Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Master → Israel → Predicting movement pathways using non-invasive methods : the reintroduced Asiatic wild ass in the Negev

Ben Gurion University of the Negev (2011)

Predicting movement pathways using non-invasive methods : the reintroduced Asiatic wild ass in the Negev

Davidson, Achiad

Titre : Predicting movement pathways using non-invasive methods : the reintroduced Asiatic wild ass in the Negev

Auteur : Davidson, Achiad

Université de soutenance : Ben Gurion University of the Negev

Grade : Master of Arts (MA) 2011

Résumé
Our knowledge of large scale animal movements in complex landscapes is very limited. Understanding mechanisms affecting animal movement patterns is essential for predicting and conserving movement corridors between population core areas. Successfully reintroduced species offer a unique opportunity to study the effect of different landscape features on population range expansion. My research goal was to predict the movement pathways of the reintroduced Asiatic wild ass (Equus hemionus) between population core areas in the Negev desert by understanding the effect of landscape factors on their movement patterns. I developed a set of 40 a-priori alternative models to explain the effect of landscape factors on the movement patterns of the wild ass. Then I constructed six GIS (Geographic information system) models that were based on my basic a-priori models that took into account only the main landscape factors without the interactions between them : topography (slope, canyons), vegetation, water sources and 4WD trails. The outputs were least cost pathways that connect between six population core areas. I surveyed dung density using transects along each least-cost pathway, as an index of pathway use. I combined the transects, their feces counts and their landscape characteristics, and grouped them into three data sets : a) “Entire terrain” : transects from the whole study site (n=128) ; b) “Landscape barriers” : only transects from mountain ridges (n=54), which were used in order to study long distance movement pathways ; c) “Open landscape” : transects that are not from the mountain ridges (n=74). I constructed General Linear Models (GLMs) from the three data sets. I selected the best a-priori models that fit the empirical data the most using an information-theoretic approach. For the “open landscape“ and “entire terrain” data sets the GLMs that gave the best fit to the data included vegetation coverage and distance from water sources. However, in the “Landscape barriers” the best GLMs included canyons indicating that the wild ass pass mountain ridges through canyons. The presence of man-made 4WD trails was also found to be an important factor in landscape barriers, but their effect is a function of the terrain : There was a positive effect on movement when crossing a mountain ridge along relatively steep slopes ; However, a pronounced negative effect was found when 4WD trails go through narrow canyons which are a natural movement corridor for wild ass movement. Among the GIS least cost models, the most used pathway in the entire area and in the open landscape was the vegetation coverage model (Kruskal-Wallis test ; P=0.02, P=0.08, respectively). In the “landscape barriers” the most used least cost pathway was the slope pathway (P=0.059). I conclude that a number of landscape factors have considerable effect on the wild ass movement, but the type and magnitude of the effect is a function of the terrain. In open landscapes – vegetation and water sources are preferred whereas in mountain ridges (landscape barriers) – canyons with no 4WD trails are preferred. Our method based on least-cost models, feces surveys and strategy of the informationtheoretic approach enabled the assessment of complex landscape factors that facilitate wild ass movements. This new original methodology could be used for non-invasive ecological studies of animal movement. The models that had the best fit to the empirical data can be used to predict movement pathways of the wild ass between population core areas. The identification of landscape factors that affect movement, as well as important pathways, could facilitate the selection of appropriate movement corridors for the wild ass. These findings should be considered in the conservation and management of the endangered Asiatic Wild Ass.

Mots clés : Onager — Israel — Negev ; Onager — Israel — Negev — Habitat

Présentation (BGU)

Version intégrale

Page publiée le 31 mars 2015, mise à jour le 22 novembre 2018