Transport in the stratosphere involves meridional overturning and isentropic mixing (mixing along surfaces of constant potential temperature). These components together represent the Brewer-Dobson circulation, proposed by Brewer in 1949 and Dobson in 1956 to explain why the lower stratosphere is extremely dehydrated and why tropical air has less ozone than polar air, even though the tropical stratosphere is where most atmospheric ozone is produced. The overall pattern of the Brewer-Dobson circulation, characterized by upward mass transport in the tropics and poleward-downward mass transport in the extra-tropics (overturning) as well as by horizontal mixing along isentropic surfaces, is reflected in long-lived constituent distributions. The mixing ratio isopleths of all such tracers, e.g., nitrous oxide (N₂O), methane (CCH₄), sulfur hexafluoride (SF6), CFC-11, etc., bulge upward in the tropics and slope downward in the subtropics and at the edge of the winter hemisphere polar vortex. The mixing ratio isopleths are quite flat in the mid-latitudes of both hemispheres. This was evident from some of the earliest global observations of CH₄ and N₂O from the Nimbus-6 satellite and has been confirmed in many subsequent aircraft and especially satellite observations and modeling studies of many such tracers. The distribution of long-lived stratospheric tracers just results from the competition between vertical advection by diabatic (overturning) circulation acting to steepen isopleths, and isentropic mixing acting to flatten isopleths in the stirring region, but producing very strong gradients at the equatorward and poleward edges of the stirring region.

The aim of the project DIMITRI is a better understanding of transport and mixing processes in thestratosphere, especially in the regions of strongest meridional tracer gradients, as inferred from longlived tracer observations from satellite. The particular focus of this project is the analysis of tracer gradients in the subtropics, at the equatorward edge of the mid-latitude “surf-zone”. This edge, or subtropical barrier, defines a clear contrast between tropical and extra-tropical air, both chemically and dynamically. The position of the subtropical barrier can be diagnosed simply looking at the passive (e.g., long-lived) tracer fields and their meridional gradients. In the lower stratosphere, for instance, the latitude-time cross section of the passive tracer mixing ratio concentrations is characterized by a back and forth shifting of the tropical maximum region in passive tracer (N₂O, CH₄) concentrations which follows a clear annual periodicity. A periodicity of approximately two-years is superimposed to the annual periodicity as an effect of the Quasi-Biennial Oscillation (QBO) in the zonal mean zonal winds. At higher levels, the atmosphere shows other modes of variability that can affect tracer distributions, such as the semi-annual oscillation, SAO). The stratospheric subtropical barrier in both hemispheres is collocated with the narrow region of maximum meridional tracer gradient change and its intensity and position is affected by the processes that contribute to the variability of atmospheric transport (such as the QBO in the lower stratosphere, or the SAO above).

Within DIMITRI, N₂O and CH₄ observations from four different satellite-borne sensors, namely, the Halogen Occultation Experiment (HALOE) on board UARS, the Microwave Limb Sounder (MLS) on board Aura, the Michelson Interferometer for Passive Atmospheric Sounding (MIPAS) on board Envisat, and the Sub-Millimeter Radiometer (SMR) on board Odin, are used to diagnose the position of the subtropical barrier and its variability. An alternative approach to the analysis of tracer meridional gradients which allows to determine the position of dynamical barriers in the stratosphere was proposed by Sparling in 2000, and consists in calculating the probability distribution functions (PDFs) of tracer concentrations. This approach, used within DIMITRI, allows to quantify stratospheric mixing regions and mixing barriers, the former being associated with maxima in tracer-concentration PDFs, the latter with minima. The winter hemisphere PDF has three peaks associated to the polar vortex, the tropics, and mid-latitude surf-zone and two valleys defining the vortex edge and the subtropical boundary regions. The summer hemisphere PDF has two peaks corresponding to the tropics and the summer extratropics with a broader minimum between them corresponding to the subtropical barrier. An important and consistent part of the activity performed within DIMITRI has included the analysis of the consistency among the different sets of data and of their capability to identify mixing and barrier to transport regions in the stratosphere, particularly the subtropical barrier. A careful intercomparison among the one-dimensional, two-dimensional and conditional tracer PDFs morphology from the four sensor data has been made, to find to which extent transport characteristics are reproduced by sensors with very different instrumental and sampling characteristics. The position of the subtropical barrier is calculated on a seasonal basis, using the statistics of the support region applied to the tracer PDFs from the four data sets. Taking advantage of the long time series available through the combination of the different sets of satellite data, the time evolution of the subtropical barrier position from 1992 to 2009 is analyzed. This is done at different potential temperature surfaces and the role of the QBO as the most important factor contributing to the year-to-year variability of lower stratospheric transport is identified. Forthcoming DIMITRI activities will include the possibility of recognizing trend features in the long time series of the subtropical barrier position and the analysis of transport processes across dynamical barriers in coupled Chemistry-Climate model (CCM) data, also aimed at performing a process-oriented validation of CCMs.