M. Isabirye1*, D. Kimaro2, O. Semalulu1 , A. De Meyer3, M. Magunda1, J.Poesen3, J. Deckers4

1National Agricultural Research Organization, National Agricultural Research Laboratories, P.O. Box 7065, Kampala, Uganda

2Sokoine University of Agriculture, P. O. Box 3003, Morogoro, Tanzania

3Katholieke Universiteit Leuven, Physical and Regional Geography Research Group, Geo-Institute, Celestijnenlaan 200E, 3001 Heverlee, Belgium

4Katholieke Universiteit Leuven, Laboratory for Soil and Water Management, Geo-Institute, Celestijnenlaan 200E, 3001 Heverlee, Belgium

Soil erosion, among others, is major land degradation process decreasing productivity in the Lake Victoria basin. Identification of sources of sediment sources can help in directing soil conservation interventions. Studies on sources of sediment and their connectivity to water bodies are done in the northern shore line and Rwizi catchments in the Lake Victoria basin. Vegetation types are also evaluated for sediment trapping efficiency. Preliminary observations indicate that settlement areas generate more sediment compared to the crop fields. Results show that settlements generate significantly higher sediment yields i.e. between 17 – 87 ton ha-1 yr-1 whereas agricultural land use types produced between 0 – 27 ton ha-1 yr-1. In Mayuge study area, the average soil erosion rate in compounds amounts to 107 Mg ha-1 year-1 and in the landing sites to 207 Mg ha-1 year-1. Although compounds and landing sites occupy a small area of the study area (1.1 %), they are a major source of sediment to Lake Victoria (63 %). After the correction for compaction is carried out, the soil erosion rate on footpaths amounts to 34 Mg ha-1year-1 and on main roads to 35 Mg ha-1year-1. Also footpaths and main roads occupy a small area of the study area (1.1 %), but contribute disproportionately to the total soil loss in the catchment (22 %). In accordance to an adaptation of the model of McHugh et al. (2002), 32 % of the sediment that is generated in the catchment, is deposited in Lake Victoria (i.e. 2 209 Mg-1year-1 or 0.7 Mg ha-1year-1). Sediment delivery decreased exponentially with increasing vegetative filter widths for all tested vegetation types with correlation coefficient (R2) ranging from = 0.7 to more than 0.9. The trapping efficiency of the tested vegetative filters increased logarithmically with increasing vegetative filter widths. Sediment trapping efficiency increased from 24 % at 2.5 m to over 80 % at 10 m vegetative filter widths for all tested vegetation types. The high sediment yield from settlements justifies the need to conduct further investigations on the contribution of settlements to sediment production in catchments with different soil – landscape and climatic setting in the Lake Victoria catchment. This study demonstrates that the amount of sediment trapped by Vegetative Buffer Strip Filters and their corresponding widths are significant up to a width of 10 metres. Also, the results demonstrate that Vegetative Buffer Strip Filters can be effective filters in removing sediment inputs to surface waters of Lake Victoria by restricting a minimum land area not exceeding 10 m under the tested VBSFs. Therefore, it may be necessary to rethink the 100 – 200 m buffer zone legislative requirements in the riparian zones of Lake Victoria basin.

Keywords: Sediment generation, settlements, agricultural fields, vegetation buffer, Lake Victoria


To be a centre of excellence generating and promoting appropriate agricultural technologies


To generate and promote agricultural technologies and improve productivity, value addition, income and food security


Hosted Institutions