Jensen, S.S., Ketzel, M., Khan, J. (2023). Kortlægning af luftkvalitet langs statsvejene i Danmark. Aarhus Universitet, DCE – Nationalt Center for Miljø og Energi, 51 s. - Videnskabelig rapport fra DCE - Nationalt Center for Miljø og Energi nr. 537
http://dce2.au.dk/pub/SR537.pdf
DCE – Danish Centre for Environment and Energy, Aarhus University, has previously mapped air quality in 2012 along the state road network in Denmark for the Danish Road Directorate (Jensen et al., 2015). In the present report, this mapping has been updated with data for 2019 in addition to an improved version of the OML-Highway model.
The Danish Road Directorate has funded the development and application of OML-Highway, a GIS-based tool for calculating air quality along roads in open terrain.
Road authorities may use the nationwide air quality mapping results as a screening tool in connection with EIA studies of road projects. Citizens may use the results as an information source on air quality. The Danish Road Directorate will publish the air quality maps on their home page in the near future.
The air quality maps show selected air pollutants related to health effects: Annual mean concentrations in 2019 of NO2 (nitrogen dioxide), and the mass of particulate matter of PM10 and PM2.5, that are the total mass of particles with a diameter less than 10 and 2.5 microns. PM2.5 represents the largest health burden in Denmark, followed by NO2 according to DCE's health impact calculations for air pollution in Denmark (Ellermann et al., 2021).
The calculations are carried out for calculation points up to 1,000 m from the state road network, where the contribution from the road is less important.
The OML-Highway model requires information about the road network, traffic data, background concentrations, meteorological data and calculation points.
The road network is based on vejman.dk, which includes all state roads. As a new development, GPS-based travel speed data from INRIX has been used and obtained from the Danish Road Directorate. Previously, SpeedMap data from the Danish Road Directorate was used.
Background concentrations are calculated using the regional Danish Eulerian Hemispheric Model (DEHM) (5.6 km x 5.6 km for Denmark) and the Urban Background Model (UBM) with associated emission and meteorology data for Denmark in a 1x1 km2 grid. For Denmark, emissions are based on the emission model SPREAD, which has emissions for Denmark from all sources at a 1 km x 1 km resolution. Models and data are developed by AU/DCE.
Applied meteorological data is generated with the meteorological model WRF (Weather Research and Forecasting Model).
OML-Highway has a tool for generating calculation points along the road network, as well as forming contiguous buffer zones for visualizing of results.
In the calculations, the influence of noise barriers, dams and bridges are not taken into account, as it is difficult to link this data to the road network in such a way that the effect can be calculated with the OML-Highway, and also because the impact on air quality is limited (Jensen et al., 2013).
Road emissions are included in the SPREAD, i.e. emissions from state and municipal road networks. To avoid double counting of emissions from state roads, an extra separate calculation with OML-Highway, where emissions from the entire state road network are laid out on 1 km x 1 km grid cells, was carried out and considered "background concentrations". The final result is determined by concentrations from the "full calculation" (for each individual receptor) minus the "extra calculation" (for the nearest 1 km x 1 km grid cell) to avoid double counting.
For non-reactive substances such as PM2.5 and PM10, the above calculation procedure can be carried out without adjustments. NOx includes NO and NO2. Since NO2 is part of photochemistry, there is not a linear relationship between NOx and NO2, and the NO2 contribution from the state roads is therefore calculated using a simplified chemistry model, which is based on the annual mean of NOx, NO2 and ozone concentrations as well as information about the directly emitted NO2 emissions from traffic.
To handle the many separate OML-Highway calculations as well as the correction for double counting, the data flow in this project was controlled by newly developed programming scripts based on the R programming language (R Core Team, 2022). These R-scripts generate and reformat the creation and reformatting of input data, initiation of the OML-Highway calculations and post-processing of the results.
The calculated air quality in 2019 has been compared with current limit values; the proposed revised EU limit values from 2022; the WHO's guidelines for air quality from 2005; as well as the new WHO recommendations from 2021.
The number and type of affected homes within 1,000 m of the state road network has been calculated based on data from the Building and Housing Registry (BBR) as well as the number of people living in these homes based on the Central Civil Registration System (CPR), allowing for an estimation of air pollution population exposure.
