Agriculture is the main non-point polluter source of groundwater in irrigated areas where fertilizers and other agrochemicals are the main contaminants in the water that drains out of the root zone to recharge the aquifer. Generally, the nitrate concentrations occur from fertilizers that are dissolved in percolation losses in farmlands. The concentration of nitrates in the percolated water depends on distributed water availability and nitrogen balances over the region. Its concentration in the groundwater depends on the total recharge, pollution loading, groundwater flow and solute traort within the aquifer. In this study, we compare the performance of predicting nitrate distribution in Arak alluvial groundwater resources using Fuzzy logic, geostatistical spatio-temporal and Support Vector Regression (SVR) methods. The available input data were the nitrate concentration, pH, electrical conductivity, total dissolved solids, total coli-form, BOD and COD measured on 40 groundwater samples taken from spatially distributed wells at every season during autumn 1385 to summer 1386. In order to evaluate and validate the output predicted data produced by aforementioned methods, all wells were re-sampled in summer 1391. In the first step, the nitrate concentrations in the aquifer were predicted as a function of input variables using SVR and Fuzzy logic methods. The predicted output results from SVR method gave rise to 65% correlation with observed values whereas this measure amounts to just 50% for Fuzzy logic method. Employing geostatistical spatio-temporal improved the prediction accuracy amounting to more than 93% correlation between observed and predicted values. As is seen from prediction results obtained using geostatistical spatio-temporal method, the variation of nitrate concentrations ranges from local to regional scales depending on seasonal rainfall variations. In addition, the spatial variation of nitrate contamination plume moves to the East and South of the plain, which coincides with the measured values in summer 1391. Finally, the prediction results reveal the existence of high nitrate concentrations and other pollutants such as TDS and EC suggesting urgent attention and action of authorities to the issue of rapid contaminant development at Arak alluvial aquifer.