Sensitivity of tropical tropospheric composition to lightning NOx production as determined by replay simulations with GEOS‐5. Issue 16 (20th August 2015)

• Record Type: Journal Article
• Authors: Liaskos, Christina E.; Allen, Dale J.; Pickering, Kenneth E.
• Is Part Of: Journal of geophysical research. Volume 120:Issue 16(2015:Sep.)
• Page Start: 8512
• Page End: 8534
• Publication Date: 2015-08-20
• Subjects: Atmospheric physics -- Periodicals; Geophysics -- Periodicals; 551.5
• DOI: 10.1002/2014JD022987
• Languages: English
• ISSNs: 2169-897X

Abstract

<abstract abstract-type="main" id="jgrd52218-abs-0001"> <title>Abstract</title> <p id="jgrd52218-para-0001">The sensitivity of tropical tropospheric composition to the source strength of nitrogen oxides (NO<sub>x</sub>) produced by lightning (LNO<sub>x</sub>) is analyzed for September through November 2007 using the NASA GEOS‐5 model constrained by MERRA fields, with full GMI stratospheric‐tropospheric chemistry and an LNO<sub>x</sub> algorithm that is appropriate for use in a climate modeling setting; satellite retrievals from OMI, TES, and OMI/MLS; and in situ measurements from SHADOZ ozonesondes. Global mean LNO<sub>x</sub> production rates of 0 to 492 mol NO flash<sup>−1</sup> and the subsequent responses of NO<sub>x</sub>, ozone (O<sub>3</sub>), hydroxyl radical (OH), nitric acid (HNO<sub>3</sub>), peroxyacetyl nitrate (PAN), and NO<sub>y</sub> (NO<sub>x</sub> + HNO<sub>3</sub> + PAN) are investigated. The radiative implications associated with LNO<sub>x</sub>‐induced changes in tropospheric O<sub>3</sub> are assessed. Increasing the LNO<sub>x</sub> production rate by a factor of 4 (from 123 to 492 mol flash<sup>−1</sup>) leads to tropical upper tropospheric enhancements of greater than 100% in NO<sub>x</sub>, OH, HNO<sub>3</sub>, and PAN. This increase in LNO<sub>x</sub> production also leads to O<sub>3</sub> enhancements of up to 60%, which subsequently yields a factor‐of‐three increase in the mean net radiative flux at the tropopause. An LNO<sub>x</sub> source of 246 mol flash<sup>−1</sup> agrees reasonably well with measurements, with an approximate factor‐of‐two uncertainty due to the short length of the study, inconsistencies in the observational data sets, and systematic biases in modeled LNO<sub>x</sub> production. Further research into the regional dependencies of lightning flash rates and LNO<sub>x</sub> production per flash, along with improvements in satellite retrievals, should help resolve the discrepancies that currently exist between the model and observations.</p> </abstract>