Traffic-related Air Pollution - Benzol, Nitrogen Oxide and Diesel Particulates 1993

Statistical Base

Cadastre of Motor Traffic Emissions

The Berlin Department of Urban Development, Environmental Protection and Technology (SenStadtUmTech – Senatsverwaltung für Stadtentwicklung, Umweltschutz und Technologie) maintains a cadastre of emissions for the major groups of polluters, including the “motor traffic” polluter group.

The 1993 Cadastre of Motor Traffic Emissions gives the first unified picture of air pollutant emissions produced by motor traffic for the entire city of Berlin.

This cadastre uses a new method to calculate emissions. This method is also a suitable basis for dispersal calculations which can ascertain pollutant loads on roads. This far-reaching restructuring of calculation methodology enables only limited comparison with previous emission investigations based on much simpler methods.

Investigation of Motor Traffic Pollution

The basis is the first comprehensive traffic count, performed in 1993. This count included the primary road network as well as scheduled bus routes. This count resulted in the availability of the following data for every road segment in the primary road network:

  • average daily motor traffic (DTV) in motor vehicles/day,
  • average daily truck traffic in trucks/day for heavy trucks,
  • percentage of busses in regular traffic.

This data was supplemented with extensive analyses of vehicle type and total travelled distance of registered motor vehicles in Berlin. The data was also supplemented by emission factors that describe these cars and utility vehicles (cf. Map 07.01 SenStadtUm 1995).

Methodology of Emission Studies

Pollutant emissions produced by motor traffic include the exhaust and abrasions of moving traffic, the evaporative emissions of stopped traffic, and evaporative emissions at fuel stations. Figure 2 presents an overview of the methodology for investigating emissions. Fuel station emissions are listed under light industry.

Fig. 2: Methodology of the 1993 Traffic Emission Cadastre
Fig. 2: Methodology of the 1993 Traffic Emission Cadastre
Image: Umweltatlas Berlin

Emission models aided the calculation of pollutants and CO2 for line sources (primary roads), area sources (secondary roads, and evaporative emissions).

Exhaust and abrasion emissions appear as line sources on primary and secondary roads. These emissions are calculated as line sources only for the primary road network because only these streets had data available from previous counts for average daily traffic values (DTV) and hourly capacity. Emissions from line sources are classified as area values in the grid system. Emissions for the secondary road network, however, are directly deduced from the separate grids from assumptions made about traffic volumes and amounts of trucks.

Hydrocarbon evaporative emissions occur from pressure differences between the fuel tank and the carburetor float chamber. They occur

  • in non-moving motor vehicles resulting from daily temperature fluctuations (“tank respiration emissions”),
  • in hot engines after long distances,
  • in warm engines after short distances.

Evaporative hydrocarbon emissions and benzol fractions are also determined for the grids. Evaporative emissions resulting from refueling are also calculated. Evaporative emissions from moving traffic could be neglected because they are very low.

Emissions Models for Primary Road Networks (Line Sources) and Secondary Road Networks (Area Sources)

The emission simulation model EMISS helped calculation of pollutants, CO2 emissions (cf. Map 08.03 CO2 Emissions, SenStadtUmTech, in preparation), and fuel consumption for traffic on primary road networks.

Figure 3 shows the individual model parameters, including total travelled distance factors, stop-and-go formulas, cold start factors, etc., and the results. The methodological background is described in detail in Liwicki, Garben 1993.

Link to: Vergrößern
Fig. 3: EMISS - Emmission Model for Primary Roads (Line Sources)
Image: Liwicki, Garben 1993

Emissions for motorized two-wheel vehicles could not be ascertained because traffic counts do not exist. Two-wheel vehicle contribution to total emissions are calculated on the basis of national total travelled distances, and on available emission factors.

Road segments in areas of varying topography are classified according to longitudinal inclines; but this is not necessary for Berlin.

