Jensen, S.S. & Ketzel, M. 2006. Research Notes from NERI, No. 232. 43 pp.
Summary
New stringent European Union emission standards for particles are expected to be implemented around 2010 for diesel-powered passenger cars and vans. Provided that such standards are agreed upon diesel-powered passenger cars and vans are expected to be equipped with particle filters that effectively will remove particle emissions.
Against this background, the Danish Environmental Assessment Institute wanted to carry out a cost-benefit analysis of a scenario where the emission standards came into force in 2007 instead of 2010.
The objective of the project is to assess the effects on particle emissions and air quality of PM2.5 (particles less than 2.5 micrometer) of moving forward the emission standards for passenger cars and vans. An assessment of a scenario is carried out that assumes that all new diesel-powered passenger cars and vans entering the vehicle fleet during January 1st, 2007 and December 31st, 2009 will be equipped with particle filters. The National Environmental Research Institute has carried out this assessment.
In the business-as-usual scenario and the particle filter scenario in 2010 the share of passenger cars and vans that are diesel-powered and the share of passenger cars and vans that are equipped with particle filters are important assumptions for assessment of emissions. Data for these assumptions has been provided by the Danish Environmental Assessment Institute.
In the business-as-usual scenario some passenger cars and vans will be equipped with particle filters even if the emission standards are not brought forward. Some expensive cars already have particle filters and the newly introduced tax reduction on particle filters is also expected to lead to more cars with particle filters. It is assumed that the share of diesel-powered passenger cars will increase from 11 % to 33 % and from 7 % to 37 % for vans between the business-as-usual scenario and the particle filter scenario.
It is assumed the share of diesel-powered passenger cars will increase to 15 % by 2010, and to 84 % for vans.
The air quality of PM2.5 in urban background has been selected as an indicator for the environmental assessment since it is one of the few indicators for which it is possible to carry out a cost-benefit analysis.
Calculations of PM2.5 in the business-as-usual and the particle filter scenario have been carried out the Urban Background Model (UBM) developed by the National Environmental Research Institute. Calculations are carried out on a 1x1 km2 grid for the Greater Copenhagen Area. This region encompasses the Municipality of Copenhagen, the Municipality of Frederiksberg, the County of Copenhagen, the County of Roskilde and the County of Frederiksborg. The population in this region is about 1.8 million people. Results from this region has been extrapolated to the rest of the country to provide for an assessment of Denmark (5.3 million people).
A particle filter will only reduce the number and mass of particle exhaust and not the non-exhaust part. The non-exhaust part includes particle emissions related to mechanical processes (wear of brakes, tires, and road surface, and re-suspension of dust).
To assess the uncertainty of the model results a comparison of modeled and measured data was carried out for an urban background monitor station at the roof of the H.C. Ørsted Institute in Copenhagen.
The UBM model estimated the urban background contribution from vehicle particulate emissions to be 0.7 µg/m3 for PM2.5 in Copenhagen. Measurements between a regional background monitor station and the monitor station at H.C. Ørsted Institute showed similar levels (20 µg/m3) suggesting that the urban background contribution is small. For PM2.5 the regional background contribution is dominant.
Model results showed good correlation with measurements of NOx, NO2, and O3 indicating that the model describes the dispersion well.
For further testing of the applied emission factors calculations were performed for street concentrations at a busy street in Copenhagen (H.C. Andersen’s Boulevard) using the Operational Street Pollution Model (OSPM) developed by the National Environmental Research Institute. The street contribution was modeled with the OSPM model and compared with measurements. The street contribution is the concentration monitored in the street minus the urban background concentration at the H.C. Ørsted Institute. The comparison showed that modeled average diurnal profile of NOx correlated well with measurements but were overestimated by 10-15 % for PM2.5.
The exhaust emission of PM2.5 will be reduced by 18 % for passenger cars and 35% for vans as an average in the Greater Copenhagen Area. However, the total reduction in PM2.5 exhaust emissions is on by 12 % for all vehicle categories since the emissions from heavy duty vehicles are unchanged.
