Parameter Sensitivity of a Non-Linear, Process-Based Soil Erosion Model

See related content

by Mario Tiscareno-Lopez, Univ of Arizona, Tuczon, United States,
V. L. Lopes, Univ of Arizona, Tuczon, United States,
J. J. Stone, Univ of Arizona, Tuczon, United States,
L. J. Lane, Univ of Arizona, Tuczon, United States,

Document Type: Proceeding Paper

Part of: Management of Irrigation and Drainage Systems: Integrated Perspectives

Abstract: Process-based, parameter distributed models have become widely accepted tools to predict the impacts of land use changes. Unfortunately these computer models are becoming more complex because the representation of the temporal and spatial variability of the systems' behavior. However, increases in model complexity usually increase the uncertainty of model predictions. Sensitivity analysis is a technique to assess model uncertainties related to errors in parameter estimation. Sensitivity analysis based on the Monte-Carlo method provides a criterion by which to judge uncertainties in model predictions due to errors in parameter estimation when the system variability is represented in probabilistic terms. The benefits of the Monte-Carlo method for sensitivity analyses are illustrated with a complex, non-linear, process-based, parameter distributed soil erosion model designed to predict water erosion at watershed scale.

Subject Headings: Sensitivity analysis | Erosion | Parameters (statistics) | Computer models | Soil analysis | Errors (statistics) | Uncertainty principles | Monte Carlo method | Europe | Monaco | Monte Carlo

Services: Buy this book/Buy this article


Return to search