Domestic cats (Felis catus) are a conservation concern because they kill billions of native prey each year, but without spatial context the ecological importance of pets as predators remains uncertain. We worked with citizen scientists to track 925 pet cats from six countries, finding remarkably small home ranges (3.6 AE 5.6 ha). Only three cats ranged > 1 km 2 and we found no relationship between home range size and the presence of larger native predators (i.e. coyotes, Canis latrans). Most (75%) cats used primarily (90%) disturbed habitats. Owners reported that their pets killed an average of 3.5 prey items/month, leading to an estimated ecological impact per cat of 14.2-38.9 prey ha À1 yr À1. This is similar or higher than the peranimal ecological impact of wild carnivores but the effect is amplified by the high density of cats in neighborhoods. As a result, pet cats around the world have an ecological impact greater than native predators but concentrated within~100 m of their homes.

Domestic cats preying on wildlife is a frequent conservation concern but typical approaches for assessing impacts rely on owner reports of prey returned home, which can be biased by inaccurate reporting or by cats consuming prey instead of bringing it home. Isotopes offer an alternative way to quantify broad differences in animal diets. By obtaining samples of pet food from cat owners we predicted that we would have high power to identify cats feeding on wild birds or mammals, given that pet food is thought to have higher C isotope values, due to the pervasive use of corn and/or corn by-products as food ingredients, than native prey. We worked with citizen scientists to quantify the isotopes of 202 cat hair samples and 239 pet food samples from the US and UK. We also characterized the isotopes of 11 likely native prey species from the southeastern US and used mixing models to assess the diet of 47 cats from the same region. Variation in C and N isotope values for cat food was very high, even within the same brand/flavor, suggesting that pet food manufacturers use a wide range of ingredients, and that these may change over time. Cat food and cat hair from the UK had lower C values than the US, presumably reflecting differences in the amount of corn used in the food chains of the two countries. This high variation in pet food reduced our ability to classify cats as hunters of native prey, such that only 43% of the animals could be confidently assigned. If feral or free ranging cats were considered, this uncertainty would be even higher as pet food types would be unknown. Our results question the general assumption that anthropogenic foods always have high C isotope values, because of the high variability we documented within one product type (cat food) and between countries (US vs. UK), and emphasize the need to test a variety of standards before making conclusions from isotope ecology studies. How to cite this article McDonald BW, Perkins T, Dunn RR, McDonald J, Cole H, Feranec RS, Kays R. 2020. High variability within pet foods prevents the identification of native species in pet cats' diets using isotopic evaluation. PeerJ 8:e8337 http://doi.Lindblad-Toh K. 2013. The genomic signature of dog domestication reveals adaptation to a starch-rich diet. Nature 495(7441):360-364 . Caut S, Angulo E, Courchamp F. 2009. Variation in discrimination factors (δ15N and δ13C): the effect of diet isotopic values and applications for diet reconstruction.

No abstract