MODELLING OF KANGIMI DAM WATERSHED HYDROLOGICAL PROCESSES USING GIS AND SWAT MODEL
No Thumbnail Available
Date
2019-01-11
Journal Title
Journal ISSN
Volume Title
Publisher
Ahmadu Bello University Zaria
Abstract
This study focused on application of physically based hydrological model, Soil and Water
Assessment Tool (SWAT) interfaced with ARCGIS software over the Kangimi dam sub-
watershed, located in Kangimi river sub-basin, in Igabi Local Government Area, about 37km
away from Kaduna metropolis, Kaduna State, Nigeria.The watershed was delineated with 10
sub-basins, 39 hydrological response units (HRUs) were defined, which are areas with similar
land use, soil, and slope characteristic, the watershed has a total surface area of 349.94km2
and a
corresponding perimeter of 156.82km.The maximum and minimum elevation in the study area
were determined to be 784m and 512m respectively.The program SUFI-2 in SWAT-CUP
package was used for sensitivity analysis, the parameters found to be most sensitive are curve
number (CN2), threshold water depth aquifer (GWQMN) followed by, soil available water
capacity (SOL_AWC),groundwater delay time (GW_DELAY), groundwater „„revaporation‟‟
coefficient (ESCO), effective hydraulic conductivity (SOL_K) and base flow alpha factor
(ALPHA_BF.gw) as relative to the determination of surface runoff.The model was executed
from 1979 to 2014using SCS curve number method for estimation of surface runoff, Hargreaves
method for potential evapotranspiration and Variable-storage method for channel routing.The
calibration and validation of the model produced good simulation results based on the objective
functions (p-factor=0.77, r-factor=0.71) and (p-factor=0.83, r-factor=0.75) for calibration and
validation respectively, after achieving 500 simulations. The model performance was evaluated
and found to be very good for both calibration and validation period of historical discharge
datawith R
2
and NSE to be 92% and 82%, for calibration, and 93% and 86%, for validation
respectively. The watershed hydrology was simulated in response to different LULC and climate
changes, the surface runoff, evapotranspiration, contribution of groundwater to surface runoff,
deep aquifer recharge and total average annual water yield at the watershed outlet for the
simulation period were 387.37mm, 509.3mm, 248.22mm, 15.19mm and 655.51mm respectively.
This interesting performance obtained with the ArcSWAT model suggests that SWAT model
could be a promising decision support tool for sustainable management of water resources.