Academic journal article The American Midland Naturalist

Relationships between Landscape Pattern and Space Use of Three Mammalian Carnivores in Central Mississippi

Academic journal article The American Midland Naturalist

Relationships between Landscape Pattern and Space Use of Three Mammalian Carnivores in Central Mississippi

Article excerpt

ABSTRACT.-

Space use and habitat selection have been studied extensively in mammalian carnivores, and it is widely accepted that many factors influence these parameters. Although landscape characteristics are perceived to be important in how carnivores use their environment, relatively few studies have evaluated influences of landscape features on carnivore behavior. We examined relationships between landscape pattern and space use in sympatric bobcats (Lynx rufus), coyotes (Canis latrans) and gray foxes (Urocyon cinereoargenteus) in central Mississippi during 1989-1997. We quantified landscape features within three different spatial scales (3000, 4500 and 6000 m) centered on home ranges of 46 bobcats, 12 coyotes and 22 gray foxes. We evaluated a suite of class- and landscape-level variables and their influence on space use of each species. Linear mixed models suggested that habitat interspersion (intermixing), patch shape (complexity) and edge contrast were important predictors of space use in bobcats. We failed to detect any relationships between landscape pattern and space use of coyotes and gray foxes. We offer that the relationships we observed are related to, and can be explained by, ecology and behavioral adaptations of each species.

INTRODUCTION

Anthropogenic and natural disturbance regimes, variation in the physical environment and local biotic processes interact at all spatial scales to create spatial heterogeneity (Levin, 1992; Turner et al., 2001; Bissonette, 2003). At the landscape scale, heterogeneity (i.e., pattern) can be denned generally as the composition and spatial configuration of a landscape (Li and Reynolds, 1994; McGarigal and McComb, 1995). The fundamental premise of landscape ecology is that landscape pattern can have important constraints on biotic (e.g., movement patterns of individuals) and abiotic (e.g., nutrient flows) processes in ecological systems (Li and Reynolds, 1994; Turner et al, 2001).

In general, we know little of how mammalian carnivores respond to changes in landscape pattern (Sargeant et al., 1998; Crooks, 2002). Most landscape-scale studies of carnivores have focused on edge-effects and the probability of nest predation (e.g., Chalfoun et al., 2002; Dijak and Thompson, 2002; Kuehl and Clark, 2002; Stephens et al, 2003). Relationships between space use and other aspects of landscape pattern have not been well-studied. For example, home range size can be an important indicator of food density and efficiency of movement (Schoener, 1971), but constraints imposed on home range size by the spatial arrangement of resources and shelter are largely unknown (Ims, 1995; Kie et al., 2002).

Bobcats (Lynx rufus), coyotes (Canis latrans) and gray foxes (Urocyon cinereoargenteus) are sympatric over much of North America and typically show some overlap in diet and space use (Bekoff, 1977; Fritzell and Haroldson, 1982; Larivière and Walton, 1997; Chamberlain, 1999; Neale and Sacks, 2001). Nevertheless, differences in body size and life history requirements suggest that home range size in the three species would not show equivalent shifts with respect to changes in landscape pattern (Crooks, 2002). Moreover, responses by the three species may differ depending on the generality of the landscape metrics used. The most inclusive descriptors of spatial pattern are landscape-level metrics, which represent the structure of the entire landscape mosaic (McGarigal et al., 2002); we would expect these to be suitable for generalist species (i.e., coyotes). However, landscape-level metrics may contain non-crucial information (Tischendorf, 2001) that would obscure the relationship between pattern and process in species with more restricted diet and habitat requirements (i.e., bobcats). Therefore, home range size may be predicted better by class-level metrics, which represent amount and spatial distribution of a single patch type (McGarigal et al., 2002; Tischendorf, 2001). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.