Tropical deep convection and its impact on composition in global and mesoscale models – Part 2: Tracer transport

The tropical transport processes of 14 different models or model versions were compared, within the framework of the SCOUT-O3 (Stratospheric-Climate Links with Emphasis on the Upper Troposphere and Lower Stratosphere) project. The tested models range from the regional to the global scale, and include numerical weather prediction (NWP), chemistry transport, and climate chemistry models. Idealised tracers were used in order to prevent the model's chemistry schemes from influencing the results substantially, so that the effects of modelled transport could be isolated. We find large differences in the vertical transport of very short lived tracers (with a lifetime of 6 hours) within the tropical troposphere. Peak convective outflow altitudes range from around 300 hPa to almost 100 hPa among the different models, and the upper tropospheric tracer mixing ratios differ by up to an order of magnitude. The timing of convective events is found to differ between the models, even among those which source their forcing data from the same NWP model (ECMWF). The differences are less pronounced for longer lived tracers, however they could have implications for the modelling of the halogen burden of the lowermost stratosphere through species such as bromoform, or for the transport of short lived hydrocarbons into the lowermost stratosphere. The modelled tracer profiles are found to be strongly influenced by the convective transport parameterisations, and boundary layer mixing parameterisations of the models. The location of rapid transport into the upper troposphere is similar among the models, and is mostly concentrated over the western Pacific, the Maritime Continent and the Indian Ocean. In contrast, none of the models indicates significant enhancement in upward transport over western Africa. The mean mixing ratios of an idealised CO like tracer in the upper tropical troposphere are found to be sensitive to the surface CO mixing ratios in the regions with the most active convection, revealing the importance of correctly modelling both the location of convective transport and the geographical pollutant emission patterns.