Land cover change of arid environment in Tunisia based on analysis of Landsat images

Abstract

The impact of Land use/land cover (LULC) change was assessed through monitoring the distribution of ecological indicators and tracking the aeolian deposits, which provides valuable information on desertification and climate change in Tunisian arid regions. This study was conducted in Oum Zessar area, in southeastern Tunisia. Both visual interpretation and automated classification approach were developed to extract sand features using Landsat images for 2000, 2008 and 2014. The automated classification includes a decision tree classifier (DT) and an unsupervised classification applied to the principal components extracted from Knepper ratios composite. The validation of the classification methods showed that the DT had an overall accuracy over 84%. The results of the change detection have shown an increase in the classes of Agriculture behind tabia by 10.68%, the rangelands and croplands by 24.37% and the mountain rangelands by 14.93%, and a decrease in the classes of Agriculture behind jessour by 33.65%, sand encroachments by 12.93% and halophyte rangelands by 3.4%, respectively. These resulting maps seem to be the suitable decision-support tools for management of land use in arid regions of Tunisia, in particular, for land degradation assessment and water and soil conservation.Keywords change detection, land cover, Landsat imagery classification, Tunisia Résumé L'impact des changements de l'occupation des sols a été évalué à travers le suivi de la distribution des indicateurs écologiques et le suivi des dépôts éoliens, afin de fournir des informations utiles sur la désertification et les changements climatiques dans les régions arides tunisiennes. Cette étude a été réalisée dans la région d'Oum Zessar, au Sud-Est de la Tunisie. L'interprétation visuelle et l'approche de classification automatisée ont été développées pour extraire les accumulations sableuses à l'aide d'une série d'images Landsat (2000, 2008 et 2014). La classification automatisée par Arbre de Décision et une classification non supervisée a été appliquée aux principaux Knepper Ratios Composites. La validation des méthodes de classification a montré une précision globale supérieure à 84%. Les résultats de la détection des changements ont montré une augmentation des classes d'Agriculture derrière Tabia de 10,68%, des Parcours et Terres Cultivées de 24,37% et des Parcours de Montagne de 14,93%, et une diminution des classes d'Agriculture derrière Jessour de 33,65%, | 747 HENCHIRI Et al.

Monitoring Seasonal Fluctuations in Saline Lakes of Tunisia Using Earth Observation Data Processed by GRASS GIS

“…Inspired by the existing examples of the RS data analysis in environmental monitoring [61][62][63][64][65], this study integrated the advanced methods of the GRASS GIS with applied techniques of its image processing modules for assessment and mapping of changes in landscapes of saline lakes in northern Tunisia, Figure 1. The main purpose of this study is to evaluate the response of land cover types around salt lakes of Tunisia to seasonal fluctuations in temperature and the increase in salinisation.…”