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

Accueil du site → Doctorat → Israel → The Impact of landscape level attributes on livestock distribution in a semiarid environment

Ben-Gurion University of the Negev (2014)

The Impact of landscape level attributes on livestock distribution in a semiarid environment

Hertz, Maya Zehavi

Titre : The Impact of landscape level attributes on livestock distribution in a semiarid environment

Auteur : Hertz, Maya Zehavi

Etablissement de soutenance : Ben-Gurion University of the Negev

Grade : Doctor of Philosophy (PhD) 2014

Résumé partiel
There is a broad, long-term trend of increasing grazing pressure on many of the world’s semi-arid rangelands. More studies that combine long-term and detailed data collection, analysis and modeling are required to develop sustainable grazing strategies for these areas. The issue concerns not only the total number of animals maintained in a given region but also the spatial dimension. Uneven distribution of grazing pressure across the landscape is one of the main causes of desertification in arid regions. In this dissertation, we studied spatio-temporal variability of herd presence, and the timeline of grazing behavior as affected by different landscape attributes. We documented the grazing patterns of small ruminant herds of sheep and goats and generated quantitative maps of herd presence and grazing pressure. We analyzed the uniformity of herd presence across the landscape, site-use preferences and animal behavior as affected by biotic (herbaceous biomass, shrub cover) and abiotic (geo-spatial) characteristics of the landscape. The general aim of this research was to quantify and understand the spatio-temporal distribution of grazing animals as affected by landscape attributes. This general aim led to the development of the following four specific objectives : (1) To determine the effect of abiotic landscape attributes on animal movement ; (2) To determine the effect of biotic landscape attributes, expressed by herbaceous primary production and shrub cover, on animal movement and grazing-site preference ; (3) To quantify grazing intensity in a semiarid rangeland in a spatially explicit way ; and (4) To study the detailed time-line of animal grazing behavior as affected by landscape attributes and temporal changes. Combining GPS and GIS technologies with an innovative technology to monitor grazing activity can improve our ability to gather data on grazing animals in remote locations and across time. This should foster a better understanding of how spatial and temporal variability of animal presence, primary production and landscape features affect grazing preferences and rangeland utilization over long periods of time. Likewise, it should provide new insights into rangeland/animal management to improve efficiency of use and minimize negative effects of grazing animals in semiarid ecosystems. The study was conducted on two farms located in the northern Negev : (1) Bedouin demonstration farm, a Long Term Ecological Research (LTER) site, near Lehavim ; and (2) the Ella farm near Kibbutz Shomriya. At both sites animals are managed as separate grazing herds, each accompanied by a shepherd. The movement of four herds was monitored during the years 2003/04 to 2012/13 using custom-built GPS harnesses and the data were analyzed using spatial models and geographic information systems (GIS). In one of the study years, foraging behavior of two herds at the Ella farm was monitored using both GPS and acoustic monitoring. The first part of this study focused on the Ella site, where data were collected during 2008/09, 2009/10 and 2010/11. We monitored two herds : a smaller herd of 320 head (designated S) comprising lactating goats (wet), and a larger herd of 550 head (designated L) of sheep and dry goats. A GPS device carried by one animal recorded the location of the herd during the course of each daily foraging excursion at intervals of 1 minute. Throughout the study we documented 541 foraging routes (herd–day combinations), and these data were imported to GIS. We analyzed the uniformity of presence across the rangelands, the abiotic factors associated with herd presence, and grazing spatial efficiency. Herd distribution was neither random nor uniform across the landscape, with many areas not visited at all, and others receiving a higher number of visits than expected. Herd presence on the north-facing slopes and wadi shoulders was less than expected on the basis of the area of these topographically-defined habitats. Herds visited less than expected areas of steep topographic slopes (> 10°–12°) and more than expected areas of more moderate slope. In relation to expected herd presence based on distance from the central corral, herd presence was elevated near the corral (within about 1.2 km), and reduced in areas exceeding this distance. The mean velocity of movement was 0.51 m/s but velocity was affected strongly by path angle. Animals moved relatively slowly on steeper slopes, both uphill and downhill, and at quicker paces on moderate slopes

Présentation (BGU)

Version intégrale

Page publiée le 17 décembre 2017, mise à jour le 21 novembre 2018