Investigating vegetation of saline lands around Urmia Lake using satellite images

TThis study aims to survey vegetation canopy cover of saline lands around Urmia lake using satellites data in 2014. This study was conducted in the area about 353150 hectares in saline lands around Urmia lake. In this study, various pre-processing including geometric correction using topographic maps and atmospheric correction was the dark object subtraction. Field data was collected in 1393.Samplings were done in plant units with stratified random sampling method. In this study, multiple regression was used to examine the relationship between variables. In this model, first by calculating the correlation matrix, the presence or absence of collinearity between the independent variables (vegetation indices and bonds) was studied. In this study, the best model was chosen due to the higher value of R2.On the selected model several variables have been chosen from among all independent variables and were introduced as the most important factors in determinig the vegetation cover map of the study area. Finally the produced maps and sampling points were controlled in order to validate the accuracy of results. The results showed that the indices of MIRVI, PD312, PD322, TVI, VNIRI, INFRARED, MID-IR-INDEX and bounds of b2, b3, b5, b6 have significant correlation at the 5% level with the field data and by using backward regression models the canopy cover map of vegetation was estimated. According to results much of saline lands around Urmia lake have canopy cover of 0-20 percent therefore we concluded that the rangelands in the region are not in good condition.