low crop in 2021) for both spring and fall (Table S3).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSquirrel diet consisted predominantly of fungi (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e) (\u0026gt;\u0026thinsp;57% in spring and \u0026gt;\u0026thinsp;62% in fall), with a small conifer cone contribution (~\u0026thinsp;30% in spring and fall) (Table S4 and Figure S2). However, consumption of different food groups varied annually (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Filled seed production significantly affected the consumption of most alternate food sources (fungi: z = -4.68, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, pseudo r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.661; berry: z = -4.22, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, pseudo r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.442). However, lichen consumption was not affected by changes in cone production (z = -1.31, p\u0026thinsp;=\u0026thinsp;0.054, pseudo r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.118).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study demonstrated that red squirrels, often considered seed specialists, can exhibit a diet predominantly composed of fungi rather than seeds, which to our knowledge has not been found elsewhere. The sporadic and low cone production in our study area, along with a low number of filled seeds within cones, is reflected in this predominantly fungal diet. In interior boreal forest, many studies have recorded extensive white spruce cone production in mast years over thousands of cones per tree (Krebs et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e, McAdam et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, Leeper \u0026amp; LaMontagne \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2021\u003c/span\u003e, Krebs et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2023\u003c/span\u003e, Wishart \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). While cone production quadrupled in our mast year (2022), the total number of cones per tree rarely exceeded 600 cones per tree, with a maximum of 1288 cones. Given the low tree density and small tree size in our study area compared to interior areas of squirrel habitat (Young et al. \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2002\u003c/span\u003e, Sharma \u0026amp; Parton \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Boonstra et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2008\u003c/span\u003e), food availability per hectare remains limited. Although we experienced two years of low crops in our 4-year study period, low cone crops succeeding a mast year are not uncommon due to the immense energy requirements of a mast year (Krebs et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). The single masting event we observed, as well as the 2018 mast year from our study area (Kucheravy et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), did not achieve crop yields to the same extent as other studies (Table\u0026nbsp;2). Unfavourable environmental and climatic conditions at distributional edges can affect tree growth, resulting low cone production, and a reduction or failure to produce viable seeds. In particular, the dynamics of tree populations at northern latitudinal limits can be linked to the limiting effect of low seasonal temperatures and short growing seasons on pollen and seed production and the subsequent germination and establishment of seeds and seedlings (Henttonen et al. \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e1986\u003c/span\u003e, Zasada \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e1988\u003c/span\u003e, Sirois \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). In conifers, while long-term cone production is frequently linked to tree size, for species such as white spruce, climatic limitations in the past and current years rather than endogenous factors often dictate cone crop size (Messaoud et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Krebs et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e, LaMontagne \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Seasonal temperature changes in our study region typically exhibit a delay of approximately one month compared to boreal forest regions farther south of the northern treeline (Environment Canada data from Haines Junction Airport, Yukon, and Churchill Airport, Manitoba). Additionally, snow cover can persist for up to two months longer (Scott et al. \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e1993\u003c/span\u003e). The variable nature of the climate at the northern boreal forest treelines reduces the likelihood of having consecutively abundant years for spruce cones (Messaoud et al. \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Additionally, life history theory predicts individuals will prioritize limited energy budgets toward growth, survival, and reproduction to maximize fitness. In the harsh growing conditions presented at the northern treeline, trees may allocate energy toward growth instead of reproduction.\u003c/p\u003e \u003cp\u003eIn addition to fluctuation in spruce cone production, the quality and quantity of seeds per cone can vary with cone size and year (Waldron \u003cspan citationid=\"CR99\" class=\"CitationRef\"\u003e1965\u003c/span\u003e, Zasada \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e1988\u003c/span\u003e), potentially making cone numbers a poor predictor of food availability. Without the embryonic tissues, empty seeds have little nutritional value for granivores (Verd\u0026uacute; \u0026amp; Garc\u0026iacute;a-Fayos \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). As the proportion of empty seeds increases within cones, the handling cost of finding highly nutritious seeds also increases (Verd\u0026uacute; \u0026amp; Garc\u0026iacute;a-Fayos \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e1998\u003c/span\u003e, \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). In our study area, the percentage of filled seeds per cone fell from 8.5% (2021) to 1% in the mast year (2022). Such observations are lower than those observed in other white spruce forests. In northern Ontario, O\u0026rsquo;Connell (2005) reported that the percentage of filled seed per cone ranged from 26\u0026ndash;42% in a non-mast year. Furthermore, in contrast to our observations, Waldron et al. (1965) noted an increase in filled seed in mast years compared to non-mast years, from 12\u0026ndash;58% in southern Manitoba. Reduced pollination is commonly associated with empty seeds in conifers (O'Connell et al. \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). In many spruce species, unpollinated ovules can develop into