Presentation of Suitable Model to Estimate Vegetation Fraction Using Satellite Images in Arid Region (Case Study: Sadough- Yazd, Iran)

Abstract

One of the influential tools concerning the rangeland and vegetation sciences is<br />the technology of remote sensing and satellite data. Satellite data have played essential<br />roles in preparing the needed information for studying the vegetation. Vegetation has been<br />widely recognized as one of the best indicators for determining the land conditions. Using<br />vegetation indices is one of the techniques of remote sensing to study the vegetation. The<br />purpose of this research is to study the vegetation indices and also provide an appropriate<br />mathematical model for estimating the vegetation fraction using the mentioned indicators.<br />In this research, different vegetation indices were calculated by Landsat TM (2006)<br />satellite images in order to evaluate their capability to estimate the vegetation in the arid<br />regions. At the outset, vegetation fraction was measured using the sample quadrates and<br />then the measured data were compared with the values of digital numbers for the site<br />pixels. Multiple regression analysis was utilized for the actual values and parameters to<br />select the validation and optimization models. Indices which have been calculated include<br />MSR, DVI, TRVI, SARVI, SR, RDVI, DVI, NDVI, SAVI, ARVI and MSAVI. The results<br />showed that the model in which ARVI was used had the highest accuracy (R2=0.86) and<br />therefore, it was chosen as the most suitable model for the estimation of vegetation fraction<br />in the study area. One accomplishment of the above results is to suggest some<br />mathematical equations with those indices to estimate the amount of vegetation. These<br />results provide a strong foundation for the use of some vegetation indices for assessing the<br />vegetation cover in central Iran.