Abstract:
Land use and land cover changes in river basins are among the main drivers of global
environmental change in many developing countries. In Mozambique and Eswatini, land use
and land cover changes cause serious challenges to natural resource management in river
basins, especially in the Umbeluzi River basin, which is jointly managed by both countries.
These challenges occur in the form of water resource depletion, land use conflicts and are also
a key driver of climate change. Remote sensing and Geographic Information Systems (GIS)
are effective and frequently used tools for comprehensively understanding the dynamics of
land use and land cover changes. The aim of this study was to use remote sensing and GIS
techniques to analyse variations in land use and land cover changes in the Umbeluzi River
basin for the years 2019 and 2024. For this purpose, satellite images from Sentinel-2 were used,
on the Google Earth Engine platform, to compute supervised classification. Supervised
classification was performed using machine learning Random Forest (RF) algorithm and the
accuracy of the classification was evaluated using confusion matrix. Image processing and
mapping were performed based on the ArcGIS Pro 3.2 software. The results showed changes
in land use and land cover of the Umbeluzi River basin during the study period. The urban
construction decreased by 8%; bare soil or pasture increased by 16.7%; water bodies increased
by 77.4%; evergreen vegetation or woodlands decreased by 8.1%; shrublands decreased by
12%; irrigated agricultural area decreased by 30.8%; mangroves increased by 5.6%; small-
scale agriculture increased by 4.6%. Classification accuracy consistently achieved values
above 0.8 for the indices generated through the confusion matrix. The results of this study have
the potential to contribute significantly to the effective management of the Umbeluzi River
basin, providing an appropriate analysis of the potential and demand for water resources,
integrating the needs of human activities with the ecological sustainability activities of aquatic
and riparian habitats.
