Deforestation area estimation in China based on Landsat data

In this study, we used 30m Landsat TM data to develop regression models that estimate the deforestation area of China. At first, the Maximum Likelihood Classification method was used to achieve the image classification based on the priori knowledge from the field measured data over China. Secondly, high resolution SPOT-5 data was used to validate the classification precision of Landsat TM data and build the deforestation area models. All built model highly accurate, and the R² are bigger than 0.85. Finally, the deforestation area over China was estimated according to regression model developed.