seed coats that contain degenerated gametophytes (Owens \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e1995\u003c/span\u003e). Factors such as small stand sizes, low tree density, and environmental stresses such as temperature, water availability, and nutrients, can restrict pollination, increase inbreeding, and raise seed abortion rates (Owens \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e1995\u003c/span\u003e, O'Connell et al. \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2006\u003c/span\u003e, Benjamin et al. \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). For plants residing at the northern limit of their distribution, shorter and cooler growing seasons can result in low seed quality and quantity (Zasada \u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e1992\u003c/span\u003e, Lavoie \u0026amp; Payette \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e1994\u003c/span\u003e). Specifically, studies have suggested low seasonal temperatures during spring pollination can lead to seed abortion and cone damage (Zasada \u003cspan citationid=\"CR105\" class=\"CitationRef\"\u003e1992\u003c/span\u003e, Owens \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e1995\u003c/span\u003e) and inhibit seed maturation (Sirois \u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). In white spruce, multi-year reproductive cycles increase susceptibility to such factors affecting seed production. In our study, while the number of filled seeds per cone was low, the total number of filled seeds per tree exceeded those in low cone crop years due to the abundance of cones. However, extracting filled seeds from many empty cones likely increases predator handling costs, influencing seed predator abundance and foraging efficiency (Perea et al. \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). The low seed quality in our study area may further explain why squirrels in this area predominantly consumed fungi.\u003c/p\u003e \u003cp\u003eIn our study area, fungi appeared from late August to early September with an annual variation typical for boreal forest fungal crops (Luoma et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, Krebs et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Mean dry biomass was comparable with other studies in coniferous forests, including interior boreal forest (2.5 kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) (Krebs et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) and Douglas-fir forests in the Pacific Northwest (2\u0026ndash;5 kg ha\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) (Luoma et al. \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Fungal reproduction can also experience widespread synchrony (Mehus \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e1986\u003c/span\u003e). Our estimates of spruce cone and fungal abundance could be biased by squirrels sampling and caching food prior to our surveys. Despite conducting surveys at consistent times annually, fluctuations in fungal emergence due to rainfall could mean peak abundance was missed.\u003c/p\u003e \u003cp\u003eAt the northern treeline, red squirrels must constantly adjust to fluctuating seasonal temperatures and food availability. Optimal foraging theory states that species prioritize high-energy foods to maximize fitness (Stephens \u0026amp; Krebs \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e1986\u003c/span\u003e), exhibiting specialized diets when preferred resources are abundant and broader diets when resources are scarce. Our results support foraging theory, with a narrower dietary niche breadth following the mast year, although broader than expected if spruce seeds were the dominant food source, likely due to the low quality of seeds at the treeline. In non-mast years, dietary niche breadths generally expanded, with broadest niche observed during the low cone crop in 2021.\u003c/p\u003e \u003cp\u003eFungi were the primary contributor to squirrel diet (~\u0026thinsp;70%) across all seasons and years, although other food sources, including berries, were more prominent during the low cone crop year (2021 but reflected in the 2022 diet). Although not a major component, berries increased during low cone years. Like fungi, berries are typically abundant in late summer and early fall and can act as an important source of carbohydrates when spruce seeds are scarce (Stephens et al. \u003cspan citationid=\"CR91\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Lichen was negligible and showed no variation with cone abundance, consistent with red squirrel\u0026rsquo;s limited use of this resource (Currah et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, Dubay et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eRed squirrels are also known nest predator of understory-nesting birds, such as passerines, in northern coniferous forests (Reitsma et al. \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e1990\u003c/span\u003e, Sieving \u0026amp; Willson \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e1998\u003c/span\u003e, Willson et al. \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). However, passerines were excluded from our stable isotopes models because they arrive for the mating season in early June, after the spring moult has occurred. Snowshoe hares were also excluded as a potential food source since predation by red squirrels is rare and mainly involves young hares (\u0026lt;\u0026thinsp;2 weeks old) in spring and summer (O'Donoghue \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e1994\u003c/span\u003e). In more interior northern regions of red squirrel habitat, snowshoe hares typically have three litters (late May, late June - July after the spring molt, and early August before the fall moult) (O'Donoghue \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e1994\u003c/span\u003e, Oli et al. \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). But delayed spring conditions in our study area shift reproduction into late August, overlapping with fall molt hair growth. However, during August and September, red squirrels are intensely focused on harvesting spruce cones and fungi, reducing their likelihood of consuming alternative food sources. Further, other studies reporting hare consumption by red squirrels have observed squirrels scavenging on hare carcasses in winter (Peers et al. \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), which was not covered by the timeline of hair growth in our study.