As one of the most destructive weather phenomena, drought has a natural but complicated nature that influences all aspects of the human life in all climates in a hidden way. Considering many drought occurred in Isfahan province, this study aimed to detect the effects of drought on vegetation changes in a 905000 ha area in the western part of the province in past four decades using remote sensing and climate data. For this purpose, climate data and Landsat time series satellite images including MSS 1973 and 1976, TM 1985, 1992, 1998 and 2009, ETM+ 2000 and 2006 and a 2013 Landsat 8 image which were a good representatives of dry, normal and wet conditions, were selected. Then with use of standardized precipitation index (SPI) and image differencing of normalized vegetation index (NDVI) reclassifying NDVI values, the amount of vegetation changes were determined in eight periods based on satellite images. Finally, spatial correlation technique was used to investigate relationships between SPI and NDVI indices. The results of SPI in the scales of 3, 6 and 12 months leading up to the month of taking the satellite images, have demonstrated that the year 2000 was the driest and the year 1992 the wettest year among the available stations in the study area. The results of change detection analysis showed that the maximum increase in vegetation has occurred from 1973 to 1998. In this period, vegetation cover increased 117893 ha and the maximum decrease in vegetation of the area has occurred from 1992 to 2013 in which 99976 ha decrease was observed in vegetation cover. An analysis of mostly influenced areas by drought showed that nearly after the year 2000, the vegetation cover of these areas has experienced a great decrease until 2013 and this status has reached its peak from 2009 to 2013 in most of the regions including Isfahan, Tiran-Karvan and Shahinshar. Considering vegetation changes inside and around the urban areas, it seems that population increase and urban development have an important role in vegetation changes in comparison with drought in the study area. The spatial correlation between SPI and NDVI ranged from 0.76 to 0.98 in the entire study area and naturally vegetated areas, respectively, which indicates the high potential of the indices in natural lands. Overall, the results of this study indicated that the integration of remote sensing and climatological techniques is a very suitable methodology for drought mapping and monitoring and it can be used as one of the most applicable techniques in optimal management of vegetation cover and drought phenomenon in broad areas. Key Words: SPI, NDVI, Drought monitoring, Change detection, Image differencing, Spatial correlation
