Monitoring of rangeland forage production at specified spatial and temporal scales is necessary for grazing management and also for implementation of rehabilitation projects in rangelands. This study focused on the capability of statistical models combined with GIS techniques to explore the relative importance of environmental variables for predicting forage production and the impact of climate change on rangeland forage production in Isfahan Province. Systematic randomized sampling in 16 ecological sites was performed during 10 years (1998-2007). Four and six parallel transects with 400 and 200 meters lengths were established in each site of steppe (11 ecological sites) and semi steppe (5 ecological sites) regions respectively. Fifteen plots with size of 1 by 2 meters and ten plots with size of 1 by 1 meter were established along each transect in steppe and semi steppe regions respectively. The production of plant species was recorded in each plot using double sampling method. Specifically, a dataset of 115 monitored records of forage production were collected from 16 rangeland sites during 10 years. Neural network models (ANN), generalized additive model (GAM), generalized linear model (GLM) and nonparametric multiplicative regression (NPMR) were designed using the monitored forage production values and available environmental data (including climate, topography, soil, plant vegetation indices and soil moisture index), and the performance of each model was assessed using correlation coefficient(r), root-mean-square error (RMSE) and model efficiency factor(MEF). The best neural network model was then selected and further applied to predict the forage production of rangelands in the future (in 2030 and 2080) under A1B climate change scenario using Hadley Centre coupled model. The present and future forage production maps were also produced. This study applied three different methods, i.e. the moving window regression technique, NPMR, and GLM, to downscale the annual mean temperature and annual precipitation in central Iran from layers with a pixel size of 1 × 1 km to layers with a pixel size of 250 m. Elevation was used as the covariate in all three methods. According to the results of Canonical Correspondence Analysis (CCA), minimum temperature of coldest month, annual mean temperature, electrical conductivity, calcium carbonate and sand percentage in steppe region and precipitation of wettest month, annual precipitation, elevation, organic matter and potassium in semi steppe region were identified as the most important ecological factors affecting plant species production. Using ANN model based on climatic and topographic data, forage production varied from 25.6 to 574.1 kg/h. Using GAM model based on climatic, topographic and soil data, forage production varied from 0 to 890.2 kg/h. Using NPMR model based on climatic variables, topographic data, vegetation indices and soil moisture index, forage production varied from 20.5 to 458.2 kg/h. Using NPMR model based on all environmental variables (climate, topography, soil, vegetation indices and soil moisture index) forage production varied from 30.9 to 531.2 kg/h. Elevation, slope, aspect, mean annual temperature, minimum temperature of coldest month, calcium carbonate percentage, phosphorus, potassium and soil moisture index were the most important predictors affecting rangeland forage production. Downscaling of bioclimatic variables showed that moving window regression technique had the best performance and highest accuracy in downscaling annual precipitation data. The NPMR had better performance and higher accuracy than the GLM in this regard. all three models (moving window regression technique, NPMR and GLM) could downscale the mean annual temperature data with similar statistical index values. Under climate change scenario (HadCM3/A1B), the forage production was predicted to vary from 25.4 to 567.5 kg/ha and 26.2 to 553.2 kg/ha in 2030 and 2080 respectively. The prediction maps of forage production in the future indicated that the area with low level of forage production (0–100 kg/h) will increase while the areas with other levels of forage production (?100 kg/h) will decrease both in 2030 and in 2080. It was predicted that forage production of rangelands will decrease in the next couple of decades, especially in the western and southern parts of Isfahan Province. These changes are more pronounced in elevations between 2200 and 2900 m. The results of this research can help rangeland managers to advance their management and rehabilitation goals by predicting forage production and grazing capacity in present and under climate change effect. Keywords : Climate change, Rangeland production, Ordination,Spatial downscaling, Statistical models