Assessment of Land Use and Land Cover Changes and their effect on Rainwater Harvesting.

Abstract

This study examined how changes in land use and land cover (LULC) affected rainwater harvesting (RWH) in the Opeta-Bisina catchment, a sub-catchment of the Lake Kyoga system in Eastern Uganda, between 2000 and 2020. The research aimed to quantify LULC changes, assess their hydrological effects on RWH, and evaluate management strategies to optimize RWH amidst these changes. Using QGIS 3.34.15 for remote sensing, Landsat images from 2000, 2010, and 2020 were analyzed through supervised classification, achieving overall accuracies of 80.03% (2000), 83.57% (2010), and 86.67% (2020), with Kappa coefficients of 0.77 (2000), 0.78 (2010), and 0.82 (2020). A Soil and Water Assessment Tool (SWAT) model, calibrated and validated with streamflow data from 2003 to 2005, and 2006 to 2007, respectively, was used to simulate the hydrological effects of the LULC changes. The model showed satisfactory performance (R² of 0.642 and NSE of 0.630 for calibration, and R² of 0.778 and NSE of 0.543 for validation). The results of the study revealed significant LULC changes in the study period (2000-2020): built-up areas increased from 0.4% (632.1 ha) to 1.46% (2,337.9 ha), agricultural land expanded from 19.42% (31,044.6 ha) to 29.03% (46,396.2 ha), a decrease in grassland decreased from 31.77% (50,782.4 ha) to 22.47% (35,911.3 ha), forestland from 3.18% (5,076.4 ha) to 1.45% (2,323.7 ha), and wetlands from 36.63% (58,542.4 ha) to 32.83% (52,470.9 ha) was also recorded. These LULC changes were mainly driven by population growth in the catchment districts, which increased from about 800,000 in 2002 to over 2 million by 2014. The hydrological impacts of these LULC changes were substantial, as simulated by the SWAT model. Surface runoff increased by 36.01%, from 113 mm/yr in 2000 to 165 mm/yr in 2020, and sediment yield increased by over 11%, from 5.54 t/ha/yr to 6.5 t/ha/yr, due to increased impervious surfaces from urban expansion and reduced vegetative cover from deforestation and agricultural growth. These changes have several impacts on RWH. The rise in sediment yield and pollutants from agricultural and urban runoff necessitates pretreatment infrastructure like settling basins and first-flush diverters to ensure water quality. The increased runoff supports surface and rooftop RWH systems, but higher peak flows require robust storage designs to manage flood risks and prevent system failure. To optimize RWH under these LULC changes, four management scenarios were modeled using SWAT: reforestation of agricultural land, conversion of built-up areas to grassland, no-tillage on cropland, and cultivation/cropping on previously bare soil. All the modeled scenarios effectively reduced both surface runoff and sediment yield in the catchment by over 8%. This indicates the need for effective water resources management strategies in the catchment.