Evaluating the vegetation changes upon vegetation index by using remote sensing

The monitoring of vegetation changes has a fundamental role in planning and management of environment. There are various methods to determine the changes in a region using satellite images that each has advantages and limitations. The use of vegetation indices is one of the methods to detect the changes. The aim of this study was to evaluate four vegetation indices including NDVI, SAVI, RVI and WAVI. This research was performed in Qeshm Island using Landsat images during 2001 and 2014. In this research, ETM and OLI data were used. After calculating each indicator, 100 sample training points were used to assess the accuracy of indicators by ENVI.5.3. Four classes including bare land, mangrove forests, agriculture and water were classified. Based on Dlapyan & Smith method, the product accuracy and user accuracy for each class were evaluated. The results showed that the SAVI index with the highest kappa coefficient, 0.93 in 2014 and 0.83 in 2001, had the best results and WAVI index with the lowest kappa coefficient, 0.43 in 2001 and 0.81 in 2014, had the weakest results. To evaluate the changes, crosstab method was used .The results showed that during 13 years the area of mangrove forests and agricultural lands and natural vegetation of Qeshm Island increased up to 21% and 60%, respectively.