Estimating Lightning NOx Emissions for Regional Air Quality Modeling

Lightning emissions have long been recognized as an important source of nitrogen oxides (NOx) on a global scale, and an essential emission component for global atmospheric chemistry models. However, only in recent years have regional air quality models incorporated lightning NOx emissions into simulations. The growth in regional modeling of lightning emissions has been driven in part by comparisons with satellite-derived estimates of column NO₂, especially from the Ozone Monitoring Instrument (OMI) aboard the Aura satellite.

We present and evaluate a lightning inventory for the EPA Community Multiscale Air Quality (CMAQ) model. Our approach follows Koo et al. [2010] in the approach to spatially and temporally allocating a given total value based on cloud-top height and convective precipitation. However, we consider alternate total NOx emission values (which translate into alternate lightning emission factors) based on a review of the literature and performance evaluation against OMI NO₂ for July 2007 conditions over the U.S. and parts of Canada and Mexico. The vertical distribution of lightning emissions follow a bimodal distribution from Allen et al. [2012] calculated over 27 vertical model layers.

Total lightning NO emissions for July 2007 show the highest above-land emissions in Florida, southeastern Texas and southern Louisiana. Although agreement with OMI NO₂ across the domain varied significantly depending on lightning NOx assumptions, agreement among the simulations at ground-based NO₂ monitors from the EPA Air Quality System database showed no meaningful sensitivity to lightning NOx. Emissions are compared with prior studies, which find similar distribution patterns, but a wide range of calculated magnitudes.