Impacts of boundary layer mixing on pollutant vertical profiles in the lower troposphere: Implications to satellite remote sensing

Mixing in the planetary boundary layer (PBL) affects vertical distributions of air tracers in the lower troposphere. An accurate representation of PBL mixing is critical for chemical-transport models (CTMs) for applications sensitive to simulations of the vertical profiles of tracers. The full mixing assumption in the widely used global CTM GEOS-Chem has recently been supplemented with a non-local PBL scheme. This study analyzes the impact of the non-local scheme on model representation of PBL mixing, consequences for simulations of vertical profiles of air tracers and surface air pollution, and implications for model applications to the interpretation of data retrieved from satellite remote sensing. The non-local scheme significantly improves simulations of the vertical distributions for NO2 and O-3, as evaluated using aircraft measurements in summer 2004. It also reduces model biases over the U.S. by more than 10 ppb for surface ozone concentrations at night and by 2-5 ppb for peak ozone in the afternoon, as evaluated using ground observations. The application to inverse modeling of anthropogenic NOx emissions for East China using satellite retrievals of NO2 from OMI and GOME-2 suggests that the full mixing assumption results in 3-14% differences in top-down emission budgets as compared to the non-local scheme. The top-down estimate combining the non-local scheme and the Lin et al. inverse modeling approach suggests a magnitude of 6.6 TgN yr(-1) for emissions of NOx over East China in July 2008 and 8.0 TgN yr(-1) for January 2009, with the magnitude and seasonality in good agreement with bottom-up estimates. (C) 2010 Elsevier Ltd. All rights reserved.