Aarhus University Seal / Aarhus Universitets segl

No. 64: Health effects and related external costs of air pollution in Copenhagen

Brandt, J., Jensen, S.S., Plejdrup, M. (2013): Sundhedseffekter og relaterede eksterne omkostninger af luftforurening i København. Aarhus Universitet, DCE –Nationalt Center for Miljø og Energi, 46 s. - Videnskabelig rapport fra DCE - Nationalt Center for Miljø og Energi nr. 64. http://www.dmu.dk/Pub/SR64.pdf

Summary

Background and Purpose

Air pollution has significant negative effects on human health and wellbeing and this has significant socio-economic consequences.

The purpose of the project is to calculate the health impacts and external costs related to air pollution in the Municipalities of Copenhagen and Frederiksberg using the integrated model system EVA (Economic Valuation of Air pollution). The EVA-system calculates health impacts and related external costs based on information on locations of air pollution sources, the dispersion of air pollution and related population exposure, dose-response relationship between exposure and health effects, and valuation of health effects, also called external costs related to health effects of air pollution.

The study

Health effects and related external costs are first calculated for the total air pollution in the Copenhagen metropolitan area (approx. 1.8 million inh.) with separate estimates for the Municipalities of Copenhagen and Frederiksberg (approx. 0.7 million inh.). The total air pollution is a result of all sources in the Copenhagen metropolitan area and all other sources in Denmark and abroad. This gives a picture of how air pollution contributes to health impacts whether it's from local sources in Copenhagen or other sources.

Additionally, the calculations are carried out for each major type of emission source in Copenhagen providing the contribution from each source within the municipalities of Copenhagen and Frederiksberg which are the sources that the cities are able to regulate. Emissions used for the calculation are equivalent to those used in the report "Kildebidrag til sundhedsskadelig luftforurening i København" (Jensen et al. 2013).

The health impacts and associated external costs are first calculated for the Copenhagen metropolitan area based on emissions from the municipalities of Copenhagen and Frederiksberg and the impacts are distributed to the different emission sources. Then similar calculations are performed but only including impacts within the municipalities of Copenhagen and Frederiksberg and based on local emissions within these two municipalities. The different calculations are summarized in Table 3.1.

The calculations are carried out with the integrated model system EVA, Economic Valuation of Air pollution (Brandt et al., 2011a, b; 2013a, b), which is based on the so-called "impact pathway" approach. The calculations are based on the regional air pollution model DEHM with a resolution of 16.7 km x 16.7 km for Denmark. This solution is too coarse for an assessment for the municipalities of Copenhagen and Frederiksberg, and there has been a significant development of the EVA-system to include the Urban Background Model (UBM) to be able to perform calculations on a 1 km x 1 km resolution in the Copenhagen metropolitan area. This development has achieved a unique high spatial resolution among systems for health impact and external costs assessment in an international context.

The urban background pollution in a city is the general air pollution in the city, and reflects the concentration in a park or on the roof of buildings. Urban background concentrations differ from street concentrations where also the direct contribution from the traffic in the specific street is included.
Calculations are performed for 2010 which is the latest year where updated emissions are available for Denmark at 1 km x 1 km resolution.

Main Findings

Premature deaths and morbidity

The total annual number of premature deaths in 2010 in the Copenhagen metropolitan area (model domain) is approx. 1500 of which approx. 540 are within the municipalities of Copenhagen and Frederiksberg due to all domestic and foreign emission sources.

Premature deaths are almost entirely due to so-called chronic deaths caused by long-term exposure to air pollution in contrast to so-called acute deaths due to short-term exposure during periods with elevated concentrations (episodes).

Those who die prematurely due to air pollution are especially elderly weakened persons who already suffer from cardiovascular and respiratory diseases. Deaths among infants represent a very small part.

In addition to premature deaths, there are many cases of morbidity. This include chronic bronchitis and discomfort for children and adults with asthma (use of bronchodilator, cough and respiratory symptoms), hospital admissions for respiratory disease and stroke, heart failure, lung cancer, and many with reduced activity (sick days).

Health Impacts in Copenhagen metropolitan area due to local emissions sources in municipalities of Copenhagen and Frederiksberg

The purpose of these calculations is to quantify how much the local emission sources in Copenhagen contribute to health impacts in the Copenhagen metropolitan area. The calculations include distances up to approx. 20 kilometers from Copenhagen.

The local emission sources in Copenhagen contribute to approx. 80 premature deaths in Copenhagen metropolitan area while all air pollution in Copenhagen (domestic and foreign sources) contributes to about 540 premature deaths. This means that local emission sources in Copenhagen contribute to about 15% of all premature deaths in Copenhagen metropolitan area.

