SPRINGTIME ozone depletion over Antarctica is thought1,2 to be due to
catalytic cycles involving chlorine monoxide, which is formed as a res
ult of reactions on the surface of polar stratospheric clouds (PSCs).
When the PSCs evaporate, ClO in the polar air can react with NO2 to fo
rm the reservoir species ClONO2. High concentrations of ClONO2 can als
o be found at lower latitudes because of direct transport of polar air
or mixing of ClO and NO2 at the edges of the polar vortex. ClONO2 can
take part in an ozone-depleting catalytic cycle18, but the significan
ce of this cycle has not been clear. Here we present model simulations
of ozone concentrations from March to May both within the Arctic vort
ex and at a mid-latitude Northern Hemisphere site. We find increasing
ozone loss from March to May. The ClONO2 cycle seems to be responsible
for a significant proportion of the simulated ozone loss. An importan
t aspect of this cycle is that it is not as limited as the other chlor
ine cycles to the timing and location of PSCs; it may therefore play a
n important role in ozone depletion at warm middle latitudes.