Classification methods for detecting and evaluating changes in desertification-related features in arid and semiarid environments

Land cover, land use, soil salinization, and sand encroachment, which are desertification-indicating features, were integrated in a diachronic assessment, obtaining quantitative and qualitative information on the ecological state of the land, particularly degradation tendencies. In arid and semi-arid study areas of Algeria and Tunisia, sustainable development requires the understanding of these dynamics as it withstands the monitoring of desertification processes. Both visual interpretation and automated classification approaches have been set up for salt and sand features extraction using historical and present Landsat imagery. The automated one includes a decision tree classifier and an unsupervised classification applied to the principal components extracted from Knepper ratios composite. New spectral indices are employed in the decision tree classifier for the extraction of features of interest. The validation of the classification methods showed that the decision tree had an overall accuracy over 85% in both areas. Integrating results with ancillary spatial data, we could identify driving forces and estimate the metrics of desertification processes. In the Biskra area (Algeria), it emerged that the expansion of irrigated farmland in the past three decades has been contributing to an ongoing secondary salinization of soils, with an increase of over 75%. In the Oum Zessar area (Tunisia), there has been substantial change in several landscape components in the last decades related to increased anthropic pressure and settlement, agricultural policies, and national development strategies. One of the most concerning aspects is the expansion of sand-encroached areas over the last three decades of around 27%.