The main local sources to the approx. 80 premature deaths in the Copenhagen metropolitan area is non-industrial combustion which primarily consists of combustion from wood stoves and boilers (about 50 premature deaths) and road transport (about 20 premature deaths). The same pattern is observed for morbidity.

Similarly, it was examined to what extent local emission sources in Copenhagen contribute to health impacts in Copenhagen.

The local emission sources in the municipalities of Copenhagen and Frederiksberg contribute to approx. 67 premature deaths in the municipalities of Copenhagen and Frederiksberg while sources in Copenhagen contributed to about 80 premature deaths in Copenhagen metropolitan area as described above. This means that the municipalities of Copenhagen and Frederiksberg contribute to approx. 13 premature deaths to surrounding municipalities within a distance of approx. 20 kilometers from Copenhagen.

The local emission sources in the municipalities of Copenhagen and Frederiksberg contribute to about 12% of all premature deaths in the municipalities of Copenhagen and Frederiksberg.

The main local sources are also non-industrial combustion which primarily consists of combustion in wood stoves and boilers followed by road traffic. The same pattern is observed for morbidity.

External costs due to all air pollution

The total external costs in the Copenhagen metropolitan area in 2010 due to all air pollution from both domestic and foreign emissions sources are about DKK 12 billion (DKK 109) per year, of which about DKK 4 billion per year in Copenhagen and Frederiksberg.

The external costs are almost entirely related to particles that here include primary emitted particles, and secondary inorganic particles formed in the atmosphere (nitrate, sulfate, ammonium), secondary formed organic particles (SOA) and "unknown mass". Unknown mass is likely to include water bound to the particles. Since sea salt is not yet included in the DEHM model sea salt is also part of "unknown mass" in this context, and the levels of SOA and unknown mass are included in the calculation with the EVA-system based on analysis of measurements.

External costs related to ozone have been calculated to about DKK 250 millions in Copenhagen metropolitan area, including about DKK 60 million for the municipalities of Copenhagen and Frederiksberg. Ozone is not directly emitted but is formed in the atmosphere from emissions of NOx, hydrocarbons and CO. Ozone is directly harmful to health, and therefore associated with external costs.

Contributions from CO to external costs are very small with about DKK 4 millions in Copenhagen metropolitan area including about DKK 2 million in the municipalities of Copenhagen and Frederiksberg.

Most of the external costs are due to premature deaths.

External costs in the Copenhagen metropolitan area related to emission sources in Copenhagen

It is investigated how much the different local emission sources in Copenhagen contribute to health impacts in the Copenhagen metropolitan area.

The total external costs are about DKK 600 millions in Copenhagen metropolitan area due to emissions sources in municipalities of Copenhagen and Frederiksberg in 2010. The total external costs in the Copenhagen metropolitan area due to all air pollution from domestic and foreign emissions sources are approx. DKK 12 billion and local emissions in municipalities of Copenhagen and Frederiksberg contribute only about DKK 600 million corresponding to around 5%. This comparison is not entirely justified since we only include the external costs up to 20 km from the municipal border of Copenhagen and compares this with the total external costs within Copenhagen metropolitan area where we should compare with the total external costs within 20 km. However, this is not at present possible because of the way the model is set up. 5% is therefore an underestimation.

The external costs are due almost entirely to PM2.5 as the local sources only are primarily emitted PM2.5 as secondary particles are not formed on the short transport distances within the Copenhagen metropolitan area. The external costs of ozone are negative (DKK -60 million) which in this context shows that ozone counts positively. This is due to local emissions of nitrogen oxides (NOx = NO + NO2) in municipalities of Copenhagen and Frederiksberg and contribute to the reduction of O3 levels in the city (NO reacts with O3 to form NO2), and therefore there is a "positive" effect of NOx since it reduces harmful O3.

External costs related to CO are negligible.

The main local source in Copenhagen that contributes to the external costs in the Copenhagen metropolitan area is non-industrial combustion (which consists primarily of combustion in wood stoves and boilers) which accounts for about 2/3 of the external costs followed by road transport with 1/4. Other types of sources which give a notable contribution is the use of products (emissions from solvents and use of products) with approx. 5% and mobile sources (non-road mobile machinery, etc.) with approx. 4%.

External costs in Copenhagen related to emission sources in Copenhagen

Similarly, external costs for the municipalities of Copenhagen and Frederiksberg in 2010 due to local emissions in the municipalities of Copenhagen and Frederiksberg have been calculated.

