| Front page | | Contents | | Previous | | Next |

OMIT – Manual for environmental calculation of international freight transport

Enclosure B

Enclosure B.1
Energy consumption and pollutant emissions from rail freight transport

This chapter gives the details of the approach and data sources for the consumption and emission figures for rail transport in Europe. This includes the emission factors from electricity production in different countries, the allocation rules between the good under consideration and the whole train and the standardised distances used.

7.1 General method

Most international rail freight transport to and from Denmark is confined to a few main lines. The attempt is to capture these typical and dominant transports while staying simple and user friendly. Therefore a limited number of corridors for rail transport is defined. Data apply for these conditions in particular. For destinations off these main corridors the user can use a delivery train recommended not to be used for more than 100 kilometres transport distance.

As trains differ a lot in size, weight of cargo and average utilisation it is recommended to get specific data from the transport operator, not the least when corridors outside the main lines are used.

The European main line corridors for rail transport to and from Denmark are specified in table 1 based on data from Kombi Dan and DSB Cargo division - 2000.

Table 1.
Final destinations for main lines from DK. main stations in between are not mentioned.

International and long distance rail transport is carried out almost exclusively by electric traction. Therefore diesel traction is treated here only as an option for delivery to and from the main lines and consequently not treated in the same detail.

Furthermore international rail transport mostly goes in block trains [KombiDan 2000]. Hence data are typical for this train configuration.

The user is not expected to know about the rail transport operation in detail. He/she is only asked to specify the transport distance along a given corridor and the transported mass. All other parameters for the calculation are given as default values.

There are two steps needed in order to determine the energy consumption and emissions from electric trains:

1. Determine the consumption of electricity of the train.
2. Determine the conversion efficiency and pollutant emissions for electricity production at power stations.

The multiplication of both values gives the respective consumption of energy and emissions from train transport¹.

Each step is considered on its own in the following.

7.2 Factors influencing the energy consumption of trains

Numerous factors influence the energy consumption of a train. Most obvious are the transport distance and the total train mass. To abstract from these only specific values, i.e. per kilometre travelled and per gross hauled mass, are discussed in the following. Further factors for freight trains are [e.g. Anderson 2000, Brunner & Pelli 1998, Meyer et al. 2000, Schwannhäusser et al. 1986, EuroTC 1997]

	Aerodynamic resistance, depending on cross sectional shape, length and body resistance and the velocity by the square,
	Rolling resistance, depending on the bearing resistance and the mass,
	Inertia, depending on mass and acceleration,
	Topology and route characteristics, in particular inclinations, tunnels and bends,
	Driving characteristics, i.e. speed and acceleration,
	Electric equipment and efficiency of the locomotive, including recuperative brakes,
	Auxiliary energy consumption, e.g. for cooling, lighting, etc.
	Weather conditions, e.g. wind and outside temperature.

For the application here a number of factors however are fixed and not influenced by the transport under consideration:

	We consider given corridors, therefore topology and route characteristics are fixed.
	The trains usually run on schedule, hence the driving characteristics are predetermined;
	We consider annual averages: Therefore we assume typical values for the train configuration, determining aerodynamic and rolling resistance, for the locomotives with their electrical and mechanical characteristics, for weather conditions etc.

In consequence when the train’s configuration and operation as well as its route are given the total mass - thus indirectly also the train length - remain as parameters determining the specific energy consumption.

Empirical data show that the specific energy consumption of trains becomes less with increasing gross hauled mass, i.e. that energy consumption increases much less as the weight increases. The main task is to determine the form of this dependence.

7.3 Specific electricity consumption of trains

Existing data for the specific energy consumption of trains are either representative for a whole network, but not differentiated or differentiated for trains but not representative for the whole system. Therefore we combine the findings from both approaches.

Representative data are derived from top down values, typically the total electricity consumption by trains in a period divided by the transport volume in that period. Values derived this way are averaged over all driving conditions, train and locomotive types, configurations, routes etc in that period, exactly as needed for our task. They are taken as reference points for the absolute value. The absolute values differ from railway to railway. For the purpose here it is sufficient to differentiate between flatland networks, e.g. Denmark, midland networks, e.g. Germany, France, Italy, and mountainous networks as Switzerland and Austria.

