Photochemical grid models are used to evaluate air pollution control strategies by simulating the physical and chemical processes that influence pollutant concentrations. Their accuracy depends on the accuracy of input data used for anthropogenic and biogenic emissions, land surface characteristics, initial and boundary conditions and meteorological conditions. Evaluation of model performance requires sufficient ambient data. This work develops approaches for applying satellite data to allow more frequent and timely estimates of parameters required to estimate emissions and pollutant removal processes for regional air quality modeling.

Land use and land cover (LULC) data prepared from remote sensing satellite data were evaluated for use as inputs to photochemical grid models for estimating dry deposition velocities and biogenic emissions. The results indicated that satellite-based data derived from the Moderate Resolution Imaging Spectroradiometer instrument can be used to provide periodic updates to LULC information used in photochemical models. The sensitivity of predicted ozone concentrations to LULC data used for biogenic emission estimates was examined by comparing the database currently used for modeling in southeastern Texas with a new database prepared from Landsat satellite imagery and field data. The satellite data and image classification techniques provide useful tools for mapping and monitoring changes in LULC. However, field validation is necessary to link species and biomass densities to the classification system needed for accurate biogenic emissions estimates, especially in areas that have dense concentrations of species that emit high levels of biogenic hydrocarbons.

The application of NO₂ measurements from the Ozone Monitoring Instrument (OMI) to validation of NO_x emission estimates and identification of emission sources for regional air quality modeling for Texas was examined. OMI observations can be used to identify regions with changes in emissions over time or where estimates have large uncertainties and to evaluate the effectiveness of emission reduction strategies. For example, in the Dallas-Fort Worth area, observed NO₂ column densities from OMI indicate that emission controls are less effective than anticipated due to increased area source emissions. The techniques developed in this work have broad applicability in the advancement of methods for including satellite remote sensing data in regional air quality modeling.