ABSTRACT
This study presents a comparison of SO2 concentrations simulated by a dispersion model, using as input both emission data inventories of the Mexican petroleum company Petroleos Mexicanos and emissions estimated by the Ozone Monitoring Instrument on board of the Aura satellite, contrasting results with data measured by an on-ground sensor located in the city of Villahermosa, Tabasco, Mexico, during the month of February 2013. The emissions used came from both oil extraction facilities located in the Gulf of Mexico and oil installations located on land. The days selected to be modeled were those when the winds had a northern component, and there was an observable emission detected by the Aura satellite coming from the emission sources. Evaluation analysis of the results showed that satellite data overestimate SO2 concentration, while the inventories of Petroleos Mexicanos underestimate it. Good correlations (r, r2, and concordance index) were observed between simulated and measured concentrations when the winds had a high frequency of directions of northern component and not-so weak speeds. It is recommended that Mexican oil company calculates more precise and updated emission inventories, considering not only a higher sulfur content of the gas emitted to the atmosphere and its combustion process but also from other existing sources of SO2 emission in this region.
