Spatiotemporal assessment of mining impact on land use land cover in Bole District: a performance evaluation of support vector machine (SVM) and random forest (RF) classifiers with future impact prediction

The recurring upsurge of illegal mining activities in the Bole District has led to the rapid disruption of the existing biodiversity stemming from the absence of periodic studies in the area. Also, little attention has been paid to the performance evaluation of modern classification algorithms such as SVM and RF in African tropical regions, of which Ghana is no exception. This paper aims to assess the impact of mining activities in the Bole District (2005–2020) using LULC trend analysis and future impact predictions of 2030. The research further focuses on the performance evaluation of SVM and RF classifiers, showcasing how different machine learning algorithms capture LULC variations and contribute to the advancement of mapping techniques. The results indicate the superiority of the SVM in pixel-based image classification tasks. Notable changes in LULC between 2005 and 2020 include a significant increase in built-ups by 454.77%, attributed to an influx of people due to the growth of small-scale mining activities. Bare lands increased by 88.79% due to land clearance for farming purposes and small-scale illegal mining. This study will thus serve as a guide for individuals, developers, the government, and other stakeholders in biodiversity conservation for further research and decision making.