OMI NO2 column densities over North American urban cities: the effect of satellite footprint resolution

H. C. Kim, P. Lee, L. Judd, L. Pan, B. Lefer
2015 Geoscientific Model Development Discussions  
Nitrogen dioxide vertical column density (NO<sub>2</sub> VCD) measurements via satellite are compared with a fine-scale regional chemistry transport model, using a new approach that considers varying satellite footprint sizes. Space-borne NO<sub>2</sub> VCD measurement has been used as a proxy for surface nitrogen oxide (NO<sub><i>x</i></sub>) emission, especially for anthropogenic urban emission, so accurate comparison of satellite and modeled NO<sub>2</sub> VCD is important in determining the
more » ... in determining the future direction of NO<sub><i>x</i></sub> emission policy. The National Aeronautics and Space Administration Ozone Monitoring Instrument (OMI) NO<sub>2</sub> VCD measurements, retrieved by the Royal Netherlands Meteorological Institute (KNMI), are compared with a 12 km Community Multi-scale Air Quality (CMAQ) simulation from the National Oceanic and Atmospheric Administration. We found that OMI footprint pixel sizes are too coarse to resolve urban NO<sub>2</sub> plumes, resulting in a possible underestimation in the urban core and overestimation outside. In order to quantify this effect of resolution geometry, we have made two estimates. First, we constructed pseudo-OMI data using fine-scale outputs of the model simulation. Assuming the fine-scale model output is a true measurement, we then collected real OMI footprint coverages and performed conservative spatial regridding to generate a set of fake OMI pixels out of fine-scale model outputs. When compared to the original data, the pseudo-OMI data clearly showed smoothed signals over urban locations, resulting in roughly 20–30 % underestimation over major cities. Second, we further conducted conservative downscaling of OMI NO<sub>2</sub> VCD using spatial information from the fine-scale model to adjust the spatial distribution, and also applied Averaging Kernel (AK) information to adjust the vertical structure. Four-way comparisons were conducted between OMI with and without downscaling and CMAQ with and without AK information. Results show that OMI and CMAQ NO<sub>2</sub> VCDs show the best agreement when both downscaling and AK methods are applied, with correlation coefficient <i>R</i> = 0.89. This study suggests that satellite footprint sizes might have a considerable effect on the measurement of fine-scale urban NO<sub>2</sub> plumes. The impact of satellite footprint resolution should be considered when using satellite observations in emission policy making, and the new downscaling approach can provide a reference uncertainty for the use of satellite NO<sub>2</sub> measurements over most cities.
doi:10.5194/gmdd-8-8451-2015 fatcat:5ui6bhdzz5ciboqze3hn4hthyq