Comparison of Probit Logistic Regression and Discriminant Analysis Methods for Identification of Suitable Areas for Walnut Planting (Case Study: Baft District)

The aim of this study was to identify similar areas of physiography and climatic potential of walnut in Baft city. Logistic regression and discriminant analysis were used for the purpose of classifying of study area using a set of predictor variables to find the suitable areas to grow walnut. For this purpose, environmental conditions affecting walnut trees used included Annual Mean Temperature, Annual Precipitation, Precipitation of Driest Month, Precipitation of Wettest Month, Elevation, Slope Gradient and Aspect. 70% of the data, as training data and the remaining 30%, are used as model validation data. Using probit logistic regression and discriminant analysis, the effect of variables used on walnut in Baft County was calculated and then map of walnut was calculated. The results showed effect of physiographic and climatic variables on walnut. The results of validation data set in this study showed that probit logistic regression with 75% and discriminant analysis with 86% able to accurately explain the favorable areas for walnut cultivation. Suitable physiographic areas and climatic potentials for planting walnut trees are located in the central, northern, and northwestern regions of Baft area.