TERM EU Emissions per passenger-kilometre and per tonnekilometre

Indicator fact sheet TERM 2001 28 EU Emissions per passenger-kilometre and per tonnekilometre for NO x, NMVOCs, PM 10 and SO x by mode Specific emissions of all pollutants from cars, buses and trucks have decreased significantly over the past decade. Projections show that further reductions for all pollutant emissions can be expected from all modes, as a result of more stringent emission limits and fuel quality standards and improved technologies. Figure 1: NO x Modelled specific emission of NO x, VOC, PM and CO per passenger-km (cars and buses) and tonne-km (trucks) VOC 120 120 Index (1990=100) 100 80 60 40 Trucks Cars Buses Index (1990=100) 100 80 60 40 Trucks Cars Buses 20 20 0 1990 1995 2000 2005 2010 0 1990 1995 2000 2005 2010 PM CO 120 120 Index (1990=100) 100 80 60 40 Trucks Cars Buses Index (1990=100) 100 80 60 40 Trucks Cars Buses 20 20 0 1990 1995 2000 2005 2010 0 1990 1995 2000 2005 2010 NB: Cars and buses/coaches are based on grams pollutant emissions per passenger-kilometre; trucks are based on grams pollutant emissions per tonne-kilometre. Source: EEA ETC/ACC, 2002. 17-12-2002 1

Results and assessment Policy relevance By reducing emissions per transport unit (passenger-km or tonne-km), less environmental damage occurs for transportation of the same amount of goods and passengers. Such reduction can be realised by either increasing occupancy rates or load factors (see TERM 2002 29 EU and 30 EU), or by decreasing the emissions per vehicle-kilometre (i.e. setting higher emission standards, see TERM 2002 34 EU Proportion of the vehicle fleet meeting certain emission standards). No explicit targets exist on the European level directly addressing specific emissions. Policy context For policy on occupancy rates, load factors and emissions standards, the reader is referred to TERM 2002 29 EU Occupancy rates, TERM 2002 30 EU Load factors, TERM 2002 34 EU Proportion of the vehicle fleet meeting certain emission standards. Environmental context Specific emissions are defined as emission (of air pollutants) per transport unit (tonne-km or passenger-km), specified by mode (car, bus/coach, truck, etc.). Since these emissions are expressed per transport unit, occupancy rates and load factors play an important role with respect to specific emissions. Therefore, variations can be wide, especially for buses. Doubling the number of passengers results in almost a halving of specific emissions per passengerkilometre, since the extra passengers do not significantly change vehicle performance. For trucks, extra load changes the performance (i.e. energy uptake and specific emission of air pollutants) more significantly: higher load factors may not result in a proportionally better performance. The average age of the vehicle fleet is of great importance for specific emissions of the whole vehicle fleet, as it reflects the technology level. ForeMove, the model used for calculating specific emissions, therefore takes the vehicle fleet composition into account. In order to compare the emissions of different transport modes, ideally, difference in transport distance between the origin and destination by different modes, and the routes taken, should be taken into consideration. If a ship has to travel twice as far as an aircraft between the same origin and destination, both transport modes will emit around the same amount of pollutants when this ship emits half of the emissions per transport unit compared with an aircraft (the impact of more or less emissions also depends on the resulting concentrations of pollutants, which in cases of longer distance will be spread over larger areas, leading to lower concentrations). Lastly, emissions in highly populated or ecologically sensitive areas can be much more harmful than the same amount (or even more) emissions emitted in less sensitive areas or lowly populated areas. Assessment Specific NO x emissions from passenger cars fell significantly (47 % compared to 1990), mainly as a result of the introduction of catalytic converters in order to meet emission limits set by legislation. For heavy- and light-duty trucks, specific NO x emissions also decreased markedly: by 20 % between 1990 and 1998. Specific NO x emissions from buses also decreased, but less (12 %) during the same period, mainly because of decreases in occupancy rates (see TERM 2002 29 EU Occupancy rates). Specific NO x emissions are projected to continue to decline. The directives on emission standards for new passenger cars and trucks should result in significant reductions of specific NO x emissions from 2000 up to 2010: 64 % for cars, 50 % for trucks and 44 % for buses and coaches. In order to achieve these reductions, appropriate inspection and maintenance of the fleet will be necessary (particularly for cars), as will efficient loading of trucks and sufficient occupancy rates for buses and coaches. The specific emission of particulate matter (PM) from passenger cars increased up to 1985 but has since been declining, mainly as a result of improved technology and the introduction of limit 17-12-2002 2

