Sensitivity analysis and auto calibration using swat model: a case study of 1dd1 Catchment
Loading...
Date
2006
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Dar es Salaam
Abstract
A complex hydrologic model is generally characterized by a multitude of parameters. The success of a manual calibration essentially depends on the experience of the modeler and their knowledge of the basic approaches and interactions in the moreover, it can be extremely time consuming. Parameters identifiability was conducted using Sensitivity analysis based on sampling strategy that is a combination of Latin-Hypercube and one-factor-at a-time sampling which built-in a physically based model SWAT. An automatic calibration based on SCE-UA which is incorporated into SWAT model, was used for parameter optimization. Available daily streamflow data for six years (i.e. 1977-1982) and four years (i.e. 1970-1973) were used for model calibration and validation respectively. The sensitivity analysis has identified six (6) most sensitive parameters. Such parameters are Curve Number (CN2), Surface Runoff lag time(Surlag), minimum threshold depth of water for ground flow to occur(GWQMN), Recharge to deep aquifer (RCHRG-DP), Overland Slope (SLOPE), and Soil depth (SOL-Z). Besides, it was found that surface runoff parameters are mostly influencing rainfall-runoff transformation in the catchment. Calibration and validation on a daily basis gave satisfactory results with Nash-Sutcliffe coefficient of efficiency R^2 = 54% and R^2=64% respectively. The model performance on monthly basis during one hydrological year gave up to R^2= 77%. The modeling exercise has suggested that using processed or adequate and reliable spatial rainfall data and setting up a distributed watershed model could improve the results of this study.
Description
Available in print form, EAF Collection, Dr. Wilbert Chagula Library, (THS EAF TC409.J85)
Keywords
Watershed management, Calibration, Hydrology
Citation
Juma, K, A (2006) Sensitivity analysis and auto calibration using swat model: a case study of 1dd1 Catchment, Masters dissertation, University of Dar es Salaam, Dar es Salaam