Accounting for local meteorological effects in the ozone time-series of Lovozero (Kola Peninsula) O. A. Tarasova1 and A. Yu. Karpetchko2,* 1Atmosphere Physics Department, Faculty of Physics, Moscow State University, Moscow, Russia 2Polar Geophysical Institute, Apatity, Russia *Present address Finnish Meteorological Institute, Sodankyla, Finland
Abstract. The relationship between local meteorological conditions and the surface ozone variability was
studied by means of statistical modeling, using ozone and meteorological parameters measured
at Lovozero (250 m a.s.l., 68.5°N, 35.0°E, Kola Peninsula) for the period of
1999-2000. The regression model of daily mean ozone concentrations on such meteorological parameters as
temperature, relative humidity and wind speed explains up to 70% of day-to-day ozone
variability in terms of meteorological condition changes, if the seasonal cycle is also considered.
A regression model was created for separated time scales of the variables. Short-term, synoptical
and seasonal components are separated by means of Kolmogorov-Zurbenko filtering. The
synoptical scale variations were chosen as the most informative from the point of their mutual
relation with meteorological parameters. Almost 40% of surface ozone variations in time periods
of 11-60 days can be explained by the regression model on separated scales that is
30% more efficient than ozone residuals usage. Quantitative and qualitative estimations of the relations
between surface ozone and meteorological predictors let us preliminarily conclude that at the
Lovozero site surface ozone variability is governed mainly by dynamical processes of various
time scale rather than photochemistry, especially during the cold season.
Citation: Tarasova, O. A. and Karpetchko, A. Yu.: Accounting for local meteorological effects in the ozone time-series of Lovozero (Kola Peninsula), Atmos. Chem. Phys., 3, 941-949, doi:10.5194/acp-3-941-2003, 2003.