Watershed sediment yield modelling for data scarce areas; a case study, Awash River Basin, Ethiopia

The main goal of the research was to device an alternative solution for watershed sediment yield modelling for data scarce areas where the existing physically based models can not be applicable. Awash River Basin in Ethiopia was selected as case study area.

GIS data on soil, land use, precipitation, temperature, stream flow and suspended sediment yield was collected from the Federal Ministry of Water Resources of Ethiopia (FMWRE) and from the National Metrology Service Agency (NMSA) offices. Soil data obtained from FMWRE and Food and Agriculture Organization (FAO) world soil 1974 database was used for derivation of the soil erodibility factor (ERFAC) estimation equation. The ratio of silt to sand and clay content was considered as the governing factor for soil erodibility in developing the ERFAC equation.

The SWAT2005 model was selected for calibration and validation of stream flow and sediment yield. A sensitivity analysis was carried out to prioritize model calibration parameters. From the sensitivity analysis, curve number II (CN2), soil water available to plants (SOL_AWC) and ground water base flow factor (ALPHA_BF) were selected as major stream flow calibration parameters. Similarly CN2, SURLAG (surface lag), slope and sediment routing factor (SPCON) were taken as the major sediment calibration parameters. Parameters related to the soil properties and river channel characteristics were given special attention during the model calibration.

Eleven years (1990-2000) stream flow and sediment data were used for model calibration and six years data (2001-2006) were used for model validation. Calibration has been done at three gauging stations located in the Awash River basin. The statistical indicators, Coefficient of determination (R2), Nash-Sutclife efficiency (NSE),Root mean square error observations standard deviation (RSR were applied to evaluate the calibration and validation results. The values of these indicators were used to rate the performance of the model.

Watershed geomorphologic and topographic factors were extracted from the SWAT2005 watershed configuration, using a GIS tool and empirical equations. The relative importance of the factors was determined using Pearson's correlation coefficient based on the sediment yield output obtained from the SWAT2005 model calibration. The results show that, the sediment yield is highly correlated with stream flow, watershed area and watershed slope. Based on the identified parameters and the SWAT2005 model output, an alternative sediment yield estimation equation was derived and checked for its validity.

Silke Wieprecht

Walter Marx
Achim Beylich

Habtamu Itefa Geleta

LWW

Von: 01.10.2007

Bis: 30.09.2010

International Postgraduate Studies in Water Technologies (IPSWaT) Scholarship