Assessing Urban Expansion of Algiers with Random Forest-Based Multi-Index Using Landsat Imagery

Urban sprawl is a global phenomenon characterized by the uncontrolled expansion of human settlements, negatively impacting the environment, economy, and society. Moreover, the evolution of urban sprawl had been identified as a significant factor contributing to various issues, including the intensification of the urban heat island effect in cities. In this study, we proposed an approach to analyze urban expansion based on the Random Forest Multi-Index classification (RaFMIC) using the Google Earth Engine (GEE) platform. The Multi-Index approach integrates multiple indices derived from remote sensing data to enhance the accuracy and comprehensiveness of the classification process. For this purpose, multi-temporal Landsat data from 1995 to 2023 were used to generate classification maps of land use and land cover. The results obtained from the multi-index approach showed better classification performance when compared to SVM and CART methods. With an outstanding overall performance of 93.20% accuracy and a kappa value of 0.87 for all LULC classes, the methodology was applied to analyze urban expansion in Algiers (Algeria).