Ozone loss driven by nitrogen oxides and triggered by stratospheric warmings can outweigh the effect of halogens

Ozone loss in the lower and middle stratosphere in spring and summer, in particular over polar regions, is driven mainly by halogens and nitrogen oxides (NOx). Whereas the stratospheric chlorine levels are expected to decrease in the future, the role of NOx for the O-3 budget in a changing climate is not well quantified. Here we combine satellite measurements and model simulations to diagnose the accumulated O-3 loss during winter and spring 2002-2003 in the Arctic polar stratosphere. We show that in a winter stratosphere strongly disturbed by warmings, O-3 loss processes driven by halogens and NOx can significantly overlap within the polar column and become comparable in magnitude even if a significant, halogen- induced O-3 loss has occurred. Whereas, until the beginning of March 2003, polar column O-3 loss was mainly caused by the halogen chemistry within the vortex at an altitude around 18 km, the chemical O3 destruction in March and April was dominated by the NOx chemistry in O3-rich air masses transported from the subtropics and mixed with the polar air above the region affected by the halogens. This NOx-related O-3 loss started around mid-December 2002 in subtropical air masses above 30 km that moved poleward after the major warming in January, descended to 22 km with an increasing magnitude of O-3 loss and reached surprisingly high values of up to 50% local loss around the end of April. To some extent, the NOx-driven O-3 loss was enhanced by mesospheric air trapped in the vortex at the beginning of the winter as a layer of few km in the vertical and transported downward within the vortex. The effect of NOx transported from the subtropics dominated the O-3 loss processes in the polar stratosphere in spring 2003, both relative to the effect of the halogens and relative to the contribution of the mesospheric NOx sources. A comparison with the 1999/2000 Arctic winter and with the Antarctic vortex split event in 2002 shows that wave events triggered by stratospheric warmings may significantly enhance O3 loss driven by NOx when O-3- and NOx- rich air masses from the subtropics are transported poleward and are mixed with the vortex air.