The transition zone between the Tellian and Saharan Atlas of the Tiaret region in Algeria is believed to be undergoing land degradation processes because of the increase in population needs, which causes a weakening of the biological potential and generates ecological and socio-economic imbalances. The lack of periodically acquired data on land cover is a major handicap to understand and diagnose the state of this transition zone. The goal of our work was to reconstruct past annual land cover changes in order to assess the best monitoring tools for the current and future situations. We used an annual series of Landsat images, acquired between 1984 and 2017 via the Google Earth Engine platform. We then classified relevant land covers by applying a SVM (Support Vector Machine) classification algorithm to this multi-spectral data set. We also integrated other derived data such as vegetation, water and color indices, texture measures and TC transformationsin order to improve classification accuracy. The resulting 34 high-resolution land cover maps for the 1984–2017 period precisely characterize the recent dynamics and status of the transition zone of the Tellian and Saharan Atlas in the steppe region of Tiaret, and allow to explore land cover and climatic trends.