Abstract:
Sedimentation caused by soil erosion, transportation and deposition from poor land practices and
improper management system is a critical problem in many water resources projects. Therefore,
modelling of runoff and sediment yield at a watershed level is very essential. A conceptual, semidistributed
river basin model and continuous time step, SWAT2012 (Soil and Water Assessment
Tool) model was selected for the simulation of the runoff and sediment yield from Geba watershed
(SWAT delineated area 3,576 km2), in north eastern upper Baro Akobo river basin, south western
Ethiopia. The main objective of the study was modeling of stream flow and Sediment load,
assessing the spatial and temporal distribution, from the Geba watershed. The simulation was done
by dividing the watershed (total area of 3,576km2) into 138 Hydrologic Response Units (HRUs)
established with 23 sub basins. A 30m spatial data of Digital Elevation Model (DEM), Land use
map with 1km spatial resolution and 10km spatial resolution of soil map was used. The time series
data (1990-2015) from 4 meteorological stations, in which one of them synoptic station, were used.
After the simulation has implemented, the Calibration and Uncertainty analysis was carried out
with SWAT-Cup the Sequential Uncertainty Fitting-2 (SUFI-2) and ArcSWAT2012, using
monthly observed discharge and estimated sediment load values through the application of SUFI-
2 procedure at Chora gauging stations from 2003-2008 and 2009-2011 for validated. The model
performance was checked by statistical model performance ratings such as coefficient of
determination (R2), Nash-Sutcliffe model Efficiency (ENS), and Observation Standard Deviation
Ratio (RSR) and percent bias (PBIAS). Goodness of fit and the degree to which the calibrated
model accounted for the uncertainties were assessed with the P-factor and the R-factor of the SUFI-
2 algorithm. The results of the model show that calibration and validation for streamflow performed
good for the sub catchment and for sediment loads estimation, calibration performed better than
validation. After calibration has performed the simulated average annual sediment yield waswas
estimated to (50154.6ton/yr.) at the outlet, and an average spatial distribution of 3.4ton/ha/yr. The
sediment yield distribution analysis showed that11sub basins were relatively high sediment yield
rate (> 3ton/ha/yr.) among the 23 sub basins. Two watershed management scenarios were simulated
as sediment inflow rate reduction option from the existing baseline condition. The result showed
that the mean annual sediment yield at the outlet can be reduced by 24% by applying filter strips, 26.7% by applying grassed waterway
