Mapping the Eucalyptus spp woodlots in communal areas of Southern Africa using Sentinel-2 Multi-Spectral Imager data for hydrological applications

Invasive alien plant species (IAPs) drive anthropogenic global environmental change by threatening native species communities and ecosystems. For instance, the Eucalyptus species are notoriously known for releasing toxic chemicals which impedes the growth of indigenous plants which usually results in severe erosion. It is, therefore, important and necessary to quantify and understand the spatial extent and distribution of these IAPs to identify critical hotspots for restoration. This study, therefore, tested the utility of Sentinel 2 Multispectral Imager (MSI) data in distinguishing Eucalyptus community-woodlots from other land cover types with an aid of the Support Vector Machines (SVM) algorithm. The results show that eucalyptus species were successfully discriminated from other land cover types with an overall accuracy (OA) of 85.07%. Eucalyptus covered an area of 2378.08 ha, constituting 11.6% of the study area. Furthermore, about 23% of Eucalyptus wood plots were found at a closer range of 1 km from the rivers. Sentinel 2 MSI Band 3 (green), 7 (vegetation red edge) and 8A (narrow near-infrared (NIR)) proved to be effective in characterizing the spatial distribution and extent of Eucalyptus wood plots. The findings of this study serve as a footstool towards the establishment of a framework for monitoring these IAPs and understanding their impact on the diversity of indigenous tree species.