The region that was the focus of the Southern oxidant study (SOS) was
divided into five 5 degrees latitude by 5 degrees longitude areas (i.e
., areas of about 500 km by 500 km) and a 10-year (1981-1990) time ser
ies of the diurnal maximum ozone concentration (DMOC) was created in e
ach area which was used to examine and analyze the subregional charact
eristics of ozone. The time series for the DMOC was divided into four
terms: the long-term mean (i.e. the 10 yr mean in this case); the intr
a-annual perturbation; the inter-annual perturbation; and the synoptic
perturbation. The results indicated that all four terms in the time s
eries, as well as the number and characteristics of exceedances, Varie
d substantially from region to region in the SOS region. Certain aspec
ts of the regional variation of the long-term mean could be explained
by variations in solar radiation. The variations of the intra-annual p
erturbation appear to be correlated with the variations of emissions,
and that for the interannual perturbation, with climate fluctuations.
A strong relationship between synoptic perturbations in specific regio
ns was also found which was not expected. The synoptic perturbation is
forced by synoptic weather systems. Since the spatial scale of synopt
ic weather systems is about 1000 km and the weather characteristics in
the systems can change markedly over 1000 km, strong relationships be
tween synoptic perturbations from region to region are highly unlikely
over a 10 yr period. This suggest that possibly factors other than sy
noptic weather systems may have played a role in producing the relatio
nship (C) 1998 Elsevier Science Ltd. All rights reserved.