Land Use and Vegetation Density across Altitudinal Gradients in Sterkspruit Nature Reserve, South Africa: A Drone-Based Analysis

The continued growth of human populations and per capita consumption has resulted in unsustainable exploitation of the Earth’s biological diversity, exacerbated by climate change, ocean acidification, and other anthropogenic environmental impacts. Studies are crucial for classifying vegetation types, understanding their differences, and supporting ecological research, biodiversity preservation, and environmental monitoring. This study investigated Land Use and Land Cover (LULC) classification and the relationship between vegetation density and altitude in the Sterkspruit Nature Reserve (STNR) in South Africa, using remote sensing techniques. A supervised classification was conducted on aerial photography images using the Maximum Likelihood Algorithm (MLA) with 58 ground-truthed points, resulting in four land cover classes: wooded area, rock and mountain summit, bare soil and built-up area, and grassland. The classification achieved an overall accuracy of 0.97 and a Kappa value of 0.96, indicating a highly reliable model. To assess vegetation density, the Normalized Difference Vegetation Index (NDVI) was computed and analyzed against altitude. The Spearman correlation coefficient of -0.157 showed a weak negative association, while linear regression revealed a slight, statistically significant negative relationship between NDVI and altitude (p = 0.038), though altitude explained only 2.1% of NDVI variance. These findings suggest that altitude impacts vegetation density, but other environmental factors likely play a more significant role. This study demonstrates the effectiveness of remote sensing for land cover classification and provides insight into vegetation patterns in mountainous ecosystems in Sterkspruit Nature Reserve.