Analysis of SWAT 2005 Parameter Sensitivity with LH-OAT Method

The Soil and Water Assessment Tool (SWAT) is widely used as watershed simulation tool, and the parameter sensitivity of the SWAT model is the foundation of model calibration. For the latest version of the SWAT model (SWAT 2005), the LH-OAT automated sensitivity analysis tool uses a stratified sampling approach and attributes the output changes to the change of a single parameter at the sampling point. But there is still some uncertainty of sensitivity result made by this tool. A SWAT 2005 model for the Jiyunhe Basin in Tianjin of China was constructed and the LHOAT method was applied for sensitivity analysis. The sensitivity of the predefined hydrologyrelated parameters was analysed. 16 sensitive parameters showed a disparity of sensitivity rank under different scenarios. The scenarios were configured considering the parameter varying methods, parameter bound, the number of intervals within parameter bound, and the random seed number. Some parameters showed evident disparity of sensitivity when different varying methods or LH-OAT configurations were applied. Under some scenarios, a single parameter’s sensitivity rank changed from 14th to second and five parameters’ sensitivity rank changed more than five. The nonlinearity of the relationship between the output and the parameters, the spatial distribution of the parameters and the bound inconsistency are the main causes of the sensitivity disparity of each individual parameter in the proposed scenarios. The conclusion is that the sensitivity of the hydrological-related parameters determined by the LH-OAT method is strongly influenced by the parameter varying methods and the random seed number that were required as the configuration of the LH-OAT method.

[1]  Hans-Georg Frede,et al.  Comparison of two different approaches of sensitivity analysis , 2002 .

[2]  Ronald L. Iman,et al.  Risk methodology for geologic disposal of radioactive waste: small sample sensitivity analysis techniques for computer models, with an application to risk assessment , 1980 .

[3]  Saltelli Andrea,et al.  Global Sensitivity Analysis: The Primer , 2008 .

[4]  Jeffrey G. Arnold,et al.  The Soil and Water Assessment Tool: Historical Development, Applications, and Future Research Directions , 2007 .

[5]  Richard J. Beckman,et al.  A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output From a Computer Code , 2000, Technometrics.

[6]  Steffen L. Lauritzen,et al.  Some modern applications of graphical models , 2003 .

[7]  Max D. Morris,et al.  Factorial sampling plans for preliminary computational experiments , 1991 .

[8]  Sang Xue-feng.  Application of improved SWAT model to area with strong human activities , 2008 .

[9]  John R. Williams,et al.  LARGE AREA HYDROLOGIC MODELING AND ASSESSMENT PART I: MODEL DEVELOPMENT 1 , 1998 .

[10]  Stephen R. Workman,et al.  SIMULATION OF DAILY AND MONTHLY STREAM DISCHARGE FROM SMALL WATERSHEDS USING THE SWAT MODEL , 2000 .

[11]  P. Vanrolleghem,et al.  Hydrodynamic modelling with Soil and Water Assessment Tool (SWAT) for predicting dynamic behaviour of pesticides. , 2004 .

[12]  Raghavan Srinivasan,et al.  Regional estimation of base flow and groundwater recharge in the Upper Mississippi river basin , 2000 .

[13]  M. Crawford,et al.  Theory and methods , 1980 .

[14]  J. Arnold,et al.  SWAT2000: current capabilities and research opportunities in applied watershed modelling , 2005 .

[15]  MARGARET GOWING,et al.  A Model of Development , 1973, Nature.

[16]  Fred Worrall,et al.  Sensitivity analysis and identification of the best evapotranspiration and runoff options for hydrological modelling in SWAT-2000 , 2007 .

[17]  G. Whittaker,et al.  Sensitivity Analysis And Interdependence Of The SWAT Model Parameters , 2007 .

[18]  R. Srinivasan,et al.  A global sensitivity analysis tool for the parameters of multi-variable catchment models , 2006 .