Evaluation of Forest Degradation Using Google Earth Engine (case study forest of Sidi Bel Abbes)

Cloud-computing engine using various classification techniques that include Random Forest (RF), Support Vector Machine (SVM), and algorithms to produce a multi-date land use land cover map of forest Sidi Bel Abbes region (2000-2020). Landsat images (ETM+ and OLI) will be used for the thematic classification of forest formations. The outcome demonstrated a reliable GEE-based mapping of the region's forest with satisfactory classification accuracy. It revealed the overall accuracy values and the Kappa coefficients to be above 77% during the different time nodes under study. The research further revealed that the Random Forest performed the best machine-learning models tested in this study for mapping the forest and other land cover classes. At the end of this work, change maps will be drawn up. This approach will enable us to detect spatiotemporal changes in the forest landscape, establish conversion relationships between different land-use categories, and determine the direction, speed, and rate of land-use change for each defined period. This work will provide forest managers and other stakeholders with a decision-making tool that will enable them to plan the efficient and sustainable management and development of their respective areas.