Persistence of stratospheric ozone layers in the troposphere

On 20th March 1994 at 12Z two dry layers with high ozone mixing ratio were encountered in the free troposphere by an ozonesonde ascent at Aberystwyth (52.4 degrees N, 4.1 degrees W). The layers resembled tropopause folds, but were not associated with any significant synoptic development, and did not have enhanced static stability. We use isentropic trajectory analysis, wit h winds taken from ECMWF analyses, to look for the origin of the layers. Th e analysis suggests that the two layers had different source regions. The l ower layer spent 10 days over the Atlantic before reaching Northern Europe, and appears to have originated in the break-up of a stratospheric streamer between the 5th and 9th of March. The upper layer seems to have originated over the Western USA where the trajectories passed through a region of low Richardson number above the Rocky Mountains. In both cases, the low-water- vapour/high-ozone air mass had been advected with little mixing in the trop osphere for at least 10 days. We argue that the air must be of stratospheri c origin, since photo-chemistry cannot generate large amounts of ozone at t his time of year and the trajectories do not, in any case, point to a bound ary-layer origin. If, as the analysis suggests, the upper layer reached the troposphere by mixing across the jet then this may be evidence for stratos phere-troposphere exchange (STE) taking place other than by folding of the tropopause. The persistence of layers with anomalous chemical content in th e troposphere for so long indicates that an accurate model of tropospheric transport and mixing is needed to assess the chemical impact of STE on trop ospheric chemistry, in addition to a representation of the dynamical behavi our near to the tropopause. (C) 2000 Elsevier Science Ltd. All rights reser ved.