The accuracy of the data is however considerably hampered by two circumstances: There is no physical way to differentiate the electricity consumption of freight trains from passenger trains when taken from the same electric net. Hence the allocation of the respective amounts to freight and passenger transport is up to the discretion of the railways. Second, the transport volume is usually not known exactly but is a nominal figure taken from freight papers or bills. To what extent these so-called tariff ton-kilometres coincide with the actual ton-kilometres performed, e.g. due to a different loading or diversions of the line, is not known. The related uncertainty of the top down values is estimated to be at least ±30 percent.

Differentiated data, e.g. from dynamical models, are taken to determine the shape of the functional dependence of the specific energy consumption on the gross hauled mass. A functional dependence in the form

q = 675 * M^-0,5          M in Gt, q in Wh/Gt*km

could be derived for block trains from [Schwanhäusser et al. 1983], which is compatible with the functional dependence derived by DSB [TEMA 2000] and with empirical data from DB, (fig. 1).

Look here!

Fig. 1:
Average specific energy consumption of freight trains depending on gross-hauled mass. Formulas used here.

The same functional dependence of the specific energy consumption on the gross hauled mass is assumed for all networks. The absolute value is decreased for flat countries and increased for Switzerland and Austria, i.e. for Alpine-crossing transport (table 2). Modern traction technology and recuperative brakes are positively taken into account. Yet in 2000 this modern technology is not in place for all lines, not all energy recuperated on the locomotive can be used again (and will partly disseminate therefore) and heavy freight trains must use conventional brakes as well, therefore always loosing part of the kinetic energy [Meyer & Aeberhard 1999, Meyer et al. 2000].

Table 2:
Specific electricity consumption of electric freight trains, averaged over the lines in the different countries (Gt: Gross hauled mass = weight on the hook).

7.4 Emission factors for electricity production

The specific electricity consumption is multiplied with the energy efficiency of the electricity chain and the emission factors from the power plant to give the total emissions.

The emission factors and the efficiency of conversion for the electricity production (in g/kWh primary energy of fossil input fuels) are derived from a standard inventory [Frischknecht et al. 1996] (France, Italy and the UCTPE-Mix) and by IFEU [Knörr et al. 2000] (for Germany). For Austria and Switzerland additional information is used [ÖBB 1998, SBB 1998].

As result the data for France, Italy and the UCTPE-Mix are based on the public electricity production; for Austria the electricity mix based on the public electricity production according to [Frischknecht et al. 1996] and in addition 30% electricity produced by hydro power from ÖBB-power plants [ÖBB 1998].The Swiss railways use only hydro power for electricity production [SBB 2001]. For Germany the electricity mix of the German Railways (DB AG) for 1998 from the TREMOD-model is used [Knörr et al. 2000].

The following tables (3-4) show the share of input energies, the average efficiency factor for fossil power plants and the emission factors which are used in the model.

Table 3.
Share of primary input energies and overall efficiency factor for fossil power plants (weighted average over technologies)

Table 4.
Emission factors for fossil power plants (weighted average over technologies)

All countries important for Danish rail freight transport are thus treated with a country specific energy mix. The UCPTE mix is applied for all other countries, which is a simplification justified by the small transport volume.

In OMIT cogeneration of heat and power are taken into the calculations. specifically for Denmark and Norway, the latter importing 20 % of the power from Denmark [Andersen 2001]. The efficiency factor for heat for Danish fossil fuel is 32 %, thus reaching an overall efficiency of 69 % for heat and power.

The allocation of primary energy consumption and emissions on power and heat can be done by two different methods.

	The model recommended by the Danish Energy Agency, and default in the model is the so called 200 % method. This allocation model takes the quality of the produced energy into consideration and therefore uses an efficiency factor for heat of 200 %.
	The other model that can be used for allocating primary energy and emissions in OMIT on power and heat is the energy content method, the allocation is then based only on the energy content produced.

As seen from the Danish example with efficiencies of 37 % and 32 %, using cogeneration and employing the latter method for allocation of primary energy consumption and emissions, almost halves the environmental load from train transport.

7.5 Data for delivery trains (diesel)

To account for rail transport off the main lines the user can calculate a diesel driven train, typical for marginal lines. Usually they have much less capacity utilisation, frequent stops, older stock and higher aerodynamic resistance. All these factors lead to a higher specific energy consumption. On the other hand the speed is lower, reducing the specific consumption. Reliable and representative data on this transport form do not exist. The values here are not differentiated by country or network due to the low share and thus low importance of diesel trains in long distance transport.

