Monitoring the impacts of mining activities on land cover using remote sensing technologies and machine learning in the Musonoi copper‒cobalt deposit in the Democratic Republic of Congo

This study examined the impact of land use land cover (LULC) change across the various geomorphological zones in relation to mining activities using remote sensing and random forest (RF) in the Musonoi copper‒cobalt deposit of Democratic Republic of Congo (1994–2024). The RF yielded an overall classification accuracy ranging from 77.96% to 86.20% and Cohen’s Kappa of 0.73 to 0.83. Results of this study, shows that expansion mining activities signicantly affected LULC features overtime, especially agricultural land. The hydrothermal alteration zones are dominated by chlorite, ferrous silicate and kaolinite - highlighting areas affected copper mineralisation. The findings of the study offer baseline information critical for understanding mining-induced landscape dynamics and identifying altered zones potentially linked to copper deposits. Therefore, this study further contributes to international imperatives, such as the United Nations Sustainable Development Goals (SDGs), which focus on responsible resource extraction and environmental sustainability.