| dc.description.abstract | This study aims to assess the impacts of land use and land cover (LULC) changes on runoff in the Luvuvhu catchment in Limpopo, South Africa. To achieve this aim, the study had two specific objectives. The first objective was to determine past and future spatial trends of LULC change, using remote sensing data. To determine the past spatial trends of LULC, Landsat images from 1990 to 2020 were classified using the Support Vector Machine (SVM). The results revealed that there were significant LULC changes in the Luvuvhu catchment during the selected study period. The results of the LULC classification showed that built-up area increased by 4.8% from 1990 to 2020. Plantation and natural vegetation decreased by 0.3% and 18.9%, respectively. Bare land and waterbodies increased by 15.2% and 0.4%, respectively. Agricultural land decreased from 4.7% to 3.5%. The classification results showed high overall accuracies (93% to 98%) and Kappa coefficients (>0.9). These results showed good classification. The artificial neural networks-cellular automata (ANN-CA) based simulation model was implemented to predict future LULC in 2025 and 2030. The predicted LULC for 2025 shows that the built-up area will increase by approximately 120 km2, plantation by 17 km2, agricultural land by 51 km2 and bare land by 168 km2. Natural vegetation may lose up to 356 km2 in 2025. Meanwhile, a waterbody may lose close to 2 km2 by 2025. The trends observed in 2025 are expected to continue until the year 2030. The second objective was to establish the response of runoff to LULC changes, using a process-based semi-distributed approach. The study adopted the Soil and Water Assessment Tool (SWAT) model for the simulation of the hydrological response of the Luvuvhu catchment. The calibration and validation for monthly flow between the observed and simulated data showed a good relationship. Model performance for the calibration and validation for both the 1990 LULC and 2020 LULC period show satisfactory results. On average, overall surface runoff increased by 7.3% from 169.14 mm to 181.49 mm between 1990 and 2020. The results show that the surface runoff of the Luvuvhu catchment was altered due to the significant LULC changes in bare land and natural vegetation with a 15.2% increase and an 18.9% decrease, respectively. Therefore, these changes contributed significantly to the runoff response of the Luvuvhu catchment. This study provides baseline information on the impacts of LULC changes on surface runoff within the Luvuvhu catchment in Limpopo, South Africa. | en_US |
