Measurement results
Comparison to EU values

The measurements were performed as campaigns, with three as "full" campaigns and nine as "reduced" campaigns. The number of samples in each of the classes in the two different types of campaigns are presented in table below, which is valid for both NO₂ and SO₂. VOC was measured at 9 or 10 sites during the three "full" campaigns and during one of the "reduced" campaigns.

Number of sampling sites in each of the classified environments during the campaigns.

Classes	Number of sampling sites, "full" campaigns	Number of sampling sites, "reduced" campaigns
A B C D	5 30-34 24-29 15-18	1-2 8-10 3-9 3-6

Background measurements of NO₂ and SO₂ were made in the outskirts of Tallinn at two or four different geographical directions during most of the campaigns.

Measurement results

The results from the different classified environments and from background measurements are presented as a total for all 12 campaigns in table below, and for each campaign event in Appendix. It is evident from the differences in the results that the selection of sites, mainly regarding the traffic situation, has been successful. As was expected, there is a falling average concentration in the order A, B, C, D and background sites. The minimum monthly mean in the A sites, 27 µg/m³, is way higher than in the other local environments. The difference between the background measurements and the "local background", represented by the D sites is obvious, showing a clear influence also on D-sites from urban traffic.

Results from all campaigns, monthly averages of NO₂, µg/m³.

Class	Max	Average	Min	Number of results
A B C D Background	48 50 33 30 16	36 29 18 14 8	27 10 6 5 1	32 173 136 95 26

In the figures below the levels of NO₂ are presented as the average, maximum and minimum value for each month and for the four classified environments respectively. The darker bars represent the "full" measurement campaigns.




Residential areas and areas of "local background" (C and D sites) generally exhibited lower exposure levels of NO₂ than more traffic exposed areas (A and B). This is applicable for all measurement periods.

The concentration levels were quite uniform throughout the whole measurement period, with mean values between 30 to 40 mg/m³ at the most exposed (A) sites down to 10-15 mg/m³ at the class D sites. A seasonality is however visible at the C and D sites with generally somewhat lower levels measured during summer months than in wintertime. The lowest mean values were measured in July 1996, at all types of sites. For the more traffic exposed A and B sites the highest mean values of NO₂ were measured in April 1996 and in August of 1997 respectively. These results do not follow the expected variations in pollution levels, which are of higher levels during wintertime than in summer. At the less traffic exposed C and D sites, the highest average values observed were, as for the A sites, in April 1996 but also in January 1997. As discussed further, the highest SO₂ levels were also measured in April 1996. In July of 1996, when the lowest mean values were observed, the predominant wind direction was from the west, which was the only campaign with this dominant wind direction. The occasions with high average levels of NO₂ were dominated by northerly, south westerly or north-easterly winds. Besides the dominant wind direction, also the average temperature and wind speed influences the levels of pollutants in air. The average calculated results might also be influenced by the number of sampling sites, as the "reduced" campaigns only used approximately 20% of the total number of stations (see table ).

The geographical variations in concentration levels are illustrated in figures below for the three "full" measurement campaigns, for AB and CD sites respectively. All monitoring sites are marked as black triangles. The differences in concentration load between the exposure classes are clearly shown, as well as the difference between the periods studied. The central parts of the town, where the traffic load can be expected to be high, exhibited the highest monthly means of NO₂.

Comparison to EU limit values

The EU limit values are not directly applicable to the measurements performed, but the average values calculated from all campaigns can be compared to the EU annual limit values (see Appendix). For NO₂ the annual limit value for protection of human health is 40 µg/m³ and the upper and lower assessment thresholds are 32 and 26 µg/m³ respectively. The limit value was not exceeded in Tallinn according to the measurements, but the upper assessment threshold was exceeded at the A points (36 µg/m³ as an average), and the lower assessment threshold also at the B points (29 µg/m³) (table). The average NO₂-levels at C, D and background measurement sites were all below the lower assessment threshold for protection of human health.