Kawa et al. [1993] showed that agreement between
measured and modeled NO could be improved if the detailed histories of
individual airmasses were taken into account. However, important discrepancies
still remained. It is unclear whether these differences are due to an incomplete
understanding of the reactivity of N
O
on SSAs, to an incomplete
description of NO and NO
exchange, or to some missing gas-phase or
heterogeneous process. Laboratory studies by Fried et al.
[1994] found that the reaction probablility of N
O
on SSAs
varies with aerosol composition, which is dependent on stratospheric
temperature and humidity, whereas Tolbert et al. [1993] showed
that CH
O is efficiently taken up by SSAs. Although it remains to be seen if
these studies will have an significant impact on our understanding of
stratospheric chemistry, at the very least they point out that our
understanding of heterogeneous chemistry can be improved.
Upper stratospheric odd-oxygen (O + O
) balance remains problematic
[ Eluszkiewicz and Allen, 1993]; models routinely underpredict
ozone abundances above 40 km, which implies that modelled production
rates are too slow, or loss rates are too fast compared to the real world. Allen
and Delitsky [1991] proposed that uncertainties in the
parameterization of O
photolysis in the Schumann-Runge bands (175 to 205
nm) might contribute to part of the discrepancies, but Siskind et al.
[1994], using models by Minschwaner and coworkers [1992,1993]
show that differences in resolution can not account for the ozone
balance problems. Eluszkiewicz and Allen [1993] suggest some
possibilities, including systematic errors in adopted rate constants. They
offer an excellent review of this issue.