Theses and Dissertations

Permanent URI for this collection

https://hdl.handle.net/20.500.14155/2231

Browse

Browsing Theses and Dissertations by Author "Khaba, Liphapang"

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Impacts of climate change on streamflow and hydrological extremes in the South Phuthiatsana Catchment, Lesotho
    (National University of Lesotho, 2025) Maphutseng, Makoanye; Khaba, Liphapang
    Global climate change is predicted to significantly modify hydrological processes, which will have a big impact on ecosystem sustainability, flood risk, and water availability. Knowing how future climate changes may impact river systems is especially important in southern Africa, where population increase and fluctuating rainfall are already placing a strain on water resources. One such critical system is the South Phuthiatsana catchment, which serves as a vital supply of water for Maseru and the other metropolitan areas. This study explores how projected climatic shifts may influence streamflow behavior and the occurrence of hydrological extremes within the South Phuthiatsana watershed, an essential source of water for Maseru and surrounding urban communities. Bias-adjusted data from the MPI-ESM1- 2-LR global climate model, along with two sample emissions trajectories for the mid-21st century (2041–2080), were used to project climate inputs. Suboptimal performance indicators showed that the process-based model (SWAT+), which was used to simulate streamflow, was not very reliable in capturing historical daily flow patterns. SWAT+ was therefore thought to be insufficient for predicting future hydrological reactions in this context. A machine learning approach using the XGBoost algorithm was adopted to address this challenge. This data-driven model was trained on bias-corrected climate variables and observed streamflow, providing a more reliable tool for future streamflow prediction. The results from XGBoost revealed substantial and complex hydrological shifts. A consistent warming trend combined with highly variable seasonal precipitation patterns was evident across both emission scenarios. Extreme high flows, represented by the 98th percentile (Q98), are projected to decline by more than 52% compared to historical values, suggesting a reduced risk of flooding. In contrast, low flows are expected to increase dramatically; the 1st percentile (Q1) flow is projected to rise from near-zero values historically to approximately 9.0 m³/s, indicating a significant shift toward more perennial flow conditions. Mid-range flows (Q25, Q50, and Q75) are also expected to increase substantially, depending on the flow percentile and scenario. While the absolute magnitude of low flows improves, the number v of days with historically low flow conditions may still increase during certain months, highlighting a shift in the intra-annual flow variability. These findings point to a future with altered hydrological regimes in the South Phuthiatsana catchment characterised by diminished flood peaks, elevated baseflows, and more frequent low- flow conditions during critical periods. Despite initial limitations with the process-based model, the machine learning approach provided robust insights that form a valuable foundation for developing adaptive, forward-looking water resource management strategies. These results underscore the need for resilient planning to ensure long-term water security under evolving climate conditions
  • Thumbnail Image
    ItemOpen Access
    Water demand forecasting using machine learning approach
    (National University of Lesotho, 2025) Mahamo, Qenehelo; Khaba, Liphapang
    The increasing challenges related to water security, exacerbated by rapid urbanization, population growth, and climate variability, necessitate accurate and reliable forecasting methodologies to support sustainable water resources planning. This study explores water demand forecasting in Ha-Foso, Lesotho, by evaluating three machine learning models: Multiple Linear Regression (MLR), Support Vector Regression (SVR), and Artificial Neural Networks (ANN). Utilizing time series data sets covering meteorological inputs (2012-2022), population, and water consumption records (2017-2024), the study assesses the influence of climatic and demographic variables, specifically precipitation, maximum and minimum temperatures, and population, on domestic water consumption. The research first used MLR to assess the influence of population, maximum temperature, minimum temperature, precipitation, and other factors on water demand. Subsequently, the study evaluated the predictive performance of MLR, SVR, and ANN models. Performance was evaluated using performance metrics, including the coefficient of determination (R2), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The regression analysis consistently identified population as the only statistically significant predictor of water demand (p<0.001), while climatic variables showed no significant influence during the study period. In the comparative evaluation, the SVR demonstrated thehighest accuracy and generalization capacity, outperforming ANN and MLR, with the least error metrics in both the training phase and the testing phase. The 2-year forecast highlighted the distinct behaviours of each model, with the SVR and ANN models providing more moderate growth projections compared to the steep, linear increase predicted by the MLR model. This study presents the potential of machine learning, particularly SVR and ANN, in addressing the intricate, non-linear relationships inherent in water demand forecasts, delivering precise and actionable water demand forecasts for peri-urban settings in Lesotho. The findings suggest the adoption of advanced architectures and the incorporation of socio- demographic variables to strengthen predictive capacity. The outputs are expected to support utility companies such as Water and Sewerage Company (WASCO) in strategic planning, conservation, and infrastructure investment decisions.