Variation in food availability is a major factor affecting breeding densities and reproduction success, with ultimate consequences on population dynamics1. Availability of food resources is expected to act as a limiting factor more for specialist predators, which use one or few resources, than for generalist predators, which opportunistically exploit the most abundant and easily accessible food resources available at any one time and locality2.
Whether the two feeding tactics are equally profitable is still debated3,4, especially for terrestrial predators5. According to optimal foraging models, trophic niche breadth should depend on the diversity and abundance of available prey, and net energy gain6, that is the ratio between the caloric value of each potentially available prey and the amount of energy spent for finding, pursuing, killing and consuming it.
Notwithstanding, in the last two decades, a growing body of evidence has outlined that foraging is not exclusively driven by energy acquisition and many species tend to regulate the macronutrient composition of their diet to a target ratio (“intake target”)7–9. As the macronutrient composition of the diet has been demonstrated to affect many fundamental fitness-related traits, including growth7,10, fecundity11 and lifespan12,13, we may expect that, for any species, suitable habitats are those which offer food resources capable of satisfying the nutritional requirements of a number of individuals in all the phases of their life cycle.
Specialist predators possess morphological and behavioural adaptations which are supposed to increase their foraging efficiency14 and feed on foods relatively invariant in their nutrient composition, which coincides with the predator’s target macronutrient ratios. In contrast, generalists must be capable of shifting between resources before the opportunity occurs, which implies preexisting behavioural and physiological adaptations15, and need to combine several nutritionally complementary foods to achieve their intake target.
The few available studies16 suggest that, based on their tolerance towards carbohydrates, mammalian predators can be aligned along a carnivore–omnivore continuum, ranging from obligate carnivores, such as wolves (Canis lupus)17 to poorly specialized ursids18. The ability of using fat or carbohydrates as sources of non-protein energy may be expected to be a physiological prerequisite for generalist predators, allowing them to rely on a wide variety of food resources19.
Geographic and seasonal variation in the composition of generalist predators’ diets makes it difficult to compare the diet of populations of widespread species. However, using nutritional geometry Gazzola and Balestrieri20 have recently demonstrated that using a wide variety of food resources does not imply as much variation in the nutritional composition of diets: although using a wide range of fruit and small mammals, widespread carnivores such as martens (Martes martes and Martes foina) can be considered macronutrient specialists (i.e. the macronutrient compositions of the diets of different populations are similar21).
Among carnivore mammals, the red fox (Vulpes vulpes) is considered a prototypical generalist predator: its feeding habits vary widely spatially, temporally and in response to human influence, reflecting the biogeographical patterns of distribution and abundance of food resources22–24. Records of local specialization, due to the disproportionate profitability of anthropogenic resources, reflect the highly opportunistic behaviour of this species25.
This dietary flexibility allows foxes to occur in a wide variety of habitats, from sea level up to 4500 m, including several cities26. Its geographical range is the widest of any member of the order Carnivora (ca. 70 million km²), including most of the Northern Hemisphere, from the Arctic Circle to northern Africa, and Australia27, where it was introduced in the 1870s28.
Such a wide distribution rises an interesting question, that is whether different populations persist on diets that vary widely in macronutrient composition or are capable of using complementary foods to gain the same nutrient intake throughout the species’ range.
The first nutritional strategy has been reported for the wild boar (Sus scrofa), which is a dietary generalist and tolerates a wide range of macronutrient ratios across its whole range, particularly in terms of proportion of energy from protein29. In contrast, mustelids, such as martens (Martes spp.) and the Eurasian badger (Meles meles), tend to keep constant the percent protein energy, while showing a gradient of tolerance towards carbohydrates16,20.
Laboratory experiments on Drosophila melanogaster11 and mice12 suggest that unbalanced diets may have profound effects on life span and reproduction. While the broad fundamental macronutrient niche of wild boars has been suggested to enhance their invasion success, increasing the reproductive output of sows29, we still do not know whether an excess of carbohydrates may affect the individual fitness of free-ranging carnivores.
To assess the macronutrient niche of the red fox, we applied right-angle mixture triangles (RMT)21,30, in the framework of nutritional geometry. Data were extracted from published reports following the approach proposed by Remonti et al.19,31. Based on the wide variety of foods used by foxes, we expected a wide degree of inter-population variation in the percent energy provided by carbohydrates, as so as the recording of clusters of unbalanced diets.
Secondly, we made an attempt to highlight the effects of unbalanced diets on fox density, which was assumed as a proxy for Darwinian fitness. As diet is only one of several factors that may affect population density, samplings were carried out in five areas belonging to the same biogeographical region, in a radius of ca. 30 km (western Italian Alps). We aimed to assess both the yearly diet of each population and their correspondent density, calculated through a faecal DNA-based genetic census.
We expected macronutrient ratios to affect individual fitness, and, therefore, populations showing nutritional balances close to the intake target to achieve higher densities than those with unbalanced diets.