\u003c/p\u003e \u003cp\u003eWe observed little seasonal variation in diet during our study period. In early summer, red squirrels will feed on spruce buds and new growth as their winter food caches become depleted, shifting to primary resources, such as spruce seeds and fungi, as they become more abundant in late summer and early fall (Ren et al. \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). As spruce buds were isotopically similar to seeds, we were unable to differentiate between the two food items in spring diet. However, squirrels were observed eating spruce buds in early summer during our study. By late summer and fall, as primary resources become available, squirrels likely consume the same food items they are caching.\u003c/p\u003e \u003cp\u003eThroughout their distribution, red squirrels primarily consume conifer seeds and are considered conifer specialists (Smith \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e1968\u003c/span\u003e, Fletcher et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, McAdam et al. \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, Wishart \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In northern interior areas of boreal forest, red squirrels rely exclusively on white spruce seeds (McAdam \u0026amp; Boutin 2003, Boutin et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2006\u003c/span\u003e) with individuals hoarding up to 20,000 cones for winter (Smith \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e1968\u003c/span\u003e, Hurly \u0026amp; Lourie \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). While spruce cones may not be abundant at the treeline, our results suggest squirrels still prefer conifer seeds, with seed availability influencing the consumption of alternate foods like fungi and berries. Fungi are a known alternate food source for red squirrels, often comprising over half of squirrel\u0026rsquo;s diet in low cone years (Steele \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e1998\u003c/span\u003e, Currah et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, Koprowski \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2005\u003c/span\u003e, Teron \u0026amp; Hutchison \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2013\u003c/span\u003e, Derbridge \u0026amp; Koprowski \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, Pauli et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Fungal hoarding by red squirrels fluctuated with cone production, peaking during the low cone crop year (2021) and declining in the mast year (2022). In the northern boreal forest, spruce cone hoarding can be highly variable, and red squirrels will switch from hoarding cones to fungi in years of low production, hanging mushrooms in trees to dry prior to caching (Currah et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, Krebs et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Fletcher et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) noted that fungi comprised over 60% of observed caching events when spruce cone production was low, but only 1% of cached items during abundant cone years. The availability and minimal foraging effort needed make fungi a valuable secondary food source for squirrels, and squirrels play a key role in fungal spore dispersal (Krebs et al. \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, the production of white spruce cones varies greatly, and red squirrels have had to adjust their diet according to the availability of primary resources. Compared to similar studies conducted elsewhere in the northern boreal forest (LaMontagne et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2005\u003c/span\u003e, LaMontagne \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Lamontagne \u0026amp; Boutin \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Krebs et al. \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2012\u003c/span\u003e, Wishart \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), spruce cone production was generally lower at the sub-Arctic treeline. As a consequence, red squirrels living at the sub-Arctic edge of their range rely on fungi as their primary food source, even during mast years. While the caching and consumption of fungi by red squirrels is not a new phenomenon (Currah et al. \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2000\u003c/span\u003e, Fletcher et al. \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, Pauli et al. \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), fungi have not previously been documented as the primary food source for red squirrels, which emphasizes how marginal treeline is for squirrels.\u003c/p\u003e \u003cp\u003eClimate change is altering vegetation patterns in the sub-Arctic and within the treeline, leading to increased tree density and recruitment (Payette et al. \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2001\u003c/span\u003e, Mamet \u0026amp; Kershaw \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2011\u003c/span\u003e, Mamet \u0026amp; Kershaw \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). In recent decades, changes in masting dynamics have been largely attributed to climate change. In the northern boreal forest, the intervals between mast years have shortened from 5\u0026ndash;7 to 2\u0026ndash;4 years since 2006 (Krebs et al. \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). As a result, the restrictions placed on squirrels by limited food availability may lessen in the future, increasing the success of future populations and the likelihood of range expansion. However, given the low proportion of filled seed within cones, it is unclear if these shorten intervals between mast years represent increased seed availability. Understanding how factors operate within range boundaries to restrict a species' success and how species conform to these limitations can provide insight into adaptations along latitudinal gradients (Gaston \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) and species\u0026rsquo; potential for future range expansion.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eThis study was funded by the Natural Sciences and Engineering Research Council of Canada, the Northern Scientific Training Program (NSTP), the University of Manitoba Fieldwork Support Program, and the Churchill Northern Studies Centre (CNSC) Northern Research Fund. We thank the CNSC for their logistical support and all our field assistants for their invaluable assistance in data collection.\u0026nbsp;We are also thankful to Drs. Collin Garroway and Jane Waterman for their advice and loan of equipment.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was funded by the Natural Sciences and Engineering Council of Canada, the NSTP, the University of Manitoba Fieldwork Support Program, the Oakes Riewe Environmental Studies research award from the University of Manitoba, and the Northern Research Fund from the CNSC.