Predicting the Climatic Ecological Niche of Artemisia aucheri Boiss in Central Iran using Species Distribution Modeling

Changes in the geographical distribution of plants are one of the major impacts of the climate change. This study was aimed to predict the potential changes in the distribution of Artemisia aucheri Boiss in Isfahan rangelands. Therefore, six bioclimatic variables and two physiographic variables were used under the Generalized Linear Model (GLM), Flexible Denotative Analysis (FDA), Surface Range Envelope (SRE), Random Forest (RF), and Multivariate Adaptive Regression Spline (MARS), as well as in an ensemble model framework. Among the environmental factors, the mean temperature of the coldest quarter and the mean annual precipitation played the most important role or exhibited relative importance in model fitting. Based on the model evaluation, the ensemble model and then MARS presented the most accurate prediction in the determination of the climatic habitat. There was also a significant difference between area under curve (AUC) of the SRE model with a good performance and other models with a high performance. The maximum presence of the species occurred at the 2000-3000 m elevation and the 100-250 mm annual precipitation. In both CCSM4 and MPI-ESM-LR climatic models, the highest amount of habitat displacement is likely to happen under the RCP8.5 scenario in 2070, while the lowest amount may happen under the RCP2.6 scenario in 2050. Also, the habitat of species will be displaced more in 2070 than in 2050. The results of this study can be used to prioritize conservation measures or species introduction into areas with similar ecological conditions.