Vegetation is an essential element of the land surface system that links climate change. Water availability is a determining factor in vegetation dynamics. Both environmental and anthropogenic factors are effected on vegetation dynamics. There are several remote sensing indices for indicating vegetation. Normalized Difference Vegetation Index (NDVI) is a common and widely used index. NDVI is highly sensitive to ecosystem conditions; therefore, it can be representative of detecting changes in vegetation activity. The visible and near infrared bands on the satellite multi spectral sensors have monitored the greenness of vegetation. In this research, we present the trend analyses of 31 years (1982-2012) remote sensing vegetation index based on long-term time series of NDVI observation from the Global Inventory Modeling and Mapping studies (GIMSS) group derived from NOAA AVHRR imagery with 0.08 degree spatial resolution over Iran (40 t0 65 East, 25- 45 North). Iran has a dry climate characterized by long, hot, dry summers and short, cool winters. NDVI has a strong seasonal cycle, and In this research we only used the mean growth season NDVI. We depict greening (NDVI increase) and browning (NDVI decrease) regions. NDVI, an indicator of vegetation growth and coverage, has been described the characteristics of land use and land cover. In this research, we also have investigated NDVI changes in different land use to to find out the relationship between land use and NDVI trend. The existence of positive autocorrelation in the time series increases the probability of detecting trends when actually none exist, and vice versa. In this study, the effect of autocorrelation on the variance of the Mann-Kendall trend test statistic is considered, therefore we have used modified Man-Kendall method to do trend analyses. According to the results, almost regions have negative NDVI trend and Poor range became browner and forest became greener
