Sediment and nutrient loading into Lake Kivu: A case study lwiro micro-catchment, Democratic Republic of Congo

Lake Kivu is one of the deepest lakes in Africa with a maximum depth of 485 m and the most fragile and highly sensitive ecosystems in the African Great Lakes region in term of gaseous content. During the last decade, an increase in pollution loading was observed in the lake contributing to eutrophication, decreased dissolved oxygen and increased methane (CH4) and carbon dioxide (CO2) levels. Estimation of pollution loading including nutrient and sediment yield in the micro-catchment is a prerequisite for developing the best micro-catchment management plan for the Lake-wide basin. The objectives of this study were to; determine land use/land cover changes in Lwiro micro-catchment for the last 25 years and; to estimate the contribution of the different nutrient and sediment sources to the pollution loading into the River Lwiro. Time series of the Landsat images (1987, 2001, 2010) were classified into four categories: forest, buildup areas, wetland and small scale farmland area. SWAT model was used to estimate pollution loading from different sources into the River. Results show that land use/ cover in the study area have changed substantially from 1987 to 2010. The forests decreased by 23%, while wetlands, small scale farmland and Built-up areas increased by 2.11; 0,88 and 2.29 % respectively. A good agreement between observed and simulated discharge, sediment yield and nutrient load during the study period was observed using SWAT model. The coefficient of determination (R2) for the monthly runoff, TSS, TP and TN were obtained as 0.56; 0.79; 0.66 and 0.86 respectively, during the period of study. ArcSWAT simulated results show that sediment and runoff high yield was generated by small scale farmland and forest area respectively. Sensitization about best management practices should be conducted in order to reduce nutrient and sediment loads due to erosion into the River Lwiro micro-catchment.