METEOROLOGICAL AND SULPHUR DIOXIDE DISPERSION MODELLING FOR AN INDUSTRIAL COMPLEX NEAR MEXICO CITY METROPOLITAN AREA

Abstract

Major sulphur dioxide emissions in Mexico are due largely to fuel of oil refining and coal combustion. In Tula-VitoApasco industrial corridor (TVA) are located two important sources of SO2: the “Miguel Hidalgo†refinery and the “Francisco Perez Rios†power plant. Due to from March 25 to April 22 of 2006 a major field campaign took place as part of a collaborative research program called MILAGRO. Data collected around the Industrial Complex were used to: a) evaluate the air quality to local and regional scale; b) study the structure of the atmospheric boundary layer (BL); and c) validate meteorological and dispersion models. In this study we presented the behaviour of daytime BL, and the results of meteorological and dispersion modelling for selected episodes of high sulfur dioxide (SO2). The Regional Atmospheric Modeling System (RAMS) and the Hybrid and Particle Concentration Transport Model (HYPACT) were used to evaluate the impact of SO2 emissions to regional scale. For modelling, we selected the days where higher mean daily levels of SO2 surface concentrations were observed, these corresponded to March 31 and April 6. The results indicate that: (1) The daytime BL in TVA, exhibited a normal behavior, a stable layer or thermal inversion close to surface was observed at 0800 LST (up to 80% of the cases), then the mixing height (MH) growths, with a growth rate of 313 m h-1 (between 0800 to1200 LST). The most rapid MH growth happened between 1200 to 1500 LST; (2). The maximum MH was observed at 1500 LST (90% of the cases); the mean maximum MH was close to 2794 m AGL; (3) Potential temperature and humidity profiles showed a normal behavior; (4) High persistence in wind direction (> 0.6) close to surface up to 500 m AGL, was observed at 1500, and 1800 LST, at the same time, a low level jet, penetrating from the NE, with wind speed between 6 to 8 m s-1 was observed. Meteorological modelling was used to determine the circulation patterns in the region, and as input to dispersion modelling. Validation of the meteorological model was focused in the variables used as input in the dispersion model, consequently simulated and observed values of profiles of wind speed and direction, potential temperature, and specific humidity and were compared. RAMS mixing heights were overestimated at late morning (0800 LST), but during the afternoon were found to be in good agreement with observations. Simulated potential temperature and specific humidity profiles showed good agreement with the corresponding observed profiles. Simulated wind speed profiles and surface winds presented similar behavior that observations, but surface wind speed was overestimated at late morning and underestimated at the early evening. HYPACT simulations indicate that air quality standard for SO2 (=0.13 ppm or 341µgm-3 in 24 hours) was exceed close to source emissions. Models simulations indicate that TVA emissions can reach the Mexico City. HYPACT performed well the general behaviour of surface concentrations of SO2 but fails to simulate the observed peaks of the pollutant.