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of interest/competing interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare they have no conflict of interest.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data and samples collected were in accordance with the ethical standards of the University of Manitoba (animal care protocol F20-010) and approved by Government of Manitoba Wildlife and Fisheries branch (Permit no: WB2433).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and material\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used for this study are available from the corresponding author upon request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCode availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe code used for this study is available from the corresponding author upon request.\u003c/p\u003e"},{"header":"References","content":"\u003cp\u003eArchibald, D. W., Q. E. Fletcher, S. Boutin, A. G. McAdam, J. R. Speakman, and M. M. Humphries. 2013. Sex-specific hoarding behavior in North American red squirrels (\u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e). Journal of Mammalogy 94:761-770.\u003c/p\u003e\n\u003cp\u003eArchibald, D. W., A. G. McAdam, S. Boutin, Q. E. Fletcher, and M. M. Humphries. 2012. Within-season synchrony of a masting conifer enhances seed escape. The American Naturalist 179:536-544.\u003c/p\u003e\n\u003cp\u003eBayne, E. M. and K. A. Hobson. 2002. Effects of red squirrel (\u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e) removal on survival of artificial songbird nests in boreal forest fragments. The American Midland Naturalist 147:72-79.\u003c/p\u003e\n\u003cp\u003eBearhop, S., C. E. Adams, S. Waldron, R. A. Fuller, and H. Macleod. 2004. Determining trophic niche width: a novel approach using stable isotope analysis: Stable isotopes as measures of niche width. Journal of Animal Ecology 73:1007-1012.\u003c/p\u003e\n\u003cp\u003eBenhamou, S.\u0026nbsp;1996. Space use and foraging movements in the American red squirrel (T\u003cem\u003eamiasciurus hudsonicus\u003c/em\u003e). Behavioural processes 37:89-102.\u003c/p\u003e\n\u003cp\u003eBenjamin, J. S., J. D. Roth, and J. H. Markham. 2024. Red foxes increase white spruce seed production at its northern range limit. Basic and Applied Ecology.\u003c/p\u003e\n\u003cp\u003eBoonstra, R., L. Desantis, C. J. Krebs, and D. S. Hik. 2008. Climate and nutrient influences on the growth of white spruce trees in the boreal forests of the Yukon. Climate Research 36:123-130.\u003c/p\u003e\n\u003cp\u003eBoutin, S., L. A. Wauters, A. G. McAdam, M. M. Humphries, G. Tosi, and A. A. Dhondt. 2006. Anticipatory reproduction and population growth in seed predators. Science 314:1928-1930.\u003c/p\u003e\n\u003cp\u003eBrown, J. H., D. W. Mehlman, and G. C. Stevens. 1995. Spatial variation in abundance. Ecology 76:2028-2043.\u003c/p\u003e\n\u003cp\u003eCallaghan, T. V., et al. 2004. Biodiversity, distributions and adaptations of Arctic species in the context of environmental change. AMBIO: A Journal of the Human Environment 33:404-417.\u003c/p\u003e\n\u003cp\u003eCorkery, C. A., E. Nol, and L. Mckinnon. 2019. No effects of asynchrony between hatching and peak food availability on chick growth in Semipalmated Plovers (\u003cem\u003eCharadrius semipalmatus\u003c/em\u003e) near Churchill, Manitoba. Polar Biology 42:593-601.\u003c/p\u003e\n\u003cp\u003eCurrah, R., E. Smreciu, T. Lehesvirta, M. Niemi, and K. Larsen. 2000. Fungi in the winter diets of northern flying squirrels and red squirrels in the boreal mixedwood forest of northeastern Alberta. Canadian Journal of Botany 78:1514-1520.\u003c/p\u003e\n\u003cp\u003eDehling, D. M., et al. 2021. Specialists and generalists fulfil important and complementary functional roles in ecological processes. Functional Ecology 35:1810-1821.\u003c/p\u003e\n\u003cp\u003eDerbridge, J. J. and J. L. Koprowski. 2019. Experimental removals reveal dietary niche partitioning facilitates coexistence between native and introduced species. Ecology and Evolution 9:4065-4077.\u003c/p\u003e\n\u003cp\u003eDubay, S., G. Hayward, and C. Martinez del Rio. 2008. Nutritional value and diet preference of arboreal lichens and hypogeous fungi for small mammals in the Rocky Mountains. Canadian Journal of Zoology 86:851-862.\u003c/p\u003e\n\u003cp\u003eFerguson, S. H. and P. C. Elkie. 2004. Seasonal movement patterns of woodland caribou (\u003cem\u003eRangifer tarandus caribou\u003c/em\u003e). Journal of zoology 262:125-134.\u003c/p\u003e\n\u003cp\u003eFletcher, Q. E., et al. 2010. The functional response of a hoarding seed predator to mast seeding. Ecology 91:2673-2683.\u003c/p\u003e\n\u003cp\u003eGaston, K. J.\u0026nbsp;2003. The structure and dynamics of geographic ranges. Oxford University Press on Demand.\u003c/p\u003e\n\u003cp\u003eGaston, K. J.\u0026nbsp;2009. Geographic range limits: achieving synthesis. Proceedings of the Royal Society B: Biological Sciences 276:1395-1406.\u003c/p\u003e\n\u003cp\u003eGelman, A., J. Hwang, and A. Vehtari. 2014. Understanding predictive information criteria for Bayesian models. Statistics and computing 24:997-1016.\u003c/p\u003e\n\u003cp\u003eGeweke, J.\u0026nbsp;1991. Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments. Federal Reserve Bank of Minneapolis.\u003c/p\u003e\n\u003cp\u003eGross, S. J. and T. D. Price. 2000. Determinants of the northern and southern range limits of a warbler. Journal of Biogeography 27:869-878.\u003c/p\u003e\n\u003cp\u003eHaines, J. A., et al. 2022. Sex-specific effects of capital resources on reproductive timing and success in red squirrels. Behavioral Ecology and Sociobiology 76:142.\u003c/p\u003e\n\u003cp\u003eHarper, K. A., et al. 2011. Tree spatial pattern within the forest\u0026ndash;tundra ecotone: a comparison of sites across Canada. Canadian Journal of Forest Research 41:479-489.\u003c/p\u003e\n\u003cp\u003eHarper, K. A., A. A. Lavallee, and P. Dodonov. 2018. Patterns of shrub abundance and relationships with other plant types within the forest\u0026ndash;tundra ecotone in northern Canada. Arctic Science 4:691-709.\u003c/p\u003e\n\u003cp\u003eHenttonen, H., M. Kanninen, M. Nygren, and R. Ojansuu. 1986. The maturation of \u003cem\u003ePinus sylvestris\u003c/em\u003e seeds in relation to temperature climate in northern Finland. Scandinavian Journal of Forest Research 1:243-249.\u003c/p\u003e\n\u003cp\u003eHo, R. H.\u0026nbsp;1984. Seed-cone receptivity and seed production potential in white spruce. Forest ecology and management 9:161-171.\u003c/p\u003e\n\u003cp\u003eHobbie, E. A., et al. 2017. Stable isotopes and radiocarbon assess variable importance of plants and fungi in diets of Arctic ground squirrels. Arctic, Antarctic, and Alpine Research 49:487-500.\u003c/p\u003e\n\u003cp\u003eHumphries, M. M. and S. Boutin. 