Work Package 3: The effect of STE on the oxidizing capacity of the troposphere and on chemistry in the tropopause region


 Activity 1: Mixing and its role for chemistry
The study of the meteorological processes leading to mixing (see workpackage 2) will be an essential input for studying the effect of mixing on chemical processes. In this context, it is important to distinguish between mixing and stirring, because chemical reactions to take place requires two volumes of air with different chemical characteristics to come into direct contact. In the presence of non-linear chemical reactions, the average chemical composition of two volumes of air with different chemical composition after some time depends on whether they are mixed before or after chemical reactions occur, i.e. on the rate of mixing. In the process of STE, air masses of contrasting chemical composition may be located particularly close to each other.

After mixing, the combination of high ozone levels from the stratosphere and high water vapor content from the troposphere will increase the production of OH radicals. This, in turn, may lead to a stronger destruction of CO, methane, and higher hydrocarbons. Furthermore, the downflux of nitrogen oxides from the stratosphere adds to ozone production in the troposphere. Hence, STE may have a significant impact on the oxidation capacity of the troposphere.

As a first approach, BOKU will perform photochemical model calculations applying a 1-dimensional model with a few grid boxes in order to simulate the transition zone between air masses of different origin. The outer boxes of the model will be used as reservoirs for tropospheric and stratospheric air, respectively. Mixing between the boxes will be simulated with diffusion coefficients, selected in accordance with results from workpackage 2. In all boxes, an adjusted version of the "Regional Atmospheric Chemistry Mechanism" (RACM) will be integrated.

The stratospheric air transported into the troposphere contains reactive halogenated species. Their effect on the tropospheric chemistry is considered to be of minor importance. However, it might be worthwhile to check this. During stratospheric intrusion episodes, BOKU will try to find signatures of the above qualitative findings in the measurements of chemical species, which are performed at high mountain stations.


Activity 2: The influence of STE on the oxidizing capacity of the troposphere
Global chemistry models suggest that downward transports of relatively ozone-rich stratospheric air contribute significantly to tropospheric ozone levels, down to the surface where STE of ozone contributes to the ozone spring time maximum observed at remote locations in the northern hemisphere. However, STE also implies exchange of air with specific tropospheric and stratospheric signatures other than ozone. A climate change may alter the characteristics of STE, which in turn will affect the chemical composition of the troposphere and the tropopause region both directly through a different exchange rate and indirectly through different chemical reactant concentrations. In order to realistically simulate global atmospheric chemistry, models must have an adequate representation of both the large-scale Brewer-Dobson circulation and the synoptic-scale processes, and must parameterize mixing suitably. These models, with suitable resolution, are just becoming available.

UTRECHT UNIVERSITY will perform state-of-the art simulations with a coupled climate-chemistry model. At present, their model has 19 vertical levels. For STACCATO, they will first enhance the vertical resolution of their model in the tropopause region and test this improvement with the available measurement data. The trace species represented by the chemistry model will be "coloured", meaning that chemical species may be subdivided according to their origin, e.g., stratospheric vs. tropospheric ozone or natural vs. anthropogenic NOx. Next, the contribution of STE of ozone to the tropospheric ozone budget will be studied, with focus on the situation over Europe. Finally, UTRECHT UNIVERSITY will evaluate the impact of STE on the oxidation efficiency of the troposphere by considering the impact on chemistry also of the downward transport of species other than ozone. Due to the strong non-linearities in atmospheric chemistry, for this it is important to have a realistic representation of mixing in the model. Sensitivity analyses will be carried out to assess the contribution of STE related processes to the oxidation efficiency of the troposphere and the tropopause region, to study the annual variability of this contribution, and to investigate the specific role of natural and anthropogenic emissions. Furthermore, the chemical tendencies will be studied in detail along selected trajectories, creating chemical histories of air parcels.

THE MET OFFICE will run their Lagrangian chemistry model for three years at a time starting at 1990, 2030, 2060 and 2100, and will determine the changes in column ozone and ground level ozone concentrations. They will also test the sensitivity of their results to the mixing parameterization.

KNMI will run their global chemistry model based on analyzed meteorological data. One advantage of this model is its higher resolution compared to the others. It will be interesting to compare the different representations of STE in all three global chemistry models.


Activity 3: Air Traffic and STE
Civil aviation traffic has increased dramatically over the past decades and is expected to grow further, as do the aircraft emissions of carbon dioxide, carbon monoxide, hydrocarbons, nitrogen oxides and aerosols.

Photochemical models predict that aircraft emissions of nitrogen oxides cause an increase in tropospheric ozone, which could become important in the future as it scales approximately linearly with the aircraft emissions, while the impact on stratospheric ozone should be small. However, the impact of aircraft emissions on the chemistry of the tropopause regions depends also critically on STE. The residence time of emissions deposited in the stratosphere (order of 50 days) is considerably longer than that of emissions deposited in the troposphere. Presently, approximately 20-30% of the jet fuel is consumed above the tropopause. Also of importance is that species emitted into the stratosphere are largely confined to midlatitudes over typical stratospheric lifetimes and cause significant increases in NOx concentrations along the major flight corridors.

