Assessing the impact of artisanal and semi-industrial gold mining in tropical forests using remote sensing : a case study of the Okapi Wildlife Reserve in the Democratic Republic of Congo

Details

Supervisors

Defourny, Pierre

Faculty

Faculté des bioingénieurs

Degree label

Master [120] : bioingénieur en sciences et technologies de l'environnement, à finalité spécialisée

Abstract

The growing global demand for minerals and the consequential environmental impacts of mining necessitate effective monitoring strategies. Remote sensing provides opportunity for this monitoring. This study focuses on artisanal and semi-industrial gold mining in tropical regions, with the Muchacha gold mine in the Okapi Wildlife Reserve (Democratic Republic of Congo) as a case study. Leveraging satellite imagery, this research assesses the performances of remote sensing techniques to detect land cover changes associated with gold mining activities. The research investigates multiple approaches, including convolutional neural networks, object-based image analysis, and continuous change detection and classification. High-resolution data from Sentinel-2, PlanetScope, and Pleiades satellites enable detailed analysis of the mining area. The study reveals the potential of remote sensing to track mining-induced land cover changes accurately. By assessing spatial organization and landscape dynamics of mining sites, this study contributes to understanding their ecological and socio-political implications. This help in providing valuable information in a context of mining-conflict-conservation nexus encountered in some developing countries. Challenges for an effective monitoring of mining areas are addressed, and avenues for improvement are suggested. In conclusion, this research underscores the significance of remote sensing in monitoring and managing mining impacts. The outcomes highlight the urgency of sustainable practices in mineral extraction and underscore the pivotal role of remote sensing in supporting informed decision-making