Habitat selection along an environmental gradient: Theoretical models with an example of Negev Desert rodents

Research output: Contribution to journalArticlepeer-review

20 Scopus citations

Abstract

We present a set of models of density-dependent habitat selection along an environmental gradient. The models are based on the assumption of ideal-free distribution and use an approach similar to Morris' isodar analysis technique. It was assumed that habitats along a gradient differ either qualitatively (efficiency of resource acquisition) or quantitatively (resource abundance) or in both parameters. Depending on the distribution of resource abundance and habitat quality along a gradient, the models predict density-independent or density-dependent habitat distribution with or without density-dependent habitat shifts. The models were tested using data on a desert rodent community of the central Negev Desert, Israel. In six of the nine rodent species (Gerbillus dasyurus, G. gerbillus, G. henleyi, Meriones crassus, Acomys cahirinus and A. russatus), direct estimations of habitat breadth and position fit well the predictions of one of the models based on density relationships. Of these nine species, four (Jaculus jaculus, Gerbillus dasyurus, G. gerbillus and Mus musculus) were found to be density-independent habitat selectors, four (Meriones crassus, Psammomys obesus, Acomys cahirinus and A. russatus) were density-dependent habitat selectors and one (Gerbillus henleyi) was a density-independent habitat selector in winter and a density-dependent habitat selector in summer.

Original languageEnglish
Pages (from-to)257-277
Number of pages21
JournalEvolutionary Ecology Research
Volume2
Issue number3
StatePublished - 1 Mar 2000

Keywords

  • Density-dependent habitat selection
  • Habitat selection model
  • Negev desert
  • Rodents

Fingerprint

Dive into the research topics of 'Habitat selection along an environmental gradient: Theoretical models with an example of Negev Desert rodents'. Together they form a unique fingerprint.

Cite this