1Department of Chemistry, University of Leicester, University Road, Leicester, UK
2National Centre for Atmospheric Science, Department of Chemistry, University of Leicester, University Road, Leicester, UK
3Laboratoire de Météorologie Dynamique, École Polytechnique, 91128 Palaiseau, France
4Laboratoire de Sciences du Climat et de l'Environnement, Gif-sur-Yvette, 91191 France
5Empa, Swiss Federal Laboratories for Materials Testing and Research, Dübendorf, Switzerland
6Institute of Applied Physics, Russian Academy of Sciences, Nizhniy Novgorod, Russia
7Laboratoire Inter-Universitaire de Systèmes Atmosphériques, CNRS, Université Paris-Est and Université Paris 7, Créteil, France
Abstract. National and European legislation over the past 20 years, and the modernisation or removal of industrial sources, have significantly reduced European ozone precursor emissions. This study quantifies observed and modelled European ozone annual and seasonal linear trends from 158 harmonised rural background monitoring stations over a constant time period of a decade (1996–2005). Mean ozone concentrations are investigated, in addition to the ozone 5th percentiles as a measure of the baseline or background conditions, and the 95th percentiles that are representative of the peak concentration levels. This study aims to characterise and quantify surface European ozone concentrations and trends and assess the impact of the changing anthropogenic emission tracers on the observed and modelled trends.
Significant (p < 0.1) positive annual trends in ozone mean, 5th and 95th percentiles are observed at 54 %, 52 % and 45 % of sites respectively (85 sites, 82 sites and 71 sites). Spatially, sites in Central and Northwestern Europe tend to display positive annual ozone trends in mean, 5th and 95th percentiles. Significant negative annual trends in ozone mean 5th and 95th percentiles are observed at 11 %, 12 % and 12 % of sites respectively (18 sites, 19 sites and 19 sites) which tend to be located in the eastern and south-western extremities of Europe. European-averaged annual trends have been calculated from the 158 sites in this study. Overall there is a net positive annual trend in observed ozone mean (0.16 ± 0.02 ppbv yr−1 2σ error)), 5th (0.13 ± 0.02 ppbv yr−1) and 95th (0.16 ± 0.03 ppbv yr−1) percentiles, representative of positive trends in mean, baseline and peak ozone. Assessing the sensitivity of the derived overall trends to the constituent years shows that the European heatwave year of 2003 has significant positive influence and 1998 the converse effect; demonstrating the masking effect of inter-annual variability on decadal based ozone trends.
The European scale 3-D CTM CHIMERE was used to simulate hourly O3 concentrations for the period 1996–2005. Comparisons between the 158 observed ozone trends to those equivalent sites extracted from regional simulations by CHIMERE better match the observed increasing annual ozone (predominantly in Central and Northwestern Europe) for 5th percentiles, than for mean or 95th ozone percentiles. The European-averaged annual ozone trend in CHIMERE 5th percentiles (0.13 ± 0.01 ppbv yr−1) matches the corresponding observed trend extremely well, but displays a negative trend for the 95th percentile (−0.03 ± 0.02 ppbv yr−1) where a positive ozone trend is observed. Inspection of the EU-averaged monthly means of ozone shows that the CHIMERE model is overestimating the summer month O3 levels.
In comparison to trends in EMEP emissions inventories, with the exception of Austria-Hungary, we find anthropogenic NOx and VOC reductions do not appear to have a substantial effect on observed annual mean O3 trends in the rest of Europe.