not-yet-known not-yet-known not-yet-known unknown 4 | DISCUSSION Predator population dynamics, behavior, and diet are typically associated with changes in prey populations, which means that predation and prey interactions tend to be the primary concerns when designing conservation and management programs. Our results suggest the opposite can be true— even for a large apex predator. Brown bear diet changed in response to primary productivity, regardless of moose availability, indicating that landscape changes that affect berry production will have larger impacts on bears than changes in moose populations. We identified trends in brown bear diet from 25 years of δ 15N and δ 13C stable isotope data. Brown bear hair δ 15N values were closely related to the proportion of moose in the diet but did not respond to indices of moose availability. Rather, δ 15N values were explained by bilberry production. Variation in δ 13C values were also explained by bilberry production, though some variation in δ 13C was also explained by annual moose calf production. Over 25 years, the proportion of moose and ants in bear diet decreased while bilberry increased, suggesting a general shift from clustered protein-rich foods towards dispersed carbohydrate-rich foods, even with increasing moose numbers. Many species operate on tight metabolic budgets, and it is difficult to determine the full impact of small changes in diet. For brown bears, the 2% decrease in animal derived foods is unlikely to have significant impacts on population health and persistence because they result in small changes in calorie content (Mikkelsen et al., 2023) and even lactating mothers can recoup these costs with minimal increases in foraging (Farley & Robbins, 1995; Welch et al., 1997). But it certainly highlights that changes in diet in response to the environment can be subtle and difficult to interpret in the full context of life-history. Continued changes in primary productivity may reach a tipping point in which populations decline and researchers are likely to miss the connection so long as we exclusively refer to bears, and other predators exclusively “apex predators” or “omnivorous carnivores”. This is particularly noteworthy considering that bilberry production is expected to decrease with climate change (García-Rodríguez et al., 2021). The unintuitive relationship in which δ 13C values are more similar to plant-derived foods in years with high moose calf observations and more similar to animal-derived foods during years of low moose observations, may be an artifact of timing. Bears tend to predate moose neonates in the spring soon after calving (Swenson et al., 2007), while calf observations are reported by the public in October. Brown bears can be a substantial source of mortality for moose calves in Sweden (~25%; Swenson et al., 2007), and so this relationship may arise from bears predating (and eating) more moose calves in the spring, resulting in fewer calves to be counted in the fall. Prior to 2011, annual moose calf numbers in the study area fluctuated around 3000–4000, then rapidly increased until numbers restabilized in 2014 around 13000 (Statistik älgdata). Yet, the proportion of moose in the diet does not increase with calf observations, rather, it appears to decrease through time. This indicates that either moose calves are not a preferred food for bears, or calves have never been a limited food resource within this study population. The population-wide, cyclic pattern of the δ 15N values indicates large-scale processes, most likely related to climate and primary productivity. In the case of many predators, primary productivity is assumed to have an indirect effect on population dynamics, typically through prey responses (Stoner et al., 2018). However, our results indicate that diet is driven by berry production, rather than moose availability, and berry productivity also affects reproduction and body condition in our study population (Hertel et al., 2018). These results add to the growing body of evidence that meat may be overemphasized in diets of some polyphagous species (Deacy et al., 2017; Robbins et al., 2022; Rode et al., 2021), which minimizes the other roles these species play, such as nutrient cyclers and seed dispersers (Borchert & Tyler, 2011; Harrer & Levi, 2018; Reimchen, 2017). Beyond ursid species, modern monitoring tools, such as DNA metabarcoding, camera traps, and stable isotopes analysis, have documented “novel” foraging behaviors in a wide variety of predators, such as bonnethead sharks (Sphyrna tiburo ) eating seagrasses (Leigh et al., 2018), crocodilians (order Crocodilia) eating fruit and dispersing seeds (Platt et al., 2013), or wolves (Canis lupus ) foraging on berries (Homkes et al., 2020). It is unclear how many other polyphagous (omnivorous) species are sensitive to changes in primary production, and hopefully more research will focus on polyphagous species in addition to herbivores. While much work has been done on trophic cascades via top-down effects, more research is needed on trophic transcendence via bottom-up effects. For example, within Scandinavia, annual weather patterns explain little variation in annual berry production (Hertel et al., 2018; Selås 2000). This is because fruit production may be negatively affected by short but extreme weather events or a complex interaction of conditions over the full growing season (Orsenigo et al., 2014). Human modifications to the landscape, such as the conversion of wildlands to agriculture or timber production, changes the structure of ecological communities and can interact with climate change to exacerbate changes in primary productivity (Pirotta et al., 2022). For instance, timber harvest removes the physical protection and thermal buffering the overstory provides to understory plants (Santos et al., 2024), which makes them more vulnerable to extreme weather (Gallinat et al., 2015), and may turn a historically reliable food source into an unpredictable, stochastic resource. Ultimately, polyphagous species may be doubly affected by changes in plant phenology and productivity, once through a direct change in primary productivity as a food source (Hertel et al., 2018), then again as prey species also respond to changes in primary production (Deacy et al., 2017). In this scenario of global change, polyphagous megafauna, such as the brown bear, can be pivotal in maintaining ecosystem stability (Gutgesell et al. 2022).