Distribution Modeling of Protective and Valuable Plant Species in the Tourist Area of Polour Using Generalized Linear Model (GLM) and Generalized Additive Model(GAM)

The prediction models of geographical distribution of the plant species are probabilistic and static models. They determinate the mathematical equations governing on geographical distribution of species with their current environment and environmental factors effective on the distribution of species. The aim of the current Study is to review the performance of Generalized Linear Model (GLM) and Generalized Additive Model )GAM(in determination of the relationship between vegetation and environmental factors in Polour rangeland. Stratified random sampling method was used. Five dominant species include Astragalus ochrodeucus, Ferula gumosa, Thymus kotschyanus,Onobrychis Cornata and Agropyron Sp were identified. the studied environmental factors were including 13 soil characteristics, 3 topographic factors and 3 climatic factors. The analyses were performed using Presence-Absence and GRASP package of R software. Also for evaluating the performance of prediction model , the statistical coefficients including AUC, AIC, RMSE and R2 were used. The results showed that the highest R2 in GLM model is related to presence of Agropyron Sp with 0.98 and also the lowest RMSE and AIC is related to Astragalus ochrodeucus with 0.29 and 12 respectively. Also in GAM model the highest R2 is related to Thymus kotschyanus with 0.98 and the lowest RMSE and AIC is related to Astragalus ochrodeucus and Ferula gumosa with 0.22 and 18.12 respectively. .. Also the highest AUC in GLM model is related to Onobrychis Cornata and in GAM model is related to Agropyron Sp with 0.86.Given the value of protecting and valuable plant species studied, the results of these models can be used in programs of conservation and improvement of the tourist area.