Long-term land use and land cover dynamics in the Okavango River Basin: impacts on wetlands ecohydrological conditions using satellite data and machine learning
| dc.contributor.author | Sigopi, Maria | |
| dc.contributor.author | Moropane, Lebogang Mmasechaba | |
| dc.contributor.author | Dube, Timothy | |
| dc.date.accessioned | 2026-01-12T07:30:13Z | |
| dc.date.available | 2026-01-12T07:30:13Z | |
| dc.date.issued | 2025 | |
| dc.description.abstract | The Okavango River Basin (ORB), one of sub-Saharan Africa’s most ecologically significant and well-preserved endorheic system, is critical for sustaining biodiversity and providing ecosystem services. However, increasing anthropogenic pressure and environmental change demand continuous and precise monitoring to safeguard its natural assets. This study utilized Google Earth Engine (GEE) to present a robust 34-year (1989–2023) analysis using Landsat 5 and 8 at 30 m resolution. The study examined the relationship between the normalized difference vegetation index (NDVI), soil-adjusted vegetation index (SAVI), and normalized difference phenology index (NDPI), coupled with Climate Hazards Group Infrared Precipitation (CHIRPS), ERA5 Land, and TerraClimate products. Utilizing a Random Forest (RF) classifier, we achieved accuracies of 95-98% across nine intervals. Wetlands maintained 3% coverage from 1989-2004, while forest occupied 20-26%. Water bodies declined from 1989-2016, then gained 6419 km2 (2017–2020). Wetlands gained 19144 km2 (1989–1992) and 8406 km2 (2017–2020), but lost -10986 (1993–1996) and -7734 km2 (2009–2012). Higher temperatures are correlated with NDPI (β = 0.05, p = 0.003, R2 = 0.32) and NDVI (β = 0.106, p = 0.0045, R2 = 0.29), while precipitation and evapotranspiration were not significant. SAVI presented no significant relationship (R2 = 0.27, p = 0.027). These findings underscore the urgent need for continuous LULC monitoring to inform adaptive management strategies for the ORB. | |
| dc.identifier.citation | Sigopi, M., Shoko, C., Moropane, L.M. and Dube, T., 2025. Long-term land use and land cover dynamics in the Okavango River Basin: impacts on wetlands ecohydrological conditions using satellite data and machine learning. South African Geographical Journal, pp.1-26. | |
| dc.identifier.uri | https://doi.org/10.1080/03736245.2025.2575951 | |
| dc.identifier.uri | https://hdl.handle.net/10566/21629 | |
| dc.language.iso | en | |
| dc.publisher | Routledge | |
| dc.subject | Artificial Intelligence | |
| dc.subject | Climate Change | |
| dc.subject | Environmental Monitoring | |
| dc.subject | Remote Sensed Data | |
| dc.subject | Southern Africa | |
| dc.title | Long-term land use and land cover dynamics in the Okavango River Basin: impacts on wetlands ecohydrological conditions using satellite data and machine learning | |
| dc.type | Article |