FOREST LOSS ANALYSIS AND ESTIMATION OF FUTURE FOREST PRESENCE BY GEOGRAPHICAL INFORMATION SYSTEMS AND REMOTE SENSING
Project no. 652 · Status: Kapanmış · 2015
Summary
The forests are one of the most important sources of human life and ecological system because of their various uses from live to energy supply and their contribution to reduce environmental impacts globally from soil erosion to climate change. However, forest loss (deforestation) is a growing global problem with factors as urbanization, agriculture, industrial facilities, mining activities, uses for paper / furniture and / or energy, and fires. Strategies to be developed in order to solve this problem require that the change of forest entity be known. In the project, deforestation detection and future estimation of Forest distribution will be realized by Geographic Information Systems (GIS) and Remote Sensing (RS) techniques. The multi-temporal satellite images will be classified, and the spatial features will be visualized. Changes of the spatial features will be identified by the Land Change Modeler (LCM), a GIS technique. Since the study is related with forests, "forest" and "non-forest" legends will be used as spatial features in the classification. Thanks to this module which is used efficiently for ecological sustainability, the change between the maps of different time periods with the same legend can be analyzed both graphically, numerically and visually. Thanks to LCM running on a neural network system, deforestation can be detected and mapped very efficiently if it exists. After all these determinations, future estimation of the forested areas will be done. MARKOV formulation embedded in the LCM module will be used for estimation. Thus, near-real-time stochastic probability and estimation model will be created.