MODELED AND MEASURED TRANSBOUNDARY ACIDIFYING POLLUTION IN EUROPE - VERIFICATION AND TRENDS

The routine acid deposition model at MSC-W EMEP has been run for 5 years. Routine measurements taken by countries participating in EMEP are so far available from CCC EMEP for four of these years. This paper aims at demonstrating the usefulness of having access to simultaneous sets of data, such as measurements and model-calculated quantities for which one can vary the determining parameters. By comparing the data sets with emissions used in the calculations, one can identify both important model shortcomings as well as probable errors in measurements and emissions. Furthermore, trends in annual deposition and concentration levels can be investigated by controlling the meteorological input in order to filter out the effect of meteorological variability. In general, errors in modelled concentrations can only be identified with confidence if a statistically sufficient number of stations agree. This is the case for NO2 in air for which the model seems to underestimate systematically the measured concentrations, and for sulphate and nitrate in precipitation for which the model gives persistently lower concentrations than measured over peripheral areas in Europe. A majority of the ''outliers'' in the scatter plots comparing measurements and calculations are seen to be connected with clear misfits between emission data and measured values. Furthermore, at a few sites measurements show much larger variations from year to year than anticipated. The trend experiments emphasize the importance of meteorological variability as compared to emission changes. Even if the impacts of emission trends can be larger than those of meteorological variations at some sites, the complex spatial pattern caused by the latter completely masks the former. Measurements, being influenced by changes in both conditions, are thus of limited value for a year-to-year monitoring of emission reduction effects. Keeping the meteorology the same from year to year in the model calculations, reveals that calculated air concentrations and depositions correspond to the reported emissions. Trends in measurements and calculations generally agree well, except that for sulphur in air in 1989 measured concentrations are smaller than calculated. This is believed to be a consequence of too high emission numbers for SO2 in 1989. The conclusions that can be drawn from this study are direct consequences of the distribution of measurement sites in Europe as part of EMEP. This distribution favours central European areas north of the Alps. The statistical confidence and representativity of the results would be much improved by a more evenly distributed measurement network.