Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Autriche → Modelling soil erosion and sediment delivery to reservoirs at a large scale domain, a strategy for catchment management : the case of Masinga catchment, Kenya

Universität für Bodenkultur Wien (2005)

Modelling soil erosion and sediment delivery to reservoirs at a large scale domain, a strategy for catchment management : the case of Masinga catchment, Kenya

Mutua, Benedict Mwavu

Titre : Modelling soil erosion and sediment delivery to reservoirs at a large scale domain, a strategy for catchment management : the case of Masinga catchment, Kenya

Auteur : Mutua, Benedict Mwavu

Université de soutenance : Universität für Bodenkultur Wien

Grade : Doctoral Thesis 2005

Soil erosion and sediment yield from catchments are key limitations to achieving sustainable land use and maintaining water quality in streams, lakes and other water bodies. Kenya is one country threatened heavily by land degradation due to increasing anthropogenic pressure on its natural resources. This is more so mainly in the river basins where the rural communities encroach to open up new land for agricultural activities. As land degradation has become more evident with increasing changes in land use within Masinga catchment, it has become necessary to quantify spatially soil erosion and sediment yield over the entire catchment. A tool for planning soil conservation strategies at a large scale to control soil erosion and sediment delivery to the reservoirs is therefore required. This study presents a first attempt of the application of Geographical Information System (GIS) technology to simulate soil erosion and sediment yield within Masinga catchment. It demonstrates the integration of an empirical model, the Revised Universal Soil Loss Equation (RUSLE) within the GIS environment to estimate average annual soil losses in a spatial domain. Refinements of slope length and rainfall erosivity factors of the RUSLE model were introduced to improve its applicability in an area of varied terrain under a tropical climatic regime. The study employed the upslope drainage substitution method to generate LS-values, and the R factor was based on a developed daily rainfall model. With these factor refinements, the RUSLE was adapted to function on a semi-distributed basis, which more comprehensively considered the cumulative effects of overland flow on erosivity. Current land use/land cover and management practices and selected, feasible, potential management practices were evaluated to determine their effects on average annual soil loss. The results show that only 9.3% of the total catchment area is experiencing soil erosion within the tolerable rates under the current land use. By simulating the best viable management practices (BMPs), the results show that the area under tolerable soil erosion rate could be increased to 36.5%. The achieved results show that substantial reduction in soil erosion can be realised when conservation support practices and viable land use/land cover are applied. Controlling sediment loading requires adequate prediction of soil erosion and sedimentation. However, sediment yield is usually not available as a direct measurement but it is estimated using a sediment delivery ratio (SDR). An accurate prediction of SDR is important in controlling sediments for sustainable natural resources development and environmental protection. There is no precise procedure to estimate SDR, although the USDA has published a handbook in which the SDR is related to the drainage area. This research presents a new approach for estimating spatial sediment delivery ratio (SDR) for large rural catchments. The SDR was predicted using a Hillslope Sediment Distributed Delivery (HSDD) model in conjunction with a physically distributed hydrological model in a GIS environment. A physically based geospatial Hydrological Model, the Stream Flow Model (SFM) was incorporated in the ArcView graphical user interface to estimate the catchment hydrologic parameters in a spatial domain. The hydrological model was validated using predicted and observed daily stream flows for 1992. Model results show a spatial variation of sediment yield even within the same sub-catchment. The results also show that sediment yield does not depend only on the size of the sub-catchment, but more on the sub-catchment properties. The overall predicted mean annual sediment yield for the whole catchment based on the land use and management practices for the year 2003 is about 31 t ha-1 yr-1. This translates to about 14 million cubic metres (Mcm) per year as the predicted overall sediment volume from the catchment (6,255 km2) reaching Masinga reservoir. This is within the same order of magnitude when compared with the estimate of 11 Mcm of sediment estimated by KenGen (2000). The simulated results show that by applying the BMPs, the sediment yield reaching the reservoir could be reduced to 12 t ha-1 yr-1. In this study, the critical soil erosion and sediment yield source areas within the catchment were identified. The developed model is not only conceptually easy and well suited to the local data needs, but also requires few parameters in its application while meeting the intended purpose. The integrated modelling approach allows for relatively easy, fast, and cost-effective estimation of spatially distributed soil erosion and sediment yield. It thus provides a useful and efficient tool for predicting long-term soil erosion potential and assessing downstream impacts under different land management options.

Mots clés : Technik:Hydraulik Landdegradation Bodenerosion Modell sedimentertrags RUSLE-GIS — ENGINEERING, HYDRAULIC Soil erosion model Sediment yield RUSLE-GIS Land degradation

Présentation (österreichischen Bibliothekenverbundes)

Page publiée le 21 novembre 2015, mise à jour le 13 mars 2019