values for PM emissions from diesel engines by Directive 88/436/EEC ( 1 ). For trucks, the specific emission of PM is also decreasing, but this process started somewhat later than with passenger cars (see Figure 1). Benefits from the introduction of the clean lorry directive (91/542/EEC) ( 2 ), reducing limit values for emissions in two phases, are becoming visible and clearly show the delay in effect. This is due mainly because new trucks replace older models relatively slowly (see TERM 2002 32 EU Average age of the vehicle fleet). The specific PM emission of buses has improved over the past decade, after worsening at the beginning of the 1990s. This improvement in PM emissions, however, went slower for buses than for trucks. Since the same emission standards apply to buses and to trucks, the difference in development might be explained by differences in average age between trucks and buses (though no statistics are available to support this statement), or by different changes in occupancy rates and load factors ( 3 ). The introduction of even more stringent standards for gaseous emissions and PM (Euro IV and Euro V classes) together with the environmentally enhanced vehicle (EEV) standard for heavy-duty vehicles are expected to further decrease specific PM emissions. Similar trends can be seen with respect to VOC and CO emissions. The enormous reduction in VOC emissions from passenger cars (51 % between 1990 and 1998) result from improvements in engine technology (Figure 1). Similar improvements can also be seen in CO emissions (Figure 1). Specific emissions of NMVOC from motorcycles (not shown in the graph), increased markedly during the 1960s and fell again only in the early 1990s. Motorcycles still have high specific emissions. The introduction of the directive on certain components and characteristics of two- or three-wheel vehicles (COM 2000 (314)) aims to alter this. Specific emissions from aircraft, trains and ships Between 1976 and 1988, specific emissions of HC and CO for freight and passenger transport by air fell by 85 and 70 % respectively as more-efficient aircraft entered the fleet (Airline Handbook). Emissions of NO x increased in the same period by around 12 %, mainly due to higher engine temperatures required to increase fuel efficiency and reduce other emissions. However, emissions of NO x, CO and HC are expected to decrease significantly in 2010 compared with 1995 levels: by 5 15 % for NO x, 3 9 % for CO and 2 12 % for HC (Airline Handbook). These reduction potentials assume specific fuel consumption and emission technology improvements, aircraft efficiency improvement and specific future fleet compositions. Data from Austria (EEA, 2000) shows a dramatic reduction in NO x, NMVOC and PM emissions per passenger-kilometre for heavy rail during the period 1950 80, mainly because of electrification and the use of hydropower. Specific emissions from trains depend critically on the technical level and the method of energy production used. Hence, there can be significant differences in specific emissions from trains in different countries. Specific emission data from ships are still poor and mostly reported as fuel- or energy-specific emission factors, in units of kilograms per tonne of fuel or grams per kilowatt hour, respectively (EEA, 2000). These data cannot be compared with data from other modes. In the future, emission reductions are expected from waterborne transport, mainly as a result of improved fuel quality and engine technology. EU legislation setting more stringent limits for sulphur content in fuel oils will greatly contribute to the expected emission reductions. To this end, the European Commission is developing a Community strategy to reduce air pollution from seagoing ships (expected in the summer of 2002) ( 4 ). ( 1 ) OJ L 214, 6.8.1988, p. 1. ( 2 ) OJ L 295, 25.10.1991, pp. 1 19. ( 3 ) The average occupancy of buses/coaches decreased by 7 % between 1990 and 1999, while the load factors of trucks increased by around 5 % between 1990 and 1998 (last year for which data is available data refers to EU-4 only and might not be valid for the entire EU). ( 4 ) See also http://www.europa.eu.int/comm/environment/air/future_transport.htm 17-12-2002 3