Since the non-exhaust part is unchanged the total reduction in PM2.5 emissions are 8 % for passenger cars and 30 % for vans. The relative reduction is larger for vans compared to passenger cars. This is because the exhaust emission is higher for vans since the share of diesel-powered vehicles is higher for vans.
The total reduction for exhaust and non-exhaust for all vehicle categories is 9 % as a consequence of bringing forward the particle emission standards for passenger cars and vans.
The geographic distribution of PM2.5 concentrations was modeled with the UBM model for the Greater Copenhagen Area on a 1x1 km2 grid. The maximum model contribution to the urban background from traffic was 1.14 µg/m3 or 6 % in relation to the regional background level of 20 µg/m3. As expected, the contribution from traffic was highest in the major transport corridors and in highly populated areas.
Annual mean of urban background concentrations of PM2.5 at the H.C. Ørsted Institute in Copenhagen was modeled to 20.7 µg/m3 in the business-as-usual scenario in 2010. The contribution from traffic was 0.7 µg/m3 as the regional background level is 20 µg/m3.
The effect of the particle scenario was a reduction of 0.047 µg/m3 in the urban background at the H.C. Ørsted Institute in Copenhagen. This corresponds to a reduction of 6 % of the contribution from traffic to PM2.5 urban background concentrations. However, in relation to the urban background levels the reduction is only 0.2 % since it is dominated by the regional background level.
The largest modeled reduction was 0.075 µg/m3. This corresponds to 10 % of the contribution from traffic to urban background PM2.5 concentrations or 0.4 % in relation to the urban background levels. The largest concentration reductions were in Copenhagen and along the major transport corridors in the Greater Copenhagen Area whereas reductions were less in smaller cities and in rural areas.
Hence, the reductions of PM2.5 concentrations in the urban background were marginal due to the particle filter scenario. The reductions are so small that it is not possible to measure the difference in urban background concentrations since the uncertainty on measurements are much higher than the modelled reductions.
The reduction in population exposure due to the particle filter scenario depends on the reduction in PM2.5 concentrations and the population density. The largest reduction in population exposure was found in the most densely populated urban areas.
Previous Danish studies have not assessed the effect of particle filters on passenger cars and vans but have focused on heavy duty vehicles. One study assessed the impacts of a requirement for all trucks and buses to be equipped with particle filters as part of a proposed environmental zone in the central part of Copenhagen. Another study has assessed the impact of equipping all heavy duty vehicles in Denmark with particle filters.
The overall findings of the present study are consistent with the previous studies.
The environmental assessment has been limited to assess PM2.5 in urban background as this is the indicator used for the subsequent cost-benefit analysis. Dose-response relations exist for this indicator for the relation between exposure and health effects. However, other particle fractions or other air pollution indicators could also be relevant. PM2.5 is a complex indicator and the methodology assumes that dose-response relations established for PM2.5 also is applicable for reduction of soot particles that are targeted by particle filters.
The modelled reductions of PM2.5 concentrations in the urban background are marginal due to the particle filter scenario. These marginal reductions are the basic input to a series of subsequent multiplications involving population density to get exposure, dose-response relations to get health effects and finally unit costs to get benefits. In this process it is assumed that all relations are linear and even marginal reductions will lead to benefits.
The impact of the particle filter scenario on street concentrations have not been quantified as the assessment has focused on urban background. However, the relative reduction in street concentrations will be higher than for urban background.
The total PM2.5 emission from traffic is uncertain but the distribution between vehicle categories and between exhaust and non-exhaust is even more uncertain. These uncertainties obviously influence the assessment of a particle filter scenario focused on the exhaust of passenger cars and vans.
Particle filters should theoretically increase fuel consumption and hence CO2 emissions but this effect is marginal and not measurable in practice.
The share of directly emitted NO2 has been seen to increase for certain filter technologies for heavy duty vehicles. However, the filter technology applied for factory equipped particle filters for passenger cars and vans is different and does not exhibit this problem.
As described above, there is still a need for more knowledge to reduce the uncertainty in relation to assessment of the effects of control measures against particulate air pollution.
Full report in pdf-format (846 kB).