Emission Model for Secondary Road Network (Area Sources)

Link to: Vergrößern
Fig. 4: EM-NEBEN - Emission Model for Secondary Road Network (Area Sources)
Image: Liwicki, Garben 1993

Emissions in the secondary road network are not calculated for each specific road segment. They are calculated from a grid area of 1 km2 per segment. Travelled distances within the grid surface area is estimated based on the following data:

  • predominant use of the area, subdivided into
    • residential living in outer areas of the city,
    • small business and industrial areas,
    • inner city and suburban areas,
  • number of inhabitants and positions of employment, differentiated according to
    • commercial and service industry,
    • manufacturing industry,
  • and motor traffic source-goal-matrices derived from the above.

It can be assumed that significant traffic jams do not occur in the secondary road network. “Stop and go” supplements are not added to the calculations. Daily, weekly and annual matrices for the secondary road network are not necessary.

Further input variables needed for determining total emissions for each pollutant component in each grid area correspond to input variables used in calculating total emissions in the primary road network.

Pollutions – Road Monitoring Results

Five road monitoring stations were set up and random sampling programs (1990,1991, 1993/4, 1996) were carried out in the framework of the Air Quality Monitoring System BLUME. Table 4 depicts results from the 1996 random sampling programs. Benzol and particulates were monitored at 16 sampling points on primary roads and at 4 points in inner city residential areas.

The results show that pollution loads on the primary roads monitored are 2-3 times higher than in inner city residential areas.

Link to: Vergrößern
Tab. 4: Annual Parameters of Road Monitoring in the Random Sampling Monitoring Program, 1993/94 and 1996
Image: Umweltatlas Berlin


  • Data for traffic volume based on 1993 count.
  • The 98 % values for nitrogen dioxide in the 1993/94 random sampling monitoring program were calculated from yearly average values (Factor 2.2).

Concentration levels are not solely dependent on the number of vehicles and these emissions. They are also dependent on air exchange conditions determined by meteorological parameters such as wind, as well as by the size and type of surrounding constructions. That is why high pollution loads are registered in streets lined on both sides with buildings, such as Silbersteinstrasse in Neukölln and Schildhornstrasse in Steglitz. The city expressway has a much higher traffic volume but registers lower pollutant concentrations. Figure 1 depicts typical pollutant distribution in a street lined with buildings. This distribution occurs when the wind (from over the roof) blows from the point of sampling towards the street and creates a whirling current within the street lines with buildings. This current sweeps motor vehicle emissions to the roadside which has the monitoring station.

The results of the 1997 monitoring program at primary roads show

  • Nitrogen dioxide loads (only at permanent monitoring stations) show a slight decrease since 1990. The 23rd Regulation concentration value was exceeded at the Silbersteinstrasse monitoring stations.
  • Benzol loads have decreased compared to previous measurements in random sample monitoring programs in 1990 and 1993/94. This development is partly due to significant decreases in the number of motor vehicles with two-stroke engines (mainly East German “Trabbis”). These engines release extremely high hydrocarbon emissions. Benzol emissions also noticeably fell as a result of increasing numbers of cars equipped with regulated catalytic converters. Concentration values for benzol set at 10 µg/m3 as of 1998 were exceeded on three streets in the 1996 random sample monitoring program. The concentration value of 15 µg/m3 effective 1 July 1998 was not reached, however. Automatic monitoring procedures for benzol have only been available for a few years. Data from permanent monitoring stations exists only after 1995.
  • Particulates show constant high loads at permanent monitoring stations and in random sample monitoring programs. The concentration value of 8 µg/m3 in effect since 1 July 1998 was exceeded at almost every monitored primary road. The concentration value of 14 µg/m3 was exceeded at the two monitoring points Silbersteinstrasse and Grünauer Strasse.

Particulate load is mainly produced by diesel motor vehicles and tire abrasion. A perceptible reduction is not to be expected in the next few years.

