Modeling some of the Factors Affecting Rill Erosions Using Fuzzy Inference System

Hosseini, Seyyedeh Motahareh; Mosaedi, Abolfazl

Modeling some of the Factors Affecting Rill Erosions Using Fuzzy Inference System

Document Type : Complete scientific research article

Authors

¹ Former M.Sc student of Rangeland Management, Faculty of Natural Resource and Environment, Ferdowsi University of Mashhad. Mashhad - Iran

² Professor, Faculty of Natural Resources and Environment, Ferdowsi University of Mashhad

Abstract

One of the major challenges in planning and management of natural ecosystems such as water and soil resources is lack of understanding the complexities and dynamics of these ecosystems in conventional models. That based on basic idea of fuzzy theory, this limit can be entered in the model. In this study, the fuzzy logic was used to model rill erosion (as a complex natural phenomenon). Various fuzzy models and different methods of defuzzification mechanism were constructed in Mamdany method to determine the optimal model to forecast tracks per unit area of rill erosion in the study area. In all these models, the parameters such as the surface gravel, silt percentage and mutual effects between length and degree of slope (L*S) were considered as input factors of model and cross-section of rills per unit length were considered as the output factors. The results showed that the Gaussian membership function with the defuzzification of smallest maximum is the optimal model for predicting rill erosion in Ahmad-Abad area of Mashhad. The choice of this model, the statistical combination of Ideal point error (IPE), the generalizability of the model was given. According to the results obtained, in the erosion modeling using the softer functions such as Gaussian, will lead to better results than the sharp and simple functions such as trapezoidal and triangular. It is suggested that further research on the use of fuzzy techniques, neural networks and artificial intelligence in the modeling and management of natural resources is done.

Keywords