Utilization of satellite observation of ozone and aerosols in providing initial and boundary condition for regional air quality studies

Abstract

[1] To demonstrate the efficacy of satellite observations in the realization of the background and transboundary transport of pollution in regional air quality modeling practices, satellite observations of ozone and aerosol optical depth were incorporated in the EPA Models-3 Community Multiscale Air Quality (CMAQ) model (http://www.cmascenter.org). Observations from Ozone Monitoring Instrument (OMI) aboard NASA's Aura satellite and AOD products from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard Terra (EOS AM) and Aqua (EOS PM) satellites were used to specify initial and lateral boundary conditions (IC/BC) for a simulation that spanned over August 2006. The tools and techniques using the satellite data were tested in the context of current regulatory air quality modeling practices. Daily satellite observations were remapped onto the modeling domain and used as IC/BC for daily segments of a month-long simulation and the results were evaluated against surface and ozonesonde observations. Compared to the standard application of CMAQ, OMI O3 profiles significantly improved model performance in the free troposphere and MODIS aerosol products substantially improved PM2.5 predictions in the boundary layer. The utilization of satellite data for BC helped in the realization of transboundary transport of pollution and was able to explain the recirculation of pollution from Northeast Corridor to the southeastern region. Ozone in the mid- to upper-troposphere was largely dominated by transport and thus benefited most from satellite provided BC. The ozone within the boundary layer was mostly affected by fast production/loss mechanisms that are impacted by surface emissions, chemistry and removal processes and was not impacted as much. A case study for August 18–22 demonstrated that model errors in the placement of a stationary front were the main reason for errors in PM2.5 predictions as the front acted as a boundary between high and low PM2.5 concentrations.