2000. The determinants of optimal litter size in free‐ranging red squirrels. Ecology 81:2867-2877.\u003c/p\u003e\n\u003cp\u003eHumphries, M. M., et al. 2005. Expenditure freeze: the metabolic response of small mammals to cold environments. Ecology Letters 8:1326-1333.\u003c/p\u003e\n\u003cp\u003eHurly, T. A. and S. A. Lourie. 1997. Scatterhoarding and larderhoarding by red squirrels: size, dispersion, and allocation of hoards. Journal of Mammalogy 78:529-537.\u003c/p\u003e\n\u003cp\u003eJackson, A. L., R. Inger, A. C. Parnell, and S. Bearhop. 2011. Comparing isotopic niche widths among and within communities: SIBER - Stable Isotope Bayesian Ellipses in R: Bayesian isotopic niche metrics. Journal of Animal Ecology 80:595-602.\u003c/p\u003e\n\u003cp\u003eKayes, I. and A. Mallik. 2020. Boreal forests: distributions, biodiversity, and management. Pp. 1-12, Springer International Publishing Cham.\u003c/p\u003e\n\u003cp\u003eKelly, D. and V. L. Sork. 2002. Mast seeding in perennial plants: why, how, where? Annual review of ecology and systematics 33:427-447.\u003c/p\u003e\n\u003cp\u003eKoprowski, J. L.\u0026nbsp;2005. The response of tree squirrels to fragmentation: a review and synthesis. Animal Conservation 8:369-376.\u003c/p\u003e\n\u003cp\u003eKranowski, P. V.\u0026nbsp;1969. Aspects of red squirrel (\u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e) population ecology in winter in interior Alaska. University of Alaska Fairbanks.\u003c/p\u003e\n\u003cp\u003eKrebs, C., P. Carrier, S. Boutin, R. Boonstra, and E. Hofer. 2008. Mushroom crops in relation to weather in the southwestern Yukon. Botany 86:1497-1502.\u003c/p\u003e\n\u003cp\u003eKrebs, C., J. LaMontagne, A. Kenney, and S. Boutin. 2012. Climatic determinants of white spruce cone crops in the boreal forest of southwestern Yukon. Botany 90:113-119.\u003c/p\u003e\n\u003cp\u003eKrebs, C. J., et al. 2023. Long-term monitoring in the boreal forest reveals high spatio-temporal variability among primary ecosystem constituents. Front Ecol Evol 11:1187222.\u003c/p\u003e\n\u003cp\u003eKucheravy, C. E., J. D. Roth, and J. H. Markham. 2021. Red foxes increase reproductive output of white spruce in a non-mast year. Basic and Applied Ecology 51:11-19.\u003c/p\u003e\n\u003cp\u003eLafleur, P. M.\u0026nbsp;1999. Growing season energy and CO\u003csub\u003e2\u003c/sub\u003e exchange at a subarctic boreal woodland. Journal of Geophysical Research: Atmospheres 104:9571-9580.\u003c/p\u003e\n\u003cp\u003eLaMontagne, J. M.\u0026nbsp;2007. Spatial and temporal variability in white spruce (\u003cem\u003ePicea glauca\u003c/em\u003e) cone production: individual and population responses of North American red squirrels (\u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e). Doctor of Philosophy, University of Alberta Edmonton, Canada.\u003c/p\u003e\n\u003cp\u003eLaMontagne, J. M.\u0026nbsp;2020. Terrestrial ecology: natural selection for mast seeding. Current Biology 30:996-998.\u003c/p\u003e\n\u003cp\u003eLamontagne, J. M. and S. Boutin. 2007. Local‐scale synchrony and variability in mast seed production patterns of \u003cem\u003ePicea glauca\u003c/em\u003e. Journal of Ecology 95:991-1000.\u003c/p\u003e\n\u003cp\u003eLaMontagne, J. M., S. Peters, and S. Boutin. 2005. A visual index for estimating cone production for individual white spruce trees. Canadian Journal of Forest Research 35:3020-3026.\u003c/p\u003e\n\u003cp\u003eLaMontagne, J. M., C. T. Williams, J. L. Donald, M. M. Humphries, A. G. McAdam, and S. Boutin. 2013. Linking intraspecific variation in territory size, cone supply, and survival of North American red squirrels. Journal of Mammalogy 94:1048-1058.\u003c/p\u003e\n\u003cp\u003eLane, J., et al. 2015. Post‐weaning parental care increases fitness but is not heritable in North American red squirrels. Journal of Evolutionary Biology 28:1203-1212.\u003c/p\u003e\n\u003cp\u003eLavoie, C. and S. Payette. 1994. Recent fluctuations of the lichen-spruce forest limit in subarctic Quebec. Journal of Ecology:725-734.\u003c/p\u003e\n\u003cp\u003eLayne, J. N. 1954. The biology of the red squirrel, \u003cem\u003eTamiasciurus hudsonicus loquax\u0026nbsp;\u003c/em\u003e(Bangs), in central New York. Ecological Monographs 24:228-267.\u003c/p\u003e\n\u003cp\u003eLeeper, A. C. and J. M. LaMontagne. 2021. Cone characteristics and insect predation levels vary across years in mast seeding white spruce. Canadian Journal of Forest Research 51:1550-1557.\u003c/p\u003e\n\u003cp\u003eLepage, P. and G. Parker. 1988. Copper, nickel, and iron levels in pelage of red squirrels living near the ore smelters at Sudbury, Ontario, Canada. Canadian journal of zoology 66:1631-1637.\u003c/p\u003e\n\u003cp\u003eLuoma, D. L., J. M. Trappe, A. W. Claridge, K. M. Jacobs, and E. Cazares. 2003. Relationships among fungi and small mammals in forested ecosystems. Mammal Community Dynamics in Western Coniferous Forests: Management and Conservation Cambridge University Press, Cambridge, United Kingdom:343-373.\u003c/p\u003e\n\u003cp\u003eLynch, H. J., M. Rhainds, J. M. Calabrese, S. Cantrell, C. Cosner, and W. F. Fagan. 2014. How climate extremes\u0026mdash;not means\u0026mdash;define a species\u0026apos; geographic range boundary via a demographic tipping point. Ecological Monographs 84:131-149.\u003c/p\u003e\n\u003cp\u003eMamet, S. D. and G. P. Kershaw. 2011. Radial-growth response of forest-tundra trees to climate in the Western Hudson Bay Lowlands. ARCTIC 64:446-458.\u003c/p\u003e\n\u003cp\u003eMamet, S. D. and G. P. Kershaw. 2013. Multi-scale analysis of environmental conditions and conifer seedling distribution across the treeline ecotone of northern Manitoba, Canada. Ecosystems 16:295-309.\u003c/p\u003e\n\u003cp\u003eMcAdam, A. G. and S. Boutin. 2003. Effects of food abundance on genetic and maternal variation in the growth rate of juvenile red squirrels. Journal of evolutionary biology 16:1249-1256.\u003c/p\u003e\n\u003cp\u003eMcAdam, A. G., S. Boutin, B. Dantzer, and J. E. Lane. 2019. Seed masting causes fluctuations in optimum litter size and lag load in a seed predator. The American Naturalist 194:574-589.\u003c/p\u003e\n\u003cp\u003eMcAulay, J., P. J. Seddon, D. J. Wilson, and J. M. Monks. 2020. Stable isotope analysis reveals variable diets of stoats (\u003cem\u003eMustela erminea\u003c/em\u003e) in the alpine zone of New Zealand. New Zealand Journal of Ecology 44:1-13.\u003c/p\u003e\n\u003cp\u003eMehus, H.\u0026nbsp;1986. Fruit body production of macrofungi in some North Norwegian forest types. Nordic Journal of Botany 6:679-702.\u003c/p\u003e\n\u003cp\u003eMessaoud, Y., Y. Bergeron, and H. Asselin. 