Sub-indicator: Comparison of specific emissions for different modes of transport The specific emissions of Euro II gasoline passenger cars are generally lower than those of other modes. Emissions of particulates from diesel cars, despite significant improvements, are high compared with other modes, especially when driving in urban areas. Shipping is the least polluting mode for freight transport, except for specific SO 2 emissions. Setting higher quality standards for bunker fuels can significantly improve the environmental performance of the shipping sector. Diesel freight trains are almost as polluting as Euro II diesel trucks. The environmental performance of diesel trains needs to be improved dramatically before shifting freight transport from road to rail will have an environmental benefit (in terms of emissions). Figure 2: Specific NO x emissions in passenger transport for different modes (mid- 1990s) 10.0 gramme NOx emissions per passenger-km 1.0 0.1 0.18 0.18 0.03 0.03 0.90 0.18 0.54 0.11 1.50 0.36 1.28 1.28 0.21 0.39 0.10 0.05 0.33 0.16 0.42 0.25 0.18 0.11 1.52 0.95 0.63 0.32 0.0 Urban car, petrol (w/o cold start) Highway car petrol Urban car, diesel (w/o cold start) Highway car diesel Urban train diesel Regional train diesel High speed train diesel Urban motorcycles (250-750 cm3) Highway motorcycles (250-750 cm3) Urban buses Highway buses Aviation short haul Aviation long haul NB: Range of considered occupancy rates: cars: 1 5 passengers per vehicle; motorcycles: 1 2 passengers per vehicle; coaches: 20 40 passengers per vehicle; trains: 50 100 % loaded; aircraft: 65 100 % loaded. Aviation: long haul refers to 125 2 500 nautical miles (232 4 630 km), short haul refers to a maximum of 125 nautical miles (232 km). Source: Eurostat, 2000. 17-12-2002 4

Figure 3: Specific NO x emissions in freight transport for different modes (mid-1990s) 100.0 gramme NOx emissions per passenger-km 10.0 1.0 0.22 0.40 1.36 0.19 0.20 0.88 0.31 1.15 0.24 18.22 8.48 2.35 0.47 0.1 0.11 Bulk carrier Container ship Diesel freight train Rural truck EURO II Highway truck EURO II Aviation short haul Aviation long haul NB: Range of considered load factors: bulk carrier: 50 100 % loaded; container ship: 65 100 % loaded; diesel freight train: 65 100 %; trucks: 50 100 % loaded; aircraft: 65 100 % loaded. Aviation: long haul refers to 125 2 500 nautical miles (232 4 630 km), short haul refers to a maximum of 125 nautical miles (232 km). Source: Eurostat, 2000. Assessment of the sub-indicator Due to limited data availability and methodological difficulties, caution is needed in comparing specific emissions by mode. For example, the specific VOC emission of aircraft excludes ground emissions. Additionally, when specific emissions by mode are compared, the actual distance covered per mode should also be taken into account: distances by aircraft can be much smaller than distances by ship. Even though the actual emission per kilometre for aircraft is higher, if the distance between origin and destination by air is much smaller than by ship, the environmental impact of the trip by air might come close to that of ships. Also, the location of emissions is an important aspect, as emissions in sensitive or urban areas will affect people and nature more than emissions in remote and non-sensitive areas. Figure 2 clearly shows the influence of occupancy rates and driving circumstances (in urban areas or on highways) on specific emission. A fully loaded urban diesel train still emits almost twice as much NO x per passenger-kilometre than a fully occupied Euro II diesel passenger car driving in an urban area. With only one passenger in the car, the specific emissions will, however, increase almost fivefold. When cold-start excessive emissions are also included, diesel trains may become cleaner in terms of NO x emissions per passenger vehicle ( 5 ). With respect to particulate emissions (not shown in the graphs, but available in Table 1 and Table 2), buses and coaches (both in urban and inter-urban situations) generally emit the least PM per passenger-kilometre. Euro II diesel passenger cars, particularly in urban areas, emit most PM per passenger-kilometre (with low occupancy). Diesel car emissions are considerably reduced in Euro III and will be further reduced with Euro IV. This, together with the introduction of particulate traps/filters and gasoline direct injection engines (higher PM emissions than current engines), means that a shift to diesel cars in the future is likely to have only a small effect on overall PM emissions, which are in any case dominated by emissions from commercial ( 5 ) No statistics are available for the average seat occupancy in trains. Cars carry on average 1.6 persons (see TERM 2002 29 EU Occupancy rates). 17-12-2002 5

