A chemical amplifier based radical detector was used to make measureme
nts of radical (RO(x) = HO2 + HO + RO + RO(2)) concentrations at a rur
al site in the summer of 1992, as part of the Southern Ontario Oxidant
Study (SONTOS). The average maximum daytime radical concentration was
around 7 pptv, with the maximum value recorded being 23 pptv. There w
as no distinguishable radical signal on any of the cloudy days, wherea
s there was a clear diurnal variation, maximizing in the mid afternoon
on sunny days. The highest radical signals were observed on days of t
he highest ozone, which were days that showed photochemical ozone prod
uction. The local ozone production was calculated From the radical and
NO concentrations but, even in the absence of loss terms, was insuffi
cient to explain the ozone increase at the site. This indicates that p
art of the measured ozone increase is due to ozone production elsewher
e followed by transport to the site. Peroxyacetyl nitrate (PAN) decomp
osition appears to provide about 30% of the radicals measured at the s
ite so that the contribution of PAN as a radical reservoir needs to be
considered. Fluctuations in the measured parameters prevented the mea
ningful determination of radical concentrations from the photostationa
ry state at the low levels encountered. Copyright (C) 1996 Elsevier Sc
ience Ltd