Functional feeding responses of coyotes, Canis latrans, to fl uctuating prey abundance inthe Curlew Valley, Utah, 1977â•fi1993

We investigated interactions between coyotes (Canis latrans Say, 1823) and prey in the Curlew Valley, Utah, by comparing prey abundances with prey consumption rates. Previous studies reported a cyclic trend in black-tailed jackrabbit (Lepus californicus Gray, 1837) density with a period of 10 years and >150-fold amplitude, as well as short-term fl uctuations among some rodent species that exceeded an 8-fold difference in amplitude over 2 years. Our results suggest changes in coyote diets mainly refl ect the fl uctuations in jackrabbit abundance. Prey switching to rodents during periods of low jackrabbit abundance also was evident. We used the initial feeding pattern analysis to compare prey consumption rates to prey abundance. Coyotes demonstrated a type II (hyperbolic) functional feeding response to changes in jackrabbit abundance. Functional feeding responses to rodent abundances were more diffi cult to assess because of the strong infl uence of jackrabbits. In most comparisons, we visually detected a linear functional feeding response to varying rodent abundances; yet this was not statistically supported by Akaike’s Information Criterion corrected for small sample sizes (AICc) to assess different models. Résumé : Nous avons étudié les interactions entre les coyotes (Canis latrans Say, 1823) et leurs proies dans la vallée Curlew, Utah, en comparant l’abondance des proies et les taux d’ingestion de ces proies. Des études antérieures ont signalé des tendances cycliques dans la densité des lièvres de Californie (Lepus californicus Gray, 1837) avec une période de 10 années et une amplitude de >150 fois, ainsi que des fl uctuations à court terme de certaines espèces de rongeurs avec des différences d’amplitude de plus de 8 fois sur 2 années. Nos résultats indiquent que les changements dans le régime alimentaire des coyotes refl ètent surtout la fl uctuation d’abondance du lièvre. Il y a aussi des indications de changement de proies favorisant les rongeurs durant les périodes de faible abondance des lièvres. Une analyse des patrons d’alimentation initiale (initial feeding pattern analysis) nous a permis de comparer les taux de consommation des proies et l’abondance de celles-ci. Les coyotes ont une réponse fonctionnelle alimentaire de type II (hyperbolique) en réaction aux changements d’abondance des lièvres. Les réponses fonctionnelles aux abondances de rongeurs sont plus diffi ciles à déterminer à cause de la forte infl uence des lièvres. Dans la plupart des comparaisons, nous détectons visuellement une réponse fonctionnelle alimentaire linéaire aux variations d’abondance des rongeurs, mais ce n’est pas appuyé statistiquement par le critère d’information d’Akaike (corrigé pour les petits échantillons; AICc) utilisé pour évaluer les différents modèles. Published in Canadian Journal of Zoology 83: 569–578 (2005) doi: 10.1139/Z05-039 © 2005 NRC Canada Received 14 December 2004. Accepted 15 March 2005. Published on the NRC Research Press Web site at http://cjz.nrc.ca on 8 June 2005. R.A. Bartel: Department of Forest, Range, and Wildlife Sciences, Utah State University, Logan, UT 84322-5230, USA; Corresponding author (e-mail: [email protected]); Present address: Department of Zoology, North Carolina State University, Raleigh, NC 27695-7617, USA. F.F. Knowlton: United States Department of Agriculture, Wildlife Services, National Wildlife Research Center, Logan, UT 84322-5295, USA. 569 570 Bartel & Knowlton in Canadian Journal of Zoology (2005) depiction is similar to type II (hyperbolic), except that the intersection of the increasing line and horizontal line is sharp versus the smoother transition in type II. Most authors refer to type I as linear without the asymptote (Turchin 2003). It has been suggested that this type of response is rare and only occurs when prey handling time is trivial or predation is a matter of chance encounter (Knowlton and Stoddart 1992; Messier 1995). In a type II response, the number of prey killed increases, but the proportion of overall prey consumed per predator decreases at higher densities (Murdoch 1973; O’Donoghue et al. 1998). This deceleration may result from predator satiation (Murdoch 1973) or an adaptive adjustment of search rates (Abrams 1990). Type II responses have been previously suggested to describe one of the relationships between coyotes and black-tailed jackrabbits, Lepus californicus Gray, 1837 (Hoffman 1979; Stoddart et al. 2001). Type III responses are sigmoidal in shape and are typical of generalist predators (Keith et al. 1977). Type III responses could result from a number of possible interactions: (i) predators learning to recognize, capture, and (or) handle prey more effi ciently as prey density increases; (ii) predators switching prey types; (iii) adaptive variation in foraging rates; or (iv) changes in prey behavior or vulnerability (O’Donoghue et al. 1998).