2007. Reproductive potential of balsam fir (\u003cem\u003eAbies balsamea\u003c/em\u003e), white spruce (\u003cem\u003ePicea glauca\u003c/em\u003e), and black spruce (\u003cem\u003eP. mariana\u003c/em\u003e) at the ecotone between mixedwood and coniferous forests in the boreal zone of western Quebec. American Journal of Botany 94:746-754.\u003c/p\u003e\n\u003cp\u003eNelson, B. A.\u0026nbsp;1945. The spring molt of the northern red squirrel in Minnesota. Journal of Mammalogy 26:397-400.\u003c/p\u003e\n\u003cp\u003eNewbury, R. K. and K. E. Hodges. 2018. Regional differences in winter diets of bobcats in their northern range. Ecology and evolution 8:11100-11110.\u003c/p\u003e\n\u003cp\u003eNygren, M., K. Rissanen, K. Eerik\u0026auml;inen, T. Saksa, and S. Valkonen. 2017. Norway spruce cone crops in uneven-aged stands in southern Finland: a case study. Forest Ecology and Management 390:68-72.\u003c/p\u003e\n\u003cp\u003eO\u0026apos;Connell, L., A. Mosseler, and O. Rajora. 2006. Impacts of forest fragmentation on the mating system and genetic diversity of white spruce (\u003cem\u003ePicea glauca\u003c/em\u003e) at the landscape level. Heredity 97:418-426.\u003c/p\u003e\n\u003cp\u003eO\u0026apos;Donoghue, M.\u0026nbsp;1994. Early survival of juvenile snowshoe hares. Ecology 75:1582-1592.\u003c/p\u003e\n\u003cp\u003eOli, M. K., C. J. Krebs, A. J. Kenney, R. Boonstra, S. Boutin, and J. E. Hines. 2020. Demography of snowshoe hare population cycles. Ecology 101:e02969.\u003c/p\u003e\n\u003cp\u003eOstfeld, R. S. and F. Keesing. 2000. Pulsed resources and community dynamics of consumers in terrestrial ecosystems. Trends in Ecology and Evolution 15:232-237.\u003c/p\u003e\n\u003cp\u003eOwens, J.\u0026nbsp;1995. Constraints to seed production: temperate and tropical forest trees. Tree Physiology 15:477-484.\u003c/p\u003e\n\u003cp\u003ePagani‐N\u0026uacute;\u0026ntilde;ez, E., C. Barnett, H. Gu, and E. Goodale. 2016. The need for new categorizations of dietary specialism incorporating spatio‐temporal variability of individual diet specialization. Journal of Zoology 300:1-7.\u003c/p\u003e\n\u003cp\u003eParnell, A. C., et al. 2013. Bayesian stable isotope mixing models. Environmetrics 24:387-399.\u003c/p\u003e\n\u003cp\u003ePauli, J. N., et al. 2019. Quantifying niche partitioning and multichannel feeding among tree squirrels. Food Webs 21:e00124.\u003c/p\u003e\n\u003cp\u003ePayette, S., M.-J. Fortin, and I. Gamache. 2001. The subarctic forest\u0026ndash;tundra: the structure of a biome in a changing climate: the shifting of local subarctic tree lines throughout the forest\u0026ndash;tundra biome, which is linked to ecological processes at different spatiotemporal scales, will reflect future global changes in climate. BioScience 51:709-718.\u003c/p\u003e\n\u003cp\u003ePeers, M. J., et al. 2020. Prey availability and ambient temperature influence carrion persistence in the boreal forest. Journal of Animal Ecology 89:2156-2167.\u003c/p\u003e\n\u003cp\u003ePerea, R., M. Venturas, and L. Gil. 2013. Empty seeds are not always bad: simultaneous effect of seed emptiness and masting on animal seed predation. Plos One 8:e65573.\u003c/p\u003e\n\u003cp\u003ePhillips, D. L.\u0026nbsp;2012. Converting isotope values to diet composition: the use of mixing models. Journal of Mammalogy 93:342-352.\u003c/p\u003e\n\u003cp\u003eR\u0026eacute;ale, D., D. Berteaux, A. McAdam, and S. Boutin. 2003. Lifetime selection on heritable life‐history traits in a natural population of red squirrels. Evolution 57:2416-2423.\u003c/p\u003e\n\u003cp\u003eRehm, E. M., P. Olivas, J. Stroud, and K. J. Feeley. 2015. Losing your edge: climate change and the conservation value of range‐edge populations. Ecology and Evolution 5:4315-4326.\u003c/p\u003e\n\u003cp\u003eReitsma, L. R., R. T. Holmes, and T. W. Sherry. 1990. Effects of removal of red squirrels, \u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e, and eastern chipmunks, \u003cem\u003eTamias striatus\u003c/em\u003e, on nest predation in a northern hardwood forest: an artificial nest experiment. Oikos:375-380.\u003c/p\u003e\n\u003cp\u003eRen, T., et al. 2017. Seasonal, spatial, and maternal effects on gut microbiome in wild red squirrels. Microbiome 5:1-14.\u003c/p\u003e\n\u003cp\u003eRettie, W. J. and F. Messier. 2000. Hierarchical habitat selection by woodland caribou: its relationship to limiting factors. Ecography 23:466-478.\u003c/p\u003e\n\u003cp\u003eScott, P. A., R. I. Hansell, and W. R. Erickson. 1993. Influences of wind and snow on northern tree-line environments at Churchill, Manitoba, Canada. Arctic:316-323.\u003c/p\u003e\n\u003cp\u003eSexton, J. P., P. J. McIntyre, A. L. Angert, and K. J. Rice. 2009. Evolution and Ecology of Species Range Limits. Annual Review of Ecology, Evolution, and Systematics 40:415-436.\u003c/p\u003e\n\u003cp\u003eSharma, M. and J. Parton. 2007. Height\u0026ndash;diameter equations for boreal tree species in Ontario using a mixed-effects modeling approach. Forest Ecology and Management 249:187-198.\u003c/p\u003e\n\u003cp\u003eShipley, L. A., J. S. Forbey, and B. D. Moore. 2009. Revisiting the dietary niche: when is a mammalian herbivore a specialist? Integrative and comparative biology 49:274-290.\u003c/p\u003e\n\u003cp\u003eSieving, K. E. and M. F. Willson. 1998. Nest predation and avian species diversity in northwestern forest understory. Ecology 79:2391-2402.\u003c/p\u003e\n\u003cp\u003eSir\u0026eacute;n, A. P. K. and T. L. Morelli. 2020. Interactive range‐limit theory (iRLT): An extension for predicting range shifts. Journal of Animal Ecology 89:1-15.\u003c/p\u003e\n\u003cp\u003eSirois, L.\u0026nbsp;2000. Spatiotemporal variation in black spruce cone and seed crops along a boreal forest-tree line transect. Canadian Journal of Forest Research 30:900-909.\u003c/p\u003e\n\u003cp\u003eSmith, M. C.\u0026nbsp;1968. Red squirrel responses to spruce cone failure in interior Alaska. The Journal of Wildlife Management:305-317.\u003c/p\u003e\n\u003cp\u003eSteele, M. A. 1998. \u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e. Mammalian Species:1-9.\u003c/p\u003e\n\u003cp\u003eStephens, D. W. and J. R. Krebs. 1986. Foraging theory. Princeton University Press.\u003c/p\u003e\n\u003cp\u003eStephens, R. B., E. A. Hobbie, T. D. Lee, and R. J. Rowe. 2019. Pulsed resource availability changes dietary niche breadth and partitioning between generalist rodent consumers. Ecology and Evolution 9:10681-10693.\u003c/p\u003e\n\u003cp\u003eSteury, T. D. and D. L. Murray. 2003. Causes and consequences of individual variation in territory size in the American red squirrel. Oikos 101:147-156.\u003c/p\u003e\n\u003cp\u003eSullivan, T. P. and D. S. Sullivan. 1982. Influence of fertilization on feeding attacks to lodgepole pine by snowshoe hares and red squirrels. The Forestry Chronicle 58:263-266.\u003c/p\u003e\n\u003cp\u003eSzumski, C. M., J. D. Roth, and D. L. Murray. 