At the Mediterranean region, the problems related to the agriculture, the defense of the soil and watershed management, are one of the most significant environmental problems. To solve this problem it is necessary to study, analyze and understand the impact of the erosive action in these semi-arid areas. There are different methods that estimate soil erosion at the plot or watershed level; these methods range from simple to complex. For this work, the analysis has focused on modelling of water erosion by the RUSLE model (Revised Universal Soil Loss Equation) consisting of a modified version of the universal soil loss equation of (USLE) originally developed by Wischmeier and Smith in Predicting rainfail erosion losses—a guide to conservation planning. Agricultural handbook No. 537, USDA, Washington, 1978. A methodology of assessing the various risk indices of erosion around the geographic information system (GIS-Arc Gis 9.3). The RUSLE equation was applied by overlaying, using ArcGIS software, the various appropriate thematic maps for each factor. According to the RUSLE model, over 70% of the study areas are subject to different soil loss levels: low (<5) to very high (50–200). The average loss by water erosion across all study sites is about 8.73 t/ha/year. Total annual losses for the six selected townships are of the order of 33,995.4 t/ha/year. In light of these figures, we can conclude that the unequal distribution of areas with potential risk of erosion in the Aurès center results from the variability of the characteristics of the factors involved in the process and the model used. These results will be used for territorial planning, in reference to the project to reduce the vulnerability of these areas.

La présente étude s’inscrit dans le cadre d’une caractérisation de la dynamique spatiotemporelle de l’occupation du sol entre 1987 et 2014 d’une zone forestière située dans l’Aurès oriental (Algérie). La méthode de comparaison diachronique des images satellitaires Landsat a été adoptée. Les résultats ont permis de révéler une évolution régressive de la forêt aurasienne au cours des 27 dernières années (de 1987 à 2014). Cette régression de la couverture végétale est localisée essentiellement au niveau des peuplements de Pin d’Alep qui prédominent dans la région d’étude. La dégradation actuelle serait liée aux changements climatiques récents et amplifiée par l’action anthropique.

Desertification is the gradual and sustained reduction in the quantity and quality of the biological productivity of arid and semi-arid land.

The study area is located in the North Eastern part of Algeria, it has a rich heritage in its biodiversity, however weather conditions and adverse human reality, induce a degradation of the physical environment in the form of a regression of vegetation cover. To assess desertification in our study area map of desertification sensitivity is a tool for decision support.

For the realization of this Map we used the ArcGis software applied a methodology which is inspired by the concept MEDALUS (Mediterranean Desertification and Land Use, 1999) by crossing four thematic layers that may have an impact on the process of desertification.

The results of Cartography and statistical analysis permit the classification of our region in terms of sensitivity to desertification in four very important classes. (Not affected, Insensitive, Sensitive and highly sensitive).

More than 69.92% of the surface area were classified sensitive to very sensitive, For against 30.07% is classified in unallocated insensitive.

Planning restoration work and the fight against desertification are expected to limit the risk of desertification in the study area perspectives.