H. Elbern et al., DEEP STRATOSPHERIC INTRUSIONS - A STATISTICAL ASSESSMENT WITH MODEL GUIDED ANALYSES, Atmospheric environment, 31(19), 1997, pp. 3205-3224
A statistical assessment of deep intrusions of stratospheric air based
on records of two adjacent mountain stations of the northern Alps at
different altitudes is presented. Ten years recordings of beryllium ac
tivity, ozone concentrations, and relative humidity at the Zugspitze s
ummit (2962 m a.s.l.), as well as ozone and relative humidity at the W
ank summit (1776 m a.s.l., 15 km distance) were analyzed 195 stratosph
eric intrusion events could unambiguously be identified for the Zugspi
tze, whereas 85 intrusion events were found for the Wank. No event cou
ld be reliably identified at the valley floor station at Garmisch-Part
enkirchen (740 m a.s.l.). There is a pronounced seasonal cycle of the
frequency of events showing highest activity during fall, winter, and
spring, whereas low activity is found during summer. By assessing aver
age events it was possible to infer the monthly mean enrichment rate o
f the lower tropospheric ozone concentration by deep stratospheric int
rusions. It was found that at least 3% of the ozone burden is replaced
every month on an annual average. Three events of moderate strength w
ere taken to be further analyzed by mesoscale meteorological model sim
ulations with subsequent trajectory studies. In two cases the intrusio
n of stratospheric air was induced by tropopause foldings. In the thir
d case a cut-off low with an associated fold was responsible, for the
increased exchange. All three cases revealed that the ingress of strat
ospheric air observed at the mountain station is a non-local process i
nduced more than 2000 km apart. Transport over these distances took ab
out 2-4 days. Along the pathways through the tropopause the air parcel
s are shown to subside from the tip of the folds at 400-500 hPa down t
o about 700 hPa to reach the Zugspitze measurement station. (C) 1997 E
lsevier Science Ltd.