چکیده:
The aim of this research was to study the relationships between presence of plant species and environmental factors in Garizat rangelands of Yazd province and providing their predictive habitat models. After delimitation of the study area, sampling was performed using randomized-systematic method. Accordingly, vegetation data including presence and cover percentage were determined in each quadrate. The topographic conditions were recorded in plot locations. Soil samples were taken at depths of 0-30 and 30-80 cm in each plot. The measured soil variables included texture, lime, saturation moisture, gypsum, acidity (pH), ECe and soluble inos (Na+, K+, Mg2+, Cl-, Co3 2-, HCo3
- and So4 2-). Logistic regression technique was used to analyze the collected data. The results showed that the vegetation distribution is mainly related to soil characteristics such as texture, gravel, EC, gypsum, lime and OM. The presence of Artemisia sieberi- Zygophyllum eurypterum has relation with gravel, lime, available water and pH. Ephedra strobilaceae-Zygophyllum eurypterum has positive relation with gypsum. Rheum ribes-Artemisia sieberi has relation with clay and OM. Cornulaca monacantha has also relation with elevation above sea, gravel and gypsum. The presence of Seidlitzia rosmarinus has relation with lime. Electrical conductivity is the most factors effect on presence of Tamarix ramosissima.
خلاصه ماشینی:
Zare Chahoukib a Assistant Professor, Faculty of Natural Resources, University of Tehran, Iran b MSc. Graduate, Faculty of Natural Resources, University of Tehran, Iran Received: 29 November 2009; Received in revised form: 17 October 2010; Accepted: 10 November 2010 Abstract The aim of this research was to study the relationships between presence of plant species and environmental factors in Garizat rangelands of Yazd province and providing their predictive habitat models.
Introduction Predictive modeling of plant species’ distributions based on their relationship with environmental variables is important for a range of management activities.
g. Salisbury, 1926; Cain, 1944; Good, 1953; McArthur, 1972; Box, 1981; Stott, 1981; Walter, 1985; Woodward, 1987; Ellenberg, 1988).
In the other hands, logistic regression is one of the methods that can predict the probability of occurrence of each plant species related to site condition factors.
Model evaluation The best measure of agreement between observed (actual vegetation types) and predicted presence/absence is Kappa (Cohen 1960; Monserud and Leemans, 1992; Bell and Fiedling, 1997; Zimmermann and Guisan, 2000; Moisen and Frescino, 2002; Robertson et al.
In this study, predictive models of vegetation types were provided based on absence-presence of species, using logistic regression.
Comparing results of prediction with real vegetation map of the study area shows that in logistic regression method absence-presence of species, as a dependent variable, is an effective factor for predictive species modeling.
Based on the prediction models, it is possible to estimate the probability of presence/absence of plant species in response to environmental factors (He et al, 2007).