Estimation of the land use changes using satellite data in an under management area of mountain forest

Total RMSE (Root-Mean-Square Error) in geometric correction was obtained as 0.42 and 0.36 for images of 1999 and 2013, respectively. Then, the best band combination was specified in the images of 1999 and 2013 which was the combination of 2, 3, and 5 bands in the image of 1999 and the combination of 3, 6, and 7 bands in the image of 2013). The classification was performed and results showed a reduction in the area of residential use classes from 372.87 to 329.31, a reduction in dense forest from 1126.44 to 844.11, an increase in tea gardens from 1225.17 to 1519.92, and an increase in semi-dense forest from 121.59 to 152.73 hectares. Investigating the maximum likelihood classification also showed that the images of 1999 with kappa coefficient of 95.51 and the overall accuracy of 96. 80 % have higher accuracy than the images of 2013 with kappa coefficient of 89.30 and overall accuracy of 92.80%. During the recent decades, deciduous forests have been molested by human intervention. Easy access, abundance and diversity of valuable forest products have led to increased population, creating new residential areas and deforestation activities and use satellite data suggest to management natural resources.