1NCAS-Climate, Centre for Atmospheric Science, University of Cambridge, Cambridge, UK
2Laboratoire de Physique et Chimie de l'Environnement et de l'Espace, CNRS and University of Orléans, Orléans, France
3Department of Geosciences, University of Oslo, Norway
4Institute for Atmospheric and Climate Science, ETH Zurich, Zurich, Switzerland
5Centre National de Recherches Météorologique/Groupe d'étude de l'Atmosphère Météorologique, Météo-France and CNRS, Toulouse, France
6NCAS-Weather, Centre for Atmospheric & Instrumentation Research, University of Hertfordshire, Hatfield, UK
7Institute for Climate and Atmospheric Science, School of Earth and Environment, University of Leeds, UK
8Centre for Atmospheric Science, University of Cambridge, Cambridge, UK
9Lancaster Environment Centre, Lancaster University, UK
*now at: British Antarctic Survey, Cambridge, UK
Abstract. Tropical convection is a very important atmospheric process acting on the water cycle, radiative budget of the atmosphere and air composition of the upper troposphere and lower stratosphere (UTLS), and it affects a broad range of spatial and temporal scales. The fast vertical transport in convective plumes can efficiently redistribute water vapour and pollutants up to the Tropical Tropopause Layer (TTL), and therefore affect the composition of the lower stratosphere. Chemistry Climate Models and Chemistry Transport Models are routinely used to study chemical processes in the atmosphere. In these models convection and convective transport of tracers are parameterised, and due to the interplay of chemical and dynamical processes, it has proven difficult to evaluate the convective transport of chemical species by comparison with observed chemical fields.
In this work we investigate different characteristics of tropical convection by using convective proxies from many independent observational datasets (including surface precipitation rates, cloud top pressure and OLR). We use observations to analyse the seasonal cycle and geographical preferences of convection, and its impact on water vapour. Using highly temporally resolved cloud top data we calculate the frequency distribution of high clouds in three tropical regions. The observational data is used as a benchmark for a number of numerical models, with a view to assess the ability of models to reproduce the seasonality, preferential location and vertical extent of tropical convection. Finally we discuss the implications of our findings on modelling the composition of the upper troposphere and lower stratosphere.