Future Air Quality in Danish cities

7. Comparison with EU Predictions

7.1 Comparison with EU emission predictions
7.2 Comparison with EU air quality predictions

In this chapter, a comparison between the results of the present study and work carried out by the EU Commission is undertaken for emissions and air quality.

The EU Commission has carried out an impact assessment of the Auto-oil II programme on the total vehicle emissions. The task has been undertaken by Senco - Sustainable Environment Consultants - based on the TREMOVE model (AOPII Emissions Base Case). For further details see Senco (1999). The emission factors of the present study are derived for the different vehicle categories (g/km) based on COPERT 3 and Danish traffic characteristics. To be able to compare these emission factors with the results of the EU Commission, Danish national emissions were calculated by Risoe National Laboratory, Denmark and NERI (Fenhann 2000). The comparison is shown in Table 7.1.

Table 7.1
Comparison of Total Vehicle Emission Predictions

The comparison shows that the predicted total emissions are more or less similar. The observed differences may be due to differences in approach and assumptions about emission factors and traffic development.

Comparison of the results of the present study with EU predictions and another Danish study has been carried out. The comparison includes urban background and street predictions.

EU and Danish Studies

Urban Impact Assessment Study, AOPII

The EU Commission has undertaken the Urban Impact Assessment study including ten larger European cities to estimate the impact of the Auto-oil II proposals for vehicle emission reductions on urban air quality 1995-2010 (EU Commission 1999). The urban background air quality has been modelled by different urban scale models: CALGRID, Urban Airshed Model (UAM) v. IV, and the European Urban Airshed Model (EUAM). The regional background is the boundary conditions for the urban models. Data on the regional background has been derived from work carried out by IIASA with the RAINS model as part of the ongoing EU strategy to reduce trans-boundary air pollution.

As part of the project, street canyon modelling has been carried out in two of the cities: Milan and Berlin using the MICROCALGRID and the Danish OSPM model. The results for OSPM calculations are given in Berkowicz (1999a). Copenhagen is not included in the case studies.

Impact Assessment at Seven Streets in Copenhagen

A minor Danish study has carried out an impact assessment of the future air quality related to the new EU limit values in 2010 for seven streets in Copenhagen with reference year 1997 . This study did not model future vehicle emissions nor did it model the urban background. Crude assumptions were applied for future vehicle emission factors and the urban background (Berkowicz & Palmgren 1999).

Air Quality in the Urban Background

In Table 7.2, findings of the present study are compared with results from the urban background predictions in ten European cities based on preliminary results from the Urban Impact Assessment Study.

Table 7.2
Urban Background Predictions for NO₂ and Benzene in Ten European Cities (EU Commission 1999)

Comparison with Ten European Cities

For the ten European cities, the results are preliminary based on EU Commission (1999). For each city, the modelled concentrations reflect the highest modelled levels in a grid cell of all the grid cells considered. It is seen that the annual NO₂ urban background levels in Copenhagen are the lowest compared to the other European cities. There is a tendency to high NO₂ levels in Southern Europe cities compared to Northern European cities due to high ozone levels in Southern Europe. Modelled annual benzene levels in the urban background are also relatively low in Copenhagen in 1995 and the lowest in 2010.

Comparison with Milan, Berlin and Danish Study

The urban background was also modelled as part of the project on street canyon modelling in Milan and Berlin (Berkowicz 1999a), and as part of a Danish study (Berkowicz & Palmgren 1999). Results of the present study are compared with these studies in Table 7.3.

Table 7.3
Comparison of Annual Levels of Urban Background Predictions

¹ based on Berkowicz (1999a), ² based on Berkowicz & Palmgren (1999)

Berlin and Milan

It is seen that the results of the present study are similar to findings for Berlin and that much higher levels are modelled for Milan except for ozone due to depletion of ozone by the high NO_x levels in Milan. All pollutions decreases except ozone from 1995 to 2010.

Danish Study

It is surprising that predicted urban background levels in 2010 for the Danish study are very close to the predictions of the present study because crude assumptions were applied in Berkowicz & Palmgren (1999). The study assumed that urban background levels decreased 70% for NO_x, CO and benzene, and that regional background levels of NO₂ and ozone decreased by 50%. The NO_x emissions from trucks were assumed to decrease by 50% and 90% of passenger cars were assumed to have catalyst converters in 2010. The present study modelled a reduction in the urban background of NO_x, CO and benzene of 61%, 43% and 82%, respectively (see Table 5.6). The regional NO₂ background was modelled to decrease by 36%, and for ozone by only 2% (see Table 4.3). NO_x emissions from trucks were modelled to decrease by 61-67% (see Table 3.6). A combination of under- and overestimations that neutralizes each other explains why Berkowicz & Palmgren (1999) estimate similar results as the present study. Underestimation of the regional and urban background levels of NO_x and overestimation of NO_x emission explains the similar NO_x levels. NO₂ levels are similar although regional NO₂ and also urban NO_x are assumed to be lower. The reason is the combination of overestimated NO_x emissions and overestimated ozone levels raising NO₂ levels in the urban background.

Air Quality at Street Level

In Table 7.4, findings of the different studies are shown at street level. The selected roads in Milan, Berlin and Copenhagen are different with respect to traffic loads, vehicle composition, fraction of catalyst cars, and street configurations, and therefore air quality levels will vary.

Table 7.4
Comparison of Predicted Annual Levels at Street Level

¹ based on Berkowicz (1999a), ² based on Berkowicz & Palmgren (1999)

Berlin and Milan

It is seen that the results of the present study are similar to findings for Berlin (except for NO_x)and that much higher levels are modelled for Milan.7

Danish Study

Compared to the present study, the assessment of the 7 Copenhagen streets finds higher NO_x levels due to overestimation of NO_x emissions. NO₂ levels are similar for the same reasons given as for the urban background. Benzene levels are overestimated because the reduction of the content of benzene in petrol to 1% in 2010 is not fully taken into account in the assumptions.

All in all, the findings of the present study for Copenhagen are in accordance with the EU Urban Impact Assessment study for a comparable city like Berlin.