Abstract
Forecasting models of Land Use/ Cover changes are the main resources for managers and policy-makers in order to develop a sustainable land management plan. Changes of Orchard-lands have an effect on water resources as well as soil permeability. Thus simulation of this land use changes, in areas where there is a shortage of water resources, can provide more information about the occurred changes during a specified time scales along current management. The present study was carried out to simulate and predict the spatial-temporal changes of the orchard by 2026. For this purpose, Geomod method was used to simulate spatial changes of the orchard. Due to the lack of ability of this model in temporal simulation, Markov chain analysis method was used to solve the mentioned problem with the error proportional of 0.012. Orchard was extracted using Landsat 5, 7, and 8 satellite data after necessary corrections as well as SVM in 1987, 2000, and 2013. Then, to understand the impact of each of the criteria used to change this type of land use, instead of Delphi methods, logistic regression, Fuzzy standardization and, after all, WLC were used. The ROC index was used to validate the model. The results showed, this model has a good performance to simulate spatial changes because of area under Curve 0.91 for both of the 2000 and 2013. In the 26-years period, there are 294 hectares of orchard development, and the hybrid model showed that this land use will increase to 304 hectares till 2026
)