The chemical and dynamical processes in the tropical tropopause layer (TTL) control the amount of radiatively active species like water vapour and ozone in the stratosphere, and hence turn out to be crucial for atmospheric trends and climate change. Chemistry transport models and chemistry climate models are suitable tools to understand these processes. But model results are subject to uncertainties arising from the parametrization of model physics. In this thesis the sensitivity of model predictions to the choice of the vertical transport representation will be analysed.
Therefore, backtrajectories are calculated in the TTL, based on different diabatic and kinematic transport representations using ERA-Interim and operational ECMWF data. For diabatic transport on potential temperature levels, the vertical velocity is deduced from the ERA-Interim diabatic heat budget. For kinematic transport on pressure levels, the vertical wind is used as vertical velocity. It is found that all terms in the diabatic heat budget are necessary to cause transport from the troposphere to the stratosphere. In particular, clear-sky heating rates alone miss very important processes. Many characteristics of transport in the TTL turn out to depend very sensitively on the choice of the vertical transport representation. Timescales for tropical troposphere-tostratosphere transport vary between one and three months, with respect to the chosen representation. Moreover, for diabatic transport ascent is found throughout the upper TTL, whereas for kinematic transport regions of mean subsidence occur, particularly above the maritime continent. To investigate the sensitivity of simulated trace gas distributions in the TTL to the transport representation, a conceptual approach is presented to predict water vapour and ozone concentrations from backtrajectories, based on instantaneous freeze-drying and photochemical ozone production. It turns out that ozone predictions and vertical dispersion of the trajectories are highly correlated, rendering ozone an interesting tracer for aspects of transport in the TTL where water vapour is not sensitive. Consequently, dispersion and mean upwelling have similar effects on ozone profiles, with slower upwelling and larger dispersion both leading to higher ozone concentrations. Analyses of tropical upwelling based on mean transport characteristics (e.g., mean ascent rates) and model validation have to take into account this ambiguity. Predicted ozone concentrations for kinematic transport are robustly higher than for diabatic transport, due to larger trajectory dispersion caused by the larger inhomogeneity in the kinematic vertical velocity field. During the tropical SCOUT-O3 campaign, kinematic ozone predictions show an extreme high bias compared to in-situ observations.
The high sensitivity of many characteristics of transport to the choice of the transport representation, demonstrates the need to better constrain transport in the TTL. Consequently, estimates of exact numbers from models, e.g., for timescales of transport, are not reliable and only a range of values can be given. However, there are robust features of tropical transport, not depending on the transport representation, as for example, a signifcant impact of monsoon driven horizontal in-mixing from the extratropics on the composition of the TTL. In fact, the annual cycle of ozone above the tropical tropopause is attributed, at least in ‘ECMWF-world’, to in-mixing of ozone-rich extratropical air during summer.