Use of Computed Trajectories in Aerobiology and Air Pollution Studies

  • Daniel Schneiter
  • Regula Gehrig
  • Dominique Ruffieux


Trajectories can be considered as a powerful tool to study transport of aerosols and tracer gases into the atmosphere. The TRAJEK Model (Fay et al. 1994) developed by the DWD (Deutscher Wetterdienst) is used twice daily by MeteoSwiss to compute hundreds of trajectories. Forward trajectories can be used to forecast the transport of nuclear or chemical pollutants in case of civil or military accidents. They are computed at various levels from locations corresponding to existing nuclear installations. Forecasters can use these results any time to predict direction and speed of nuclei transport in case of accidental releases.

Backward trajectories determine the origin of air masses passing over Switzerland at different levels. They help to understand and to better forecast the weather from the point of view of air masses origin and transport. A set of pollen measurement stations is also taken as arrival points. In these cases, backward trajectories indicate the various regions, where the measured pollen was probably emitted. When they are observed at places where they can not be emitted, because of absence of these plant species in the surroundings, pollens can be considered as medium or long range tracers. For instance Castanea sativa (chestnut-tree) pollen emitted in the Ticino and in North Italy can be detected, by southern wind, in the Engadin (Peeters and Zoller 1998), at Giitsch (2282 m, near the St. Gotthard Pass) (Frei 1997) and on the northerly side of the Alps (Vuillemin et al. 2001). They attest for low level currents crossing the alpine barrier. A peak of Castanea pollen concentration originated from South East of France has been measured at Geneva and two hours later at Lausanne (at a distance of 60-km), after overflying the Lake of Geneva (Gehrig and Schneiter in prep.). Most suitable stations for detecting transported pollen are elevated sites like Giitsch (2282 m) or La Chaux-de-Fonds (1018 m), where no chestnut trees are present and therefore no local pollen influences the measurement. Due to the building of mixing height after sunrise, the peak of pollen concentration is normally measured during the day, with sometimes two peaks, one forenoon and one afternoon (Jäger, 1990). Therefore high concentrations of pollen appearing suddenly during the night are due to medium or long range transport. Castanea pollen associated at times with dust coming from the Sahara, proves the southern origin of these air masses.

Backward trajectories have been operationally computed for three years (1998 to 2000), in relation with ozone concentration measurements at the Jungfraujoch (46°39’N / 07°59’E 3580-m). They can partially explain 03 concentration variations due to the origin and history of air masses. Analytical studies (Forrer et al. 1999, Fricke et al. 2000) can extract trajectories, which have passed at least 4 hours during the last 48 hours below the 850 hPa level (about 1500-m asl). This criterion can be considered as a filter, allowing the distinction of weather situations with air masses strongly influenced by pollutant emitted into the boundary layer (for instance over the Pô valley) from relatively pure air, remaining above the 850-hPa level all the time over the last two days.

During the Mesoscale Alpine Programme (MAP), trajectories observed by constant volume balloons (CVB) were compared with computed trajectories. Strong vertical wind components occurring sometimes on trajectories have also been observed by a wind profiler located next to the Julier Pass at 2233-m asl (Ruffieux and Stübi., 2001). The rather rough resolution (14-km) of the Swiss Model (developed in collaboration with the DWD) cannot take into account the fine influence of local complex topography. We observe in some weather conditions with southern wind, that strong vertical motions can last for hours. They are probably connected with orography induced waves.

The recent exploitation, since January 2001, by MeteoSwiss of the new LM (Local Model developed in collaboration with the DWD) produces trajectories based on windfield calculated with a resolution of 7-km. The better representation of the topography will allow the computing of regional wind characteristics due to the Jura Mountains and large alpine valleys like the Rhone or the Rhine Valleys.


Wind Profiler Jura Mountain Range Transport Chestnut Tree Rhine Valley 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Fay B., H. Glaab, I. Jacobsen, R. Schrodin, 1994. Radioactive Dispersion Modelling and Emergency Response System at the German Weather Service. Air Pollution Modeling and its Application X. Editied by S.-V. Gryning and M.M. Millán. Plenum Press, New York. pp. 395–403.CrossRefGoogle Scholar
  2. Forrer J., P. Hofer, D. Schneiter, 1999. Variabilität von Spurengasen an der Hochalpinen Station Jungfraujoch. Oesterreichische Beiträge zu Meteorologie und Geophysik 21, 125–134.Google Scholar
  3. Frei T. 1997. Pollen distribution at high elevation in Switzerland: Evidence for medium range transport. Grana 36, 34–38.CrossRefGoogle Scholar
  4. Fricke W., A. Fischer, J. Forrer, S. Gilge, P. Hofer, P. Jeannet, A. Kaiser, K. Kenndoff, R. Nemeth, L. Ries und P. Winkler, 2000. Berichte des Deutschen Wetterdienst 211. Filterung luftchemischer Messreihen im Alpenraum zur Charakterisierung ihrer Repräsentanz. GAW — DACH — Projekt.Google Scholar
  5. Gehrig R. and Schneiter D., 2001. Castanea pollen on the Swiss Plateau: use of trajectories for demonstrating medium and long range transport. Aerobiologia, in prep.Google Scholar
  6. Jäger S., 1990. Tageszeitliche Verteilung und langjährige Trends bei allergiekompetenten Pollen. Allergologie, Jahrgang 13, Nr 5, S. 159–182.Google Scholar
  7. Peeters A. and H. Zoller, 1988. Long range transport of Castanea sativa pollen. Grana 27:203–207.CrossRefGoogle Scholar
  8. Ruffieux D., R. Stübi 2001. Wind profiler as a tool to check the ability of NWP models to forecast winds above highly complex topography. Meteorologische Zeitschrift (accepted).Google Scholar
  9. Schneiter D. 1998. Trajectory and Dispersion Models Applied to Emergency Cases and to Ozone Research Studies. In Air Pollution Modeling and its Application XII. Plenum Press. New York p. 709.Google Scholar
  10. Vuillemin F., A.G. Peeters, R. Gehrig, B. Clot, Th. Herren, M. Udriet, M. Hauser and C. Hess, 2001. Luftpollengehalt in der Schweiz 2000. Herausgegeben von MeteoSchweiz, Zürich.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Daniel Schneiter
    • 1
  • Regula Gehrig
    • 2
  • Dominique Ruffieux
    • 1
  1. 1.MeteoSwissPayerneSwitzerland
  2. 2.MeteoSwissZurichSwitzerland

Personalised recommendations