The same functional dependence as for electric traction (for secondary lines) is taken and has to be divided by the efficiency of the diesel-electric conversion of about 37% [Feihl, 1997]:

q = 675810/0,37* M^-0,5 M in Gt, q in Wh/Gt*km

The emission factors for diesel trains are linked to the fuel consumption. The following values are taken [Borken et al. 1999, TEMA 2000]:

Table 5.
Diesel emission factors g/kWh.

Because the system boundary here had to be chosen equal to TEMA 2000’s boundary the emission factors and conversion efficiencies of the refinery are disregarded.

7.6 Standardised distances for international rail freight transport

The distance data are taken from the standardised international rail freight distance tables DIUM for each country [DIUM 2000]. These data are generally used by transport operators. Moreover these standardised rail distances are also the basis for the top down values of transport volume and hence correspond well to these values. These data were modified such that the distance AC equals the sum of the distances AB and BC (assuming B is a location on the standard track between A and C).

Literature

[Andersen 2001] Andersen, O: personal communication, Vestforsk 29/3-2001.

[Andersson 2000] Andersson, E.: Improved energy efficiency in future rail traffic. In: UIC Railway Energy Efficiency Conference, Paris 10/11 May 2000.

[Borken et al. 1999] Borken et al. : Basisdaten für ökologische Bilanzierungen. Einsatz von Nutzfahrzeugen in Transport, Landwirtschaft und Bergbau. Vieweg Verlag, Wiesbaden/Braunschweig 1999.

[Brunner & Pelli 1998] Brunner, C.U. and T. Pelli: Simulationsprogramm für den Energieverbrauch von Reisezügen. Elektrische Bahnen 96 (1998) 11, 346-354.

[DIUM 2000] Distancier international uniforme marchandise. Available at http://www.railcargo.at/, ->Kundenservice ->Serviceleistungen ->DIUM.

[EuroTC 1997] Euro Transport Consult: Energy Saving potential in rolling stock and train operation. Commissioned by UIC, Utrecht 1997.

[Feihl 1997] Feihl, J.: Die Diesel-Lokomotive; Technik, Aufbau, Auslegung. Transpress, Stuttgart 1997.

[Frischknecht et al. 1996] Frischknecht, R. et al: (ETHZ): ECOINVENT - Ökoinventare für Energiesysteme; commissioned by Bundesamt für Energiewirtschaft and Nationaler Energie-Forschungs-Fonds; Bern 1996

[Knörr et al. 2000] Knörr, W. et al. (IFEU) : TREMOD - Transport Emission Estimation Model ; commissioned by the Environmental Protection Agency ; Berlin 2000

[KombiDan and DSB Cargo 2000] note on main corridors, cargo weight and balance.

[Meyer & Aeberhard 1999] Meyer, M. and M. Aeberhard: Vom Gratisstrom zur Energiesparlokomotive – Energieverbrauch bei elektrischen Bahnen, Eisenbahn-Revue 1-2/1999, 28-39.

[Meyer et al. 2000] Meyer, M. et al. : Einfluss der Fahrweise und der Betriebssituation auf den Energieverbrauch von Reisezügen, Eisenbahn-Revue 8-9/2000, 360-366.

[ÖBB 1998] ÖBB: Environmental Report 1998

[SBB 1998] Geschäftsbericht 1998 der Schweizer Bundesbahnen, Bern 1998.

[SBB 2001] personal communication H. Kuppelwieser, SBB Environment Center, 2001.

[Schwannhäusser et al. 1986] Spezifischer Energieeinsatz im Verkehr. Ermittlung und Vergleich der spezifischen Energieverbräuche. Commissioned by Bundesminister für Verkehr, Aachen 1986.

Enclosure B.2

Information for the OMIT project about freight trains across the Danish borders

The information from extract "International trains (trains that have crossed the border in Padborg respectively Copenhagen) during the period 2.7.2000 to 23.1.2001."

The extract gives information about the maximum size at train has had in Denmark, and not the size it had when crossing the border.

Shipment weight is weight of freight (i.e. exclusive of weight of wagon) for container wagons the weight of the container is included.

Train

	50% of the wagons are found in trains consisting of more than 22 wagons.
	50% of the freight (assessed as weight of shipment) is transported on trains with a total weight on the hook (not including locomotive) of more than 1,037 tons.
	50% of the freight (assessed as shipment weight) is transported on trains with a total shipment weight of more thanover 550 tons.

Wagons