Stereotypic route tracing in captive Carnivora is predicted by species-typical home range sizes and hunting styles

Abstract

General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Full terms of use are available: http://www.bristol.ac.uk/pure/about/ebr-terms 1 Stereotypic route-tracing in captive Carnivora is predicted by species-typical 1 home range sizes and hunting styles 2 3 Abstract 4In captive conditions (e.g. zoos), some Carnivora species typically show negligible 5 stereotypic behaviour (SB) and reproduce successfully, while others tend to 6 reproduce poorly and be very stereotypic. We used comparative methods to identify 7 species-level risk factors for SB and captive infant mortality (CIM). Candidate 8 predictor variables were natural ranging behaviour, territoriality, aspects of natural 9 foraging, wild activity levels, cranial volume, and IUCN Red List status. Previous 10 research had identified naturally long daily travel distances, and being large-bodied 11 and wide-ranging, as SB risk factors. We nearly doubled the size of this original SB 12 database, and then imposed stricter quality controls (e.g. on minimum sample sizes 13 for inclusion). Analysing the resulting 23-species dataset confirmed naturally large 14 ranges and travel distances as risk factors. It also showed that the range size effect: 15 is independent of body mass (although body mass and range size together predicted 16 SB most strongly); is stronger for stereotypic route-tracing (e.g. pacing) than for all 17 SB forms combined; and explains the apparent daily travel distance effect (which 18 vanished when range size was controlled for). Furthermore, a new finding emerged: 19 that naturally long chase distances during hunts also predicted more severe route-20 tracing. Turning to CIM, previous research had also identified naturally long travel 21 distances and large home ranges as risk factors. We failed to replicate this, or to 22 confidently identify any species-level risk factor (despite CIM significantly varying 23 between related species, at least for Canidae and Ursidae

Phylogenetic comparative methods: Harnessing the power of species diversity to investigate welfare issues in captive wild animals

“…Capitalizing on new PIC software, a larger database, and an updated phylogeny, Kroshko et al () replicated this work. They confirmed the relationship between route‐tracing and large daily travel distances/home range sizes, but found the latter no longer depended on body size, and that the daily travel distance effect was a mere by‐product of home range size.…”

“…Practical ways to minimize such effects of intra‐specific variation are to use data from as many individuals from as many locations as feasible; and/or to impose a minimum sample size of individuals per species (on determining suitable intra‐specific sample sizes and for analyses overall see: Garamszegi, ). For example, in the Kroshko et al () and McDonald Kinkaid () studies, for inclusion each species had to be represented by ≥5 individuals; while in Müller et al (, ) the minimum was 45. Another possible solution, though not always feasible, is to statistically account for intra‐specific variation and sampling error (Symonds & Blomberg, ), which then improves the estimation of parameters when models are run (Ives et al, )—an approach touched on further below.…”

“…CIM values in the Carnivora studies (Clubb & Mason, , ; Kroshko et al, ) were extracted from International Zoo Yearbooks , which published IM reports. This practice stopped in 2000, however, making this data source increasingly historic.…”

“…Using ready‐published data is efficient, and can also allow coverage of diverse collections worldwide. The Carnivora studies extracted observational data on stereotypic behavior from 173 studies using this method (Clubb & Mason, , ; Kroshko et al, ), yielding data on over 1,300 individuals across 51 species from collections worldwide. For each species, values were summarized across individuals to calculate a single statistic: median % observations spent stereotyping by affected animals.…”

Predator size and prey size–gut capacity ratios determine kill frequency and carcass production in terrestrial carnivorous mammals

“…Kill frequencies that apply only to a specific prey species (e.g. the number of moose killed by wolves, irrespective of other prey taken in the same time period) were not taken into account since these estimates did not consider all prey species hunted by the carnivore (Kroshko et al 2016). Per carnivore species, the average of most common prey mass (M prey ), the average N Pack and the average actually observed kill frequency were calculated.…”