The evaluation of vegetation variations trend using linear regression methods and change vector analysis


Tarbiat Modares University


Vegetation is one of the vital indexes in the ecological life in the environment. Vegetation nearly covers 70% of the world land surfaces. The purpose of this research is to analyze the trend and attribute of vegetation change in our case study during a specific period of time using CVA (Change Vector Analysis) and linear regression methods. In this work, NDVI composites of MODIS have been used. After doing required preprocess, we used CVA to detect change magnitude and direction, and also and linear regression is used to shed light on the trend of the change. We study he change’s nature of 11 sequential years for four predetermined time periods (sixteen days, monthly, seasonal, and yearly) applying the mentioned methodologies. The slope change map, in warm periods, shows reduction of vegetation in all areas except in southern region and small part of north. While temperature rises, in the regions experiencing positive change in the beginning of the warm season, the vegetation decreases and tends to zero. The trend of change is lying across the down part of Tigris and Euphrates rivers, and in the spring season the north and border of Turkey with Iraq and Syria. The maximum of negative changes in this season is in the No. 1 dust source while in the No. 2 the change is positive and the changes lie in the No. 1 dust source in the summer. Maximum negative change is experienced in Mesopotamia and across the border of Iraq and Syria. The reason perhaps is because of rain shortage, construction of dams in Turkey, and manmade causes including war and immigration in the border of Iraq and Syria. Maps derived from CVA shows that most considerable changes occurred in the north strip of the image, the boundary parts of Turkey, Iraq, and Syria, and some years also in southern areas.