The hourly O3 concentrations at the surface level and the tropospheric NO2 column contents observed by the satellites are compared with corresponding CMAQ-predicted concentrations obtained at 36km resolution. June 1-3, 2006, were selected as design days, given that ozone episodes have been reported over three pairs of cities along the U.S./Mexico border during this period. The satellite-based tropospheric columns were from SCIAMACHY (The Scanning Imaging Absorption Spectrometer for Atmospheric Chartography) and OMI (Ozone Monitoring Instruments), provided by collaborators. The surface data were obtained from the EPA's Aerometric Information Retrieval System (AIRS). The model calculations were performed using the MM5/SMOKE/CMAQ system, together with emission from the National Emissions Inventory (NEI) database of 2001 (for the US) and 1999 (MEXICO). Some traditional statistics, e.g. correlation coefficient, mean bias (MB) and normalized mean bias (NMB) were then calculated. Process analysis, together with back-trajectory analysis, was used to assess the potential sources of ozone for these three places and their contributions.

The results show that CMAQ captures diurnal variation of surface ozone concentrations at most observational sites. Over 70% of the model cells containing surface observations showed a correlation larger than 0.6 and more than 50% showed a NMB between -25% and 25% when (AIRS) and CMAQ (lowest grid cell) are compared. The model reproduces the hot spots of high tropospheric NO2 columns well, when compared with satellite measurements.

The results of process and back-trajectory analyses indicate that the main sources of ozone are from the local chemistry and regional transport from the coastal area of the northwestern California in the case of San Diego; from the east or coastal area of northwestern California for Imperial Valley; and mid-eastern part of Texas for the case of El Paso.