Assessing the photochemical impact of snow NOx emissions over Antarctica during ANTCI 2003

Surface and aircraft measurements show large amounts of reactive nitrogen over the Antarctic plateau during the ANTCI 2003 experiment. We make use of 1-D and 3-D chemical transport model simulations to analyze these measurements and assess the photochemical impact of snow NOx emissions. Boundary layer heights measured by SODAR at the South Pole were simulated reasonably well by the polar version of MM5 after a modification of ETA turbulence scheme. The average of model-derived snow NOx emissions (3.2-4.2 x 10(8) molec cm(-2) s(-1)) at the South Pole is similar to the measured flux of 3.9 x 10(8) molec cm(-2) s(-1) during ISCAT 2000. Daytime snow NO, emission is parameterized as a function of temperature and wind speed. Surface measurements of NOx HNO3 and HNO4, and balloon measurements of NO at the South Pole are reasonably simulated by 1-D and 3-D models. Compared to Twin Otter measurements of NO over plateau regions, 3-D model simulated NO concentrations are at the low end of the observations, suggesting either that the parameterization based on surface measurements at the South Pole underestimates emissions at higher-elevation plateau regions or that the limited aircraft database may not be totally representative for the season of the year sampled. However, the spatial variability of near-surface NO measured by the aircraft is captured by the model to a large extent, indicating that snow NOx emissions are through a common mechanism. An average emission flux of 0.25 kgN km(-2) month(-1) is calculated for December 2003 over the plateau (elevation above 2.5 km). About 50% of reactive nitrogen is lost by deposition and the other 50% by transport. The 3-D model results indicate a shallow but highly photochemically active oxidizing canopy enshrouding the entire Antarctic plateau due to snow NOx emissions. (C) 2007 Elsevier Ltd. All rights reserved.