Modelling the spatial distribution of Lantana Camara in the inkomati catchment in Mpumalanga, South Africa
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Lantana Camara is a highly problematic invasive species, extensively studied for its spatial distribution. The present study was conducted to investigate Lantana Camara's spatial patterns across the topographic variations within the Inkomati catchment, located in Mpumalanga, South Africa, using the Random Forest and Maxent algorithms with Sentinel-2 data within Google Earth. Lantana Camara covered 34.86% of the study area, with a user's accuracy of 91% and producer's accuracy of 84%. Elevation strongly influenced the species' spatial distribution, while the Topographic Wetness Index had minimal impact. MaxEnt vulnerability maps revealed higher vulnerability to Lantana Camara in the central western part of the study area than the eastern part. The model using topographic variables achieved the highest accuracy (AUC = 0.88), surpassing the predictive model with Sentinel-2 bands (AUC = 0.81). The study suggests that the Sentinel-2 Red Edge and NIR bands, combined with Random Forest, offer accurate insights into Lantana Camara distribution, aiding in identifying regions vulnerable to its invasion.
Description
A research report submitted in fulfilment of the requirements for the degree of Master of Science to the Faculty of Science, School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Johannesburg, 2023
Keywords
Lantana Camara, MaxEnt, Topographic variable