Chemistry transport models play a crucial role in the evaluation of the effect of anthropogenic emissions on the atmosphere and climate, but they come with high computational costs and require specialized know-how. This renders them impractical for applications in multidisciplinary optimisation, or regulatory and operational decision-making processes where environmental effects are to be considered. Such applications require computationally efficient surrogate models of the complex chemistry transport models. Here we investigate the use of data-driven discovery and reduced-order modelling methods for this purpose. Specifically, we examine the dynamic mode decomposition (DMD) and proper orthogonal decomposition coupled with the sparse identification of non-linear dynamics (POD-SINDy). We evaluate their ability to reconstruct and forecast changes in the distribution of ozone in response to the introduction of supersonic aircraft as modelled by the GEOS-Chem chemistry transport model. Of the tested methods, we find that optimized DMD and bagging optimized DMD with constrained eigenvalues perform best. These methods can reconstruct and forecast full-atmospheric ozone responses for up to several years without losing stability, at smaller errors than estimates using the spatio-temporal mean of the data. On average, the constrained optimized DMD method reduces the reconstruction error by 63.5 % and that of forecasting by 25.8 % compared to the spatio-temporal mean. For the constrained bagging optimized DMD these reductions are 45.0 % and 23.1 %, respectively. The resulting change in global ozone column, calculated from the reconstructed atmospheres, has an error smaller than 10 %. This is achieved while reducing the computational and storage requirements by several orders of magnitude, which may be a worthwhile tradeoff for some applications.

Civil supersonic aviation may return in the near future. Their emissions have been found to lead to changes in the composition of the stratosphere, affecting the ozone layer and climate. To keep up with the rapid developments in supersonic aircraft technology and alternative fuels there is an increasing need for the development of surrogate modeling methods, which requires knowledge of the sensitivities to these emissions. We present a parametric study which evaluates the first- and second-order sensitivities of the ozone column and radiative forcing (RF) to supersonic emissions across two flight corridors and three altitudes. For a given increase in global fuel burn, we find that the increase in emission of (Formula presented.) is the main driver of both the changes in the global ozone column and RF, the latter of which is linked through changes in the ozone distribution. Followed by the increase in the emission of (Formula presented.), which leads to (Formula presented.) loss and has a cooling effect. The ozone column and climate are least sensitive to increases in (Formula presented.) emissions. We also show that interactions between (Formula presented.), (Formula presented.), and (Formula presented.) emissions lead to non-linear behavior in the atmospheric response. The effect of these interactions can lead to (Formula presented.) 5% differences in the ozone column impacts and up to 7.3% increases in RF. Our results demonstrate that the majority of second-order sensitivities may be neglected in surrogate models for small errors, which could greatly simplify their development. Our results also indicate that reductions in flight altitude and fleetwide (Formula presented.) emissions may effectively reduce the environmental footprint of supersonic aviation emissions.