2023. Canada lynx foraging strategies: Facultative specialists become obligate generalists toward the distribution edge. Ecosphere 14:e4629.\u003c/p\u003e\n\u003cp\u003eTeron, J. N. and L. J. Hutchison. 2013. Consumption of truffles and other fungi by the American red squirrel (\u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e) and the eastern chipmunk (\u003cem\u003eTamias striatus\u003c/em\u003e)(\u003cem\u003eSciuridae\u003c/em\u003e) in northwestern Ontario. The Canadian Field-Naturalist 127:57-59.\u003c/p\u003e\n\u003cp\u003evan der Veen, B., J. Mattisson, B. Zimmermann, J. Odden, and J. Persson. 2020. Refrigeration or anti-theft? Food-caching behavior of wolverines (\u003cem\u003eGulo gulo\u003c/em\u003e) in Scandinavia. Behavioral Ecology and Sociobiology 74:1-13.\u003c/p\u003e\n\u003cp\u003eVerd\u0026uacute;, M. and P. Garc\u0026iacute;a-Fayos. 1998. Ecological causes, function, and evolution of abortion and parthenocarpy in \u003cem\u003ePistacia lentiscus (Anacardiaceae)\u003c/em\u003e. Canadian Journal of Botany 76:134-141.\u003c/p\u003e\n\u003cp\u003eVerd\u0026uacute;, M. and P. Garc\u0026iacute;a-Fayos. 2001. The effect of deceptive fruits on predispersal seed predation by birds in Pistacia lentiscus. Plant Ecology 156:245-248.\u003c/p\u003e\n\u003cp\u003eWaldron, R.\u0026nbsp;1965. Cone production and seedfall in a mature white spruce stand. The Forestry Chronicle 41:316-329.\u003c/p\u003e\n\u003cp\u003eWillson, M. F., T. L. D. Santo, and K. E. Sieving. 2003. Red squirrels and predation risk to bird nests in northern forests. Canadian Journal of Zoology 81:1202-1208.\u003c/p\u003e\n\u003cp\u003eWishart, A. E.\u0026nbsp;2023. Variation in resource acquisition in a food-caching mammal, the North American red squirrel (\u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e). Doctor of Philosophy, University of Saskatchewan.\u003c/p\u003e\n\u003cp\u003eYang, L. H., J. L. Bastow, K. O. Spence, and A. N. Wright. 2008. What can we learn from resource pulses. Ecology 89:621-634.\u003c/p\u003e\n\u003cp\u003eYang, L. H., K. F. Edwards, J. E. Byrnes, J. L. Bastow, A. N. Wright, and K. O. Spence. 2010. A meta‐analysis of resource pulse\u0026ndash;consumer interactions. Ecological Monographs 80:125-151.\u003c/p\u003e\n\u003cp\u003eYoung, P. J., V. L. Greer, and S. K. Six. 2002. Characteristics of Bolus Nests of Red Squirrels in the Pinaleno and White Mountains of Arizona. The Southwestern Naturalist 47:267-275.\u003c/p\u003e\n\u003cp\u003eZasada, J.\u0026nbsp;1992. The reproductive process in boreal forest trees. A systems analysis of the global boreal forest:211-233.\u003c/p\u003e\n\u003cp\u003eZasada, J. C. 1988. Embryo growth in Alaskan white spruce seeds. Canadian Journal of Forest Research 18:64-67.\u003c/p\u003e"},{"header":"Tables","content":"\u003cp style='margin:0in;text-indent:0in;line-height:200%;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eTable 1: Comparison of white spruce cone counts and the percentage of seeds that were filled seeds within spruce cones near Churchill, MB, Canada, from interior forests within red squirrel distribution.\u0026nbsp;\u003c/p\u003e\n\u003ctable style=\"border-collapse: collapse;border: none;width: 624px;\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;border-top: 1pt solid windowtext;border-left: none;border-bottom: 1pt solid windowtext;border-right: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eLocation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;border-top: 1pt solid windowtext;border-left: none;border-bottom: 1pt solid windowtext;border-right: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eTotal cones/tree\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;border-top: 1pt solid windowtext;border-left: none;border-bottom: 1pt solid windowtext;border-right: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eYears\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;border-top: 1pt solid windowtext;border-left: none;border-bottom: 1pt solid windowtext;border-right: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eReference\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;border: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u003cstrong\u003eMast year\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;border: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;border: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;border: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eTreeline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e74 \u0026ndash; 863\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003ethis study\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eTreeline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e0 \u0026ndash; 407\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2018\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eKucheravy et al. 2021\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eYukon\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1,000 \u0026ndash; 2,500\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1993 \u0026ndash; 2010\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eKrebs et al. 2012\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eYukon\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e750 \u0026ndash; 3,000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2005 \u0026ndash; 2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eKrebs et al. 2023\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eYukon\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1,000 \u0026ndash; 3,500\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2008 \u0026ndash; 2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eWishart 2023\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eMichigan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2,400 \u0026ndash; 2,700\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2012 \u0026ndash; 2017\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eLeeper \u0026amp; LaMontagne 2021\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u003cstrong\u003eNon-mast year\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eTreeline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e0 \u0026ndash; 123\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2020 - 2023\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003ethis study\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eTreeline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e0 \u0026ndash; 93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2019\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eKucheravy et al. 2021\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eYukon\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e0 \u0026ndash; 200\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1993 \u0026ndash; 2010\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eKrebs et al. 2012\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eMichigan\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e216\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2012 \u0026ndash; 2017\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eLeeper \u0026amp; LaMontagne 2021\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eWisconsin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e6 \u0026ndash; 186\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2012 \u0026ndash; 2014\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eCorona et al. 2022\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eFilled seeds/cone (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;border: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u003cstrong\u003eMast year\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;border: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;border: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;border: none;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eTreeline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1.2\u0026nbsp;\u003cspan style=\"font-family:Symbol;\"\u003e\u0026plusmn;\u003c/span\u003e 0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2022\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003ethis study\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eSouthern Manitoba\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1954 \u0026ndash; 1963\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eWaldron 1965\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eAlaska\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e60.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1957 \u0026ndash; 1959\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eZasada \u0026amp; Gregory 1969\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u003cstrong\u003eNon-mast year\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eTreeline\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e6.4\u0026nbsp;\u003cspan style=\"font-family:Symbol;\"\u003e\u0026plusmn;\u003c/span\u003e 1.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2021 \u0026ndash; 2023\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003ethis study\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eSouthern Manitoba\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e12 \u0026ndash; 48\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1954 \u0026ndash; 1963\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eWaldron 1965\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eNorthern Quebec\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e2001 \u0026ndash; 2004\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eMessaoud et al. 2007\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eNorthern Ontario\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e26 \u0026ndash; 42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1994\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eO\u0026rsquo;Connell 2006\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 119.1pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eAlaska\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 119.45pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e22 \u0026ndash; 62\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 76.45pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003e1954 \u0026ndash; 1963\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 153pt;border-top: none;border-right: none;border-left: none;border-image: initial;border-bottom: 1pt solid windowtext;padding: 0in 5.4pt;vertical-align: top;\"\u003e\n \u003cp style='margin:0in;font-size:16px;font-family:\"Times New Roman\",serif;'\u003eZasada \u0026amp; Gregory 1969\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University of Manitoba","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":"American red squirrel, diet, facultative specialist, stable isotopes, range boundaries, Tamiasciurus hudsonicus","lastPublishedDoi":"10.21203/rs.3.rs-6624068/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6624068/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRange boundaries limit local populations, which may experience pronounced fluctuations in resource availability, particularly at higher latitudes, often seen as resource pulses. In boreal forests, conifers undergo pulses of seed production followed by intervals of low seed production, profoundly affecting consumers dependent on these resources. Red squirrels (\u003cem\u003eTamiasciurus hudsonicus\u003c/em\u003e) are considered seed specialists across the boreal forest. We evaluated how annual changes in white spruce (\u003cem\u003ePicea glauca\u003c/em\u003e) cone production at the sub-Arctic treeline near Churchill, MB, Canada, influenced squirrels\u0026rsquo; use of alternative food sources, predicting that low cone production would increase reliance on alternate foods. Cone crops varied from 2020\u0026ndash;2023, with a mast year in 2022 of 471 cones per tree, approximately 70\u0026ndash;80% lower than mast years elsewhere, and lower crops in other years (6-115 cones per tree). Furthermore, the number of filled seeds (containing an embryo) per cone was low, ranging from 0.6 \u0026plusmn; 0.03 (mean \u0026plusmn; SE) in 2022 to 3.6 \u0026plusmn; 2.6 in 2023. Using stable isotope ratios of hair and Bayesian mixing models, we found that squirrels primarily consumed fungi (~\u0026thinsp;70% of diet), even in mast years, with other food sources varying with cone production. The dominance of fungi in squirrel diet even in mast years, highlights the dietary plasticity of red squirrels beyond seed specialization challenging the seed specialization paradigm. Flexible foraging strategies likely allow populations to persist in resource-limited environments and may facilitate range expansion as climate change reshapes habitats.\u003c/p\u003e","manuscriptTitle":"Presumed seed specialists rely on fungi as their primary food source at the sub-Arctic treeline","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-22 14:44:58","doi":"10.21203/rs.3.rs-6624068/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":"1bae1c98-9cc7-4296-bdc9-11caf3807769","owner":[],"postedDate":"May 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":48282157,"name":"Population Biology"}],"tags":[],"updatedAt":"2025-05-22T14:44:58+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-22 14:44:58","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6624068","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6624068","identity":"rs-6624068","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}