Document Type


Publication Date



On 28 February 2000, a volcanic cloud from Hekla volcano, Iceland, was serendipitously sampled by a DC-8 research aircraft during the SAGE III Ozone Loss and Validation Experiment (SOLVE I). It was encountered at night at 10.4 km above sea level (in the lower stratosphere) and 33–34 hours after emission. The cloud is readily identified by abundant SO2 (≤1 ppmv), HCl (≤70 ppbv), HF (≤60 ppbv), and particles (which may have included fine silicate ash). We compare observed and modeled cloud compositions to understand its chemical evolution. Abundances of sulfur and halogen species indicate some oxidation of sulfur gases but limited scavenging and removal of halides. Chemical modeling suggests that cloud concentrations of water vapor and nitric acid promoted polar stratospheric cloud (PSC) formation at 201–203 K, yielding ice, nitric acid trihydrate (NAT), sulfuric acid tetrahydrate (SAT), and liquid ternary solution H2SO4/H2O/HNO3 (STS) particles. We show that these volcanically induced PSCs, especially the ice and NAT particles, activated volcanogenic halogens in the cloud producing >2 ppbv ClOx. This would have destroyed ozone during an earlier period of daylight, consistent with the very low levels of ozone observed. This combination of volcanogenic PSCs and chlorine destroyed ozone at much faster rates than other PSCs that Arctic winter. Elevated levels of HNO3 and NOy in the cloud can be explained by atmospheric nitrogen fixation in the eruption column due to high temperatures and/or volcanic lightning. However, observed elevated levels of HOx remain unexplained given that the cloud was sampled at night.

Publisher's Statement

© 2006 American Geophysical Union. Publisher's version of record:

Publication Title

Journal of Geophysical Research


Publisher's PDF