Background: Air quality modeling can be considered as a useful tool to predict air quality in future and determine the control strategies of emissions abatement. In this study, AERMOD dispersion model has been applied as a tool for the analysis of the values of CO emissions from the stacks and flares of South Pars fourth Gas Refinery located in Asaluyeh.

Methods: First, the values of CO  emissions from the refinery's stacks and flares were investigated by measurement and using the emission factors in four seasons of 2013. Then, dispersion of pollutants was predicted by using the AERMOD model in the region with area of 10×10 km²  in each direction of x and y, in average times of 1, 3, 8, 24-hours and for the annual statistical period. Then the predicted and field measurement values in 9 receptors have been compared.

Results: Statistical evaluation showed that the correlation coefficient values for CO were 0.85 in spring, 0.89 in summer, 0.96 in fall, and 0.95 in winter. The maximum concentration of CO was occurred in local scale of 10×10 km².

Conclusion: Comparison of maximum 1-hour and 8-hour concentrations of the predicted results with the national and international standards showed that CO concentration is higher than standard values. In total, according to the evaluation of the predictions made, the performance of AERMOD model was acceptable in prediction of CO concentrations in the study area.