Skip to main content
ARS Home » Pacific West Area » Tucson, Arizona » SWRC » Research » Publications at this Location » Publication #195226
Title: A NEW SENSITIVITY ANALYSIS FRAMEWORK FOR MODEL EVALUATION AND IMPROVEMENT USING A CASE STUDY OF THE RANGELAND HYDROLOGY AND EROSION MODEL 1835

Author
item WEI, H. - UNIVERSITY OF ARIZONA
item Nearing, Mark
item Stone, Jeffry

Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/15/2007
Publication Date: 7/25/2007
Citation: Wei, H., Nearing, M.A., Stone, J.J. 2007. A new sensitivity analysis framework for model evaluation and improvement using a case study of the rangeland hydrology and erosion model. Transactions of the ASABE. 50(3): 945¿953.

Interpretive Summary: Complex computer models are used in the field of natural resource management for many purposes. Because of the complexity of the numerical models used, and the large numbers of input factors there is a high risk for these models of having problematic or nonsensical model responses in certain applications. Sensitivity analysis (SA) is a useful tool for ascertaining whether model response is logical and reasonable. This paper describes a new method for conducting this type of analysis and it describes how to use the method for identifying model deficiencies and improving model function. The method was applied to the Rangeland Hydrology and Erosion Model (RHEM), using soil erosion response as a case study. Results of the study showed that the sensitivity of the model to changes in site conditions was interdependently related to all the other the input parameter values in a complex manner. The paper also shows how this method, combined with techniques such as statistics and scatter plots, can be used effectively to know which of the input variables are most critical to proper functioning of the model and to identify incorrect relationships in the model. The paper shows how this method can be used as an element of the iterative modeling process whereby model response can be surveyed and problems identified and corrected in order to construct a robust model. This work will help us to develop better tools for the management of natural resources across the United States.

Technical Abstract: The complexity of numerical models and the large numbers of input factors result in complex interdependencies of parameter sensitivities on input parameter values, and high risk of having problematic or nonsensical model responses in localized regions of the input parameter space. Sensitivity analysis (SA) is a useful tool for ascertaining model response to input variables. One popular method is local SA, which calculates the localized model response of output to an input factor. This paper describes a new SA method that utilizes a local sensitivity equation to build a sensitivity matrix for a model. The paper further describes how to use this sensitivity matrix for identifying model deficiencies and improving model function. The method was applied to the Rangeland Hydrology and Erosion Model (RHEM), using soil erosion response as a case study. Results of the sensitivity matrix quantified localized sensitivity, which varied and was interdependently related to the input parameter values. The paper also shows that the sensitivity matrix, combined with techniques such as correlation analysis and scatter plots, can be used effectively to compare the sensitivity of different inputs, locate sensitive regions in parameter space, decompose the dependency of the model response on the input factors, and identify nonlinear and incorrect relationships in the model. The paper shows how this method can be used as an element of the iterative modeling process whereby model response can be surveyed and problems identified and corrected in order to construct a robust model.