STACCATO will study the relation between aircraft emissions and STE in two ways: TUM will carry out passive tracer simulations. With their particle dispersion model, they will initialize an aircraft emission tracer along the major flight corridors, study the residence times in the stratosphere and troposphere and establish age spectra of the aircraft tracer in the different reservoirs adopting the methodology described in workpackage 2.

UTRECHT UNIVERSITY will implement aircraft emissions into their chemistry-climate model and study their effect relative to that of other anthropogenic and natural sources by "colouring" aircraft effluents in their model. Again, it will be studied where and when the aircraft emissions mix with the surrounding (tropospheric or stratospheric) air, and the relation to STE is of particular interest.




 Scientific Achievements to February 2001

Chemistry transport model simulations with and without consideration of NOx emissions from aircraft have been carried out. For the representation of aircraft emissions in a GCM a new, revised emission database (ANCAT II), in which aviation NOx emissions are significantly lower than the previous estimates, have been implemented. The model results indicate a clear increase in NOx mixing ratios (50-60 pptv) in the upper troposphere at northern mid latitudes due to the aircraft emissions. In relative terms, changes are more pronounced in winter (> 50%) than in summer (~20%) and are restricted to the North Atlantic Flight Corridor (NAFC). Ozone perturbations appear to be limited in the northern hemisphere and are more pronounced during summer (~3ppbv). The ozone production efficiency of NOx emitted in the UT/LS region depends strongly on the concentration of the peroxy radicals HO2 and RO2, which favor ozone production. Acetone is an important source of HO2 in this region. On the other hand, oxidation of higher hydrocarbon species produces RO2. To assess the influence of the latter, the simulations were repeated with and without aircraft emissions, but higher hydrocarbon emissions were replaced by "pseudo-higher hydrocarbon" emissions, in the form of CO. Comparison with the other simulations show that when higher hydrocarbon chemistry is considered, the aircraft-induced ozone perturbations increase by ~10% in the flight corridors of the northern hemisphere during summer. The corresponding increase to the net ozone production rate is ~15%.

Global (chemistry) modelers use aircraft emission data on the basis of annual mean emission fields, but in reality the flight tracks are adjusted according to the actual synoptical situation. Since the impact of aircraft emissions on STE and the residence times of the emissions depend on whether the flights take place in the upper troposphere or lower stratosphere, there might be significant systematic errors if a mean emissions inventory is used instead of the emissions along the actual flight tracks. This problem is being investigated using MOZAIC data in conjunction with a particle dispersion model. A passive aircraft tracer is initialized along the major flight corridors in the northern hemisphere. Two model runs over a period of one year are performed, one with an annual mean emissions field and the other with emissions along the actual single flight tracks. The emissions inventories for both runs are established from MOZAIC aircraft position data. The residence time of the emissions in the stratosphere / troposphere is studied and age spectra of the concentrations are established. For young tracers, the mean concentrations in the stratosphere are higher by a factor of two for the single-run, while maximum concentrations lie further north, than for the mean-run. Obviously, in the single-run a greater amount of the aircraft emissions is released further northward in the stratosphere. This can be explained by the fact that the single flight tracks are shifted according to the meteorological situation such that more flights take place in the stratosphere. Net fluxes across the tropopause of the single-run are greater by about 5 % for the younger tracers, whereas for older tracers the fluxes of the single-run are greater by about 20 %. Hence, using a mean emissions inventory climate modellers might introduce a significant error, as they do not account for the exact location and the exact concentrations of the emissions, which is particulary important for atmospheric chemistry on short time scales.

In co-operation with the MOZAIC-III project, 10-day three-dimensional backwards trajectories for every aircraft position in the period November 1999 to October 2000 have been calculated, based on high-resolution wind field analyses from the ECWMF. MOZAIC ozone climatologies are extended by attributing a measured ozone concentration to the whole path of the corresponding back trajectory. This yields concentration distributions that are not restricted to the actual flight corridors, but have an extended coverage and show from where high or low ozone concentrations are advected towards the flight corridors. The North American and Eurasian continental boundary layers are identified as the major sources of the ozone observed in the upper troposphere in spring and summer, while the African boundary layer appears to be the greatest ozone source in fall and winter. This result underlines the importance of ozone formation in the continental boundary layers and its subsequent export to the upper troposphere for the global ozone budget.


Simulated concentration changes for a) NOx (pptv) and b) Ozone (ppbv) caused by current aircraft emissions at 250 hPa for July.

Ozone concentration fields obtained with trajectory statistics for purely tropospheric data. Shown in each panel are the results for altitudes above 400 hPa (top), between 400 and 700 hPa (middle) and below 700 hPa (bottom). Grid boxes containing less than 100 trajectories are left blank.