The total external cost is about DKK 500 million in Copenhagen and Frederiksberg due to emission sources in these two municipalities in 2010. The total external costs in Copenhagen due to the total air pollution from domestic and foreign emission sources are approx. DKK 4 billion and the local emissions in Copenhagen and Frederiksberg contribute only about DKK 500 million equal to 11%. On the other hand, this means that about 89% of all external costs related to air pollution in Copenhagen are caused by emissions outside Copenhagen.

If we compare the difference between external costs from the contribution of emission sources in the municipalities of Copenhagen and Frederiksberg, and Copenhagen metropolitan area, it represents approx. DDK 100 million. Consequently, the emission sources in the municipalities of Copenhagen and Frederiksberg contribute to external costs in the order of DKK 100 million outside the municipal border of Copenhagen at distances up to about 20 km.

The main local source in Copenhagen that contribute to the external costs is non-industrial combustion (which consists primarily of combustion in stoves and boilers) which accounts for about 2/3 of the external costs followed by road transport (1/5). Other types of sources with a notable contribution is the use of products (emissions from solvents and use of products) with approx. 5% and mobile sources (non-road mobile machinery, etc.) with approx. 4%. The distribution of external costs between these sources is roughly equivalent to the distribution in the Copenhagen metropolitan area due to emission sources in Copenhagen, as demonstrated above.

Uncertainties

Impact pathway method

The EVA-system is based on the impact pathway approach covering emissions from sources, the dispersion and chemical transformation in the atmosphere, exposure of the population, health effects, and economic valuation of these health effects. There are uncertainties in all these elements.

Emission sources in Copenhagen

The calculations of health effects and external costs of emissions in the municipalities of Copenhagen and Frederiksberg are based on the official national emission inventory broken down by different geographical distribution keys.

The main local source in Copenhagen in relation to the external costs of Copenhagen is non-industrial combustion (which consists primarily of combustion in wood stoves and boilers) which accounts for about 2/3 of the external costs. However, there is considerable uncertainty in this estimate related to the level of emissions and their geographic distribution.

The other main local source of emissions in Copenhagen in relation to external costs is road transport which accounts for about 1/4 of the external costs. There is less uncertainty in emissions from road traffic since the knowledge base and the potential for validation of emissions is much greater than for wood-burning stoves.

Calculation of dispersion of air pollution

The current calculations do not include the contribution of the health effects and related external costs from emissions in Copenhagen on a regional scale, that is, outside the Copenhagen metropolitan area. Emissions of air pollution will disperse over greater distances leading to population exposure further away - in terms of the primarily emitted PM2.5 and secondarily formed nitrate particles from NOx emissions in Copenhagen. It has not been possible to take into account this effect within the framework of this project since it would require development of a feedback from the urban background model (UBM) to the regional model (DEHM). This means that the estimated external costs are underestimated since it does not include the impacts outside the Copenhagen metropolitan area. It also leads to an underestimation of NOx sources e.g. road transport compared to wood stoves.

Health hazard of particles

There is great uncertainty about which types of particles (both size and chemical composition) cause health hazards. The international knowledge base, however, shows a clear correlation between the concentration of particles smaller than 2.5 microns (PM2.5 measured as mass) and premature deaths. With the current knowledge, it is not possible to quantitatively distinct between health effects of particles of different sizes and chemical composition, and therefore these are treated equally in the calculations. However, recent studies (Rohr & Wyzga 2012) suggest that carbonaceous particles (e.g. directly emitted) are more harmful than non-carbonaceous particles, such as the inorganic secondary particles (e.g. particles that are emitted as gas and by chemical processes in the atmosphere are converted into particles).

There is still a need for more information about the relationship between air pollution and health effects, and how it depends on different types of particles and different types of emission sources.

Spatial resolution

The spatial resolution of the model system may affect the results of the calculations of the health impacts, especially for local sources of road traffic and wood stoves. At low spatial resolution will likely underestimate population exposure for persons living close to local sources since air concentrations are greater than predicted by the model and correspondingly overestimated for those who live further away. In order to partly account for this the present project has developed a higher spatial resolution of the model where the resolution is increased from 16.7 km x 16.7 km to 1 km x 1 km for Copenhagen metropolitan area by integration of the urban background model (UBM) into the EVA-system.

Population exposure is now determined with a resolution of 1 km x 1 km for urban background concentration as an indicator but it is likely that an even higher spatial resolution using the street addresses with associated higher resolution in concentration calculations (e.g. integration of street model) would give even more reliable results. Studies have shown that people who live within 50 meters of a busy urban street and 100 meters from a highway have increased risk of premature death from heart and lung diseases (Hoek et al. 2002). However, at present the knowledge about dose-response relationships is not comprehensive enough to quantify the overall health effects of a population on the basis of exposure at address level.