Urbanisation critically alters the availability of resources and the nature of risks for wildlife. Despite these challenges, carnivores frequently persist in and around urban environments, where novel opportunities, such as anthropogenic food, may outweigh associated ecological risks. Here, we investigate the responses of an urban adapter to novel resources, using stable isotope analysis of vibrissae (whiskers) to understand the spatiotemporal foraging patterns of caracals on the fringes of the city of Cape Town, South Africa. Caracals (Caracal caracal) are medium-sized felids and the largest remaining predators on the Cape Peninsula. Using isotopic niche metrics and home range estimates we assess the comparative effects of demographics, seasonality and urbanisation on variation in individual foraging behaviour from GPS-collar monitored caracals (n=28) across an urban gradient. Despite a wide isotopic niche at the population level, we observed high levels of individual specialisation. Adult and male niches were wider likely due to larger home ranges, which facilitates exploitation of diverse prey across trophic levels. Patterns in δ¹³C were seasonal, with increases during the warmer, drier summer months across the Peninsula irrespective of habitat use. Taken together with niche contraction for caracals in urban areas, our findings suggest higher individual reliance on anthropogenic resources in summer. Caracals using areas dominated by wildland cover had higher δ¹⁵N values and larger niches than those using urban-dominated areas. Across the study area, δ¹⁵N values varied spatially, with increased enrichment in caracals using more coastal and wetland areas and consuming more marine and wetland-adapted prey, particularly in winter. Individual foraging flexibility in caracal is clearly a key strategy for their success in this rapidly transforming landscape. Understanding spatiotemporal shifts in dietary niche and trophic ecology in adaptable urban carnivores, like the Cape Peninsula caracals, is fundamental for understanding the ecological needs of wildlife in and around rapidly growing cities.

Sample collection

The dataset analysed here represents 28 unique individual caracals from the Cape Peninsula, South Africa, some of which (n = 11) were sampled more than once due to recaptures or subsequent mortality sampling. One male was collared twice, once as a subadult (TMC16a) and again 10 months later as an adult (TMC16b; Fig. S1) and is therefore considered as two separate ‘individuals’. The whiskers prioritised for analysis were those for which we had GPS data of the individual caracal, where the whisker represented growth during the collaring periods, allowing the isotopic data to be linked with location data by date. Whiskers were collected from live individuals under sedation during capture for GPS-collaring (n = 25 caracals), and from post-mortem individuals (n = 10 caracals; e.g., roadkill mortalities). Following Mutirwara et al. (2018) the longest whisker sampled for each animal was selected to standardise the whisker position and to allow the longest period of diet assimilation for comparative analysis.

Sample preparation and stable isotope analysis

Sectioned whiskers were degreased for 24 h using a cleaning solution of chloroform, methanol and distilled water in a ratio of 2:1:0.8 as described in Bligh and Dyer (1959) and following Lee-Thorp et al. (1989). The whisker segment was then rinsed with 1 ml distilled water three times and dried at 40 °C for at least 24 h in a drying oven (Lasec Series 2000). Cleaned samples were processed in the Archaeology Department’s Isotope Laboratory at the University of Cape Town. The whisker samples were weighed (0.3 – 0.5 mg) using a micro balance (Sartorius, model M2P micro balance), where more distal sections were combined to make up the weight if required, before being folded into tin cups to 1 mcg accuracy. These were then combusted in a Delta V Plus organic elemental analyser/isotope ratio mass spectrometer (IRMS) via a Conflo IV gas control unit (Thermo Scientific, Germany). The calibrated standards used by both laboratories were DL Valine, Merck Gel, sucrose and Choc, which are calibrated against International Atomic Energy Agency (IAEA) standards. Nitrogen is expressed relative to atmospheric nitrogen, carbon is expressed relative to Pee-Dee belemnite (Ben-David and Flaherty 2012).