The Monitoring and Dynamic Analysis for Otindag Sandy Land based on Object-Oriented Method

Abstract

Otindag Sandy land is one of the major areas of the Beijing and Tianjin sandstorm resource project. The dynamic change analysis for desertification land is conducive to the whole project implementation and adjustment. Therefore, in order to provide decision support for desertification land monitoring, fractal network evolution (FNEA) segmentation was applied to GF-1 data, and decision tree and support vector machine (SVM) classification methods were used to analyze the sandy land dynamic change in Otindag sandy land. The results showed that the total area of Otindag sandy land was 20542 km2 in 2013, which reduced by 5.48%, compared with that in 2002. It demonstrated that the Beijing and Tianjin sandstorm resource project had played a positive effect in nearly a decade, but some areas such as the hinderland of the Otindag sandy land were still deteriorated. The second phase of the Beijing and tianjin sandstorm resource program has been launched, it was wished to be useful for the further improvement and providing basic decision-making.