vehicles. Hence, increasing dieselisation of the passenger car fleet will have a negative effect on air quality, but this effect will be small. Specific VOC and CO emissions are by far highest from motorcycles. The adopted directive on emission standards for powered two- and three-wheel vehicles aims to reduce these emissions and those of other pollutants. Aviation is by far the most polluting freight transport mode (see Figure 3), except for specific PM emissions. The high specific NO x emissions from aircraft are due mainly to the higher engine temperatures needed to improve fuel efficiency and reduce emissions of other pollutants. Road freight transport emits almost equal quantities of NO x as diesel freight trains. The introduction of PM traps in trucks will, to a large extent, even further reduce specific PM emission. The specific emissions of other pollutants (CO, VOC and SO 2 ) show more or less the same picture, i.e. Euro II trucks can to a large extent compete with diesel trains in terms of emissions per tonne-kilometre. Shifting freight transport to rail (one of the pillars of the CTP to reduce congestion) without improving the environmental performance of (diesel) trains might therefore have an adverse effect on transport emissions. Shipping is by far the cleanest mode of transport, except for specific SO 2 emissions, which are highest for shipping (and aviation), mainly because of the high sulphur content of bunker fuels. References Airline Handbook, web site of the Air Transport Association (http://www.airtransport.org/public/publications/display1.asp?nid=961). EEA, 2000, Are we moving in the right direction? Indicators on transport and environment integration in the EU, TERM 2000, European Environment Agency, Copenhagen, Denmark, February 2000. EEA ETC/ACC, 2002, National and central estimates for air emissions from road transport, Technical report No 74, European Environment Agency (EEA), European Topic Centre on Air and Climate Change (ETC/ACC), Copenhagen, Denmark, 2002 (http://reports.eea.eu.int/technical_report_2002_74/en/technical%20report%2074%20high%20f or%20the%20www.pdf). European Commission, 1999, Air transport and the environment Towards meeting the challenges of sustainable development, COM(1999) 640 final, Commission of the European Communities, Office for Official Publications of the European Communities, Luxembourg, 1999. Eurostat, 2000, Transport and environment: Statistics for the transport and environment reporting mechanism (TERM) for the European Union Data 1980 1998, European Commission in cooperation with Eurostat, Luxembourg, 2000. Eurostat, 2001, Transport and environment: Statistics for the transport and environment reporting mechanism (TERM) for the European Union Data 1980 1999, 19 December 2000 version, European Commission in cooperation with Eurostat, Luxembourg, 2001. LAT/TÜV/KTI, 1999, Study on transport-related parameters of the European road vehicle stock Final report, Laboratory of Applied Thermodynamics (LAT), Thessaloniki, Greece; Institut für Umweltschutz und Energietechnik (TÜV), Cologne, Germany; Institute of Transport Sciences Ltd (KTI), Budapest, Hungary; Thessaloniki, December 1999. 17-12-2002 6

Data Table 1: Estimated PM emissions per passenger-kilometre Unit: grams per passenger-km Non-urban High Low Diesel regional train, 50 100 % loaded 0.054 0.042 Diesel high-speed train, 50 100 % loaded 0.032 0.036 Euro II petrol PC, 1 5 passengers per vehicle Euro II diesel PC, 1 5 passengers per vehicle 0.043 0.009 Motorcycles (250 750 cm 3 ), 1 2 passengers per vehicle Euro II coaches, 20 40 passengers per vehicle 0.035 0.018 Urban Diesel urban train, 50 100 % loaded 0.000 0.000 Diesel regional train, 50 100 % loaded 0.000 0.000 Euro II petrol PC, 1 5 passengers per vehicle (w/o cold start) - - Euro II diesel PC, 1 5 passengers per vehicle (w/o cold start) 0.109 0.109 Motorcycles (250 750 cm 3 ), 1 2 passengers per vehicle - - Euro II urban buses, 20 40 passengers per vehicle 0.000 0.000 Source: Eurostat, 2000. Table 2: Estimated PM emissions per tonne-kilometre Unit: grams per tonne-km High Low Bulk carrier, 100 % loaded 0.002 0.002 Bulk carrier, 50 % loaded 0.004 0.004 Container ship, 100 % loaded 0.005 0.004 Container ship, 65 % loaded 0.007 0.005 Diesel freight train, 100 % loaded 0.050 0.011 Diesel freight train, 65 % loaded 0.077 0.017 Euro II road HGV, highway driving, 100 % loaded 0.030 0.005 Euro II road HGV, highway driving, 50 % loaded 0.056 0.010 Euro II road HGV, rural driving, 100 % loaded 0.042 0.008 Euro II road HGV, rural driving, 50 % loaded 0.078 0.014 Source: Eurostat, 2000. File: TERM 2001 28 EU Specific emissions.xls Metadata Technical information 1. Data source: specific NO x, VOC, PM and CO emissions are from EEA ETC/ACC, 2002; estimated emissions are extracted from Eurostat, 2000; emissions from aviation from Airline Handbook; passenger-km and tonne-km statistics (1980 98) are from Eurostat, 2001. Passenger-km and tonne-km projections based on ForeMove (EEA ETC/ACC, 2002), which is consistent with Primes (European Commission, 1999). 17-12-2002 7

