Evaluation of stratospheric chlorine chemistry for the Arctic spring 2005 using modelled and measured OClO column densities

• Published in: Atmospheric Chemistry and Physics
• Main Authors: H. Oetjen, F. Wittrock, A. Richter, M. P. Chipperfield, T. Medeke, N. Sheode, B.-M. Sinnhuber, M. Sinnhuber, J. P. Burrows
• Format: Article in Journal/Newspaper
• Language: English
• Published: Copernicus Publications 2011
• Subjects: Physics; QC1-999; Chemistry; QD1-999; Arctic; Ny-Ålesund; Ny Ålesund
• Online Access: https://doi.org/10.5194/acp-11-689-2011; https://doaj.org/article/ee200ffd88884b108a2f328cab667f88

Chlorine dioxide, OClO, column amounts retrieved from measurements of the SCIAMACHY satellite instrument are presented and validated by comparison with simultaneous ground-based DOAS observations. In addition, the measurements are compared to model calculations taking into account the photochemical change along the light path. Although OClO does not participate directly in the destruction of ozone, its accurate measurement as well as modelling is crucial to understand the highly perturbed chlorine chemistry in the polar vortices. SCIAMACHY OClO slant columns retrieved during spring 2005 have been quantitatively validated by comparison with slant columns retrieved from measurements made in Ny-Ålesund (79° N, 12° E) and Summit (73° N, 38° W) as well qualitatively for Bremen (53° N, 9° E). Fair to good agreement is found depending on location as well as time of year. OClO slant column densities modelled with a set of stacked box models and considering the light path through the atmosphere are also included in this comparison. The model predictions differ significantly from the measured quantities. OClO amounts are underestimated for conditions of strong chlorine activation and at large solar zenith angles. Sensitivity studies for several parameters in the stacked box model have been performed and it is inferred that using the chemistry known to date, the observed OClO cannot be adequately reproduced within the range of uncertainties given for the various model parameters.