The atmospheric response to perturbations in NO x emissions from global air traffic is investigated by performing a coherent set of sensitivity experiments. The importance of cruise altitude, size of the emission perturbation and geographical distribution of emissions is systematically analyzed using two global chemistry transport models and an off-line radiative transfer model. NO x emissions from a contemporary aircraft inventory have been used to assess the impact of global air traffic on ozone and methane. In further experiments the NO x emissions are perturbed, in turn, in 16 cruise altitude bands between 5 and 15 km altitude. In the p-TOMCAT model we diagnose an annual mean ozone increase of up to 6 ppbv and a decrease in the methane lifetime of 3% due to global air traffic in 2002. Associated radiative forcings of 30 mWm−2 for ozone and −19 mWm−2 for methane are diagnosed; a simple method is used to estimate the forcing due to the methane-induced ozone change and this yields an additional −11 mWm−2. Results show that up to the tropopause, ozone production efficiency and resulting impacts increase per emitted mass of NO x with the altitude of the perturbation. Between 11 and 15 km we find that the geographical location of the NO x emissions plays a crucial role in the potential O3 impact and lifetime change of CH4. We show that changes in flight routing in this altitude range can have significant consequences for O3 and CH4 concentrations. Overall, we demonstrate a linear relationship in the atmospheric response to small emission changes which can be used to predict the importance of perturbations about the reference aircraft emissions profile, provided the geographical distribution of the emissions is not altered significantly.