2. Description of data: emission of NO x, PM, CO and VOC per passenger-km or tonne-km: original data (from LAT/TÜV/KTI, 1999) can be downloaded from http://vergina.eng.auth.gr/mech/lat_data.exe. The relevant data has been copied to ForeMove_Basedata.xls and complemented with data from Eurostat, 2001. The estimated emission data is extracted from Eurostat, 2000. The data presented is based on the work done for developing the Trends database, since it covers all 15 Member States while Auto-Oil covers only nine. Furthermore, the Trends database is more disaggregated than Auto-Oil, allowing the analysis to focus on passenger cars, buses and coaches, and heavy- and light-duty trucks. The Trends data are to a large extent consistent with the Auto-Oil study findings. The existing ForeMove model is used to assess the possible future evolution of the basic transport parameters. ForeMove was developed for the European Commission and has also been applied in the Auto Oil I and II programmes (LAT/TÜV/KTI, 1999). ForeMove has been superseded by Trends, and will be further developed as part of Trends. 3. Geographical coverage: EU-15 (Belgium, Denmark, Germany, Greece, Spain, France, Ireland, Italy, Luxembourg, the Netherlands, Austria, Portugal, Finland, Sweden and the United Kingdom). 4. Temporal coverage: 1981 98 and projections for 2000, 2005 and 2010. 5. Methodology and frequency of data collection: unknown. 6. Methodology of data manipulation, including making early estimates : Total tonnes divided by total passenger-kilometres or tonne-kilometres for the corresponding mode. Light- and heavy-duty vehicles are summed. Quality information 7. Strength and weakness (at data level): full data series for all countries (strong); modelled rather than measured, based on estimates of the vehicle fleet composition and on emissions from different vehicle classes (weak). 8. Reliability, accuracy, robustness, uncertainty (at data level): data can be considered fairly reliable, but it must be kept in mind that, for projections, the assumption is that vehicles will gradually comply with the highest emission standards. 9. Overall scoring (give 1 to 3 points: 1 = no major problems, 3 = major reservations): 2 Relevancy: 1 Accuracy: 3 (The data is modelled rather than measured, and based on different studies and measurements of different types of vehicles. Errors in these studies and measurements will greatly influence the calculation of specific emissions.) Comparability over time: 3 (The used model should be updated on a regular basis to generate time series (model runs) that are comparable, but based on the most recent emission factors the data resulting from one model run are not considered to be comparable over time as these represent one set of assumptions only.) Comparability over space: 1 (Same input data used for all Member States.) Further work required More work is needed to provide data at EU level. Eurostat and the Energy and Transport DG are jointly developing a database system (Trends) that links transport and other data with methodologies for estimating emissions and other environmental pressures. An important aim is to produce a consistent set of estimates to be used for EU policy purposes, including TERM. Both absolute and specific emissions are calculated. Trends enables the effects of specific policy measures on emissions and other environmental pressures to be monitored (EEA, 2000). This project and a number of research projects under the Commission s transport RTD programme (in particular the Artemis project) are expected to develop an improved emission model for road transport, based on a harmonised methodology for national and regional inventories and forecasts. Artemis will also include emission estimates for aircraft, motorcycles and probably even ships. Most resources are, however, still allocated to road. 17-12-2002 8

To remove the effect of different distances per mode, specific emissions should be based on a standardised theoretical distance rather than the actual one. An indicator on primary emission intensities would provide a better basis for comparing modes. This would require a life-cycle analysis to take account of energy used and emissions generated by the production of electricity and fuels, and by the production and disposal of vehicles. This would, however, require extensive methodological development and data collection. 17-12-2002 9