Pollutions – Need Assessment for Dispersal Calculation Studies at Highly Stressed Road Segments

Road monitoring results indicate that 23rd Regulation concentration values have been exceeded at a significant number of primary roads; especially particulate loads. The monitoring of every road in the urban area is not financially feasible. This is why the pollution load for the entire primary road network in Berlin has been assessed with an emissions and dispersal calculation. This calculation method helps determine roads where loads definitely exceed or remain close to concentration values given by the 23rd Regulation.

An estimate for every primary road (screening) is enabled by the IMMIS-Luft method, a calculation instrument developed specifically for this purpose.

The IMMIS-Luft screening program was designed to assess road pollutant loads in the context of the 23rd Regulation. IMMIS-Luft helps to quickly calculate pollution loads for individual roads as well as for extensive road networks, if necessary variables are known.

Pollution loads at both sides of the road were calculated at a height of 1.5 m and at a distance of 1.5 m to the building edge (cf. Fig. 1). The calculated mean pollutions at these two points are considered to be the characteristic pollution loads at that road segment.

This model requires variables at 3,000 primary road segments to be prepared. These variables include

  • average daily traffic volume, determined from 1993 counts,
  • amounts of heavy utility vehicles (trucks) and busses,
  • street width,
  • building heights,
  • coordinates of road ends and
  • ratio of buildings to road length (porosity).

The following additional variables valid for the entire urban area of Berlin had to be ascertained:

  • previous existing air loads: yearly average for particulates (4.1 µg/m3), yearly average for benzol (4.0 µg/m3), and the 98 % value for nitrogen dioxide (72 µg/m3), as a short-term characteristic value,
  • average wind speed (3.2 m/s),
  • average emission factors for cars and trucks at different driving speeds. The specific emission factors for each motor vehicle type, including tire abrasion, were ascertained from the “Manual of Vehicular Traffic Emission Factors” (Umweltbundesamt – Federal Environmental Agency) and weighted with the composition of motor vehicle fleet determined by a monitoring project at Schildhornstrasse.

The composition of car fleet was:

  • 57 % with reguleted catalytic converter (gKat),
  • 17 % with diesel engines,
  • 1 % with 2-stroke engines,
  • 25 % with conventional internal combustion engines (as of 1994).

Light utility vehicles of less than 3.5 tons are included in the car fleet.

Benzol loads were derived from hydrocarbon emissions; an average of about 6 % benzol. Particulate loads were calculated from particle emissions. It was assumed that 40 % of particle emissions consist of particulates.

A flat rate of 10 % for traffic jams was added to screening calculations.

Map 03.10.9 presents Calculated Benzol and Particulate Pollutions at Current, and a Projection for the Year 2010. Motor vehicle load on primary road networks was ascertained by model calculations. The projection for the year 2010 assumed all cars to be equipped with regulated catalytic converters.

This kind of screening process cannot take into account extremely differing local variations in emission and dispersal conditions. The following 3 categories were formed to assess results:

  • “green area” reference value not exceeded,
  • “yellow area” reference value possibly exceeded,
  • “red area” reference value exceeded.

The assessment of road segments with calculated concentration values within the yellow or red areas was modified 1) if prerequisites for the screening calculations were fulfilled, and 2) if people were in affected areas for more than just short periods of time. Some expressway segments where people spent longer periods of time are no longer included for further assessment.

Results are summarized in Figures 2, 3, and 4, and in Table 5. They show that the 23rd Regulation concentration values effective July 1998 will be exceeded at many road segments. Concentration values due to high particulate concentrations can be expected to exceed guide values at about 20 % of primary road segments in Berlin; this means at more than 500 locations. Benzol and particulate concentrations have been monitored at some road segments since April, 1997, following a priority list.

Link to: Vergrößern
Tab. 5: Calculated Concentrations of the Berlin Primary Road Network Compared to the Reference Values of the 23rd Regulation of Federal Pollution Control Law. The Total Length of Primary Roads is 1,145 km; there are a total of 2,967 Road Segments.
Image: Umweltatlas Berlin