Main Conclusions
The purpose of mapping air quality along the state road network is to describe the geographic variation. It can be considered a screening of air quality. The uncertainty can be considerable on specific points due to uncertainty in input data, but also due to the general uncertainty of the air quality models.
The aim is not to try to calculate the number of exceedances of the limit value along with state road network in Denmark for NO2. Model calculations should be seen as a complementary tool for preliminary assessment of ambient air quality and assessment of potential exceedances in locations where there are no measurements.
It is the Danish EPA which has the overall responsibility for compliance with the limit values for air quality in Denmark. The official announcement of exceedances of the limit values is done within the annual reporting from the Sub-program on air quality under the NOVANA program (Ellermann et al. 2022b). The assessment of exceedances is based on measurements from the Danish monitoring stations, and partly based on model calculations for selected urban streets in Copenhagen and Aalborg, for which calculations are made, and where it is possible to get quality assured input data from the municipalities about traffic.
The limit value does not apply to the carriageway of roads, but where people are staying, e.g. where people live and work.
There are no air quality monitoring stations along motorways since traffic stations in the larger cities are prioritized. Furthermore, former air quality campaigns at the Køge Bugt Motorway and the Holbæk Motorway have only indicated possible exceedances of the NO2 limit value very close to the motorway.
NO2
The geographical variation of NO2 in 2019 is as expected, with the highest concentrations along the busiest motorway sections in the Greater Copenhagen area and along parts of the E20 across Funen and E45 through Jutland. The highest modelled NO2 concentration (34.5 μg/m3) occurs close to the Limfjord’s tunnel, which is caused by a very high modelled background concentration due to contributions from Aalborg Portland, which are probably overestimated. Otherwise, some of the highest concentrations are found along the Køge Bugt Motorway. The maximum calculated concentration of NO2 complies with the current EU limit values (40 μg/m3) but exceeds the proposed limit value for NO2 (20 μg/m3). It also exceeds the WHO guideline from 2021 (10 μg/m3).
Pm2.5
The geographical variation of PM2.5 in 2019 is very different from NO2, as the background pollution is dominant, and hence the contribution from the state road network is less influential. There is a clear gradient from south to north with higher concentrations in the south and lower to the north, caused by emission sources in Central Europe that contribute to the background pollution in Denmark. The highest calculated value is 12 μg/m3 and complies with the current EU limit value (25 μg/m3) but exceeds the proposed limit value for PM2.5 (10 μg/m3). It also exceeds the WHO guideline from 2021 (5 μg/m3), and even the lowest calculated PM2.5 concentration exceeds the WHO guideline from 2021.
PM10
The geographical variation of PM10 in 2019 is similar to that of PM2.5, but it also differs as PM10 is significantly influenced by sea salt, which causes higher concentrations along the west and south-facing coasts due to the dominant south-westerly wind direction in Denmark. The highest calculated concentration is 23 μg/m3 and it complies with the limit value for PM10 (40 μg/m3) but exceeds the proposed limit value for PM10 (20 μg/m3). It also exceeds the WHO guideline 2021 (15 μg/m3).
Population exposure
There are 496,978 addresses, including 727,320 housing units within 1,000 m of the state road network, with a population of 1,493,386. Regarding population exposure to NO2, for all the people living within 1,000 m of the state road network the concentrtion level is below the limit value. Only about 500 people (0.04%) live at addresses where the proposed limit value is exceeded. On the other hand, about 49% live at addresses that exceed the WHO guidelines from 2021.
For all people living at addresses within 1,000 m of the state road network the concentrations of PM2.5below the current limit value. On the other hand, about 49% live at addresses with PM2.5 concentrations that exceed the proposed limit value, and all addresses exceed the WHO guidelines from 2021.
The population exposure to PM10 shows that all people living at addresses within 1,000 m of the state road network live at addresses that are below the limit value for PM10, but about 2% live at addresses with PM10 concentrations that exceed the proposed limit value, and about 97% live at addresses that exceed the WHO guidelines from 2021.
Project Results
The geographical variations of NO2, PM2.5 and PM10 in 2019 are shown in Figure 3.1 and population exposure in relation to limits values and WHO guidelines are shown in Figure 3.2.
